In the Company of Many Good Poets. Collected Papers of Franco Montanari: Vol. I: Ancient Scholarship 9783110772371, 9783110772180

Citation preview

In the Company of Many Good Poets. Collected Papers of Franco Montanari

In the Company of Many Good Poets. Collected Papers of Franco Montanari Vol. I: Ancient Scholarship Edited by Antonios Rengakos

ISBN 978-3-11-077218-0 e-ISBN (PDF) 978-3-11-077237-1 e-ISBN (EPUB) 978-3-11-077241-8 Library of Congress Control Number: 2023939389 Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2023 Walter de Gruyter GmbH, Berlin/Boston Cover image: © pictore / DigitalVision Vectors / Getty Images Printing and binding: CPI books GmbH, Leck www.degruyter.com

Foreword I finally gave in to Antonios Rengakos’ friendly insistence and agreed to publish my Collected Papers, of which he wanted to be the editor. I would like to thank him wholeheartedly for this further demonstration of a long friendship, which has gone hand in hand with close scientific collaboration for many years. I would also like to thank the publisher De Gruyter and all the staff who have followed this publication with competence and skill, in particular Serena Pirrotta and Carlo Vessella. In these two volumes, I have gathered together most of my articles published in different venues up to a couple of years ago. Among the few items excluded are the numerous papyri containing Homeric text, of which I have published the editio princeps over the years and which I did not feel it would be useful to reproduce here. The articles have not been updated, except for sporadic and limited additions and minor corrections on points of detail. They must therefore be read bearing in mind the time in which they were originally published. For this, the List of the Original Publication Venues, found in each of the two volumes, is indispensable. I left the University three years ago, with only one regret: not for the research, because this can be continued assiduously anyway (at least in our field), but for the relationship with the students and in general with the young people who are beginning their education and research and who are the greatest source of joy in our work. Thanks to Simone Fiori for his invaluable help in preparing these Kleine Schriften.

https://doi.org/10.1515/9783110772371-202

Preface The Collected Papers of Franco Montanari, his Kleine Schriften, represent a rather modest part of his very rich scholarly production over the last 50 years (1973– 2021).1 The first volume (62 chapters) is understandably larger, since it covers the field of Ancient Scholarship where Franco Montanari is unanimously regarded as one of the world's leading scholars. Ancient Scholarship is here present in its entire range: Montanari treats with vast erudition and path-breaking results its theoretical premises, the practical ways in which it was practiced by ancient grammarians, as well as its history from the earliest beginnings in the context of the Peripatos to late antiquity. Emphasis is laid on the edition and interpretation of the Homeric epics in the Hellenistic period, but equally important publications illustrate the ancient critics' treatment of other authors (Hesiod, Pindar, Aeschylus, historians, Alexandrian poets etc.). Readers will find among these chapters some truly epoch-making publications concerning the form and the content of the Alexandrian ekdoseis, especially those of the Homeric text by the three major Hellenistic grammarians (Zenodotus, Aristophanes of Byzantium, and Aristarchus), the so-called Mythographus Homericus, the ancient scholia on the Iliad and the Odyssey or the traces of ancient scholarship in the papyri. The second volume (42 chapters) includes Montanari’s publications on ancient authors treated from a variety of angles: interpretation, historical context, reception, but also text criticism or the papyrological tradition. Homer takes again pride of place — Montanari’s earliest Homeric essay on Karl Reihardt (1975) foreshadows in a magisterial way his future expertise in epic poetry; other papers deal with Pindar, ancient drama, historiography, Aristoteles, Lucian, various philological subjects, and include, last but not least, three warm portraits of fellow colleagues, Giuliana Lanata, Giovanni Pugliese Carratelli and Luigi Enrico Rossi. It is my privilege to thank the editorial team at De Gruyter, past and present, especially Serena Pirrotta, Marco Aquafredda, Carlo Vessella and Anne Hiller for their early and constant support of this project. Finally, to Franco goes my gratitude for 25 years of unfailing friendship and very fruitful cooperation. Thessaloniki, August 2023 Antonios Rengakos

 1 Fausto Montana sketches Montanari’s life and career in “Homo faber. Franco Montanari: the Scholar (and the Man)”, in: A. Rengakos, P. Finglass, and B. Zimmermann, More than Homer Knew – Studies on Homer and His Ancient Commentators, Berlin/Boston 2020, 1–6. https://doi.org/10.1515/9783110772371-203

Contents Forward  V Preface  VII Abbreviations  XV List of the Original Publication Venues  XVII List of Figures  XXI

Part I: Theory, Practice, and History of Philology 

Scholarship, Philology, and Grammar  3



Ancient Scholarship and Classical Studies  44



Ancient Scholarship Today  57



History of Ancient Greek Scholarship from the Beginnings to the End of the Byzantine Age: Introduction  65



Evoluzioni del coro e movimenti celesti  72



Citazione di un’opera di Aristonico in Strabone  84



Ekdosis alessandrina: Il libro e il testo  88



Correcting a Copy, Editing a Text: Alexandrian Ekdosis and Papyri  106



From Book to Edition: Philology in Ancient Greece  120



Ekdosis: A Product of the Ancient Scholarship  144



Aristarchus’ Conjectures (once) Again  175



Remarks on the Citations of Authors and Works in Ancient Scholarship  184



Scoliografia e teatro greco: qualche appunto  192

X  Contents 

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  205



Filologia ed erudizione antica  229



Glossario, parafrasi, ‘edizione commentata’ nei papiri  243



Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica  249

Part II: Homer 

Commentary on Odyssey 21: P.Fay. 312 (with M. Haslam)  267



Glossary to Iliad 2.201—18: P.Oxy. 3832  276



A Re-examination of PBerol. 13282: The Homeric Historiae Fabulares on Papyrus  280



The Mythographus Homericus  295



Ancora sul Mythographus Homericus (e l’Odissea)  323



Zenodotus, Aristarchus and the Ekdosis of Homer  339



A New Papyrus Commentary on the Iliad  360



Towards a New Critical Edition of the Scholia to the Iliad: A Specimen (with Fausto Montana, Davide Muratore, Lara Pagani)  383



Omero, Eubulo, i pesci e i chorizontes  402



Una glossa omerica pre-alessandrina? (Sch. Il. XVI 235 – Eubulo fr. 139 K.)  408



Aristarchus on Odyssey II 136–7: A Case Study in Ancient Homeric Philology  416

Contents  XI



On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401  430



The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing  440



P.Oxy. 574 verso riconsiderato: frammento di hypotheseis dell'Iliade  455



The Homerica on Papyrus: Notes Toward a Classification of Genres  459



Ancient Homeric Philology as Attested in the Papyri  473



Note sulla tradizione manoscritta degli scholia D in Iliadem: Un caso di errore da archetipo  482



Aristarchus in the D-Scholia to the Iliad  488



The Wrong Verses of Homer and Ancient Philology  494



Antichi commenti a Omero. Alcune riflessioni  510



Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis  524



Discussioni su «Olimpo» nella filologia omerica antica e il grammatico Leogoras  532



I percorsi della glossa: Traduzioni e tradizioni omeriche dall’antichità alla cultura bizantina  536



La filologia omerica antica e la storia del testo omerico  543



The Fragments of Hellenistic Scholarship  554



Die Berliner Wachstafeln P. 10508 – 10512: Scholia minora und grammatikalische Passagen (zusammen mit Davide Muratore und Fabian Reiter)  569

XII  Contents 

Homeric Papyrology: Themes, Problems, and Perspectives  582



Dialect Glossai in the Scholia to Homer  597



Penelope at the Symposium: Od. 1, 328–335 and Dicaearchus  616



Alexandrian Homeric Philology  621

Part III: Other Authors 

Ancient Scholarship on Hesiod  643



Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201  673



Citazioni di Pindaro negli scoli omerici  682



L’esegesi antica di Eschilo da Aristotele a Didimo  691



Alcuni passi di Eustazio e il Prometeo di Eschilo  730



Gli storici greci e la filologia di età ellenistico-romana  734



Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319  756



Demetrius of Phalerum on Literature  762



The Peripatos on Literature Interpretation, Use and Abuse  780



Un nuovo frammento di commentario a Callimaco  796



Callimaco e la filologia  805



Satiro in P.Oxy. 2506 (fr. 26 II)?  831



Un papiro di Eschine con correzioni (P.Oxy. 2404): Considerazioni sull’ekdosis alessandrina  834



Iperide e Cratino in due Glosse dell’Αntiatticista  845

Contents  XIII



Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries”  849

Bibliography 863 General Index 899 Index of Passages Discussed 905

Abbreviations CLGP CPF FGrHist LGGA

PMG PMGF SH SOD Supp. SH TrGF 1 TrGF 2 TrGF 3 TrGF 4 TrGF 5

Commentaria et Lexica Graeca in Papyris reperta, Berlin/Boston 2004– Corpus dei Papiri Filosofici greci e latini, Firenze 1989– Jacoby, F. Die Fragmente der griechischen Historiker, Berlin/Leiden 1923– Lessico dei Grammatici Greci Antichi (http://www.aristarchus.unige.net/lgga) = Lexicon of Greek Grammarians of Antiquity (https://referenceworks.brillonline. com/browse/lexicon-of-greek-grammarians-of-antiquity). Page, D.L. 1962. Poetae Melici Graeci, Oxford. Davies, M. 1991. Poetarum melicorum Graecorum fragmenta. Vol. 1, Alcman, Stesichorus, Ibycus, Oxford. Lloyd-Jones, H./Parsons, P. (eds.) 1983. Supplementum Hellenisticum, Berlin/ New York. Stork, P./Ophuijsen, J.M. von/Dorandi, T. (eds.) 2000. ‘Demetrius of Phalerum: The Sources, Text and Translation’, in: Fortenbaugh/Schütrumpf 2000, 1–310. Lloyd-Jones, H. (ed.) 2005. Supplementum Supplementi Hellenistici, Berlin/New York. Snell, B./Kannicht, R. 1986. Tragicorum Graecorum Fragmenta, Vol. 1, Göttingen. Kannicht, R./Snell,B. 1981. Tragicorum Graecorum Fragmenta, Vol. 2, Göttingen. Radt, S. 1985. Tragicorum Graecorum Fragmenta, Vol. 3, Aeschylus, Göttingen. Radt, S. 21999. Tragicorum Graecorum Fragmenta, Vol. 4. Sophocles, Göttingen. Kannicht, R. 2004. Tragicorum Graecorum Fragmenta Vol. 5, Euripides, Göttingen.

https://doi.org/10.1515/9783110772371-205

List of the Original Publication Venues 1. 2.

3.

4.

5. 6.

7.

8.

9.

10.

11.

12. 13. 14. 15. 16.

17.

L’erudizione, la filologia, la grammatica, in: “Lo spazio letterario della Grecia antica”, vol. I, Tomo II, Roma, 1993, 235–281. [transl. Gregory Mellen] Ancient Scholarship and Classical Studies, in: Ancient Scholarship and Grammar. Archetypes, Concepts and Contexts, Edited by Stephanos Matthaios/Franco Montanari/ Antonios Rengakos, TCSV 8, Berlin/New York, 2011, 11–24. Afterword. Ancient Scholarship today, in: Approaches to Greek Poetry. Homer, Hesiod, Pindar, and Aeschylus in Ancient Exegesis, Edited by Marco Ercoles/Lara Pagani/ Filippomaria Pontani/Giuseppe Ucciardello, Berlin/Boston, 2018, 345–354. S. Matthaios/F. Montana/A. Novokhatko/Filippomaria Pontani, History of Ancient Greek Scholarship. From the Beginnings to the End of the Byzantine Age, Edited by Franco Montanari, Introduction by Franco Montanari, Leiden/Boston, 2020, 1–8. Evoluzioni del coro e movimenti celesti, in: “Scena e spettacolo nell’antichità. Atti del Convegno Internazionale di Studio, Trento, 28–30 marzo 1988”, Firenze 1989, 149–163. Due note esegetiche. 1. Selene e Armonia in Nonno, Dionisiache IV 196; 2. Citazione di un’opera di Aristonico in Strabone, in: Synodia. Studia humanitatis A. Garzya septuag. dicata, Napoli, 1997, 687–694. Ekdosis alessandrina: il libro e il testo, in: M. Sanz Morales/M. Librán Moreno, eds., Verae Lectiones. Estudios de Crítica Textual y Edición de Textos Griegos, Cáceres/Huelva, 2009, 143–167, tavv. 1–8 385–392. Correcting a Copy, Editing a Text. Alexandrian Ekdosis and the Papyri, in: From Scholars to Scholia. Chapters in the History of Ancient Greek Scholarship, Edited by Franco Montanari/ Lara Pagani, TCSV 9, Berlin/New York, 2011, 1–15. From Book to Edition. Philology in Ancient Greece, in: World Philology, Edited by Sheldon Pollock, Benjamin A. Elman, and Ku-ming Kevin Chang, Cambridge Mass./London 2015, 48–97. Ekdosis. A Product of the Ancient Scholarship, in: Brill’s Companion to Ancient Greek Scholarship, Ed. by Franco Montanari, Stephanos Matthaios and Antonios Rengakos, Leiden/Boston 2015, 641–672. Aristarchus’ Conjectures (once) again, in: Lemmata. Beiträge zum Gedenken an Christos Theodoridis, hrsg. Von Maria Tziatzi/Margarethe Billerbeck/Franco Montanari/Kyriakos Tsantsanoglou, Berlin/Boston 2015, 119–129. Remarks on the citations of authors and works in ancient scholarship, Trends in Classics 8, 2016, 73–82. Scoliografia e teatro greco: qualche appunto, in: Dal teatro greco al teatro rinascimentale: momenti e linee di evoluzione, Trento 1992, 73–87. Alessandria, Cirene e Pergamo, in: “Lo spazio letterario della Grecia antica”, vol. I, Tomo II, Roma, 1993, 625–655. Filologia ed erudizione antica, in: Da AIΩN a Eikasmos. Atti della giornata di studio sulla figura e l’opera di Enzo Degani, Eikasmos – Studi 8, Bologna 2002, 73–88. Glossario, parafrasi, ‘edizione commentata’ nei papiri, in: I classici greci e i loro commentatori, a cura di G. Avezzù/P. Scattolin, Memorie della Accademia Roveretana degli Agiati, s. II vol. X, Rovereto 2006 (ma 2007), 9–16. Le tortuose strade del frammento. Citazioni d'autore nell'erudizione antica, in: Fragments in Context – Frammenti e dintorni, a cura di Virginia Mastellari, Göttingen 2021, 23–38.

https://doi.org/10.1515/9783110772371-206

XVIII  List of the Original Publication Venues 18. Commentary on Odyssey 21 (P. Fayum 312). BASP 20, 1983 (1985), 113–122. 19. 3832. Scholia minora to Iliad 2.201–18 etc, in: “The Oxyrhynchus Papyri. Vol. LVI, London 1989”, 48–51. 20. Revisione di P. Berol. 13282. Le Historiae fabulares omeriche su papiro, in: “Atti del XVII Congresso Intern. Di Papirologia. Napoli 1984”, Napoli 1985, vol. II, 229–242. [transl. Gregory Mellen] 21. The Mythographus Homericus, in Greek Literary Theory after Aristotle. A Collection of Papers in Honour of D. M. Schenkeveld, Amsterdam, 1995, 135–172. 22. Ancora sul Mythographus Homericus (e l’Odissea), in: La mythologie et l’Odyssée. Hommage à Gabriel Germain, Textes réunis par A. Hurst et F. Létoublon, Genève 2002, 129– 144. 23. Zenodotus, Aristarchus and the Ekdosis of Homer, in: Editing Texts / Texte edieren, ed. by G.W. Most, “Aporemata” B. 2, Göttingen, 1998, 1–21. 24. A New Papyrus Commentary on the Iliad, in: Fragments of the Past. Ancient Scholarship and Greek Papyri, Ed. by Franco Montanari/Serena Perrone, Trends in Classics 1.2, 2009, 177–182. 25. Franco Montanari/Fausto Montana/Davide Muratore/Lara Pagani, Towards a new critical edition of the scholia to the Iliad: a specimen, Trends in Classics 9, 2017, 1–21. 26. Omero, Eubulo, i pesci e i Chorizontes, SCO 25 (1976), 138–150. 27. Una glossa omerica pre-alessandrina? (Sch. Il. XVI 235 – Eubulo fr. 139 K.), “Rendiconti Istituto Lombardo – Classe di Lettere”, 110 (1976), 202–211. [transl. Gregory Mellen] 28. Aristarco ad Odissea II 136–7. Appunti di filologia omerica antica, MD – Materiali e discussioni per l’analisi dei testi classici 3 (1979), 157–170. [transl. Gregory Mellen] 29. L’hypothesis di Odyss. β in P. Strasb. Gr. 1401 e in P. Oxy. XLIV 3160, in: “Miscellanea Papyrologica”, Papyrologica Florentina VII, Firenze 1980, 273–279. [transl. Gregory Mellen] 30. Il grammatico Tolomeo Pindarione, i poemi omerici e la scrittura, in: “Ricerche di Filologia Classica I. Studi di letteratura greca”, Pisa 1981, 97–114. [transl. Sergio Knipe] 31. P. Oxy. 574 verso riconsiderato: frammento di hypotheseis dell’Iliade, ZPE 48, 1982, 89–92. 32. Gli Homerica su papiro: per una distinzione di generi, in: “Ricerche di Filologia Classica II. Filologia e critica letteraria della Grecità”, Pisa 1984, 125–138. [transl. Gregory Mellen] 33. Filologia omerica antica nei papiri, in: “Proceedings of the XVIII International Congress of Papyrology. Athens 25–31 May 1986”, Atene 1988, 337–344. [transl. Gregory Mellen] 34. Note sulla tradizione manoscritta degli Scholia D in Iliadem. Un caso di errore da archetipo, in: “Storia Poesia e Pensiero nel mondo antico. Studi in onore di Marcello Gigante”, Napoli, 1994, 475–481. 35. Aristarco negli Scholia D all’Iliade, in: Studia Classica Iohanni Tarditi oblata (Biblioteca di Aevum Antiquum 7), Milano 1995, 1151–1159. [transl. Gregory Mellen] 36. I versi ‘sbagliati’ di Omero e la filologia antica, in: Struttura e storia dell’esametro greco, a cura di M. Fantuzzi/R. Pretagostini, Roma 1995, 265–287. [transl. Sergio Knipe] 37. Antichi commenti a Omero. Alcune riflessioni, in: Omero. Gli aedi, i poemi, gli interpreti, a cura di Franco Montanari, Firenze, 1998, 1–17. 38. Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis, in: Poesia e religione in Grecia. Studi in onore di G. Aurelio Privitera, Napoli, 2000, 479–486. 39. Discussioni su «Olimpo» nella filologia omerica antica e il grammatico Leogoras, in: POIKILMA. Studi in onore di Michele R. Cataudella in occcasione del 60° compleanno, La Spezia, 2001, 937–941.

List of the Original Publication Venues  XIX

40. I percorsi della glossa. Traduzioni e tradizioni omeriche dall’antichità alla cultura bizantina, in: Erudizione scolastico-grammaticale a Bisanzio, a cura di Paola Volpe Cacciatore, Napoli 2003, 81–88. 41. La filologia omerica antica e la storia del testo omerico, in: Antike Literatur in neuer Deutung, Festschr. Für J. Latacz, hrsg. Von A. Bierl/A. Schmitt/A. Willi, München, 2004, 127– 143. 42. The Fragments of Hellenistic Scholarship, in: Collecting Fragments / Fragmente Sammeln, Ed. by G.W. Most, “Aporemata” B. 1, Göttingen, 1997, 273–288. 43. (con Davide Muratore e Fabian Reiter), Die Berliner Wachstafeln P. 10508–10512: Scholia Minora und grammatikalische Passagen, in: Actes du 26e Congrès international de papyrologie, Genève 16–21 août 2010, Ginevra 2012 549–557. 44. La papirologia omerica: temi, problemi, prospettive, in: I papιri omerici. Atti del Convegno Internazionale di Studi, Firenze 9–10 giugno 2011, a cura di Guido Bastianini e Angelo Casanova, Firenze 2012, 1–16. [transl. Sergio Knipe] 45. Glosse dialettali negli Scholia omerici, in: Homer, gedeutet durch ein großes Lexikon, hrsg. Von Michael Meier–Brügger, Abhandlungen der Akademie der Wissenschaften zu Göttingen, N.F. 21, Berlin/Boston, 2012, 123–139. [transl. Gregory Mellen] 46. Penelope al simposio. Od. 1, 328–335 e Dicearco, in: Mythologeîn. Mito e forme di discorso nel mondo antico. Studi in onore di Giovanni Cerri, a cura di Antonietta Gostoli e Roberto Velardi, Pisa/Roma 2014, 51–53. 47. Alexandrian Homeric Philology. The form of the ekdosis and the variae lectiones, in: Epea Pteroenta. Beiträge zur Homerforschung, Festschrift für Wolfgang Kullmann zum 75. Geburtstag, hrsg. Von M. Reichel/A. Rengakos, Stuttgart, 2002, 119–140. 48. Ancient Scholarship on Hesiod, in: Brill’s Companion to Hesiod, ed. by Franco Montanari/ Antonios Rengakos/Christos Tsagalis, Leiden/Boston 2009, 313–342. 49. Osservazioni sul commentario a Pindaro P.Oxy. LXXIX 5201, in: Ε me l’ovrare appaga. Papiri e saggi in onore di Gabriella Messeri (P.Messeri), a cura di Guido Bastianini/ Francesca Maltomini/Daniela Manetti/Diletta Minutoli/Rosario Pintaudi, Firenze 2020, 319–327. 50. Citazioni di Pindaro negli Scholia omerici, in: Il potere della parola. Studi di Letteratura Greca per Maria Cannatà Fera, a cura di G.B. D’Alessio/L. Lomiento/C. Meliadò/G. Ucciardello, Alessandria 2020, 279–287. 51. L’esegesi antica di Eschilo da Aristotele a Didimo, in: Eschyle à l’aube du théâtre occidental, Entretiens sur l’antiquité classique, tome LV. Entretiens préparés et présidés par Jacques Jouanna et Franco Montanari, Fondation Hardt, Vandoeuvres/Genève, 2009, 379–433. 52. Alcuni passi di Eustazio e il Prometeo di Eschilo, MD – Materiali e discussioni per l’analisi dei testi classici 5 (1980), 173–176. 53. Gli storici greci e la filologia di età ellenistico-romana, in: Le età della trasmissione. Alessandria, Roma, Bisanzio, Atti delle Giornate di Studio sulla storiografia greca frammentaria: Genova, 29–30 maggio 2012, Roma 2013, 1–32. 54. Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319, in: Studi offerti ad Alessandro Perutelli, Roma, 2008, 237–244. 55. Demetrius of Phalerum on Literature, in: Demetrius of Phalerum. Text, Translation and Discussion, Ed. by W.W. Fortenbaugh and E. Schütrumpf, “Rutgers Univ. Stud. In Class. Humanities” vol. IX, New Brunswick/London 2000, 391–411. 56. The Peripatos on Literature. Interpretation, Use and Abuse, in: Praxiphanes of Mytilene and Chamaeleon of Heraclea. Text, Translation, and Discussion, ed. By Andrea Martano/

XX  List of the Original Publication Venues

57. 58.

59. 60. 61. 62.

Elisabetta Matelli/David Mirhady, “Rutgers Univ. Stud. In Class. Humanities” vol. XVIII, New Brunswick/London 2012, 339–358. Un nuovo frammento di commentario a Callimaco, Athenaeum 54 (1976), 138–150. Callimaco e la filologia, in: Callimaque, Entretiens sur l’antiquité classique, tome XLVIII. Entretiens préparés et présidés par Franco Montanari et Luigi Lehnus, Fondation Hardt, Vandoeuvres/Genève, 2002, 59–97. Satiro in P. Oxy. 2506 (fr. 26 II)?, ZPE 62 (1986), 46–48. Un papiro di Eschine con correzioni (P.Oxy. 2404). Considerazioni sull’ekdosis alessandrina, APF 55, 2009 (Festschrift G. Poethke) 401–411. Iperide e Cratino in due glosse dell’Antiatticista, in: ΟΔΟΙ ΔΙΖΗΣΙΟΣ. Le vie della ricerca. Studi in onore di F. Adorno, Firenze, 1996, 357–360. Filologi alessandrini e poeti alessandrini. La filologia sui ‘contemporanei’, Aevum Antiquum 8, 1995 (Atti del Congresso Poeti e filologi, Filologi-poeti – Brescia 26–27 aprile 1995), 47–63. [transl. Sergio Knipe]

List of Figures Fig. 1: Fig. 2: Fig. 3: Fig. 4: Fig. 5: Fig. 6: Fig. 7: Fig. 8:

Timoteo, Persiani, P. Berol. inv. 9875 col. IV.  102 Timoteo, Persiani, P. Berol. inv. 9875 col. V.   102 Posidippo, P. Mil. Vogl. 309, col. XI.  102 Eschine, Contro Ctesifonte §§ 51–3, P. Oxy. 2404.  103 Eschilo, Diktyoulkoi, P. Oxy. 2161.  103 Commentario al Teeteto di Platone, P. Berol. inv. 9872.  104 Hypotheseis di varie tragedie di Eschilo, P. Oxy. 2256.  104 Vangelo di Giovanni, P. Bodmer 2.  105

https://doi.org/10.1515/9783110772371-207



Part I: Theory, Practice, and History of Philology

 Scholarship, Philology, and Grammar Philology, Grammar, and the Transmission and Interpretation of Texts: Scholarly Literature If we imagine a single bookcase containing all of ancient Greek literature, in addition to the canonical works of the great authors whom everyone knows and reads, there would also be a sizeable quantity of large volumes that were not consulted very often — not even by specialists in classical philology. In this section of our imaginary bookcase, one would find collections of scholia on poets and prose authors, lexica and etymologica both large and small, grammatical treatises of different lengths and degrees of thoroughness, compilations of proverbial sayings: in short, that vast, complex body of works that we may define as scholarly literature. Moreover, modern studies of the transmission of classical texts often speak of ancient variant readings, which are known to us from indirect testimony or even from papyrus fragments that pre-date the era of Byzantine manuscripts; and we learn that there existed ancient editions of classical texts, which incorporated scholarly emendations alongside readings carefully selected from among multiple variants, as well as ancient commentaries, which discussed the constitution of the text together with its correct interpretation.1 After epic, lyric, drama, historiography, and oratory, after philosophical and scientific writing, within the vast literary heritage of Greek civilization there emerged the disciplines of philology and grammar. At this moment in history, in certain particularly thriving cultural centers, scholars began to dedicate their attention to a specific set of problems with an unprecedented depth of analysis and intellectual rigor. For the first time, they framed — and attempted to resolve — the question of how to establish an exact and reliable text for the ancient works that had been passed down and continued to be regarded as the touchstones of thought and culture. They also set themselves the task of interpreting those same works as completely and as accurately as they could, and in the process they developed a range of highly refined exegetical methods and they cultivated an overall approach that can truly be called historical. In this context, the study of language — which previously had been the remit of rhetoric and  1 For the Hellenistic age, the fundamental work is Pfeiffer 1968; also still useful (for the Imperial period it remains authoritative), Sandys 1920; for Late Antiquity and the Byzantine era, Wilson 1990. Briefer surveys may be found in: Reynolds/Wilson 1987; Fraser 1972, I, 447 ff.; Carlini 1977; La Penna 1985; Montanari 1987; Degani 1987; Kennedy 1989, 200 ff. Interesting observations on scholarly literature can be found in Tosi 1988. https://doi.org/10.1515/9783110772371-001

  Scholarship, Philology, and Grammar philosophy — was gradually expanded and rendered more precise, via the consideration of grammatical phenomena as facts in and of themselves, until finally we can say that the scientific study of grammar, with its rich explanatory power, was established once and for all. In this period, then, numerous new areas of scholarly research were developed; commentaries and critical editions were composed for the major works of the past; and the interpretation of texts assumed various innovative forms, resulting in the birth of entire new genres of literary production. We have already introduced the overall category of scholarly literature: this term covers a wide range of different texts and subject areas — from philology, to grammar, to lexicography, to paroemiography and biography, along with the production of scholarly inventories and antiquarian and doxographic compilations, not to mention various other kinds of anthologies and miscellanies.2 Taken as a whole, this genre of literature constitutes an important chapter in the history of Greek culture in the Hellenistic and Imperial periods. It is not difficult to see how all these different strands of scholarly activity belong to a single cultural and intellectual current, whose object was to study the great authors of the past and to interpret their works in every possible light and with every possible tool, since only in this way was it possible to fully grasp the culture of which those authors were held to be the authentic depositaries. For a long time, however, all of these various works of scholarly literature were pillaged by modern academics in search of information about ancient authors or fragments of lost works, rather than being interpreted as autonomous cultural products, each of which represents one aspect of a larger intellectual trend in the context of a specific historical moment: a historical moment that was tenaciously dedicated to studying, collecting, summarizing, and conserving the treasures of the past — a glorious intellectual heritage accumulated over the course of many centuries — in order to define and to fix a cultural identity, so that this identity could live on among future generations. Biography, too, belongs to this category of scholarly activity — at least insofar as it is concerned with reconstructing the lives of poets and other writers by gathering and interpreting the relevant information contained in their works. Of

 2 As for example the Deipnosophistai (“The Sophists at the Banquet”) of Athenaeus of Naucratis (2nd–3rd cent. CE), an invaluable miscellany containing different kinds of literary, grammatical, and antiquarian material; Athenaeus cites over a thousand authors and titles of works, and includes more than 10,000 lines of poetry. Another example is the miscellany (Varia Historia) of Claudius Aelianus (2nd–3rd cent. CE); a similar work by Favorinus of Arelate (2nd cent. CE) is now lost.

Scholarship, Philology, and Grammar  

course, biography was cultivated in other contexts and developed in other directions as well: above all, there was political biography, like the Parallel Lives of Plutarch, dedicated to the lives and works of famous statesmen; and there were also collections of biographies of men famous in other disciplines, such as the Lives of the Sophists by Philostratus or the Lives of the Philosophers by Diogenes Laertius. That being said, in the Hellenistic period this genre saw an important development precisely in the composition of biographies that discussed the personalities of poets and other literary figures, and these biographies were intimately related to the exegetical efforts dedicated to explaining the works of these authors. For instance, the interpretation of specific pieces of information included in literary works often allowed scholars to reconstruct — of course, with a certain degree of improvisatory deduction — specific moments in the life of the author (examples of this are found even in the scholia and in the fragments of ancient commentaries found on papyrus);3 vice versa, the author’s life was considered a crucial component of the interpretative toolkit that could be employed in order to understand a given work. In this way, biography, hermeneutics, and literary criticism appear to have been closely linked; biography thus had an entirely organic role within the overall category of scholarly literature.4 In this space we can only cast a quick glance at rhetoric, even if it too played a significant part within this field of cultural and intellectual activity. Indeed, the principal strand of Greek rhetorical studies, from Aristotle down to the beginning of the Imperial period, was characterized by constant reference to the texts of named authors and by a form of stylistic analysis that was firmly based upon philological and historical knowledge. In other words, it was nourished upon intense hermeneutic activity; one might even say that its lifeblood consisted precisely in the interpretation of literary works. In this connection, one thinks, for example, of the heated debates concerning stylistic models; of Asianism and Atticism with its various trends; of figures like Dionysius of Halicarnassus; or of works like the treatise On the Sublime. It is only later, in Hermogenes, that rhetoric is untethered from concrete examples taken from the texts  3 Excellent examples are extant for lyric: for instance, in the scholia we find reconstructions of the polemics between Pindar and Bacchylides and of the relationships between the choral lyricists and their patrons, along with accounts of the composition of the odes and their performance contexts (see also the following footnote). In a commentary on Anacreon partially restored to us by P.Oxy. 3722 (2nd cent. CE), we find comments of an explicitly biographical nature (frr. 15 and 73); moreover, one could also consider ‘biographical’ the information that the commentary attempts to provide regarding the individuals who are mentioned or addressed in the poetry. 4 On these questions, cf. Arrighetti 1987, 139 ff., with bibliography at 254 ff.

  Scholarship, Philology, and Grammar of specific authors and thus comes to constitute a kind of abstract science, the purpose of which is to provide an all-encompassing system of rarefied stylistic ideals; before then, it was entirely normal to find even an entire ode of Sappho cited within a rhetorical treatise, alongside dozens of more or less extensive citations from a wide range of authors.5 It is useful at this point to mention a particular fact, the significance of which has only recently come to be fully appreciated.6 In the history of the transmission of Greek literature, the Alexandrian scholars of the Hellenistic period (roughly from the third to the first century BCE) were responsible for a first round of selection of texts, insofar as their preferences effectively rendered certain authors “canonical” (and others not); in this way, in addition to their contributions to philology and to the interpretation of texts, these scholars definitively influenced the choice of which works would be preserved and which would be lost. The study of material finds from the ancient world (i.e., papyrus fragments, the remains of books once in circulation) has shown that there was a second major round of selection, which occurred in the period from the end of the first to the beginning of the third century CE, with a large influence played by the tastes associated with the well-known cultural efflorescence of the second century. Now, if we examine extant papyrus fragments from a statistical point of view, we learn that the highest density of finds belongs to this latter period, and this testifies to a significant increase in the production and circulation of books; but if we turn to an examination of the texts themselves and of the interpretative strategies found in them, we can make two further observations. On the one hand, the philological approach to literature formulated by the Alexandrian scholars, with its scientific apparatus and its methodological premises, continued to play an influential role in determining how authors were read; on the other hand, there seems to have been a revival of interest in some works that had been overlooked in the Hellenistic period. This revival led to a broader selection of books in circulation and also had an influence on the formation of corpora and on the subsequent shape of the textual tradition. For instance, it was in this phase that definitive selections were made of the dramatic authors (both tragedy and comedy), and this proved decisive in determining which works would later be preserved in the Byzantine manuscript tradition. In the third century, however, this literary activity diminished, in keeping with the general crisis that affected cultural life in this period, and as a result many works were lost to us; some additional losses, however, were prevented  5 In general, cf. Luzzatto 1988, 220 ff., with bibliography. 6 Cavallo 1986; 1987. Cf. also Carlini 1990; Cavallo 1989a.

Scholarship, Philology, and Grammar  

by the literary revivals in the fourth century, thanks especially to the role of institutions like the Library of Constantinople and the various schools of Late Antiquity. To sum up: the major creative phase of philology and hermeneutics occurred above all in the Hellenistic era, that is to say, from the generation of Philitas and Zenodotus down to the death of Aristarchus (around 144 BCE). The work of these scholars was then continued by a few generations of disciples and successors, all the way to the Augustan age, which can be effectively represented by scholars such as Aristonicus and Theon and above all by the imposing figure of Didymus Chalcenterus of Alexandria, whose nickname (“BronzeGuts”) has become a symbol for the voracious attempt to collect and digest all the products of a vast tradition. After the second century CE, it is fair to say that this creative phase had definitively come to an end; but scholarship continued to play an important role for many subsequent centuries: scholars now dedicated their efforts especially to conserving, epitomizing, and recasting earlier works; they set themselves the task of collecting and condensing commentaries and various other scholarly works of the preceding generations. Thus, over the course of the Imperial era and the era of Late Antiquity, the foundations were laid for the preservation of ancient scholarship, which gradually began to assume the shape in which it would later arrive to us via the Byzantine miscellanea. The majority of the original works of scholarly literature, produced in the Hellenistic and Imperial periods, have been lost; in general, we now possess only compilations that were assembled much later, and from these compilations we can piece together fragments of information, which are at times rather substantial, at other times extremely meagre. In any case, these compilations are the source of much of our knowledge of ancient scholarly literature; in large part, we can glimpse the activity and the works of the earlier scholars only through the filter of the later compilations. Therefore, it seems necessary to begin our survey with an examination of these later genres of texts.

Editions, Commentaries, and Scholia The works of many Greek authors have come down to us in manuscripts that are equipped with ancient comments presented in the form of marginal notes, sometimes extremely brief, sometimes quite extensive: these notes are called scholia (Greek σχόλια, Latin scholia). The corpora of scholia pertaining to different authors vary significantly in the quality and the quantity of the information that they provide; their value depends in great part upon the ancient sources

  Scholarship, Philology, and Grammar from which they derive, and, as a rule, they present more reliable information when they are based upon the exegetical works of the scholars of the Hellenistic era. Generally speaking, the scholia are the principal source in which we can find the remnants of the interpretative efforts carried out by ancient grammatikoi in the various phases of scholarly literature outlined above; effectively, after several intermediate stages, the scholia have inherited the content of the exegetical works produced by the Alexandrian scholars. In their original forms — which assumed their typical contours precisely in the Hellenistic period — these works include: hypomnemata (commentaries that provide annotation sequentially, following the order of the text being studied); syngrammata (treatises on specific issues, generally in monographic form); and collections of lexeis (which would later constitute the basis for subsequent lexicography). Among these, perhaps one can say that the hypomnema (commentary) is the most characteristic and most significant form in which the Alexandrian exegetical efforts were produced. This genre of scholarly text established itself gradually, reaching its apex first in the copious production of Aristarchus and then in that of his students down to Didymus and Aristonicus; later it would become widely employed in the Imperial period, when rich miscellanea were often put together. However, in discussing the history of the commentary as a literary form, one cannot now ignore the existence of the Derveni Papyrus, a substantial fragment of a papyrus scroll discovered in Macedonia and dated to the end of the fourth century BCE. This is one of the most ancient extant papyrus documents currently known, and it is one of the few found outside of Egypt. It presents parts of a commentary, primarily dedicated to philosophical and religious questions, on an Orphic Cosmogony composed in hexameters; the commentary itself is arranged in the typical succession of lemmata followed by exegesis, which will become the norm in the analogous works produced in the Alexandrian era; moreover, the text of the commentary itself in the Derveni papyrus shares with the later works certain elements of style and terminology.7 Alongside the commentary, another typical scholarly product of the Alexandrian era was the ekdosis, i.e., edition. The production of a new edition of an earlier work entailed both the emendation of the transmitted text and the selection of readings from among the variants found in different copies of the same work; the term used to describe these interventions was diorthosis. In modern scholarship these two aspects of diorthosis have received differing emphases: indeed, some modern scholars have held that the greatest Alexandrian philologists did not compare different copies of the same work or select from among  7 Henry 1986; Lamedica 1991; 1992; M.S. Funghi 1995.

Scholarship, Philology, and Grammar  

multiple variant readings; by this view, the Alexandrian philologists were instead engaged merely in emending (and without great scruples at that) a text which they assumed for various reasons to be already corrupt and untenable. In truth, even if it is true that the Alexandrians often modified the text with a certain degree of nonchalance (not more so, however, than has occurred even in more recent, and therefore much “more scientific”, periods in the history of classical philology), still, it seems impossible to deny that, at least to a certain degree, they also collated exemplars and selected between variants when they were faced with a tradition that presented divergent readings. Thus we can say that they had an authentically philological idea of textual criticism and textual transmission, even if they did not perhaps apply this idea with complete consistency. It is already an achievement to have conceived of an idea and to have put it into practice; in this way, a contribution to intellectual progress has been made once and for all. In sum, in the works of the most important Alexandrian philologists, the organic unity that comprises both textual criticism and interpretation is well-established and richly productive (with all the benefits and the risks that this entails). Modern scholarship has long debated the nature of Alexandrian ekdosis: What exactly were the components of a Hellenistic edition? What methods were employed, and what resulting product was obtained, when an ancient grammaticus undertook the edition of, say, Homer, or Pindar, or Herodotus? One possibility is that the final product was a completely new copy containing the specific text established by a specific grammaticus; another possibility is that an ancient grammaticus carefully selected a pre-existing copy and worked with that as his basis, consigning his own editorial judgements to annotations written in the margins or else to the philological discussion in a commentary.8 However the case may be, in the manuscript tradition of the texts that have come down to us the presence of readings chosen by the great Alexandrian philologists is, in general, fairly uncommon; and this suggests that their ekdoseis did not give rise to an influential strand within the textual tradition. (It seems that their only decisive influence consisted in definitively fixing the number of lines of the Homeric poems, which became stable more or less in the first century CE, i.e., after Aristarchus.) In any case, the connection between the text and the commentary was ensured by the use of critical signs, which were set alongside the text itself and specified the nature of a given textual intervention, or the type of exegetical problem that was addressed in the commentary: expunction, rearrangement of  8 Discussion of the problem may be found in Pfeiffer 1968, 215 ff., with bibliography: cf. Erbse 1959; Turner 1984, 131 ff.; van Thiel 1991, x ff. and xxviii ff.

  Scholarship, Philology, and Grammar parts, interpretative difficulties, dubious readings adopted by a predecessor, alteration of linguistic forms. The systematic use of the critical signs was gradually improved and adapted to different kinds of texts, and it constituted a technical innovation of fundamental importance for the ongoing work dedicated to the textual tradition and the interpretation of earlier writings.9 From the various materials we currently have at our disposal, it is clear that exegetical problems, and occasionally even textual problems, were also discussed in other kinds of works. Among such works were the monographic treatises known as syngrammata, which generally were dedicated to the discussion of a specific issue posed by a single text or author, and the collections of lexeis, in which a key word was employed as a lemma and thus provided the link between a given literary text and problem being addressed (etymology, semantics, linguistic or grammatical difficulties, and so on). Material deriving from these syngrammata and collections of lexeis has made its way into the scholia we possess, alongside the material deriving from hypomnemata and ekdoseis. In addition to what we can glimpse within the scholia, however, significant remnants of scholarly literature have also been restored to us via papyrus finds. Dating roughly from the third century BCE to the seventh century CE (with the greatest number belonging to the first three centuries CE), these papyrus documents have given us bits of various kinds of works: copies of texts with variant readings and marginal annotations; fragments of hypomnemata and syngrammata; and portions of lexica of different types, produced in the Hellenistic and Imperial eras and all the way down through Late Antiquity. We can thus now hold in our own hands fragments of these early works of scholarship — fragments which vary greatly in length and attest to numerous different interests: from the most basic kind of glossographic explanation of individual words, to the discussion of philological problems, exegetical difficulties, antiquarian details, mythographic issues, linguistic usage, lexical questions, rhetorical figures, stylistic observations, to proverbs, anecdotes, and biographical material, and even aesthetic and moral judgments regarding an author’s work. Fragments have even appeared containing allegoresis, an ancient method of explaining the content of myths (or of imagery), which in the Hellenistic period was practiced above all in the Stoic-influenced group of scholars working at Pergamum.10 In extant fragmentary papyrus documents, sometimes a single subject-matter is discussed alone or is at least prevalent, but often observations related to various kinds of content are grouped together in the manner of a  9 Gudeman 1922; Pfeiffer 1968, passim; Montanari 1979, 43 ff.; McNamee 1992, with bibliography. 10 Cf. below, sections 2 and 3.

Scholarship, Philology, and Grammar  

miscellany; sometimes one specific interpretative orientation is dominant, but commonly we find a mixture of material employing different scholarly or doxographical approaches, not infrequently in reduced or abbreviated form. Even within the same genre of text the structure of different documents is not always the same, but in general we can say that hypomnemata had a structure that alternated lemmata and exegesis and followed the order of the work being interpreted; the extant fragments of hypomnemata thus present us with a part (which by this point has grown rather substantial) of what was read alongside the text of the great authors in order to understand them better. I have already mentioned that the original works of the scholars of the Hellenistic and Imperial periods are in large part lost to us: what we now possess, instead, are fragmentary documents that each represent one part of a variegated congeries of scholarly material, which was passed down through the centuries, fused together into compilations assembled from heterogenous sources, epitomized into smaller and smaller notes, but also enriched at times by new contributions, reorganized into more efficient structures, reduced or amplified as deemed necessary — a process of evolution that continued until the corpora of scholia were copied into the margins of the Byzantine manuscripts so that they might function as commentaries on the ancient texts.11 A few scholarly works of the Imperial period, however, have in fact been preserved down to our time. As an example, one may cite the extensive medical commentaries of Galenus (his philosophical commentaries have been lost), whose importance for the history of philology has not yet been fully clarified and ought to be investigated further in future research.12 The history of the interpretation of Homer in antiquity also offers a few interesting cases, which involve the Homeric Questions of Porphyry and the brief allegorizing treatise written by a certain Heraclitus. Extracts from the work by Porphyry are fairly common within the marginal scholia found in the manuscripts of Homer, but the first book of this work has also been preserved autonomously, and comparison between the two traditions reveals very significant divergences. The same can be said regarding the interpretations of Heraclitus: the excerpts found in the margins of the manuscripts are noticeably different from the corresponding

 11 Cf. Gudeman 1921; Wilson 1983; Arrighetti 1987, 190 ff., with bibliography. On the origin of the scholia, cf. Wilson 1990, 87 ff. (along with Wilson 1967; 1971a; 1977; 1983). 12 Cf. Bröcker 1985; von Müller 1891; Manuli 1984; Manetti/Roselli 1994, with bibliography.

  Scholarship, Philology, and Grammar passages in the complete text, which has been preserved in several exemplars.13 These two examples are exceptional but highly instructive: they reveal the extent to which ancient works of scholarship were modified during the process that led to their being employed as sources for the scholiographic corpora. Papyrus fragments have restored to us exegetical materials related to many authors that did not survive in a Byzantine manuscript tradition and therefore do not possess a corpus of scholia. Indeed, without the fragments of hypomnemata and syngrammata that have appeared in papyrus finds, we would have a much poorer understanding of philological activity in antiquity14 — for instance the philological work done on the lyric poets, apart from Pindar, who is the only lyric poet for whom a corpus of scholia is extant. In general, however, there are quite a few collections of scholia that have survived in Byzantine manuscripts, usually copied in the margins of the volumes containing the original texts, although we do possess a few codices that contain only scholia (accompanied by lemmata, of course).15 The most significant collections of scholia are those related to the Iliad and the Odyssey, which probably constitute the richest source we currently have of information about Alexandrian philology and hermeneutics; this is primarily due to the remnants they contain of the works of Didymus, Aristonicus, Nicanor, and Herodian, but it is also thanks to other material deriving from anonymous exegetical commentaries of the Roman period, as well as from glossographic compilations and mythographic commentaries, which in turn incorporate and reutilize earlier scholarly materials.16 After Homer, the

 13 For Porphyry: Schrader 1880 (Iliad) and Schrader 1890 (Odyssey); Sodano 1970; cf. also (various authors) Porhpyre, in «Entretiens Hardt» 12, Vandoeuvres 1965. For Heraclitus: Buffière 1989. Cf. below, section 3. 14 Cf. Turner 1984, 120 ff.; Del Fabbro 1979; Arrighetti 1987, 190 ff.; Montanari 1988b (= this volume, ch. 33); McNamee 1992. 15 This is the case of the principal manuscripts of the D-Scholia to the Iliad (cf. the following footnote). 16 The scholia deriving from Didymus, Aristonicus, Nicanor, and Herodian have been transmitted via the so-called “Four-Man Commentary” (VMK), which incorporated their works; VMK is the origin of the majority of the scholia in the MS Venetus A for the Iliad and of the corresponding scholiastic material in various manuscripts of the Odyssey. For the scholia to the Iliad: the extracts of VMK and the scholia exegetica are the two classes of material included in the edition of Erbse 1969–1988; for the D-Scholia (not included in Erbse’s edition), we have the editio princeps of J. Lascaris (1517, Rome), often reprinted (in addition, the portions of the D-Scholia that made their way into Venetus A — especially the scholia related to mythographic content — can be found in the old edition of the A-Scholia produced by W. Dindorf, Oxford 1875); a new edition of the D-Scholia is currently being prepared by the author of the present chapter. For the scholia to the Odyssey: Dindorf 1855; Ludwich 1888–1890; for the D-Scholia (the material belonging

Scholarship, Philology, and Grammar  

most significant scholiastic corpora are those on Aristophanes and Pindar; next come those dedicated to the three tragedians Aeschylus, Sophocles, and Euripides; and last, those on Hesiod.17 As for the major poets of the Hellenistic era, we have substantial collections of scholia on Lycophron, Theocritus, and Apollonius of Rhodes;18 a smattering of scholia is also preserved for the Hymns of Callimachus, thanks to their transmission via medieval manuscripts.19 Lastly, a not insignificant group is represented by the scholia dedicated to a few didactic poets: above all those on Aratus, but also those on the two surviving poems by Nicander, and on Dionysius Periegetes and Oppian.20 In general, the collections of scholia dedicated to prose works are less common and less rich than those dedicated to poetry. A handful of scholia on Herodotus are extant (for the most part very brief), whereas the corpus of scho-

 to this class is indicated by the siglum V in Dindorf’s edition) we have the editio princeps of F. Asulanus (1521, Venice); A. Dyck has announced that he is preparing a new edition of the scholia to the Odyssey. The bibliography on the Homeric scholia is too large to be summarized in a footnote: for a general survey one may consult Schmidt 1976, 9 ff.; Montanari 1979; 1992, 159 f. and 173 f.; Snipes 1988. 17 For Aristophanes: Dübner 1877; the scholia of the Ravenna MS have been edited separately by Rutherford 1896–1905; the scholia on Birds have been edited by White 1914; a new edition of all the ancient and Byzantine scholia to Aristophanes is in preparation under the direction of D. Holwerda and J.W. Koster so far the scholia vetera have been published only for Acharnians, Knights, Clouds, Wasps, Peace, and Birds, Groningen, E. Forsten, 1969–1991; cf. also Zuntz 1975. For Pindar: Drachmann 1903–1927; Tessier 1989; Arrighetti/Calvani/Montanari 1991; cf. Deas 1931; Lefkowitz 1985. For Aeschylus: Dindorf 1851; the scholia vetera to Prometheus Bound have been edited by Herington 1972, those to The Persians by Dähndardt 1894, those to Seven Against Thebes by Morocho Gayo 1989; of the new edition being prepared by O.L. Smith two volumes have appeared, with the scholia to Agamemnon, Choephoroi, Eumenides, Suppliants, Seven Against Thebes, Leipzig 1976–1982. For Sophocles: Papageorgiou 1888; the scholia vetera to Oedipus at Colonus have been edited by de Marco 1952, those to Ajax by Christodoulou 1977; cf. de Marco 1936; 1937. For Euripides: Schwartz 1887–1891. For Hesiod: Pertusi 1955; Di Gregorio 1975. 18 For Lycophron: Scheer 1908. For Theocritus: Wendel 1914. For Apollonius of Rhodes: Wendel 1935. 19 For the scholia to Callimachus’ Hymns, cf. Pfeiffer 1949–1953, II, lxxvi ff. and 41 ff.; in addition to the scholia deriving from the manuscript tradition of the Hymns, Pfeiffer has also included the fragments of scholia found on papyrus; further papyrus fragments of commentary on Callimachus can be found in Lloyd-Jones/Parsons 1983, frr. 238 ff.; cf. also Montanari 1976b (= this volume, ch. 57). 20 For Aratus: Martin 1974. For Nicander: Crugnola 1971; Geymonat 1974. For Dionysius Periegetes, the scholia can be found in Müller 1861, 427–457. For Oppian: Bussemaker 1847.

  Scholarship, Philology, and Grammar lia on Thucydides is more considerable;21 we also have a small group of scholia on Xenophon, Strabo, and Plutarch.22 For Thucydides, we also now have a papyrus fragment dating to the late second century CE which preserves a substantial portion of a commentary on Book 2.23 The fragment contains nineteen columns of text, organized in the usual form of alternating lemmata and exegetical comments, and covering the Thucydidean text from the beginning of the book down to II 45.2. However, toward the beginning, several entire columns are occupied by a single comment (on one lemma), which addresses the criticisms that Dionysius of Halicarnassus had made regarding the structure Thucydides adopted for the exposition of his history. This document, then, is a very interesting case, since it demonstrates how much the content of hypomnemata could vary, and it serves as a stark caution against hasty conclusions: for if we had only found a few of the first columns of this text, we might very naturally have supposed that the fragment was part of a treatise on historical composition, or else perhaps a polemical pamphlet; whereas in fact we are dealing with a typical example of a commentary composed of lemmata and explanations (some of which are in fact extremely concise, and which even cite poets, such as Callimachus, Euripides, Homer, Pindar). By a fortunate turn of events, we also have a small papyrus fragment of the third century CE which bears the title “Hypomnema of Aristarchus on Herodotus Book 1”.24 This title represents the first secure piece of evidence we have for an Alexandrian commentary on a prose author, but this surely cannot be taken

 21 We do not have a complete collection or a systematic edition of the scholia to Herodotus. What has been edited so far may be found in Stein 1871, II, 431–482, which contains the scholia along with a collection of Λέξεις Ἡροδότου (reprinted in Latte/Erbse 1965, XII f. and 191–230). This material has been employed in the apparatus criticus of the new Teubner edition of Herodotus: Rosén 1987–1997, vol. I (cf. Praef. p. XLIX; see also Rosén 1962, 217–231); the scholia have also been reproduced, along with an Italian translation, in the appendices to the volumes of the edition of Herodotus that is being published (by various editors) in the series «Scrittori Greci e Latini» of the Fondazione Lorenzo Valla (see, thus far, volumes I, II, III, Milan, Mondadori, 1988, 1989, 1990). For the scholia to Thucydides: Hude 1927; cf. Luschnat 1954; Tosi 1983; a new edition is being prepared by A. Kleinlogel, and promises to be a notable advance on previous scholarship, especially regarding the history of the how the corpus of scholia was formed. 22 For Xenophon: Dindorf 1855, 381–396. For Strabo: Diller 1954 (subsequently also included in Diller 1983). For Plutarch: Manfredini 1979. 23 POxy. VI (1908) 853; cf. Luschnat 1954, 25–29; Canfora 1974. 24 PAmherst II (1901) 12: cf. Paap 1948, 37–40; Pfeiffer 1968, 224 f.; and, providing a rather eccentric assessment, Hemmerdinger 1981, 154 ff. The text of the papyrus is reproduced in the apparatus of the edition by Rosén 1987–1997, 1, 135 f.

Scholarship, Philology, and Grammar  

to mean that earlier scholars paid no attention at all to prose: they may well have done so, of course, without producing proper hypomnemata. Rudolf Pfeiffer, in any case, is right to emphasize how notable it is that an exemplar of a commentary by Aristarchus has been copied, with its title included, four centuries after the time in which Aristarchus lived: from what little has been preserved, it is clear that the commentary circulated in a reduced format, but this of course is not at all surprising, given that we find numerous fragments of hypomnemata dating to earlier centuries (especially on poetic works) which are clearly epitomized and/or aggregate products. Indeed, the process of transformation of the scholarly works of the Alexandrians began very early, and it was carried out in a fairly unscrupulous fashion: it was considered legitimate to make use of scholarly literature with a freedom and a flexibility that were impossible in the case of the texts of the “Authors”; each copy could be reduced or increased or recast as needed and as dictated by individual preferences; and thus every single copy could easily represent an entirely different redaction. As far as oratory is concerned, the most studied author was Demosthenes. In antiquity, the exegesis dedicated to the works of Demosthenes was considerable (a systematic commentary on his entire oeuvre was written by Didymus, who in turn must have relied on predecessors);25 unfortunately, the scholia on Demosthenes preserved in the medieval manuscripts are by comparison rather meager,26 and those on his contemporaries Aeschines and Isocrates are extremely scant indeed. Scholia are not at all abundant for philosophical works.27 For later prose authors, we have collections of scholia to Lucian, Aelius Aristides, Hephaestion, and Hermogenes.28 A special case is that represented by the substantial and highly interesting corpus of scholia of a grammatical nature, deriving from various time periods and various sources, on the Techne grammatiké of Dionysius Thrax.29

 25 Much the same can be said (with the necessary precautions and reservations) for Thucydides: cf. Pfeiffer 1968, 225; Tosi 1983, 9. 26 Dilts 1983–1986; a papyrus fragment has restored a substantial portion of Didymus’ commentary on the Philippics: Pearson/Stephens 1983; cf. also Lossau 1964. 27 Dindorf 1852; Dilts 1992. For Plato and Aristotle: Greene 1938; Brandis 1961. (The scholia to Plato and Aristotle represent a special case, given that, all told, their number is extremely small, especially in comparison with the enormous quantity of philosophical commentaries that have been preserved for the works of both philosophers.) 28 For Lucian: Rabe 1906. For Aelius Aristides: Dindorf 1829; cf. Lenz 1934. For Hephaestion: the scholia may be found in Consbruch 1906, 79 ff. 29 Hilgard 1901.

  Scholarship, Philology, and Grammar I have already mentioned that the quality of the different corpora of scholia is highly variable. These corpora were first assembled, for the most part, within a chronological arc that stretches from the seventh to the ninth centuries CE, that is to say, from the period immediately preceding, to the period immediately following the so-called “dark ages” — or, if one prefers, between the end of Late Antiquity and the beginning of the Byzantine period. The variable quality of the scholia thus derives from the nature of the sources that were employed, and the state of their conservation, at the moment when the collections of scholia were composed.

Lexicography The scholia have very close connections with another genre of scholarly literature, namely, lexicography.30 This is because information often passed, by a kind of osmosis, back and forth from the one genre of text to the other, and also because both of these forms of scholarship ultimately derive, generally speaking, from the same original cultural context and had the same processes of development. Moreover, from the first collections to the great Byzantine compilations, the routes by which lexicographical material was preserved and transmitted were analogous to those already described in the case of the scholia. The roots of lexicography stretch back to the earliest modes of Greek glossographic interpretation. We have unequivocal evidence of the fact that, at least from the fifth century CE, schoolteaching involved the explanation of individual words: that is to say, terms that were difficult to understand (because they were foreign from ordinary usage, or obsolete, or technical, or deriving from a different dialect) were paraphrased or translated into the contemporary Greek spoken by teachers and students. This practice was adopted especially for the teaching of Homer — the dominant, indeed inescapable, schooltext — but also in the study of early laws (for instance, those of Solon) with their obscure phraseology.31 This phenomenon surely has its origins even further back in the history of Greek culture, and it continued without interruption: glossography and paraphrase — the “translation” of obscure Greek into contemporary spoken Greek —

 30 The following studies, though dated, remain essential: Cohn 1913; Tolkiehn 1925. More recent general introductions: Serrano 1977; Degani 1987; Alpers 1990; cf. also Tosi 1988, 115 ff. 31 The essential evidence comes from Aristophanes’ Δαιταλῆς, fr. 233 Kassel-Austin: cf. Pfeiffer 1968, 15 and 79 (in general, chapters I and II for the pre-Alexandrian phase); Cassio 1977, 75–77.

Scholarship, Philology, and Grammar  

are attested in the Hellenistic and Imperial periods, not to mention the continuous paraphrastic works of Late Antiquity and the Byzantine era. This type of literary activity — the explanation of the meaning of individual words — was initially associated with two specific cultural contexts: on the one hand, the teaching done in schoolroom, and on the other, the reflections on language developed, first of all, by the poets themselves, and later by philosophers like Democritus, Gorgias, and the Sophists, not to mention Plato’s interest for the true meaning of words as attested in his Cratylus.32 In the Hellenistic period it passed into the hands of the philologists, too (who were often harshly critical of the earlier, non-scientific forms of glossography, as is clear from Aristarchus’ polemic against the glossographoi),33 where it assumed different forms and addressed different kinds of questions, with a continual expansion of its range of interest. The result of all this work was a set of scholarly material related to glossai (‘difficult’ words found in the classical authors and requiring context-specific interpretation, in accordance with Aristotle’s taxonomy of poetic language and the methodology he developed for its study, as established in his Poetics)34 and lexeis (words which for any reason provided some stimulus for discussion, whether linguistic, lexical, philological, or antiquarian, and thus could be employed as Stichwörter or lemmata in lexicographic collections, which were originally thematic, but over time increasingly arranged in strictly alphabetical order). Lexicography thus became an autonomous scholarly activity, practiced with particular fondness by the Alexandrian philologists; we have evidence for collections produced already by Philitas, Zenodotus, and Callimachus, not to mention Philemon (the forerunner of Atticist lexicography), Xenocritus of Cos (the author of the first Hippocratic lexicon), and Neoptolemus of Parium (a compiler of Homeric and dialect glossai). A decisive role was played by Aristophanes of Byzantium: his Lexeis, an imposing work divided into numerous sections organized thematically, became a fundamental model for future lexicographers, because of its expert methodology (historical, philological, linguistic), its vast learning in every subject area, and the richness of its contents. From Aristophanes the work passed to his students; then, from Aristarchus to his successors down to Didymus; and then into the Imperial era. Ancient literary scholars never stopped cultivating and developing lexicography, and

 32 On these problems, see especially Pfeiffer 1968, Part One; Arrighetti 1987, Part One; with these authors, we get close to the problematic concept of the “prehistory” of philology, on which see section 2 below. 33 Dyck 1987. 34 § 21, 1457 b 1 ff. and § 25, 1461 a 10 ff.

  Scholarship, Philology, and Grammar the overall production was enormous: dialect lexica (among which the Atticist lexica stand out), lexica of synonyms, lexica of etymologies, lexica dedicated to individual authors (like the Homeric lexica of Apion and Apollonius Sophista, 1st cent. CE) or to literary genres (tragedy, comedy, historiography, oratory, and so on), onomastic lexica (like that of Julius Pollux, 2nd cent. CE), compilations of words related to specific subject-areas (like the lexicon of Stephanus of Byzantium, 6th cent. CE, on ethnic and geographical topics), and even epimerismi.35 New compilations of lexeis were assembled, incorporating the results of earlier studies more and more as time went on. One of the most famous was that of Pamphilus of Alexandria (1st cent. CE), in 95 volumes, which may stand with dignity next to the work of Didymus as a summa of the accumulated learning of the previous centuries. Pamphilus’ work was epitomized by Julius Vestinus (2nd cent. CE), who in turn was a major source (along with others) for the work of Diogenianus of Heraclea (2nd cent. CE); Diogenianus then constituted the principal basis for the work of Hesychius of Alexandria (5th-6th cent. CE). The lexicon of Hesychius has come down to us, in a redaction, however, that is both incomplete and yet also notably interpolated; still, in its current form, with its over 50,000 lemmata, it is one of the richest sources we possess of lexical and antiquarian information regarding Greek antiquity.36 Just as for the scholia, in the case of lexicography too a large part of the works produced in the Alexandrian and Imperial periods has been lost. Our knowledge of this genre of scholarly literature, therefore, stems once again largely from the great compilations assembled in Late Antiquity and the Byzantine era; these compilations inherited and redacted the lexicographic materials of the earlier Greek tradition, and they often gathered around a single lemma various kinds of information, with the ambition and the intent of an encyclopedia. These compilations are, above all: the so-called Glossary of Cyrillus (dating to the 5th cent. CE, and later expanded), the Lexicon of Photius (9th cent.), the Suda (10th cent.), the lexicon of Pseudo-Zonaras (end of the 12th cent.), and the  35 In this peculiar genre of text, an original work is divided word-by-word, and for each word is provided a grammatical analysis (and sometimes a semantic analysis). We know of epimerismi written by Herodian, but the collection attributed to him that has come down to us is not authentic. The most famous epimerismi are the ones of Homer (see now Dyck 1983, with information on previous collections and editions) and the ones of the Psalms. Originally epimerismi were compiled following the order of the text (the so-called scholiastic epimerismi), subsequently they came to be re-arranged in alphabetical order (alphabetic epimerismi). 36 For a thorough discussion and bibliography, cf. Degani 1987. For Hesychius we have a complete edition by Schmidt 1858–1868; and a new edition undertaken by Latte 1953–1966 (thus far only complete through the letter o).

Scholarship, Philology, and Grammar  

four major Etymologica along with numerous other texts of similar kind but smaller dimensions.37

Grammar The study of grammar — at least starting with Aristophanes of Byzantium and Aristarchus — developed hand-in-hand with textual criticism and the exegesis of literary works. Subsequently, Dionysius Thrax and the later generations of post-Aristarchan scholars began to consolidate earlier research on grammar and to systematize it as a set of linguistic norms. The culmination of all this scholarly activity is found in the second century CE, in the works of the two most significant figures in the history of Greek grammar, Apollonius Dyscolus and Herodian. We have already seen38 that philosophers like Democritus, Gorgias, the Sophists, and Plato were all interested in questions of language: indeed, they can be said to stand at the origins of that particular form of metalinguistic reflection that we call grammar. These thinkers began to debate problems such as the relationship between words and objects; they made observations on homonyms and synonyms; and they framed the question of linguistic ‘correctness’. Later, in Aristotle (Poetics 20, 1456 b 20 ff.), we find a series of considerations regarding the letter, the syllable, conjunctions, nouns, verbs, and inflection. It was in Stoic contexts, however, and especially in the thought of Zeno and Chrysippus, that grammar was first formalized as a science: especially important was the analysis and the definition of many aspects of morphology and verbal tense. Other topics that were investigated include vowels and consonants, the parts of speech, the ambiguity of language, and the relationship between concepts and  37 For more information and further bibliographic details, cf. Degani 1987. We do not have a complete edition of Cyrillus. For Photius: for older editions, which still contain the lacunae present in the Codex Galeanus (the largest of which goes from ἀδιάκριτος to ἐπώνυμοι), cf. Porson 1823 and Naber 1864–1865 to a very small extent, towards the beginning, the passages in these editions that are lost in the lacunae are supplied by the material published in Fredrich/Wentzel 1896 and in Reitzenstein 1907; thanks to the MS Zavordensis 95, which is not marred by lacunae, a new edition is in preparation: Theodoridis 1983. For the Suda: Adler 1928–1938. Regarding the etymologica: the Etymologicum Genuinum (second half of the 9th cent. CE) is for the most part still unpublished; Etymologicum Graecae Linguae Gudianum (11th cent.): Sturz 1818 (a new edition by De Stefani 1909–1920 was never completed); Etymologicum Magnum (first half of the 12th cent. CE): Gaisford 1848; Etymologicum Symeonis (the work of Symeon, first half of the 12th cent. CE), for the most part still unpublished; [Iohannis Zonarae] Lexicon: Tittmann 1808. 38 Cf. above, Section Lexicography, p. 616.

  Scholarship, Philology, and Grammar linguistic forms (the fact that linguistic forms often do not neatly correspond to the real content of concepts was termed anomaly). In the process, many grammatical terms were coined that later remained in use;39 indeed, notwithstanding their differences, we should not underestimate the influence that Stoic thought had on later Alexandrian grammatical studies after Dionysius Thrax. One fundamental difference, however, remained firm: the Stoics strongly emphasized linguistic anomaly and the essential role of usage (Greek synetheia, Latin consuetudo); the Alexandrians, instead, focused on the description of regular phenomena in morphology, inflection, and word-formation, and they thus gave privilege to the principle of analogy. This latter strand of thought lies at the birth of a normative grammar of the Greek language. The systematic foundations for such a grammar were laid in the Hellenistic period, with Aristophanes of Byzantium and Aristarchus each playing a decisive role. Modern scholarship has debated when exactly grammar was established as an autonomous discipline, separated from its philosophical origins and identified definitively as an independent techne; more precisely, scholars have questioned whether a system of normative rules for grammar was first defined and drawn up in the second or in the first century BCE. According to some, with the observation that similar or identical linguistic phenomena reoccur in a regular fashion, and with the increasingly refined application of the principle of analogy, Aristophanes and Aristarchus were able to conceive of the existence of morphological regularity, but they did not yet arrive at the abstract formulation of a system of rules. According to others, however, it was precisely Aristophanes and Aristarchus who established a set of rules for grammatical inflection, and this doctrine would then assume its definitive form during the first generation of the students of Aristarchus; by this view, normative grammar would have been born by the end of the second century BCE (and not in the time of Varro). Apart from the question of how to evaluate the evidence for Aristophanes and Aristarchus’ grammatical observations, one of the main difficulties in answering this question has to do with the Techne grammatiké of Dionysius Thrax, a student of Aristarchus who lived roughly from 170 to 90 BCE: for if we were certain that the grammatical treatise that has come down to us under this name is in fact the work of Dionysius Thrax, then we could settle the question of Aristophanes and Aristarchus’ contributions to grammatical studies and their relative chronology. But the authenticity of the Techne grammatiké is debated; some scholars maintain  39 Cf. Schmidt 1839; Barwick 1957; Collart 1962; Lloyd 1971; Frede 1978; Schenkeveld 1984 and 1990 (with bibliography); Sluiter 1990, 5 ff. (with bibliography); see also the works cited in the following fn. and in section 3, fn. 99, on Crates.

Scholarship, Philology, and Grammar  

that the text we have is in fact a compilation of the fourth century CE (or even later), and they assign the foundation of normative grammar to the first century BCE or later, emphasizing the role played by figures such as Tyrannion, Philoxenus, and Tryphon. That being said, the question could be analyzed from another point of view: even if the authenticity of the Techne remains in doubt, what we know about the morphological observations made by Aristophanes and Aristarchus has a certain value in and of itself; and these observations, even if only partially preserved, could be considered evidence for a first attempt to establish a normative grammar. Moreover, one aspect of the question regarding Dionysius Thrax is worth emphasizing: even if one admits that the Techne, in the form in which it has come down to us, is not authentic in its entirety, it is certain that the first four chapters are in fact authentic; and this means that Dionysius Thrax40 must have written a grammatical treatise with the intent of employing a systematic, textbook-like structure for the exposition of the various elements of grammatiké — and what is more, he must have done so with a genuinely Alexandrian outlook, seeing as the work begins with the definition of grammar as the experiential knowledge (empeiria) of what is contained in the works of the poets and prose writers.41 To sum up: the hypothesis that (at least) an inchoate system of rules of normative grammar was defined and assembled by the scholars of the generations of Aristophanes and Aristarchus and their direct successors (such as Dionysius Thrax) — that is to say, in the course of the second century BCE — remains the most probable; subsequent generations then continued on the same path, down to the time of Herodian.42 When the reflection upon linguistic phenomena passed from a philosophical and rhetorical context to a philological one, it acquired an essential and organic link with the interpretation of texts. For it is in connection with textual exegesis dedicated to literary works that the foundations were laid for the study of grammar, on the basis of an increasingly broad and increasingly profound observation of the linguistic phenomena contained in the authors whose works  40 On Dionysius Thrax, see below, section 4, fn. 95. In general, for discussions regarding the problems related to the first phases of Greek grammatical studies, in addition to the works cited in the previous footnote, one may also consult: Fuhrmann 1960, 29 ff. and 144 ff.; Pinborg 1975; Siebenborn 1976; Wouters 1979; Erbse 1980; Ax 1982; Leszl 1985; Ax 1986; Taylor 1986; Di Benedetto 1990; Schenkeveld 1982; 1984; 1990; 1990a; Sluiter 1990 (with bibliography); Nicolai 1992, 178 ff. 41 I do not intend to suggest that, at this stage, there was a comprehensive and systematic treatment of morphology and the parts of speech; these levels of grammatical presentation would only be achieved later. 42 Cf. section 4 below.

  Scholarship, Philology, and Grammar were being interpreted and edited.43 Even later, however, when the study of grammar established itself as an autonomous discipline, it never lost its link with philology: this can be readily observed in the works of Asclepiades of Myrlea, Tyrannion, Philoxenus, Ptolemaeus of Ascalon, and Herodian; but even in figures such as Tryphon and Apollonius Dyscolus, whose work (at least what we currently know of it) appears to be strictly grammatical, the interpretation of texts is nonetheless present, and indeed remains the basis for the observation and analysis of linguistic phenomena.

Paroemiography In all probability, paroemiography has its roots in ancient collections of wise sayings, aphorisms, and proverbs attributed to charismatic and authoritative figures. Gnomai of this sort were also present in literary works and played an important role in poetic style, and this in turn stimulated scholars to collect and interpret them. In truth, the type of content gathered in paroemiographic compilations was never precisely defined or rigidly delimited; it was instead rather elastic, embracing various kinds of gnomai, sententious sayings, proverbs, and traditional expressions. Indeed, it is not really feasible to draw a firm distinction between proverbial sayings of a truly popular origin, on the one hand, and gnomai culled from literary works, on the other; as a result, it is also not practicable to maintain a firm distinction between two different scholarly orientations in the selection of materials, with two associated kinds of collections, paroemiographical and gnomological. This is especially evident if one considers that popular proverbs were employed in poetry and even became literary topoi; vice versa, particularly striking poetic expressions could be transformed into sayings that were often repeated and eventually circulated at various levels of society. Thus, in the compilations that have come down to us we find apophthegms of the Seven Wise Men and lines of poetry alongside sayings that seem to betray a genuinely popular character and origin.44 An explicit interest in proverbs is attested for Aristotle, who wrote a work entitled Proverbs, which unfortunately has not come down to us.45 Aristotle’s

 43 Cf. Tosi 1988, 173 ff. 44 Cf. in general Rupprecht 1949; Kindstrand 1978; Pfeiffer 1968, 83 f., 208 f.; Ieraci Bio 1979; Tosi 1988, 197 ff. For an idea of gnomological literature, one may begin by consulting Spoerri 1975, which has an excellent bibliography. 45 Doubts have been raised on this issue, but cf. Pfeiffer 1983, 83 f.

Scholarship, Philology, and Grammar  

interest in proverbs was then continued by the Peripatetics Theophrastus, Clearchus of Soli, and Chamaeleon.46 The Stoic Chrysippus also wrote a work entitled On Proverbs, of which a few fragments are extant.47 Another work On Proverbs was written by the Attic historian Demon, a near contemporary of Aristotle.48 In philosophical circles, traditional maxims were viewed as the remnants of ancient wisdom, the first steps toward philosophical thought, worthy of being recovered and analyzed. It was thus, in all likelihood, in the Peripatetic tradition that traditional sayings first became the subject of research, not only as part of a “history of philosophy,” but also for their antiquarian interest; and this research will have paved the way, first, for the work of the philologists, and then for independent paroemiographic studies. In the context of Alexandrian philology, Aristophanes of Byzantium studied this material with a scientific approach and through the lens of an interpreter of texts. We have a few fragments of two collections he assembled, one dedicated to metrical proverbs, the other to non-metrical proverbs.49 Thus, the connection with philology and textual exegesis was established and solidly put into practice, and it would be much developed subsequently. The next important phase is that represented by Didymus, who compiled a substantial collection of paroimiai (in thirteen books), in which it appears that there were many citations from comedy; it also appears that Didymus’ paroemiographical researches were carried out with a polemical stance (though in what way exactly is not entirely clear) vis-à-vis his predecessors in the genre (Aristophanes himself?).50 After the compilation of Lucillus of Tarrha (1st cent. CE), this strand of scholarly research found its apex in the great sylloge of Zenobius (2nd cent. CE). This latter work has come down to us in modified redactions, and in it one can observe how the paroimiai are arranged in series according to the various literary genres from which they have been excerpted. As with other genres of scholarly literature, our knowledge of earlier paroemiography derives especially from the later compilations (structured by alphabetical order) that were transmitted. Among these are those that circulated under the names of Diogenianus and of Plutarch; Byzantine-era compilations were

 46 For Clearchus, Chamaeleon, and also Hermippus, see Montanari 1989 (= Ancient Authors, ch. 30); and also, for Hermippus, ibidem, I 1**, 59 2T, 258 ff. 47 Cf. SVF III, 202. 48 Cf. FGrHist 327 F 3, 7 ff. 49 Cf. Slater 1986, 124 ff. 50 Cf. Schmidt 1854, 396 ff.; Pfeiffer 1968, 279.

  Scholarship, Philology, and Grammar assembled by Gregory II of Cyprus, Makarios Chrysokephalos, Michael Apostolius, and Arsenius Apostolius.51

Exegesis before Alexandria: The Role of Aristotle and the Peripatetics The earliest known examples of exegetical activity dedicated to literary texts involve epic poetry (i.e., Homer). They date back at least to the fifth or sixth century BCE — thus, long before the birth and development of Hellenistic philology — and they are associated with two specific aspects of interpretation. On the one hand, we find the explanation of individual words or expressions, i.e., glossography,52 which has as its aim the basic, literal comprehension of the poetic text. On the other hand, we find the interpretation of myths, which aims above all to discover and to describe the supposedly true meaning of the mythical narratives found in epic, and generally takes the form of various types of allegoresis. Allegorical interpretation, in fact, often provided a response to the moral and theological criticisms that had been made regarding the representation of divinities in epic: for if one downplayed the literal sense of the text and promoted a hidden meaning, the poetry could become morally or religiously acceptable. This fact has given rise to the long-standing querelle concerning the original nature of allegory: Was allegory born from apologetic aims, precisely in reaction to moral and religious criticism, and with the intent of absolving Homer? Or was its original character disinterestedly exegetical and philosophical, a sincere attempt to render explicit and illustrate what readers believed to be the true meaning of the poetry? Put in these terms, the distinction perhaps creates an artificial set of alternatives: in practice, the two objectives probably coexisted and cooperated, with now one, now the other being emphasized. In any case, notwithstanding the limited quantity of the evidence and the early date of much of it, we can identify the general outline of this particular genre of Homeric criticism, which was applied to the ‘meaning’ of myths in epic — a meaning that was considered different from the literal one, insofar as it is hidden and more profound. The earliest allegorical interpretations of Homer that are currently known are those of Theagenes of Rhegium in the sixth century BCE. Subsequently, this  51 Von Leutsch/Schneidewin 1839–1851, 1961 (reprint of several fundamental essays); Bühler 1982–1987. 52 Cf. section 1.2 above.

Scholarship, Philology, and Grammar  

method of exegesis was taken up by Anaxagoras and his disciples, and later it was also practiced in Stoic contexts and by the scholars working at Pergamum;53 still later, it would find widespread employment in the philosophical schools of Late Antiquity (especially in Neo-platonic contexts) and in many areas of Christian literature and culture. Even Plato, however, was fully aware of allegory as an interpretative strategy, and he refers to it in the Ion, the Phaedrus, and the Republic, although it is clear that, as one would expect, he had a low opinion of an activity that lacked a precise techne and was far from constituting a truly ‘scientific’ (i.e. epistemologically sound) method.54 Aristotle’s opinion, on the other hand, seems to have been rather different: his Homeric Questions has not come down to us, but in a fragment cited in the scholia to Odyssey XII 128–129 and in Eustathius, Aristotle explains the meaning of the seven herds of the Sun, each formed of fifty cows, by appealing to the number of days in the lunar year: this constitutes a bona fide allegory, and it is hard to believe that it was a single, isolated example within Aristotle’s thought.55 It seems scholars now agree that the first individuals to provide interpretations of epic poetry, when the text and the audience required it, were the rhapsodes themselves during their recitations. In this way, the interpretation of poetry was born, and took its first steps, within the very context of poetic performance itself. After the rhapsodes, the next intellectual current that focused on poetic exegesis, and on linguistic questions more broadly, is found in the line of philosophical development that connects such thinkers as Protagoras, Prodicus, Democritus, and Plato (although, of course, I do not mean by this to posit any specific genealogies of teachers and students). Next we find another poet, the comedian Aristophanes, who intervened directly in the cultural and political debates of the fifth century BCE, and he did so in part with observations

 53 Cf. section 3 below. For Theagenes, cf. Diels-Kranz, Fragm. Vors. 8; Lanata 1963, 104 ff.; Pfeiffer 1968, 9 ff.; De Lacy 1948; Long 1992. 54 For a general discussion of this issue, cf. Tulli 1987. 55 Aristotle, fr. 175 Rose = 398 Gigon (in the scholion, note the adverb φυσικῶς): cf. Hintenlang 1961, 131 ff. In addition, a passage from the Metaphysics seems to provide a kind of theoretical justification for the practice of allegorical interpretation: Metaphysics XII 1074 b 1 ff.: “A tradition has been passed down from the ancient thinkers of the earliest periods to later generations in the form of a myth, which holds that these divine bodies are gods, and the Divine imbues all of nature. The rest of their tradition has been added later in mythological form, in order to convince the common people and to be used as an expedient to the advantage of the law and the general interest. They say that these gods have a human appearance, or that they resemble certain animals, and they make other corresponding declarations, similar to the ones we have cited” (cf. also Metaphysics II 995 a 4 ff.). On this subject, see Most 1992, VI.

  Scholarship, Philology, and Grammar on poets and poetry that establish him as an authentic precursor of literary criticism. Then there is the poet Antimachus of Colophon (5th/4th cent. BCE), who not only studied the language of epic poetry, but also — and well before the Alexandrian philologists — produced an edition of Homer which we find cited in the scholia.56 There is a tradition that attributes to Aristotle a genuine diorthosis of the Homeric text, which would have served as the basis for the copy of the Iliad that Alexander the Great carried with him; the authenticity of this tradition is certainly open to doubt, but the fact that such a tradition exists is perhaps noteworthy, and the fact that it employs the technical term diorthosis should at least give us pause.57 Thus, poets, sophists, and philosophers — not to mention the teachers who explained difficult words in poetry to their students — were all at work in this period, examining what was to be investigated in literary works; and their activity can be considered the antecedent of the philological, exegetical, and grammatical studies that would be further developed in the Hellenistic period to the point of becoming an autonomous discipline.58 When the poet Philicus, in the first decades of the third century BCE, dedicated one of his learned compositions (probably the Hymn to Demeter) to the grammatikoi, he was addressing himself to the erudite readers of poetry who constituted the privileged audience of his work. Thus it is clear that by this point the term grammatikos had acquired a technical meaning and referred specifically to the philologist who specializes in the interpretation of poetry; previously, in the classical period, the term had designated an elementary school teacher, the instructor who taught students how to read and write. The semantic shift — from schoolteacher to scholar — did not pass unnoticed among the ancients themselves, who explicitly posed the question of who had been the first to receive the epithet in its new meaning. One tradition held that it was

 56 The fragments of Antimachus are collected in Wyss 1936: cf. Praef. p. xxix, to which one should now add Lloyd-Jones/Parsons 1983, frr. 52 ff., 20 ff.; cf. Pfeiffer, History cit., 93 ff. 57 Cf. section 1.2 above, as well as section 3 below; cf. Pfeiffer 1968, 71 ff. 58 The most extensive discussion of the “prehistory” of Hellenistic philology is that in Pfeiffer 1968, Part One; cf. also Arrighetti 1987, Part One; and cf. section 1.2 above, with fn. 32. I would like to clarify that when I use terms such as “prehistory”, “antecedent”, or “predecessor”, I do not at all intend to make or to imply any judgements regarding relative value. My intent is merely to identify different phases, attitudes, and results, dating to different historical contexts, within such an important sphere of cultural activity like that of interpreting literature, authors, and styles; the reflection that this sphere of culture always existed in an organic connection with literary production helps us not to posit ruptures where in fact it is better to see a natural evolution within the history of Greek literary civilization.

Scholarship, Philology, and Grammar  

Praxiphanes, a student of Theophrastus; and the evidence for this tradition can be connected to that which speaks of Aristotle as both the founder of grammatiké (in the sense “advanced study of literature”) and, for that very reason, the intellectual ancestor of the Alexandrian grammatikoi.59 And indeed, the role of Aristotle and the Peripatetics — with their interests in the history of literature, in the lives of the poets, and in works of poetry themselves (e.g. in the form of monographs dedicated to problemata of various kinds raised by the texts) — should not be underestimated. They were the true predecessors who inspired the later developments of Hellenistic philology. The seed that flowered in Alexandrian soil was sown by Aristotle and the Peripatetics. It is traditional to identify the central activity of Hellenistic philology in the edition and interpretation of texts, in grammatical studies and lexicography and paroemiography, and such a view is certainly not unfaithful to reality. However, it is equally true that, for a complete picture of this moment in the history of scholarship, we should also take into consideration three other areas of study: biography, the theory of poetry, and the doctrine of literary genres.60 We have already seen above that biography fully belongs to the sphere of scholarly literature (research into the lives of poets was closely connected to research into their works); and as far as biography is concerned, the link between Aristotle and Hellenistic scholarship is fundamental.61 With regard to poetic theory, although reflections on the nature of poetry always formed part of the interests of poets and philosophers, it was Aristotle who, with the dissemination of his Poetics, became the inescapable interlocutor in this field of study. Of course, it is easy to point out the differences between a classical aesthetics of poetry, as evidenced in Aristotle’s treatise, and an Alexandrian one, as revealed in the literary works of (e.g.) Callimachus and Theocritus; but in any case, this is not the perspective that interests us here. Rather, for our purposes it is important to remember that Aristotle, in his Poetics, reflected on and analyzed precisely that poetry which the Alexandrians themselves would later study with intense dedication and with every scholarly means at their disposal. And it is worth remarking that the  59 The text of Philicus is now in Lloyd-Jones/Parsons 1983, fr. 677, 321. The tradition (Hephaestion) states that the apostrophe was contained in the proem, though some have proposed that it was contained in a sphragis at the end. The evidence regarding Praxiphanes is found in Clement of Alexandria, Stromata I 16. 79, 4 = fr. 10 Wehrli. For the various terminological questions, cf. Sandys 1920, 1 ff.; Pfeiffer 1968, esp. 156 ff. (with bibliography); Kuch 1965; Nicolai 1992, 178 ff. 60 These aspects of the history of philology were overlooked by Pfeiffer, and this has had a negative effect on the correct evaluation of the relation between Aristotle and Hellenistic philology (see below, fnn. 63 and 67). Cf. Arrighetti 1987; Rossi 1976. 61 Cf. section 1 above.

  Scholarship, Philology, and Grammar Alexandrian philologists (and not only those who were also poets) simply could not have afforded to neglect the theory of poetry. Indeed, it seems absurd even to think that a scholar such as Aristarchus did not have his own views on the nature of poetry; put differently, it is too facile, indeed specious, to point to the scarcity of theoretical and epistemological commentary within the extant fragments of Hellenistic philology as if this were evidence for the lack of such reflection on the part of the Alexandrians.62 Finally, there was the question of literary genres, the classification of which constituted a problem the Alexandrians took great pains to resolve. Here too the connection they maintained with the thought of the past, and in particular that of Aristotle, emerges in a very significant way.63 And in this regard, if we broaden our perspective slightly, we can reflect that the theory of literary genres fully belongs to what may be defined as “literary history” (the quotation marks are used advisedly, of course). More research is required to illuminate this subject in a rigorous fashion and within a proper framework, but there can be no doubt that the Alexandrians cultivated certain interests that can be called “literary-historical”; and here too they were working in a terrain that had been tilled by their predecessors.64 In short, the kind of research and discussion that was widespread in Peripatetic contexts, along with the theoretical framework that found expression in Aristotle’s Poetics, definitively changed the way that the works of literature were consumed.65 No longer principally relegated to dramatic competitions, to poetic recitations during the festivals of the polis, or to the various social contexts that involved the singing of lyric, literature — especially poetic literature, but subsequently also scientific literature, historiography, and oratory — was consumed by reading, and it became the object of interpretation, which entailed the study of authors and their works. A certain conviction became more and  62 Cf. further below in section 2. The fact that Aristotle’s Poetics is nowhere cited in the scholia is banal and proves nothing: why must it have been cited? in the comment on what passage? (Consider furthermore that the material in the scholia is the reduction of a vast amount of earlier work.) Rather than being cited explicitly, Aristotle’s Poetics will have provided theoretical tools and frameworks for interpretation. 63 Cf. Rossi 1971, 1976; see also, in the present volume (“Lo spazio letterario della Grecia antica”, vol. I, Tomo II), Ch. III 2 La ricodificazione dei generi (M. Fantuzzi). 64 A few examples only, from among the many possible hints of “doing literary history” that can be found among the texts that have come down to us: Aristarchus’ distinction between what is Homeric and what is cyclic or neoteric; Callimachus’ activity of classification and cataloguing; the insertion of literary events within works of chronology; and one can probably also include the observations on the history of style and poetic language. For some discussion of even more possible elements, see Montanari 1992, 15 ff. 65 Cf. the introduction by Lanza 1987.

Scholarship, Philology, and Grammar  

more widespread, namely that the present generation had inherited from the past a set of cultural models and profound literary creations, which were the repository of that patrimony of civilization which the Greeks called paideia; and along with this conviction arose the desire to understand, to interpret, and to pass on this patrimony, since it was felt to be the foundation of Greek cultural identity, which connected the present generation to the past without any sense of rupture. One of Theophrastus’ students, Demetrius of Phalerum, made his way to Alexandria around 295 BCE, perhaps at the invitation of Ptolemy I. There we find this Peripatetic, the author of treatises on Homer perfectly in line with the style of his school, present at the foundation of the city under Ptolemy I and during its splendid development under Ptolemy II, with the inauguration of the great institution that comprised the Museum and the Library — the centers, respectively, of scholarly activity and of the preservation of a precious literary heritage.66 Unfortunately, we know almost nothing about Demetrius’ own activity there, or his possible cultural influence; but his presence represents a link that cannot be ignored.67

From Zenodotus to Aristarchus Between the fourth and the third centuries BCE, there lived an elegant poet who was also a kritikos — Philitas of Cos.68 Philitas’ role in the history of philology is most associated with his collection of difficult expressions found in the works of the poets. This collection was entitled Ataktoi glossai (or Atakta), and it was notable for its innovative critical and exegetical approach, which constituted a significant methodological advance if compared to the elementary glossography of the previous era.

 66 Cf. Canfora 1993. Ch. III 1 La Biblioteca e il Museo (L. Canfora). 67 Pfeiffer 1968, esp. 67 ff. and 87 ff., tends to diminish the role of Aristotle and the Peripatetics, but cf. Rossi 1976; Carlini 1977; Podlecki 1969; Blum 1977, ch. 2; Richardson 1983; Meijering 1987; Nicolai 1992, 178 ff.; Richardson 1992. The fragments of the Peripatetics can be found in Wehrli 1967–19692 (+ Suppl. I–II, 1974–1978); a general survey can be found in Wehrli 1983. The fragments of Demetrius of Phalerum are in Wehrli 1968; on this figure and questions related to the Library and the Museum, cf. Pfeiffer 1968, 96 ff.; Fraser 1972, I, 320 ff.; Blum 1977, 99 ff. and 140 ff., with bibliography. 68 He is thus defined by Strabo XIV 657; cf. Pfeiffer 1968, 88 f.; for the frr., see Kuchenmüller 1928.

  Scholarship, Philology, and Grammar Philitas’ most important student was Zenodotus of Ephesus (ca. 330–ca. 260 BCE), who can be called the true founder of the Alexandrian school of philology.69 Zenodotus served as tutor to the royal family, and he was the first to obtain the role of head librarian, thus becoming the director of the innovative, momentous institution that had been created by the Ptolemies. He assembled a compilation of Glossai, arranged alphabetically (unlike that of Philitas), and he studied Hesiod, Pindar, and perhaps Anacreon. Of this work we possess only meager fragments. We are better informed of his work on Homer, which was for Zenodotus (as for nearly all of his successors) the true testing ground. Faced with numerous copies of the Homeric poems, each presenting divergent readings, Zenodotus posed the question of the authenticity and the correctness of the transmitted text. It appears that he did not place much credit in the “eccentric” copies of the poem, the copies full of those lines that are now called “plusverses”; in addition, he deemed many lines inauthentic on the basis of internal criteria (repetitions, contradictions, or other improprieties). In the latter cases, he took one of two approaches: either he deleted the lines entirely (i.e., he did not copy them out in his edition: his text of Homer had at least 140 lines less than that of Aristarchus), or else he included the lines in the text, but with a symbol appended, in order to indicate that they probably ought to be deleted. The symbol that he employed was the obelos (a horizontal dash placed in the left-hand margin alongside the verse in question), the first of the famous critical signs introduced by the Alexandrians; and a verb was coined to describe this practice of marking lines for probable deletion, namely athetein, from which was also coined the substantive athetesis. As the basis for his interventions in the text, Zenodotus must have adopted two procedures: on the one hand, he will have collated different exemplars and then selected from among variant readings; on the other hand, he will have often had recourse to his own conjectures (motivated, among other things, by concepts such as literary decorum and propriety).70 Both of these editorial procedures must be considered at work in Zenodotus’ philological activity. The name given to this kind of editorial activity, overall, was diorthosis, and its final product was an ekdosis (both of which we have already discussed above).71 It is certain that Zenodotus did not write commentaries in which he defended his reasons for his interventions in the text; some of this argumentation might have been found in his Glossai, but much of it  69 Nickau 1972, 1977; Pfeiffer 1968, 105 ff.; Pasquali 1952, 215 ff.; van der Valk 1963–1964, II, 1 ff.; Nicolai 1991; van Thiel 1992. 70 Cf. Nickau 1977, 183 ff. 71 Cf. section 1.1 above (also for discussion of the philological methods of the Alexandrians).

Scholarship, Philology, and Grammar  

must have remained oral doctrine passed down in discussions among Alexandrian scholars, with all the uncertainties that this likely entailed. There is no definitive evidence that it was his decision to divide the Homeric poems into 24 books; it is well attested, however, that he dedicated one study to a calculation of the number of days in which the action of the Iliad takes place. Zenodotus did not himself work on drama; this was instead the subject area privileged by two of his contemporaries, poet-philologists both, Alexander Aetolus and Lycophron. The former studied tragedy, the latter comedy, but unfortunately we know almost nothing concrete about their scholarly activity.72 We have some information indicating that an edition of Homer was made by the epic poet Rhianus of Crete. Rhianus was active in the second half of the third century; as a philologist, he was certainly influenced by Zenodotus, and his Homeric diorthosis must have built upon the foundations laid by his illustrious predecessor. It is noteworthy that the scholia preserve over forty readings attributed to Rhianus, which is testimony of his reputation as a scholar.73 Two other scholar-poets, working in exemplary solidarity, were Callimachus and Apollonius of Rhodes.74 Callimachus of Cyrene (ca. 320–ca. 240 BCE) must be remembered for the great undertaking that was his Pinakes: although in part a kind of catalogue of the literary patrimony contained in the Library of Alexandria, the Pinakes surely was intended to stand as an autonomous work, and it constituted effectively an annotated inventory of Greek literature. Authors were listed in alphabetical order, with a biography; works were cited by title and incipit, with an indication of their length; and other information was included, as Callimachus deemed necessary. Callimachus also wrote numerous scholarly monographs and works of literary criticism, such as his Against Praxiphanes. Apollonius of Rhodes (ca. 300–ca. 220 BCE) was Zenodotus’ successor as director of the library. His assiduous philological activity can be seen at work in the poetry of his Argonautica, which, among other things, contains a vast harvest of well-informed interpretatio of Homer. He tackled exegetical problems in his Against Zenodotus, which is the first known example of the debates that arose within the Alexandrian school itself. He also studied Hesiod and wrote a monograph on Archilochus.  72 Cf. Pfeiffer 1968, 106 ff. and 119 f.; Carlini 1977, 347. 73 In our sources there is some uncertainty as to his origins (some sources indicate that he was from Messenia); in terms of dating, he is said to have been a contemporary of Eratosthenes. Cf. Pfeiffer 1968, 148 f.; Poli 1989; Zumbo 1987; 1988. 74 For Callimachus, cf. Pfeiffer 1968, 123 ff.; Blum 1977; Brink 1946; Meillier 1979; Lehnus 1993. For Apollonius of Rhodes, cf. Pfeiffer 1968, 140 ff.; Michaelis 1875; Rossi 1968; Fantuzzi 1988; Fusillo 1993.

  Scholarship, Philology, and Grammar The next head librarian was Eratosthenes of Cyrene (ca. 295–ca. 215 or ca. 275–ca. 195 BCE), an extraordinary figure who combined studies in grammar, philology, and science, and even wrote poetry, if to a marginal degree.75 His scientific research was prodigiously vast, and in this space we can only mention his works on chronology and geography. In the first subject, he had a decisive influence on future studies: in his Chronographiai, improving upon previous research, he fixed many key dates both for historical events in the stricter sense, and also for the history of literature, such as: the fall of Troy in 1184; the akmé of Homer’s poetic activity 100 years later; Hesiod’s relative date compared to Homer (posterior); the first Olympiad (776); all the way down to the death of Alexander the Great, in 323. In his Geographika, he began his history of geography with Homer, although he maintained that the setting of the Homeric epics was for the most part imaginary, since (in his view) the role of poetry is to please rather than to teach. With regard to more strictly literary study, his great work was his research on Old Comedy, which grew to at least twelve books. In the few extant fragments, we can observe his interest in language and in problems of stage representation. He also undertook lexicographical research; he studied the lyric poets; and more — in works that are almost entirely lost. The apex of Alexandrian philology occurs with Aristophanes and Aristarchus. Aristophanes of Byzantium (ca. 265–ca. 190 BCE) was the successor to Eratosthenes as director of the Library of Alexandria.76 As a philologist, with his edition of Homer he distinguished himself from Zenodotus for his greater editorial prudence: instead of taking the drastic measure of not including spurious lines, he preferred more often simply to mark verses as probably spurious by means of the obelos; in this way, many lines that had been deleted by Zenodotus made their way back into the text, even if they now bore the stigma of the obelos. It seems probable that Aristophanes also collated texts more extensively than his predecessor and consulted numerous copies from which he selected his readings. These choices suggest that his editorial outlook was more conservative, and that he tended to treat the text with greater scruples than others had before him — although this should not be taken to mean that he  75 We lack a comprehensive collection of all the fragments of Eratosthenes. Information on this figure can be found in Pfeiffer 1968, 152 ff., along with Aujac 1977; Dihle 1986, 200 ff.; Meijering 1987, passim. 76 The traditional collection is Nauck 1848; we now also have a new edition, Slater 1986. Cf. also Pfeiffer 1968, 171 ff.; Pasquali 1952, 199 ff., 205 ff., 234 ff., 264 ff.; Cantarella 1969; Slater 1976 and 1982; Blank/Dyck 1984; Meijering 1985 and 1987, passim; Lasserre 1986–1987; Callanan 1987; Schenkeveld 1990a; Ax 2000. On Callistratus, student of Aristophanes, see R. Schmidt, Commentatio de Callistrato Aristophaneo, in Nauck 1848; Gudeman 1919; Barth 1984; Holwerda 1987.

Scholarship, Philology, and Grammar  

never had recourse to conjecture as an editorial option. He added to the number of critical signs in use (which at this point had become an essential part of the philological toolkit): whereas Zenodotus had only employed the obelos, Aristophanes also used the asteriskos (for repeated verses), and the pair sigma and antisigma (for two consecutive verses with the same content). He also made substantial progress in the notation of accents, and he enriched the system of punctuation that was in use. All of these improvements in editorial technique guided him in the completion of his critical editions, which were numerous and of great importance. In addition to Homer, he also edited Hesiod, and he was responsible for key innovations in the philology of lyric and drama. His classification of Pindar’s odes in 17 books is well-known, and he improved Pindar’s text by employing paragraphoi to distinguish strophes and the koronis to separate odes. His division of lyric poetry into rhythmic kola marked a decisive advance in the metrical analysis of these works. We have evidence that he also worked on the texts of Sophocles and Euripides; everyone is familiar with his hypotheseis to dramatic works, which included the essential information regarding the occasion of the first performance, the setting of the play, the characters, and the plot. It is certain that he studied comedy as well, producing at least an edition of Aristophanes, and perhaps more. It appears to be the case that, like his predecessors, Aristophanes of Byzantium did not write genuine hypomnemata. We have already seen above that he also dedicated himself to lexicography (in his work Lexeis), and that he studied grammar, in particular making important observations regarding certain regularities of inflection.77 After Aristophanes, the Library was directed by a certain Apollonius, who acquired the epithet Eidographos (“the Classifier”). His scholarly identity is something of a mystery: all we possess are a few pieces of evidence that point to his undertaking a classification of literary genres, as his nickname clearly suggests.78 Direction of the library then passed to Aristarchus of Samothrace (ca. 215– ca. 144 BCE).79 Aristarchus stands out, above all, because he wrote a substantial  77 Cf. sections 1.2 and 1.3 respectively. 78 Cf. esp. Rossi 1971, 81 f.; Pfeiffer 1968, 172 fn. 2, 184 and 210. 79 The bibliography on Aristarchus is extremely vast: for a general survey, cf. Pfeiffer 1968, 210 ff. On Aristarchus and Homer: Lehrs 1882; Ludwich 1884–1885; Pasquali 1952, 205 ff.; Erbse 1959; van der Valk 1963–1964, II 84 ff.; Severyns 1928; Schenkeveld 1970; Apthorp 1980; McNamee 1981 and also 1992; Ax 1982; Meijering 1987, passim; van Thiel 1992; Lührs 1992; Porter 1992; Raffaelli 1992 (collection of fragments with introduction and commentary). On Aristarchus and Pindar: Irigoin 1952, 51 ff. On Aristarchus and Aeschylus: Wartelle 1971, 164 ff. On Aristarchus and Aristophanes: Boudreaux 1919, 52 ff.

  Scholarship, Philology, and Grammar number of hypomnemata, the great commentaries which brought the interpretation of texts to the forefront of scholarship. He also discussed specific topics in numerous monographs (syngrammata), which sometimes had a polemical character: a sizeable number of titles and fragments are extant, and it is clear that in general they dealt with Homeric subjects. Scholars disagree on the exact nature of Aristarchus’ editorial work on Homer (for him too, Homer was the preeminent author), i.e., whether he published an ekdosis and a hypomnema together, or separately, or else perhaps only a hypomnema, in which however he outlined his opinion on what form the text should take (= ekdosis). Aristarchus added further critical signs to the ones already used by Aristophanes: characteristic of his usage were the diplé (to mark various kinds of interpretative comments and scholarly information) and the diplé periestigmene (to indicate points of disagreement with Zenodotus).80 It was Aristarchus who most thoroughly put into practice the exegetical principle by which an author is his own best interpreter — in other words, the text-critical axiom that parallels in support of an argument ought to come first, and predominantly, from elsewhere within the author’s own work.81 Armed with this theoretical guideline (corrected, when appropriate, by counterbalancing notion of the hapax legomenon), Aristarchus applied himself to the analysis of the language and the content of the Homeric poems. He believed that Homer was the author of both the Iliad and the Odyssey, and he engaged in polemic against the chorizontes, the scholars who ‘distinguished’ the author of the Iliad from that of the Odyssey. His unified investigation into all aspects of the Homeric poems — language, Realien, customs, society — led him to define ‘the Homeric’ (τὸ Ὁμηρικόν), i.e., what was specific to Homer and different from what was cyclic and neoteric (and even Hesiodic), i.e., chronologically later. In this way, an exegetical principle — which imposed a critical procedure based on the internal analysis and the analogical comparison of elements from within a given author’s text, and which could be applied to aspects of both form and content82 — led Aristarchus to a specific view of  80 On these questions, cf. section 1.1 above. 81 Pfeiffer 1968, 226 f. includes a thorough reevaluation of the famous maxim Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν, in which he maintains that its formulation should not be attributed to Aristarchus, even if its content is in completely line with his thinking. This has given rise to an extensive debate: Wilson 1971; Montanari 1974, 120 f.; Lee 1975; Wilson 1976; Schäublin 1977. F. Montanari, Aristarco e il σαφηνίζειν, forthcoming. 82 It is important to be wary of the hasty accusation, often levelled at Aristarchus, of a lack of antiquarian knowledge caused by a philological approach dedicated predominantly to textual issues; this accusation is often formulated in explicit contrast to the imposing erudition displayed by Didymus.

Scholarship, Philology, and Grammar  

literary history.83 In general, his philological activity was characterized by a prudent examination of manuscript witnesses to the text, though he certainly did not refrain from intervening via conjecture and thus making use of his extraordinary interpretative akribeia, which would become proverbial.84 Methodological rigor and a vast store of knowledge guided his extensive work on classical texts. After Homer, we are best informed regarding his work on the text of Pindar (he is cited about 70 times in the scholia, and it seems probable that he wrote a commentary on Pindar’s complete works, building upon the foundation laid by Aristophanes of Byzantium); but we also know that he studied the odes of Bacchylides, and he must have produced a commentary on them as well. We also have evidence for commentaries on Hesiod, Archilochus, and Alcman; it is certain that he produced an edition of Alcaeus, probably also one of Anacreon; and there are hints that intervened in the texts of Hipponax and Semonides and perhaps Mimnermus. As for Attic drama, we know that he worked on Aeschylus and Sophocles, whereas the evidence is less clear for Euripides; from the sources available we can conclude that he commented on at least eight of Aristophanes’ comedies, but we do not have enough information to assert that he produced a commentary on all of that comedian’s extant works. Aristarchus is, moreover, the first ancient scholar who is known with certainty to have worked on prose authors: we possess a fragment of his hypomnema on Herodotus (there is no extant indication that he studied Thucydides).85 For his editorial and exegetical activity, Aristarchus’ main predecessor was Aristophanes of Byzantium: the earlier scholar laid the groundwork for the later one’s research, and it is only because the two were so often in agreement that much less material of Aristophanes has been preserved. Even in the field of grammatical studies Aristarchus built upon the observations of his precursor: he produced a more precise formulation of linguistic and grammatical analogy86 (a concept which, as we have seen above, he also applied in his exegesis); this, along with his rejection of allegorical interpretation (to which we shall return shortly), formed the basis for his disagreement with, and his polemics against, the Pergamene philologists. At Pergamum there was never a philological tradition in the same way as at Alexandria; no school was ever formed there of comparable strength and duration. Scholarly activity did however flourish under Attalus I (241–197 BCE) in the

 83 Cf. the brief discussion in section 2 above. 84 The still-debated issue of how the Alexandrian philologists worked has been addressed in section 1.1 above. 85 Cf. section 1.1 above. 86 Cf. section 1.3 above.

  Scholarship, Philology, and Grammar figures of Antigonus of Carystus (who produced a work on painters and sculptors, assembled a compilation of paradoxography, and wrote biographies of philosophers) and of Polemon of Ilium (who composed an important periegesis full of antiquarian details and notable especially for its use of inscriptions). Polemon was surpassed, at least in the field of Homeric topography, by Demetrius of Skepsis, who lived perhaps one generation later and was the author of an enormous commentary on the Catalogue of the Trojans (Iliad II 816–877). Probably modelled upon the Alexandrian hypomnema, this work was written around the middle of the second century BCE (certainly after the time of Crates of Mallus, with whom Demetrius takes issue in one of the extant fragments), and it constituted a veritable summa of topographic and antiquarian scholarship: in total, 30 volumes of commentary on just over sixty verses! Demetrius’ work would later be consulted by Apollodorus when the latter compiled his own commentary on the Catalogue of Ships: thus, while Demetrius is cited a few times in the Homeric scholia, it is probable that further material of his has reached us without being attributed to him, through the filter of Apollodorus.87 Skepsis was a city not far distant from Ilium, a few kilometers inland along the course of the Scamander, and it appears that Demetrius lived his whole life there; given this relatively withdrawn lifestyle, his connections with the scholarly world of Pergamum cannot have been particularly strong. We are now in the time of Eumenes II (197–159 BCE): a king of notable qualities, who founded a rich library and hosted numerous scholars, thus rendering Pergamum, almost miraculously, a cultural center capable of competing with Alexandria.88 During Eumenes’ reign, then, and contemporaneous with the activity of Aristarchus, Pergamum had its true moment of scholarly prestige.89 It was in this period that the most important Pergamene philologist flourished, invited by the king and hosted in the capital: Crates of Mallus. Crates was by formation a Stoic philosopher: “In contrast to the Alexandrian scholar poets he and his pupils approached the literary heritage as philosophers, and in particular as orthodox Stoics.”90 He differentiated himself from the Alexandrian grammatikoi by defining himself as a kritikos. In his exegetical methodology, he made broad use of  87 Cf. Pfeiffer 1968, 249 ff. The fragments can be found in Gaede 1880; cf. also Schwartz 1957, 106 ff.; Gabba 1974. 88 On Pergamum, cf. Montanari 1993d = this volume ch. 14. 89 On scholarship and philology at Pergamum, see the survey in Pfeiffer 1968, 234 ff. with bibliography. Cf. also in this volume, ch. 14 Pergamo, with discussion of Crates and his students. For Crates: Wachsmuth 1860; Helck 1905; 1914; Kroll 1922; Mette 1931; 1936; 1952; 1984; Nicolai 1992, 180 ff., 198 ff., 285; Porter 1992. On Telephus: Wendel 1934a. 90 Cf. Pfeiffer 1968, 237.

Scholarship, Philology, and Grammar  

allegorical interpretation, which had originated long before and had been continued in philosophical contexts, being adopted especially by the Stoics; but it had also been vigorously rejected at Alexandria, and thus it came to represent a first cause for polemic between Crates and the Alexandrians. A second area of polemic was the theory of language: whereas Aristophanes and Aristarchus had made analogy the basis for their understanding of language, Crates followed Chrysippus in his emphasis on the Stoic concept of anomaly, which held that the free development of language as seen in concrete usage (consuetudo) could not simply be overridden by norms dictated by analogy (ratio).91 The polemics waged by Crates and Aristarchus then found fertile terrain in the field of Homeric studies. Of Crates’ Homeric scholarship, the titles of two works have come down to us (Diorthotika and Homerika) along with a few fragments containing examples of his allegorical interpretation (his cosmological reading of the shield of Achilles is well-known); these will certainly have inspired objections from Aristarchus and from the Alexandrian school. In addition to Homer, Crates dedicated his attention also to Hesiod (we have some notices of his work in the scholia), to lyric authors, to drama, and even to the Phenomena of Aratus. Thanks to the broad popularity of Stoicism, Crates’ influence in the GrecoRoman world was quite notable, and it was increased in part by his long stay in Rome in 168 BCE. The Alexandrians repudiated the allegorical method: in their view, myths constituted an integral aspect of the content of poetic works and therefore deserved to be studied as such, within the context of the broader system in which they played a part (from a “scholarly” point of view, one might say). Quite differently, for the Stoics allegory had become extremely common as a method for approaching both myth and literature; and this, naturally, could not have failed to influence scholars such as Crates and his followers, who employed allegorical readings extensively in their exegesis. Thus, for Crates, Homer was a poet who intentionally included allegory in his poetry; for Aristarchus, Homer did not intend allegorical meanings, and therefore allegory ought not to be employed when interpreting his works. In the end, therefore, we are dealing with a methodological disagreement which derives from two different views of the nature of poetry: those interpreters who apply allegory as a tool of literary criticism must, necessarily, assume that the poet himself had deliberately intended to employ an allegorical style when composing his works. Despite the Alexandrians’ opposition, the allegorical interpretation of Homer became extremely widespread — to the extent, in fact, that it would later  91 Cf. section 1.3 above.

  Scholarship, Philology, and Grammar be extensively included in the margins of manuscripts containing every sort of commentary on the Homeric poems. As examples of extended allegorical exegesis, we may mention here the Homeric Problems attributed to a certain Heraclitus and dated to the beginning of the Imperial period, as well as passages from the writings of Porphyry. But even in such a work of literary criticism as the anonymous On the Sublime, we read the following remark apropos of Homeric divinities and their actions: “... these things are blasphemous and indecent, unless understood allegorically”. Allegory was, then, a profoundly entrenched idea, with a history that spanned several centuries. It could rely on a basic conception that was, in essence, extremely clear: the opposition between, on the one hand, a literal meaning which was morally bad, and on the other, a deeper meaning which was morally good. This permitted allegory to develop over the centuries into a rich tradition, which found its way into the works of the pagan philosophers of Late Antiquity; and the Christians, too, made use of allegory, exploiting it as a precious tool by which to appropriate the patrimony of the ancient world, but also to keep it under control, and to construct an idea of continuity between past and present.92 As I have mentioned above, it is not really legitimate to speak of a “school” at Pergamum, because we do not have evidence for a regular series of teachers and students. That said, the intellectual descendents of Crates continued to engage in controversy with those of Aristarchus (regarding the topics discussed above), even if at times there were signs of rapprochement, for instance in the field of grammar. Here, the Pergamene context later produced another fascinating figure, Telephus, who lived in the second century CE, and was therefore a contemporary of Galen — the most illustrious figure of Pergamene origins in all of intellectual history.

After Aristarchus In addition to his own scholarship, Aristarchus is rightly famous also for having produced a series of students who continued his style of scholarship.93 Unfortunately, of their works often only meager fragments have survived, and in this space we can only mention most of them very briefly. The two most important

 92 The bibliography on this subject is notoriously vast, and it seems pointless here to include a long list. For an introduction to the issue, one may consult Pépin 1976, which contains a rich bibliography at pp. 517–548. Cf. also Hahn 1967; Perret 1982; Steinmetz 1986; Lamberton 1986. 93 On Aristarchus’ students, see Blau 1883; cf. Pfeiffer 1968, 252 ff.

Scholarship, Philology, and Grammar  

were, without doubt, Apollodorus and Dionysius Thrax, both of whom were involved in the political crisis of the years 145–144 BCE related to Ptolemy VIII’s rise to power. Both are figures of great importance in the history of philology. They still belong entirely within the Aristarchan tradition, and their work marks major developments in the cultural debates of their epoch. Apollodorus of Athens (ca. 180–ca. 115 BCE)94 is notable for the unique richness of his intellectual experience (even if we cannot be unequivocally certain of all the facts that have been reconstructed from the evidence). At Athens he was the student of the Stoic philosopher Diogenes of Babylonia; at Alexandria he was the student and colleague of Aristarchus; and then it seems that he first fled to Pergamum and finally returned to his native Athens. Apollodorus was the true successor of Eratosthenes in the field of chronology: his Chroniká, which began with the fall of Troy and went down to his own time, was written in iambic trimeters — a decision clearly made in an effort to facilitate memorization. Although his work was naturally based on that of his predecessor, he made significant modifications and he attempted to establish many dates on the basis of more authoritative evidence; his work also contributed notably to constructing the basic chronological outline that stands as the foundation of literary history. Apollodorus also wrote an extensive treatise entitled On the Gods, in which he carried out a rigorous analysis of religious deities, studying their epithets in particular detail and employing etymology as an interpretative tool to this end. (Another treatise of his was dedicated entirely to etymology; here, it is difficult not to see a Stoic influence at work, as is reasonable given his background.) On the Gods was founded upon a solid understanding of epic poetry and of the representations of the gods found therein. Another work on Homer — an enormous exegetical effort — was his On the Catalogue of Ships, which built upon both Eratosthenes and Demetrius of Skepsis in addition to Aristarchan philology. In this work the essential problem, of a historical and a literary nature, was the geography of the Homeric epoch, with all that this implies in terms of both general questions and the innumerable problems relating to the interpretation of individual place-names. Lastly, Apollodorus also produced important work on Doric comedy: he studied Epicharmus and the mimes of Sophron. It is worth emphasizing95 that the origins from which all of this research sprang, and the foundations upon which it was constructed, are to be found in Apollodorus’

 94 Pfeiffer 1968, 253 ff., which contains the essential bibliography, to which however one may add Theodoridis 1979. 95 As does Pfeiffer 1968.

  Scholarship, Philology, and Grammar combination of the roles of philologos and grammatikos — essentially, interpreter of texts and scholar of literature. Notwithstanding his nickname (actually related to his father’s name), Dionysius Thrax (ca. 170–ca. 90 BCE) was from Alexandria.96 He received his intellectual formation there during the last years of Aristarchus’ teaching. He dedicated his attention above all to Homer, and following the example of his teacher he produced hypomnemata and monographs (such as the polemical pamphlet Against Crates). He also worked on Hesiod, and perhaps Alcman. In addition, we possess a few fragments of writings of a grammatical nature. But in this latter field Dionysius Thrax is renowned above all for the problematic Techne grammatiké, which has been mentioned above; if it is his work, it would be the first treatise of an Alexandrian scholar transmitted down to our time in more or less complete form.97 Dionysius fled from Alexandria — perhaps in 144 BCE, in connection with the political events that shook the city and the royal court — and established a school at Rhodes. One of his students there was Tyrannion, who from 71 BCE lived at Rome, where it was his fate to be the first to examine the writings of Aristotle that had been brought from Athens by Sulla; a complete edition would be carried out in the next generation, by Andronicus of Rhodes. Both Tyrannion and his contemporary Philoxenus wrote brief treatises on the Latin language. Among the latter’s works we also find commentaries on Homer, but his main importance was in the field of grammar: here he produced a work on the monosyllabic verbal root, and in general his research is characterized by a fondness for etymology and for the principle of analogy. Another significant figure is Asclepiades of Myrlea (in Bithynia, 1st cent. BCE). It is not clear if he was a student of Dionysius Thrax; but we know that as a scholar and philologist he was active at Rome and then in Spain. In his exegetical writings on Homer one can detect an influence deriving from scholars associated with Pergamum, especially Crates. He also studied Theocritus and Aratus, and in addition to investigating issues of grammar himself, he composed a work entitled On the Grammarians.98

 96 Linke 1977. For the Techne grammatikè, the fundamental edition is that of Uhlig 1883; Italian edition by Pecorella 1962; new edition by Lallot 1989. Cf. Di Benedetto 1958; 1959; 1973; Fuhrmann 1960; Pfeiffer 1968, 266 ff.; Erbse 1980; Di Benedetto 1990; Nicolai 1992, 186 ff.; and see also section 1.3 above, with fnn. 39–40. 97 Cf. section 1.3 above. 98 On Tyrannion: Pfeiffer 1968, 273 f.; Haas 1977. On the beginnings of the philological study of the text of Aristotle: Moraux 1973; 1986. On Philoxenus: Pfeiffer 1968, 273 f.; Theodoridis 1976. On Asclepiades of Myrlea: Müller 1903; Pfeiffer 1968, 272 f.

Scholarship, Philology, and Grammar  

In the field of grammatical studies one should also mention: Tryphon of Alexandria, who lived in the Augustan era; Alexion, active during the first century CE; and Ptolemaeus of Ascalon, whose date is uncertain.99 These are the scholars whose work leads recta via to the great systematization of Greek grammatical doctrine carried out in the second century CE by Apollonius Dyscolus and Herodian, whose work, in both cases, is still testified by significant remains.100 Apollonius Dyscolus lived in Alexandria, apart from a brief sojourn in Rome. He wrote numerous works in all areas of Greek grammar and dialectology. Of this work, we possess treatises on the parts of speech, especially pronouns, adverbs, and conjunctions; he also studied nouns and verbs. It is worth emphasizing in particular his treatise on syntax, since Greek grammatical studies were for the most part dedicated to phonetics and morphology, whereas syntax was discussed only rarely; for this reason, Apollonius Dyscolus’ syntax was highly influential for later scholars, in part filtered through Priscian. Apollonius found a worthy successor in his son Aelius Herodianus, who was active in Rome in the time of Marcus Aurelius (161–180 CE). Herodian is remembered above all for his great work on the general principles of accentuation (Katholikè prosodia) and his research into problems of prosody in Homeric language (which is known via copious excerpts present in the scholia); but he also wrote on various other topics, such as orthography, verbal and nominal inflexion, and other grammatical issues. After a few generations of Aristarchus’ students, the tradition of research that had its main focus in textual criticism, exegesis, and scholarly commentary culminated in the colossal scholarly production of Didymus. Didymus was active in the second half of the first century BCE in Alexandria, and already in antiquity he was famous for the enormous quantity of monographs and commentaries he wrote, dedicated to a vast range of authors. His work, overall, can be considered a massive reservoir, collecting and organizing the results of earlier research and transmitting them to future compilations (which eventually will flow into the various corpora of scholia). Didymus was perhaps a fairly unoriginal thinker (as is often stated), but he conducted his research with extreme care and

 99 For Tryphon: De Velsen 1853; Wendel 1939; cf. also West 1965. For Alexion: Berndt 1906. For Ptolemaeus of Ascalon: Baege 1882. 100 Cf. section 1.3 above. For Apollonius Dyscolus: edition by Schneider/G. Uhlig 1878–1911; cf. Thierfelder 1935; Erbse 1960, 311 ff.; Householder 1981; Blank 1982; Sluiter 1990, 39 ff. For Herodian: edition by Lentz 1867–1870; cf. Lehrs 1848; Egenolff 1894; Dain 1954; Schulz 1912; Erbse 1960, 344 ff.; van der Valk 1963–1964, I, 592 ff.; Hunger 1967; Wouters 1979, 216 ff., 231 ff.

  Scholarship, Philology, and Grammar scrupulousness.101 More or less contemporaneous to Didymus was Aristonicus, who dedicated much scholarly effort to explaining Aristarchus’ use of critical signs in the text of Homer, thus acting as a kind of mediator between his contemporaries and his intellectual predecessor. Much of this work can now be found in the scholia to Homer; but Aristonicus also studied Hesiod and Pindar and wrote several monographs.102 In these same decades flourished Theon of Alexandria, a commentator who dedicated himself not only to the canonical classical authors, but also to Alexandrian poets such as Callimachus, Theocritus, Apollonius of Rhodes, and Nicander. We now know that he was not the first to do so (he was preceded at least by Asclepiades of Myrlea, for instance), but it is certain that his work enriched philological scholarship by making major advances in the study of Hellenistic poetry.103 Ties with traditional exegesis are also found in the Homeric lexicography of Apion and Apollonius Sophista (who lived in the first and the second halves of the first century CE, respectively), which was nourished by the results of Aristarchan philology.104 In the second century CE, scholars continued to produce significant contributions to philology. Nicanor developed a sensitive system of punctuation and is known above all for the writings in which he discussed, specifically, the punctuation of the Homeric text (notable excerpts are found in the scholia). Hephaestion became the most competent and authoritative scholar in the study of meter. Valerius Harpocration was the author of a Lexicon of the Ten Orators. And we have already mentioned, if briefly, Galen, the doctor and philosopher who also produced commentaries and treatises on topics ranging from literary criticism, to linguistics, to rhetoric.105  101 Cf. section 1.1 above. A general survey can be found in Pfeiffer 1968, 274 ff., with essential bibliography; an edition, by now extremely obsolete, by Schmidt 1854. The most remarkable addition to our knowledge is that represented by the papyrus containing commentary on Demosthenes, mentioned in section 1.1 above; cf. also S. West 1970; van der Valk 1963–1964, I, 536 ff. 102 For a general survey, see L. Cohn, in «RE» II 1, 1895, coll. 964 ff. For his Homeric philology: Friedländer 1853; Carnuth 1869; cf. also van der Valk 1963–1964, I, 553 ff. 103 Guhl 1969; Montanari 1974, 119 ff. 104 Cf. section 1.2 above. On Apion: Neitzel 1977. A Homeric glossary has come down to us, falsely attributed to Apion but containing some genuine portions, and has been published by Ludwich 1917 and 1918 (= Latte/Erbse 1965, 283 ff.). On Apollonius Sophista: Bekker 1833; Steinicke 1957 (new edition covering Α to Δ); Henrichs/Müller 1976. 105 For Nicanor: the fragments may be found in FGrHist, n. 628, with commentary; cf. Wendel 1936a; Blank 1983. On his Homeric philology: Friedländer 1857; Carnuth 1875. On Hephaestion: the original treatise on meter in 48 books was reduced by the author himself into the Encheirid-

Scholarship, Philology, and Grammar  

The third century CE marks a decline in original scholarship; but in this period various compilations were assembled, which are precious for the earlier material they contain. Precious too are the miscellaneous works of the period, such as the Deipnosophistai (“The Sophists at the Banquet”) by Athenaeus of Naucratis (2nd–3rd cent. CE), which preserves innumerable bits of information on literature, grammar, and Realien (Athenaeus cites over a thousand authors and titles of works and more than 10,000 lines of poetry). Also in this era, Diogenes Laertius (3rd cent. CE) produced his Lives of the Philosophers in ten books. These last two works are marginal to the main stream of literary scholarship and philology; more directly relevant were the writings of Porphyry (233–ca. 305), such as his Homeric Questions and his allegorical interpretation of the cave of the nymphs in the Odyssey, not to mention his smaller treatises (unfortunately lost) on grammatical and rhetorical topics.106 At this point we have reached the threshold of Late Antiquity. For the vast majority of works of scholarly literature (commentaries, critical monographs, lexicographical collections, paroemiographical compilations, grammatical treatises) this was the period in which began the arduous process of selection, modification, and redaction, which would lead to the great Byzantine compilations. And it is precisely from these Byzantine compilations that we derive most of our knowledge about the scholarship of the preceding centuries.

 ion which has come down to us; the edition of Consbruch 1906 also includes the ancient scholia to the text; cf. Hense 1912; van Ophuijsen 1987. On Harpocration: Keaney 1991; Cunningham 1986. On Galen, cf. section 1.1 above. 106 On Porphyry, cf. section 1.1 above.

 Ancient Scholarship and Classical Studies The definition of ‘ancient scholarship’ includes numerous phenomena that belong to the literary civilization both of the ancient Greek and the Latin world. However, I should start by stating that the focus of my attention refers to ancient Greek culture, although experts will have no difficulty transferring to Latin culture many of the concepts we will be dealing with. The term ‘scholarship’ refers first and foremost to all written works that aim specifically and directly to provide an interpretation of the literary works on various levels. Thus in this sense ‘scholarship’ refers in the first place to the different forms of commentary on the texts (from the extended ὑπόμνημα to short annotations) and to exegetic treatments of a monographic nature (συγγράμματα). For this set of works, it is appropriate to take into account the difference between a syntagmatic and a paradigmatic procedure in the treatment of the text or texts to be interpreted. But one immediately realizes that the term ‘scholarship’ also covers many other genres: in primis the impressive phenomena of lexicography and linguistic-grammatical studies. Thus the numerous forms of lexicographic works come to mind, ranging from the most ancient collections of glosses and λέξεις to the great Etymologica of the Byzantine age, through a long history of materials traced, studied and ordered. To this category there also belong various collections of linguistic data and of Realien classified by lexical form (including the onomastics). Equally lively is the broad array of studies on language: reflections on human language, springing at first from the sphere of philosophy (testified by Democritus, perhaps started even earlier), were further developed in the cultural framework of Alexandrian scholarship, where they gave rise to an independent linguistic-grammatical science. In the work by Aristophanes of Byzantium and Aristarchus the solid basis was already laid for the subsequent extraordinary flowering of this field, which would culminate a few centuries later in the epoch-making work of Apollonius Dyscolus and Herodian. And the shelf ‘scholarship’ can be expanded to include by no means accessory materials, such as studies on prosody and metrics, biography, paroemiography, mythography, not to mention the investigations into the form of the ancient book, with the peculiarities represented specifically by phenomena of graphic layout linked to ‘scholarship’. The vast field of rhetoric must also be taken into account, at least to the extent that rhetorical works may be found to include texts of different authors, such texts being used, quoted and interpreted. Nor  English translation by Rachel Barritt Costa. https://doi.org/10.1515/9783110772371-002

Ancient Scholarship and Classical Studies  

should one overlook the reflections on poetics, another theme that deserves a passing mention here. Indeed, Aristotle’s Poetics focused specifically on the poetry of the previous centuries, to which the Alexandrians later devoted themselves with a profusion of effort and means. The Alexandrian philologists, and not only those among them who were poets in their own right, could hardly have neglected theoretical thought on poetry: in my view, it would be absurd to imagine that Aristarchus had no ideas of his own on poetics (we will return to this point later), while it is far too obvious to point to the fact that only very scanty evidence on theoretical and epistemological questions can be traced in the philological-exegetic fragments that have come down to us, as such fragments are of quite a different kind and have quite a different purpose. The kind of research and treatises on the poets and their works that was widespread in Peripatetic circles, and the thought expressed by Aristotle in the Poetics, are the concrete intellectual and cultural precedent of the scholarship of the Hellenistic age, and made a crucial contribution to encouraging a fresh approach to literature, in primis poetry: it was through this new vision that poetry became a privileged object of investigation and interpretation, of care and conservation. There had arisen an awareness of standing on the shoulders of an era that had forged decisive and influential cultural models, great poetic creations that embodied an overwhelming portion of the Greek παιδεία, and this treasure had to be understood, known, and handed down.1 The resulting picture is that of a comprehensive panorama, populated by a wide variety of remarkably diverse products, from treatises displaying profound critical-literary insight to the most unadorned and bare collections of erudite materials. However, there is a fundamental element that is common to the entire range of this variegated archipelago: namely, that the overwhelming majority of the contents present in works classifiable as ‘scholarship’ derive essentially and materially from the immense store-house consisting of literary and linguistic phenomena. That is to say, the contents of ‘scholarship’ depend crucially on the overall body of works composed by the authors of literature and on the phenomenon of language as the common tool utilized for their creation. Thus on the one hand we have, in the actual and direct sense, the edition, conservation and interpretation of texts (i.e. ecdotics and exegesis), while on the other we have the study of language phenomena (grammar, lexicon, rhetoric),  1 An extensive reconsideration of these problems was made on the occasion of the XL Entretiens Hardt, held in 1993; cf. Montanari 1994a. Its starting point was the proposal to rethink, a quarter of a century on, the themes addressed in Rudolf Pfeiffer’s by now classical treatment, dating from 1968. Another fifteen years have gone by since that moment.

  Ancient Scholarship and Classical Studies which are the tool of literature. But the two spheres are organically linked: an understanding of the texts is indispensable in order to utilize them for purposes of describing and explaining linguistic and rhetorical phenomena, but at the same time an understanding of linguistic and rhetorical phenomena is indispensable for a good and satisfactory exegesis of the texts themselves. This is true, for instance, even for biography, which is based mainly on the interpretation of the statements made by the authors themselves in their compositions, but it is just as true for erudite genres such as paroemiography and mythography. The figurative arts and the observation of real life may also constitute important sources of information, yet they too are often filtered through a literary or paraliterary form and so they give rise to commentaries and interpretations, enriching in this way the contents of the works of scholarship. ‘Language‒Text‒ Literature’ are the keywords which rightly inspired the conference. Over the course of history the phenomena of ‘scholarship’ became considerably diversified, assuming different functions. Thus on the one hand scholarship presented the characteristics and fulfilled the function of direct interpretation of texts, i.e. a variably broad and organic discourse offering further developments on a pre-existing discourse or at least partly based on and motivated by it. In other cases scholarship presented the characteristics and fulfilled the function of a repertory of reference and consultation materials, at times with encyclopaedic breadth and pretensions. This theoretical distinction is of help in understanding the products by abstract categories, although in the concrete reality of the works the balance was more likely to be tipped in one of the two directions, but without completely excluding the other. However, to give a perhaps somewhat brutal example, there is an evident difference as regards the aim and treatment of materials between a study with contents of a criticalliterary nature devoted to and based on one or more texts, versus a general lexicographic collection in alphabetical order. By contrast, the differentiation appears to be less marked when a comparison is drawn between a commentary or a series of annotations on a poetic text versus a specific glossary of that same text which follows its exact wording and order. Nor should it be overlooked that one of these two characteristics and functions may have developed secondarily compared to the original intentions of the composition, as a result of the different ways a work of ‘scholarship’ happened to be used on different historical occasions. For example, a commentary supplied with a capitulatio by topics (such as the commentary or monograph of Didymus on Demosthenes preserved by a Berlin papyrus, P.Berol. 9780) could be utilized through a consultation by subject, even independently of the text to which the commentary referred. In contrast, lexicographic works of general scope (like the major lexicons and the

Ancient Scholarship and Classical Studies  

great encyclopaedic dictionaries) may have served a useful function in exegetic commentaries on various texts and may thus be abundantly utilized within ὑπομνήματα or scholiographic corpora. It would be interesting — although beyond the scope of this paper — also to consider the relation between antiquarian studies and historiography, in order to examine how far, and with what differences, this may constitute a parallel for the arguments being put forward here. The terms of comparison could be as follows: does the relation between the erudite and well ordered collection of materials and critical (philological-exegetic) interpretation or theoretical (poetic) reflection in the field of scholarship stand parallel to the relation between the collection of antiquarian data and of Realien of all kinds and historiographic reconstruction or theoretical reflection on history? In a slightly different sense, albeit not too far-removed on a plane of pronounced abstraction, it can be noted that in the case of Aristotle and the early generations of the peripatetic school the treatises on individual authors and works of poetry are to the theoretical reflections of the Poetics as the collections of data constituted by the πολιτεῖαι are to the theoretical reflections of Politics. But all this would distract us excessively from our basic theme. Works of a specifically exegetic character, such as a commentary or a monographic treatise, were naturally influenced by the contents of the text that was the object of interpretation, from which, however, they may diverge considerably, showing a notable degree of independence in content and arguments. Since interpreting authoritative texts was a highly favoured working practice, the literary forms taken by exegesis as well as the occasions prompting an exegetic attitude were strikingly varied. Passages of an exegetic character may be present in works of a different genre or objective: for instance philosophical treatises may include considerations on previous works dealing with the same theme or may focus on the interpretation of passages from literary authors. The stratification may be twofold: it may on the one hand involve cases of a comment on a comment, as occurs when Aristotelian commentators find within the text of Aristotle commented citations of previous authors; on the other hand, an exegete may quote a parallel passage and also add a particular comment upon it. The element all such works have in common is that their subjects and arguments are built on a previous subject and argument, the text is constructed on another text, the importance and authoritative nature of which is known and recognized. A very general definition of such works could be ‘text on a text’ or ‘text about a text’. One need only think of the enormous literature produced on Homer over the centuries to gain an eloquent picture of what I am conveying. But illustrious and grandiose examples are also to be found in philosophy (suffice

  Ancient Scholarship and Classical Studies it to mention the quantity and value of the commentaries on Plato and Aristotle), in science (the Corpus Hippocraticum and Galen) and in religion (the case of the Bible is emblematic). Even more than in other sectors of the literary civilization, scholarship can be identified as a cultural phenomenon in which literature grew directly on literature and drew its nourishment from itself, increasing over the course of history by virtue of its necessary and inescapable reference to previous exegetic products as well as to the texts that were the object of interpretation. Particularly important is the relation holding between such activity of commenting and interpreting literary works and the reference canon: the influence is reciprocal because scholarship on the one hand certainly reflects an acquired canon, since it deals with the things that belong to it, while on the other hand scholarship contributes to shaping and consolidating the canon, exerting a non secondary influence on the image a culture acquires over time. If we use the concept of ‘text on a text’ or ‘text about a text’, then the attempt to give a specific definition of certain exegetic products becomes less dramatic, while remaining important. This is the case, for instance, of fragmentary texts that one hesitates to label as ὑπόμνημα or as σύγγραμμα. Under the skies of Thessaloniki the Derveni Papyrus immediately comes to mind. This is certainly an exegetic text that deals with and explores another text, namely an Orphic cosmogonic poem, though it is not clear whether the author is proceeding in syntagmatic order as would be expected in a commentary, or in paradigmatic order like in a treatise. Nevertheless, it is definitely an argument that starts out from another argument, and its religious aim is pursued by exploiting the approach and techniques characteristic of text interpretation, which thus set it in the framework of scholarship. The religious sphere suggests a parallel with Christian homilies, which present a doctrinal argument developed with varying degrees of independence starting out from interpretation of passages from the Old and New Testament. Accordingly, this suggestion can be expanded to include the biblical catenae, which consist of collections of extracts taken from different sources, including the homilies themselves, and gathered together for exegetic purposes. Note in addition that the catenae existed both in the form of independent commentaries and also as annotations in the margins of the text being commented on. There is a compelling parallel with the form-content relationship and the exchange of materials between the ὑπομνήματα and the συγγράμματα found in the Homeric scholarship of the Alexandrian grammarians, as is well documented in the sources. The considerations put forward so far on ancient scholarship prompt the introduction of another highly significant distinction, despite the difficulties and epistemological uncertainties with which the path is certainly strewn. A

Ancient Scholarship and Classical Studies  

commentary and its arguments can have an aim which could be defined as philological-grammatical-historical, namely that of an exegesis designed to explain the intrinsic and genuine content of the text properly speaking, and to discover what it really meant and sought to express in its authentic and original linguistic and historical-cultural context. In this perspective, scholarship appears as a science that is independent with regard to its ends, and its exegetic aim is important and valid in its own right and not as part of an overarching purpose. The objective is to explain the text without any other motive than that of correctly reconstructing and preserving historical evidence, which has its own intrinsic value. But in a different perspective, a commentary starting out from an authoritative text may present an interpretation that seeks above all to put forward the interpreter’s own arguments, focusing on specific doctrinal points, often with the ideological intention of ‘enlisting’ the author who is the object of analysis. The interpreter utilizes the work as a starting point, not infrequently seeking to unveil meanings that may have remained implicit or even ‘hidden’, and if necessary putting forward the possibility of uncovering important meanings the author did not consciously intend to include. Ultimately, this went as far as implying that it is legitimate to attribute to the author meanings that are generated and motivated only by the cultural and critical history that has been built up around the work during its reception. A classic case, which happens in fact to be the first of which we have knowledge, is the analysis of the song of Simonides, carried out by Socrates, and Protagoras in Plato’s Protagoras (339a ff.). What is also significant is that this experiment was prompted by the sophist Protagoras. For it was precisely the Sophists, at least in their representation by Plato, that claimed to be the representatives of a practice that had always existed, but which had only recently come to light: the wise of earlier times, namely the poets, were likewise σοφισταί, except that they ‘masked’ their wisdom so as to avoid being envied by the overwhelming majority. The mentality seems to be precisely that of an interpreter, who is capable of uncovering the meanings hidden in the authoritative texts because he possesses the tools to do so. What we have here is a duality which comes to the fore throughout the history of the exegesis of texts recognized as important, a duality which displays features at times distinctive and spectacularly brandished as banners of a particular approach, at other times rather faint or even uncertain: features and elements in many cases intermingled and overlaid, in other cases difficult to disentangle and often the object of sharp polemics. Once again, it should be underlined that strictly speaking this is a theoretical distinction, an intellectual tool helpful in understanding phenomena through abstract categories, whereas

  Ancient Scholarship and Classical Studies in concrete reality pure and distilled exemplars are a much rarer occurrence. Rather, it is more usually the case that the balance is tipped more or less decidedly towards one of the two aspects, but without the other being necessarily absent or excluded, perhaps making itself felt in the facts and the results more than in the declarations and intentions. There should be clear recognition that the boundaries between the two approaches are very problematic, sometimes labile, and above all debatable and debated. The concrete contrapositions and intersections between the two modes of critical discourse and the attendant polemics give rise to the irregular and multiform evolution of the reception of the great works and the great authors. A fine and much cited declaration by Aristarchus, preserved in a scholion to Iliad 5.385, seems to be almost a classroom case designed to illustrate the concept, starting from ancient criticism. In the passage of the Iliad reference is made to the episode of the aggressive brothers Otos and Ephialtes, who were said to have imprisoned no less than the god Ares. Aristarchus warns that the statement by Homer should be taken as mythical content springing from poetic license, and one should not try to excogitate contents that are alien to the poet’s words, or to invent something he did not say. The Aristarchean observation is followed by the exposition of the myth to which the cited episode belongs, evidently with the aim of thus providing the mythographic explanation of the brief allusion present in the poetic text. Then the scholion introduces three interpretations of a completely different nature. The first, defined as ‘more convincing’, sees the myth as a misrepresentation of historical facts that did really occur; the second interprets it as a physical allegory concerning astronomic questions; the third (the most extensive) broadens out into a complex philosophical exegesis having ethical content. The latter interpretation appears to be the one preferred by the author of this doxography, since he introduces it by saying it is better to maintain that at this point the poet wants to φιλοσοφεῖν, as in fact is the case in the whole poem.2 We will not dwell in detail here on these three forms of allegoresis, which effectively belong to the most well-known and widespread typologies: they are interpretations which have in common the idea that the real meaning of the words of the text is different, is ‘something other’ than what a strictly

 2 Sch. D ad Il. 5.385: Ἀρίσταρχος ἀξιοῖ τὰ φραζόμενα ὑπὸ τοῦ ποιητοῦ μυθικώτερον ἐκδέχεσθαι κατὰ τὴν ποιητικὴν ἐξουσίαν, μηδὲν ἔξω τῶν συμφραζομένων ὑπὸ τοῦ ποιητοῦ περιεργαζομένους. Ὦτος οὖν καὶ Ἐφιάλτης γόνῳ μὲν ἦσαν Ποσειδῶνος, [...]. ἄλλοι δὲ πιθανώτερον ἐξηγούμενοι τὰ Ὁμήρου τοὺς Ἀλωείδας τούτους φασὶ γενέσθαι βασιλεῖς τῆς Ἑλλάδος. [...]. οἱ δέ φασιν, ἐν τῇδε τῇ ἱστορίᾳ περὶ τῶν μετεώρων διαλέγεσθαι τὸν Ὅμηρον, μαθηματικὸν ὄντα [...]. βέλτιον λέγειν δὲ ὅτι φιλοσοφεῖν βούλεται διὰ τῆσδε τῆς ῥαψῳδίας, ὡς καὶ δι’ ὅλης τῆς ποιήσεως.

Ancient Scholarship and Classical Studies  

philological method can derive from the author’s words, which, Aristarchus contends, should instead be adhered to more strictly. The above distinction has the consequence that the choice of exegetic method requires the critic to take an important and decisive position on the author’s poetics (whatever the degree of awareness in this regard). Only the conviction that the author deliberately intended to express his thought by allegory, symbols, metaphors, enigmas, obscurities of sundry types, justifies the interpreter’s adoption of decoding techniques and tools (accepted and shared or completely idiosyncratic) designed to highlight meanings attributable to the author and purportedly introduced by the author himself into his works in a veiled, or even intentionally obscure and concealed manner. To take an example, one may contrast the Homeric interpretation allowed by Aristarchus and the Alexandrian grammarians (nothing that goes beyond the words of the poet) with that abundantly represented in the Homeric problems written by one Heraclitus of the imperial age. Heraclitus believed that Homer wrote his poem using allegory as a means of expression, making deliberate use of this rhetorical tool: that is to say, the act of decoding the poetic creation enables the interpreter not only to defend the poet against any charge of impiety, but also to discover in his lines an enormous quantity and an extremely elevated quality of knowledge and doctrines, which became common knowledge later on. Techniques of this kind, or similar procedures, led to the idea of Homer as the source of all learning and all knowledge, widely pursued by ancient critics whose attitude was considerably different from that of Aristarchus. Aristarchus himself believed that Homer did not express his thought by means of allegories: thus he had no intention of attributing to Homer forms of knowledge which, for strictly historical reasons, the poet could not possibly have had; rather, he saw Homer merely as availing himself of the poetic licence to construct legendary stories based on the matter contained in myth. Before him, Eratosthenes had said the poet should concern himself with delighting the spirit and not seek to educate in some discipline or other; and even earlier Aristotle theorized that the art of poetry is not required to ensure correctness in reference to any τέχνη that happens to be the object of representation: poetry can say and teach true things with regard to a different sphere of knowledge, but it is not necessarily obliged to. One further observation on a related question. We should maintain a careful distinction between the idea of allegorical interpretation of the mythic content of a literary work, and the allegorical interpretation of the myth as such. The fact that the myth constitutes a semiotic system of an allegorical character, which conveys its meanings through this semantic tool, is a problem that concerns religious-anthropological studies, whereas what a poet intended to convey by

  Ancient Scholarship and Classical Studies utilizing mythic content is a problem of exegesis and text interpretation. This is a delicately problematic point, but the two planes should not be confused, nor, much less, should they be identified, however intertwined or indeed overlaid they may come to be in actual practice. Moreover, it should also be clearly borne in mind that the problems pertaining to allegorical expression-interpretation do not concern merely the mythic content, even though myth is typically the sphere most intensely involved. One need only mention political allegory, such as the celebrated allegory of Alcaeus’ ship, to realize that the overall issue of allegory goes beyond the problem of interpreting the mythic content of poetry. The twofold concept of allegory/allegoresis already had a far wider scope even in antiquity, let alone in subsequent eras, from late antiquity and throughout the medieval age in the Orient and in the West. Over the course of history, the very term ‘allegory’ itself has covered an extremely broad and variegated range of meanings, both as regards the rhetorical-expressive mean and also the critical-exegetic tool. Thus consideration of ‘allegory’ in the strict sense should be expanded to consideration of all types of interpretation that seek real or supposed ‘hidden’ meanings beneath the letter of the text. But let us not yield to the temptation of casting the net of our enquiry too wide, which would draw us towards theory-oriented ambitions both on literary criticism and poetics. Rather, let us restrict ourselves, almost playfully, to citing a passage from Alcibiades II, Platonic if not by Plato, in which it is asserted that all poetry is by its very nature αἰνιγματῶδες and cannot be understood by a mere ordinary person,3 and then let us set this passage alongside the declaration by Northrop Frye4 that ‘it is not often realized that all commentary is allegorical interpretation, an attaching of ideas to the structure of poetic imagery’. The exegete stands in contrast to the ‘mere ordinary person’ and is seen as the one who is capable of understanding, who has the cultural and intellectual tools to perform the interpretation; furthermore, he attributes to his work and to himself the right to extract all kinds of meaning from a text and to construct his own line of argument, focusing on what he regards as useful and important to develop. The exegete not only enjoys great freedom (which can go as far as arbitrary discretion), but he also wields potentially enormous cultural influence. So powerful can his influence be that he can sway widely held opinions and shape general attitudes, all the more so if the text he is dealing with is recognized as a highly authoritative work and if the interpreter himself enjoys great authority.  3 Alc. 2, 147b 5–10. 4 Frye 1957, 89 (my attention was drawn to the passage by Filippomaria Pontani).

Ancient Scholarship and Classical Studies  

The critical currents linked to philosophical, political and religious ideologies belong to this general and generic framework, in which even the most blatant anachronisms are present and are in effect admitted. The ancient interpreters could tenaciously trace in Homer numerous forms of knowledge and discoveries actually belonging to later ages, and they had no hesitation in attributing to Homer references to historical events, or elements of Platonic or Stoic or Christian thought, leaving aside the mordant observation by Seneca (Epist. 88), according to whom the evidence that Homer was not a philosopher resides precisely in the fact that his works are supposed to contain the entire gamut of doctrines, even though in contradiction to one another. It was, and it is, better to have an open door through which to enlist within one’s own army any text viewed as authoritative and important. History right up to the modern times is studded with manifestations of this attitude and these cultural operations (and each of us must decide individually whether to consider them acceptable or not, and to what extent). The crystal-clear methodological position in Homeric scholarship proposed by Aristarchus provides a fairly specific characterization of Alexandrian philology, but it does not represent the most common and widespread situation of criticism over the course of the centuries. Basically, allegoresis or any form of ideological, philosophical or religious interpretation has only rarely been excluded: in fact, I believe that it can be said for Homer that after the Alexandrians such a kind of exegesis was no longer excluded, right up until the modern age. Again, under the skies of Thessaloniki, we can cite the case of the archbishop Eustathius, whose lengthy Homeric commentaries gathered up the philological heritage of the Alexandrian grammarians but combined it with sweeping allegorical interpretations, capable of forging a link between Christian culture and the most venerated monument of ancient pagan culture. The methodological choice adopted by Aristarchus in the field of Homeric criticism rests on the idea that the exegetic method utilized by the interpreter must correspond to the author’s poetics, i.e. the instrument must be in tune with the object. Careful reflection suggests that this theoretical aspect accords well with the principle of internal and analogical criticism which holds that the author himself is his own best interpreter (no matter what was the exact formulation and to whom we owe it). This principle is recognized as essential and basic in Aristarchean exegetical practice. The picture that emerges is one of an intellectual framework of historical-philological exegesis, wherein studying and understanding a text is important and valid in its own right, and the objective is to correctly reconstruct and preserve a piece of historical evidence which has intrinsic value (for Homeric criticism, the nature and character of τὸ Ὁμηρικόν).

  Ancient Scholarship and Classical Studies Such an objective stands in opposition to any critical methodology that fails to respect these two principles, i.e. that admits the possibility of applying to some work a method that is unsuited to its poetics, or condones the idea of seeking within the work meanings that are quite alien to it in terms of cultural background, spatial location or time. All this identifies Alexandrian philology as a sort of ‘parenthesis’ in the history of ancient scholarship, since the predominant attitude throughout the long history of ancient scholarship right up to the modern age has been marked by critical attitudes of a sharply different orientation, according to the typology outlined above. It would certainly be of great interest to analyze this distinction and contraposition in modern scholarship from the Eighteenth century to the present day, but I am happy to avoid venturing into a province that is not my own, at least in this moment. Certainly, one can readily understand that within this parenthesis the practice of producing new editions of the works of the ancient παιδεία originated and became consolidated. In fact the Alexandrian scholars introduced a new idea in the sphere of scholarship, formulating for the first time the problem of the correctness of classical authors’ texts as they could read them, and addressing the question of the damage the ancient texts had suffered in the course of their transmission. Such an approach involved collating different copies, examining and evaluating the variants of the textual tradition, deleting parts judged as spurious and emending errors in order to reconstruct the authentic text. This was enough to create a most peculiar intellectual and scientific situation, but it also compelled the scholars to consider the criteria of critical interpretation in a new light, and even to create new standards and principles arousing needs that were different and deeper: because, as Giorgio Pasquali wrote, ‘constituting a text [...] requires the same learning and knowledge as interpreting [...] constituting a text and interpreting it are, ultimately, one and the same thing’.5 What seems remarkable indeed is the fact that in comparison to the prolonged history of ancient scholarship, which has been built up over the centuries from the origins right up to modern time, this short parenthesis lasting no more than the period from Zenodotus to the Augustan age, or possibly to the 2nd century AD, is the very period which we regard as the cradle of philological science in the modern sense, of course with all the mutatis mutandis necessary both for a correct historical evaluation and also to placate the concerns displayed  5 Pasquali 1920, repr. 1998, 26: ‘Insomma, a costituire un testo gli occhi servono così come le mani a scrivere un poema; occorre invece la stessa preparazione che a interpretare e a gustare: costituire un testo e interpretarlo sono, in fondo, tutt’uno’.

Ancient Scholarship and Classical Studies  

by a certain hypercritical hyperscepticism, which denies similarities perhaps because it is unable to evaluate the differences properly. Let us now recall that the publication of the Venetian Homeric scholia by Villoison, in which (and this was the great novelty) the materials of the Alexandrian scholarship were mainly to be found, gave rise to the famous Prolegomena ad Homerum by F.A. Wolf (a pupil of C.H. Heyne), always considered as one of the first and most significant turning points at the end of the Eighteenth century leading towards the birth of modern classical philology. It will therefore hardly be considered intellectual snobbery to point out that Alexandrian philology once again played a vital role and exerted decisive influence at a crucial moment in the history of classical studies. In the second half of the twentieth century, studies on Alexandrian philology and in general on the history of erudition, exegesis and grammar in the cultural panorama of the ancient world, experienced a renewed period of great flowering, which continues unabated in this opening stage of the twenty-first century. As compared to the state and tendencies of studies in the first half of the last century, today the picture appears radically changed. The by now classic work of Rudolf Pfeiffer dating from 1968, the two volumes of Kurt Latte’s incomplete edition of Hesychius dating from 1953 and 1966, Hartmut Erbse’s edition of the Scholia Vetera to the Iliad begun in 1969, the launch in 1974 of a series entitled Sammlung griechischer und lateinischer Grammatiker (SGLG), can be remembered as significant symptoms and stimuli of a new season. These research themes have progressively grown in importance and presence in the current panorama of classical studies, and now rest on different cultural foundations and orientations compared to the manner in which they were considered and treated, albeit with abundance and attention, in the context of nineteenth-century philology. As new and adequate working tools are devised and editions of texts become available, the effort to construct solid bases for research in this sector is acquiring more concrete form, while studies and essays continue to shed light in greater depth on a number of themes that are relevant for these aspects of the ancient literary civilization. Furthermore, such progress on the one hand can be seen as forming part of movements towards a positive reappraisal, which is by now complete and consolidated, of the postclassical historical phases of ancient Greek culture, from the imperial to the Byzantine age. On the other hand, this can rightly be described as one of the important aspects of a definitive transition away from the aestheticizing and intuitionist tendencies of the misguided and often ahistorical classicism that characterised a great part of the twentieth century. One fundamental element is that the investigations on the philology and erudition of the ancients no longer have an exclusively or predominantly ancillary

  Ancient Scholarship and Classical Studies value, and are no longer considered essentially or only as a repository of fragments of lost works, antiquarian curiosities or potential aids to modern philology. Ancient erudite and philological-grammatical production, in a word ‘ancient scholarship’, has acquired an independent meaning of its own, inasmuch as it is now seen as an expression and manifestation of a precise intellectual sphere and as an important aspect of ancient civilization. The exegetic observations and the erudite knowledge of the ancients are no longer considered only for what they tell us about a work or a phenomenon they aim to explain: rather, today they are and must be perceived as useful and interesting for what they tell us about themselves, i.e. about the ideas and the culture of which they are an expression. Yet even today one still too often notes the tendency to discuss the data of ancient philology and grammar on the basis of the principle of what is ‘right’ or ‘wrong’ from the point of view of modern science; in other words, the tendency to try to gauge how far the ancients had drawn close to the ‘correct’ interpretation and to what extent they missed the point, whether they were good or bad philologists, with regard to their textual choices as well. These are evaluations that distort the historical perspective. Moreover, too often the criterion for selection of materials considered worthy of interest and study remains based essentially on what appears to be useful or useless for the specific purpose of interpreting today, according to our criteria and for our own ends, the ancient author who is the focus of attention. In other words, too often the body of knowledge represented by ancient scholarship is viewed as potentially interesting and significant (only) when it is of aid in helping to solve a problem of modern scholarship. But this is a drastically limited and reductive viewpoint. Instead, everything that is of no aid in specifically interpreting Homer or Pindar or Aristophanes from our own point of view, is of the greatest aid in interpreting Zenodotus, Aristophanes of Byzantium, Aristarchus, and in understanding their cultural context and their intellectual milieu. This is a perspective that has become consolidated over recent decades and has contributed to the undeniable progress in the general historical vision of the ancient world: the products of scholarship have begun to be subjected to investigation for the purpose of discerning the critical principles, the ideas on literature and language, the interests, the thought of the scholars themselves in their cultural context. Of this important revival of studies on ancient scholarship in the current framework of classical studies, this conference, with its rich array of themes, is the most recent episode. Certainly it will not be the last, because the future path of this field of research promises to be long and rich in results.

 Ancient Scholarship Today This volume (Montanari 2018) is the final output of a project begun in 2013 (“Homer, Hesiod, Pindar, Aeschylus: Forms and transmission of the ancient exegeses”) and concluded in 2016, within the program FIRB — Futuro in Ricerca 2012, funded by the Italian Ministero dell’Istruzione, dell’Università, della Ricerca. The research was conducted by four groups directed by (in alphabetical order) Marco Ercoles (Università di Bologna), Lara Pagani (Università di Genova), Filippomaria Pontani (Università Ca’ Foscari di Venezia), and Giuseppe Ucciardello (Università di Messina), who also carried out the task of coordinating the work of the various groups. The research project has been fulfilled thanks to studies that have addressed the four great poets in various ways, exploring multiple themes concerning the exegesis of Homer, Hesiod, Pindar, and Aeschylus. It may not be useless to recall that all of them are outstanding authors of ancient Greek poetry, whose interpretation, in the broad sense of the term, has a historical and cultural significance of enormous importance and of special significance for illuminating the meaning of culture and thought in antiquity. Actually the research, while it has already been quite productive, is anything but finished, since, in addition to the conferences1 and this volume, the project has produced a conspicuous quantity of studies: beside those already published, others are in print, yet more are still in progress, and seeds for new trends of research have been planted. Thanks, therefore, to the effort and commitment of those who have lavished their energies and abilities on more than four years of work within this framework, the wave of research will last much longer and will continue to bear fruit in scholarly investigations. The result of the project is clearly evident and of great importance and utility for advancing our knowledge: this is what happens (obviously, one might add) when resources for research are allocated well. Alas, this is not always the case.2 Given this background, I would not say there is a need to ‘conclude’ in some way in relation to this project and its development, but rather to express some reflections on research into ancient scholarship today: the current picture, the changes in recent decades, the results gained, the paths along which studies are  1 A series of meetings were held in itinere, at Venice (16 May 2013), Genoa (12 May 2014), and Messina (11 May 2015), concluded by the final conference at Bologna (23–25 May 2016), where the most part of the papers collected in the present volume were delivered. 2 I have in mind some oddities that have emerged from ERC calls in recent years, but these are certainly not the only cases of perplexing funding decisions, which produce results of rather dubious value and are thus a very poor use of resources. https://doi.org/10.1515/9783110772371-003

  Ancient Scholarship Today proceeding and should now proceed. If we look broadly at the past half century, it is undeniable that there have been important changes in this field of research on the ancient world. The decisive turning point was undoubtedly marked by Rudolf Pfeiffer’s History, by its diffusion among scholars, and by the discussion raised by this seminal book.3 In short, through Pfeiffer’s account ancient scholarship definitively escaped the essentially ancillary role that it had traditionally occupied (abetted by the perspective of an often aestheticizing and ahistorical classicism) and rightly took on (thanks to the progressive abandonment of an increasingly jaded classicism) a role and function as one of the essential historical and cultural aspects of the ancient world, one that can no longer be overlooked. Since Pfeiffer’s book exactly half a century has passed, and I think we may say that fifty years ago such a large and challenging work as the recently published Companion to Ancient Greek Scholarship would not even have been conceivable.4 In the current research trends of these studies, I believe that some basic ideas, which are essential for an accurate view and valuation of cultural and intellectual history, should be accepted by scholars as established. The fundamental themes which we should approach in the proper way can be identified as follows: 1) What was the role of Aristotle and the Peripatos in the birth of Alexandrian philology, and what was their influence on the intellectual turning point that this entailed? 2) What questions should we ask of the products and results of the ancients’ critical and exegetical activities, what answers should we expect, and what is their significance for our research? 3) What is the historical and cultural value and role of the Alexandrians’ critical and philological activity, and on what is it based?

From the Peripatos to Alexandria: a new cultural outlook One of the aspects of Pfeiffer’s History mostly discussed is the reduced role that he ascribed to Aristotle and his school in the process towards the birth of philology at Alexandria. Pfeiffer criticized the view, held already in antiquity, and traditionally, even if automatically, repeated in modern studies, according to which Aristotle was the ‘father’ or ‘founder’ of Alexandrian philology; in doing so, Pfeiffer opened up a problem and a still ongoing debate which continues to

 3 Pfeiffer 1968; a general reassessment in Montanari 1994a. 4 Montanari/Matthaios/Rengakos 2015.

Ancient Scholarship Today  

stimulate new analyses and studies in ever greater depth.5 It seems worth underlining one more time that in Pfeiffer’s view the elements that link Aristotle and the Peripatos with Alexandria are in fact present, and in good number, and are even noted explicitly. Pfeiffer by no means omits to mention Aristotle and the work of the Peripatetics when the topic leads him to do so, but he then downplays any deeper connection: Aristotle was not the master of the first philologists, the Alexandrian philologists were not Aristotelians, Aristotle was not the founder or father of philology. I believe that it is clear that Pfeiffer’s approach originates from having given too much support to the relationship between poetry and philology, by refraining from a much more complex and nuanced picture involving the erudite activity as a whole, with its search for documentation on literature and language. Actually I think that, if we put together all the elements that imply deep and concrete connections between Aristotle/Peripatos and the work of the scholars of the Hellenistic period, we are led, rather, to emphasize and realize ever more fully that it was that environment and that line of development which provided the key impulses and inspirations. Aristotle took a new kind of interest in what we call ‘literature’ as a whole.6 Firstly, it is connected with his accentuated systematic interest in the history of the various disciplines: a large space was given to scholarly research and to antiquarian collecting, with serious effort made towards historical documentation in the domain of thought in which Aristotle himself developed his own doctrine. The link between the ordered collection of opinions expressed by predecessors (which gave rise to doxography) and theoretical reflection seems to be a characteristic intellectual trait: for the scientific foundation of a discipline, a conscious knowledge of its history is indispensable, and this is true also of rhetoric and poetry, the human activities that use words. This aspect of Aristotle’s approach cannot be underestimated or treated as secondary in scholarship. For example, most information available on the earlier techne rhetorike is owed to Aristotle, even though his collection of Technai is lost. For our purposes, special consideration is due to everything can be linked to researches on the great poetry of the past, on the traditional paideia of the Greeks. The great collections of historical and antiquarian erudition by Aristotle and his school undoubtedly had an organic connection to the field of literary history, to the study of literary works and the reconstruction of the lives of the authors, for which they provided indispensable notices and materials. Converging with this, literature itself was a source of information and an object of commentary and  5 See Montanari 2012 (= this volume, ch. 56), 2014, and 2017, to which I refer for further bibliography. 6 Montanari 2017.

  Ancient Scholarship Today exegesis, and there was a profound link between erudite documentation and the interpretation of texts, the legacy of which was a fertile training in work and method; this had an important continuation both in the activity of some exponents of the Peripatetic school and also in the critical and exegetical activity of Alexandrian philology, thus constituting the decisive intellectual inspiration for its development. Alongside his scholarly researches on works and authors, Aristotle was committed to theoretical reflection and produced his own doctrine on the techne poietike: as in other fields of knowledge, here too the two aspects cannot be detached from each other, since his view of the sphere of poetic art within the framework of human activities sharply altered the standard approach. The emergence of varied interests in what we call ‘literature’ arose from the importance Aristotle accorded to the products of verbal art as a sphere of human activity, and was deeply rooted in the attention that he gave to the historical and cultural topics connected to them. The shift in approach, compared to the past, was firm and decisive. While Plato did not assign to poetry, as the imitation of sensible things, a value as knowledge, for Aristotle it produced its knowledge precisely as mimesis of nature, which is in fact “reality”: it does not imitate the particular accidental, but the universal, because there is no true knowledge except of the universal. In other words, what concerns the end and the effect is firmly linked with the level of epistemology. In Aristotle’s view, one no longer imposes to the poetic art the condition of educating towards the good or of teaching things that are good and proper: its aim and function are of a cognitive order in relation to the intellect, and of a psychological-emotive order in relation to the passions. The specificity of the poetic art and the autonomy of poetry from reality in Aristotle’s conception are two elements that have emerged in a recent reconsideration of the historical relation between the school of Aristotle and the Museum of Alexandria, which systematically compares the two realities in order to investigate their shared theoretical outlook in the field of techne poietike.7 Through his reflections, activity, and teaching Aristotle brought about a cultural shift and gave to the Peripatos a direction that is clearly encountered among his pupils, despite the near-total loss of their writings, of which unfortunately only fragments survive and even these usually rather meager and unsatisfying. But in what remains of the works of figures such as Demetrius of Phalerum, Dicaearchus, Praxiphanes, Chamaeleon, and others, we find that the study of literature and the poets is present to an increasingly important degree: as soon as the lessons of Aristotle were digested ever more thoroughly, the investigation  7 Bouchard 2016, to be read with the help of Montana 2017, which focuses on the main issues.

Ancient Scholarship Today  

of poetic works and research into their authors now becomes a common and productive field that is of primary importance for interpreting and understanding the reality of humanity and human life in the world. A line of research that seems to me productive, and which I have pursued in some studies in recent years, is based on identifying Homeric passages about which Aristotle himself or one of the Peripatetics of the first generation raised problems of interpretation of various types, that were picked up again and had a continuation among the Alexandrian philologists in a different form — we could openly say, in a philological form. In short, an exegetical difficulty pointed out by Aristotle or one of his followers can be found transformed into an Alexandrian athetesis, or addressed with other tools developed by the post-Aristotelian scholars. In its specificity, this type of research can give substance and content to forms of continuity that are precise and concrete, reinforcing the picture of a more general intellectual and cultural continuity and relationship, in the sense outlined above. This therefore seems to be a research line that should be pursued further,8 also broadening the scope of observation beyond Homeric philology.9

The significance of ancient scholarship: what the texts tell us about themselves For a long time the extant remains of ancient exegesis and scholarship were considered and studied essentially (even if the particular focuses were of course varied) for two reasons and with two approaches: 1) as testimonia to fragments of lost works and to information otherwise unknown (for example on Realien, historical facts, institutions, and so on); 2) as a source of information for the interpretation and comprehension of the commented work (or occasionally of other works). The first case is immediately comprehensible: all editions of fragmentary works (Hesiod, tragedy, lyric, Hellenistic poetry, as well as prose works) teem with quotations found in scholiographic corpora, grammatical works, and lexicographical collections; not infrequently studies of various aspects of different historical periods benefit from scholarly sources (the one example of the FGrHist of Felix Jacoby can stand for them all). On the other hand, it sometimes happens that the exegesis of the surviving works of major authors of ancient

 8 Montanari 2000b (= this volume, ch. 55); 2012 (= this volume, ch. 56); 2014; Bacigalupo 2018. A slightly different approach in Pagani 2018. 9 E.g. Tocco 2019 studies the Peripatetic interest in the history of music as a scholarly field of inquiry.

  Ancient Scholarship Today literature gain some clarification from a scholion or a lexicographical item. These are two essential and important aspects which should certainly not be underestimated, or brushed aside, given the exceptionally important function that they exert. But for some time the approach has been changing, and this shift should now be consolidated as an established advance in knowledge. We can formulate it as follows: in addition to being important for what they tell us about everything other than themselves in their own right, the products of ancient scholarship are important, indeed fundamental, for what they tell us about themselves. We can appreciate the great importance of an unknown fragment of a lost work or an otherwise unknown fact about the ancient world; however, at least as important and significant, or perhaps even more so, is what these texts, difficult and complicated to understand, tell us about the methods of the ancient exegetes, the cultural assumptions, ideas, and intentions of their times and their settings. It is a fact — and we can no longer deny or ignore it — that the exegesis of ancient authors, the scholarship, grammar, reflection on language, all that we usually define under the general term ‘ancient philology’, should be accepted as one of the essential and indispensable aspects of the historical and cultural framework of the ancient world, and also as the final major step in the escape from an aestheticizing classicism (for which the products of ancient scholarship were usually late trivia of little or no value in themselves), whose scientific findings are irremediably ephemeral. Let us therefore ask of the texts of ancient scholarship, above all, what they tell us about themselves; let us adopt as a basic principle that it does not matter whether what they tell us is correct or mistaken, whether their interpretations are good or bad from our point of view and with our methods. What matters is instead what they imply and what they mean in their own right: that a scholion to Homer or Aeschylus chooses a mistaken reading is entirely secondary as compared to our understanding of the methods and assumptions that are brought to bear in making the selections. To defend the ‘usefulness’ of the texts of ancient scholarship does not mean to observe that sometimes they are right and interpret well according to our philology and ideas; it means to understand the reasons why they have interpreted in a certain way and have made certain choices; in short, to understand what they tell us about themselves, their age, and their setting.

The intellectual turning point: a question of method and ideas The other point is whether we should think of the phase of Alexandrian philology as a decisive intellectual turning point in the cultural history of our civilization or not, and whether it stands out as the foundation act of a way of studying

Ancient Scholarship Today  

literary texts that lies at the origin of the discipline that we today call (classical) philology. From this viewpoint a debate has developed that has become especially lively in recent times. Some scholars have denied to the Alexandrian philologists any kind of activity involving collating copies and selecting variants, and maintained that their readings were only conjectures, thus substantially downgrading their historical and cultural significance in the discipline of philology. This is the central aspect in the evaluation of Alexandrian scholarship from the point of view of cultural history. It would perhaps be all too easy to begin by pointing out that conjecture is one of the fundamental tools of the practice of philology, which would highlight an inexplicable contradiction in this perspective, and be sufficient to discredit it. There are, however, some clear and indisputable testimonia to the fact that collation was done from copies and choices were made among variant readings; to the objection that this was an occasional phenomenon and not an established practice, we may reply that this is a problem involving principles and methods, not the quantity of data (the number of copies collated and variants discussed), or the quality of the results (whether correct or mistaken from our point of view). We should not be trying to establish a minimum number of copies to be compared to each other or of variants to be considered, nor to determine how many ‘correct’ readings or ‘good’ interpretations (see above) would be necessary in order to speak of philology. Rather, in a historical approach, all that is necessary for a crucial step forward, in terms of an intellectual advance, is the very fact that the problem is understood and addressed, even if in a partial, desultory, or incoherent manner: a literary text had lived a multifaceted history of transmission, during which it could have been distorted; it was possible to restore the correct text (that is, which was an authentic line of verse and which a spurious one, and what was the original wording) by conjecture or by choosing the best reading among those offered by a discordant tradition. Without any doubt the work of the Alexandrian philologists encompassed both variants drawn from the comparison of copies and also conjectures ope ingenii, that is exactly the working tool of modern philology. A further consideration has decisive force: it concerns the invention of the critical sign termed obelos by Zenodotus. He performed two different operations, which mark an important intellectual point: on the one hand the material deletion of lines regarded as certainly not authentic, and on the other hand the indication that a line could be suspected of being spurious, but without sufficient certainty to eliminate it physically and permanently from the text. Such a line therefore remained in the text, with a mark of doubt that leaves to the readers the possibility of forming their own opinion. It was the codification of philological doubt,

  Ancient Scholarship Today which we indicate in our critical editions with the marks of expunction, which signal the part of the text that is regarded as uncertain and debatable, but yet remains in the text, available to its reader. The idea that a textual damage can be detected, and that a way needs to be found to repair it, reveals that the organic unity between interpretation and textual criticism was deemed to be achieved. Although much remained to be done, and ‘Wolfian’ philology, critical editions, and scientific commentary still were far into the future, our point of view — far from being an anachronism — is the historical assessment that a crucial step was made in the period between Zenodotus and Aristarchus.10 By these brief remarks I certainly do not claim to have drawn a complete picture of the research themes and problems that shape the field of ancient scholarship in its many and multifaceted aspects: a picture that has now become large and complex, in view of the great number of ancient grammarians who are at stake, often known only through meagre and problematic fragments, or through several texts of exegesis and erudition that constitute its essential sources, texts that not infrequently are difficult and still relatively unexplored due to the long periods in which there was little interest in them. I hope merely to have addressed some key-points about the history of ancient scholarship, in which there have been truly decisive advances in recent decades: advances and changes of approach that have set in their true light their historical-cultural and intellectual value and significance, by establishing them definitively as part of a legacy of achieved knowledge, which should never again be lost in studies of the ancient world. Many studies and more in-depth investigations must still be made, of course, and many texts await an adequate critical edition, several figures of primary or secondary importance in ancient culture must still be examined to shed more light on their activities, their position, and their role.11 Yet today we look at this sector of research not only in an entirely new light, but also in a different, and now more accurate perception of its role, importance, and influence in the context of ancient culture and thought.

 10 I have developed these ideas in some recent works: Montanari 2015a (= this volume, ch. 9); 2015b (= this volume, ch. 11); 2015c, to which I refer for a more detailed discussion and for the bibliography; cf. Montana 2012 and Pagani 2020. 11 A prosopographical collection of figures important in the sphere of philology and scholarship in a broad sense in the ancient Greek world is presented within the Lexicon of Greek Grammarians of Antiquity (LGGA): Montanari/Montana/Pagani (in progress).

 History of Ancient Greek Scholarship from the Beginnings to the End of the Byzantine Age: Introduction The history of classical scholarship is classical scholarship in the making Rudolf Pfeiffer

This book began from the idea of detaching the first four chapters, which formed the historiographical part, of Brill’s Companion to Ancient Greek Scholarship (published in 2015) as a single independent volume, to produce a History of Ancient Greek Scholarship from the Beginnings to the End of the Byzantine Age. For this it was obviously necessary first of all to update the bibliography,1 but also to reorganize and rethink as a unified treatment all that we usually understand under the general term “ancient scholarship” or “ancient philology.” This includes the exegesis of ancient authors and the editing of their texts, the orderly collection of materials useful for exegetical purposes, the study of grammar, reflection on language as an instrument of literature, and everything that can be linked to this sphere. It is easy to observe that no such work covering a period from the origins of ancient scholarship to the fall of Byzantium (i.e. a couple of millennia of history, a chronological span that is hard even to conceptualize in a unified perspective) currently exists in the research landscape on ancient scholarship, and that it is today hard to imagine such a work being undertaken by a single person. In any case, while a single-authored study, should it ever be possible, would have the advantage of unity of perspective and cultural outlook, it would lose the benefits offered by a volume with multiple voices in a field so complex and multiform. It would be hard to deny that the history of ancient scholarship has been one of the most significant innovations of the past century in the panorama of studies on the ancient world, both in the importance achieved by this sector of research in just a few decades, and in the quantity, breadth, and depth of new critical editions of the texts of ancient philological erudition, which were at first rather neglected in modern research but are now the subject of new and philologically robust editions. Indeed we can say without risk of contradiction that by now, in our century, the history of ancient scholarship has gained an indisputable standing within the field of research on ancient culture. Only a myopic  1 The present volume contains an extensive bibliography that covers all the topics treated in this introduction, so I choose not to repeat it here. https://doi.org/10.1515/9783110772371-004

  History of Ancient Greek Scholarship and — thankfully — fading aestheticizing approach and outlook could imagine taking the backward step of denying or downplaying its importance. It is now accepted that the turning point, both intellectually and historically, was the publication of Pfeiffer’s History in 1968. The comprehensive discussion and reflection dedicated to the book’s findings a quarter century after its publication, in the Entretiens Hardt on this topic,2 demonstrated that Pfeiffer’s work was already securely established as an indispensable point of reference, while also indicating possible expansions, supplements, and adjustments of its position. As regards the temporal range under consideration, many of us hoped even back then that Pfeiffer would give us a second volume dedicated to the Roman Imperial period, but sadly this did not happen.3 Since Pfeiffer did not do what the scholars expected, the world of research felt as particularly necessary to have a historical framework for the centuries of the Imperial period in continuity with the Hellenistic period and until the end of the Byzantine millennium. The problem of continuity between the Hellenistic and the imperial period is captured by the very title of the Entretiens of 1994 cited above, and is discussed explicitly in their Introduction: “We wanted to avoid a sharp cut-off date at the end of the Hellenistic period, which would be artificial, and to broaden the view as far as possible towards a Hellenistic-Roman chronological span. We are convinced that a unified examination of this period would strongly benefit our understanding of many phenomena and of their historical consequences. These include decisive interventions in the transmission of classical texts; the history of production of Hellenistic commentaries and treatises up to the early stages of the formation of scholiographic corpora; the consolidation of grammatical theory up to Apollonius Dyscolus and Herodian; the development of the lexicographical collections that would flow into the Byzantine compilers; the foundations of the paroemiographical research up to the first major collections; and the evolution of rhetorical thought.”4 We believe that the chronological limits marked by the present volume, from the origins to the end of the Byzantine era, are plausible and comprehensible and indicate the crucial historical importance of the Roman Imperial period (including the presence and role played by the Latin world, in dialogue with Greek culture), and of the full contribution of the Byzantine millennium. If these limits open up a path for a historical treatment of subsequent eras, it would certainly constitute a further chapter of “classical  2 Montanari 1994a. 3 Pfeiffer chose instead to turn his attention to a different period, which he perhaps felt was more in need of historical reflection by a classical philologist of his era: Pfeiffer 1976. 4 Montanari, Introduzione in Montanari 1994a, 4, here translated from the original Italian.

History of Ancient Greek Scholarship  

scholarship in the making”, to borrow Pfeiffer’s phrase. The passage just quoted from the introduction to the Entretiens of 1994 highlights another series of themes that can, or at least could, be added to the contents of scholarship as represented in Pfeiffer’s book, and which subsequent studies have taken into consideration in various ways. On that occasion a further problem was raised, which is by no means of secondary importance: alongside a treatment with a primarily historical-chronological framework, such as that of Pfeiffer and of the present volume (in which, however, the chronological span extends many centuries beyond the end of the Hellenistic era), is it also possible to envisage a thematic structure? The latter should, so far as possible, follow the separate lines of development of the various currents of research within this field, the vastness and complexity of which is becoming ever clearer. However, I shall not expand on this and repeat what was said in 1994. Scholars interested in the history of classical scholarship can return to those Entretiens and reflect once again on the tangle of problems that were put on the table then and remain there still, and which have been stimulating research of enormous value and quantity ever since. I would like instead to devote a few pages to recalling some fundamental and indispensable concepts, which both highlight and embody the progress made in roughly half a century of studies and which, I firmly believe, give concrete form to the value of the historical development described by Pfeiffer. The surviving products of ancient exegesis and scholarship were long considered and studied essentially for two reasons and with two basic approaches, though the particular focuses of course varied: 1) as testimonia to fragments of lost literary works and as constituting otherwise unknown information (for example on Realien, historical facts, institutions, and so on); and 2) as a source of information for the modern interpretation and understanding of the work commented upon (or occasionally on other works). The first case is immediately comprehensible: all editions of fragmentary works teem with quotations found in scholiographic corpora, grammatical works, and lexicographical collections; and often studies of various aspects of different historical eras benefit from erudite sources (the example of the FGrHist of Felix Jacoby may here stand for them all). And, in the second case, it does sometimes happen that the exegesis of surviving works of major authors of ancient literature gains some clarification from a scholium or a lexicon lemma. These are two indispensably important aspects which should certainly not be underestimated, far less brushed aside, given the exceptionally important function that they fulfill. But for some time the approach has been deeply changed, and this shift should now be consolidated as an established advance

  History of Ancient Greek Scholarship in research and knowledge. We can formulate it as follows: as well as being important for what they tell us about everything other than themselves in their own right, the products of ancient scholarship are important, indeed fundamental, for what they tell us about themselves. We can appreciate the great importance of an unknown fragment of a lost work or an otherwise unknown fact or phenomenon about the ancient world, but at least as important and significant, or perhaps even more so, is what these texts, albeit difficult and complicated to understand, tell us about the methods of the ancient exegetes, the cultural assumptions, ideas, settings and intentions of their times and environments. It is a fact — and we can no longer deny or ignore it — that all that we usually include under the general term “ancient philology” or “ancient scholarship” should be accepted as an essential and indispensable aspect of the historical and cultural framework of the ancient world, and also as the final major step in the escape from an aestheticizing classicism (for which the products of ancient scholarship were usually late trivial works of little or no value in themselves), whose scientific findings are irremediably ephemeral and entirely subjective. Let us therefore ask of the texts of ancient scholarship, above all, what they tell us about themselves, and let us adopt as a basic principle that it does not matter whether what they tell us is correct or mistaken in a modern view and interpretation, whether their interpretations are good or bad from our point of view and with our “scientific” methods: what matters, rather, is what they imply and what they mean in their own right. That a scholium to Homer or Aeschylus selects a mistaken reading is entirely secondary as compared to our understanding of the methods and assumptions that are brought to bear in making the selections. To defend the ‘usefulness’ of the texts of ancient scholarship does not mean to observe that sometimes they are right and interpret well according to our philology, methods and ideas. It means understanding the reasons why they have interpreted in a certain way and have made certain choices: in short, again, to understand what they tell us about themselves, their era, and their setting, to understand their culture. In Pfeiffer’s book scholars’ attention (some of it critical) was drawn at once by, among other things, the title and content of the first part of his book: “Prehistory of Greek Scholarship,” divided into three chapters that run from the epic poets to Aristotle and the Peripatus (that is to say, the first generations of students of Aristotle). The concept of historical development which Pfeiffer thus applied very explicitly to the sphere of scholarship as a cultural field, by assigning to it a long and complex phase of preparation and, in a sense, of progressive incubation, had crucial intellectual implications. To disentangle and illuminate

History of Ancient Greek Scholarship  

these aspects and phenomena, which are by no means of secondary importance in Greek cultural history, was to appreciate in a new way the fact that ancient scholarship did not spring miraculously out of nothing, nor was it a series of independent or only loosely connected events and personalities. Rather, it was the result of a deep cultural need, linked to the most important and essential developments of a civilization, which took note of the importance and value of its own heritage in literature and thought, and acknowledged the need to acquire tools to study it, preserve it, interpret it, and understand it for its own sake and for that of future generations.5 And this is true of the whole two millennia chronological span considered in this book, until the end of the Byzantine era marked a further historical and cultural transition. In this context, one of the aspects of Pfeiffer’s History that have been debated the most is the reduced role that he ascribed to Aristotle and his school as bearers of the decisive impulses in the birth of philology at Alexandria. Pfeiffer criticized the view, held already in antiquity, and traditionally, if rather automatically, repeated in modern studies, which identified Aristotle as the “father” or “founder” of Alexandrian philology. With this Pfeiffer opened up a problem and a debate which continue to stimulate new analyses and studies in ever greater depth. It seems worth underlining one more time that in Pfeiffer the elements that link Aristotle and the Peripatus with Alexandria are in fact present, and in good number, and are even noted explicitly. That is, Pfeiffer by no means omits to mention Aristotle and the work of the Peripatetics when the topic leads him to do so, but he then downplays any deeper connection: Aristotle was not the master of the first philologists, the Alexandrian philologists were not Aristotelians, Aristotle was not the founder or father of philology. I believe it is clear that Pfeiffer’s approach arises from having privileged to an excessive degree and having even, as it were, in a sense isolated the relationship between poetry and philology (philology was born from poetry, and is in a certain sense its child), instead of the much more complex and nuanced picture given by the whole sphere of erudite activity, with its search for documentation concerning literature and language and its interpretation.

 5 Pfeiffer showed that he was perfectly aware of this fact (and how could it be otherwise?) when he wrote in the Preface (p. viii): “There have, of course, been earlier attempts in this field since the days of Henri Estienne who wrote in 1587 De criticis veteribus Graecis et Latinis. But only one really comprehensive book exists: Sandys 1920, in three volumes of 1.629 pages … But as a whole Sandy’s work is rather a catalogue of classical scholars, century by century, nation by nation, and book by book than a real history of scholarship itself; there is no leading idea, no coherent structure, no sober discrimination between the transient and the perennial.”

  History of Ancient Greek Scholarship In reality I think that, if we put together all the elements that imply deep and concrete connections between Aristotle/Peripatus and the work of the scholars of the Hellenistic period, we are led, rather, to emphasize and realize ever more fully that it was that environment and that line of development which provided the key impulses and inspirations. This is what has been stressed and confirmed by sound arguments in many studies since Pfeiffer, and today I believe it can be regarded as established and as a significant change in perspective, or perhaps better a change of emphasis. Through his reflections, activity, and teaching Aristotle brought about a cultural shift and gave to the Peripatus a direction that is clearly evident among his pupils, despite the near-total loss of their writings, of which unfortunately only rather meager fragments survive. In substance, thanks to a renewed and reliable historical outlook, the historical and intellectual problem of whether we should hold that in our civilization the phase of Alexandrian philology was or was not a decisive intellectual turning point (based on a question of method and ideas established once and for all) and whether or not it constituted the birth of a way of studying literary texts that lies at the origin of the discipline that we today call (classical) philology, finds an incontrovertibly positive reply. On this dimension a debate has developed and has become especially lively in recent times. Some scholars have wanted to deny to the Alexandrian philologists any kind of activity involving collating copies and selecting variants, and have maintained that their readings were only conjectures, thus substantially downgrading their historical and cultural significance in the discipline of philology. This is the central aspect in the evaluation of Alexandrian scholarship from the point of view of cultural history. It would perhaps be all too easy to begin by pointing out that conjecture is one of the fundamental instruments of the practice of philology, which would highlight an insoluble contradiction in that point of view, and be sufficient to discredit it. But, anyway, there are some clear and indisputable testimonia to the fact that collation was done from copies and choices were made among variant readings. To the objection that this was an occasional phenomenon and not an established practice, we may respond that this is a problem of principles and methods, not of the quantity of data (the number of copies collated and the number of variants discussed), or of the quality of the results (whether correct or mistaken from our point of view). We should not be trying to establish a minimum number of copies to be compared to each other or of variants to be considered, nor to determine how many ‘correct’ readings or ‘good’ interpretations would be necessary in order to speak of philology. Rather, in a historical approach, all that is necessary for there to be a crucial step forward — a “turning point,” we said — in terms of an intellectual advance, is the very fact that the

History of Ancient Greek Scholarship  

problem is addressed and understood, even if in a partial, desultory, or incoherent manner: a literary text had had a multifaceted history of transmission, during which it could have been distorted; it was possible to restore the correct text (that is, which was an authentic line of verse and which a spurious one, and what was the original wording) by conjecture or by choosing the best reading among those offered by a discordant tradition. Without any doubt the work of the Alexandrian philologists encompassed both variants drawn from the comparison of copies and also conjectures ope ingenii, that is, exactly the instrumentarium of modern philology, despite all the defects that may be found. A further consideration has decisive force: it concerns the invention of the critical sign termed obelos by Zenodotus. He performed two different operations, which mark an important intellectual point: on the one hand the material deletion of lines regarded as certainly not authentic, and on the other hand the indication that a line could be suspected of being spurious, but without sufficient certainty to eliminate it physically and permanently from the text: such a line therefore remained in the text, with a mark of doubt that drew attention to the textual problem and left to the readers the possibility of forming their own opinion. It was the codification of philological doubt, which we indicate in our critical editions with the marks of expunction that signal the words that are regarded as uncertain and debatable, but which remain in the text, available to its reader. The idea of recognizing damage, and that a way and instruments need to be found to repair it, reveals that the organic unity between interpretation and textual criticism had been achieved. This point of view, far from being an anachronism, is the historical assessment that a crucial step was made in the period between Zenodotus and Aristarchus. Just as Pfeiffer argued, this is precisely a discrimination between the transient and the perennial.6 The wider chronological perspective encompassed by this volume has reinforced the ideas summarized above, and it has confirmed in a definitive way the significance and historical and cultural value of a History of Ancient Greek Scholarship with a vision from the Beginnings to the End of the Byzantine Age Having reached the end of the work, at the moment of submitting the volume to the press I cannot fail to thank wholeheartedly Fausto Montana and Valeria Bacigalupo for the active and intelligent commitment with which they have assisted the editor and the authors in the preparation of the final text. Marco Comunetti took care of the indexes.

 6 Cf. above, n. 5.

 Evoluzioni del coro e movimenti celesti Devo cominciare con una apologia, per due ragioni. La prima è che l’argomento che tratterò ha con il teatro antico un rapporto che definirei assai astratto, e comunque labile: ma questo forse non è troppo grave. Probabilmente più grave è che presento qui soltanto qualche piccola idea e alcune considerazioni, che avrebbero bisogno di vari approfondimenti, suggeritemi di recente da un testo nel quale mi sono imbattuto occupandomi di grammatici greci minori (cfr. bibliografia). Nel 1832 J. Fr. Boissonnade trovò nel codice Par. Gr. 2551 e pubblicò (Boissonnade 1832, IV, 458 sg.) un breve testo, chiaramente un excerptum da un’opera erudita, nel quale si parlava di poeti e poesia lirica, canti monostrofici e triadici, strofe antistrofe epodo e relativi movimenti del coro nell’esecuzione. L’editore indicò in nota il sospetto, naturalmente espresso in forma dubitativa, di trovarsi di fronte a un frammento estratto dal celebre (e perduto) trattato Πεϱὶ λυϱιϰῶν ποιητῶν del grammatico Didimo, attivo nella seconda metà del I secolo a. C. e all’inizio del I d. C. Su questa base, alcuni anni dopo, non mancò di far menzione di quel frammento M. Schmidt nella sua raccolta dei resti delle opere di Didimo Calcentero (pp. 395 sg.), con un breve commento che vale la pena riprendere. A un certo punto dell’excerptum si cita come fonte un «Tolomeo», senza epiteti né ulteriori specificazioni del nome: su questo nome Boissonnade non aveva detto nulla, ma Schmidt intese con evidente naturalezza trattarsi di Claudio Tolomeo, il famoso scienziato vissuto nel II secolo d. C., e osservò di conseguenza che ciò impediva, per ovvie ragioni cronologiche, di attribuire a Didimo il materiale, o almeno tutto il materiale, restituito dal breve testo; a meno di non pensare, all’inverso, che sia stato Tolomeo a usare materiale del trattato di Didimo sulla lirica, cioè che l’excerptum provenga, attraverso ignoti intermediari eruditi, da un’opera di Tolomeo, che a sua volta ci trasmetterebbe Didimo. In appendice all’edizione degli Scholia Vetera agli Epinici di Pindaro, il Drachmann (1903–1927) raccolse un gruppetto di brevi testi, che trattano temi relativi al canone dei poeti lirici, alla poesia lirica stessa, la sua struttura, la sua metrica e simili, raggruppando diverse notiziole. Il più cospicuo di essi è costituito dal proemio dell’arcivescovo bizantino Eustazio al suo perduto commentario a Pindaro. Gli altri, raggruppati sotto la definizione di Capitula ad praefationem pertinentia, sono brevi excerpta, concepiti un po’ come rapide ed assai sommarie introduzioni generali alla lettura dell’opera di un poeta lirico e provenienti da qualche manoscritto pindarico. Fra questi trovò posto anche il nostro frammento, indicato come Cap. f (III, pp. 310 sg.), che l’editore, oltre che nel https://doi.org/10.1515/9783110772371-005

Evoluzioni del coro e movimenti celesti  

codice Par. Gr. 2551 del Boissonnade, aveva rinvenuto in altri due manoscritti: completo nel codice Marc. gr. 475 (= Y, dove peraltro è frammischiato con il Cap. c, cfr. Drachmann, ibid. p. 282); solo l’ultima parte, quella di cui ci occupiamo qui, nel codice Urb. gr. 151 (scritta nel margine accanto al Cap. c, cfr. Drachmann, loc. cit.). Per questi Capitula, Drachmann ci tenne a precisare (p. 281) che non si tratta, nella sostanza, solo di tarde e autoschediastiche composisizioni bizantine: abbiamo in verità a che fare con resti provenienti da attività grammaticaleerudita antica, rimasugli esili, naturalmente, e malmenati dalle tortuose vicende della loro trasmissione, ma nei quali labili tracce di buona fonte possono ancora trasparire, e talvolta traspaiono realmente. Come si può ben capire, la trasmissione di testi di tale genere non è molto solida e neppure troppo definita: si evince facilmente che sono state assai facili aggiunte e omissioni, scomposizioni e ricomposizioni, che creano incertezze non soltanto sul dettato, ma anche sulla configurazione complessiva, i limiti e le dimensioni stesse di ciascun excerptum. Il nostro frammento, nella forma rinvenuta primamente da Boissonnade e ripresa da Schmidt e Drachmann, può essere analizzato in diversi modi, di cui il più comodo e adatto per noi in questa sede, ma direi anche il più plausibile, è il seguente. Si possono grosso modo distinguere due parti. Nella prima (p. 310, 18–311, 6 Dr.), al canone dei poeti lirici seguono cenni assai rapidi e scarni sulla poesia lirica in sé (autori, strumenti, metrica), dove tuttavia compare il nome (che non si può certo dire dei più usuali) del grammatico Lisania (forse di Cirene, vissuto nel III secolo a. C. e citato da Suida come uno dei maestri di Eratostene; autore di un trattato Πεϱὶ ἰαμβοποιῶν; cfr. Porfirio ad Il. I 378; sch. Il. Π 558b, Φ 262e; sch. Eur., Andr. 10; Etym. Mag. 779, 9 (+ sch. Od. ψ 3, Ap. Soph. 158, 8 sgg.); Athen. VII 304b, XI 504b, XIV 620c); poi si introduce una generale ripartizione (con qualche esempio) dei canti lirici in due grandi gruppi, quelli monostrofici e quelli triadici costituiti da un insieme di strofe, antistrofe, epodo. Si può far iniziare a questo punto la seconda parte del nostro testo (p. 311, 6–14 Dr.: è quella che, come dicevamo sopra, compare isolata nel cod. Urb. gr. 151), che viene introdotta con il riferimento a Πτολεμαῖος ἐν τῷ Πεϱὶ στατιϰῆς ποιήσεως, al quale pare con ogni evidenza si faccia risalire il materiale che segue. ϰέϰληται δὲ ἡ μὲν στροφή, ϰαθό φησι Πτολεμαῖος ἐν τῷ Περὶ στατιϰῆς ποιήσεως, διὰ τὸ τοὺς ᾄδοντας ϰύϰλῳ ϰινεῖσθαι περὶ τὸν βωμόν, σημαίνοντας τὴν τοῦ †βίου† ϰίνησιν. ἀντίστροφος δὲ διὰ τὸ ἀναστρέφοντας αὐτοὺς εὐρύθμως ϰινεῖσθαι, ἄχρις ἂν ἔλθωσιν ἐπ᾿ ἐϰεῖνον τὸν τόπον, ἀφ᾿ οὗ ἤρξαντο πρῶτον ϰινεῖσθαι· ἠνίσσοντο δὲ διὰ τούτου τὴν τοῦ ἡλίου ϰίνησιν, ἐπειδὴ τὴν ἐναντίαν οὗτος δοϰεῖ τῷ ϰόσμῳ ποιεῖσθαι. ἐπῳδὸς δὲ ἐπειδὴ ἱστάμενοι ἐπῇδον, διὰ τούτου τὸ ἔμμονον ϰαὶ στερρὸν τῆς γῆς παριστῶντες.

  Evoluzioni del coro e movimenti celesti 3 βίου corruptum: ἡλίου perperam Boissonnade; ϰόσμου (in textu) dub. Crusius (cf. 6 infra); οὐϱανοῦ (in app.) dub. Crusius, Drachmann; ὅλου (in app.) dub. Crusius

Dunque si dice qui che la strofe si chiama in questo modo, secondo quanto afferma Tolomeo nel suo trattato Πεϱὶ στατιϰῆς ποιήσεως, perché coloro che cantano si muovono in circolo attorno all’altare, raffigurando così il moto del cosmo o del tutto [la lezione tramandata sarebbe della vita, certamente sbagliato: il confronto col r. 6 suggerirebbe ϰόσμου, i testi paralleli di cui infra piuttosto παντός, cfr. 306, 18 e 307, 14 Dr.]. L’antistrofe invece si chiama così perché coloro che cantano si muovono ritmicamente girando in direzione contraria, sino a che giungono al luogo dal quale ha avuto inizio il movimento. In tal modo fanno allusione, simboleggiano il movimento del sole, poiché esso percorre un cammino inverso a quello del cosmo. L’epodo infine si chiama così perché lo cantano rimanendo fermi, indicando con questo la saldezza immobile della terra. Nonostante il carattere chiaramente riassuntivo e ridotto, il senso generale del discorso risulta abbastanza chiaro: il movimento che il coro esegue intorno all’altare cantando la strofe e l’antistrofe, rispettivamente prima in un senso e poi in quello inverso, simboleggia, allude, riproduce i movimenti celesti: quello del cosmo, inteso evidentemente come sfera esterna dell’universo o cielo delle stelle fisse (anche οὐϱανός oppure ζωδιαϰὸς ϰύϰλος nei testi paralleli di cui infra), in una direzione; quello del sole (anche dell’insieme dei pianeti nei testi paralleli di cui infra), nella propria sfera più bassa e dotata di movimento autonomo, nella direzione contraria; lo star fermi nell’esecuzione dell’epodo è infine immagine della fissità della terra al centro dell’universo. Non credo ci sia bisogno di essere esperti di teorie astronomiche antiche e di storia dell’astronomia per rendersi conto di come il discorso in termini astronomici sia in effetti grossolano e superficiale: d’altra parte, la cosa non sorprende dato il tipo di testo che abbiamo di fronte, e non è certo responsabilità della fonte originaria. In ogni caso, è sufficientemente noto che la concezione dell’universo come formato da una serie di sfere omocentriche, ciascuna dotata di un proprio autonomo movimento, prevedeva tra gli altri l’aspetto secondo cui le sfere del sole e dei pianeti possiedono un movimento autonomo e contrario a quello del cielo delle stelle fisse, che è il più esterno. Più precisamente, il cielo delle stelle fisse compie un movimento da oriente a occidente, mentre il sole nella sua sfera più interna viene sì trascinato da oriente a occidente dal predominante movimento del cielo esterno delle stelle fisse, ma di per sé avrebbe un movimento inverso da occidente a oriente, la cui «resistenza» nei confronti dell’altro spiega il fatto che il sole rimanga indietro rispetto alle stelle. Non credo ci sia possibilità di approfondire ulteriormente da un punto di vista storicoscientifico, al di là di questo generico riferimento: per il nostro discorso mi pare

Evoluzioni del coro e movimenti celesti  

sufficiente quello che abbiamo notato, cioè che nel nostro testo veniva stabilito un parallelo fra le evoluzioni del coro e i movimenti dei corpi celesti, nei termini che abbiamo detto. Vediamo ora di affrontare il problema dell’identificazione del Tolomeo citato come fonte nell’excerptum. Come abbiamo visto sopra, già lo Schmidt si era tranquillamente riferito a Claudio Tolomeo, senza mostrare dubbi al riguardo. Il problema in realtà poteva scaturire dal rapporto che si voleva istituire con materiale proveniente da Didimo: è chiaro che, se si potesse dimostrare che tutto l’excerptum nel suo insieme deriva recta via dal Πεϱὶ λυϱιϰῶν ποιητῶν del grammatico attivo nella seconda metà del I secolo a. C. e all’inizio del I d. C., dovremmo pensare a un altro Tolomeo anteriore a Didimo. L’ipotesi inversa, cioè di avere di fronte un excerptum da Claudio Tolomeo che a sua volta avrebbe usato materiale del trattato didimeo, non può essere scartata a priori, almeno per la parte di testo che abbiamo preso in considerazione. Sembra che il problema, posto in questi termini, diciamo pure piuttosto meccanici ed esclusivi, sia irresolubile. Ma bisogna ricordare che il piccolo testo, come tutti quelli analoghi per carattere e per tradizione, ha conosciuto una trasmissione impervia e per nulla «rispettosa», per cui i numerosi rimaneggiamenti hanno sicuramente allontanato parecchio il dettato del testo non solo dalle sue fonti prime, ma anche dall’aspetto del primo excerptum che fu costituito. Dunque, la cosa più probabile è che in un primo tempo un anonimo erudito, più tardo di tutti i personaggi implicati, abbia usato e fuso materiali di diversa provenienza in un più o meno breve excursus, a sua volta poi oggetto di rimaneggiamenti e riutilizzazioni disinvolte. La parte del testo che riporta materiale di «Tolomeo» può essere trattata senza ipoteche preventive, anche se vogliamo pensare che la parte precedente possa in ultima analisi e attraverso mediazioni varie risalire davvero a Didimo. Invece, nella voce della R.E. (risalente al 1959) dedicata al grammatico Tolomeo Epitete, A. Dihle ha attribuito a questo grammatico alessandrino vissuto nel II sec. a. C. (un po’ più giovane di Aristarco e quindi anteriore a Didimo), del quale conosciamo pochissimi frammenti, anche il nostro testo, con la sola giustificazione che Tolomeo Epitete è l’unico erudito di nome «Tolomeo» citato nel corpus degli scoli a Pindaro. Questa inclusione ha di fatto riaperto il problema e obbligato a riesaminarlo proprio in occasione dell’edizione dei frammenti di Tolomeo Epitete: cfr. infatti Montanari 1988, Ptolem. Epith. T 4* e F 6*, pp. 87, 97 sg., 107 sgg., dove il frammento è indicato come spurio. In realtà, l’argomento per attribuirlo a Tolomeo Epitete è del tutto fallace: primo, per la sua completa esteriorità e casualità; secondo, perché considera legato organicamente al corpus degli scoli pindarici un testo che non ne è affatto una parte

  Evoluzioni del coro e movimenti celesti costitutiva (cfr. Drachmann 1903–1927, loc. cit., pp. 282 sg.), ma è stato loro associato con altre funzioni in qualche manoscritto e poi nelle edizioni moderne: e infatti la sezione che ci interessa sembra muoversi su un piano molto generale, riferendosi astrattamente a un qualsivoglia movimento di un qualsivoglia coro, come vedremo oltre (e l’utilizzazione «pindarica» appare piuttosto secondaria; giusto il dubbio di Drachmann 1903–1927, loc. cit.: «hoc capitulum num ad scholia Pindarica pertineat incertum est»); terzo perché non tiene conto del suggerimento offerto dal contenuto stesso del testo, come vediamo subito. Più produttiva infatti si è rivelata l’osservazione, fatta a suo tempo da O. Crusius (1888, 9 sgg.: un lavoro su cui torneremo fra poco) e poi ripresa da H. Färber (1936, 16 sgg., partic. 21 sg.), proprio sul contenuto del frammento. L’analogia cosmologica, chiamata in causa per «spiegare» la struttura della triade strofica, fa riferimento alle sfere delle stelle fisse e dei pianeti e all’immobilità della terra al centro dell’universo: questo tipo di argomento non può non parlare in favore di Claudio Tolomeo. A ciò si aggiunga il fatto che una citazione così naturale di «Tolomeo» senza specificazioni, e per di più in un simile contesto, ben difficilmente da chiunque avrà potuto essere percepita altrimenti che come riferita a Claudio Tolomeo (scienziato e astronomo che, tra l’altro, come sappiamo bene, si occupò anche di musica), piuttosto che a un grammatico certo assai meno noto e citato (oggi ne rimangono cinque frammenti, che trattano questioni di critica testuale: cfr. Montanari 1988, 75 sgg.). A questo punto, benché esuli dal nostro interesse primario in questa sede, non dobbiamo trascurare il fatto che per questa via sono stati recuperati il titolo e un frammento di un trattato perduto di Claudio Tolomeo, cioè il Πεϱὶ στατιϰῆς ποιήσεως. Per quanto ho potuto vedere, pur senza compiere una ricerca sistematica, non mi pare che quest’opera sia tenuta in considerazione a proposito di Tolomeo: fra l’altro, essa non è registrata nei, peraltro non vecchi, articoli su di lui della R. E. (citt. in bibliografia), dei quali soprattutto quello di B.L. van der Waerden presenta un quadro di riferimento globale. Mi pare difficile avventurarsi a immaginare il contenuto dello scritto, dal momento che già il titolo di per sé risulta problematico e di difficile interpretazione. Crusius, in una rapida annotazione (p. 10, nota 1), confrontando στατιϰή con il latino stataria (detto della commedia con uno svolgimento pacato e non concitato), intendeva: «στατιϰή ποίησις heißt also wohl die ruhige ernste Poesie». Su una simile linea interpretativa, Färber (1936, 21) rimandava a Kranz (1933, 114), dove (richiamando la celebre definizione aristotelica, Poet. 12, 1452b 23, dello στάσιμον come canto del coro senza ritmo anapestico e trocaico) lo stasimo è inteso come «‘ruhig gehaltenes Lied’... im Gegensatz zum ‘Marschlied’». Infine, nel dizionario L.S.J., s.v. στατιϰός, si legge: «πεϱὶ σ. ποιήσεως, composition of στάσιμα, title of work

Evoluzioni del coro e movimenti celesti  

by Ptolemaeus». Possiamo credere, tutto sommato e senza addentrarci in ulteriori speculazioni, che l’opera di Tolomeo trattasse di questioni relative alla composizione di quelle particolari forme poetiche definite στάσιμα (cfr. anche la testimonianza dell’Etym. Magn. citata subito sotto). Il tema di quello che possiamo dunque chiamare frammento di Tolomeo trova alcuni paralleli in un piccolo gruppo di testimonianze, che sono state riunite ed esaminate in parallelo da O. Crusius nell’articolo già menzionato (pp. 10 sg.). Si tratta in primo luogo di altri due brevi excerpta appartenenti al gruppo dei Capitula ad praefationem pertinentia riuniti da Drachmann in calce agli scoli pindarici, per la precisione dei Capp. b (p. 306, 16–24 Dr.) e c (p. 307, 7–18 Dr.). A questi si aggiungono: Etym. Magn. 690, 42 sgg. s.v. πϱοσῴδιον, nella parte sugli στάσιμα (690, 49 sgg.); lo sch. B a Efestione, p. 172, 29 sgg. Westphal (p. 200, 17 sgg. Gaisford, om. Consbruch; su questo breve estratto cfr. Hoerschelmann 1881, 272); uno sch. byz. a Euripide, Hec. 640. E infine due testi latini: Elio Festo Aftonio, p. 58, 5 sgg. e p. 60, 1 sgg. Keil, Gramm. Lat. VI [da Crusius (e Färber) citato come Mario Vittorino, a causa dell’ormai noto guasto tradizionale per cui alla mutila Ars grammatica di Mario Vittorino nei codici segue senza soluzione di continuità l’acefalo trattato metrico di Aftonio] e Atilio Fortunaziano, p. 294, 28 sgg. Keil, Gramm. Lat. VI. In queste fonti, fatte salve alcune differenze di dettaglio e di terminologia (e anche qualche problema testuale), si esplicita in modo inequivocabile, per quanto ancora una volta certamente grossolano, lo stesso concetto di base relativo al «significato» astronomico dei movimenti del coro: la strofe «raffigura» il movimento della sfera più alta, quella delle stelle fisse, da oriente a occidente; l’antistrofe il movimento in senso contrario delle sfere del sole e dei pianeti, da occidente a oriente; l’epodo la fissità della terra. Tuttavia, in nessuno di questi altri testi si trovano indizi sulla fonte di questa idea: l’analogia così stabilita in essi è data come totalmente anonima, e la citazione di Tolomeo da cui siamo partiti rimane unica nel gruppo delle testimonianze. Il riferimento al movimento in un senso eseguendo la strofe e nel senso opposto eseguendo l’antistrofe, con il noto fondamento etimologico che fa appello a στϱέφω e ἀντιστϱέφω, in alcuni testi (non nel nostro che cita Tolomeo) sembra precisarsi, oltre a quelle astronomiche, con indicazioni direzionali che chiamano in causa le nozioni di destra e sinistra. Nel Cap. b (p., 306, 19 sg. Drachmann) troviamo i due ordini di idee collegati sulla base di un riferimento a Omero. Si veda p. 306, 16 sgg. Drachmann: ἰστέον ὅτι οἱ λυριϰοὶ ἐν τοῖς ποιήμασιν αὐτῶν ἐχρῶντο στροφῇ, ἀντιστρόφῳ ϰαὶ ἐπῳδῷ. στροφῇ μέν, ὅτι ἀπὸ τῶν δεξιῶν ἐστρέφοντο ἐπὶ τὰ ἀριστερὰ, ᾗτινι ἀναλογεῖ ἡ τοῦ παντὸς ϰίνησις ἡ ἀπὸ τῶν ἀνατολιϰῶν ἐπὶ τὰ δυτιϰά, δεξιὰ γὰρ ϰαλεῖ ὁ Ὅμηρος τὰ ἀνατολιϰά,

  Evoluzioni del coro e movimenti celesti ἀριστερὰ δὲ τὰ δυτιϰά, ἀντιστρόφῳ δὲ, ὅτι ἐϰινοῦντο ἀπὸ τῶν ἀριστερῶν ἐπὶ τὰ δεξιά, ᾗτινι ἀναλογεῖ ἡ τῶν πλανήτων ϰίνησις ἡ ἀπὸ δύσεως εἰς ἀνατολάς. ἐπῳδῷ δὲ, ὅτι ἵσταντο ἐν ἐνὶ τόπῳ ϰαὶ ἔλεγον τὰς ᾠδὰς, ᾗτινι ἀναλογεῖ ἡ στάσις τῆς γῆς.

Dunque il movimento da oriente a occidente va da destra a sinistra, e viceversa. L’identificazione geografico-cosmologica della destra con l’oriente e la sinistra con l’occidente è riportata alla massima autorità della cultura greca, e per questo non mancano i fondamenti. A Iliade M 239 sg. si parla di trarre auspici da uccelli «che vanno a destra verso l’aurora e il sole, oppure a sinistra verso l’ombra nebbiosa»: e il commento di Aristonico dice che Omero conosce due dimensioni cosmiche, oriente e occidente, «ϰαὶ δεξιὰ μὲν τὰ πρὸς ἀνατολάς, ἀριστερὰ δέ τὰ πρὸς δυσμὰς λέγει». L’osservazione trova diversi paralleli negli scoli omerici (vedi per esempio sch. D ad M 239; sch. β 154), fra cui però vale la pena considerare lo sch. θ 29, anch’esso risalente ad Aristonico, dove si dice che Omero in quel verso menziona soltanto oriente e occidente senza aggiungere settentrione e mezzogiorno, perché gli antichi (οἱ ἀρχαῖοι) dividevano il cosmo semplicemente in oriente e occidente. Tutto questo è con buona probabilità pensiero di Aristarco, e comunque risalente alla maggiore filologia alessandrina: ma non possiamo prolungare oltre questa digressione. Insomma, in parallelo a quelle astronomiche relative all’asse oriente-occidente, vengono fornite le indicazioni direzionali di destra e sinistra, basate in sostanza (anche se non mancano imprecisioni e confusioni disinvolte, per cui destra e sinistra sono talora addotte in modo incoerente: ma le divergenze sono verosimilmente dovute soltanto a errori, come riteneva Crusius 1888, p. 11 nota 1) sull’identificazione destra-oriente e sinistra-occidente. È facile tuttavia osservare che le nozioni di destra e sinistra non offrono né aggiungono in realtà alcun dato concreto né indicazioni fattuali riscontrabili nella prassi esecutiva ma rimangono su un piano totalmente astratto (cfr. infra). Le divergenze che riscontriamo nel nostro gruppo di testi possono benissimo essere dovute a rimaneggiamenti di eruditi ed epitomatori, ma il fondo concettuale è sempre lo stesso, e autorizza chiaramente a pensare che alla base di questo insieme ci sia una sola fonte, identificabile con Claudio Tolomeo, utilizzata da qualche erudito tardo-antico per i suoi scopi e poi incanalatasi in rivoli strani. Era questa l’opinione di Crusius (1888, 12) e di Färber (1936, 21), che io credo sia da accogliere. Naturalmente non possiamo dire se l’idea sia stata proprio di Tolomeo o se egli l’abbia ripresa da altri, e mi sembra inutile, quando già difficoltà e incertezze sono molte, avventurarsi a speculare oltre. Accontentiamoci, per riassumere, di aver acquisito con buona probabilità che l’idea di un rapporto simbolicoanalogico fra le evoluzioni del coro e i movimenti celesti era in qualche modo

Evoluzioni del coro e movimenti celesti  

espressa da Claudio Tolomeo, in un’opera perduta di cui conosciamo il titolo. Se vogliamo, poi, è anche verisimile che Tolomeo non abbia inventato tale parallelismo, ma lo abbia ripreso da qualche precedente. Crusius (1888, 12) parlava di possibili origini dell’idea di fondo da teorici di orientamento pitagorico, ma Färber (1936, 22) ha giustamente sottolineato che non ci sono elementi per provarlo: e di fatto tali indicazioni restano nel vago. Credo solo si possa dire che, se Tolomeo ha avuto delle fonti, non si può comunque andare molto indietro rispetto a lui: mi pare indubitabile che abbiamo a che fare con una tarda speculazione. Dico tarda speculazione per affermare soprattutto che non possiamo evidentemente pensare a resoconti basati su osservazione effettiva, bensì ad elucubrazioni teoriche e libresche lontane da una prassi concreta (cfr. anche Reisch 1899, 2384 sg.). Färber riteneva che questi testi offrissero una testimonianza che strofe e antistrofe originariamente fossero un concetto relativo alla danza e alle evoluzioni del coro, e solo in seguito anche musicale: tuttavia, se i testi sono di origine tarda come sembra, bisognerebbe ammettere in essi un autorevole fondamento antico su questo piano e una reminiscenza di lunga durata, che non appaiono altamente probabili (cfr. ancora Reisch, loc. cit.). Temo, in conclusione, che in verità ci sia ben poco in questi testi da utilizzare per parlare della realtà orchestica e/o musicale dei cori antichi, siano essi cori drammatici o cori lirici: mi pare che alla base ci sia piuttosto un’idea del tutto teorica e — torno a dire — libresca del «coro» in astratto, un’idea speculativa in sostanza piuttosto generica. Al di fuori del gruppo di testi che abbiamo fin qui considerato, cioè quelli raccolti da Crusius, la stessa idea-base della corrispondenza analogica fra evoluzioni dei cori e movimenti celesti fa la sua comparsa in un passo di Macrobio, Commentarιi in Somnium Scipionis, II 3, 5. Macrobio sta parlando dell’armonia delle sfere celesti in rapporto con l’armonia musicale e il canto delle Muse «che rappresentano il canto del cosmo», e aggiunge: In ipsis quoque hymnis deorum per stropham et antistropham metra canoris versibus adhibebantur ut per stropham rectus orbis stelliferi motus, per antistropham diversus vagarum regressus praedicaretur, ex quibus duobus motibus primus in natura hymnus dicandus deo sumpsit exordium.

Dunque, dice Macrobio, la strofa esprime il movimento della sfera delle stelle fisse (zodiaco), l’antistrofe il moto retrogrado dei pianeti: da questi due movimenti avrebbe preso avvio e spunto il primo inno dedicato alla divinità. Manca il riferimento all’epodo, forse perché si tratta di un argomento introdotto rapidamente en passant senza soffermarsi a parlarne distesamente: ma per il resto

  Evoluzioni del coro e movimenti celesti abbiamo ritrovato gli stessi elementi già visti per il parallelo fra strofe-antistrofe e i movimenti dei corpi celesti (cfr. Wille 1967, 627). Il problema delle fonti dell’opera di Macrobio è sicuramente intricato, ma la sua dipendenza (diretta o indiretta) dal perduto commento di Porfirio al Timeo di Platone sembra ormai assodata (cfr. Scarpa 1981, 28 sg. e comm. a II 3, 5, p. 464; Regali 1983, p. 25). Nell’edizione appunto dei frammenti superstiti del commento di Porfirio al Timeo, Sodano include fra i frammenti certi, con i numeri LXVIII e LXXII, parti consistenti dello stesso capitolo II 3 (cui appartiene il nostro passo) dell’opera di Macrobio, ma il fr. LXVIII si ferma a II 3, 3 e il fr. LXXII inizia da II 3, 12, per cui II 3, 5 resterebbe escluso. Invece, nel suo commento all’opera di Macrobio, lo Scarpa sembra convinto dell’ascendenza porfiriana anche del contenuto di II 3, 5 (cfr. comm. ad loc., p. 464; diversamente Regali 1989, comm. ad loc.). Che l’idea della corrispondenza analogica fra strofe-antistrofe e movimenti celesti provenga davvero da Porfirio, deve dunque restare in dubbio: ci limitiamo a osservare che Porfirio, vissuto nel III secolo d. C., ci farebbe risalire abbastanza vicino all’epoca di Claudio Tolomeo, sopra riconosciuto come la fonte prima cui ricondurre il primo gruppo di testimonianze. La lignée (che non vuol dire per forza dipendenza diretta) Tolomeo (-Porfirio) Macrobio ci porta fino al pieno dell’età tardo-antica. Successivamente, nel Medio Evo e poi nel Rinascimento, la fortuna di questo parallelo analogico avrà conosciuto riutilizzazioni e rifunzionalizzazioni varie, che credo sarebbe interessante e divertente inseguire: ma anche molto impegnativo, lungo e soprattutto fuori dalle mie possibilità al momento. Però non voglio privarmi del gusto di menzionare un paio di occorrenze, che mi ha segnalato Lina Bolzoni. Cominciamo con Francesco Patrizi, filosofo e poeta cortigiano, partecipe del dibattito tardo cinquecentesco sulla poesia. Egli riprende e commenta nella sua Poetica (Deca istoriale, 1969, 258) proprio il passo di Macrobio di cui sopra, ricavandone l’intenzione degli antichi «e sacerdoti e poeti» di «sottontendere alcuno simbolico sentimento» non solo nei miti e nei misteri, ma anche «nelle figure, e ne’ gli ornamenti delle ceremonie, e ne’ movimenti, e balli». La stessa idea sembra poi alla base di un passo del poemetto Eridano, pubblicato nel 1557, dove Apollo è circondato dal coro delle Muse, che sono messe in relazione all’armonia delle sfere celesti e alla loro musica, il che pare chiaramente collegarsi con quanto Macrobio dice nei paragrafi di II 3 appena precedenti al passo sopra riportato (cfr. Bolzoni 1980, 51 e 61 nota 12). Più interessante ancora mi sembra la comparsa del tema in Tommaso Campanella: lo troviamo nei Commentaria, scritti da Campanella fra il 1627 e il 1629, alle poesie latine del papa Urbano VIII, e precisamente nel commento all’ode

Evoluzioni del coro e movimenti celesti  

intitolata De Sancto Ludovico Francorum Rege, dove Campanella scrive: (uso l’edizione in corso di stampa di Bolzoni): In odarum vero celebratione movebantur cantores iidemque instrumentorum pulsatores concinnis saltationibus a dextris ad sinixtras altaris partes: et hunc actum et melos vocabant stropham. Tunc e sinixtris ad dextras chorum ducentes, musica agitatione eisdem consonantiis metrisque agitabantur. Et hoc vocabant antistropham, idest contra stropham. Alii e medio ad aras prospectantes, canebant consistentes et hunc repraesentationem soni vocis et actus dicebant epodum. Nec easdem metricas rationes iste tertius servabat actus. Quis autem non videt hoc esse imitamen coelestium, quorum instar divinus Moses, nostri vatis typus, et tabernaculi et mensarum et chori tripudium instituit, iubente Domino: «Inspice, et fac omnia secundum exemplar, quod tibi in monte monstratum est.»... Esse autem in coelestibus motiones varias, non quidem qualiter ab Aristotile ponuntur sibi invicem adversantes, sed iuxta systematum ingenia, palam facit Sol, qui arae instar a Mercurio et Venere circumeuntibus et ab ignotis stellulis ... stropham et antistropham captat mediusque motus in supremo et infimo abside epodum repraesentat.

Il fatto che i movimenti «in odarum celebratione» siano visti come un «imitamen coelestium» (e si osservi che le indicazioni da destra a sinistra per la strofe e viceversa per l’antistrofe sono coerenti con il passo riportato sopra, basato sull’identificazione destra-oriente e sinistra-occidente, e quindi stelle fisse per la strofe e pianeti per l’antistrofe) è subito rapportato, con il richiamo a Mosè, all’idea biblica per cui le cose umane sono modellate «secundum exemplar». Nel prosieguo, non è più la terra come nei testi antichi visti sopra, bensì il Sole ad assumere la posizione centrale corrispondente all’epodo: strofe e antistrofe sono rappresentate dai pianeti che girano intorno ad esso, secondo il nuovo sistema galileiano. Non è questo il luogo per disquisire su Campanella, tanto più che è totalmente fuori dalle mie possibilità e competenze: mi pare evidente comunque che dall’idea di base egli abbia tratto sviluppi autonomi e «applicazioni» proprie che suscitano curiosità, sia per il ricondursi alla visione biblico-cristiana sia per il rimodellamento scientifico moderno di stampo galileiano. Se nel caso del Patrizi la fonte è dichiaratamente Macrobio, in Campanella gli elementi sembrano più ricchi e non si può escludere, credo, che il filosofo calabrese abbia conosciuto altre fonti antiche, come induce a sospettare il riferimento a Pindaro che si trova all’inizio e alla fine del commento citato: si può pensare a testi come quelli che abbiamo utilizzato sopra, letti magari in manoscritti pindarici corredati di excerpta eruditi oppure in luoghi di grammatici latini o altro ancora che ci sfugge. Abbandoniamo questa divagazione, per seguire un’altra strada che l’argomento ci apre. Come parallelo per l’idea della corrispondenza simbolica fra evoluzioni dei cori e movimenti celesti, già Crusius (1888, 12) aveva indicato quella costruzione che stabilisce una simile analogia fra il cosmo e, questa volta,

  Evoluzioni del coro e movimenti celesti il circo. La testimonianza più antica a questo proposito è costituita da un frammento (citato nella cronaca di Giovanni Malalas) di Carace di Pergamo (II secolo d. C.), dove si dice chiaramente che l’ippodromo è costruito sulla base del modello fornito dalla struttura generale dell’universo. Testimonianze successive possono essere confrontate con quella di Carace: l’esame degli elementi del parallelo strutturale che viene istituito sarebbe troppo lungo e complesso in questa sede, e rimandiamo alla recente riconsiderazione, con fonti e bibliografia, che si trova nell’articolo di E.B. Lyle citato in calce. Dunque anche per il circo l’idea di un simbolismo fondato sul parallelo analogico con il macrocosmo appare in testi che vanno dal II secolo d. C. in poi: e anche in questo caso possiamo affermare con la stessa probabilità di prima che ci troviamo di fronte a una elucubrazione piuttosto tarda; e se si vuole pensare a precedenti più antichi di Carace, sarà difficile risalire indietro di molti secoli. Potremmo fermarci qui, ma forse vale la pena tirare qualche conclusione ancora su un piano diverso, diciamo pure quello dei modelli culturali. I testi relativi al coro mostrano la costruzione di un modello analogico basato sull’immagine della struttura dell’universo (le sfere e i corpi celesti, la loro armonia, i loro movimenti), modello che dava interpretazione e senso all’idea che si aveva (se pur in astratto) della struttura delle evoluzioni di danza e dei comportamenti spettacolari dei cori nell’esecuzione dei loro canti. Nel caso del coro, tale modello forniva motivazioni e strumenti per capire fenomeni con ogni evidenza ormai lontani ed estranei, ma nella sua generalità il modello simbolico astratto poteva applicarsi anche ad una realtà presente e ben viva come il circo; e inoltre potè trasformarsi e rimanere vitale in epoche e contesti culturali molto diversi, come abbiamo visto. L’argomento rientra in una tematica generale molto ampia e ricca di sfaccettature, quella che riguarda la generale relazione macrocosmo/microcosmo: strumento di pensiero utilizzato per interpretare l’uomo e la natura, istituendo, grazie all’identificazione di un parallelismo fra le componenti, relazioni strutturali fra due totalità organiche che si corrispondono fra di loro. Fra i due termini del confronto si istituisce (e si costruisce) una determinazione analogica, che normalmente è molto lata o è empirica. Pensiamo per esempio all’ambito dell’antica medicina, alle teorie che spiegavano l’organismo umano sulla base del rapporto con gli elementi fondamentali della natura, ponendo un principio di corrispondenza e di sintonia fra individuo e natura nel suo complesso. Se nelle teorie mediche è il macrocosmo che fonda la spiegazione del microcosmocorpo umano, nelle antiche teorie cosmologiche filosofiche è di solito il microcosmo umano a fondare le spiegazioni del macrocosmo-universo; mentre diverso è l’atteggiamento dell’astrologia, che in generale pone come assunto quello

Evoluzioni del coro e movimenti celesti  

di una connessione diretta, che vuol dire influenza del macrocosmo sul microcosmo rappresentato dagli eventi umani. Ma torniamo al nostro caso, rappresentato dall’analogia strutturale istituita fra le evoluzioni dei cori e i movimenti celesti. L’allestimento e l’affermazione di un simile modello analogico ci induce ad un’ultima riflessione. L’insieme di metafore e immagini utilizzate dagli antichi filosofi greci fino a Platone e Aristotele per rappresentarsi l’universo ed esprimere le loro teorie cosmologiche sono state ben analizzate da Lloyd nel suo libro Polarity and Analogy (1966, cap. IV). Egli raggruppa queste immagini concrete in tre gruppi, vale a dire politiche, biologiche e tecnologiche: uno stato, un essere vivente, un artefatto offrivano il riferimento noto per rappresentarsi con una immagine l’ignoto che si voleva spiegare. Prevale dunque l’orientamento per cui si pensa nei termini: il cosmo è come a, b, c. Dopo la scienza ellenistica, e più che mai con Tolomeo, l’universo, le sfere, i movimenti celesti sono ormai stati oggetto di nutrite indagini e la loro conoscenza è via via molto cresciuta: il cosmo è diventato assai meno ignoto, anzi è ormai disvelato e la sua conoscenza è un possesso acquisito. Allora è proprio il cosmo, unità fornita di ordine e razionalità scientificamente espressa, che all’occorrenza può offrire un modello per interpretare altre cose, o realtà oscure ed estranee per ragioni cronologiche come i cori, oppure fenomeni attuali come il circo e i suoi spettacoli. Si potrà dunque ragionare nei termini per cui a, b, c è come il cosmo. E quando l’immagine del cosmo sarà cambiata, sarà il sole e non più la terra a simboleggiare la ‘fissità’ dell’epodo.

 Citazione di un’opera di Aristonico in Strabone La figura del grammatico Aristonico, contemporaneo di Strabone e di Didimo, è studiata principalmente — se non unicamente — per le opere sui segni critici di Aristarco a Omero (e a Esiodo), di cui sono rimasti cospicui frammenti soprattutto negli scoli all’Iliade e all’Odissea. Egli dedicò molte fatiche a esporre dettagliate spiegazioni dei segni critici aristarchei al testo omerico, facendosi mediatore del materiale risalente al maestro: l’implacabile selezione tradizionale ha fatto sí che la maggior parte di quanto si è conservato del suo lavoro provenga dall’opera Περὶ (τῶν) σημείων (τῶν) (τῆς) Ἰλιάδος καὶ (τῆς) Ὀδυσσείας,1 per cui Aristonico di solito ha suscitato e suscita interesse prevalentemente in quanto importante fonte di trasmissione dei materiali esegetici di Aristarco. Nel campo della critica omerica, pare che egli abbia prodotto anche hypomnemata all’Iliade e all’Odissea, distinti dallo scritto sui segni critici aristarchei: benché le testimonianze su questi commentari, per vero dire, non siano abbondanti e neppure esenti da dubbi, la citazione di opere definite ὑπόμνημα sembra indicare che si tratti davvero di opere diverse rispetto al o ai Περὶ (τῶν) σημείων κτλ. Nel quadro dei prodotti di esegesi omerica di Aristonico, la citazione in Strabone I 2, 31 pone un problema che vale la pena esaminare. Testo e traduzione: Ἐπεί δὲ καὶ τὰ περὶ τῆς πλάνης τῆς Μενελάου λεχθέντα συνηγορεῖν δοκεῖ τῇ ἀγνοίᾳ τῇ περὶ τοὺς τόπους ἐκείνους, βέλτιον ἴσως ἐστὶ τὰ ἐν τοῖς ἔπεσι τούτοις ζητούμενα προεκθεμένους ἅμα ταῦτά τε διαστεῖλαι καὶ περὶ τοῦ ποιητοῦ ἀπολογήσασθαι καθαρώτερον. φησί δὴ πρὸς Τηλέμαχον ὁ Μενέλαος θαυμάσαντα τὸν τῶν βασιλείων κόσμον· ἧ γὰρ πολλὰ παθὼν καὶ πόλλ᾿ ἐπαληθεὶς ἠγαγόμην ἐν νηυσὶ καὶ ὀγδοάτῳ ἔτει ἦλθον, Κύπρον Φοινίκην τε καὶ Αἰγυπτίους ἐπαληθείς, Αἰθίοπάς θ᾿ ἱκόμην καὶ Σιδονίους καὶ Ἐρεμβοὺς καὶ Λιβύην (δ 81–85). ζητοῦσι δὲ πρὸς τίνας ἦλθεν Αἰθίοπας πλέων ἐξ Αἰγύπτου· οὔτε γὰρ ἐν τῇ καθ᾿ ἡμᾶς θαλάττῃ οἰκοῦσί τινες Αἰθίοπες, οὔτε τοῦ Νείλου τοὺς καταράκτας ἦν διελθεῖν ναυσί· τίνες τε οἱ Σιδόνιοι· οὐ γὰρ οἵ γε ἐν Φοινίκῃ· οὐ γὰρ ἂν τὸ γένος προθεὶς τὸ εἶδος ἐπήνεγκε· τίνες τε οἱ Ἐρεμβοί· καινὸν γὰρ τὸ ὄνομα. Ἀριστόνικος μὲν οὖν ὁ καθ᾽ ἡμᾶς γϱαμματικὸς ἐν τοῖς πεϱὶ

 1 Quale che sia il titolo esatto; edizioni: Friedländer 1853, poi naturalmente nell’edizione degli Scholia Vetera di Erbse 1969–1988; Carnuth 1869. Bibliografia: Cohn 1895; Lehrs, 1882, 1–15; Ludwich 1884–1885, passim (v. indice a p. 725); Erbse 1960, 174ss.; van der Valk 1963–1964, I 553ss.; Pfeiffer 1973, 334, 340, 342. n. 59, 344, 352, 355, 360, 368; Montanari 1993, 279 (= this volume, ch. 1). https://doi.org/10.1515/9783110772371-006

Citazione di un’opera di Aristonico in Strabone  

τῆς Μενελάου πλάνης πολλῶν ἀναγέγραφεν ἀνδρῶν ἀποφάσεις περὶ ἑκάστου τῶν ἐκκειμένων κεφαλαίων· ἡμῖν δ᾿ ἀρκέσει κἂν ἐπιτέμνοντες λέγωμεν. «Ora, poiché si ritiene che anche i racconti [di Omero] intorno alle peregrinazioni di Menelao rivelino l’ignoranza di quei luoghi, forse è meglio passare prima in rivista i problemi dibattuti a proposito di questi versi e nel contempo distinguere tali questioni e difendere il poeta con maggiore chiarezza. Quando Telemaco guarda ammirato le decorazioni del palazzo reale, Menelao gli dice: Dopo aver molto sofferto e molto vagato le portai nelle navi e arrivai all’ottavo anno, dopo aver vagato per Cipro e la Fenicia e fra gli Egizi, giunsi fra gli Etiopi e i Sidoni e gli Erembi e in Libia (IV 81–85). Ci si chiede presso quali Etiopi mai arrivò navigando dall’Egitto: non ci sono Etiopi stanziati nel nostro mare e non era possibile attraversare con le navi le cateratte del Nilo; e chi sono i Sidoni: non certo quelli della Fenicia, perché non avrebbe detto prima la stirpe e poi aggiunto la specie; e chi sono gli Erembi: il nome è altrimenti ignoto. «Aristonico, il grammatico nostro contemporaneo, scrivendo sulle peregrinazioni di Menelao ha registrato le cose che molti uomini hanno sostenuto su ciascuno degli argomenti suddetti: per noi basterà dire qualcosa in breve».

Il problema suscitato dalla citazione di Aristonico in questo passo è se περί τῆς Μενελάου πλάνης sia il titolo preciso di un’opera di Aristonico, che dunque sarebbe una monografia, un syngramma particolare dedicato all’argomento delle preregrinazioni di Menelao. Sembra questa l’opinione che emerge da due diffuse traduzioni di Strabone, quella di H.L. Jones nella Loeb:2 «Now Aristonicus, the grammarian of our own generation, in his book On the Wanderings of Menelaus, has recorded opinions of many men...»; e quella di G. Aujac per Les Belles Lettres:3 «Vu que le grammarien actuel Aristonicos, dans son ouvrage Sur le périple de Ménélas, a collectionné diverses interprétations...». Invece L. Cohn, nell’articolo dedicato a Aristonico sulla RE, riteneva che la citazione provenisse dall’hypomnema all’Odissea, e precisamente dalla parte dedicata al IV libro:4 non dunque un syngramma particolare bensí una sezione del commentario all’Odissea, citata in questo modo per il suo specifico contenuto, per gli argomenti che trattava commentando alcuni versi omerici del IV libro.

 2 Jones 1917, I 139. 3 Aujac/Lasserre 1969, 126. 4 Cohn 1895, 966.

  Citazione di un’opera di Aristonico in Strabone Forse si può osservare che, all’inizio della porzione di testo riportata, Strabone introduce il tema parlando di τὰ περὶ τῆς πλάνης τῆς Μενελάου λεχθέντα, cioè di «quello che Omero dice intorno alle peregrinazioni di Menelao», il che farebbe pensare che poi abbia ripreso questa semplice indicazione di contenuto citando Ἀριστόνικος ... ἐν τοῖς περὶ τῆς Μενελάου πλάνης. Tuttavia, in mancanza di altri elementi, è senz’altro difficile dire quale sia la verità e la questione mi pare di quelle che devono restare sostanzialmente aperte: precisa citazione del titolo di una monografia specifica di Aristonico oppure solo un modo della fonte (Strabone) per fare riferimento al contenuto di una sezione del suo commentario? Devo confessare una qualche propensione per la seconda ipotesi: ma devo anche dire che si tratta di un’inclinazione intuitiva e poco concretamente motivata, che condivido con il vecchio articolo di Cohn. La questione in sé, invece, mi pare evocare una tematica importante, cui vale la pena accennare brevemente. La tematica è quella della cosiddetta πεϱίLiteratur o Problemata-Literatur e delle sue caratteristiche entro il quadro dei rapporti fra opere biografiche e opere esegetico-erudite, per quanto riguarda sia i materiali che le forme letterarie e i modi della tradizione.5 Un aspetto di questa problematica si colloca propriamente all’interno della letteratura esegetica e riguarda forme, materiali e caratteristiche dei syngrammata e degli hypomnemata, cioè di due delle produzioni piú specifiche della tradizione filologicogrammaticale ellenistico-romana. L’immediata identificazione di un titolo-περὶ con un syngramma, vale a dire con forma e contenuto di tipo monografico e saggistico, rischia assolutamente di essere troppo meccanica: anche in considerazione del fatto che il titolo-περὶ fornito dalla fonte a nostra disposizione può benissimo non rispecchiare letteralmente il titolo originale, e quindi in sostanza non essere affatto il vero titolo dell’opera, ma essere una formulazione dovuta alla fonte medesima oppure a qualcuno degli intermediari precedenti, talvolta numerosi.6 Il problema, credo, deve essere ripreso in considerazione ed esaminato piú a fondo, analizzando caso per caso le citazioni fornite dalla letteratura erudita, con molta attenzione al modo della citazione e alla tradizione che sta

 5 Classica trattazione in Leo 1901, ripresa nelle recensioni al De Demosthene di Didimo e alla Vita di Euripide di Satiro, rispettivamente in NGG 1904, 254ss. e in NGG 1912, 273ss. = Leo 1960, 387ss. e 365ss.; cfr. Pfeiffer 1973, 238s.; il tema è stato riconsiderato da Arrighetti 1987, 180ss., con ampia bibliografia; recentemente cfr. anche I. Gallo, «Nascita e sviluppo della biografia greca: aspetti e problemi», in Gallo/Nicastri 1995, partic. 21s. 6 Su questi problemi di exact wording ho attirato l’attenzione nell’articolo Montanari 1997 (= this volume, ch. 42).

Citazione di un’opera di Aristonico in Strabone  

dietro la fonte che la fornisce. Sono molti gli esempi imbarazzanti che vengono alla mente, dal ben noto Περὶ Δημοσθένους di Didimo7 alla forma dei titoli delle opere utilizzate dal cosiddetto Commentario dei Quattro (VMK):8 e fra questi va considerato anche il caso della citazione di Aristonico in Strabone che abbiamo visto.

 7 Cfr. Arrighetti 1987, con bibliografia. 8 Oltre allo stesso Aristonico, sono Didimo, Nicanore, Erodiano: si vedano le sottoscrizioni ai vari libri dell’Iliade conservate nel celebre manoscritto Ven. A, riprodotte alla fine di ogni libro nell’edizione degli Scholia Vetera di H. Erbse.

 Ekdosis alessandrina: Il libro e il testo Nel periodo da Zenodoto ad Aristarco l’ekdosis si impone nel quadro della cultura antica come risultato tipico dell’opera dei filologi alessandrini, accanto allo hypomnema, al syngramma, alla raccolta di lexeis e altri prodotti esegeticoeruditi. Cosa effettivamente significasse la realizzazione della ekdosis di un’opera letteraria, quale fosse la sua forma concreta e quale il modo in cui essa veniva materialmente allestita da un grammatico, è da tempo oggetto di discussioni: il problema peraltro ha conseguenze anche sulla ricostruzione e valutazione dei metodi e dei risultati dell’attività filologica di cui l’ekdosis è prodotto, come vedremo. Me ne sono occupato negli ultimi anni in alcuni lavori, a cui rimando per documentazioni e argomentazioni qui omesse o riassunte:1 ma voglio tornare in questa occasione a rifletterci, anche per sottolineare meglio alcuni aspetti della questione. Premetto che mi sembra da valorizzare l’importanza centrale del rapporto fra l’aspetto librario e il testo oggetto di edizione con i vari elementi di contorno “paratestuale”, quali annotazioni e semeia. Bisogna chiamare in causa con il dovuto peso quanto sappiamo della prassi consolidata per la realizzazione di nuove copie dei testi (negli scriptoria da parte di scribi professionali oppure privatamente da parte di singole persone) e quanto possiamo constatare negli esemplari pervenuti. Problemi decisivi si sono posti a partire dall’interpretazione del modo di lavorare di Zenodoto, il primo dei maggiori filologi alessandrini. Scriveva Rudolf Pfeiffer: «It is not improbable that Zenodotus, examining manuscripts in the library, selected one text of Homer, which seemed to him to be superior to any other one, as his main guide; its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Διόρθωσις can be the term for either kind of correction. It is hard to imagine any other way».2 A Zenodoto ha dedicato importanti studi K. Nickau, che riassume a questo proposito: «Dann ist zu fragen, ob Z(enodotos) nicht einen durch Recensio ermittelten Homertext zugrundelegte (der jedoch nicht seinen Vorstellungen von der genuinen Form der Epen entsprach), diesen mit Obeloi versah und zu ihm Textvorschläge sowie deren Begrundung mitteilte. Z(enodotos) selbst wie auch seine Hörer machten sich entsprechenden Notizen, die, wären sie von Z(enodotos) schriftlich veröffentlicht worden, ‘Hypomnemata’ hätten heißen können. Aber

 1 Vedi Bibliografia in fondo. 2 Pfeiffer 1968, 110 = 1973, 188–189. https://doi.org/10.1515/9783110772371-007

Ekdosis alessandrina: Il libro e il testo  

die Zeit der schriftlich publizierten Homer-Kommentare begann erst mit Aristarchos. So würden sich auch die späteren Unsicherheiten in der Berichterstattung über Z(enodotos)s Ausgabe erklären».3 L’idea di fondo, condivisa dai due autorevoli studiosi,4 esclude la possibilità che l’ekdosis zenodotea consistesse in un nuovo esemplare recante il testo continuo voluto dal grammatico, cioè l’intera opera completamente riscritta da lui (o per lui), con le sue lezioni incorporate, con gli obeloi nei margini a indicare le atetesi, eliminati e quindi assenti i versi giudicati da omettere. Come dice Pfeiffer, è difficile immaginare un modo diverso dal lavoro di correzione eseguito su una copia già esistente, opportunamente scelta fra quelle a disposizione, utilizzata come testo-base sul quale operare gli interventi voluti e via via annotata dal grammatico con il procedere dei suoi studi e del suo lavoro di diorthosis. Anch’io ritengo che questa sia la visione da accogliere e sulla quale fondare i successivi ragionamenti. Per ampiezza di documentazione e di conoscenze, la discussione riguarda specificamente la storia del testo omerico. Per questo aspetto basilare, tuttavia, mi pare si possa pensare agevolmente a una procedura grosso modo simile per gli autori oggetto di cura filologica da parte dei grammatici alessandrini. Un simile modo di produzione dell’ekdosis fu adottato da Zenodoto e rimase in uso con i grammatici posteriori. Un filologo sceglieva, secondo le proprie preferenze, un esemplare che gli risultava adatto come base di lavoro: quando il testo non incontrava la sua approvazione, in corrispondenza del luogo interessato egli annotava la lezione preferita negli spazi liberi o nell’interlinea. Il suo testo risultava dall’insieme rappresentato dal testo-base più le indicazioni di modifica contenute nel “contorno” paratestuale. Accanto ai versi si mettevano i semeia opportuni: cominciò Zenodoto con il solo obelos per la proposta di atetesi, poi il sistema si arricchì e si differenziò considerevolmente. Resta oggetto di dubbio definire in quale misura la copia di lavoro fosse corredata anche di brevi e sporadiche annotazioni esplicative, che pure dovevano esserci:5 qui ci può essere stata una evoluzione, nel senso indicato sopra da Nickau, fino all’esteso hypomnema separato. Dobbiamo immaginare tutto questo come esito di anni di studio, che producevano via via svariati interventi sullo stesso esemplare: istoriato e martoriato dal lavoro di diorthosis, esso era materialmente l’ekdosis omerica di quel filologo, la sua copia personale, recante il suo nome per identificazione  3 Nickau 1972, 30–31. 4 Cfr. anche alla n. 18 per l’opinione di van Thiel. 5 Per i marginalia su papiro vedi McNamee 2007: cfr. General Index s.v. correction(s), corrector(s).

  Ekdosis alessandrina: Il libro e il testo e contenente il frutto delle sue riflessioni e indagini, ekdosis in quanto appunto ἐκδοθεῖσα, cioè a disposizione di dotti, scolari, poeti e intellettuali. Fino a quando non fu introdotto l’uso di scrivere commentari, con spazi più ampi che consentivano di addurre ragionamenti e materiali di supporto, l’esemplare di lavoro recante in margine i semeia, le lezioni divergenti ed eventuali brevi note fu la fonte essenziale per conoscere le opinioni di un filologo sul testo da lui studiato. Ad esso si potevano affiancare opere di diverso genere, come trattazioni a tema o raccolte lessicografiche, oppure anche la memoria dell’ambiente e di “scuola”. Doveva essere così per Zenodoto, per Aristofane e in generale per i grammatici prearistarchei, con gli elementi di incertezza sul loro operato che conosce bene chi si occupa di queste cose. Con Aristarco certamente rimase l’uso di produrre una ekdosis, costituita dall’esemplare di lavoro con il suo “contorno”, ma ad essa si affiancò spesso lo hypomnema, che divenne il recipiente ideale per sviluppare le argomentazioni su una quantità e varietà di temi filologico-esegetici. Mi risulta difficile trovare argomenti contro questa idea di fondo, secondo cui il lavoro per realizzare una ekdosis di carattere filologico si basava di fatto su modalità e pratiche di lavoro sicuramente non nuove e certamente per nulla stravaganti nel mondo intellettuale e colto, ormai abituato ad essere una “civiltà del libro” da almeno un paio di secoli. Non c’è dubbio (e lo sostengo da tempo) che con il lavoro filologico dei grammatici alessandrini, a partire dalla prima generazione, qualcosa di nuovo sia comparso nella storia culturale e sia stato compiuto un progresso intellettuale importante. D’altra parte, mi pare che questa “rivoluzione” acquisti realtà ancor più solida se ne vengono evidenziati i precedenti e le basi che hanno fornito avvio e nutrimento a sviluppi e applicazioni nuovi nei metodi e negli scopi. È un dato di fatto che esemplari di opere letterarie erano normalmente oggetto di rilettura e correzione, grazie a un nuovo confronto con l’antigrafo, talvolta anche in base a una collazione con altre copie. Gli studiosi di papirologia conoscono bene questi fatti. Ci interessano naturalmente le testimonianze più antiche, ma siamo condizionati dal fatto che i più antichi papiri greci a noi pervenuti, della seconda metà del IV sec. a.C., sono pochissimi. Uno di essi, forse il più antico, è il celebre frammento dei Persiani di Timoteo, P.Berol. inv. 9875. Alla col. IV, v. 133, dopo avere scritto βορεαιαραισον|ται, lo scriba ha inserito un δ supra lineam per ripristinare la corretta lezione βορέᾳ διαραίσον|ται (tavola 1). Alla col. V, v. 196, dopo avere scritto πλουτουοδε, lo scriba ha inserito uno ι supra lineam per ripristinare la corretta lezione πλούτου oἱ δὲ (tavola 2). Due interventi di correzione (διόρθωσις) realizzati con ogni probabilità in seguito a una rilettura: non sono una cosa particolarmente vistosa, ma sono le testimonianze più antiche

Ekdosis alessandrina: Il libro e il testo  

a disposizione e si collocano poco prima o in contemporanea con l’attività di Zenodoto (330 ca.–260 ca.). Occasionali correzioni della prima mano si trovano anche nel celebre Papiro di Derveni, anch’esso degli ultimi decenni del IV secolo.6 Al III sec. a.C. appartiene il papiro milanese con gli epigrammi di Posidippo, P.Mil.Vogl. 309, pubblicato di recente. Il rotolo presenta numerose correzioni: la maggior parte sono opera dello stesso scriba del testo; dopo di lui, altre due mani sono intervenute con ulteriori emendamenti.7 Si tratta di un esemplare assai rilevante per il nostro discorso, per la quantità degli interventi e per l’epoca alta cui appartiene. Vediamo un solo caso (tavola 3). Alla col. XI, r. 30 si legge κεντρακαιεξω[; nel margine superiore, una delle due mani intervenute dopo quella dello scriba principale ha scritto καικεντρα (le ultime tre lettere non si vedono in fotografia ma si possono leggere nell’originale): è assai probabile che si tratti di una correzione o di una variante, probabilmente per il κέντρα καί del testo si propone una inversione καὶ κέντρα, ma il tutto non è chiaro, anche perché non è conservato il resto del verso.8 I ritrovamenti papiracei sono particolarmente abbondanti nei secoli II–III della nostra era e nei testimoni di questo periodo troviamo una ricca documentazione. Proponiamo soltanto qualche esempio significativo. Il P.Oxy. 2404, datato fine II – III sec. d.C., contiene i §§ 51–53 dell’orazione Contro Ctesifonte di Eschine (tavola 4). Una seconda mano ha corretto il testo originale, con aggiunte sopra e nel rigo e con cancellature. Queste ultime sono realizzate in vario modo: punti sopra le lettere al r. 5, un tratto di penna sopra le lettere al r. 17 e sulle lettere al r. 18. Al r. 5 e al r. 15 si vedono aggiunte supra lineam, ai rr. 17–18 il testo (§ 53) è stato riveduto e corretto con diversi interventi, ottenendo infine un assetto testuale del tutto diverso da quello tradito nei codici. Si ritiene che la revisione sia stata effettuata per mezzo della collazione con un secondo esemplare: a giudicare da quello che abbiamo, essa appare sistematica.9 P.Oxy. 2161, del II sec. d.C., contiene i Diktyoulkoi di Eschilo (tavola 5). Lo scriba ha occasionalmente corretto egli stesso alcuni errori: per esempio al v. 831 aveva scritto ηδη, ma poi lo ha cancellato con un tratto obliquo sopra ciascuna lettera e ha scritto supra lineam la lezione corretta o]ιoν. P.Berol. inv. 9872 (BKT II), del II sec. d.C., è un lungo rotolo (75 colonne più diversi frammenti non collocati) che restituisce un commentario al Teeteto di

 6 Turner/Parsons 1987, 92; edizione in Kouremenos/Parassoglou/Tsantsanoglou 2006. 7 Bastianini/Gallazzi 2001, 15. 8 Bastianini/Gallazzi 2001, ad loc. 9 Cfr. il testo in P.Oxy. 2404 e nell’edizione di Dilts 1997; per un nuovo frammento del rotolo e per la datazione cfr. Messeri Savorelli/Pintaudi 1997, 172–174.

  Ekdosis alessandrina: Il libro e il testo Platone con molti interventi correttivi. Così descrivono gli ultimi editori (tavola 6): «Il volumen è stato sottoposto a revisione e corretto in più punti: parole o lettere omesse nella stesura originaria sono state reintrodotte, lettere o parole superflue sono state cancellate, lettere ritenute errate sono state sostituite con quelle giudicate esatte. Tutti questi interventi non sembrano presupporre necessariamente una collazione con un esemplare diverso da quello di copia (l’affermazione contraria di McNamee 1981, 90 non appare sufficientemente fondata) ... La varietà del modo con cui le emendazioni sono state attuate può indurre il sospetto che il rotolo sia stato corretto a più riprese: una prima mano (quella di un diorthotès dello scriptorium) ha aggiunto le parole saltate, che sono riportate nel margine superiore ... o inferiore... oppure sono poste a proseguire il rigo direttamente nell intercolunnio ... Una mano successiva, o forse più mani, sembrano poi avere ripercorso tutto il testo, cancellando da capo con un tratto d’inchiostro tutte le lettere ritenute errate».10 Per esempio, alla col. LXIII r. 6 lo scriba aveva scritto προσαλλαουτεσχη, omettendo qualche parola: nell’intercolunnio a sinistra il correttore ha posto il segno dell’àncora a puntare verso l’alto e nell’interlinea fra αλλα e ουτε ha scritto ἄνω; nel margine superiore si leggono le parole θεωρειται ουτε γαρ χρω|μα κα(τω), che probabilmente erano precedute da un’àncora ora in lacuna. Il testo ripristinato è dunque πρὸς ἄλλα θεωρεῖται οὔτε γὰρ χρῶμα οὔτε σχῆ|μα. P.Oxy. 2256, del II–III sec. d.C., contiene hypotheseis di varie tragedie di Eschilo (tavola 7). La hypothesis frammentaria del fr. 3 ricorda la vittoria, con la trilogia di cui facevano parte le Danaidi, contro Sofocle e un altro autore, probabilmente Mesato (r. 5). Dopo il nome di quest’ultimo e all’inizio del successivo r. 6, si vedono chiaramente le parentesi, generalmente usate come segno di espunzione, sia in testi letterari che in testi documentari. Qui è evidente che le parentesi sono poste in scribendo, il che si può spiegare solo pensando che lo scriba copiasse da un esemplare dove le espunzioni erano già presenti a indicare che i drammi messi tra parentesi erano stati erroneamente posti dopo il nome di Mesato.11 La copia del Vangelo di Giovanni restituita da P.Bodmer 2 risale al III sec. d.C. (Tavola 8). Lo scriba ha corretto il proprio testo in vari modi. Ci sono aggiunte supra lineam (rr. 2 e 12) e riscritture sopra cancellazioni con una spugna: ai rr. 9–10 εταραχθη è stato scritto su una parola “lavata” via e che continuava nel rigo seguente, dove è rimasto un vuoto nel quale si intravvede σατο; la seconda parte del r. 10 è riscritta; all’inizio del r. 11 τον è un resto di lezione eliminata,  10 Bastianini/Sedley 1995, 243–244. 11 Arata/Bastianini/Montanari 2004.

Ekdosis alessandrina: Il libro e il testo  

successivamente segnato con punti di espunzione e anche piccole parentesi supra lineam. La documentazione per la fase più antica non è abbondante, come è naturale perché scarsi sono i reperti in generale, ma con il proseguire del tempo essa diventa cospicua ed estremamente significativa: i pochi esempi che abbiamo visto possono essere facilmente incrementati e arricchiti. I papiri dunque ci testimoniano bene l’uso di diverse modalità per migliorare un esemplare librario, vale a dire correggere la copia di un testo, cancellare quello che si riteneva sbagliato per sostituirlo con ciò che si riteneva corretto, scrivendolo sopra la riga, nei margini e negli intercolunni (talvolta con segnali di richiamo), oppure anche sopra le parole precedenti. Per cancellare qualcosa, si poteva tirare un tratto orizzontale o obliquo (anche molto vistoso) sulle lettere o le parole da eliminare, oppure esse potevano venire contrassegnate da punti o da linee al di sopra o al di sotto, oppure ancora essere racchiuse entro una sorta di parentesi tonde in coppia, o addirittura essere erase o lavate via con una spugna.12 Come scrivono Turner e Parsons: «One of the questions the palaeographer should ask about any literary manuscript is whether it has been adequately compared against its antigraph (the exemplar from which it was copied), a task which, in a publishing house, was the duty of the diorthotès, corrector, or whether it has been collated with a second exemplar a procedure often carried out by private individuals to secure a reliable text». «But several of our surviving papyrus manuscripts, and especially those which are beautifully written, contain such serious un-noted errors that it is clear their ‘proof-reading’ was of a summary, superficial kind, if done at all ... Those ancient themselves who set store by having a dependable copy (persons like Strabo and Galen) were aware of this weakness and adopted a routine to counter it: they themselves (or their secretaries) checked the copy to be used against another exemplar. If, therefore, the text had been checked against its first exemplar, and was later collated with a second, it may well bear the marks of this double checking».13 La migliore pratica di produzione libraria prevedeva operazioni di confronto fra copie e interventi di correzione, diòrthosis, realizzati da un diorthotès di professione o occasionale, che aveva valide risorse pratiche per cancellare, aggiungere, sostituire, segnalare vari aspetti e caratteri del testo allo scopo di migliorarlo e renderlo più affidabile. Le analogie con il lavoro filologico sono evidenti  12 Turner/Parsons 1987, 15–16, con rinvio a esempi nelle tavole; cfr. Turner 1980, 93 e pl. VIII (= 1984, 112–113 e tav. VIII); Bastianini 2001. 13 Turner/Parsons 1987, 15–16; Turner 1980, 93 (= 1984, 112–113).

  Ekdosis alessandrina: Il libro e il testo e devono essere valorizzate: i metodi e la tecnologia libraria offrirono una base per così dire tecnico-artigianale, che trovò applicazione e sviluppo nel lavoro del grammatico. Un procedimento che non doveva apparire particolarmente strano o stravagante, ma implicava un principio innovativo straordinario: per Zenodoto, ricordato come il primo diorthotès di Omero, la diorthosis del correttore dello scriptorium divenne la diorthosis del filologo.14 Possiamo pensare di concettualizzare il passaggio riflettendo nei termini seguenti. Lo scopo che guidava il lavoro del correttore di uno scriptorium era quello di produrre una copia corretta da vendere a un compratore-committente, vale a dire un esemplare realizzato nel modo migliore possibile sul piano artigianale. L’idea di fondo del grammatico nel correggere il suo testo era invece quella di trovare la forma giusta dell’opera di cui si stava occupando: lavorava su una copia con lo scopo e l’intenzione di realizzare l’esemplare modello, nel quale si potesse trovare quella che era secondo lui la forma esatta dell’opera letteraria, compresa l’indicazione quanto possibile codificata di dubbi e aporie testuali, un aspetto che certo non apparteneva al sistema mentale e all’orizzonte operativo dello scriptorium e dei suoi artigiani. Così alla cancellazione drastica (operazione tipica del correttore artigianale della produzione libraria) si affiancò dapprima il segno del dubbio filologico, l’obelos, che segnò il cambiamento intellettuale decisivo nel senso di pensare all’opera in quanto tale invece che alla singola copia. Poi il progresso della disciplina andò sempre più nella direzione di incrementare e sviluppare i segnali di intervento critico, di riflessione e di indagine, non tanto su una singola copia con intento produttivo artigianale, quanto sul recupero della volontà autentica dell’autore, dunque anche sull’interpretazione corretta del testo oggetto di studio. La diorthosis del grammatico non aveva di mira il singolo esemplare, pensava invece alla forma dell’opera in se stessa: alcuni strumenti pratici collaudati dalla produzione libraria certamente risultarono comodi e furono utilizzati, ma con finalità diverse e atteggiamenti intellettuali di altro genere. Zenodoto lavorò dunque su un esemplare da lui scelto,15 che però gli suscitava non poche riserve, riguardanti sia il numerus versuum che una certa quantità  14 Nickau 1977, 10 sg. 15 Sulla scelta della copia di lavoro da parte di Zenodoto cfr. West 1998–2000, VII; West 2001a, 33–45, id., West 2001b e West 2002 (approvato da Janko 2002). Egli ipotizza che la particolare eccentricità del testo di Zenodoto non possa essere interamente dovuta soltanto al suo giudizio e alle sue opinioni: almeno in parte deve riflettere una eccentricità della tradizione da cui egli dipende e su cui si basa. L’idea è che egli abbia lavorato su un esemplare rapsodico prodotto in un contesto ionico probabilmente nel IV secolo, esemplare nel quale si rifletteva dunque una linea tradizionale diversa rispetto a quella diffusasi in seguito e di origine sostan-

Ekdosis alessandrina: Il libro e il testo  

di lezioni. Aveva dubbi sull’autenticità di alcuni versi e quindi adottò il segno dell’obelos per la proposta di espunzione e lo tracciò accanto ad essi. Questo era lo ἀθετεῖν, la ἀθέτησις, un’azione del tutto nuova, che implicava un principio intellettualmente raffinato: l’idea di un dubbio testuale che poteva essere indicato e lasciato come tale al giudizio di altri. Ma le informazioni parlano anche di versi che egli riteneva da eliminare decisamente dal testo, evidentemente perché li considerava senz’altro spuri, ed è risultato più difficile capire come andassero le cose a questo proposito. Si trattava di versi che comunque dovevano essere presenti nel suo testo-base, proprio come quelli di cui proponeva lo ἀθετεῖν per mezzo dell’obelos: altrimenti non si capirebbe il senso di indicarne la eliminazione. Per questi casi nelle fonti scoliografiche troviamo usate le espressioni οὐ γράφειν (la più frequente), oppure οὐκ εἶναι, οὐ φέρεσθαι, ed è sempre stato un problema ricostruire esattamente e concretamente la differenza di operazione indicata dalla differente terminologia.16 Quanto abbiamo visto sopra offre la chiave per capire. La copia usata da Zenodoto presentava sia versi segnati con obelos che versi decisamente cancellati con mezzi grafici. Accanto a questi ultimi, egli poteva annotare qualcosa come οὐ γράφειν o altro termine equipollente;17 oppure affidarsi al solo segno di cancellazione senza annotazioni “verbali”, nel qual caso la terminologia per l’eliminazione di versi può essere stata riportata da chi seguì il suo insegnamento al Museo, oppure essere stata creata o modificata o ampliata dalla tradizione successiva per dar conto dei suoi interventi (concepibile soprattutto per οὐκ εἶναι e οὐ φέρεσθαι). Mi sembra la spiegazione più plausibile della terminologia usata per gli interventi di Zenodoto di atetesi e di omissione di versi, compreso il problema della differenza materiale, libraria e grafica fra i versi atetizzati e i versi eliminati.18  zialmente attica: è plausibile che esso provenisse da Efeso, la sua città di nascita, e che egli se lo fosse portato con sé ad Alessandria. Cfr. Montanari 2002a (= questo volume, cap. 47). 16 Cfr. Nickau 1972; 1977, 1–30. Non si avvede del problema Ercolani 2006, 224–225. 17 I verbi περιγράφω e διαγράφω sono termini tecnici per “cancellare” con i mezzi materiali di cui sopra: cfr. Turner/Parsons 1987, 16; alcuni esempi sono rimasti negli scoli: per περιγραφή cfr. Nickau 1977, 10–12 e 29; Dickey 2007, 174 e 254. 18 In Montanari 1998b, (= questo volume, cap. 37), ho discusso le posizioni di van Thiel (cfr. 1992; 1997), che ritiene le ekdoseis alessandrine “annotierte Homertexte”. La mia idea dell’ekdosis di Zenodoto, per quanto riguarda il punto di partenza e la base di lavoro, non si discosta da quella che abbiamo visto emergere da Pfeiffer e Nickau, e anche dai presupposti di van Thiel. Preciso alcuni aspetti di dettaglio. Rispetto a Nickau, io ritengo che nella copia usata come testo-base Zenodoto scrivesse di più e che sostanzialmente questa copia recasse alla fine tutti gli interventi sul testo voluti dal grammatico: gli obeloi per le atetesi, le cancellature, le lezioni ritenute giuste annotate in margine. Rispetto a van Thiel, io non credo assolutamente

  Ekdosis alessandrina: Il libro e il testo La copia usata da Zenodoto doveva dunque recare tutti gli interventi sul testo voluti dal grammatico: le atetesi per mezzo degli obeloi e le eliminazioni per mezzo delle cancellature, mentre negli spazi liberi erano scritte le lezioni ritenute giuste e talvolta anche stringate annotazioni esegetiche. È difficile dire fino a quando l’esemplare zenodoteo sia rimasto disponibile. Penso che quello che si è conservato delle argomentazioni e dei ragionamenti sul testo omerico del primo grammatico diorthotès della poesia epica provenga essenzialmente da tre fonti: memoria, più o meno aiutata da appunti, delle sue lezioni nel Museo; ricostruzioni a posteriori, aiutate da una breve annotazione (se e quando c’era) oppure dedotte solo dagli interventi testuali; materiali reperibili in altre sue opere. È spiegabile la nota incertezza, già in età aristarchea, non solo sulle argomentazioni ma anche talvolta sulle lezioni stesse di Zenodoto: se il suo esemplare era andato perduto, è facile pensare a riprese parziali e imprecise, a informazioni contraddittorie, a una tradizione anche orale e non completamente affidabile. Ma anche se il suo esemplare si fosse conservato a lungo, non è inverosimile che in taluni punti esso non risultasse del tutto chiaro, che certe annotazioni ponessero difficoltà di interpretazione, che talvolta le informazioni fossero discordanti.19 La forma esteriore, materiale e libraria, dell’ekdosis omerica di Aristofane di Bisanzio e di Aristarco doveva essere molto simile a quella di Zenodoto: una copia opportunamente scelta fra quelle a disposizione, sulla quale lavorare e annotare i propri interventi sul testo. La scelta del testo-base da parte di Zenodoto (oltre a molti dei suoi interventi) apparve assai discutibile e criticabile e per questo Aristofane e Aristarco scelsero esemplari con caratteristiche nettamente diverse:20 così, grazie anche al testo delle copie di lavoro usate dai grammatici posteriori a Zenodoto, si diffuse una linea tradizionale che fu decisiva soprattutto per il numerus versuum, mentre le lezioni dei singoli grammatici in

 che questo apparato di annotazioni marginali fosse una sorta di stringato commentario fatto in quantità preponderante di passi paralleli e commenti: ritengo che esso contenesse come elemento prevalente un vero corredo di interventi sul testo, con i quali Zenodoto intendeva (per collazione o per congettura) correggere il testo-base che aveva scelto per ‘trasformarlo’ in quello che egli riteneva il testo omerico autentico. 19 Su questi discussi problemi cfr. Pfeiffer 1968, 108 sgg. (= 1973, 185 sgg.); Nickau 1972, 28–31 e Nickau 1977, cap. I (ritiene che Aristarco non avesse già più l’originale di Zenodoto ed è scettico anche sull’esistenza di copie di esso); Montanari 1988, 83 sgg. e 98–105 (in particolare sull’opera di Ptolemaios Epithetes); Rengakos 1993, 12 sgg.; van Thiel 1997, 20; West 2001a, cap. I.2–3. 20 Da ultimo West 2001a, 36: «Clearly Aristophanes and Aristarchus were not dependent on Zenodotus’ text but followed another source or sources more similar to the vulgate»; cfr. West 2002, 138.

Ekdosis alessandrina: Il libro e il testo  

genere non si imposero nella vulgata. I plus-versi presenti nel testo zenodoteo non erano sue interpolazioni, ma tipici di esemplari correnti al suo tempo:21 sparirono perché con Aristarco si impose un testo con un numero di versi molto simile a quello della nostra vulgata. È importante che già Aristofane non abbia più praticato quell’intervento radicale che era l’eliminazione di versi, il che significa che non fece più cancellazioni materiali nei modi descritti. L’atetesi con l’uso dell’obelos (l’invenzione zenodotea) divenne lo strumento principale o addirittura unico per esprimere un dubbio prudente sul testo: l’οὐ γράφειv scomparve, lasciando solo lo ἀθετεῖν.22 Anche Aristarco si comportò alla stessa maniera. Per questo molti versi che Zenodoto voleva eliminare, ma erano presenti nelle copie scelte dai grammatici posteriori, si conservarono nel numerus versuum che grosso modo si affermò nella tradizione dopo l’età aristarchea e rimase nella nostra vulgata.23 L’abbandono della drastica cancellazione materiale fa capire anche come sia accaduto che molti versi ‘cancellati’ da Zenodoto siano stati ‘ripristinati’ (cioè non più cancellati)24 e così non furono obliterati. Il numerus versuum risultò grazie a copie che non avevano quelli che oggi a noi appaiono come plus-versi e non cancellavano versi eliminati da Zenodoto. Con Aristarco cominciò l’epoca in cui la produzione filologica comprendeva la stesura di ampi hypomnemata: la comunicazione e la conservazione degli argomenti e motivazioni dei grammatici furono molto più ricche e agevoli, per cui la tradizione ci ha preservato un materiale più abbondante. L’ekdosis di Aristarco doveva avere anch’essa la forma di una copia annotata, la pratica doveva essersi consolidata: tuttavia la possibilità di discutere con un certo agio e spazio svariati argomenti di critica del testo e di esegesi nello hypomnema costituiva una risorsa importante, che cambiò sensibilmente le cose. Prima degli hypomnemata separati c’era più bisogno di scrivere sull’esemplare di lavoro:25 con Aristarco invece, anche se le varianti e le lezioni da adottare potevano essere comunque annotate nei margini e negli interlinei, divennero essenziali i segni critici posti accanto ai versi (e il sistema, già arricchito dopo Zenodoto, divenne

 21 West 2001a, 40. 22 Se qualche effettiva cancellazione fu ancora fatta, essa fu di portata così irrilevante che se ne perse notizia. 23 Haslam 1997, 85; West 1998–2000, VII; Führer/Schmidt 2001, 7. 24 Spesso si dice invece che furono “recuperati” o “reintrodotti”: questo comporterebbe operazioni molto più complicate. 25 Su questo dunque è verisimile che Zenodoto scrivesse relativamente di più rispetto a Aristarco.

  Ekdosis alessandrina: Il libro e il testo molto articolato),26 mentre la più distesa trattazione filologico-esegetica trovava posto nel commentario. Il numero delle ekdoseis di Aristarco e il loro rapporto filologico e cronologico con il commentario o i commentari suscitano ancora qualche difficoltà. Non ripeto qui né la storia della questione né la trattazione già esposte altrove: le presuppongo e mi limito a riassumerne i risultati, per fissarli nel quadro che si sta delineando.27 Da una parte abbiamo i frequenti, inequivocabili riferimenti al plurale che si trovano nelle fonti per il lavoro editoriale di Aristarco su Omero: αἱ Ἀριστάρχου (scil. ἐκδόσεις ο διορθώσεις), ἡ ἑτέρα τῶν Ἀριστάρχου e simili. Dall’altra, i titoli di due opere del grammatico Ammonio, successore diretto di Aristarco: Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως e Περὶ τῆς ἐπεκδοθείσης διορθώσεως, il primo in apparente contraddizione con il secondo e con le citazioni scoliastiche che indicano due edizioni. Le soluzioni proposte arrivano comunque a ipotizzare una ἐπεκδοθεῖσα διόρθωσις fatta dagli immediati scolari di Aristarco, probabilmente dal suo successore Ammonio, che comunque la conosceva. Per quanto riguarda gli hypomnemata, credo che sia difficile negare che Aristarco ne abbia fatte due versioni successive: un primo commentario basato sull’ekdosis di Aristofane di Bisanzio è esplicitamente citato nello sch. B133 a: ἐν τοῖς κατ’ Ἀριστοφάνην ὑπομνήμασιν Ἀριστάρχου. A questo si contrappone la citazione di hypomnemata perfezionati (ἠκριβωμένα) nello sch. B 111 b: è del tutto verisimile che Aristarco abbia prodotto una seconda versione degli hypomnemata nella quale tenere conto dei progressi fatti dal suo lavoro di filologo omerico col passare del tempo.28 Abbiamo due passi omerici a proposito dei quali gli scoli ai vv. K 397–399 e T 365–368 ci informano di un ripensamento e cambiamento di opinione da parte di Aristarco rispetto alla sua prima scelta testuale. È una situazione che trova abbondanti paralleli nella documentazione scoliografica, in cui si parla appunto di ripensamenti e seconde scelte aristarchee:29 tuttavia in quei due casi possiamo dire, malgrado incertezze di dettaglio, che i filologi posteriori cercavano notizie sui termini di tale ripensamento e sulla situazione testuale determinata 26 Potevano anche essere ripetuti nell’hypomnema accanto ai lemmi, come accade per esempio in P.Oxy. 1086 (pap. II Erbse). 27 Montanari 1998a (= questo volume, cap. 23) e 2000a (= questo volume, cap. 38); cfr. West 2001a, 61–7. 28 Approfitto dell’occasione per rimandare a Montanari 1997, 285–6 (= questo volume, cap. 42), a proposito del celebre Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν. 29 Due altri esempi sono esaminati in Montanari 2000a (= questo volume, cap. 5).

Ekdosis alessandrina: Il libro e il testo  

si, che risultavano loro non chiari e incerti, consultando opere di Ammonio, allievo e diretto successore di Aristarco “nella scuola”. Per spiegare questa situazione e conciliare l’apparente contraddizione fra i due titoli di Ammonio, si è immaginato che in effetti Aristarco avesse fatto personalmente una sola ekdosis, come presupposto dal primo, e che la ἐπεκδοθεῖσα διόρθωσις menzionata nel secondo fosse stata fatta in seguito, dopo la morte del maestro, nella cerchia dei suoi primi e più anziani allievi, magari proprio da Ammonio o da altri. Credo che un’ipotesi un poco diversa possa farci capire meglio cosa sia accaduto. Il fatto che non ci siano state più edizioni della διόρθωσις aristarchea e che la διόρθωσις medesima sia stata riedita (ἐπεκδοθεῖσα), mi pare si possa concretamente conciliare per davvero soltanto supponendo che la ἐκδοθεῖσα διόρθωσις e la ἐπεκδοθεῖσα διόρθωσις fossero in un certo senso (fisicamente) la stessa cosa, però modificata, cioè riveduta, corretta e ulteriormente “lavorata”. Ripercorriamo in sintesi l’insieme della storia. Aristarco cominciò scrivendo hypomnemata basati sul testo di Aristofane; poi si dedicò alla sua διόρθωσις e produsse la propria ekdosis, vale a dire scelse un esemplare come testo-base e vi annotò i suoi interventi testuali (segni critici e lezioni) a rappresentare il suo testo omerico; a questo punto scrisse nuovi hypomnemata, basati sulla propria ekdosis e definiti come più approfonditi e accurati. Ma la ricerca e la riflessione di un filologo come Aristarco su un testo come Omero difficilmente trovano una fine assoluta, una conclusione definitiva: oltre che a riflettere e studiare, Aristarco continuò a insegnare, Omero era sempre tra le sue mani. Così egli riprese la sua διόρθωσις, riprese in mano la sua ekdosis e su quella stessa copia si mise ad annotare ripensamenti e nuove scelte sul testo: la stessa copia, che era stata la ἐκδοθεῖσα διόρθωσις, divenne poi la ἐπεκδοθεῖσα διόρθωσις. Poiché non fece ancora nuovi hypomnemata, non potè redigere per esteso i motivi di questi suoi ripensamenti e definire il punto a cui essi si arrestarono, per cui in taluni punti rimasero delle incertezze sullo stadio finale del suo pensiero.30 Anche su questo punto, quanto detto sulla forma dell’ekdosis come copia di lavoro e sulla prassi libraria della diorthosis offre una buona chiave per capire. Un solo esemplare di Omero recante successivi strati di interventi spiega perfettamente il fatto che nella tradizione si sia parlato comunemente di αἱ Ἀριστάρχου, ἡ ἑτέρα τῶν Ἀριστάρχου e simili: in un certo senso, si poteva parlare di due edizioni, cioè di due assetti diversi del testo; e nella maggior parte dei casi la prima e la seconda versione dovevano essere leggibili e distinguibili, così da permettere alla tradizione filologica successiva di rendere conto di entrambe.  30 Questa soluzione del problema ekdosis / ekdoseis di Aristarco è approvata da Slings 1999. Non mi è chiaro se l’abbia percepita West 2001a, 61–67.

  Ekdosis alessandrina: Il libro e il testo Nuovi interventi su una copia già annotata spiegano anche situazioni di incertezza, in cui per varie ragioni (magari materiali) poteva non risultare chiaro quali fossero i due assetti testuali indicati dal maestro, quale il suo ultimo pensiero e se egli avesse davvero raggiunto una decisione. Ammonio aveva tutti i motivi per spiegare come la ekdosis fosse in realtà una sola, che a un certo punto però era stata “riedita” con cambiamenti: è probabile che i titoli Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως e Περὶ τῆς ἐπεκδοθείσης διορθώσεως si riferiscano a due parti di una stessa opera oppure a due operette collegate, il cui scopo doveva essere spiegare come erano andate le cose e soprattutto dare chiarimenti sui casi dubbi, passi omerici in cui la decisione finale del maestro non era chiara perché non era stata annotata bene o il punto si era danneggiato o anche perché la sua riflessione era stata interrotta dalle circostanze e non si avevano resoconti evidenti. Conferma questa idea il fatto che non tutto doveva essere ben chiaro a Didimo, che consulta Ammonio in casi dubbi.31 Non abbiamo affrontato ancora il problema relativo alla vera natura delle lezioni attribuite dalla tradizione erudita ai grammatici alessandrini: congetture ope ingenii e basate solo su criteri soggettivi, varianti di fonte documentaria derivate da collazione di copie e oggetto di scelta, oppure una commistione dei due procedimenti? Si tratta del punto nodale a proposito del lavoro dei filologi alessandrini, rispetto al loro ruolo nella storia intellettuale e culturale della nostra civiltà: su questo punto ci sono state recentemente le maggiori discussioni. Ne ho trattato altrove nei lavori citati in bibliografia e ne farò qui solo un cenno in chiusura. Credo che il contesto evocato in precedenza, anche con i suoi aspetti squisitamente materiali e tecnici, aiuti a chiarire il problema in modo meno astratto del solito. Personalmente, resto convinto che la produzione di un’ekdosis da parte dei filologi alessandrini, con il relativo lavoro di interpretazione, comportasse sia il lavoro di emendamento congetturale, sia la scelta fra varianti testuali rinvenute grazie alla collazione di copie diverse. I due aspetti sono stati differentemente enfatizzati e c’è stato chi ha voluto sottrarre pressoché totalmente all’impostazione critico-filologica e intellettuale degli alessandrini l’idea e la pratica di confrontare copie diverse e scegliere fra varianti testimoniate. Per alcuni essi erano soltanto o soprattutto congetturatoti, volti a correggere senza molti scrupoli un testo ritenuto corrotto e insostenibile sulla base solo di criteri  31 Pfeiffer 1968, 217 (= 1973, 339): «Whether Didymus was able to work on copies of these originals διορθώσεις and ὑπομνήματα of Aristarchus and of his monographs, the συγγράμματα, is an insoluble problem»; cfr. West 2001a, 61–67.

Ekdosis alessandrina: Il libro e il testo  

di giudizio ‘interni’, come supposte incoerenze, inopportunità e inappropriatezze, ripetizioni materiali, preferenza per un testo più conciso, amore di standardizzazione e di uniformità.32 Non c’è dubbio che spesso i grammatici intervenissero congetturando senza supporto tradizionale, ma ritengo impossibile sottrarre loro un certo lavoro di confronto fra copie diverse e di scelta fra varianti di fronte a una tradizione non univoca. Quanto abbiamo visto fa ritenere che questa operazione non avesse di per sé un carattere particolarmente curioso o stravagante: come pensare che intellettuali eruditi trovassero strano e inusitato quello che faceva un buon artigiano di scriptorium e non pensassero a utilizzarlo per i loro scopi, ampliarlo, approfondirlo? Questa pratica comportava per loro una vera idea di critica testuale e di storia del testo, anche se magari non applicata con totale coerenza metodologica: come abbiamo detto sopra, correggere la singola copia divenne, sotto il cielo del grammatico, ripristinare la forma corretta e autentica dell’opera in se stessa. Basta aver concepito e affrontato il problema anche in modo embrionale e tentennante, perché un progresso intellettuale decisivo sia stato acquisito per sempre: un testo letterario aveva una storia della trasmissione, nel corso della quale poteva essersi, anzi si era di certo guastato in più punti; il suo dettato corretto si poteva ripristinare per congettura oppure scegliendo la lezione giusta fra quelle offerte da una tradizione divergente; bisognava dotarsi dei metodi per ottenere questo risultato. L’idea del riconoscimento di un guasto rispetto alla forma autentica e di un intervento e un metodo per sanarlo rivela come sia stabilita e operante l’unità organica e solidale fra critica del testo e interpretazione.

 32 Discussione in Montanari 2004 (= questo volume, cap. 41); un caso interessante adduce Fantuzzi 2005. M.L. West segue una linea di svalutazione della filologia alessandrina risalente a M. van der Valk e ripresa più recentemente da H. van Thiel, ma criticata decisamente, oltre che dal sottoscritto, da M. Schmidt. R. Führer, J.-F. Nardelli, A. Rengakos e G. Nagy: riferimenti nella Bibliografia.

  Ekdosis alessandrina: Il libro e il testo

Fig. 1: Timoteo, Persiani, P. Berol. inv. 9875 coll. IV.

Fig. 2: Timoteo, Persiani, P. Berol. inv. 9875 coll. V.

Fig. 3: Posidippo, P. Mil. Vogl. 309, col. XI.

Ekdosis alessandrina: Il libro e il testo  

Fig. 4: Eschine, Contro Ctesifonte §§ 51–3, P. Oxy. 2404.

Fig. 5: Eschilo, Diktyoulkoi, P. Oxy. 2161.

  Ekdosis alessandrina: Il libro e il testo

Fig. 6: Commentario al Teeteto di Platone, P. Berol. inv. 9872.

Fig. 7: Hypotheseis di varie tragedie di Eschilo, P. Oxy. 2256.

Ekdosis alessandrina: Il libro e il testo  

Fig. 8: Vangelo di Giovanni, P. Bodmer 2.

 Correcting a Copy, Editing a Text: Alexandrian Ekdosis and Papyri In the period from Zenodotus to Aristarchus, ekdosis confirmed its place within ancient culture as a typical product of Alexandrian philologists along with hypomnema, syngramma, and the collection of lexeis and other exegetical-erudite products. How the ekdosis of a literary work was effectively carried out, what form it took and the way in which it was prepared in actual practice by the grammarian have for some time been the object of debate. As we shall see, the question also has effects on the reconstruction and assessment of the method and results of philological activity, of which ekdosis is a part. I have over the last few years analysed these questions,1 emphasising the importance of the relationship between the library artefact on one hand and the text as an object of editing, with its various paratextual elements such as annotations and semeia, on the other hand.2 We must take into account and give the right prominence to what we know regarding the creation of new copies of texts (in the scriptoria by professional scribes or privately by individuals) along with insights that can be gleaned from surviving examples. Significant problems for instance have arisen as to understanding the method of work adopted by Zenodotus, the first of the major Alexandrian philologists. According to Rudolf Pfeiffer, “It is not improbable that Zenodotus, examining manuscripts in the library, selected one text of Homer, which seemed to him to be superior to any other one, as his main guide; its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Διόρθωσις can be the term for either kind of correction. It is hard to imagine any other way”.3 K. Nickau, the author of important works on Zenodotus, states: «Dann ist zu fragen, ob Z(enodotos) nicht einen durch Recensio ermittelten Homertext zugrundelegte (der jedoch nicht seinen Vorstellungen von der genuinen Form der Epen entsprach), diesen mit Obeloi versah und zu ihm Textvorschläge sowie deren Begrundung mitteilte. Z(enodotos) selbst wie auch seine Hörer machten sich entsprechenden Notizen,

 English translation by Justin Rainey.  1 Montanari 1998a (= this volume, ch. 23); 2002a (= this volume, ch. 47); 2004 (= this volume, ch. 41); 2009b and 2009c (= this volume, ch. 60), with extensive bibliography. 2 The question is discussed by Jacob 1999, as part of a wider overview of the development of philology (see on the Alexandrians in particular p. 80). 3 Pfeiffer 1968, 110. https://doi.org/10.1515/9783110772371-008

Correcting a Copy, Editing a Text  

die, wären sie von Z(enodotos) schriftlich veröffentlicht worden, ‘Hypomnemata’ hätten heißen können. Aber die Zeit der schriftlich publizierten Homer-Kommentare begann erst mit Aristarchos. So würden sich auch die späteren Unsicherheiten in der Berichterstattung über Z(enodotos)s Ausgabe erklären».4

The fundamental position, supported by the two eminent scholars,5 excludes the possibility that the ekdosis of Zenodotus consisted in a new copy bearing the continuous text wanted by the grammarian i.e. the whole text completely rewritten by him (or for him), with his readings incorporated, with the obeloi in the margins indicating the athetesis, without the verses which in his opinion had to be omitted. As Pfeiffer says, it is hard to imagine another way of working that did not consist in carrying out corrections on an already existing copy, appropriately chosen from those available and used as the basic text on which the grammarian would over time make changes and add annotations as part of his studies and the work of diorthosis. I share this position and regard it as the one on which to base further discussions. Owing to the size of the documentation and knowledge available, these discussions will focus specifically on the Homeric text. However, we can assume a substantially similar approach also for other authors subject to philological analysis by the Alexandrian grammarians. A similar method of producing the ekdosis was utilised by Zenodotus and continued to be used by later grammarians. A philologist chose, according to his own preferences, an exemplar that he considered suitable as a basis for his work. When he rejected the text, he noted in the place in question the preferred reading in the free spaces or between the lines. His own text resulted from the original text chosen together with the changes suggested and contained in the paratext created. Next to the verses were placed the appropriate semeia: Zenodotus began just with an obelos for his proposal of athetesis, with the system subsequently becoming considerably richer and more differentiated. Doubts remain as to what extent the working copy may have contained also explanatory annotations. Such annotations, however, must have been present6 and, in this respect, a development may have taken place (as proposed by Nickau), leading eventually to the separate hypomnema. We can imagine this as a product of years of study that led to over time a series of interventions to the same copy. This copy, bearing the traces of the work of diorthosis, resulted materially in the philologist’s own ekdosis of Homer. This was his own personal copy, it bore his name for purposes of identification and contained the fruit of his work  4 Nickau 1972, 30–31. 5 See also infra, n. 38 for the position of H. van Thiel in this regard. 6 For a discussion of papyrus marginalia, see McNamee 2007.

  Correcting a Copy, Editing a Text and insights; ekdosis in that it was ekdotheisa, i.e. available for consultation by scholars, poets and intellectuals. Until the writing of commentaries became standard practice, with larger spaces available for illustrating the arguments proposed and the materials used, the working copy containing in the margins the semeia, the divergent readings along with any brief notes was the key source for knowing the philologist’s opinion of the text he had worked on. This copy could also be accompanied by other types of works, such as essays on a particular subject, lexicographic collections or even notes from scholar/school milieu, transmitted orally or by memory. This was probably the case for Zenodotus, Aristophanes and the preAristarchean grammarians in general, with the inevitable degrees of uncertainty well-known to those working in this field. With Aristarchus, the practice certainly remained of producing an ekdosis made up by the working copy and its paratextual surrounds. However, this ekdosis often came with the hypomnema, which became the ideal means with which the philologist was able to develop his arguments on a quantity and variety of philological-exegetical themes. I find it difficult to raise objections to the view that the production of a philological ekdosis was based on working methods and practices that were certainly by no means new or unusual in an intellectual environment that had been used to be a “book civilisation” for at least two centuries. There is little doubt (and I have supported this position for some time) that the philological work of the Alexandrian grammarians, starting from the first generation, represented something new in cultural history and marked significant intellectual progress. The reality of this revolution, I believe, becomes more evident and tangible if we highlight the precedents and foundations that initiated and nurtured developments as well as uses that were new in methodology and above all in scope. It is an accepted fact that examples of literary works were normally re-read and corrected thanks to additional further comparison with the antigraph, at times even on the basis of a collation with other copies. Naturally, we are interested in the most ancient evidence. However, we are to an extent conditioned by the fact that the most ancient Greek papyri that we are aware of, from the second half of the IVth century B.C., are extremely limited in number. One of these, possibly the oldest, is the well-known fragment of the Persians of Timotheus, PBerol. inv. 9875. At col. IV, l. 133, after having written βορεαιαραισον|ται, the scribe inserted a δ supra lineam in order to restore the correct reading βορέᾳ διαραίσον|ται. At col. V, l. 196, after having written πλουτουοδε, the scribe inserted an ι supra lineam to restore the correct reading πλούτου οἱ δέ. These two corrections (διορθώσεις) were made either in scribendo or following a rereading of the text. In the renowned Derveni Papyrus, dated around the end of the

Correcting a Copy, Editing a Text  

IVth century B.C., at col. XXI, l. 11 a first hand corrects ]ρμοναδε by inserting an ι supra lineam to restore the correct Α]ρμονία δέ, whilst at col. VI, l. 5 the omission of a letter in τοιδε (instead of τοῖς δέ) at the end of the line is not corrected. Such examples suggest that the corrections were not the product of a systematic analysis, but were made by the scribe, probably in scribendo, if he should notice a mistake.7 These corrections of material errors, though not classifiable as a significantly evident phenomenon, certainly represent the most ancient, tangible and visible evidence of a concern for the creation of a correct text and can be placed just before or at the same time as Zenodotus (330 ca.–260 ca.). From the IIIrd century B.C. comes the Milan papyrus with epigrams by Posidippus, PUniv.Milan. 309, which represents important evidence not only because of the ancient epoch to which it belongs, but also in view of the quantity of corrections and annotations the text presents. The majority of the corrections were made by the same scribe, but subsequently two other hands intervened with further emendations and the differences in approach should be recognised. “Il medesimo scriba è responsabile della maggioranza delle correzioni ... Gli interventi sono tutti assai limitati (in genere coinvolgono una sola lettera e mai più di tre) e sono tutti diretti ad emendare banali errori di stesura, cioè fraintendimenti, sostituzioni accidentali ed omissioni. La maggior parte di essi è stata manifestamente eseguita in scribendo; e tutti quanti sono stati compiuti con studiata accuratezza ... infatti, se gli errori emendati fossero apparsi evidenti, avrebbero dato immediatamente l’impressione che il testo fosse scritto senza la necessaria precisione ed avrebbero svilito il lavoro di chi lo aveva scritto. Dopo quella del copista, altre due mani hanno inserito emendamenti nel rotolo, operando entrambe in pochi punti e senza avere la precauzione di occultare i loro interventi, come invece cercava di fare il copista ... Si potrebbe pensare che questa [scil. la seconda mano, m. 2] sia la mano di un revisore del centro di copia in cui fu realizzato il rotolo. L’ipotesi non è del tutto esclusa; ma non è nemmeno sostenuta da indicazioni certe. Anzi, il fatto che le correzioni di m. 2 siano tutte concentrate in due sole colonne consecutive, induce a ritenere che chi le apportò fosse un lettore particolarmente interessato a quella sezione dello scritto, più che un revisore obbligato a controllare tutto il rotolo. Fu sicuramente un lettore la terza persona che intervenne sul testo [scil. m. 3] ... i suoi interventi sono concentrati tutti nella col. XI. Lì egli segnalò una variante di lettura per la l. 30, annotandola nel margine superiore”.8 Let us take one example. At  7 Turner/Parsons 1987, 92; text in Kouremenos/Parassoglou/Tsantsanoglou 2006. 8 Bastianini/Gallazzi 2001, 15: “The same scribe is responsible for most of the corrections … Interventions are extremely limited (in general amounting to one and never more than three

  Correcting a Copy, Editing a Text col. XI, l. 30 we can read κεντρακαιεξω[; in the upper margin, one of the two hands working on the text after the original scribe has written καικεντρα (the last three letters are not visible in the photograph but can be seen in the original document).9 It is extremely likely that this is a correction or a variant, probably for the κέντρα καί of the text, an inversion — καὶ κέντρα — is proposed, but it is not clear owing also to the fact that the rest of the verse has not been preserved.10 The papyrus findings of the IInd and IIIrd centuries of our era are sizeable and the evidence of the period provides us with valuable and abundant documentation. The following significant examples will suffice for our purposes, although these could be easily added to. POxy. 2161, of the IInd century A.D., contains Aeschylus’ Diktyoulkoi. The scribe has occasionally corrected some of his own errors. For instance, at l. 831 he wrote ηδη, but then crossed this out with an oblique line through each letter, writing supra lineam the correct reading ο]ιον. PBerol. inv. 9872 (BKT II), of the IInd century A.D., is a long papyrus roll (75 columns plus various fragments) that contains a commentary on Plato’s Theaetetus with a substantial number of corrections. The most recent editors of the roll, Bastianini and Sedley, write: “Il volumen è stato sottoposto a revisione e corretto in più punti: parole o lettere omesse nella stesura originaria sono state reintrodotte, lettere o parole superflue sono state cancellate, lettere ritenute errate sono state sostituite con quelle giudicate esatte. Tutti questi interventi non sembrano presupporre necessariamente una collazione con un esemplare diverso da quello di copia (l’affermazione contraria di McNamee 1981, 90 non appare sufficientemente fondata) ... La varietà del modo con cui le emendazioni  letters) and are all aimed at correcting minor slips in the drafting stage i.e. misunderstandings, accidental substitutions and omissions. Most of these are clearly made in scribendo; and all of these were carried out with considerable accuracy … in fact, if the errors corrected had appeared visible, they would have given the impression that the text had been written without due precision so devaluing the importance of the work of its writer. After the copyist’s corrections, two further hands inserted revisions in the roll, both operating in only a few places and without any attempt to hide their corrections, unlike the copyist … It is possible that this [scil. the second hand, m. 2] is the hand of a proof reader of the workshop where the roll was created. This hypothesis is not completely to be rejected; however, it lacks reliable evidence. Indeed, the fact that the corrections of m. 2 are concentrated in only two consecutive columns leads one to believe that who made these corrections was a reader with particular interest in that section of the script rather than a proof reader having to check the entire roll. A reader was certainly the third person to make changes to the text [scil. m. 3] … his amendments are concentrated all in col. XI. There he recorded a variant on the reading of l. 30, noting it in the upper margin”. 9 Bastianini/Gallazzi 2001, 76–77. 10 Bastianini/Gallazzi 2001, ad loc.

Correcting a Copy, Editing a Text  

sono state attuate può indurre il sospetto che il rotolo sia stato corretto a più riprese: una prima mano (quella di un diorthotès dello scriptorium) ha aggiunto le parole saltate, che sono riportate nel margine superiore ... o inferiore ... oppure sono poste a proseguire il rigo direttamente nell’intercolunnio ... Una mano successiva, o forse più mani, sembrano poi avere ripercorso tutto il testo, cancellando da capo con un tratto d’inchiostro tutte le lettere ritenute errate”.11 For example, at col. LXIII, l. 6 the scribe had written προσαλλαουτεσχη, omitting some words. In the intercolumnium to the left, the corrector has put the sign of an upwards-pointing ancora and in the space between αλλα and ουτε has written ἄνω; in the upper margin, one can read the words θεωρειται ουτε γαρ χρω|μα κα(τω), which were probably preceded by an ancora now lost in lacuna. The corrected text is therefore πρὸς ἄλλα θεωρεῖται οὔτε γὰρ χρῶμα οὔτε σχῆ|μα. POxy. 2256, of the IInd–IIIrd centuries A.D., contains hypotheseis of various tragedies by Aeschylus. The fragmentary hypothesis of fr. 3 recalls the victory, with the trilogy of which the Danaids was a part, against Sophocles and another author, probably Mesatos (l. 5). After the name of the latter and at the beginning of the following l. 6, round brackets can be clearly seen, which are generally used as a sign to indicate expunction in literary texts and non-literary documents. It is clear here that the round brackets were placed in scribendo, which can be explained solely by imagining that the scribe copied from an exemplar where the expunctions were already present to indicate that the plays placed between brackets had been mistakenly placed after the name of Mesatos.12 The copy of the Gospel according to St. John contained in PBodmer 2 dates to the IIIrd century A.D. The scribe has corrected the text in a variety of ways. There are supra lineam additions (ll. 2 and 12) and words rewritten above parts of the text cancelled with a sponge: at ll. 9–10 εταραχθη has been written over a word that has been scrubbed out and which continued in the following line,

 11 Bastianini/Sedley 1995, 243–244: “The volumen has been proof read and corrected in many places: letters or words omitted in the original drafting stage have been restored, superfluous letters or words have been cancelled, letters judged to be mistakes have been replaced by those considered correct. All these changes do not appear necessarily to presuppose a collation with an exemplar different from that of the copy (the contrary view held by McNamee 1981, 90 does not seem sufficiently well-grounded) … The variety of ways the corrections have been made may lead one to suspect that the roll had been corrected on various occasions: the first hand (a diorthotès in the scriptorium) added the missing words, which are marked in the upper margin … or lower … or are placed after the line directly in the intercolumnium … A later hand, or perhaps hands, appears to have gone through the whole text, cancelling with a line in ink all the letters judged to be wrong”. 12 Arata/Bastianini/Montanari 2004, 39, 47–48.

  Correcting a Copy, Editing a Text where the letters σατο can be made out in the remaining space; the second part of l. 10 has been rewritten; at the beginning of l. 11 τον is the remains of an eliminated reading, subsequently punctuated with dots as well as small round brackets supra lineam. I turn now to a manuscript that, I believe, provides us with what can be termed an anthology of the techniques and methods available for correcting and improving a text: POxy. 2404 + PLaur. inv. III/278, a fragment of a papyrus roll (late IInd century–early IIIrd century A.D.) containing a part of §§ 51–53 (POxy. 2404) and of §§ 162–163 (PLaur. III/278) of Aeschines’ oration Against Ctesiphon.13 We can see that the work of proof reading was not limited solely to correcting minor errors as discretely as possible in order to reduce the possibly negative impact of emendations on the appearance of the text (see above the case of the papyrus of Posidippus); in fact, more evident corrections, albeit written with care and precision, have been made, with the apparent aim of improving the text and enabling it to be read according to the intention of the corrector or correctors. As regards punctuation, the scribe provided the text only with paragraphoi, whilst copious punctuation was added (at least it is thought) by a later hand.14 Most of these are dots, placed slightly higher than the letters, which had already been written, making sure that they were not above a letter but in the narrow space between the end of the preceding word and the beginning of the next. A lower dot can also be seen at col. I, l. 17. The system can be described as follows: the upper dot combined with the paragraphos marks the end of a sentence; the upper dot on its own distinguishes the cola of the sentence; the lower dot indicates a weaker pause.15 If we examine the second column (preserved almost in its entirety),16 which includes the portion of the text (relatively small as the columns are narrow and not high) which goes from § 52.7 τριάκοντα to § 53.7 ὥστε (Dilts 1997 edition, 212– 213), we find six upper dots that correctly mark out all the cola of the passage (ll. II 5, 8, 12, 15, 16, 22 = ll. 33, 36, 40, 43, 44, 50 in the numbering of the editio princeps); two of these dots  13 Editio princeps of POxy. 2404: Turner 1957; see also Turner 19802, Pl. VIII and p. 212; editio princeps of PLaur. inv. III/278: Messeri Savorelli/Pintaudi 1997, 172–174; see also Neri 2003, 511–514; Esposito 2004, 3–4; Colomo 2008, passim. 14 See Turner 1957, 130; 19802, 212; Colomo 2008, 15–16. On punctuation marks in papyri, see Turner 19802, 92–93; Turner/Parsons 1987, 9–10. 15 Colomo 2008, 15–16. 16 Only one line is missing at the beginning while the other lines are complete, thereby allowing for a well-founded evaluation. The upper and lower margins remain in the first column so guaranteeing 28 lines per column. The column is however mutilated both on the right and on the left (approximately half the line remains).

Correcting a Copy, Editing a Text  

are also combined with the paragraphos (ll. 8 = 36 and 22 = 50) and mark precisely the end of the two sentences that occur in these lines (i.e. they correspond with the two full stops of the modern edition). In the first column, where the beginnings of the lines are lost and the absence of the left margin does not allow us to know whether there were paragraphoi, one can note two upper dots at ll. 14 and 22 in addition to the already mentioned lower dot at l. 17. Proof therefore of a serious attempt to highlight the syntactic and rhetorical structure of the text that leads us to consider the role of punctuation in Alexandrian philological exegesis (rather than the complex and idiosyncratic system created by Nicanor, one can mention the simpler and more widely-used system of the three stigmai of Dionysius Thrax).17 Starting from the editio princeps of POxy. 2404, all corrections have been attributed to a single second hand, although D. Colomo in a recent work states that three subsequent hands can be identified following that of the scribe’s.18 As I feel insufficiently competent on this point, believing that further autopsy on the original is required, I will not go into the question, focusing my attention on the changes made in particular in col. II of POxy. 2404. At l. 6 (= 34 of the continuous numbering) the first hand wrote ἐν τοῖς διονυσίοις and the proof reader makes the appropriate correction ἐν διονύσου by putting four dots above the letters of the τοις to be eliminated and writing simply ου above ιοις, without cancelling out these letters, but paying attention to place ου exactly above ιο, so as to avoid possible misinterpretations (e.g., σιου), and to clearly mark the correct reading σου, obviously counting on the fact that the residual ις did not create problems. At l. 21 (= 49) λείαν has been corrected to λίαν by cancelling ε with two oblique lines through the letter; at l. 26 (= 54), the first hand wrote ὅπως, which the corrector revises with οὕτως by deleting the incorrect π with an oblique line through and writing υτ exactly above π.19 More interesting and considerably more intriguing is the situation at ll. II 16– 20 (44–48 in the numbering of the editio princeps). This is the text of the passage in the codices and in modern editions (Blass 1908,20 Adams 1919, Leone 1977, Dilts 1997):

 17 See Colomo 2008, 15–22; Montana 2009a; D.T., in GG I/I 7.3–8.2. 18 Colomo 2008, 24–27; see also Neri 2003, 511–514. 19 According to Neri 2003, 512, the two deletions at ll. 21 and 26 are assignable to the first hand i.e. the scribe, whilst the corrector later added the correct letters at l. 26; furthermore, at col. I, l. 28 Neri suggests that the scribe tried to correct an erroneous ε, but with imprecise results, which led the corrector to add the correct ο supra lineam. See Colomo 2008, 26. 20 Unchanged in Blass/Schindel 1978.

  Correcting a Copy, Editing a Text ἀλλ᾿ ἐκεῖνο φοβούμενος, μή μοι παρ᾿ ὑμῶν ἀπαντήσῃ τὸ δοκεῖν ἀληθῆ μὲν λέγειν, ἀρχαῖα δὲ καὶ λίαν ὁμολογούμενα.

First, a small point of interest. The medieval codices agreed on δοκεῖν μὲν ἀληθῆ λέγειν, whilst C.G. Cobet had suggested the transposition δοκεῖν ἀληθῆ μὲν λέγειν, rightly accepted by Blass 190821 and later editions.22 Our papyrus confirms this transposition: the correct reading ἀληθῆ μέν is in the primary text and is left unchanged by the corrector. Apart from this, the textual situation of the passage is complex in that neither the primary text nor the text the corrector produced from it correspond to medieval tradition. The primary text appears to have been: ἀλλ᾿ ἐκεῖνο φοβούμενος, μή μοι παρ᾿ ὑμῶν ἀπαντήσῃ τι τοιοῦτον καὶ τὸ δοκεῖν ἀληθῆ μὲν λέγειν κτλ.

Turner observes that “This first reading appears to mean ‘lest I be greeted on your part by some such thing as the impression of telling the truth but...’, as if τι τοιοῦτον οἷον or ὡς καί stood in the Greek”.23 Later τις has been added supra lineam, positioned carefully between μή and μοι (l. 16 = 44); τι τοιοῦτον (l. 18 = 46) has been eliminated with two horizontal lines (one drawn through the two words, now faded, the other clearly visible supra lineam) and next to that on the right (in the intercolumnium) in its place has been written θορ.], which is generally integrated as θόρυ]βος. In the next line (l. 19 = 47) καί has been left, whilst τὸ δοκεῖν has been eliminated with a horizontal line through the letters (double lines through οδο, probably a first, shorter line which was later gone over by a longer line through the whole segment) and δόξω added in smaller writing not in the margin (as was the case for θόρυ]βος) but in an empty space in the final part of the line in which the four letters of smaller size fit into the space with only a minimal part extending outside the space on the right (see below). The resulting text is as follows: ἀλλ᾿ ἐκεῖνο φοβούμενος, μή τις μοι παρ᾿ ὑμῶν ἀπαντήσῃ ⟦τι τοιοῦτον⟧ θόρυβος καὶ ⟦τὸ δοκεῖν⟧ δόξω ἀληθῆ μὲν λέγειν κτλ.

 21 Confirmed in Blass/Schindel 1978, XXIX. 22 μὲν ἀληθῆ remains solely in the Martin/de Budé 1928 edition. 23 See Turner 1957, 132, ad loc.; according to Colomo 2008, 24, here “È stato operato, inoltre, un intervento di correzione attraverso la collazione di un altro esemplare diverso dall’antigrafo, ad opera di due mani diverse, la terza e la quarta” (“a correction was made by the collation of another exemplar different from the antigraph carried out by two different hands, i.e. a third and fourth hand”).

Correcting a Copy, Editing a Text  

The most straightforward scenario is to imagine that the first was the text written by the copyist24 and the second, resulting from the sum of corrections, was the one wanted by the corrector or correctors. A doubt remains regarding the correction at l. 19 (= 47): why did the copyist leave an empty space after τὸ δοκεῖν at the end of l. 19 = 47, a space which was used for adding δόξω? There are certainly some irregularities in right side justification, but the space left empty here is perhaps too big (corresponding to around three letters of the normal size). Neri suggests that the scribe himself may have left the space, making the correction later (if however this occurred in scribendo, why should the scribe write in a smaller and more compact size?) or he may have deliberately left the space for the corrector, who would fill the space afterwards.25 Colomo, on the other hand, thinks that δόξω is attributable to a fourth hand, different from the one who wrote τις between the lines and θόρυ]βος in the margin.26 Nevertheless, none of the two textual arrangements correspond to the one passed down by the codices. On publishing the papyrus, Turner wrote, with regards to this passage: “at a place where the text had not even been suspected, there are two alternative readings, both superior to that of the manuscript tradition, no doubt both deriving from different ancient editions”.27 Merkelbach rejects this position, defending rather the correctness of the reading contained in the codices, which he regards as superior both to those contained on papyrus. He also attempts to explain that these papyrus readings are the fruit of conjecture.28 This opinion was clearly shared by the editors of Aeschines as no edition after the publication of the papyrus questions the readings of the codices (see Leone 1977, Dilts 1997 ad loc.).29 However, the intrinsic value of the reading(s) of POxy. 2404, its position and the importance of the papyri in general in the manuscript tradition of Aeschines is not within the scope of this paper.30

 24 With an error/omission after τοιοῦτον, as suggested by Turner 1957, 132. However, see Merkelbach 1959. 25 Neri 2003, 512. 26 Colomo 2008, 25. 27 Turner 1957, 130. 28 Merkelbach 1959; Turner 19802, 212 appears less convinced about his earlier opinion; Colomo 2008, 30. 29 In Blass/Schindel 1978, XXIX, as regards this passage in POxy. 2404, oddly only the reading ἀληθῆ μέν is mentioned, whilst no mention is made of the rest. 30 On this question, in addition to the introduction of Dilts 1997, see Monaco 2000 (the papyrus in discussion is examined on pp. 38 and 52, but only as regards the reading ἀληθῆ μέν); Colomo 2008, 30 ff.

  Correcting a Copy, Editing a Text The methods of cancellation used in the papyrus are therefore: the use of dots above a letter, an oblique (single or double) line through a letter in question and, for longer sequences, a line above or through the letters to be deleted, or by a combination of these methods.31 We have also seen the widespread practice of simply writing the correct letters above those judged incorrect as way of indicating a deletion. Another form of correction is the addition of words between the lines or in the margin. Significant in this discussion is that the exemplar of the oration of Aeschines has been the object of detailed and systematic correction which seeks to: identify textual structure by distinguishing cola and periods; correct copying errors and make them evident for the benefit of the reader; emend the text in those places judged unsatisfactory. The view offered by Turner in the editio princeps that the work has been collated with a second exemplar is plausible and one that I accept.32 I wish to stress at this stage why I have drawn attention to these manuscripts and their characteristics. The point here is not that they may provide us with insights as to their use and erudite destination33 or even be considered as an exemplar of a grammarian’s ekdosis. Rather, I view their value in terms of their highlighting the importance of the techniques adopted in the workshop for book production and the effect such craftsmanship had on the development of a philological practice that sought to improve and emend texts regarded as unsatisfactory due to the errors they contained. The papyri in fact provide ample evidence of the different methods used to correct the copy of a text, cancel what was regarded as erroneous and replace it with what was judged to be correct, by writing it above the line, in the margins and in the intercolumnia (at times with specific signs), or also above the preceding words. To delete, a horizontal or oblique line could be drawn through the letters or words to be cancelled, or these letters or words could be marked by dots or lines above or below or enclosed within round brackets or even erased with a sponge.34 As Turner and Parsons write: “One of the questions the palaeographer should ask about any literary manuscript is whether it has been adequately compared against its antigraph (the exemplar from which it was copied), a task  31 Neri 2003, 512; Colomo 2008, 24–25. 32 Turner 1957, 130: “The second hand not only revised the text for errors but collated its readings with an exemplar different from that from which it was copied”; see also Colomo 2008, 24. In spite of the arguments contained in Merkelbach 1959, I find it hard to accept that all is the result of conjecture. On related problems, see above all Turner 19802, 92–93. 33 This aspect has been extensively shown also for instance in Neri 2003 and Colomo 2008. 34 Turner/Parsons 1987, 15–16, with reference to examples in plates; see also Turner 19802, 93 and Pl. VIII; Bastianini 2001.

Correcting a Copy, Editing a Text  

which, in a publishing house, was the duty of the diorthotès, corrector, or whether it has been collated with a second exemplar (a procedure often carried out by private individuals to secure a reliable text)” … “But several of our surviving papyrus manuscripts, and especially those which are beautifully written, contain such serious un-noted errors that it is clear their ‘proof-reading’ was of a summary, superficial kind, if done at all ... Those ancient themselves who set store by having a dependable copy (persons like Strabo and Galen) were aware of this weakness and adopted a routine to counter it: they themselves (or their secretaries) checked the copy to be used against another exemplar. If, therefore, the text had been checked against its first exemplar, and was later collated with a second, it may well bear the marks of this double checking”.35 Best practice in the publishing house (scriptorium) consisted in a comparison between copies and corrections, carried out by a professional or occasional diorthotès, who had adequate resources for deleting, adding, replacing and marking various aspects and features of the text in order to improve it and increase its reliability. Analogies with philological practice are evident and need to be stressed: the methods and techniques adopted in the publishing house offer the skills that were applied and developed by grammarians. A procedure that probably did not appear particularly strange or extravagant, but which implied an extraordinarily innovative principle: in the case of Zenodotus, remembered as the first diorthotès of Homer, the diorthosis of the corrector of the scriptorium became the diorthosis of the philologist; concerns and emendments of a specifically publishing and commercial nature became those of a critical and philological-grammatical nature.36 We can imagine this transformation in the following terms. The aim of the corrector in a scriptorium was to produce a corrected copy for sale to a client in terms of handicraft in the best possible way. The grammarian’s aim on the other hand was to identify the right form of the text he was working on: he worked on a copy in order to produce a model exemplar, in which the form of the literary work he judged exact could be found, including an as far as possible codified indication of doubts and textual aporia. Such an approach did not belong certainly to the intellectual and working world of the scriptorium and its craftsmen. In this way, a drastic deletion (typical of the scriptorium corrector) was flanked at first by a sign marking a philological doubt, the obelos, which represented a fundamental intellectual change because now the work itself rather than the single copy was taken into account. The next steps were the increase in and  35 Turner/Parsons 1987, 15–16; Turner 1980, 93. 36 Nickau 1977, 10–11.

  Correcting a Copy, Editing a Text development of signs indicating textual criticism, study and analysis, essentially not for use in the correction of a copy as a craft product, but as a means of recovering the intention of the author and therefore interpreting correctly the text in question. The diorthosis of the grammarian was not concerned with the single exemplar, but the form of the work itself. Some tried and trusted instruments of the scriptorium were certainly useful and were used, but the aim and intellectual approach adopted by those who now used them were very different. This reconstruction, also based on purely material and technical aspects, helps to clarify, on a more solid basis than usual, also the problem of the real nature of the readings attributed by the erudite tradition to the Alexandrian grammarians: conjectures ope ingenii based solely on subjective criteria; choice among variants attested by documentary sources and deriving from the collation of copies; a combination of both? These questions are crucial in an evaluation of the work of the Alexandrian philologists and their role in the intellectual and cultural history of our civilisation and have been the object of the majority of recent discussions. I have addressed these questions several times37 and here will make only brief reference to them in conclusion. Personally, I am convinced that the production of an ekdosis by Alexandrian philologists, with the work of interpretation this implied, consisted in not only making conjectural emendations, but also in having to choose from the textual variants that came from the collation of different copies. These two aspects have received differing emphasis, with some suggesting that the idea and practice of comparing different copies and choosing from variants generated by collated texts was alien to the Alexandrian critical-philological mindset. Some sustain that the Alexandrians solely or mostly conjectured with the aim of correcting without too many scruples a text judged to be corrupted and unacceptable on the basis of a raft of subjective criteria, such as supposed inconsistency, inappropriateness, material repetitions, preference for greater textual concision, standardisation and uniformity.38 Undoubtedly, the grammarians often conjectured without the support of textual tradition. However, I believe it is impossible not to assign to them the work of comparing different copies and choosing from variants when the textual

 37 Montanari 1997 (= this volume, ch. 42); 1998a (= this volume, ch. 23); 2000a (= this volume, ch. 38); 2002a (= this volume, ch. 47); 2004 (= this volume, ch. 41). 38 Discussion in Montanari 2004 (= this volume, ch. 41). West 2001a; 2001b follows a trend that minimizes the importance of Alexandrian philology, going back to Valk 1963–1964 and returned to recently by Thiel 1992; 1997, criticised not only by myself but also by Schmidt 1997; Führer/Schmidt 2001; Nardelli 2001; Rengakos 2002a; 2002b and Nagy 2000; 2003; 2010.

Correcting a Copy, Editing a Text  

tradition was not univocal. What we have seen leads us to believe that the work of comparison and selection was not particularly unusual or extravagant. Indeed, how is it possible to imagine that erudite intellectuals considered strange or unorthodox the techniques of the craftsman in the scriptorium, which of course they knew very well? And, given that they were familiar to these techniques, would not they want to use, improve and personalise them, for their own needs? This practice implied for the grammarians a real idea of textual criticism as well as of history of the text, even though it lacked methodological rigour. As stated above, correcting a single copy becomes for the grammarian the act of restoring the correct and authentic form of the work itself. The conception of the problem and an attempt to address it even in a primitive and hesitant way represents a decisive and permanent step in intellectual progress: a literary text had its own history of transmission, during the course of which it probably — or better surely — has been deteriorated in several parts. Restoring the text to its correct form could be achieved either via conjecture or by choosing the correct reading from those offered by a non univocal tradition. To obtain such a result required tools and methods. The recognition of transmissioninduced damage to the authentic text along with steps and procedures to restore it is proof of how the mutual dependency of textual criticism and textual interpretation became established and operational.39

 39 See Pasquali 1920: «a costituire un testo … occorre la stessa preparazione che a interpretare …; costituire un testo e interpretarlo sono, in fondo, tutt’uno» (“constituting a text … requires the same learning and knowledge as interpreting …; constituting a text and interpreting it are, ultimately, one and the same thing”), citation from the reprint of 1998, 26.

 From Book to Edition: Philology in Ancient Greece The Birth of Philology in the Hellenistic Age The scientific discipline that for many centuries has been known by the name of “philology” is believed to have first arisen in Greece during the Hellenistic age, that is to say, over the period from the third to the first century B.C.E. This view recurs with some regularity in studies and overviews of the history of philology, and can be said in general to represent an established and accepted fact.1 When, however, we seek to define its exact meaning and limits, above all in relation to philology (not only classical) in the modern age, a number of problems arise and non-negligible divergences come to light. From a historical perspective, we may speak, in this regard, of the Alexandrian scholars and the Alexandrian period (early Hellenistic age) as the essential and decisive stage, which began under the reigns of Ptolemy I Soter (305– 283 B.C.E.) and Ptolemy II Philadelphus (285–246 B.C.E.) and extended up to the second century B.C.E., centering on Alexandria with its institutions (chiefly the Library and the Museum) and its remarkable cultural ferment. The major personalities were Zenodotus of Ephesus (c. 325–260), Eratosthenes of Cyrene (c. 280–195), Aristophanes of Byzantium (c. 260–185), Aristarchus of Samothrace (c. 215–144), and the great scholar-poets Callimachus (c. 310–240) and Apollonius Rhodius (c. 300–220). The second century B.C.E. also saw the rise of the kingdom and city of Pergamum, which became a flourishing cultural center and rivaled Alexandria, also excelling in the field of philology and text interpretation: the figure of one of the greatest Pergamene representatives, Crates of Mallos,2 is usually mentioned for his controversies with Aristarchus, the “supreme authority as critic and interpreter” of the Alexandrian school.3  English translation by Rachel Barritt Costa. I would like to thank Fausto Montana, Lara Pagani, and Alessia Ferreccio for their valuable help.  1 For a survey of the history of classical scholarship see Pfeiffer 1968; Montanari 1993 (= this volume, ch. 1), 1994a; Montana 2012; Matthaios, Montanari, and Rengakos 2011; Montanari, Matthaios, and Rengakos 2015; LGGA is a specific lexicon of the figures of the ancient scholars. Dickey 2007 provides an overview of the materials of ancient scholarship. For an outline of the ideas and concepts of literary criticism present in these materials, see Meijering 1987; Nünlist 2009 (with the review by Pagani 2009). 2 Broggiato 2001. 3 Pfeiffer 1968, 232. https://doi.org/10.1515/9783110772371-009

From Book to Edition: Philology in Ancient Greece  

A second stage can be identified for the period extending from the generation of the first pupils of Aristarchus up to the work of the scholars of the Augustan age, such as Didymus, Aristonicus, and Theon — in other words, up to the end of the Hellenistic age. The attention of grammarians focused, first and foremost, on poets, obviously awarding priority to Homer, but also directed toward the works of lyric and scenic poets; prose writers also were objects of study, above all historians and orators (from a papyrus of the third century C.E. we have a fragment of Aristarchus’s commentary on Herodotus);4 finally, interest also began to center on “contemporaries” in a broad sense, that is, the major poets of the Hellenistic age. To assess the weight and importance of this cultural phenomenon it should be borne in mind that, by the age of Augustus, scholarship covered an extremely wide range of literary genres and took into consideration a vast chronological time span (from Homer to the Hellenistic age). For the purposes of this essay, I will focus on the Alexandrian period up to the second century B.C.E., a period we regard as the decisive phase that led to essential innovations. On the cultural and intellectual plane, we can endeavor to bring the problem into sharper focus (setting aside numerous other aspects) by inquiring into the type of activity the Alexandrian grammatikoi (the ancient term for “scholars”) effectively undertook, what purposes they had in mind and what they aimed to accomplish by examining the texts of their cultural heritage, and what genuine change and intellectual progress effectively ensued from their ideas and activities. I believe that the question hinges fundamentally on the form and content of what was defined as the ekdosis, “edition,” of a text carried out by an ancient scholar. What the Alexandrian philologists’ production of the ekdosis of a literary work really meant and truly involved is a problem that raises at least two main points, closely linked and mutually illuminating. That is to say, it is crucial to determine (1) how the ekdosis was performed, what material form it took, and how it was concretely built up — in other words, what a grammarian actually did when he set to work on producing an ekdosis,5 and (2) what was the real nature of the readings attributed by the erudite tradition to Alexandrian grammarians: conjectures ope ingenii and based only on subjective criteria, or variants deriving

 4 P.Amh. 12: Grenfell/Hunt 1901, 3–4 and pl. 3; Paap 1948, 37–40; cf. Pfeiffer 1968, 224–225. 5 This problem is connected to a number of aspects, which cannot be addressed here. Cf. Montanari 1998a (= this volume, ch. 23); 2000b (= this volume, ch. 55); 2002a (= this volume, ch. 47); 2004 (= this volume, ch. 41); 2009a (= this volume, ch. 7); 2009c (= this volume, ch. 60); 2011; Rengakos 2012.

  From Book to Edition: Philology in Ancient Greece from collation of copies and thus the result of a selection, or a mixture of both proceedings? Indeed, this is the most crucial and central node of the work of the Alexandrian scholars. We will see that it is vital to be aware that we face a problem of principles and method, not of quantity of the data or quality of the results. The ancients had a very clear idea of the concept of “textual reading,” and their technical terminology in this field made reference to the basic ideas of “reading” and “writing.” In Greek, the most widespread term in the erudite material known to us (scholia and grammatical works) is graphe (what is written) and the related verb graphein (to write), but the ancients also made use of anagnosis (what is read, a reading) and the associated verb anagignoskein (to read). In Latin one also finds scriptura, based on scribere (to write), but the most frequent term is certainly lectio, with the verb legere (to read). The concept of “variant,” expressed by varietas or more commonly by varia lectio, appears only later, in the Latin of humanism, and it has become established in modern philological terminology, where the term generally used is varia lectio/variae lectiones.6

Ekdosis: Correcting a Copy, Editing a Text The Hellenistic age has rightly been seen as a civilization based on books, a society in which the spread of written copies of poetic-literary works gradually increased and became customary. Possession of books and personal reading took on a much greater role than in the past, even though the use of written books had already begun to be significant in the preceding two centuries. As stated by Rudolf Pfeiffer: “it is obvious that we have reached the age that we called — hesitatingly — a ‘bookish’ one; the book is one of the characteristic signs of the new, the Hellenistic, world. The whole literary past, the heritage of centuries, was in danger of slipping away in spite of the learned labors of Aristotle’s pupils; the imaginative enthusiasm of the generation living towards the end of the fourth and the beginning of the third century did everything to keep it alive. The first task was to collect and to store the literary treasures in order to save them for ever.”7 The idea that scholars should be concerned with preserving the magnificent culture and education (paideia) of previous centuries was certainly not restricted to the material aspect of book production and the collection of exemplars. The decisive cultural impetus came from Aristotelian and

 6 Rizzo 1973, 209–215. 7 Pfeiffer 1968, 102.

From Book to Edition: Philology in Ancient Greece  

Peripatetic circles:8 intellectuals and men of culture realized that preserving the cultural heritage of a priceless and incomparable past could not be achieved without an understanding of its true worth and a proper interpretation of its content, and this called for the creation of appropriate and effective tools. In a logical order, which was also a chronological development, the first problem concerned the actual text of the great writers of the past, and the place of honor could not fail to be assigned to Homer, who had constituted the basis of the Greek paideia since the very beginning. In the period from Zenodotus to Aristarchus and his direct pupils (i.e., roughly in the third and second centuries B.C.E.), the Alexandrian ekdosis confirmed its place within ancient culture as a typical product of Hellenistic philology along with the continuous commentary (hypomnema), the monograph (syngramma), the collection of words peculiar in form or significance or rare and obsolete (lexeis or glossai), and other exegetical-erudite products. Zenodotus was chosen by King Ptolemy as the first head of the Library of Alexandria. In the source from which this piece of information is derived, he is defined as the first diorthotes of Homer.9 The term diorthotes is highly significant and is confirmed in another source, which states that during the reign of Ptolemy II Philadelphus (285–246 B.C.E.) two philologist-poets, Alexander Aetolus and Lycophron, “dealt with” plays (the former with tragedy, the latter with comedy),10 while Zenodotus “dealt with” Homer and the other poets. I have deliberately paraphrased the Greek verb with a neutral and imprecise phrase, “dealt with,” although in actual fact it is a precise and specific Greek term, that is, diorthoo, namely “straightening up, revising,” more precisely “correcting”: it is the verb from which is derived the designation diorthotes, used to characterize Zenodotus, literally “corrector.” The term that indicated the operation of correcting a text was, naturally, diorthosis (correction, emendation), which is used here in connection with both Zenodotus and Aristarchus. As Pfeiffer pointed out in this regard: “it is not improbable that Zenodotus, examining manuscripts in the library, selected one text of Homer, which seemed to him to be superior to any other one, as his main guide; its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Diorthosis can be the term for either kind of correction. It is hard to

 8 On the role of Aristotle and of the Peripatos see Montanari 2012 with the bibliography (= this volume, ch. 56), and in particular Montanari 1994a; Montanari 2000a (= this volume, ch. 38). 9 Suidas, Zenodotos Ephesios (zeta 74 Adler). 10 Tzetzes, Prolegomena de comoedia, Prooem. I 1–12, Prooem. II 1–4, 33.22–39 Koster; Alexander Aetolus TrGF 1, 100 T 6 = T 7 Magnelli; Lycophron TrGF 1, 101 T 7; cf. Pfeiffer 1968, 101, 105–106.

  From Book to Edition: Philology in Ancient Greece imagine any other way.”11 Klaus Nickau, who has produced fundamental studies on Zenodotus,12 is in agreement with the vision put forward by Pfeiffer on an important point, which in my view is to be made the basis of subsequent lines of reasoning: Zenodotus selected a copy he considered to be suitable and worked on it in various ways. Helmut van Thiel likewise believes the Alexandrian ekdosis consisted of the copy chosen by the grammarian from among those available, provided with a series of annotations.13 Martin West suggests that the particular eccentricity of Zenodotus’s text could not have been due merely to his judgment and opinions, but must in part have reflected an eccentricity of the tradition he followed: he may have worked on a rhapsodic exemplar produced in an Ionian context (perhaps brought with him to Alexandria from Ephesus, his native city), which thus reflected a line of tradition idiosyncratic and different from that which subsequently became widely accepted and was predominantly of Attic origin.14 Of course, this is no more than a mere hypothesis, which, however, is based on the same vision with regard to the manner of working of the first diorthotes of Homer: choosing a copy and performing a diorthosis, that is, carrying out corrections, emendations on the copy in question, in order to produce his own ekdosis. By pondering on these themes over the years, I have come to the conclusion that the problem of the characteristics of the Alexandrian ekdosis can be profitably addressed by starting from its material form as a book, on the basis of the following presupposition: in order to understand the nature of what we call a grammarian’s ekdosis of a text and what it contained, it is crucial to examine the way it was materially constructed. I have therefore tried to emphasize the importance of the relationship between the bookshop artifact on the one hand and the text as an object of philological editing, with its various paratextual elements such as annotations and critical signs (see below) on the other.15 We must take into account and award suitable prominence to what we know regarding the creation of new copies of texts, in the scriptoria by professional scribes or privately by individuals, along with insights that can be gleaned from surviving examples. To look at the problem in this perspective, the papyri are an essential

 11 Pfeiffer 1968, 110. 12 Nickau 1972, 30–31; cf. Nickau 1977. 13 Van Thiel 1991, IX–XIII; van Thiel 1996, 20102, V–VI; van Thiel 1992, discussed by Schmidt 1997, with a reply in van Thiel 1997. 14 West 2001b, 39; cf. Montanari 2002a, 123 (= this volume, ch. 47). 15 Montanari 1998; 2000b (= this volume, ch. 55); 2002a; 2004 (= this volume, ch. 41); 2009a (= this volume, ch. 7); 2009c (= this volume, ch. 60); 2011, and 2015, with extensive bibliography.

From Book to Edition: Philology in Ancient Greece  

source of information that cannot be disregarded; we will thus start from the papyri to search for data helpful to illuminate these issues. It is an accepted and well-documented fact that, in book production, new copies of literary works were normally reread and corrected through additional further comparison with the antigraph, at times even on the basis of a collation with other exemplars. Numerous types of evidence for this can be adduced on the basis of papyrus fragments of literary texts, and papyrologists are fully aware of the phenomena of corrections introduced in order to improve an exemplar in the framework of book production. Naturally we are particularly interested in the most ancient evidence, although we are hampered by the fact that the papyri datable to the period between the last decades of the fourth and the first half of the third century B.C.E. (the era of Zenodotus) are very limited in number. Notwithstanding, some small corrections of material errors can already be observed in the two most ancient surviving literary papyri, the Persians of Timotheus (PBerol. inv. 9875) and the renowned Derveni Papyrus,16 dated to the last decades of the fourth century B.C.E. (recall the dates of Zenodotus: c. 325–260). Such examples suggest that the corrections were not the result of a systematic revision but were made by the scribe, perhaps in scribendo. Though not a highly striking phenomenon, these occasional corrections of small errors certainly represent the most ancient and visible evidence of a concern for a correct text, or better, of the intention to correct a text in which an error could be perceived.17 A few decades later we already find some considerably richer and more significant evidence. One noteworthy witness is the Homeric roll P.Ilias 12, of which substantial parts are preserved, pertaining to books 21, 22, and 23 of the Iliad, dated between 280 and 240 B.C.E., thus still in the Zenodotean era or shortly thereafter, and in any case decidedly pre-Aristarchean.18 Apart from the fact — not particularly strange, given its chronological position — of having numerous plus-verses,19 this exemplar of the Iliad shows a particularly abundant quantity of corrections performed on the base text (in agreement or at variance with the so-called vulgata,  16 Turner/1987, 92; text in Kouremenos/Parassoglou/Tsantsanoglou 2006. 17 Montanari 2009a, 146–147 (= questo volume, cap. 7); 2011, 3–4. 18 P.Heid.Lit. 2 (inv. 1262–1266) + P.Hib. 1.22 (Bodl.Libr. inv. Ms.Gr.Class.b3(P)/2) + P.Grenf. 2.4 (Bodl.Libr. inv. Ms.Gr.class.b.3(P)) = MP3 979; cf. S. West 1967, 136–191; Sforza 2000. 19 It is well known that the witnesses (both direct and indirect) of the Homeric text that date from the early Hellenistic age (roughly up to the second century B.C.E.) show the presence of numerous additional lines as compared to the numerus versuum fixed at a later stage, which corresponds to that of the modern editions; cf. Haslam 1997 for an effective overview.

  From Book to Edition: Philology in Ancient Greece that is, the Homeric text that prevailed in the transmission) and of marginal signs (which are numerous, if one takes into account that the left-hand margin of the columns is often lost, and that the signs are considerably problematic). This is a witness that should certainly be the object of an in-depth reexamination from all points of view, including from the perspective of paleography, above all to determine the time gap between the base text and the subsequent interventions. Of a slightly more recent date, but equally significant, is P.Odyssey 31, dated to between 250 and 200 B.C.E., which contains parts of books 9 and 10 of the Odyssey.20 There are plus-verses and marginal signs, most of which are probably of a stichometric character, but the most important fact is that the roll underwent a twofold process of collation and correction: the first scribe would seem to have had two exemplars available, and he often corrected his text on the basis of another manuscript, after which a second hand inserted readings that are in agreement with the vulgata. This is of great importance for the question we are examining here: we have two Homeric exemplars from the mid-third century B.C.E., therefore definitely and decidedly pre-Aristarchean, which in addition to various kinds of — often somewhat problematic — critical signs and the expected plus-verses, also show clear evidence of collation with other copies and a conspicuous number of interventions performed on the base text at various times and in different ways.21 Michael Haslam has pointed out that of the over forty known Homeric manuscripts datable up to the middle of the second century B.C.E., in which the readings of the vulgata coexist with “eccentric” readings, quite a few certainly show traces of having been collated with another exemplar and consequently present interventions consisting of corrections and annotations of variants: “the Homer of readers in the 3rd and early 2nd century ... was appreciably more flaccid than the Homer of subsequent readers.” This was the situation that Zenodotus and his earliest successors found themselves facing.22 Another significant piece of evidence from the third century B.C.E. is the Milan papyrus with epigrams by Posidippus, which is rightly held to be of major importance in view of the quantity of corrections and annotations the text presents. The majority of the corrections were made by the same scribe, clearly in scribendo (in general amounting to one and never more than three letters,  20 P.Sorbonne inv. 2245 A = MP3 1081; cf. S. West 1967, 223–224. 21 Cf. Haslam 1997, 64–66; Rengakos 2012, 241–242. 22 Haslam 1997, 64–66; on the references to the ekdoseis of Homer in the papyri, see Pagani/ Perrone 2012.

From Book to Edition: Philology in Ancient Greece  

and all aimed at correcting minor slips in the drafting stage), but subsequently two other hands intervened with further emendations. The third person to make changes to the text in column 11 recorded a variant on the reading of line 30, noting it in the upper margin.23 The papyrus fragments of the following centuries, and in particular of the three centuries of our era (with regard to which the papyrus findings are most abundant) provide rich and valuable documentation of exemplars with interventions of deletion, addition, and correction of all types. Elsewhere I have examined and described various examples;24 here I will mention only one case, which I consider to be particularly significant: POxy. 2404 + PLaur. inv. III/278, a fragment of a papyrus roll (late second–early third century C.E.) containing a part of paragraphs 51–53 (POxy. 2404) and of paragraphs 162–163 (PLaur. III/278) of Aeschines’s oration Against Ctesiphon. It seems quite evident that this copy has been collated with a second exemplar and has been the object of detailed and systematic correction, seeking to identify the textual structure by distinguishing cola and periods and to correct copying errors for the benefit of the reader, and to emend the text in places judged unsatisfactory by means of various methods of deletion and by writing the alternative readings above or beside the wrong interpretations.25 I wish to emphasize at this stage why I have drawn attention to these manuscripts and their characteristics, with a choice of significant examples, to which others could easily be added.26 The point is not that they may be considered exemplars of a grammarian’s ekdosis: there is absolutely no evidence for such a suggestion. Rather, in my view they are of value because they highlight the importance of the techniques adopted in the workshop for book production and the effect such craftsmanship had on the development of a philological practice that sought to ameliorate and emend texts regarded as unsatisfactory due to the (real or supposed) errors they contained. The papyri provide ample evidence of the different methods used to “improve” an exemplar of a book, in other words to correct the (new) copy of a text. It was considered appropriate to add, remove, or modify letters or words that had been omitted or written erroneously, or cancel what was regarded as erroneous and replace it with what was judged  23 P.Univ.Milan. 309: see Bastianini/Gallazzi 2001, 15, 76–77; Montanari 2009a, 147 (= this volume, ch. 7); Montanari 2011, 4–5. 24 Montanari 2009a; 2009c (= this volume, ch. 60); 2011; also Montanari 2015 (= this volume, ch. 10). 25 Detailed analysis in Montanari 2009c with bibliography; on related problems, see above all Turner 1980, 92–93. 26 Other useful material can be found by using S. West 1967 and Haslam 1997, 63–69.

  From Book to Edition: Philology in Ancient Greece to be correct by writing the correction above the line, in the margins, and in the intercolumnia. At times specific markings were used to indicate the position referred to; often the correction was introduced in replacement of the previous words once these had been materially eliminated; sometimes the correct letters or words were simply written between the lines or in the interlinear space above the form judged to be incorrect, as a way of indicating, as it were, a self-evident deletion without the need for other material indications. On occasion, a horizontal or oblique line could be drawn through the letters or words to be deleted, or these letters or words could be marked by dots or lines above or below or enclosed within round brackets or even erased with a sponge.27 Thus there was a veritable toolkit for diorthosis. Often the interventions were carried out by the diorthotes (corrector) of the scriptorium (the book production atelier), whose task was to re-read and correct the text, also by comparing the copy with the model, in other words, through a practice of collation. Turner and Parsons write: One of the questions the palaeographer should ask about any literary manuscript is whether it has been adequately compared against its antigraph (the exemplar from which it was copied), a task which, in a publishing house, was the duty of the diorthotes (corrector), or whether it has been collated with a second exemplar (a procedure often carried out by private individuals to secure a reliable text). But several of our surviving papyrus manuscripts, and especially those which are beautifully written, contain such serious un-noted errors that it is clear their “proof-reading” was of a summary, superficial kind, if done at all. Those ancients themselves who set store by having a dependable copy (persons like Strabo and Galen) were aware of this weakness and adopted a routine to counter it: they themselves (or their secretaries) checked the copy to be used against another exemplar. If, therefore, the text had been checked against its first exemplar, and was later collated with a second, it may well bear the marks of this double checking.28

Best practice in book production consisted in a comparison between copies and corrections of mistakes, carried out by a professional or occasional corrector, who had adequate resources for deleting, adding, replacing, and marking various aspects and features of the text in order to improve it and increase its reliability. Even a private copy could be subjected to the same kind of treatment, with the use of the same tools and procedures, for personal reasons springing from cultural or research interests. Analogies with what we understand by

 27 Turner/Parsons 1987, 15–16, with reference to examples in plates; see also Turner 1980, 93 and pl. 8; Bastianini 2001. 28 Turner/Parsons 1987, 15–16; Turner 1980, 93 and pl. 8.

From Book to Edition: Philology in Ancient Greece  

“philological practice” are evident and need to be stressed: the methods and techniques adopted in the craft of book production honed the skills that were applied and developed by grammarians. A procedure that probably did not appear particularly strange or extravagant among those who used books every day developed into an extraordinarily innovative principle: the diorthosis of the corrector of the publishing house became the diorthosis of the philologist, diorthotes not of an individual copy of Homer but diorthotes of Homer. Effectively, concerns and emendations of specifically commercial book production took on a critical and philological-grammatical nature.29 The corrector of a publishing house aimed to produce an exemplar that would represent the best possible workmanship, a good copy suitable for sale on the book market or to a client, perhaps destinated to be the personal copy of a scholar or an educated man, who did his own corrections and annotations (we will note the case of Galen). In contrast, the grammarian’s underlying objective in correcting the text of his personal copy was more ambitious, because he sought to find the true and proper form of the work he was dealing with. He worked on a copy with the aim and intention of achieving, as it were, the model exemplar, which would display what in his view was the genuine form of the literary work in question. This conception led to the possibility of indicating doubts or a textual aporia, a perspective that certainly did not belong to the mental system and operational horizon of the craftsmen of the scriptorium. Thus in Zenodotus, drastic and univocal deletion (a typical action of the craftsman in book production, meaning “don’t write these words in the new copy”) for the first time was accompanied by the sign of philological uncertainty, namely the obelos, a simple horizontal stroke on the left of the line. This “critical sign” (semeion) marked a decisive intellectual change: attention began to focus on the work in its own right rather than merely on perfecting an individual copy. It is vital not to underestimate or downplay the invention of this critical sign, which had a momentous impact because it could also be applied systematically to poems of great length and cultural importance such as the Iliad and the Odyssey. By means of his simple semeion, the obelos, Zenodotus was able to indicate his suspicion that a given line might not be a genuine line of Homer, but that he was not sufficiently sure to be able to proceed with clearcut and definitive deletion of the element in question. Later, the discipline gradually progressed and further developed the system of semeia and the markings of exegetic reflection and erudite comment. By the time of Aristarchus the system of semeia had become complex and refined, but it had all begun with  29 Cf. Nickau 1977, 10–11.

  From Book to Edition: Philology in Ancient Greece Zenodotus’s small obelos and its radically new meaning for a reader of his texts.30 I believe that the philological work of the Alexandrian grammarians, starting from the first generation, represented something new in cultural history and marked significant intellectual progress. The reality and significance of this “revolution” become more evident and tangible if we grasp a fundamental chain of circumstances: the aspects and procedures of book production had molded a material and, in a sense, “craft-oriented” base of tools and working procedures that were adopted and utilized by grammarians for different purposes and in a different perspective. Thus the tools and methods of book production became the tools and methods of scholarship by virtue of an innovative and decisive intellectual change, which signaled a transition from the aims of pure craftsmanship, namely correcting an individual copy in the scriptorium so as to create a good product, to an intellectual aim of a philological nature, namely producing the exemplar that would contain what was held to be the correct text of the work. Thus, no longer would the copy be an exemplar of the work: rather, it would be the text of the work in itself, and this implied a sharp difference between “correcting a (single) copy” and “editing a text itself.”31 Let us recapitulate. Zenodotus worked on an exemplar of Homer that was available to him and that he deliberately chose for the specific purpose of producing his ekdosis. However, he had more than a few reservations about it, concerning both the numerus versuum and a certain quantity of readings. He had doubts about the authenticity of some lines, and adopted a sign indicating his suggestion that they should be expunged, the obelos, which he marked beside the lines; the technical term for this operation was athetein, the operation was the athetesis. But he also believed that some lines should be deleted from the text as definitely spurious and to be completely rejected. Such lines would have been present in his base text, and he must have used one of the graphic methods mentioned above for lines that were clearly and definitely intended to be deleted and omitted from the text (with or without verbal annotations accompanying the deletion sign).32 Zenodotus’s ekdosis was a corrected copy, and  30 Conte 2013, 63 mentions in general terms a basic principle of good philology: “it should not be forgotten that the cases of athetesis proposed ... should not fill us with terror almost as if they were irreparable death sentences. The athetized text survives, thank goodness, it remains readable, it can always come back from momentary exile.” This intellectual breakthrough was achieved by Zenodotus with his little obelos. 31 Montanari 2011. 32 This is the most plausible explanation for the different terminology used in the scholia for Zenodotus’s text alterations of athetesis by the obelos on one side and line deletion on the other

From Book to Edition: Philology in Ancient Greece  

for this reason the operation has been called diorthosis, correction, and he has been called diorthotes, corrector, of Homer. It is not unlikely that the paratextual apparatus on the working copy may have given rise to problems of comprehension and readability, especially with the accumulation of interventions over time, and in places where the multiple interventions on the text became interlaced with one another. The copy bearing the work of diorthosis resulted materially in the philologist’s own ekdosis, and we can conceive of this as a product of years of study that led over time to a series of interventions on the same exemplar. Together with critical semeia, explanatory annotations must have been present in the working copy starting from Zenodotus onward, and probably continued to be used by grammarians in their editorial and exegetic work. I therefore feel it is far more plausible to assume that the ekdosis became available for consultation by scholars, poets, and intellectuals as soon as the grammarian himself, or someone working on his behalf, had had a copy made that followed the indications in the base text on which diorthosis had been performed. An exemplar would thus be created that was a correct and “fair copy” of the work,33 but still bore the name of the grammarian who was the author of the copied diorthosis, with the marginal annotations that would still be necessary once the text had been properly prepared. In short, first there was a working copy belonging to the diorthotes, with all his interventions and annotations, after which it was possible to proceed to reproducing it as a “fair copy” of his ekdosis. Thus it was a stepwise production, which we should obviously imagine to have been done not only for Zenodotus but also for his successors. This would also explain the conservation and transmission of the interventions and textual choices made by the grammarians.34 Zenodotus’s choice of the base text of Homer seemed highly debatable and was open to criticism, which is why Aristophanes of Byzantium and Aristarchus

 side. For cases of deletion, one finds the expressions ou graphein (do not write: the most frequent), ouk einai (is not there), ou pheresthai (is not handed down) and a few others (cf. Ludwich 1884–85, 2:132–135; Nickau 1977, 1–30). In Montanari 1998a (= this volume, ch. 23); 2000b (= this volume, ch. 55); 2002a (= this volume, ch. 47); 2002b (= this volume, ch. 58), I discuss some possible objections to this reconstruction; Rengakos 2012 expresses some reservations with regard to this vision: reply in Montanari 2015 (= this volume, ch. 39). 33 Perhaps ou graphein (do not write: see previous note), which is the most frequent expression for elimination, may go back to Zenodotus himself and may have been an indication to whoever transcribed his ekdosis that the element in question was not to be copied. 34 Helpful confirmation comes from a testimony by Galen, mentioned further on.

  From Book to Edition: Philology in Ancient Greece chose exemplars with noticeably different characteristics.35 Consequently, a line of tradition predominantly of Attic origin gradually spread, partly by virtue of the base text of working copies used by grammarians who were active at a later period than Zenodotus. That base text of post Zenodotean grammarians proved decisive above all as regards the numerus versuum, whereas the readings suggested by individual grammarians generally did not become standard in the vulgata. The plus-verses present in the Zenodotean text were not his own interpolations but were typical of exemplars current in his day:36 they disappeared because the work of Aristarchus led to general recognition of a text that had a very similar number of lines to our vulgata. It is significant that, after Zenodotus, grammarians like Aristophanes did not go so far as to carry out the drastic act of line deletion: in other words, that Aristophanes stopped performing material deletions on his own copy with the graphic techniques mentioned above. The obelos became the prime tool for expressing a cautious doubt concerning parts of the text of the working copy: ou graphein (do not write, eliminate from the text) disappeared, leaving only athetein (suggestion that the line may be spurious and then should be expunged).37 Aristarchus followed the same procedure. This explains why many of the lines Zenodotus had decided to eliminate from the Homeric text — but that were present in the copies chosen by later grammarians — were preserved in the numerus versuum that became the generally accepted tradition after the Aristarchean age and thus remained in our vulgata.38 The abandonment of the drastic practice of material deletion highlights the increasing sense of caution that had developed, and accounts for the fact that many of the lines “deleted” by Zenodotus were in effect no longer deleted39 and thus were not obliterated from the tradition. The work of Aristarchus marked the period in which Alexandrian philological production included the drafting of extensive commentaries (hypomnemata). The great continuous commentary, which followed the text step by step, greatly facilitated and enriched the communication and preservation of the arguments and motivations put forward by the grammarians, so the material that has come  35 Montanari 2002a, 123–125 (= this volume, ch. 47); West 2001b, 36: “clearly Aristophanes and Aristarchus were not dependent on Zenodotus’ text but followed another source or sources more similar to the vulgate”; cf. West 2002, 138. 36 Haslam 1997; West 2001b, 40; cf. previous note. 37 Or else, if genuine deletions were still carried out, they were of such minor relevance that all knowledge of them was lost; cf. notes 32 and 33 here. 38 Haslam 1997, 85; West 1998–2000, vol. 1: VII; Führer/Schmidt 2001, 7. 39 It is sometimes stated, instead, that they were “recovered” or “reintroduced”: this would have involved far more complicated operations.

From Book to Edition: Philology in Ancient Greece  

down to us from this tradition is much more substantial. Yet the ekdosis as an annotated working copy by no means went out of use, as clearly testified by the information on the Aristarchean edition(s).40 However, the hypomnema on arguments pertaining to text criticism and exegesis constituted an important resource. In practice, the need to write on the copy chosen as the base text was no longer so strongly felt, especially as regards philological-exegetic arguments. Previously, before the rise of separate commentaries, there had been a greater need to write on the working exemplar, but with Aristarchus the particularly elaborate system of critical signs placed next to the lines41 as well as the variants and the readings to be adopted must have been present in the margins and interlinear spaces, while the philological-exegetic treatment was mostly developed in the commentary. Marginal annotations continued to be utilized whenever they were felt to be of practical use, for example, for short notes and textual proposals.

Variants and Conjectures An interesting testimony concerning these problems can be found in the recently discovered De indolentia by Galen, an author of major importance in the history of ancient philology, not only on account of his activity and thought but also by virtue of the information Galen’s text provides. It has begun to be studied and appreciated from this point of view, but certainly much fruitful investigation remains to be done.42 The new text is preserved in a manuscript that, overall, has many incorrect forms and results in considerable uncertainty of interpretation, also affecting the points of interest here, but it is worth commenting on the material and singling out several pieces of information.43 In the work in question Galen relates that in the 192 fire of Rome he lost, among other things, all the books he possessed; he talks extensively about his activity as a scholar and about his books. Those lost included texts “corrected in my own hand” (par. 6); there were also rare books that were not available else 40 As regards Aristarchus’s Homeric ekdos(e)is here I will restrict myself to referring the reader to the arguments put forward in Montanari 1998a (= this volume, ch. 23); 2000b (= this volume, ch. 55), and 2002a (= this volume, ch. 47), of which approval has been expressed by Nagy 2003 and 2010, 21–37; Rengakos 2012, 252 (cf. West 2001b, 61–67). 41 They could also be repeated in the hypomnema beside the lemmas, as was the case for instance in P.Oxy. 1086 (pap. II Erbse). 42 Manetti/Roselli 1994; Manetti 2006; 2012a; 2012b; 2015; Roselli 2010; 2012. 43 Editions: Boudon Millot/Jouanna 2010; Kotzia/Sotiroudis 2010; Garofalo/Lami 2012.

  From Book to Edition: Philology in Ancient Greece where, and books that, while not rare, constituted unrepeatable exemplars due to the particularly accurate and carefully written text, such as the Plato by Panaetius and two Homers by Aristarchus, and others of this kind (par. 13). There follows a rather tortuous passage, which may contain a reference to copies with marginal annotations and bearing the name of the person who had (originally) made the marginal jottings.44 A little further on, Galen relates that he also lost books he himself had worked on, in which he had corrected various errors in order to compose an ekdosis of his own. The task he had set himself, he says, involved careful attention to textual readings, to ensure that nothing was added or left out and that all the appropriate signs were present to distinguish the structural parts of the text,45 as well as the punctuation — the latter being so important, especially in obscure works, that it could even substitute for the exegete himself (par. 14). It has been rightly emphasized that here a new aspect of the personality of Galen emerges: already known as an exegete and commentator, he can now also be seen as a text editor (not only of medical works but also of the works of numerous philosophers). Thus he was the author of editions (ekdoseis) designed for his own personal use: in preparing them, he worked on the text in order to identify lacunae and interpolations and to highlight the structural framework of the work with appropriate diacritical signs, as well as to indicate the punctuation as an important aid to text comprehension.46 Elsewhere Galen cites a number of ekdoseis, including that of Hippocrates by Bacchius (a partial contemporary of Zenodotus), dating from the third century B.C.E., and by Dioscorides and by Artemidorus (from the age of Hadrian, one generation earlier than Galen). Especially for the ekdosis by Dioscorides, Galen offers important insight into its material form: it presented diacritic signs and punctuation, the obelos was used to indicate doubtful authenticity, and variants were marked in the blank spaces (lower and upper margins and the intercolumnia).47 The information Galen provides on the ekdoseis of medical texts that he himself had performed or that had been carried out by his predecessors is in agreement with the arguments put forward so far. The philologist selected a copy on which to work and thus produced his own edition by correcting it; he then personalized this copy, on which all his markings were visible, by writing his own name on it, after which the copy was allowed to circulate for essentially

 44 Cf. Roselli 2010; 2012; Stramaglia 2011; Manetti 2012b, 14–16; 2015. 45 Paragraphoi and coronides are mentioned. 46 Cf. Boudon Millot/Jouanna 2010, xxxiii–xxxiv; Manetti 2012a; Roselli 2012; Manetti 2015, with bibliography; on lectional signs, punctuation, and accentuation, cf. Pfeiffer 1968, 178–179. 47 Cf. Manetti/Roselli 1994, 1625–1633; Manetti 2012a; 2015; Roselli 2012.

From Book to Edition: Philology in Ancient Greece  

private use, or for school and teaching purposes (like the Homer of Zenodotus or of Aristarchus). On request or for various reasons, the copy itself could then be copied, that is, reproduced as complete exemplars of the work, corrected and presented as “fair copies.” This could be done either by the editor himself or by someone entrusted with the task. In paragraph 14 Galen mentions precisely the case of books transcribed as fair copies after undergoing correction (diorthosis; see above).48 Our reconstruction of the manner of carrying out the Alexandrian ekdosis, based on observation of well-documented and purely material and technical facts, helps to clarify, on a more solid basis than usual, the problem of the real nature of the readings attributed by the erudite tradition to the Alexandrian grammarians. Were they conjectures ope ingenii based solely on subjective criteria, or deliberate choices among variants attested by documentary sources and deriving from the collation of copies, or a combination of both? Was there a practice of comparing a variety of exemplars of the Homeric text to identify the differences and thereby offer an opportunity for choice? This is a problem of fundamental importance — indeed, it is one of the most disputed issues in the history of ancient philology, not only for the history of the Homeric text in antiquity — inasmuch as these questions are crucial in evaluating the work of the Alexandrian philologists and their role in intellectual and cultural history. Let us now take another look at the passage from Pfeiffer cited at the beginning: “it is not improbable that Zenodotus ... selected one text of Homer ... its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Diorthosis can be the term for either kind of correction. It is hard to imagine any other way.” 49 Considering the terminology used, it is in fact almost impossible to imagine any other way, which means that the Alexandrian philologists’ production of an ekdosis must have involved both conjectural emendations and choice among variants detected through the collation of copies: “Zenodotus’ text is shown to be based on documentary evidence,” says Pfeiffer.50 The two aspects, namely conjecture and comparative assessment of copies, have received differing emphasis, with some scholars suggesting that the idea and practice of comparing different copies and choosing among variants generated by collated texts was alien to the Alexandrian critical-philological mindset, at least until Aristarchus, perhaps until Didymus. They maintain that the  48 Roselli 2012, 64–67. 49 Pfeiffer 1968, 110. 50 Pfeiffer 1968, 114.

  From Book to Edition: Philology in Ancient Greece Alexandrians conjectured exclusively (or at least mostly) with the aim of changing, without too many scruples, a text judged to be unacceptable on the basis of a series of subjective criteria, such as supposed inconsistency, inappropriateness, moral principles, material repetitions, preference for textual concision, standardization, and uniformity. I would argue that it is not appropriate either to assert that all the readings espoused by the Alexandrian grammarians were merely arbitrary conjectures or, alternatively, to claim that they were all readings derived from collated exemplars. One may far more reasonably think of a mixture of conjectures and critical assessment of variants, complete with the work of interpretation this implied. Naturally, given the evidence available to us today, it will be very difficult to distinguish case by case whether a reading represented a personal and subjective conjecture or rested on a documentary source, unless we resort to hypotheses and deductions that may not always be reliable and show a high degree of arbitrariness. But this is our own problem in interpreting individual cases and readings — it is not a problem regarding the modus operandi and the method of the Alexandrian philologists. The two levels must not be superimposed, and the fact that we have a serious lack of definitive criteria to distinguish coherently between what springs from a conjecture and what represents a variant by no means implies that one of the two categories is misleading and should be excluded. The idea that the Alexandrian philologists offered only arbitrary conjectures (“Konjekturalkritik”) and did not carry out any collation of copies (no documentary basis) has had a number of supporters.51 This tendency leads recta via to a (quite unfair) underestimation of the importance and the value of the work

 51 First of all, van der Valk 1949; 1963–64, whose line of interpretation was also adopted (naturally with individually differentiated stances) by H. van Thiel and by M.L. West (West 2001a; 2001b; 2002, cf. 1998–2000, Praef. VI–VIII: discussion in Montanari 2002a [= this volume, ch. 47]; 2004 [= this volume, ch. 41]; 2015 [= this volume, ch. 10]). Sharp criticism of van der Valk’s ideas has been put forward: for example, Rengakos 1993, 17–48; Haslam 1997 (see 70 n. 31); Rengakos 2002a, 146–148. Van Thiel 1992 and 1997 (see also 1991, Einleit.; 1996, 20102, Einleit.) has argued that the readings attributed to the Alexandrian grammarians were actually exegetic glosses or mere indirect references or reminiscences of parallel passages, written in a “Rand- und Interlinearapparat,” which Didymus, Aristonicus, and others then wrongly interpreted as textual variants: as far as I know, no one has accepted this rather idiosyncratic vision (discussed in Montanari 1998a, 4–6 [= this volume, ch. 23]). West 2001b, 36 and West 2002, 140: “in fact, the first scholars known to have cited manuscript authority for variant readings are Aristarchus’ contemporaries Callistratus and Crates. Didymus is the first author known to have compiled anything in the nature of a ‘critical apparatus.’ It is entirely unjustified to project his methods back onto Aristarchus or Zenodotus.”

From Book to Edition: Philology in Ancient Greece  

performed by the Alexandrians. Arguments against it have been adduced by various scholars.52 On the question of the “Konjekturalkritik” often and abundantly attributed to the Alexandrians, it can easily be observed that this is a theory based on the false presupposition that we have general criteria for distinguishing between conjectures and genuine variants, when we are faced with the overall set of readings contained in the erudite sources. But in actual fact no such criteria exist (except in a distinctly rationalistic and perhaps somewhat naïve illusion). Furthermore, in the sources there is no explicit testimony referring to conjectural interventions, and it is impossible to demonstrate that a given reading is the fruit of a conjecture by the philologist to whom the textual choice is attributed (see above). On the contrary, there is actually a considerable amount of plausible evidence of the Alexandrian philologists’ knowledge of variants deriving from a comparison among copies.53 In addition to the arguments already illustrated above, based on the papyri and on the general practice of book production, Antonios Rengakos rightly invokes the testimony offered by the poets of early Hellenism, that is, of the Zenodotean age, who reveal knowledge of different preexisting Homeric readings: “do Hellenistic poets offer cases which prove beyond doubt that they made use of different Homeric manuscripts? In other words, do their works display Bindefehler which point to the older Homeric tradition? The answer is clearly ‘yes.’”54 Indeed, we may confidently maintain that some of the Homeric variants testified in the lines of the philologist-poets of the Zenodotean age derived from the consultation of manuscripts and collation of copies. To this should be added cases in which it can be demonstrated, by finding veritable conjunctive errors, that the variants chosen by the Alexandrians already existed in a more ancient Homeric tradition.55 Indeed, Pfeiffer himself explicitly supported this argument, reaching the following conclusion: “these three examples from the fifth to the third centuries, in which Zenodotus’ text is shown to be based on documentary evidence, show how unjustly he was

 52 Haslam 1997; Schmidt 1997; Führer/Schmidt 2001, 6–7; Nagy 2000; 2003; 2004, 3–24; 2010, 1–72; Nardelli 2001a (especially 52–70) in direct opposition to West’s theories; Rengakos 2002a; 2002b; and 2012; Montanari 2002a (= this volume, ch. 47); 2004 (= this volume, ch. 41); 2015 (= this volume, ch. 10); Conte 2013, 44–50. 53 Lately Rengakos 2012 (on “Konjekturalkritik”, 247). 54 Rengakos 2002a, 149; cf. Rengakos 1993; 2001; 2002b; 2012; an interesting case pertaining to Zenodotus is highlighted by Fantuzzi 2005. 55 Lately Rengakos 2012.

  From Book to Edition: Philology in Ancient Greece charged by ancient critics, and by those modern scholars who followed them, with making arbitrary changes for wrong internal reasons.”56 Besides this indirect evidence, direct evidence can be found, and I believe that it is decisive. Explicit testimony is supplied by the scholia, where one finds several undeniable references to the fact that Aristarchus consulted a number of different ekdoseis and found them to contain divergent readings: in other words, he certainly availed himself of the direct tradition of the copies he had at hand. The most evident and irrefutable case is that of sch. Il. 9.222 b, where Didymus reports that Aristarchus accepted a reading (graphe) because he found that it appeared in this form in many ekdoseis. Equally significant is sch. Il. 6.4 b, where Didymus states that Aristarchus at first accepted a certain reading, but later changed his mind because he had found another reading that he deemed to be preferable.57 This is a clear testimony that, when engaging in text criticism, the Alexandrians — starting with Zenodotus, the most refined method being reached with Aristarchus — based themselves not only on text-internal conjectural proposals but also on external and diplomatic resources, consisting in choice among variants they found or noticed in a nonunivocal tradition composed of the copies they had available and were thus able to consult. It would seem, therefore, that the burden of proof is on whoever seeks to strip the Alexandrian grammarians of any knowledge of variants deriving from collation of copies, attributing to them only arbitrary conjectures, rather than the opposite: the fact is that we have, at the very least, convergent evidence in favor of knowledge of variants — and I would go so far as to say that we have real proof.58 Martin West warns against a travesty of the situation: “The misapprehension, which goes back at least to the time of Wolf, is that Zenodotus, Aristophanes

 56 Pfeiffer 1968, 110–114: the citation is on 114; the three examples adduced by Pfeiffer concern Il. 1.5, Il. 1.225–233, and Il. 16.432–458, Il. 4.88. Pfeiffer normally attributed the collation of copies to the great philologists who succeeded Zenodotus: cf. for example 173. Pfeiffer’s arguments should have been awarded greater consideration. 57 Cf. Rengakos 2012, 244–248, with bibliography. The abovementioned evidence of Didymus in sch. Il. 9.222 b is rightly underlined by various scholars (Nagy, Janko, Rengakos, and myself) and cannot be dismissed out of hand, as does West (2001b, 37 n. 19). On Aristarchus’s second thoughts, see Montanari 1998 and 2000b (= this volume, ch. 55). 58 I believe that it is simply begging the question to claim that Didymus’s method is “projected back onto Aristarchus” or Zenodotus (see note 51 here): in actual fact there is absolutely no evidence that Didymus was the first to apply this method rather than having inherited it from his predecessors. West’s “Didymean” hypothesis is rejected by Nagy 2000; Nardelli 2001a, 61– 64; Janko 2002; Rengakos 2012.

From Book to Edition: Philology in Ancient Greece  

and Aristarchus were all editors in the modern sense, who wanted to establish a good text of Homer and who approached the task as a modern editor does, by collecting manuscripts and comparing their readings.”59 But what is likely to have been the aim in carrying out emendations (even only arbitrary conjectures!) on the Homeric text? Are we thus to believe that Zenodotus had a conscious premeditated idea of “modeling” Homer according to his own taste, that is: “I’m going to set about reworking Homer and I’m going to make it the way I think it ought to be”? This paradoxical possibility is by no means easy to accept, but actually this is the only alternative to the view that “he wanted to establish a good text,” which is the natural goal of anyone who starts working on a text, whatever the value of the result. The fact that Zenodotus used the obelos in the margins to signal the critic’s suspicion that certain verses were un-Homeric60 obviously means that Zenodotus felt he had to tackle the problem of how to discriminate the authentic from the spurious. “The critic’s suspicion that certain verses were un-Homeric” suffices, for once a critical approach toward the way the text presents itself has been acquired, the problem at hand resides in the opposition between authentic/correct versus spurious/damaged and in seeking to identify the proper text. By addressing the issue of the authentic text and how to devise the criticalmethodological tools to obtain it, Zenodotus achieved a major breakthrough: it was a crucial intellectual step, identified above as residing in the difference between “correcting a single copy” and “editing a text.” I fear that misunderstandings arise from the fact that there is no clear definition of the guidelines for our judgment on the work of the Alexandrian philologists. By adopting our own point of view concerning the “competence” on which they based their opinions and arguments, so as to ascertain whether and when they were right or wrong in comparison with the “truth” according to scientific philology, we risk producing unfounded and pointless judgments. Naturally, evaluation of the quality of their choices is the proper perspective for the interpreter and editor of Homer as a modern philologist. In contrast, maintaining conscious awareness of historical distance and taking care not to overlay our criteria on their behavior is the proper perspective for the historian of philology as a cultural and intellectual phenomenon and for the reading of

 59 West 2001b, 34; 2002, 138. 60 West 2001b, 38; 2002, 140. The idea of the possible use of signs for material deletion placed on the base copy was clearly put forward already in Montanari 1998, 6, but West does not seem to be concerned with the distinction between athetein and ou graphein in the terminology on the textual interventions of Zenodotus (see above, notes 32 and 33).

  From Book to Edition: Philology in Ancient Greece Homer in ancient civilization. Perhaps it is hard to admit that Zenodotus’s aim (however incoherent and unsophisticated) was precisely “to establish a good text of Homer,” inasmuch as the testimony that has come down to us indicates that his text was far from good, from our point of view — in fact it was dreadful, and incoherent seen through the filter of the requirements and knowledge of modern scientific philology. And even as regards the successors of Zenodotus, including the great Aristarchus, we can hardly claim always to agree with their text choices. The viewpoint from which a Homeric scholar approaches his task is the need to decide whether the text Zenodotus, Aristophanes, or Aristarchus judged to be the best is indeed the one to print in a present-day critical edition,61 and whether their interpretations should be espoused as valid in a scientific commentary. A historian of ancient philology, by contrast, starts out by seeking to understand their methods, arguments, principles, and knowledge — in a word, their historical and intellectual position. The tendency to scoff at the opinions of the Alexandrian philologists in terms of modern Homeric studies should by no means translate into discrediting their historical significance, which needs to be correctly positioned and contextualized. It is a mistake to blur the distinction between the two planes.62 It is impossible to escape the fact that by inventing the obelos and setting himself the task of emending and restoring the text he had at hand, Zenodotus had lit upon an idea that, however embryonic and crude it may appear, would undergo further development among his successors, eventually becoming the germ of the discipline we call classical philology. But even if one were to suppose that  61 Whatever idea of a critical edition of Homer, whatever idea of spurious and authentic lines one has in mind: G. Nagy and M.L. West have two very different conceptions, but it would be beyond the scope of this essay to address the issue here (but see Montanari 2004 [= this volume, ch. 41]; Conte 2013, 38–78). 62 Janko renders this concept rather more clearly. He believes that the majority of the readings of the Alexandrian philologists are indeed arbitrary conjectures, but he does not go so far as to deny recourse to manuscripts and comparison among copies as part of their work (Janko 1992, 23 and 27; 2002). On the one hand, Janko argues, there stands the problem of the origin of their proposed text choices (subjective emendation, documentary sources) and therefore of their working procedures; on the other, he points out, “my own concern, as a Homerist, has always been whether such readings are authentic.” Cassio 2002, 132 offers an extremely apt remark, on the issue of pre-Alexandrian criticism but perfectly applicable to Alexandrian philology as well: “the earliest scholarly approach to the Homeric text is totally foreign to us ... we do right to think along very different lines, but we should never forget that it was the commonest approach to the Homeric text in the times of Socrates, Plato and Aristotle. As a consequence, we ought to be wary of looking at it with a superior smile, and ought to try to understand its motives in more depth instead.”

From Book to Edition: Philology in Ancient Greece  

he acted purely on the basis of conjectures, could it be denied that conjecture is one of the emblematic and representative tools of philology aimed at restoring the correct text? A further comment by West is surprising: “Consider what we know of Aristarchus’ methods, for which we have plenty of material in the scholia. Of course he had the text of his teacher Aristophanes before him. He also kept an eye on that of Zenodotus,63 and took up critical positions against it. But the arguments he used were always based on the internal evidence of contextual coherence or general Homeric usage. Not once does he appeal to the authority of manuscripts.” 64 Be that as it may, the picture implies he made a certain small comparison among copies, but that he took great care to avoid consulting any other exemplar than his own, together with that of Zenodotus and that of Aristophanes, although these alone already presented him with various divergences: a paradoxical Aristarchus who, despite his concern for the Homeric text, made every effort not to look at other copies he may have come across, not to note the points where they departed from his own copy, and not to ask himself any questions about those differences.65 It seems to me far more likely that he noticed the differences, in both the number of lines and individual readings, and decided to write them down and express his opinions.66 However, a subtle ambiguity needs to be eliminated: when speaking of “other” or “various” copies of Homer that were actually available and utilizable, one should not be misled into thinking that hundreds of exemplars were concretely at hand and ready for consultation, thereby transforming the idea of a comparison into an exaggerated undertaking that becomes totally implausible. Obviously, it would be a pure anachronism to assume that the Alexandrian philologists had conceived the idea of a collatio codicum of the entire known manuscript tradition, after a recensio in the style of the so-called Lachmann method: but who would dare advance such a ludicrously naive proposal? In actual fact the problem should be considered in a rather different fashion, embracing a perspective that is perfectly reliable in historical terms. More specifically: can one begin to speak of comparison among copies only when a certain number (how many?) is reached, or was it sufficient to compare a few, to detect

 63 My italics: the phrase is insidiously reductive, given the hundreds of cases preserved by a very incomplete tradition. 64 West 2001b, 37: at least the case of Il. 9.222 clearly contradicts this; cf. above. 65 West 2001b, 38: “No doubt it would have been easy for him [scil. Zenodotus] to collect several copies if he had taken the trouble”: a somewhat lazy philologist? 66 Rengakos 2012 provides highly cogent arguments in this regard.

  From Book to Edition: Philology in Ancient Greece variants when the textual tradition was not univocal, and then address the problem of which text was correct and which ones were wrong? 67 Overall, we must recognize that we owe to the Alexandrian grammarians an idea of text philology aiming to establish a good text, to restore the correct text, freeing it from errors and damages. From the age of Zenodotus onward, progress was gradually made in refining the method, which achieved its highest accomplishment with Aristarchus. The grammarians realized that a text had its own history of transmission, during the course of which it deteriorated in various ways; it could be restored to its correct form either via conjecture or by choosing the correct reading from among those offered by a nonunivocal tradition.68 The recognition of transmission-induced damage that had affected the authentic text, along with steps and procedures to restore it, is proof of how the mutual dependency of textual criticism and textual interpretation became established and operational.69

Conclusions This essay started by stating that we would concentrate on the elements we regard as essential for an appropriate vision of intellectual and cultural history, namely the birth of an idea, an approach to literary texts and their interpretation, an initial seed, which came to represent the dawn of that discipline we now call (classical) philology. To bring to a conclusion the various points outlined in the preceding pages, it is worth noting once more that we are dealing with a problem of principles and methods, not of the quantity of the data (number of collated copies or of variants discussed) or of the quality of the results (right or wrong from our point of view). We are not concerned with establishing the minimum number of copies to be subjected to comparison or of variants to be considered before one can even begin to speak of philology, nor with determining how many “correct” readings or “good” interpretations are needed before it makes sense to speak of philology.

 67 Montanari 1998a, 2 (= this volume, ch. 23); Rengakos 2002a, 146. 68 Naturally Homer comes to mind, but also the texts of the tragic and comic poets. 69 Pasquali 1920 (citation from the reprint of 1998, 26): “a costituire un testo ... occorre la stessa preparazione che a interpretare ... ; costituire un testo e interpretarlo sono, in fondo, tutt’uno” (“constituting a text ... requires the same learning and knowledge as interpreting ... ; constituting a text and interpreting it are, ultimately, one and the same thing”).

From Book to Edition: Philology in Ancient Greece  

Rather, in a historical perspective, all that was needed for there to be a decisive step forward in intellectual achievement was the very fact of understanding and addressing the problem, even if only partly, erratically, and incoherently: a literary text had a multifaceted history of transmission, during which it could become distorted at various points; the correct text (i.e., what is authentic versus, what is spurious and what was the original wording) could then be restored by conjecture or by choosing the best reading among those offered by a divergent tradition.70 The idea of the recognition of damage and of finding a way to repair it reveals that the organic unity between interpretation and textual criticism had become established. Although much progress still remained to be made, and Wolfian scientific philology, the modern critical edition, and the scientific commentary were still in the distant future, our viewpoint — far from being an anachronism — is the historical evaluation that a nodal step had been taken in the period from Zenodotus to Aristarchus.

 70 Cf. now Conte 2013, 44–50.

 Ekdosis: A Product of the Ancient Scholarship The Form of the Alexandrian Ekdosis The Hellenistic age has rightly been seen as a civilization based on books, that is to say, a society in which the spread of written copies of poetic-literary works gradually intensified and became customary. Possession of books and personal reading became considerably more significant than in the past, even though use of written books had already begun to play an increasing role in the preceding two centuries.1 As stated by R. Pfeiffer: “It is obvious that we have reached the age that we called — hesitatingly — a ‘bookish’ one; the book is one of the characteristic signs of the new, the Hellenistic, world. The whole literary past, the heritage of centuries, was in danger of slipping away in spite of the learned labours of Aristotle’s pupils; the imaginative enthusiasm of the generation living towards the end of the fourth and the beginning of the third century did everything to keep it alive. The first task was to collect and to store the literary treasurers in order to save them for ever”.2 The idea that scholars should be concerned with preserving the magnificent culture and education (paideia) of previous centuries was certainly not restricted to the material aspect of book production and the collection of exemplars. The decisive cultural impetus came from Aristotelian and Peripatetic circles:3 intellectuals and men of culture realized that preserving the cultural heritage of a priceless and incomparable past could not be achieved without an understanding of its true worth and proper interpretation of its content, and that such a task called for the creation of appropriate and effective tools. In a logical order, which however was also a chronological development, the first problem concerned the actual text of the great writers of the past, and the place of honour could not fail to be assigned to Homer, who had constituted the basis of the Greek paideia since the very beginning.

 1 For a survey of the history of classical scholarship see Pfeiffer 1968; Montanari 1993 (= this volume, ch. 1); Montanari 1994a; Matthaios/Montanari/Rengakos 2011; Montana 2012; Montanari 2015a (= this volume, ch. 9), and Montana 2015; LGGA is a specific lexicon of the figures of the ancient scholars; Dickey 2007 provides an overview of the materials of ancient scholarship, see also Dickey 2015; for an outline of the ideas and concepts of literary criticism present in these materials, see Meijering 1987; Nünlist 2009a (with the rev. by Pagani 2009). 2 Pfeiffer 1968, 102. 3 On the role of Aristotle and of the Peripatos, see Montanari 2012 (= this volume, ch. 22) with the bibliography (in particular Montanari 1994a; Montanari 2000a [= this volume, ch. 38]); see also Montana 2015; Hunter 2015; and Nünlist 2015. https://doi.org/10.1515/9783110772371-010

Ekdosis: A Product of the Ancient Scholarship  

In the period from Zenodotus to Aristarchus and his direct pupils (i.e. roughly in the 3rd–2nd c. BC), the Alexandrian ekdosis confirmed its place within ancient culture as a typical product of Hellenistic philology along with the hypomnema, the syngramma, the collection of lexeis and other exegetical-erudite products.4 Zenodotus was chosen by King Ptolemy as the first head of the Library of Alexandria and in the source of this piece of information he is defined as the first diorthotes of Homer.5 The term is highly significant and is also confirmed in another source, which states that during the reign of Ptolemy II Philadelphus (285–246 BC) two philologist-poets, Alexander Aetolus and Lycophron, dealt with plays (the former with tragedy, the latter with comedy),6 while Zenodotus dealt with Homer and the other poets. I have deliberately paraphrased the Greek verb with a neutral and imprecise term, dealt with, although in actual fact it is a precise and specific term, diorthoo, namely straightening up, revising, more precisely correcting; it is the verb from which is derived the designation diorthotes, used to characterize Zenodotus, literally corrector. The term that indicated the operation of correcting a text was, naturally, diorthosis, which is indeed used here in connection both with Zenodotus and Aristarchus. As Pfeiffer pointed out, in this regard: “It is not improbable that Zenodotus, examining manuscripts in the library, selected one text of Homer, which seemed to him to be superior to any other one, as his main guide; its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Diorthosis can be the term for either kind of correction. It is hard to imagine any other way”.7 So the first diorthotes of Homer selected a copy he considered to be suitable and worked on it in various ways. On this important point Pfeiffer and K. Nickau, who has produced fundamental studies on Zenodotus,8 are in agreement. H. van  4 On the typology of philological writings, see Dubischar 2015 and Tosi 2015. 5 Suidas, Ζηνόδοτος Ἐφέσιος (ζ 74 Adler). 6 Tzetzes, Prolegomena de comoedia, Prooem. I 1–12, Prooem. II 1–4, 22–39 Koster; Alexander Aet. TrGF 1, 100 T 6 = T 7 Magnelli; Lycophron TrGF 1, 101 T 7; cf. Pfeiffer 1968, 101, 105–106; and Montana 2012. 7 Pfeiffer 1968, 110. 8 Nickau 1972, 30–31: “Dann ist zu fragen, ob Z(enodotos) nicht einen durch Recensio ermittelten Homertext zugrundelegte (der jedoch nicht seinen Vorstellungen von der genuinen Form der Epen entsprach), diesen mit Obeloi versah und zu him Textvorschläge sowie deren Begründung mitteilte. Z(enodotos) selbst wie auch seine Hörer machten sich entsprechende Notizen, die, wären sie von Z(enodotos) schriftlich veröffentlicht worden, ‘Hypomnemata’ hätten heißen können. Aber die Zeit der schriftlich publizierten Homer-Kommentare begann erst mit Aristarchos. So würden sich auch die späteren Unsicherheiten in der Berichterstattung über Z(enodotos) Ausgabe erklären”; see also Nickau 1977.

  Ekdosis: A Product of the Ancient Scholarship Thiel likewise believes the Alexandrian ekdosis consisted of the copy chosen by the grammarian from among those available, and provided with a series of annotations.9 M. West suggests that the particular eccentricity of Zenodotus’ text could not have been due merely to his judgment and opinions, but must in part have reflected an eccentricity of the tradition on which he based himself: he may have worked on a rhapsodic exemplar produced in an Ionian context, which thus reflected a line of tradition differing from that which subsequently became widely accepted and which was predominantly of Attic origin. Thus it may have been an exemplar he had perhaps brought with him to Alexandria from Ephesus, his native city. According to this hypothesis, Zenodotus worked on a Homeric text characterized by idiosyncratic aspects: consequently his Homeric text, resulting from the combination of the base-text plus the annotations in the margins, was necessarily influenced by this circumstance.10 Of course, this is no more than a mere hypothesis, which, however, is based on the same vision with regard to the manner of working of the pioneer of Hellenistic philology: namely, choosing a copy and performing a diorthosis, i.e. carrying out corrections on the copy in question, in order to produce his own ekdosis of Homer. By pondering on these themes over the years, I have come to the conclusion that the problem of the characteristics of the Alexandrian ekdosis can be profitably addressed by starting from its concrete form in terms of its production as a book, on the basis of the following presupposition: in order to understand the nature of what we call a grammarian’s ekdosis of a text and what it contained, it is crucial to examine the way in which it was materially constructed. I have therefore tried to emphasize the importance of the relationship between the bookshop artefact on one hand and the text as an object of ‘philological’ editing, with its various paratextual elements such as annotations and semeia (critical signs), on the other.11 We must take into account and award suitable prominence to what we know regarding the creation of new copies of texts, in the scriptoria by professional scribes or also privately by individuals, along with insights that can be gleaned from surviving examples. To look at the problem in this perspective, the papyri are an essential source of information that cannot

 9 Van Thiel 1991, IX–XIII; van Thiel 1996, 2010 V–VI; van Thiel 1992, discussed by Martin Schmidt 1997, with a reply in van Thiel 1997: see below and n. 68. 10 West 2001a, 39, cf. Montanari 2002a, 123 (= this volume, ch. 47). 11 Montanari 1998a (= this volume, ch. 23); 2000a (= this volume, ch. 38); 2002a; 2004 (= this volume, ch. 41); 2009a (= this volume, ch. 7); 2009b; 2011a (= this volume, ch. 8), and 2015a (= this volume, ch. 9), with extensive bibliography.

Ekdosis: A Product of the Ancient Scholarship  

be disregarded; we will thus start from the papyri to search for data helpful to illuminate these issues. It is an accepted and well documented fact that new examples of literary works were normally re-read and corrected through additional further comparison with the antigraph, at times even on the basis of a collation with other copies. Numerous types of evidence for this can be adduced on the basis of papyrus fragments of literary texts, and papyrologists are fully aware of the phenomenon of corrections introduced in order to improve an exemplar in the framework of book production. Naturally we are particularly interested in the most ancient evidence, although we are hampered by the fact that the papyri datable to the period between the last decades of the 4th and the 3rd century BC (the era of Zenodotus) are very limited in number. This notwithstanding, some small corrections of material errors can already be observed in the two most ancient surviving literary papyri, the Persians of Timotheus (PBerol. inv. 9875) and the renowned Derveni Papyrus,12 dated to the last decades of the 4th century BC (recall the dates of Zenodotus: ca 325–ca 260). Such examples suggest that these corrections were not the result of a systematic revision, but were made by the scribe, perhaps in scribendo. Though not classifiable as a highly striking phenomenon, these occasional corrections of small errors certainly represent the most ancient and visible evidence of a concern for a correct text, or better, of the intention to correct a text in which an error could be perceived.13 A few decades later we already find some considerably richer and more significant witnesses, which I will now briefly summarize. One noteworthy witness is the Homeric roll P.Ilias 12, of which substantial parts are preserved, pertaining to books XXI, XXII and XXIII of the Ilias, dated between 280 and 240 BC, thus still in the Zenodotean era or shortly thereafter, and in any case decidedly pre-Aristarchean.14 This exemplar shows a particularly abundant quantity of corrections performed on the base text, written in a “neat, legible and attractive hand ... A second, rather cursive, hand (m. 2) has in many places corrected mistakes and inserted variants. It is not clear whether the latter all come from a single text, or are a selection from various texts, a kind of primitive apparatus criticus ... A third hand, which it is convenient to refer to as m. 3, although it may be earlier than m. 2, has inserted the reading of the Vulgate, ὦρτο, at Ψ 214, and was probably

 12 Turner/Parsons 1987, 92; text in Kouremenos/Parassoglou/Tsantsanoglou 2006. 13 See Montanari 2009a, 146–147 (= this volume, ch. 7); Montanari 2011a, 3–4 (= this volume, ch. 8). 14 P.Heid.Lit. 2 (inv. 1262–1266) + P.Hib. 1.22 (Bodl.Libr. inv. Ms.Gr.Class.b3(P)/2) + P.Grenf. 2.4 (Bodl.Libr. inv. Ms.Gr.Class.b.3(P)) = MP3 979: cf. S. West 1967, 136–191; Sforza 2000.

  Ekdosis: A Product of the Ancient Scholarship responsible for part of the double attempt to correct Βορεαι at Ψ 195 ... In several places there are marginal signs. Points are prefixed to ... The remaining marginal signs are even more puzzling ...”.15 Given its chronological position, the fact of having at least 21 plus-verses in the preserved parts16 is not particularly strange, but what interests us here is above all the conspicuous quantity of variants and marginal signs. According to the analysis of S. West, the range of signs in the text is extremely varied. Sometimes the second hand inserts readings of the vulgata17 in passages where the base text is different, whereas on other occasions it inserts readings that differ from the vulgata, although the latter is actually in agreement with the base text; in yet other places neither the base text nor the readings inserted by the second hand agree with the vulgata; the third hand has inserted a reading taken from the vulgata in a passage where the base text differs from the latter.18 The marginal signs are numerous, if one takes into account that the left-hand margin of the columns is often lost, and are considerably problematic. In five or six places the line is marked by a dot, a sign that always gives rise to problems of interpretation with regard to whether it should be considered as having stichometric value or as a critical sign, and even in the latter case its meaning is doubtful.19 At the side of l. 23.157 “there are traces of a sign rather like a diple”:20 if it were genuinely a diple, then one would have to raise the question of when the sign was marked on the papyrus. That is to say, one would have to endeavor to ascertain whether or to what extent it dates from a time later than the base text, given that it is normally believed that the diple was introduced by Aristarchus21 (born around 215 BC), and in effect a sch. of Aristonicus to this line provides information on an Aristarchean observation (and draws a parallel with 2.278).22 This papyrus is a  15 S. West 1967, 136–137. 16 It is well known that the witnesses (both direct and indirect) of the Homeric text which date from the early Hellenistic age (roughly up to the 2nd c. BC) show the presence of additional lines as compared to the numerus versuum fixed at a later stage, which corresponds to that of the modern editions: cf. further on and Haslam 1997 for an effective overview. 17 The so-called vulgata can be defined as the Homeric text that prevailed in the transmission, cf. shortly below. 18 S. West 1967, 137. 19 S. West 1967, 133, 137; cfr. Pfeiffer 1968, 218; Nickau 1977, 261; McNamee 1992, 9 nn. 5–6, 15 n. 31, Table 1 p. 28, Table 2 p. 38, Table 3 p. 43; Montanari 1998a, 16 (= this volume, ch. 23), and Montanari 2012c, 28–29, with additional bibliography. 20 S. West 1967, 138. The usual form of the diple is >. 21 Pfeiffer 1968, 218. 22 The observation concerns the use of the verb in the plural with a singular but collective subject (σχῆμα πρὸς τὸ νοητόν), as is also found in Il. 2.278: cf. Matthaios 1999, 384.

Ekdosis: A Product of the Ancient Scholarship  

witness that should certainly be the object of an in-depth re-examination from all points of view, including from the perspective of paleography, above all to determine the time gap between the base text and the subsequent interventions. Of a slightly more recent date, but equally significant, is P.Odyssey 31, dated to between 250 and 200 BC, which contains parts of books IX and X of the Odyssey.23 This too has been examined by S. West: “The text had undergone a double process of correction and collation. The original scribe appears to have had two MSS. at his disposal, and in several places he has cancelled readings correct in themselves in favour of readings which are no better and are sometimes obviously worse. Presumably he had more faith in the MS. which he used in order to correct than in that from which he originally copied it ... Corrections have also been inserted by a second hand, which can usually be distinguished without difficulty from that of the original scribe. The readings inserted by m. 2 agree with the Vulgate, but in several places where the text diverges considerably from the Vulgate there is no trace of a correction. There are no marked affinities between this text and that of any of the Alexandrian critics”. There are plusverses (they are quantitatively fewer as compared to the previous exemplar) and marginal signs, most of which are probably of a stichometric character (but the left-hand margin is often lost). The most important fact is that the roll underwent a twofold process of collation and correction: the first scribe would seem to have had two exemplars available, and he often corrected his text on the basis of another manuscript, after which a second hand inserted readings that are in agreement with the vulgata. These are witnesses of great importance for the question we are examining here: we have two Homeric exemplars from the mid-3rd century BC, therefore definitely and decidedly pre-Aristarchean, which in addition to various kinds of — often somewhat problematic — critical signs and the expected plus-verses, also show rather clear evidence of collation with other copies and a conspicuous number of interventions performed on the base-text at various times and in various different ways.24 M. Haslam says: “Our earliest Homeric manuscripts, those of the 3rd cent. B.C., are characterized by their startling degree of difference from the text that prevailed later, sometimes known as the ‘vulgate’ ... We now have fragments of about forty Homer manuscripts written c. 150 B.C. or earlier ... Several of these early manuscripts give evidence of having been collated with

 23 P.Sorbonne inv. 2245 A = MP3 1081: cf. S. West 1967, 223–224. 24 Cf. Haslam 1997, 64–66; Rengakos 2012, 241–242.

  Ekdosis: A Product of the Ancient Scholarship another exemplar25 (so ‘wild’ is hardly the word for them) and sometimes reveal that ‘vulgate’ readings coexisted alongside ‘eccentric’ ones ... Of the variants entered from a second exemplar in the most extensive of the early Iliad papyri (P 12 [scil. P.Heid.Lit. 2 + P.Hib. 1.22 + P.Grenf. 2.4]) four coincide with the vulgate ... A similar picture is presented by the most extensive of the early Odyssey texts, P 31 [scil. P.Sorbonne inv. 2245 A] ... The Homer of readers in the 3rd and early 2nd century ... was appreciably more flaccid than the Homer of subsequent readers”.26 This was the situation that Zenodotus and his earliest successors found themselves facing.27 From the 3rd century BC we have the Milan papyrus with epigrams by Posidippus, P.Univ.Milan. 309, another important piece of evidence in view of the quantity of corrections and annotations the text presents.28 The majority of the corrections were made by the same scribe, clearly in scribendo (in general amounting to one and never more than three letters and all aimed at correcting minor slips in the drafting stage), but subsequently, after the copyist’s corrections, two other hands intervened with further emendations and the differences in approach should be recognised.29 The third person to make changes to the text in col. XI recorded a variant on the reading of l. 30, noting it in the upper margin: at col. XI 30 we can read κεντρακαιεξω[ and in the upper margin it is written καικεντρα (the last three letters are not visible in the photograph but can be seen in the original document).30 It is extremely likely that this is a correction or a variant, probably for the κέντρα καί of the text, an inversion — καὶ κέντρα — has been proposed, but it is not clear, owing also to the fact that the rest of the verse has not been preserved.31 The papyrus fragments of the following centuries, and in particular of the first three centuries of our era (the era with regard to which the papyrus findings are most abundant) provide us with rich and valuable documentation of exemplars with interventions of deletion, addition and corrections of all types. The

 25 And consequently they present interventions consisting of corrections and annotations of variants. 26 Haslam 1997, 64–66. 27 Haslam 1997, 64–66; on the references to the ekdoseis in the papyri of Homeric exegesis, see Pagani/Perrone 2012. 28 P.Univ.Milan. 309: Bastianini/Gallazzi 2001, 15, 76–77; Montanari 2009a, 147 (= this volume, ch. 7); Montanari 2011, 4–5. 29 Bastianini/Gallazzi 2001, 15. 30 Bastianini/Gallazzi 2001, 76–77. 31 Bastianini/Gallazzi 2001, ad loc.

Ekdosis: A Product of the Ancient Scholarship  

following significant examples will suffice for our purposes;32 they could be easily increased with fragments from various periods. POxy. 2161, of the 2nd century AD, contains Aeschylus’ Diktyoulkoi. The scribe has occasionally corrected some of his own errors: for instance, in l. 831 he wrote ηδη, but then crossed this out with an oblique line through each letter, writing supra lineam the correct reading ο]ιον. PBerol. inv. 9872 (BKT II), of the 2nd century AD, is a long papyrus roll (75 columns plus various fragments) that contains a commentary on Plato’s Theaetetus with a substantial number of corrections. The most recent editors of the roll, G. Bastianini and D.N. Sedley, write: “The volumen has been proof read and corrected in many places: letters or words omitted in the original drafting stage have been restored, superfluous letters or words have been cancelled, letters judged to be mistakes have been replaced by those considered correct. All these changes do not appear necessarily to presuppose a collation with an exemplar different from that of the copy ... The variety of ways the corrections have been made may lead one to suspect that the roll had been corrected on various occasions: the first hand (a diorthotès in the scriptorium) added the missing words, which are marked in the upper margin ... or lower ... or are placed after the line directly in the intercolumnium ... A later hand or perhaps hands, appears to have gone through the whole text, cancelling with a line in ink all the letters judged to be wrong”.33 For example, at col. LXIII, l. 6 the scribe had written προσαλλαουτεσχη, omitting some words. In the intercolumnium to the left, the corrector has put the sign of an upwards-pointing ancora (something similar to an arrow) and in the space between αλλα and ουτε has written ἄνω; in the upper margin, one can read the words θεωρειται ουτε γαρ χρω|μα κα(τω), which were probably preceded by a downwards-pointing ancora now lost in lacuna. The corrected text is therefore πρὸς ἄλλα θεωρεῖται οὔτε γὰρ χρῶμα οὔτε σχῆ|μα. POxy. 2256, of the 2nd–3rd centuries AD, contains hypotheseis of various tragedies by Aeschylus. The fragmentary hypothesis of fr. 3 recalls the victory, with the trilogy of which the Danaids was a part, against Sophocles and another author, probably Mesatos (l. 5). After the name of the latter and at the beginning of the following l. 6, round brackets can be clearly seen, which are generally used as a sign to indicate expunction in literary texts and non-literary documents. It is clear here that the round brackets were placed in scribendo, which can be explained solely by imagining that the scribe copied from an exemplar

 32 Montanari 2009a; 2009c (= this volume, ch. 60); 2011. 33 Bastianini/Sedley 1995, 243–244.

  Ekdosis: A Product of the Ancient Scholarship where the expunctions were already present to indicate that the plays placed between brackets had been mistakenly placed after the name of Mesatos.34 The copy of the Gospel according to St. John contained in PBodmer 2 dates to the 3rd century AD. The scribe has corrected the text in a variety of ways. There are supra lineam additions (ll. 2 and 12) and words rewritten above parts of the text cancelled with a sponge: in ll. 9–10 εταραχθη has been written over a word that has been scrubbed out and which continued in the following line, where the letters σατο can be made out in the remaining space; the second part of l. 10 has been rewritten; at the beginning of l. 11 τον is the remains of an eliminated reading, subsequently punctuated with dots as well as small round brackets supra lineam. I turn now to a manuscript which, I believe, provides us with what can be termed an anthology of the techniques and methods available for correcting and improving a text: POxy. 2404 + PLaur. inv. III/278, a fragment of a papyrus roll (late 2nd–early 3rd century AD) containing a part of 51–53 (POxy. 2404) and of 162–163 (PLaur. III/278) of Aeschines’ oration Against Ctesiphon.35 It seems quite evident that this copy of the oration of Aeschines has been collated with a second exemplar and has been the object of detailed and systematic correction seeking to identify the textual structure by distinguishing cola and periods and to correct copying errors for the benefit of the reader, and to emend the text in places judged unsatisfactory, by means of various different methods of deletion and by writing the alternative readings above or beside the wrong interpretations.36 We can see that the work of proof reading was not limited solely to correcting minor errors as discretely as possible in order to reduce the possibly negative impact of emendations on the appearance of the text. In fact, more evident corrections, albeit written with care and precision, have been made, with the apparent aim of improving the text and enabling it to be read according to the intention of the corrector or correctors. The methods of corrections and cancellation used in this papyrus are: the use of dots above a letter, an oblique (single or double) line through a letter in question and, for longer sequences, a line above or through the letters to be deleted, or by a combination of these methods. We have also seen the widespread practice of simply writing the cor-

 34 Arata/Bastianini/Montanari 2004, 39, 47–48. 35 Editio princeps of POxy. 2404: Turner 1957; cf. Turner 1980, Pl. VIII and 212; editio princeps of PLaur. inv. III/278: Messeri Savorelli/Pintaudi 1997, 172–174; see also Neri 2003, 511–514; Esposito 2004, 3–4; Colomo 2008, passim. 36 Detailed analysis in Montanari 2009c with bibliography (= this volume, ch. 60); on related problems see above all Turner 1980, 92–93.

Ekdosis: A Product of the Ancient Scholarship  

rect letters above those judged incorrect as way of indicating a deletion, as it were, ‘automatically’ without the need for other material indications. Another form of correction is the addition of words between the lines or in the margins. As regards punctuation, the scribe provided the text only with paragraphoi, whilst copious punctuation was added (at least, so it is thought) by a later hand.37 Most of these are dots, placed slightly higher than the letters, which had already been written, making sure that the dots were not above a letter but in the narrow space between the end of the preceding word and the beginning of the next. A lower dot can also be seen at col. I, l. 17. The system can be described as follows: the upper dot combined with the paragraphos marks the end of a sentence; the upper dot on its own distinguishes the cola of the sentence; the lower dot indicates a weaker pause.38 This constitutes proof of a serious attempt to highlight the syntactic and rhetorical structure of the text, leading us to consider the role of punctuation in Alexandrian philological exegesis (rather than the complex and idiosyncratic system created by Nicanor, one can mention the simpler and more widely-used system of the three stigmai of Dionysius Thrax).39 I wish to emphasize at this stage why I have drawn attention to these manuscripts and their characteristics, with a choice of significant examples, to which others could easily be added.40 The point is not that they may be considered as exemplars of a grammarian’s ekdosis: there is absolutely no evidence for such a suggestion. Rather, in my view they are of value because they highlight the importance of the techniques adopted in the workshop for book production and the effect such craftsmanship had on the development of a philological practice that sought to ameliorate and emend texts regarded as unsatisfactory due to the (real or supposed?) errors they contained. The papyri provide ample evidence of the different methods used to ‘improve’ an exemplar of a book, in other words to correct the (new) copy of a text. It was considered appropriate to add, remove or modify letters or words that had been omitted or written erroneously, or cancel what was regarded as erroneous and replace it with what was judged to be correct by writing the correction above the line, in the margins and in the intercolumnia,  37 Turner 1957, 130: “The second hand not only revised the text for errors but collated its readings with an exemplar different from that from which it was copied”; cf. Turner 1980, 212; Colomo 2008, 15–16, 24. On punctuation marks in papyri, see Turner 1980, 92–93; Turner/ Parsons 1987, 9–10. 38 Colomo 2008, 15–16. 39 Dion. T., Ars Gram. § 4; cf. Colomo 2008, 15–22; Montana 2006b. For the punctuation system employed by Dionysius Thrax and Nicanor, see also Montana 2015, Matthaios 2015, Dickey 2015, and Valente 2015. 40 Other useful material can be found in S. West 1967, passim and Haslam 1997, 63–69.

  Ekdosis: A Product of the Ancient Scholarship at times with specific markings to indicate the position referred to. Sometimes the forms presumed to be correct were introduced in replacement of the previous words once these had been materially eliminated, or at times by simply writing the correct letters or words between the lines or in the interlinear space above the form judged to be incorrect, as a way of indicating, as it were, an ‘automatic’ deletion without the need for other material indications. On occasion, a horizontal or oblique line could be drawn through the letters or words to be deleted; another method was to mark these letters or words by dots or lines above or below or enclosed within round brackets, or even to erase them with a sponge.41 Thus there was a veritable ‘tool kit’ for diorthosis. Often the interventions were carried out by the diorthotes of the scriptorium, whose task within the atelier was to re-read and correct the text, if necessary also by comparing the copy with the model, in other words through a practice of collation. G. Turner and P.J. Parsons write: “One of the questions the palaeographer should ask about any literary manuscript is whether it has been adequately compared against its antigraph (the exemplar from which it was copied), a task which, in a publishing house, was the duty of the diorthotes, corrector, or whether it has been collated with a second exemplar (a procedure often carried out by private individuals to secure a reliable text)” ... “But several of our surviving papyrus manuscripts, and especially those which are beautifully written, contain such serious un-noted errors that it is clear their ‘proof-reading’ was of a summary, superficial kind, if done at all ... Those ancients themselves who set store by having a dependable copy (persons like Strabo and Galen) were aware of this weakness and adopted a routine to counter it: they themselves (or their secretaries) checked the copy to be used against another exemplar. If, therefore, the text had been checked against its first exemplar, and was later collated with a second, it may well bear the marks of this double checking”.42 Best practice in the book production consisted in a comparison between copies and corrections of mistakes, carried out by a professional or occasional diorthotes, who had adequate resources for deleting, adding, replacing and marking various aspects and features of the text in order to improve it and increase its reliability. Even a private copy could be subjected to the same kind of treatment, with the use of the same tools and procedures, for personal reasons springing from the cultural or research interests of the owner. Analogies with what we understand by philological practice are evident and need to be  41 Turner/Parsons 1987, 15–16, with reference to examples in plates; see also Turner 1980, 93 and Pl. VIII; Bastianini 2001. 42 Turner/Parsons 1987, 15–16; Turner 1980, 93 and Pl. VIII.

Ekdosis: A Product of the Ancient Scholarship  

stressed, as the methods and techniques adopted in handicraft book production honed the skills that would gradually be developed and applied by grammarians. Little by little, a procedure that probably did not appear particularly strange or extravagant among those for whom use of books was an everyday practice developed into an extraordinarily innovative principle: the diorthosis of the diorthotes of the scriptorium became the diorthosis of the philologist, diorthotes not of an individual copy of Homer but diorthotes of Homer. Effectively, concerns and emendments of a specifically book production and commercial nature became those of a critical and philological-grammatical nature.43 The aim pursued by the corrector of a publishing house was to produce an exemplar that would represent the best possible workmanship, a good copy suitable for sale on the book market or to a client, perhaps intended to be the personal copy of a scholar or an educated man, who did his own corrections and annotations (we will note the case of Galen). In contrast, the grammarian’s underlying objective in correcting the text of his personal copy was more ambitious, because he sought to find the true and proper form of the work he was dealing with. He worked on a copy with the aim and intention of achieving, as it were, the model exemplar, which would display what in his view was the genuine form of the literary work in question. This conception led to the possibility of indicating doubts or a textual aporia, a perspective that certainly did not belong to the mental system and operational horizon of the craftsmen of the scriptorium. Thus with Zenodotus, drastic and univocal deletion (a typical action of the craftsman in book production, meaning “don’t write these words in the new copy”) for the first time was accompanied by the sign of philological uncertainty, namely the obelos, a simple horizontal stroke on the left of the line. This marked a decisive intellectual change: attention began to focus on the work in its own right rather than merely on perfecting the individual copy. It is vital not to underestimate or downplay the invention of this critical sign, which had a momentous impact because it could also be applied systematically to poems of the great length and cultural importance of the Iliad and the Odyssey. By means of his simple semeion Zenodotus was able to indicate his suspicion that a given line might not be a genuine line of Homer, but that he was not sufficiently sure of his judgment to be able to proceed with clear-cut and definitive deletion of the element in question. Later, the progress of the discipline gradually increased, with further development of the system of critical signs and the markings of exegetic reflection and erudite comment. By the time of Aristarchus the

 43 Cf. Nickau 1977, 10–11.

  Ekdosis: A Product of the Ancient Scholarship system of semeia had become complex and refined, but it had all begun with Zenodotus’ small obelos and its radically new meaning for a reader of his texts.44 I believe that the philological work of the Alexandrian grammarians, starting from the first generation, represented something new in cultural history and marked significant intellectual progress. The reality and significance of this ‘revolution’ becomes more evident and tangible if we succeed in grasping a fundamental chain of circumstances: namely, the aspects and procedures of book production had moulded a material and, in a sense, ‘craft-oriented’ base of tools and working procedures that were subsequently adopted and utilized by grammarians for quite different purposes and in a different perspective. Thus the tools and methods of book production became the tools and methods of scholarship by virtue of an innovative and decisive intellectual change, which signalled a transition from the aims of pure craftsmanship, i.e. from correcting an individual copy in the scriptorium and thereby creating a good product, to an intellectual aim of a ‘philological’ nature, namely producing the exemplar that would contain what was held to be the correct text of the work. Thus no longer would the copy be an exemplar of the work: rather, it would be the text of the work in itself, and this implied a sharp difference between “correcting a (single) copy” and “editing a text itself”.45 Let us recapitulate. Zenodotus worked on an exemplar of Homer that was available to him and which he deliberately chose for the specific purpose of producing his ekdosis. However, he had more than a few reservations about it, concerning both the numerus versuum and also a certain quantity of readings. He had doubts about the authenticity of some lines, and adopted a sign indicating his suggestion that the line should be expunged, the obelos, which he marked in besides the lines: this was athetein, the athetesis. But it has always been more difficult to determine how he proceeded with lines he believed should most certainly be deleted from the text as definitely spurious and really to be rejected. Normally, such lines would have been present in his base-text (as were those for which he proposed the athetesis by means of the obelos). For such cases of deletion, in the scholia one finds the expressions ou graphein (do not write: the most frequent), ouk einai (is not there), ou pheresthai (is not handed down) and a few others.46 The task of reconstructing exactly and concretely what was the difference between the operations indicated by this terminology has always been problematic. A good idea of what was meant can  44 For the evolution of the system of semeia see also Montana 2015. 45 Montanari 2011. 46 Cf. Ludwich 1884–1885, II 132–135; Nickau 1977, 1–30.

Ekdosis: A Product of the Ancient Scholarship  

be gained by examining more specifically the practices used in literary papyri for different modes of deleting something that is present in a text. Thus Zenodotus, as we have seen, marked some lines with the obelos on his copy, but he also used one or other of the graphic methods mentioned above for lines that were clearly intended to be deleted from the base-text. But the grammarian could equally well jot down another similar or equivalent term for line deletion;47 alternatively, he could rely purely on a deletion sign without verbal annotations, in which case the terminology for line deletion may have been noted down by those who followed his teachings at the Museum, or it may have been created, modified or extended by the subsequent tradition (this is conceivable above all for ouk einai and ou pheresthai).48 I would argue this is the most plausible explanation for the different terminology used in the scholia for Zenodotus’ text alterations of athetesis and line omission, including the problem of the material difference in the copies between athetised and deleted lines.49 A somewhat skeptical attitude towards this vision has recently been expressed by A. Rengakos, who puts forward some objections. 1) The expressions in the scholia in reference to the lines deleted by Zenodotus more clearly describe the situation of lines genuinely not included and therefore absent; 2) it is strange that Zenodotus invented a critical sign for athetesis, namely the obelos, whereas there is no trace of a sign of the same type for deletion, even though it has a more radical impact on the text; 3) with regard to passages where Zenodotus eliminates some lines but information concerning his interventions on the deleted lines is preserved, Rengakos believes there is nothing to confute the idea that those lines were effectively absent in his text. Thus the Zenodotean ekdosis may have been a veritable continuous text, with the obeloi and the variants and without the deleted lines.50 My opinion is that the variability of expressions used in  47 The verbs perigrapho and diagrapho are technical terms for ‘deleting’ using the material means mentioned: cf. Turner/Parsons 1987, 16; some examples have remained in the scholia: for perigrapho cf. Nickau 1977, 10–12 and 29. 48 One may mention in this regard the diadoche of ‘Zenodotean’ grammarians testified by Suda (π 3035 Adler); their last representative was Ptolemaios Epithetes, who worked on the text of Zenodotus and entered into a polemical argument with Aristarchus: see Montanari 1988. 49 In Montanari 1998a (= this volume, ch. 23); 2000a (= this volume, ch. 38); 2002a (= this volume, ch. 47); 2002b (= this volume, ch. 58) I discussed some possible objections to this reconstruction, which I will not repeat here. 50 Two cases that have given rise to particularly extensive discussion are Il. 2.111–118 and Il. 2.156–169: the scholia (Aristonicus and Didymus) say that Zenodotus deleted the lines for a shortened version of the passage, but on the other hand they preserve text interventions by Zenodotus himself on these very lines (in actual fact, the intervention on the first passage strikes me as debatable, and calls for further clarification): cf. Rengakos 2012, 250–252. I see no

  Ekdosis: A Product of the Ancient Scholarship the scholia to indicate line deletion suggests that, at least in part, these were devices developed (perhaps much) later within the tradition to put into written form what were predominantly (with the possible exception of ou graphein) the material modes of deletion used in book production, as described above. These devices were the signs for line deletion: there was no need to invent any ex novo, because they were well known and available in the practice. But the real innovation was the athetesis, which did indeed call for a special sign. I still find it difficult to imagine that Zenodotus began his task of diorthosis by writing a continuous and definitive text: to do so, he would have had to begin once he had already made all the decisions, with poor opportunity for second thoughts and new corrections, and the definition of diorthosis/ diorthotes does not point in this direction. However, it is not unlikely that the paratextual apparatus on the working copy may have given rise to problems of comprehension and readability, especially with the accumulation of interventions over time, and in places where the multiple interventions on the text became interlaced with one another. The copy bearing the work of diorthosis resulted materially in the philologist’s own ekdosis, and we can imagine this as a product of years of study that led over time to a series of interventions on the same exemplar. Together with critical semeia, explanatory annotations must have been present in the working copy starting from Zenodotus onwards, and are likely to have continued to be used by grammarians in their editorial and exegetic work. I therefore feel it is far more plausible to assume that the ekdosis became ekdotheisa, i.e. ‘published’ and therefore available for consultation by scholars, poets and intellectuals, as soon as the grammarian himself, or someone working on his behalf, had had a copy made that followed the indications in the base-text on which diorthosis had been performed, so as to create an exemplar that was a correct and ‘fair copy’ of the work,51 but still bearing the name of the grammarian who was the author of the copied diorthosis, with the marginal annotations which would still be necessary once the text had been properly prepared. In short, first there was a working copy belonging to the diorthotes, with all his interventions and annotations, after which it was possible to proceed to reproducing it as a ‘fair copy’ of his

 difficulty in thinking that with regard to the lines in question Zenodotus’ copy had one of the well-known deletion signs and that a possible variant for one of the deleted lines was indicated in the margin. See below and n. 85. 51 Perhaps ou graphein, which is the most frequent expression for line elimination, may go back to Zenodotus himself and may have been an indication to whoever transcribed his ekdosis that the element in question was not to be copied.

Ekdosis: A Product of the Ancient Scholarship  

ekdosis. Thus it was a stepwise production, which we should obviously imagine to have been done not only for Zenodotus but also for all his successors. This can also explain the conservation and transmission of the interventions and textual choices made by the grammarians.52 The material form of the ekdosis after Zenodotus must have remained very similar: a grammarian chose, according his own preferences, an exemplar that he considered suitable as a basis of his work. But Zenodotus’ choice of the basetext of Homer seemed highly debatable and was open to criticism, which is why Aristophanes and Aristarchus chose exemplars with noticeably different characteristics.53 Consequently, a line of tradition predominantly of Attic origin gradually spread, partly by virtue of the base-text of working copies used by grammarians active in a later period than Zenodotus. The latter base-text proved decisive above all as regards the numerus versuum, whereas the readings suggested by individual grammarians generally did not become standard in the vulgata. The plus-verses present in the Zenodotean text (as well as in several pre-aristarchean copies) were not his own interpolations but were instead typical of exemplars that were current in his day:54 they disappeared because the work of Aristarchus led to general recognition of a text that had a very similar number of lines to our vulgata. The Aristarchean numerus versuum, which became the standard of the vulgata, was essentially the outcome of the particular working copy selected by Aristophanes and, finally, by Aristarchus. It is significant that Aristophanes did not go so far as to carry out the drastic act of line deletion: in other words, it is significant that he abandoned the use of performing material cancellations on his own copy with the graphic techniques mentioned above. The obelos became the prime tool for expressing a cautious doubt concerning parts of the text; ou graphein disappeared, leaving only athetein.55 Aristarchus followed exactly the same procedure. This explains why many of the lines Zenodotus had decided to eliminate from the Homeric text once and for all — but which were present in the copies chosen by later grammarians — were preserved in the numerus versuum that became the generally accepted tradition after the Aristarchean age and thus remained in our vulgata.56 The  52 Helpful confirmation comes from a testimony by Galen, quoted a little further on. 53 Montanari 2002a, 123–125 (= this volume, ch. 47); West 2001a, 36: “Clearly Aristophanes and Aristarchus were not dependent on Zenodotus’ text but followed another source or sources more similar to the vulgate”; cf. West 2002, 138. 54 Haslam 1997; West 2001a, 40; cf. above. 55 Or else, if genuine deletions were still carried out, they were of such minor relevance that all knowledge of them was lost: this is possible, but not demonstrated. 56 Haslam 1997, 85; West 1998–2000, vol. I, p. VII; Führer/Schmidt 2001, 7.

  Ekdosis: A Product of the Ancient Scholarship abandonment of the drastic practice of material deletion highlights the increasing sense of caution that had developed in the meantime, and accounts for the fact that many of the lines ‘deleted’ by Zenodotus were in effect no longer deleted57 and thus were not obliterated from the tradition. The work of Aristarchus marked the period in which Alexandrian philological production included the drafting of extensive hypomnemata. The great continuous commentary, which followed the text step by step, notably facilitated and enriched the communication and preservation of the arguments and motivations put forward by the grammarians, so that the material which has come down to us from this tradition is much more substantial.58 Yet the ekdosis as an annotated working copy by no means went out of use, as clearly testified by the information on the Aristarchean edition(s). On the other hand, the possibility of dwelling at length in the hypomnema on arguments pertaining to text criticism and exegesis constituted an important resource. In practice, the need to write on the copy chosen as the base-text was no longer so strongly felt, especially as regards philological-exegetic arguments. Previously, before the rise of separate hypomnemata, there had been a greater need to write on the actual text of the working exemplar, but with Aristarchus the particularly elaborate system of critical signs placed next to the lines59 as well as the variants and the readings to be adopted must have been present in the margins and interlinear spaces, while the philological-exegetic treatment was mostly developed in the commentary, although marginal annotations continued to be utilized whenever they were felt to be of practical use, e.g. for short notes and textual proposals. The number of Aristarchus’ ekdoseis of Homer and their philological and chronological relation to the commentary or commentaries is still a subject of dispute. I will not go over the entire background here, nor report the treatment already given elsewhere: I will restrict myself to summarising the results, in order to set them within the framework that is being delineated. On the one hand we have the frequent unequivocal references to the plural for Aristarchus’ editorial work on Homer: αἱ Ἀριστάρχου (scil. ἐκδόσεις or διορθώσεις), ἡ ἑτέρα τῶν Ἀριστάρχου and similar. On the other, the titles of two works by the grammarian Ammonius, the direct successor of Aristarchus: Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως and Περὶ τῆς ἐπεκδοθείσης

 57 It is sometimes stated, instead, that they were ‘recovered’ or ‘reintroduced’: this would have involved far more complicated operations. 58 For hypomnema features, see Montana 2015 and Dubischar 2015. 59 They could also be repeated in the hypomnema beside the lemmas, as was the case for instance in P.Oxy. 1086 (pap. II Erbse).

Ekdosis: A Product of the Ancient Scholarship  

διορθώσεως, the former in apparent contradiction with the latter and with the scholiastic citations that indicate two editions. Nevertheless, the solutions proposed go as far as to hypothesize a ἐπεκδοθεῖσα διόρθωσις carried out by Aristarchus’ immediate pupils, probably by his successor Ammonius, who in any case was familiar with it.60 As regards the hypomnemata, I think it is difficult to deny that Aristarchus made two successive versions: a first commentary based on the ekdosis by Aristophanes of Byzantium is explicitly cited in sch. Il. 2.133 a: ἐν τοῖς κατ᾽ Ἀριστοφάνην ὑπομνήμασιν Ἀριστάρχου. In contrast to this stands the citation of perfected (ἠκριβωμένα) hypomnemata in sch. Il. 2.111 b: it is perfectly plausible to assume that Aristarchus produced a second version of the hypomnemata in which he took into account the progress achieved over time by his work as a Homeric philologist.61 We have two Homeric passages on which Aristarchus is known to have changed his mind in comparison to his first text choice, as reported in the scholia to Il. 10.397–399 and 19.365–368. Such a situation has many parallels in the scholiographic documentation, describing Aristarchean second thoughts and changes of heart.62 However, in the case of these two Homeric passages, it can confidently be stated (despite uncertainties about details) that later philologists were searching for information on the reasons and circumstances for his change of mind and on the text situation that had ensued, and since they were far from certain, they consulted the cited works by Ammonius. In the attempt to explain this situation and reconcile the apparent contradiction between the two titles of the successor in the school, it has been suggested that Aristarchus himself personally composed only one ekdosis, as suggested by the first title, and that the ἐπεκδοθεῖσα διόρθωσις mentioned in the second one was actually composed later, after the master’s death, in the circle of his first and most senior pupils, possibly even by Ammonius or else by other followers. However, I believe that a slightly different hypothesis can lead to a better understanding of what really happened. The paradox that there were no more editions of the Aristarchean διόρθωσις and that this work itself was issued again (ἐπεκδοθεῖσα) can only be resolved, in my view, by supposing that the ἐκδοθεῖσα διόρθωσις and the ἐπεκδοθεῖσα διόρθωσις were in a sense truly (physically) the same thing, but modified, revised,  60 On this topic, cf. Montana 2015. 61 Allow me to take this opportunity to refer to Montanari 1997, 285–286 (= this volume, ch. 42), on the question of the famous Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν. 62 Another two examples are examined in Montanari 2000a (= this volume, ch. 38).

  Ekdosis: A Product of the Ancient Scholarship corrected and further worked on. So let us briefly trace out the entire story. Aristarchus began by writing hypomnemata based on Aristophanes’ text; then he devoted himself to his διόρθωσις and produced his own ekdosis; at this point he then wrote new hypomnemata, based on his own ekdosis, presenting what he felt was a more careful and refined analysis. But the research and reflection of a philologist like Aristarchus on a text like Homer was unlikely to come to a definitive conclusion, for not only did Aristarchus continue to meditate and study, but he also continued to teach, and Homer was constantly in his hands. And so he would resume work on his διόρθωσις, pick up his ekdosis again and jot down his second thoughts and new ideas on the text: thus the very copy that had been the ἐκδοθεῖσα διόρθωσις, then became the ἐπεκδοθεῖσα διόρθωσις. And since he wrote no further new hypomnemata, he could not write at length on the reasons for changing his mind and explain what stage his thought had reached, and so on some points there was and there is considerable uncertainty as to the final stage of his thought.63 Thus the assumption that there existed a single exemplar of Homer, bearing successive layers of alterations and jottings, provides a perfect explanation for the fact that the tradition commonly spoke of αἱ Ἀριστάρχου, ἡ ἑτέρα τῶν Ἀριστάρχου, διχῶς Ἀρίσταρχος (cf. particularly Didymus in sch. Il. 2.517 a) and so forth. In a certain sense, one could indeed speak of two editions, i.e. one and the same exemplar displayed a dual set-up: in the majority of cases the first and second version must both have been legible and distinguishable, thereby permitting the subsequent philological tradition to be familiar with them and discuss them both. But if new alterations were made on an already densely annotated copy, it is hardly surprising that uncertainties may have arisen, where for some reason (at times possibly due to material conditions) it may not have been clear which of the two types of text set-up represented the master’s final position, which was his ultimate decision and whether indeed he had actually come to any definitive conclusion. Ammonius certainly had perfectly valid reasons for explaining that there was effectively one and only one ekdosis, and that at a certain point it had been ‘re-issued’ with changes: in fact it is quite likely that the titles Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως and Περὶ τῆς ἐπεκδοθείσης διορθώσεως referred either to two parts of one and the same work or to two very closely linked works, which were designed to explain  63 See Montanari 1998a (= this volume, ch. 23); 2000a (= this volume, ch. 38) and 2002a (= this volume, ch. 47): this solution of the problem ekdosis / ekdoseis of Aristarchus is approved by Slings 1999; Nagy 2003 and 2009, 21–37; Rengakos 2012, 252; I am unsure whether it was perceived by West 2001a, 61–67.

Ekdosis: A Product of the Ancient Scholarship  

what had really happened and above all to clarify the doubtful points. In some passages the master’s final decision was not clear: perhaps because it had not been annotated properly, or because there was physical damage at that particular point, or perhaps because his reflections had been interrupted by unforeseen circumstances. In any case, Didymus evidently felt that not everything was perfectly clear, because he consulted Ammonius on doubtful points.64

Conjectures and/or Variae Lectiones Our reconstruction of the manner of carrying out the Alexandrian ekdosis, based on observation of well documented technical facts, helps to clarify, on a more solid basis than usual, the problem of the real nature of the readings attributed by the erudite tradition to the Alexandrian grammarians such as Zenodotus, Aristophanes of Byzantium and Aristarchus: were they conjectures ope ingenii based solely on subjective criteria, deliberate choice among variants attested by documentary sources and deriving from the collation of copies, or a combination of both? Was there a practice of comparing a variety of exemplars of the Homeric text to spot the differences and thereby offer an opportunity for choice? This is a problem of fundamental importance — indeed, it is one of the most disputed issues in the history of ancient philology (not only for the history of the Homeric text in antiquity) — inasmuch as these questions are crucial in an evaluation of the work of the Alexandrian philologists and their role in intellectual and cultural history (see also Montana in this volume). Let us now take another look at the passage by Pfeiffer cited at the beginning: “It is not improbable that Zenodotus, examining manuscripts in the library, selected one text of Homer, which seemed to him to be superior to any other at hand, as his main guide; its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Diorthosis can be the term for either kind of correction. It is hard to imagine any other way”.65 It is indeed almost impossible to imagine any other way, which means that the Alexandrian philologists’ production of an ekdosis must have involved both conjectural emendations and choice among variants detected through the

 64 Pfeiffer 1968, 217: “Whether Didymus was able to work on copies of these originals διορθώσεις and ὑπομνήματα of Aristarchus and of his monographs, the συγγράμματα, is an insoluble problem”; cf. West 2001a, 61–67. 65 Pfeiffer 1968, 110.

  Ekdosis: A Product of the Ancient Scholarship collation of copies: “Zenodotus’ text is shown to be based on documentary evidence”.66 My own position, already expressed and argued several times elsewhere, can be summarized in the conviction that the Alexandrian philologists’ production of an ekdosis, with the work of interpretation this implied, involved both conjectural emendations on the text that had come down to them and choice among text variants they had discovered through the collation of different copies. This overall work on the text was referred to by the term διόρθωσις and this was the procedure to which they adhered from Zenodotus onward. The two aspects have received differing emphasis, with some suggesting that the idea and practice of comparing different copies and choosing from variants generated by collated texts was alien to the Alexandrian critical-philological mindset. Some maintain that the Alexandrians solely or mostly conjectured with the aim of correcting without too many scruples a text judged to be corrupted and unacceptable on the basis of a raft of subjective criteria, such as supposed inconsistency, inappropriateness, material repetitions, preference for greater textual concision, standardisation and uniformity. I do not feel it is appropriate to use absolute and exclusive categories — that is, either to assert that all the readings espoused by the Alexandrian grammarians were merely arbitrary conjectures devoid of a documentary source, or alternatively to claim that they were all readings deriving from exemplars that had been collated and thus resulted from a choice among variants. I think it is far more likely that their work involved both conjectural criticism and also choice among variants deriving from collation. Naturally, with the evidence available to us today it will be difficult to distinguish case by case whether a reading represented a personal and subjective conjecture or whether it rested on a documentary source, unless we resort to hypotheses and deductions that may not always be reliable. But this is our own problem in interpreting individual cases: it is not a problem regarding the modus operandi and the method of the Alexandrian philologists. The two levels must not be superimposed, and the fact that we lack definitive criteria to consistently distinguish what is a conjecture and what is a variant by no means implies that one of the two categories is in abeyance. The idea that the Alexandrian philologists from Zenodotus to Aristarchus known as authors of ekdoseis of Homer did not carry out any collation of copies, but offered only conjectures without any documentary basis, and practised only a completely arbitrary ‘Konjekturalkritik’ without comparison among copies, has had a number of supporters, starting above all from the positions of M. van der  66 Pfeiffer 1968, 114.

Ekdosis: A Product of the Ancient Scholarship  

Valk,67 whose line of interpretation was also adopted (of course with individually differentiated stances) by H. van Thiel68 and most recently by M.L. West, to whom we will return later.69 This tendency leads recta via to a (quite unfair) underestimation of the importance and the value of the work performed by the Alexandrians. Arguments against it have been put forward by M. Haslam, Martin Schmidt, G. Nagy, J.-F. Nardelli, A. Rengakos and myself.70 On the question of the ‘Konjekturalkritik’ often and abundantly attributed to the Alexandrians, Rengakos observes that it is a theory based on the false presupposition that we have general criteria for distinguishing between conjectures and genuine variants when we are faced with the overall set of readings contained in the erudite sources, whereas such criteria do not exist at all. Furthermore, in the sources there is no explicit testimony referring to conjectural interventions, and it is impossible to demonstrate that a given reading is the fruit of a conjecture by the philologist to whom the textual choice is attributed. Rengakos has very clearly recapitulated that, on the contrary, there is actually a considerable amount of plausible evidence of the Alexandrian philologists’ knowledge of variants deriving from a comparison among copies.71 In addition to the arguments already illustrated above, based on the papyri and on the general practice of book production, Rengakos has dwelt on this problem, presenting very precise and cogent arguments concerning the testimony offered by the poets of early Hellenism, i.e. of the Zenodotean age, who reveal knowledge of different pre-existing Homeric

 67 Sharp criticism of van der Valk’s ideas (van der Valk 1949 and 1963–1964) has been made in a number of papers: for ex. Rengakos 1993, 38–48; Rengakos 2002, 146–148; Haslam 1997, 70 n. 31: “... he does not concern himself with the transmission. In categorizing readings he operates with an opposition between ‘original, old readings’ and ‘only subjective conjectures’ ... a schematization that is surely too simple to cope successfully with the complex vicissitudes of the Homeric text”. 68 H. van Thiel 1992 and 1997 (see also 1991, Einleit., and 1996, 20102, Einleit.) has argued that the readings which the tradition attributes to the Alexandrian grammarians were actually exegetic glosses or mere indirect references or reminiscences of parallel passages, written in a “Rand- und Interlinearapparat,” which Didymus, Aristonicus and others then wrongly interpreted as textual variants; I discuss this rather idiosyncratic vision in Montanari 1998a, 4–6 (= this volume, ch. 23); van Thiel 1992 is discussed by Schmidt 1997, with a reply in van Thiel 1997: see above and n. 9. 69 West 2001a, 2001b and 2002: discussion in Montanari 2002a (= this volume, ch. 47); 2004 (= this volume, ch. 41); 2009c (= this volume, ch. 60); 2011a (= this volume, ch. 8) and 2015a (= this volume, ch. 9); see further on. 70 Haslam 1997; Schmidt 1997; Führer/Schmidt 2001, 6–7; Nardelli 2001a (partic. 52–70, in direct opposition to West’s theories); Nagy 2000, 2003, 2004, 2009; Rengakos 2002, 2002a, 2012. 71 Rengakos 2012.

  Ekdosis: A Product of the Ancient Scholarship readings: “Do Hellenistic poets offer cases which prove beyond doubt that they made use of different Homeric manuscripts? In other words, do their works display Bindefehler which point to the older Homeric tradition? The answer is clearly ‘yes’.”72 Indeed we may confidently maintain that some of the Homeric variants testified in the lines of the philologist-poets of the Zenodotean age derived from the consultation of manuscripts and collation of copies. To this should be added cases in which it can be demonstrated, by finding veritable conjunctive errors, that the variants chosen by the Alexandrians already existed in a more ancient Homeric tradition.73 Indeed, Pfeiffer himself had explicitly supported this argument, reaching the following conclusion: “These three examples from the fifth to the third centuries, in which Zenodotus’ text is shown to be based on documentary evidence, show how unjustly he was charged by ancient critics, and by those modern scholars who followed them, with making arbitrary changes for wrong internal reasons.”74 Besides this indirect evidence, direct evidence can be found and I believe that it is decisive. Explicit testimony is supplied by the scholia, where one finds several undeniable references to the fact that Aristarchus consulted a number of different ekdoseis and found them to contain divergent readings: in other words, he certainly availed himself of the direct tradition of the copies he had at hand. The most evident and irrefutable case is that of sch. Il. 9.222 b, where Didymus reports that Aristarchus accepted a reading (graphe) because he found that it appeared in this form in some ekdoseis. Equally significant is sch. Il. 6.4 b, where Didymus states that Aristarchus at first accepted a certain reading, but later changed his mind because he had found another reading which he deemed to be preferable.75 All this evidence indicates that when engaging in text criticism, the Alexandrians — starting with Zenodotus and reaching the most refined method with

 72 Rengakos 2002, 149; cf. Rengakos 1993; 2001; 2002a; 2012; an interesting case pertaining to Zenodotus is highlighted by Fantuzzi 2005. 73 Lately Rengakos 2012. 74 Pfeiffer 1968, 110–114: the citation is on p. 114; the three examples adduced by Pfeiffer concern Il. 1.5, Il. 1.225–233 and Il. 16.432–458, Il. 4.88. Pfeiffer normally attributed the collation of copies to the great philologists who succeeded Zenodotus: cf. for example p. 173. Pfeiffer’s arguments should have been awarded greater consideration. 75 Cf. Rengakos 2012, 244–248, with bibliography. The above mentioned evidence of Didymus in sch. Il. 9.222 b is rightly underlined by various scholars (Nagy, Janko, Rengakos and myself) and cannot be dismissed out of hand, as does West (2001a, 37 n. 19): I examine closely meaning and importance of this sch. in 2015b. On Aristarchus’ second thoughts, see Montanari 1998a (= this volume, ch. 23) and 2000a (= this volume, ch. 38).

Ekdosis: A Product of the Ancient Scholarship  

Aristarchus — based themselves not only on text-internal conjectural proposals but also on external and diplomatic resources, consisting in choice among variants they found or noticed in a non-univocal tradition composed of the copies they had available and were thus able to consult. It would seem, therefore, that the burden of proof is on whoever seeks to strip the Alexandrian grammarians of any knowledge of variants deriving from collation of copies, attributing to them only arbitrary conjectures, rather than the opposite: the fact is that we have, at the very least, convergent evidence in favor of knowledge of variants — and I would go so far as to say that we have real proof. An interesting testimony concerning these problems can be found in the recently discovered De indolentia by Galen, an author of major importance in the history of ancient philology, not only on account of his activity and his thought, but also by virtue of the information Galen’s text provides. It has begun to be studied and appreciated from this point of view, but certainly much fruitful investigation remains to be done.76 The new text is preserved in a copy which, overall, has many incorrect forms and results in considerable uncertainty of interpretation, also affecting the points of interest here, but it is worth commenting on the material and singling out several pieces of information.77 In the 192 fire of Rome, Galen lost, among other things, all the books he possessed, and in this work (composed in epistolary form as an answer to a Pergamene friend) he talks extensively about his activity as a scholar and about his books. Those that had been lost included texts “corrected in my own hand” (§ 6); there were also rare books that were not available elsewhere, and books which, while not rare, constituted unrepeatable exemplars due to the particularly accurate and carefully written text, such as the Plato by Panaetius and two Homers by Aristarchus, and others of this kind (§ 13). There follows a rather tortuous passage, which may perhaps contain a reference to copies with marginal annotations and bearing the name of the person who had made the marginal jottings.78 A little further on Galen relates that he had also lost books he himself had worked on, in which he had corrected various errors in order to compose an ekdosis of his own. The task he had set himself, he says, had involved careful attention to textual readings, so as to ensure nothing was added or was missed out, and to make sure that all the appropriate signs were present to distinguish the structural parts of the text (paragraphoi and coronides), as well as the punctuation, the latter being so important, especially in obscure works, that it could  76 Manetti/Roselli 1994; Manetti 2006; 2012a; and 2012b; Roselli 2010; 2012; see Manetti 2015. 77 Editions: Boudon/Millot/Jouanna 2010; Kotzia/Sotiroudis 2010; Garofalo/Lami 2012. 78 Cf. Roselli 2010; 2012; Stramaglia 2011; Manetti 2012b.

  Ekdosis: A Product of the Ancient Scholarship even substitute for the exegete himself (§ 14). It has been rightly emphasized that here a new aspect of the personality of Galen emerges: already known as an exegete and commentator, he can now also be seen as a text editor (not only of medical works but also of the works of numerous philosophers). Thus he was the author of editions designed for his own personal use, and in preparing such editions he worked on the text in order to identify lacunae and interpolations, and to highlight the structural framework of the work with appropriate critical signs, and to indicate the punctuation as an aid to text comprehension.79 Elsewhere Galen cites a number of ekdoseis, including that of Hippocrates by Bacchius, dating from the 3rd century BC (a partial contemporary of Zenodotus), and by Dioscorides and by Artemidorus (from the age of Hadrian, one generation earlier than Galen). With regard above all to the ekdosis of Dioscorides, Galen offers important insight into its material form: it presented critical signs and punctuation, the obelos was used to indicate doubtful authenticity, and variants were marked in the blank spaces (lower and upper margins and the intercolumnae).80 The information Galen provides on the ekdoseis of medical texts, which he himself had performed or had been carried out by his predecessors, is in agreement with the arguments we have put forward so far. The philologist selected a copy on which to work and thus produced his own ekdosis; he then personalized this copy, on which all his markings were visible, by writing his own name on it, after which the copy was allowed to circulate for essentially private use, or for school and teaching purposes (like the Homer of Zenodotus or of Aristarchus). Upon request or for various reasons, the copy itself could then be copied, i.e. reproduced as complete exemplars of the work, corrected and presented as ‘fair copies’. This could be done either by the editor himself or by someone entrusted with the task. In § 14 Galen mentions precisely the case of books transcribed as fair copies after undergoing correction (diorthosis).81 Over the past few years, the arguments put forward mainly by myself and by A. Rengakos82 have prompted a debate above all with the positions of M. West, the most radical advocate of the theory that the Alexandrian philologists from Zenodotus to Aristarchus known as authors of ekdoseis of Homer did not carry out any collation of copies: “In fact, the first scholars known to have cited manuscript

 79 Cf. Boudon/Millot/Jouanna 2010 xxxiii–xxxiv; Manetti 2012a; Roselli 2012; on lectional signs, punctuation and accentuation cf. Pfeiffer 1968, 178–179. 80 Cf. Manetti/Roselli 1994, 1625–1633; Manetti 2012a; Roselli 2012; see Manetti 2015. 81 Roselli 2012, 64–67. 82 But see also above and n. 69.

Ekdosis: A Product of the Ancient Scholarship  

authority for variant readings are Aristarchus’ contemporaries Callistratus and Crates. Didymus is the first author known to have compiled anything in the nature of a ‘critical apparatus’. It is entirely unjustified to project his methods back onto Aristarchus or Zenodotus, or to assume that all the various copies available to Didymus in the time of Augustus were already part of the library’s holdings in the early third century”.83 I feel I must express some misgivings upon reading that “it is entirely unjustified to project his methods back onto Aristarchus” immediately after the statement that “the first scholars known to have cited manuscript authority for variant readings are Aristarchus’ contemporaries”. It is simply begging the question to claim that Didymus’ method is projected back onto Aristarchus or Zenodotus, for in actual fact there is absolutely no evidence that Didymus was the first to apply this method rather than having inherited it from his predecessors, unless it be the preconception that Zenodotus, Aristophanes and Aristarchus did not apply it at all.84 In M. West’s view, Zenodotus’ readings are to be regarded either as conjectural emendations or as peculiarities of his base-text, but do not result from any form of comparison among copies. Zenodotus is claimed to have lacked the concept of variants, his only critical concern being the existence and identification of spurious lines. It is West’s contention that “The one kind of textual criticism we know Zenodotus practised (my italics) was not concerned with choices between variant readings but with the identification of spurious lines and passages. The one feature of his text that marked it out as a critical diorthosis was the presence of obeloi in the margins (and perhaps brackets enclosing longer passages) to signal the critic’s suspicion that certain verses were unHomeric”.85 This, however, obviously means that Zenodotus felt he had to tackle the problem of how to discriminate the authentic from the spurious: for whereas an entire line held to be spurious could be eliminated from the text, a part of the line (a word or an expression) cannot be removed without replacing it with something else. Sometimes, by eliminating or accepting a line, the meaning and  83 West 2001a, 36; West 2002, 140. 84 West’s ‘Didymean’ hypothesis is rejected by Nagy 2000; Nardelli 2001, 61–64; Janko 2002; Rengakos 2012; and myself (see above n. 68). The already mentioned evidence of Didymus in sch. Il. 9.222 b (see above and n. 75; below nn. 88 and 89) is rightly underlined by various scholars (Nagy, Janko, Rengakos and myself): in that scholium Didymus does indeed state that Aristarchus found a reading in some exemplars. 85 West 2001a, 38; West 2002, 140. The idea of the possible use of signs for material and graphic deletion placed on the base-copy was clearly put forward already in Montanari 1998a, 6 (= this volume, ch. 23), but West does not seem to be concerned with the distinction between athetein and ou graphein in the terminology on the textual interventions of Zenodotus.

  Ekdosis: A Product of the Ancient Scholarship syntax called for the alteration of some word or words before or after the line itself; at times, the alteration of a word called for or allowed the elimination or addition of a line. But Zenodotus’ “one kind of textual criticism” and “the critic’s suspicion that certain verses were un-Homeric” do suffice, for once a critical approach towards the way in which the text presents itself has been acquired, it is inconceivable for there to be a theoretical and essential separation which would discriminate between line athetesis and single word alteration and would thereby justify the assumption that the philologist’s concern focused only on athetesis of whole lines and not on shorter text alterations.86 In either case, the problem at hand for the philologist resides in the opposition between authentic/ correct vs. spurious/damaged and in seeking to identify the proper text. By addressing the issue of the authentic text and how to devise the criticalmethodological tools to obtain it, Zenodotus achieved a major breakthrough: it was a crucial intellectual step, which we identified above as residing in the difference between “correcting a single copy” and “editing a text”. M. West warns against a travesty of the situation: “The misapprehension, which goes back at least at the time of Wolf, is that Zenodotus, Aristophanes and Aristarchus were all editors in the modern sense, who wanted to establish a good text of Homer and who approached the task as a modern editor does, by collecting manuscripts and comparing their readings”.87 Now, if Zenodotus had at least “one kind of textual criticism”, what is likely to have been his aim in carrying out emendations on the Homeric text? Are we thus to believe that Zenodotus had a conscious premeditated idea of ‘modelling’ Homer according to his own taste, i.e. “I’m going to set about reworking Homer and I’m going to make it the way I think it ought to be”? This possibility is by no means easy to accept, but actually this is the only alternative to the view that “he wanted to establish a good text”, which is the natural goal of anyone who starts working on and correcting a text. In effect the aim of the Alexandrian philologists cannot but have been “to establish a good text”, whatever the value of the result according to modern scholars. I fear that the misunderstandings arise from the fact that there is no clear definition of the guidelines for our judgment on the work of the Alexandrian philologists. By adopting our own point of view concerning the ‘competence’ on which they base their opinions and arguments, so that it can be ascertained  86 All the more so since expunction at times involves variants in the part which remains in the text: cf. Montanari 1998a, 7 n. 17 (= this volume, ch. 23), on the subject of Il. 2.156–168 (sch. 2.156–169); cf. Rengakos 2012, 250–252; see above and n. 50. 87 West 2001a, 34; 2002, 138.

Ekdosis: A Product of the Ancient Scholarship  

whether and when they are right or wrong in comparison with the ‘truth’ according to scientific philology, we risk producing unfounded and pointless judgments. Naturally, evaluation of the quality of their choices is the proper perspective for the interpreter and editor of Homer as a modern philologist; on the other hand, maintaining conscious awareness of historical distance and taking care not to overlay our criteria on their behavior is the proper perspective for the historian of philology as a cultural and intellectual phenomenon and for the ‘reading’ of Homer in ancient civilization. Perhaps it is hard to conceive that Zenodotus’ aim (however incoherent and unsophisticated) was precisely “to establish a good text of Homer” because the testimony that has come down to us indicates that his text was far from good — in fact it was dreadful, and incoherent seen through the filter of the requirements and knowledge of modern classical philology. And even as regards the successors of Zenodotus, or even the great Aristarchus, we can hardly claim always to agree with their text choices. The viewpoint from which a modern Homeric scholar approaches his task is the need to decide whether the text Zenodotus, Aristophanes or Aristarchus judged to be the best is indeed the one to print in a present-day critical edition, and whether their interpretations should be espoused as valid in a scientific commentary. In contrast, the viewpoint from which a historian of ancient philology starts out is that of seeking to understand their methods, arguments, principles, knowledge — in a word, their historical and intellectual position. The tendency to scoff at the opinions of the Alexandrian philologists in terms of modern Homeric studies should by no means translate into the tendency to discredit their historical significance, which needs to be correctly positioned and contextualized. It is mistaken to blur the distinction between the two planes.88 It is impossible to escape the fact that by inventing the obelos and  88 Janko 2002 seems to render this concept explicit rather more clearly. His position on the methods of the Alexandrian philologists is not an extremist unilateral stance: he believes that the majority of their readings are indeed arbitrary conjectures (by the Alexandrians themselves or possibly of more ancient origin), but he does not go so far as to deny the recourse to manuscripts and comparison among copies as part of their ekdosis work (for Zenodotus, Janko 1992 23: “His caution was salutary, given the abundance of interpolated texts; he certainly had MS authority for some omissions”; for Aristarchus, Janko 1992, 27, and Janko 2002: “This [i.e. sch. Il. 9.222, cf. above n. 75 and 84] certainly implies that Aristarchus did check manuscripts for variant readings”). On the one hand, Janko argues, there stands the problem of the origin of their proposed text choices (subjective emendation, comparison among copies) and therefore of their working procedures; on the other, he points out, “my own concern, as a Homerist, has always been whether such readings are authentic”. Perfectly clear: modern philologists can to some extent be severe regarding the opinions of the Alexandrians, considering them to be fairly acceptable or fairly unacceptable from their own point of view (Janko is very negative),

  Ekdosis: A Product of the Ancient Scholarship setting himself the task of emending and restoring the text he had at hand, Zenodotus had lit upon an idea which, however embryonic and crude it may appear, would undergo further development among his successors, eventually becoming the germ of the discipline we call classical philology. But even if one were to suppose that he acted purely on the basis of conjectures, could it be denied that conjecture is one of the emblematic and representative tools of philology aimed at restoring the correct text? A further comment by M. West is surprising. “Consider what we know of Aristarchus’ methods, for which we have plenty of material in the scholia. Of course he had the text of his teacher Aristophanes before him. He also kept an eye on [my italics] that of Zenodotus, and took up critical positions against it. But the arguments he used, as reported by Aristonicus and Didymus, were always based on the internal evidence of contextual coherence or general Homeric usage. Not once does he appeal to the authority of manuscripts”.89 So Aristarchus compared his working copy with that of Aristophanes and that of Zenodotus; but the phrase “kept an eye on” is insidiously reductive, given that the number of cases preserved by the tradition runs into the hundreds — and the tradition itself is patchy and incomplete. Be that as it may, the picture delineated above implies he made a certain small comparison among copies, but that he took great care not to let his eye stray onto any further copies: in other words he did study and interpret Homer, but he painstakingly avoided consulting any other exemplar than his own, together with that of Zenodotus and that of Aristophanes, although these alone already presented him with not inconsiderable divergences. Frankly, this seems to me like yet another paradoxical portrayal. Certainly, what is clear is that Zenodotus used a copy of Homer as the basetext to work on: but are we to believe that Zenodotus’ copy was the only existing exemplar of Homer among the circles of those frequenting the Museum and the Library of Alexandria, or is it quite likely that there were others around, with

 but they cannot downplay the fact that the ancient Alexandrians emended and compared exemplars to correct the Homeric text, a method that combined interpretation of the text with awareness of the history of the tradition. An extremely apt remark, perfectly applicable to Alexandrian philology as well, is offered by Cassio 2002, 132, on the issue of pre-Alexandrian criticism: “The earliest scholarly approach to the Homeric text is totally foreign to us ... we do right to think along very different lines, but we should never forget that it was the commonest approach to the Homeric text in the times of Socrates, Plato and Aristotle. As a consequence, we ought to be wary of looking at it with a superior smile, and ought to try to understand its motives in more depth instead”. 89 West 2001a, 37: at least the mentioned case of sch. Il. 9.222 clearly contradicts this, cf. above and nn. 75, 84, 88.

Ekdosis: A Product of the Ancient Scholarship  

different text characteristics? Should we conceive of a paradoxical Zenodotus who, despite his taste and his concern for the Homeric text, made every effort not to look at other copies he may have come across, not to make a note of the points where they departed from his own copy and not to ask himself any questions about those differences?90 It seems to me far more likely that he noticed the differences, both as regards the number of lines and certain individual readings, and that he decided to write them down and express his opinions. However, a subtle ambiguity needs to be eliminated: when speaking of ‘other’ or ‘various’ copies of Homer that were actually available and utilizable, one should not be misled into thinking that hundreds and hundreds of exemplars were concretely at hand and ready for consultation, thereby transforming the idea of a comparison into an exaggerated undertaking that immediately becomes totally implausible. Obviously, it would be a pure anachronism (and not very intelligent) to assume that the Alexandrian philologists had conceived the idea of a collatio of the entire known manuscript tradition, a recensio after the style of the so-called Lachmann method: but who would dare advance such a ludicrously naive proposal? In actual fact the problem should be considered in a rather different fashion, embracing a perspective that is perfectly reliable in historical terms. More specifically: can one begin to speak of comparison among copies only when a certain number (how many?) is reached, or was it sufficient to compare a few, to detect variants when the textual tradition was not univocal and then address the problem of which text was correct and which ones were wrong? Overall, we must recognize that we owe to the Alexandrian grammarians an idea of text philology aiming to restore the correct text, freeing it from errors and imperfections. From the age of Zenodotus onwards, progress was gradually made in refining the method, which achieved its highest accomplishment with Aristarchus. The grammarians realized that a text had its own history of transmission, during the course of which it deteriorated in various ways; it could be restored to its correct form either via conjecture or by choosing the correct reading from among those offered by a non univocal tradition.91 The recognition of transmission-induced damage that had affected the authentic text, along with

 90 West 2001a, 38: “No doubt it would have been easy for him [scil. Zenodotus] to collect several copies if he had taken the trouble”. Are we to assume that Zenodotus was a somewhat lazy philologist? 91 Naturally the works of Homer come to mind, but also of the tragic and comic poets.

  Ekdosis: A Product of the Ancient Scholarship steps and procedures to restore it, is proof of how the mutual dependency of textual criticism and textual interpretation became established and operational.92

Conclusions To bring to a conclusion the various points outlined in the preceding pages, it is worth pointing out, first and foremost, that the material form of the Alexandrian ekdosis is a question closely linked to debate on the textual criticism performed by the Alexandrian philologists, but also to enquiry into the provenance of their readings (variants and conjectures); in short, it constitutes the core issue in the historical-cultural assessment of Alexandrian scholarship. This is the basic point and we must make it clear once and for all that we are dealing with a problem of principles and methods, not of the quantity of the data (number of collated copies or of variants discussed) or of the quality of the results (right or wrong from our point of view). We are not concerned with establishing the minimum number of copies to be subjected to comparison or of variants to be taken into consideration before one can even begin to speak of philology, nor with determining how many ‘correct’ readings or ‘good’ interpretations are needed before it makes sense to speak of philology. Rather, in a historical perspective, all that was needed in order for there to be a decisive step forward in intellectual achievement was the very fact of understanding and addressing the problem, even if only partially, erratically and incoherently: a literary text had a multifaceted history of transmission, during which it could become distorted at various points; the correct text (i.e. what is authentic versus, what is spurious and what was the original wording) could then be restored by conjecture or by choosing the best reading among those offered by a divergent tradition.93 The idea of the recognition of damage and of finding a way to repair it reveals that the organic unity between interpretation and textual criticism had become established. Although much progress still remained to be made, and Wolfian scientific philology, the modern critical edition and the scientific commentary were still in the distant future, our viewpoint — far from being an anachronism — is the historical evaluation that a nodal step had been taken in the period from Zenodotus to Aristarchus.  92 See Pasquali 1920 (citation from the reprint of 1998, 26): “a costituire un testo ... occorre la stessa preparazione che a interpretare ...; costituire un testo e interpretarlo sono, in fondo, tutt’uno” (“constituting a text ... requires the same learning and knowledge as interpreting ...; constituting a text and interpreting it are, ultimately, one and the same thing”). 93 See now Conte 2013, 44–50.

 Aristarchus’ Conjectures (once) Again The idea that the Alexandrian philologists from Zenodotus to Aristarchus, known as authors of ekdoseis of Homer, did not engage in collation of copies but merely offered conjectures without documentary basis, and practised only a completely arbitrary “Konjekturalkritik” without comparison among copies, has had a number of supporters, starting above all from the positions of M. van der Valk.1 The line of interpretation espoused by the latter was also adopted (of course with individually differentiated stances) by H. van Thiel2 and most recently by M.L. West.3 However, it is an approach that leads recta via to a (quite unfair, I believe) underestimation of the importance and the value of the work performed by the Alexandrians. Arguments against it have been put forward by M. Haslam, M. Schmidt, G. Nagy, J.-F. Nardelli, A. Rengakos and myself.4 On the question of the “Konjekturalkritik” often and abundantly attributed to the Alexandrians, Antonios Rengakos observes that it is a theory based on the false presupposition that we have general criteria for distinguishing between conjectures and genuine variants when we are faced with the overall set of readings contained in the erudite sources (chiefly scholia), whereas there are in fact no such criteria. Furthermore, in the sources there is no explicit testimony referring to conjectural interventions, and it is impossible to demonstrate that a  English translation by Rachel Barritt Costa.  1 Valk 1949 and 1963‒1964; sharp criticism of van der Valk’s ideas has been made in a number of papers: for ex. Rengakos 1993, 38‒48; 2002, 146‒148; 2002; 2012; Montana 2012; Haslam 1997, 70 n. 31: “… he does not concern himself with the transmission. In categorizing readings he operates with an opposition between ‘original, old readings’ and ‘only subjective conjectures’ … a schematization that is surely too simple to cope successfully with the complex vicissitudes of the Homeric text”. 2 Van Thiel 1992 and 1997 (see also 1991, Einleit., and 1996, 20102, Einleit.) has argued that the readings which the tradition attributes to the Alexandrian grammarians were actually exegetic glosses or mere indirect references or reminiscences of parallel passages, written in a “Randund Interlinearapparat,” which Didymus, Aristonicus and others then wrongly interpreted as textual variants; I discuss this rather idiosyncratic vision in Montanari 1998, 4‒6; van Thiel 1992 is discussed by Schmidt 1997, with a reply in van Thiel 1997. 3 West 2001a, 2001b; 2002: discussion in Montanari 2002 (= this volume, ch. 21); 2004 (= this volume, ch. 41); 2009a (= this volume, ch. 7); 2011a (= this volume, ch. 8); 2015 (= this volume, ch. 10); 2015a (= this volume, ch. 9). 4 Haslam 1997; Schmidt 1997; Führer/Schmidt 2001, 6‒7; Nardelli 2001a (partic. pp. 52‒70, in direct opposition to West’s theories); Nagy 2000; 2003; 2004; 2010; Rengakos 2002; 2002a; 2012; Montanari: see n. 3. On Janko’s position, see infra n. 26. https://doi.org/10.1515/9783110772371-011

  Aristarchus’ Conjectures (once) Again given reading is the fruit of a conjecture by the philologist to whom the textual choice is attributed. Rengakos has very clearly recapitulated that, on the contrary, there is actually a considerable amount of plausible evidence of the Alexandrian philologists’ knowledge of variants deriving from a comparison among copies.5 In addition to the arguments based on the papyri and on the general practice of book production,6 Rengakos has dwelt on this problem in depth, presenting very precise and cogent arguments concerning the testimony offered by the poets of early Hellenism, i.e. of the Zenodotean age, who reveal knowledge of different pre-existing Homeric readings: “Do Hellenistic poets offer cases which prove beyond doubt that they made use of different Homeric manuscripts? In other words, do their works display Bindefehler which point to the older Homeric tradition? The answer is clearly ‘yes’”.7 Indeed we may confidently maintain that some of the Homeric variants testified in the lines of the philologist-poets of the Zenodotean age derived from the consultation of manuscripts and collation of copies. To this should be added cases in which it can be demonstrated, by finding veritable conjunctive errors, that the variants chosen by the Alexandrians already existed in a more ancient Homeric tradition.8 Pfeiffer himself explicitly supported this argument, reaching the following conclusion: “These three examples from the fifth to the third centuries, in which Zenodotus’ text is shown to be based on documentary evidence, show how unjustly he was charged by ancient critics, and by those modern scholars who followed them, with making arbitrary changes for wrong internal reasons”.9 Besides this indirect evidence, direct evidence can be found and I believe that it is decisive. Explicit testimony is supplied by the scholia, where one finds several undeniable references to the fact that Aristarchus consulted a number of different copies and found them to contain divergent readings: in other words, he certainly availed himself of the direct tradition of the copies he had at hand. The most evident and irrefutable case is that of sch. Il. 9.222 b, where Didymus

 5 Rengakos 2012. 6 Montanari 2009a (= this volume, ch. 7); 2009c (= this volume, ch. 60); 2011; 2015 (= this volume, ch. 10); 2015a (= this volume, ch. 9). 7 Rengakos 2002, 149; cf. Rengakos 1993; 2001; 2002; 2002a; 2012; an interesting case pertaining to Zenodotus is highlighted by Fantuzzi 2005. 8 Lately Rengakos 2012. 9 Pfeiffer 1968, 110‒114: the citation is on p. 114; the three examples adduced by Pfeiffer concern Il. 1.5, Il. 1.225‒233 and Il. 16.432‒458, Il. 4.88‒89. Pfeiffer normally attributed the collation of copies to the great philologists who succeeded Zenodotus: cf. for example p. 173. Pfeiffer’s arguments should have been awarded greater consideration. Cfr. now Montana 2012, 31‒32 and 49‒53.

Aristarchus’ Conjectures (once) Again  

reports that Aristarchus accepted a reading because he found that it appeared in this form in some ekdoseis. Equally significant is sch. Il. 6.4 b, where Didymus states that Aristarchus at first accepted a certain reading, but later changed his mind because he had found another reading which he deemed to be preferable.10 Let we first take into consideration Il. 9.222 and Didymus’ scholia ad loc. Sch. Il. 9.222 b1 (Did.) αὐτὰρ ἐπεὶ πόσιος καὶ ἐδητύος 〈ἐξ ἔρον ἕντο〉: φαίνονται καὶ παρ᾿ Ἀγαμέμνονι πρὶν ἐπὶ τὴν πρεσβείαν στείλασθαι δειπνοῦντες· φησὶ γοῦν “αὐτὰρ ἐπεὶ σπεῖσάν τ᾿ ἔπιόν θ᾿, ὅσον ἤθελε θυμός, / ὡρμῶντ᾿ ἐκ κλισίης” (Ι 177‒8). ἄμεινον οὖν εἶχεν ἄν, φησὶν ὁ Ἀρίσταρχος, 〈εἰ〉 ἐγέγραπτο “ἂψ ἐπάσαντο” ἢ “αἶψ᾿ ἐπάσαντο”, ἵν᾿ ὅσον χαρίσασθαι τῷ Ἀχιλλεῖ γεύσασθαι μόνον καὶ μὴ εἰς κόρον ἐσθίειν καὶ πίνειν λέγωνται. ἀλλ᾿ ὅμως ὑπὸ περιττῆς εὐλαβείας οὐδὲν μετέθηκεν, ἐν πολλαῖς οὕτως εὑρὼν φερομένην τὴν γραφήν. b2 (Did.) ἄμεινον, φησίν, εἶχεν, Ἀρίσταρχος, εἰ ἐγέγραπτο “ἂψ ἐπάσαντο”. b3 (Did.) {αὐτὰρ ἐπεὶ πόσιος καὶ ἐδητύος ἐξ ἔρον ἕντο:} Ἀρίσταρχος γράφει “ἂψ ἐπάσαντο”· ἤδη γὰρ ἦσαν παρ᾿ Ἀγαμέμνονι κορεσθέντες· οὐ μετέθηκε δὲ τὴν γραφήν.

Agamemnon has summoned the army chiefs to his tent and has prepared a banquet for them (9.90), and the sumptuous meal is eaten before the ambassadors start out on their mission (9.174‒178). Then the ambassadors make their way to Achilles’ tent, where they again tuck into the lavish spread put on for them, and Odysseus’ speech begins only “after they had eaten and drunk their fill”: αὐτὰρ ἐπεὶ πόσιος καὶ ἐδητύος ἐξ ἔρον ἕντο (v. 222). The line evidently raised a number of queries: given that, before setting out for their embassy, the ambassadors had already feasted at Agamemnon’s banquet ὅσον ἤθελε θυμός (v. 177), how could they feel the pangs of hunger and thirst again shortly afterwards and how could they be expected to gorge themselves on the food and quaff the libations once more (in the manner of a typical Homeric scene)? Sensing this incongruity, Aristarchus says it would have been better if the poet had written ἂψ ἐπάσαντο or αἶψ᾿ ἐπάσαντο, therefore αὐτὰρ ἐπεὶ πόσιος καὶ ἐδητύος ἂψ ἐπάσαντο “after they had once again partaken in the drink and food” or αὐτὰρ ἐπεὶ πόσιος καὶ ἐδητύος αἶψ᾿ ἐπάσαντο “after they had rapidly sipped the drink and tasted the food”.11 Had it been worded in this manner, the line would have portrayed the ambassadors as honouring the grand banquet laid out for them merely so they could please Achilles, rather than devouring the food voraciously as if they were ravenous.

 10 Cf. Rengakos 2012, 244‒248, with bibliography. 11 The verb πατέομαι with the gen. as in Il. 19.160, 24.641 and a number of times in the Odyssey.

  Aristarchus’ Conjectures (once) Again There follows an interesting judgment by Didymus on Aristarchus’ philology: “But nevertheless he ὑπὸ περιττῆς εὐλαβείας did not change anything, since in many ekdoseis he had found the reading handed down in this form”. How should one interpret ὑπὸ περιττῆς εὐλαβείας unless it be “through extraordinary prudence” or possibly even “as a result of excessive cautiousness, beset by excessive scruples”? In short, Aristarchus was not convinced that the text he found in many editions was genuinely reliable and he had a conjecture (or rather, two possible conjectures) as to how to improve it and make it more plausible, but he was so cautious — perhaps even too cautious — that he made no attempt to change a text he regarded as incoherent by introducing what would effectively have been a conjecture of his own. This piece of evidence is rightly underlined by various scholars12 and cannot be dismissed out of hand, as does M.L. West in a truly surprising manner:13 This does not mean that Aristarchus said ‘I find this reading in many manuscripts, and therefore do not venture to alter it’; it is Didymus’ way of saying that Aristarchus refrained from putting the reading he would have preferred into the text because he had no manuscript authority for it. Didymus knew that Aristarchus consulted more than one text, because he cited different scholars’ readings on different occasions, but it is just his own assumption that Aristarchus systematically checked ‘many’ copies before discussing any reading. I have already highlighted, elsewhere, the methodological error that leads West to maintain that Aristarchus (like other Alexandrian philologists) did not collate copies: as compared to an ideal “systematically checking ‘many’ copies”, the established fact of consulting “more than one text” is rendered meaningless in this context, as if to say that the collation of copies is performed only when one makes a systematic recensio in modern terms, searching for all the known and available copies. When dealing with such a superficially quantitative vision, it is vital to be aware that we are facing a problem of principles and method, not of quantity of the data or quality of the results (see below). The principle which holds that comparisons are made among various copies and the differences are noted is certainly present, though not in terms of a systematic procedure, which was obviously unthinkable at the time, but it does signify a genuine idea of textual philology which had the aim of establishing the correct text.14

 12 Nagy, Janko, Rengakos and myself: cf. Rengakos 2012, 244‒248, with bibliographical information. 13 West 2001a, 37 n. 19. 14 I will not repeat here the arguments already spelled out in Montanari 2011; 2015 (= this volume, ch. 10); 2015a (=this volume, ch. 9), with the previous bibliography.

Aristarchus’ Conjectures (once) Again  

Furthermore, I must confess that West’s considerations on the meaning of this scholium strike me as completely unmotivated and I feel obliged to restrict myself to what the scholium actually states, namely that Aristarchus pointed out a difficulty in the text but, according to Didymus, he was too cautious to change what he found in the copies he consulted (who can say how many he examined? Only Aristarchus himself could tell us) if such a modification sprang purely from his own conjectural assumption. Personally, for reasons of caution I would not venture to argue that this philological mode of operating was the only one applied by Aristarchus with total and absolute rigor. It cannot be ruled out, indeed it is perhaps quite probable, that he sometimes did introduce his own conjectures even if they were devoid of documentary support, but it is an established fact that we have no general criteria for distinguishing coherently between what springs from a conjecture and what represents a testified variant. Yet this by no means implies that one of the two categories is misleading and should be excluded when we are faced with the overall set of readings contained in the erudite sources. The Alexandrian philologists’ production of an ekdosis must have involved both aspects: conjectural emendations as well as choice among variants detected through the collation of copies. And while there is no explicit testimony referring to conjectural interventions and it is impossible to demonstrate that a given reading is the fruit of a conjecture by the philologist to whom the textual choice is attributed, there is, nevertheless, a considerable amount of plausible evidence of the Alexandrian philologists’ knowledge of variants deriving from a comparison among copies.15 Again, West goes so far as to state: “Not once does he [scil. Aristarchus] appeal to the authority of manuscripts”.16 However, this fails to take into account that there are several cases clearly testifying to the fact that Aristarchus did check different copies and effectively made use of them. One of the most explicit and evident instances is found in the scholia of Aristonicus and Didymus on Il. 6.4: Sch. Il. 6.4a (Ariston.) μεσσηγὺς Σιμόεντος 〈ἰδὲ Ξάνθοιο ῥοάων〉: ὅτι ἐν τοῖς ἀρχαίοις ἐγέγραπτο “μεσσηγὺς ποταμοῖο Σκαμάνδρου καὶ Στομαλίμνης”· διὸ καὶ ἐν τοῖς ὑπομνήμασι φέρεται. καὶ ὕστερον δὲ περιπεσὼν ἔγραψε· μεσσηγὺς Σιμόεντος ἰδὲ Ξάνθοιο ῥοάων· τοῖς γὰρ τοῦ ναυστάθμου τόποις ἡ γραφὴ συμφέρει, πρὸς οὓς μάχονται.

 15 Most recently Rengakos 2012 (on “Konjekturalkritik” p. 247); see also Montanari 2011; 2015; 2015a (= this volume, ch. 9), with the previous bibliography. 16 West 2001a, 37.

  Aristarchus’ Conjectures (once) Again Sch. Il. 6.4b (Did.) μεσσηγὺς Σιμόεντος 〈ἰδὲ Ξάνθοιο ῥοάων〉: πρότερον ἐγέγραπτο “μεσσηγὺς ποταμοῖο Σκαμάνδρου καὶ Στομαλίμνης”. ὕστερον δὲ Ἀρίσταρχος ταύτην17 εὑρὼν ἐνέκρινεν.18 Χαῖρις (fr. 2 Β.) δὲ γράφει “μεσσηγὺς ποταμοῖο Σκαμάνδρου καὶ Σιμόεντος”.

It is quite clear that Didymus is reporting a change of opinion on the part of Aristarchus, who had been having second thoughts and subsequently (ὕστερον) decided to change the reading he had previously accepted (and commented on, as part of his hypomnemata), because in the meantime he had found another that he regarded as better. This is a perfect counterbalance to the case of Il. 9.222: there he refrained from changing the only reading he found in his copies (scil. in all the ones he had seen), whereas here he opted in favour of change because he had, in the meantime, come across a copy with a reading he believed was preferable.19 Two interesting cases in which Aristarchus had second thoughts and changed his textual choice can be found in the scholia to Iliad 18.207 and 9.464.20 Let us start with the observation put forward by Didymus in sch. Il. 18.207a: Sch. Il. 18.207a (Did.): ὡς δ᾿ ὅτε καπνὸς ἰὼν 〈ἐξ ἄστεος αἰθέρ᾿ ἵκηται〉: οἱ περὶ Διονύσιον τὸν Θρᾷκα φασὶν Ἀρίσταρχον πρῶτον ταύτῃ χρώμενον τῇ γραφῇ μεταθέσθαι καὶ γράψαι “ὡς δ᾿ ὅτε πῦρ ἐπὶ πόντον ἀριπρεπὲς αἰθέρ᾿ ἵκηται”· ἐμφατικῶς τὸ ἐν πολέμῳ πῦρ ἐπιτεθὲν τῷ Ἀχιλλεῖ παρέβαλε τῷ ἐν πολεμουμένῃ ἁπτομένῳ. Sch. Il. 18.207b (Did.): Ἀρίσταρχος “ὡς δ᾿ ὅτε πῦρ ἐπὶ πόντον ἀριπρεπὲς αἰθέρ᾿ ἵκηται”· καὶ γὰρ ἄτοπόν φησι πῦρ εἰκάζεσθαι καπνῷ.

Dionysius Thrax tells us here that Aristarchus had initially accepted the reading ὡς δ᾿ ὅτε καπνὸς ἰὼν ἐξ ἄστεος αἰθέρ᾿ ἵκηται, presupposed by the scholium as a lemma (Ἀρίσταρχον πρῶτον ταύτῃ χρώμενον τῇ γραφῇ), but later changed his mind (μεταθέσθαι καὶ γράψαι), adopting the reading ὡς δ᾿ ὅτε πῦρ ἐπὶ πόντον ἀριπρεπὲς αἰθέρ᾿ ἵκηται. Very interesting in terms of the issues it raises is the case where Didymus once again cites Dionysius Thrax in sch. Il. 9.464b: Sch. Il. 9.464b (Did.): ἀμφὶς ἐόντες: Διονύσιος ὁ Θρᾷξ ἐν τῷ Πρὸς Κράτητα διὰ τῆς ἱπποδρομίας φησὶ γεγραμμένου “ἀντιόωντες” μεταθεῖναι τὸν Ἀρίσταρχον ἀμφὶς ἐόντες.

 17 Erbse ad loc.: ταύτην τὴν γραφήν b (fort. rectius). 18 ἐπέκρινε T, ἐπέκρινεν b: ἐνέκρινεν Bekker (vd. Erbse ad loc.). 19 Cfr. Rengakos 2012, 245. 20 Cfr. Montanari 2000a, 483‒486, with bibliography (= this volume, ch. 38).

Aristarchus’ Conjectures (once) Again  

According to Didymus, in the treatise Πρὸς Κράτητα Dionysius Thrax indicated that Aristarchus had changed the earlier reading ἀντιόωντες (γεγραμμένου ἀντιόωντες: whereas previously it was written ἀντιόωντες), replacing it (μεταθεῖναι) with the reading ἀμφὶς ἐόντες. It is worth noting the parallel in the use of the verb μετατίθημι to underline the textual change: μεταθεῖναι here, μεταθέσθαι in the previous case. There is still (in my view) a serious problem in interpreting exactly what is meant by the expression διὰ τῆς ἱπποδρομίας, but this is not of crucial importance for the question at hand and it can be set aside, postponing its examination to another occasion.21 What is relevant here is that the scholium informs us that this Aristarchean textual choice and the reasons lying behind it are discussed by Dionysius Thrax in his monograph Against Crates. In the scholiastic-erudite tradition, with its brachylogic annotations, information on Aristarchus’ second thoughts is in most cases restricted to the fact that two different choices are provided for the same place. Rather less frequently (probably on account of reduction of the exegetic material during formation of the scholiographic corpora and the choices made by the epitomators) traces remain of a discussion among later philologists who were seeking to determine at what stage of the situation Aristarchus had ceased his investigations, what had prompted his decision in favour of one choice or another, or in what circumstances he had second thoughts or changes of opinion on textual choices.22 In the case of Il. 6.4 the textual change is explicitly stated as resulting from the discovery of a better reading, and Il. 9.222 mentions that Aristarchus’ cautious attitude made him reluctant to be swayed purely by his own conjecture and he was therefore unwilling to change the reading found in many copies he had consulted.23 In philological terms, this is effectively equivalent to use of the  21 Discussion of the problem in the bibliography quoted by Montanari 2000a, 485, n. 13 and 14 (= this volume, ch. 38). It was apparently believed that Aristarchus favoured ἀμφὶς ἐόντες on the basis of a comparison with one or more lines belonging to the episode of the chariot race during the funeral games held in honour of Patroclus in Book XXIII (Erbse, ad loc., indicates Ψ 330 vel 393): but in my view the terms of the Aristarchean comparison seem somewhat unclear and further investigation is required. 22 In previous works I examined some issues concerning the form and characteristics of the Alexandrian ekdosis, where I focused in particular on the long-standing and hotly debated question of the number and sequence of Aristarchus’ editions and commentaries of Homer, and also devoted attention to the problems raised by Ammonius’ testimony on the second ekdosis of Aristarchus (Il. 10.397‒399 and 19.365‒368, with the associated scholia): cf. Montanari 1998; 2000a; 2002 (= this volume, ch. 21); 2004 (= this volume, ch. 41); 2009a (= this volume, ch. 7); 2011; 2015 (= this volume, ch. 10); 2015a (= this volume, ch. 9); Montana 2012, 49‒53; 2014. 23 Cf. Rengakos 2012, 244‒248.

  Aristarchus’ Conjectures (once) Again direct tradition for text constitution; the question of utilisation of the indirect tradition has already been addressed above. The situation is aptly summarized by the words of Rengakos:24 Der Rückgriff der Alexandrinischen Philologie auf die direckte und die indireckte HomerÜberlieferung ist also in den Scholien selbst bezeugt. Ausdrückliche Zeugnisse der Konjekturalkritik der Alexandriner sind leider in den Scholien nicht mit Sicherheit zu identifizieren.

This overall evidence indicates that when engaging in text criticism, the Alexandrians — whose methods reached their most refined achievements with Aristarchus — based themselves not only on text-internal conjectural proposals but also on external and diplomatic resources, consisting in choice among variants they found or noticed in a non-univocal tradition composed of the copies they had available and were thus able to consult. It would seem, therefore, that the burden of proof is on whoever seeks to strip the Alexandrian grammarians of any knowledge of variants deriving from collation of copies, attributing to them only arbitrary conjectures, rather than the opposite: the fact is that we do have, at the very least, convergent evidence in favour of knowledge of variants — and I would go so far as to say that we have real proof.25 In a nutshell: we must make it clear once and for all that we are dealing with a problem of principles and methods, not of the quantity of the data (number of collated copies or of variants discussed) or of the quality of the results (right or wrong from our own point of view). We are not concerned with establishing the minimum number of copies to be subjected to comparison or of variants to be taken into consideration before one can even begin to speak of philology, nor with determining how many “correct” readings or “good” interpretations are needed before it makes sense to speak of philology. The tendency (inappropriate, in my opinion) to scoff at the opinions of the Alexandrian philologists in terms of modern Homeric studies should by no means translate into the tendency to discredit their historical significance, which needs to be correctly positioned and contextualized. It is mistaken to blur the distinction between the two planes.26

 24 Rengakos 2012, 247. 25 See most recently Rengakos 2012; Montana 2012. 26 Janko 2002, 658‒662, seems to render this concept explicit rather more clearly. His position on the methods of the Alexandrian philologists is not an extremist unilateral stance: he believes that the majority of their readings are indeed arbitrary conjectures (by the Alexandrians themselves or possibly of more ancient origin), but he does not go so far as to deny the recourse

Aristarchus’ Conjectures (once) Again  

Rather, in a historical perspective, all that was needed in order for there to be a decisive step forward in intellectual achievement was the very fact of understanding and addressing the problem, even if only partially, erratically and incoherently. The crucial milestone was the realization that a literary text had a multifaceted history of transmission, during which it could become distorted at various points; the correct text (i.e. what is authentic versus, what is spurious, and what was the original wording) could then be restored by conjecture or by choosing the best reading among those offered by a divergent tradition.27 The idea of the recognition of damage in a text and of finding a way to repair it reveals that the organic unity between interpretation and textual criticism had become established. Although much progress still remained to be made, and Wolfian scientific philology, the modern critical edition and scientific commentary were still in the distant future, our viewpoint — far from being an anachronism — is the historical evaluation that a nodal step was taken in the period from Zenodotus to Aristarchus.

 to manuscripts and comparison among copies as part of their ekdosis work (for Zenodotus, Janko 1992, 23: “His caution was salutary, given the abundance of interpolated texts; he certainly had MS authority for some omissions”; for Aristarchus, Janko 1992, 27; 2002: “This [i.e. sch. Il. 9.222] certainly implies that Aristarchus did check manuscripts for variant readings”). On the one hand, Janko argues, there stands the problem of the origin of their proposed text choices (subjective emendation, comparison among copies) and therefore of their working procedures; on the other, he points out, “my own concern, as a Homerist, has always been whether such readings are authentic”. Perfectly clear: modern philologists can to some extent be severe regarding the opinions of the Alexandrians, considering them to be fairly acceptable or fairly unacceptable from their own point of view (Janko is very negative: “what counts is that they are conjectures, and nearly all bad”: 1992, 24; 2002, 661), but they cannot downplay the fact that the ancient Alexandrians emended and compared exemplars to correct the Homeric text, a method that combined interpretation of the text with awareness of the history of the tradition. An extremely apt remark, perfectly applicable to Alexandrian philology as well, is offered by Cassio 2002, 132, on the issue of pre-Alexandrian criticism: “The earliest scholarly approach to the Homeric text is totally foreign to us … we do right to think along very different lines, but we should never forget that it was the commonest approach to the Homeric text in the times of Socrates, Plato and Aristotle. As a consequence, we ought to be wary of looking at it with a superior smile, and ought to try to understand its motives in more depth instead”. 27 See now Conte 2013, 44‒50.

 Remarks on the Citations of Authors and Works in Ancient Scholarship One well-known characteristic of the commentaries that originate from the exegetic-philological work of the erudite grammarians of the Hellenistic-Roman age is their frequent and abundant inclusion of citations of texts by writers who are distinct from the authors of the commentaries themselves. This widespread practice came about for the most disparate reasons, that can be ascribed to three major categories: in some cases, because the cited text was helpful for interpreting and understanding the passage that was being commented on, and therefore the citation fulfilled a practical and direct purpose; or because the cited text contained an item that was of interest from the point of view of content, the Realien, the lexicon or the language, and therefore the citation formed part of more general and decidedly erudite approach and was not prompted merely by a strictly exegetic and more immediate line of investigation; or else because the cited text — which at times underwent distortion and misrepresentation — served to corroborate a certain position. Thus what we are dealing with are loci similes properly speaking or in a broader sense, or sometimes passages that are helpful for a better understanding of the text forming the object of the commentary, or elements that may provide additional information and erudite knowledge of a highly varied nature, or even passages adduced in support of an “ideological” aim. In the latter case, the “ideologically oriented” citations take as their starting point the actual text being commented on but then use it as a pretext for launching into a broader argument, which may on occasion even involve tweaking or refashioning the cited texts.1 This phenomenon prompts an important historical-cultural observation, which is worth taking up again in its entirety. I will begin by repeating a remark that has been made on numerous occasions and is by now obvious: if we did not have fragment P.Amh. 2.12 (MP3 483) we would be completely unaware that Aristarchus of Samothrace dealt specifically with Herodotus, and that far from being sporadic and occasional, Aristarchus’ work was so thorough and systematic as to inspire him to write a veritable hypomnema, at least on the first book.2 Had we no knowledge of the existence of this work by the prince of the Alexan 1 With regard to the citations in ancient erudite works, the work of Tosi 1988 remains indispensable; more recently, see Perrone 2010. 2 To the second as well, or so it would appear: cf. Matijašić 2013; see also the contribution by Montana 2009. https://doi.org/10.1515/9783110772371-012

Remarks on the Citations of Authors and Works in Ancient Scholarship  

drian philologist-grammarians, there would be consequences for the idea that would take shape in our minds concerning the development of philology in the Hellenistic age. In short, we would end up with a rather mistaken assessment and an inadequate understanding of the historical-cultural phenomenon in itself, but also of its overall implications: specifically, we would fail to appreciate the role of Alexandrian philology within the history of the fortune of Greek historiography in particular, but also of Greek literature in a general perspective, because historiography is known to have been the first prose genre taken into consideration by the Alexandrian philologists after the other genres of poetry. This small papyrus became known in 1901 thanks to the editio princeps of Grenfell and Hunt.3 Their presentation allows insight into a textbook example of the change in perspective that has taken place over the last hundred years or so, characterised by greater awareness as well as enhanced utilisation and more sophisticated appraisal of the works of erudite ancient Greek literature. The short early twentieth-century introduction by the two papyrologists concludes matter-of-factly as follows: “The chief points of interest in the fragment are the proof of a hitherto unknown variant ἅμιπποι for ἄνιπποι in chap. 215 and a new quotation from the Ποιμένες of Sophocles”. For Grenfell and Hunt, the commentary is of interest with regard to two aspects: 1) what it tells us about the text forming the object of the commentary, namely Herodotus: textual variants and in some cases exegetic elements of particular relevance; 2) the discovery of new fragments of lost works. On the commentary itself, as a work of criticism in its own right, only one remark is offered: “Unless the papyrus gives only a series of excerpts from Aristarchus’ commentary, which is not very likely, that work must have been extremely brief”. Thus the commentary is held to have been short and methodologically unsystematic, which obviously implied that it was only of limited value. Today, however, the progress of research has shown that the commentary was actually composed of excerpta of the work of Aristarchus,4 and was originally far more  3 Grenfell/Hunt 1901, 3–4 and Plate III; cf. Viljoen 1915, nr. V, 17–22; Paap 1948, nr. 10, 37–40. The text of the papyrus is reproduced in the apparatus in the Teubnerian edition of Herodotus of Rosén 1987, I 135–136; there will be a new edition, to appear in the section on Herodotus, edited by Fausto Montana, of the Commentaria et Lexica Graeca in Papyris reperta (CLGP). On the fragment, cf. Pfeiffer 1973, 348–349; Dickey 2007, 54; Montana 2009, 166–167; West 2011, 77–81; a somewhat offbeat position is taken by Hemmerdinger 1981, 154 ff. See also the bibliography cited in the following n. 4 Grenfell/Hunt 1901 defined the attribution to Aristarchus as “almost certain” (p. 3); Haslam 1994, 45 n. 169, states that he is sceptical, while in The Oxyrhynchus Papyri, vol. LXV, London 1998, nr. 4455, he seems to be fairly convinced (cf. p. 55 and p. 57, note on Col. I 12); no doubt is

  Remarks on the Citations of Authors and Works in Ancient Scholarship extensive and richer. We have the remains of an erudite linguistic-literary commentary, which proceeds to explain the Herodotean text by adducing parallels from poetry (Homer, implicitly, in the form that has been conserved, Sophocles transcribed explicitly) both as regards references to content and on facts pertaining to form. This was a veritable kind of Sachphilologie that was designed to assist the literary interpretation, with careful attention to the contents of the text and to Realien, and with citations taken from other authors.5 In short, one can convincingly argue that the Alexandrian philologists’ treatment of the historians (at least the major figures, starting with Herodotus) was strikingly similar to their work on the poets: that is to say, they adopted the Sprach- and Sachphilologie approach.6 The early editors said nothing about the history of Alexandrian philology as a historical-cultural phenomenon, despite coming face to face with the evidence of a previously quite unknown work — moreover, one by none other than Aristarchus himself. In 1901 R. Pfeiffer’s History of Classical Scholarship was still in the distant future, and would not come out until 1968; interest thus focused above all on the two elements mentioned above: namely useful information concerning the author of the work commented upon, and the new fragments of lost works. Since then this papyrus (like others of the same kind) has frequently received attention, not only from scholars of Herodotus but also and perhaps above all from scholars whose interests focus mainly, as noted, on the History of Classical Scholarship. In the world of classical scholarship, erudite literature has long been regarded as having more than simply — or predominantly — ancillary relevance that would be of interest only in connection with the works of the various authors mentioned in the text, or with the texts forming the object of commentaries or cited in the commentary. Rather, erudite literature is seen and studied as a set of works and literary genres that testify to — and thereby identify — a cultural and intellectual phenomenon of primary importance in the overall panorama of the ancient Greek and Latin world. Today we are actively working  expressed by Montana 2009, 166–167; 2012a, 72–76, with all the bibliographical references, to which the material cited in n. 2 should be added. I myself can see no reason why there should be any doubt concerning the explicit subscriptio of the papyrus, especially once it has been definitively ascertained that what we are dealing with is not the original work but rather excerpta: this realization should calm the nerves of those who would wish to have something that could be judged, as it were, to be of a level appropriate to the prince of ancient philologists, or rather, as one imagines that figure to have been, although in the case in question such an assessment completely overlooks that we have no original work of Aristarchus. 5 Montana 2012a, 72 and n. 5. 6 Cf. Pfeiffer 1973, 349–350; Montanari 2013 (= this volume, ch. 53).

Remarks on the Citations of Authors and Works in Ancient Scholarship  

in this cultural-historical perspective, which also enhances awareness concerning the function of the presence of citations from other authors in the exegeticerudite materials. No longer are such citations considered to be of value purely for their assistance in restoring fragments of otherwise lost works or in enriching the interpretation of the work being commented on: there is now a more general realization that the phenomenon of cited passages should more properly be seen as an intrinsic feature shaping the development of Alexandrian philology and its historical-cultural influence. Accordingly, in-depth attention should focus carefully on their presence and function. The Alexandrian philologists directed their interest first and foremost to the poets. Pfeiffer believed this was due essentially to the fact that philology was regarded as a creation of poetry, but other factors are very likely to have played an equally important role: for example, the fact that poets were felt to be the main repository of the Hellenic paideia, the element that Aristotle had identified as the culture and education handed down from the past and constituting a unique heritage of paramount importance. During the Alexandrian age the problem arose of conserving and interpreting the fundamental texts of this paideia, starting from Homer. Thus the philologists began with the poets (this was the procedure adopted by the pioneers Philitas and Zenodotus, and then by all their successors), but they soon also began to take an interest in the prose writers: the historians (Herodotus and Thucydides) and the orators (most prominently, Demosthenes, but also Isocrates, Aeschines and others), as well as the scientists (in primis medicine with Hippocrates). As I think I showed some time ago (prompted by a few remarks pointing to an interest that can be noted in Aristarchus), attention finally turned to the “contemporary writers”, i.e. to the great Hellenistic poets, on many of whom — or perhaps on all of the main figures — the grammarian Theon, a contemporary of Didymus, is known to have worked and written commentaries.7 Thus by the age of Augustus, philology and grammar had over a period of three centuries achieved a vast range of activities that covered almost every genre of authors and works of poetry and prose of Greek literature of the previous eras, from Homer up to and including the Hellenistic age. This achievement allows insight into the meaningfulness of this intellectual sphere and the enormous amount of work produced; it also allows us to gain an idea, however vague, of what we have certainly lost, an awareness that should caution us against fallacious deductions ex silentio. Interpreting texts and language with the armaments of philology and the science of language had become an established working procedure, the importance of which  7 On this cf. Montanari 1995a; 2002b (= this volume, ch. 58).

  Remarks on the Citations of Authors and Works in Ancient Scholarship clearly emerges not only from the conscious and stable use of the tools belonging to the philological-exegetic method, but also from the breadth of its scope, the objects of its interest and the materials utilised and discussed.8 Given the broad range of aspects composing this background, the observation we started out from, namely the noticeable presence of citations of other authors and of countless kinds of works in all the texts that have come down to us, should be no cause for surprise. What these citations contain and transmit in various ways are materials deriving from the philology and ancient erudition of the Hellenistic-Roman age: scholiographic corpora, lexicographic collections, grammatical treatises, the various works that gathered together erudite collections. The passages contained in the citations were certainly considerably more numerous and more abundant than the material we have today, because, as becomes clear by examining the sweeping centuries-long tendency to epitomation that took place during the transmission process, the first element to fall was precisely the citations of other authors, which were gradually reduced and eventually eliminated. The extent of the evidence we have available today gives us a rather limited picture of the intensity and massive body of knowledge of poetic-literary texts and the enormous effort made to utilize them for a range of exegetic, informational and erudite purposes. Yet we should have clear awareness that this is one of the important aspects for assessing the level achieved by ancient philology and above all its significance as a historical-cultural phenomenon. And this is the theme I am trying to highlight here. As Serena Perrone has written: “One of the most natural procedures for anyone wishing to study and interpret a literary work is certainly that of comparing the text in hand with the content of other works, first and foremost of the same literary genre, in order to take note of similarities and differences. By checking for resemblances and divergences in relation to other literary products of the same kind, one can distinguish the features that are proper to that particular work, recognise the continuity that links it to the tradition and pinpoint the novel elements that differentiate it. The use of such an approach appears to have been more than simply episodic even as early as in ancient Greek exegesis, where the σύγκρισις among different authors and works and the technique based on collecting parallels represented an interpretive tool, not least for problems concerning the attribution and authenticity of texts. Furthermore, not only did this give rise to veritable monographic treatises, but the comparative approach and the search for literary parallels must have operated also, and above all, in the drawing up of

 8 Montanari 2011, 11–24; 2015a, 25–44 (= this volume, ch. 9); 2015, 641–672 (= this volume, ch. 10).

Remarks on the Citations of Authors and Works in Ancient Scholarship  

detailed commentaries on texts, and it could certainly also have consisted in explicit references to other authors and other works and in citations as well”.9 I recalled at the beginning that citations can broadly be divided into three categories: those intended for an exegetic purpose in the strict and immediate sense; those having a fundamentally erudite aim, when the cited text offered an interesting comparison from the point of view of the content, lexicon or language; those fulfilling an ideological aim, when a passage is cited — and sometimes manipulated — in support of a given position. Thus, in a more extended perspective, these were passages that were felt to be helpful at various levels in order to improve the understanding of the text commented on, but also to supply additional elements of cultural information, taking a feature of the reference text as a starting point. Thus such a scheme outlines a first preliminary classification of the materials contained in the three types mentioned. In general, as is made clear by these remarks, it is by no means easy — indeed, it can be an arduous task — to survey and give a concrete overview of the passages taken from the major authors that are scattered around in innumerable different places in the exegetic-erudite texts. The many hurdles inevitably cast a shadow of grave doubt even on any attempt merely to estimate simple quantitative data, whatever means of computation is adopted. A very superficial assessment will show that Homer is the overwhelmingly predominant source of citations, and will confirm, at least as a general tendency, the rule that the author commented upon is the one most cited in the scholia concerning him. Taking our cue from this observation as an opportunity to go back to the commentary on Herodotus from which we started out, we can say that with regard to exegetic-erudite texts preserved on papyrus, our quest is facilitated by the fact that the MP3 on line database10 allows a search by author cited in the papyri, yielding a result that is fairly manageable for a detailed and in-depth analysis. But for the complex and extensive collections of exegetic-erudite works transmitted by the medieval tradition, the situation is far more difficult and complicated. To give just one example, a search on the citations of any given author in one or more scholiographic corpora or lexica carried out with TLG (for instance the citations of Euripides in the scholia to the Iliad or in the scholia to various authors, in the Etymologica and other lexica) is likely to yield a fairly unreliable quantitative result, because it will include, at the very least, doubles due to different redactions of scholia, conjectural occurrences and possibly even citations adduced as parallels by the editor (depending on how the text was codified in the database).  9 Perrone 2010: the passage in question is on p. 85. 10 Website: http://promethee.philo.ulg.ac.be/cedopal. See the case studied by Perrone 2010.

  Remarks on the Citations of Authors and Works in Ancient Scholarship It is possible to use the indexes of the editions of the various exegeticerudite collections, but it is widely known that a considerable number of scholiographic corpora, lexica and grammatical works are still available for consultation only in old or sometimes very old editions, the indexes of which may be unreliable; thus the quality of the editions and their indexes often does not allow a precise computation without specific checks on the actual texts themselves. This step, however, soon makes it clear that what we are dealing with is a typology that allows cases in which the reader is referred to author + work + text of the passage as well as cases in which one or two of these three elements are absent, generally because they have been lost over the course of the tradition. For a given author whose name is mentioned, we will have the citation of works that have been preserved or of fragments of works that have not been preserved: such information is of great interest for the history of the reception of the author in question. There will also be cases in which the reference relies on quintessential archetypal features: Homer cited as ὁ ποιητής or Aristophanes as ὁ κωμικός. An example that can be mentioned without major difficulties is that of the citations of Aristophanes in the scholia to the Iliad, where the comic playwright is cited 12 times by name and 15 times (i.e. more often) without any name. In this case we are helped by the existence of Erbse’s edition with good and well-structured indexes, which, however, does not include the so-called Sch. D and other material of the Iliadic scholiography.11 Moreover, it is only too easy, at this point, to raise the question of cases where omission of the author’s name leads to a doubt concerning attribution and, on the other hand, cases where all that is left is the name or little more, so that it becomes quite impossible to grasp the implication and sense of the citation. At present we can have, at best, a very approximate idea of what was evidently a vast and impressive phenomenon, which we cannot appreciate in its true original significance inasmuch as it appears before us in a dramatically impoverished form, enormously reduced in comparison to the proportions it assumed in the original works of the Hellenistic-Roman era. Moreover, it must not be forgotten that any overall examination should also carefully assess the contribution of Late Antique and Byzantine erudition, which independently continued to extract and cite passages taken from the works that had been preserved. At the current state of knowledge, a purely quantitative picture of the  11 I am referring to class h scholia, of which Erbse made incomplete use, and only for book I, whereas for all the rest of the poem they are yet to be examined. Davide Muratore is carrying out an overall examination of the manuscript tradition of h, in the framework of a new complete edition of the Scholia in Iliadem directed by this writer.

Remarks on the Citations of Authors and Works in Ancient Scholarship  

presence of individual authors and works in the erudite literature can be constructed in a highly superficial and approximate manner, by very patiently gathering and carefully putting together the data obtainable from the various scholiographic corpora, the lexicographic collections and the grammatical works, while keeping in mind the by no means minor problem of the reliability — or otherwise — of the various editions and their indexes. That is to say, a mere search by name using TLG would, at the very least, lead to numerical results with a high likelihood of being “inflated”, not to mention the loss of the citations lacking the author’s given name, and this is certainly a feature not to be overlooked. But a purely quantitative picture would also be no more than a preliminary result, a starting point for a more detailed and more reliable analysis, though I would argue that this is feasible only for individual works of erudition and for individual authors (for example, the citations of Aristophanes in the scholia to the Iliad, the case already mentioned above) because such an approach would require each citation to be examined and understood in detail. In short, what the above remarks offer is a basis for delineating a typology of citations of authors and works in the erudite literature, a reference grid on which a more allencompassing picture can be built up.12 Notwithstanding these reservations, this is a theme that is well worth taking into consideration in the overall framework of research on exegesis and erudition in the ancient world, and it certainly deserves new studies to address these numerous issues. The presence of citations of authors and works of various types in the texts that have come down to us, which transmit materials deriving from the philology of the Hellenistic-Roman era, also seen through the filter of Byzantine culture, constitutes one of the most important testimonies that allow an in-depth and correct assessment of the level achieved by philology in the Hellenistic and imperial ages, together with a clearer awareness of its impact as a historical-cultural phenomenon. I believe one can legitimately assert that while this is not the aspect most commonly invoked in connection with the question of citations, it should be awarded greater attention and studied with a more carefully focused approach.

 12 See the concluding remarks of Perrone 2010, 100–102.

 Scoliografia e teatro greco: qualche appunto Sono molto contento (e naturalmente pieno di gratitudine per gli amici che mi hanno invitato) del fatto che questa occasione mi abbia stimolato, direi addirittura costretto, a concentrare per un poco la riflessione su un tema, al quale (preso da altri interessi) ho pensato più volte in modo desultorio e francamente casuale, senza mai obbligarmi a delineare con un minimo di organicità per lo meno un rudimentale quadro di riferimento. Cercando di fissare qualche elemento, mi sono naturalmente accorto di quanto c’è di provvisorio, incompleto e forse anche caduco nello stato attuale cui sono giunto per ora. Il tema comunque, a prescindere dal relatore, è molto ricco di contenuti e capace di offrire svariate possibilità di analisi e di risultati: varrà senz’altro la pena riprenderlo e studiarlo nel suo insieme e in tutti i suoi numerosi aspetti. Questi sono soltanto pochi e limitati appunti. Voglio precisare che, benché il titolo parli per brevità solo di «scoliografia», nel mio pensiero intendo comprendere (almeno come riferimento ideale complessivo) un campo di osservazione esteso a tutto quanto possiamo chiamare «letteratura erudita» in generale: dunque, ai corpora di scolî d’ogni tipo si affianca per lo meno tutto quanto rientra nell’ampio dominio della lessicografia antica; e poi si può aggiungere ancora ogni altra opera che a buon diritto possa rientrare in quella categoria, come per esempio trattazioni grammaticali, raccolte di paremiografi, compilazioni e scritti miscellanei di vario genere, e così via. Il primo posto comunque, nella ricerca di cui sto per parlare, spetta al materiale rappresentato dalla scoliografia e dalla lessicografia. E si badi bene che, nonostante una certa abitudine — per vero dire, ormai in via di sparizione — a considerare tutto ciò come letteratura di secondo piano e di minor valore, si tratta invece di una più che considerevole fetta della civiltà letteraria greca, la cui importanza storico-culturale è assai rilevante: e per fortuna mi pare sia sempre meglio riconosciuta, sempre più studiata e messa in luce. Gli studî sulla letteratura erudita scoliografica, grammaticale e lessicografica stanno conoscendo, se non mi inganno, sviluppi notevoli e rilevanti fortune, sulle cui ragioni sarebbe interessante riflettere, se fossimo nella sede adatta. Senza esagerarne la portata, parrebbe questo uno degli antidoti, o forse meglio uno (dico uno, non certo il solo) dei salutari correttivi a certi mali degli ultimi decenni: da una parte, quelle che si sono talora rivelate come incontrollate fughe in avanti metodologiche, cariche del rischio di approdare a scatole vuote;  Ringrazio Fausto Montana per l’aiuto nella raccolta dei materiali. https://doi.org/10.1515/9783110772371-013

Scoliografia e teatro greco: qualche appunto  

ma anche attardati neopositivismi, occupati nello sforzo assiduo di trarre dai testi antichi il meno possibile di succo culturale e intellettuale, magari a braccetto di neoestetismi un po’ penosi oppure di ideologismi ammantati di scienza sbandierata e carichi di livore puerile. Il fatto che la letteratura erudita sia un campo difficoltoso e ostico, nel quale non è facile muoversi con disinvoltura fino dai primi approcci, rende ancora più importante avere a disposizione buone edizioni, con testi attendibili e ricco materiale di confronto, sulle quali basare gli sforzi di ricostruzione di un quadro molto ricco e pieno di interrelazioni da ritrovare. La fatica per approntare queste edizioni di solito è grande, e la competenza per affrontarle è frutto di sforzi lunghi e complicati. Esiste una filologia specifica di questo genere di testi, soprattutto per quanto riguarda l’erudizione scoliografica e lessicografica, e se ne possono delineare dei principi e forse una «normativa», certamente non identica a quella che concerne le opere degli Autori (con la maiuscola), se non altro perché si tratta di opere «aperte», caratterizzate nel corso della tradizione da continue fluttuazioni, osmosi e manipolazioni di materiali trattati con libertà non piccola, con tutte le conseguenze che questo comporta. Sotto questo profilo, c’è ancora moltissimo da fare, e spesso si lamenta per testi importanti la mancanza di edizioni buone o almeno attendibili: ma è confortante vedere che molto si è fatto e si sta facendo, non poche edizioni di corpora di scolî, di lessici ed etimologici, di testi e raccolte di frammenti grammaticali sono pubblicati oppure sono in corso o in preparazione. Tutti sappiamo bene come l’avere a disposizione un’edizione critica attendibile sia il presupposto irrinunciabile per approfondire validamente lo studio, e molti desiderata di vecchia data trovano o sono sulla via di trovare soddisfazione. Credo si possa dire senza tema di smentite che, in questo settore degli studî sul mondo antico, la nostra è sicuramente un’epoca di progresso e di fioritura, e ci sono i presupposti perché tale progresso continui: molte nuove ricerche si producono nel campo della storia della filologia, della grammatica e dell’erudizione antica in diverse parti del mondo, sia analisi di problemi circoscritti che sintesi su problematiche più o meno ampie. Non posso certo produrmi qui in una bibliografia, ma il fenomeno è ben visibile scorrendo ciò che si viene pubblicando nel piccolo mondo degli studiosi dell’antichità. Una riflessione sui temi appena accennati sarebbe, credo proprio, un interessante capitolo di storia della filologia, ma non è possibile farla in questa sede. Ancora una considerazione voglio aggiungere. Da qualche tempo il Thesaurus Linguae Graecae (TLG) dell’Università di Irvine in California ha cominciato a inserire nella sua banca dati (ora disponibile — se pur non per intero — anche su un CD-ROM periodicamente aggiornato e agevolmente utilizzabile con un PC

  Scoliografia e teatro greco: qualche appunto da singoli studiosi) anche i corpora di scolî, il che significa che fra non molti anni questa considerevole fetta della letteratura greca sarà a disposizione su un supporto magnetico, che ne permetterà un uso domestico e un esame sistematici, prima difficili da ottenere senza enormi dispendi di tempo e di fatica. Intanto (scusate l’autocitazione, anche se solo parziale) ha visto finalmente la luce la prima (salvo errori) concordanza in volume di un corpus di scolî, cioè quella degli Scholia Vetera a Pindaro, realizzata a Pisa. Cominciammo a lavorarci nel 1975, quando questo genere di lavori era molto più raro e iniziale di ora, e quando una concordanza realizzata con il computer per un corpus di scolî era certo un lavoro pionieristico, irto di difficoltà: fa piacere constatare come ora si trovi in totale sintonia con la Stimmung dei tempi.1 Non voglio dilungarmi troppo: mi pare che il nostro secolo abbia raccolto e stia raccogliendo frutti copiosi in un terreno dissodato tenacemente dalla filologia dell’Ottocento e del primo Novecento. Ma se facciamo attenzione alle date di pubblicazione delle edizioni e degli studî, ci accorgiamo che molto è avvenuto proprio in questi ultimi decenni, il panorama è considerevolmente cambiato diciamo nell’ultimo venticinquennio, c’è stato un fiorire evidente,2 che è ancora pienamente vivo: molti grossi e importanti lavori sono attualmente in corso, ed è più che probabile che questo scorcio del Novecento e l’inizio del prossimo millennio ci portino ancora diverse acquisizioni, sia per quanto riguarda le edizioni di testi, sia per quanto riguarda gli studî e le ricerche, che vivono in simbiosi con il lavoro editoriale, lo stimolano e ne sono stimolate. Dopo tutta questa noiosa premessa, che vi ho inflitto giusto per suscitare immagini nella vostra «memoria di biblioteca», cerchiamo dunque ora di aver presente davanti agli occhi quanti corposi tomi, quale cospicua porzione, anche soltanto in termini quantitativi, di cultura letteraria greca si celi dietro la discreta etichetta di «letteratura erudita»; e quante informazioni, quanti dati utili alla comprensione storica della civiltà greca antica. Come è stato considerato e utilizzato questo patrimonio nella tradizione dei nostri studî? Volendo schematizzare per sommi capi, possiamo dire che della letteratura erudita sono stati dati due tipi di lettura, che rispondono a una duplice funzione, e hanno portato e portano entrambi frutti copiosi e importanti. Da una parte, la letteratura erudita è stata utilizzata in funzione di altro, cioè in primo luogo in funzione degli Autori (con la maiuscola): è stata dunque vista essenzialmente come una miniera di  1 Arrighetti/Galvani/Montanari 1991. A questo proposito bisogna ricordare anche: Lomant/ Marinone 1990. 2 Non credo di far torti a nessuno dicendo che molto appare dovuto all’attività di uno studioso come Hartmut Erbse e all’impulso che essa ha dato in varî ambienti.

Scoliografia e teatro greco: qualche appunto  

citazioni, cioè soprattutto come serbatoio di frammenti di opere perdute e anche di tradizione indiretta dei testi conservati; e inoltre come portatrice di dati esegetici antichi più o meno utili all’interpretazione filologica moderna; in funzione di altro, infine e ovviamente, anche in quanto tesoro di dati linguistico-lessicali e storico-antiquarî d’ogni genere. L’altra possibilità è quella per cui la letteratura erudita viene letta, studiata e valutata anche per se stessa, e riveste un valore autonomo in quanto testimonianza storico-culturale di un intero settore della civiltà letteraria greca antica, quello della letteratura critico-esegetica, dell’interpretazione dei testi, del pensiero grammaticale, dello studio della lingua, della filologia nel senso più ampio del termine; e inoltre prodotto letterario in parte coincidente con e in parte contiguo a settori quali la retorica e la critica letteraria, per arrivare fino alla poetica teorica e applicata. Se dopo il II sec. d.C. si può dire che la fase davvero metodologicamente creativa della filologia e della grammatica greca sia praticamente conclusa, l’attività erudita esplicò invece ancora a lungo la sua funzione specifica. Nei secoli successivi ci si dedicò soprattutto al lavoro di conservazione, epitome e rifusione, raccolta e compendio: commentarî e opere erudite di vario genere venivano riassunti e sintetizzati in nuovi prodotti miscellanei, riunificati in raccolte, repertorî, compilazioni manualistiche, ridotti in gruppi di annotazioni dossografiche talvolta ricche, talvolta assai scarne. Nell’arco dell’età imperiale romana e poi di quella tardoantica si svolgono le tappe decisive della conservazione di tutto questo materiale, che assume gradatamente le forme e le fisionomie che preludono alla formazione delle grosse raccolte di età bizantina, e quindi determinano ciò che noi ne conosciamo. L’ambito dell’erudizione è anche quello dei compendî, delle sillogi antiquarie ed enciclopediche, delle raccolte dossografiche, della produzione di «tesori» cui ricorrere, e quindi anche per certi aspetti manifestazione di «enciclopedismo» come abito mentale: tutti prodotti e tappe, dall’età alessandrina in poi, di uno sforzo e di una volontà assiduamente e tenacemente volti a studiare, raccogliere, riassumere, capire e conservare tesori di dottrina che costituivano la gloriosa eredità intellettuale e culturale di molti secoli, delineare e fissare un’identità culturale che doveva vivere, il portato di un’epoca che si sentiva passata ma anche presente con la sua forte e decisiva continuità culturale. E ora veniamo davvero allo specifico del nostro tema. Ci sono sicuramente molti modi per cercare, selezionare, organizzare, studiare, utilizzare, far parlare la massa di informazioni sul teatro greco che ci può venire dalla letteratura erudita. Qui farò due esempi estremamente diversi fra loro, dai quali vengono informazioni e argomenti di riflessione assai lontani. Ci sono naturalmente, come gli studiosi sanno bene, molte altre possibilità di indagine.

  Scoliografia e teatro greco: qualche appunto Ho deciso di cominciare con dei numeri, forse di per sé un po’ aridi e magari anche noiosi: tuttavia mi pare che non sia inutile fare qualche piccola considerazione anche solo quantitativa sull’entità del fenomeno. I numeri riguardano proprio la presenza delle opere dei maggiori autori teatrali, cioè Eschilo, Sofocle, Euripide e Aristofane, in importanti corpora di scolî, confrontata con la presenza nelle stesse raccolte di alcuni poeti particolarmente significativi, vale a dire Omero, Esiodo e Pindaro. Si tratta in sostanza di un primo tentativo di ricognizione statistica, che ci offre dei dati per il momento ancora del tutto parziali e soprattutto un po’ brutali, cioè ancora poco approfonditi, ma che già mi pare facciano intravvedere qualche elemento significativo. Se è così, sarà utile sviluppare questo lavoro di ricognizione per costruire un quadro sempre più ampio, e infine il più possibile vicino alla completezza, delle risorse offerte dalla letteratura erudita per i testi del teatro greco antico. Sono necessarie alcune avvertenze, a proposito dei numeri che sto per fornire. Si tratta di numeri sottoposti a diversi motivi di dubbio e che, nei loro termini assoluti, sono certamente imprecisi: essi derivano dagli indici che troviamo nelle edizioni degli scolî presi in esame, e quindi in primo luogo dipendono dalla qualità delle edizioni e degli indici stessi (che non sempre sono tali da permettere un computo preciso e sicuro senza un controllo puntuale, cosa che richiede molto tempo e non si è potuta fare per questa occasione). Poi, siccome in tali indici di solito non si riesce a distinguere chiaramente fra la vera e propria citazione e la semplice testimonianza, ho deciso, nei computi globali, di considerare tutto insieme, senza distinzione. Infine, bisognerebbe verificare una per una le citazioni, soprattutto quelle relative a frammenti e testimonianze di opere perdute, con le raccolte più recenti dei frammenti degli autori (penso ai tragici, ai comici, a Pindaro, a Esiodo). Dunque i numeri assoluti sono in realtà da prendersi di fatto come approssimativi, e a questo stadio del lavoro bisogna accontentarsi: ottenere dati precisi richiederebbe in realtà un esame assai lungo e difficoltoso, che si potrà condurre in fasi ulteriori del lavoro. I numeri dati qui di seguito conservano comunque, a mio parere, tutto il loro valore indicativo, perché mostrano senza dubbio delle proporzioni relative e delle tendenze che non potranno essere del tutto smentite o ribaltate. Avendo una griglia di riferimento, sarà poi agevole procedere pazientemente alle correzioni, sia precisando il più possibile i dati, che allargando la considerazione ai frammenti degli altri autori tragici e comici (il primo a cui penserei è Menandro), e naturalmente esaminando ancora altri scolî e altre fonti lessicografiche.

Scoliografia e teatro greco: qualche appunto  

Gli scolî presi in considerazione sono quelli a Omero, Esiodo, Pindaro,3 Eschilo, Sofocle, Euripide, Aristofane,4 Teocrito, Apollonio Rodio; a tutti questi ho aggiunto l’Etymologicum Magnum, per avere anche la testimonianza di almeno un testo lessicografico importante. Degli indici usati, i meno attendibili mi sono parsi quelli degli scolî all’Odissea e quelli degli scolî a Eschilo. I numeri in colonna danno la quantità di citazioni trovata per ogni autore: per Omero, il conto è spesso soltanto indicativo, dato il numero altissimo di occorrenze. Scolî a Omero - Iliade

Omero oltre . Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

- Odissea

Omero oltre  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Scolî a Esiodo

Omero  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Scolî a Pindaro

Omero  Esiodo  Pindaro  Eschilo  Sofocle 

 3 Per gli Scholia Vetera agli epinici di Pindaro ho sostituito i dati forniti dagli indici dell’edizione di Drachmann 1903–1927 con quelli provenienti dall’opera citata qui sopra alla n. 1: salvo un netto aumento del numero delle citazioni di Omero e di Pindaro stesso, negli altri casi le differenze non sono state grandi. 4 È stata usata l’edizione di Dübner 1877 per avere un dato d’insieme omogeneo, dato che l’edizione intrapresa sotto la direzione di Koster 1960 è ancora parziale.

  Scoliografia e teatro greco: qualche appunto Euripide  Aristofane  Scolî a Eschilo

Omero oltre  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Scolî a Sofocle

Omero oltre  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Scolî a Euripide

Omero oltre  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Scolî a Aristofane

Omero oltre  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Scolî a Teocrito

Omero  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Scolî a Apollonio Rodio

Omero  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Scoliografia e teatro greco: qualche appunto  

Etymologicum Magnum

Omero oltre  Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Somma complessiva

Omero oltre . Esiodo  Pindaro  Eschilo  Sofocle  Euripide  Aristofane 

Abbiamo dunque esaminato, se pur in modo un po’ artigianale, otto dei maggiori corpora di scolî ai poeti e un grande etimologico: in verità non mi sembra che, in generale, i dati che sono venuti fuori provochino sorprese particolari. Si possono fare comunque alcune osservazioni di un certo interesse. La somma totale delle citazioni rivela un’ovvia prevalenza di Omero, prevalenza di cui i numeri ci fanno percepire concretamente le proporzioni veramente notevoli (e questo malgrado il fatto che il dato omerico sia particolarmente impreciso: il numero straordinariamente elevato rendeva di fatto inutile perseguire un computo più scrupoloso). Esiodo si attesta al secondo posto, grazie anche all’apporto dell’Et. Magn., dove appare molto più citato dei tre maggiori tragici. Meno lontano di quanto forse ci si poteva aspettare è Pindaro, il lirico dallo stile oscuro. Fra gli autori di teatro, Aristofane deve il suo numero notevolmente alto pressoché esclusivamente agli scolî ad Aristofane stesso e all’Et. Magn. (dove è citato quanto Esiodo e molto più dei tre tragici e di Pindaro), mentre negli altri casi ha quasi sempre una posizione nettamente inferiore. Per quanto riguarda i tre tragici maggiori, la classifica è quella, credo proprio, che ci si apettava: al primo posto (e terzo in assoluto dopo Omero e Esiodo, al quale è peraltro molto vicino) c’è il notoriamente più diffuso e conosciuto Euripide, all’ultimo il più ostico Eschilo — ultimo in assoluto, con occorrenze che sono quasi la metà dell’altro autore «difficile», cioè Pindaro. Qualche osservazione salta subito all’occhio, considerando partitamente e comparativamente i diversi dati. Intanto, vediamo come Omero sia sempre l’autore prevalente, tranne che nel caso degli scolî ad Aristofane dove è superato dalle citazioni «interne», cioè tratte dallo stesso autore commentato: il divario non è grande e sono tentato di considerare questa piccola anomalia come sospetta e bisognosa di verifica, una volta che sia completata la nuova e più

  Scoliografia e teatro greco: qualche appunto attendibile edizione, ma per il momento rimane un’eccezione in cui prevale la citazione «interna» su quella del sempre amatissimo Omero. A parte Omero, risulta costante la regola per cui l’autore commendato è il più citato negli scolî (come abbiamo visto, nel caso di Aristofane questa sembra prevalere, allo stato attuale, anche sul consueto primato omerico): pare dunque di trovare realizzata in concreto la ben nota tendenza alessandrina a spiegare un autore in primo luogo con se stesso. Una piccola osservazione sugli scolî a Eschilo, per cui è stato usato l’indice della vecchia edizione del Dindorf:5 il materiale che se ne ricava mi sembra un po’ povero per quanto riguarda proprio le citazioni «interne», cioè di Eschilo stesso, ma anche qui ritengo probabile che il dato possa cambiare almeno un poco con un’edizione più attendibile. Abbiamo già detto che Aristofane (se si escludono gli scolî ad Aristofane stesso e l’Et. Magn.) di solito è citato meno dei tre tragici. Fra questi ultimi, Euripide prevale (oltre che nel totale, come abbiamo visto) pressoché costantemente (il dato anomalo degli scolî all’Odissea può essere, anche questo, piuttosto sospetto e forse destinato a cambiare con indici migliori): non ci sorprende che negli scolî a Sofocle sia superato da Sofocle stesso, secondo la tendenza già sottolineata; per il dato degli scolî a Eschilo, vale quanto ho detto sopra.6 Ma da questo materiale, per quanto brutto, voglio ricavare anche un’osservazione distinta e particolare a proposito della tradizione delle tragedie conservate di Euripide. Qualche anno fa, pubblicando un frammento papiraceo dell’Ecuba, sulla scia di altri studî sullo stesso argomento, facevo un computo dei papiri delle tragedie conservate in relazione alle selezioni della tradizione euripidea.7 Come è noto, mentre un gruppo di nove tragedie è trasmesso praticamente da un unico codice, la «scelta delle dieci» o «tragedie con scolî» è maggiormente rappresentata nella tradizione; ma di queste dieci, sette (Alcesti, Medea, Ippolito, Andromaca, Troiane, Baccanti, Reso) sono conservate in pochi manoscritti, mentre la cosiddetta triade bizantina, cioè Fenicie, Oreste, Ecuba, è conservata in numerosissimi manoscritti; e anzi queste tre tragedie sono in assoluto le più diffuse nei manoscritti medievali, testimonianza appunto

 5 La nuova edizione teubneriana di Smith 1976, 1982 è in corso e non ha ancora indici. Ancora per ragioni di omogeneità (come nel caso di Aristofane, cfr. nota prec.) non è stato usato Herington 1972. 6 Un gradino ulteriore nell’analisi di questo materiale è costituito, naturalmente, dalla distinzione fra le citazioni dalle opere conservate e quelle dalle opere perdute. Per ottenere un computo attendibile, bisognerebbe operare una verifica completa sia nei testi degli scolî stessi che nelle più recenti edizioni dei frammenti, onde evitare dei dati statistici troppo aleatori. 7 Montanari 1987.

Scoliografia e teatro greco: qualche appunto  

dell’ultima selezione. Bene, il computo del numero dei papiri noti fino ad allora dava questa classifica: – Fenicie 22 – Oreste 20 – Medea 13 – Ecuba 11 – Baccanti 11 – Ippolito 8 – Andromaca 8 Dunque, le tragedie più rappresentate nei papiri fino all’età tardo antica appartengono tutte alla scelta delle dieci, e ai primissimi posti compaiono proprio quelle della triade bizantina (Fenicie, Oreste, Ecuba), che dunque cominciava probabilmente a delinearsi già assai presto, anche se l’Ecuba in realtà appare in una posizione un po’ incerta. Vediamo cosa succede nelle citazioni degli scolî più l’Et Magn. da noi esaminati. Facendo la somma complessiva (le singole raccolte possono offrire situazioni particolari), la classifica delle tragedie più citate è la seguente: – Fenicie 62 – Oreste 53 – Ippolito 36 – Medea 33 – Baccanti 26 – Ecuba 21 – Andromaca 16 Le altre seguono con numeri inferiori. Sono dunque esattamente le stesse sette tragedie, che sono le più rappresentate anche nei papiri: l’ordine è differente, ma il primato delle Fenicie e dell’Oreste è in entrambi i casi ben saldo. In relazione alla triade bizantina, continuiamo a trovare l’Ecuba in una posizione piuttosto oscillante,8 in competizione (confrontando i due dati) soprattutto con la Medea e le Baccanti. Ma adesso basta davvero con i numeri. Lasciamo questo tipo di considerazioni, in attesa di possibili sviluppi e approfondimenti, e passiamo a un altro genere di ricerca, cioè ad analizzare modi del tutto diversi di presenza del «teatro» nella scoliografia. Facciamo naturalmente solo qualche esempio, pescato qua e là senza pretesa di essere originali.  8 Cfr. gli articoli citati alla nota precedente, con la bibliografia.

  Scoliografia e teatro greco: qualche appunto Platone nel III libro della Repubblica (392 e–394 c) distingue la poesia in tre tipi o modalità: puramente narrativa (diegesis), quando è sempre solo il poeta a parlare, che egli esemplifica con il ditirambo; mimetica, quando il poeta si nasconde sempre dietro ai personaggi (cioè si immedesima e parla attraverso di loro), come nelle opere teatrali; mista, quando ci sono parti narrate dal poeta e parti in cui egli si nasconde dietro i personaggi (vale a dire i discorsi diretti), come nell’epica, nella quale, togliendo le parti del poeta che stanno fra le parti dei personaggi, si ottiene la forma della tragedia. E infatti più avanti, Repubblica 595 c, Omero è definito da Platone il primo maestro dei poeti tragici: Omero «sembra sia stato primo maestro e guida di tutti questi bravi poeti tragici». Aristotele nella Poetica riprende di passaggio anche questa prospettiva, laddove (3, 1448 a 20 sgg., un luogo peraltro assai problematico) ripresenta il paragone fra l’epica (Omero), in cui il poeta un po’ parla per se stesso e un po’ diventa un’altra persona (cfr. 24, 1460 a 5 sgg.), e il dramma, in cui sono solo i personaggi a parlare e il poeta non compare. Sul piano storico-letterario Aristotele delinea (anche qui con il suddetto precedente platonico) la derivazione per così dire genetica da Omero delle forme drammatiche, la tragedia da Iliade e Odissea, la commedia dal comico Margite: cfr. 4, 1448 b 34 sgg.: «Come dunque Omero fu il massimo poeta nel serio (fu unico non solo per bravura, ma anche perché produsse imitazioni di tipo drammatico), così per primo fece intravvedere anche la forma della commedia, drammatizzando non il motteggio ma ciò che è ridicolo. Come l’Iliade e l’Odissea sono il corrispondente della tragedia, così il Margite lo è della commedia» (trad. D. Lanza). L’epica per Aristotele è imitazione di fatti nobili eseguita con la parola: poesia narrativa composta in metro di un solo tipo (l’esametro eroico, che l’esperienza ha mostrato essere il più adatto, tanto che altri metri non risulterebbero adatti a questo); diversamente dalla tragedia, essa tende ad ampliare la dimensione del tempo rappresentato (Poetica 5, 1449 b 9 sgg.: cfr. 23, 1459 a 17 sgg.), ma come la tragedia deve rispondere al requisito fondamentale dell’unitarietà dell’azione. A questa prospettiva si ricollegano numerose osservazioni presenti negli scolî omerici, che scaturiscono da una lettura dei poemi sulla base di categorie simili a quelle utilizzate per la tragedia, e che non è difficile esemplificare. Questo argomento particolare poi rientrò nella tematica generale, ben nota e assai diffusa, di Omero iniziatore e origine di ogni conoscenza, protos euretes di ogni dato culturale. Diamo soltanto qualche esempio, tratto appunto dagli scolî omerici.9 Compiutasi la famosa lite all’inizio dell’Iliade imbarcata Criseide per rimandarla al padre, Agamennone manda gli araldi Taltibio ed Euribate alla tenda di  9 Le citazioni degli scolî all’Iliade rimandano naturalmente all’edizione di Erbse 1969–1988.

Scoliografia e teatro greco: qualche appunto  

Achille per dare compimento alla minaccia e portar via Briseide. Gli araldi trovano Achille seduto presso la tenda e la nave, intimoriti e rispettosi «si fermarono, non gli dicevano nulla e non parlavano» (v. 332); Achille naturalmente capisce, saluta i due che, dopo un breve discorso dell’eroe, se ne vanno con la fanciulla. Commenta lo scolio a A 332b. «Dire qualcosa infatti sarebbe stato irritante, prestare omaggio sarebbe stato servile. Omero per primo introdusse nella tragedia i personaggi muti (πρόσωπα κωφά)». Dunque, l’analisi riferita dallo scoliasta porta a ritrovare in Omero l’inventore del κωφόν πρόσωπον teatrale. La prova dell’esercito nel II libro dell’Iliade, conseguente al fallace sogno di Agamennone, sta per portare a una fuga in massa degli Achei dall’assedio di Troia: «E allora contro il destino sarebbe avvenuto il ritorno degli Achei, / se Era non si fosse rivolta ad Atena con queste parole» (vv. 155–56). Atena scende dall’ Olimpo e convince Odisseo ad arrestare la fuga, che infatti è scongiurata. Lo scolio a B 156 commenta: «Il poeta spinge le “peripezie” (τὰς περιπετείας) fino a un punto tale, che solo il dio può cambiarne il corso. Per primo Omero insegnò ai tragediografi gli artifici delle macchine teatrali (μηχανάς)». Qui a Omero si attribuisce l’introduzione del deus ex machina. Superata la possibile sedizione e conclusa l’assemblea, dopo sacrifici di rito, si chiama a raccolta l’esercito per la battaglia; prima del celebre catalogo delle forze in campo, una splendida serie di similitudini cattura la possente immagine dell’armata in mezzo alla quale spicca il condottiero Agamennone, «negli occhi e nel capo simile a Zeus fulminatore, / simile ad Ares nella cintura, nel petto a Posidone» (vv. 478–79). Lo scolio a B 478–9a commenta: «I pittori perseguono il vero, i tragediografi il superiore (più nobile), i commediografi l’inferiore, cose che si trovano tutte in Omero: la commedia come nell’episodio di Tersite, la pittura come per esempio nel verso I 209: “Automedonte teneva fermo, Achille glorioso tagliava”, mentre in questa descrizione di Agamennone l’immagine è plasmata più bella e più sontuosa del reale». Qui l’idea della derivazione omerica delle forme teatrali è esplicitamente operante, ma vale la pena aggiungere un’osservazione. Se rimane costante, ovviamente, il rapporto della tragedia con l’epica, la derivazione omerica della commedia si arricchisce, in aggiunta al rapporto con il Margite già istituito da Aristotele, grazie alla messa in luce di elementi «comici» rintracciabili all’interno degli stessi poemi epici: l’episodio di Tersile è naturalmente l’esempio preferito.10 Gli scolî all’inizio dell’Iliade si soffermano molto sulla prima parola del poema, «l’ira funesta del Pelide Achille»: l’incipit con quella parola chiave, menin, la cui fama nella storia della letteratura è forse insuperata. Attira la nostra attenzione  10 Su questi aspetti cfr. Montanari 1988c.

  Scoliografia e teatro greco: qualche appunto lo scolio 1a, rr. 9 sgg.: «Omero iniziò dalla menis, poiché essa fu presupposto e principio degli avvenimenti; e soprattutto ha inventato il proemio tragico per le tragedie: infatti il racconto delle sventure ci rende attenti e partecipi...» Abbiamo esemplificato due modi molto differenti di rivolgersi alla letteratura erudita cercandovi testimonianze relative al teatro. Altre analisi, sicuramente interessanti, potrebbero vertere da una parte su quanto viene citato per commentare e fare un discorso critico a proposito dei testi teatrali, dall’altra sulle citazioni tratte da autori di teatro per commentare opere di alto genere. Poi alla letteratura erudita possiamo chiedere informazioni sulla storia della filologia dei testi teatrali, sulle interpretazioni degli antichi filologi, sulle idee e metodologie che sono state adoperate, sulle rappresentazioni e sulle tecniche della scena, e così via.11 È tutto un discorso complesso e ricco: in realtà, La letteratura erudita e il teatro non è un solo tema bensì una serie di temi da studiare.

 11 Sui vari aspetti del teatro greco cfr. ora Albini 1991.

 Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica Alessandria Alessandria nuova città: le istituzioni e il patrocinio dei Tolomei Alessandro, poiché aveva deciso di fondare in Egitto una grande città, alle persone lasciate là con questo compito diede ordine di costruirla fra il lago e il mare. Dopo aver delimitato il luogo e diviso accuratamente in quartieri lo spazio della città, egli la chiamò Alessandria dal proprio nome [...]. Alessandro ordinò anche di edificare un palazzo reale, meraviglioso per la grandezza e l’imponenza della realizzazione. Non solo Alessandro, ma anche pressoché tutti i re dell’Egitto dopo di lui fino ai nostri tempi accrebbero la reggia con sontuose costruzioni. Insomma, la città conobbe in seguito un tale progresso, che molti la considerano come la prima città del mondo: e infatti essa prevale di molto sulle altre per bellezza e per grandezza e per abbondanza delle entrate e di quant’altro serve al vivere bene. Anche il numero dei suoi abitanti supera quello delle altre città (Diodoro Siculo, XVII 52 I–6).

Nuova fondazione (nel 332 a.C.) alle soglie di quell’età ellenistica segnata dalla vasta diffusione della cultura greca conseguente alle conquiste di Alessandro, la città sul delta del Nilo sembra subito improntata a un futuro di fortuna e splendore. Del resto i Tolomei non furono grandi urbanizzatori: la loro attenzione si concentrò in modo particolare sulla capitale, Alessandria.1 Dopo la morte di Alessandro nel 323 a.C. fu Tolomeo figlio di Lago,2 uno dei generali che aveva partecipato alle imprese del condottiero, a diventare satrapo dell’Egitto e a farne il proprio regno, assumendo il titolo di re probabilmente intorno al 305 con il nome di Tolomeo I Soter. Egli morí intorno al 283: come altri generali, fu anche storico delle guerre di Alessandro e scrisse un’opera che fu utilizzata da Arriano.3 Sotto il suo regno e quello del suo successore, Tolomeo II Filadelfo (associato al trono nel 285, re dal 283 al 246), prese impulso e fiorí in Alessandria uno dei piú importanti, attivi e vivaci centri culturali non solo dell’età ellenistica ma in generale del mondo antico. Lo sviluppo in vari settori

 1 Sull’Alessandria tolemaica disponiamo di una cospicua monografia, alla quale è d’obbligo rinviare in apertura per una sintesi assai piu ampia di questa: Fraser 1972; inoltre per le trattazione di molti temi, qui soltanto accennati per evitare inutili ripetizioni, si vedano i capitoli in Cambiano/Canfora/Lanza 1992. 2 Per questo la dinastia dei Tolomei è detta anche dei Lagidi. 3 Cfr. FGrHist, 138; Musti 1989, 687 sg. e 639. https://doi.org/10.1515/9783110772371-014

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica della cultura fu caratterizzato segnatamente dal fenomeno del patrocinio esercitato sulla cultura dal potere politico, fenomeno in sé non peculiare né del luogo né del periodo ma che nell’Alessandria tolemaica raggiunse caratteristiche particolarmente profonde e intense. Gran parte della produzione intellettuale di Alessandria deriva direttamente dal patrocinio dei Tolomei [...]. Il patrocinio letterario è un aspetto molto antico della cultura greca: cominciò con i tiranni del periodo arcaico e continuò a esistere dovunque tirannide e monarchia perdurarono nell’epoca romana [...]. Dunque il patrocinio dei re ellenistici non era nulla di nuovo: a quanto pare l’istituzione era un accompagnamento costante dello splendore regale. La differenza, almeno in Egitto, sta nell’estensione e nell’orientamento del suo sviluppo. Nell’epoca arcaica e classica, come piú tardi nella Roma Imperiale, il patrocinio era ricompensato con la poesia, e l’obbligo a quanto pare arrivava fino a questo punto. Nel mondo ellenistico questo aspetto individuale rimase fino a un certo punto valido, ma la determinazione del potere di sostenere la cultura per il proprio interesse era di maggiore significato. A tale scopo i diversi re, avendo a loro disposizione capitali maggiori rispetto ai governanti dei tempi passati, istituirono fondazioni permanenti per la cultura. Questo sviluppo, che implica una nuova concezione sia del mondo fisico che delle creazioni del passato come doverosi oggetti di studio, è di per sé una cosa di considerevole interesse.

Questo è l’efficace inizio della presentazione, da parte di uno dei maggiori studiosi dell’Alessandria ellenistica,4 dell’aspetto — come dicevamo — forse piú eclatante dell’Egitto tolemaico dal punto di vista della storia socio-culturale: uno sviluppo e una fìoritura di tale portata da avere pochi confronti, uniti a un legame particolarmente profondo e stretto con il patrocinio del potere centrale incarnato dalla dinastia dei Tolomei. Si può dunque capire che la crisi politica seguita alla morte di Tolomeo VI Filometore nel 145 a.C., che vide l’uccisione di Tolomeo VII e l’ascesa al trono da parte di Tolomeo VIII Evergete II con le conseguenti violenze e vendette, sia cosí commentata da Pfeiffer: I migliori discepoli di Aristarco5 e molti altri filologi della generazione piú giovane fuggirono in vari luoghi che non erano sotto il dominio egiziano, Rodi, Pergamo, Atene. Da questa secessio doctorum si ebbe la prima crisi nella storia della filologia. Che la carica di capo della biblioteca sia toccata in sorte a un oscuro ufficiale militare, chiamato Cida, ἐϰ τῶν λογχοφόρων,6 è un fatto che parla da sé.

 4 Fraser 1972, I 305. 5 Aristarco si rifugiò a Cipro, dove probabilmente morí non molto tempo dopo. 6 Cioè: appartenente a un corpo militare di lancieri.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

Turbamenti politico-istituzionali e culturali si connettono immediatamente; e gli stessi motivi che avevano causato crisi e momentaneo declino portarono a una diffusione che fu in sostanza una rinascita: Da Alessandria i γραμματιϰοί, esiliati e senza un centesimo, diffusero la loro τέχνη per le isole e le città e stimolarono la vita spirituale — fra loro i due eminenti discepoli di Aristarco: Apollodoro da Alessandria probabilmente si rifugiò a Pergamo e verisimilmente ritornò più tardi nella sua città natale Atene, Dionisio Trace si trasferì nell’isola di Rodi, ed entrambi continuarono vigorosamente il lavoro filologico, e anche (aggiungiamo) il loro insegnamento ricco di discepoli.7

La più vistosa e rinomata realizzazione del patrocinio dei Tolomei sulla cultura, e anche quella più ricca di durature influenze, fu la fondazione, all’interno dell’immensa reggia, del complesso formato dalla Biblioteca e dal Museo, luogo di raccolta e di conservazione per gli strumenti e gli uomini della cultura, che là vivevano coltivando i loro studi nella sicurezza del supporto dello stato. L’avvio dell’impresa dovette essere di grande efficacia, se già sotto il regno di Tolomeo II Filadelfo la Biblioteca raccoglieva quasi 500.000 volumi.8 Queste pubbliche istituzioni culturali segnarono della loro presenza l’Alessandria tolemaica e giocarono un ruolo fondamentale nella storia della trasmissione dei testi antichi, anche per quanto riguarda alcuni degli aspetti che essi assunsero.9 In tale contesto si affermò una nuova figura: quella del bibliotecario, letterato e filologo, talvolta anche grande scienziato (come Eratostene) o grande poeta (come Apollonio Rodio), personaggio importante della corte e precettore della casa reale, maestro influente e uomo di spicco della produzione di cultura e non solo della sua conservazione. La prima e più grande stagione della filologia alessandrina, per quell’irripetibile secolo e mezzo che va da Zenodoto ad Aristarco, reca l’impronta di queste figure.10 È difficile negare l’influsso operante del modello peripatetico sull’avvio e lo sviluppo in Alessandria di questa forma complessiva di organizzazione della cultura, con il tratto caratterizzante della raccolta e studio dei testi e dei materiali e il conseguente atteggiamento di riflessione sul passato. Non si può di 7 Cfr. Pfeiffer 1973, 330 sgg. e 385 sgg.: le due citazioni si trovano rispettivamente a pp. 331 sg. e a pp. 386 sg. Per la filologia alessandrina cfr. in questo vol. il saggio di Montanari 1993 (= questo volume, cap. 1). 8 Cfr. Canfora 1993. 9 Caso tipico è il numerus versuum dei poemi omerici, fissatosi pressappoco con la seconda metà del II sec. a.C., oppure il delinearsi di alcune “scelte” canoniche di opere, come quelle teatrali: anche per questo cfr. Montanari 1993. 10 Per questi aspetti cfr. ancora Montanari 1993.

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica menticare che Tolomeo di Lago senz’altro conosceva benissimo il rapporto di Alessandro con Aristotele, e questo spiega la sua assiduità nel tentativo di chiamare ad Alessandria un grande peripatetico. Non riuscì il tentativo fatto con Teofrasto, il primo scolarca dopo Aristotele,11 ma il successore di Teofrasto, Stratone di Lampsaco, rimase qualche tempo ad Alessandria e fu precettore di Tolomeo II Filadelfo.12 Infine, anche se manca una testimonianza esplicita per un suo intervento nell’attività di organizzazione culturale, un ruolo lo ebbe senz’altro Demetrio Falereo, l’allievo di Teofrasto imposto da Cassandra nel 317 al governo di Atene: il suo regime fu abbattuto da Demetrio Poliorcete nel 307, Demetrio Falereo fu costretto a fuggire da Atene e qualche anno dopo fu accolto con onore e ospitato in Alessandria, dove visse sino alla morte. È un ripetersi di relazioni, una linea di comunicazione troppo evidente di Alessandria con Atene, e in particolare con la scuola di Aristotele, per poter essere svalutata.13 Il ridimensionamento tentato da Pfeiffer dell’influenza peripatetica sul nascere di Alessandria come centro culturale ha il suo punto debole proprio nella valutazione degli stadi iniziali del fenomeno. In realtà, Pfeiffer non negò del tutto l’influsso peripatetico sulla storia della filologia alessandrina (cosa che sarebbe stata impossibile, oltre che infelicemente semplifìcatoria), ma sopravvalutò la linea della poesia rispetto a quella piú generalmente storico-culturale e filosofica e della organizzazione del sapere (e anche del legame con le istituzioni): I seguaci di Filita non erano peripatetici [...]. La linea Filita–Zenodoto–Callimaco, di cui abbiamo messo in rilievo il carattere non-aristotelico, s’incontrò in Alessandria con la genuina linea peripatetica proveniente da Atene; ma questo è il secondo stadio dell’intero processo, non il suo inizio. Gli alunni di Aristotele potevano portare un inestimabile aiuto ai ποιηταὶ ϰαὶ ϰριτιϰοί che già esistevano in Alessandria: infatti, trasferirono raccolte di materiale erudito dalla loro sede ateniese, provocarono nuove ricerche antiquarie, stimolarono la nuova critica letteraria, spesso in contrasto con le opinioni del loro maestro, e insegnarono ad organizzare le istituzioni per il progresso della filologia. Troveremo non pochi elementi aristotelici anche negli scritti in prosa di Callimaco [...].14

Ma in verità quanto abbiamo ricordato sopra ci parla di un legame col Peripato proprio nelle fasi iniziali dello sviluppo culturale alessandrino. Del resto, quei legami personali fra ambiente peripatetico e ambiente alessandrino non sono che l’esplicitazione concreta di un fatto storico-culturale, sul quale ci siamo già soffermati: il pensiero e la visione di Aristotele e della sua scuola contribuirono  11 Cfr. Diogene Laerzio, V 37; Fortenbaugh et al. 1992, I 20–21. 12 Cfr. Diogene Laerzio, V 58. 13 Cfr. Canfora 1993. 14 Pfeiffer 1973, 168 sg.; cfr. anche pp. 164 e 249.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

in modo decisivo a porre la considerazione della letteratura sotto una luce diversa, a farla diventare oggetto di interpretazione e di studio sulle personalità e sulle opere, di riflessione sui modelli culturali in cui era concentrato il patrimonio della paideia greca e la base della propria identità antropologico-culturale. In questo senso, Aristotele e il Peripato sono davvero il presupposto, il fermento decisivo che muove gli orientamenti e le idee di Alessandria.15

Produzione e circolazione di cultura La straordinaria fioritura culturale dell’Alessandria tolemaica comprende una quantità di aspetti e un cospicuo numero di personalità eminenti, di cui si è già abbondantemente parlato. Non c’è ragione dunque di indulgere a inutili ripetizioni, non soltanto riguardo alla Biblioteca e al Museo, ma anche sulla problematica dei generi letterari, su Callimaco, Apollonio Rodio, Teocrito e l’epigramma alessandrino, sulla storia della filologia e della grammatica da Filita a Didimo, sulle singole discipline scientifiche e altro ancora che nel crogiuolo di Alessandria trovò spazio e nutrimento.16 Alessandria è un centro nel quale la cultura circola con grande intensità, ma entro un pubblico ristretto e ben diverso da quello della polis dei secoli passati. La letteratura è prodotta per una élite, la poesia si rivolge a un pubblico di fruitori dotti, agli eruditi addetti ai lavori, agli uomini del potere, a chi è educato alla cultura, consumatore di letteratura, lettore colto provvisto di un bagaglio sempre piú cospicuo: all’interno della città, la cerchia limitata che ruota intorno alla reggia rappresenta la maggior parte di coloro che partecipano realmente del fenomeno (come accade piú o meno nelle altre città ellenistiche). In questo quadro rientra un nuovo tipo di fruitore dei testi letterari, vale a dire il filologo indagatore, studioso e antiquario, che riflette su di essi, li cura, li preserva e interpreta come monumenti di una cultura passata con la quale intrattiene un rapporto da storico, caratterizzato da una razionalità esegetica armata di tecniche complesse e raffinate. Tra chi produce e chi interpreta la poesia c’è spesso identificazione totale, quasi sempre comunque la contiguità dell’ambiente definito, del gruppo limitato. Il poeta-filologo, la poesia-dotta nel suo senso piú pieno sono davvero emblematici del periodo e del contesto. Intorno al perno della reggia dei Tolomei, la città sul delta del Nilo è centro di richiamo e di scambio. Alla sua brillante vita intellettuale contribuirono molte  15 Cfr. Montanari 1973. 16 Cfr. in Cambiano/Canfora/Lanza 1993 l’insieme della parte III. La Biblioteca e il Museo.

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica aree del mondo greco, come risulta facilmente percorrendo col pensiero i luoghi di nascita dei maggiori intellettuali che vi operarono: Zenodoto di Efeso, Callimaco di Cirene, Aristofane di Bisanzio, Aristarco di Samotracia; e diverse isole dell’Egeo e Atene e Siracusa e altri luoghi ancora. Si capisce anche il fenomeno della traduzione in greco di testi da lingue diverse, fenomeno culturale e politico legato in generale ai rapporti fra l’elemento greco e quello delle popolazioni non greche negli stati ellenistici.17 Come aveva fatto ogni tentativo per far venire ad Alessandria filosofi peripatetici, Tolomeo I Soter cominciò anche a chiamare alla sua corte i poeti: Filita di Cos, squisito poeta e insieme antesignano dei filologi alessandrini, fu scelto da Tolomeo I come primo precettore per il figlio Tolomeo II, che ebbe poi come secondo maestro Zenodoto, allievo di Filita. I tre maggiori poeti dell’età ellenistica furono legati ad Alessandria. Callimaco in giovane età lasciò Cirene per Alessandria, fu introdotto a corte sotto il regno di Tolomeo II Filadelfo e vi rimase con grande prestigio sotto Tolomeo III Evergete (salito al trono nel 246 a.C.), per la cui sposa Berenice compose la famosa Chioma di Berenice. Non ricoprí la carica di bibliotecario, ma ebbe l’incarico di fare un catalogo della letteratura conservata nella biblioteca e scrisse i ponderosi Pinakes in 120 libri. Apollonio Rodio era proprio di Alessandria e assunse la carica di bibliotecario dopo Zenodoto: sotto il regno di Tolomeo II Filadelfo fu precettore dell’erede al trono, ma dopo l’ascesa al regno di Tolomeo III Evergete (che avvenne nel 246) lasciò per qualche ragione la capitale e si ritirò a Rodi (da cui l’appellativo). Teocrito invece era originario di tutt’altra area geografica, quasi certamente di Siracusa, ma si trasferì ad Alessandria durante il regno di Tolomeo II Filadelfo e vi soggiornò per un imprecisato numero di anni, durante i quali frequentò certamente l’ambiente della corte. Quasi certa è la datazione fra il 274 e il 270 a.C. dell’Idillio XVII, che contiene un encomio del Filadelfo: il migliore degli uomini, sovrano pieno di meriti e di fortune di un paese felice nella pace e nella prosperità, sposo di una regina in tutto degna del suo valore; al suo elogio si accomuna il padre Tolomeo di Lago, uomo capace di compiere grandi imprese e che ora siede tra gli dei, accanto ad Alessandro e di fronte al trono del loro capostipite Eracle. Il quadro che risulta dall’esame delle relazioni fra i letterati e la corte appare dunque quello di un organico consenso e di una consonanza generalizzata in un’atmosfera di reciproco sostegno. Questo è in effetti il “colore” dominante che si impone nel rapporto fra intellettuali e potere, considerando anche altre sfere dell’attività culturale nel suo complesso, segnatamente quelle delle varie  17 Cfr. Canfora 1993.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

discipline scientifiche. In effetti, i progressi nell’organizzazione razionalizzata dello stato, sia nel senso della gestione del potere che nello sfruttamento e incremento delle risorse economiche, conferirono per lo piú all’Alessandria tolemaica quel carattere di generalizzata fioritura che accomuna alto tenore economico e sviluppo scientifico-culturale legato alla struttura politica.18 Ma anche nella più felice Alessandria, naturalmente, non poteva mancare qualche voce discordante, qualche isolato dissenso. È il caso del poeta Sotade di Maronea, che reagì in toni fortemente negativi al matrimonio incestuoso fra Tolomeo II e la sorella Arsinoe, la coppia celebrata da Teocrito. Il poeta pagò caro il suo giudizio critico: fu chiuso in una cassa di piombo e gettato in mare, a suggellare l’eliminazione delle note stonate da una sinfonia luccicante.19 Il Museo non fu solo poesia e belle lettere, filologia ed erudizione: Alessandria fu anche grande centro di studio e insegnamento della scienza e di produzione di letteratura scientifica. Matematica, astronomia e medicina contarono su figure e relazioni di primo piano. Non si sa dove sia nato Euclide, ma certo è che egli operò e insegnò ad Alessandria sotto Tolomeo I, in stretti rapporti col Museo e col sovrano. Archimede di Siracusa fece i suoi studi in Alessandria, dove fu in relazione con il celebre matematico Conone e con il suo allievo Dositeo e dove intrattenne rapporti anche con Eratostene.20 Impulso notevole fu dato, come si può ben capire, a discipline tecnico-scientifiche come la tecnica militare, la geografia e l’etnografia, l’agricoltura: tutte di grande importanza per le fortune dello stato e per un governo prospero ed efficace.21 Nel vivace quadro caratterizzato dalle tendenze dogmatiche e da quelle empiriche della medicina, l’ambiente di Alessandria giocò sicuramente un ruolo di primo piano, se non altro grazie alla figura di Erofilo con la sua scuola (che continuò fino al I sec. d.C.): egli proveniva da Calcedone, cittadina sul Bosforo di fronte a Bisanzio, e non sappiamo quando spostò la sua attività e il suo insegnamento nella capitale dei Tolomei. Se non si può citare una testimonianza esplicita sulla presenza in Alessandria dell’altro grande medico dell’epoca, Erasistrato (nativo dell’isola di Ceo), è certo che medici della sua scuola vi furono attivi, e sarà quindi difficile che il maestro non vi abbia mai messo piede.22

 18 Cfr. Gabba 1984. 19 Cfr. Pretagostini 1988, 318 sg. 20 Cfr. Canfora 1993 e Repellini 1993. 21 Cfr. Gabba 1984. 22 Cfr. in questo vol. [= Lo spazio letterario della Grecia antica, vol. I, Tomo II, Roma, Salerno Ed., 1993] il saggio di I. Garofalo, Figure della medicina ellenistica.

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica Questi medici alessandrini furono famosi per la pratica della dissezione e il connesso impulso dato allo studio dell’anatomia. Un caso come quello del grammatico Tolomeo Epitete (contemporaneo piú giovane di Aristarco e suo avversario in quanto erede di una linea zenodotea) è veramente notevole, e si direbbe emblematico. Tra i titoli delle sue opere si conserva quello di una monografìa sulle ferite in Omero, nel quale evidentemente venivano in qualche modo studiate le rappresentazioni e descrizioni omeriche di ferite in battaglia. Il tema rientra perfettamente nell’ambito di osservazione dei Realien omerici, sulla vita materiale degli eroi, presente e ben testimoniato nel lavoro dei filologi;23 d’altra parte, dal punto di vista della storia della medicina, non sono mancate le considerazioni sulla medicina omerica come trattamento delle ferite e quelle sull’attività dei medici epici Podalirio e Macaone (visti come iniziatori mitici della chirurgia e della farmacologia, anche nell’esegesi omerica: si veda lo scolio omerico a Λ 513 c Erbse; cfr. Celso, De medicina, Proemio, I sgg.). Il tema dei vulnera poteva dunque valere sia per la stilistica e l’antiquaria omerica, sia per un aspetto di storia della scienza medica, che in Alessandria non era certo estraneo: la pratica della dissezione, legata agli interessi anatomici e chirurgici, di cui si è detto, poteva ben contemplare un aspetto di storia delle conoscenze nel campo e un gusto per l’analisi di descrizioni di ferite, non di rado tutt’altro che povere di dettagli anatomici. Se lo abbiamo analizzato correttamente, il titolo dell’opera di Tolomeo Epitete ci parla di un tema che, nell’attività di un filologo, riunisce un interesse stilisticoletterario e antiquario volto all’interpretazione del testo omerico e un interesse di storia della medicina: una unione di filologia e scienza che ricorda il modello di Eratostene.24 Si colloca bene in questo quadro il sorgere di una ermeneutica e una filologia dei testi classici della medicina greca.25 Non ci sono testimonianze sicure che Erofilo abbia davvero scritto qualcosa di esegesi ippocratica; ma la figura di Bacchio di Tanagra (275 ca.–200 ca. a.C.) è carica di tutto il suo significato. Medico ed erudito, egli compilò una raccolta di Lexeis ippocratiche divisa in tre sezioni (per cui secondo Galeno utilizzò anche materiali di Aristofane di Bisanzio), fece l’edizione di almeno un trattato ippocratico, scrisse diversi commentari e un’importante opera dossografica: una vera figura di filologo alessandrino

 23 Con particolare abbondanza per Aristarco; questi aspetti si svilupperanno nella grande antiquaria posteriore, come quella di Didimo in età augustea. 24 Cfr. Montanari 1988, 81 sgg. e 95. 25 Cfr. Deichgräber 1930, 317 sgg.; von Staden 1989, 427 sgg.; Manetti/Roselli 1994, con bibliografia.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

nel campo della medicina.26 Prima di Bacchio un glossario ippocratico sembra abbia scritto Senocrito di Cos (contemporaneo di Filita);27 dopo Bacchio, e, a quanto pare, in polemica con lui, un lessico ippocratico è testimoniato per l’iniziatore della medicina empirica Filino di Cos.28 Qualche parola infine va spesa per l’empirico Zeuxis (vissuto nella prima metà del II sec. a.C.), autore di commenti a trattati ippocratici, citato da Galeno tra l’altro per notizie relative a copie di trattati ippocratici acquistati e presenti nella biblioteca di Alessandria e utilizzati in vario modo da studiosi della medicina.29 A quanto sappiamo, non fu mai ad Alessandria Timone di Fliunte, che però doveva essere ben al corrente di quello che avveniva nella città egiziana, se poteva dirigere i suoi sarcasmi contro i filologi rinchiusi nella gabbia delle Muse, dunque contro il modo di vivere e operare nelle rinomate istituzioni culturali del luogo, e poteva dare al poeta Arato il consiglio di utilizzare le vecchie copie di Omero e non quelle recentemente corrette.30 Ciò che accadeva in Alessandria era evidentemente ben noto nel mondo grecizzato e oggetto di valutazione e di commento. Centro di produzione culturale e anche di grande richiamo, luogo di circolazione del sapere e fucina di progresso, l’Alessandria tolemaica ebbe una vera importante rivale in Pergamo, il regno degli Attalidi in Asia Minore, dove intorno a una biblioteca di grande importanza un’altra dinastia reale cercava di attrarre forze intellettuali.

Cirene I rapporti di Cirene con il resto del regno d’Egitto, e dunque con Alessandria, si assestarono soltanto dopo la metà del III sec. a.C., grazie a sviluppi dinasticomatrimoniali. Per vero dire, le città della Cirenaica (la regione di Cirene delimitata a ovest dalla Grande Sirte) si sottomisero a Alessandro Magno e di conseguenza entrarono a far parte della satrapia d’Egitto di Tolomeo Lago. Tuttavia l’inquietudine della regione e le rivolte spinsero Tolomeo a inviare il figliastro Magas come governatore, ma questi rivendicò (probabilmente intorno al 274) una sorta di indipendenza da Alessandria: quando la figlia di Magas, Berenice,

 26 Cfr. von Staden 1989, 484 sgg. 27 Cfr. Deichgräber 1930, 221 fr. 311; Pfeiffer 1973, 164 n. 26. 28 Cfr. Deichgräber 1930, 221 fr. 311; Pfeiffer 1973, 164 n. 26. 29 Cfr. Deichgräber 1930, 221 fr. 343; von Staden 1989, 481 e n. 4; Manetti/Roselli 1994. 30 Cfr. Pfeiffer 1973, 171 sg., 203, 278; Di Marco 1989, T I 113 (= Diogene Laerzio, IX 113) e F 12, con comm.; Canfora 1993.

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica sposò Tolomeo III Evergete I e la coppia salí al trono nel 246, la Cirenaica fu definitivamente compresa nel regno d’Egitto. A differenza di Alessandria, Cirene non solo non era una nuova fondazione, ma anzi aveva origini assai antiche e anche una lunga storia di presenza nella letteratura.31 La città, posta sulla parte più a nord del promontorio a est della Grande Sirte, era costruita a circa otto chilometri dal mare e collegata con una strada al porto di Apollonia, distante circa 19 chilometri. Il primo diffuso racconto delle vicende connesse alla fondazione di Cirene e alla sua storia arcaica lo troviamo in Erodoto, IV 154 sgg. La fondazione risaliva al VII sec. a.C. (la data tradizionale è il 631), ad opera di coloni Dori di Tera al seguito del re Batto, che diede inizio alla dinastia dei Battiadi: l’ultimo re Battiade, Arcesilao IV, vincitore nei giochi Pitici del 462, fu celebrato da Pindaro nella Pitica IV e V. Anche la Pitica IX è dedicata a un vincitore di Cirene, Telesicrate, e racconta l’antica leggenda dell’amore di Apollo per la ninfa eponima della città: leggenda che si trova accennata nelle Argonautiche di Apollonio Rodio, II 500–10. Callimaco, nell’Inno ad Apollo, collega in altro modo Apollo alle origini della sua patria: «La mia città fertile terra a Batto indicò Febo e in sembianze di corvo guidò la gente che entrava nella Libia, alla destra del fondatore, e giurò di donare le mura ai nostri re: Apollo mantiene sempre il giuramento» (VV. 65–68). Da Cirene (nato probabilmente intorno al 430 a.C.) veniva il filosofo Aristippo, che andò ad Atene per ascoltare Socrate (del quale fu discepolo piú giovane di Platone e contemporaneo di Antistene) e che fu all’origine della scuola socratica dei Cirenaici;32 Diogene Laerzio (II 83) ci dice che scrisse anche un’opera storica sulla Libia.33 Nello stesso passo, Diogene Laerzio ci informa inoltre dell’esistenza di due omonimi cirenaici: un Aristippo di Cirene nipote del precedente e chiamato Metrodidatto perché istruito dalla madre alla filosofia dello zio; e un Aristippo di Cirene filosofo della Nuova Accademia.34 Alla scuola cirenaica appartengono anche Egesia e Teodoro detto “l’ateo” (nato forse intorno al 340 a.C.), che per anni visse ad Atene e ad Alessandria, ma tornò in vecchiaia nella città natale.35 Lacide di Cirene fu il successore di Arcesilao a capo dell’Accademia e (forse intorno al 241 a.C.) rifiutò l’invito di Attalo I a trasferirsi a Pergamo.

 31 Gentili 1990, con bibliografia; cfr. anche Nicolai 1992a. 32 Per tutti i problemi connessi alla definizione e localizzazione della scuola, dei discepoli e del ruolo di Aristippo, cfr. Giannantoni 1990, II 1 sgg., e IV 135 sgg. 33 Cfr. FGrHist, 759. 34 Cfr. Dörrie 1979. 35 Cfr. Giannantoni 1990, II 113 sgg. e 119 sgg., IV 189 sgg., con bibliografia.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

Almeno due grandi personaggi della cultura alessandrina sono figli di Cirene, Callimaco ed Eratostene. Abbiamo già ricordato come Callimaco abbia lasciato Cirene per Alessandria in giovane età. Per questo è molto incerta, e probabilmente insostenibile, la notizia secondo cui egli sarebbe stato maestro di Eratostene nella stessa Cirene:36 se è un autoschediasma biografistico, vi si può ben vedere la volontà di legare con un rapporto maestro-allievo i due grandi conterranei. Sappiamo invece che Eratostene, nella sua città natale, aveva avuto come maestro Lisania, un grammatico di cui sono rimasti pochi frammenti (sappiamo che scrisse sui giambografi, è citato per interpretazioni omeriche e per interessi glossografici), ma che dovette godere di un prestigio non piccolo.37 Si conoscono diversi altri uomini di cultura di origine cirenaica, come il grammatico Apollodoro,38 gli storici Filostefano (III sec. a.C.) e Giasone (II sec. a.C., FGrHist, 182), un Tolomeo medico citato da Galeno (II–I sec. a.C.), un Teeteto poeta epigrammatico (III sec. a.C.). Incerta è invece la notizia che fosse di Cirene il grammatico, biografo e storico Istro (FGrHist, 334), detto “Callimacheo”, dunque probabilmente in qualche modo scolaro di Callimaco, per cui nacque forse l’idea che fosse anche suo conterraneo.

Pergamo La città, gli Attalidi e la biblioteca Il visitatore dei Musei di Berlino resterà affascinato da molte meraviglie, una delle quali sarà senza dubbio la grandiosa ricostruzione dell’altare di Zeus di Pergamo, con l’originale dello splendido fregio raffigurante la lotta degli Dei Olimpi contro i Giganti. Questo capolavoro dell’arte ellenistica fu voluto dal re Eumene II, che regnò fra il 197 e il 158 a.C., e si trovava sulla rocca della città, accanto alla Stoa, al santuario di Atena, al teatro e, poco più a nord, alla famosa biblioteca: uno spettacolare insieme cui si aggiungeva il ginnasio, situato più in basso verso sud, che aveva fama di essere il più grande del mondo greco. La città (che oggi si chiama Bergama) era costruita su una scoscesa collina rocciosa alta poco più di 300 metri, nella piana attraversata dal fiume Kaikos (parte nord-ovest dell’Asia Minore), a circa trenta chilometri dalla costa anatolica che sta di fronte all’isola di Lesbo. La prima menzione di Pergamo si trova in

 36 Per incongruenze cronologiche, su cui cfr. Pfeiffer 1973, 251. 37 Gudeman 1927; Pfeiffer 1973, 238 n. 146 e 251. 38 Dyck 1981.

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica Senofonte, Elleniche, III 1 6, e riguarda un episodio del 399 a.C.: successivamente al disastro della spedizione di Ciro il Giovane (morto nella battaglia di Cunassa del 401), le città greche della Ionia chiedono l’aiuto di Sparta (fresca vincitrice della guerra del Peloponneso) contro il satrapo Tissaferne; l’intervento spartano in Asia Minore si concreta con lo sbarco nel 400 di un contingente comandato da Tibrone, cui si uniscono i resti delle truppe di Ciro; Tibrone, intenzionato ad affrontare Tissaferne, conduce la sua campagna nelle zone del nord-ovest dell’Asia Minore e si impadronisce di alcune città, fra cui Pergamo, che gli si sottomette volontariamente. Cosí racconta Senofonte: in quel momento, Pergamo è dunque una città di importanza già rilevante, degna di essere menzionata nel contesto di una campagna militare. Nel periodo in cui esercitava maggior potere su una consistente parte dell’Asia Minore, il re della Tracia Lisimaco — uno degli ufficiali di Alessandro Magno, che aveva partecipato alla spartizione dell’impero — insediò a Pergamo come proprio luogotenente il tesoriere Filetero. Approfittando di un’occasione favorevole, questi si ribellò nel 282 e, dopo la sconfitta e la morte di Lisimaco nella battaglia di Curupedio (281 a.C.), riuscì a mantenersi indipendente anche dal vincitore Seleuco, al quale si era opportunisticamente avvicinato. Di conseguenza gli fu possibile conservare in seguito a se stesso e a Pergamo una posizione autonoma rispetto al potente regno seleucidico di Siria, posizione che progressivamente si consolidò sotto il suo successore Eumene I (263–241 a.C.) fino alla totale autonomia del territorio pergameno. Il primo signore di Pergamo ad assumere formalmente il titolo di re fu Attalo I, che regnò dal 241 al 197 a.C.: la dinastia degli Attalidi aveva preso definitivamente il suo volto.39 Le origini di Filetero non erano esattamente quelle che si addicevano a un re e le voci avverse non mancarono di sfruttare questo motivo. Non era certo nobile la sua nascita: se il padre era un militare macedone di nome Attalo (da cui la dinastia fu detta degli Attalidi), la madre era una cortigiana paflagone di nome Boa. Oltre a questo, l’inizio delle sue fortune restava legato all’opportunistico atteggiamento nei confronti di Lisimaco, con la conseguente accusa di essersi in quell’occasione impadronito di tesori non suoi. Ma Filetero stesso e soprattutto il suo successore Eumene I riuscirono egregiamente a dare alla figura del tesoriere di Lisimaco il profilo di iniziatore e capostipite di una dinastia regale, che vantava origini risalenti addirittura a Eracle, dunque divine: un tema che sarà costante nella propaganda celebrativa, accanto a quello della gloria militare (soprattutto per le lotte contro i Galati).  39 Su Pergamo e gli Attalidi cfr. Virgilio 1993: ho letto il lavoro in dattiloscritto e utilizzato i risultati in queste pagine.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

Dopo il periodo di maggiore fioritura coinciso con i regni di Attalo I, Eumene II e Attalo II, sul quale torneremo fra poco, nel 133 a.C. Attalo III Filometore (dipinto da alcune fonti, che probabilmente recepiscono opinioni popolari scontente e ostili, come un re crudele e poco sollecito nei doveri di governo) decise di lasciare con il suo testamento il regno in eredità ai Romani: in pratica scelse di essere l’ultimo re di Pergamo e pose fine con questo alla dinastia degli Attalidi come casa regnante (dinastia di non lunga durata, ove si pensi che dopo il fondatore ci furono soltanto cinque re e che erano trascorsi non piú di 150 anni dalla ribellione di Filetero nel 282). Una volta scomparso Attalo III, alla donazione in favore di Roma cercò di opporsi suo fratello Aristonico, tentando di farsi riconoscere come erede al trono (con il nome di Eumene III): ma non trovò grandi sostegni proprio nella città, dove le classi alte si sentivano meglio garantite dai Romani. Fu questa per Roma la prima esperienza di una provincia al di là dell’Egeo e il Senato reagì con interventi legislativi rapidi ed efficienti, con l’invio di ambasciatori e con campagne militari (nel 130 e nel 129) che stabilizzarono la situazione e posero i Romani come i veri e legittimi continuatori della dinastia attalide (eredi dell’evergetismo, dei culti e degli onori prima tributati ai dinasti e ai loro familiari): Pergamo fu riconosciuta città libera e poi città federata. Una volta divenuta possesso di Roma, dunque, Pergamo non perse affatto la sua fama e la sua importanza e restò (ancora in eta imperiale) una delle principali città della provincia d’Asia. L’ambizione di porsi in una linea di continuità pergamena compare nell’operato di Mitridate VI Eupatore, re del Ponto dal 121 al 63 a.C., il rappresentante forse piú significativo di quelle forme di opposizione a Roma che emersero progressivamente con il venir meno dei regni ellenistici e l’affermarsi della potenza romana. Mitridate si scontrò con eserciti romani a più riprese dall’anno 88 all’anno 64, presentandosi come il campione della grecità e dei regni ellenistici nella via del riscatto dall’oppressione e dal dominio di Roma. In questa vicenda egli volle stabilire la sua capitale proprio a Pergamo, facendone il centro del proprio potere e il simbolo della propria figura storica, con l’idea di sostituirsi ai Romani come erede degli Attalidi. La sconfitta di Mitridate segnò un grave momento di crisi, che fu superato riportando la città in buoni rapporti con Roma all’interno della provincia d’Asia. Nelle guerre civili Pergamo aveva parteggiato decisamente per Ottaviano e a partire da Augusto la politica degli imperatori fu quella, ancora una volta, di avallare una ideale continuità con gli Attalidi, che restavano sempre sullo sfondo come l’ideale riferimento storico all’epoca della grandezza e delle fortune. L’antico prestigio non venne meno e valse spesso alla città, da parte di Roma, particolare attenzione e privilegi, culminati con gli interventi urbanistici del

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica filelleno Adriano. Durante il periodo mitridatico e nei primi secoli dell’età imperiale contribuirono al bene della città soprattutto figure di aristocratici che si dedicavano alla vita pubblica e ricoprivano cariche nell’amministrazione romana: i riconoscimenti che venivano loro attribuiti si riassumono nella qualifica di evergete, al quale non di rado si accompagnavano onori regali e un vero e proprio culto. Furono il culto e le onorificenze per gli imperatori ad ereditare a poco a poco tutto quanto era stato tributato prima ai dinasti e poi agli evergeti pergameni (anche se il culto degli Attalidi proseguí a lungo dopo la fine del loro regno). Come abbiamo già detto, il periodo di maggior auge del regno di Pergamo è racchiuso nell’arco di tempo che comprende i regni di Attalo I (241–197), Eumene II (197–158) e Attalo II (158–138). Per circa un secolo e mezzo (considerando anche parte del regno di Eumene I e i pochi anni di Attalo III) il regno prosperò dal punto di vista politico, economico e culturale, con grande vantaggio per le scienze, le arti e le lettere. «Anche se prendiamo in considerazione tutta l’energia, l’ambizione e l’abilità della famiglia degli Attalidi, sembra ancora un evento miracoloso che Pergamo fu portata da loro a tale importanza per un secolo e mezzo».40 Oltre ad essere stato il primo dinasta degli Attalidi ad assumere formalmente il titolo di basileus di Pergamo, Attalo I recò alla sua città anche la gloria militare con le prime vittorie contri i Galati, ottenute fra il 241 e il 238, e contro il siriaco Antioco Ierace (nel 238). Queste fortunate campagne militari gli valsero l’epiteto di Soter e offrirono un buon fondamento per proseguire con efficacia la politica (iniziata, come dicevamo, da Filetero e da Eumene I) di costruzione dell’immagine della propria sovranità e del prestigio della dinastia, soprattutto agli occhi del mondo greco. La celebrazione di tale gloria culminò con l’erezione di un grande monumento sull’acropoli pergamena: il donario di Attalo I subí spoliazioni di statue per volontà di Nerone nel 64 d.C., danni cui si cercò di riparare presto con restauri e risistemazioni di quello che era sentito come uno dei luoghi principi della memoria storica del regno. Una costante nella politica degli Attalidi fu la volontà assidua di intrecciare rapporti con uomini, luoghi e istituzioni della Grecia, presentandosi come partecipi e anzi membri organici dell’antica cultura che si era irradiata nei regni ellenistici. Già Eumene I aveva avuto relazioni con l’Accademia e il Peripato, mostrando la volontà di attirare artisti e intellettuali nella sua città: tuttavia il peripatetico Licone e l’accademico Arcesilao di Pitane, pure assai ben disposti nei confronti della corte attalide, non lasciarono Atene per Pergamo. Le dediche celebrative di Attalo I a Delo e a Delfi sono un segno evidente della sua volontà  40 Cfr. Pfeiffer 1973, 361.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

di istituire legami con due luoghi di enorme significato religioso tradizionale per il mondo greco: in particolare a Delfi il complesso monumentale di Attalo si inseriva nel contesto delle vittorie dei Greci sui barbari, in ideale consonanza e continuità. Le guerre contro Filippo V di Macedonia videro l’alleanza di Pergamo con Atene, la lega Etolica e Roma: in questo quadro, la visita di Attalo ad Atene nel 200 consacrò il suo prestigio nei confronti della città piú simbolica della cultura greca (dove fu addirittura istituita una tribù Attalide). I rapporti con Atene erano ben vivi e consolidati, naturalmente, anche sul piano strettamente culturale, principalmente con le due grandi scuole accademica e peripatetica: oltre a Licone, Attalo corteggiò, ancora senza successo, il nuovo scolarca dell’Accademia, Lacide di Cirene. Eumene II prosegui tenacemente su queste strade, e con migliori risultati. Dei rapporti con Atene è documento grandioso la Stoa di Eumene, a sud dell’Acropoli fra il teatro di Dioniso e l’Odeion di Erode Attico; ma anche Delfi è ricca di testimonianze della propaganda del re pergameno. Con i Romani invece, vincitori a Pidna nel 168, i rapporti diventarono tiepidi se non diffidenti. Eumene affrontò da solo i Galati, contro i quali ottenne una nuova vittoria (168– 166), celebrata con tanto maggiore impegno quanto piú era stata diffìcile: il re e il fratello Attalo (che sarà poi il re Attalo II) dedicarono per questo monumenti vari a Pergamo e altrove. Sull’acropoli di Atene fu realizzato quello che fu definito «il piccolo donario di Attalo», nel quale erano rappresentate le lotte vittoriose degli Dei Olimpi con Eracle contro i Giganti, degli Ateniesi contro le Amazzoni e contro i Persiani, dei Pergameni contro i Galati: l’evidente parallelismo rivela chiaramente ancora una volta l’intenzione di inserirsi nel quadro della tradizione storica e culturale greca. Altra realizzazione di Eumene fu il già ricordato imponente altare di Zeus Soter (la cui costruzione è collocata fra il 166 e il 156 a.C.). Anche qui nel fregio era raffigurata la vittoria degli Dei Olimpi contro i Giganti, cui si affiancavano le vicende di Telefo, che riprendevano il motivo delle origini della dinastia da Eracle: Telefo infatti era figlio di Eracle e di Auge, secondo alcune fonti il figlio che assomigliava di più al padre; ma c’era anche, dall’altra parte, il legame di Telefo con la guerra di Troia, che permetteva una connessione con il ciclo forse piú prestigioso e significativo della storia mitica dei Greci.41

 41 Telefo affrontò i Greci nel loro primo sbarco sulle coste asiatiche, uccise molti nemici ma poi fu ferito da Achille a una coscia: la ferita non guariva e Apollo vaticinò che Telefo sarebbe stato guarito soltanto da ciò che lo aveva ferito. Telefo si recò allora in Aulide, dove i Greci erano radunati per la seconda partenza alla volta di Troia, e Achille acconsentí a guarirlo spargendo sulla ferita pezzi di ruggine della propria lancia. Se questa è la connessione piú

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica Al regno di Eumene II si fa risalire la fondazione della Biblioteca, perché nelle fonti antiche egli è l’unico dei re di Pergamo ad essere esplicitamente collegato a questa istituzione. La fonte è Strabone (XIII 624): «[...] egli poi abbellì la città con ulteriori monumeriti e biblioteche e portò il centro urbano di Pergamo fino alle condizioni in cui si trova ora». Nacque cosí, accanto al grande santuario di Atena sull’acropoli, una delle piú famose e importanti istituzioni culturali del mondo ellenistico, che doveva porsi come contraltare, concorrente e rivale della celebre biblioteca dei Tolomei ad Alessandria (che peraltro dovette fungere, almeno fino a un certo punto, da modello).42 Gli sforzi per raccogliere libri furono intensi, la volontà di richiamare artisti e intellettuali non si arrestò. Una notizia della Suda ci dice che Aristofane di Bisanzio, non sappiamo per quale ragione, a un certo punto della sua vita progettò di fuggire a Pergamo presso Eumene, ma fu imprigionato: se la notizia è vera, la rivalità fra i due centri appare con chiarezza. Certo è che più tardi, ai tempi di Attalo I, a Pergamo si rifugiò Apollodoro, fuggito da Alessandria perché coinvolto nella crisi politica del 145–144 legata all’ascesa al trono di Tolomeo VIII: la sua Cronaca andava dalla caduta di Troia (1184/3) fino al 144/3 ed era dedicata appunto ad Attalo II.43 Alla fine Eumene riuscì a portare a Pergamo un filosofo stoico di Mallo (città della Cilicia nel sud dell’Asia Minore), Cratete: cosí Pergamo diventò un centro di cultura stoica, con ogni probabilità senza premeditazione e senza nessuna intenzione iniziale di contrapporsi su questo piano ad Alessandria. È assai probabile che Cratete abbia svolto un ruolo importante nell’organizzazione della biblioteca, ma nessuna fonte ci parla di lui come “bibliotecario”. Abbiamo poche citazioni di pinakes della biblioteca pergamena, con ogni probabilità esemplati sul modello alessandrino di Callimaco: ma non abbiamo nessuna idea su chi possa averli redatti. Per Pergamo non abbiamo notizia di una successione di letterati-bibliotecari come per Alessandria: l’unico nome che conosciamo è quello di un altro stoico, Atenodoro di Tarso, che a Roma fu amico di Cicerone e maestro di Augusto. Diogene Laerzio (VII 34) riferisce che il retore Isidoro di Pergamo parlava dell’espunzione di passi dalle opere di Zenone operata dallo stoico Atenodoro nel tempo in cui gli era stata affidata la biblioteca di Pergamo: se l’informazione è corretta, questa figura di filosofo-filologo del I sec. a.C. non

 nota, non bisogna trascurare un altro legame di Telefo con Troia: secondo quella che pare la versione più antica della sua nascita, la madre Auge si sarebbe trovata alla corte del re di Troia Laomedonte (padre di Priamo) e sarebbe stata ingravidata da Eracle quando egli conquistò la città in seguito alle offese patite appunto da Laomedonte. 42 Cfr. Canfora 1993 e Montanari 1993 [= questo volume, cap. 1]. 43 Cfr. Pfeiffer 1973, 276 sg. e 387.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

risulta poi tanto dissimile dai bibliotecari alessandrini, anche se si occupa di testi differenti e piú consoni al suo ambiente. Divenne una leggenda comune nell’antichità che la pergamena fu “inventata” in Pergamo quando Tolomeo V, contemporaneo di Eumene II, bloccò l’esportazione del papiro. Però in realtà scrivere su rotoli di pelle era cosa del tutto comune nel vicino Oriente nei tempi piú antichi e fu adottato dai Greci della costa occidentale dell’Asia Minore prima del V secolo a.C. [...] non siamo più autorizzati a dire che i Pergameni furono i primi a produrla nella qualità piú fine. Però sembra che l’abbiano prodotta in piú larga quantità, probabilmente perché l’importazione del papiro per gli scriptoria della biblioteca che si accresceva divenne troppo cara [...]. Se ci fu realmente per qualche tempo un embargo egiziano sul papiro, resta una questione aperta. La pergamena, in ogni caso, aveva un futuro glorioso [...].44

Della biblioteca si parlava ancora con ammirazione nel I sec. a.C.: Plutarco racconta che nel 32 a.C., nel contesto di un tentativo di Ottaviano di screditare Antonio presso il senato romano, un tale Calvisio, amico di Ottaviano, affermò che Antonio aveva regalato a Cleopatra la biblioteca di Pergamo, contenente circa 200.000 volumi (Ant., 58, 9). Questa è l’unica informazione che abbiamo sulla consistenza della biblioteca pergamena:45 ma subito dopo (59, 1) Plutarco avverte che la maggior parte delle accuse portate in quell’occasione da Calvisio ad Antonio pare fossero menzognere, e in effetti, a parte il numero dei volumi, la notizia (cosí caratterizzata in partenza e priva di qualunque conferma) che la biblioteca di Pergamo sia stata passata in blocco ad Alessandria è di quelle che lasciano molto perplessi e anzi decisamente scettici.46

La cultura pergamena Un buon quadro della cultura pergamena nel contesto dell’ellenismo è stata tracciata da Rudolf Pfeiffer, che ne ha ben delineato molti elementi caratterizzanti, non soltanto dal punto di vista della storia della filologia:47 Alessandria fu in parte il modello di Pergamo, in parte la sua rivale [...] nella filologia Cratete fu la figura predominante in Pergamo; in contrasto con i poeti filologi alessandrini, egli e i suoi discepoli si avvicinarono all’eredità letteraria come filosofi, e in particolare come stoici ortodossi. Questa differenza nell’approccio generale non impedí ai piú giovani

 44 Cfr. Pfeiffer 1973, Storia, cit., 363 sg. 45 Cfr. Canfora 1993. 46 Cfr. Pfeiffer 1973, 364. 47 Cfr. Pfeiffer 1973, 361 sgg.

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica studiosi di Pergamo di mettere a profitto i progressi che la filologia alessandrina aveva fatto nel corso di un secolo o piú.48

Pergamo come centro culturale prende avvio e si sviluppa dopo Alessandria, dunque in qualche modo può sfruttare un precedente già avanzato: ma ben presto diventa concorrente e rivale della città dei Tolomei, sia come polo di attrazione per intellettuali e artisti, sia per le attività connesse alla biblioteca come la ricerca di libri, sia per la produzione di cultura e il lavoro di interpretazione dei testi letterari. Tuttavia a Pergamo non si costituí una tradizione come quella alessandrina e non si ebbe una scuola paragonabile per ricchezza e per durata: questo è un aspetto fondamentale che la differenzia, cioè la mancanza (a parte il periodo di Cratete e dei suoi discepoli) di una lunga e feconda successione maestro-allievo legata a una specifica eredità metodologica e intellettuale. Un’altra differenza importante è data dall’assenza di grandi figure di poetifilologi come Callimaco o Apollonio Rodio o anche di scienziati-filologi come Eratostene: Pergamo fu piuttosto caratterizzata, come abbiamo già visto, dalla figura del filologo-filosofo nel solco della tradizione stoica. In tale cornice Cratete si distinse dal grammatikos alessandrino preferendo per se stesso la definizione di kritikos. Questa differenza di fondo, diciamo pure di radici culturali, porta con sé due tratti qualificanti, propri della tradizione stoica, di Pergamo in opposizione ad Alessandria: l’adozione dell’esegesi allegorica (respinta decisamente dagli Alessandrini) e dell’anomalia grammaticale (di contro alle concezioni analogistiche chiaramente profilatesi già con Aristofane di Bisanzio e Aristarco). Per quanto riguarda l’allegoresi, l’ermeneutica letteraria pergamena si ricongiunge all’alveo di una lunga tradizione filosofica, che risale all’età arcaica, fu accolta dallo Stoicismo e proseguirà per molte strade nei secoli seguenti; l’anomalia nella concezione della lingua era un portato delle riflessioni, già cospicue, della grammatica stoica, e i suoi sviluppi influenzeranno notevolmente il pensiero grammaticale successivo.49 La figura dominante della cultura pergamena durante il periodo attalide fu dunque, ai tempi di Eumene II, Cratete di Mallo. Egli sviluppò in forme di evoluta ermeneutica l’antica esegesi allegorica e riprese il concetto di anomalia secondo cui nella concezione della lingua alla ratio della normativa analogica si contrapponeva il piú libero sviluppo dettato dalla consuetudo. Su queste basi si contrapposero polemicamente Cratete e Aristarco nei loro scritti: soprattutto i

 48 Cfr. Pfeiffer 1973, 365. 49 Su questi aspetti cfr. Montanari 1993 (= questo volume, cap. 1).

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

versi dei poemi omerici furono il campo privilegiato del dibattito e non piccola documentazione si è conservata (nella scoliografìa e in altre fonti erudite o critico-letterarie). Filosofo stoico di formazione, Cratete era insieme filologo e interprete di testi: il fine della sua allegoresi, dunque, non era piú quello di portare nell’ambito di una dottrina i testi dei piú autorevoli maestri di pensiero della cultura greca, cioè di fare di Omero e Esiodo degli Stoici ante litteram, nelle cui opere le verità di pensiero erano presenti sotto immaginifici velami; il fine adesso semmai era quello di scoprire il vero significato delle parole dei poeti, era quello di interpretarne al meglio i racconti e le immagini. La distinzione non è cosí sottile: la filosofia qui appare come uno strumento esegetico piuttosto che uno scopo o un ideale cui piegare le parole dei poeti. Come dicevamo, non è esatto parlare di “scuola” pergamena, perché non è documentabile una regolare serie di maestri e allievi: comunque, i seguaci di Cratete polemizzarono ancora con quelli di Aristarco sui temi suddetti, anche se non di rado vi furono avvicinamenti, per esempio nel campo della grammatica. Siamo assai meglio informati sui discepoli di Aristarco (assai numerosi, e fra i primi ci sono personaggi di spicco come Dionigi Trace e Apollodoro)50 che sui seguaci di Cratete. Nelle fonti il nome complessivo plurale Krateteioi compare piú raramente di Aristarcheioi: in realtà, entrambi non offrono di per sé grandi informazioni e probabilmente sono indicazioni riassuntive e generiche nate in seguito sulla base di una ricostruita contrapposizione “di scuola”. Per esempio in uno scolio omerico l’epiteto di Krateteios è riferito a uno Zenodoto, che viene identificato con il grammatico Zenodoto di Mallo (ΙΙ/I sec. a.C.?) citato in altri passi e autore di uno scritto polemico contro le atetesi di Aristarco nel testo omerico.51 Ma di altri personaggi, probabilmente grammatici, identificati come Krateteioi sappiamo in genere assai poco e conosciamo solo qualche sporadico frammento.52 Un influsso pergameno, e segnatamente proprio di Cratete, è stato supposto negli scritti esegetici su Omero di Asclepiade di Mirlea in Bitinia (I sec. a.C.), grammatico e filologo attivo anche a Roma e poi in Spagna, autore fra l’altro di un libro Sui grammatici.53 Abbiamo già detto sopra del passaggio di Apollodoro a Pergamo presso Attalo II, il re cui era dedicata la Cronaca, invece Dionigi Trace (che si era rifugiato a Rodi) proseguí evidentemente la polemica, poiché scrisse un’opera dal titolo

 50 Cfr. Montanari 1993. 51 Cfr. Nickau 1972: lo scolio in questione è sch. ex. Ψ 79 b (V 380, 95 Erbse). 52 Cfr. per es. Kroll 1922; Pfeiffer 1973, 377. 53 Müller 1903; Pfeiffer 1973; cfr. Montanari 1993 (= questo volume, cap. 1).

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica Contro Cratete, nella quale trattava problemi omerici.54 Per un altro grammatico di scuola alessandrina, Parmenisco (datazione incerta, utilizzato da Didimo) è testimoniata un’opera Contro Cratete di argomento omerico.55 Interessante è anche il caso di Tolomeo Ascalonita,56 che a quanto pare visse e insegnò a Roma in età augustea57 e che le fonti collocano in una linea aristarchea-alessandrina: una sola volta si cita il titolo della sua opera Sulla dottrina di Cratete, nella quale probabilmente considerava e valutava le scelte e le opinioni del critico pergameno.58 Problematica la figura di Demetrio di Adramittio in Misia, detto Issione: la sua iniziale appartenenza alla scuola alessandrina post-aristarchea pare assodata, ma dai frammenti risultano anche le sue polemiche contro Aristarco e le opinioni alessandrine, per cui si ritiene (sulla scorta di una testimonianza della Suda) che in seguito sia diventato seguace di Cratete; incerta è la sua datazione, da alcuni posta immediatamente dopo Aristarco, mentre una fonte antica lo colloca in età augustea. Anche se le testimonianze che abbiamo non sono moltissime,59 bastano per intravedere un dibattito che rimase vivo per qualche tempo e diede sicuramente spazio ad approfondimenti esegetici e analisi letterarie di vario tipo. L’influenza esercitata da Pergamo su Roma fu cospicua: diversa nelle modalità e nei canali, certamente non inferiore a quella esercitata da Alessandria. In primo luogo, questo processo fu favorito dall’ampia diffusione dello stoicismo nel mondo romano: una corrente entro la quale anche le idee critico-letterarie e grammaticali pergamene scorrevano e circolavano agevolmente. I rapporti politici degli Attalidi con Roma sono stati accennati sopra: ed è ovvio che i legami divennero sempre piú organici quando Pergamo divenne città della provincia romana d’Asia. È significativo della posizione di Cratete in Pergamo il fatto che egli sia stato scelto per una missione presso il Senato di Roma nel I68: la sua permanenza fu prolungata da un incidente a una gamba che lo costrinse a una convalescenza, durante la quale ebbe contatti con il mondo intellettuale romano e tenne lezioni. Non c’è dubbio che questo episodio favorí molto il prestigio culturale di Pergamo nei confronti di Roma e aprí la strada a rapporti intensi. Panezio di Rodi (185– 110 a.C. circa) ascoltò Cratete a Pergamo, poi andò ad Atene e infine approdò a  54 Cfr. Linke 1977, 10 e fr. 15 col. commento. 55 Cfr. Blau 1883, pp. 48; Breithaupt 1915. 56 Cfr. Blau 1883, 25 sgg.; Baege 1883. 57 Fu dunque contemporaneo di Didimo e Aristonico: suo materiale è utilizzato da Nicanore e da Erodiano. 58 Nicanore in sch. Γ 155 b (I 387, 35 Erbse). 59 E non sono sistematicamente raccolte e valutate.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

Roma (dopo il 150), dove entrò nel circolo degli Scipioni: la sua formazione stoica e il legame con Cratete non gli impedirono di riconoscere talmente la grandezza di Aristarco, da gratificarne l’eccellenza di interprete con l’epiteto di «indovino».60 Abbiamo ricordato sopra Atenodoro di Tarso, che andò a Roma intorno al 70 a.C., fu amico di Cicerone e maestro di Augusto; parleremo fra poco di Galeno, il più illustre pergameno dopo Cratete, che nella Roma imperiale fu un personaggio di primo piano. Nel campo della grammatica, in età imperiale l’ambiente di Pergamo ebbe ancora un personaggio significativo in Telefo, vissuto nel II sec. d.C., che fu in Roma (assieme ad Efestione) maestro di Lucio Vero.61 In un quadro della cultura pergamena non possiamo trascurare una singolare personalità di erudito antiquario, Demetrio di Scepsi (II sec. a.C.), una città non lontano da Ilio, alcuni chilometri verso l’interno dell’Asia Minore sul corso dello Scamandro, nella quale Demetrio, a quanto pare, visse sempre. Per vero dire, i suoi legami con il centro di Pergamo non furono particolarmente forti, data la vita appartata che egli condusse. L’opera che lo ha reso famoso era un imponente scritto sulla topografia della regione di Ilio, al quale egli diede la forma di un commento (evidentemente sul modello dell’hypomnema alessandrino) al Catalogo dei Troiani (Il., 2, 816–77), in ben trenta libri. Questa summa di erudizione topografica e antiquaria fu scritta certamente dopo Cratete, nei confronti del quale Demetrio polemizza in uno dei frammenti conservati, e fu utilizzata da Apollodoro nel suo commento al Catalogo delle navi del II libro dell’Iliade: alcune citazioni si trovano nella scoliografia omerica, ma è probabile che altro materiale suo sia filtrato attraverso Apollodoro.62 «È una singolare coincidenza che la filologia fu rappresentata nella stessa generazione in luoghi differenti da Aristarco, Cratete di Mallo e Demetrio di Scepsi».63 In effetti, il II sec. a.C. fu un periodo particolarmente ricco e intenso: la contemporaneità di Cratete e Aristarco favorí il nascere di una polemica vivace, che prosegui in modo fecondo almeno per qualche decennio; Demetrio di Scepsi discusse opinioni di Cratete e il suo lavoro fu molto utilizzato da Apollodoro; la diaspora degli studiosi di Alessandria portò una diffusione del sapere con maggiore circolazione e confronto; fra Pergamo e Alessandria aumentarono i rapporti: della presenza a Pergamo di Apollodoro abbiamo già detto, e nel

 60 Cfr. Pfeiffer 1973, 377. 61 Wendel 1934a. 62 Cfr. Pfeiffer 1973, 382 sgg.; i frammenti sono in Gaede 1880. 63 Cfr. Pfeiffer 1973, Storia, cit., 385.

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica campo della grammatica l’influsso stoico-pergameno fu recepito e sentito dai continuatori della linea alessandrina. Su alcuni altri aspetti della cultura pergamena possiamo solo soffermarci brevemente. In primo luogo, quel tipo di ricerca antiquaria nel quale almeno in parte rientra anche Demetrio di Scepsi, era stato già coltivato negli ambiti dell’Asia Minore, grosso modo a partire dal regno di Attalo I, con Antigono di Caristo (autore di un’opera su pittori e scultori e di una raccolta paradossografica) e con Polemone di Ilio (autore di una periegesi antiquaria).64 Nell’ambito della letteratura tecnico-scientifica rientra il trattato di poliorcetica che un certo Biton dedicò a un Attalo, forse Attalo I.65 Ne sappiamo ben poco, ma esistette certamente anche una poesia celebrativa degli Attalidi. Nicandro di Colofone, importante poeta ed erudito e sacerdote del tempio di Apollo a Claros (località poco a sud della stessa Colofone), scrisse per un re di Pergamo un inno celebrativo. La Suda menziona: un Arriano poeta epico di età imperiale (datazione ignota), che scrisse poemi su Alessandro Magno e su un Attalo;66 il poeta epico Leschide, definito «celeberrimo», che partecipò alle spedizioni di Eumene II assieme al pittore Pitia e al medico Menandro.67 Ateneo ci menziona un Lisimaco (che sarebbe stato allievo, secondo Callimaco, del matematico Teodoro di Cirene, secondo Ermippo di Teofrasto), autore di un’opera Sull’educazione di Attalo (si tratta di Attalo I).68 Un personaggio di non secondaria importanza fu il geografo Menippo di Pergamo (seconda metà del I sec. a.C.).69 Purtroppo per gran parte perduta è la storiografia pergamena: conosciamo i nomi e pochi frammenti di Artemone (II sec. a.C.)70 e di Caristio (seconda metà del II sec. a.C.);71 sappiamo di Charax, uomo politico e storico (II sec. d.C.), autore di una storia universale e di una cronaca.72 Anche del retore Neante di Cizico (III–II sec. a.C.) è ricordata un’opera storica su Attalo I.73 Con tutto quanto abbiamo rapidamente ricordato siamo arrivati cronologicamente ben all’interno dell’età imperiale ed è quindi d’obbligo concludere il  64 Cfr. Montanari 1993 (= questo volume, cap. 1). 65 Ateneo, XIV 634 A: cfr. Gabba 1984, 21. 66 Suda, α 3867 Adler. 67 Suda, λ 311 Adler. 68 FGrHist. 170. 69 Lasserre 1981: cfr. GGM, I 566, 6 sgg. 70 FGrHist. 569. 71 Jacoby 1929; FHG, IV 356 sgg. 72 FGrHist. 103; Andrei 1984. 73 FGrHist. 171.

Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica  

nostro quadro della cultura pergamena dedicando uno spazio adeguato a Galeno, medico, filosofo, filologo. Di Galeno abbiamo esempi di estesi commentari di argomento medico (altri su temi filosofici sono perduti), la cui importanza e il cui significato nella storia della filologia devono ancora essere messi pienamente in luce attraverso una approfondita analisi anche dei suoi scritti su temi critico-letterari, linguistici e retorici.74 Egli nacque a Pergamo nel 129 o 130 d.C.: il padre Nikon, architetto e proprietario terriero di buoni mezzi e ampie vedute, gli diede una buona educazione filosofica (con esponenti delle quattro scuole allora attive: platonica, peripatetica, stoica, epicurea), ma un sogno lo avrebbe indotto ad avviare il figlio alla medicina.75 L’apprendistato anatomico iniziò dunque ben presto nella città natale, quando Galeno aveva diciassette anni, con il medico Satiro (discepolo di un celebre Quinto, a noi noto soltanto dagli scritti di Galeno stesso), che soggiornava allora a Pergamo al seguito di un importante uomo politico romano, Cuspio Pattumeio Rufino (console nel 142). Egli stesso ci racconta che la sua abilità tecnica fu presto messa a prova nell’epidemia di anthrax che colpí l’Asia Minore, e inoltre che scrisse un’operetta Sulla dissezione dell’utero per una ostetrica. Nel 150 d.C. circa, alla morte del padre, Galeno si trasferì a Smirne per seguire le lezioni di un altro anatomista, Pelope; poi volle andare a Corinto per sentire personalmente l’allora famoso anatomista Numisiano, ma non potè ascoltarlo, probabilmente perché era morto nel frattempo. Decise allora di passare ad Alessandria, che era in quei tempi il maggior centro di studi di anatomia. Vi si fermò un certo tempo e poi tornò a Pergamo, dove fu per alcuni anni medico dei gladiatori con grande successo. Si trasferì a Roma nel 161 o 162, all’inizio del regno di Marco Aurelio. Suo protettore era Eudemo, peripatetico, legato a un ambiente di eminenti personaggi di tendenza peripatetica, che seguivano un corso di anatomia privata tenuto da Galeno stesso: si trattava di Severo, maestro di Marco Aurelio, Demetrio di Alessandria, discepolo di Favorino, Barbaro, zio di Lucio Vero, Sergio Paolo, che sarà poi prefetto di Roma, Flavio Boeto, un ex-console a cui Galeno si legherà (gli curò la moglie) e dedicherà molte delle sue opere. La piú importante di esse è I procedimenti anatomici in due libri, che in anni più tardi (a partire dal 177 circa) fu riscritta in 15 libri. Nel primo soggiorno romano (162–166) lavorò ai primi sei libri del trattato Sulle

 74 Cfr. Montanari 1993, con la bibliografia (= questo volume, cap. 1). 75 L’unica edizione completa delle opere è quella di C.G. Kühn, in 20 volumi, Leipzig, Knobloch, 1821–1833; alcune opere mediche vengono edite nel Corpus Medicorum Graecorum pubblicato dall’Accademia di Berlino. La complessità del personaggio e la quantità della sua produzione rendono assai arduo tracciare una sintesi: per un orientamento cfr. Moraux 1985.

  Alessandria, Cirene e Pergamo: centri di cultura in età ellenistica dottrine di Ippocrate e Platone e al primo libro del grande trattato di anatomofisiologia De usu partium (gli altri sedici libri furono composti fra il 169 e il 176). Tornò ancora a Pergamo per un certo periodo; ma infine, a partire dal 169, rimase a Roma fino alla morte, che si colloca forse nel 199–200. L’opera di Galeno ci è rimasta in buona parte, anche se molti scritti, soprattutto di carattere non medico, sono andati perduti (una parte di essi bruciò nell’incendio del tempio della Pace a Roma del 169): egli stesso ci fornisce varie informazioni bibliografiche nella singolare operetta Sui propri libri, che si è conservata. La sua produzione letteraria abbraccia diversi campi, come la filosofia (etica e logica), l’analisi grammaticale (Sui nomi della medicina, Parole attiche notevoli, Lessico di Ippocrate), l’attività esegetica su Ippocrate, Platone e Aristotele (scrisse commenti a molti trattati ippocratici, al Timeo platonico, alle Categorie di Aristotele, che ci sono pervenuti solo in parte), l’anatomofisiologia, la terapia (soprattutto opere farmacologiche) e la patologia. Va ricordato infine, per il suo carattere propedeutico, il Protrettico all’arte medica.

 Filologia ed erudizione antica Il tema di questo intervento prende spunto da un dialogo, che purtroppo non c’è stato. Ne ripropongo, almeno in parte, gli argomenti in questa sorta di riconsiderazione a posteriori, nella quale assumono piuttosto le forme di una riflessione monologica, priva di un interlocutore che risponda nel modo inconfondibile con cui l’avrebbe fatto: ma l’interlocutore ideale è presente dappertutto e lo si vede benissimo in filigrana, con i suoi richiami al rigore del metodo e delle sovrane competenze filologiche, condite dai caratteristici scoppi della vis di un ragionatore implacabile, un erudito sovrano, un filologo appassionato e passionale. Chi prende in mano l’edizione teubneriana di Ipponatte di Enzo Degani, sul cui valore come punto di riferimento non è necessario dilungarsi,1 a proposito del fr. 129 (118 West,2 113 Sousa Medeiros,3 118 Masson4) ha occasione di imbattersi in alcuni di quei particolari eruditi, ai quali non sono molti a fare attenzione e dai quali sono attirati soprattutto gli studiosi con spiccati interessi per la storia della filologia e dell’erudizione antica. Pochi versi ipponattei sono ricostruiti grazie ai frammenti di un commentario al poeta, restituito da P.Oxy. 2176, edito per la prima volta da Lobel e datato al lI sec. d.C.5 Resti problematici di un hypomnema, sui quali molti hanno dato contributi di vario genere e con i quali hanno dovuto fare seriamente i conti gli editori di Ipponatte appena citati, con l’aggiunta di una più recente edizione di Slings, basata su una riconsiderazione completa del papiro.6 Un testo dunque molto studiato, che non voglio certo riesaminare per intero in questa occasione, con tutte le numerose questioni che hanno impegnato vari studiosi, e sul quale esiste una bibliografia ben accessibile: mi serve soltanto per spigolare qualche osservazione, da utilizzare come esempio e paradigma per una tematica di riflessione generale. Nel fr. A del commentario7 sono conservati i due versi iniziali del carme, nel quale il poeta si rivolge a un uomo di nome Sanno (oppure Sanna,8 comunque

 1 Degani 1991. 2 West 1989. 3 de Sousa Medeiros 1961. 4 Masson 1962; per l’edizione di Slings, cf. infra. 5 Lobel 1941, nr. 2176. 87–96. add. 184. 6 Slings 1987, 70–94. 7 L’indicazione dei frammenti del papiro con lettere dell’alfabeto è invalsa posteriormente all’ ed. pr., dove i frammenti erano indicati con numeri arabi: usano le lettere dell’alfabeto Sousa Medeiros, Masson, West, Degani; Slings invece usa numeri romani. 8 Degani, ad l.: «utrum Σάννος an Σάννας [ο Σαννάς ο Σαννᾶς?] fuerit, incertum». https://doi.org/10.1515/9783110772371-015

  Filologia ed erudizione antica giocato sul significato di ‘idiota’, ‘imbecille’), che è dotato di un naso sacrilego, blasfemo (forse metafora per indicare un fallo indecente) e non riesce a dominare gli istinti del ventre (rr. 1s.). Segue il commento relativo e più avanti nello stesso frammento si distinguono parti di altri due versi ipponattei (rr. 12s.), con il commento che li riguarda. Il fr. B ci restituisce un verso intero (λαιμᾶι κτλ.), che gli editori collocano in maniera diversa rispetto ai versi presenti nel fr. A: mentre Degani e Slings lo lasciano separato dai primi quattro,9 Masson e West invece lo stampano come v. 3 del componimento in continuità con i vv. 1s., il che li obbliga a supporre la caduta totale di un verso 4 (colon più breve) prima degli altri due versi del fr. A. Riportiamo per chiarezza la disposizione dei versi nell’edizione di Degani e in quella di West. Degani 129 a ὦ Σάνν’, ἐπειδὴ ῥι̑να θε͜ό[συλιν ]εις, καὶ γαστρὸς οὐ κατακρα[τεῖς, τοὗς μοι παράσχες, ὦ[ ].ν· σύν τοί τι βουλεῦσαι θέ[λω. b λαιμᾶι δέ σοι τὸ ⸤χει̑⸥λος ὡς ⸤ἐρωι⸥διου̑ West 118 ὦ Σάνν᾽, ἐπειδὴ ῥι̑να θε ͜ό[συλιν φορ]εῖς, καὶ γαστρός οὐ κατακρα[τεῖς, λαιμᾶι δέ σοι τὸ ⸤χει̑⸥λος ὡς ⸤ἐρωι⸥διου̑ [ ] 5 τοὖς μοι παράσχες [ ] σύν τοί τι βουλεῦσαι θέ[λω.

fr. A, r. 1 fr. A, r. 2 fr. A, r. 12 fr. A, r. 13 fr. B, rr. 2s.

fr. A, r. 1 fr. A, r. 2 fr. B, rr. 2s. fr. A, r. 12 fr. A, r. 13

Nella disposizione di West (e di Masson), il verso restituito dal fr. B (λαιμᾶι κτλ.) viene ‘incastrato’ fra quelli restituiti dal fr. A; in quella di Degani (e di Slings) i quattro versi restituiti dal fr. A restano nella successione testimoniata e il verso restituito dal fr. B rimane staccato, il che vuol dire a una distanza non precisabile (verosimilmente ritenuto più vicino da Degani, più lontano da Slings). Credo che Degani abbia ragione a comportarsi così nell’edizione dei resti ipponattei, perché nel fr. A il commentario papiraceo, dopo i vv. 1s. e il relativo commento (fr. A, rr. 1–11), riporta quelli che in Degani (e Slings) sono i vv. 3s., mentre in West (e Masson) sono i vv. 5s. (fr. A, rr. 12s.); solo nel fr. B (rr. 1s.) esso cita il verso collocato come v. 3 da West (e Masson) e lasciato invece separato da  9 Più precisamente: Degani stampa i quattro versi ipponattei del fr. A del commentario come fr. 129 a e stampa poi separato il verso restituito dal fr. B del commentario come fr. 129 b: Slings stampa i quattro versi restituiti dal fr. A del papiro, per lui fr. 1 (cf. sopra n. 7), come fr. I 1–4, e il verso restituito dal fr. B del papiro, per lui fr. III, come fr. III ipponatteo.

Filologia ed erudizione antica  

Degani e da Slings. Per accettare la sequenza stampata da Masson e West (sulla scia di un intervento di Maas)10 bisogna supporre in primo luogo che, commentando i vv. 1s., nel papiro venga anticipata la citazione e l’esegesi dei vv. 5s. del componimento ipponatteo (forse allo scopo di elucidare la costruzione e la connessione logico-contenutistica del discorso) e solo in séguito si passi a commentare il v. 3; e inoltre supporre che ci sia stata la perdita completa di un verso (colon più breve), che sarebbe il v. 4, il cui commento poteva collocarsi dopo quello del v. 3 nel fr. B, ma la cui supposta esistenza allontana ancora di più i vv. 1s. dagli altri due (3s. Degani–Slings; 5s. Masson–West), che compaiono insieme in successione nel fr. A del commentario. Dunque l’estensore dei testo esegetico avrebbe dapprima saltato due versi, passando da 1s. a 5s. non per ragioni di pura epitomazione (come capita e non farebbe meraviglia), cioè volendo omettere il commento di questi due versi per abbreviare, bensì solo a scopo esegetico, perché poi tornerebbe indietro per continuare sistematicamente: anche questo procedimento, in verità, non ha nulla di inammissibile di per sé, ma certo comporta l’onere della prova e il reperimento di argomenti forti che inducano a sovvertire la sequenza testimoniata.11 Lobel diceva che «there is not much doubt» sul fatto che il fr. A e il fr. B del commentario appartengano alla stessa colonna, ma fra di loro c’è una lacuna, la cui entità non c’è modo di conoscere;12 Degani e West scrivono che B «subter A stetit», il primo aggiunge «quo intervallo nescimus». Per supporre che il verso commentato nel fr. B sia il v. 3 del componimento (ipotesi Masson–West), bisogna dare per certa l’appartenenza dei due frammenti del commentario alla stessa colonna e immaginare che fra A e B ci sia una lacuna abbastanza piccola: solo così si può pensare che il commentatore abbia anticipato il commento della seconda coppia di versi in A, per tornare poco dopo in B a commentare il verso precedente del componimento. Dopo la sua attenta revisione, Slings mostra maggiore cautela e adotta una prassi conseguente. In primo luogo, osservando che l’affermazione di Lobel non si basava su prove papirologiche, bensì probabilmente solo su un’ipotesi ricostruttiva del poema, Slings si limita ad affermare che il fr. B = III appartiene allo stesso kollema dei frr. A e C = I e II (sulla base di considerazioni papirologiche materiali). Poi stampa i frammenti nella sequenza I = A, II = C, III = B (dato che il fr. II = C è costituito dalla seconda colonna del

 10 Indicazioni bibliografiche in Degani, ad l. ed in Degani 1984, 271s. 11 Non tocco la questione se tutti i frammenti restituiti da P.Oxy. 2176 appartengano allo stesso componimento ipponatteo: la cosa non appare in dubbio per quanto si ricava dai frr. A, B e C, cioè frr. 129 a b c Degani = frr. I. II. III Slings: cf. Slings 1987, 83s. 12 Lobel 1941, 184.

  Filologia ed erudizione antica pezzo di papiro in cui il fr. A è la prima colonna), prendendo definitivamente le distanze dalla «certezza» che A = I e B = III siano in sequenza ravvicinata, cioè che B = III costituisca la parte bassa del fr. A = I, vale a dire della prima colonna dell’insieme A–C = I–II.13 Questi elementi, con ogni evidenza, riducono ancora le probabilità dell’ipotesi ricostruttiva Masson–West. Si può infine aggiungere che se anche fosse assodata l’appartenenza dei frr. A e B = I e III alla stessa colonna, cioè se fossimo sicuri che B = III costituisca la parte bassa del fr. A = I (vale a dire della prima colonna dell’insieme A–C = I–II), questa sarebbe comunque un’indicazione in favore della sequenza stampata da Degani, fatto salvo il dubbio irrisolto sulla distanza fra 129 a e 129 b (e considerando anche che il fr. 129 b potrebbe, in questo caso, essere esattamente il v. 5 del componimento e seguire immediatamente i vv. 1–4 restituiti dal fr. A). Ripeto che in ogni caso spostare il verso λαιμᾶι κτλ. del fr. B incastrandolo come v. 3 fra le due coppie restituite dal fr. A, rifiutando la sequenza presentata dal commentario, avrebbe bisogno di una argomentazione molto solida. Preferibile dunque, a mio avviso, l’atteggiamento prudente scelto da Degani e appoggiato da Slings, che considera i quattro versi restituiti dal fr. I = A come i vv. 1–4 del componimento, senza intrusioni e separati dagli altri.14 Non ho elementi nuovi da portare sulla questione: ho voluto solo ragionare e prendere posizione, e soprattutto servirmi di un esempio per introdurre il problema della relazione fra testo poetico e testo del commento, che è sicuramente uno degli aspetti più interessanti e spinosi del lavoro da fare con questo tipo di tradizione erudita. I testi dell’esegesi antica assai spesso ci portano in due direzioni, diverse ma strettamente intrecciate: una è quella dell’opera poetico-letteraria oggetto dell’interpretazione, vale a dire la prospettiva del discorso commentato; l’altra è quella del discorso che commenta, cioè la prospettiva della storia della filologia nell’antichità come specifico àmbito intellettuale. Quando lavoriamo con questo genere di fonti, tenere separati i due aspetti è sicuramente sbagliato nel metodo e può essere causa di errori e fraintendimenti. Il discorso non vale naturalmente soltanto per i frammenti di commentari su papiro, su cui stiamo esemplificando, ma riguarda in misura maggiore o minore tutti i testi esegetico-eruditi che si sono conservati, dalla lessicografia alle opere grammaticali, dalla scoliografia alle trattazioni retoriche. Il caso e le diverse posizioni che abbiamo visto rappresentano l’esito non solo di una competente analisi relativa al poeta oggetto di commento, Ipponatte, ma  13 Cf. Slings 1987, 83: coerentemente con le sue osservazioni sul commentario, stampa il v. 129 b Dg., restituito dal fr. B = III, dopo i versi restituiti dal fr. C = II. 14 Cf. le argomentazioni in Degani 1984, 27 ss. e nn.; Slings 1987, 84–86.

Filologia ed erudizione antica  

della combinazione di questo elemento con un’attenta considerazione della specificità imposta dai difficoltosi frammenti di un hypomnema su papiro nella valutazione della struttura del discorso esegetico. Possiamo dedicarci ancora un poco a un’altra minuzia riguardante il fr. A. Il v. 3 Degani–Slings = v. 5 Masson– West mostra una piccola differenza, questa volta fra West e gli altri editori, che scrivono: τοὖς μοι παράσχες, ὦ[

].ν·

West invece (vedi sopra) omette l’ultima lettera prima della lacuna, cioè ω e l’ultima lettera superstite del verso, cioè ν. La ragione di questo comportamento editoriale è il modo in cui egli ricostruisce la struttura del commentario nel punto interessato, cioè fr. A = 1, rr. 10–14. Riportiamo anche qui l’edizione di Degani e quella di West (i lemmi ipponattei sono messi in corsivo grassetto). Degani

θεόσυλιν λω᾽

]. μεν[. . . ἀ]λλ᾽ οὐδὲ του ]ν· ‘ὦ Σά[ν]ν̣ ᾽, ἐπειδὴ ῥι̑να ]ς τοὖς μ̣οι παράσχες, ὦ ]. ν σὺν τοί τι βουλεῦσαι θέ-

10

14

West καλο]υ̣̑ μεν[. διὰ πο]λλοῦ δὲ τοῦ ὑπερβατοῦ φησι]ν “ὦ Σά[νν᾽, ἐ]πειδὴ ῥι̑να θεόσυλιν φορεῖ]ς, τοὖς μ̣οι παράσχες”. ὦ Σάννε, ἄκουσο]ν·“σύν τοί τι βουλεῦσαι θέλω”.

10

14

West in primo luogo congettura e integra διὰ πο]λλοῦ δὲ τοῦ | ὑπερβατοῦ φησι]ν ai rr. 10s. (e con un certo coraggio, bisogna dire, lo mette addirittura nel testo), ‘legittimando’ in questo modo l’idea che i versi citati subito dopo (a parte l’evidente ripresa a scopo esegetico da parte del commentatore dell’incipit ὦ Σάνν᾽ κτλ.), cioè τοὖς μ̣οι – θέλω ai rr. 12–14, siano distanti dai vv. 1s. e non li seguano immediatamente come vv. 3s.: questo è il problema che abbiamo discusso prima. Immagina dunque che il commentatore ricostruisca la ‘logica’ del discorso ipponatteo che, riprendendo dall’incipit, sarebbe in questi termini: «O Sanno, poiché hai questo naso sacrilego, ascoltami»: introduce per congettura ὦ | [Σάννε, ἄκουσο]ν come parafrasi scoliastica di τοὖς μοι παράσχες e solo a questo punto fa riprendere la citazione delle parole del poeta con σὺν τοί τι βουλεῦσαι θέλω. La conseguenza di questa ricostruzione è che la lettera ω alla

  Filologia ed erudizione antica fine del r. 12 e le lettere ]. ν al r. 13 non appartengono al poeta commentato bensì al discorso esegetico.15 Tutto questo è eccessivamente ricostruito, a mio avviso, sia per poter essere accolto nel testo del commentario, sia per accettarne tranquillamente le conseguenze nella ricostruzione del frammento di Ipponatte. Una minuzia, quest’ultima, che riguarda solo l’appartenenza di due lettere al testo commentato oppure al testo del commento: sulla quale comunque si gioca l’idea che il poeta riprendesse anaforicamente l’allocuzione al vocativo del v. 1 con un altro vocativo (uguale o diverso) già al v. 3 oppure al v. 5, oppure non la riprendesse per nulla (West), dunque una piccola considerazione sullo stile del poeta e sulla movenza iniziale del componimento. Torniamo ora al verso ricostruibile per intero dal fr. B = III del commentario: verso che è il fr. 129 b per Degani, che è separato dai vv. 1–4 da una distanza imprecisabile per Slings (fr. III), che è invece il v. 3 per Masson e West. Il testo riportato risulta da una collazione fra quello di Degani e quello di Slings: ometto le annotazioni introdotte da una seconda mano, che sono presenti sopra i rr. 2 e 4 e nel margine destro in corrispondenza dei rr. 2s.16 ]ηη [ ]φο[ ].η[ ]λαιμα̑ ι δέ σοι τὸ χεῖ]λος ώς [ἐρω]διοῦ [ἀπὸ] τοῦ λαιμοῦ ωσαν ...]. ωσει. [..]ε λέγε[ι. ἁρ]πακτικὸν δὲ τὸ ὄρνεο]ν ὁ ἐρωδιό[ς, διὸ] κ[αὶ] τοι̑ς περὶ τὸν Ὀδυ[σσ]έα ἐν τ[ῆι] νυκτ[ηγρ]εσίαι Ἀθηνᾶ ἐπιπέμπ[ε]ι τ[οῦτο]ν τὸν [οἰ]ωνὸν ἁρπασομένοις δηλονοτι π[ να ὥσπερ καὶ γει[ ἐρωδιόν Παλ(αμήδης) γρά[φει εὖ· ὡς ἐρωδιοῦ ω[ ος ἐκτιθείς τα.[ καθηργησαμεν[ νη̑ σον ταύτην[ μεν[ κα.[ ὡσ[ απε[ .εδ.[

5

10

15

Ipponatte continua a rivolgersi a Sanno: λαιμᾶι δέ σοι τὸ χεῖλος ὡς ἐρωδιοῦ, «il tuo labbro è vorace come quello (scil. come il becco) di un airone». Il commento  15 Cf. Degani 1984, 301 n. 95. 16 Masson 1962, 85 ha tralasciato i resti delle rr. 15–19 e (forse da lui influenzati) hanno fatto la stessa cosa Degani e West; Degani 1991 però le ha ripristinate negli Addenda, a p. 232.

Filologia ed erudizione antica  

per prima cosa indica la derivazione del verbo λαιμάω da λαιμός, «gola», poi c’è una parte non ricostruibile (alla fine del r. 3 abbiamo ωσαν, che può essere ὡς ἄν oppure ὡς ἀν-) e la frase termina con un verbo di dire al r. 4, il cui soggetto è con ogni probabilità il poeta (λέγε[ι oppure ελεγε[): dunque, il poeta usa il verbo λαιμάω da λαιμός, poi forse c’era «come» o «come se» o «in quanto» a introdurre un parallelo o una motivazione. Abbiamo poi una spiegazione della scelta della similitudine con l’airone: si tratta di un uccello rapace, tanto che anche nell’Iliade, X 274s., Atena manda un airone da destra come segno di buon auspicio a Odisseo e Diomede, appena partiti per la loro impresa notturna:17 per quanto segue, ἁρπασομένοις δηλονότι seguito da interpunzione mi sembra la struttura di frase più probabile. Il commento ha buoni paralleli negli scoli al passo omerico, dove nello schol. Il. X 274s. si trova τὸ ὄρνεον . . . ἁρπακτικόν ἐστιν e poco più avanti, nello schol. Il. X 299a, i due compagni sono indicati con la perifrasi οἱ περὶ Ὀδυσσέα.18 La struttura e il contenuto del commento diventano poi più difficili da interpretare: un punto particolarmente tormentato sono i rr. 10s., con la citazione del nome Palamede. A proposito di Παλ(αμήδης) γρά[φει dice l’apparato di Degani: «suppl(evit) et c(om)p(en)d(ium) solvit Lobel 95», scil. nell’editio princeps del papiro. Negli Addenda (p. 232) si legge: «de Palamede Eleatico, Pindari commentatore, f(or)t(asse) agitur, cf. M(a)ss(on) 162 adn. 1»:19 una breve annotazione già presente nella prima edizione20 e in una nota del volume di studi su Ipponatte che l’aveva accompagnata (praticamente in contemporanea).21 Nel papiro si trova soltanto πα con un piccolo λ in esponente ad indicare abbreviazione, cioè παλ, cui segue immediatamente γρ e poi un puntino «on the line consistent with α» (ed. pr.).22 Lobel integrò e commentò: «Perhaps Παλ(αμήδης) γρά[φει. Palamedes of Elea is known as the author of three works concerned with verbal criticism. His variant may have been δὲ] | εὖ» (cf. r. 11). Dunque, risalgono al primo editore sia lo scioglimento dell’abbreviazione in Παλ(αμήδης) sia il suggerimento di pensare a Palamede di Elea e a una sua variante testuale.  17 Il titolo della prima parte di Il. X è normalmente νυκτεγερσία (cf. schol. Il. X 1, vol. III I Erbse; Allen 1931, 269, ad Il. X 1; Strab. IX 5.18): la forma νυκτηγρεσία, qui restituita da Lobel (p. 184), era finora attestata solo in testi latini (cf. LSJ9 s.v.), mentre ci sono esempi del verbo νυκτηγρετέω accanto a νυκτεγερτέω. 18 A questo passo omerico Ipponatte fa allusione nel fr. 23 Dg. 19 Masson 1962, 162 n. 1.  Degani 1983, Addenda 222. 21 Degani 1984, 94 n. 76. 22 È sufficiente la buona riproduzione nell’ed. pr., Lobel 1941, pl. XII.

  Filologia ed erudizione antica Seguendo K. Latte,23 Degani respinge decisamente la possibilità di attribuire qui a un grammatico una variante ipponattea δὲ εὖ, evidentemente per δέ σοι tràdito nel lemma del papiro (cf. r. 2), che però comporterebbe uno iato intollerabile: ma non accetta nel testo neppure la possibilità di integrare οὐκ] | εὖ, proposta da Latte medesimo e invece accolta nel testo da Sousa Medeiros,24 Masson,25 West26 e Slings,27 seguendo la strada per cui εὖ del r. 11 sarebbe non un resto di variante ipponattea, cioè possibile parola poetica, bensì parte di un giudizio (positivo o negativo) sull’operato (quale che sia) del grammatico citato, vale a dire parola del commento. Degani rende conto di tutto con scrupolo in apparato, ma non si compromette per nulla nel testo: giusta prudenza, perché i problemi sono insoluti e probabilmente insolubili, a meno di nuove acquisizioni. L’interpretazione del passo poggia anche sulla struttura del periodo, oltre che sull’ampiezza delle lacune. La ricostruzione della lunghezza della riga rende impossibile che il r. 10 fosse completo con γρά[φει οὐκ e ci fosse semplicemente Παλ(αμήδης) γρά[φει οὐκ | εὖ, seguito da una interpunzione, riferito a qualcosa prima e perciò senza interpunzione prima di Παλ(αμήδης), come stampava Sousa Medeiros. Invece Musson, West e Slings accettano οὐκ | εὖ, ma giustamente considerano che in questo caso doveva esserci altro fra γρά[φει e οὐκ a causa della lunghezza del rigo:28 Musson e Slings lasciano un prudente spazio bianco; West propone (nel testo ma stavolta con un punto interrogativo) di integrare l’oggetto dell’osservazione (la γραφή del grammatico citato)29 nello spazio fra γρά[φει e οὐκ. In ogni caso, appare abbandonata l’idea che εὖ del r. 11 fosse una variante del verso di Ipponatte. Tutti gli editori citati interpungono dopo εὖ, intendendo dunque che a questo punto termini una frase. Non c’è invece uguale concordia sul mettere un’interpunzione prima di Παλ(αμήδης): lo fanno Degani e West; non lo fanno Sousa Medeiros, Masson e Slings. Nel primo caso abbiamo una frase che termina con ἐρωδιόν del r. 10 e che riguarda presumibilmente ancora il parallelo omerico citato appena prima, alla quale segue una frasetta autonoma nella quale si dice che Palamede scrive qualcosa bene o non bene, cioè approvato o no. Nel secondo caso, la menzione del grammatico appartiene a una frase più  23 Latte 1948.  De Sousa Medeiros 1961, 176.  Masson 1962, 85. 26 West 1989, IEC 153 (fr. 118); cf. West 1980, 119 (fr. 118). 27 Slings 1987, 72: nel commento ad l. dice «probable». 28 Masson inoltre stampa γρ(άφει) [. cioè supponendo che la parola fosse abbreviata come il precedente Παλ(αμήδης). 29 Che secondo lui potrebbe essere ὡς ἐρωδιός per ὡς ἐρωδιοῦ.

Filologia ed erudizione antica  

ampia, il cui inizio non è possibile definire: la struttura del periodo sarebbe dunque diversa e la menzione di Palamede potrebbe anche riferirsi a qualcosa di precedente la comparsa del suo nome (e quindi forse appartenente al contesto della citazione omerica), terminando sempre la frase con εὖ del r. 11. Il commento ipponatteo di P.Oxy. 2176 sembra piuttosto erudito, anche se nei limiti comuni ai resti di opere di questo tipo e di questa epoca (II–III sec.), che rivelano in genere il carattere di miscellanee compilate da fonti esegetiche precedenti, anche autorevoli ma epitomate e compendiate: per questo fenomeno e per la scarsità dei resti, non tutte tali fonti sono riconoscibili e valutabili.30 Nel fr. A si commenta il significato del nome Σάννος (così il commentatore, r. 3, intende la forma del nome Σάνν’ eliso dell’incipit ipponatteo)31 in modo sicuramente conciso rispetto ai dati che dovevano essere a disposizione:32 in compenso viene citato e discusso, rr. 6ss., un passo di Polemone di Ilio tratto dall’opera Πρὸς Ἀντίγονον καὶ Ἀδαι̑ον (di cui sono noti pochi frammenti citati da Ateneo con il titolo Πρὸς Ἀδαι̑ον καὶ Ἀντίγονον, cioè con l’ordine inverso dei nomi), a proposito del fatto che i Cretesi chiamerebbero σαννάδες le capre selvatiche, che sarebbero insensate e sciocche; evidentemente una preziosità erudita che ha suscitato curiosità.33 Nel fr. D, r. 22, si cita con ogni probabilità Aristofane di Bisanzio e forse altri nomi di filologi antichi si nascondono nei frustuli più piccoli (forse Ermippo di Smirne nei frr. H e I = frr. 11 e 12 Lobel). Dunque, nomi di grammatici ed eruditi ci sono, anche se non in quantità particolarmente rilevante, e nulla di strano che ce ne sia anche uno di nome Palamede.

 30 I giudizi di valore che si danno su questi prodotti dell’esegesi di età imperiale sono troppo spesso inquinati da due fattori: non si tiene conto del loro carattere miscellaneo ed epitomato e si opera un (magari inconscio) indebito confronto con un commento scientifico moderno. Degani 1984, 36: «Una eco diretta di questi commentari ellenistici è poi possibile cogliere nel già ricordato papiro di Ossirinco (nr. 2176), anonima compilazione del secondo secolo dopo Cristo, ma che tuttavia si rifa alla dottrina di autorevoli predecessori, da Aristofane di Bisanzio ad Ermippo [la cui citazione però è assai dubbia], a Polemone di Ilio (fr. A 6), citato, quest’ultimo, per la spiegazione della glossa cretese σαννάδας...»; Slings 1987, 70: «It is a commentary of the type called hvpomnema: lemmas followed by exegesis (in this case, a mixture of detailed. mostly irrelevant, information and close paraphrasis)». Il primo giudizio mi pare colga il valore del testo e lo identifichi meglio: non dimentichiamo che sono pochi frustuli lacunosi! Del resto, Pfeiffer 1968, 199, lo definiva «very learned commentary» (= Pfeiffer 1973, 313). 31 Cf. supra n. 8. 32 C’era sicuramente, tra l’altro, anche materiale di Aristofane di Bisanzio (fr. 1 Slater: cf. Degani 1991, ad l.; Pfeiffer 1973, 313), che è citato altrove nello stesso commentario, fr. D 22: cf. fr. 130 Dg.; Slings 1987, 82. 33 Pfeiffer 1973, 380.

  Filologia ed erudizione antica Ma chi è questo Palamede? Ed è proprio sicuro che un Palamede ci sia? Come dicevo sopra, nel papiro leggiamo l’abbreviazione παλ e Lobel suggerì: «Perhaps Παλ(αμήδης) γρά[φει». In effetti, lo scioglimento suscita qualche pensiero. La pratica di abbreviare i nomi dei grammatici negli hypomnemata è ben nota, però nei frustuli conservati di questo papiro sia Polemone (A 6) che Aristofane di Bisanzio (D 22) sono scritti per intero e per di più questa sembra l’unica abbreviazione almeno nelle parti di testo conservate.34 Palamede non è un nome di grammatico frequentemente citato, per cui bastino poche lettere a evocarne agevolmente il nome: e ci si potrebbe chiedere chi lo identificherebbe da un semplice παλ. D’altra parte, il fatto che non sia comune nell’àmbito dell’esegesi antica potrebbe essere invocato a favore della facilità di identificarlo, anche solo da poche lettere: specialmente se nel commentario era già stato menzionato prima. In verità, il dubbio era presente nel «perhaps» di Lobel, rimase nell’espressione «quisquis ille fuit» di Latte e nel punto interrogativo dopo il nome stampato da Sousa Medeiros, Masson e West. Degani e Slings invece eliminano il punto interrogativo e accettano decisamente lo scioglimento del nome: in effetti, quale altro nome si potrebbe invocare o come sciogliere altrimenti l’abbreviazione?35 Pensare a una forma dell’aggettivo παλαιός (παλαιὰ γραφή? οἱ παλαιοί?) ο agli avverbi πάλαι ο παλαιῶς (γράφεται?) può essere una possibilità;36 un’altra può essere πάλιν:37 ma certo anche per questi non ho trovato paralleli per l’abbreviazione e non saprei quale soluzione proporre e quale interpretazione del passo con uno di questi scioglimenti. Per il momento Palamede, che si trova in tutte le edizioni, rimane difficile da scalzare, pur con tutti i dubbi del caso. L’identificazione con Palamede di Elea è un passo in più, che Lobel suggerì con un discreto accostamento non troppo compromettente: Slings sembra crederci decisamente,38 mentre Degani conserva esplicitamente il dubbio: «de Palamede Eleatico, Pindari commentatore, f(or)t(asse) agitur». Attribuire un lavoro critico-filologico su Ipponatte a Palamede di Elea, di cui sappiamo che commentò Pindaro, compose una Λέξις τραγική e una Λέξις κωμική e fu autore

 34 A parte le annotazioni aggiunte negli interlinei e nei margini. 35 Cf. McNamee 1981, 77: l’abbreviazione πα(λ) o simili esiste con il significato di παλαιστή ο παλαστή, palmo (misura), in testi documentari. 36 Schol. Il. XVI 106b (ex.) ἄμεινον “καὶ φάλαρα” γράφειν, ὡς τῶν παλαιῶν οἱ πλείους. 37 Schol. Il. ΧΙ 451b (Did.) Ζηνόδοτος δὲ γράφει πάλιν, cui segue la lezione di Zenodoto. 38 Slings 1987, 79: «παλ interpreted by Lobel as Παλαμήδης (of Elea; date unknown but earlier than Athenaeus ...)».

Filologia ed erudizione antica  

di un onomastico,39 farebbe senz’altro guadagnare un paragrafo interessante nella storia dell’esegesi ipponattea antica. Tuttavia, un’insidia metodologica che questo genere di ricerche porta con sé è quella di spingersi molto avanti nell’allestire ricostruzioni plausibili ma con fondamenti incerti, laddove ciò che rimane è così esiguo e impervio. Degani ha ripreso questa notizia erudita con il puntuale interesse del filologo che vuole un’attenzione precisa a ogni particolare capace di gettare luce su un testo: l’identificazione era allettante come una sirena, ma egli ha mantenuto la cautela necessaria. Anche nel volume di Studi ha assunto una posizione prudente: nel capitolo dedicato all’antica esegesi ipponattea, ricorda e valuta con attenzione ed equilibrio il commento ipponatteo di P.Oxy. 2176 e le notizie che offre, ma menziona solo in nota la citazione di Palamede e la possibilità che si tratti di Palamede di Elea.40 Fermiamoci qui con queste minuzie erudite. Non ho cercato per l’occasione particolari forse un po’ eccentrici: ho ritrovato e ripreso in mano appunti che mi ero annotato tempo fa, interessandomi un poco del commento ipponatteo di P.Oxy. 2176 e proponendomi di discuterne con Enzo Degani quando ce ne fosse stata la possibilità. Avevamo parlato dell’edizione con commento di questo hypomnema da includere nella pubblicazione del progetto Commentarla et Lexica Graeca in Papyris reperta (CLGP), che ho promosso assieme ad altri studiosi italiani e stranieri e che anche in lui aveva suscitato interesse.41 Il momento di Ipponatte sarebbe venuto fra qualche anno, dato che allo stato attuale si lavora al vol. I, comprendente gli autori inizianti per A-B. Purtroppo l’occasione di discutere insieme dei problemi di questo commentario e della filologia ipponattea antica non c’è stata e oggi mi ritrovo a un monologo espositivo, la cui tristezza è  39 Su Palamede di Elea (se l’epiteto Ἐλεατικός si riferisce effettivamente alla città di provenienza), collocabile forse nel II/ΙII sec. d. C., cf. Wendel 1942; Gärtner, 1972. Si veda inoltre la scheda relativa nel Lessico dei Grammatici Greci Antichi (LGGA), consultabile in rete sul sito dell’Università di Genova, all’indirizzo e poi andando alla Home page della Facoltà di Lettere e Filosofia, oppure direttamente alla Facoltà di Lettere con l’indirizzo : nell’elenco dei Dipartimenti scegliere il «Dipartimento di Archeologia e Filologia Classica e loro Tradizioni (D.AR.FI.CL.ET.). poi e infine appunto . Si troverà una pagina introduttiva con le notizie essenziali, qualche esempio di scheda e infine l’elenco completo delle schede disponibili, che possono essere richieste in formato elettronico all’indirizzo e-mail indicato. Nel corso del 2002, le schede cominceranno ad essere via via direttamente leggibili in rete in formato pdf. 40 Studi cit. 36 e n. 76. 41 Cf. Montanari 2001a. Il progetto CLGP, al quale collaborano diversi studiosi, rientra ora nell’attività dell’Istituto Papirologico Vitelli di Firenze: il Comitato Editoriale è composto da Guido Bastianini (Presidente dell’Istituto Vitelli), Franco Montanari, Michael Haslam, Herwig Maehler, Cornelia Roemer.

  Filologia ed erudizione antica lenita solo dalla convinzione che Enzo Degani avrebbe apprezzato le discussioni che il destino ci ha sottratto: come apprezzava il rigore della ricerca, che impone sia di evitare di arrendersi troppo presto di fronte alle difficoltà del testo, sia di rinunciare all’eccesso di ricostruzione ipotetica non sufficientemente fondata. II suo articolo Filologia e storia (un titolo che certamente non manca di allusività) rappresenta adesso (è già stato detto) un testamento spirituale e metodologico di grande chiarezza e spessore.42 Sarà il caso che sia letto e fatto leggere come una medicina da cui trarre giovamento quando, accanto a studi improntati alla severità del metodo storico-filologico basato sulle più solide conoscenze, senza le quali interpretazione, ricostruzione e sintesi sono fragili e inconsistenti, ci si imbatte in ‛stimolanti’ divagazioni caratterizzate da spericolati e confusi dilettantismi (magari recitati ad hoc), che ottengono il favore e la risonanza del proscenio, suscitando la curiosità meno competente o più furba e nuocendo, oltre che alla scienza dell’antichità, alle stesse serie discipline che chiamano a congresso. Certo non è difficile comprendere che le ampie sintesi storico-culturali, le aperture metodologiche, le intuizioni di vasta portata sono acquisti di conoscenza importanti e di grande interesse: ma solo quando sono fondate su nozioni e competenze adeguate e su ricerche pazienti e tenaci per ogni elemento utilizzato. Altrettanto facile è rendersi conto che acquisire l’erudizione necessaria e impegnarsi nello studio puntuale dei particolari è duro e faticoso e richiede un tempo avaro di soddisfazioni, mentre la via delle ‛stimolanti’ nuove sintesi e divagazioni è senz’altro più comoda e gratificante: e poco male se i castelli in aria cadono in tempi brevi, hanno già svolto la loro funzione.43 Abbiamo esemplificato spigolando nei meandri di un frammentario e impervio hypomnema papiraceo su un autore, Ipponatte, che a Degani fu particolarmente caro. A chi utilizza non superficialmente la sua edizione, non sfuggirà l’attenzione, l’abbondanza e la competenza con cui vi sono messe a frutto le notizie variamente fornite dalla letteratura erudita antica, un àmbito per il quale egli provò sempre un interesse primario e profondo. Se adesso ci soffermiamo brevemente sulla lessicografia, non sarà difficile scoprire linee di pensiero, problemi e metodi che già sono emersi nelle considerazioni precedenti. L’ articolo Problemi di lessicografia greca costituisce un’esemplare incarnazione dell’idea di scienza dell’antichità di cui abbiamo parlato, per il modo in cui la sintesi metodologica e storica è presentata efficacemente grazie all’esame puntuale di casi rappresentativi e allo scavo filologico tenace di ogni elemento  42 Degani 1999. 43 Tenere dritto il timone su questa rotta forse eviterebbe qualche ondivago e viscerale errore, che rischia di mettere da parte nobilissime tradizioni.

Filologia ed erudizione antica  

utile.44 Con questo lavoro Degani si poneva subito come uno studioso che di questo àmbito di ricerca — la lessicografia greca in particolare e la letteratura erudita antica in generale — conosceva perfettamente le difficoltà peculiari (che lo rendono un terreno accidentato e poco gradito a molti) e comprendeva bene la portata storico-culturale autonoma (anche se piuttosto difficile da conquistare) e non solo ancillare. La problematica presentata da alcuni tipi di glosse esichiane ‛anomale’ vi è affrontata45 richiamandosi a due principi ben precisi, che emergono con valore di insegnamento e orientamento per la ricerca nel settore. Il primo è che bisogna comunque impostare correttamente, ancora una volta, la questione del rapporto fra discorso oggetto dell’esegesi e discorso esegetico, vale a dire fra la parola poetico-letteraria interpretata e la parola interpretante (in questo consiste l’anomalia di molte glosse difficili): cosa appartiene all’una e cosa all’altra, che cosa è la prima e che cosa è la seconda, da quali fonti provengono, con quali criteri e scopi sono state scelte e accostate. Il secondo è di evitare l’eccesso di rischio ricostruttivo e di azzardo nell’ipotizzare esegesi e interventi testuali, quando siano privi dei necessari fondamenti e dunque fuorvianti nei risultati proposti: questo però senza rinunciare a una motivata e documentata interpretazione, applicata con il necessario equilibrio basato su ragionamento e documentazione.46 Considerare le opere della letteratura erudita come produzione di secondo piano, dalla quale prelevare chirurgicamente il prezioso frammento poetico da trapiantare in una edizione critica, è un atteggiamento sbagliato, sul piano sia metodologico che storico-culturale, e genera errori insidiosi: i frammenti di hypomnemata e di trattati vari, la scoliografia, la lessicografia sono ricchi di esempi per questo. I resti di opere perdute, offerti dalle diverse forme di esegesi e di erudizione antica, sono ben utilizzabili e ben utilizzati qualora si capisca bene la fonte che li porta e il suo modo di ragionare e funzionare, qualora se ne intenda il dettaglio e il carattere generale. Queste fonti erudite hanno un importante significato storico-culturale di per sé e non soltanto una funzione ancillare come serbatoio di frammenti perduti o di notizie storico-antiquarie. Abbiamo già espresso sopra il concetto per cui i testi dell’esegesi antica ci portano in due direzioni legate e connesse fra loro: la prospettiva del discorso commentato, cioè quella dell’opera oggetto di interpretazione, e quella del discorso che

 44 Degani 1977–1978. 45 In reazione a Marzullo 1968, 70–87. 46 In questo campo l’insegnamento di Degani ha prodotto una scuola di assoluto rilievo: i suoi rappresentanti, ben noti a tutti gli studiosi, mi perdoneranno se in questa nota risparmio nomi e titoli bibliografici.

  Filologia ed erudizione antica commenta, cioè la storia della filologia antica come specifico àmbito intellettuale. Degani sapeva benissimo che c’è una dialettica ineliminabile fra i due elementi, per cui, se si vuole capire l’uno, si deve capire anche l’altro e fare ogni sforzo in questa prospettiva ‛unitaria’ nei confronti del testo. Il suo modo di affrontare le glosse esichiane ne è un esempio eloquente, tanto quanto il suo modo di usare le fonti erudite per il testo di Ipponatte. La serrata e tenace interpretazione puntuale sottende sempre la consapevolezza dello sguardo d’insieme, la coscienza dello sviluppo storico: come è giusto e sano, e garanzia di buoni risultati, i due aspetti si supportano a vicenda e le analisi di testi della letteratura erudita sono contributi di storia della filologia antica. Così è per Ipponatte, a proposito del quale non poteva mancare, nel volume di Studi un capitolo d’insieme non solo sulla storia della fortuna del poeta nell’antichità, ma anche specificamente sulla sua presenza nella filologia antica. Così è per la lessicografia, per la quale egli ha fornito anche una efficace e utilissima sintesi, sostenuta da quell’informazione ricca e necessaria che ne fa una valida introduzione al tema consultabile per informazioni su un quadro che pochi conoscono nel loro insieme.47 La posta in gioco in questo campo riguarda almeno due sfere importanti. Da una parte — la prospettiva del testo interpretato — si tratta di una giusta valutazione della tradizione indiretta dei testi antichi e dell’uso corretto da farne nell’edizione critica e nell’interpretazione. Dall’altra — la prospettiva del discorso esegetico — si tratta di capire come e con quali mezzi la civiltà antica ha compiuto lo sforzo di interpretare se stessa e conservare i propri contenuti, gli strumenti intellettuali che si è data per questo e i percorsi storici con cui li ha messi all’opera e ne ha ricavato i risultati voluti. La dottrina e la consapevolezza metodica con cui Enzo Degani ha lavorato anche in questi campi della scienza dell’antichità lasciano un insegnamento importante, che non si può dimenticare: l’ebbrezza sconfinata del documentarsi e dello studiare, l’incontenibile passione per l’indagare e per il ragionare.

 47 Degani 1987; 1995.

 Glossario, parafrasi, ‘edizione commentata’ nei papiri L’ecdotica e l’esegesi erudita antica utilizzano forme e tipologie librarie diverse, la cui analisi registra progressi, precisazioni e raffinamenti via via che la documentazione aumenta e gli studi si approfondiscono. In questo quadro, prenderemo in esame due casi: l’osservazione privilegia la documentazione dei papiri, ma risulterà evidente che all’occorrenza si può applicare anche ai codici di epoca bizantina. Il primo caso riguarda la definizione/descrizione di glossario e parafrasi e la loro differenza, non sempre chiarita con attenzione dagli studiosi, che si cercherà di puntualizzare al meglio. Il secondo esamina la possibilità di utilizzare la categoria di ‘edizione commentata’, cercando di definirne le caratteristiche e i limiti in base alla documentazione disponibile. Bisogna avvertire che il discorso procede, spero entro limiti accettabili e con risultati utili, per semplificazioni e generalizzazioni, anche un poco teoriche, che hanno lo scopo essenziale di costruire una griglia di riferimento. La realtà concreta e fattuale, come sempre, può presentare una poikilia più variegata e qualche volta anche in parte refrattaria a entrare nelle categorie predefinite: ma questa dialettica è così antica e produttiva, che la sua applicazione non solo non deve costituire un freno, ma anzi deve offrire conforto. Cominciamo dalla considerazione di glossario e parafrasi e dei loro rapporti. Essi hanno una caratteristica in comune: il tipo di materiale esegetico che offrono e il fatto di fare riferimento a un testo specifico, che seguono nel suo ordine sintagmatico. Per quest’ultimo aspetto differiscono in modo essenziale dai lessici, che sono caratterizzati da un ordinamento alfabetico, più o meno raffinato, ma pur sempre di carattere paradigmatico. Dedichiamo una piccola parentesi a ricordare che i lessici, a loro volta, si distinguono fra lessici d’autore (come il lessico omerico di Apollonio Sofista o ad esempio il frammento di lessico P.Oxy. 2393, del I sec. d.C., che contiene un lessico di Alcmane) e lessici di portata più ampia, che possono essere lessici generali (ci sono importanti esempi nei papiri, ma si pensi anche a Esichio, Fozio, gli Etimologici) oppure lessici di genere (come i frammenti di lessici comici su papiro o ad esempio le raccolte lessicografiche sulla commedia e la tragedia testimoniate per Didimo). Il glossario e la parafrasi presentano sostanzialmente lo stesso tipo di esegesi e di materiale volto a spiegare un testo. Il loro scopo essenziale è la spiegazione del significato letterale, vale a dire di offrire una ‘traduzione’ di un testo ‘difficile’ in una forma di greco più semplice, più facilmente comprensibile e anche,

https://doi.org/10.1515/9783110772371-016

  Glossario, parafrasi, ‘edizione commentata’ nei papiri entro i dovuti limiti, più vicina alla lingua parlata. Nel processo ermeneuticoesegetico, dunque, identificano l’operazione di primo approccio e presentano lo stesso tipo di materiale. Differiscono però nel modo di presentarlo, e qui dobbiamo richiamare l’attenzione degli studiosi perché talvolta la pubblicazione di frammenti su papiro comporta definizioni imprecise. In sintesi, il glossario seziona il testo in singole parole e espressioni e lo organizza affiancando a ciascuna la sua ‘traduzione’ (più spesso in due colonne, ma non sempre); la parafrasi costruisce un discorso continuo e autonomamente leggibile, che ‘ricopre’ totalmente il testo ‘tradotto’, laddove il glossario può lasciare dei buchi e saltare anche molte parole, come succede. Questa distinzione deve essere tenuta ferma. La ben nota regola secondo cui ogni tipo di esegesi a un testo comincia fitta e va diradandosi via via che procede (per diventare talvolta assai saltuaria verso la fine), porta a qualche incertezza quando ci si trova di fronte a un glossario della parte iniziale di un testo (tipico il caso dell’Iliade, naturalmente) e si vede che il testo è tutto glossato e tradotto, senza lacune, comprese anche le particelle connettive: se sono conservati pochi versi, la tentazione è forte; ma se il pezzo è un po’ più consistente, non bisogna aspettare molto per rilevare i primi salti e ben presto il discorso formato dalla successione delle glosse non è più leggibile come discorso compiuto e sintatticamente autonomo. Questo è un glossario, mentre per esempio il PSI 1276, del I sec. a.C. (dunque piuttosto antico in questo campo), contiene una parafrasi: a ogni esametro segue, rientrato, un rigo appunto di parafrasi, leggibile come una prosa continua che ‘traduce’ il discorso in versi del testo omerico. Non è detto che la parafrasi debba essere per forza interlineare e presentata insieme al testo: altre tipologie librarie si possono trovare, la parafrasi può anche esistere da sola e presentarsi come un testo separato da quello parafrasato. Manca nella parafrasi una relazione biunivoca e necessaria con un qualcosa che faccia funzione di lemma, necessario invece nel caso del glossario. La parte glossografica degli Scholia D ad Omero risponde precisamente a questa griglia concettuale: nei manoscritti principali tali scoli si presentano come una sequenza di lemma e interpretamentum separato dal testo omerico, ma appunto con l’essenziale presenza dei lemmi. Se la si guarda accanto alle parafrasi bizantine continue (non interlineari), quali quelle attribuite a Psello o altre analoghe, la differenza è del tutto chiara. Vale la pena avvertire che è capitato di trovare (non in edizioni recenti, se non mi inganno) l’etichetta di parafrasi attribuita anche a hypotheseis di un canto omerico, che rappresentano invece una tipologia di testi assolutamente ben definita e a sé stante.

Glossario, parafrasi, ‘edizione commentata’ nei papiri  

Passiamo ora a prendere in esame quella che da qualche tempo preferisco chiamare ‘edizione commentata’, definizione che ritengo più idonea a spiegare le caratteristiche di alcuni prodotti esegetici, che non si lasciano ben descrivere come parafrasi o commentari o altro. Preciso anche che ritengo piuttosto aleatorio individuare come una categoria il ‘commento a lemmi continui’, la cui tipologia risulta assai poco delineata. Per questo saranno chiamati in causa il P.Fay. 3 dei Topica di Aristotele e il famoso P.Lille degli Aitia di Callimaco. Un solo testimone papiraceo conosciuto di esegesi aristotelica appartiene all’arco cronologico della prima fase dell’esegesi aristotelica antica, che va da Andronico (I sec. a.C.) fino all’opera di Alessandro di Afrodisia (II – inizi III sec. d.C.): il commento ai Topici restituito dal P.Fay. 3, della fine del I sec. d.C. Si tratta di un prodotto librario definito ‘commento’ oppure ‘parafrasi’, mentre un’attenta osservazione mostra che entrambi i termini non lo descrivono bene.1 L’unica parte di cui si riesce a capire qualcosa del contenuto sono le righe di esegesi che seguono il primo lemma nella col. I (rr. 10–14), nelle quali si trovava senz’altro almeno qualche elemento che va al di là della semplice trascrizione parafrastica del testo aristotelico a scopo di comprensione letterale e che lascia intravvedere un certo sforzo di spiegazione del senso del discorso (σκέπτου μὴ κτλ., r. 10). La nuda parafrasi è interrotta da un richiamo del ‘commentatore’ al lettore su un punto particolarmente problematico. Dato che questa è la sola almeno parzialmente comprensibile delle parti rimaste, sarà azzardato pensare che fosse un caso isolato o eccezionale in un contesto generalmente del tutto elementare e modesto: sarà più plausibile che elementi di questo genere fossero presenti con una qualche regolarità. È vero che siamo abbastanza lontani dal commentario di Alessandro di Afrodisia, il primo ai Topici che sia pervenuto e che era stato preceduto da quelli perduti del suo maestro Ermino e del più antico Sozione. Per questo papiro, Sozione (I sec. d.C.) è l’unico, a nostra conoscenza, che potrebbe teoricamente entrare in gioco per motivi cronologici, ma non è davvero il caso di avanzare spericolate attribuzioni.2 L’opera testimoniata nel frammento appartiene a una fase piuttosto iniziale della filologia sulle opere esoteriche e ancora abbastanza vicina alle vicende ‘librarie’ di Andronico: dunque nel pieno del periodo, anteriore ad Alessandro di Afrodisia, che appare caratterizzato da uno sforzo filologico ed esegetico di base,

 1 Cfr. CLGP 1, 1, 4, pp. 241–244 e 260–264. 2 Cfr. Moraux 2000, II, 1, pp. 215–216; Funghi/Cavini, in CPF, III, p. 18.

  Glossario, parafrasi, ‘edizione commentata’ nei papiri volto a ordinare e chiarire il difficoltoso materiale ritrovato.3 Due caratteristiche appaiono essenziali. 1) Il testo dei Topici è presente per intero, diviso in pericopi il cui peso è quantitativamente pressappoco paritario in confronto alla relativa esegesi, che viene intercalata in eisthesis rispetto alla giustezza piena del testo aristotelico: questa mi pare la visione corretta, piuttosto che il contrario, cioè testo aristotelico in ekthesis rispetto alle righe di commento, è il testo aristotelico che detta la larghezza base della colonna. 2) Il discorso esegetico è caratterizzato da una parafrasi, nel corso della quale si aggiungono occasionalmente elementi di spiegazione del contenuto alla decodifica più o meno letterale del testo. Un’opera alla quale si adatta appunto la definizione di ‘edizione commentata’, veicolo di un lavoro che credo si collochi bene in quella prima fase della filologia aristotelica, caratterizzata dall’impegno a mettere ordine editoriale nelle opere esoteriche di Aristotele, per mezzo anche di basilari sforzi esplicativi di accompagnamento, per i quali, dato il tipo di opere, esito a usare il termine ‘elementari’, soprattutto se inteso con un valore unicamente ed eccessivamente riduttivo. Naturalmente, Moraux poteva ben dire che questo singolo ritrovamento non deve far concludere che i commentari ad Aristotele dei secoli I a.C.–I d.C. fossero tutti di questo tipo, cioè «fossero di misura modesta e passassero solo di poco la lunghezza dello scritto commentato, mentre lo sviluppo verso analisi testuali estremamente ampie si sarebbe verificato solo nel II sec. d.C., forse con Aspasio e Alessandro».4 Tuttavia in queste parole si percepisce quella tendenza a svalutare eccessivamente l’esegesi testimoniata nel papiro, di cui già abbiamo detto e da cui è opportuno guardarsi. Credo che abbiamo di fronte un prodotto che non accompagna l’edizione come qualcosa di aggiuntivo per spiegarla, ma è l’edizione stessa accompagnata da materiali esplicativi. Possiamo addurre un parallelo significativo. Il ben noto papiro di Lille di Callimaco, contenente versi della parte iniziale del III libro degli Aitia ‘con commento’. Il testo callimacheo è riportato per intero e si alterna con parti esegetiche in eisthesis; l’esegesi consiste per lo più in glosse o parafrasi, ma occasionalmente offre notizie di ordine storico o linguistico (è notevole fra l’altro una dettagliata nota prosopografica su una questione dinastica, rr. 3 ss.).5 Sottolineo ‘occasionalmente’: non c’è alcuna sistematicità e nemmeno abbondanza nell’aggiungere alla parafrasi continua del testo qualche osservazione che appare  3 Cfr. su questa linea le osservazioni di Abbamonte 2004, in partic. 29–33. 4 Cfr. Moraux 2000, II, 1, p. 211. 5 Testo in SH 254–265 (add. in Supp. SH 254–265).

Glossario, parafrasi, ‘edizione commentata’ nei papiri  

particolarmente utile o necessaria, e che riguarda solo una piccola porzione della pericope in oggetto. Per questo papiro ho usato per la prima volta, quando uscì, la definizione di «edizione commentata».6 Il volumen fu prodotto in Egitto una o due generazioni dopo la morte del poeta, essendo il papiro datato verso la fine del III sec. a.C.: la filologia callimachea era dunque proprio ai primi passi e le opere del poeta avevano conosciuto anche problematiche di sistemazione editoriale.7 Le analogie fra le situazioni in cui furono prodotti il commento ai Topici di Aristotele del I sec. d.C. e il commento agli Aitia di Callimaco della fine del III sec. a.C. appaiono abbastanza chiare per ritenere che la formula e presentazione libraria della ‘edizione commentata’, con le caratteristiche che abbiamo visto, rispondesse a esigenze simili. Forse, nei primi tempi di una filologia aristotelica alle prese con i testi procurati da Andronico di Rodi, si ritenne utile l’adozione di un modello che aveva dato prova di sé in condizioni confrontabili e paragonabili, pur con le evidenti differenze. In verità, non vorrei che questi due esempi inducessero a connettere per forza il modello della ‘edizione commentata’ alle fasi iniziali del percorso esegetico su un autore: una connessione possibile, ma non necessaria. Più prudentemente, mi accontenterei di ritenere individuata una tipologia libraria ed esegetica nella quale si presenta un testo completo, diviso in pericopi accompagnate e intercalate da un’esegesi di tenore prevalentemente parafrastico, con la possibile presenza aggiuntiva di altri contenuti esplicativi, determinati dal carattere del testo commentato: un prodotto evidentemente diverso sia dall’hypomnema che dalla nuda parafrasi interlineare e dal glossario (anche continuo); un prodotto che poteva rispondere a specifiche esigenze in varie situazioni, anche, ma non necessariamente, nelle fasi iniziali del percorso critico su un autore o un’opera. Concludiamo aggiungendo che i contorni non del tutto precisati e definiti di questa tipologia, unitamente a scarsità ed elementi di incostanza della documentazione, sono da ritenersi fisiologici: accade lo stesso per hypomnemata, lessici e marginalia, che presentano caratteristiche singole anche assai oscillanti e non riconducibili a categorie troppo rigorose e astratte, alle quali credo che in questo campo sia bene rinunciare. Ma forse si può tenere presente questa possibilità anche per tentare di capire frammenti di prodotti esegetici talvolta di ardua definizione. Posso aggiungere che i testimoni da esaminare in questa prospettiva sono in primo luogo, a mio parere, il ben noto papiro con un commentario  6 Così la definivo in Montanari 1976b, 147 (= questo volume, cap. 57); cfr. Del Fabbro 1979, 70–71. 7 Cfr. Montanari 2002b, 76–77 (= questo volume, cap. 58); Bastianini 2006.

  Glossario, parafrasi, ‘edizione commentata’ nei papiri all’epigramma dell’ostrica P.Louvre inv. 7733 = SH 983–984 (Suppl. SH 983–984), il commento a Nicandro, Theriaca, P.Oxy. 2221, e il P.Oxy. inedito con un testo riconducibile all’Athenaion politeia di Aristotele CPF Aristoteles 9 (T ?).8

 8 Cfr. Chambers 1971, 43; Montanari 1993c.

 Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica Una parte molto grande, certo maggioritaria, della Letteratura Greca antica ci è pervenuta solo in frammenti: basta pensare ai Fragmente der griechischen Historiker di Felix Jacoby, alle raccolte dei frammenti dei poeti tragici e dei poeti comici, ai frammenti dei grammatici e dei filosofi. Una poderosa mole di materiale che, salvo l’apporto minoritario di reperti papiracei (non di rado peraltro di notevole importanza), proviene da tradizione indiretta del più vario tipo. Una di queste fonti è la cosiddetta letteratura erudita, sulla quale è doveroso spendere alcune considerazioni preliminari. Quando si parla di letteratura erudita, di filologia, esegesi e commento ai testi nell’antichità, si tende a pensare essenzialmente in primis ai materiali della scoliografia e alla lessicografia, i due generi che si identificano immediatamente con questo concetto. Ma assieme a scoliografia e lessicografia dovremmo porre mente intanto alla grammatica e alla riflessione sulla lingua e poi — per cercare una definizione riassuntiva — a ogni testo variamente legato all’interpretazione dei prodotti letterari, comprese le raccolte erudite di materiali utilizzabili a scopo esegetico, i testi antichi sulla metrica, gli gnomologi e tutto quanto è in qualche modo legato allo studio dei testi letterari e dello strumento per crearli, cioè la lingua. Si tratta evidentemente di un fenomeno complessivamente di vasta portata culturale: una civiltà si rende conto dell’importanza dei testi della sua paideia e fa un poderoso sforzo per dotarsi degli strumenti per conservarne il testo corretto e per interpretarli, considerando la necessità di approntare strumenti idonei a tutte le operazioni necessarie e a comprendere lo strumento con cui questi testi sono fatti, cioè la lingua. Questo contributo si muove in questo ambito: la Ancient Scholarship, per evocare una definizione che è anche il titolo di un libro epocale di Rudolf Pfeiffer. Si tratta di un ambito di ricerca particolare (in tempi andati ancor più particolare di oggi, forse persino peregrino), estremamente vasto e multiforme, nel quale un importante cambio di prospettiva nella ricerca si è realizzato nell’ultimo mezzo secolo circa. È un dato di fatto accettato che il turning point sia stato la publicazione della History di Pfeiffer nel 1968. La generale e complessiva discussione dedicata a questo volume negli “Entretiens Hardt” del 1993, un quarto di secolo dopo la sua pubblicazione, dimostrava che il libro di Pfeiffer era già riconosciuto come un punto di riferimento, rispetto al quale si

https://doi.org/10.1515/9783110772371-017

  Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica potevano indicare ampliamenti e aggiunte alla sua trattazione, non solo dall’ovvio punto di vista cronologico.1 Oggi la domanda essenziale che uno studioso deve rivolgersi a proposito di questo ambito tematico è la seguente:2 cosa ci dicono i testi della filologia ed erudizione antica a proposito di se stessi? I prodotti che possediamo dell’erudizione antica sono stati a lungo considerati e studiati essenzialmente (pur variando singoli e specifici oggetti di indagine) per due ragioni e con due approcci: a) come testimoni di frammenti di opere perdute e per ricavare informazioni altrimenti ignote, per esempio su Realien, fatti storici, istituzioni e così via; b) come fonte di informazioni per l’interpretazione e la comprensione dell’opera che commentano (oppure, occasionalmente, di altre opere chiamate in causa dal testo esegetico). Il primo caso è immediatamente comprensibile: tutte le edizioni di opere frammentarie abbondano di citazioni che si trovano nei corpora scoliografici, nelle opere grammaticali e nelle raccolte lessicografiche. Non è raro che studi su vari aspetti di diverse epoche storiche traggano vantaggio da fonti erudite (bastino per tutti gli esempi dei FGrHist di Jacoby, dei Vorsokratiker di Diels-Kranz, delle raccolte di frammenti di tragici e comici). Dall’altro lato, accade talvolta che l’esegesi delle opere dei principali autori della letteratura antica sia chiarita da uno scolio o dal lemma di un lessico, che l’interpretazione di un passo controverso tragga vantaggio da una fonte erudita antica. Questi sono certo due aspetti rilevanti, che non devono essere né sottovalutati né messi da parte, dato che svolgono una funzione importante che non si può certo negare. Già da qualche tempo, però, l’approccio sta cambiando e questo cambiamento dovrebbe consolidarsi come un progresso acquisito nella conoscenza del mondo antico. Possiamo esprimerlo così: quanto sono significativi per quello che dicono su tutto ciò che è diverso da loro stessi e che ci viene da loro apportato, almeno altrettanto i prodotti dell’antica filologia sono importanti per quello che dicono a proposito esattamente di se stessi. Possiamo apprezzare il valore di un frammento sconosciuto di un’opera perduta o di un dato relativo al mondo antico altrimenti ignoto, ma non meno rilevante e significativo (talvolta forse anche più di questo) è ciò che tali testi, difficili e complessi, ci dicono sui metodi degli antichi esegeti, sui loro assunti intellettuali, le idee, gli orientamenti, gli intenti dei loro tempi e dei loro contesti, in una parola sulla loro cultura. È un dato di fatto — e non possiamo più negarlo o ignorarlo — che l’esegesi degli autori antichi, l’erudizione, la grammatica, la riflessione sulla lingua, tutto ciò  1 Montanari 1994a; cf. anche Montana 2012. 2 Cf. Montanari 2018 (= questo volume, cap. 3).

Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica  

che di solito definiamo genericamente come Ancient Scholarship, deve oggi essere inteso come uno degli aspetti essenziali del quadro storico-culturale del mondo antico (nonché come il passo finale nell’uscita da un classicismo estetizzante, le cui osservazioni e opinioni critiche sono irrimediabilmente effimere e per il quale i prodotti della filologia antica erano banalità di epoca tarda, di scarso o inesistente valore in se stessi). Domandiamo dunque ai testi della filologia antica, anzitutto, che cosa possono dirci su se stessi e adottiamo come principio base il fatto che non importa se quanto ci dicono è giusto o sbagliato secondo noi, se le loro interpretazioni sono buone o cattive dal nostro punto di vista e secondo i nostri metodi. Importa piuttosto ciò che implicano e ciò che significano di per sé: che uno scolio a Omero o a Eschilo scelga una lezione o un’interpretazione per noi sbagliata è del tutto secondario rispetto alla nostra comprensione dei metodi e degli assunti a cui si faceva ricorso per compiere quelle scelte testuali ed esegetiche. E del tutto inutile e ozioso è sottolineare quanto siano lontani dal giusto nelle scelte testuali, nelle interpretazioni dei passi, nelle ricostruzioni etimologiche: è ovvio e non serve a niente. Per difendere l’‘utilità’ dei testi della filologia antica non importa sforzarsi di segnalare che a volte interpretino bene secondo la nostra filologia e le nostre idee: importa invece comprendere le ragioni per cui hanno interpretato in un certo modo e hanno compiuto certe scelte, i loro presupposti intellettuali e culturali, per comprendere insomma ciò che dicono riguardo a se stessi, la loro epoca, il loro contesto, la loro cultura, i loro orientamenti critici e metodologici. Un punto essenziale diventa dunque il seguente. La svolta intellettuale della filologia antica è fondamentalmente una questione di metodo e di idee, non certo di qualità o quantità dei risultati. Un’altra questione sostanziale consiste nello stabilire se la fase storica della filologia alessandrina debba essere considerata come un momento decisivo di svolta intellettuale nella storia culturale della nostra civiltà oppure no, cioè se abbia determinato la nascita di un metodo di studio dei testi letterari che sta all’origine della disciplina che noi oggi chiamiamo filologia (classica). A questo proposito negli ultimi decenni c’è stato un acceso dibattito. Alcuni studiosi non hanno voluto riconoscere ai filologi alessandrini nessun tipo di attività comprendente la collazione di copie e la scelta fra varianti, ritenendo che le loro lezioni fossero solo congetture e con questo pensando di svalutare fortemente il loro significato storico e culturale nella nascita e nello sviluppo della disciplina filologica. Si tratta di un aspetto centrale nella valutazione della filologia alessandrina dal punto di vista della storia culturale. Sarebbe fin troppo facile far notare immediatamente che appunto la congettura è uno degli strumenti fondamentali della prassi filologica, cosa che

  Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica genererebbe una contraddizione insanabile e sufficiente a screditare una prospettiva che ritiene di svalutare i filologi antichi sostenendo che facevano solo congetture (come se questo non significasse appunto che erano filologi). Esistono d’altra parte, in ogni caso, testimonianze chiare e indiscutibili del fatto che veniva effettivamente praticata la collazione fra diverse copie e che si sceglieva fra varianti testuali testimoniate in diversi manoscritti. All’eventuale obiezione che poteva essere un fenomeno occasionale piuttosto che una prassi consolidata, si può facilmente rispondere che si tratta di un problema di principio e di metodo, non di quantità di dati (cioè numero di copie collazionate e di varianti discusse) o di qualità dei risultati (giusti o sbagliati dal nostro punto di vista di filologi moderni). Sarebbe del tutto ridicolo tentare di stabilire un numero minimo di copie da confrontare le une con le altre o di varianti da prendere in considerazione, tentare di determinare quante lezioni ‘corrette’ o interpretazioni ‘buone’ sono necessarie per poter parlare di filologia. Piuttosto, in una prospettiva storica, ciò che è necessario perché si compia un decisivo passo in avanti, in termini di progresso intellettuale, è il fatto stesso di comprendere e affrontare il problema, anche se in modo parziale, desultorio o incoerente: un testo letterario ha avuto una variegata storia della sua trasmissione, nel corso della quale può aver subito distorsioni e danni; è possibile ricostituire il testo corretto per congettura oppure scegliendo la lezione migliore tra quelle offerte da una tradizione discorde. L’attività dei filologi alessandrini comprendeva senza dubbio sia varianti ricavate dal confronto tra più copie sia congetture ope ingenii, vale a dire in linea di principio gli stessi strumenti della filologia moderna. Assume valore decisivo un’ulteriore considerazione di metodo, che riguarda l’invenzione del segno critico chiamato obelos da parte di Zenodoto. Egli mise in pratica due diverse operazioni che segnano un’importante svolta intellettuale di fronte a una tradizione testuale non univoca: da una parte la cancellazione materiale dei versi ritenuti sicuramente non autentici e da eliminare dal testo, dall’altra la possibilità di segnalare il sospetto che un verso fosse spurio, quando non si era abbastanza sicuri per eliminarlo in modo permanente e definitivo. Tale verso rimaneva così nel testo, accompagnato da un segno di dubbio che lasciava al lettore la possibilità di formarsi una propria opinione. Fu la codificazione del dubbio filologico, che nelle moderne edizioni critiche indichiamo con i segni dell’espunzione (o magari mettendola in apparato) per la parte di testo ritenuta incerta e discutibile, ma lasciandola a disposizione del lettore. Il riconoscimento di un guasto nel testo e della necessità di ripararlo rivela come l’unità organica fra interpretazione e critica testuale fosse un dato acquisito. Benché rimanesse ancora molta strada da fare e la moderna filologia “Wolfiana”, le edizioni critiche e il commento scientifico cui siamo

Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica  

avvezzi fossero in un futuro ancora lontano, il punto di vista che stiamo esponendo — lungi dall’essere un anacronismo — è la valutazione storica che un passo cruciale fu compiuto nel periodo tra Zenodoto e Aristarco e consegnato per sempre alla storia.3 All’inizio abbiamo accennato alla letteratura erudita come uno dei maggiori testimoni di frammenti di opere perdute. Questo tema rientra a pieno titolo nel discorso che stiamo facendo e ne rappresenta uno degli aspetti più significativi e importanti. Lo potremmo riassumere, forse genericamente e forse persino superficialmente, come la problematica dei loci similes o passi paralleli che si incontrano nelle argomentazioni delle opere esegetico-erudite. Anch’esso deve dunque essere guardato da una prospettiva del tutto nuova: in sintesi, non tanto come una fonte di frammenti di opere perdute, bensì soprattutto come testimonianza di una prassi esegetica e come prova delle conoscenze, della cultura, degli orientamenti degli antichi interpreti. Una caratteristica dei commentari provenienti dal lavoro filologico-esegetico degli eruditi grammatici dell’età ellenistico-romana e tardo-antica è infatti la frequente presenza di citazioni prese da autori e opere che sono differenti dall’autore e dall’opera oggetto del commentario stesso, oppure che appartengono allo stesso autore ma a un’opera diversa da quella commentata. A questa diffusa prassi della citazione di loci paralleli nel lavoro filologicoesegetico si faceva ricorso per le ragioni più disparate, che possono tuttavia essere ricondotte sostanzialmente a tre categorie principali. In alcuni casi un testo parallelo è citato perché giova all’interpretazione e alla comprensione del passo commentato: la citazione adempiva perciò ad uno scopo pratico di carattere esegetico e aveva una utilità diretta nel lavoro del commentatore. Talvolta, invece, si ricorreva ad essa perché il testo citato contiene una notizia o un’informazione interessante per il testo commentato dal punto di vista contenutistico, per i Realien, o sotto il profilo lessicale o linguistico: la citazione rientrava dunque in un approccio più generale e decisamente erudito, non era dettata da una linea di indagine solo rigorosamente esegetica ma piuttosto da curiosità conoscitiva ampliata e diversificata. In altri casi, infine, il testo citato (andando anche incontro, a volte, a distorsioni e travisamenti) serviva a corroborare una certa opinione sostenuta dal commentatore, più o meno legata al testo commentato. Abbiamo dunque a che fare, in linea di massima, con loci similes in senso ampio, cioè passi utili per una migliore comprensione del testo oggetto del commentario, oppure talvolta con materiali in grado di fornire ulteriori informazioni e notizie  3 Cf. da ultimo Montanari 2015 (= questo volume, cap. 10), 2015b (= questo volume, cap. 11) e 2018 (= questo volume, cap. 3), con bibliografia.

  Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica erudite di varia natura, o ancora con passi addotti per uno scopo ‘ideologico’. In quest’ultimo caso le citazioni ‘ideologicamente orientate’ prendevano come punto di partenza il testo commentato, che veniva però utilizzato come pretesto per dare avvio ad una più ampia argomentazione.4 Queste considerazioni sono valide principalmente per i testi di carattere esegetico, sostanzialmente la scoliografia, ma non bisogna dimenticare che la letteratura erudita nel suo complesso comprende svariate opere appartenenti ad altri generi, opere di carattere non esegetico o non direttamente esegetico, come tutto il grosso insieme della lessicografia e delle opere grammaticali. Tutti questi testi pullulano di citazioni di opere perdute di autori della letteratura antica, addotte a scopo di esemplificazione, di parallelo o di discussione concettuale. È dunque un serbatoio enorme, del quale indagare e cercare di capire il significato storico-culturale, evitando di fermarsi all’acquisizione di nuovi frammenti di opere perdute.5 Mi piace molto la formulazione del titolo di questa raccolta come “Frammenti e dintorni”, o “Fragmente im Kontext”. Voglio sottolineare una cosa, che ritengo sia della massima importanza. Noi parliamo di ‘frammenti’, che troviamo raccolti, studiati e commentati soprattutto nelle edizioni famose e canoniche di testi frammentari (FGrHist, tragici, comici, filosofi, grammatici), ma dobbiamo avere ben presente e ben chiaro che quelli che noi chiamiamo ‘frammenti’ sono in realtà citazioni all’interno di un testo che ce li restituisce e ce li fa conoscere, con tutto quello che questo comporta, prima di tutto che questi frammenti si trovano in un contesto, nel quale sono stati inseriti dall’autore per uno scopo ben preciso (che non è quello di essere poi raggruppati e inseriti in raccolte di testi frammentari come quelle a cui facevamo riferimento). Le citazioni, in un determinato contesto, si fanno con uno scopo specifico, che ha molte conseguenze e molte implicazioni. Certamente i passi d’autore citati erano assai più numerosi e abbondanti di quanto possiamo vedere oggi, perché, nel pesante processo di epitomazione prodottosi storicamente nel corso della trasmissione della letteratura erudita, è stato osservato come l’elemento che tende a cadere per primo sia proprio quello delle citazioni di passi paralleli, che vengono via via contratte e ridotte (magari riducendosi addirittura al solo nome dell’autore citato) fino a scomparire. L’entità delle testimonianze che oggi abbiamo di fronte ci dà quindi un’immagine assai limitata della poderosa mole di conoscenze dei testi poetico-letterari e dell’impegno profuso nell’utilizzarli a vari scopi esegetici, informativi ed eruditi: ma dobbiamo renderci conto che questo è uno degli  4 Cf. da ultimo Montanari 2016 (= questo volume, cap. 12), con bibliografia. 5 Cf. da ultimo Montanari, forthcoming.

Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica  

aspetti importanti per valutare il livello raggiunto dalla filologia antica, dalla cultura dei filologi antichi e soprattutto per capire la sua portata come fenomeno storico-culturale. Questo è il tema che dobbiamo e vogliamo indagare. Trattando di questi problemi, più volte ho scelto di partire da un esempio che ritengo emblematico e che può valere benissimo per introdurre e capire la questione. Lo ripeterò qui in forma sintetica. Si tratta del piccolo frammento PAmh. 2.12 (MP3 483), contenente le ultime righe del commento di Aristarco al I libro di Erodoto, con tanto di sottoscrizione Ἀριστάρχου | Ἡροδότου | α | ὑπόμνημα. Per i primi editori il frammento era interessante solo per due aspetti: 1) quello che ci dice sul testo oggetto di commento, cioè Erodoto (una variante testuale ed elementi esegetici di un certo valore); 2) il ritrovamento di nuovi frammenti di opere perdute.6 Però non dicono niente sulla storia della filologia come fenomeno storico-culturale, benché si trovino di fronte alla testimonianza di un’opera della massima importanza e prima sconosciuta (nulla si sapeva di commenti degli Alessandrini su testi storici). Il modo diverso, per quanto stringatissimo, di considerare il testimone in Pfeiffer 1968 è una prova lampante del cambiamento di punto di vista.7 Negli ultimi tempi la varia problematica relativa alla copiosa presenza di citazioni d’autore nella letteratura erudita si sta sviluppando significativamente, con risultati di notevole interesse. Certo questa è un’altra prova del cambiamento di prospettiva nella ricerca di cui dicevamo sopra.8 In generale, come si può ben comprendere, non è facile, anzi è assolutamente difficile fare un rilevamento e un panorama concreto dei passi d’autore sparsi come citazioni nei testi esegetico-eruditi e da noi di solito trattati come ‘frammenti’: gli ostacoli sono troppi e troppo gravi, gli strumenti del tutto inadeguati per non gettare un dubbio assai pesante anche sull’estrapolazione di semplici dati quantitativi, che si potrebbe tentare di ricavare in qualche modo, ma si rivelerebbero peraltro di utilità molto limitata, al massimo una base di partenza per un’analisi più approfondita e affidabile. Un’osservazione molto  6 Edito da Grenfell/Hunt 1901, 3–4 (Plate III): “The chief points of interest in the fragment are the proof of a hitherto unknown variant ἄμιπποι for ἄνιπποι in chap. 215 and a new quotation from the Ποιμένες of Sophocles”: cf. Montanari 2016, 74–76 (= questo volume, cap. 12), e Montana 2019 con bibliografia. 7 Pfeiffer 1968, 224 (1973, 348–349); nuova edizione con introduzione e commento in Montana 2019. 8 Di seguito alcuni riferimenti bibliografici, che documentano il consolidarsi dell’interesse per questo tema in generale: Montanari 1992a; Grisolia 1992; Montana 1996; Scattolin 2007; Nünlist 2009, 8–11 e passim; Perrone 2010; Montanari 2016; Vergados 2017; Pagani 2018; Phillips 2013; Gennari Santori 2018; Comunetti 2020; Montanari 2020 (= questo volume, cap. 50).

  Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica superficiale mostrerà come Omero sia l’autore largamente prevalente e come risulti (almeno tendenzialmente) costante la regola per cui l’autore commentato è il più citato negli scolî che lo riguardano.9 Per le complesse ed estese raccolte di opere esegetico-erudite trasmesse dalla tradizione medievale le cose sono molto più difficili e complicate. La ricerca delle citazioni di un qualunque autore in uno o più corpora scoliografici o lessici fatta con il TLG (per esempio le citazioni di Euripide negli scolî all’Iliade oppure negli scolî a diversi autori, negli Etimologici e in altri lessici) darà un risultato puramente quantitativo e per di più ben poco affidabile, perché includerà quanto meno doppioni dovuti a diverse redazioni di scolî, a occorrenze congetturali e magari anche citazioni addotte come parallelo dall’editore moderno (a seconda di come il testo è stato codificato nella banca dati e dello stato editoriale del testo stesso). Un quadro puramente quantitativo si rivelerebbe naturalmente solo un primo risultato, che mi sembra realizzabile solo per singole opere di erudizione e per singoli autori (per esempio le citazioni di Aristofane negli scolî all’Iliade), perché comporta che poi ogni citazione sia esaminata e capita puntualmente nel suo contesto (se conservato) e nelle sue ragioni. Le considerazioni svolte sopra offrono alcune basi per delineare una tipologia delle citazioni di autori e opere nella letteratura erudita, una griglia di riferimento da utilizzare.10 Ci si può servire degli indici delle edizioni delle varie raccolte erudite, ma sappiamo bene che diversi corpora scoliografici, lessici e opere grammaticali sono disponibili in edizioni anche molto vecchie,11 i cui indici sono anch’essi solo parzialmente affidabili: la qualità delle edizioni e dei loro indici spesso non permette un computo preciso senza una verifica puntuale di ogni citazione (certo un lavoro enorme). Si scoprirà ben presto che abbiamo a che fare con una tipologia che contempla casi in cui troviamo un rimando a autore + opera + testo in extenso (o parafrasi) del passo e casi in cui uno o due dei tre elementi sono assenti, in diversi casi probabilmente perché perduti nel corso della tradizione. Per un singolo autore chiamato in causa avremo la citazione di opere conservate oppure di frammenti di opere non conservate: un dato di grande interesse per la storia della fortuna di quell’autore. Ci saranno poi i casi in cui il riferimento è per antonomasia e dunque non compare in una ricerca automatica per nome.

 9 Cf. Montanari 1992a. 10 Cf. anche le considerazioni conclusive di Perrone 2010, 100–102. 11 Negli ultimi decenni le nuove edizioni delle opere della letteratura erudita si vanno moltiplicando, con grande vantaggio della ricerca in questo campo così vasto e dificile.

Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica  

Soprattutto è facile evocare, a questo punto, i casi in cui l’omissione del nome dell’autore comporta un dubbio di attribuzione e per contro i casi in cui rimane solo il nome o poco più, per cui non è agevole capire il motivo e il senso della citazione da parte del grammatico antico. Un passo avanti reale e attendibile può essere fatto soltanto dedicandosi a studiare casi concreti e ben individuati, tipo le citazioni di x negli scolî a y, e/o nella lessicografia o in altro genere di opere erudite. Ogni passo avanti della ricerca acquisterà così la concretezza dell’esame puntuale e preciso, con risultati attendibili. Se prendiamo per esempio le citazioni di Aristofane o di altri poeti comici negli scolî all’Iliade, siamo facilitati dall’esistenza dell’edizione di Erbse, che però non comprende i cosiddetti Sch. D e altro ancora della scoliografia iliadica (mi riferisco agli scolî della classe h, che Erbse ha utilizzato non completamente e sistematicamente solo per il I canto, mentre per tutto il resto del poema devono ancora essere esaminati).12 In questa occasione ho provato dunque a fare un esempio di questi possibili studi, osservando le citazioni in particolare di Eupoli e Cratino, con un accenno ad Aristofane, negli scolî all’Iliade, e inoltre alcune citazioni di Pindaro sempre negli scolî all’Iliade. Non mi soffermerò, anche per ragioni di spazio, sulla campionatura di scolî con citazioni pindariche, che sono trattati in un altro lavoro:13 ne riassumo sinteticamente i risultati e gli argomenti emersi. Abbiamo innanzi tutto un paio di semplici casi in cui Pindaro è citato per versioni differenti di elementi del mito rispetto a Omero.14 Disaccordi fra gli autori antichi su particolari di questo genere sono piuttosto comuni e spesso vengono notati, rivelando peraltro le puntuali conoscenze dei testi da parte dei grammatici antichi (sch. ex. Iliade 8.365–68 e sch. ex. Iliade 10.435: una variante mitica nella vicenda di Reso). Ancora uno scolio omerico (sch. ex. Iliade 13.20) cita un verso della Pitica 3 di Pindaro in parallelo con Il. 13.20. Il critico antico sostiene che Pindaro qui si sarebbe “ispirato” all’Iliade e aggiunge una serie di considerazioni basate sul confronto fra i due passi, per chiudere con un problema testuale pindarico.15 Più complesso e problematico è il caso, basato anch’esso su un parallelo fra Omero e Pindaro, trattato nello sch. di Aristonico a Iliade 2.670, nello sch. D a Iliade 2.670 e nello sch. Pind. Ol. 7.63 = 34, che si riferisce all’insediamento a Rodi dei discendenti di Eracle, guidati dall’Eraclide Tlepolemo, esiliato da Argo: il passo pose alcuni problemi alla filologia antica (come si legge negli scolî) e diversi altri ne pone agli studiosi moderni, sui quali ora non possiamo soffermarci (anche  12 Cf. Montanari/Montana/Muratore/Pagani 2017 (= questo volume, cap. 25). 13 Cf. Montanari 2020 (= questo volume, cap. 50). 14 Cf. Merro 2015. 15 Del problema testuale e metrico ho trattato in Montanari 1993b (= questo volume, cap. 50).

  Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica perché la nostra prospettiva in questa sede è del tutto diversa). La base del ragionamento e delle osservazioni dei critici antichi è senza dubbio l’evidenziazione e l’analisi del parallelo fra il passo di Iliade 2 e quello della Olimpica 7 di Pindaro. Ci basti sottolineare che in questo punto siamo praticamente certi che il materiale conservato risale ad Aristarco: per la filologia antica è stato essenziale rilevare e sottolineare il rapporto fra i due passi e la dipendenza di Pindaro da Omero, e analizzarne le differenze, sia dal punto di vista stilistico sia da quello del contenuto e della trattazione del mito. Il resto è un problema degli studi moderni. Chiariamo bene che non c’è una gerarchia di importanza fra le due prospettive e che sono importanti due aspetti: 1) tenerle ben distinte, 2) non subordinare la seconda alla prima, liquidando la critica filologica antica come irrilevante o sbagliata perché non risolve i problemi della ricerca moderna. Passiamo alle citazioni di frammenti dei comici. Nell’esame che segue ci basiamo sull’indice dei nomi propri dell’edizione di Erbse degli scolî all’Iliade. Le testimonianze sono presentate nell’ordine dei versi dell’Iliade: si dà prima il riferimento allo scolio iliadico (indicando la classe dello scolio), seguito dal numero del frammento dell’autore antico nell’edizione di riferimento. Per Cratino ci sono 6 citazioni, tutte con il nome, nessuna anonima.16 In due casi Cratino è citato assieme ad Aristofane, in un caso assieme a Eupoli. 1. Sch. Iliade 2.56c (sch. ex.), fr. 363 K. = 331 K.–A. (incertae fabulae). Con ogni probabilità, una testimonianza di carattere linguistico-grammaticale. Un neutro con valore avverbiale, per cui Cratino sembra offrire un parallelo per il passo omerico. Il testo del frammento è problematico: Erbse mette in apparato il testo dei mss. b e quello dello sch. h, che ha la forma probabilmente più vicina a quella corretta. 2. Sch. Iliade 7.76 (POxy 1087, col. II 36–38), fr. 101 K. + CGFP 72 = 108 K.–A. (παρὰ Κρατίνῳ, ἐν Μαλθακοῖς). Si tratta assai probabilmente di una testimonianza lessicale. Questo papiro (un tempo datato al I sec. a.C., oggi piuttosto al I sec. d.C.) restituisce un ampio frammento di commentario a Iliade 7.75–83, caratterizzato dalla tendenza alla disquisizione erudita di argomento lessicale. Si conserva una lunga nota (40 righi) a Iliade 7.76 ἐπὶ μάρτυρος ἔστω, nella quale sono elencati numerosi paronimi, intesi come nomi della seconda declinazione del tipo μάρτυρος, -ου, un nominativo derivato dal genitivo del corrispondente nome della terza declinazione, del tipo μάρτυς, μάρτυρος (paralleli in testi grammaticali). Fra i numerosi autori citati come testimoni di simili paronimi (Eschilo, Alceo, Antimaco, Euripide, Esiodo, Licofrone, Pindaro, Simonide,  16 È sempre possibile, per gli autori perduti di cui abbiamo solo frammenti, che ci siano delle citazioni nella letteratura erudita che non riconosciamo (magari sfuggono anche come citazioni).

Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica  

Sofocle, Stesicoro, Senofane, Archiloco e altri ancora) troviamo anche Cratino ἐν Μαλθακοῖς)* e poi Eupoli senza titolo della commedia (vedi infra Eupoli nr. 2). La ricostruzione del fr. di Cratino è problematica (una parte è in lacuna), ma certamente la citazione — come del resto tutte le altre della nota del commentario — offriva una testimonianza lessicale (vedi K.–A. 1983, 176–77 ad loc.). Sch. Iliade 9.77b1 (sch. ex.), fr. 158 K. = 164 K.–A. (ἐν Πανόπταις Κρατῖνος). Si tratta di una testimonianza di carattere linguistico-stilistico. Lo scolio dice che non è necessario sottintendere nel passo omerico un participio congiunto al verbo principale (cioè sottintendere ὁρῶν): si tratta invece di una παλαιὰ συνήθεια appunto di carattere stilistico. Per espressioni simili in parallelo al passo omerico vengono citati questo frammento di Cratino e Aristofane, Acarnesi 13: curioso è che nello scolio per la citazione di Aristofane siano caduti sia il nome dell’autore che il titolo dell’opera, entrambi conservati invece per Cratino. È dunque una delle citazioni di Aristofane senza nome, cf. infra. Sch. Iliade 16.428b (sch. ex.), fr. 406 K. = 444 K.–A. (incertae fabulae). Testimonianza di carattere linguistico-grammaticale, probabilmente di un uso attico. A commento di una forma di nominativo femminile plurale di un aggettivo composto, che si trova in questo verso omerico (ἀγκυλοχεῖλαι da ἀγκυλόχειλος, anziché ἀγκυλοχειλεῖς da ἀγκυλοχείλης), Cratino è citato come parallelo morfologico per un altro aggettivo con forma morfologicamente analoga (δωδεκ(α)έται anziché δωδεκ(α)έτεις, da δωδεκ(α)έτης). Per la stessa forma δωδεκ(α)έται in un passo di Cherobosco (che parte dallo stesso verso omerico Iliade 16.428: Choerob. in Theodos. Can., GG IV 1, p. 167.37ss.) si citano οἱ Ἀττικοί: probabile dunque che il parallelo di Cratino fosse addotto come testimonianza di un uso attico. Sch. Iliade 18.521b1 (sch. Hrd.), fr. 364 K. = 397 K.–A. (incertae fabulae). A quanto pare abbiamo un parallelo di carattere linguistico-prosodico. Si tratta di un problema di quantità di sillabe, per cui Erodiano cita Aristofane, Cavalieri 96 = 114, ma anche questa volta il nome dell’autore e il titolo dell’opera sono perduti, come nel successivo sch. 18.521b2 (cf. infra, Aristofane, citazioni senza nome), e poi Cratino. È incerto se nel frammento di Cratino si debba leggere Ἄρης / Ἄρην oppure ἀρήν: vedi discussione in K.–A. 1983, 308–309 ad loc. Sch. Iliade 23.361a1 (testimonia). Questo frammento è tramandato da alcuni mss. degli sch. h all’Iliade, che presentano materiali in più rispetto agli scolî principali, e si trova solo nell’apparato dei testimonia di Erbse, che nell’indice scrive: “Ψ 361 α1 (test. [error])”. C’è stata una lunga discussione sull’attribuzione del frammento a Cratino o a Cratete, facendo seguito alla collocazione sotto

  Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica Cratete in Kock fr. 50 (CAF I 143).17 K.–A. lo includono come Cratin. fr. 472, ma tra parentesi rimandano a Cratete; sotto Cratete però non ne parlano per niente (cf. K.–A. 1983, 107–109, dove il rinvio al frammento 50 K. manca del tutto). Ne discutono e ne trattano Bonanno 1972 e Olson/Seaberg 2018 nel commento al fr. 472 di Cratino e sembra che sia stata raggiunta una concordia sull’attribuzione a Cratino.18 Si tratta comunque di una testimonianza di carattere linguisticogrammaticale, presumibilmente sulla forma attica di una forma verbale, per la quale ci sono oscillazioni nei mss. Per Eupoli abbiamo in tutto 8 citazioni, 6 con il nome, 2 senza nome (nrr. 7–8), più una congettura respinta (vedi nr. 9). In un caso (nr. 2) è citato assieme a Cratino (vedi Cratino nr. 2). 1. Sch. Iliade 2.333a (sch. ex.), fr. 314 K = 342 K.–A. (incertae fabulae). Sono stati avanzati dubbi anche se il frammento non venga da una tragedia invece che da Eupoli (come dice lo scolio: κατὰ Εὔπολιν), magari da Euripide per erroneo scioglimento di abbreviazione (vedi K.–A. 1986, 493 ad loc.).19 In ogni caso, il senso della citazione non è immediatamente chiaro: forse il frammento è addotto come un parallelo per il contenuto del verso omerico. 2. Sch. Iliade 7.76 (POxy 1087, II 45–47), CGFP 100 = fr. 352 K.–A. (incertae fabulae). Vedi sopra Cratino nr. 2, la citazione di Eupoli ha perduto il titolo della commedia. Si tratta di una testimonianza lessicale, inserita come esempio in una serie di παρώνυμοι. 3. Sch. Iliade 13.353 (sch. ex.), fr. 43 K. = 49 K.–A. (Εὔπολις ἐν Αὐτολύκοις). Per il plurale ἐν Αὐτολύκοις vd. K.–A. 1986, 321 (Αὐτόλυκος α´ β´, test. ii): “de Eupolide fabularum retractatore vid. test. 14 [sc. Sud. δ 756], hinc explicandus videtur num. plur. ἐν Αὐτολύκοις fr. 49”. Si tratta di una testimonianza di carattere linguistico-stilistico, analoga a quella di Cratino nr. 3: non è necessario sottintendere nel passo omerico un participio congiunto al verbo principale (cioè sottintendere ὁρῶν), ἔστι δὲ τὸ τῆς φράσεως ἀρχαῖον; in Cratino nr. 3 si usava l’espressione παλαιὰ συνήθεια. 4. Sch. Iliade 14.241c (Porph. vel sch. ex.?), fr. 435 K. = 472 K.–A. (incertae fabulae). Cratino è citato come testimone di una forma di ottativo, è dunque una testimonianza di carattere linguistico-grammaticale. Nello stesso scolio è citato anche un frammento di Saffo, per confronto dialettale.

 17 Probabilmente il problema risale a Gaisford (E.M. 579,1), che scelse lo scioglimento Κράτης per un’abbreviazione che invece indicava Κρατῖνος, come poi hanno confermato altri mss. degli sch. h all’Iliade (vd. Erbse, test. ad loc.). 18 Bonanno 1972, 169–170; Olson/Seaberg 2018, 305–307; cf. anche Perrone 2019, 17. 19 Cf. Olson 2014, 51 ad loc.

Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica  

5. Sch. Iliade 15.412b1 (sch. ex.), fr. 447 K. = 483 K.–A. (incertae fabulae). Si tratta di una testimonianza lessicale. Sia lo sch. di Aristonico (412a) che lo sch. ex. rilevano l’unicità dell’uso della parola σοφία in questo passo (ὅτι ἅπαξ ἐνταῦθα), usata nel senso di τέχνη e riferita alla τεκτονικὴ τέχνη, cioè l’arte del carpentiere. Non è conservato il nome, ma la formulazione nello sch. ex. rivela che dovrebbe trattarsi di uno degli argomenti addotti dai Chorizontes, perché la parola σοφία non è usata nell’Odissea (Chorizontes fr. *XIX Kohl, collocato fra i Fragmenta probabilia et dubia). Lo sch. ex. annota che ogni τέχνη è chiamata σοφία e che i τεχνίται sono chiamati σοφοί: come prova di questo vengono addotti due passi, uno di Sofocle (fr. 820 N.2 = 906 P. = 906 R.), nel quale il citarodo è detto σοφιστής, e uno di Eupoli, nel quale è il rapsodo a essere definito σοφιστής. In Eupoli fr. 309 K.–A., si trova ὦ ῥαψῳδέ, ma senza specificazioni; Cratino fr. 2 K.–A. testimonia l’uso di σοφιστής per i poeti. 6. Sch. Iliade 16.353b (sch. ex.), fr. 14 K. = 22 (et 13) K.–A. (Αἶγες). Anche questa è una testimonianza lessicale. Lo sch. osserva che il termine μῆλον significa sia capra (πρόβατον) che pecora (αἴξ), tanto è vero che Eupoli chiama προβατικὸν χορόν quello formato di capre: questo permette di attribuire la citazione alla commedia Αἶγες. 7. Sch. Iliade 17.463 (sch. ex.), fr. 94.3 K. = 102.3 K.–A. (Δῆμοι, senza il titolo e senza il nome dell’autore). Si tratta di una testimonianza/parallelo lessicale. Il verso di Eupoli, qui citato senza il nome dell’autore, appartiene a un ben noto frammento dei Demi, di 7 versi in tutto, riportato per intero da uno sch. a Elio Aristide e parzialmente da altre fonti (vedi infra il nr. 8), nel quale si fa allusione all’eloquenza di Pericle.20 In questo sch. omerico ne viene citato un solo verso, senza il nome di Eupoli, evidentemente caduto nella trasmissione. Il verso di Eupoli serve come parallelo lessicale per il fatto che lo scoliasta intende sostenere per l’omerico ᾕρει il significato di κατελάμβανεν, cioè “prendere, uccidere”. Vedi il frammento seguente. 8. Sch. Iliade 24.85a (sch. ex.), fr. 94.7 K. = 102.7 K.–A. (Δῆμοι; ὡς ὁ κωμικός φησι, senza il titolo e senza il nome dell’autore). Il nome di Eupoli si è probabilmente perduto nella tradizione ed egli è designato solo come ὁ κωμικός, epiteto che più spesso viene usato per antonomasia per Aristofane.21 Eupoli è citato come parallelo per il passo omerico, per un’espressione metaforica presente nel passo dei Demi, nel quale a quanto pare il comico sosteneva che, nel suo stile  20 Vd. il commento di Telò 2007, ad loc., e di Olson 2017, ad loc. 21 Il caso per es. di sch. ex. a Iliade 24.617a: Φιλήμων ὁ κωμικός non è un caso di antonomasia, ma sembra piuttosto di distinzione fra omonimi, anche se qui non sappiano se e quale omonimo avesse in mente il commentatore.

  Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica oratorio, Pericle spesso τὸ κέντρον ἐγκατέλειπε τοῖς ἀκροωμένοις. Commentando il passo omerico in cui si presenta Teti che piange la sorte di suo figlio che è vicino alla morte, per ora futura, lo scolio osserva che in questo modo il poeta anticipa il successivo compianto e dolore per Ettore e “lascia il pungiglione negli ascoltatori”, come dice appunto Eupoli (τὸ κέντρον ἐγκατέλειπε τοῖς ἀκροωμένοις). 9. Sch. Iliade 13.289–291 (sch. ex.). Congettura respinta. Nello sch., dopo una citazione di Lisia, c’è una parte di testo corrotta, dove Meineke (1841, 687) indicava “comici vel tragici versus latere”: M. Runkel (ap. Kock 1880, 267, sub fr. 41) ipotizzava Eupoli, ma Kock (ibid.)22 sosteneva “sine idonea causa” e respingeva l’attribuzione; Preller (1848, 522) proponeva Menandro, ma Körte sub fr. 942 “versus mihi potius tragicum quam comicum colorem habere videtur”: cf. Men. fr. *662 K.–A. Il frammento è incluso fra i tragica adespota da Nauck2 sub fr. 450 e accolto fra i Tragica Adespota da Kannicht–Snell (fr. 450). Non abbiamo né tempo né spazio per passare in rassegna puntualmente le citazioni di Aristofane negli scolî all’Iliade. Ci basti dire che abbiamo di lui 27 citazioni: 12 con il nome, 15 (dunque di più) senza nome. Prevalgono di gran lunga quelle contenenti testimonianze di carattere linguistico-grammaticale o lessicale o stilistico, rispetto a quelle nelle quali pare essere addotto un parallelo di contenuto, il che conferma quanto abbiamo visto analizzando le citazioni di Eupoli e Cratino.23 Emerge comunque anche un dato significativo per i testimoni della storia della filologia omerica antica, che non riguarda direttamente il discorso che stiamo facendo. La maggior parte delle testimonianze di questo tipo proviene dai cosiddetti scholia exegetica, piuttosto che da quelli risalenti agli autori del Viermännerkommentar o dagli Sch. D: questo serve certamente a gettare luce sia sull’origine storica che sui dati di contenuto di questa classe di scoli, che ancora devono essere indagati a fondo. Ma questo — come dicevamo — rientra in un altro discorso rispetto al tema di cui ci stiamo occupando. Nella scoliografia omerica le citazioni dei comici tendono dunque, almeno per quanto si è conservato fino a noi, a servire ed essere addotte come parallelo linguistico, lessicale o anche stilistico, il che sembra peraltro del tutto verosimile. Dal punto di vista storico-culturale (che fin dall’inizio abbiamo messo al centro del nostro discorso) si impone un dato significativo e certamente importante. I filologi antichi, tra le numerose altre cose, avevano fatto uno studio  22 Probabilmente il problema risale a Gaisford (E.M. 579,1), che scelse lo scioglimento Κράτης per un’abbreviazione che invece indicava Κρατῖνος, come poi hanno confermato altri mss. degli sch. h all’Iliade (vd. Erbse, test. ad loc.). 23 Così già Perrone 2010, 100–102.

Le tortuose strade del frammento: Citazioni d’autore nell’erudizione antica  

approfondito e accurato anche della lingua dei comici, quanto meno dei comici della archaia: sappiamo bene che studi sul lessico comico risalgono almeno a Licofrone ed Eratostene. Le citazioni degli scolî confermano dunque l’interesse per la lingua dei comici e ne testimoniano la persistenza, probabilmente da un certo punto in poi in connessione anche con l’affermarsi dell’Atticismo. Altre tappe dell’indagine e ulteriori sviluppi della ricerca, nella stessa o in altre simili direzioni, potranno certamente allargare e arricchire i risultati e le conoscenze almeno per quello che abbiamo potuto vedere fino a questo punto. Nello stesso quadro generale rientra poi evidentemente anche Pindaro, sulle cui citazioni abbiamo detto sopra solo poche parole, rinviando ad un lavoro in corso di stampa. E tanti altri autori. Insomma, uno dei risultati importanti di questo tipo di ricerche è di mettere in luce come la filologia di età ellenistico-romana avesse raggiunto una conoscenza della letteratura greca arcaica e classica (e poi anche di età ellenistica) di grande ampiezza, di altissimo livello e di sorprendente profondità: un dato storico-culturale che ancora attende di essere indagato e messo in luce in tutta la sua portata. Bisogna essere ben consapevoli che la posta in gioco di queste ricerche, se considerate nel loro insieme e non come ‘pezzi’ separati (il quadro complessivo è fuori dalla portata di un singolo studioso), non è di significato secondario e limitato, anzi è in grado di far emergere un aspetto di grande importanza per quanto riguarda la filologia nel mondo antico. Si tratta di rendersi conto lucidamente che i filologi antichi, a partire da quelli attivi nella ‘fondativa’ età ellenistica per continuare con quelli di tutto il periodo dell’età romana, acquisirono tenacemente e furono in grado di utilizzare, nelle loro molteplici attività esegetico-erudite, una conoscenza della letteratura greca dei secoli precedenti fino ad Aristotele e poi anche di quella degli autori dell’età ellenistica (i loro ‘contemporanei’, come li ho definiti una volta),24 assolutamente ampia, dettagliata e approfondita, tale da permettere loro di mettere in campo conoscenze vastissime e imponenti per l’esegesi e il commento dei testi di cui si occupavano. L’analisi delle citazioni di autori nella letteratura erudita, inoltre, dimostra ancora una volta come la filologia ellenistica sia figlia di Aristotele e della sua consapevolezza che bisogna studiare e capire il passato, le grandi figure di autori dei secoli precedenti e le loro opere, facendo ogni sforzo per attrezzarsi al meglio possibile per questo compito.

 24 Cf. Montanari 1995b (= questo volume, cap. 62); 2002b (= questo volume, cap. 58).



Part II: Homer

 Commentary on Odyssey 21: P.Fay. 312 (with M. Haslam) P.Fay. 312

5 x 19 cm.

Second century

(P.Cairo Inv. G.H. 10848)

Here is published P.Fay. 312 (descr.) = Pack2 1213: remains of a single column written in a good round and upright hand of medium size showing affinity with the so-called Roman uncial style (on which see Cavallo 1967, 209–220, together with Turner 1987, at no. 13). Bilinearity is strict, only φ (and presumably ψ, not represented) being exempt from vertical confinement, μ has deep bow and curved legs, ω is tri-pronged, ε ϑ o ϲ ω are well-rounded, there is contrast between thick and thin strokes of the pen, there are decorative serifs and finials; but α takes its less stiff form, with apex at lower left, and the cross-bar of ε and ϑ is very high and sometimes ligatured to the next letter; there is no vertical distension. More informal versions of the same type of hand are P.Oxy. XVIII 2161 etc. (Turner 1987, no. 24) and P.Oxy. XLIV 3152; cf. also P.Oxy. XXXVII 2814, XLII 3010. The script may be assigned with some assurance to the second century, perhaps around the middle of that century. Such a hand as this is more usually employed for literary texts than for commentaries. But a less formal hand, whether or not the copyist’s own, has made corrections and filled textual gaps (see on line 18); evidently the exemplar was not easy to read. The commentary is conventional in type, and unsurprisingly fuller than the existing Odyssey scholia, especially exiguous in the later books. (On the other hand, the second century Florence papyrus edited by Bartoletti 1966, 1–4 evidences abridgement of Odyssey scholia already in the first centuries of our era). It was a scholarly enough product, adducing other Odyssean passages and naming authorities. Beyond that little in general can be said: the fragment is too small. There are a few points of contact with the extant exegetic material, but no wholesale affinities can be discerned. For the stretch of Homeric text covered by the papyrus — about fifteen verses — the surviving scholiastic corpus offers little, and still less that is in common with the papyrus. One gloss in the Dscholia (on ἀποέργαϑεν 221) finds a counterpart in the papyrus, another (on ἐὺ γνῶτον 218) does not; a note in schol. Q on 230 προμνηστῖνοι shows little correspondence with the papyrus’ more substantial note on the same word. Of the variant readings recorded in the various scholia, none appear in the papyrus’ lemmata or notes, and there seems to be no significant affinity between our commentator’s Odyssey text and that of P.Ryl. I 53 (Π28). But there is at least one https://doi.org/10.1515/9783110772371-018

  Commentary on Odyssey 21: P.Fay 312 point of interest for the Homeric text: the apparent lemma of 225 seems to show a reading otherwise attested only in cod. Vindob. 133 (see on line 13) — a reminder of the paltriness of our textual as well as our scholiastic evidence for the Odyssey as compared with the Iliad. One Demetrius is named, with regard to the ‘hyperbaton’ at 231 ff. Identification is uncertain, and the context damaged, but there seems to be a significant coincidence here with the treatment of hyperbaton offered by ‘Hermogenes’ π.μεϑ.δειν.: see on line 29 f. Our commentator was doubtless one of the large brood of post-Aristarchean compilers. All the indications are consistent with first-century composition, but we see no means of attaining precision. Just as there are fewer papyrus texts of the Odyssey than of the Iliad, so is scholiastic material on the Odyssey scarcer. Other remnants of Odyssey commentaries (as distinct from scholia minora), in addition to the Florence papyrus mentioned above, are P. Yale Inv. 551 (Hellenica 28 [1975] 60–65; cf. Würzb. Jahrb. N.F. 2 [1976] 99–104), P. Alex. Inv. 198 (Papiri letterari greci, no. 8, Pack2 2614), and an Oxyrhynchus papyrus published as P.Oxy. LIII 3710; cf. also P.Oxy. XXXIX 2888, and P. Med. Inv. 210 (Aeg. 58 [1978] 110–114); all of similar date.1 The transcription has been made from a photograph, kindly supplied by the International Photographic Archive of the Fondation Égyptologique Reine Élisabeth, courtesy of Prof. J. Bingen. The papyrus itself is held by the Egyptian Museum in Cairo; we are most grateful to the Director of the Museum, Dr. M. Saleh, for permission to publish. There is no information whether or not the back is blank. Fr. 1

5

]ϲ̣τοτ[ ]λ̣ ειτε̣ τ̣ α[ ]ϑαιφη̣ϲ̣[ ]ο̣ μενευ̣[ ]ουcκε̣ ̣ [ ] ̣ ιλοιτιουβουκ̣[ ]δεϲ μεγαλωϲ ̣ [ ο

=

] ̣α̣ϑενυληϲαπε [

 1 A photographic reproduction of the papyrus in the ed. princeps: F. Montanari/M.W. Haslam, Commentary on Odyssey 21 (P. Fayum 312), “Bulletin American Society Papyrologists” 20, 1983 (1985), 113–122.

Commentary on Odyssey 21: P.Fay 312  

10

15

] ̣ε̣υτε̣ φραϲϲα̣ [ ]τ̣ ε̣ ϲ̣ε̣π̣εγνω̣[ ] ̣ ̣ ̣ ̣ ̣ε̣το[ ]ν̣ α̣ ̣ ̣ ̣αζο[ ] ̣ ε̣ρ̣α̣ϲ̣ ̣ ̣ ̣ [ ] ̣ ϰεφαλαϲϰ ̣ [ ]μ̣οιωϲτουϲ[ ]παζεταιδια̣[ ]ϲιϰαιειϲω [ ι

]ϲ τωομενα̣[ 20

25

]εαϲαμφι ̣ [ ] ̣τεινοιειϲ̣ελϑ̣[ ]ε̣ ν ινευλογωϲτο[ ]εϲ επιτριω̣ντε̣ ϑ̣ [ ]τει ̣ οιδεϰαϑεν[ ]τοϲειϲελϑηαν[ ]ϲομοιωϲτωαι̇ π̇ρο̣ [ ] ̣ιεϲπηιϲαν̲̅π̲̅ρωτοϲ[ δυμμεϲ

30

] υ ̣ μων μετα μει̣ [ ]μ̣ατετυχϑωτιτο[ ] ι δημητριοϲ --[ ] ̣ ϰοϲ υπερβατον[ ] ̣ ιϲτ̣ [ ̣ ̣] ̣ιδαυτο̣ [ ]τ̣ ι̣ [ _____

2 ]λ̣, or α ε̣ , o not excluded? 5 ̣[, speck on line, λ, α, δ? 6 ] ̣, arc suitable for φ 7 ̣[, traces on twisted fibres 8 supralinear o by m.2 9 Beneath the supralineation, a heavy oblique rising from left to letter-top level 10–14 Obscured by dirt or stain 11 ] ̣ ̣ , ηε, τε? 13 ]., χ suggested ̣ ̣ ̣ [, see comm. 14 ] ̣, speck on line ̣[, α or λ 18 τω and `ι´ by m.2? 19 ̣[, ε or ϑ 23 ̣ , two feet on line, e.g. ν, π, ϑε by m.2, ε occupying space for two letters 25 Above ιπρ, illegible supralineation; if δε, another trace follows? 26 ηιϲ by m.2 27 ] ̣ , see comm. Perhaps μετα as well as δυμμεc by m.2 30 ] ̣ , top of apparent upright τον, o diminished 31 ] ̣ prim., traces at upper right suggesting ε or ϑ ] ̣ alt., upper arc, ε,ϑ,o,ϲ

  Commentary on Odyssey 21: P.Fay 312 Fr. 1 5

10

15

20

25

30

Fr. 1 1–6 3 4 5

] Φ̣ιλοιτίου βουκ̣ [ολ ἀριφρα]δές· μεγάλως ̣ [ ἀποέρ]γαϑεν οὐλῆς· ἀπε [ ] ̣εὖ τʹ έφράσσα[ντο ]τ̣ε̣ς̣ έ̣π̣έγνω̣[ ] ̣ ̣ ̣ ̣ ̣ε̣το[ ϰαὶ ϰύνεο]ν̣ ἀγ̣α̣π̣αζό[μενοι ]χ̣έ̣ρ̣α̣ς̣ ̣ ̣ ̣[ ] ̣ ϰεφαλὰς ϰα̣[ ὁ]μοίως τους[ ]πάζεται δια̣[ εἴπη]σι ϰαὶ εἴσω·[ ὸμοίω]ς τῷ “οἱ μὲν ἀ[ποστρέψαντες ἔβαν ν]έας ἀμφιε[λίσσας” (3.162). ἀλλὰ προμνη]στεῖνοι ἐ{ι}σέλϑ[ετε· ]εν ἵν᾽ εὐλόγως το[ ]ε̣ ς ἐπὶ τριῶν τε̣ θ̣ [ ]τειν̣ οι δὲ ϰαϑ᾽ ἕν[α ]τος εἰσέλϑη̣ αν[ 230 ]ς ὁμοίως τῷ “αἱ `δ̣ὲ̣´ προμνηστεῖν]α̣ι ἐ{σ}πήισαν” (11.233). πρῶτος έγώ ] ̣ὑμῶν μετὰ δ᾽ ὔμμες· ἄταρ τόδε σῆ]μα τετύχϑω· τί τὸ σῆμα ]ι Δημήτριος ] ̣ ϰος ὑπερβατὸν ] ̣ ιϲτ̣ [ ̣ ̣] ̣ιδ᾽ αὐτο̣ [ ] τ̣ ι̣ [ _____

217 221 222

224 225?

229

230

231 231

No lemma can be identified, but the context is evidently Odysseus’ speech to Eumaeus and Philoetius, 21.207–220. Perhaps -σ]ϑαι φησ[ίν (sc. Homer, Odysseus, a commentator?). ὁ μὲν Εὔ[μαιος one possibility among many. ϰελ̣ [ευ- possible.

Commentary on Odyssey 21: P.Fay 312  

7

8 9

9–10

11 12

13

μεγάλως φ̣[ανερόν probable for the gloss. Ap. Soph. 44.8 ἀριφραδέα· μεγάλως φανερά, cf. Hesych. 7260 ἀριφραδές· ἄγαν φανερόν. Otherwise δ̣[ῆλον (Hesych. 7259 ἀριφραδέα· δῆλα) ν. sim. There is no indication that v. 219f., absent from Π28 (P. Ryl. 53, 3rd–4th cent.), Mon. (Ludwich’s U) and U6, were known to the commentator. ἀπε[ϰάλυψε? ἀφεῖλεν, ἀπεϰάλυψεν D-schol.; cf. also D-schol. in Il. 21.599 ἀποέργαϑεν· ἀπεῖργεν, ἀπεχώριζε. For the superior dashes see at 26. εἰσιδέτην as the beginning of the lemma on v. 222 would give a linebeginning in alignment with 7 and 8. But we cannot explain the supralineation: apparently not a letter; conceivably a sign of separation before the new lemma, in which case the first part of 9 may have been occupied by a continuation of the gloss of line 8. E.g. εὖ τ’ ἐφράσσα[ντο ἕϰα|στα· ἰδόν]τες ἐπέγνω[ν πάντα, if our commentator’s text had ἕϰαστα with the vulgate rather than ἄναϰτα as the a f j families and schol. Marc. 613. (Π28 is in lacuna.) For ἐπέγνων cf. e.g. D-schol. in 19.391 ἀμφράσσαιτο· ἐπιγνῶ. Obscured by dirt or stain, which extends from line 10 to line 13. ἑκ̣α̣σ̣το[ may be possible. Apparently ϰαὶ ϰύνεο]ν ἀγαπ̣αζό[μενοι, v. 224 lemma. If the neighboring letter identifications are correct there seems to be not enough room for the π, but that must have been the intention, whatever was written. If χέρας is to be recognized, as we think it must, a variant may be indicated: ϰεφαλήν τε χέρας τε for the end of v. 224? The first letter after χ̣έ̣ρ̣α̣ς̣ is obscured by dirt or stain, then there is apparent abrasion before the papyrus breaks off: τ̣ ε̣ , while not particularly suggested, does not seem to be excluded; if so, perhaps τε —̣, cf. 29. The received text is ϰεφαλήν τε ϰαὶ ὤμους (= 17.35 and 22.499), but Ludwich reports two variants: χεῖρας τε as a γρ. variant for ϰεφαλήν τε in cod. Ven. Marc. 613 (Ludwich’s M, Allen’s U5), and χέράς τε for ϰαὶ ὤμους in textu in cod. Vindob. 133 (Ludwich’s X, Allen’s V4). Allen’s apparatus is discrepant with this: as the γρ. variant in cod. Ven. Marc. 613 he reports τε χέρας τε ϰαὶ ὤμους, assigning this reading to several other mss. too; he has no mention of either of the readings reported by Ludwich. The discrepancy is eliminated if τε χέρας (which is unintelligible) was a mistake on Allen’s part for χεῖρας. But the reading that concerns us seems to be that reported by Ludwich for cod. Vindob. 133. (It may be suspected, moreover, that it is this variant that underlies the other: χέρας τε originally a v.l. of ϰαὶ ὤμους, mistaken for a v.l. of ϰεφαλήν τε, hence χεῖράς τε.) Π28 apparently had the vulgate reading here (ωμο]υς is all

  Commentary on Odyssey 21: P.Fay 312

15–16

17 18

20–26

that survives). χέρας not in Homer, but cf. χερί, χέρα; for the line-end pattern ϰεφαλήν τε χέρας τε, cf. χεῖράς τε πόδας τε. With ϰεφαλήν τε χέρας τε Odysseus’ greeting reciprocates his servants’ exactly, cf. next n. 12 f. then will run ϰαὶ ϰύνεο]ν ἀγαπαζό[μενοι ϰε|φαλήν τε] χέρας τε (224). And that was followed directly, it seems, by quotation of v. 225: [ὡς δ᾽ αὔτως | Ὀδυσεὺ]ς ϰεφαλὰς ϰα[ὶ χεῖρας ἔ|ϰυσσεν (225)· ὁ]μοίως ϰτλ. That at least is suggested by the remains of 14, and would fit the space. ὁ]μοίως in paraphrase of αὔτως? Cf. e.g. D-schol. in Il. 3.339, ΣΤ Il. 4.17, Et. Mag. 172.36, Apion s.v. αὕτως. “He greets his servants (τοὺς [δούλους or [οἰϰέτας, or perhaps τοὺς [ἑαυτοῦ | δούλους ἀσ]πάζεται) in the same way (sc. as they greet him)”? διὰ [ϰολαϰείας e.g., cf. Eustath. 1907.10 (ϰολαϰιϰώτερον), 1908.2 (ϰολαϰεύων), 1907.40, 42; but διὰ [φιλίας looks better for the space. The lemma seems to have started at ἀτὰρ εἴπη]σι. ὁμοίω]ς τῷ probable, cf. 25. τω is irregularly written and may be a later addition to a c. 3-letter gap originally left between ς and ομεν (sic a.c.). Perhaps the exemplar was damaged or otherwise illegible at this point (cf. P. Oxy. XLIV 3151.2.6); similarly at 23, 26, 27 below. Od. 3.162 is apparently adduced. Why? It exhibits nothing obviously comparable with 21.229. The only suggestion we can make is that μεγάροιο, governed by ἐξελϑών, is held to be understood with εἴσω: as in 3.162 νέας may be understood with one or other of ἀποστρέψαντες and ἔβαν. But unless a lacuna is to be postulated the comment must have been highly elliptical: [ἀπὸ ϰοι|νοῦ˙ ὁμοίω]ς ϰτλ? εἰσέλϑ[ετε will have ended the line. In 21 ]ε̣ν is fairly secure; τοῦτʹ ἐϰέλευσ]εν would bring the beginning into alignment with the previous line, but if that was in ecthesis something shorter will be called for, e.g. οὕτως εἶπ]εν or ἔλεγ]εν. τὸ, unless τὸ[ν v. sim. is postulated, will have ended the line. In 22 μηδ᾽ ἅμα πάντ]ες, the end of v. 230, is a possibility: ἵν᾽ | εὐλόγως τὸ “μηδ᾽ ἅμα πάντες” ἐπὶ τριῶν τεϑ[ῇ? In 23 ]τειν̣οι strongly suggests προμνησ]τεῖνοι: e.g. τὸ ‘προμνησ]τεῖνοι’ δὲ ϰαϑ᾽ ἕν[α, | ἵν᾽ εἷς ἕϰασ]τος εἰσέλϑη̣, ἀν[τὶ | τοῦ ‘ἐφεξῆς,’ ϰτλ. But is the position of δέ acceptable? Montanari adduces schol. B (Porph.) in Il. 20.67, which contains τὰ πάντα δέ, τὸν ἀέρα δέ, and τῷ λόγῳ δέ. ἕϰασ]τος εἰσέλϑῃ, ἀν[τὶ | τοῦ ‘ἐφεξῆς,’ ϰτλ. But the position of δέ is against this. Another line of approach would be -]τειν· οἱ δὲ ϰτλ, or, retaining προμνησ]τεῖνοι, δὲ. Nor can the other elements of the reconstruction be regarded as se-

Commentary on Odyssey 21: P.Fay 312  

26f.

27 28f. 29f.

cure, ϰαϑ᾽ ἕν[α| seems safe enough (cf. Ap. Soph. and Et. Mag., citt. infra) but αν[ after εἰσέλϑῃ could be ἄν. προμνηστῖνος occurs only here and in the Nekuia, Od. 11.233, adduced by our commentator at 25f. προμνηστῖνοι is glossed by schol. Q here as εἷς παρ᾽ εἷς, τὸ λεγόμενον, ἄλλος πρὸ ἄλλου, ἀλλεπάλληλοι, ὅ ἐστι παραμένοντες ἀλλήλους, cf. Et. Mag. ad loc., ἀντὶ τοῦ παραδεχόμενοι ἀλλήλους ϰαϑ’ ἕνα (plus etymology). Eustathius 1907.16–22 gives the lemma as ἀλλὰ προμνηστῖνοι εἰσέλϑετε (sic, as pap.), and continues, ὅ ἑρμηνεύων φησὶ “μὴ δ᾽ ἅμα πάντες,” ϰαὶ ἔτι μᾶλλον σαφέστερον “πρῶτος ἐγώ, μετὰ δ᾽ ὔμμες,” ἤγουν ‘μετὰ δὲ ἐμὲ ὑμεῖς·’ ὅ ϰαὶ γίνεται. εἰσὶ δὲ προμνηστῖνοι, τουτέστιν ὡς οἷον εἰπεῖν προμενετῖνοι, οἱ προμένοντες ἀλλήλους ϰαὶ μὴ ἅμα πάντες προϊόντες ϰατὰ τὸν ποιητὴν ἀλλὰ πρῶτος οὗτος εἶτα ἄλλος ϰαὶ ἐφεξῆς. Cf. D-schol., schol. BQ and Eust. in Od. 11.233, Ap. Soph. 135.23 and Hesych. 3581–2 in προμνηστῖναι. The short horizontal bars added above and below the left side of the π of πρωτος are perhaps intended to articulate the text: a gap should have been left by the scribe. Cf. 8. Presumably πρῶτος | [ἐγώ, v. 231 init. Then a gloss, πρότερος (or πρότερον) ὑμῶν? Before ὑμῶν in the papyrus a couple of traces survive at letter-top level; the first is a curve very appropriate for ϲ, but then the latter, an apparent stop, is unwanted: casual? The correction, and apparently μετα too, is by the second hand. τί τὸ | [σῆμα λέγε]ι; or τοῦτ]ί; v. sim. Δημήτριος | [ὁ γραμματ]ιϰός? Of the twenty eminent Demetrii listed by Diogenes Laertius (5.83–85) three are grammatici: one from Adramyttium, nicknamed Ἰξίων; one from Cyrene, nicknamed Στάμνος; and one from Erythrae (PW 101, 104, 105). To these may be added the grammatici Demetrius of Gonypeus, Demetrius ὁ πύϰτης, and Demetrius of Troezen (PW 102, 103, 106); and there were no doubt others. Our best candidate might be thought to be Ixion, frequently cited in the Homeric A-scholia. (Testimonia are collected by Staesche 1883.) Demetrius of Scepsis and Demetrius of Gonypeus, the only other Demetrii mentioned in our scholia, apart from an isolated reference to Demetrius of Phalerum, barely come into question. That Ixion should be termed Δημ. ὁ γραμματιϰός is a bit odd, however (in the scholia he is referred to sometimes simply as Δημήτριος, more often as ὁ Ἰξίων), and the possibility of some other identification remains open. One such might be the author of the extant On Style, though that work, while it does deal with figures (σχήματα, cf. next note), does not treat of hyperbaton.

  Commentary on Odyssey 21: P.Fay 312

31

Fr. 2

References to ὑπερβατόν in the Iliad scholia are given by Erbse, 1969–1988, VI, Index III s.v.; cf. Baar 1952, 4, 17, 93 ff. A partial treatment of scholiastic references to ὑπερβατόν is offered by Degenhardt 1909, 37 f., who bases himself on ‘Tryphon’ π. τρόπων, where hyperbaton is defined as λέξις μεταϰεϰινημένη ἀπὸ τῆς ἰδίας τάξεως (iii 197.20– 22 Spengel). But more relevant to the present instance would seem to be definitions of hyperbaton as a schema, such as that offered by Anon. π. σχημ. iii 188.5–7 Sp., ὑπερβατὸν δέ, ὅταν τῆς συνεχείας τοῦ νοήματος διαϰοπείσης ὑπὸ μεταξυλογίας ἐμβληϑείσης ἕτερον (leg. ὕστερον?, cf. Zonaeus π. σχημ. iii 170.14–16 Sp.) ἀποδοϑῇ τὸ λεῖπον: for we take it that here the term is being used precisely in the sense defined and discussed by ‘Hermogenes’ π.μεϑ.δειν. 14 (p. 430. 1–4 Rabe): γίνεται δὲ ὅταν τὴν αἰτίαν τοῦ λεγομένου, ἣν μέλλει ποϑεῖν ὁ ἀϰροατής, μέσην ὁ λέγων τάξῃ (‘the speaker interposes the reason for his statement, which the hearer is going to desiderate’); hyperbaton is thus a σαφηνείας ὄργανον. (An unorthodox standpoint, this: cf. Auct. ad Her. 4.44, [Longin.] de subl. 22.1–4, Quint. 8.2.14 f., 8.6.65, Theon ii 82.19–23 Sp.) Cited in exemplification is Od. 10.190–192, ὦ φίλοι, οὐ γάρ τ᾽ ἴδμεν ...· ἀλλὰ φραζώμεϑα ϑᾶσσον. Precisely so here: 232f., ἄλλοι μὲν γὰρ ϰτλ, is the interposed explanation, while 234 f., ἀλλὰ ϰτλ, resumes what is begun with τόδε σῆμα τετύχϑω. (The γάρ clause is not strictly a parenthesis, because of the manner of resumption; cf. Hermog. π.ἰδ. p. 305f. Rabe.) Just as in the Od. 10 example διὰ μέσου τὰς αἰτίας τοῦ φράζεσϑαι ϑᾶσσον ἔταξε (sc. Odysseus; p. 430.10 Rabe), so here Odysseus interposes the reason why the bow must be handed to him by Eumaeus. This seems to us a remarkable conjunction. For general discussion of hyperbaton see Volkmann 1872, 436 f., Lausberg 1973, §§ 716–718, § 860, Martin 1974, 265 f., 308 f. Perhaps ọἱ δ᾽ αὐτό (sc. τὸ σῆμα); otherwise e.g. ϰα]θ̣-, με]θ̣ίστ[η]σι.

______ ]ρ̣η ̣ αλ ̣ [ ] ̣ ̣ρωναναδωμα[

Commentary on Odyssey 21: P.Fay 312  

Fr. 2 Column foot, presumably the same column as fr. 1, since it gives us v. 233 f.: φαρέτ]ρην̣ . ἀλλ[ὰ σὺ δῖ᾽ Εὔμαιε] φ̣έ̣ρων ἀνὰ δώμα[τα. The two pieces do not appear to join (they do not combine to give φαρε]τρην), and without first-hand examination nothing more can be ventured as to their physical relationship.

 Glossary to Iliad 2.201—18: P.Oxy. 3832 No inv. no.

9–5 × 15 cm.

Second century

Written on the back of the Euphorion published as XXX 2525, and upside down in relation to it; a brief description is given there (pp. 66 f.). There are parts of two columns, with a surviving upper margin of 2 cm but broken off at the foot; of the second column only some initial letters remain. Lobel assigned the hand, ‘a coarse medium-sized uncial’ with cursive tendencies (note i 8, beta written as an open-topped square), to the second century. There are no lectional signs. For a list of similar texts, see Raffaelli 1984, where 3832 is noted on p. 150 as no. 025. The following conventions are used in the commentary: – Simple introductory formulae, like ἀντὶ τοῦ, τουτέϲτιν etc, are usually disregarded in indicating coincidences, similarities and differences. – = indicates complete coincidence with the papyrus, both in the lemma and in the gloss. – + (placed after a siglum) indicates that this source offers more material under this lemma than the papyrus. – = [ ] means that the lexical choice of gloss is the same, but lemma and the gloss are in a different form (i.e. nouns and adjectives show differences of case, gender or number, verbs differences of mood, tense or person). – cf. refers to similarities, of various import and interest, with the entry in the papyrus. – ≠ notes, whenever relevant, a different explanation for the same lemma, ≠ [ ] for the same lemma in a different form. col. i

5

10

top απτολεμοϲ ] απολεμοϲ αναλκιϲ ] αϲθενῃϲ εναριθμιοϲ ] ϲυνκατηρι̣ θμη ]μενοϲ πολυκοιρα]νιη πολυα̣ρ̣χ̣ι̣[α αγκυλομητ]εω ητοι ϲκολ̣ [ιο]βου λου η περ] ι̣ των αγκυλων π̣ρα γματω]ν̣ αριϲτ̣ α̣ βουλευο̣ μενου ] ηχει εϲτιν δε ϊ̣διω ] μ̣α ϕωνηϲ ] [̣ ̣ ̣ ̣ ̣ ̣ ]̣

https://doi.org/10.1515/9783110772371-019

col. ii 2. 201 202 204 205

210

.[ .[ .[ [ .[ .[ [ μ[ ιδ[ θ[ ζ[

5

10

Glossary to Iliad 2.201–18: P.Oxy. 3832  

15

20

col. i

αμετ ροε]πηϲ ] εκολωα ] επεα ] ακοϲμα ] κατα κοϲμ]ο̣ ν ] ειϲαιτο ] αιϲχιϲτοϲ ] ϕολκοϲ ] ϲυνοχωκο]τε

[ ̣ ̣ ̣ ̣ ̣ ̣ ]̣ εν ̣ ̣ [̣ ̣ ̣ ̣ ̣ ̣ ̣ ̣] ̣ ̣ ν̣ ω̣ν̣ εθορ̣[υβει] λογο[υϲ ] αδια[τακτα κ̣[ατ]α̣ τ̣ρ̣ο̣[πον κα τα το κ̣αθηκο[ν ϕανειη [ εχθροτ[ατοϲ ϲτ`ρʹαβοϲ [ ϲυνπεπ̣[τωκοτε(ϲ)

.....

212

213 214

α[ .[ .[ .[ .[ .[ .[

15

215 216 217 218

1 ≠ D; ≠ Lex. Hom. α 697; ≠ Hesych. α 6877; ≠ Par. A; ≠ Par. Bekk. 2 = D + ; = Par. Bekk.; = Su. α 1941 + ; = E.M. 97.25 + ; = Par. A; = [Lex. Hom. α 482]; = [Hesych. α 4415]; ≠ Lex. Hom. α 465; cf. Hesych. α 4413–4 (ἀναλκέϲ, ἀναλκήϲ), Phot. α 1541, PsZon. 189. 3–4 καταριθμούμενοϲ D, Par. A, Par. Bekk.; καταριθμηθείϲ Hesych. ε 2671. 5 = D; = Lex. Hom. (mss); = Eust. 199.33, 202.28, 203.4, 203.8; = Hesych. π 2873; = E. Gud. 475.4 Stu.; = Ps-Zon. 1565 + ; = Par. A; = Par. Bekk.; = Par. Mosch.; = Par. Gaz. 6–8 τοῦ ϲκολιοβούλου Par. Bekk.; ἀγκυλομῆται· ϲκολιόβουλοι Σa ed. Boysen p. viii a 15, Σb An. Ba. I p. 12. 17, Phot. α 188, Su. α 253 (see Paus. att. α *16); ... ϲκολιόβουλοϲ. οἱ δὲ τὸν δυνάμενον περὶ τῶν ἀγκύλων καὶ ϲκολιῶν εὖ βουλεύεϲθαι Ε. Μ. 11. 14; ἤτοι ἀγκύλα καὶ ϲκολιὰ βουλευϲάμενοϲ ... ἢ ὁ τὰ ἀγκύλα καὶ δυϲχερῆ πράγματα τῇ μήτι περιλαμβάνων D (cf. also D on Il. 4. 59); τοῦ ἀγκύλα καὶ ϲκολιὰ βουλευομένου Lex. Hom. α 39; ... τοῦ δυναμένου περὶ τῶν ἀγκύλων καὶ ϲκολιῶν εὗ βουλεύεϲθαι Ap. S. 4.24 (α 40); τὸν ἐπὶ τῶν ἀγκύλων καὶ ϲκολιῶν πραγμάτων ὀρθῶϲ βουλευόμενον Ε. Gud. 13.4 De Ste., Εp. Hom. AO 9.7; ὁ ἀγκύλα καὶ ϲκολιὰ βουλευϲάμενοϲ Ε. Gud. 13.19 De Ste.; ϲκολιὰ βουλευομένου Hesych. α 569, Ps-Zon. 16, 22. 9–10 ηχει corrected from ηχι? Lemma βρέμεται or ϲμαραγεῖ: ἠχεῖ (for either) = D + ; = Par. A; = Par. Bekk.; = Hesych. β 1090 + , [β 1088 + ], ϲ 1224 + ; = Lex. Hom. β 100 + and s.v. ϲμαραγεῖ mss; for βρέμεται sch. D (T11) on Il. 2.210b + (p. 227.32 Erbse); cf. βρέμει· ἠχεῖ Et. Sym. p. 134.8 Berger +, Ε.M. 212.19 + ; βρέμει· ταράϲϲει, ἠχεῖν ποιεῖ Σb An. Ba. I p. 181. 30, Phot. β 269, Su. β 531; cf. ϲμαραγεῖν ἠχεῖν Ps.-Zon. 1660 + ; ≠ Ε.M. 720.51 s.v. ϲμαραγεῖ, E. Gud. 506. 5 ff. Stu., Ep. Hom. AO 385.16 ff.; βρέμεται: ἦχον ποιεῖ, ϲμαραγεῖ: ἠχεῖ Par. Mosch. Only the lemma ϲμαραγεῖ in PHamb. inv. 736 vso (Raffaelli no. 023) ii 11. 11 .[, a round letter, probably ε or θ. εζοντο] ε̣[καθεζοντο (so Par. Bekk.; cf. ἕζετο· ἐκαθέζετο D on Il. 1.246, Lex. Hom. ε 58, Hesych. ε 603, Σb An. Ba. I p. 207.12; cf. Su. ε 278) is a possibility, but the lacuna seems too short. Other words which might be glossed here are ἐρήτυθεν (211), καθ᾿ ἕδραϲ (211) or Θερϲίτηϲ (212). 12–13 The explanation of ἀμετροεπήϲ occupied two lines. In 13, εν...[, the first is perhaps η or ει or even τ (if the right-hand side of the trace is just ink running down a fibre), the second a round letter, perhaps ο, the third parts of an upright on the edge; at the end ] ̣α̣ν̣ω̣ν̣ is likely but not certain. The gloss in D, ἄμετροϲ ἐv τῷ λέγειν, ϕλύαροϲ (so also Hesych. α 3619; cf. ἄμετροϲ ἐv τῷ λέγειν Lex. Hom. α 378, Par. Bekk.; τὸν ἐv τῷ λέγειν ἄτακτοv Eust. 205.3) sug-

  Glossary to Iliad 2.201–18: P.Oxy. 3832 gests possibilities, but, supposing that we can restore [αμετροϲ] in 12 (a little long for the lacuna), we still have the difficulty that in 13 the space is too great for ἐv τῷ λέγειν· and the traces at the line-end do not suit it. In itself, ] ̣α̣ν̣ω̣ν̣ suggests a participle, but the first trace goes against most of the obvious verbs (seemingly not β γ δ θ π τ χ). One possibility (suggested by Dr Rea) might be εν τ̣ ο̣ ι̣ [ϲ λογοιϲ λ]ε̣ ι̣ α̣ν̣ ω̣ν̣ , where λίαν would qualify an adjective in 12; but the supplement would again be a little long, and the hyperbaton odd. 14 = D; = sch. Il. 2.212–6 (Nic.); = Ap.S. 65.24 + ; = Eust. 205.6; = Hesych. ε 1551 + ; = Σb An. Ba. I p. 213.32; = Su. ε 526; = Ε.M. 323.19 + ; = Par. A; = Par. Bekk.; = Par. Mosch.; = Par. Gaz.; cf. E.Gud. 445.13 De Ste.; cf. κολω̣όν· θόρυβον D on Il. 1.575 + , al.; PColon. 2281 (Raffaelli no. 016) v 20: κολω̣όν· θ̣όρυβον, E.Gud. 334.46 Stu., Ep.Hom. AO 135.3. 15 = D on Il. 2.109; = Lex. Hom. ε 556; = Hesych. ε 4287; = Ps.-Zon. 808 + ; = Par. Mosch.; = Par. Gaz.; = [D on Il. 1.77, 108, 150, 201; 2.7 etc]; = [Hesych. ε 4285] + ; = [Su. ε 2816] + ; = [Ε.M. 367.55) + ; = [Σb An. Ba. I p. 234.11 ] + ; PStrasb. 33 (Raffaelli no. 011) iv 22 ἔ̣π̣ε̣α̣ πτερ̣ό̣ε̣ν̣τα· τ̣α̣χεῖϲ̣ [λ]ό̣γουϲ cf. Ε.M. 335.2, 383. 50; Hesych. ε 4286; ≠ Par. Bekk. (ῥήματα). 16 ἄκοϲμα· ἀδιάτακτα, ἀπρεπῆ Lex. Hom. in ms Ο; ἄκοϲμα· ἀπρεπῆ, ἄτακτα D in ms C (not in D and Lex. Hom. as published), Hesych. α 2501, Σb An. Ba. I p. 59.3, Phot. α 798, Su. α 933, Ε.M. 51.22 + ; ἀπρεπῆ Par. Bekk.; ἄτακτα sch. Il. 2.212-6 (Nic.); ἀτάκτουϲ (λόγουϲ) Par. Mosch., Par. Gar. 17–18 = Hesych. κ 1180 + ; κατὰ τὸ πρέπον D + , Par. Bekk.: but a reading π̣ρ[ does not suit the traces; κατὰ τὸ προσῆκον D on Il. 5.759, 17.205 + . 19 = sch. ex. Il. 2.215b + ; ἐϕαίνετο D, Par. A; = [Lex. Hom. ε 183, ε 94]; = [Apio Ludw. s.v. εἵϲαϲθαι] + ; = [Ap. S. 63.24], cf. 62.25; = [Ep. Hom. AO 135.7] + ; = [Su. ει 235] + ; = [E.Gud. 433.10 De Ste.] + ; = [Ε.M. 306.36] + ; cf. sch. Il. 2.212–6 (Nic.), Eust. 205.18, 729.57, Ε.M. 296.10 ff.; εἴδεται· ϕαίνεται PStrasb 33 (Raffaelli no. 011) v 13, D on Il. 8.559, Lex. Hom. ε 94, Ap.S. 63.24, Eust. 729.57; εἴϲαιτο· δόξειεν Hesych. ε 1084 + ; εἴϲατο ... ἐϕάνη Ps.-Zon. 647 + . 20 αἰϲχρότατοϲ D, Lex. Hom. α 181, Hesych. α 2148 + , Ε.M. 39.57 + , Par. A, Par. Bekk.; αἰϲχρόϲ E.Gud. 56.11 De Ste. + : is ἐχθρότ[ατοϲ of the papyrus a simple error for αἰϲχρότατοϲ? (Cf. also Il. 2.220 αἰϲχρότατοϲ· ἐχθρότατοϲ D, Lex. Hom. ε 1027, Par. Bekk.) 21 = D + ; = Lex. Hom. (mss); = Ap.S. 164.17 + ; = Ep. Hom. AO 423.27 + ; = Eust. 206.18; = Hesych. ϕ 730 + ; = Σb An. Ba. I p. 407.32; = Su. ϕ 565; = E.Gud. 555.34 Stu. + ; = Ε.M. 798.2 + ; = Ps.-Zon. 1817 + ; = Par. A; ≠ Par. Bekk. The rho of ϲτραβοϲ was added later, with the loop higher than the other letters and a thin tail inserted between τ and α. 22 For orthographic variation between ϲυνοχωκότε and ϲυνοκωχότε see Chantraine, Dict. Etym. 1070b with bibliography, ϲυμπεπτωκότεϲ D + , sch. ex. on Il. 2.218b, Eust. 206. 47 (cf. also 206.44), Ps.-Zon. 1693 + , Par. Bekk.; ἐπιϲυμπεπτωκότεϲ Hesych. ϲ 2675 + ; ϲυμπεπτωκότε Ε.M. 735.46 + , Par. A; (ϲυμπεπακότεϲ [sic] E.Gud. 516.14 Stu.); cf. also Su. ϲ 1604 ϲυνοχωκότε· παρ’ Ὁμήρῳ, διὰ τὸ μέτρον; differently Ps.-Zon. 1694 ϲυνοχωκότα· εἰϲδεδυκότα. col. ii 6 ̣[, α or λ? 14 ̣[, probably α. 15 ̣[, a round letter (ε θ ο).

I have looked for a possible sequence of lemmata which would fit these traces, but found nothing that was not open to substantial objection. Col. i covers 18

Glossary to Iliad 2.201–18: P.Oxy. 3832  

verses of Homer in 22 lines; at that rate, we expect no more than 40 verses to be covered in the lines lost between i 22 and ii 6. μ[ετάφρενον, ἰδ[νώθη, θ[αλερόν, at 265–266 would be very attractive. But the first occurrence of a word beginning with ζ comes only much later and is the not so attractive Ζεύϲ at 324 (the same word also at 371, 375, 412), perhaps already too far from the verses covered by col. i; and the next alternative appears as late as ζώνην in 479.

 A Re-examination of PBerol. 13282: The Homeric Historiae Fabulares on Papyrus PBerol. 13282 is a small fragment of a papyrus codex (5.5 cm high by 7.2 cm long), currently conserved in the Staatliche Museen of East Berlin (DDR) and published for the first time by W. Müller.1 The scrap of papyrus is mutilated on three edges; there is a margin, about 2 cm wide, along the right-hand edge of the side with horizontal fibers and along the left-hand edge of the side with vertical fibers; on each side are conserved, respectively, the final portion and the initial portion of eight lines of literary text. There is only one lectional sign, namely the two dots on the iota in l. 7↓; there are no visible breathings, accents, or other diacritics; the mute iota is not adscripted. On paleographic grounds the papyrus has been dated to the second half of the third century CE. In the editio princeps, the fragment was described as “Kommentar zu Homer, Ilias XX”, with the specification that it seemed to present a commentary with a prevailing interest in sanctuaries and rituals;2 and it has generally been thought (not without some uncertainty, of course) that the extant portion of the commentary covers the Iliadic text from Υ 147 to Υ 404 — in other words, a rather sparse commentary, even if it is true that we are near the end of the poem. However, a closer examination of the text, along with a more apt comparison with similar documents, permits us to ascertain with greater precision the category of “commentary” to which PBerol. 13282 belongs, in such a way that we can even classify it within a specific type; and such an examination will also allow us to make further contributions to the reconstruction of the text itself. The corpus of scholia in which one can find the most fruitful parallels for PBerol. 13282 is that of the Scholia D in Iliadem, for which (as is well known) we lack a modern critical edition; it is therefore still necessary to consult the editio princeps of Janus Lascaris published in Rome in 1517 (and later reprinted), with the  1 “Forsch. Berichte” X (1968), 118 f., which includes a photograph of the side with horizontal fibers. I have been able to check the transcription of the entire text against a high-quality photograph taken by the Photographische Abteilung of the Staatliche Museen zu Berlin (DDR), which I hereby thank for its collaboration. I must also thank my friend J. Rusten for several valuable suggestions. Papyrus codices dated to the third century are common: cf. Turner 1977, 91 ff., esp. 111 for the Homerica. Unfortunately the remains of our fragment are not such that they permit us to reconstruct the dimensions of the pages of the codex. Among the papyrus documents with Homeric ἱστορίαι, PSI 1173 is another papyrus codex of the third century (see the inventory in the Appendix). 2 Cf. p. 119. https://doi.org/10.1515/9783110772371-020

A Re-examination of PBerol 13282  

exception of the excerpta of these scholia which are found in the Ven. A MS of the Iliad and which Dindorf included in his edition of the A-Scholia. In order to have a reliable basis upon which to make a comparison, I myself have established a critical text of that part of the D-Scholia which is relevant to our present purpose — namely (and here I must necessarily anticipate the results of my reexamination of the papyrus) the mythographic ἱστορίαι that pertain to the relevant portion of the Homeric text, i.e., Υ 147–Y 404. For this purpose I have made use of the five principal MSS, along with Lascaris’ editio princeps; the text of the scholia is given here in the Appendix, and it should be referred to during the discussion that follows. Let us begin with the aspects that seem most secure. In ll. 3–4↓ the two verses Υ 403–404 have been identified with certainty (they are the only lemma that is conserved on the document), and this fact permits us to arrive at two conclusions: first, the lemmata on this papyrus are placed in ekthesis; and second, the line-length for this document consists of 38 letters on average.3 It is difficult to determine whether or not there was any sign indicating the separation of lemma and commentary: blank space, or two dots, or other methods are all equally possible. That said, it is clear that, after the lemma, the comment was not begun on the next line but rather followed immediately: as can been seen from the reconstruction of the text, the commentary certainly begins at the end of l. 4. What follows thereafter can be inferred rather easily and indeed can be in large part recovered (hypothetically, of course), thanks to comparison with the ἱστορία on Υ 404 in ll. 5–8 of the scholion.4 In the text of the papyrus document, l. 6 presents a problem that renders it difficult to reconstruct the period as a whole: it seems scarcely possible that the lacuna has enough space for anything more than a statement of the destination of the ἀποικία led by Neleus. In the scholion we find εἰς Μίλητον καὶ τὴν Καρίαν, which is certainly too long; if it were inserted, it would be impossible to explain the following τῆς γῆς; and there are other difficulties at the beginning of l. 6, because if we conjecture (on the basis of the scholion) καὶ τ[ῆ]ς [Ἀχαϊκῆς Ἑλίκης, it becomes difficult to fit within the remaining space both the necessary mention of the place where the colony settled and the required verb that would connect with the beginning of l. 7. It is clear, then, that in this passage the papyrus text was certainly more

 3 Cf. ed. pr., p. 118. 4 The subscriptio that attributes the ἱστορία to Cleitophon has caused difficulties: cf. Müller, FHG, IV, 368; Jacoby 1921; Idem FGrHist 293, Kleitophon von Rhodos (esp. the commentary) and FGrHist 490, Klytos von Milet (esp. the commentary); van der Valk 1963–1964, I, 359–361. For the parallels, cf. Lünstedt 1961, 150.

  A Re-examination of PBerol 13282 concise than that of the ἱστορία found in the scholion. A solution might be found (exempli gratia) by supposing that τῆς γῆς was governed by a verb like ἅπτω5 in participial form (ἁψάμενος, similar to παραγενόμενος in the scholion) and that the first part of the line had only τ[ῆ]ς [Ἑλίκης. This would provide the space to restore a mention of Miletus or Caria as the destination point, for instance as follows: ἐξ]/ Ἀθηνῶν καὶ τ[ῆ]ς [Ἑλίκης. περὶ Μίλητον δὲ ἁψάμενος]/ τῆς γῆς κτλ. All of this naturally remains purely hypothetical. Working backwards from this point, it is natural to suppose that in ll. 1–2 we must look for the end of the previous ἱστορία. It is hard to believe that the remaining portion of l. 2 can be supplemented in any other way than as Πρια]μιδῶν, which is supported by an excellent parallel in the ἱστορία in the scholion on Υ 307 (l. 2): τῆς τῶν Πριαμιδῶν ἀρχῆς. It is also important to note that in the D-Scholia, between Υ 307 and Υ 404 we do not find any other mythographic ἱστορίαι. Both of these facts, in my opinion, support the conclusion that at l. 2↓ the ἱστορία to Υ 307 came to an end; and perhaps we can have some confidence in the supplement proposed for the rest of l. 2 on the basis of the end of the corresponding DScholion (l. 8), since the formula ἡ ἱστορία παρὰ τῷ δεῖνα is not only regular in the scholia but also well attested in the papyri too.6 The ἱστορία found in the scholion states that, since she was aware that the kingdom of the Sons of Priam would pass over to the descendents of Anchises, Aphrodite first coupled with the latter in order to give birth to Aeneas, and then she caused the destruction of Troy by arousing Alexander’s desire for Helen; with this in mind, we can suppose that the papyrus’ account concluded by saying more or less, as a final summing-up, that in this way Aphrodite caused the destruction (e.g. κατέλυσεν: cf. sch. ll. 3, 5) of the kingdom of the Sons of Priam. The beginning of the line remains obscure: before the α there is a vertical stroke that could be an ι or else part of an η and is preceded by very small traces which could agree with τ (Ἀφροδί]/τ̣η̣??); after the λ one can see the lower part of a round stroke connected to what remains of a vertical stroke (the οι in the editio princeps seems difficult; ει may agree better with the visible traces). More than this, at present, it is impossible to say. Let us examine now the remains of the side with horizontal fibers, which turn out to be even more problematic. In his editio princeps, W. Müller sees a probable reference to Υ 147, at which point in the Homeric text there is a brief  5 Cf. Diodorus Siculus IV 48, 7: ὅταν τῆς γῆς ἅψωνται. 6 Cf. esp. PSI 1173, where it is always included, as well as POxy. 3003 I 2, and also PSchubart 21, l. 22 (reproduced below). On the ἱστορία attributed to Acusilaus of Argos, cf. Jacoby, FGrHist 2 F 39 (with commentary); Lünstedt 1961, 149.

A Re-examination of PBerol 13282  

allusion to the monster that was sent by Poseidon to punish Laomedon and was then killed by Herakles; as we will see, this link can now be considered secure. Indeed, in the ἱστορία of the D-Scholia to Υ 145–147, this event is narrated at length: it begins with the service rendered by Poseidon and Apollo to Laomedon, who then refused to pay the agreed-upon wages, as a result of which Poseidon, enraged, sent a monster that caused death and destruction; an oracle informed Laomedon that he could rid himself of the monster by offering to it his daughter Hesione, which he then proceeded to do; but he also promised the immortal horses (a gift of Zeus) to whoever killed the monster and saved Hesione; this task was undertaken by Herakles, who, however, was in turn deceived when Laomedon gave him mortal horses instead of the immortal ones he had promised; Herakles then destroyed Troy and took for himself the horses that were his due. The subscriptio attributes the material found in the scholion to Hellanicus.7 There does not appear to be a close correspondence between the text of the scholion and that found in the papyrus, in which nonetheless one can identify certain key elements of the story: the κῆτος in l. 1 (cf. sch. ll. 7, 10, 11, 15), Laomedon in l. 5, his daughter in l. 6 (Hesione, cf. sch. ll. 9, 11), Herakles in l. 7 (cf. sch. l. 13). One element of the papyrus text that may be of some interest is to be found in the word πλευρά (l. 3): in the scholion (l. 15 f.), we read that Herakles “having penetrated, through its jaws, into the stomach of the monster, sliced through its flanks”, λαγόνας; and this detail prompts one to suppose that in ll. 3–4 the papyrus recounted, precisely, how Herakles sliced through the flanks of the monster (e.g. Ἡρακλῆς δὲ διέφθειρεν αὐτοῦ] τ̣ὰ πλευρά: cf. sch. l. 16) and in this manner (καὶ οὕτως) he prevented the monster from abducting Hesione (ἀναρπάσῃ τὴν/[Ἡσιόνην). This element, indeed, seems decisive: whereas the other sources8 in general simply state that Herakles killed the monster, in the ἱστορία contained in the scholion (deriving from Hellanicus) we find a description of the manner of combat (which will later be taken up by Lycophron, Alex. 33 ff.).9 A privileged connection, then, emerges between the scholion and our papyrus, as a result of which the parallel between λαγόνας in the scholion and πλευρά in the papyrus acquires significance. What remains of the latter document, therefore, ought to form part of the second half of the narrative,

 7 Cf. Jacoby, FGrHist 4 F 26 a-b (with commentary); Ambaglio 1980, fr. 72 a-b, pp. 74 f. and 122 f.; cf. also the scholion to Φ 444 and Erbse ad loc. 8 They can be found in Lünstedt 1961, 148. 9 In Lycophron too one finds the detail whereby Herakles killed the monster by penetrating into its viscera and then slicing it open from the inside: Alex. 35 ἔμπνους δὲ δαιτρὸς ἡπάτων... Cf. Jacoby, commentary on FGrHist 4 F 26.

  A Re-examination of PBerol 13282 more or less beginning at the moment when Herakles kills the monster: we can perhaps glimpse the description of Laomedon who, having received his daughter safely, did not grant to Herakles the promised reward (ll. 5–7); but it is pointless to spend effort in a reconstruction that must remain largely speculative. Perhaps, however, one can legitimately observe that, if the detail in l. 3 does in fact correspond to l. 16 in the scholion, then the papyrus ἱστορία must have appeared in a redaction that was not only different, but perhaps even somewhat richer and more extensive than the (relatively hurried) conclusion which we read in the scholion. I now present the text of the papyrus in its entirety, incorporating the new conclusions at which we have arrived in the above discussion.

1

→

3 5 7

[ [ [ [ [ [ [ [

± 29 ± 22 ± 22 ± 25 ± 20 ± 22 ± 23 ± 27

— – — – — ]τ̣ ο̣ ῦ̣ κή̣τ̣ ο̣ υ̣ς̣ το]ῦ̣ θ̣ ηρίου τὸν θεόν ]τ̣ ὰ πλευρὰ καὶ οὕτως ]τε ἀναρπάσῃ τὴν τὸ]ν Λαομεδόντα καρπω τὴ]ν̣ θυγατέρα μηδου Ἡ]ρ̣ακλεῖ τὸ ἔπαθλο(ν) ἔ]ν̣ θεν δ̣οῦναι — – — – —

3 ex. gr. Ἡρακλῆς δὲ διέφθειρεν αὐτοῦ] τὰ πλευρά, καὶ οὕτως: vd. comm. /[Ἡσιόνην 5 καρπω-/[σάμενον? ed. pr. (cf. Sch. D r. 8?) 6 μὴ δοῦ-/[ναι?

↓ 1 3 5 7

ad Υ 147

4 μή]τε?

– — — – — ..απλ..[ ± 21 τὴν τῶν Πρια] μιδῶν ἀρχήν. ἡ̣ [ἱστορία παρὰ Ἀκουσιλάῳ (?) ] αὐτὰρ ὁ θυμὸν ἄισθ[ε καὶ ἤρυγεν, ὡς ὅτε ταῦρος ἤρυ] γεν ἑλκόμενος Ἑ̣ λ̣ ικών̣ [ιον ἀμφὶ ἄνακτα () Νηλεὺς ὁ] Κ̣ ό{ν}δρου μα[ν]τεία[ν λαβὼν ἀποικίαν ἔστειλεν ἐξ] Ἀθηνῶν καὶ τ[ ̣ ]ς[ ± 26 ] τῆς γῆς ἱερὸν Π̣[οσειδῶνος ἱδρύσατο καὶ ἀπὸ τοῦ] ἐν τῇ Ἀχαϊκ̣[ῇ Ἑλίκῃ τεμένους Ἑλικώνιον προση]/[γόρευσε… – — — – —

4–5 τὴν

ad Υ 307 ad Υ 403–4

1 τ̣η̣ απλοι̣[ ed. pr. ex. gr. (Ἀφροδίτη) κατέλυσε τ. τ. Π. ἀ. 2 cf. Sch. D ad Υ 307, r. 8 4–8 suppl. coll. Sch. D ad Υ 404, rr. 5–8 6–7 ex. gr. καὶ τ[ῆ]ς [Ἑλίκης. περὶ Μίλητον δὲ ἁψάμενος]/ τῆς γῆς κτλ., cf. Sch. D ad Υ 404, rr. 6–7.

A Re-examination of PBerol 13282  

Overall, then, the fragment restored by PBerol. 13282 appears to form part of a collection of mythographic ἱστορίαι on Υ 147, 307, and 404. In the corresponding section of the D-Scholia we find another ἱστορία, on Υ 215 (see the Appendix), which in the papyrus codex was perhaps omitted (unless it too was originally to be found in the lacuna that has swallowed up the final part of Υ 147 and almost all of Υ 307); in the scholion, this ἱστορία has a subscriptio that attributes it to Lycophron.10 In any case, the principal element for establishing the relationship between the papyrus document and the D-Scholia is to be found in the characteristics of the ἱστορίαι. The text of the ἱστορία on Υ 403–404 can be restored, to a large extent, thanks to comparison with the corresponding scholion; as a result it seems legitimate to assume that, although it may have differed in the parts that have been lost, the papyrus text in this case must have been overall rather similar to that of the scholion, and it must have run in fairly close parallel with it. (We have of course noted a small point in which the texts differ, at l. 6, where the papyrus was surely more concise.) The opposite is true, however, of the remnants of the ἱστορία on Υ 147, where, although we can be certain that the papyrus and the scholion treated the same mythological narrative, our comparison has revealed that, if anything, there is a substantive difference between the two texts: as mentioned above, among other things, the final part of the papyrus ἱστορία was perhaps more extensive than the somewhat hurried conclusion found in the scholion. Regarding the meagre remains of Υ 307, there is little that can be said: what I have speculated above would tend to suggest that the scholion presented a redaction that was somewhat more substantial. To summarize our findings: the relationship between the papyrus text and the text of the corresponding ἱστορίαι found in the D-Scholia is characterized by a mixture of elements; on the one hand, we find similarity and parallelism, but, on the other hand, we also find fairly clear signs of differentiation. Although it is not very numerous, the group of extant papyri that present mythographic ἱστορίαι related to Homer is by now substantial enough that we can ascertain the character of these texts and we can observe their (by no means trivial) diffusion within the ancient world. Where once we only had a vague idea  10 The scholia attributed to Lycophron are discussed in Lünstedt 1961, 15 ff.; for the scholion to Υ 215, cf. also van der Valk 1963–1964, I, 329 ff. One suspects that the ἱστορίαι with subscriptiones attributing them to Lycophron did not belong to the original redaction of the Mythographus Homericus, but rather that they were inserted in the scholia from a different source, which employed Lycophron with scholia (discussion in Lünstedt, ibid.). The issue should be reexamined in greater detail: it is relevant to the overall question of the formation of the Mythographus Homericus (that is to say, to the question of its sources) and to the issue regarding its relationship with the D-Scholia of the medieval tradition.

  A Re-examination of PBerol 13282 of the earlier redaction of the Mythographus Homericus, these papyrus documents have given us a clearer view of the ancient form of this material, and an overall re-examination of all the aspects of these documents would certainly lead to many interesting contributions to knowledge, especially if these documents are to be compared with the facies that this material assumes within the D-Scholia. At present, the list of documents that have been published so far (see the Appendix) counts five items that are certain, and two that are doubtful; to these one must add the hitherto unpublished PHamb. III 199,11 which brings the total number of secure fragments to six. The chronological distribution of these documents covers a range that goes, more or less, from the first to the fifth century CE: the oldest item is POxy. 418, dated to the first or second century CE; the most recent is PSchubart 21, a parchment codex of the fifth century CE; the others date to the second and third centuries CE. The most substantial text, and perhaps the most emblematic of the group, belongs to the Odyssey: this is PSI 1173, eight fragments of a papyrus codex containing ἱστορίαι related to books γ, λ, μ, ν, and ξ. PSI 1173 is also the only certain document that treats the Odyssey; as is generally the case, the finds containing texts related to the Iliad are more numerous: in total, five are certain, and they treat parts of books A (2), N, T, Y. Leaving aside the question of PLit. Lond. 142 (whose classification within this category of texts is currently doubtful and seems destined to remain only possible),12 the item that most requires an attentive re-examination and further study is without doubt PSchubart 21. It was R. Merkelbach (building upon remarks of B. Snell, apud ed. pr.) who recognized that PSchubart 21, a fragment of a parchment codex, contained Homeric ἱστορίαι, and he was able to present a reconstruction of the ἱστορία to Υ 53 on the basis of the material found in the D-Scholion to Υ 3, restoring in l. 22 the attribution-formula present in the scholion: ἡ δὲ ἱστορία] παρὰ Δημητρίῳ τῷ Σκηψ̣[ίῳ.13 From l. 23 onward we have only meagre scraps of text related to Υ 147; these must of course belong to the initial part of the ἱστορία, and it is likely that l. 23 itself was occupied by the lemma. Perhaps in this case the narrative of Laomedon’s vicissitudes did not begin with the service rendered to him by Poseidon  11 I thank the editor B. Kramer for the authorization to report on this document before its being published. Meanwhile, it has been published: Kramer/Hagedorn 1984, 25 ff. 12 Cf. most recently Erbse, 1969–1988, II, 392, Pap. VII. 13 Cf. R. Merkelbach, “Archiv Papyr.” XVI (1958), 118. I have been able to check a photograph (for which I thank the Papyrus-Sammlung of the Staatliche Museen zu Berlin), and with great effort (the text is extremely faded) I have been able to recover a few more letters in l. 22, which confirm Merkelbach’s supplementation; one should merely add that the mute iota is not adscripted.

A Re-examination of PBerol 13282  

and Apollo (in accordance with Zeus’ wishes) as we read in the D-Scholion on this verse,14 but rather at a point earlier in the mythical background, as we find in the ἱστορία to A 399 present in the D-Scholia and in POxy. 418: Zeus had seized control of the realm of the heavens, and he had been acting arrogantly; at this point, Hera, Poseidon, and Apollo (and Athena?) formed a plot against him, but with the help of Thetis Zeus discovered the conspiracy and punished the conspirators, with Hera being placed in chains and Poseidon and Apollo being forced to render service to Laomedon; at which point the rest of the story follows.15 I reproduce here what remains of PSchubart 21, ll. 23–28. 23 25

28

[± 18 [± 20 [± 30 [± 30 [± 32 [± 35

] ̣ ̣ ̣ην ̣ ̣ ̣ ̣ ̣ ̣ ̣ ̣ ̣ ̣ ̣ ̣ ] ̣ ̣ ̣ νιοσ ̣ ̣[ ̣ ̣ ̣ ̣ ̣ ̣]σ ̣ ευς παραστ[ ]ραο καὶ Ἀπολλω[ δ]εσμοῖς ἠσφαλισμ[έν]ν προσέταξ[εν ]συμπ[

I have consulted a photograph of the document, but this has not led to many improvements in the transcription; I have only been able to note that, in l. 24, the reading ]σσευς of the editio princeps should be called into question, because the second σ is actually entirely swallowed up by a hole, apart from (perhaps) a tiny trace at the top right-hand side. We can, therefore, with confidence read instead ]ς Ζ̣ευς. It is perhaps possible, then, to imagine that Zeus, faced with the conspiracy of certain gods (among them Apollo, cf. l. 25), is able to have the better of them, and he punishes one at least (probably Hera) by placing her/him in chains (cf. l. 26), whereas he commands (cf. l. 27) others (most likely Poseidon and Apollo)... scil. to go to serve Laomedon. For each of these elements one can rely on comparison with the ἱστορία to Α 399, mentioned above.16 If this hypothesis is valid, at

 14 We have already mentioned this ἱστορία à propos of PBerol. 13282. The D-Scholion to Υ 147 is found in the Appendix below. 15 Cf. fn. 7 above; and see Jacoby, commentary on FGrHist 4 F 26; Ambaglio 1980. 16 POxy. 418 on A 399 (ll. 25 ff.): … Διὸς ἐπικρατέστερον χρωμένου [τῇ τῶν] θεῶν βασιλείᾳ, Ποσειδῶν τε καὶ Ἥρα κα[ὶ Ἀπόλ]λων ἐπεβούλευσαν αὐτῷ· Θέτις δὲ γνοῦσα [παρὰ] Νηρέως τοῦ πατρός, ὃς μάντις ἦν, δηλοῖ τῷ [Διὶ] τὴν ἐπιβουλὴν καὶ σύμμαχον παραδίδωσι τ[ὸν] Αἰγαίωνα ἑκατόγχειρον, Ποσειδῶνος παῖδα. Ζεὺς δὲ Ἥραν μὲν ἔδησεν, Ποσειδῶνι δὲ κ[αὶ Ἀπόλλω]νι προστάσσει θητεῦσαι Λαομέδον[τι. – D-Scholion on A 399: Ζεὺς παραλαβὼν τὴν ἐν οὐρανῷ διοίκησιν, περισσῶς τῇ παρρησίᾳ ἐχρῆτο, πολλὰ αὐθάδη διαπρασσόμενος. Ποσειδῶν δὲ καὶ Ἥρα καὶ Ἀπόλλων ἐβούλοντο αὐτὸν δήσαντες ὑποτάξαι. Θέτις δὲ ἀκούσασα παρὰ τοῦ πατρὸς Νηρέως (ἦν γὰρ μάντις) τὴν Διὸς ἐπιβουλὴν, ἔσπευσε πρὸς αὐτὸν ἐπαγομένη Αἰγαίωνα φόβη-

  A Re-examination of PBerol 13282 Υ 147 PSchubart 21 contained a ἱστορία regarding Laomedon that appeared in a redaction that differed from the corresponding D-Scholion — more precisely, a redaction that began its account at an earlier point in the myth and was, therefore, more extensive. Let us recall that even in the case of PBerol. 13282, regarding the remnants of the ἱστορία on this very same verse (Υ 147), we had arrived at the conclusion that, in all probability, we were dealing with a redaction that was more extensive than that of the scholion; in that case we have remnants of the final portion, whereas here we have instead the initial part of the myth. From this comparison of the various fragments, we can conclude that the defining and always-present characteristics of this category of texts are effectively two: 1) the presence of Homeric lemmata (sometimes in ekthesis) arranged in a sequence that follows the order of the Homeric narrative (this fact guarantees the close relationship between these collections and the Homeric text, and it establishes that their original purpose was to provide “commentary” on Homer with a specific reference to the mythographic content, even if it is true that they later acquired a kind of autonomous role as reservoirs of mythographic material); 2) the exclusive and specialized17 focus on content of a mythographic nature, with the intent of explaining the mythological references in the Homeric text, which are treated in a series of ἱστορίαι that are independent of each other (and graphically distinguished by means of various methods, such as the paragraphos, or by beginning a new line, and so on). A third characteristic should be mentioned and ought to be considered definitive, in the sense that, if it is present, we can be absolutely certain that the text belongs to this category, but its absence is proof of nothing, since, although frequent, it is not always present: this is the attribution (subscriptio), which assigns the material to a specific author, in the form ἡ ἱστορία παρὰ τῷ δεῖνα (vel sim.). It is precisely with regard to the attributions — which are a key element for studying the origin and the tradition of these collections — that one may make some interesting observations concerning the relationship between the ἱστορίαι  τρον τῶν ἐπιβουλευόντων θεῶν· ἦν δὲ θαλάσσιος δαίμων οὗτος καὶ τὸν πατέρα Ποσειδῶνα κατεβράβευεν. ἀκούσας δὲ ὁ Ζεὺς Θέτιδος, τὴν μὲν Ἥραν ἐν τοῖς καθ’ αὑτοῦ δεσμοῖς ἐκρέμασε, Ποσειδῶνι δὲ καὶ Ἀπόλλωνι τὴν παρὰ Λαομέδοντι θητείαν ἐψηφίσατο ... 17 Only in the ἱστορία on Α 399 in POxy. 418 (ll. 23 ff.) does one find mention of a textual variant, which corresponds to what we find in the D-Scholia and also in the A-Scholia (Ariston. on A 400) and in bT (sch. ex. ad A 400): cf. Erbse 1969–1988, I, 114. The case is interesting, because the variant (which occurs precisely in l. 400) directly involves the myth itself, and the ἱστορία has a subscriptio that attributes it to Didymus: this fact should be taken into consideration by anyone conducting further research into the origins of the collection of the Homeric ἱστορίαι.

A Re-examination of PBerol 13282  

on papyrus and those in the scholia. It is immediately clear that there is a high degree of agreement between the earlier and later texts, especially in the case of PSI 1173, in which all of the attributions find exact counterparts in the scholia; then there is the parallel case of the ἱστορία to Α 263, which lacks an attribution both in POxy. 418 and in the corresponding D-Scholion. On the other hand, there are also divergences: POxy. 3003 omits the attribution for N 302, which is found in the D-Scholion; conversely, PHamb. III 199 preserves the attribution for Α 38 Τενέδοιο, whereas the D-Scholion omits it.18 We thus find a panorama that contains every possible variation: agreement in presence or absence of an attribution; or else greater detail either in the papyrus document or in the scholion. It is important to note, however, that when the citation of the source is preserved both in the papyrus ἱστορία and in the corresponding scholion, there is always agreement: that is to say, there have been found (at least up to this point) no cases of the same ἱστορία attributed to two different sources, one named in the papyrus and another named in the scholion. This fact, naturally, bears witness to a strong link between the earlier and later texts, and it is a valid piece of evidence attesting to just how well the attributions present in the papyrus documents of the Mythographus Homericus have been preserved within the corpus of the scholia.19 An analysis, then, of the texts of the papyrus ἱστορίαι in comparison with those found in the scholia — an analysis which I do not have enough space at present to conduct in detail — leads to results that are mixed and variegated. It is by now certain that there are close connections between the two groups of texts, if nothing else from the three distinguishing characteristics above, which were outlined with regard to the papyrus texts, but are valid for the scholia as well. In addition to this, one also finds that a general agreement in subjectmatter (partly determined, of course, by reference to the Homeric text) is the

 18 Yet another situation is found in the case of the ἱστορία on A 264: in POxy. 418 it does not have any attribution, whereas in the corresponding D-Scholion we find the following: μέμνηται δὲ αὐτοῦ καὶ Ἀπολλώνιος ἐν τοῖς Ἀργοναυτικοῖς λέγων οὕτως (I 59–63). This, however, is not a true attribution but rather the citation of a parallel. 19 I mean simply to point out that the scholia, in this regard, generally seem to preserve their ancient source in a reliable manner, insofar as they maintain the attributions (subscriptiones). Whether or not the subscriptiones to the ἱστορίαι are themselves reliable (that is to say, how faithfully or not the ἱστορίαι reproduce the source that they claim to follow, and therefore how valid, or not, we should consider the original subscriptiones) is another question, and it has been examined only partially (by Lünstedt 1961, who arrives at the conclusion that the attributions are generally valid: ibid., with bibliography, to which one should add van der Valk 1963– 1964, I, 303 ff.). The question ought to be re-examined in its entirety.

  A Re-examination of PBerol 13282 rule. The two groups, then, generally overlap in content; but if we look more precisely, there is a range of possible relationships that can be differentiated. Indeed, we find a greater or lesser richness of contents and of detail sometimes in one group, sometimes in the other; at times the texts correspond with each other right down to the level of the letter, whereas at other times the linguistic form is different but the content is clearly the same and is developed in a parallel fashion; in some cases substantial portions of a lacunose papyrus ἱστορία can be reconstructed via comparison with the scholion, in other cases the two redactions are clearly divergent and the scholion offers little help in restoring the papyrus text. In these pages I have had the opportunity to give examples, at least to a limited degree, of the various possible situations. In short, the comparison of the Mythographus Homericus as attested on papyri and the corpus of mythographic ἱστορίαι contained in the scholia reveals a complex and varied set of similarities and differences. As a result, we may conclude that the collections of ἱστορίαι found on papyrus are, indeed, the ancient antecedent of the analogous material present in the scholia, but the relationship between them is not that of a direct line of transmission. Therefore, unless new finds contradict our current understanding of these texts (and frankly it seems improbable that this will happen), our next task will be to understand better the process that led from the earlier to the later redactions, by identifying possible intermediate stages and by investigating the ways in which, in this case as in others, the compilers of the medieval corpora reworked the ancient exemplars that they had at hand.

Appendix I present here the text of the mythographic ἱστορίαι that are contained in the corpus of the Scholia D in Iliadem for verses Υ 147–Υ 404. These are the texts upon which I based my comparisons between the ἱστορίαι in the D-Scholia and those conserved in PBerol. 13282. The text of the D-Scholia, at present, must be established on the basis of five principal MSS along with the editio princeps of Janus Lascaris (cf. most recently Montanari 1979, with bibliography). For the section with which we are currently concerned, the MS H is lacking, due to a lacuna resulting from the loss of a quaternion between f. 282 and f. 283, which has caused the loss of the scholia from Υ 68 to the hypothesis of Φ (cf. Montanari 1979, 100 fn. 16); as a result, the text that follows is based only upon A C R V La.

A Re-examination of PBerol 13282  

Sigla codicum A Venet. Marc. Gr. 454 Zan. (coll. 822), saec. X C Rom. Bibl. Naz. gr. 6 + Matrit. 4626 (71 Ir.), saec. X in. H Vat. gr. 2193, saec. XII R Vat. gr. 32, saec. XII V Vat. gr. 33, saec. XI ex. La J. Lascaris, ed. pr., Romae, 1517. The stemma codicum established by De Marco (cf. the bibliography in Montanari, op. cit., p. 9) is still valid and is as follows:

Υ 145–147 τεῖχος ἐς ἀμφίχυτον Ἡρακλῆος θείοιο, / ὑψηλόν, τὸ ῥά οἱ Τρῶες καὶ Παλλὰς Ἀθήνη / ποίεον, ὄφρα τὸ κῆτος ὑπεκπροφυγὼν ἀλέαιτο: Ποσειδῶν καὶ Ἀπόλλων, προστάξαντος Διὸς Λαομέδοντι θητεῦσαι ἐπὶ μισθῷ τεταγμένῳ, τὸ τεῖχος κατασκευάζουσι. Λαομέδων δὲ παραβὰς τοὺς ὅρκους καὶ τὰς συνθήκας, μὴ δοὺς τὸν μισθόν, ἀπήλασεν αὐτούς. ἀγανακτήσας δὲ Ποσειδῶν ἔπεμψε τῇ χώρᾳ κῆτος, ὃ τούς τε παρατυγχάνοντας ἀνθρώπους καὶ τοὺς γιγνομένους καρποὺς διέφθειρε. μαντευομένῳ δὲ Λαομέδοντι χρησμὸς ἐδόθη Ἡσιόνην τὴν θυγατέρα αὐτοῦ βορὰν ἐκθεῖναι τῷ κήτει καὶ οὕτως ἀπαλλαγήσασθαι τοῦ δεινοῦ. προθεὶς δὲ ἐκεῖνος τὴν θυγατέρα, μισθὸν ἐκήρυξε τῷ τὸ κῆτος ἀνελόντι τοὺς ἀθανάτους ἵππους δώσειν, οὓς Τρωῒ Ζεὺς ἀντὶ Γανυμήδους ἔδωκεν. Ἡρακλῆς δὲ παραγενόμενος ὑπέσχετο τὸν ἆθλον κατορθώσειν καὶ Ἀθηνᾶς αὐτῷ πρόβλημα ποιησάσης τὸ καλούμενον ἀμφίχυτον τεῖχος, εἰσδὺς διὰ τοῦ στόματος

  A Re-examination of PBerol 13282 εἰς τὴν κοιλίαν τοῦ κήτους, αὐτοῦ τὰς λαγόνας διέφθειρεν. ὁ δὲ Λαομέδων ὑπαλλάξας θνητοὺς δίδωσιν ἵππους. μαθὼν δὲ Ἡρακλῆς ἐπεστράτευσε καὶ Ἴλιον ἐπόρθησε, καὶ οὕτως ἔλαβε τοὺς ἵππους. ἡ ἱστορία παρὰ Ἑλλανίκῳ. (FGrHist 4 F 26b) ACRVLa 1–4 le ita CRVLa (ὑψηλὸν om RVLa; pro κῆτος: τεῖχος perper. R): Τρῶες καὶ Παλλὰς Ἀθήνη A 7 ἔπεμψε δὲ V τε om R 9 τῷ Λαομ. RVLa αὐτοῦ AC: Λαομέδοντος La: om RV 10 ἐκθεῖναι τῷ κήτει codd: ἐκδοῦναι La 11 δ’ ἐκεῖνος RV 15 καὶ αὐτοῦ AC: ὃ καὶ αὐτοῦ RV: om La

Υ 215–216 Δάρδανον αὖ πρῶτον τέκετο : Δάρδανος ὁ Διὸς καὶ Ἠλέκτρας τῆς Ἄτλαντος θυγατρὸς διϊὼν τὴν Σαμοθράκην, ἐπομβρίας γενομένης, κατασκευάσας σχεδίαν καὶ ἀσκὸν ἑαυτῷ περιθείς, διεκομίσθη εἰς τὴν Ἴδην τῆς Τρῳάδος· καὶ ἀποβὰς Διὸς ὑποθεμένου, κτίζει πόλιν, ἣν καὶ ἀφ’ ἑαυτοῦ Δαρδανίαν ἐκάλεσεν. ἡ ἱστορία παρὰ Λυκόφρονι. CRVLa e x . ( D)

Δάρδανον αὖ πρῶτον : Δάρδανος υἱὸς Ἠλέκτρας — τὴν Δαρδανίαν ἔκτισεν. AT

5 Τρωάδου R καὶalt om R 6 (ὡς R) ἱστορεῖ Λυκόφρων RVLa 7–9 sch. ex. edid. Erbse ad loc.

Υ 3 0 7 ν ῦ ν δ ὲ δ ὴ Α ἰ ν ε ί α ο β ί η Τ ρ ώ ε σ σ ι ν ἀ ν ά ξ ε ι : Ἀφροδίτη, χρησμοῦ ἐκπεσόντος ὅτι τῆς τῶν Πριαμιδῶν ἀρχῆς καταλυθείσης οἱ ἀπ’ Ἀγχίσου Τρώων βασιλεύσουσιν, Ἀγχίσῃ παρηκμακότι συνῆλθε. τεκοῦσα δ’ Αἰνείαν καὶ βουλομένη πρόφασιν κατασκευάσαι τῆς τῶν Πριαμιδῶν καταλύσεως, Ἀλεξάνδρῳ πόθον Ἑλένης ἐνέβαλε καὶ μετὰ τὴν ἁρπαγὴν τῷ μὲν δοκεῖν συνεμάχει τοῖς Τρωσὶ, ταῖς δὲ ἀληθείαις παρηγόρει τὴν ἧτταν αὐτῶν, ἵνα μὴ παντελῶς ἀπελπίσαντες ἀποδῶσι τὴν Ἑλένην. ἡ ἱστορία παρὰ Ἀκουσιλάῳ. (FGrHist 2 F 39) A C R V La 1 le βίη — ἀνάξει om A 2 ante Ἀφροδίτη extat in A sch. Ariston. (307 a1 σημειοῦνται — ἀρχήν Erbse) τῶν om RVLa 3 Ἀγχ. παρ. A: Ἀγχ. καὶ παρ. C: Ἀγχ. δὲ Ἀφροδίτη παρ. RV: Ἀγχ. ἤδη παρ. La 6 ἐνέβαλε ACLa: συνέβαλε RV 7 μὴ om AC 8 πίσαντες C ἱστορεῖ Ἀκουσίλαος RVLa.

Υ 4 0 4 Ἑ λ ι κ ώ ν ι ο ν ἀ μ φ ὶ ἄ ν α κ τ α : τὸν Ποσειδῶνα, ἤτοι ὅτι καὶ ἐν Ἑλικῶνι ὄρει τῆς Βοιωτίας τιμᾶται ἢ ἐν Ἑλίκῃ, μᾶλλον οὖν παρὰ τὸν ἐν Ἑλίκῃ θεόν. διαφέρει γὰρ Ἑλικὼν καὶ Ἑλίκη, ὅτι Ἑλικὼν μὲν Βοιωτίας ὄρος, Ἑλίκη δὲ νῆσος τῆς Ἀχαΐας ἱερὰ Ποσειδῶνος. ἡ δὲ ἱστορία αὕτη. Νηλεὺς ὁ Κόδρου, χρησμὸν λαβών, ἀποικίαν ἔστειλεν εἰς Μίλητον καὶ τὴν Καρίαν ἐξ Ἀθηνῶν καὶ τῆς Ἀχαιικῆς Ἑλίκης. παραγενόμενος δὲ εἰς τὴν Καρίαν, ἱερὸν Πο-

A Re-examination of PBerol 13282  

σειδῶνος ἱδρύσατο καὶ ἀπὸ τοῦ ἐν Ἑλίκῃ τεμένους Ἑλικώνιον προσηγόρευσε. δοκεῖ δὲ, ἐπὰν θύωσι τῷ θεῷ, βοησάντων μὲν βοῶν προσδέχεσθαι τὸ θεῖον τὴν θυσίαν· σιγώντων δέ, λυποῦνται μηνίειν νομίζοντες. ἡ ἱστορία παρὰ Κλειτοφῶντι. (FHG IV 368, cf. Jacoby, RE XI, 661) ACRVLa 1 le ὡς δ’ ὅτε ταῦρος (ἤρυγεν ἑλκόμενος A) Ἑλ. ἀ. ἄν. ACRVLa καὶ om RVLa 3 θεόν — Ἑλίκη om A 4 ὄρ. Βοι. RVLa τῆς om RVLa ἱερὰ AC: ἱερὸν RV: ἐν ᾗ ἱερὸν La 4 ἡ δὲ ἱστορία — 10 Κλειτοφῶντι om R 5 Νείλεως C ἔστελλεν VLa 6 τὴν om La 6 ἐξ Ἀθηνῶν — 7 τὴν Καρίαν om VLa (ἔνθα ἱερὸν Π. κτλ. La) 8 ἐν om V Ἑλ. τὸν θεὸν προσηγ. La ἐπὰν AC: ἐπειδὰν VLa 9 τῷ θεῷ CV: τὸν θεὸν A: om La τῶν βοῶν La 10 σιγῶντα A λυπ. μην. νομ. AC: λυπεῖσθαι καὶ μηνιᾶν νομίζεσθαι VLa παρὰ om C

Inventory of Papyri Containing Mythographic Ἰστορίαι Related to Homer Iliad 1

Iliad 2

Iliad 3

Iliad 4

Iliad 5

(?) Iliad 6

Odyssey 1

PHamb. III 199 (Inv. Nr. 81 recto); pap. roll; saec. II p Α 38, 39 Ed. pr. B. Kramer, PHamb. III (Bonn, 1984), pp. 25 ff. POxy. 418 (Pack2 1164); pap. roll; saec. I-II p Α 263, 264, 399 Ed. pr. B.P. Grenfell – A.S. Hunt, POxy. III, 1903, pp. 63 ff. POxy. 3003; pap. roll; saec. II p Ν 301 (?), 302, 459; Ξ 319; Ο 229 Ed. pr. P.J. Parsons, POxy. XLII, 1974, pp. 15 ff. PSchubart 21 = PBerol. inv. 13930 (Pack2 1203); parch. cod.; saec. V p Τ 332; Υ 53, 147 Ed. pr. W. Schubart, Gr. Lit. Pap. 1950, p. 45 ss.; recogn. (coll. B. Snell apud ed. pr.) R. Merkelbach, «Archiv Papyr.» XVI (1956), pp. 117 f. PBerol. 13282; pap. cod.; saec. III p Υ 147, 307, 404 Ed. pr. W. Müller, «Forsch. Bericht.» (1978), p. 118 ff.; recogn. Montanari 1985a (= this volume, ch. 20). PLit. Lond. 142 (Brit. Mus. inv. 1605, Pack2 1188); pap. roll; saec. II p Ι 447 Ed. pr. H.J.M. Milne, Catalogue, 1927, p. 121; cf. Pfeiffer 1937, 16– 18; Erbse, 1969–1988, II, 392 f. PSI 1173 (Pack2 1209); pap. cod.; saec. III p γ 4, 91; λ 321, 322, 326, 519, 582; μ 70, 85; ν 96, 259; ξ 327 Ed. pr. G. Coppola, PSI X, 1932, pp. 131 ss.; add. Pfeiffer 1937, 14–16.

  A Re-examination of PBerol 13282 (?) Odyssey 2 PVindob. Gr. Inv. 29784 (PRainer I 17, Pack2 2447); pap. roll; saec. III p λ 308 ss., 576 ss. (?) Ed. pr. H. Gerstinger, “Mitt. Pap. Samml. Wien” n.s. I (1932), pp. 130 ff.; cfr. Pfeiffer 1937, 16 n. 25.

 The Mythographus Homericus Introduction The so-called Mythographus Homericus (henceforth MH) is one of the many interesting issues in the field of ancient Homeric scholarship and mythography. In the light of the progress of research and the new evidence obtained above all from papyri, I would argue that it is of interest to reconsider this text globally and to review the main questions it raises. I will thus mainly be concerned with its origin and its historical-cultural position. Before providing a definition of what is meant by the term MH, a few words are required on the so-called D-scholia to Homer, an important corpus of Homeric scholia which includes a considerable amount of glossographical and paraphrastic material as well as a good deal of zetemata, mythographical historiae and some other material.1 The D-scholia are the most widespread scholia corpus in Byzantine manuscripts: they were first edited by Janus Lascaris (Rome 1517, the Iliad), and by Franciscus Asulanus (Venice 1528, the Odyssey) and remained almost the only scholia widely used and known until Villoison published the Venetian scholia in 1788, though subsequently falling into neglect. After the editiones principes of Lascaris and Asulanus, the D-scholia were reprinted several times in different forms, but no modern critical edition has so far been produced. The great edition of the Scholia vetera to the Iliad by H. Erbse includes the scholia derived from VMK (Viermännerkommentar, a compilation of materials taken from Didymus, Aristonicus, Nikanor, Herodianus) together with the so-called Scholia exegetica: the element still needed to complete the set of the Scholia vetera to the Iliad is precisely the corpus of D-scholia. For about a century now, interest in this class of scholia has been growing.2 I myself have been working on an edition of the D-scholia to the Iliad for several years and here I use texts from Iliadic D-scholia drawn from my provisional collation. The D-scholia to the Iliad are much richer and quantitatively more extensive than

 English translation by Rachel Barritt.  1 Moreover, the D-scholia are usually provided in the codices by the hypotheseis (book summaries). 2 Status quaestionis in Montanari 1979; cf. Schmidt 1976; Snipes 1988; Montanari 1994b (= this volume, ch. 34). https://doi.org/10.1515/9783110772371-021

  The Mythographus Homericus those to the Odyssey (I thus believe that the corpus of D-scholia to the Odyssey can be included in a new edition of the complete Odyssey scholia).3 The glossographical material of the D-scholia, providing a ‘translation’ of Homeric lemmata represented by single words or short expressions into an easier and more understandable form of Greek, carries on the ancient Scholia minora or glossaries, which are widespread in many papyri from the first century BCE to the seventh century CE. The same glossographical material is also found in the so-called Lexeis Homerikai (very close to the D-scholia), in Apollonius Sophista and in many streams of Greek lexicography. Indeed, this is in itself a highly interesting feature of the D-scholia, but it is not the subject I intend to focus on at this time. The mythographical material is represented by a fairly large group of short stories of varying length (historiae), providing mythographical accounts whenever the Homeric text offers a suitable opportunity. Such historiae usually have the following structure: they are introduced by the Homeric lemma providing the link with the text of the Iliad or the Odyssey, which is followed by the main body of the comment with a mythological account; finally, at least in most cases, one finds a subscription attributing the content of the historia to some authority: ἡ ἱστορία παρὰ τῷ δεῖνα or (ὡϛ) ἱστορεῖ ὁ δεῖνα, οὕτωϛ ὁ δεῖνα.4 At times, however, the subscription is missing. These mythographical historiae are precisely what is meant by ‘MH’ in the corpus of the Homeric D-scholia transmitted by the codices. Its ancient predecessor is represented by similar mythographical historiae on papyrus fragments, for which the name MH is also used. Thus when we talk about the MH, we are dealing with that whole set of mythographical historiae on Homer found in papyrus fragments as well as in the D-scholia. The corpus of Iliad D-scholia contains some two hundred mythographical historiae, while a significantly smaller number are to be found in the D-scholia to the Odyssey. As for the papyri, they provide a markedly richer testimony today than was available twenty or even only ten years ago and are composed of about ten pieces, eight for the Iliad and two (of which one is doubtful) for the Odyssey. The list at the end of this article stands to my knowledge as the complete repertory of MH fragments available to date. However it should be kept in mind that there exists a multitude of papyri containing remains of mythographical narrations, among which it is quite likely that some fragment of the MH may  3 It is reported that an edition of the Scholia to the Odyssey is being prepared by A.R. Dyck. 4 But one can also have ὡς γὰρ ἱστορεῖ ὁ δεῖνα and then the historia (e.g. Hellanicus in D-sch. on Iliad 3.144).

The Mythographus Homericus  

be hidden. If fate has resulted in loss of the Homeric lemma and of the subscription in the fragment and if the preserved historia turns out to be noticeably different from that present in the D-scholia or indeed if it is not present at all in the D-scholia, then identification of the papyrus text will be extremely difficult and always open to doubt. Fortunately, however, the undisputed pieces do offer sufficient material for satisfactory evaluation. Let us begin with a few general observations on the papyrus fragments of the MH. First, we will consider their chronological distribution. The oldest piece dates back to the period between the end of the first century and the beginning of the second century CE (Il. 002, cf. also 004), but most of the pieces belong to the second century CE. From the third century CE we have one papyrus of the Iliad (008) and the only certain piece of evidence of the Odyssey (001); the latest find, a parchment codex referring to the Iliad (007), dates from the fifth century. As for the book formats, the largest group consists of the papyrus rolls datable roughly to the second century CE (Il. 001, 002, 003, 005, 006);5 there are also two papyrus codices of the third century CE (Il. 008; Od. 001) and a parchment codex from the fifth century CE (Il. 007). Taken together, these pieces provide a range of findings stretching from about the first to the fifth century CE and presenting the various book formats that were widespread in the imperial age. Of particular interest, from this point of view, is the ostrakon Iliad 004 (PSI 1000), dating from the 1st–2nd century CE, in which a small remnant of the historia to Iliad 13.217–8 has been identified.6 In all likelihood this was no more than a limited copying exercise from a text that must obviously have borne the title of the book, the Homeric lemma and the corresponding historia (or possibly a group of historiae). N 216–8

εἰσάμενος ϕθογγὴν Ἀνδραίμονος υἷι Θόαντι ὅς πάσῃ Πλευρῶνι καὶ αἰπεινῇ Καλυδῶνι Αἰτωλοῖσιν ἄνασσε

Ostrakon PSI 1000. 1 3

]... τῆς Ν ὃς πάσῃ] Π̣λευρῶνι καὶ αἰπεινῇ Καλυδῶνι Αἰτωλὸς ὁ Ἐ]ν̣ δυμίωνος παῖς κατήντησεν εἰϛ Ἆπιν ?]

Ν 217

 5 I will disregard Od. 002, a papyrus roll of the third century CE, of uncertain identification. 6 The text had previously been given a doubtful attribution to an unknown tragedy; cf. TrGF II F 709 (= adesp. 323 d Nauck/Snell).

  The Mythographus Homericus 1 τῆς Ν (poss. (Ἰλιάδoς) ἱστο]ρ̣ί̣ α̣ι) Salvadori 2–4 suppl. Salvadori: εἰς Ἐπειόν, ὃϛ ἦν Ἐ]ν̣ δυμίωνος παῖς, κατήντησ | ἡ βασιλεία ed. pr. Sch. D Ν 217. Πλευρῶνι καὶ Καλυδῶνι: ἀντὶ τοῦ Πλευρῶνος καὶ Καλυδῶνος, εἰσὶ δὲ πόλεις Αἰτωλίας. CHV La P.Ryl. 536: Scholia minora N 198–562 Recto → col. 1.12–13 Π]λευρὼν καὶ [Καλυδὼν πόλεις Αἰτωλίας [

Ν 217

Sch. D N 218. Αἰτωλοῖσιν ἄνασσε: Αἰτωλὸς παῖς μὲν ἦν Ἐνδυμίωνος, ὃς ἀκούσιον ϕόνον δράσας ἔϕυγεν εἰς τὴν ἀπ᾿ αὐτοῦ προσαγορευθεῖσαν Αἰτωλίαν κἀκεῖ τεκνοῦται Πλευρῶνα, ἀϕ᾿ οὗ ἡ ἐν Αἰτωλίᾳ πόλις Πλευρὼν ἐκλήθη. τούτου δὲ γίνονται δύο παῖδες Κούρης τε καὶ Καλυδών, ἀϕ᾿ ὧν ἄλλαι δύο πόλεις ἐν Αἰτωλίᾳ προσαγορεύθησαν. ACHRV La Sch. ex. Ν 217–8. ὃς πάσῃ Πλευρῶνι : Αἰτωλὸς ὁ Ἐνδυμίωνος, Ἠλεῖος τὸ γένος, Ἇπιν ἀκουσίως τὸν Φορωνέως ἀνελὼν φεύγει εἰς τὴν ἀπ᾿ αὐτοῦ Αἰτωλίαν προσαγορευθεῖσαν, ἴσχει δὲ παῖδα Πλευρῶνα, οὗ ἐγένοντο Κούρης καὶ Καλυδών, ἀϕ᾿ ὧν αἱ πόλεις. οὕτω Δηΐμαχος (FGrHist 65 F 1). Τ

The tiny remnant of the ostrakon is sufficient to indicate that it is not the remains of a glossary, as can be seen by comparing it with the D-Sch. to 13.217 and with P.Ryl. 536: on the contrary it was a genuine historia, as confirmed by comparing it with D-Sch. to 13.218 and with Sch. ex. to 13.217–218, which preserve two different versions that can fairly certainly be traced back to the same original source (unfortunately, no evidence on papyrus is available for this particular historia). The presence of the verb κατήντησεν suggests that the story of the fight between Endymion’s son Aetolus and Apis, followed by the involuntary killing, was more extensive in the ostrakon source than the more concise form in the scholia. In any case, this utilisation is eloquent testimony to the popularity of the MH already in the second century. The second observation is that the material in our possession allows us to state that the MH provided a systematic mythographical commentary to the entire Iliad and the entire Odyssey. PSI 1173 (Od. 001), dating from the third century CE, is the only certain papyrus for the Odyssey, but it is also one of the most interesting fragments and the most extensive one: from this papyrus codex parts of eight leaves are extant, in which historiae to books 3, 11, 13 and 147 can be recognised. This was therefore an exemplar that undoubtedly covered the entire poem (in forty or so pages?). For the Iliad the documentation is richer: 003 (P.Oxy. 3830 + P.Lit.Lond. 142), a papyrus roll of the second century CE, has preserved parts of books 7, 8, 9; 005 (P.Oxy. 3003), also a papyrus roll of the  7 One of the eight fragments is unidentified and could belong to yet another book.

The Mythographus Homericus  

second century CE, offers remains of books 13, 14, 15; 006 (P.Oxy. 4096), another papyrus roll of the second century CE, covers books 18–24;8 the other pieces of evidence present historiae to books 1, 13, 19, 20, and among these there are the remains of a papyrus codex of the third century CE (008) and the remains of a parchment codex of the fifth century CE (007). For the Iliad we have no extant exemplar similar to papyrus codex PSI 1173 of the Odyssey, which covers no fewer than twelve books (from 3 to 14), but Iliad 006 (P.Oxy. 4096) is even more extensive as far as the quantity of text preserved is concerned and offers fragments of a papyrus roll with remains from books 18–24, covering a total of seven books and extending to the end of the poem. The overall set of findings ranges from book 1 to book 24 and provides unequivocal evidence that this type of erudite exegesis existed for the whole of the Iliad as well. The evidence available to date also indicates that the MH was in circulation quite separately from any other kind of exegetic material. This is a point worth noting, since the imperial age is dotted with examples of rich miscellaneous commentaries of highly diversified content: discussions on text constitution intermingled with exegetic material, grammatical disquisitions, matters of antiquarian interest and so on. In the framework of Homeric scholarship, there is certainly no lack of hypomnemata offering a rich array of contents, yet it would appear that the MH circulated quite independently, unaccompanied by other material, appearing in copies containing only the isolated historiae. The findings that have come to light so far have provided ample confirmation that this was indeed the general rule. A similar case is that of the glossographical material called Scholia minora, mentioned at the beginning of this paper: this material is present in a large number of glossaries (taking both the Iliad and the Odyssey together, the total is close to one hundred), in which Homeric lemmata represented by single words or short expressions are accompanied by a ‘translation’ into an easier and more understandable form of Greek. Indeed the numerous fragments of glossaries (Scholia minora) and a considerable number of papyri of the MH demonstrate that during the imperial age these two streams of Homeric scholarship each enjoyed considerable autonomy and were widespread at several different levels, even in schools. Literal translation into a currently used language and mythographic explanation (in different forms and with different intentions) are the two most ancient forms of Homeric exegesis, traceable back to at least the fifth century BCE: at the same time they also supplied the most

 8 The edition in the P.Oxy. series will be prepared by P. Schubert, to whom I would like to express my heartfelt thanks for the information concerning the papyrus.

  The Mythographus Homericus substantial amounts of materials that eventually contributed to forming the corpus of Homeric D-scholia.

Structure and origin of the MH How was the MH built up? How did it originate? Some time ago the MH was regarded merely as a sort of mythographical handbook, differing little from many other similar works in circulation, which could have been a school compilation and was certainly not considered a high level work. Such a view was derived from the opinions expressed by E. Schwartz, E. Maass, E. Bethe and J. Panzer towards the end of the last century. Schwartz went as far as to deny that the subscriptions of the historiae in the Homeric D-scholia could be awarded any reliability at all and he rejected the indications of sources contained therein. When discussing the indirect use of sources in Ps.-Apollodorus Bibliotheca, he refused to accept the view that, on account of the subscriptions, the historiae in the D-scholia should be regarded as genuine excerpts from the authors cited. Rather, he argued, they had derived from the use of compendia similar to Ps.-Apollodorus Bibliotheca and the subscriptions were no more than quotations placed at the end, which could refer to a single variant or to a part of the story, and only in a few fortunate exceptions, when the compiler had taken the entire story from an author and had cited the latter, did they encompass the entire excerptum without substantial error.9 Such a view was also shared by E. Bethe, who identified the origin in a ‘compendium fabularis historiae’,10 and by E. Maass, who then claimed that the historiae were not the fruit of Homeric exegesis, but derived instead from a ‘mythologisches Handbuch’, probably dating from the second century CE and similar both to Ps.-Apollodorus and to Ps.-Hyginus.11 Following a similar line of thought, J. Panzer contended that the origin of this material was to be found in a separate handbook of mythology: its textual tradition as mythographical D-scholia, he maintained, was of a later date and wholly secondary: Hoc igitur ex iis quae supra disputavimus pro certo opinor habemus, aliunde ἱστορίας reliquiis scholiis D adscriptas esse ... ad explicanda Homeri verba ἱστορίας primitus non fuisse destinatas contendo ... Compendio

 9 Schwartz 1881; synthesis in Schwartz 1894, 2878. 10 Bethe 1887, 80–90. 11 Maass 1884, 537 n. and 563–564.

The Mythographus Homericus  

debentur mythologico Apollodori bibliothecae simillimo.12 Thus the true origins of the MH were kept at one remove from the framework of Homeric scholarship13 and high erudition, on the assumption that its historical-cultural status was that of a fairly late and rather low-level mythographical compendium, based on indirect knowledge gleaned from manuals, and only subsequently used to enrich collections of Homeric scholia with mythographical notes: none of these could be viewed as particularly erudite, despite the presence of subscriptions containing many great names of the noblest Greek literature, for the subscriptions were not held to be valid. The first observation that needs to be made at this point is that the view outlined so far was based exclusively on an examination of the historiae in the Homeric D-scholia and could not benefit from any evidence from papyri. The first MH papyrus was edited in 1903 (Il. 002), but taken in itself, it was still far from providing a clarification of the true nature of the problem. The richest one, PSI 1173 (Od. 001), was not published until 1932. Naturally, comparisons with the Homeric scholia were made within the general framework of speculation on ancient precedents of Byzantine scholiastic corpora, but little progress was achieved on this particular problem and the specific question of what had now become known as the MH was not taken up again. It is my belief that we now need to adopt a different approach, grounded on more careful and painstaking analysis of the materials at hand: it needs to be emphasised that the papyri have supplied crucial evidence on this matter, evidence which was not available earlier and failed to be awarded due consideration subsequently. Without taking into consideration the MH fragments on papyrus, Lünstedt 1961 likewise devoted a dissertation to the subscriptions of the historiae in the D-scholia: to his mind, examination of the verifiable subscriptions, i.e. those referring to extant authors, tended to lead to a substantially positive assessment concerning their validity. The essence of the results achieved by Lünstedt deserves to be emphasised at this point. First and foremost, it was shown that the subscriptions are not without grounds, and that there exists a genuine link with the author cited in each subscription: elements are shown to come frequently not only from the author’s text but also from the scholia to this text and can therefore be traced back to ancient commentaries. One should not make the mistake, he reasoned, of thinking that the general intention of the subscriptions  12 Panzer 1892, 61–63; cf. also Wilamowitz 1925, 76–77. 13 I should add that a different opinion was in fact expressed: Schwarz 1878 believed that the historiae derived from a mythological commentary on Homer, but his opinion and arguments were decidedly refuted by Panzer 1892, 61–62.

  The Mythographus Homericus is always that of indicating a veritable ‘source’ for the historia as a whole, for it may be also a suggestion of a parallel text or a learned comparison (indeed, the very concept of trying to evaluate them strictly as ‘sources’ has constituted a false presupposition of research on these subscriptions). He therefore claims that the basic reliability of the subscriptions resides in the fact that the reference to the author cited does actually exist in some way, inasmuch as the authors they quote deal with at least a part of the relevant historia. On the other hand, the conclusion that they should not be taken as an indication of global ‘sources’ of the scholion is corroborated by the presence of elements derived from other sources as well, i.e. ancient comments on the cited author himself or parallel texts by other authors. This is a rather crucial difference in approach, although somewhat cautious.14 The fact that the subscription does not indicate the ‘source’ of the whole of the material of the historia is in itself no longer seen by Lünstedt as the crucial factor: what is decisive is that the author cited (and at times also the ancient exegesis to the author) should have a genuine direct link with the content of the scholion. Along the same lines, M. van der Valk15 claims that the material of the mythographical D-scholia does not appear to be worse or more recent than that found in parallel versions contained above all in the exegetical scholia to Homer. In fact, he goes as far as to cast out Schwartz’s hypothesis according to which the MH was built up exclusively with the use of handbooks and therefore contains only low-level and late material. Overall, then, although the author cited is often pertinent only to a part of the content of the historia, van der Valk concludes in favour of the basic reliability of the subscriptions of the scholia, despite revealing skepticism in more than a few cases.16 The last couple of decades have seen the publication of several studies by G. Arrighetti,17 who has been able to make use of the evidence furnished by the papyri and thus to set up an explicit linkage between the papyrus fragments of the MH and the historiae in the Homeric D-scholia. The term ‘commentary’18 is now confidently used for the MH, the origins of which are seen as lying in the  14 Lünstedt (1961, 36) concluded that the last word had not yet been said as regards the negative judgement passed by Schwartz and urged further investigation of those subscriptions that were not directly verifiable, i.e. those pertaining to lost authors. 15 Van der Valk 1963–1964, I 7 (The mythographical D scholia), 303–413. 16 For subscriptions citing the Kyklos, cf. also Janko 1986, 51–55, esp. 52–53. 17 Arrighetti 1968; 1977a; 1977b, 52–54; 1987, 204–210. 18 Turner 1968, 119 (= 138), had defined PSI 1173 (Od. 001) as a hypomnema: in contrast, for PSI 1173 and for P.Schubart 21 (Il. 007), Turner 1977, nr. 207 and nr. 206 respectively, uses the term scholia.

The Mythographus Homericus  

exegetic activity carried out in the highest circles of Hellenistic-Roman scholarship. I myself put forward the same opinion several years ago, as I began to investigate the D-scholia,19 and my conviction is constantly being strengthened by acquisition of new data and new arguments. The papyrus fragments show that, even in the most ancient forms that have come down to us, what we now call MH is properly a commentary on the Homeric poems, following the text step by step and linked to it by the quotation of lemmata: each historia takes as its starting point some feature mentioned in the poetic text, which may even be just a single word or an expression that is felt to call for an explanation or an extended comment. It was a mythographical commentary, by which I mean a commentary dealing with a specific subject-matter, and to continue calling it a ‘handbook’ of Homeric mythology would be misleading: it was a commentary which had a fairly widespread circulation from the first century CE onwards and continued in circulation throughout the imperial age. In all identified MH fragments the order follows that of the Homeric text and the lemmata are always present; moreover in P.Oxy. 3830 (Il. 003), in P.Schubart 21 (Il. 007) and in P.Berol. 13282 (Il. 008) the lemmata are in ekthesis, a characteristic of the commentary. In some cases one also finds a form of title in passing from one book to another: PSI 1173 (Od. 001) fr. 5 recto preserves the end of book 11 and the beginning of book 12, which is indicated by Ὀδ(υσσείας) μ. Forms of title are also found at two points in P.Oxy. 4096 (Il. 006); a title is also present in the ostrakon Il. 004, see above. By chance, one piece suggests an extremely interesting line of inquiry. Mention has already been made several times of codex PSI 1173 (Od. 001), with historiae to several books of the Odyssey, a large part of which deal with the heroines Odysseus encounters in 11.235–332: in frgs. 2–4 there remain parts of six historiae to 11.321–322 and 326, followed directly by a piece (in fr. 4 verso) of the historia on Eurypylus at 11.519–520. A papyrus of Odyssey book 11, P.Mil.Vogl. 259, a small fragment dating from 1st–2nd century CE, provides evidence of the initial parts of 11.292–310 and definitely has a diple before 11.298 and perhaps the trace of another diple before 11.305.20

 19 Montanari 1979, 14–15; cf. Janko 1986, 51–55, with interesting observations on the D-sch. on Il. 23.346 (subscription to Kyklikoi) and the story of Cycnus. 20 Ed. pr. by D. Del Corno, P. Mil. Vogliano, VI, Milano 1977, nr. 259, 1–3; corrections to the ed. pr. (in which the transcription of one line was omitted) in Moretti 1993, 96. Through the kind assistance of G. Bastianini I was able to see the original of the fragment preserved at the Istituto di Papirologia of the University of Milan.

  The Mythographus Homericus Tab. 1 Odyssey 

historiae in PSI 

historiae in D-sch.

 Pero

XXX Pherecydes

 Iphicles

XXX missing

 Leda

XXX neoteroi

 Iphimedea

P.Mil.Vogl. 

---------diple diple ? ----------

 Phaedra

---------XXX fr.  Asclepiades

XXX Asclepiades

Procris

XXX fr.  missing

XXX Pherecydes

– Ariadne

XXX fr.  missing

XXX Pherecydes

 Maera

XXX fr.  Pherecydes

XXX Pherecydes

Clymene

XXX fr. – Hesiod

XXX Hesiod

Eriphyle

XXX fr.  Asclepiades

XXX Asclepiades

– Eurypylus

XXX fr. v missing ----------

XXX Acusilaus

The situation is summarised in the chart, in which the broken lines indicate the fractures in the fragments, leading one to regret that so little is preserved of PSI 1173 and P.Mil.Vogl. 259. The use of the diple by Alexandrian philologists to indicate discussion of exegetic questions, including questions of antiquarian interest, is well known: thus even if for 11.298 (καὶ Λήδην εἶδον, τὴν Τυνδαρέου παράκοιτιν) the scholia and codices highlight a grammatical problem concerning the form of the genitive Τυνδαρέου or Τυνδάρεω, I regard it as likely that the presence of this diple or of these two diplai in the tiny Milan fragment is to be related to the presence

The Mythographus Homericus  

of a mythographical comment on the character cited in the line.21 This fascinating suggestion admittedly appears to rest on somewhat shaky foundations: however, it will be seen below that, when considered together with certain other elements, the idea begins to acquire a little more plausibility.22

The Papyri Thus what we know of the MH has come down to us through two different forms of transmission: the group of papyrus fragments dating from Roman times traceable to the first through fifth centuries CE and the historiae contained in the corpus of the Homeric D-scholia preserved in a few codices of the Byzantine age. If we consider the papyrus fragments and compare them with the text of the D-scholia, we observe first and foremost that no papyrus fragment shows exactly and totally the same text as the D-scholia manuscripts, no papyrus fragment is totally and absolutely different from the D-scholia manuscripts: thus it is not unknown for the text of a given historia on a shattered and fragmentary papyrus in one case to be reconstructable on the basis of the D-scholion, whereas in another case one may well find that comparison with the corresponding D-scholion does not allow the lacunae of the papyrus to be integrated. Sometimes the papyrus texts are richer and more extensive, whereas at other times it is the papyri that present an abridged and impoverished version. But in general, despite the often considerable differences, there is a basic similarity among all the sources of evidence, showing beyond a shadow of doubt that the papyri preserve what must be regarded as the ancient precursor of the historiae of the Homeric D-scholia, i.e. that the historiae in D can be traced back to a mythographical commentary on Homer of the imperial age, which is known thanks to papyrus fragments and is called MH. It is therefore fully legitimate to regard the papyrus fragments and the Homeric D-scholia as phases of the transmission of the MH, as we have been doing so far, in a line of transmission within which the characteristics — similarities as well as differences — can all be provided with a plausible explanation.  21 In PSI 1173 the historiae are separated from one another by a long horizontal line, which one hesitates to define as a paragraphos: it may be purely by coincidence that the same is seen in P.Mil.Vogl. 259, where before 11.298 and before 11.305 there remains a piece of interlinear separation line which, even allowing for the small part of column preserved, would appear to be rather too long for a normal paragraphos. 22 Of course, inquiry into diple preservation for lines provided with historia in the scholia is what is needed at this point.

  The Mythographus Homericus The possibility of comparing two extant papyri both presenting the same historia is for the moment extremely limited. This is indeed regrettable, as it would be highly instructive to be able to benefit from at least a few such comparisons. We have only one case in which a historia to the same Homeric line appears in two papyri, namely the historia at 20.403–4 found both in P.Oxy. 4096 (Il. 006) and in P.Berol. 13282 (Il. 008): the two versions are similar and parallel but not identical, and a few differences can be spotted. For the historia at 20.147, P.Schubart 21 (Il. 007) and P.Berol. 13282 (Il. 008) have a notably more extensive version compared to the corresponding D-scholion, but the former preserves the beginning of the historia and the latter the final part: consequently, we cannot tell whether the two editions had identical texts.23 However, I believe it is doubtful that exactly the same text will ever be found in two papyri, that is, that we will ever find two papyrus copies of exactly the same version of MH. For as so often happens in this kind of text, each individual copy represented a different version and copyists or redactors felt relatively free to modify or adapt the text as they produced a new exemplar. Furthermore I would also argue that it is highly probable that the first compilation of the D-scholia was not composed by copying one single exemplar of MH, but rather the redactor D may well himself have produced a new version based on more than one single copy at his disposal (as indeed he did when using his sources for the glossographic material). Now, because a new version always involves a certain amount of changes, this can explain why one finds the above described situation of similarities and differences with the papyrus versions. Even the fact that at times the author cited in the subscription may strictly speaking be associated with only a part of the historia is not always to be taken as representing the original situation, for it could be due to the process of transmission and manipulation of the exegetic material starting from learned products of the Hellenistic and Roman age up to the Byzantine scholia. As far as the crucial element represented by the subscriptions is concerned, the whole range of possibilities can be observed.

 23 Cf. Montanari 1985, 235–236.

The Mythographus Homericus  

Tab. 2 N 

om. P.Oxy. 

Pherecydes D-Sch.

A  Τενέδοιο

Myrsilus & Euripides (?) P.Hamb. 

om. D-sch.

Θ 

Euphorion P.Oxy. 

om. D-sch.

A 

om. P.Oxy. 

om. D-sch.

H 

Anticleides P.Oxy. 

Anticleides D-sch.

Y 

Demetrius of Scepsis P.Schubart 

Demetrius of Scepsis D-sch. (ad Y )

γ

Hellanicus PSI 

Hellanicus D-sch.

λ  Phaedra

Asclepiades PSI 

Asclepiades D-sch.

λ 

Pherecydes PSI 

Pherecydes D-sch.

λ 

Hesiod PSI 

Hesiod D-sch.

λ 

Asclepiades PSI 

Asclepiades D-sch.

λ 

Asclepiades PSI 

Asclepiades D-sch.

v 

Lycophron PSI 

Lycophron D-sch.

N 

subscr.??? P.Oxy. 

Apollodorus D-sch.

Y 

subscr.??? P.Berol. 

Acusilaus D-sch.

First of all one notices a large number of agreements between the papyri and the scholia manuscripts. But on the other hand there are also differences: for instance P.Oxy. 3003 (Il. 005) omits the subscription at the end of the historia at 13.302, while the D-scholion preserves the subsciption to Pherecydes. On the other hand, P.Hamb. 199 (Il. 001) shows the subscription (a double one, to Myrsilus and possibly Euripides)24 at 1.38 Τενέδοιο, while the corresponding Dscholion has lost it. The same happens in P.Oxy. 3830 (Il. 003), which preserves the subscription to Euphorion at 8.479, lost in the corresponding D-scholion.25 For 1.263 the subscription is lost both in P.Oxy. 418 (Il. 002) and in the manuscripts; but in contrast: P.Oxy. 3830 (Il. 003) and the D-scholion preserve the same subscription to Anticleides for 7.44; P.Schubart 21 (Il. 007) and the D-scholion

 24 Ed. pr. 25–26: ἱστοροῦσιν] / Μύρτιλος καὶ Ἐ[υριπίδης; Luppe (ZPE 56 (1984), 31–32): οὕτως] / Μύρτιλος καὶ Ἐ[υριπίδης. For the second name Euripides is preferred because he wrote a tragedy on Tennes; I wonder however whether the association with Myrsilus of Methymna might not lead one to prefer — since the topic focused on is Tenedos — another author of Lesbiaka, so possibly: Μύρτιλος καὶ Ἑ[λλάνικος? (cf. P.Oxy. 3711? see comm. to fr. 1, I 10–7; II 15–7). 25 For the subscription to Eratosthenes in the historia to 9.447 of the same papyrus, cf. infra.

  The Mythographus Homericus preserve the same subscription to Demetrius of Scepsis for the historia at 20.53 (which in the D-scholion is found at 20.3); in PSI 1173 (Od. 001) seven subscriptions are preserved and all of them agree perfectly with the corresponding Dscholion (3.4 Hellanicus; 11.321 [Phaedra] Asclepiades; 11.326 Pherecydes, Hesiod, Asclepiades; 11.582 Asclepiades; 13.259 Lycophron); in P.Oxy. 3003 (Il. 005) there was a subscription at 13.301 but the name is lost in lacuna, and the Dscholion attributes the historia to Apollodorus; in P.Berol. 13282 (Il. 008) there was a subscription at 20.307 but the name is lost in lacuna, and the D-scholion attributes the historia to Acusilaus. Thus when analysing the subscriptions, it can be seen that the range of possiblities includes agreement between papyri and manuscripts of D both in presence and in omission, as well as differences involving greater richness either in the papyri or in the manuscripts of D (subscription preserved in the papyrus and lost in D, subscription preserved in D and lost in the papyrus). It is remarkable, however, that when the subscription has survived both in the papyri and in the manuscripts, there is always agreement between them. That is, at least so far, there has in effect never been a discrepancy in the attribution of the same historia, in that the same historia has never been ascribed to different authorities in different pieces of evidence. This is of course a most important connection as well as a clear sign of the reliability of the tradition by which the subscriptions have survived within the corpus of the D-scholia, where such subscriptions cannot be considered a later addition. Generally the subscriptions belonged to the original form of the MH and have been well preserved in the D-scholia. There is one exception to this general rule that is well worth taking into consideration. In P.Oxy. 3830 (Il. 003), for the historia at 9.447 (the Phoenix story), the first editor26 suggested that it might be possible to reconstruct in the papyrus a subscription to Eratosthenes from the traces of fr. 3 II 21: thus she argues that ἡ δ᾿ ἱστορία] | παρ᾿ Ἐρ̣[ατοσθέ]ν̣ει (11. 20–21) is to be preferred to a possible παρ᾿ Εὐ̣[ριπίδηι].ει, which would leave us with a highly problematic ].ει at the end, but she prudently refrains from integrating the text. The reconstruction of the name Eratosthenes is put forward rather more confidently by Haslam,27 who however also compares it with the different subscription of the same historia in the corresponding D-scholion: ἡ ἱστορία παρὰ τῷ ποιητῇ, παρὰ γὰρ

 26 M.A. Harder in P.Oxy. LVI. 27 Haslam 1990, 34–36: cf. W. Luppe, Gnomon 64 (1992), 292.

The Mythographus Homericus  

τοῖς τραγικοῖς διαλλάσσει.28 This is undoubtedly an anomalous case. Haslam’s explanation is as follows: ‘Ιn the light of the papyrus subscription it is clear that the (formally atypical) subscription in A [scil. in ACH, that is in D-sch.] is not a preservation lost in the other parts of the tradition, but an addition consequent on the loss of the Eratosthenes subscription.29 Therefore the copy or copies of the MH available to the redactor of the Iliadic D-scholia lying at the origin of our tradition had lost the subscription for the Phoenix-historia at 9.447: then somewhere along some branch of the tradition30 an unknown hand filled the gap with a subscription created ad hoc. No connoisseur of scholiastic literature will be startled to find that the editor of the Byzantine corpus or the copyist/user of a manuscript did not hesitate to boldly insert or add material or blend together materials of varying origin or epitomise materials with differing degrees of precision. Even if the form of the subscription does not strike me as being particularly atypical,31 I regard this as a plausible explanation. However, one may still wonder whether, given the very small part extant, restoration of the name of Eratosthenes in the papyrus should be considered as definite, or whether Euripides could still be an outside possibility, or even something else.32 In any case, this kind of exception certainly does not invalidate the general observation of absence of discrepancies between  28 For the subscription Haslam knows only the text of A, but it is also found in C and H, and therefore in branch u of the tradition, while it does not appear in branch ν: therefore either it was present in the archetype and ν omitted it, or else it was not present in the archetype and was added only in u. For the manuscripts and the tradition of D-scholia to the Iliad cf. Montanari 1979; 1984 (= this volume, ch. 32); 1994b (= this volume, ch. 34). 29 Haslam 1990, 34 n. 1. 30 Cf. n. 28. 31 For the reference to Homer himself, cf. the subscriptions (all to the Iliad) above all to 13.66, but also to 3.443, quoted infra. Double subscriptions are by no means rare: P.Hamb. 199 (Il. 001) has been cited for the subscription to Myrsilos and Euripides (?) at 1.38 Τενέδοιο; cf. infra the subscriptions at 2.145, 3.242, 13.66. For a generic reference such as οἱ τραγικοί cf. κατὰ τοὺς τραγικούς to 1.7, but one can also make a comparison with οἱ κυκλικοί (e.g. 3.242 quoted infra, and 18.486; 19.332; 20.346; 20.660), οἱ ῥητορικοί to 1.50 or possibly even πολλοὶ ἐμνήσθησαν to 6.130 (quoted infra). Comparisons between Homer and tragedy are also not without parallels: e.g. to 1.7 concerning Agamemnon’s family. Cf. Lünstedt 1961, 1–6. 32 Cf. Sch. ex. (vel Porph.) Iliad 9.453 c = Sosiphanes 92 F 6 TrGF, Euripides p. 621 TGF2: the link between sch. 9.453 c and the historia to 9.447 is found in TGF2, l.c., and is also mentioned by Lünstedt, l.c. Thus one might think of something like παρ᾿ Εὐ̣ριπίδηι καὶ Σωσιϕάν̣ει, which would perhaps give a name to the τραγικοί cited in the D-sch.: however, this would appear to be too long for the lacuna in the papyrus fr. 3 II 21 unless there were some abbreviations, for which in this same papyrus cf. fr. 1–2 Ι 1, 13; therefore perhaps e.g. παρ’ Εὐ̣ρ(ιπίδηι) κ(αὶ) Σωσιϕάν̣ει.

  The Mythographus Homericus the papyri and the D-scholia in attribution of the same historia and the assertion that the subscriptions belonged to the original form of the MH and have been well preserved in the D-scholia.33

Sources and subscription A peculiar and interesting case is represented by the historia at Iliad 1.264, which is worth taking into consideration. This historia is preserved in P.Oxy. 418 (Il. 002) as well as in the D-scholia and the form is quite similar: Sch. D A 264. Καινέα τ᾿: καὶ τὸν Καινέα. ὁ δὲ Καινεὺς Ἐλάτου μὲν ἦν παῖς, Λαπίθων βασιλεύς. πρότερον ἦν παρθένος εὐπρεπής, μιγέντος δὲ αὐτῇ Ποσειδῶνος, αἰτησαμένη μεταβαλεῖν εἰς ἄνδρα ἡ νεᾶνις34 ἄτρωτος γίνεται γενναιότατος τῶν καθ᾿ αὑτὸν ὑπάρξας. καὶ δήποτε πήξας ἀκόντιον ἐν τῷ μεσαιτάτῳ τῆς ἀγορᾶς, θεὸν τοῦτο προσέταξεν ἀριθμεῖν· δι᾿ ἣν αἰτίαν ἀγανακτήσας ὁ Ζεὺς, τιμωρίαν τῆς ἀσεβείας παρ᾿ αὐτοῦ εἰσεπράξατο. μαχόμενον γὰρ αὐτὸν τοῖς Κενταύροις καὶ ἄτρωτον ὄντα, ὑποχείριον ἐποίησε· βαλόντες οἱ προειρημένοι δρυσί τε καὶ ἐλάταις ἤρεισαν εἰς τὴν γῆν. μέμνηται δὲ αὐτοῦ καὶ Ἀπολλώνιος ἐν τοῖς Ἀργοναυτικοῖς λέγων οὕτως· “Καινέα γὰρ δὴ πρόσθεν ἔτι κλείουσιν ἀοιδοὶ | Κενταύροισιν ὀλέσθαι, ὅτε σϕέας οἶος ἀπ᾿ ἄλλων | ἤλας᾿ ἀριστήων· οἱ δ᾿ ἔμπαλιν ὁρμηθέντες | οὔτε μιν ἀγκλῖναι προτέρω σθένον οὔτε δαΐξαι, | ἀλλ᾿ ἄρρηκτος ἄκαμπτος ἐδύσατο νειόθι γαίης | θεινόμενος στιβαρῇσι καταΐγδην ἐλάτῃσιν” (A.R. 1.59–64).

ACHV P.Oxy. 418.9–2235 A 264

[Καινέα τ᾿ Ἐξάδιό]ν τε καὶ ἀντίθεον Πολύϕημον [ὁ Καινεῦς Ἐλάτου] μὲν παῖς, Λαπίθων δὲ βασιλε[ύς, [πρότερον παρθέ]νος εὐπρεπής ἐγένετο· δ[ [ μιγέν]τος αὐτῇ Ποσειδῶνος, αἰτησα[με[νη μεταβαλεῖν νε]ανίας ἄτρωτος γίνεται· γ[εν[ναιότατος δὲ τ]ῶν καθ᾿ ἑαυτὸν ὑπάρξας, τον[

10

 33 Thus Haslam 1990, 34 n. 1: ‘Discrepancy in attribution between the D-scholia and a papyrus witness, apart from mere presence or absence of the subscription, would be unique. It becomes increasingly clear that the subscriptions were part and parcel of the original work, and trustworthy, but were subject to omission and to mistaken reparation.’ 34 Probably νεανίαϛ legendum: cf. P.Oxy. 418.13 and n. 36. 35 The papyrus itself is lost and no photograph exist. I give the text of the ed. pr. by Grenfell & Hunt, adding punctuation, breathings, accent, iota mutum. Besides: 12 ποσιδωνοϛ pap.—13 ]ανειαϛ, corr. ]ανιαϛ pap.; γεινεται pap.—20 ατροτο[, corr. ατρωτο[ pap.—22 ηρισαν pap.

The Mythographus Homericus  

[ τῆς ἡγε]μ̣ονίας οὐκ ἐβάστασεν. ἐξευ[τε[λίσας δὲ καὶ το]ὺς θεοὺς παρ᾿ οὐ[δὲν ἐποιή[σατο καί ποτε πή]ξας ἀκόντιον ἐν [μέση̣ τῆ̣ [ ] ἀγορᾷ, τοῦτο θεὸν π̣[ροσέτα[ξε νομίζειν· Ζεὺ]ς δὲ ἀγανακτήσα[ς Κενταύ[ροις πολεμοῦν]τα, καίπερ ἄτρωτο[ν ὄντα ὑπο[χείριον ἐποίησε]ν· ἐλάταις γὰρ κα[ὶ δρυσίν [οἱ Κένταυροι] αὐτὸν ἤρεισαν εἰς [γῆν.36

15

20

The papyrus text seems to contain something more in ll. 11–2 and ll. 14–6, but at the end of the story (ll. 17–22) it is a little shorter than the D-scholion (ll. 4–7).37 But while the papyrus text stops when the story is finished,38 in the scholion we find the quotation of Apollonius Rhodius, 1.59–64, that is to say the explicit reference to the authority, not a simple reference of the kind we can read for instance at Iliad 3.237 ὡς Ἀπολλώνιός ϕησιν ἐν δευτέρῳ Ἀργοναυτικῶν. This explicit quotation could just as easily have been replaced by one of the normal forms of subscription, something like ἡ ἱστορία παρὰ Ἀπολλωνίῳ or (ὡς) ἱστορεῖ Ἀπολλώνιος or οὕτως Ἀπολλώνιος, but in actual fact it was not, and this shows that here the text of the scholia preserves a more learned and richer version of the MH than this particular papyrus: I think this version was a more ancient one, since it is well known that the citations of the authorities are the first to be lost in transmission of erudite material (as is frequently observed in lexicographical works). It is therefore far more probable that the quotation was originally present and was then omitted in certain copies, rather than being absent originally and added at a later date. What we have, in fact, is a typical case of the process of impoverishment of exegetic material, from a more erudite version to one that was reduced and simplified. Such an example should be carefully kept in mind. The historia of the MH gives more mythological material than that which can be strictly extracted from Apollonius’ lines: this could be due to the cultural level of the comment at the origin and some material may possibly have been borrowed  36 11–2 δ[ιὰ | δὲ τὸ κάλλος Blass apud ed. pr. —14–5 τὸν [μέ | γαν ὄγκον Blass apud ed. pr. —17– 8 ἐν [μεσαι | τάτῃ τῇ] ἀγορᾷ brevius: ἐν [τῷ με | σαιτάτω τῆς ἀγορᾶ?—18–9 poss. τοῦτο θεὸν π̣[ροσέταξεν ἀριθμεῖν· Ζεὺ]ς δὲ κτλ. 37 Cf. ed. pr.: ‘This [scil. the D-sch.] is almost identical with the papyrus, but is more compressed in some parts and more expanded in others. As before, the papyrus exhibits the better text, (1) by avoiding the repetition of ἦν in the first sentence, (2) by having νεανίας in place of ἡ νεᾶνις which is detrimental to both sense and construction, and in the light of the papyrus should be corrected to νεανίας.’ 38 At least as far as we can see in the edition, since the papyrus itself is lost: but it seems to me quite unlikely that the papyrus edition could have committed such a blunder as to fail to indicate whether there was some more text between ll. 22 and 23.

  The Mythographus Homericus from commentaries or scholia to the passage. For instance in the collection of Apollonius’ scholia we have a mythographical comment giving the detail — not present in Apollonius’ lines — of Caeneus’ transformation from a woman into an invulnerable man, an element that the historia does furnish. It is well to recall Lünstedt’s remarks concerning historiae containing material which derives jointly from the text of the author cited in the subscription and from ancient comments to the given author. But investigation into this case can go beyond the confines of what is strictly speaking the evidence of the MH. There is an extremely interesting parallel in P.Oxy. 1611, a substantial fragment of a hypomnema to an unknown work,39 of which there remain erudite comments deprived of the lemmata of the work to which the commentary refers, but still introduced by the typical ὅτι (scil. τὸ σημεῖον ὅτι): thus this was probably already an excerptum from the original hypomnema. A section of this text, ll. 38–96, has a close bearing on the questions under discussion here, though naturally we do not know what sparked the comment. The comment opens with the quotation of a passage from Theophrastus, On Kingship 2, on the subject of Caeneus’ spear.40 It then goes on to say that Theophrastus’ statement must be elucidated with the help of the story related by Acusilaus of Argos and at this point Acusilaus’ text itself is given,41 and it shows a number of elements similar or parallel to the Caeneus’ historia of the D-scholia and of P.Oxy. 418, to which it is in fact much closer than that which one finds, say, in the Apollonius Rhodius scholia. Afterwards, the passage taken from Acusilaus is used by the author of the hypomnema to explain the proverbial expression τῷ δόρατι ἄρχειν τὸν Καινέα, to which he referred in the passage from Theophrastus, and it is further used as the launch pad for discussion of a passage from the Alcmeon in Corinth by Euripides.42 It can thus be seen that the excerptum offered by P.Oxy. 1611 is extremely erudite and far removed from the simplicity of the historiae of the MH: but in any case it does give an idea of the possible origin of this material. In the exegesis of Iliad 1.264 the passage from Acusilaus could well have been utilised together with Apollonius Rhodius (and maybe other material as well?) in an erudite commentary rich in references and doctrine such as that constituted by P.Oxy. 1611, and the whole could therefore represent a typical source used by the MH. Certainly, one

 39 Attention to this text and its relevance for the purposes of the present discussion is due to Arrighetti 1968: later in Arrighetti 1977a; 1977b; 1987, 204–210. 40 Theophrastus fr. 600 FHS&G. 41 FGrHist 2 F 22. 42 Fr. 73a TGF2 Suppl.

The Mythographus Homericus  

could perfectly well have found, as indeed occurred elsewhere, that the MH had preserved the historia with a subscription like ἡ ἱστορία παρὰ Ἀκουσιλάῳ (as for instance is observed in Iliad 20.307 and Odyssey 11.519–520) or (as mentioned above) like ἡ ἱστορία παρὰ Ἀκουσιλάῳ or else ἡ ἱστορία παρὰ Ἀκουσιλάῳ καὶ Ἀπολλωνίῳ. But instead, in this case, in one exemplar of the MH (P.Oxy. 418) we find the historia shorn of any subscription, and in another (the D-scholia) we have the historia with an explicit citation, this time not lost as so often happens but preserved in literal form, similarly to what one finds in excerpta from commentaries such as P.Oxy. 1611. Another rather interesting case is offered by the historia on Iliad 1.399–400, still preserved in P.Oxy. 418 and in the D-scholia. A 399–400

ὁππότε μιν ξυνδῆσαι Ὀλύμπιοι ἤθελον ἄλλοι ‘Ήρη τ᾿ ἠδὲ Ποσειδάων καὶ Παλλὰς Ἀθήνη

Ariston. A 400 α. Παλλὰς Ἀθήνη: ὅτι Ζηνόδοτος γράφει “ϕοῖβος Ἀπόλλων”. ἀϕαιρεῖται δὲ τὸ πιθανόν· ἐπίτηδες γὰρ τοὺς τοῖς Ἕλλησι βοηθοῦντας . A Sch. ex. A 400 b. Ἥρη τ᾿ ἠδὲ Ποσειδάων καὶ Παλλὰς Ἀθήνη: πιθανῶς τοὺς Ἀχαϊκοὺς θεοὺς ἠχθρευκέναι Διΐ ϕησιν, ἵνα ἀκούοιτο Θέτις. τινὲς δὲ γράϕουσι “καὶ ϕοῖβος Ἀπόλλων”· τοῦτον γὰρ εἰκὸς ἠχθρευκέναι μὴ φέροντα τὸν ζῆλον τῆς ἐκείνου τυραννίδας· ὅθεν Ἥραν μὲν δεῖ, τοὺς δὲ Λαομέδοντι ὑποτάσσει. οἱ δὲ Ἥραν μὲν διὰ τὸ πολλαῖς μίγνυσθαι, Ποσειδῶνα δὲ διὰ τὸ πλεονεκτεῖσθαι εἰς τὴν διανομήν, Ἀθηνᾶν δὲ διὰ τὸ ἀναγκασθῆναι ζευχθῆναι Ἡϕαίστῳ. bT

Cf. sch. ex. 399–406 and sch. ex. 400 c. Sch. D A 399–400. ὁππότε μιν ξυνδῆσαι < Ὀλύμπιοι ἥθελον ἄλλοι | Ἥρη τ᾿ ἠδὲ Ποσειδάων καὶ Παλλὰς Ἀθήνη>: γράϕεται “καὶ Φοῖβος Ἀπόλλων”. Ὀλύμπιοι δὲ ἄλλοι οἱ Τιτᾶνες· καὶ γὰρ ἐν ἄλλοις “μάλα γάρ κε μάχης ἐπύθοντο καὶ ἄλλοι οἵ περ ἐνέρτεροί εἰσι θεοὶ Κρόνον ἀμϕὶς ἐόντες” (Ο 224–5). CHRV Ζεὺς παραλαβὼν τὴν ἐν οὐρανῷ διοίκησιν περισσῶς τῇ παρρησίᾳ ἐχρῆτο, πολλὰ αὐθάδη διαπρασσόμενος. Ποσειδῶν δὲ καὶ Ἥρα καὶ Ἀπόλλων καὶ Ἀθηνᾶ ἐβούλοντο αὐτὸν δήσαντες ὑποτάξαι. Θέτις δὲ ἀκούσασα παρὰ τοῦ πατρὸς Νηρέως (ἦν γὰρ μάντις) τὴν Διὸς ἐπιβουλὴν, ἔσπευσε πρὸς αὐτὸν, ἐπαγομένη Αἰγαίωνα ϕόβητρον τῶν ἐπιβουλευόντων θεῶν· ἦν δὲ θαλάσσιος δαίμων οὗτος, καὶ τὸν πατέρα Ποσειδῶνα κατεβράβευεν. ἀκούσας δὲ ὁ Ζεὺς Θέτιδος, τὴν μὲν Ἥραν ἐν τοῖς καθ᾿ αὑτοῦ δεσμοῖς ἐκρέμασε, Ποσειδῶνι δὲ καὶ Ἀπόλλωνι τὴν παρὰ Λαομέδοντι θητείαν ἐψηφίσατο, τῆ δὲ Θέτιδι τὴν Ἀχιλλέως τιμὴν εἰς τὰ μετὰ ταῦτα ἐταμιεύσατο. ἱστορεῖ Δίδυμος (Schmidt p. 179 fr. 1). ACHRV

  The Mythographus Homericus P.Oxy. 418.9–2243 A 399

ὁππ[ότε μιν] ξ[υν]δῆσαι Ὀλύμπιοι ἤ[θε]λ[ον ἄλλοι γράϕ[ου]σί τινες “καὶ Φοῖβος Ἀπόλλων”· ϕ[ασὶ γὰρ ὅτι Διὸς ἐπικρατέστερον χρωμένου [τῆ τῶν θεῶν βασιλείᾳ, Ποσειδῶν τε καὶ Ἥρα κα[ὶ Ἀπόλλων ἐπεβούλευσαν αὐτῷ. Θέτις δὲ γνοῦσα [παρὰ Νηρέως τοῦ πατρὸς, ὃς μάντις ἦν, δηλοῖ τῷ [Διὶ τὴν ἐπιβουλὴν καὶ σύμμαχον παραδίδωσι τ[ὸν Αἰγαίωνα ἑκατόγχειρον Ποσειδῶνος παῖδα [ Ζεὺς δὲ Ἥραν μὲν ἔδησεν, Ποσειδῶνι δὲ κ[αὶ [Ἀπόλλω]ν̣ ι προστάσσει θητεῦσαι Λαομέδον[τι44

25

30

For line 400 we have (as is shown in Aristonicus’ scholion and in the Scholia exegetica) an ancient varia lectio of Zenodotus, which replaces Athena by Apollon as the third god involved in the plot against Zeus. The scholia show that Aristarchus rejected Zenodotus’ text, saying that Homer had purposely chosen three pro-Greek gods to plot against Zeus because Zeus had listened to Thetis’ prayers against the Greeks (explaining Homer by Homer). The sch. ex. reports Zenodotus’ reading, ascribing it to τινες, but on account of the presence of Apollo among the gods hostile to Zeus the sch. ex. preserves an element that is lost in Aristonicus’ scholion, namely, that Apollo is portrayed as having rebelled because he could no longer bear Zeus’ tyranny: such an explanation would appear to allude to some sort of plot against the king of the gods, who is then said to have punished them by putting Hera in fetters and sending Apollo and Poseidon off to be servants of Laomedon. The mythographical account of P.Oxy. 418 is from the very beginning closely related (ϕασὶ γάρ) to the variant (without Zenodotus’ name, but with τινες, l. 24, as happens in the sch. ex.), and consequently later on it names Poseidon, Hera and Apollo as the three gods in the plot (ll. 26–27). Similarly in the Dscholia45 the quotation of the variant is connected with the historia, after which four gods, with a degree of unsophisticated syncretism, are named as enemies of Zeus, namely Poseidon, Hera, Apollo and Athena (l. 6). Yet curiously, in its last part the scholion says that the well known punishments were inflicted upon

 43 Cf. n. 35. 44 26 ποσιδων pap.—30 αιγεωνα, εκατονχειρον, ποσιδωνος pap.—31 ποσιδωνι pap. 45 Despite the fact that in some manuscripts the first part of the scholia to the passage is separated from the historia: it can be observed that A omits the first part of the D-sch. (present in CHRV) probably because information concerning the variant, in A itself or more properly in its model, was already given in Aristonicus’ scholion (which unfortunately is mutilated).

The Mythographus Homericus  

Hera, Poseidon and Apollo (ll. 9–11), and this is in accordance with Zenodotus’ text. This discrepancy suggests a rather rough and ready summarised version made by someone who may perhaps have shied away from a longer version and from more detailed discussion of how interpretation of the passage could depend on which text was accepted. The subscription to Didymus is preserved in D (l. 12): but unfortunately the papyrus is damaged here and we cannot state whether there was one or not. In any case, the presence of the variant is also an important feature indicating a learned origin, certainly preserved in the MH on papyrus in a linkage which I think is very likely to be the original one. The text of Iliad 1.399–400 takes on a different meaning depending on whether the name is Athena or Apollo: in the former case, the ill-will of three deities towards Zeus deals with issues strictly internal to the Iliad, while in the latter case there is an allusion to an ancient story of a struggle among the gods, following which the walls of Troy were built at the time of Laomedon and then the first destruction of Troy was carried out by Heracles. It therefore seems to me perfectly reasonable to deduce from this picture that Didymus told the story in the commentary to 1.399–400, thereby discussing the textual variant: as a result, the historia became closely linked to Didymus in later arrangements of the material, and in the transmission of the MH the subscription ἱστορεῖ Δίδυμος is found. It should not be forgotten that Aristonicus’ scholion makes it clear beyond a shadow of doubt that the rejection of Zenodotus’ reading was due to Aristarchus. The scholion begins with ὅτι, scil. τὸ σημεῖον ὅτι, or rather ἡ διπλῆ περιεστιγμένη ὅτι, as was the norm for scholia illustrating an Aristarchean argument against Zeonodotus’ text.46 It is thus Aristarchus who preferred the text naming the three enemy deities as Hera, Poseidon and Athena because this is pithanon, inasmuch as it is linked to the promise made to Thetis by Zeus at the beginning of the Iliad: any idea of a reference here to an ancient plot by Hera, Poseidon and Apollo, harking back to the old story of Laomedon and all its consequences, seems unacceptable to Aristarchus. This is an application of the criterion of ‘explaining Homer by Homer’ and the same criterion guides the Aristarchean interpretation of the Διὸς βουλή at Iliad 1.5, as reported by the scholia to this line and indeed by a particularly rich D-scholion. This D-sch. states that it is thought by some that the Διὸς βουλή refers to Zeus’ intention of relieving the earth of the excessive quantity of men and their wickedness, by causing many  46 A small detail can be added: the monster Briareos or Aegaeon, whom Thetis summons to the aid of Zeus, in the Homeric text, l. 402, is called ἑκατόγχειρον, an epithet that is absent in the D-sch. but preserved in the papyrus, l. 30: no scholia are extant for this word, but Apollonius Sophista 65.15 says: ὁ Ἀρίσταρχος κατὰ τὸ μυθικὸν ἑκατὸν χεῖρας ἔχοντα κτλ.

  The Mythographus Homericus to perish in the two great wars of Thebes and Troy; on the other hand, according to Aristarchus and Aristophanes the Διὸς βουλή at 1.5 should be taken merely as Zeus’ intention to keep the promise to Thetis of honouring Achilles by bringing about the defeat of the Acheans. Here too we find one interpretation referring to the mythical macrocontext and a contrary one that remains internal to the Iliad. This problem is discussed in the scholia, which point out that Aristarchus and Aristophanes opted for the second approach: particularly extensive among these scholia is the above mentioned D-Sch. to 1.5, endowed with a learned quotation from the Cyclic poem Cypria.47 That this scholion belongs to the MH is in my view highly probable, even though we do not have the typical ἡ ἱστορία παρὰ τῷ δεῖνα or similia. The presence of a grammarian such as Didymus in the subscriptions to the historiae, alongside poets, historians and mythographers, constitutes a feature we would do well to reflect on.48 The name of Didymus also appears for the historia to Iliad 22.126–127 οὐ μέν πως νῦν ἐστὶν ἀπὸ δρυὸς οὐδ᾽ ἀπὸ πέτρης | τῷ ὀαριζέμεναι. The D-sch. contains an explanation of the proverbial expression in this line (and such a situation is by no means surprising for the ancient exegesis of poets),49 with a reference to the nomadic life of mankind in ancient times, when women would give birth in rocky caves and hollow oak trees: the scholion then closes with the subscription ἡ ἱστορία παρὰ Διδύμῳ.50 It is clear that here the ‘anthropological’ explanation of the origins of the proverb is in no way strictly connected with the Homeric contextual use, where Hector uses it with paradoxical force as if to say: this is certainly not the right moment to meet Achilles the way a boy and a girl have a tender meeting behind an oak-tree or a rock. However in the D-sch. a much more general explanation is given,51 which could have belonged to an extensive learned excursus in a commentary on the Homeric line: indeed, the material of such an excursus may well have been available to Didymus, who was the author of the Περὶ παροιμιῶν. This is therefore a highly plausible subscription, despite the fact that its relation with the Homeric text is generic rather than specific. Furthermore, a case such as this shows that the

 47 Fr. 1 Bernabé. 48 Schmidt 1854, 182 and again 213–214, claimed it was possible to go as far as to hypothesise that the scholia with the subscription ἡ ἱστορία παρὰ τῷ δεῖνα could be traced back to Didymus (this is also mentioned by Lünstedt 1961, ii). 49 Cf. above the proverb concerning Caeneus’ spear in P.Oxy. 1611. 50 Schmidt 1854, 182 fr. 15. 51 Which, if anything, calls to mind Hesiod Theogony 35 (and this could perhaps have been cited as a parallel).

The Mythographus Homericus  

historiae of the MH could also involve notable variety of content, even though the strictly mythographical material was by far the most prominent. It is also worth noting that even the mythographical content was not always of the same standard type: a divergence from the simple short story to which the historiae generally restrict themselves is found for example in the D-Sch. to Iliad 2.103, a historia dedicated to the epithet ἀργεϊϕόντης, which begins with an etymological exegesis, continues with a lengthy aetiological explanation starting from Zeus’ love for Io, and closes with a subscription to Apollodorus: ἡ δὲ ἱστορία πλατύτερον κεῖται παρὰ Ἀπολλοδώρῳ ἐν τῇ α᾿ (ἐν δευτέρῳ Α).52 The brief scholion immediately preceding it on the same Homeric line contains an etymological explanation of the epithet διάκτορος, which in this line is, as always, used together with ἀργεϊϕόντης: I contend that what we have at this particular point in the scholia is the separate presence of two pieces deriving from a unified treatment composed of etymological interpretations and mythological aetiologies on the epithets of Hermes διάκτορος ἀργεϊϕόντης. This content cannot fail to remind us of the Περὶ θεῶν of Apollodorus of Athens. In order to offer a few more examples of subscriptions taken from the D-scholia that I consider to be particularly noteworthy, I would also like to recall the following items. Space prevents me from commenting on them one by one, but I believe that even the short text of these subscriptions stands as a testimony of traces of true scholarly investigation. Β 145 ... ἱστορεῖ Φιλοστέφανος καὶ Καλλίμαχος ἐν Αἰτίοις. Γ 242 ... ἡ ἱστορία παρὰ τοῖς Κυκλικοῖς, καὶ ἀπὸ μέρους παρ᾿ Ἀλκμᾶνι τῷ λυρικῷ. Γ 443 ... ὡς δηλοῖ ὁ ποιητὴς διὰ τῶν ἑαυτοῦ λόγων. Ζ 130 ... τῆς ἱστορίας πολλοὶ ἐμνήσθησαν, προηγουμένως δὲ ὁ τὴν Εὐρωπίαν πεποιηκὼς Εὔμηλος. Ν 66 ... ἡ ἱστορία παρὰ Καλλιμάχῳ ἐν Αἰτίοις καὶ παρ᾿ αὐτῷ τῷ ποιητῆ ἐν τῇ δ᾿ τῆς Ὀδυσσείας (488 ff.) παχυμερῶς.

Conclusion This list could indeed be extended, but I believe that the evidence provided so far will certainly be sufficient to allow constructive conclusions.

 52 On this historia cf. Lünstedt 1961, 26–28, who correctly interprets πλατύτερον with the meaning of ‘more extensive and detailed’ (ausführlicher), i.e. as a specific observation concerning the comparison.

  The Mythographus Homericus I believe the MH derives from reliable sources belonging to the field of Homeric scholarship and of learned mythography. Some compiler built up the MH by borrowing from sources with mixed material, like hypomnemata or other learned products, and shaping it into a veritable specialised commentary which, from a certain point of time onwards, circulated separately. This is a by no means unusual case in which reduced, epitomised and simplified or specialised forms gained favour at the expense of the products of the great scholarly tradition, inasmuch as many of the fragments of hypomnemata or syngrammata that have come down to us on papyri dating from the early centuries of the imperial age are already in epitomised form. The MH may have assumed its present form, as shown in the papyri, just after the Augustan age: Didymus is quoted, the most ancient papyrus still preserved dates back to the 1st–2nd century CE and the ostrakon discussed above is of the same date. It was then presumably copied in this form all through the Roman age, finally ending up in the first compilation of our D-scholia.53 Naturally, along the way it underwent important and substantial modifications and, as was to be expected, its quality declined, at least from a scholarly point of view, while yet retaining traces of its high-level origins. Quite frequently, breaks in the continuity and complexity of exegetical analysis must have occurred, with the result that some elements are now found in different scholia, so that it is not always possible to reconstruct the integrity of the line of argument and the links between different parts and types of materials. I am of course aware that having overall learned origins does not provide an exception-free rule and does not necessarily imply absolute reliability of all the final results. I consider the subscriptions in general to be reliable in the sense we have seen in this discussion, but it is clear that case-by-case examination must be carried out with possibly different outcomes. However, my overall assessment is that we can be reasonably sure that the material contained in the MH must not be viewed as deriving from a fairly late and low-level mythographical compendium, based merely on indirect manualistic knowledge. On the contrary, it appears to me that there is strong evidence that somebody made a selection from high-quality learned commentaries of the Alexandrian age, thereby producing this mythographical commentary to Homer, which was transmitted in later centuries in a number of versions. Most of the material, in my view, probably comes from the Alexandrian tradition, but there is no reason to take this as a strict constraint on interpretation of each single element. As is  53 Put together in the Protobyzantine period or at the very beginning of the Byzantine Renaissance: the most ancient manuscript dates from the second half of the ninth century: cf. bibliography at n. 28.

The Mythographus Homericus  

only natural in this sort of product, materials from different sources may have been added or conflated at varying points in time. For instance, it is impossible to say at which stage elements of allegorical interpretation were included in the main body of mythological erudition, but this is hardly surprising in the eclectic and miscellaneous products of the Roman age. As a matter of fact, within the corpus of the D-scholia, allegorism does not figure prominently, but its presence should be noted in a few important passages.54 In the evidence available from the papyri, on the other hand, to date no passages of allegorical exegesis have been discovered. I would like to conclude by pointing out that the hypothesis of the origin and transmission of the MH sketched here can be set within a typology that is far from unusual in the history of (Homeric) scholarship. An interesting parallel I would mention, mutatis mutandis, is the Homeric lexicon of Apollonius Sophista, on which an interesting state-of-the-art survey was given recently by M. Haslam. Apollonius has come down to us in epitomised form from only one complete manuscript, the codex Coislinianus gr. 345 of the 10th century and from a handful of papyrus fragments from the Roman age (1st–6th centuries). These papyri vary considerably in form and version, and clearly each copy is the outcome of an individually made selection: none of them is identical with the Coislinianus text, but similarities and differences are intermingled. The text has undergone sundry modifications over time and in the different exemplars: ‘How different can two texts be, and still be regarded as texts of the same work? ... In that sense they are different works. But each of them may still have a right to be called Apollonius’ lexicon. That lexicon was a discrete textual entity, originating as such at a particular time and place, and its subsequent textual instability did not compromise its ontological integrity. This is what distinguishes the lexicon from what modern scholars collectively label scholia minora. As a popular reference work, it could change without losing its identity, and could exist simultaneously in multifarious forms ... Adaptability and success went hand in hand’.55 So also for Apollonius we have a series of fragmentary papyri covering different parts of the lexicon and the instability of the text can be seen from a comparison with what is found in the Coislinianus: here too, at times it is the tenth century codex that presents the richer text and a form closer to the original, whereas at other times it is the exemplar on papyrus. More complex glosses  54 I would like to remind for instance of D-sch. on Il. 5.385, where one finds allegorical interpretations adduced in opposition to Aristarchus; equally important is D-sch. on Il. 20.67, where an articulated allegorical explanation of the battle of the Gods is given. 55 Haslam 1994, 108–109; 114.

  The Mythographus Homericus were simplified and among the first elements to be lost were literal citations of authors; text impoverishment created imprecision affecting names and arguments of scholars who supported this or that interpretation. The materials for setting up an alphabetic Homeric lexicon were available in the whole gamut of Homeric scholarship (hypomnemata and syngrammata as well as the rich and ancient production of glossaries), from whence they could successfully be extracted and organised: ‘The full range of the lexicon’s material was available in ready-packaged form, requiring only reassembly’.56 All in all, there are many elements in the history of the origin and tradition of Apollonius Sophista that bear a high similarity to what we have reconstructed for the MH. Perhaps the greatest differences lie firstly in the fact that an authorial name is known for the lexicon of Apollonius Sophista, whereas the MH was probably composed by an obscure grammarian working with the aid of good erudite sources; and secondly by the fact that the lexicon of Apollonius has been preserved as an independent entity in a Byzantine codex, even though in a strongly epitomised form, while the MH has been preserved within the Byzantine corpus of the D-scholia, forming one of its main sources. Ilias

001.

A 38, 39. P.Hamb. 199 r—Vol. pap., IIp. Ed. pr. B. Kramer, P.Hamb. III, 1984, 25–34; cf. W. Luppe, ZPE 56 (1984), 31–32.

002.

A 263, 264, 399. P.Oxy. 418 (Pack2 1164)—Vol. pap., I–IIp. Ed. pr. B.P. Grenfell/A.S. Hunt, P.Oxy. III, 1903, 63–65.

003.

H 8 (?), 44, 86; Θ 479; I 447. P.Lit.Lond. 142 + P.Oxy. 3830 (Pack2 1188) —Vol. pap., IIp. Ed. pr.: P.Lit.Lond. 142 (Brit. Mus. inv. 1605): H.J.M. Milne, P.Lit.Lond., 1927, 121; cf. R. Pfeiffer, Philologus 92 (1937), 16–18 = Ausgew. Schr., München 1960, 39–41; H. Erbse, Sch. Iliad. II, 1971, 392–393, Sch. Iliad. VII, 1988, 266, 300–P.Oxy. 3830 (+ P.Lit. Lond. 142): M.A. Harder, P.Oxy. LVI, 1989, 37–44; cf. M. Haslam, BASP 27 (1990), 31–36; W. Luppe, Gnomon 64 (1992), 291–293, APF 39 (1993), 9–11.

 56 Haslam 1994, 44.

The Mythographus Homericus  

Θ 479: v. 003. I 447: v. 003. 004.

N 217. PSI 1000 (Pack2 2463)–Ostrakon, I–IIp. Ed. pr. P. Viereck e G. Vitelli, PSI VIII, 1927, 209; v. L. Salvadori, CCC 9 (1988), 259–262.

005.

N 301 (?), 302, 459; Ξ 319; O 229. P.Oxy. 3003—Vol. pap., IIp. Ed. pr. P.J. Parsons, P.Oxy. XLII, 1974, 15–19. Ξ 319: v. 005. O 229: v. 005.

006.

Σ [, T], Y, Φ, X, Ψ, Ω. P.Oxy. 4096—Vol. pap., IIp. Ed. pr. P. Schubert, P.Oxy. LXI, 1995

007.

T 326/32; Y 53, 147. P.Schubart 21 = P.Berol. 13930 (Pack2 1203)—Cod. perg., Vp. Ed. pr. W. Schubart, Gr. Lit. Pap., Berlin 1950, 45–7; cf. B. Snell, apud ed. pr.; R. Merkelbach, APF 16 (1956), 117–118; Turner 1977, nr. 206; Montanari 1985, II, 235–236; W. Luppe, APF 31 (1985), 5–11. Y: v. 006. Y 53, 147: v. 007.

008.

Y 147, 307, 404. P.Berol. 13282—Cod. pap., IIIp. Ed. pr. W. Müller, Forsch. u. Berichte 10 (1968), 118–119; cf. Montanari 1985. Φ, X, Ψ, Ω: v. 006.

001.

Odyssea γ 4, 91; λ 321–322, 326, 519–520, 582; μ 70, 85; ν 96, 259; ξ 327. PSI 1173 (Pack2 1209)–Cod. pap., IIIp. Ed. pr. G. Coppola, PSI Χ, 1932, 131–140; cf. R. Pfeiffer, Philologus 92

  The Mythographus Homericus (1937), 14–16 = Ausgew. Schr., München 1960, 37–39; Turner 1977, nr. 207 λ 321, 322, 326, 519–520, 582: v. 001. 002.

λ 308, 576 (?). P.Vindob.Gr. 29784 (Pack2 2447)–Vol. pap., IIIp. Ed. pr. H. Gerstinger, MPER. N.S. 1 (1932), 130–131; cf. R. Pfeiffer, Philologus 92 (1937), 16 n. 25 = Ausgew. Schr., München 1960, 39–40 n. 25. μ 70, 85: v. 001. v 96, 259: v. 001. ξ 327: v. 001.

 Ancora sul Mythographus Homericus (e l’Odissea) Il cosiddetto Mythographus Homericus (da ora in avanti MH) costituisce un interessante tema di indagine nel campo dell’esegesi omerica antica e in quello dell’erudizione mitografica. Tenendo conto dei progressi nella ricerca e delle nuove testimonianze fornite soprattutto dai papiri, qualche anno fa (nel 1995) ho riconsiderato globalmente questo «testo» e i problemi principali che lo riguardano, vale a dire essenzialmente la sua origine, la sua definizione e le sue caratteristiche (di che cosa si tratta) e anche, di conseguenza, la sua posizione storico-culturale e la valutazione delle notizie che offre. Una riconsiderazione complessiva del problema ha fatto in seguito Monique van Rossum-Steenbeek in un capitolo di un interessante libro dedicato a diversi generi di letteratura erudita nei papiri, trovando nel suo pregevole esame delle varie questioni motivi di consenso e altri di dissenso nei confronti delle idee da me sostenute.1 Non posso affrontare di nuovo in questa sede tutti gli aspetti del problema, sul quale penso di tornare in seguito in modo sistematico: ne toccherò soltanto qualcuno, non senza esprimere soddisfazione per il fatto che il MH sia oggi oggetto di ben maggiore interesse rispetto a quanto accadeva in passato e dunque anche oggetto di una comunicazione in questo colloquio. In primo luogo mi sembra opportuno ricordare qui ciò che si intende con la denominazione di MH. Essa comprende insieme materiale presente nella scoliografia omerica e in alcuni frammenti papiracei, materiale che si intende risalire a un unico testo. Il materiale contenuto negli Scholia D a Omero è rappresentato da un considerevole gruppo di ἱστορίαι, che forniscono informazioni mitografiche dove il testo omerico ne offre l’occasione. Esse sono introdotte dal lemma, che costituisce il legame preciso con il testo dell’Iliade o dell’Odissea, seguito da un resoconto mitografico più o meno esteso; infine abbiamo, almeno nella maggior parte dei casi, una sottoscrizione che attribuisce la ἱστορία a una fonte: ἡ ἱστορία παρὰ τῷ δεῖνα, (ὡς) ἱστορεῖ ὁ δεῖνα, οὕτως ὁ δεῖνα;2 quando la sottoscrizione manca, possiamo ragionevolmente pensare che sia andata perduta nella trasmissione. Questo è ciò che si intende per MH nel corpus degli scoli omerici, segnatamente della classe D: il suo precedente antico è rappresentato  1 Montanari 1995 (= questo volume, cap. 22); van Rossum-Steenbeek 1998, cap. III. Mythographus Homericus, 85–118, e testi alle 278–309. 2 Si può avere anche: ὡς γὰρ ἱστορεῖ ὁ δεῖνα e poi il racconto della ἱστορία (per es. Ἑλλάνικος in sch. D a Γ 144). https://doi.org/10.1515/9783110772371-022

  Ancora sul Mythographus Homericus (e l’Odissea) da simili ἱστορίαι mitografiche ritrovate in frammenti di papiro. Quando parliamo di MH, abbiamo dunque a che fare con l’intero insieme di ἱστορίαι mitografiche omeriche restituite dai papiri e dagli Scholia D a Omero, intendendo — come si diceva sopra — che esso provenga da uno stesso testo, che ha subito modificazioni anche cospicue nel corso della tradizione. Nel corpus degli Scholia D all’Iliade le ἱστορίαι mitografiche sono una presenza abbondante e costante; un numero sensibilmente inferiore si trova negli Scholia D all’Odissea, in accordo peraltro con la minore ricchezza in generale degli scoli odissiaci (anche tenuto conto dell’arretratezza dell’edizione disponibile e degli studi sulla tradizione manoscritta). Per quanto riguarda i papiri, la loro testimonianza è oggi assai più ricca rispetto a venti o anche solo dieci anni fa e comprende allo stato attuale una decina di pezzi, otto per l’Iliade e due (però uno è dubbio) per l’Odissea.3 Il materiale in nostro possesso ci permette ora di affermare con certezza che il MH è nato con lo scopo di provvedere un commento mitografico sistematico all’intera Iliade e all’intera Odissea. Il PSI 1173, datato al III sec. d.C., è l’unico papiro sicuro per l’Odissea ma anche uno dei reperti più interessanti ed estesi: sono conservati parti di otto fogli di un codice papiraceo, nel quale si riconoscono resti di ἱστορίαι relative ai libri III, XI, XII, XIII, XIV (uno degli otto frammenti non è identificato e potrebbe appartenere a un altro libro ancora),4 dunque un esemplare che (forse in una quarantina di fogli?) doveva senz’altro coprire l’intero poema. Il POxy 4096, datato al II sec. d.C., è anche più cospicuo per quantità di testo conservato: sono numerosi frammenti di un volumen, che interessano in tutto sette libri dell’Iliade nella parte finale del poema e restituiscono parti di ἱστορίαι dal libro XVIII al libro XXIV.5 L’insieme dei testimoni per  3 Elenco in Montanari 1995 (= questo volume, cap. 22) e van Rossum-Steenbeek 1998. Bisogna tuttavia tenere presente che esistono molti papiri contenenti resti di narrazioni mitografiche, fra i quali è possibile che si nasconda qualche frammento del MH: se il caso ha fatto in modo che nel frammento siano perduti il lemma omerico e la subscriptio e se la ἱστορία conservata è sensibilmente differente da quella presente negli Scholia D (oppure negli Scholia D non è conservata affatto), allora l’identificazione è molto difficile e soggetta a dubbio. Inoltre bisogna considerare la possibilità di inediti giacenti nelle collezioni papirologiche, non sempre del tutto collaborative nel segnalare pezzi di interesse per qualche specifica ricerca. Fortunatamente, i pezzi sicuri e pubblicati offrono abbastanza elementi di giudizio, ma l’incremento di documentazione è sempre benvenuto: l’aggiunta di POxy 4096 è un caso emblematico di un acquisto recente di grande importanza, cfr. poco oltre. 4 Non è edito nella ed. pr.: cfr. G. Coppola, PSI X, Firenze 1932, 131 n. 1, pubblicato da Luppe 1997, 13–18. Alla bibliogr. cit. in Montanari 1995 e van Rossum-Steenbeek 1998 va aggiunto anche R.L. Fowler, ZPE 97, 1993, 29–42, a proposito del fr. 2 r–v (Prokris). 5 Ed. pr. di P. Schubert, POxy LXI, London 1995. Il testo ha suscitato una profluvie di interventi (che in un caso potevano anche essere raggruppati): Haslam 1996; van Rossum-Steenbeek

Ancora sul Mythographus Homericus (e l’Odissea)  

l’Iliade oggi disponibile offre così porzioni di testo che vanno dal libro I al libro XXIV e conferma inequivocabilmente che questo genere di esegesi erudita esisteva per l’intero poema. In tutti i frammenti identificati del MH l’ordine è quello del testo omerico e ci sono sempre i lemmi che collegano direttamente la ἱστορία al verso oggetto di commento.6 Talvolta una ἱστορία prende spunto da un particolare accennato nel testo poetico, anche da una singola parola o espressione, che richiedono o almeno suscitano il desiderio di una spiegazione, di un commento di contenuto mitografico. La forma del MH risulta dunque inequivocabilmente quella del commentario mitografico, cioè un commentario con uno scopo e un contenuto specifico (continuare a chiamarlo un «manuale» di mitografia omerica non è una buona definizione e può risultare fuorviante), che aveva una certa diffusione a partire probabilmente dal I sec. d.C., almeno per quanto si può dire allo stato attuale: i frammenti conservati coprono un arco temporale che va dal sec. I–II al sec. V; nella ἱστορία a Iliade 1.400 (conservata parzialmente in POxy 418 e completa negli Scholia D) è citato Didimo,7 il che offre un valido terminus post quem. I reperti papiracei ci dicono che il MH almeno in età imperiale, prima di confluire nell’archetipo degli Scholia D, circolava isolato da altri materiali esegetici. La cosa è degna di essere notata. Nel quadro della Homeric Scholarship dall’età alessandrina all’età tardoantica non mancano certo, anzi prevalgono esempi di commentari che presentano una commistione (più o meno ricca) di contenuti diversi: discussioni sull’assetto del testo unitamente a materiale esegetico vario, glossografico-parafrastico, grammaticale, antiquario. Invece il MH compare a sé stante, in copie che contenevano soltanto la serie isolata delle ἱστορίαι: i ritrovamenti sono ormai più che sufficienti per assicurare che questa era la regola; solo negli Scholia D troveremo il MH mischiato ad altri elementi esegetici. Questo può voler dire due cose: o il MH è nato in questa forma, cioè è stato scritto espressamente come un commentario specializzato; oppure (relati-

 1996; Gärtner 1998; contributi vari di Luppe: 1995; 1996; 1996b; 1996c; 1996d; 1996e; 1996/97; 1997a. 6 Talvolta i lemmi sono in ekthesis, elemento caratteristico dell’assetto librario del commentario. In alcuni casi abbiamo anche una forma di titolo quando si passa da un canto all’altro: in PSI 1173 il fr. 5 recto conserva la fine del canto XI e l’inizio del canto ΧII, che viene segnalato con Ὀδ(υσσείας) μ (vedi il testo in appendice); forme di titolo si trovano in due luoghi di POxy 4096; un titolo è presente anche nell’ostrakon PSI 1000. 7 A proposito di Il. 1.400, in POxy 418 viene indicata l’esistenza di una variante testuale (cfr. lo scolio di Aristonico e lo sch. ex. al verso), cui è legata la ἱστορία che segue: nel papiro è perduta la parte finale, mentre lo Sch. D conserva la subscriptio a Didimo. Per la datazione vedi più avanti n. 14.

  Ancora sul Mythographus Homericus (e l’Odissea) vamente presto, anche se comunque dopo Didimo, dunque dopo l’età augustea) esso è stato prodotto estraendo e rielaborando l’elemento mitografico da commentari più ricchi e miscellanei, dove questo tipo di esegesi conviveva frammischiato al resto. Una certa analogia per questo può essere indicata (con le dovute prudenze, limitazioni e differenze) nel materiale di tipo glossograficoparafrastico, che si trova spesso frammischiato a vari altri contenuti, ma che in età imperiale comincia a presentarsi anche nella forma autonoma dei numerosi frammenti di scholia minora (definiti anche glossari):8 dunque due modalità differenti convivono in modo non esclusivo, il che peraltro appare ragionevole e persino ovvio per qualcosa di così magmatico e facilmente distribuibile come pezzi di parafrasi e glosse, lasciando aperta la possibilità di trattare questo tipo di spiegazione del testo secondo la volontà e le esigenze del prodotto che si perseguiva. La stessa distribuzione poteva caratterizzare anche l’elemento mitografìco nei prodotti di esegesi omerica. Per quanto riguarda invece specificamente il MH, come dicevamo, esso si presenta sempre in forma autonoma: il che fa pensare che sia nato autonomamente e pone decisamente il problema se sia nato isolato perché un autore lo ha scritto di sua creazione oppure perché un «compilatore» lo ha messo insieme in modo più o meno meccanico da altre fonti. Nel lavoro citato all’inizio parlavo di «compiler»,9 mentre con voluta differenziazione M. van Rossum-Steenbeek usa il termine «author»:10 «The name Mythographus Homericus is generally used to indicate the author of a collection of mythical stories...», e in nota alla parola «author»: «The term author is used and not compiler — the latter occurs in Montanari (1995, 165) and in the ed. pr. of 53 [scil. POxy 4096] p. 15 — because the collection of the MH does not consist of extracts verbally adopted from previous literature but rather of stories sharing specific characteristics, even though their contents are based on earlier writings. As most catalogues (cf. ch. 4)11 consist of names that are extracted from previous literature and do not allow us to distinguish a certain style, they will be called ‘compilations’ written by ‘compilers’». Queste considerazioni sono giuste e accettabili, corroborate anche dall’analisi dello stile delle historiae che la studiosa presenta subito dopo.12 Ma non credo che le nostre idee e concezioni

 8 Cfr. Montanari 1995, 165–168 (= questo volume, cap. 22); 1984 (= questo volume, cap. 32); 1988b (= questo volume, cap. 33); Haslam 1990; 1994, 43–44, 108–109, 114. 9 Così anche Schubert, POxy LXI 4096, 15. 10 1998, 85 e n. 1. 11 Si riferisce al cap. IV del libro cit., dedicato appunto ai Catalogues. 12 1998, 86–92.

Ancora sul Mythographus Homericus (e l’Odissea)  

siano molto lontane: io non intendevo dare al termine «compiler» un significato pregnante e un valore brutalmente meccanico, di pedissequo ricopiatore e assemblatore di pezzi. Condivido l’opinione che quelle del MH siano «stories sharing specifìc characteristics, even though their contents are based on earlier writings»: il nocciolo del problema sta nel significato che diamo al fatto che i contenuti di queste historiae siano basati su opere precedenti. Così mi esprimevo nell’articolo citato (p. 165): «I believe the MH derives from reliable sources belonging to the field of Homeric scholarship and of learned mythography. Some compiler built up the MH by borrowing from sources with mixed material, like hypomnemata or other learned products, and shaping it into a veritable specialized commentary which ... had isolated circulation. This is a by no means unusual case in which reduced, epitomized and simplified or specialized forms gained favour at the expense of the products of the great scholarly tradition, inasmuch as many of the fragments of hypomnemata or syngrammata that have come down to us on papyri dating from the early centuries of the imperial age are already in epitomized form. The MH may have assumed its present form, as shown in the papyri, just after the Augustan age». Forse non tutto è assolutamente esplicito in queste righe e credo di dover precisare meglio il mio pensiero, compresi i dubbi che ne fanno parte. Una possibilità è che il MH abbia lavorato per così dire di prima mano: leggeva il testo omerico e, quando riteneva di dover provvedere un commento mitografico, consultava autori come Erodoto, Ferecide, Acusilao, Ellanico, Demetrio di Scepsi, Prosseno Epirota, Apollodoro e diversi altri, oppure poeti come (oltre a Omero stesso) Esiodo, Euripide, Euforione, Apollonio Rodio, Licofrone e altri, traendone i contenuti che riteneva utili e scrivendo sulla base di questi le proprie ἱστορίαι, provviste di lemma omerico e di sottoscrizione per indicarne almeno la fonte principale. L’altra possibilità è che il MH abbia lavorato essenzialmente e principalmente su esistenti opere di interpretazione omerica a sua disposizione, in prevalenza hypomnemata ma anche syngrammata (e forse anche — perché no? — occasionalmente qualche altro testo, come opere di erudizione mitografico-antiquaria),13 nei quali trovava molti materiali mitografici già selezionati e rielaborati a scopo esegetico, e da questi abbia tratto i contenuti delle ἱστορίαι che scriveva, provviste naturalmente del lemma omerico e di sottoscrizione per indicarne almeno la fonte principale. In entrambi i casi non ho  13 E qualche volta poteva metterci anche la propria cultura personale, quale che fosse. In generale, credo si debba pensare al modo di composizione di queste opere con maggiore flessibilità, pensando a un processo con qualche varietà e differenziazione di metodo di lavoro da parte dell’erudito che costruisce la sua opera.

  Ancora sul Mythographus Homericus (e l’Odissea) difficoltà ad attribuirgli la definizione di «autore», anche se non si può negare che il termine assuma un significato diverso. Propendo per la seconda ipotesi, che mi sembra molto più verisimile. In suo favore parla per esempio la presenza della citazione di Didimo come fonte nella ἱστορία a Iliade 1.400 (conservata completa negli Scholia D e parzialmente, per frattura meccanica, in POxy 418). In questo caso il MH può aver avuto di fronte l’opera di Didimo e aver trovato che nei materiali forniti nessuna fonte utilizzata dal grammatico era così prevalente e decisiva rispetto all’elaborazione fattane, per cui ha deciso per una subscriptio a Didimo stesso. In questo senso va decisamente il fatto che la ἱστορία sia preceduta da e collegata alla presentazione di una variante testuale: questo si trova proprio nel papiro, che dunque doveva contenere anch’esso (nel finale perduto) la sottoscrizione a Didimo. O tutto questo è stato fatto secondariamente, cioè è né più né meno che un’aggiunta al testo originale del signor MH, oppure parla in favore dell’origine da un commentario.14 Rimando ancora al troppo citato articolo del 1995 per gli argomenti relativi alla ἱστορία a Il. 1.264 (compreso il parallelo fornito dal frammento di hypomnema POxy 1611), che considero ancora utili a questo proposito. Utili anche in relazione a un altro stadio del problema, che è il seguente: i papiri e gli Scholia D ci conservano sempre o spesso o talvolta la versione originale scritta da MH oppure abbiamo normalmente una versione già ridotta ed epitomata? Penso che ci siano indizi per ritenere che in genere la versione a nostra disposizione (anche nei papiri) sia già frutto di manipolazioni, anche se è senz’altro molto difficile dire in che misura e in che modo: credo che questo spieghi bene, tra l’altro, il fatto che talvolta l’opera appaia di livello piuttosto basso.15  14 van Rossum-Steenbeek 1998, 86, dice: «The historiae do not deal with grammar, metre, or matter of textual criticism»: questo è vero per il corpo delle ἱστορίαι stricto sensu, ma non si deve sottovalutare questo caso con il legame che vi si trova (e cfr. più avanti). Lo stesso vale anche per il problema della datazione: a p. 115, van Rossum-Steenbeek considera la citazione di Demetrio di Scepsi (ca. 200–130 a.C.) il terminus post quem, il che significa evidentemente che non considera originaria la citazione di Didimo (su cui io attiravo e attiro l’attenzione). Cfr. anche Haslam 1990, 31 n. 2: a questo proposito osservo che il fatto che il MH sia circolato isolato fin dall’inizio (come credo anch’io e ho ripetuto più volte) non è affatto in contraddizione con il fatto che il suo «autore» lo abbia scritto prendendo da precedenti commentari, quest’ultima ipotesi riguarda solo il modo in cui il signor MH ha lavorato. 15 Montanari 1995, 154–158 (= questo volume, cap. 22); cfr. van Rossum-Steenbeek 1998, 94. A proposito della citazione di Apollonio Rodio presente nella ἱστορία dello sch. D a Il. I 264 ma assente in POxy 418, è vero che io stesso dicevo (p. 153) che negli scoli si possono anche trovare materiali aggiunti posteriormente, ma mi pare difficile per una rilevante citazione di alcuni versi di Apollonio Rodio (aggiunti da un tardo scoliasta?), e continuo a credere che in questo caso sia lo sch. D a conservare la versione plenior: questo fa capire come non sia così sicuro che

Ancora sul Mythographus Homericus (e l’Odissea)  

Abbiamo menzionato sopra l’importante codice PSI 1173 e ora diremo qualcosa a proposito di questo papiro e della parte di MH che è interessata. Esso contiene brandelli di ἱστορίαι su alcuni libri dell’Odissea, una buona parte delle quali riguarda le eroine incontrate da Odisseo in Od. 11.235–332: nei frr. 2–4 rimangono parti di sei ἱστορίαι ai versi 321–322 e 326, cui segue direttamente un pezzo (nel fr. 4 verso) della ἱστορία su Euripilo, v. 519; nel fr. 5 r abbiamo la ἱστορία su Tantalo, v. 11.582, cui segue l’indicazione del passaggio al libro XII; il resto del fr. 5 e i frr. 6–7 interessano i libri XII–XIV, il fr. 1 riguarda il libro III. Un frammento papiraceo del libro XI dell’Odissea, P.Mil. Vogl. 259, datato al I–II sec. d. C., restituisce le parti iniziali dei vv. 292–310 e reca con certezza una diple davanti al v. 298 e forse la traccia di un’altra diple davanti al v. 305.16 La situazione è riassunta nel prospetto in appendice, nel quale le linee tratteggiate indicano le fratture dei frammenti e fanno molto rimpiangere che sia di PSI 1173 che di P.Mil. Vogl. 259 non si sia conservato almeno qualcosa in più, soprattutto almeno una parte corrispondente. È ben noto l’uso della diple da parte dei filologi alessandrini per indicare la discussione di diverse questioni esegetiche, anche di tipo antiquario. Per il v. 298 gli scoli segnalano un problema grammaticale sulla forma del genitivo Τυνδαρέου ο Τυνδάρεω e offrono una ἱστορία relativa a Leda: per ipotesi possiamo pensare che, nel commento esegetico, al problema testuale potesse seguire l’introduzione del materiale mitografico, cioè della ἱστορία che leggiamo nello Sch. D, così come in POxy 418 per Iliade 1.400 dalla variante di Didimo si passa alla ἱστορία mitografica. La traccia di diple davanti al v. 305 è molto dubbia: se il segno esisteva, gli scoli (almeno come possiamo leggerli, ma non sono davvero escluse sorprese da una nuova e vera edizione degli scoli all’Odissea) non danno elementi che lo giustifichino e una possibilità ipotetica è che esso si riferisse a una, oggi perduta, ἱστορία mitografica

 i papiri rappresentino per forza sempre la versione più vicina all’originale rispetto agli scholia D. Per quanto riguarda l’uso che faccio del parallelo con POxy 1611, continuo a pensare che sia utile, solo per dare un’idea di come potevano andare le cose nel processo di estrazione e epitomazione. Il quadro rende ragione anche dei tipi di rapporto che si rilevano fra il testo degli scoli e quello dei papiri (più ricchi gli uni oppure gli altri) e permette di affrontare correttamente il problema delle sottoscrizioni: problemi che qui non riprendo. A proposito della ἱστορία a Il. 9.447, van Rossum-Steenbeek si stupisce (p. 96 n. 36) che per il testo degli scoli io stampi διαλλάσσει invece di παραλλάσσει; questo deriva dal mio lavoro in corso sui manoscritti degli Scholia D in vista dell’edizione: διαλλάσσει è la lezione di CH, mentre παραλλάσσει è la lezione solo di A. Voglio aggiungere che mi pare un risultato interessante e da tenere presente quello raggiunto dalla studiosa olandese a proposito delle historiae dello Ps. Apollodoro: cfr. op. cit., 103–108. 16 Cfr. Montanari 1995, 146–148 (= questo volume, cap. 22).

  Ancora sul Mythographus Homericus (e l’Odissea) relativa al personaggio di Ifimedea, sposa di Aloeo e madre di Oto ed Efìalte.17 È un peccato che la porzione di testo omerico di P.Mil.Vogl. 259 e quella di MH di PSI 1173 non si sovrappongano neppure un poco, ad esempio per i vv. 321–326. Non c’è dubbio che l’indizio sia labile, ma certo è stimolante: e soprattutto non è l’unico. Il prospetto che si trova in appendice presenta: nella prima colonna l’indicazione dei personaggi dell’Odissea che sono o potrebbero essere oggetto di commento mitografico; nella seconda e nella terza la situazione rispettivamente di PSI 1173 e degli Scholia D all’Odissea (sulla base dell’edizione disponibile); la terza colonna serve solo per la piccola porzione di testo interessata da P.Mil. Vogl. 259. Il segno XXX segnala la presenza di una ἱστορία; sotto è indicata la presenza o meno della sottoscrizione, con il nome dell’autore-fonte. Il testo di PSI 1173 relativo al libro XI comincia con il fr. 2 (il fr. 1 riguarda il libro ΙII) e precisamente con il v. 321, la ἱστορία su Fedra: per i versi precedenti, a partire dal v. 287, abbiamo solo gli scoli e non il papiro (mentre abbiamo già parlato sopra del P.Mil. Vogl. 259). Qualche considerazione interessante la possiamo fare se prendiamo la porzione di testo che va da 11.321 a 14.327. Per i vv. 321–520 il parallelo fra papiro e scoli è perfetto: entrambi contengono ἱστορίαι sugli stessi personaggi; entrambi da Erifile del v. 326 passano a Euripilo del v. 519 (per il papiro questo è assicurato dalla continuità del fr. 4), saltando le eventuali «possibilità» offerte dalla menzione nei vv. 387–506 di Agamennone, Achille, Patroclo, Antiloco, Aiace, Achille di nuovo, Neottolemo: probabilmente in quanto personaggi ben noti della saga e forse già trattati in precedenza, a quanto pare (salvo perdite anche negli scoli) essi non hanno stimolato il desiderio di una ἱστορία mitografica. Per quanto riguarda invece la porzione di testo fra 11.519 (Euripilo) e 11.582 (Tantalo), negli Scholia D non troviamo ἱστορίαι ma non possiamo sapere con certezza se il papiro avesse qualcosa perché si passa dal fr. 4 al fr. 5 e non è dato sapere cosa poteva esserci nel mezzo: è probabile che procedessero in parallelo, come accade sicuramente fino alla fine del libro XI, dato che nel fr. 5 r del papiro si legge addirittura esplicitamente il passaggio al libro ΧII e anche negli scoli non si trovano ἱστορίαι fra 11.582 (Tantalo) e 12.69 (la nave Argo). Per vero dire, nella porzione di testo che va da 11.519 (Euripilo) fino alla fine del libro XI alcune omissioni sono ben comprensibili (Aiace e Eracle, per le ragioni dette sopra a proposito dei personaggi), mentre altre (per esempio Tizio e Sisifo) lasciano aperta la domanda se una ἱστορία non ci fosse per nulla e in questo caso quale fosse il criterio di lavoro (se ce n’era uno). Poi il papiro e gli scoli sembrano procedere in parallelo: è possibile che il fr. 6 e il fr. 7  17 Cfr. il caso di Od. 1.276 che ho considerato in 1995a, 35.

Ancora sul Mythographus Homericus (e l’Odissea)  

di PSI appartengano a due fogli consecutivi, considerando anche quello che si trova negli scoli per la parte di testo omerico fra 12.85 e 13.96. Proseguendo con il fr. 7 fino a 14.327, l’unico dubbio, indicato per scrupolo nel prospetto, riguarda il v. 13.408 (Korax), sicuramente saltato nel papiro, mentre negli scoli si legge in effetti un commento che però nella forma pubblicata non ha esattamente l’aspetto di una ἱστορία e dunque non è detto che appartenga al MH. Per quanto riguarda le sottoscrizioni, si osserva facilmente che non ci sono divergenze fra papiro e scoli: quando l’indicazione di fonte è conservata in entrambi, la coincidenza è totale, come peraltro accade con tutti gli altri frammenti di MH su papiro, che riguardano l’Iliade. Ho sottolineato il valore di questo fatto nell’articolo del 1995 e non torno qui sul problema.18 In appendice si trova anche la presentazione su due colonne di una parte del testo di PSI 1173 (a sinistra) e degli scoli corrispondenti (a destra), non soltanto per comodità del lettore. Ho voluto presentare in questo modo il materiale delle due versioni del MH con lo scopo di sottolineare il parallelo fra i papiri e gli scoli nella forma specifica e nell’arrangiamento di un commentario al testo omerico. Avendo la pazienza di fare questo per tutte le porzioni di MH conservate dai papiri, si guadagnerà in concretezza nel vedere che cosa era il MH, malgrado le deformazioni che può avere subito nei diversi rivoli della sua trasmissione.

Appendice Odissea

ἱστορίαι PSI 

ἱστορίαι Sch. D

XI  Però

XXX

 Ificlo

XXX

 Leda

XXX

 Iphimedeia

P.Mil. Vogl. 

diple diple?

 Phaidra

XXX fr.  r Asklepiades

XXX Asklepiades

Prokris

XXX fr.  r–v perduta

XXX Pherekydes

– Ariadne

XXX fr.  r perduta

XXX Pherekydes

 Maira

XXX fr.  v Pherekydes

XXX Pherekydes

 18 Cfr. Montanari 1995, 150 sgg (= questo volume, cap. 22).

  Ancora sul Mythographus Homericus (e l’Odissea)

Odissea

ἱστορίαι PSI 

ἱστορίαι Sch. D

Klymene

XXX fr. v–r Hesiodos

XXX Hesiodos

Eriphyle

XXX fr.  r Asklepiades

XXX Asklepiades

– Eurypylos

XXX fr.  v perduta

XXX Akusilaos

 Aias

?

 Minos

?

 Orion

?

 Tityos

?

 Tantalos

XXX fr.  r Asklepiades

XXX Asklepiades

XII – nave Argo

XXX fr.  r–v / fr.  r perduta

XXX Asklepiades

XII  Skylla

XXX fr.  v perduta

XXX Dionysios

XIII  Phorkys

XXX fr.  r perduta

XXX Herodoros

XIII  Idomeneus

XXX fr.  v Lykophron

XXX Lykophron

 Agamemnon  Achilleus Patroklos Antilochos Aias  Achilleus  Neoptolemos

 Sisyphos  Herakles

XIII  Korax XIV  Dodone

? assente XXX fr.  v Proxenos

XXX Proxenos

P.Mil. Vogl. 

Ancora sul Mythographus Homericus (e l’Odissea)  

PSI 

Scholia Odyss.

Fr.  Recto ———— ]μεταλα[ ]. ντης πελ[ ]δε δό̣ ντων[ ].ρουντω[ εἰ]ς τὴν Πύλ̣ [ον ]νεπομεν̣[· ].ε Χλωρι[ ] ἐξ ἧς αὐτῶι κ[ ]το. ἡ δὲ ἱστορ[ία παρὰ Ἑλλανί]κωι. [ε̣ἴ τε καὶ ἐν π]ελάγει με[τὰ κύμα [σιν Ἀμφιτρί]της. ἐν τῆ[ι Νάξωι ———— Verso ———— ].τη...[ ].ος Ὀδυσσέω̣[ς ]υσιως ουχ̣[ ]ιοςθαλασσ[ ]υζην αυτ[ ]νεψαι την .[ ]. Τηλέμαχος κ[ ]ι̣ ς ὑβρίσαι τον̣ [ ]ο̣ υ μητέρα α[ ]τῶι ἔργωι δὲ[ ]ευσεν οκα[ ]..[ ———— Fr  Recto ———— [φαν]ερᾶς γεν[ο]μένη[ς τῆς διαβο] [λ]ῆς ἀναρ̣τ̣ η̣θῆνα[ι. ἡ δὲ ἱστορί] α] παρὰ Ἀσκληπιάδηι. [Πρ]ό̣ κ̣ρ̣ιν. Κέ{λα}φαλ[ο]ς ὁ Δηϊ[ονέως] [γή]μας Πρόκριν τὴν Ἐρεχ[θέως] [κα]τ̣ ῴκει ἐν Θορικῶι. θέλων [δὲ] [τῆς γ]υναικὸς ἀποπειρᾶσθαι ἀ [πο]δημεῖ ἐπ’ ἔτη ὀκτὼ καταλ[ι] [πὼ]ν αὐτὴν ἔτ⟦ε⟧ι νύμφην̣ . [ἔπει]

. οἱ δὲ Πύλον Νηλῆοϲ] Νηλεὺϲ μαχησάμενος μετὰ Πελίου ἐξ Ἰωλκοῦ ἧκεν εἰϲ Μεϲϲήνην καὶ τὴν Πύλον ἔκτιϲε, Μεϲϲηνίων χώραν παραϲχόντων. ἱϲτορεῖ Ἑλλάνικοϲ. . κύμαϲιν Ἀμφιτρίτηϲ] τῆϲ θαλάϲϲηϲ. λέγεται δὲ οὕτω καὶ ἡ Ποϲειδῶνοϲ γυνή. Η.Μ. Ἄλλωϲ. Ἐν Νάξῳ τὴν Ἀμφιτρίτην χορεύουϲαν ἰδὼν Ποϲειδῶν ἥρπαϲεν· ὅθεν ὑπὸ τῶν ἐγχωρίων Ποϲειδωνία ὠνομάϲθη ἡ θεὸϲ, ὡϲ καὶ ἡ Ἥρα Διώνη παρὰ Δωδωναίοιϲ, ὡϲ Ἀπολλόδωροϲ.

. Φαίδρην τε] Θηϲεὺϲ ὁ Αἰγέωϲ ἔχων παῖδα Ἱππόλυτον ..... ἐκεῖνον μὲν οὖν λέγουϲι τὸ ἅρμα γυμνάζοντα, παραϕανέντοϲ ἐξαίϕνηϲ ἀπὸ τῆϲ θαλάϲϲηϲ ταύρου καὶ ταραχθέντων τῶν ἵππων, ἑλκόμενον ἀποθανεῖν, τὴν δὲ Φαίδραν ϕανερᾶϲ γενομένηϲ τῆϲ διαβολῆϲ ἀπάγξαϲθαι. ἡ δὲ ἱϲτορία παρὰ Ἀϲκληπιάδῃ. V Πρόκριν τε] Πρόκριν Ἐρεχθέωϲ ὁ Κέϕαλοϲ ἀνὴρ ἀνεῖλεν ὡϲ ἐπὶ θηρὸϲ τὸ δόρυ καθείϲ. Η. Πρόκριϲ Ἐρεχθέωϲ θυγάτηρ Κέϕαλον τὸν ἑαυτῆϲ ἄνδρα ϲυνεχῶϲ ἐπὶ θήραν ἐπιόντα ἐτήρει λαθραίωϲ διὰ ζηλοτυπίαν, ὁ δὲ Κέϕαλοϲ, νομίϲαϲ εἶναι θηρίον ἀπὸ τοῦ τῆϲ ὕληϲ κτύπου, ἔλαθεν αὐτὴν Πρόκριν κατακοντίϲαϲ. V. Κέϕαλοϲ ὁ

  Ancora sul Mythographus Homericus (e l’Odissea)

PSI 

Scholia Odyss.

[τα κατ]α̣κοσμήσας ἑαυτὸν κ[αὶ] [ἀλλ]ο̣ ι̣ ώ[σα]ς ἔρχεται ⟦ἔρχ[εται] ⟧ [εἰς τὴν οἰκία]ν ἔχων κ[όσμον] ]..[ ————

Verso ———— [........]....[..] π̣ρ̣οσα[. . .] [........]αυρα γένοιτο καὶ τ[οῦ] [το μόνον σ]υνειδέναι αὐτῷ· ἡ [δὲ] [Πρόκρις] ἀκούσασα ἔρχεται ε[ἰς] [ταύτ]ην ` [τ]ὴν ´ κορυφὴν καὶ κατα[κρύπτε] [τ]αι καὶ κατακούουσα αὐτο[ῦ] τοῦτο λέγοντος προστρέχ[ει] ὡς αὐτόν. ὀ δὲ Κέφαλος ἐξ[αίφ] νης ἰδὼν ταύτην ἔξ[ω αὑτοῦ] [γί]νεται καὶ ὡς εἶδε βάλλ[ει ἀκον] [τί]ῳ τὴν Πρόκριν κ[αὶ κτεί] [νε]ι. μεταπεμψάμεν[ος δὲ τὸν] [Ἐρεχ]θέα τα.[ ————

Δηϊονέωϲ γήμαϲ Πρόκριν τὴν Ἐρεχθέωε ἑν τῇ Θυριέων κατῴκει. θέλων δὲ τῆϲ γυναικὸϲ ἀποπειρᾶϲθαι ἔρχεται εἰϲ ἀποδημίαν ἐπὶ ἔτη ὀκτὼ καταλιπὼν αὐτὴν ἔτι νύμϕην οὖϲαν. ἔπειτα κατακοϲμήϲαϲ καὶ ἀλλοειδῆ ἑαυτὸν ποιήϲαϲ ἔρχεται εἰϲ τὴν οἰκίαν ἔχων κόσμον, καί πείθει τὴν Πρόκριν δέξαϲθαι τοῦτο καὶ ϲυμμιγῆναι αὐτῷ. ἡ δὲ Πρόκριϲ ἐποϕθαλμίϲαϲα τῷ κόϲμῳ καὶ τὸν Κέϕαλον ὁρῶϲα κάρτα καλὸν ϲυγκοιμᾶται αὐτῷ. ἐκϕήναϲ δὲ ἑαυτὸν ὁ Κέϕαλοϲ αἰτιᾶται τὴν Πρόκριν. οὐ μὴν ἀλλὰ καταλλαγεὶϲ ἐξέρχεται ἐπὶ θήραν, πυκνῶϲ δὲ αὐτοῦ τοῦτο δρῶντοϲ ὑπώπτευϲεν ἡ Πρόκριϲ ὅτι μίϲγεται γυναικὶ ἑτέρᾳ. προϲκαλεϲαμένη οὖν τὸν οἰκέτην ἔλεγεν εἰ ϲύνοιδεν. ὁ δὲ θεράπων ἔϕη τὸν Κέϕαλον ἰδεῖν καί τινοϲ ὄρουϲ κορυϕήν, καὶ λέγειν ϲυχνῶϲ, ὦ νεϕέλη παραγενοῦ, καὶ τοῦτο μόνον ϲυνειδέναι. ἡ δὲ Πρόκριϲ ἀκούϲαϲα ἔρχεται εἰϲ ταύτην τὴν κορυϕὴν καὶ κατακρύπτεται. καὶ τὸ αὐτὸ λέγοντοϲ αὐτοῦ πυθομένη προϲτρέχει πρὸϲ αὐτόν. ὁ δὲ Κἐϕαλοϲ ἰδὼν αὐτὴν αἰϕνιδίωϲ ἔξω ἑαυτοῦ γίνεται, καὶ ὥϲπερ εἶχε βάλλει μετὰ χεῖρα ἀκοντίῳ τὴν Πρόκριν καὶ κτείνει. μεταπεμψάμενοϲ δὲ τὸν Ἐρεχθέα θάπτει πολυτελῶϲ αὐτήν. ἡ δὲ ἱϲτορία παρὰ Φερεκύδη ἐν τῇ ἑβδόμῃ.

Fr.  Recto ———— λάβη̣ κρατήσαντά τε τῆ [ς κεφα] [λ]ῆς καὶ τῶν τριχῶν σφ[άξαι τῷ] [Π]οσιδῶνι ο κ.....[.].υσαι[...] [.].ελίττοντα. ὁ δὲ [Θησ]εὺς [ἀκού] [σ]ας ταῦτα λαμβάνε[ι τ]ὴν [Ἀριά] [δ]νην ἀφικνεῖταί ` τε ´ εἰ[ς Ἀθή] [ν]ας. συντόμως δὲ διαν̣ [αστὰς] [τ]ο̣ ῦτο ποιεῖ· ἀφειμέν[ης τῆς] [Ἀρ]ιάδνης, Ἀφροδίτη ταύ[τῃ πα] [ρασ]τ̣ ᾶσα θαρρεῖν παρῄν[ει· Διόνυ] [σo]ν̣ γὰρ αὐτῇ μίσγεσθα[ι καὶ εὐ]

.– καλήν τ᾽ Ἀριάδνην] Ἀριάδνη Μίνωοϲ θυγάτηρ, γυνὴ Θηϲέωϲ, ἐξ ἧϲ Δημοϕῶν καὶ Ἀκάμαϲ. Q.V. Μίνωοϲ ὀλοόϕρονοϲ] Θηϲεὺϲ ὁ Αἰγέωϲ λαχὼν μετὰ τῶν ἠϊθέων εἰϲ Κρήτην πλεῖ τῷ Μινωταύρῳ παρατεθηϲόμενοϲ πρὸϲ ἀναίρεϲιν. ἀϕικομένου δὲ αὐτοῦ ἐρωτικώϲ πρὸϲ αὐτόν διατεθεῖσα ἡ τοῦ Μίνωοϲ θυγάτηρ Ἀριάδνη δίδωϲιν ἀγαθίδα μίτου λαβοῦϲα παρὰ Δαιδάλου τοῦ τέκτονοϲ, καὶ διδάσκει αὐτὸν, ἐπειδὰν εἰϲέλθῃ, τὴν ἀρχὴν τῆϲ ἀγαθίδοϲ ἐκδῆϲαι περὶ τὸν ζυγὸν τῆϲ ἄνω θύραϲ καὶ ἀνελίϲϲοντα ἰέναι μέχριϲ ἂν ἀϕίκηται εἰϲ τὸν μυχὸν, καὶ ἐὰν αὐτὸν καθεύδοντα μάρψῃ, κρατήϲαντα τῶν τριχῶν τῆϲ κεϕαλῆϲ τῷ Ποϲειδῶνι θῦϲαι καὶ ἀπιέναι ὀπίϲω ἀνελίϲϲοντα τὴν ἀγαθίδα. ὁ δὲ Θηϲεὺϲ λαβὼν τὴν Ἀριάδνην εἰϲ τὴν ναῦν

Ancora sul Mythographus Homericus (e l’Odissea)  

PSI 

Scholia Odyss.

[κλ]ε̣ ῆ γενέσθαι ἔλεγεν. [ὅθεν ὁ] [θεὸς] φανε[ὶς μ]ίσγεται [αὐτῇ] ————

ἐμβάλλεται καὶ τοὺϲ ἠϊθέουϲ καὶ παρθένουϲ οὐδέπω ϕθάϲανταϲ τῷ Μινωταύρῳ παρατεθῆναι. καὶ ταῦτα ποιήϲαϲ νυκτόϲ μέϲηϲ ἀποπλεῖ. προϲορμίϲαϲ δὲ τῇ Δίᾳ νήϲῳ ἐκβὰϲ ἐπὶ τῆϲ ἠϊόνοϲ μετακοιμᾶται. καὶ αὐτῷ ἡ Ἀθηνᾶ παραϲτᾶϲα κελεύει τὴν Ἀριάδνην ἐᾶν καὶ ἀϕικνεῖϲθαι εἰϲ Ἀθήναϲ. ϲυντόμωϲ δὲ διαναϲτὰϲ ποιεῖ τοῦτο. κατολοϕυρομένηϲ δὲ τῆϲ Ἀριάδνηϲ ἡ Ἀϕροδίτη ἐπιϕανεῖϲα θαρρεῖν αὐτῇ παραινεῖ· Διονύσου γὰρ ἔϲεϲθαι γυναῖκα καὶ εὐκλεῆ γενέϲθαι. ὅθεν ὁ θεὸϲ ἐπιϕανεὶϲ μίϲγεται αὐτῇ καὶ δωρεῖται ϲτέϕανον αὐτῇ χρυϲοῦν, ὃν αὖθιϲ οἱ θεοὶ κατηϲτέριϲαν τῇ τοῦ Διονύϲου χάριτι. ἀναιρεθῆναι δὲ αὐτὴν ὑπ᾽ Ἀρτέμιδοϲ προεμένην τὴν παρθενίαν, ἡ ἱϲτορία παρὰ Φερεκύδῃ. V.

Verso ———— [ἡγουμέ]νη δ[ὲ] π̣ε̣ρ̣[ὶ πλείονος τὴν] [παρθε]νείαν εἵπετο τῇ Ἀρ̣[τέ] [μιδι ἐπ]ὶ τὰς κυνηγεσίας. τα[ύ] [της] ἐρα[σθεὶς] Ζεὺς ἀφικνε[ῖ] [ται] ὡς αὐτὴν καὶ λαθὼν φ[θεί] [ρει· ἔγκ]υος δὲ γενηθεῖσα τίκ̣[τει] [Λοκρὸ]ν τὸ ὄνομα, ὃς Θήβας [με] [τὰ Ἀ]μφίονος καὶ Ζήθου κ̣[τί] [ζει. τὴ]ν δὲ Μαῖραν λέγεται [ὑπὸ] [Ἀρτέμι]δος τοξευθῆναι δι[ὰ τὸ] [μηκέτι] παραγείνεσθαι εἰς [τὰ κυ] [νηγέσι]α̣. ἡ δὲ ἱστορία παρὰ [Φε] [ρεκύ]δ̣ει Kλυ[μένην ————

Fr.  Recto ———— [δὲ] α̣ὐτὴν τὴν Κλ[υμένην προ] [γαμη[θῆναί φασιν ὑφ᾽ [Ἡλίου ἐξ ἧς]

. Μαῖράν τε] Φερεκύδηϲ ἐκ Μαίραϲ καὶ Διὸϲ τὸν Λοκρὸν εἶναι ϕηϲί. Gl. Η. παρθένοϲ ἀπέθανεν ἡ Μαῖρα. ἔϲτι δὲ Προίτου καὶ Ἀντείαϲ θυγάτηρ· ἡ δὲ Κλυμένη Μινύου θυγάτηρ καὶ Εὐρυάληϲ. Η.Τ. Μαῖρα Προίτου καὶ Ἀντείαϲ θυγάτηρ. ἐτελεύτηϲε δὲ παρθένοϲ. V. Μαῖρα ἡ Προίτου τοῦ Θερϲάνδρου θυγάτηρ καὶ Ἀντείαϲ τῆϲ Ἀμϕιάνακτοϲ ἐγένετο κάλλει διαπρεπεϲτάτη. ἡγουμένη δὲ περὶ πλείονοϲ τὴν παρθενίαν εἵπετο τῇ Ἀρτέμιδι ἐπὶ τὰ κυνηγέϲια. ταύτηϲ ἐραϲθεὶϲ Ζεὺϲ ἀϕικνεῖται ὡϲ αὐτὴν καὶ λανθάνων διαϕθείρει. ἡ δὲ ἔγκυοϲ γενομένη τίκτει Λοκρὸν τὸ ὄνομα παῖδα, ὃϲ Θήβαϲ μετ᾽ Ἀμφίονοϲ καὶ Ζήθου οἰκίζει. τὴν δὲ Μαῖραν λέγεται τοξευθῆναι ὑπὸ Ἀρτέμιδοϲ διὰ τὸ μηκέτι παραγίνεϲθαι εἰϲ τὰ κυνηγέϲια. ἡ δὲ ἱϲτορία παρὰ Φερεκύδη. V. Κλυμένην τε ἴδον] Κλυμένη Μινύου θυγάτηρ. V. Κλυμένη Μινύου τοῦ Ποϲειδῶνοϲ καὶ Εὐρυανάϲϲηϲ τῆϲ ὑπέρϕαντοϲ γαμηθεῖϲα Φυλάκῳ τῷ Δηΐονοϲ Ἴϕικλον τίκτει ποδώκη παῖδα. τοῦτον λέγεται διὰ τὴν τῶν ποδῶν ἀρετὴν ϲυναμιλλᾶϲθαι τοῖϲ ἀνέμοιϲ ἐπί τε τῶν ἀϲταχύων διέρχεϲθαι, καὶ διὰ τοῦ τάχουϲ τὴν κουϕότητα μὴ περικλᾶν τοὺϲ ἀθέραϲ. ἔνιοι δὲ αὐτὴν τὴν Κλυμένην προγαμηθῆναί ϕαϲιν Ἡλίῳ, ἐξ ἧϲ Φαέθων ἐγένετο παῖϲ. ἡ δὲ ἱϲτορία παρὰ

  Ancora sul Mythographus Homericus (e l’Odissea)

PSI 

Scholia Odyss.

[Φαέθ]ων ἐγένε[το] παῖς. ἡ δὲ [ἱστο] [ρία] παρ᾽ Ἡσιόδωι. στυγερήν γ᾽ Ἐριφύλην. Ἀμφ[ιάρα] ος γήμας Ἐριφύλην τὴν Ταλ[αοῦ καὶ] διενεχθεὶς ὑπέρ τινων π̣ρ̣[ὸς Ἄ] δραστον, καὶ π[ά]λιν διαλυθ[εὶς ὤ] μνυεν ὑπὲρ ὧν ἂν διαφ[έρων] τ̣ αι πρὸς ἀλλήλους αὐτός τε κ[αὶ Ἀδρα] στος ἐπιτρέψειν Ἐριφύλη̣[ν κρί] νειν καὶ π❬ε❭ισθήσεσθαι αὐτῇ· [μετὰ] [δὲ τα]ῦ̣ τα̣ γενομ̣[έ]νης 〈τῆς〉 ἐπὶ Θή[βας] [στρατείας ὁ μὲν Ἀ]μφιάρα[ος ἀπέ] [τρεψε τοὺς Ἀργείους] καὶ τὸ[ν ————

Ἡϲιόδῳ. H.Q.V.

Verso ———— πρ]ὶν ἀποκτεῖν[α]ι τ[ὴν μητέ] [ρα.] ταῦτα δὲ πάντα δρᾶ[σαι λέ] [γ]ε̣ ται ⟦δε⟧ {παρα} τὸν Ἀλκ[μαίωνα] [καὶ] διὰ τὴν μητροκτονίαν [μα]νῆναι. {δε} τοὺς` δὲ ´ θεοὺς ἀκού [σα]ντας τῆς νήσου μεταλλά [ξαι] διὰ τὸ ὁσίως ἐπαμύνον [τα] τῷ πατρὶ τὴν μητέρα ἀπο [κτε]ῖναι. ἡ δ’ ἱστορία παρὰ Ἀσκλη [πι]άδηι. [τὸν] Τηλεφίδην. Εὐρύπυλος ὁ [Τηλ]έφου καὶ Ἀστυόχης Ἡρακλέ[α πά]π’πον εχ̣[ ]ἀρχὴ[ν ————

Fr.  Recto ———— [συνδι]ατρίβων τοῖς θ[εοῖς καὶ συν] [ε]στιώμενος ἀπλήστ[ως διετέθη.] κλέψας γὰρ τὸ νέκταρ κα[ὶ τὴν ἀμ]

.–. ϲτυγερήν τ᾽ Ἐριφύλην – τιμήεντα] Ἀμϕιάραοϲ ὁ Οἰκλέουϲ γήμαϲ Ἐριφύλην τὴν Ταλαοῦ καὶ διενεχθεὶϲ ὑπέρ τινων πρὸϲ Ἄδραϲτον, καὶ πάλιν διαλυθεὶϲ ὁρκούμενοϲ ὡμολόγηϲαν ὑπὲρ ὧν ἂν διαϕέρωνται πρόϲ ἀλλήλουϲ αὐτόϲ τε καὶ Ἄδραϲτοϲ ἐπιτρέψειν Ἐριϕύλην κρίνειν καὶ πείθεϲθαι αὐτῇ. μετὰ δὲ ταῦτα γινομένηϲ τῆϲ ἐπὶ Θήβαϲ ϲτρατείαϲ ὁ μὲν Ἀμϕιάραοϲ ἀπέτρεπε τοὺϲ Ἀργείουϲ καὶ τὸν ἐϲόμενον ὄλεθρον προεμαντεύετο. λαβοῦϲα δὲ ἡ Ἐριϕύλη τὸν ὅρμον παρὰ τοῦ Πολυνείκουϲ τὸν τῆϲ Ἀρμονίαϲ, προέθετο τοῖϲ περί τὸν Ἄδραϲτον βιαζομένοιϲ τὸν Ἀμϕιάραον ἰδόντα τὴν τῶν δώρων ὑποδοχὴν καὶ πολλὰ τὴν Ἐριϕύλην αἰτιαϲάμενον, αὐτὸν μὲν ἐξορμῆϲαι πρὸϲ τὴν ϲτρατείαν, Ἀλκμαίωνι δὲ προϲτάξαι μὴ πρότερον μετὰ τῶν ἐπιγόνων ἐπὶ Θήβαϲ πορεύεϲθαι πρὶν ἀποκτεῖναι τὴν μητέρα. ταῦτα δὲ πάντα δρᾶϲαι λέγεται τὸν Ἀλκμαίωνα καὶ διὰ τὴν μητροκτονίαν μανῆναι. τοὺϲ δὲ θεοὺϲ ἀπολῦϲαι τῆϲ νόϲου αὐτὸν διὰ τὸ ὁϲίωϲ ἐπαμύνοντα τῷ πατρὶ τὴν μητέρα κατακτεῖναι. ἡ ἱϲτορία παρὰ Ἀϲκληπιάδῃ. V. .— Εὐρύπυλον] Εὐρύπυλοϲ ὁ Ἀϲτυόχηϲ καὶ Τηλέϕου τοῦ Ἡρακλέουϲ παῖϲ λαχὼν τὴν πατρῴαν ἀρχὴν τῆϲ Μυϲίαϲ προΐϲτατο. πυθόμενοϲ δὲ Πρίαμοϲ περὶ τῆϲ τούτου δυνάμεωϲ ἔπεμψεν ὡϲ αὐτὸν ἵνα παραγένηται ϲύμμαχοϲ. εἰπόντοϲ δὲ αὐτοῦ ὡϲ οὐκ ἐξῆν αὐτῷ διὰ τὴν μητέρα, ἔπεμψεν ὁ Πρίαμοϲ τῇ μητρὶ αὐτοῦ δῶρον Ἀϲτυόχῃ. χρυϲῆν ἄμπελον· ἡ δὲ λαβοῦϲα τὴν ἄμπελον τὸν υἱὸν ἔπεμψεν ἐπὶ ϲτρατείαν, ὃν Νεοπτόλεμοϲ ὁ τοῦ Ἀχιλλέωϲ υἱὸϲ ἀναιρεῖ. ἡ δὲ ἱϲτορία παρὰ Ἀκουϲιλάωι. Q.V. .. Τάνταλον εἴϲιδον κρατερὰ ἄλγε’ ἔχοντα] Τάνταλοϲ Διὸϲ καὶ Πλουτοῦϲ ϲυνδιατρίβων τοῖϲ νέοιϲ καὶ ϲυνεϲτιώμενοϲ αὐτοῖϲ ἀπλήϲτωϲ διετέθη.

Ancora sul Mythographus Homericus (e l’Odissea)  

PSI 

Scholia Odyss.

[β]ροσίαν, οὐκ ἐξὸν αὐτῶι, ἔδω[κε] τ[ο]ῖς φίλοις· ἀγανακτήσαντα [δὲ τὸν] Δία ἐπαρτῆσαι αὐτῷ τὴν κ[......] τιμωρίαν, ἐκβαλόντα τῶ[ν.....] καὶ τὴν Σίπυλον ἐπικαταστήσαν τα, ἔνθα ἐκηδεύετο. ἡ δ’ ἱ[στορία] [π]αρὰ Ἀσκληπιάδηι. Ὀδυσσείας μ [Ἀ]ργὡ πασιμέλουσα παρ’ Ἀ[ιήταο.] [Τ]υρὼ ἡ Σαλμωνέως ἔχ[ουσα δύο] [παῖδ]ας ἐκ Ποσιδῶνο[ς, Νηλέα τε] [καὶ Πελία]ν, ἐπέγημε Κ[ρηθέα ————

κλέψαϲ γὰρ τὸ νέκταρ καὶ τὴν ἀμβροϲίαν, οὐκ ἐξόν αὐτῷ, ἔδωκε τοῖϲ ὁμήλιξιν. ἐϕ’ οἷϲ ἀγανακτήϲοντα τὸν Δία ἐκβαλεῖν αὐτὸν τῆϲ ἐν οὐρανῷ διαίτηϲ καὶ ἐξαρτῆϲαι ἐπ’ ὄρουϲ ὑψηλοῦ ἐκδεδεμένον τῶν χειρῶν, καὶ τὴν Cίπυλον, ἔνθα ἐκεκήδευτο, ἀνατρέψαι. ἡ ἱϲτορία παρὰ Ἀϲκληπιάδῃ. V.

Verso ————

]ια̣δ[ πέμπ]ει παρὰ Χ[είρων]α. [παρ[.]] [τραφ]ε̣ ὶς δὲ παρὰ τούτωι καὶ ἠ[ΐ] [θε]ος γενόμενος ἦλθεν εἰς [Ἰω]λκὸν ἀπαιτῶν τὴν πα [τρώι]αν ἀρχὴν τὸν Πελίαν. ὁ [δ]ὲ [ἔφασ]κεν αὐτὸν δεῖν πρότε [ρον δια]κομίσαι τὸ χρυσοῦν δέ [ρος ἐ]κ Κόλχων. ἀκούσας δὲ τα[ῦ] [τα ὁ] Ἰάσων λέγει τῷ Χείρωνι· [καὶ α]ὐτῷ (ὁ) Χείρων συνεξέ [πεμ]ψεν τοὺς ἠϊθέους. κατα [σκευάζ]ει δ’ ἡ Ἀθηνᾶ τὴν Ἀρ [γώ. πλέ]οντες δὲ ἀφι[κνοῦν] [ται πρὸς τὰς Πλα]γ'κτὰ[ς καλου] [μένας πέτρας ————

Fr.  Recto ———— Φι]νεύς. [τοῦτον δὲ τὸν] [Φιν]έα ἰδόντε[ς οἱ περὶ τὸν] [Ἰάσο]να παρεκάλο[υν ὑποθέσ] [θαι π]ῶς διαπλεύσωσι τὰ[ς Πλαγ]

.–. οἴη δὴ κείνη γ’ ἐπέπλω ποντόποροϲ ναῦϲ Ἀργὼ παϲιμέλουϲα] Τυρὼ ἡ Ϲαλμωνέωϲ ἔχουϲα δύο παῖδαϲ ἐκ Ποϲειδῶνοϲ, Νηλέα τε καὶ Πελίαν, ἔγημε Κρηθέα. καὶ ἴϲχει παῖδαϲ ἐξ αὐτοῦ τρεῖϲ, Αἴϲονα καὶ Φέρητα καὶ Ἀμυθάονα. Αἴϲονοϲ δὲ καὶ Πολυμήλαϲ καθ’ Ἡϲίοδον γίνεται Ἰάϲων, κατὰ δὲ Φερεκύδην ἐξ Ἀλκιμέδηϲ. τελευτῶν δὲ οὗτοϲ καταλείπει τοῦ παιδὸϲ ἐπίτροπον τὸν ἀδελϕὸν Πελίαν, ἐγχειρίϲαϲ αὐτῷ καὶ τὴν βαϲιλείαν, ἵνα αὐξηθέντι τῷ υἱῷ παράϲχῃ. ἡ δὲ τοῦ Ἰάϲονοϲ μήτηρ Ἀλκιμέδη δείϲαϲα δίδωϲιν αὐτὸν τρέφεϲθαι Χείρωνι τῷ Κενταύρῳ. τραϕεὶϲ δὲ καὶ ἡβήϲαϲ ἔρχεται εἰϲ Ἰωλκὸν ἀπαιτῶν τὴν πατρῴαν ἀρχὴν τὸν Πελίαν. ὁ δὲ ἔϕαϲκε χρῆναι αὐτὸν πρότερον διακομίϲαι τὸ χρυϲοῦν δέροϲ ἀπὸ Κόλχων καὶ τοὺϲ πυριπνόουϲ ἀνελεῖν ταύρουϲ. ἀκούϲαϲ δὲ ταῦτα ὁ Ἰάϲων λέγει τῷ Χείρωνι. καὶ αὐτῷ ὁ Χείρων ϲυνεκπέμπει τοὺϲ ἠϊθέουϲ. καταϲκευάζει δὲ ἡ Ἀθηνᾶ τὴν Ἀργώ, πλέοντεϲ δὲ ἀϕικνοῦνται ἐπὶ τὴν τῶν Βιθυνῶν χώραν, καὶ ὁρῶϲι τὸν Φινέα πηρὸν διὰ ταύτην τὴν αἰτίαν. ἔχων γὰρ παῖδαϲ ἐκ Κλεοπάτραϲ τῆϲ Βορέου καὶ ἐπιγήμαϲ Εὐρυτίαν δίδωϲιν αυτοὺϲ τῇ μητρυιᾷ διαβληθένταϲ πρὸϲ ἀναίρεϲιν. Ζεὺϲ δὲ χαλεπήναϲ λέγει αὐτῷ πότερον βούλεται ἀποθανεῖν ἢ τυφλὸϲ γενέϲθαι. ὁ δὲ αἱρεῖται μὴ ὁρᾶν τὸν ἥλιον. ἀγανακτήϲαϲ δὲ ὁ Ἥλιοϲ Ἁρπυίαϲ ἐπιπέμπει αὐτῷ, αἵτινεϲ, εἴ ποτε μέλλοι ἐϲθίειν, αὐτοῦ διέπραϲϲον ἐμβάλλουϲαι ϕθοράν τινα. καὶ οὕτω Φινεὺϲ ἐτιμωρεῖτο. τοῦτον οὖν τὸν Φινέα θεωρήϲαντεϲ οἱ περὶ τὸν Ἰάϲονα παρεκάλουν ὑποθέϲθαι πῶϲ δεῖ διαπλεῦϲαι τὰϲ Πλαγκτὰϲ λεγομέναϲ πέτραϲ. ὁ δὲ εἶπεν, ἐὰν τὰϲ Ἁρπυίαϲ ἀπαλλάξωϲι τῆϲ πρὸϲ αὐτὸν ὁρμῆϲ. θέντεϲ δὲ ϲυνθήκαϲ ἐπαγγέλλο-

  Ancora sul Mythographus Homericus (e l’Odissea)

PSI 

Scholia Odyss.

[κτ]ὰς πέτρας. ὁ δ’ εἶπεν δ[ιαπλεύσε] [σθ]αι, ἐὰν ἀποστήσωσι τὰ[ς Ἁρ] [π]υΐας. θέντες δὲ [σ]υνθ[ήκας] αντεπετέλλονται αὐτῷ τοῦ [τ]ο δρᾶν ἐφ’ ὅσον δύνανται. [ὁ δὲ Φι] [νε]ὺς ἐν τῇ Ἀργοῖ κελεύε[ι φέ] [ρ]ειν πελιάδα καὶ α[

νται αὐτῷ τοῦτο δρᾶϲαι. ὁ δὲ λέγει αὐτοῖϲ, πόϲον δύναται ἔχειν τάχοϲ ἡ Ἀργώ; ϕάντων δὲ πελειάδοϲ, ἐκέλευϲεν ἀϕεῖναι περιϲτερὰν κατὰ τὴν ϲυμβολὴν τῶν πετρῶν, κἂν μὲν μεϲολαβηθῇ, μὴ πλεῖν, ἐὰν δὲ ϲωθῇ, τότε περαίνειν τὸν πλοῦν. οἱ δὲ ταῦτα ἀκούϲαντεϲ ποιοῦϲι. καταϲχεθείϲηϲ δὲ τῆϲ περιϲτερᾶϲ διὰ τῆϲ οὐρᾶϲ, προϲβάλλουϲι τῇ Ἀργοῖ δυοῖν .... πληγάδεϲ πέτραι ϲυνελθοῦϲαι τῆϲ νεὼϲ ϲυμμύουϲιν, αὐτοὶ δὲ ϲώζονται. οἱ δὲ Βορεάδαι Ζήτηϲ καὶ Κάλαϊϲ ἀποδιώκουϲι τὰϲ Ἁρπυίαϲ ἀπὸ τῶν Φινέωϲ δείπνων, καὶ οὕτω παραγίνονται εἰϲ Κόλχουϲ. ἡ δὲ ἱϲτορία παρὰ Ἀϲκληπιάδῃ. V.

 Zenodotus, Aristarchus and the Ekdosis of Homer From Zenodotus to Aristarchus, the Alexandrian ekdosis establishes itself within the framework of ancient culture as a typical product of the work of the grammarians, together with the hypomnema, the syngramma and the collection of lexeis. However, the question of what the production of the ekdosis of a literary work on the part of the Alexandrian philologists really and precisely meant has long been the object of considerable debate, touching on at least two fundamental aspects: 1. what is the true nature of the readings attributed by the tradition to figures such as Zenodotus, Aristophanes of Byzantium and Aristarchus? 2. what was the material form of a grammarian’s ekdosis and the manner in which it was concretely produced? In this paper I will deal mainly with the second aspect; to the former issue I will devote only a few preliminary remarks, in order to indicate my own position in the ongoing debate. I remain convinced that the Alexandrian philologists’ production of the ekdosis of a literary work involved both the work of conjectural emendation of the transmitted text and the choice between textual variants discovered through the collation of different copies. The term διόρθωσις referred to the combination of both kinds of activity. One or the other of these two aspects has been emphasized to varying degrees, and some scholars have gone so far as almost totally to deny that the major Alexandrian philologists had any critical philological or intellectual approach involving the concept and practice of comparing different copies and choosing among variants attested in them. In the view of some scholars, the Alexandrian philologists were only — or above all — conjectural critics, who had little compunction about correcting a text they regarded as corrupt and unacceptable on the sole basis of ‘internal’ criteria of evaluation, such as supposed instances of incoherence, unsuitableness and inappropriateness, material repetitions, preference for a more concise text, a love for standardization and uniformity.1 My own view is that, even if there is no doubt that the Alexandrians often did intervene into the texts with rash conjectures devoid of any support in the textual tradition, they must surely be credited with a genuine effort to compare different copies available to them and, confronted as they

 English translation by Rachel Barritt and Glenn W. Most.  1 Valk 1963–1964; Erbse 1969–1988, vol. 7, V f.; Thiel 1991, IX–XIII; Janko 1992, 22–29 (cf. Gnomon 1994, 291 f.). https://doi.org/10.1515/9783110772371-023

  Zenodotus, Aristarchus and the Ekdosis of Homer were with a tradition that was not univocal (and that included cases of omissions and additions), to choose among variants: thus, a true conception of textual criticism and of the history of the text, even if perhaps it was not applied with full methodological coherence. All that is necessary for a crucial intellectual breakthrough to become established as a permanent achievement is for the problem to have been conceived and addressed, even if only in an embryonic, partial, incoherent way: a literary text had its own history of transmission, during the course of which it could have been disrupted at several points; and the correct text could be restored by conjecture or by selecting the correct reading from among those offered by a divergent tradition. The idea of the recognition of a disruption and of a method for repairing it reveals the way in which the collaborative organic unity between textual criticism and interpretation had now become established and operative. The extent and role of either aspect of the Alexandrian grammarians’ philological work should not be taken in isolation from the other, indeed should not even be excessively emphasized: to think in terms of mere conjectures without any manuscript variants, or of no conjectures and mere manuscript variants, are both extreme and highly implausible alternatives. I believe that selection among textual variants discovered through collation of different copies and recourse to conjecture to emend a univocal text are jointly present in the Alexandrian διόρθωσις, starting from Zenodotus. In considering the surviving evidence, we lack general criteria to distinguish easily between the two types of readings, and only a case-by-case analysis will be able to incline us with more or less good reason and confidence in favor of the one or the other.2 In any case, I feel obliged to point out in general that considerable caution must be exercised in evaluating the textual and exegetic interventions of the Alexandrian philologists. For if we wholeheartedly adopt our present-day point of view concerning the “competence” on which their opinions were based, in order to assess whether and to what extent they were right or wrong in comparison with the “truth” established by modern scientific philology, then we run the risk of passing judgments that are not well founded historically or may even be anachronistic. Naturally, this does not mean that we should always refrain from judging the quality of their interventions or that we need make no effort to ascertain

 2 Pfeiffer’s opinion that the Alexandrians’ philological work combined both arbitrary conjectures and the use and collation of copies was already clearly expressed with reference to Zenodotus; cf. Pasquali 1952, 213–243; Nickau 1977, ch. II; S. West 1988, 40–48; a balanced synthesis of the problem in Rengakos 1993, 12–26, esp. 21–24; most recently, see the excellent study by Haslam 1997; for the latest position adopted by Thiel 1997, see below.

Zenodotus, Aristarchus and the Ekdosis of Homer  

whether or not the information available to us is acceptable and plausible. Rather, it means that we should remain fully aware of the historical distance and avoid superimposing our own criteria upon their behavior. In short, let us not forget that what is essential is not so much to know whether Aristarchus does or does not agree with the conception I as a modern philologist have of a passage from Homer, but rather to understand what he himself was doing, and on the basis of what principles, what methods and what knowledge. There has also been much discussion of the second aspect, viz. the real nature of the Alexandrian ekdosis. Giving a precise definition of its form and character remains an arduous task, despite the notable progress achieved in studies of ancient philology (and the increase in documentation deriving from the papyri). What procedure did a grammarian who was the author of an edition of Homer or Pindar or Herodotus follow, and what exactly did he produce? One possibility is that the result of his work was a new continuous copy bearing his text, the entire text rewritten as he believed was right and correct. The other possibility is that he worked on a copy that already existed, carefully chosen from among those at his disposal and used as a base text for all his interventions (hence without rewriting a continuous text). In the latter case one may imagine that the critical signs (however many and of whatever nature they may have been) were placed beside the lines in this very same copy; but it remains open to doubt whether that copy was also furnished with annotations, and if so to what extent and with what contents — in short, how and where the author of a specific ekdosis annotated and preserved his own readings. It has also been hypothesized that at least in certain cases (Aristarchus), the ekdosis was actually identical with the hypomnema, that is to say with all the indications about the text to be accepted that were contained in the complex of lemmas + philological and exegetic treatment of the commentary (with the selected copy acting purely as a basic reference).3 I will address this issue by isolating and focusing specifically on the case of Homer, for at least two good reasons. First, Homer (whose historical and cultural primacy is self-evident) is the text for which this problem arises at the very start of Alexandrian philology, with regard to the edition of Zenodotus, and continues to arise at least up to Aristarchus; what is more, in relation to this latter the scholarly discussion has been particularly lively and detailed because of the preeminence of his person, the influence he exerted, and the quantity of  3 The problem is discussed in detail by Pfeiffer, above all in the chapters on Zenodotus and Aristarchus; cf. Erbse 1959; Groningen 1963; Turner 1980, 109–214 = 1984, 128–143; Thiel 1991, IX–XIII, and Thiel 1992; Janko 1992, 22–29; Rengakos 1993, 12–26; Schmidt 1997; Thiel 1997.

  Zenodotus, Aristarchus and the Ekdosis of Homer documentation. Second, Homer is the text for which we have an unrivaled abundance of material available for each phase, thus providing a richer basis for discussion. But despite the exemplary status that can be attributed to Homer, it is nevertheless prudent not to generalize automatically and globally to all authors what can be proven or reasonably hypothesized for a specific case. My discussion applies to Homer, and whether or not it can be extended to other authors must be determined case by case.4 The idea that Zenodotus’ ekdosis consisted in a new copy containing the entire continuous text which he proposed, completely rewritten with his own readings included and the obeloi in the margins to indicate the atheteses, appears in the most recent studies to have given way to the other possibility, that he worked on a copy which already existed, carefully chosen from among those at his disposal, and used it as a base text on which to perform the interventions he desired. This is how R. Pfeiffer put it: “It is not improbable that Zenodotus, examining manuscripts in the library, selected one text of Homer, which seemed to him to be superior to any other one, as his main guide; its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Διόρθωσις can be the term for either kind of correction. It is hard to imagine any other way.”5 Zenodotus has been the subject of important studies by K. Nickau, who summarizes the issue thus: “Dann ist zu fragen, ob Z(enodotos) nicht einen durch Recensio ermittelten Homertext zugrunde legte (der jedoch nicht seinen Vorstellungen von der genuinen Form der Epen entsprach), diesen mit Obeloi versah und zu ihm Textvorschläge sowie deren Begründung mitteilte. Z(enodotos) selbst wie auch seine Hörer machten sich entsprechende Notizen, die, wären sie von Z(enodotos) schriftlich veröffentlicht worden, ‘Hypomnemata’ hätten heißen können. Aber die Zeil der schriftlich publizierten Homer-Kommentare begann erst mit Aristarchos. So würden sich auch die späteren Unsicherheiten in der Berichterstattung über Z(enodotos’)s Ausgabe erklären.”6 More recently, in connection with his important new editions of the Odyssey and Iliad, H. van Thiel has offered a number of new ideas and proposals on these problems of ancient Homeric philology; these have been much discussed and deserve detailed consideration. Let me remind the reader that these new ideas and proposals are not restricted to just Zenodotus but refer in general to  4 For example, lyric poetry presented problems which certainly differed from the case of Homer, and editions of lyrical poets begin to be mentioned as early as Zenodotus, cf. Nickau 1972. 5 Pfeiffer 1968, 110 = 1973, 188–189. 6 Nickau 1972, 30–31.

Zenodotus, Aristarchus and the Ekdosis of Homer  

Alexandrian ekdosis up to Aristarchus. After the preface to his edition of the Odyssey7 and a first long article,8 and in response to the reactions these sparked — in particular responding to a detailed and effective discussion by M. Schmidt9 — van Thiel returned to his hypotheses with a new article, in which he has clarified and specified many points.10 In his view, the Alexandrian ekdoseis were “annotierte Homertexte”: “In der Einleitung der Odyssee (IX–XII) habe ich die Hypothese aufgestellt, daß alle gelehrten Homerausgaben, die Aristarch benutzte, und außerdem wenigstens eine seiner eigenen beiden Bearbeitungen nicht Textausgaben nach unserem Verständnis waren.” At this point van Thiel specifies and corrects his view in a note: “Ich vermute jetzt (u.S. 29d), daß beide Bearbeitungen Texte mit Rand- und Interlinearkommentaren waren, der eine von Aristarch selbst publiziert, der andere das zugrunde liegende und auch weiterhin ergänzte Handexemplar, das von den Schülern postum mehrfach zum eigenen Gebrauch oder zur Publikation abgeschrieben wurde. Das wäre die ἐπεκδοθεῖσα διόρθωσις Aristarchs, über die sein Schüler Ammonios ein Buch verfaßt hat (Didymos zu T 365–368).” We shall return to the question of Aristarchus’ editions in the second part of this article. In his text van Thiel continues: “Allgemeine historische Gründe sprechen dafür, daß es sich um übernommene Texte handelte, denen ein Rand- und Interlinearapparat beigeschrieben war, in Form der Textscholien” (Thiel 1992, 1 and n. 1). In his latest formulation, he clarifies the point: “Bei der neuen Erklärung handelt es sich nicht um Ersatz des verfügbaren kritischen Instrumentariums, sondern um eine Ergänzung. Alle Bemerkungen zum Text können sein: a) Überlieferungsvariante, vom antiken Editor übernommen, b) Konjektur des antiken Gelehrten, c) Parallele (die natürlich immer auch kommentierende Bedeutung hat und nicht selten mit Varianten übereinstimmt) ..., d) Kommentar in konkreter Formulierung” (Thiel 1997, 15–16). On van Thiel’s hypothesis, then, the Alexandrian ekdosis was a copy chosen by the grammarian from among those available and equipped with a fairly rich set of annotations: “Beiträge der antiken ‘Ekdoseis’ zum Text standen nicht im Text sondern neben dem Text am Rande.”11 At first van Thiel not only attributed  7 Thiel 1991: cf. also Thiel 1996. 8 Thiel 1992. 9 Schmidt 1997. 10 Thiel 1997. 11 This is how van Thiel 1997, 34 summarizes his position; with the value he assigns to this Randapparat he intends inter alia to overcome the aporia pointed out by Nickau (1972, 30 f., cf. Nickau 1977, ch. I; see below and n. 19) concerning the preservation and transmission of the Zenodotean material: cf. Thiel 1992, 6.

  Zenodotus, Aristarchus and the Ekdosis of Homer to the Alexandrian philologists predominantly (if not exclusively) conjectural interventions, but also inclined too much to the view that the contents of this “Rand- und Interlinearapparat” were composed of the annotation of parallel passages and bits of commentary, which quite soon were mistaken for textual readings. Van Thiel’s later suggestion that four different types of content — variants of the tradition, conjectures, parallel passages, and bits of commentary — might have coexisted together finally provides a framework which in general is quite plausible and acceptable in its general outlines; yet the problem is thereby merely displaced and the decisive question becomes that of the relative and absolute weight to be attributed to each of these possible components. It seems clear to me that van Thiel believes that the element consisting of parallel passages and commentary was predominant (and decidedly so) over that consisting of readings indicated by the philologist as the preferred text.12 For my part, on the other hand, I am convinced that the actual interventions upon the text (the genuine διόρθωσις) composed the major portion of the marginal annotations; as regards the additional presence of parallel passages and bits of commentary, I believe this is possible as a secondary element. Furthermore, when the scholia give us incorrect information concerning ancient variants, I think that this is due in most cases to misunderstandings and corruptions that have arisen during the long history of the tradition of scholiographic material and not to parallel passages or paraphrases having already been mistaken for alternative readings in ancient times (indeed even before Didymus and Aristonicus).13 Turning now to the form of the Alexandrian ekdosis, recent studies seem to prefer the idea that it consisted of an already existing copy, chosen and gradually annotated by the philologist as he worked his way through it. To return to the specific case of Zenodotus, I will give a brief description, with a few clarifications, of the way in which I think he is likely to have performed his work of being the πρῶτος τῶν ‘Ομήρου διορθωτής. Zenodotus examined the Homeric copies that

 12 Thiel 1997, 14–15: “Ich wies auf eine andere Erklärungsmöglichkeit hin: Dass die zahlreichen ‘dunklen Lesarten’, die anderweitig nicht bezeugt sind und die zu prinzipiellen Zweifeln an der philologischen und sogar sprachlichen Kompetenz der Alexandriner geführt haben, verständlich werden, wenn man sie nicht als Lesarten versteht, sondern als ‘konkrete’ Notizen kommentierenden Charakters, die wir zu unserem Verständnis ausformulieren müssen; ihre ursprüngliche Gestalt ist in vielen kurzen Bemerkungen des Venetus A erhalten. Ein wichtiger, jedenfalls der auffälligste Teil dieser Anmerkungen besteht in Hinweisen auf andere Homerstellen, in ‘Parallelen’.” It may be added in passing here that in my opinion the “Textscholien” such as are found in Ven. A are by no means an original form, derived directly from the marginal notes of the Alexandrian ekdoseis, cf. Schmidt 1997. 13 Cf. Schmidt 1997 (see the conclusion on p. 12).

Zenodotus, Aristarchus and the Ekdosis of Homer  

were available to him, chose one from among these that seemed preferable to him and used it as the base text for his work of διόρθωσις. About the authenticity of some lines he had doubts and next to these he drew an obelos, the critical sign he invented to indicate athetesis. It has always been harder to understand how he proceeded with those lines that he was certain were spurious and hence definitely to be eliminated from the text — lines which, in any case, must have been present in his base text. It seems to me that we can derive some idea about this from the fact that literary papyri provide ample evidence of different ways of cancelling something present in a text: a horizontal or oblique stroke could be drawn above the words or letters to be eliminated, or they could be marked by dots or lines above or below them, or alternatively they could be enclosed within a pair of lines like round parentheses.14 Cancellations were generally performed, when the need arose, by the διορθωτής, whose task within the book production workshop was to reread and correct the text, often by comparing the copy with the model (a technique of comparing exemplars). The most noble scribal practice thus indicated to Zenodotus ways of performing “cancellations” that he could transfer without difficulty to his own particular διόρθωσις, the διόρθωσις of a philologist and not that of the corrector of a scribe’s work.15 Thus in the end the copy used as base text presented both lines marked with obeloi and lines that were definitely “eliminated”; next to the latter, Zenodotus may very well have annotated something like οὐ γράφειν or some other equivalent term,16 or alternatively he may not have written anything at all, in which case the entire terminology for the elimination of lines could have been written down by those who followed his lessons at the Museum or else been created by the later tradition in order to describe his interventions. Several objections can be raised against this reconstruction. First: what happened when a proposed athetesis or cancellation necessarily involved introducing changes into the remaining text in order for it to be legible? The answer: I see no difficulty in supposing that Zenodotus could have written the alternative text in the margin, next to the line in question, before and / or after the obelized or cancelled section.17 Second: why is it that in the scholiastic tradition  14 Turner/Parsons 1987, 16, with references to examples in the plates; cf. Turner 1980, 93 and Pl. VIII (= 1984, 113 and tav. VIII). 15 Cf. Ludwich 1884–1885, II 134; Nickau 1977, 10 f. 16 The verbs περιγράφω and διαγράφω are technical terms for “cancel” using the material means described above, cf. Turner/Parsons 1987, 16; some examples have remained in the scholia: for περιγράφω cf. Nickau 1977, 10–12 and 29; for διαγράφω see below. 17 The case of B 156–168 is emblematic: cf. sch. B 156–169. If Zenodotus’ copy contained this entire passage and he wanted to eliminate ll. 157–168, then he necessarily had to adopt a dif-

  Zenodotus, Aristarchus and the Ekdosis of Homer we also find οὐκ εἶναι used for Zenodotus’ eliminations, if the eliminated lines were actually there in his copy, cancelled yet present? The simplest answer is that this was a simplified and imprecise way of describing his intervention, probably employed much later and without exact knowledge of the circumstances: the terminology used by the scholia to refer to Zenodotus’ interventions causes problems for any reconstruction, which should not be too surprising. Third: how can it come about that text interventions are found for lines that have been eliminated and hence cancelled? Answer: why not? If so, we should think that the eliminated lines were left in the text, even though they were “cancelled” in some way, and could consequently become the object of attention. Moreover, Zenodotus could have had second thoughts, he could have adopted the idea of elimination at a later stage, after he had thought of intervening upon one of the lines that he later cancelled.18 Earlier in this paper I expressed my conviction that the readings, like the proposals of athetesis and the eliminations of lines, put forward by Zenodotus (and other Alexandrian grammarians) could have arisen both from conjectures designed to improve the text according to internal and subjective criteria, and also from a choice among textual variants discovered by collation. It is clear that at times he must have preferred a different text from that offered by the copy he had chosen as the base text and that he must have wished to change it by introducing either a conjecture of his own or a variant found in another manuscript. In such a case, I am convinced that he annotated his preferred reading in the margins, next to the line in question; consequently, his Homeric text resulted from this set of indications written on his own copy. We need to imagine this work on the text as the fruit of years of study and investigation, leading over time to innumerable interventions on the copy chosen as the base text. Finally this copy, marked up and disfigured by the labor of διόρθωσις, materially became the Homeric ekdosis of Zenodotus the philologist, his very own copy containing the entire fruits of his reflections and investigations, termed ekdosis precisely by virtue of being ἐκδοθεῖσα, i.e., made available to the scholars and students of the Museum. Moreover, it must surely have borne in  ferent text for l. 156 from that which introduces 157 ff.: next to the l. 156 that he had in his copy, namely εἰ μὴ Ἀθηναίην Ἥρη πρὸς μῦθον ἔειπεν, he could easily have written in the margin the text he favored for l. 156 following the cancellation of 157–68, viz. εἰ μὴ Ἀθηναίη λαοσσόος ἦλθ᾿ ἀπ’ Ὀλύμπου. Cf. Ludwich 1884–1885, II 134; Nickau 1972, 30; Nickau 1977, 6 ff. 18 Again, the passage B 156–168 (see previous n.) provides a good example: sch. B 161a tells us of a Zenodotean variant to B 161, even if we know from sch. B 156–169 that Zenodotus eliminated ll. 157–168: both scholia can be traced back to Aristonicus, who evidently was not surprised. On this problem cf. Nickau 1977, 6 ff.

Zenodotus, Aristarchus and the Ekdosis of Homer  

some way the name of the philologist who had thus personalized it, and must have been perfectly identifiable. Thus my idea of Zenodotus’ ekdosis does not differ from the one that we have seen emerging from Pfeiffer, Nickau, and van Thiel, at least as far as the starting point and working base are concerned. But in contrast to Nickau, I suggest that Zenodotus wrote far more in the copy he used as a base text and that in the end this copy contained substantially all the interventions on the text he wished: the atheteses by means of obeloi, the eliminations by means of cancellations, the readings he considered correct by means of marginal annotations. In contrast to van Thiel, I do not believe that this apparatus of marginal annotations was a sort of concise commentary composed primarily of parallel passages and comments; rather, I believe that it mainly contained a true apparatus of textual interventions (atheteses, eliminations and alternative readings) by means of which Zenodotus intended (by conjecture or by collation) to correct the base text he had chosen in order to “transform” it into what he believed to be the correct text of Homer. How long the exemplai of Zenodotus’ ekdosis remained available and accessible, how and in what form it was reproduced in copies, in what way Zenodotus’ arguments were preserved and transmitted — these are other issues that cannot be given in-depth treatment here. I believe that, given that Zenodotus did not produce a hypomnema in the “Aristarchean” sense, what has been preserved of the first διορθωτής’s arguments and lines of reasoning about the Homeric text derives essentially from three sources: first, the memory of his lectures at the Museum, aided to some extent by notes; second, a posteriori reconstructions, aided by a brief annotation (if there was any) or else deduced entirely from his textual interventions (these annotations and / or textual interventions would have been checked against the original, as long as it remained available — this seems likelier than that copies were made); third, material in lexical works (such as the Γλῶσσαι) and in monographs. Let me add that the well-known uncertainty concerning not only Zenodotus’ arguments but at times even his readings — an uncertainty attested as early as the age of Aristarchus — is not at all contradicted by this proposed reconstruction of the form of his ekdosis and of the annotations in it. If his exemplar was lost, it is easy to think of inaccurate copies (if there actually were any copies), contradictory information, a tradition that was less than completely reliable. But even if his exemplar was preserved, it is not improbable that it was not completely clear on certain

  Zenodotus, Aristarchus and the Ekdosis of Homer points, that certain annotations raised problems of interpretation, that sometimes traditional information disagreed with it.19 *** The material form of Aristophanes of Byzantium’s Homeric ekdosis must have been very similar to Zenodotus’: a copy carefully chosen from among those available, on which to work and to annotate his own textual interventions. One important difference, as far as we can tell, was that Aristophanes no longer practiced the drastic kind of intervention represented by the elimination of lines. On our hypothesis, this means that he no longer performed material cancellations. Athetesis marked by the obelos, as already practiced by Zenodotus, became the main or even exclusive tool to express a cautious doubt about parts of the text: οὐ γράφειν disappeared, leaving only ἀθετεῖν; Aristarchus too acted in the same way. It is for this reason that many of the lines which Zenodotus wanted to eliminate but were present in the copies chosen, were preserved in the numerus versuum that became more or less the standard version in the tradition of the Homeric text after the age of Aristarchus. The fact that the drastic step of material cancellation was abandoned provides a vivid illustration of the greater sense of prudence that was now felt, and helps us understand how it happened that many lines “eliminated” by Zenodotus were “restored” (i.e., no longer cancelled) in the text, and thus were not obliterated. Aristarchus marked the beginning of an era in which Alexandrian philological production included an abundance of lengthy hypomnemata. This greatly enriched and facilitated the communication and preservation of philological exegetical arguments, and as a result the tradition has preserved a much more extensive body of material. Aristarchus’ ekdosis too must have had the form of an annotated copy; the active practice of the philological tradition must have imposed itself. Yet the possibility, offered by the hypomnema, of discussing with a certain case and spaciousness numerous topics pertaining to textual criticism and exegesis represented an important resource which substantially changed matters. For in practice, writing more annotations, in addition to the textual interventions, in the margins of the copy chosen as the base text was much less  19 On these problems cf. Pfeiffer 1968, 108 ff. = 1973, 185 ff.; Nickau RE, 28–31 and Nickau 1977, ch. I (he believes that already Aristarchus no longer had the original of Zenodotus, and is also skeptical about the existence of copies of it, cf. above n. 11); Montanari 1988a, 83 ff. and 98–105 (in particular on the work of Ptolemaeus Epithetes); Rengakos 1993, 12 ff. I agree with the following words of van Thiel’s: “Im Fall Zenodots, der Direktor der Bibliothek von Alexandria war, scheint es so ausgesehen zu haben, dass sein hinterlassenes Arbeitsexemplar in der Bibliothek aufbewahrt wurde und seinen Amtsnachfolgern zugänglich war” (Thiel 1997, 20).

Zenodotus, Aristarchus and the Ekdosis of Homer  

useful, and indeed perhaps useless. All that was needed were the critical signs placed beside the lines and a note indicating the readings to be chosen, while the rest, the philological exegetical discussion, found a place and space in the commentary. Aristarchus’ edition of Homer, like Zenodotus’, continues to be the object of spirited debate. In the case of Aristarchus the problems concern not only the form of the ekdosis but also the question of how many Aristarchean ekdoseis there were and what their relation to the commentary or commentaries was. The outlines of the problem first started to become clear as the result of K. Lehrs’ recognition that it was difficult to understand concretely and precisely just what kind of works and how many of them were produced by Aristarchus’ activity of editing and commenting on the text of the Homeric poems. Let us try to summarize briefly its development.20 Lehrs’ interpretation of the material supplied by the scholia led him to the conclusion that Aristarchus had made two editions of Homer preceded by two editions of the commentary (and that Didymus had been able to make use of all this material). The question was reexamined in depth by Erbse (1959), who argued instead that Aristarchus did not produce a genuine new edition, but relied upon current texts and instead composed extensive hypomnemata (probably only once), which were furnished with lengthy lemmas and a large quantity of interventions and opinions about the text. Erbse thus conceived the hypomnema all in all as the place where Aristarchus expounded his ideas about the text, so that the hypomnema effectively coincided with his ekdosis in the sense that it incorporated it. Pfeiffer, finally, rejected this reconstruction, hypothesizing that Aristarchus first wrote hypomnemata based on Aristophanes’ text (i.e. on his ekdosis), later performed his own διόρθωσις, and only then, at this point, produced new (and more “careful”) hypomnemata based on his own ekdosis; finally his διόρθωσις provided the basis for a new ekdosis edited by his pupils (Ammonius, whom I shall discuss shortly). By and large, this is the status quaestionis regarding the works of Homeric philology by the prince of the Alexandrian grammarians, as we have seen it reflected, for instance, in the words of van Thiel quoted above.21 The problem arises, fundamentally, from the fact that the scholia frequently refer unequivocally in the plural to Aristarchus’ editorial work on Homer: αἱ Ἀριστάρχου (scil. ἐκδόσεις or διορθώσεις) and similar expressions, ἡ ἑτέρα τῶν Ἀριστάρχου and similar expressions. On the other hand stand the titles of two  20 Cf. Pfeiffer 1968, 214 ff. = 1973, 335 ff.; Ludwich 1884–1885, I (esp. 16–43) followed in Lehrs’ footsteps. 21 Cited from Thiel 1992, 1 n. 1; the same concept in Thiel 1997, 25 and 34.

  Zenodotus, Aristarchus and the Ekdosis of Homer works by the grammarian Ammonius, Aristarchus’ direct successor: Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως and Περὶ τῆς ἐπεκδοθείσης διορθώσεως, the first in apparent contradiction with the second and with the scholiastic citations that indicate two editions. The various solutions that have been proposed always end up hypothesizing an ἐπεκδοθεῖσα διόρθωσις made by Aristarchus’ immediate pupils, probably by his successor Ammonius, who was certainly familiar with it. However, I do not believe that all the aspects of the problem have been satisfactorily solved, and perhaps it is possible to go a little further. I will begin by stating that, as far as the hypomnemata are concerned, I think that Pfeiffer was right and that it is difficult to deny that Aristarchus made two successive versions of them: a first commentary based upon Aristophanes of Byzantium’s ekdosis is explicitly cited in sch. B 133 α: ἐν τοῖς κατ’ Ἀριστοφάνην ὑπομνήμασιν Ἀριστάρχου, and I do not see why it should be denied that Aristarchus began by writing what his predecessors had not written, a commentary, taking as his starting point the work on the text already accomplished by his predecessor, with whom he was often to find himself in agreement on issues of Homeric philology (especially in contrast with old Zenodotus). In contrast to this stands the citation of finished (ἠκριβωμένα) hypomnemata in sch. B 111 b, which contains one of the most erudite extracts known in Homeric scholiography. Therefore it is entirely plausible that Aristarchus produced a second version of his hypomnemata after his work as a Homeric philologist had made considerable progress. On the specific problem of Aristarchus’ ekdosis or ekdoseis, one very important aspect focuses on the testimony of Ammonius: the meaning of the two titles mentioned above (Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως and Περὶ τῆς ἐπεκδοθείσης διορθώσεως) and the interpretation of the erudite exegetical material in which these works are cited, the scholia to K 397–399 and T 365–368. It is best to start from the case of T 365–368, the content of which is very clear. Sch. T 365–68 a1 (Did. | Ariston.) ... ἀθετοῦνται στίχοι τέσσαρες· γελοῖον γὰρ τὸ βρυχᾶσθαι τὸν Ἀχιλλέα· ἥ τε συνέπεια οὐδὲν ζητεῖ διαγραφέντων αὐτῶν. ὁ δὲ Σιδώνιος ἠθετηκέναι μὲν τὸ πρῶτόν φησιν αὐτοὺς τον Ἀρίσταρχον, ὕστερον δὲ περιελεῖν τοὺς ὀβελούς, ποιητικὸν νομίσαντα τὸ τοιοῦτο. ὁ μέντοι Ἀμμώνιος ἐν τῷ Περὶ τῆς ἐπεκδοθείσης διορθώσεως οὐδὲν τοιοῦτο λέγει. | διπλῆν δὲ κτλ.

Didymus says that these four lines are expunged, and correspondingly there are four obeloi in Ven. A. As is confirmed immediately afterwards, the expunction is naturally due to Aristarchus, who felt that the description of Achilles in a rage while putting on his new armour was ridiculous and claimed that nothing

Zenodotus, Aristarchus and the Ekdosis of Homer  

would be lost in this context by eliminating the four lines.22 Then Didymus also offers some information about the “history” of this philological problem: Dionysius Sidonius, a pupil of Aristarchus, claims that Aristarchus at first (τὸ πρῶτον) expunged the four lines, but then (ὕστερον δέ) changed his mind and eliminated (περιελεῖν) the obeloi, on the grounds that the image was highly poetic and that there was no reason to condemn these lines. In contrast, Ammonius, who was a pupil and the direct successor of Aristarchus, said nothing about any of this: in other words, Ammonius does not say that Aristarchus changed his mind, and hence we are to understand that he supported the view that the great master remained firm in his decision to expunge T 365–368. The source for Ammonius’ “silence” is indicated as a work entitled Περὶ τῆς ἐπεκδοθείσης διορθώσεως: the meaning is evidently that, if Aristarchus had indeed changed his mind, resulting in a concrete philological intervention, one would have expected Ammonius to give a notice and an account of it in that work. Didymus apparently believes Ammonius, or rather his silence, and so he records the expunction without any hint of a doubt: ἀθετοῦνται στίχοι τέσσαρες. There is thus a sharp contrast between the two pupils of Aristarchus. We can suppose that Ammonius chose on his own initiative to ignore the matter because he was convinced that Aristarchus had been wrong to change his mind and had been right the first time; alternatively, we can suppose that Dionysius Sidonius made a genuine mistake or even made the whole thing up; or else, if we wish to trust both of them, we can suppose that this question had been the object of conversation and debate with the pupils but had not gone beyond the stage of exegetic discussion and hence had not ended up producing a true change of heart, a written and formalized repudiation of his earlier proposal of expunction — if so, then the result could have been a difference of opinion between the two pupils as to just what the master’s final idea really was. But all this is pure speculation; what concerns us here is not the solution to the problem but rather its existence and the manner in which it is stated. From the scholia it is clear that Didymus says: Aristarchus’ opinion is that the four lines should be expunged, Dionysius Sidonius claims that Aristarchus reversed his decision, but in Ammonius’ Περὶ τῆς ἐπεκδοθείσης διορθώσεως there is no trace of a change of mind on his part in this question. The other scholium too concerns a case of expunction, namely that of K 397–399, and here too we find a correspondence in the three obeloi placed beside these three lines in Ven. A (where for 1.398 there is also a diple, to be discussed

 22 I think διαγράφω is used here without a specific technical connotation, cf. n. 16.

  Zenodotus, Aristarchus and the Ekdosis of Homer shortly). But the situation is rather more complex and requires a more extended examination. Sch. K 397–99 a. (Did.) ἢ ἤ δ η χ ε ί ρ ε σ σ ι ν : καὶ γραπτέον οὕτως, καί ἀθετητέον τοὺς τρεῖς στίχους, εἵ τι χρὴ πιστεύειν Ἀμμωνίῳ τῷ διαδεξαμένῳ τὴν σχολήν, ἐν τῷ Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως τοῦτο φάσκοντι. καὶ παρὰ Ἀριστοφάνει δὲ ἡθετοῦντο. ἐν δὲ ταῖς Ἀριστάρχου ἐγέγραπτο “εἰ ἤδη”, ταῦτα ὁ Δίδυμος (ρ. 115 Schm.) Sch. Κ 397–99 b. (Ariston. I x) : ὅτι οὕτως γραπτέον, βουλεύουσι (398) καὶ θέλουσι (398)· τὸ γὰρ σφίσιν (398) ἐν τῷ περί τινών ἐστι λόγῳ ἀντὶ τοῦ αὐτοῖς, ᾧ ἀκόλουθα δεῖ εἶναι τὰ ῥήματα. ταῦτα ὁ Ἀριστόνικος (ρ. 180 Friedl.) περὶ τῆς γραφῆς ταύτης φησί, διπλῆν βάλλων τῷ στίχῳ. | ἐν μέντοι τῆ τετραλογίᾳ Νεμεσίωνος οὕτως εὗρον περὶ τῶν στίχων τούτων· “τῶν παρακειμένων ὀβελῶν (sc. ante versus 397–99) οὐκ ἔστιν αἰτίαν εὑρεῖν διὰ τῶν Ἀρισταρχείων ὑπομνημάτων. Ἀμμώνιος δὲ ὁ Ἀριστάρχειος πρῶτον μὲν στιγμαῖς φησι τὸν Ἀρίσταρχον παρασημειώσασθαι αὐτούς, εἶτα δὲ καὶ τελέως ἐξελεῖν, τάχα διὰ τὸ ἐπὶ δευτέρου προσώπου τὸ σφίσι (398) τετάχθαι, καὶ ἄνωθεν (sc. e Κ 310–312) μετενηνέχθαι.”

Erbse has divided scholium Κ 397–399 b into two parts. In the first (up to the vertical line), the scholiast explicitly makes use of material derived from Aristonicus; in the second (introduced by ἐν μέντοι τῇ τετραλογίᾳ Νεμεσίωνος οὕτως εὗρον περὶ τῶν στίχων τούτων κτλ.), on the other hand, the material comes from an unknown source, which Erbse indicates by an x, commenting ad loc.: “ad haec verba explicanda Oedipo, ut dicunt, opus coniectore est.” Whether the τετραλογία of Nemesion is actually VMK (the Viermännerkommentar) or whether we are dealing with another source used by the scholiast, what relation holds between these lines and the rest of the scholium on this passage and their sources Didymus and Aristonicus, who is the first person who says εὗρον — all these questions remain shrouded in mystery.23 However, this aspect is not crucial for the purposes of our investigation, and for the moment we can leave it aside and proceed to a general analysis of the information that we are provided. Here it is at the end that the reference, introduced by an expression of doubt (τάχα), appears to what may have been the reasons for suspecting K 397–399, which (as Didymus informs us in scholium 397–399 a) had already been expunged by Aristophanes of Byzantium. The problem regards ll. 310–312 and 397–399. Let us first take a look at the Homeric passages. In K 299 ff. Hector appears to be eager for action: he asks the Trojans who would be willing to be

 23 Erbse 1969–1988, ad loc., with bibliographical references; cf. Valk 1963–1964, I 107 ff.; discussion in Nickau 1977, 3, 260–263.

Zenodotus, Aristarchus and the Ekdosis of Homer  

sent on a mission to spy on the enemy and he promises a reward of a carriage and two horses to whomever 307

310 312

ὅς τίς κε τλαίη, οἷ τ᾿ αὐτῷ κῦδος ἄροιτο, νηῶν ὠκυπόρων σχεδὸν ἐλθέμεν ἔκ τε πυθέσθαι, ἠὲ φυλάσσονται νῆες θοαὶ ὡς τὸ πάρος περ, ἦ ἤδη χείρεσσιν ὑφ᾿ ἡμετέρησι δαμέντες φύξιν βουλεύουσι μετὰ σφίσιν, οὐδ᾿ ἐθέλουσι νύκτα φυλασσέμεναι, καμάτῳ ἀδηκότες αἰνῷ.

Dolon volunteers to go, but during the foray he is captured by Diomedes and Odysseus. In answer to the latter’s questions, Dolon reports that Hector incited him to go out in the night to spy on the enemy. We have a repetition involving five lines (308–312 ≈ 395–399), with a variation in the first hemistich of ll. 308 ~ 395 and a textual problem in l. 398. 395

399

395

399

ἠνώγει δέ μ᾿ ἰόντα θοὴν διὰ νύκτα μέλαιναν ἀνδρῶν δυσμενέων σχεδὸν ἐλθέμεν ἔκ τε πυθέσθαι, ἠὲ φυλάσσονται νῆες θοαὶ ὡς τὸ πάρος περ, — ἦ ἤδη χείρεσσιν ὑφ᾿ ἡμετέρησι δαμέντες >— φύξιν βουλεύουσι μετὰ σφίσιν, οὐδ᾿ ἐθέλουσι — νύκτα φυλασσέμεναι, καμάτῳ ἀδηκότες αἰνῷ. (Ludwich; Allen) ἠνώγει δέ μ᾿ ἰόντα θοὴν διὰ νύκτα μέλαιναν ἀνδρῶν δυσμενέων σχεδὸν ἐλθέμεν ἔκ τε πυθέσθαι, ἠὲ φυλάσσονται νῆες θοαὶ ὡς τὸ πάρος περ, — ἥ ἤδη χείρεσσιν ὑφ᾿ ἡμετέρησι δαμέντες >— φύξιν βουλεύοιτε μετὰ σφίσιν, οὐδ᾿ ἐθέλοιτε — νύκτα φυλασσέμεναι, καμάτῳ ἀδηκότες αἰνῷ. (Leaf; Ameis-Hentze; Mazon; van Thiel)

In the first passage, it is Hector who is speaking, and therefore the third-person verbs of l. 311 are quite natural: the spy has to find out whether the ships are still being guarded (φυλάσσονται νῆες) as they were earlier, or whether the Achaeans are planning among themselves to flee and no longer wish to keep guard during the night. In the second passage it is Dolon speaking, addressing Diomedes and Odysseus who are standing in front of him. In the corresponding l. 398 the manuscript tradition presents an alternative: the verbs in the third person indicative, βουλεύουσι – ἐθέλουσι, i.e., whether the Achaeans are planning among themselves to flee and they no longer wish to keep guard at night; or else the verbs in the second person optative, βουλεύοιτε – ἐθέλοιτε, i.e., whether you Achaeans are planning among yourselves to flee and you no longer wish to

  Zenodotus, Aristarchus and the Ekdosis of Homer keep guard at night. The latter variant evidently bears the hallmark of contextual adaptation of the repetition. Modern editions (as represented above) follow either the one alternative or the other. I believe that it is difficult to deny that the information provided by Didymus and Aristonicus indicates that in K 398 the reading βουλεύουσι – ἐθέλουσι was the one favored by Aristarchus.24 Aristonicus clearly explains that the reason for this choice lies in the fact that the pronoun σφίσιν can only be used for the third person, therefore meaning “among themselves” and not “among yourselves”. The reasoning is perfectly Aristarchean: always attentive to Homer’s characteristic linguistic usage, he rejects an anomalous and unparalleled use of σφίσιν for the second person. It has been claimed that here Aristarchus was intervening into the text conjecturally, correcting to βουλεύουσι – ἐθέλουσι a tradition that gave only βουλεύοιτε – ἐθέλοιτε, which he deemed unacceptable because it involved σφίσιν = ὑμῖν against the normal Homeric usage.25 This is possible, although I must say that I see no cogent reason for rejecting the hypothesis that he found divergences among the Homeric copies he consulted. In any case, the fact remains (or so we are told by Aristonicus at least) that Aristarchus maintained, either by personal conjecture or as a result of choice among variants, that it was necessary to write βουλεύουσι – ἐθέλουσι in K 398 because βουλεύοιτε – ἐθέλοιτε involved a linguistic usage he considered unacceptable. Aristonicus’ diple preserved at K 398 concerned this specific point. This line is actually part of a group of three, K 397–399, for which expunction had been proposed. Let us first of all reject the idea that there might be some difficulty in thinking of a discussion on a linguistic problem contained in an athetized line: it was not unusual for scholars to treat textual or exegetical issues in lines for which athetesis was proposed, i.e., lines that remained in the text marked by an obelos. Didymus informs us that the three lines had been athetized by Aristophanes of Byzantium: hence the problem had already been raised earlier and Aristarchus found himself confronted with it in some way. At the beginning of scholium 397–399a Didymus says that the three lines are to be athetized, scil. also according to Aristarchus, if we are to believe what Ammonius says (εἴ τι χρὴ πιστεύειν Ἀμμωνίῳ ... τοῦτο φάσκοντι) in his work Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως. It is by no means infrequent to

 24 So Ludwich 1884–1885, I 318–319; Chantraine 1973–1981, I 274–275; Erbse 1969–1988, ad loc.; a different view is put forward by Nickau 1977, 260–263. In sch. K 397–399 a, Didymus also tells us of a reading of Aristarchus εἰ ἤδη for l. 397: cf. B 349, see Ludwich 1884–1885, ad loc.; Chantraine 1973–1981, II 292–296. 25 Leaf 1900–1902, ad loc.; Chantraine 1973–1981, I 274–275; Valk, I 108; Nickau 1977, 261.

Zenodotus, Aristarchus and the Ekdosis of Homer  

find Didymus expressing uncertainty concerning the exact state of the text and the genuine interpretation of Aristarchus; here the expression of doubt applies, admittedly somewhat cautiously, to Aristarchus’ pupil and successor, the author of works devoted specifically to his master’s Homeric edition: εἴ τι χρὴ πιστεύειν Ἀμμωνίῳ. In the other scholium, K 397–399 b, in the part attributed to Nemesion, we find first of all the statement that in Aristarchus’ hypomnemata no explanation of the obeloi placed beside K 397–399 was to be found. This suggests that Didymus had a copy in which these lines were marked with obeloi, but could discover no justification for this athetesis in the Aristarchean exegetic material. This must have caused some uncertainty: so the only authority he could turn to were Ammonius’ works on the Aristarchean ekdosis, to which he could evidently resort in doubtful cases, as is indicated by the case of Sch. T 365–368 a1 discussed above. We know something about what Ammonius said thanks once more to the part of scholium K 397–399 b attributed to Nemesion. Ammonius claimed that Aristarchus had first (πρῶτον μέν) marked the three lines with stigmai, but subsequently (εἶτα δέ) had definitively suppressed them (τελέως ἐξελεῖν), probably (τάχα) because they contained σφίσι used for the second person (1. 398) and had been repeated from an earlier passage (i.e., from K 310–312). These statements must be examined closely. First, the mention of the stigme as a semeion: στιγμαῖς φησι τὸν Ἀρισταρχον παρασημειώσασθαι αὐτούς. The use of the stigme as a critical sign is not widely attested, and, as far as we can tell, was not very frequent. Nonetheless, there appears to be general agreement that Aristarchus sometimes marked a line with a stigme in situations in which he had reasons for doubt and misgivings, but did not possess a firm decision and cogent arguments for proposing an athetesis and marking an obelos (or else for performing some other critical intervention?); possibly, these were cases Aristarchus left temporarily unresolved in the hope of returning to the passage and studying it in greater depth later. This is what is normally assumed to have happened here: it is thought that Aristarchus first marked the lines with a sign of doubt (whatever that sign may have been), and then later decided in favor of a genuine athetesis (εἶτα δὲ καὶ τελέως ἐξελεῖν).26 The use of the term ἐξελεῖν, instead of

 26 Lehrs 1882, 340–341; Friedländer 1853, 180; Ludwich 1884–1885, I 318–319, II 138–142; Leaf 1900–1902, ad loc.; Pfeiffer 1968, 218 = 1973, 340; S. West 1967, 133; Nickau 1977, 261. Cf. sch. B 192a + 203a; sch. Θ 535–537; H. Erbse in Gnomon 28, 1956, 275.

  Zenodotus, Aristarchus and the Ekdosis of Homer the more common ἀθετεῖν, is supported by enough parallels that in itself it does not constitute a difficulty.27 If this was his final decision, it would therefore seem that Aristarchus did not have the time to write down his arguments in the hypomnemata, where, as we have seen, no explanation of this point was to be found. The justifications put forward uncertainly (τάχα) in the continuation of the scholium are evidently deductions, either by Ammonius himself (based on his master’s discussions, which he had heard?) or else even by the source that quotes Ammonius. These justifications refer to the fact that the three lines are repeated from K 310–312 and to the difficulty of the pronoun σφίσι used for the second person in l. 398 (the latter point could constitute an argument in favor of the suggestion that the transmitted reading was only βουλεύοιτε – ἐθέλοιτε). One can imagine that Aristarchus felt suspicion and doubt when faced by a repeated passage that was unacceptable with βουλεύοιτε – ἐθέλοιτε and σφίσιν = ὑμῖν but perhaps could become acceptable with the correction to (or the choice of) βουλεύουσι – ἐθέλουσι: first he would simply have placed a sign of doubt, and later he would have decided in favor of athetesis, but without having the time to write down the reasons. He could have argued that although the passage could be remedied grammatically by the reading βουλεύουσι – ἐθέλουσι, it nevertheless had to be expunged because it was a useless or verbose repetition (this is one of the justifications indicated: τάχα διὰ τὸ ἐπὶ δευτέρου προσώπου τὸ σφίσι (398) τετάχθαι καὶ ἄνωθεν (sc. from Κ 310–312) μετενηνέχθαι). In any case, leaving aside the reasons for the position Aristarchus finally adopted, this would explain the obeloi and the text of Didymus, who knows of the athetesis but can find no support for it in Aristarchus’ commentaries and can therefore refer only to Ammonius (εἴ τι χρὴ πιστεύειν Ἀμμωνίῳ ... τοῦτο φάσκοντι), who must have said that Aristarchus finally athetized the lines but was not entirely convincing about the justifications — hence, in both cases, a certain diffidence on the part of Didymus (εἴ τι χρὴ πιστεύειν). There is no contradiction between this part of the scholium  27 Cf. sch. Φ 195b; Plut. Mor. 2. 26 f; Athen. 5. 181 c; Nickau 1977, 28–29. I would reject the idea that ἐξελεῖν might refer not to the lines but rather to the stigmai, i.e., that it might mean that Aristarchus decided at a later date to eliminate the earlier signs (cf. above sch. T 365–368: ὕστερον δὲ περιελεῖν τοὺς ὀβελούς) and not to athetize the lines (having overcome all the problems by reading βουλεύουσι – ἐθέλουσι?): for one would have to suppose that for some reason the uncertainty remained (e.g. the stigmai might not have been physically eliminated, through forgetfulness or lack of time, or might have been eliminated only imperfectly) and that someone introduced the obeloi by conjecture after the time of Aristarchus, speculating in some way about the reason for the athetesis. Such a reconstruction, I would argue, is much too contorted and consequently is unlikely.

Zenodotus, Aristarchus and the Ekdosis of Homer  

and Aristonicus’ material: the latter concerns a diple, which evidently referred to a discussion by Aristarchus in the hypomnema concerning the state of the text of 1.398 and the related problem of σφίσι: this diple could perfectly well remain even once the master — despite having chosen the reading that removed the grammatical problem of σφίσι — had decided at the same time in favor of athetesis and had added the obeloi. Let me now recapitulate this reconstruction. Faced by the obeloi, in the absence of information in Aristarchus’ hypomnemata, Didymus consulted Ammonius, who informed him about the two stages of Aristarchus’ exegetical considerations on the lines in question and about Aristarchus’ final decision in favor of athetesis, in this way explaining why there was not enough time for the arguments in favor of the final position — athetesis — to be set down in the commentary, which was evidently written earlier. Despite the uncertainties, we can at least say that we have two Homeric passages with regard to which there are parallel suggestions that Aristarchus may have changed his mind from an earlier position, and information about this change of mind is sought by consulting the writings of Ammonius, his pupil and direct successor. In order to explain this situation and reconcile the apparent contradiction between the two titles of works by Ammonius, Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως and Περὶ τῆς ἐπεκδοθείσης διορθώσεως, it has been supposed that Aristarchus might actually have composed only one ekdosis (as presupposed by the first title) and that the ἐπεκδοθεῖσα διόρθωσις (mentioned in the second one) was composed by Ammonius himself or other pupils. I must confess, however, that I have never been fully satisfied by this reconstruction, above all after careful analysis of the two philological cases in relation to which Ammonius’ works are cited. I think that a slightly different hypothesis might help us understand better what happened. The fact that there were no further editions of the Aristarchean διόρθωσις seems to me to be capable of being fully reconciled with the fact that his διόρθωσις was re-edited (ἐπεκδοθεῖσα) only if we suppose that the ἐκδοθεῖσα διόρθωσις and the ἐπεκδοθεῖσα διόρθωσις were one and the same thing, but modified. My reconstruction runs as follows: Aristarchus began by writing hypomnemata based on the text of Aristophanes; then he devoted himself to his own διόρθωσις and produced an ekdosis of his own, i.e., the copy he had chosen as his base text and containing annotated in the margins all his textual interventions (critical signs and readings) so that it would represent his Homeric text. At this point he wrote new hypomnemata, which were based on his own ekdosis and felt (or defined) to be more in-depth and careful. But the research and reflection by a

  Zenodotus, Aristarchus and the Ekdosis of Homer philologist like Aristarchus on a text like Homer was hardly likely ever to come to an absolute end, a definitive conclusion. Not only did Aristarchus continue to reflect and study, he also continued to teach: Homer was never out of his hands. Thus he resumed work on his διόρθωσις, took up his ekdosis once more, and on that very same copy began to annotate changes of mind and new textual choices. The very same copy which had been the ἐκδοθεῖσα διόρθωσις went on to become the ἐπεκδοθεῖσα διόρθωσις; but because he wrote no new hypomnemata, he was unable to set down at length his reasons for changing his mind on various issues and define the point his thoughts had reached.28 This copy can well be called ἐπεκδοθεῖσα because it did indeed remain available to scholars and numerous pupils, probably when Aristarchus fled from Alexandria during the crisis of 144 or else at his death. A single copy of Homer bearing two successive layers of interventions — this is a perfect explanation for the fact that the tradition commonly speaks of αἱ Ἀριστάρχου, ἡ ἑτέρα τῶν Ἀριστάρχου and so forth. In a certain sense, one could speak of two editions, i.e., two different states of the text, and in general the first version and the second one must both have been readable and distinguishable, so that the later philological tradition could know both and take account of both. On the other hand, new interventions on an already annotated copy also provide a highly plausible explanation for situations of uncertainty in which, for whatever reason (perhaps sometimes for material reasons), it may not have been clear just what the two textual states indicated by the master were, what his final position was, and whether he had really reached a firm decision. Ammonius was certainly justified in explaining that there was really only one ekdosis, but that at some point it had been “re-edited” with changes: it is highly probable that the titles Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως and Περὶ τῆς ἐπεκδοθείσης διορθώσεως refer either to two parts of one and the same work or to two closely connected small works, whose aim would have been to explain how this had happened and above all to give clarifications about doubtful or problematic cases, Homeric passages where the master’s final decision was unclear because it had been imperfectly annotated or else perhaps because his reflections had been interrupted by circumstances. In the case of T 365–368, it is likely that Aristarchus reconsidered and discussed the passage once more, without being able finally to resolve his doubts, when he took up his διόρθωσις again, yet did not go as far as to eliminate the obeloi he had previously, marked. Thus the athetesis remained intact in the  28 He may have undergone some changes of mind from what was noted down in the ekdosis even while he was working on the second series of hypomnemata, based on his own text.

Zenodotus, Aristarchus and the Ekdosis of Homer  

ἐπεκδοθεῖσα διόρθωσις, but Dionysius Sidonius and Ammonius made different evaluations of Aristarchus’ arguments, to the point of disagreeing about the interpretation of his final decision. In the case of K 397–399, it seems that Aristarchus first indicated doubt (apparently with stigmai) but later opted for athetesis, in all probability replacing the previous sign in some way with obeloi; however, he did not have the chance to expound his arguments because he wrote no further hypomnemata. Ammonius then thought it useful to explain that the earlier signs had been replaced by the second ones and to say something about his master’s reasons for changing his mind. In any case, not everything was perfectly clear for Didymus, who appears to be reluctant to place total faith in Ammonius’ words. On the other hand, as Pfeiffer wrote:29 “Whether Didymus was able to work on copies of these original διορθώσεις and ὑπομνήματα of Aristarchus and of his monographs, the συγγράμματα, is an insoluble problem.”30

 29 Pfeiffer 1968, 217 = 1973, 339. 30 I think that the analysis can usefully be taken further by analyzing the few other fragments of Ammonius in the Homeric scholia and by examining in depth the cases in which we have information that Aristarchus changed his mind about the Homeric text: for instance, Z 4 (cf. Nickau 1977, 3), O 449–451, Σ 207, I 464 (?).

 A New Papyrus Commentary on the Iliad Some fragments of a papyrus codex to be published in P.Oxy. vol. LXXVI (Inv. 84/17 (a) + 84/32 (a)) uncover a new commentary on the Iliad of the utmost importance.1 Only frr. 1 and 2 + 3 preserve a significant portion of text. The scribe writes in a splendid example of the perfected Alexandrian majuscule of the type with contrasting broad and narrow letter forms, probably datable either to between the Vth and VIth century A.D., i.e. the period during which the canon of this type of script became established, or to the first half of the VIth century.2 The broad margins (the lower one preserved for over 4 cm, the upper one for 1.5 cm, and the lateral margin for up to 2.5 cm) and the calligraphic style reveal a high-level book product, elegant in its graphic presentation, with great care devoted to precision in the text: a rather fine exemplar, certainly destined to the bookselling trade. The height and width of the page cannot be determined with certainty: it can be said only that the lines contained roughly 37–40 letters and the writing area must have been about 14 cm wide, but there is nothing to indicate how many lines were contained in a page or to suggest the height of the writing area. The work that has come to light in the papyrus fragments forms part of a rich and extensive hypomnema on the Iliad in an independent codex, separate from the text of the poem. In the larger fragments, sections of commentary to book XII and book XV can be identified: fr. 1 preserves a series of lemmata belonging on the → side to Il. 12.91–92 and 110–111, on the ↓ side to Il. 12.136, 139– 140, 147 and 148; frr. 2+3 preserve a series of lemmata belonging on the → side to Il. 15.610–614, 618, 623, 624 and 625, on the ↓ side to Il. 15.641–642 (?), 645, 653, 659, 661. It seems obvious that the preserved remnants constituted part of a hypomnema to the entire Iliad. They clearly do not constitute sporadic annotations: rather, we are dealing with a complete and systematic work, reproduced in a fine high-quality edition. The lemmata identified on the two sides of fr. 1 go  English translation by Rachel Barritt Costa.  1 I wish to express my warm thanks to Dirk Obbink for the permission to anticipate informations concerning this unpublished papyrus, and to Davide Muratore for his important help in preparing the edition. 2 Guido Bastianini argues in favour of dating the codex to the V/VIth century, with reference to Cavallo 1975, 39–40, 46–48, 51 = Cavallo 2005, 188–190, 195–196 and 199 (with tab. XLVII ab); Cavallo/Maehler 1987, tables 22a-b, with comm. 52; Cavallo 2008, 101–105. Daniela Colomo prefers a dating set well into the VIth century A.D., drawing on a comparison with Cavallo/ Maehler 1987 tab. 37 (Paschal letter dated A.D. 577), with the references therein. https://doi.org/10.1515/9783110772371-024

A New Papyrus Commentary on the Iliad  

from l. 91 to l. 148 of book XII, and those identified on the scraps of the two sides of frr. 2+3 go from l. 610 to l. 661 of book XV. We thus may have 50/60 commented lines per page in the preserved parts, but nothing can be said about the lost parts because the page size cannot be reconstructed. It could perhaps be speculated that roughly 100 Iliad lines may have been covered per sheet as an average for the whole poem, in which case book XII might have occupied five sheets, book XV eight sheets, and the entire hypomnema to the Iliad could have been contained in a fine codex composed of roughly 140–160 sheets.3 The iota adscript is always written. Only scanty lectional signs: an apostrophe in fr. 1 →, l. 17 δ’ ομηρος; single dot over ι (fr. 1 →, l. 12) and υ (fr. 2 + 3 →, l. 5), perhaps over υ in fr. 1 ↓, l. 11 (or it could be a smooth breathing). The only visible abbreviation (κ(αι) fr. 1 →, 1. 12) appears to be motivated by respect for endof-line alignment, but in fr. 2 + 3 →, 1. 3, the end-of-line alignment is not respected (because the word is a lemma?). In general, the spelling is correct and the text is written carefully; probably there is a (simply phonetical?) mistake in fr. 1 ↓, l. 19, ἀΐσσονται for the dual ἀΐσσοντε. The sections of the commentary are well separated by means of a middle point,4 which is placed at the end of the lemma and at the end of the commentary section before a new lemma; but a middle point is also used as a punctuation sign within the sentences in the body of the exegesis (cf. fr. 1 →, ll. 11, 17 and 18; fr. 2 + 3 →, l. 6; fr. 2 + 3 ↓, l. 15). The extension of the lemmata is variable, as is usually the case: the majority are contained within one line; in one case it is certain that the lemma consists of a single word (15.645, fr. 2 ↓, l. 3); in one case the lemma is a pericope that extends over two lines, both incomplete (12.139–140, fr. 1 ↓ ll. 11–12); one lemma consists of two lines that are not fully written out but are instead indicated with a few words followed by ἕως and by the final word (12.110–111, possibly also 12.91–92, both in fr. 1 →). In the parts that have been preserved, the commentary matches the order of the Homeric text perfectly and no lemmata placed in inappropriate positions are found. No names of grammarians appear, nor are citations from other authors adduced in the exegetic arguments. A number of elements seem to suggest that P.Oxy. should be regarded as a highly unique finding, of paramount importance. The latest Homeric commentary known so far (excluding the scholia Minora and the Mythographus Homericus) was P.Mich. inv. 1206, which is dated to the IlIrd/IVth century A.D. and  3 This seems plausible according to the data given by Turner 1977, 82–84. 4 More frequent is the use of a dicolon and/or blank space; on the middle point see Salomons 1984.

  A New Papyrus Commentary on the Iliad written only on the recto, suggesting (although there is no absolute certainty) that it is a fragment of a roll.5 This means that the text under investigation here is now unquestionably the latest known Homeric hypomnema and the first to come to light that is definitely contained in a papyrus codex. We have ten or so codex commentaries on various authors, datable to between the III/IVth and the VIth century A.D., but so far none on Homer:6 somewhat strange, if one thinks of the astonishing critical fortune of the Homeric poems at all levels of education, scholarship and society. Now we have what is a late-antiquity commentary presumably on the entire Iliad, contained in a book of excellent quality both as regards its graphic appearance and the presentation of the text: it is thus a book that combines external beauty with the value of the critical-exegetic content and careful transcription, a product that must surely have been destined to an equally discriminating public.7 Let us now look at two short text portions from fr. 1 → and from fr. 2 →. As usual the complete edition will be presented in the Oxyrhynchus Papyri series. fr. 1 → 15

20

ἀλλ᾿ οὐ]χ̣ Ὑρτακίδης ἔϑε̣ λ̣᾿ Ἄ̣σιος ἕ̣ ω̣ϛ̣ ϑεράποντα ἕτερος ποιητὴς ἂ]ν τῶι Πουλυδάμαντι πάντας ἐποί ησεν πειϑομέ]νους· ὁ δ᾿ Ὅμηρος μιμούμενος τὴν αλ̣ή̣ ϑειαν ἕνα γοῦ]ν τὸ[ν ἀπε]ι̣ ϑοῦντα εἰσάγει· διὰ τί δὲ ἕνα τοῦτον; ὅτι μάλιστα το]ῖ̣ ϛ̣ ἵπποις ἠγάλλε το

12.110–111

Sch. ex. Il. 12.110a1 (T): ἀλλ᾿ οὐχ Ὑρτακίδης ἔϑελ᾿ < Ἄσιος, ὄρχαμος ἀνδρῶν>: βαρβαρικὴ ἡ ἀπείϑεια. ἕτερος μὲν ἂν ποιητὴς τῷ Πολυδάμαντι ἐποίησε πάντας πειϑομένους, ὁ δὲ Ὅμηρος μιμούμενος τὴν ἀλήϑειαν ἕνα γοῦν τὸν ἀπειϑοῦντα εἰσάγει. διὰ τί δὲ ἕνα τοῦτον; ὅτι μάλιστα τοῖς ἵπποις ἠγάλλετο· “μεγάλοι” (Μ 97) γὰρ ἦσαν· οἷς καὶ ϑαρρῶν ἀπόλλυται (cf. Ν 384–93).

 5 Luppe 1992; cf. MP3 1198.01, CPP 0485, LDAB 2078. 6 See Stroppa 2009. A few commentaries on papyrus roll are dated to the IVth century A.D.: for ex. P.Oxy. 856, commentary on Aristophanes, Ach. (Aristophanes 1 CLGP); P.Berol. inv. 13419 (MP3 1357), commentary on Pindar, P. 2.17–19. 7 On the characteristics of the manuscripts written in Alexandrian majuscule of profane content, cf. Porro 1985, who sets out a typology that easily encompasses our P.Oxy.; for the papyrus commentaries in bookhands, see also Del Fabbro 1979, 81–83.

A New Papyrus Commentary on the Iliad  

Sch. ex. Il. 12.110a2 (b): βαρβαρικὴ ἡ ἀπείϑεια. μιμούμενος δὲ τὴν ἀλήϑειαν ὁ ποιητὴς ἕνα γοῦν τὸν ἀπειϑοῦντα εἰσάγει. διὰ τί δὲ ἕνα τοῦτον; ὅτι μάλιστα τοῖς ἵπποις ἠγάλλετο. καταστέλλει οὖν τὴν τῶν πολλῶν ἀλαζονείαν. fr. 2 + 3 → 5

ἴσχον γὰρ π[υργηδὸν ἀρηρότες ἠΰτε πέτρη· ἡ μὲν ἐμβολὴ τῶν [βαρβάρων κύμα σιν εἴκασται καὶ ἀ̣νέ̣ μ̣ωι· ἡ δὲ τῶν̣ [Ἑλλήνων καρτε ρία τῶι τῆς πέτρας ἀκιν]ή̣τωι κα̣[ὶ δυσπαϑεῖ. λαμπό μενος πυρί· ] α̣λλεμ.[ ]π̣υ̣ρός· ἐν δ᾿ ἔπ̣[εσ᾿ ]ὡς ὅτε κῦ[μα

15.618

15.623 15.624

4–7 Cf. sch. ex. Il. 15.618b (bT): πυργηδὸν : ἡ μὲν ἐμβολὴ τῶν βαρβάρων τοῖς κύμασιν εἴκασται ταχὺ διαλυoμένoις· ἡ δὲ τῶν Ἑλλήνων καρτερία τῶι τῆς πέτρας ἀκινήτωι καὶ δυσπαϑεῖ. 8 ̣[ slightly leftwards arching trace, so that ε, ο, σ are possible; the curvature of φ is usually more accentuated (but ἐμφ̣[α- cannot be excluded). 7–9 Cf. sch. ex. Il. 15.623 (bTil): περιλαμπόμενος ὑπὸ τῶν ὅπλων ὡς ὑπὸ πυρός. These two sections of the commentary, as can be seen, form an almost wordperfect match to the corresponding scholia exegetica known from the medieval codices, and can easily be reconstructed in parallel with them (comparison is with the edition of Erbse). In the other parts that have been preserved, the phenomenon is not so striking, but from what remains it can be established that P.Oxy. undeniably constitutes evidence of a hypomnema, the exegetic contents and interests of which are of the same type as is found in the scholia exegetica of the medieval tradition: in fact, the contents are extensively overlapping or bear a strong similarity. None of the considerable number of Homeric papyrus commentaries known to date can be so closely likened to this traditional stream, nor do they provide such distinctively characterized exegetic materials (thus differing sharply, in other words, from the line of Aristarchean Alexandrian ancestry which, through the works of Aristonicus, Didymus, Nicanor and Herodianus and the so called VMK, i.e. Viermännerkommentar, eventually led to the pre-

  A New Papyrus Commentary on the Iliad ponderant mass of the scholia of Ven. A as well as in smaller quantity to the scholia of the bT group of mss.).8 The formation of the corpus of the scholia exegetica is one of the many aspects of the history of ancient Homeric philology in which the acquisition of new data and new bodies of evidence is essential in order to clarify a number of issues and achieve progress in knowledge.9 P.Oxy. will certainly offer much material for study and in-depth research. It contains a commentary belonging to the same stream as the sch. ex. of the medieval tradition: it may coincide verbatim with the corresponding sch. ex., or show only partial agreement with the latter, or display a clear difference while still addressing contents of the same kind; at times it may contain richer materials and preserve unknown annotations, whereas elsewhere the material seems meager and lacks portions present in the sch. ex. that have come down to us. For example, what can be read in fr. 1 → as a comment on Il. 12.91–92 is strikingly different and far richer than the material known from the corresponding sch. ex. of bT, but this annotation is then followed directly by that pertaining to 12.110–111, while a number of sch. ex. that are present in bT in the portion of text between 12.92 and 12.110 are missing in the papyrus; in fr. 1 ↓ the comment on Il. 12.147 is certainly richer than the sch. ex., but the preceding part unquestionably lacks the annotations to 12.137 and 144, which are present in bT. In short, at the present stage of research it seems possible to say that the compiler of the corpus of sch. ex. transmitted by the bT group of manuscripts, i. e. by the tradition dating back to Erbse’s archetype c, made use of three commentaries from late antiquity. It also seems likely that the material was gathered together somewhere between the IIIrd and the IXth century A.D., as the extreme dates.10 P.Oxy. is a fine exemplar of a commentary that incontrovertibly belongs to this critical-erudite line and which was composed at the very latest on the date assigned to this copy, but perhaps earlier: it can therefore be set fully within the time span (a fairly broad interval) identified by scholars as the likely range in which the formation of the corpus should be situated. What was the model of P.Oxy., and how should we propose to characterize it?

 8 For ex., the hypomnema from which the fragment P.Mich. inv. 1206 derives, mentioned above, cites the grammarians Demetrius Ixion (with the title of the work), Zenodotus and Didymus in just a few lines. 9 An overview in Schmidt 2002, partic. 170–176, with the bibliographical references; cf. infra and n. 10. 10 Cf. Schmidt 1976, 67–69.

A New Papyrus Commentary on the Iliad  

We should resist the temptation of imagining that our commentary was in fact a genuine exemplar of the corpus sch. ex. in its original form that shaped the archetype c (BT): this seems impossible, because in the sch. ex. deriving from c (i.e. in the bT manuscripts) a substantial amount of material is found which is absent in P.Oxy. (cf. above). One possibility is that the model of P.Oxy. was one of the (three) commentaries used in compiling the corpus of the sch. ex. and that it therefore represents an earlier stage than c in the tradition of this exegetic material (but even in this case the date of our papyrus copy would be of no use for the chronology of c, which could have been created earlier or later). A further possibility is that the model of P.Oxy. was a commentary belonging to the same typology as those that were at the origin of the corpus of the sch. ex. and that it featured contents which were partly the same as, and partly different from, those that eventually made up c (bT), having elements in common but intermingled with others that were subsequently lost.11 Within such a framework of reference, this overall new body of evidence will have to be analyzed with precision and subjected to in depth investigation, as soon as the material becomes fully available for scholars.

Commentary on Iliad XII and XV (P.Oxy. 5095) 84/17(a) + 84/32(a)

fr. 1 12.5 × 18 cm

fifth/sixth century

fr. 2 11.2 × 6.5 cm

Seven fragments of a papyrus codex containing a new commentary on the Iliad. A good portion of text is preserved in frr. 1 and 2+3. The → side of fr. 1 shows a right-hand margin of 2 cm, the ↓ side a left-hand margin of 2.5 cm; the → side of fr. 2 shows a left-hand margin of 2.2 cm and an upper margin of 1.5 cm, the ↓ side a right-hand margin of 2 cm and an upper margin of 1.5 cm. Fr. 3 (4 × 5.4 cm) belongs to the same leaf as fr. 2. The → side of fr. 4 (2.5 × 2.1 cm) shows a left-hand margin of 1.5 cm, the ↓ side a right-hand margin of 1 cm; the → side of fr. 5 (6 × 4.5 cm) shows a lower margin of 4.3 cm, the ↓ side a lower margin of 4 cm. Fr. 6 is 3 × 2 cm, and fr. 7 is 1 × 0.5 cm. The scribe writes in a splendid example of the perfected Alexandrian majuscule of the type with contrasting broad and narrow letter forms, probably datable either to between the fifth and sixth centuries AD, i.e. the period during which the canon of this type of script became established, or to the first half of  11 On the problems of the formation of the scholiographic corpora cf. Montana 2007 and Montana 2010, § 2.2.5, with the bibliographical references.

  A New Papyrus Commentary on the Iliad the sixth century. For the dating to the fifth/sixth century (favoured by Professor G. Bastianini), see Cavallo 1975, esp. 39–40, 46–48, 51 = 2005, esp. 188–190, 195–196, 199 (with tav. XLVII a–b); Cavallo/Maehler 1987, tav. 22a–b, with comm. p. 52; Cavallo 2008, 101–105. Dr Daniela Colomo prefers a dating set well into the sixth century AD, drawing on a comparison with Cavallo/Maehler 1987, tav. 37 (Paschal letter dated AD 577) with the references there. The broad margins (the lower one preserved for over 4 cm, the upper one for 1.5 cm, and the lateral margin for up to 2.5 cm) and the calligraphic style reveal a high-level book product, elegant in its graphic presentation, with great care devoted to precision in the text: a rather fine copy, certainly destined for the bookselling trade. The height and width of the page cannot be determined with certainty: it can be said only that the lines contained roughly 37–40 letters and the writing area must have been about 14 cm wide, but there is nothing to indicate how many lines were contained in a page or to suggest the height of the writing area. The text was a rich and extensive hypomnema on the Iliad in an independent codex, separate from the text of the poem. In the larger fragments, sections of commentary to book XII and book XV can be identified: fr. 1 preserves a series of lemmata belonging on the → side to Il. 12.91–92, 110–111, on the ↓ side to Il. 12.136, 139–140, 147, 148; frr. 2+3 preserve a series of lemmata belonging on the → side to Il. 15.610–614, 618, 623, 624, 625, on the ↓ side to Il. 15.641–642(?), 645, 653, 659, 661. It was very likely part of a hypomnema to the entire Iliad. The preserved portions clearly do not constitute sporadic annotations: rather, we are dealing with a complete and systematic work, reproduced in a fine highquality edition. The lemmata identified on the two sides of fr. 1 go from l. 91 to l. 148 of book XII, and those identified on the scraps of the two sides of frr. 2+3 go from l. 610 to l. 661 of book XV. We thus may have 50–60 commented lines per page in the preserved parts; but nothing can be said about the lost parts because the page size cannot be reconstructed. It may be conjectured that roughly 100 Iliad lines may have been covered per sheet as an average for the whole poem, in which case book XII might have occupied five sheets, book XV eight sheets, and the entire hypomnema to the Iliad could have been contained in a fine codex composed of roughly 140–160 sheets (this seems plausible, according to the data given by Turner 1977, 82–84). Iota adscript is always written. But only a few lectional signs appear: an apostrophe in fr. 1 →, l. 17 δ᾿ ομηροϲ; diaeresis appears (unusually) marked by a single suprascript dot, both inorganic (fr. 1 →, l. 12 over ι in ιππουϲ) and organic (fr. 2+3 →, l. 5 over υ in ηυτε). (GMAW2 10 with n. 44, asserts that ‘sometimes a single dot’ serves to mark diaeresis, but give no examples.) The only visible abbreviation, κ(αι) fr. 1 →, l. 12, appears to be motivated by respect for end-of-line

A New Papyrus Commentary on the Iliad  

alignment, but in fr. 2+3 →, l. 3, the end-of-line alignment is not respected (because the word is a lemma?). In general, the spelling is correct and the text is written carefully; probably there is a (simply phonetic?) mistake in fr. 1 ↓ 19, ἀΐϲϲονται for the dual ἀΐϲϲοντε. The sections of the commentary are well separated by means of a middle point (more frequent is the use of a dicolon or blank space; on the middle point, see Salomons 1984), which is placed at the end of the lemma and at the end of the commentary section before a new lemma; but a middle point is also used as a punctuation sign within the sentences in the body of the exegesis (cf. fr. 1 → 11, 17, 18; fr. 2 + 3 → 6; fr. 2 + 3 ↓ 15). The extent of the lemmata is variable, as is usually the case: the majority are contained within one line; in one case it is certain that the lemma consists of a single word (15.645, fr. 2 ↓ 3); in one case the lemma is a pericope that extends over two lines, both incomplete (12.139–140, fr. 1 ↓ 11–12); one lemma consists of two lines that are not fully written out but are instead indicated with a few words followed by ἕωϲ and by the final word (12.110–111, possibly also 12.91–92, both in fr. 1→). In the parts that have been preserved, the commentary matches the order of the Homeric text perfectly, and no lemmata placed in inappropriate positions are found. No names of grammarians appear, nor are citations from other authors adduced in the exegetic arguments. A number of elements suggest that 5095 is of unusually high importance among previously known commentaries on Homer on papyrus. The Homeric commentary of the latest known dating so far (excluding the Scholia Minora and the Mythographus Homericus) is P. Mich. inv. 1206 of the third/fourth century AD and written only on the recto, suggesting (although this is not absolutely certain) that it is a fragment of a roll: Luppe 1992; cf. M–P3 1198.01, LDAB 2078, CPP 0485). This means that 5095 is now unquestionably the latest known Homeric hypomnema and the first to come to light that is definitely contained in a papyrus codex. We have ten or so codex commentaries on various authors, datable to between the third/fourth and the sixth century AD, but so far none on Homer (Stroppa 2009, 298–327; Stroppa 2008, 54–55; a few commentaries on papyrus rolls are dated to the fourth century AD, e.g VI 856, commentary on Aristophanes, Acharn. (Aristophanes 1 CLGP); P.Berol. inv. 13419 (M–P3 1357), commentary on Pindar, Pyth. 2.17–19). This is somewhat strange, given the extensive and long-lived endurance of the Homeric poems at all levels of education, scholarship, and society. In 5095 we have a large-scale late-antique commentary, presumably on the entire Iliad, contained in a book of excellent quality both as regards its graphic appearance and the presentation of the text: a book that combines concerns for aesthetic qualities with that for high-quality

  A New Papyrus Commentary on the Iliad critical exegesis and careful transcription, a product no doubt aimed at an equally discriminating public. (On the characteristics of the non-biblical manuscripts written in Alexandrian majuscule, see A. Porro, ‘Manoscritti in maiuscola alessandrina di contenuto profano: Aspetti grafici, codicologici, filologici’, S&C 9 (1985) 169–215, with a typology that encompasses and includes 5095; for the papyrus commentaries in bookhands, see also Del Fabbro 1979, 81–83.) Some sections of the commentary form an almost word-perfect match to the corresponding scholia exegetica known from the medieval codices, and some lines can be reconstructed in parallel with them (comparison is with the edition of Erbse). In other parts the phenomenon is not so striking, but here the exegetic contents and interests are of the same type as is found in the scholia exegetica of the medieval tradition. None of the considerable number of Homeric papyrus commentaries known to date can be so closely likened to their medieval counterparts. In this respect they differ sharply, in other words, from the line of inquiry of Aristarchean Alexandrian ancestry which, through the works of Aristonicus, Didymus, Nicanor, and Herodianus and the so-called VMK, i.e. Viermännerkommentar, eventually led to their preponderance in the scholia of Ven. A, as well as in smaller quantity to the scholia of the bT group of manuscripts. By comparison, the hypomnema from which P. Mich. 1206 derives, mentioned above, cites the grammarians Demetrius Ixion (with the title of the work), Zenodotus, and Didymus in a few abbreviated lines. Thus 5095 alternately coincides verbatim with the corresponding medieval scholia, or shows only partial agreement, or displays a clear difference while still addressing contents of the same kind; at times it contains richer materials and preserves unknown annotations, whereas elsewhere the material seems meagre and lacks portions present in the scholia of the bT tradition that have come down to us. For example, what can be read in fr. 1 → 9–14 as a comment on Il. 12.91–92 is strikingly different and far richer than the material known from the corresponding scholia, but this annotation is then followed directly by that pertaining to 12.110–111, while a number of observations on the portion of Homeric text between 12.92 and l. 110 that are present in bT are missing in the papyrus; in fr. 1 ↓ 15–18 the comment on Il. 12.147 is certainly richer than the corresponding medieval scholia, but the preceding part unquestionably lacks the annotations to 12.137 and 144, which are present in bT. We are grateful to Professor F. Pontani for comments on an earlier version, and to Dr Daniela Colomo for the English translation from the Italian of the notes below.

A New Papyrus Commentary on the Iliad  

Fr. 1 → .

5

10

15

20

.

.

.

. ]. ] ̣α̣ι ̣ ] ̣ ̣ ] ̣ ̣ ] ̣ ̣ ] ε̣ϲ̣ ] α̣ϲ̣ ] ̣ α ̣ ]αιϲϕινκε ]μενγαρου ]ω̣[ ̣] ̣ [ 10–12 ] ̣ τ̣ ̣εθριππον·δια ] ̣ ̣ σελα ̣ [ 7–9 ] ̣ ̣ τ̣εινδϊππουϲκ() ] ̣ν̣ατο·δ ̣ [ 4–6 ]αλλονεπαυτωνκα ]κτωρ ̣ ̣( ̣)του ̣ ̣ [ ̣ ̣] ̣ αϲ̣τονκε̣βριονην ] ̣ υρτακιδηϲ̣εθ ̣λ̣α̣ϲιọϲ̣ ̣ ̣ϲ̣θεραποντα ]ντωπουλυδαμαντιπααντα ̣ εποι ]νουϲ·οδ᾿ομηροϲμιμουμενοϲτηνα̣ ̣ ̣ ]ντ ̣ [ 3–5 ] ̣θουνταειϲαγει·διατι ] ̣ ̣ ι̣πποιϲηγαλλε ] ̣ ̣ ̣ ̣ ̣ [ 3–4 ] ( ̣) ̣εϲ ]πρ̣ο ]τ̣ι̣ . . . . .

1 ]., part of horizontal stroke in upper part of writing space; to left possibly remains of upright, 5 ] ̣ ̣, possibly ⲏ̣ⲥ̣, less probably ⲡ̣ⲉ̣ 8 ] ̣ ̣, first, negligible; second, remains of two uprights: ⲛ or ⲡ 12 ] ̣ ̣, first, possibly ⲭ (with upper part of diagonal ascending from left to right hardly visible), ⲁ not to be ruled out; second, remains of upright with thick foot, ⲓ, ⲧ or ⲡ ιπ, single dot above ι, diaeresis 13 ̣[, upright 14 ρ ̣ ̣, first, upper part of upright; second, lower part of upright curving to right; third, round letter, ⲟ or ⲱ υ ̣ ̣, two uncertain letters ] ̣, ⲧ? 20 ] ̣ ̣ ̣ ̣ ̣[, first, remains of horizontal with, in upper part of the writing space to the right, extremity of diagonal ascending from left to right, perhaps of a ⲕ; second, apex, probably of a ⲁ; third, remains of thick vertical or narrow arc or left-hand half of rather thick horizontal, compatible with ι; the whole series can be read as ⲕⲁⲓ, but ⲕⲁⲧ is not to be ruled out ] ⲉ ̣ ⲥ, upright curving to left probably in ligature with other letter, ⲁⲓ or ⲇⲓ

ⲅ or ⲧ

10

15

κ]αί ϲϕιν Κε βριόνηϲ τρίτοϲ εἵπετο ]μεν γὰρ ου ]ω̣[ ̣] _ [ 10–12 ] ̣ τ̣ ̣εθριππον· δια ] ̣ ̣ ϲελα ̣ [ 7–9 ]. . τ̣ ειν δ᾿ ἵππουϲ κ(αὶ) ] ̣ ν̣ατο· δ ̣ [ 4–6 ] ἄλλον ἐπ᾿αὐτῶν κα τέλιπεν ὁ Ἕ]κτωρ ̣ ̣ ( ̣)του ̣ ̣ [ ̣ ̣] ̣αϲ̣: τὸν Κε̣βριονην ἀλλ᾿ οὐ]χ̣ Ὑρτακίδηϲ̣ ἔθε̣λ̣᾿ Ἀ̣ϲιοϲ ἕ̣ω̣ϲ̣ θεράποντα

XII 91–92

110–111

  A New Papyrus Commentary on the Iliad

20

ἕτεροϲ ποιητὴϲ ἂ]ν τῶι Πουλυδάμαντι πάνταϲ ἐποί ηϲε πειθομέ]νουϲ· ὁ δ᾿ Ὅμηροϲ μιμούμενοϲ τὴν ἀλ̣ή̣ θειαν ἕνα γοῦ]ν τὸ̣ [ν ἀπε]ι̣ θοῦντα εἰϲάγει· διὰ τί δὲ ἕνα τοῦτον; ὅτι μάλιϲτα το]ῖ̣ϲ̣ ἵπποιϲ ἠγάλλε το ]

9–14 The commentary apparently focuses on the τέθριπποϲ. Although the preserved scholia on these lines do not treat it, ancient exegesis preserves discussions on the presence of the four-horse chariot in the Homeric poems; see sch. Il. 8.185ab, 11.699, 19.400, Od. 4.500. In Il. 8.185 the names of Hector’s four horses are listed: Aristarchus expunges the verse mainly on the ground that the use of the four-horse chariot is not known to Homer, a fact supported by the use of the dual for the horses. (See Aristonicus in sch. 8.185a: ὅτι οὐδαμοῦ Ὅμηροϲ τεθρίππου χρῆϲιν παρειϲάγει. μάχεται δὲ καὶ τὰ ἐπαγόμενα δυϊκά, καὶ ἡ προεϕώνηϲιϲ εὐήθηϲ. The following part of the sch. seems to suggest that the four names were actually two names accompanied by two epithets; Nicanor, however, refuses this possibility.) The long sch. ex. 8.185b too states that οὐδαμοῦ δὲ τεθρίππῳ κέχρηνται ἥρωεϲ and adds that the τέθριπποϲ is quoted only ἐπὶ παραβολῆϲ in Od. 13.81 (cf. sch. ad loc.). However, Il. 11.699 apparently mentions a four-horse chariot belonging to Neleus, ὑπώπτευται ὡϲ νόθον ἢ δύο ἅρματα δηλοῦν βούλεται (p. 335.36–39 Erbse): if so, the verse is spurious or it mentions two bigae and not a quadriga (so Aristonicus in sch. 11.699a, recording Aristarchus’ interpretations: not a τέτρωροϲ but δύο ϲυνωρίδεϲ; cf. sch. ex. 699b). In any case the same sch. ex. 8.185b goes on to present other elements of the discussion. In the first place the text (p. 335.39–49 Erbse) remarks that, although the other warriors do not use the four-hourse chariot, Hector alone dares to yoke four horses in order to awe and shock the enemy. Moreover, Hector is a descendant of Tros, Ganymede’s father: in exchange for Ganymede Zeus gave Tros special horses since he was particularly keen on and good at riding horses. But afterwards the objection that the use of the dual follows (8.186, 191) is taken into consideration: the problem should be solved by referring the dual to two yoked horses (ζύγιοι) and two added laterally (παρήοροι): heroes usually use three horses, two of them ζύγιοι and one παρήοροϲ in case one of the ζύγιοι is wounded, but Hector dared to yoke a fourth horse, fearing that both ζύγιοι (p. 335.49–336.62 Erbse) would be wounded. That Hector represents an ‘exception’ among Homeric heroes allows us to understand why sch. Od. 4.590, commenting on the gift of three horses and a chariot by Menelaus to Telemachus, observes that there would be no explanation for the number ‘three’ if they knew the τέθριπποϲ: we have actually to do with a ϲυνωρίϲ (a biga with two ζύγιοι) plus a παρήοροϲ, as usual in the Iliad, πλὴν Ἕκτοροϲ: on this basis the reference to the idea that Hector may represent an exception to the rule according to which τέθριπποϲ is not used appears to be clear. In the same direction goes the sch. of Aristonicus on ll. 11.699a (see above) on the ‘possible’ four-horse chariot belonging to Neleus: it explains that it actually consists of two bigae and concludes οἱ δὲ ὡϲ Ποϲειδῶνοϲ υἱὸν τετρώρῳ ϕαϲὶ χρῆϲθαι· ἵππιοϲ γὰρ ὁ θεόϲ. To sum up, on the one hand there was an exegetical view according to which the τέθριπποϲ or τέτρωροϲ was not used by Homeric warriors, apart from two exceptions due to different reasons: Hector at ll. 8.185 and Neleus at ll. 11.699. On the other hand, the interpretation given by Aristarchus allows us to understand the two passages within the general rule, according to a typical Aristarchean critical tendency. This issue somehow emerges also in the sch. of Aristonicus at ll. 19.400, which offers a comment on Achilles’ speech to his horses Xanthos and Balios, as gifts from the gods to

A New Papyrus Commentary on the Iliad  

Peleus: ϲημειοῦνταί τινεϲ, ὅτι ἐντεῦθεν ἡ διαϲκευὴ τοῦ τεθρίππου πεποίηται ‘Ξάνθε τε καὶ ϲύ, Πόδαγρε᾿ (Θ 185). The meaning is not clear, but perhaps there are traces of a discussion making a point against those who held that this passage on Achilles’ horses could have been used as a model for the passage on Hector’s horses at Il. 8.185, dealing with the problem of the τέθριπποϲ in this instance too. As said above, the preserved sch. at 12.91–2 do not mention this problem. However, 5095 reveals that the problem of the τέθριπποϲ (see line 11) was relevant to the interpretation of this passage too, although it is difficult to understand how and to what extent. Lines 13–14 refer to 12.92 ἄλλον Κεβριόναο χερείονα κάλλιπεν Ἕκτωρ, which suggests that the lemma included this verse itself (with regard to line 10 one could think of τρίτοϲ εἵπετο ἕωϲ Ἕκτωρ; cf. line 15, but see below): exempli gratia δι̣[ὸ (δὲ) καὶ] ἄλλον or δι[ὰ τοῦτο] ἄλλον (Pontani) ἐπ᾿αὐτῶν κα|[τέλιπεν ὁ Ἕ]κτωρ or ἄλλον ἐπ᾿αὐτῶν κα|[κίω κατέλιπεν ὁ Ἕ]κτωρ (Lapini), perhaps too long, although it is difficult to count the letters in the rest of this line (ἔλιπεν instead of κατέλιπεν seems to me less probable). One may think that the focus of the exegesis here was the following: Hector adds a third person to guide the first group of warriors, apart from Cebrion, Poulydamas, and himself (see lemma in ll. 9–10), 12.92 ἄλλον Κεβριόναο χερείονα), i.e. a fourth anonymous warrior, who was left near the chariot because his inferiority to Cebrion. This may have somehow been related to the horses (ἵππουϲ on line 12). In any case it is not clear whether the argument was against or in favour of an allusion to the τέθριπποϲ: such an allusion, however, would be rather tortuous and extravagant, given that nothing in the whole passage 12.80–92 suggests anything related to the four-horse chariot, although one cannot either rule out completely that this topic was somehow mentioned, or underestimate the fact that 5095 goes in the same direction as the sch. ex., whose sources consider Hector as an exception in relation to the use of the τέθριπποϲ, as said above on ll. 8.185: a commentator holding the view that Hector dared to yoke a four-horse chariot in order to impress the enemy could have tried to find traces of this interpretation at any cost, especially with regard to the moment of the attack against the Achaean wall. Professor Pontani suggests the following reconstruction of this passage:

10

κ]αί cϕiv Κε βριόνηϲ τρίτοϲ εἵπετο· εὐλόγωϲ· ἔϕα]μεν γὰρ οὐ δαμοῦ τῆϲ ποιήϲε]ω̣[ϲ] π̣[αρειϲάγεϲθ]αι τ̣έθριππον· δια ϕερων δ᾿ ἦν οὗ]τ̣ο̣ϲ ἐλαύ̣[νειν ὄχημα, κ]ρ̣α̣τ̣εῖν δ᾿ ἵππουϲ κ(αὶ) ἥττων τιϲ ἠδ]ύ̣ν̣ατο· δι̣[ὰ τοῦτο] ἄλλον ἐπ᾿ αὐτῶν κα τέλιπεν ὁ Ἕ]κτωρ ̣ ̣( ̣)του ̣ ̣[ ̣ ̣] ̣αϲ̣ τὸν Κε̣βριόνην.

10 ἔϕα]μεν γὰρ would refer to the discussion of the use of bigae instead of quadrigae; at line 13 ἥττων τιϲ or ἕτερόϲ τιϲ can be suggested; at line 14 the idea of ‘taking with himself’ Cebrion could have been expressed. 15–20 At this point the text exactly coincides with the known sch., and therefore it is possible to reconstruct the text on the basis of the sch. ex. at this line: sch. 110a1: βαρβαρικὴ ἡ ἀπείθεια. ἕτεροϲ μὲν ἂν ποιητὴϲ τῷ Πολυδάμαντι ἐποίηϲε πάνταϲ πειθομένουϲ, ὁ δὲ Ὅμηροϲ μιμούμενοϲ τὴν ἀλήθειαν ἕνα γοῦν τὸν ἀπειθοῦντα εἰϲάγει. διὰ τί δὲ ἕνα τοῦτον; ὅτι μάλιϲτα τοῖϲ ἵπποιϲ ἠγάλλετο· “μεγάλοι” (Μ 97) γὰρ ἦϲαν· οἷϲ καὶ θαρρῶν ἀπόλλυται (cf. N 384–93); the wording is more concise in sch. 110a2: βαρβαρικὴ ἡ ἀπείθεια. μιμούμενοϲ

  A New Papyrus Commentary on the Iliad δὲ τὴν ἀλήθειαν ὁ ποιητὴϲ ἕνα γοῦν τὸν ἀπειθοῦντα εἰϲάγει. διὰ τί δὲ ἕνα τοῦτον; ὅτι μάλιϲτα τοῖϲ ἵπποιϲ ἠγάλλετο. καταϲτέλλει οὖν τὴν τῶν πολλῶν ἀλαζονείαν.

Fr. 1 ↓ .

5

10

15

20

. . . . ̣[ [̣ [̣ ϕ ̣[ ϲ ̣[ τ̣ ̣[ τ[ λα ̣( ̣)[ μίμνο ̣[ μιοντο̣υ̣ ̣[ ταϲαυτον·κα ̣[ ̣ ̣ ̣ ̣ ̣ ̣] ̣α ̣ο̣ρ ̣ [ ταδαμαντα ̣ ̣[ ̣ ̣ ̣ ̣ ̣ ̣] ̣ ̣ ̣ ̣ ̣ ̣ ( ̣) [ ανωτερωιπαρα ̣ [ ̣ ̣ ̣] ̣ τωνγαρ ̣[ ει ̣ ̣ ̣ ̣ ̣ ̣ κ̣ατ ̣α[ ̣ ̣ ̣] ̣ νυνενκαιρω[ ξι̣ωϲι̣νευ̣νε ̣ ̣ ̣[ ̣]( ̣) ̣ ναυτουϲτωι[ ηδ̣ ̣ κυνωνδεχ ̣ ται·εμϕατικη[ ̣ ̣τηνγενομενηνταραχηνκαι ̣[ γ ̣[ ̣]ϲτοιϲκωλοιϲϲ ̣ ̣ ̣ ̣[ ̣ ̣ ̣ ̣( ̣)] ̣ ̣ ̣ ̣[ τ ̣ ̣ϲϲονται·π ̣ [ ρε ̣[ ̣ ̣ ̣ ̣ ( ̣)] ̣ ̣ ̣ ̣ ̣[ ρι ̣[ τ̣[ . . . .

.

5 initial ⲥ rounded, while ⲥ in this script is usually oval: this may be due to the fact that the letter has slightly been enlarged because it is at line-beginning 6 τ̣ ̣ [, π is to be excluded because there is not the usual curving right-hand upright; of the second letter remains of left-hand arc, possibly ⲧ̣ⲱ 8 α ̣( ̣)[, remains of upright at mid-height and perhaps traces of another letter 12 ] ̣ ̣ ̣ ̣ ̣, first, two diagonals crossing one another: the upper part of both strokes and the lower part of the diagonal ascending from left to right are visible, possibly ⲭ; second, trace of horizontal stroke at mid-height touching an upright, perhaps ⲏ; third, perhaps ⲙ; fourth, right-hand angle of ⲁ or ⲇ; fifth, upright, perhaps with a horizontal protruding from the centre, as ⲏ 13 ̣[, angle to the left in lower part of writing space, possibly ⲇ 15 ] ̣ ̣ν, first, upper part of upright 18 γ ̣[ ̣], upright whose lower extremity ends with a small blob; to right, join with another letter at midheight 20 ̣[, probably ι 21 ̣[, upright slightly curved, apparently not ⲉ

10

μίμνον̣ [ἐπερχόμενον μέγαν Ἄϲιον· τὸ ἐγκώ μιον το̣ῦ̣ Ἀ[ϲίου ϕθάνει καὶ ἐπὶ τοὺϲ ὑπομένον

XII 136

A New Papyrus Commentary on the Iliad  

15

ταϲ αὐτόν· καὶ ᾿Ἰ̣αμενὸν] κ̣αὶ̣ Ὀ̣ρέ̣[ϲτην Ἀϲιάδην τ᾿ Ἀδάμαντα ̣ ̣ [ ̣ ̣ ̣ ̣ ̣ ̣] ̣ ̣ ̣ ̣ ̣ ̣ ( ̣)[ ἀνωτέρωι παρα ̣ [ ̣ ̣ ̣ ̣ ̣ ] ̣των γαρ ̣[ ει ̣ ̣ ̣ ( ̣ ̣)κ̣ατ( ̣)α[ ̣ ̣ ̣] ̣ νῦν ἐν καιρῶ[ι κατ᾿ ἀ ξι̣ωϲι̣ν ϲυ̣νε ̣ ̣ ̣ [ ̣]( ̣) ̣ν αὐτοὺϲ τωι[ ἀνδρῶν ἠδ̣ὲ κυνῶν δέχα̣ται· ἐμϕατικη[ ̣ ̣ ( ̣)τὴν γενομένην ταραχὴν καὶ η̣[ γ ̣ϲ τοῖϲ κώλοιϲ ϲ ̣ ̣ ̣ ̣ [ ̣ ̣ ̣ ̣ ( ̣)] ̣ π̣ ̣ ̣[ δοχμώ τ᾿ ἀ̣ί̣ϲϲονται· π ̣[

139–140

147

148

9–11 Cf. sch. ex. ll. 12.136 μίμνον ἐπερχόμενον μέγαν Ἄϲιον: μεγα· τὸ ἐγκώμιον Ἀϲίου ϕθάνει καὶ ἐπὶ τοὺϲ ὑπομένονταϲ αὐτόν. At line 9 it is not clear whether the commentary began with the repetition of μέγαν as in sch. (μέγαν Ἄϲιον· μέγαν τὸ ἐγκώ]|μιον) or not (μέγαν Ἄϲιον· τὸ ἐγκώ]|μιον): Professor Pontani prefers the second possibility, assuming a line of 33 letters (plus high stop); the two following lines have 36 and 35 letters respectively, but include several rather broad letters (4 M and ω). 12 The form Ἀδ- is the commonest; only a couple of MSS contain the form Ἀκ- (cf. West, app. ad loc.): v. 140 Ἀϲιάδην τ᾿ Ἀδάμαντα Θόωνά τε Οἰνόμαόν τε. After the gap, the series may be read as ϲ]χ̣ῆ̣μ̣α̣. But see below a different interpretation of the traces. 12–15 Cf. sch. ex. 139–140: οὗτοι τῶν περὶ τὸν Ἄειόν εἰϲιν ἄριϲτοι. ἐν δὲ τῇ προκειμένῃ διατάξει (M 95–97) οὐκ ἀναγκαῖον ἦν καὶ τούτουϲ καταλέγειν. εἰϲὶ γὰρ κατὰ τὸ ἴδιον τάγμα τοῦ Ἀϲίου, ἐκεῖ (M 88–104) δὲ τοὺς ἀρίστουϲ ἐξ ἁπάντων κατέλεγεν. Perhaps a similar line of thought could be recognized here: as in sch. ἐν τῇ προκειμένῃ διατάξει/ἐκεῖ contrasts with the preceding οὗτοι κτλ., in 5095 ἀνωτέρωι (a later spelling for ἀνωτέρω) contrasts with νῦν ἐν καιρῶ[ι; then probably a comment on the fact that the poet thought it appropriate to list the ἄριϲτοι companions of Asius; e.g. νῦν ἐν καιρῶ[ι ὁ ποιητὴϲ κατ᾿ ἀ]|ξίωϲιν ϲ̣υ̣ν̣ έ̣ τ̣ α̣[ξ]ε̣ v αὐτοὺς τῶι [Ἀϲίωι (as suggested by Lapini, cf. sch. D ad 12.415: ἐκαρτύναντο ϕάλαγγαϲ glossed with ϲυνετάξαντο, but ϲυνέταξεν seems to be too short and ϲυνέταϲϲεν is not very convincing). 15–16 After αὐτούϲ the commentary probably carries on (sc. αὐτοὺϲ τῶι [Ἀϲίωι?) rather than beginning a new lemma in the form αὐτούϲ τῶι [τ᾿ ἐν ὄρεϲϲιν ἀνδρῶν]|ἠδ̣ὲ κυνῶν δεχα̣ται. Such a reconstruction would lead us to assume the omission of a high stop and a mistake, τῶι for τώ; moreover line 15 would be too long. 16 All witnesses (included the oldest ones; cf. sch. ad loc.) have δέχαται; only Ambr. gr. F 205 inf. has δέχεται (cf. app. West ad loc.). In 5095 the letter after χ, crucial to deciding which reading was written on the papyrus, is damaged:

  A New Papyrus Commentary on the Iliad however, the traces are compatible only with Α, although some uncertainty remains. The concept of emphasis often occurs in Homeric sch. as a criterion for stylistic judgment: examples can easily be found through Erbse’s Index III, but the sch. to this passage do not offer parallels. It is possible that the commentary referred to the simile as having an emphatic function (ἐμϕατικὴ [ἡ παραβολή, cf. sch. Il. 15.624–625, Od. 9.292) in relation to the ταραχή produced in the battle. 17 The ο of γενομένην is particularly narrow, unless there is a mistake. 17–18 In line 18 the phrase τοῖϲ κώλοιϲ surely belongs to the comment on κολοϲυρτόν of the second half of v. 147 (cf. sch. 12.147b κολοϲυρτὸϲ παρὰ τὸ κῶλα ἐπιϲύρειν, ἢ τὸν κολῳὸν καὶ ϲυριγμόν). As the lost part of v. 18 could contain from 12 to 15 letters, it does not seem possible to assume here another lemma. 11–18 Professor Pontani proposes the following reconstruction:

15

καὶ Ἰ̣[αμενὸν] κ̣αὶ̣ Ὀ̣ρέ̣[ϲτην Ἀϲιάδην τ᾿ Ἀδάμαντα Θ̣ό̣[ωνα τε Οἰ]ν̣ό̣μ̣α̣ό̣[ν τε· τούτουϲ ἀνωτέρωι παραλ̣[είπει· α]ὐ̣τῶν γὰρ τ̣[oὺc ἡγεμόvαc εἰπ̣ὼ̣ν̣ κ̣ατ᾿ ἄν̣[δρ]α̣, νῦν ἐν καιρῶ[ι πάvτας κατ᾿ ἀ ξι̣ωϲι̣ν ϲυ̣νεπ̣ή̣γ[α]γ̣εν αὐτοὺϲ τῶι [Ἀϲίωι. ἀνδρῶν ἠδ̣ὲ κυνῶν δέχα̣ται· ἐμϕατικὴ [ἡ παραβολή· νῦν γ̣ὰ̣ρ τὴν γενομένην ταραχὴν καὶ ἠ̣[χὼ κυϲὶν κατὰ γῇϲ τοῖϲ κώλοιϲ ϲυ̣ρ̣ο̣μ̣[ένοι]ϲ̣ π̣α̣ρ̣[έβαλεν.

139–140

19 τ᾿ ἀ̣ί̣ϲϲονται·. Of ⲁ and ι after the initial ⲧ extremely scanty traces survive. The following high stop suggests that we have a lemma from v. 148, with a mistake in the ending, τ’ ἀ̣ί̣ϲϲονται for τ’ ἀ̣ΐ̣ϲϲοντε (a simple phonetic exchange αι/ε), although at fr. 2+3, line 6, the high stop is used within the commentary. It is possible that the lemma started in the preceding line and included the word δοχμώ also from the same verse, cf. sch. 148a. Muratore suggests πλ̣[αγίωϲ ὁρμῶντεϲ, cf. sch. D Il. 12.148: δοχμώ τ’ ἀΐϲϲοντεϲ· εἰϲ πλάγιον ὁρμῶvτεϲ. Fr. 2 + 3 →

5

top (̣ )̣ ̣]εργαζονταικαιπαθ[ ̣[ ̣ ]̣ ταβολ ̣ ̣ ̣ ̣ ̣ νερ ̣ ̣ ̣ [ ̣ ̣ ̣ ̣ ̣ ( ̣)η ̣ ̣ [ ̣( ̣)] ̣ δ[ ̣] ̣ κνυ̣ταικαιτου ̣[ πιϲτινεμπο[ ̣ ̣]υ̣ϲιν·ιϲχονγαρπ̣[ ηϋτεπετρη·ημενεμβολητων̣[ ϲινεικα̣οταικαιανεμ̣ωι·ηδετων[ ] ̣τωικ ̣[

A New Papyrus Commentary on the Iliad  

10

15

] ̣λλεμ ̣[ ] ̣ ̣ροϲ·ενδε ̣ [ ̣ ̣]ωϲοτεκ ̣ [ ]τομηϕαναι̣ ̣ ̣ [ ] ̣ ϕοδρανεμϕ ̣ [ ]εωνανεμ ̣ ̣[ ]τωνακουον[ ] ̣ματωνκατ ̣[ ] ̣ϲεκτυπουα ̣ [ ]υμμεν ̣ [ ] ̣υ̣[

.

.

.

.

.

3 ̣ ̣ ̣ ̣ ̣ ( ̣), first, a tiny curve at line-level, possibly lower extremity of left-hand arc; second, two extremely scanty traces at line-level and in upper part of writing space (before them perhaps small lacuna of one letter), third, upper part of two uprights (possibly ⲛ, although very doubtful) η ̣ ̣, first, probably ⲑ̣, second, curve in upper part of writing space (suiting ⲉ) υ ̣, left-hand arc, perhaps ⲥ 8 ̣[, left-hand arc suggesting an oval letter like ⲉ, ⲟ, ⲥ; ⲫ not to be ruled out 17 ] ̣υ̣, horizontal stroke, perhaps ⲧ or ⲡ, followed by traces compatible with the upper part of a ⲩ, but without the usual ligature with ⲧ/ⲡ

5

10

15

top ̣( ̣) ̣] ἐργάζονται καὶ παθ[ ̣[με]τ̣αβολῃ ̣ ̣ ̣ ̣νερ ̣ ̣ ̣[ ̣ ̣ ̣ ̣ ̣( ̣)η ̣ ̣[ ̣( ̣)] ̣δ[ε]ί̣κνυται καὶ του ̣[ πίϲτιν ἐμπο[ιο]ῦ̣ϲιν· ἴϲχον γὰρ π̣[υργηδὸν ἀρηρότεϲ ἠΰτε πέτρη· ἡ μὲν ἐμβολὴ τῶν̣ [βαρβάρων κύμα ϲιν εἴκαϲται καὶ ἀ̣νέ̣μ̣ωι· ἡ δὲ τῶν [Ἑλλήνων καρτε ρία τῶι τῆϲ πέτραϲ ἀκιν]ήτωι κα̣[ὶ δυϲπαθεῖ. λαμπό μενοϲ πυρί· ]α̣λλεμ ̣[ ]π̣υ̣ροϲ· ἐν δ᾿ ἔπ̣[εϲ]᾿ ὡϲ ὅτε κῦ̣[μα θοῆι ἐν νηὶ πέ ϲηιϲι· εὖ δὲ] τὸ μὴ ϕάναι̣ ἐ̣π̣[ῆλθεν ] ̣ϕοδρανεμϕ ̣ [ ] ϲ̣ϕοδρὰν ἐμϕα̣[ιν- τὴν βίαν ὑπαὶ νεϕ]έων ἀνεμο̣τ̣[ρεϕέϲ ]των ακουον[ ] ̣ματων κατ ̣[ ] ̣ϲεκτυπουα ̣[ κεκαλ]υμμένῃ [ ]τ̣υ̣[ . . . . . . .

XV 610–614

618

623 624

625

1–4 The remains of the commentary certainly refer to 15.610–614, but the problem related to the expunction of these verses by Aristarchus is not treated. This problem is treated both in sch. a of Aristonicus (ἀθετοῦνται ϲτίχοι πέντε) and in sch. ex. b1 and b2. The sch. ex. defend the authenticity of those verses, especially on the basis of stylistic observations. Sch. b2 observes that the reiterated occurrence of Hector’s name at v. 610 conveys

  A New Papyrus Commentary on the Iliad more emphasis (and therefore is not a pointless repetition; cf. sch. a). Moreover, sch. b1 and b2 coincide almost completely in the last section, which concerns 15.612–614: καὶ ἡ πρόληψιϲ (sc. 612–614) δέ ἐϲτι ϲχῆμα ποιητικόν. προϲεκτικὸν δὲ ταῦτα τὸν ἀκροατὴν καὶ περιπαθέϲτερον ἐργάζονται. καὶ τοῦ λέγοντοϲ †ἦθοϲ χρηοτόν, καὶ ὡϲ (ἀπὸ add. b1) ϲυναλγοῦντοϲ ἡ πίϲτιϲ (b2, sim. b1). It cannot be known whether in the lost part of our hypomnema the problem of the expunction was treated (if it was treated, probably the expunction was rejected); we can observe, however, that 5095 contains observations on three points: (a) the prolepsis in 15.612–614; (b) the psychological impact of the passage on ἀκροατήϲ (cf. l. 1); and (c) the πίϲτιϲ produced by a speaker who sympathetically shares the sorrow (cf. l. 4) in a fuller form in comparison to the sch. This is significant because these points represent the arguments that in the sch. are used to reject the expunction by Aristarchus. The μεταβολή of l. 2 is not paralleled in the sch.: it could refer to the ‘change’ announced at vv. 612–614, where Hector’s imminent death is anticipated and announced, while Hector at this very moment is victorious thanks to Zeus’ support (cf. μινυνθάδιοϲ 612, ὀλιγοχρόνιοϲ sch. b1, p. 127, r. 39 Erbse).

1 Exempli gratia [προϲεκτικὸν δὲ ταῦτα τὸν ἀκροα]|τὴν] ἐργάζονται καὶ παθ[ητικώτερον (Lapini), if not a mistake for παθ[έϲτερον; cf. the sch. quoted above. 3 Possibly ἐ̣ [π]ι̣ δ[ε]ί̣ κνυται. 4 πίϲτιν ἐμπο[ιο]ῦ̣ϲιν. On this phrase cf. sch. ex. Il. 23.670. 4–7 The text can be reconstructed on the basis of sch. ex. 15.618b: ἡ μὲν ἐμβολὴ τῶν βαρβάρων κύμαϲιν εἴκαϲται ταχὺ διαλυομενοιϲ (sc. vv. 624–628), ἡ δὲ τῶν Ἑλλήνων καρτερία τῷ τῆϲ πέτραϲ ἀκινήτῳ καὶ δυϲπαθεῖ (sc. vv. 618– 622). The length of l. 5, which seems to be of 34 letters (plus high stop) against the 40 letters of l. 4 and the 38 letters of l. 6, is doubtful; but cf. above, fr. 1 ↓ 9 with comm. at ll. 9–11: here too l. 5 contains several broad letters (unless we insert νῦν before κύμαϲιν, as Pontani suggests). In 5095 as in sch. the comment at v. 618 parallels the two similes in a reverse sequence in respect to the poetic text. The simile at vv. 624–628 is then treated separately at ll. 9 ff. 5 A single dot above υ of ηυτε marking diaeresis. 7–9 On the basis of ]π̣υ̣ροϲ of l. 9 one can suppose a comment on v. 623 ἀύτὰρ ὃ λαμπόμενοϲ πυρὶ πάντοθεν ἔνθορ᾿ ὁμίλω: cf. sch. ex. 623 περιλαμπόμενοϲ ὑπὸ τῶν ὅπλων ὡϲ ὑπὸ πυρόϲ. 8 Although the body of ϕ is usually more rounded, it may be possible to read ἐμϕ̣[α-, with reference to the emphasis (cf. at l. 11 and fr. 1 ↓ 16) of the image of Hector joining the fight λαμπόμενοϲ πυρί. 9 Here is the join between the two frr.: ]ωϲοτεκ. [ belongs to fr. 2, which here breaks off, while ] ̣ ̣ ροϲ·ενδε.[ belongs to fr. 3, which includes the following lines. 9–12 Cf. sch. ex. 624–625 ἐν δ᾿ ἔπεϲ᾿ ὡε ὅτε κῦμα : πεπύκνωται ταῖϲ παραβολαῖϲ ὁ τόποϲ πρὸϲ ἔμϕαϲιν τῶν πραγ-

A New Papyrus Commentary on the Iliad  

μάτων. ἡ δὲ ἔν (624) ἐμϕαίνει τὴν βίαν τοῦ κνματοϲ. εἰϲ ἐπίταϲιν δὲ πρόϲκειται ἀνεμοτρεϕέϲ (625). εὖ δὲ τὸ μὴ ϕάναι ἐπῆλθεν ἢ ἐπέδραμεν, ἀλλ᾿ ἐπέπεϲε, ϲϕοδρὰν τὴν βίαν δηλῶν. Possible reconstruction exempli gratia: 10

ὑπὸ] π̣υ̣ρόϲ· ἐν δ᾿ ἔπ̣[εϲ]᾿ ὡϲ ὅτε κῦ̣[μα θοῆι ἐν νηὶ πέ ϲηιϲι· εὖ δὲ] τὸ μὴ ϕάναι ἐ̣π̣[ῆλθεν ἢ ἐπέδραμεν, ἀλλ᾿ ἐνέπεϲε] ϲ̣ϕοδρὰν ἐμϕα̣[ίνων τὴν βίαν τοῦ κύματοϲ ὑπαὶ νεϕ]έων ἀνεμο̣τⲧ[ρεϕέϲ

624

625

12 ff. Here the comment on the simile of the wave falling on the ship, with the description of the ship itself during the violent storm. Cf. sch. ex. Il. 15.6256a ὑπαὶ νεϕέων ἀνεμοτρεϕεϲ· : ηὐξημένον ὑπὸ ἀνέμου, ὡϲ εἶναι ὑποκάτω τῶν νεϕῶν καὶ πληϲιάζειν αὐτοῖϲ. οἱ δὲ ἀνεμοτρεϕὲϲ τὸ εὔτονον. καὶ Ϲιμωνίδηϲ (fr. 107 P. = PMG 612) “ἀνεμοτρεϕέων πυλάων” εἴρηκεν. ὁ δὲ κόμποϲ τῶν λεγομένων καὶ ὁ ψόϕοϲ τῶν ὀνομάτων οὐκ ἐᾷ ἰδεῖν τὴν ναῦν ἀϕρῷ κεκαλυμμένην. 13 ]τῶν ἀκουόν[των? 14 Possibly ὁ ψόϕοϲ τῶν ὀν]ομάτων. 15 Probably ] ̣ ϲε κτύπου αμ̣ [. Fr. 2 + 3 ↓ top

5

10

15

]αρετηνπροϲεκλογη[ ] ̣[ ̣]ονε ̣[ ̣] ̣ ̣ ωτοι ̣[ ] ̣τηντουτ̣οδε ̣ ̣ ̣[( ̣) ̣] ̣ ( ̣)·ϲτρ ̣ϕ̣ ̣ειϲ ]καινοϲοτροποϲ ̣ [ ̣( ̣)] αρ [ ̣] ιαλλαπτω ] ̣ εϲενουτοϲϲτραϕειϲγαρι̣[ ̣]α̣ϕυλα ]αϲεαυτονπ̣ ̣ ρ̣αποδιϲ̣ ̣ ει ̣ ̣ ̣ ̣ ̣ ̣ ] ̣ ϲκατα[ ]χ̣θ̣η·ειϲ[ ]γαρπρω[ ̣ (̣ ̣)] ̣ ̣ ̣ ̣ [ ̣ ̣] ̣[ ]ϲπροϲτωι ̣[ ]ωνεωρωντ[ ]μενπεριτω ̣ [ ]α[ ̣]οι·νεϲτω[ ]ων·εντοιϲ[ ]ι̣ϲ·διατου ̣ ̣[ ] ̣ταικαιτα̣ ̣ [ ]( ̣) ̣ ̣ ·καια̣ι ̣ [ ] ̣ ̣ ι̣τ̣ ̣ ̣ [ . . . .

  A New Papyrus Commentary on the Iliad 2 ⲉ ̣[, probably ⲁ ι ̣[, ⲉ rather than ⲥ 3 ] ̣τ, probably ⲉ ̣[, upper part of upright with thicker extremity ] ̣ ̣ , two traces in upper part of writing-space, of which the second is the upper part of a upright with left-hand blob: ⲛ or ⲁⲓ 8 χ̣θ̣, first, short oblique trace ascending from left to right and slightly protruding below the line level, should belong to the ascending diagonal of a ⲭ; second, the prolongation of the lower part of the descending diagonal of ⲭ into ⲑ as in fr. 1 ↓ 16 9 ] ̣ ̣ ̣ ̣, third, ⲇ? 10 ̣ ̣[, upright bearing a crossbar: ⲧ or ⲡ 16 ] ̣, upright ̣[, diagonal stroke descending from left to right 17 ]( ̣) ̣ ̣, two traces in the upper part of the writing space, compatible with the upper part of ⲛ, or with the right-hand half of a ⲁ followed by ⲓ, or with the central and right-hand part of ⲱ 18 ̣ι̣, possibly ⲉ or ⲑ; less probably ⲁ

5

10

15

top ]ἀρετὴν πρὸϲ ἐκλογὴ[ν( ̣)] ] ̣[ ̣]ονε ̣[ ̣] ̣ ̣ ωτοι ̣[ ̣ ̣] ] ̣την τουτ̣οδε ̣ ̣ ̣[( ̣) ̣] ̣( ̣)· ϲτρε̣ϕ̣θ̣είϲ· ] ̣καινὸϲ ὁ τρόποϲ ̣[ ̣( ̣)]αρ[ ̣]·ι αλλα πτω ] ̣εϲεν οὗτοϲ ̣ ϲτραϕεὶϲ γὰρ ἵ[ν]α̣ ϕυλά ξηται ϲυϲτείλ]αϲ ἑαυτὸν π̣α̣ρ̣αποδιϲθ̣εὶϲ̣ τ̣ῶ̣ι μ̣ή̣ κει τῆϲ ἀcπίδ]ọc κατὰ [τὴν ἄντυγα ὕπτιοc πάλτο ὅ ἐcτι κατηνέ]χ̣θ̣η· εἰc[ωποὶ ]γὰρ πρω[ ̣ ̣( ̣)] ̣ ̣δ̣ [̣ ̣ ̣] ̣[ ]cπροcτωι ̣[ ]ων εωρωντ[ ]μεν περι τω ̣ [ ]α[ ̣]οι· Νέcτω[ρ αὖτε μάλι cτα Γερήνιοc οὖροc Ἀχαι]ῶν· ἐν τοιc̣[ ]ι̣c· δια του ̣ ̣[ ] ̣ ται και τα̣ ̣ [ ]( ̣) ̣ ̣·καὶ α̣ἰδ̣[ῶ ] ̣ ̣ ι̣τ̣ ̣ ̣[ . . . . . .

XV 641–3? 645

653

659

661

1 The commentary would refer to 15.641 or rather to 641–643, where Periphetes, son of Copreus, is told to be much better than his father παντοίαϲ ἀρετάϲ: the sch. ex. criticize the simile for being so unfavourable to the father in comparison to the son (cf. sch. ex. 15.641ab). Line 1 should refer to the ‘choice’ based on the ἀρετή (on the phrase πρὸϲ ἐκλογὴν cf. sch. ex. Il. 18.490).

3 ].την. Read ἀρ]ε̣ τὴν? The middle stop before ϲτρεϕθεὶϲ is very faded. The lemma slightly protrudes in the margin to the right and is followed by a dot indicating lemma-end: therefore it consisted of a single word. Didymus’ sch. 645a, διχῶϲ δὲ τὸ ϲτρεϕθείϲ, διὰ τοῦ α καὶ διὰ τοῦ ε, provides evidence for the existence of the alternative reading ϲτραϕθεὶϲ (cf. West app. ad loc.: no MSS record this read-

A New Papyrus Commentary on the Iliad  

ing): unfortunately in 5095 of the letter between ρ and ϕ only very scanty traces survive, although they seem to be compatible with A. 3–8 The comment concerns 15.645 ff.: ϲτρεϕθεὶϲ γὰρ μετόπιϲθεν ἐν ἀϲπίδοϲ ἄντυγι πάλτο, τὴν αὐτὸϲ ϕορέεϲκε ποδηνεκε᾿, ἕρκοϲ ἀκόντων· τῇ ὅ γ᾿ ἐνὶ βλαϕθεὶϲ πέϲεν ὕπτιοϲ, ἀμϕὶ δὲ πήληξ ϲμερδαλέον κονάβηϲε περὶ κροτάϕοιϲι πεϲόντοϲ.

645

648

Periphetes (cf. 1 n.) ‘turning himself back, stumbled on the edge of his shield’, and therefore fell to the ground and was easily killed by Hector. The sch. ex. offer comparable elements: sch. 645b ϲτρεϕθεὶϲ γὰρ μετόπιϲθεν < --- πάλτο>: ὥϲτε οὐδὲ τοῦτο ἀϕ᾿ ἑαυτοῦ κατορθώϲαϲ ϕαίνεται ὁ Ἕκτωρ, τῇ δὲ τοῦ Περιϕήτου δυϲτυχίᾳ ϲυγκέχρηται· τὸ γὰρ πτῶμα ϲυνέπραξε πρὸϲ τὸ εὐχερῆ γενέϲθαι τὴν ἀναίρεϲιν; sch. 645c ὅπωϲ ϕυλάξηται ϲυϲτείλαϲ ἑαυτὸν ὑπὸ τὴν ἀϲπίδα; sch. 645e: ἐν ἀϲπίδοϲ ἄντυγι πάλτο: παραποδιϲθεὶϲ τῷ μήκει τῆϲ ἀϲπίδοϲ κατὰ τὴν ἄντυγα ὕπτιοϲ κατηνέχθη. πάλτο δὲ ἐνεπελάϲθη (cf. also sch. ex. Il. 4.462a ὕπτιοϲ γὰρ ἐνέπεϲεν κτλ.). At l. 4 Pontani suggests τοῦ θανάτο]υ̣ καινὸϲ ὁ τρόποϲ; an explanatory sentence could have followed (ὁ γὰρ would be natural, but it does not seem possible to reconstruct the remainder with certainty). ἀλλὰ πτώματι would be plausible, followed by a verb like κατέ]π̣εϲεν or a similar verb (cf. sch. D 4.493 ἤριπε: κατηνέχθη, ἔπεϲεν; 14.55 κατήριπεν: κατηνέχθη, κατέπεϲεν; 15.464 παρεπλάγχθη: παρηνέχθη, παρέπεϲεν), or by βίον ἀπώ]λ̣εϲεν (see below), and preceded by something contrasting the πτῶμα, as the cause of the warrior’s death: e.g. οὐκ ἀρᾶι ἀλλὰ πτώματι, perhaps with irony on the death of clumsy Periphetes (a most inescapable and fatal end, contrasting with a completely avoidable and banal obstacle; on the phrase cf. sch. ex. Il. 6.286 οὔτε γὰρ ϲυναινεῖν τῇ Ἀλεξάνδρου ἀρᾷ δύναται ὡϲ μήτηρ, οὔτε ἀπολογεῖϲθαι κτλ.). However, in view of the traces and space, it is not possible to read οὐκ before αρ. Moreover, ἀρᾶι is not frequently used. Alternatively, one may think of a verb before ἀλλά (ending in -ρει?) and indicating something (positive?) accomplished by the character, who, however, ingloriously dies because of his fall. An alternative interpretation is offered by Lapini: ἀλλ᾿ ἄπτω|[τοϲ ὢν ἂν ἢρ]κεϲεν οὗτοϲ (but it is unclear what preceded): in this case the comment would not be ironic, but Periphetes would be pitied as a warrior who has been much better than his father in his ἀρετή, and therefore would have been able to defend himself, if a cruel destiny had not reserved an inglorious death for him from a fall (a death that in any case brings to Hector κῦδοϲ ὑπέρτερον, v. 644). On the basis of the plausible πτώ|[ματι βίον ἀπώ]λ̣ εϲεν οὗτοϲ at 11.4–5, Pontani proposes three interpreta-

  A New Papyrus Commentary on the Iliad tions, ‘each unsatisfying for one reason or another’: (a) τοῦ θανάτο]υ̣ καινὸϲ ὁ τροποϲ· ο̣ [ὐ γ]ὰρ [δ]ι᾿ ἄλλ᾿ ἢ πτώ|ματι βίον ἀπώ]λ̣ εϲεν οὗτοϲ, which makes necessary the correction of αλλα into αλλη; (b) τοῦ θανάτο]υ̣ καινὸϲ ὁ τρόποϲ π̣αρ[ὰ] τἆλλα πτώ|ματα, ὧι βίον ἀπώ]λ̣εϲεν οὗτον, with a problematic reading π̣αρ[ after τρόποϲ; (c) τῆϲ μάχηϲ ο]ὐ̣ καινὸϲ ὁ τρόποϲ π̣αρ[ῆ]ν̣ , ἀλλὰ πτώ|ματι βίον ἀπώ]λ̣ εϲεν οὗτον, with the same difficulty as above. Finally, there remains the possibility of a lacuna in l. 4. The phrase καινὸϲ ὁ τρόποϲ at l. 4 is interestingly paralleled by sch. ex. Il. 16.594 ἐτράπετο: ἐδήλωϲεν αὐτὸ ἑξῆν διὰ τοῦ “ϲτρεϕθεὶϲ ἐξαπίνηϲ” (Π 598). καινὸν δὲ ὁ τρόποϲ· ὁ γὰρ ϕεύγων ἐπιϲτραϕεὶϲ κτείνει τὸν διώκοντα. Under the attack by the Acheans, the Trojans are withdrawing, but Glaucus at first ἐτράπετ(ο) and, while being chased and caught (15.598), suddenly turns back, ϲτρεϕθεὶϲ ἐξαπίνηϲ, and kills his pursuer. As in 15.645 ff., here too we find a chased warrior who turns back to face his pursuer: while in the first case Periphetes fails in his attempt, falls and is killed by Hector, in this case, on the contrary, Glaucus succeeds against the warrior Bacticles, who is going to catch him. The sch. remarks on the contrast, ὁ ϕεύγων ἐπιϲτραϕεὶϲ κτείνει τὸν διώκοντα. The two cases have in common the motive of a chased warrior who turns himself back and faces his pursuer: in 15.645 ff. one may say that the καινότηϲ pointed out by the ancient interpreters consists in the fact that the chased warrior is killed because of a clumsy fall in turning back, while in 16.593–598 the καινότηϲ consists in the fact that the chased warrior succeeds against his pursuer. 7–8 πάλτο or ἐπᾶλτο, followed by a gloss. Alternatively Pontani: κατὰ [τὴν ἄντυγα ἐνέπεϲεν καὶ|ὕπτιοϲ ἐϲϕά]χ̣θ̣η. After η there is a high dot, probably followed by a lemma, which must be εἰϲωποί of 15.653: εἰϲωποὶ δ᾿ ἐγένοντο νεῶν, περὶ δ᾿ ἔϲχεθον ἄκραι / νῆεϲ ὅϲαι πρῶται εἰρύατο. There are problems here with the precise identification of places and movements of warriors at the moment when the Trojans reach the Achean ships. The sch. offer not many suggestions for the reconstruction of these lines: sch. ex. 653 εἰϲωποὶ δ᾿ ἐγένοντο νεῶν: ὑπέϲτειλαν ἑαυτοὺϲ ὑπὸ τὰϲ ναῦϲ· εἰϲ γὰρ τὰ μεταξὺ διαϲτήματα ϕεύγουϲι, βραχὺ μέροϲ ὑπολειπόμενοι τῶν νεῶν ὡϲ τὰϲ πρύμναϲ αὐτοὺϲ ὑποβεβηκέναι. οἱ δὲ ὅτι ὑποχωρηϲάντων τῶν Ἑλλήνων ἐν ἀπόψει γέγονε τὸ πλῆθοϲ τῶν νεῶν τοῖϲ Τρωϲίν; sch. D εἰϲωποὶ δ᾿ ἐγένοντο νεῶν: ἐν ὄψει τὰϲ ναῦϲ ἔβλεπον, ὅ ἐϲτιν εἰϲῆλθον εἰϲ αὐτάϲ, τουτέϲτιν ὑπὸ τὴν ϲτέγην αὐτῶν ἐγένοντο. 9 In this line the two frr. join: the sequence ]γαρπρω[ belongs to fr. 2, which breaks off here; the following sequence ]. .δ̣ .[. .].[ belongs to fr. 3, which in-

A New Papyrus Commentary on the Iliad  

cludes the following lines. The sch. D ἄκραι: αἱ πρῶραι suggests τὰν] γὰρ πρώ[ραϲ. 10 Muratore suggests τὰ]ν πρὸϲ τῶι τ̣ [είχει]|[ναῦν; cf. sch. ex. Il. 14.31–32, concerning the topography of the Achaean camp (on which subject Aristarchus wrote a treatise): ἔϲχαται δὲ αἱ (sc. νῆεϲ) πρὸϲ τῷ τείχει. αἱ δὲ τοῦ Ἀγαμέμνονοϲ νῆεϲ πᾶϲαι πρὸϲ τῇ θαλάϲϲῃ ἦϲαν. 11 ]ων εωρωντ[. Read ἑωρῶντο or ἑώρων. Exempli gratia τ[ὰϲ πρύμναϲ πρώραϲ τῶν νεῶν. 13–17 The dots at 13 and 14 mark the beginning and the end of the lemma, which included all of v. 659. The comment here in 5095 was rather long, although the sch. preserve nothing (apart from sch. D οὖροϲ: νῦν ϕύλαξ). Perhaps it included also the following v. 660; cf. sch. ad loc. 14 Probably ἐν τοῖϲ̣ [. 15 Probably διὰ τοῦ. . [, preceded by the trace of a high stop: however, what precedes and what follows are not part of a lemma, and therefore the punctuation must articulate the commentary. 17 After the stop, a lemma from v. 661 beginns: ὦ ϕίλοι ἀνέρεϲ ἔϲτε καὶ αἰδῶ θέϲθ᾿ ἐνὶ θυμῷ, which is commented by sch. ex. 661–662ab; the same verse occurs in 15.561, commented by sch. Ge (Erbse ad loc.). Fr. 4 →

.

.

↓

] ̣[ ρεξα[ δι[ ̣ ̣[ . .

.

. ] ̣[ ] ̣νπε ]ρ̣ωεϲ ]τ̣οτα . .

Fr. 5 →

. . ]ϲιν̣[ bottom

↓

. . ] ̣υκ[ bottom

Fr. 4 → 1 ]., because of the loss of the upper layer of fibres only a very scanty trace is visible; apparently accidental ink to the left of the preserved text, not in alignment with it 3 . .[, first, a curve in upper part of writing space; second, the thick upper extremity of an upright

  A New Papyrus Commentary on the Iliad 2 Exempli gratia ῥεξα[ντι ἐοικώϲ 15.586 or ὀ]ρέξα[ι 15.596.

↓

3 Τ]ρ̣ωεϲ ? Fr. 5 ↓ 1 ]., horizontal stroke at mid-height, probably belonging to ε.

Fr. 6 →(?)

Fr. 7 →(?)

top ] ̣ ̣ ̣ ̣ ̣[ . .

. ] ̣ ϲ̣ .

Fr. 6 Negligible traces. No traces of ink on the other side. Fr. 7 On the other side very faded traces of ink.

. .

 Towards a New Critical Edition of the Scholia to the Iliad: A Specimen (with Fausto Montana, Davide Muratore, and Lara Pagani) Hartmut Erbse’s critical edition of the scholia vetera to the Iliad unquestionably provided classical philology with a tool that is excellent from two points of view: the quality of the text constitution, and the historical reconstruction of the traditional classes of the Iliad scholia, with the textual arrangement and layout consequently adopted in the edition.1 There cannot be any doubt that Erbse achieved a fundamental step forward in the study and editing of these materials. On the other hand, as is understandable, he did not include the entire range of materials we have at hand. It is generally known that Erbse restricted his edition to what he called scholia grammatica or maiora. This definition was intended to identify the most learned streams of textual criticism and exegetical work on the Iliad in antiquity, represented by two classes of scholia. Firstly, the class derived from a late-antique compilatory commentary, which in modern studies is conventionally called Viermännerkommentar (VMK), because it is traceable back to the writings pertaining to Aristarchean scholarship on the Iliad composed between the Augustan Age and the 2nd century AD by Didymus, Aristonicus, Nicanor and Herodian. The fundamental witness of this class in the available Medieval tradition is the renowned ms. “Venetus A” (siglum A).2 The second class edited by Erbse consists of the scholia exegetica (ex.), known by this name because they are characterized by an interpretive and content-related approach to the Homeric poem; their best witnesses are five mss. collectively designated with the sigla bT.3 Only limited traces of a third class of scholia can be found in Erbse’s edition. This class was wrongly known at the beginning as scholia Didymi and therefore over time has been conventionally referred to as D-scholia. The material

 1 Erbse 1969–1988. 2 Venezia, Biblioteca Nazionale Marciana, gr. Z. 454 (coll. 822). 3 T = London, British Library, Burney 86, known as ‘Townleianus’; b = the common ancestor of the four mss. Venezia, Biblioteca Nazionale Marciana, gr. Z. 453 (coll. 821) (B); El Escorial, Real Biblioteca, y. I. 1 (294 Andrés) (E3); El Escorial, Real Biblioteca, Ω. I. 12 (513 Andrés) (E4); and Firenze, Biblioteca Medicea Laurenziana, Laur. Plut. 32.3 (C). All these mss. are datable towards the 11th–12th century, but the Escorialensis E4 has to be dated at least two centuries later than the 11th in the view of Maniaci 2006, 222–223 n. 32. https://doi.org/10.1515/9783110772371-025

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen consists of glosses, paraphrases, critical questions (ζητήματα), mythographical explanations (ἱστορίαι), and summaries of the Iliadic books (ὑποθέσεις). It is impossible to determine the date and cultural background of the different components of these scholia, but at least the glossographic and the mythographic elements have important forerunners and equivalents in a rich array of papyri of the imperial age, which thus testify to their antiquity and widespread diffusion. Erbse’s choice of the D-material to be included was deliberately selective and incomplete, essentially limited to giving a summary account of the longest Dscholia in Venetus A and to recording D-glosses definitely present in the reconstructed archetype of mss. bT (named c in Erbse’s stemma). After the editio princeps by Janos Lascaris (1517) and some preparatory studies and unfinished attempts in modern times,4 a complete proecdosis of the D-scholia was carried out by Helmut van Thiel and is now available on the web,5 based on a limited number of witnesses and mainly on the (possibly overestimated) testimony of ms. Romanus-Matritensis (siglum Z in the proecdosis).6 In addition, and finally, Erbse substantially excluded from his edition the testimony of the rich and complex branch of manuscripts known as the hfamily. In this family, explanatory materials drawn from all three of the classes are merged together in a peculiar and often problematic amalgam, intermingled with further exegesis of yet unidentified origin.7 Erbse drew only to an extremely limited and instrumental extent on the h-family of manuscripts, to improve the text of the VMK-scholia.8 These circumstances suggest the need for availability of all the classes of scholia within the same edition, as well as an extensive scrutiny of the manuscript tradition in order to recover the entire range of scholia.9 For this reason, our team — currently benefiting from the collaboration of Francesco Plebani —

 4 Schimberg 1892; de Marco 1932; 1941; Montanari 1979, 3–25, 43–75. An edition of the D-scholia was undertaken, but not finished, by V. de Marco. 5 Van Thiel 2014 (first release 2000). 6 Roma, Biblioteca Nazionale Centrale, gr. 6 + Madrid, Biblioteca Nacional de España, 4626, of the 9th century. 7 Information about the family in Vassis 1991; Sciarra 2005; Muratore 2014 (especially Appendix 2). 8 Erbse 1969–1988, 1, LVIII: Sequitur, ut scholia classis h maximi momenti sint, ubicumque codex A mendis et lacunis corruptus est. Cf. Erbse 1960, 207–208. 9 This delimitation implies the exclusion of more recent material of known origin, such as the Moschopoulean scholia or other annotations which recent hands of the manuscripts (e.g. E4, T) have drawn from extra-scholiastic sources (for example Eustathius’ commentary). In contrast, Porphyrian excerpta are included, as in Erbse’s edition.

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

has undertaken the project of a new and more comprehensive edition of the Iliad scholia.10 In particular, as far as the VMK-scholia and scholia exegetica are concerned, explorations carried out so far on the witnesses reveal that in some cases Erbse’s monumental work was not free from imperfections.11 In addition, in recent decades specialist research on ancient and Byzantine scholarship in general and on Homeric ancient exegesis in particular has taken enormous steps forward: critical-textual and interpretive contributions have multiplied, resulting in the need to extend and update text and apparatuses, as well as the bibliography, of the edition. Furthermore, it would be desirable for the complete edition of the D-scholia to be integrated, both by proceeding to a fresh collation and recensio of the witnesses used by modern scholars in their preliminary studies and ecdotic works, and also by systematically utilizing the evidence of manuscripts Venetus A and bT. Again, the impact of the h-family of manuscripts can no longer be underestimated, in order to acquire its contribution to the text constitution of the three classes as well as to ascertain the very nature of its extra-material. This is the most intriguing side of our on-going investigation. Finally, the close relations that can be recognized between the paraphrases present in the manuscripts and the different classes of scholia, especially the D-glosses, suggest that this particular and widespread form of exegetical product should also be taken into consideration as an important component of the exegetic mosaic itself.12 Another important element which will be part of the edition is represented by the Homeric lexicon known as Lexeis Homērikai (a kind of arrangement in alphabetical order of the glossographic component of the Iliad scholia).13 Moreover, Erbse included the exegesis on papyrus, reserving an independent space for it at the beginning of the scholia to each Iliadic book. The increase, during recent decades, of the papyrological findings that provide Homeric exegesis, as well as the substantial difference of “genre” between hypomnēmata and exegetical marginalia, on the one hand, and the scholia transmitted in the Medieval manuscripts, on the other, both advise against the inclusion of this  10 Cf. Montana 2020. 11 See Muratore 2012 and Montana 2018. As far as the collation of the manuscript witnesses is concerned, one should bear in mind the technological conditions in which Erbse was operating: he was working on reproductions (essentially xerox copies, at that time) and it was only as a second step that he checked the individual doubtful points by an autoptic examination of the originals. 12 See Muratore 2014 and 2016. 13 The edition of this material was undertaken by V. de Marco, but only the first volume was published: de Marco 1946. A proecdosis is available on the web: van Thiel 2005.

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen kind of material into the edition of scholia. All the more so, when its edition and explanation are planned in a special volume within the series Commentaria et Lexica Graeca in Papyris Reperta.14 The present specimen is firstly intended to give a sample of the current stage of the inquiry and to point out the resulting substantial difference between Erbse’s and the planned edition, in terms of both amount and type of edited text. Erbse’s and the new edition are synoptically displayed in the specimen, respectively on the left-hand and right-hand pages, in order to show the difference. In addition to the classes of scholia pertaining to VMK (Ariston., Did., Hrd., Nic.) and to the exegetica (ex.), the new edition will also host the critical text of the D-scholia (whose manuscript witnesses we have designated collectively as the d-family) and of other annotations which are found only in the hfamily. It is often difficult to establish whether the extra-material present in h was simply copied from sources foreign to the scholiastic tradition (typically from the Epimerismi Homerici) and interpolated into a mix of the three classes of scholia established in modern scholarship (VMK, ex., D), or whether, on the other hand, it represents the outcome of a genuine reworking and metabolization of exegetical material carried out in order to achieve a new and more comprehensive compilation or corpus of scholia. This distinction is by no means nominalistic: on this basis, we should consider (at least a part of) the amalgamation attested to by the h-family as a veritable further class of scholia. For this reason, whenever the extra-material in h is not literally traceable back to extrascholiastic sources, we would be inclined to include it in the edition and mark it as h in the margin. This would therefore mean: exegetical material present in h-manuscripts and not found in the three established classes, which could therefore be promoted to the dignity of a further class of scholia in its own right. All this is work in progress. Many aspects, especially concerning the h-manuscripts, have not yet been exhaustively explored, nor exactly defined or completely ascertained.15 This is why in the specimen the sigla of the collated mss. are listed analytically and singularly at the end of each scholium and in the critical apparatus. The progress of the collations will lead to more precise clues of kinship between the manuscripts and, as a consequence, will allow groups of manuscripts to be gathered together under collective sigla. For example, the manuscripts Roma, Biblioteca Angelica, gr. 122 (Ag), Cologny, Fondation  14 Cf. Pagani 2020. 15 Among others, the textual testimony of ms. Genavensis gr. 44 (Ge) is absent from the specimen, because — pace Erbse 1969–1988, 1, LVII — its traditional position needs further examination.

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

Martin Bodmer, 85 (Bd), and Paris, Bibliothèque Nationale, gr. 2556 (P) clearly form a consistent subgroup of witnesses within the h-family.16 The following are other features of the specimen and/or the new edition. The specimen does not include the source apparatus either of Erbse’s or of the new edition. The source apparatus, duly enriched and updated, will be present in the edition. Scholia of different classes concerning the same Homeric lemma are ordered as follows: VMK (Ariston., Did., Hrd., Nic.), ex., D, h. This rule is not imperative and is set aside whenever the manuscript tradition displays significant setups which compel the editor to adopt a different order. For instance, the sequence of the scholia to Il. Α 11c–f in the specimen (D, h, ex. | ex., h | h) mirrors their actual setup and ‘organic’ texture in the h-family, where they represent a compilatory unit endowed with a consistent exegetical sense. The different components of the D-scholia are signalled in the outer margin of the pages as follows: D = glosses and paraphrases; Dhist = historiai; Dhyp = hypotheseis; Dzet = zētēmata. The lemmata or parts of them which are not present in the manuscript tradition have been placed in italics. Erbse treated these additions as critical integrations of textual omissions, and therefore included them between < >, recording the modern authorship of the integration in the critical apparatus, e.g. (sch. A 8a. Erbse) “le. add. Frdl.”. In contrast, we are inclined to interpret the behaviour of the manuscripts in this respect as generally reflecting a choice on the part of the compiler; so we add (parts of) lemmata in italics not as textual restoration, but as an editorial aid for the reader, in consideration of the absence of the Homeric text and of other kinds of cross-references in a modern edition of scholia. Given the unceasing manipulation over time of some types of ancient annotation, typically the glossographic component of the D-scholia, it is often questionable whether a simple gloss transmitted by only one or a pair of manuscripts of a family is actually attributable to its common ancestor or is rather a posterior addition. For this reason, the siglum d placed at the end of a glossographic D-scholium must be understood not strictly as a reference to the “archetype” of the D-scholia, but rather as a conventional indication of its belonging to this family of manuscripts. The same is true for isolated annotations present only in particular branches or single manuscripts of the h-family. As in Erbse’s edition, ms. Leipzig, Universitätsbibliothek, Cod. gr. 32 (olim 1275) (Li) is not listed among the witnesses in the text at the end of the scholia

 16 Cf. Sciarra 2005, 111–128.

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen exegetica, because for the relevant part it is a copy of B(E3)T.17 Its readings are quoted in the critical apparatus only when the writing of the mentioned manuscripts is damaged or illegible. Precise documentary evidence, confirmed by examination of the text, has made it possible to eliminate a so far poorly studied witness of the D-scholia, namely ms. Venezia, Biblioteca Nazionale Marciana, gr. App. IX, 5 (coll. 1336) (U18) inasmuch as it is a copy of ms. Città del Vaticano, Biblioteca Apostolica Vaticana, gr. 33 (Q).18 In contrast, use is made in this specimen of ms. Heidelberg, Universitätsbibliothek, Pal. gr. 222 (Pal2), which is certainly closely related to Q, although further investigation is required in order to give a more precise definition of the relationship. With regard to Lascaris’ edition, we now have knowledge not only of the model on which it is based (Vat. gr. 33, Q), but also of the main sources from which it drew the greater part of the additional material it offers.19 Its readings should thus be considered as devoid of independent traditional value and should be expunged from the critical apparatus. However, on account of their historical significance, they are nevertheless cited in all cases where they diverge from the reading of Q. Finally, it is worth recalling that among the witnesses of the b branch that are mentioned above, n. 3, and in the Conspectus siglorum, there are two — Laur. Plut. 32.3 (C) and Escorial. gr. y. Ι. 1 (gr. 294 Andrés) (E3) — which lack scholia vetera to the section of the poem chosen for the present specimen (Α 6–11).20

 17 Maass 1884; Erbse 1969–1988, 1, XXIV and XLVIII with n. 65. 18 Muratore 2018/b. 19 Muratore 2018/a. 20 F. 1 of the Laurentian manuscript houses, instead of the expected scholia exegetica to Il. Α 5–12, excerpta from John Tzetzes’ Exegesis of the Iliad concerning the same lines of the poem (cf. Montana 2018). F. 1 (Il. Α 1–28) of the Escorialensis is a later restoration (15th cent.) of a material loss, lacking scholia (cf. Erbse 1969–1988, 1, XIX).

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

Conspectus siglorum A Ag B Bd Bm12 C E3 E4 La Li M1 M11 P P11 Pal2 Q T V13 V19 W3 X Y Z

Marc. gr. Z. 454 (822) Angelicus gr. 122 Marc. gr. Z. 453 (821) Bodmer. 85 Lond. Harl. 5727 Laur. Plut. 32, cod. 3 Escorial. gr. y. Ι. 1 (gr. 294 Andrés) Escorial. gr. Ω. Ι. 12 (gr. 513 Andrés) sch. D ed. princeps Lascariana, Roma 1517 Lips. gr. 32 (1275) Ambr. A 181 sup. (gr. 74) Ambr. L 116 sup. (gr. 502) Par. gr. 2556 Par. gr. 2766 (P11a = scholia f. 7rv) Heidelb. Pal. gr. 222, ff. ιζr–λδv Vat. gr. 33 Lond. Burn. 86 (Townleianus) Vat. gr. 1316 Vat. Pal. gr. 6 Vratislaviensis Rehdiger. 26 Vat. gr. 32 Vat. gr. 2193 Romanus Bibl. Nat. gr. 6 + Matrit. Bibl. Nat. 4626

b d h

BCE3E4 E4Pal2QXYZ AgBdP Bm12M1M11P11V13V19W3

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen

Erbse’s edition

ex.

6 < δ ι α σ τ ή τ η ν >: διέστησαν. ἡ δὲ τῶν δυϊκῶν κατάχρησις Ἀττική. τὸ δὲ “ἔριδι” (Α 8) ἀντὶ τοῦ μετὰ ἔριδος, ὡς μεθ᾿ ὅπλων. b(BE4)Tt

ex.

7 a. Ἀ τ ρ ε ί δ η ς: ἀντονομασία ὁ τρόπος ἀντὶ τοῦ Ἀγαμέμνων, ὡς καὶ ἡ Τριτογένεια ἀντὶ τοῦ Ἀθηνᾶ. ἐπεὶ δὲ Ἀτρείδης καὶ τὸν Μενέλαον ἐδήλου, ἀντιδιαστέλλων προσέθηκεν τὸ ἄναξ ἀνδρῶν. b(BE4)T Ἀ τ ρ ε ί δ η ς τ ε ἄ ν α ξ : Ἀγαμέμνων – – – – – Ἰφιγένειαν καὶ Ἠλέκτραν. A

D

55

60

55 le. suppl. Ddf. 58 ἀτρείδης — τρόπος om. b ]σ. ἀντονομασία[.......] T suppl. Ma. e Li 59 ὡς κ[.......]γένεια ἀ[..........]ᾷ T suppl. Ma. Li (ubi ὡς κ. ἡ τρ. ἄλλο τι τῆς ἀθηνᾶς), ὡς ἀγελείη καὶ τριτογένεια ἀντὶ τοῦ (Ag, τριτ. ἡ b) ἀθηνᾶ Ag b ἐπεὶ δὲ T ἐπεὶ γὰρ τὸ b 60 μενέλεων Li (fort. e T) ἀντιδ. Li (fort. e T), ἀντὶ τοῦ (τοῦ om. Β) ἀγαμέμνονος b τὸ om. Li

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

The new edition 6 a. ἐξ οὗ δή: ἀφ᾿ οὗ δὲ χρόνου. d(Pal2QYZ) h(AgslM11W3) 2

5

D 3

b. τὰ πρῶτα: τὸ πρῶτον καὶ τὴν ἀρχήν. d(Pal QYZ) h(W ) κατὰ τὴν ἀρχήν h(Agsl) c. διαστήτην: διέστησαν καὶ διεχωρίσθησαν. d(QYZ) h(AgBdP M11slW3)

D

d. διαστήτην: διέστησαν. ἡ δὲ τῶν δυϊκῶν κατάχρησις Ἀττική. τὸ δὲ “ἔριδι” (Α 8) ἀντὶ τοῦ μετὰ ἔριδος, ὡς μεθ᾿ ὅπλων. b(BE4)Tt h(AgBdP M1) e. ἐ ρ ί σ α ν τ ε : φιλονεικήσαντες. d(Pal2QYZ) h(AgslW3)

ex.

7 a. Ἀ τ ρ ε ί δ η ς : ὁ Ἀτρέως παῖς Ἀγαμέμνων. Ἀγαμέμνων κατὰ μὲν Ὅμηρον Ἀτρέως τοῦ Πέλοπος, μητρὸς δὲ Ἀερόπης, κατὰ δὲ Ἡσίοδον Πλεισθένους, τὸ γένος Μυκηναῖος, ὃς ἤγαγε ναῦς εἰς Ἴλιον. ἐκπορθήσας δὲ τὴν Ἴλιον καὶ ὑπο10 στρέψας οἴκαδε ἀναιρεῖται ὑπὸ Αἰγίσθου τοῦ Θυέστου δόλῳ ἐπὶ εὐωχίας. οὗτος γὰρ παρὰ τὸν καιρὸν τῆς ἀποδημίας ἐμοίχευσε τὴν Ἀγαμέμνονος γυναῖκα Κλυταιμήστραν. κατὰ δὲ τοὺς τραγικοὺς αὐτὴν τὴν Κλυταιμήστραν ἀνελεῖν αὐτὸν χιτῶνα μὴ ἔχοντα διέκδυσιν τραχήλου δοῦσαν αὐτῷ. ἔσχε δὲ ἐξ αὐτῆς υἱὸν μὲν Ὀρέστην καὶ θυγατέρας τέσσαρας, Λαοδίκην, Χρυσόθεμιν, Ἰφιγένειαν καὶ 15 Ἠλέκτραν. A d(E4Pal2QYZ) h(AgBdP Bm12M11V13V19W3) b. Ἀ τ ρ ε ί δ η ς τ ε , μία ἡ ὀξεῖα· σπονδειακὸν γάρ· χωρὶς εἰ μὴ ἐπιφέροιτο ἡ διὰ τοῦ σφῶν ἀντωνυμία· ὡς ἐπὶ τοῦ “τόξού σφεων·” (Callim. fr. 560 Pf.) “ἀλλ᾿

D

D Dhist

h

1 ἀφ᾿οὗ et M11sl 2 τὰ om. Pal2Q 4 διαστήτην le. Z διέστησαν deest in QLa καὶ om. LaM11sl διεχωρίσθησαν et Agsl 5 δὲ1 om. h Ἀττική: μάλιστα πάντων τοῖς Ἀττικοῖς ἁρμόζει h (= Epim. Hom. A 6B, r. 60 Dyck) 5-6 τὸ δὲ ἔριδι—ὄπλων deest in h 6 τοῦ—ὅπλων evan. in B 7 ἐρίσαντες Z, ἐρείσαντε Pal2 φιλονικ- Z 8 Ἀτρείδης τε ἄναξ le. A, τ᾿ post Ἀτρείδης add. Zrec, le. deest in E4 Bm12M11 ὁ Ἀτρέως παῖς Ἀγαμέμνων om. A E4 Bm12M11V13V19W3 Ἀγαμέμνων1 om. AgBdP Ἀγαμέμνων δὲ κατὰ μὲν Q 9 Ἀτρέως τοῦ Πέλοπος: πατρὸς μὲν Πέλοπος W3 τοῦ: δὲ La μητρὸς δὲ οἱ AgBdPac ἀερώπης Z, Ἀερόπῳ E4pc Πλησθ- V19 10 Μυκηναῖος: Μυκυναῖος V13, Μηκυναῖος W3 ἤγαγε: -γεν BdP V13Μ11 ναῦς: ασν΄ ναῦς Bm12M11 (cf. Dictys FGrHist 49 F 4 = Malalas Chron. 5, 11 p. 80.92 Thurn), ασν΄ ἄλλοι δὲ λέγουσιν ἀγαγεῖν αρξγ΄ ναῦς AgBdP, ας΄ ναῦς Schimberg 1891, 22–23 coll. Thuc. 1, 10, 4 Ἴλιον: ἥλιον V19 11 Αἰγίστου AgBdP θυέσθου Z ἐπὶ εὐωχίας om. E4, ἐπ᾽ 2 13 19 εὐ- Y, εὐωχίας ἔπι Pal Q, om. V V : fort. ἐπ᾽ εὐωχίαις de Marco 12 κατὰ τὸν καιρὸν AgBdP Bm12pcM11, κατὰ καιρὸν Bm12ac ἐμοίχευσεν V13, ἐμοίχευε Pal2QY Bm12M11W3 Ἀγαμέμνωνος vid. Zac 13 Κλυταιμήστραν: -μνη- Pal2QY Bm12M11V13V19W3 (semper, non amplius notatur), Κλυτεμνή-στραν ... Κλυτεμνήστραν Α, Κληταιμνήστραν ... Κληταιμνήστραν BdP αὐτὴ ἡ Κλυταιμνήστρα ἀνεῖλεν La ἀνελεῖν: ἀναιλεῖν W3 14 χιτῶνα μακρὸν ἔχοντα AgBdP, χιτῶνι μὴ ἔχοντι Pal2Q V19, χιτῶνι μὴ ἔχον V13 χειτῶνα vid. Apc, χειτόνα Z, χιτῶνι E4 διέκδυσιν: ἔνδυσιν Q, ἔνδυσι V13V19ac τραχίλου Z δοῦσαν αὐτῷ AgBdP Bm12M11 van Thiel (e M11), δοὺς αὐτῷ ut vid. W3, ἐμβαλὼν αὐτῷ παρὰ τὸν πότον Y, unde ἐμβαλοῦσαν de Marco (auct. Schimberg 1890, 435) ἔσχεν Z ἐξ αὐτῆς δὲ ἔσχεν Pal2Q V13 ἐξ non liquet W3 (fort. ἀπ᾿) 14-15 υἱὸν μὲν: υἱὸν τὸν Pal2Q V13 15 Ὀρέστιν vid. Z Ἰφιγένειαν: ante Χρυσόθεμιν transp.

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen V19, Ἰφιάνασσαν Bm12M11 17-3 post sch. D ad A 14 στέμματ᾿ ἔχων ἐν χερσίν P11a σπονδιακόν Ag, σπονδεῖος P11a 18 ἡ διὰ τοῦ σφων: διὰ τῆς φύσεως P11a τόξου BdP P11a

ex.

b. δ ῖ ο ς ὁ ἔνδοξος ἀπὸ τῆς Διὸς ὑπεροχῆς. ποσαχῶς δὲ τὸ δῖος, φησὶν ἡ κατὰ συλλογισμὸν διήγησις· Ἀχιλλεὺς μὲν γὰρ πάντων χάριν,Ὀδυσσεὺς δὲ φρονήσεως, Ἀλέξανδρος δὲ καὶ Κλυταιμνήστρα κάλλους, Εὔμαιος εὐνοίας, θάλασσα καὶ Χάρυβδις μεγέθους. τὸν δὲ Ἀχιλλέα καὶ διογενῆ δίχα τοῦ κυρίου: “αὐτὰρ ὁ διογενής” (Φ 17). b(BE4)T

ex.

8-9 ἐπὶ τὸ διηγηματικὸν μετιὼν οὐχ ὑποβάλλει τὰς διηγήσεις αὐτομάτους, ἵνα μὴ δοκῇ τοῖς ἀκούουσι προσκορὴς εἶναι, ἀλλὰ διὰ πεύσεως καὶ ἀποκρίσεως τὴν πραγματείαν πεποίηται, τὸν τῶν ἀκροατῶν νοῦν ἀναρτῶν καὶ ὑψῶν αὐτοὺς πρότερον τῇ πεύσει, εἶτα τὴν ἀπόκρισιν ἐπάγων. ἐν δὲ τῇ εἰσβολῇ τὸν λόγον ἀναρτήσας φιλοτέχνως διηγεῖται τὴν μῆνιν. τοιοῦτος δέ ἐστι, κεφαλαιώδεις τινὰς ἐκδιδοὺς περιοχὰς καὶ ἐξ ἀναστροφῆς κατὰ μέρος διηγούμενος. b(BE4)

17

65

70

75

64-8 δῖος — διογενής incertiss. T; primo quinque fere versuum litterae confusae vel detritae; tum haec ] ... ης σῶμα[, tum ] καὶ πάντων. . ριν [, denique ] πᾶσι, postremo unus versus obscuratus; e cod. Li, textum B referente, haec suppleri nequeunt 65 sq. πάντων χάριν b χάριν συνέσεως Eust. 66 φρονήσεως b συνέσεως Eust. 70 sq. le. addidi 71 οὐχ Ag, om. b 73 πεποίηται E4, ἐπιδείξη B (lectio incertissima) 74 αὐτὸν E4

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

ἤτοί σφεας κεῖθεν” (ν 276) τό τε “λοέσσαί τε· χρῖσαί τε·” (τ 320) καὶ “γενέσθαί τε· τράφεμέν τε” (H 199)· καὶ τὸ μὲν λοέσσαί τε ἵνα ἀποφύγωμεν τὸ πληθυντικὸν ῥῆμα· τὸ δὲ γενέσθαί τε ἵνα τὸν διπλασιασμόν. h(AgBdP P11a) c. Ἀτρείδης: ἀντονομασία ὁ τρόπος ἀντὶ τοῦ Ἀγαμέμνων, ὡς καὶ ἡ 5

10

15

20

25

Τριτογένεια ἀντὶ τοῦ Ἀθηνᾶ. ἐπεὶ δὲ Ἀτρείδης καὶ τὸν Μενέλαον ἐδήλου, ἀντιδιαστέλλων προσέθηκεν τὸ ἄναξ ἀνδρῶν. b(BE4)T h(AgBdP) d. ἄναξ: βασιλεύς. d(Pal2QYZ) e. καὶ δῖος Ἀχιλλεύς: δῖος ὁ ἔνδοξος ἀπὸ τῆς Διὸς ὑπεροχῆς. ποσαχῶς δὲ τὸ δῖος, φησὶν ἡ κατὰ συλλογισμὸν διήγησις· Ἀχιλλεὺς μὲν γὰρ πάντων χάριν, Ὀδυσσεὺς δὲ φρονήσεως, Ἀλέξανδρος δὲ καὶ Κλυταιμνήστρα κάλλους, Εὔμαιος εὐνοίας, θάλασσα καὶ Χάρυβδις μεγέθους. τὸν δὲ Ἀχιλλέα καὶ διογενῆ δίχα τοῦ κυρίου· “αὐτὰρ ὁ διογενής” (Φ 17). b(BE4) h(M1P11 ad A 292)

ex.

f. καὶ δῖος Ἀχιλλεύς: θεῖος, εὐγενής, ἢ ἀπὸ Διὸς ἔχων τὸ γένος, ἔντιμος· ἀπὸ γὰρ Αἰγίνης τῆς Ἀσωποῦ τοῦ ποταμοῦ Θηβῶν Αἰακός, Αἰακοῦ δὲ Τελαμὼν καὶ Πηλεύς, Πηλέως δὲ Ἀχιλλεύς. d(E4Pal2QYZ) h(AgBdP Bm12V13V19)

D

g. δῖος: τὸ δι μακρόν, ἀπὸ τοῦ Ζεύς, ὃ κλίνεται Διός· καὶ ἐξ αὐτοῦ δίιος· ὁ ἀπὸ τοῦ Διὸς καταγόμενος· καὶ κράσει τῶν δύο ιι εἰς ι μακρόν, δῖος. h(AgBdP Bm12M11V13) 8-9 a.1 τίς τάρ σφωε θεῶν ἔριδι ξυνέηκε μάχεσθαι; / Λητοῦς καὶ Διὸς υἱός: ἐπὶ τὸ διηγηματικὸν μετιὼν οὐχ ὑποβάλλει τὰς διηγήσεις αὐτομάτους, ἵνα μὴ δοκῇ τοῖς ἀκούουσι προσκορὴς εἶναι, ἀλλὰ διὰ πεύσεως καὶ ἀποκρίσεως τὴν πραγματείαν πεποίηται, τὸν τῶν ἀκροατῶν νοῦν ἀναρτῶν καὶ ὑψῶν αὐτοὺς πρότερον τῇ πεύσει, εἶτα τὴν ἀπόκρισιν ἐπάγων. ἐν δὲ τῇ εἰσβολῇ τὸν λόγον ἀναρτήσας φιλοτέχνως διηγεῖται τὴν μῆνιν. τοιοῦτος δέ ἐστι, κεφαλαιώδεις τινὰς ἐκδιδοὺς περιοχὰς καὶ ἐξ ἀναστροφῆς κατὰ μέρος διηγούμενος. b(BE4)

h ex.

1 κεῖθεν om. P11a τό τε: καὶ χωρὶς τοῦ P11a λοέσσαί τε: λοέσσαι τε vid. P11a (male leg. in imo 2 καὶ τὸ μὲν: τὸ μὲν γὰρ P11a marg. int.), λοέσσεταί τε AgBdP τοῦ ante γενέσθαι add. P11a λοέσσαί τε: λοέσσαι τε P11a, λοέσσεται AgBdP πληθυντικὸν: ποῦ εὐκτικὸν P11a 3 ἵνα τὸν διπλ.: ἵνα μὴ ἔσται διπλ. τὸ τραφέμεν P11a 4 Ἀτρείδης—τρόπος om. b ]σ. ἀντονομασία[.......]ς ἀντὶ τοῦ T suppl. Maass e Li πατρωνυμικὴ ἀντωνομασία h Ἀγαμέμνονος h 4-5 ]ων ὡς κ[.......]γένεια ἀ[. ]ᾷ T suppl. Maass e Li (ubi ὡς κ. ἡ τρ. ἄλλο τι τῆς ἀθηνᾶς), ὡς ἀγελείη καὶ τριτογένεια ἀντὶ τοῦ (h, τριτ. ἡ b) ἀθηνᾶ b h 5-6 ἐπεὶ—ἀνδρῶν deest in h 5 ἐπεὶ γὰρ τὸ b μενέλεων Li (fort. e T) 6 ἀντιδ. Li (fort. e T), ἀντὶ τοῦ (τοῦ om. Β) ἀγαμέμνονος b -θηκε B τὸ om. Li 7 ἀνδρῶν in le. et τοῦ λαοῦ in gl. add. Zrec (om. Pal2Q, de Marco), ἄναξ ἀνδρῶν le. van Thiel 2000, 2014 8-10 δῖος—φρονήσεως: ἐξόχως τῆς ἀπὸ τοῦ Διὸς εὐγενείας χάριν δῖος ὁ Ἀχιλλεύς· ὁ μέντοι Ὀδυσσεὺς δῖος χάριν συνέσεως h 10 φρονήσεως b συνέσεως Eust. δὲ2 om. h 11 τὸν δὲ: τὸν μέντοι h 12 κυρίου: κυρίως καλεῖ h δόρυ μὲν λίπεν αὐτοῦ ἐπ᾿ὄχθῃ post διογενής add. h 13 le. om. V13V19 in le. καὶ et Ἀχιλλεύς om. Pal2QY AgBdP Bm12M11 13-14 13 19 θεῖος—ἔντιμος om. V V 14 ἔντιμος om. E4, ἔντιμον Bm12ac Ἀσωποῦ: Αἴσωπος AdBdP τοῦ om. AgBdP Αἰακός om. Bd 15 Ἀχιλεύς BdP Bm12, καὶ Ἀχιλλεύς V19 16-17 ante sch. 7a. Bm12M11V13; sim. Epim. Hom. 7 D2 Dyck (ex Hdn.?), cf. Dyck 1981 16 γὰρ post ἀπὸ add. V13 ὃ om. V13 17 γίνεται ante δῖος add. V13 19-25 sch. ex. 8c. proxime seq. in b 20 οὐχ om. b, suppl. Erbse coll. Ag (sch. a.2) 22 πεποίηται: πε[ vel πεπ[ B (lectio incertissima: ἐπιδείξη leg. Erbse) 23 αὐτοὺς: αὐτὸν E4

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen

Ariston. ex.

Hrd. ex.

D

8 a. ὅτι Ζηνόδοτος “σφωι” ἔγραφεν. Aim b. σ φ ω έ: οἱ περὶ τὸν Σιδώνιον τῇ “” ἀρέσκονται, οὐ καλῶς· ἡ γὰρ σφῶϊ δύο πτώσεις ἐπιδέχεται, ὑμεῖς τε καὶ ὑμᾶς, ὧν οὐδέτερον ἁρμόζει. τὸ δὲ μᾶλλον τρίτου ἐστὶ δυϊκοῦ προσώπου καὶ σημαίνει τὸ αὐτούς. b(BE4)T c. δασύνεται τὸ ξυνέηκε. Aint d. ἔ ρ ι δ ι ξ υ ν έ η κ ε μ ά χ ε σ θ α ι : ὡς δι᾿ ὅ γε Τρωσὶ μάχεσθαι. T

9 Ζεὺς ἐρασθεὶς – — — — — βάσεως ἐρριζώθη. A

80

85

78 le. add. Frdl. ὅτι A, fort. ἡ διπλῆ (sc. περιεστιγμένη), ὅτι σφῶϊν A corr. Lehrs 79-82 pone sch. A 8–9 (coni. cum ν. διηγούμενος) in b 79 le. scripsi, τίς τ᾿ ἄρ: T, om. b oἱ μέν τοι περὶ b σιδόνιον E4 σφῶϊ bT emendavi (cf. test.) 80 ἡ γὰρ] τὸ γὰρ E4 81 οὐθέτερον b σφῶε T προσ. δυϊκοῦ E4 83 le. add. Vill. 84 locus obscurus; τρωσὶ μάχεσθαι = Δ 156. Ad rem cf. sch. A 56 a; unde post μάχ. verba συνέβη αὐτούς addenda esse censet Nickau

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

5

10

15

20

a.2 ἐπὶ τὸ διηγηματικὸν μετιὼν oὐχ ὑποβάλλει τὰς διηγήσεις αὐτομάτως, ex. ὡς μὴ τοῖς ἀκούουσι προσκορεῖς εἶναι δοκεῖν· ἀλλὰ διὰ πεύσεως καὶ ἀποκρίσεως ἀντιπεποίηται τὴν ἀνόρθωσιν τῆς προθυμίας, καὶ διψῶσιν αὐτοῖς ἐπάγει τὴν ἀπόκρισιν φιλέλλην ὤν· ὡς γὰρ τὴν στάσιν τῶν ἀριστέων ἐγκαλεῖ θεοῖς, ὡς οὐκ ἀθεεὶ τῶν τοιούτων ἀνδρῶν ἡκόντων εἰς ἅμιλλαν. h(AgBdP W3) b. δύο δὲ εἴδη θεῶν παρὰ τῷ ποιητῇ τὸ μὲν εἰς τὴν ποίησιν συντελοῦν, h καὶ πάθεσιν ὑποκείμενον παντοδαποῖς, τὸ δὲ ἀκήρατον h(AgBdP W3) περὶ οὗ φησι “θεοὶ ῥεῖα ζῴοντες” (Z 133) “θεοὶ δοτῆρες ἐάων” (θ 325). h(W3) 8 a. τίς τ᾿ ἄρ᾿ σφωε: τίς δὴ αὐτοὺς τοὺς δύο, Ἀγαμέμνονα καὶ Ἀχιλλέα· πάλιν δὲ τοῦτο ὁ ποιητὴς ἐρωτᾷ τὴν Μοῦσαν. d(Pal2QYZ) h(AgBdP) b. σ φ ω έ: ὅτι Ζηνόδοτος “σφωι” ἔγραφεν. Aim c. σ φ ω έ: οἱ περὶ τὸν Σιδώνιον τῇ “σφωι” ἀρέσκονται, οὐ καλῶς· ἡ γὰρ σφῶϊ δύο πτώσεις ἐπιδέχεται, ὑμεῖς τε καὶ ὑμᾶς, ὧν οὐδέτερον ἁρμόζει. τὸ δὲ σφωέ μᾶλλον τρίτου ἐστὶ δυϊκοῦ προσώπου καὶ σημαίνει τὸ αὐτούς. b(BE4)T d. ἔ ρ ι δ ι ξ υ ν έ η κ ε μ ά χ ε σ θ α ι: ὡς δι᾿ ὅ γε Τρωσὶ μάχεσθαι. T e. ἔ ρ ι δ ι: φιλονεικίᾳ. d(Pal2QZ) h(Agsl) f. ξ υ ν έ η κ ε: δασύνεται τὸ ξυνἕηκε. Aint g. ξ υ ν έ η κ ε: συνέβαλεν. d(Pal2QZ) h(AgslM11sl) συνέμιξε. d(Q)

D Ariston. ex.

ex. D Hrd. D

9 a. Λ η τ ο ῦ ς κ α ὶ Δ ι ὸ ς υ ἱ ό ς : ὁ τῆς Λητοῦς καὶ Διὸς παῖς Ἀπόλλων. D | Dhist πάλιν δὲ τοῦτο ὡς ἀπὸ τῆς Μούσης. | Ζεὺς δὲ ἐρασθεὶς Λητοῦς τῆς Κοίου θυγατρός, ἑνὸς τῶν Τιτάνων, καὶ Φοίβης, ἔγκυον αὐτὴν ἐποίησεν· ἥτις

1-5 post sch. 8a. AgBdP 1 μετιὼν: ἐπιὼν W3 δὲ post ὑποβάλλει add. W3 2 προσκορῒς W3 3 3 δοκεῖν: δοκῇ W 3 ἀντιπεποίηται: αὐτὴν πεποίηται W τὴν ἀνόρθωσιν τῆς προθυμίας: τῶν ἀκροατῶν ἀνορθῶν τὴν προθυμίαν W3 τὴν προθυμίαν BdpcP ἐπάγων PW3 4 φιλέλλην ὤν: ὡς φιλέλλην W3 ὡς γὰρ: ὥστε con. Erbse ἐγκαλεῖ Erbse: ἐπικλᾷ h 5 ἀθεεὶ: ἀθετεῖ W3 τῶν τοιούτων: τῶν om. W3 6-8 cum sch. a.2 coniunctum AgBdP 8 δοτῆρες: δω- Hom. 9 le. τίς τάρ σφωε Q εἰπὲ θεά et δὴ αὐτοὺς τοὺς δύο Αgsl ut vid., αὐτοὺς M11sl θεῶν post σφωε rec add. Z 10 post ἐρωτᾷ usque ad A 27 αὖτις: πάλιν deest Y foliis duobus amissis 11 ὅτι A, fort. ἡ διπλῆ (sc. περιεστιγμένη), ὅτι σφῶϊν A, corr. Lehrs 12-14 sch. 8c. cum 8-9 (post διηγούμενος) coniunxit b 12 le. τίς τ᾿ ἄρ T, om. b oἱ μέν τοι περὶ b σιδόνιον E4 σφῶϊ bT, em. Erbse ἡ γὰρ: τὸ γὰρ b 13 οὐθέτερον b 14 σφῶε T προσ. δυϊκοῦ E4 15 cum sch. 4c. coni. T; “locus obscurus; τρωσὶ μάχεσθαι = Δ 156. Ad rem cf. sch. A 56 a; unde post μάχ. verba συνέβη αὐτούς addenda esse censet Nickau.” Erbse 17 ξυνἕηκε Aint 18 μάχεσθαι in le. add. Zrec 19 le. om. Bm12M11V13V19 ἱστορία in mg. Α Pal2QΖ BdP, in textu ante Ζεὺς E4 τῆς Λητοῦς καὶ Διὸς παῖς om. E4 19-20 ὁ τῆς Λητοῦς—ἀπὸ τῆς Μούσης om. A Bm12M11V13V19W3 20 τοῦτο om. AgBdP δὲ om. A E4Pal2 Bm12M11V13V19W3 Λιτοῦς vid. Z ικοίου Z, κόου vid. A, Κυσσέως V13, τῆς τοῦ κοίου W3, 21 ἔγκυον: ἔγκοιον Z, ἔγγυον V19 ἐποίησεν αὐτήν E4

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen

ex.

διὰ τί οὐκ Ἀγαμέμνονι, ἀλλὰ βασιλεῖ φησιν; δείκνυσι διὰ τούτου ὡς οὐδὲν ὀνίνησι τύχη τοὺς περὶ τὸ θεῖον πλημμελήσαντας. b(BE4)

ex.

10 a. ψιλωτέον τὸ ἀπὸ γὰρ τοῦ ὄρω τὸ ὁρμῶ Αἰολικῶς γίνεται. b(BE4)Tt

ex.

b. καίτοι τοῦ λαός περιλημματικοῦ ὄντος πρὸς πλείονα ἔμφασιν τῷ πληθυντικῷ ἐχρήσατο. b(BE4)Tt

87 et 90 le. add. Li

87 διά τί b καλῶς Li

88 δείκν. διὰ τούτου om. Li

90

92 le. add. Ddf.

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

δεκαμηνιαίου χρόνου διαγενομένου, παρεγένετο διὰ θαλάσσης εἰς Ἀστερίαν τὴν νῆσον, μίαν οὖσαν τῶν Κυκλάδων· ἐκεῖ τε ἐλθοῦσα καὶ ἁψαμένη δύο φυτῶν, ἐλαίας καὶ φοίνικος, διδύμους ἀπεκύησε παῖδας, Ἄρτεμιν καὶ Ἀπόλλωνα καὶ τὴν νῆσον ἐκάλεσεν Δῆλον, ὅτι ἐξ ἀδήλου βάσεως ἐρριζώθη. A d(E4Pal2QZ) h(AgBdP 5 Bm12M11V13V19W3) b. Λητοῦς, ἐκ τοῦ λήθω, τὸ λανθάνω· ἐκ γὰρ τοῦ ἀφανοῦς ἐξεφάνη ὁ 12

h

11 13

ἥλιος. h(Bm M V ) c. β α σ ι λ ῆ ϊ: διὰ τί οὐκ Ἀγαμέμνονι, ἀλλὰ βασιλεῖ φησιν; δείκνυσι διὰ

ex.

4

10

τούτου ὡς οὐδὲν ὀνίνησι τύχη τοὺς περὶ τὸ θεῖον πλημμελήσαντας. b(BE ) d. ὁ γ ά ρ: οὗτος γάρ. d(Pal2QZ)

e. β α σ ι λ ῆ ϊ: τῷ βασιλεῖ. d(Pal QZ)

f. β α σ ι λ ῆ ϊ: ἤγουν κατὰ τὸν βασιλέα Ἀγαμέμνονα. h(M11sl) 2

g. χ ο λ ω θ ε ί ς: ὀργισθείς. d(Pal QZ)

15

10 a. ν ο ῦ σ ο ν: νόσον λοιμικήν. d(Pal2QZ)

c. ὦ ρ σ ε: ψιλωτέον τὸ ὦρσε· ἀπὸ γὰρ τοῦ ὄρω τὸ ὁρμῶ Αἰολικῶς γίνεται.

D

D ex.

t

b(BE )T

d. ὄ ρ σ ε: διήγειρεν, ἐνέβαλεν. d(Pal2QZ) h(AgBdP)

20

h

D 2

b. ἀ ν ὰ σ τ ρ α τ ό ν: κατὰ τὸ στρατόπεδον. d(Pal QZ) h(AgBdP) 4

D D

2

e. ὄ ρ σ ε: αἰολικῶς παρὰ τὸ ὄρρω τὸ διεγείρω. h(AgBdP) 2

f. κ α κ ή ν: κακωτικήν. d(Pal QZ) h(AgBdP) g. κ α κ ή ν: παρὰ τὸ χάζω τὸ ὑποχωρῶ· χάζος καὶ κακός. h(AgBdP) h. ὀ λ έ κ ο ν τ ο: ἀπώλλυντο. d(Pal2QZ)

i. λ α ο ί: καίτοι τοῦ λαός περιλημματικοῦ ὄντος πρὸς πλείονα ἔμφασιν τῷ 4

D h D h D ex.

t

πληθυντικῷ ἐχρήσατο. b(BE )T

1 δεκαμηναίου Q M11ac χρόνου om. M11, corr. M11sl παρεγένετο: ἐπείγετο La δὲ διὰ P εἰς: πρὸς La Ἀστερίαν E4La M11V13V19, Ἀστέριον A Pal2QZ Bm12W3 2 ἐκεῖ τε Q Bm12M11V13V19, ἐκεῖσε τε A, ἐκεῖσε δὲ La ἐλθοῦσαν A καὶ om. La ἁψαμένην A, δρεψαμένη La 2-3 φυτῶν, ἑλαίας δύο E4 3 ἀποκυῆσαι A, ἀπέτεκε W3 4 ἐκάλεσεν: -σε E4Pal2 Bm12M11V19W3, ἐκέλευσεν Z, ἐκέλευσε A βάσεως: βασιλέως Z ἐριζώθη Z, ἐρριζώται E4 Bm12V13, ἐρίζωται V19 αὕτη in fine add. E4 Bm12M11V13V19 8 le. add. Li διὰ τί b, καλῶς Li 8-9 δείκν. διὰ τούτου om. Li 10 ὃ in le. QZ 15 ἀνὰ στρατὸν: κατὰ στρατόν le. Q παρὰ τὸ στερρῶς (στερεῶς BdP) ἵστασθαι (= EM. 728.46, Et. Gud. 513.21) ἢ παρὰ τὸ ἵστημι (= Epim. Hom. 10 C [Ps], p. 88.92 Dyck) post στρατόπεδον AgBdP (cf. Dyck 1983, 39) 16 le. add. Li 18 le. ὄρσε Z, ὦρσε Q, ὤρσε van Thiel 19 καὶ παρὰ τοῦ ὅρω τοῦ ὁρμῶ οὗ ὁ μέσων παρακείμενος ὦρα post διεγείρω add. Matranga 22 ἀπόλλ- Z, ἀπώλλοντο Pal2Q 23-24 sch. partim evan. in B, post sch. 3f. in Tt

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen D (ad A 10)

λ α ο ύ ς: παρὰ τὸ λᾶαν. λᾶας δὲ κατὰ διάλεκτον ὁ λίθος. Προμηθέως παῖς — — — οὓς δὲ ἡ Πύρρα γυναῖκες. A ad A 126 λαούς posuit Erbse I, 46

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

5

10

15

j. λ α ο ί: ὄχλοι. λᾶες κατὰ διάλεκτον οἱ λίθοι λέγονται. | Προμηθέως παῖς Δευκαλίων γίνεται· οὗτος βασιλεύων τῶν περὶ τὴν Φθίαν τόπων, γαμεῖ Πύρραν τὴν Ἐπιμηθέως καὶ Πανδώρας, ἣν ἔπλασαν οἱ θεοὶ πρώτην γυναῖκα. καὶ ἐπειδὴ Ζεὺς ἠθέλησε τὸ χαλκοῦν γένος ἀπολέσαι ὑποθεμένου Προμηθέως, Δευκαλίων τεκτηνάμενος λάρνακα καὶ τὰ ἐπιτήδεια ἐνθέμενος, εἰς ταύτην μετὰ Πύρρας ἐνέβη. Ζεὺς δὲ πολὺν ὑετὸν ἀπ᾿ οὐρανοῦ χέας τὰ πλεῖστα μέρη τῆς Ἑλλάδος κατέκλυσεν, ὥστε διαφθαρῆναι πάντας ἀνθρώπους ὀλίγων χωρίς, οἵτινες συνέφυγον εἰς τὰ πλησίον ὑψηλὰ ὄρη. τότε δὲ καὶ τὰ κατὰ Θεσσαλίαν Τέμπη διέστη καὶ τὰ ἐκτὸς Ἰσθμοῦ καὶ Πελοποννήσου συνεχύθη πάντα. Δευκαλίων δὲ ἐν τῇ λάρνακι διὰ τῆς θαλάσσης φερόμενος ἐφ᾽ ἡμέρας ἐννέα καὶ νύκτας ἴσας, τῷ Παρνασσῷ προσίσχει κἀκεῖ τῶν ὄμβρων παῦλαν λαβόντων ἐκβὰς ἔθυσε Διῒ Φυξίῳ. Ζεὺς δὲ πέμψας Ἑρμῆν πρὸς αὐτὸν ἐπέτρεψεν αἰτεῖσθαι ὅ τι βούλεται. ὁ δὲ αἱρεῖται ἀνθρώπους αὐτῷ γενέσθαι καὶ Διὸς εἰπόντος ὑπὲρ κεφαλῆς ἔβαλλεν αἴρων τοὺς λίθους· οὓς μὲν οὖν ἔβαλεν ὁ Δευκαλίων ἄνδρες ἐγένοντο, οὓς δὲ ἡ Πύρρα γυναῖκες. ἡ ἱστορία παρὰ Ἀπολλοδώρῳ (1.46 ss.). A d(E4Pal2QZ) h(AgBdP Bm12M11P11aV13V19W3) 11 a. οὕνεκα τὸν Χρύσην: ἕνεκα, ἐπειδὴ τὸν Χρύσην. d(Pal2QXZ)

D | Dhist

D

1-15 sch. ad A 126 λαούς A, quod secuti sunt Vill. Bk. Ddf. Erbse, hoc loco d h ἱστορία in mg. Pal2Q BdP, in textu ante Προμηθέως E4 1 λαοί: ὄχλοι—λέγονται om. Bm12M11V13V19 λαούς le. A (ad A 126) ὄχλοι om. E4, οἱ λίθοι W3, παρὰ τὸ λάαν Α λᾶες: λᾶες δὲ P, λάας δὲ A ὁ λίθος A λέγονται om. A Προμίθεως vid. Z (sim. l. 4) 2 οὕτος βασιλεύων (βασ. οὗτος W3)— τόπων post γυναῖκα (r. 3) transp. W3 οὕτως E4Pal2Q V13V19 τῶν περὶ: τὸν περὶ Α, τῶν παρὰ Bm12 (et vid. M11 per compend.) τόπων: ποταμῶν E4 V19 3 Ἐπιμίθεως Φθίαν: Φοίαν V19 4 2 13 vid. Z Πανδῶρας Z: πανδώρης E Pal Q V πρώτην: πρῶτον ut vid. W3 ἐπεὶ δὲ A Z 4 ἠθέλησεν: -σε Pal2 BdP Bm12V19, ἠλέησεν Z χαλκ᾽ A ἀπολέσθαι A Δευκαλλίων W3 5 καὶ om. V13ac θέμενος Bm12M11W3 εἰς ταύτην: ἐν αὐτῇ A Πυρρᾶς Z 6 ἐνέβη: ἀνέβη d 13 19 AgBdPV V ὑετὸν: υἱετὸν E4 [χέας τὰ πλεῖ]στα W3 7 κατέκλεισεν Pal2ac ἀνθρώπων Z W3 ὀλίγ᾽ χωρὶς V13, χωρὶς ὀλίγων AgBdP 8 ὄρει Ag [τότε δὲ καὶ τὰ κα]τὰ W3 δὲ om. Bm12ac καὶ om. V19 κατὰ om. V13 Θεσσαλίαν: θάλασσαν V19 Τέμπη W3: στενώματα A AgBdP, στενώματα καὶ Τέμπη E4Q Bm12M11V13V19 9 ἐκτὸς: ἐντὸς V19 Πελοπονήσου AgBdPM11V19W3 4 2 10 ἐνέα Z 11 Παρνασῷ E Pal Q AgBdPBm12M11V13V19W3 ὄρει post Παρνασσῷ add. 12 11 3 AgBdPBm M W προσίσχει Apollod.: περιΐσχει E4Pal2QZ AgBdPBm12M11, περιίσχει ἤγουν ἐπλησίασε V13, περιηχεῖτο A (πε-), περιήχθη W3, περιέχει Bk. ἔθυσε: εὐθὺ ἐν Z, εὐθύς ἔθυσε AgBdPM11W3 Διῒ: δυῒ Z, τῷ Διῒ Pac? 12 Φυξίῳ: Φυξίων Pal2, Πυξίω Q, Φυξίας V19 πεμψαμ Z, ut vid. [vel -ψαν: non liquet] ἐπέτρεψεν om. W3 αἰτεῖσθαι: αἰτῆσαι La φησὶ post 3 βούλεται add. W 13 δὲ om. P αἰρεῖται: αἰτεῖται La αὐτῷ om. V13 Διὸς non liquet Z κεφαλῆς Ag: κεφαλὴν E4Pal2Q P Bm12V13V19 13-14 ἔβαλλεν αἴρων τοὺς λίθους: λίθους βάλλειν, αὐτὸς ἔβαλλεν αἴρων τοὺς λίθους AgBd, λίθους μᾶλλον αὐτὸς ἔβαλεν αἴρων τοὺς λίθους P 13 ἔβαλλεν: ἔβαλεν V19 14 τοὺς λίθους: λάας, τουτέστι λίθους La οὓς: καὶ οὓς A E4 12 13 3 2 12 11 13 19 3 AgBdPBm V , non liquet W οὖν om. A Pal Q AgBdPBm M V V W ἔβαλλεν A E4Pal2Q AgBm12M11V13V19W3 ὁ Δευκαλίων: οὗτος A ἐγίνοντο A E4 V19 15 Πυρρᾶ Z ἡ ἱστορία παρὰ Ἀπολλοδώρῳ om. A 17 ἀρητίρα τὸν χρύσην (11g): ἐπειδὴ τὸν χρύσην inc. ms. Χ οὕνεκα— ἕνεκα om. X τὸν Χρύσην in le. Zrec (om. Z) τὸν Χρύσην Zrec (om. Z)

  Towards a New Critical Edition of the Scholia to the Iliad: A Specimen

Ariston

11 a. διὰ τὴν τάξιν τοῦ ἄρθρου, τὸν ἀρητῆρα Χρύσην. Aim

ex. | ex.

b. ὑπερβατόν, ἵν᾿ ᾖ τὸν ἀρητῆρα Χρύσην· τὰ γὰρ ἐπιθετικώτερον συντασσόμενα κυρίοις ἢ προσηγορικοῖς τὰ ἄρθρα δέχονται, εἰ μὴ τὰ ῥήματα διαφόρων γνώσεων κατηγοροίη· τότε γὰρ ἀδιάφορον τὸ ἄρθρον. ἔστιν οὖν καθ᾿ ὑπερβατόν, ὡς τὸ “ἀλλ᾿ ὅτε δὴ †ὁ μοχλὸς ἐλάϊνος” (ι 378). | κατὰ συλλογισμὸν δὲ ἡ διήγησις· εἰ γὰρ θεοφιλεῖ τις ἀνδρὶ προσκρούων τοιαῦτα πάσχει, πόσῳ μᾶλλον εἰ θεῷ; καὶ εἰ βασιλεὺς οὕτω κολάζεται, τί ἄρα πείσεται ἰδιώτης; b(BE4)T

95

1

5

94 le. add. Vill. 95 τὸν] ἀντὶ τοῦ τὸν Uhlig 96 sq. usque ad v. συλλογισμὸν et post v. μᾶλλον εἰ omnia evanuerunt in T 96 (le.) οὕνεκα B (ut vid.), om. E4 2 διαφ. γνώσ. Β, διαφόρου γνώσεως E4 κατηγοροίη Ag κατηγοροῖντο E4, κατ[.......] Β 4 δὴ τάχ᾿ ὁ Li (ut Hom.) 6 μᾶλλον εἰ[ T, μᾶλλον b τί ἄρα πείσ. E4 πόσῳ μᾶλλον Li, evan. BT

Towards a New Critical Edition of the Scholia to the Iliad: A Specimen  

5

10

15

20

25

b. τὸν Χρύσην ... ἀρητῆρα: διὰ τὴν τάξιν τοῦ ἄρθρου, τὸν ἀρητῆρα Χρύσην. Aim c. τὸν Χρύσην ἀρητῆρα: ἱερέα, παρὰ τὸ ἀρᾶσθαι, ὅ ἐστι εὔχεσθαι. d(Pal2QXZ) h(AgBdP Bm12M11) d. τὰ δύο ὀνόματα καθ᾿ ἑνὸς προσώπου λέγονται· ὧν τὸ μέν ἐστι κύριον· τὸ δὲ ἐπίθετον καὶ προσηγορικόν· τὰ ἄρθρα οὖν ἐν τοῖς κυρίοις ἐν τῷ πολιτικῷ λόγῳ ἐντάττειν προσῆκεν· οἱ μέντοι ποιηταὶ καὶ ἐν τούτῳ ἐντάττουσιν· ὡς καὶ ὁ ποιητῆς ἐνταῦθα· ἢ κατὰ τὸν Χρύσην ἠτίμησεν ἀρητῆρα· ἀντὶ τοῦ εἰπεῖν Χρύσην τὸν ἀρητῆρα. h(AgBdP M1P11) e. ὑπερβατόν, ἵν᾿ ᾖ τὸν ἀρητῆρα Χρύσην· τὰ γὰρ ἐπιθετικώτερον συντασσόμενα κυρίοις ἢ προσηγορικοῖς τὰ ἄρθρα δέχονται, εἰ μὴ τὰ ῥήματα διαφόρων γνώσεων κατηγοροίη· τότε γὰρ ἀδιάφορον τὸ ἄρθρον. ἔστιν οὖν καθ᾿ ὑπερβατόν, ὡς τὸ “ἀλλ᾿ ὅτε δὴ ὁ μοχλὸς ἐλάϊνος” (ι 378). b(BE4)T h(AgBdP) | κατὰ συλλογισμὸν δὲ ἡ διήγησις· εἰ γὰρ θεοφιλεῖ τις ἀνδρὶ προσκρούων τοιαῦτα πάσχει, πόσῳ μᾶλλον θεῷ; καὶ εἰ βασιλεὺς οὕτω κολάζεται, τί ἄρα πείσεται ἰδιώτης; b(BE4)T f. ἀρητῆρα δὲ παρὰ τὸ αἴρειν τὰς χεῖρας εἰς προσευχήν· ἢ παρὰ τὸ ἀρέσκειν τῷ θεῷ. h(AgBdP) | καὶ πάλιν παρὰ τάξιν ἐχρήσατο τῷ ἄρθρῳ· ἐχρῆν γὰρ εἰρηκέναι τὸν ἀρητῆρα Χρύσην· ὁπόταν γὰρ δύο ὀνομάτων συνταττομένων τὸ μὲν κύριον ᾖ, τὸ δὲ προσηγορικόν, τὸ ἄρθρον οὐ τῷ κυρίῳ, ἀλλὰ τῷ προσηγορικῷ συντάττομεν· ὁπόταν δὲ δύο ὀνόματα, ὧν τὸ μὲν προσηγορικόν ἐστι, τὸ δὲ ἐπιθετικόν, τὸ ἄρθρον οὐ τῷ προσηγορικῷ, ἀλλὰ τῷ ἑπιθετικῷ συντάξομεν· οὐδεὶς γὰρ ἐρεῖ τὸν γραμματικὸν καλόν, ἀλλὰ τὸν καλὸν γραμματικόν, τὸ ἄρθρον τῷ ἐπιθετικῷ προτάξας, οὐ τῷ προσηγορικῷ. h(AgBdP M1P11) g. ἀρητῆρα: τὸν Χρύσην. d(X)

2

h. ἠτίμησεν: ἀτίμως περιύβρισεν. d(Pal QXZ)

Ariston. D h

ex. | ex.

h|h

D D

1 τὸν: ἀντὶ τοῦ τὸν Uhlig 3 Ἀτρείδ(ης) add. Zrec in le. post ἀρητῆρα 3-4 παρὰ—εὔχεσθαι deest in h praeter M11sl 3 ἀρᾶσθαι: ἀλῦσθαι Χ 5-9 cum praec. con. h 5-6 τὰ δύο– προσηγορικόν deest in M1P11 6-9 τὰ ἄρθρα—ἀρητῆρα post sch. 11f. r. 23 γραμμματικόν verbis καὶ ἄλλως praemissis M1P11 6 οὖν ἐν τοῖς κυρίοις: τῶν κυρίων M1 ἐν om. M1 7 προσήκει ac 1 1 Bd M ἐν τούτῳ: οὕτως M 8-9 ἢ κατὰ—τὸν ἀρητῆρα: τὸν Χρύσην ἀρητῆρα M1 10-14 ὑπερβατὸν—συλλογισμὸν evan. T 10 καὶ ἄλλως Χρύσην ἀρητῆρα ante ὑπερβατόν h, cum sch. 7d. coniungens ἵν᾿ ᾖ τὸν om. AgBdP ἐπιθετικώτερα B AgP 11 συνταττ- Bd 11-12 εἰ μὴ—κατηγοροίη: εἰ μὴ ἐπαγόμενα ῥήματα πρωτότυπα κατηγοροίη AgBdP 12 διαφόρου γνώσεως E4, non liquet B κατηγοροίη Ag, κατηγοροῖντο E4, κατ[.......] Β ἀδιαφορεῖ AgP 12-13 ἔστιν οὖν—ἐλάϊνος: τὸν Χρύσην ἀρητῆρα ὠνόμασεν τὸν ἀρητῆρα Χρύσην· ἔστιν οὖν ὡς τοῦ παιδὸς Ἀγαυοῦ ὁ μόχλος ἑλάϊνος AgBdP 13 δὴ ὁ Ε4, δὴ τάχ᾿ ὁ Li (ut Hom.), de B et T non constat 15 μᾶλλον εἰ Erbse (e T, sed post μᾶλλον pleraque evanida sunt) 15-16 post θεῷ male leg. vel omnino evan. in Li, οὕτω—ἄρα et ἰδιώτης evan. in B 18 καὶ πάλιν om. M1P11 20-22 τὸ ἄρθρον—τὸ δὲ ἐπιθετικόν om. P11 21 ὧν om. M1 ἐστι: ἦ M1 22 ἀλλὰ om. M1 ἐπιθέτῳ P11 23-24 τὸ ἄρθρον—προσηγορικῷ om. M1P11 25 ἀρητίρα X 26 ὕβρισεν Pal2QX Ἀτρείδ(ης) add. Zrec post περιύβρ.

 Omero, Eubulo, i pesci e i chorizontes Uno degli argomenti sfruttati dai Chorizontes a sostegno della ben nota tesi che l’lliade e l’Odissea non furono scritte dallo stesso poeta (l’lliade sarebbe di Omero, l’Odissea no)1 riguardava il cibarsi o meno dei prodotti del mare. La testimonianza più esplicita si trova nello Sch. A ad Il. XVI 747, nel quale si legge che i Chorizontes affermavano che il poeta dell’Iliade non fa mangiare pesci ai suoi eroi, mentre quello dell’Odissea sì:2 l’argomentazione è naturalmente rintuzzata da Aristarco (certo si tratta di lui, anche se non è menzionato esplicitamente), che sappiamo aver vigorosamente combattuto — e con successo — l’idea ‘separatista’.3 Il. XVI 746–747: εἰ δή που ϰαὶ πόντῳ ἐν ἰχθυόεντι γένοιτο, πολλοὺς ἂν ϰοϱέσειεν ἀνὴϱ ὅδε τήθεα διφῶν. Sch. A v. 747: ὅτι ἅπαξ εἴϱηϰε τήθεα. ἔστι δὲ εἶδος τῶν θαλασσίων ὀστϱέων. πϱὸς τοὺς Χ ω ϱ ί ζ ο ν τ α ς (fr. 7 Kohl)· φασὶ γὰϱ ὅτι ὁ τῆς Ἰλιάδος ποιητὴς οὐ παϱεισάγει τοὺς ἥϱωας χϱωμένους ἰχθύσιν, ὁ δὲ τῆς Ὀδυσσείας (cf. δ 368. μ 331). φανεϱὸν δὲ ὅτι, εἰ ϰαί μὴ παϱάγει χϱωμένους, ἴσασιν, ἐϰ τοῦ τὸν Πάτϱοϰλον ὀνομάζειν τήθεα. νοητέον δὲ τὸν ποιητὴν διὰ τὸ μιϰϱοπϱεπὲς παϱῃτῆσθαι. ϰαὶ μὴν οὐδὲ λαχάνοις παϱεισάγει χϱωμένους, ἀλλ᾿ ὅμως φησὶ «δμῶες Ὀδυσσῆος τέμενος μέγα ϰοπϱήσοντες» (ϱ 299).4

La questione, se gli eroi omerici si cibassero o meno di pesci, torna a più riprese nel I libro dei Deipnosofisti di Ateneo.5 In questa sede, a noi interessa soltanto la citazione che egli ci conserva a I 25bc: Ὅτι δὲ ϰαὶ ἰχθῦς ἤσθιον Σαϱπηδὼν δῆλον

 1 Procl., Vita Hom., p. 102, 2 sg. Allen: γέγϱαφε δὲ ποιήσεις δύο, Ἰλιάδα ϰαὶ Ὀδύσσειαν, ἣν Ξένων ϰαὶ Ἑλλάνιϰος ἀφαιϱοῦνται αὐτοῦ. I frammenti sono raccolti in Kohl 1917; una seconda parte contenente un saggio sui problemi generali fu annunciata ma non apparve mai; fu parzialmente sostituita da un articolo: Kohl 1921, 198 sgg., che trovo quasi sistematicamente ignorato (a parte Diehl, cit. qui sotto) e risulta ricco di stimoli, (v. infra n. 15). Cfr. Grauert 1827; Susemihl 1892, II, 149 sg.; Diehl 1938, 1 sgg.; Cohn 1899; Gudeman 1912; Fuhrmann 1967; Pfeiffer 1973, 333 e 357 n. 126; Fraser 1972, I, 467. 2 Per questo problema cfr. Weber 1889, 91, 96 sg., 132 sg., 157 sgg., 187 sgg.; Hofmann 1905, 33; Kohl 1917, 27 sgg. Per le fonti v. anche Erbse 1969–1988, IV, 295 sg., Testimonia a II 747. 3 Cfr. bibl. cit. alla n. 1. 4 Cit. da Erbse 1969–1988, IV, 295. 5 I luoghi nelle opp. citt. alla n. 2. https://doi.org/10.1515/9783110772371-026

Omero, Eubulo, i pesci e i chorizontes  

ποιεῖ, ὁμοιῶν τὴν ἅλωσιν πανάγϱου διϰτύου θήϱᾳ. Καίτοι Εὔβουλος ϰατὰ τὴν ϰωμιϰὴν χάϱιν φησὶ παίζων·6 ἰχθὺν δ᾿ Ὅμηϱος ἐσθίοντ᾿ εἴϱηϰε ποῦ τίνα τῶν Ἀχαιῶν; ϰϱέα δὲ μόνον ὤπτων, ἐπεὶ ἕψοντά γ᾿ οὐ πεποίηϰεν αὐτῶν οὐδένα. ἀλλ᾿ οὐδὲ μίαν ἄλλην ἑταίϱαν εἶδέ τις αὐτῶν, ἑαυτοὺς δ᾿ ἔδεφον ἐνιαυτοὺς δέϰα. πιϰϱὰν στϱατείαν δ᾿ εἶδον, οἵτινες πόλιν μίαν λαβόντες εὐϱυπϱωϰτότεϱοι πολὺ τῆς πόλεος ἀπεχώϱησαν, ἧς εἷλον τότε.

Nel frammento di Eubulo (120 K. = 120 Edm.) la giocosa presa in giro dell’epopea omerica prende avvio con una interrogazione retorica, il cui senso evidente è: Omero non ha mai rappresentato alcuno degli Achei nell’atto di mangiare pesce. Viene subito facile, dunque, mettere in rapporto tale ‘presa di posizione’ del comico con le discussioni sull’argomento che abbiamo visto opporre i Chorizontes ad Aristarco: i primi sostenevano che nell’Iliade non si mangia pesce e nell’Odissea sì, per cui i due poemi sono in contraddizione; Aristarco invece si studiava di mostrare come gli eroi iliadici conoscessero il cibo marino benché non fossero rappresentati nell’atto di consumarlo, per cui la contraddizione non sussiste. Potrebbe sorgere il dubbio se Eubulo si riferisca agli eroi di entrambi i poemi o solo a quelli iliadici, a proposito del cibarsi di pesce: il contesto parrebbe suggerire preferibilmente che l’affermazione del comico vada riferita soltanto all’Iliade. Se è lecito mettere in relazione i versi di Eubulo con la questione sollevata dai Chorizontes sull’uso del pesce nei due poemi, si potrà dedurne: o che Eubulo, sostenendo che nell’Iliade non si mangia pesce, porta acqua al mulino della tesi ‘separatista’; oppure che il comico — come pare forse più probabile per le parole con cui sia Ateneo sia Eustazio introducono la citazione — si prende argutamente gioco delle discussioni intorno a questo problema. Qualunque fosse la posizione di Eubulo riguardo al problema del mangiar pesci in Omero, a noi basta — per quel che vogliamo trarne — vedere nel fr. del comico un’eco, una traccia della discussione intorno a tale problema: il che, per il modo in cui esso si presenta e per la relazione che può essere istituita con lo scolio riportato, ci pare almeno probabile.7 Eubulo, poeta della commedia di  6 Cfr. Eustazio 1720, 30 sgg.: εἰ ϰαὶ Εὔβουλος ϰωμιϰώτεϱον παίζων φησίν· ἰχθὺν δὲ ποῦ Ὅμηϱος ἐσθίοντα εἴϱηϰέ τινα τῶν Ἀχαιῶν; (cfr. vv. 1–2) ϰαὶ λέγει μὲν ἐϰεῖνος ἀληθῶς. 7 In Montanari 1976 ho cercato di mostrare come Eubulo conoscesse una glossa ad Il. XVI 235 e l’abbia sfruttata in un gioco di allusione ad Omero. Il risultato di quel lavoro e quanto propongo qui si sostengono a vicenda, cooperando nel mostrare come il gioco allusivo e parodistico

  Omero, Eubulo, i pesci e i chorizontes mezzo, visse nel IV sec. a.C.; un indizio a conferma che il problema era noto a quell’epoca mi pare si possa ricavare anche da Platone, Resp. III, 404 b: Καὶ παϱ᾿ Ὁμήϱου, ἦν δ᾿ ἐγώ, τά γε τοιαῦτα μάθοι ἄν τις. Οἶσθα γὰϱ ὅτι ἐπὶ στϱατιᾶς ἐν ταῖς τῶν ἡϱώων ἑστιάσεσιν οὔτε ἰχθύσιν αὐτοὺς ἑστιᾷ, ϰαὶ ταῦτα ἐπί θαλάττῃ ἐν Ἑλλησπόντῳ ὄντας, οὔτε ἑφθοῖς ϰϱέασιν, ἀλλά μόνον ὀπτοῖς, ἃ δὴ μάλιστ᾿ ἂν εἴη στϱατιώταις εὔποϱα.

Platone sembra sostenere che nell’Iliade gli eroi non si cibano di pesce soltanto a causa delle particolari esigenze imposte dalla situazione bellica, il che vanificherebbe la possibilità di sfruttare l’argomento come «contraddizione» fra i due poemi. Ma questa testimonianza mi sembra tale da non insisterci più di tanto, e soltanto subordinatamente a quella precedente.8 Abbiamo comunque indizi che, si riferiscono al IV sec., cioè prima della nascita della filologia alessandrina, prima che i Chorizontes propriamente detti formulassero la loro tesi destinata ad essere fulminata dagli strali di Aristarco. Dunque, se il nostro discorso è valido, siamo condotti a ritrovare traccia dell’esistenza in epoca prealessandrina di elementi di una posizione ‘separatista’, elementi che sarebbero stati ripresi e sfruttati più ampiamente dai Chorizontes di età ellenistica. Il passo di Eubulo ci fornirebbe un valido indizio per ipotizzare anche in questo caso una non trascurabile continuità, nel senso che certi argomenti di una embrionale posizione ‘separatista’ avrebbero un’origine precedente alla nascita della filologia in Alessandria: i Chorizontes non avrebbero creato ex nihilo ma avrebbero ripreso qualcosa che aveva già avuto un rudimentale concepimento, naturalmente facendone una vera e propria tesi sostenuta in modo più ampiamente motivato e consapevolmente approfondito grazie agli strumenti critici forniti dalla nuova era filologica. Se il fr. 120 K. di Eubulo (con il passo di Platone e in rapporto con lo Sch. A ad Il. XVI 747) ci fornisce un indizio per formulare un’ipotesi, ci sentiremo un po’ confortati nel caso che qualche ulteriore indizio concorra a sostenerla. Un altro argomento sfruttato dai Chorizontes era quello relativo al numero delle città di Creta: siamo informati dallo Sch. A ad Il. II 649 che essi rilevavano come in tale luogo dell’Iliade Creta sia detta «dalle cento città», mentre ad Odissea XIX 174 si dice che l’isola ha novanta città (cfr. anche lo sch. al verso odissiaco);

 del comico nei confronti della poesia epica si giovasse dei primi elementi di esegesi omerica che si producevano nella cultura greca. 8 La relazione fra lo Sch. Il. XVI 747 da una parte e il fr. di Eubulo e il passo di Platone dall’altra era già stata vista da Kohl 1917, 27 sgg.; cfr. Kohl 1921, 209 (v. infra n. 15).

Omero, Eubulo, i pesci e i chorizontes  

negli scoli troviamo anche, naturalmente, la confutazione di tale argomento.9 Nel cod. Ven. B ad Il. II 649 e in HQ ad Odyss. XIX 174 si trovano due passi degli Ὁμηϱιϰὰ ζητήματα di Porfirio, nei quali si discute appunto il problema della divergenza fra i due poemi a proposito del numero delle città di Creta (Porphyr. I 48–49 Schader).10 Il. II 649: ἄλλοι θ᾿ οἳ Κϱήτην ἑϰατόμπολιν ἀμφενέμοντο. Odyss. XIX 174: (Κϱήτη) . . . . . . ἐν δ᾿ ἄνθϱωποι πολλοί ἀπειϱέσιοι ϰαὶ ἐννήϰοντα πόληες. ad Il. Β 649: διὰ τί ἐνταῦθα μὲν πεποίηϰεν «ἄλλοι θ᾿ οἳ Κϱήτην ἑϰατόμπολιν ἀμφενέμοντο», ἐν δὲ Ὀδυσσείᾳ (τ 174) εἰπὼν ὅτι ἔστιν ἡ Κϱήτη ϰαλὴ ϰαὶ πίειϱα ϰαὶ πεϱίϱϱυτος ἐπάγει· «ἐν δ᾿ ἄνθϱωποι πολλοὶ ἀπειϱέσιοι ϰαὶ ἐννήϰοντα πόληες»; τὸ γὰϱ ποτὲ μὲν ἐνενήϰοντα ποτὲ δὲ ἑϰατὸν λέγειν δοϰεῖ ἐναντίον εἶναι. Ἡ ϱ α ϰ λ ε ί δ η ς μὲν οὖν ϰαὶ ἄλλοι λύειν ἐπεχείϱουν οὕτως· ἐπεί γὰϱ μυθεύεται τοὺς μετ’ Ἰδομενέως ἀπὸ Τϱοίας ἀποπλεύσαντας ποϱθῆσαι Λύϰτον ϰαὶ τὰς ἐγγὺς πόλεις, ἅς ἔχων Λεύϰων ὁ Τάλω πόλεμον ἐξήνεγϰε τοῖς ἐϰ Τϱοίας ἐλθοῦσιν, εἰϰότως ἂν φαίνοιτο μᾶλλον τοῦ ποιητοῦ ἡ ἀϰϱίβεια ἢ ἐναντιολογία τις. οἱ μὲν γὰϱ εἰς Τϱοίαν ἐλθόντες ἐξ ἑϰατὸν ἦσαν πόλεων, τοῦ δὲ Ὀδυσσέως εἰς οἶϰον ἥϰοντος ἔτει δεϰάτῳ μετὰ Τϱοίας ἅλωσιν ϰαὶ φήμης διηϰούσης, ὅτι πεπόϱθηνται δέϰα πόλεις ἐν Κϱήτη ϰαὶ οὐϰ εἰσί πως συνῳϰισμέναι, μετὰ λόγου φαίνοιτ᾿ ἂν Ὀδυσσεὺς λέγων ἐνενηϰοντάπολιν τὴν Κϱήτην, ὥστε, εἰ ϰαὶ μὴ τὰ αὐτὰ πεϱὶ τῶν αὐτῶν λέγει, οὐ μέντοι διὰ τοῦτο ϰαὶ ψεύδεται. Ἀ ϱ ι σ τ ο τ έ λ η ς δὲ οὐϰ ἄτοπόν φησιν, εἰ μὴ πάντες τὰ αὐτὰ λέγοντες πεποίηνται αὐτῷ· οὕτως γὰϱ ϰαὶ ἀλλήλοις τὰ αὐτὰ παντελῶς λέγειν ὤφειλον. μήποτε δὲ ϰαὶ μεταφοϱά ἐστι τὰ ἑϰατόν, ὡς ἐϰ «τῆς ἑϰατόν θύσανοι» (Β 448)· οὐ γὰϱ ἑϰατὸν ἦσαν ἀϱιθμῷ· ϰαὶ «ἑϰατόν δέ τε δούϱατ᾿ ἀμάξης» (Hes. Ο. 456). ἔπειτα οὐδαμοῦ λέγει ὡς ἐνενήϰοντα μόναι εἰσίν· ἐν δὲ τοῖς ἑϰατόν εἰσι ϰαὶ ἐνενήϰοντα.

 9 Sch. A ad Il. II 649: ἄλλοι θ᾿ ο ἳ Κ ϱ ή τ η ν : πϱὸς τοὺς Χωϱίζοντας (fr. 2 Kohl), ὅτι νῦν μὲν ἑϰατόμπολιν τὴν Κϱήτην, ἐν Ὀδυσσείᾳ (cf. τ 174) δὲ ἐνενηϰοντάπολιν. ἤτοι οὗν ἑϰατόμπολιν ἀντὶ τοῦ πολύπολιν, ἢ ἐπὶ τὸν σύνεγγυς ϰαὶ ἀπαϱτίζοντα ἀϱιθμὸν ϰατενήνεϰται νῦν, ἐν Ὀδυσσείᾳ δὲ τὸ ἀϰϱιβὲς ἐξενήνοχεν, ὡς παϱὰ Σοφοϰλεῖ (fr. 813 Ν.2 = 899 Ρ.). τινὲς δέ †φασι πυλαιμένη† τὸν Λαϰεδαιμόνιον δεϰάπολιν ϰτίσαι (cit. da Erbse 1969–1988, I, 318: cfr. i Testimonia). Sch. V ad Odyss. τ 174: ἐ ν ν ή ϰ ο ν τ α π ό λ η ε ς : ἐν Ἰλιάδι (Β 649) ἑϰατόμπολιν τὴν Κϱήτην λέγει, οὐχ ὡϱισμένως ἑϰατὸν πόλεις ἔχουσαν, ἀλλὰ ἀντί τοῦ πολλάς. ἔνιοι δέ φασιν Ἰδομενέα ϰατὰ τὸν ἐξ Ἰλίου ἀνάπλουν ἀπελαυνόμενον τῆς Κϱήτης ὑπὸ Λεύϰου, ὃν θετὸν παῖδα ϰαταλελοίπει φύλαϰα τῆς βασιλείας, δέϰα πόλεις ποϱθῆσαι. μετὰ δὲ τὰ Τϱωϊϰὰ αἱ δέϰα πόλεις πϱοσεϰτίσθησαν (cit. dall’ediz. del Dindorf). 10 Cit. da: Schrader 1880, 48 sg.; cfr. Schrader 1890, 126. I due frammenti si trovano anche nell’ediz. degli Sch. B e degli Sch. Odyss. del Dindorf.

  Omero, Eubulo, i pesci e i chorizontes ad Od. τ 174: πῶς οὖν ἐν τῷ ϰαταλόγῳ φησὶν «ἄλλοι θ᾿ οἵ Κϱήτην ἑϰατόμπολιν ἀμφενέμοντο» (Β 649); ἤτοι οὖν ἐϰεῖ τὸ ἑϰατὸν ἀντὶ τοῦ πολλοῦ ϰεῖται, ὡς «τῆς ἑϰατὸν θύσανοι» (Β 448)· ἢ ἐπεὶ μετὰ τὸν ἀπόπλουν οἱ μετὰ Ἰδομενέως ἐπόϱθησαν Λύϰτον ϰαὶ τὰς πέϱιξ, ἃς ἔχων Λεῦϰος ὁ Τάλω πόλεμον ἤϱατο πϱὸς αὐτούς, οὗτος θετὸς ὢν Ἰδομενέως παῖς ἀφεθεὶς ὑπ᾿ αὐτοῦ φύλαξ τῆς Κϱήτης ἐστασίασε πϱὸς αὐτοὺς ἐπανελθόντας .... μετὰ δὲ ταῦτα πϱοσεϰτίσθησαν αἱ δέϰα.

Lo ζήτημα di Porfirio ci fornisce due testimonianze importanti. La prima è costituita dalla citazione di un Ἡϱαϰλείδης, in cui è stato riconosciuto un frammento delle Λύσεις Ὁμηϱιϰαί di Eraclide Pontico (fr. 171 Wehrli2).11 Siamo rinviati al IV sec. a.C. e ci inscriviamo nel quadro della produzione di λύσεις e di ζητήματα intorno al testo di Omero, di cui abbiamo ricche testimonianze e sulla quale è inutile che ci soffermiamo.12 Ciò che ci importa qui, è che la λύσις di Eraclide Pontico riguarda un problema di «contraddizione» fra l’Iliade e l’Odissea, che sarà ripreso e sfruttato in età tolemaica dai Chorizontes ma che deve essere stato formulato prima: anche questo dunque è indizio del fatto che un’opinione ‘separatista’ esisteva almeno nel IV sec. a.C., che argomenti pro e contro erano già oggetto di discussione. Ancor più importante ci pare però la citazione di Aristotele conservataci da Porfirio: si tratta di un frammento degli Ἀποϱήματα Ὁμηϱιϰά13 di cui non conosciamo il contesto preciso e che certamente ci è stato tramandato in forma non letterale bensì riassunta e condensata. Basta comunque, ci pare, per dire che Aristotele prendeva posizione — con una affermazione di portata generale — sulle «contraddizioni» omeriche e sulle deduzioni che se ne potevano trarre: Ἀϱιστοτέλης δὲ οὐϰ ἄτοπόν φησιν, εἰ μὴ πάντες τὰ αὐτὰ λέγοντες πεποίηνται αὐτῷ. Il fatto che Porfirio riporti il passo in relazione con la discussione e la confutazione di una argomentazione dei Chorizontes, indica con ogni probabilità che nella presa di posizione di Aristotele rientravano anche le «contraddizioni» fra i due poemi e il dedurne una posizione ‘separatista’. Ancora. Un passo di Strabone (X 4, 15; cfr. Eustazio 313, 34) ci fornisce un frammento di Eforo di Cuma sullo stesso argomento (FGrHist 70 F 146): τοῦ δὲ ποιητοῦ τὸ μὲν ἑϰατόμπολιν λέγοντος τὴν Κϱήτην (Il. Β 649), τὸ δὲ ἐνενηϰοντάπολιν (Od. τ 174), Ἔφοϱος μὲν ὕστεϱον ἐπιϰτισθῆναι τὰς δέϰα φησὶ μετὰ τὰ Τϱωιϰὰ ὑπὸ τῶν Ἀλθαιμένει τῷ Ἀϱγείῳ συναϰολουθησάντων Δωϱιέων· τὸν μὲν οὖν Ὀδυσσέα λέγει ἐνενη-

 11 Cfr. Wehrli 1969, 51 e 121 sg.; Kohl 1917, 18 n. 6 e p. 20. 12 Cfr. Lehrs 1882, 197 sgg.; Pfeiffer 1973, 133 sgg.; Wehrli 1969. 13 Fr. 146 Rose; cfr. anche Rose 1863, 148 sgg. È noto che Porfirio è la nostra fonte principale per gli Ἀποϱήματα Ὁμηϱιϰά di Aristotele: cfr. Schrader 1880, 415 sgg. e 1890, 180 sgg. Per la problematica qui accennata v. Pfeiffer 1973, 133 sgg.

Omero, Eubulo, i pesci e i chorizontes  

ϰοντάπολιν ὀνομάσαι. οὗτος μὲν οὖν πιθανός ἐστιν ὁ λόγος, ἄλλοι δ᾿ ὑπὸ τῶν Ἰδομενέως ἐχθϱῶν ϰατασϰαφῆναί φασι τὰς δέϰα.

Anche Eforo, dunque, aveva di questa difficoltà la sua λύσις per conciliare la divergenza fra i due poemi e controbattere una deduzione ‘separatista’. Ripetiamo ancora per lo storico cumano: ci troviamo in pieno IV sec. a.C.14 Gli indizi che abbiamo considerato inducono a formulare l’ipotesi che almeno nel IV sec. a.C. ci fosse della gente che pensava che l’Iliade e l’Odissea non fossero dello stesso poeta: costoro andavano alla ricerca di argomenti a sostegno della loro opinione e alcuni di questi argomenti furono ripresi e sfruttati in un quadro più ricco e filologicamente agguerrito dai Chorizontes di età alessandrina, contro i quali polemizzò duramente Aristarco. Il fatto che qualche traccia consenta di ipotizzare tale ripresa, spingerebbe a pensare — ponendo mente all’ampio ruolo giocato dalla casualità sulla nostra documentazione — che il fenomeno avesse proporzioni più ampie, che la continuità sia stata non così sporadica e desultoria bensì più organica e consapevole.15 Anche questo è un non trascurabile elemento di continuità fra «l’epoca prefilologica» e «l’epoca filologica», che si aggiunge alla ben nota linea che collega l’epoca classica a quella ellenistica nel campo della glossografia e dell’interpretazione allegorica.16 Certo gli indizi, che abbiamo fornito sono di portata limitata: il loro insieme e il loro cospirare, tuttavia, ci pare che abbia un peso non indifferente; e crediamo che altri se ne possano reperire con ulteriori ricerche. Mancando dati irrefutabili, alcuni indizi almeno non infondati sono quanto di meglio si possa ottenere.

 14 Anche il significato di queste testimonianze era stato colto in nuce da Kohl 1917, fr. 2, pp. 17 sgg. (v. infra n. 15). 15 Questa impostazione, cioè rintracciare e sottolineare una linea di continuità che collega «Chorizontes» pre-alessandrini ai Chorizontes propriamente detti di età tolemaica, era quella di Kohl nell’ art. cit., 1921, scritto in relazione con il materiale raccolto nella silloge dei frammenti (1917 cit.): il suo lavoro, per molti aspetti embrionale e bisognoso comunque di verifiche e approfondimenti, offre spunti interessanti. A tali premesse, tuttavia, non hanno arriso sviluppi adeguati e non ho trovato che si sia tenuto conto del problema quanto meno proponibile di una continuità fra età classica e filologia alessandrina in questo campo (come invece è d’uso fare per la glossografia e l’interpretazione allegorica). Gli argomenti che abbiamo qui sviluppato (muovendo per vero dire da punti di partenza autonomi, confortati poi da questa indipendente concordia) si trovano annunciati in brevi cenni di Kohl (cfr. nn. 8 e 14); e l’idea generale della continuità era già sua, anche se fu poi ignorata. 16 Una rapida sintesi su questi problemi in Montanari, art. cit. alla n. 7, con bibliografia.

 Una glossa omerica pre-alessandrina? (Sch. Il. XVI 235 – Eubulo fr. 139 K.) Nota presentata dal s. c. Emilio Gabba (Adunanza del 3 giugno 1976) Nella famosa preghiera di Achille a Zeus prima dell’ingresso in battaglia di Patroclo (Il. XVI 233 sgg.) si trova una incidentale menzione dei Selli, i sacerdotiinterpreti del celebre santuario-oracolo di Dodona: 235

ἀμφὶ δὲ Σελλοὶ σoὶ ναίουσ᾿ ὑποφῆται ἀνιπτόποδες χαμαιεῦναι.

Dell’epiteto ἀνιπτόποδες, hapax in Omero1 e di uso assai raro anche posteriormente,2 gli scolî al v. 235 forniscono numerose ed eterogenee spiegazioni. Tuttavia, per quanto riguarda il puro e semplice significato etimologico della parola, si hanno due sole divergenti interpretazioni: una di esse appare assolutamente predominante, mentre l’altra è riportata rapidamente e con scarso rilievo. Trascriviamo degli scolî solo quanto è più direttamente funzionale al nostro discorso.3 Sch. AD. ἀ ν ι π τ ό π ο δ ε ς : ... ταύτην ἔχοντες δίαιταν ὡς μηδὲ ἀπονίζεσθαι τοὺς πόδας ... τινὲς δὲ αὐτοὺς διὰ τοῦτο λέγουσιν ἀνιπτόποδας, ὅτι οὐϰ εξίασιν ἔξω τοῦ ἱεϱοῦ· διὸ οὔτε ἀπολούεσθαι ἀνάγϰην ἔχουσιν ... Sch. bT. ἀ ν ι π τ ό π ο δ ε ς : οἱ φυλασσόμενοι μή τι μιαϱὸν πατῆσαι. ἢ μὴ πϱοϊόντες τοῦ ἱεϱοῦ ὡς μὴ δεῖσθαι νίπτϱων . . . οἱ δὲ ἀνιπτόποδας ἀνιπταμένους ταῖς διανοίαις, μετεωϱολόγους. χαμαιεῦναι δὲ χαμαὶ ὄντες ϰαὶ τὰ πόϱϱω σϰοποῦντες. Sch. Ge. ἀ ν ι π τ ό π ο δ ε ς : ἐὰν μὲν ψιλῶς, ποδανίπτϱοις μὴ χϱώμενοι διὰ τὸ ϰαθαϱὸν ϰαὶ ἁγνὸν ϰαὶ διὰ τὸ μὴ ἐξιέναι τοῦ σηϰοῦ, ἐὰν δὲ δασέως, οἱ ἀναπηδῶντες ϰαὶ δίϰην πτηνῶν ἀναπετόμενοι ἐν τῷ ἐνθουσιασμῷ.

 1 Cfr. Ebeling, s.v. 2 L.S.J. s.v. registra: questo passo omerico e il fr. 139 di Eubulo di cui infra; Nonno, Dion. 40, 285; un’iscrizione di Tralles del II sec. recante una dedica a Zeus Larasios fatta da una certa L. Aurelia Emilia (cfr. Robert 1937, 406 sg., con bibliogr.): la dedicante è detta ἐϰ πϱογόνων παλλαϰίδων ϰαὶ ἀνιπτοπόδων e sarebbe interessante sapere se qui ἀνιπτόποδες deriva dall’uso religioso-cultuale oppure se è solo una reminiscenza del famoso verso omerico. 3 Gli scolî sono citati dall’ediz. di Erbse (1969–1988, vol. IV), tranne lo sch. AD per il quale ho usato la vecchia ediz. del Dindorf (vol. II, Oxford 1875). Lo sch. Ge e lo sch. min. sono riportati da Erbse fra i Testimonia; lo sch. min. si trova in De Marco 1946. https://doi.org/10.1515/9783110772371-027

Una glossa omerica pre-alessandrina?  

Sch. min. ἀ ν ι π τ ό π ο δ ε ς χ α μ α ι ε ῦ ν α ι· ἤτοι ὡς ἀγϱίων αὐτῶν ὄντων ἢ ἕνεϰα θεοῦ φυλασσομένων πατεῖν τι μιαϱὸν ϰαὶ διὰ τοῦτο μὴ χϱείαν ἐχόντων νίπτεσθαι· οὕτω γὰϱ ϰαὶ τοὺς χαμαιεύνας ἀϰουστέον ὡς διὰ τὴν θϱησϰείαν χαμαὶ ϰοιμωμένων αὐτῶν.

L’etimologia di gran lunga più seguita è quella che fa derivare l’aggettivo da α (privativo) + νίπτω + πούς = «che non si lavano i piedi»: secondo tale interpretazione, l’epiteto testimonia una specifica costumanza dei sacerdoti dodonei (accanto a quella di dormire per terra: χαμαιεῦναι), ed a questo significato di base si rifà tutta una serie di spiegazioni che cercano di dar ragione dell’usanza di non lavarsi i piedi. Abbiamo tralasciato alcune spiegazioni di natura mitologicoetnografica che si trovano nello sch. AD e nello sch. T; quelle che abbiamo riportato riferiscono questo costume dei Selli alla sfera sacrale in relazione al concetto di purità e impurità.4 I sacerdoti di Dodona erano detti ἀνιπτόποδες, secondo gli scolî, perché non avevano bisogno di lavacri (scil. purificatori) dal momento che facevano attenzione a non calpestare qualcosa di impuro oppure perché non uscivano dal tempio — evidentemente, in terreno impuro. Questo tipo di spiegazione è l’unico di cui sia rimasta una condensata testimonianza nella lessicografia: Ap. Soph. 33, 12 ἀνιπτόποδες οἱ δι᾿ ἁγνείαν μὴ πϱοεϱχόμενοι (Π 235); Hesych. ἀνιπτόποδες· δι’ ἁγνείαν χϱείαν μὴ ἔχοντες νίπτεσθαι. I Selli curavano che le loro estremità inferiori non venissero contaminate in seguito al contatto con elementi impuri: perciò usavano non sottoporle a lavacri ed erano detti «dai piedi non lavati» – Σελλοὶ ἀνιπτόποδες. L’altra interpretazione etimologica è testimoniata in bT e in Ge ed ha una assai minore rilevanza nel complesso dell’esegesi scoliastica. L’epiteto ἀνιπτόποδες (che in questo caso non testimonierebbe uno specifico costume dei sacerdoti) deriverebbe da ἀνίπταμαι + πούς e significherebbe pressappoco «coloro i cui piedi volano», «che hanno le ali ai piedi» o qualcosa di simile: in sostanza, semplicemente «che volano». Ma in senso figurato, ovviamente: quelli che volano col pensiero (ἀνιπτάμενοι ταῖς διανοίαις bT), che sanno vedere e capire le cose celesti (μετεωϱολόγοι bT); il riferimento sarebbe alle pratiche mantiche e divinatorie dei Selli nella loro attività di interpreti del dio (ἀναπετόμενοι ἐν τῷ ἐνθουσιασμῷ Ge). Lo sch. Ge avverte scrupolosamente ἐὰν δὲ δασέως (contrapposto al precedente ἐὰν μὲν ψιλῶς), a rendere più chiara l’etimologia da ἀνἵπταμαι. Questa interpretazione non ha paralleli altrove.5 Ad essa si riallaccia  4 Cfr. soprattutto Parke 1967, 7 sgg. e 20 sgg.; Poetscher 1966, 143 sgg. Per la bibliogr., oltre alle opp. citt., v. Erbse 1969–1988, IV 222 sg.; Montanari 1976b, partic. 143 n. 11 (= this volume, ch. 57). 5 Negli studi moderni essa, ovviamente, non è presa in considerazione: la sua peregrinità e insostenibilità linguistica, comunque, non ci importano affatto in questa sede.

  Una glossa omerica pre-alessandrina? con ogni probabilità anche la spiegazione di χαμαιεῦναι che in bT troviamo subito affiancata: χαμαιεῦναι δὲ χαμαὶ ὄντες ϰαὶ τὰ πόϱϱω σϰοποῦντες;6 spiegazione differente sia da quella che si legge in Sch. min. sia dalla più elementare ‘traduzione’ che si trova in D (οἱ ἐπὶ τοῦ ἐδάφους ϰοιμώμενοι).7 Nel commento a questo passo omerico, Eustazio (1058, 11 sgg.) porta a confronto alcuni versi del comico Eubulo: si tratta del fr. 139 Kock,8 restituitoci in forma più ampia da Ateneo, III 113 f. οὗτοι ἀνιπτόποδες χαμαιευνάδες ἀεϱίοιϰοι, ἀνόσιοι λάϱυγγες, ἀλλοτϱίων ϰτεάνων παϱαδειπνίδες, ὦ λοπαδάγχαι λευϰῶν ὑπογαστϱιδίων.

Temi e personaggi ben noti alla commedia: la situazione è quella del banchetto e oggetto di scherno sono i parassiti smodati e voraci. Sia da Eustazio che da Ateneo, però, apprendiamo che questi parassiti non sono gente qualsiasi, bensì dei filosofi (probabilmente in particolare dei Cinici):9 filosofi ovviamente da strapazzo, gente di bassa lega, ciarlatani e profittatori volgari, che campano alle spalle altrui magari ammantandosi con una facile veste di uomini di pensiero. La ripresa omerica del primo verso è palmare e giustifica in modo del tutto immediato il richiamo di Eustazio: la scelta lessicale presenta due epiteti già molto rari di per sé ed il cui accoppiamento è addirittura limitato — a nostra conoscenza — a questi due luoghi di Omero e di Eubulo; ed anche lo stile è consonante: il verso, infatti, è un esametro dattilico di classica fattura, scandito in quattro membri di cui tre costituiti da parole composte di scoperta risonanza epica.10 Ma è subito evidente anche l’inversione di registro operata da Eubulo nei confronti di Omero: da una parte la serietà religiosa della preghiera epica, dall’altra lo scherno sarcastico della commedia che mette alla berlina. E i due aggettivi che nell’lliade indicavano costumi cultuali di sacerdoti sono diventati insulti contro  6 La relazione fra le due è più chiara in Eustazio 1058, 5: ἕτεϱοι δὲ σεμνότεϱον τοιοῦτον συντιθέασι νόημα, χαμαιεῦναι μέν, ἀνιπτόποδες δέ, τουτέστι χαμαὶ μὲν εὐναζόμενοι, ἀνιπτάμενοι δὲ τῶν ϰάτω ταῖς διανοίαις διὰ τὴν ἐν ταῖς μαντείαις φιλοσοφίαν. Il rapporto con gli scolî è chiaro anche a livello verbale. 7 V. Erbse, 1969–1988, IV 223 nei Testimonia. 8 T. Kock, CAF II p. 212 = fr. 139 Edmonds, FAC II pp. 144 sg. 9 Kock ad loc.: Apud Eubulum cynici his verbis significantur in summa simplicitatis specie voracissimi; cfr. Webster 1953, 53, 61 (per il tema della derisione dei filosofi v. partic. 50 sgg.). Cfr. Ateneo l.c.: ὁ πϱοπάτωϱ ὑμῶν Διογένης; Eustazio l. c.: φέϱεται Εὐβούλου πεϱὶ φιλοσόφων τὸ «ἀνιπτόποδες ϰτλ». 10 Per gli esametri comici cfr. White 1912, 149 sgg.

Una glossa omerica pre-alessandrina?  

gente volgare e mal vissuta: «costoro che non si lavano i piedi, che dormono per terra ...». L’uso di epiteti femminili contribuisce a caricare ulteriormente lo scherno; e χαμαιευνάδες rimanda per un’altra via ancora ad Omero, dove (come già notava Eustazio) lo si trova usato come epiteto delle scrofe (Od. X 243 e XIV 15): quindi «che dormono per terra», in trasparenza «come scrofe».11 Troppo connotato è il verso (metro, stile, linguaggio) perché l’allusione — piena di elementi facilmente afferrabili — non sia del tutto voluta: è chiaro che vuole essere colta e arricchire il contesto sarcastico degli elementi ‘figurali’ che il meccanismo della memoria allusiva chiama in gioco.12 Non sembrerà casuale l’impiego dell’aggettivo ἀνόσιος, termine della sfera religiosa quant’altri mai, qui violentemente abbassato ad una sfera ben diversa: «empie gole, divoratori parassiti dei beni altrui», che con l’empietà religiosa ha molto poco a che fare. L’elemento religioso-sacerdotale in Eubulo non esiste più come significato immediato del testo, ma soltanto come intertesto e come ‘figura’ dovuta al meccanismo allusivo: cioè come portato dei significanti che, trasposti in un diverso contesto che fornisce loro una connotazione del tutto differente, non sono ‘sradicati’ ma recano pur sempre con sé qualcosa del contesto da cui provengono. Qui la rarità del lessico e del sintagma assicurano la percezione di quello spazio figurale nel quale risiede il significato (omerico) «religione-sacerdozio»: esso, nell’operazione di mutamento di registro, conferisce allo scherno sarcastico una maggiore e più pregnante intensità comica.13 Finora la nostra analisi del rapporto letterario fra il frammento di Eubulo e il passo omerico si è fondata sul sintagma ἀνιπτόποδες χαμαιευνάδες: non abbiamo fatto parola, a proposito di Eubulo, dell’epiteto ἀεϱίοιϰοι, letteralmente  11 Eustazio l. c.: Ὅϱα δέ, ὅτι χαμαιεῦναι μὲν ἄνθϱωποι ἐπαινετοὶ τῆς χαμεύνης ὡς ἐνάϱετοι, αἱ δὲ χαμαιευνάδες οὐ τοιοῦτον· σύες γάϱ. ἡ δὲ ϰωμῳδία τὰ συνήθη παίζουσα τοὺς χαμαιεύνας χαμαιευνάδας ϰαλεῖ πϱὸς ἀναλογίαν τοῦ «Ἀϱϰάδας ϰαὶ φυγάδας» ϰαὶ τῶν ὁμοίων. φέϱεται γοῦν Εὐβούλου πεϱὶ φιλοσόφων τό (fr. 139), ὃ παϱαπεποίηϰεν ἐϰεῖνος ἐϰ τοῦ παϱάσιτοι, εἰ δὲ φιλοσόφους ἄνδϱας οὕτως ἔσϰωψέ τις, πῶς οὐϰ ἂν ὁ τοιοῦτος ἀσφαλέστεϱον ϰαὶ γυναῖϰας χαμαιευνάδας ἐϱεῖ τινὰς ὧν ὁ βίος οὐϰ εὐπϱεπής; Per il senso degli epiteti femminili cfr. anche Kock l.c.: «χαμαιευνάδες autem pro χαμαιεῦναι eo consilio, quo apud Homerum Ἀχαιίδες, οὐϰέτ᾽ Ἀχαιοί». 12 Per la problematica relativa a questi fenomeni letterari è sufficiente rinviare a Conte 1974: interessa partic. al nostro caso il cap. I, “Memoria dei poeti e arte allusiva” (già in Strumenti Critici 16, 1971, 325 sgg.). 13 Sulle parodie di Eubulo nei confronti della poesia «seria» (epica e drammatica), cfr. A. Meineke, FCG I, Berlino 1839, 355 sgg.; Kaibel 1907; Argenio 1964. Un esempio grazioso e significativo è dato dal fr. 120 K. = 120 Edm., nel quale viene argutamente presa in giro la spedizione contro Troia degli ‘eroi’ omerici: ed Ateneo, I 25c, lo introduce con le parole «Εὔβουλος ϰατὰ τὴν ϰωμιϰὴν χάϱιν φησὶ παίζων».

  Una glossa omerica pre-alessandrina? «che abitano nell’aria», vale a dire — nel contesto — «gente con la testa in aria» o qualcosa di simile.14 Una parola che in greco si potrebbe parafrasare con ... ἀνιπτάμενοι ταῖς διανοίαις, μετεωϱολόγοι, naturalmente intendendo queste espressioni con una connotazione negativa, rovesciate di valore rispetto a quello che avevano negli scoli omerici da cui le abbiamo tratte.15 Vale a dire: «che volano col pensiero» non nel senso che conoscono le cose divine e sono vicini al dio, bensì nel senso che non hanno i piedi in terra, che hanno la mente fra le nuvole. A quanto pare, il gioco sottile dell’inversione di registro — operazione dotta di memoria poetica — ha coinvolto non solo le parole di Omero, ma anche le interpretazioni che ne venivano date. Ἀνιπτόποδες si è per così dire sdoppiato: è rimasto come significante del significato «che non si lavano i piedi» (ma con una connotazione rovesciata) ed ha ‘generato’ un nuovo significante (un hapax legomenon!), che è l’immagine fonica dell’altro suo possibile significato secondo le differenti etimologie antiche, naturalmente anche questo rovesciato di valore come vuole il contesto nel quale è stato inserito. Se è valida la nostra analisi del rapporto fra Omero ed Eubulo ed è stato ricostruito correttamente il processo genetico dell’inserimento dell’epiteto ἀεϱίοιϰοι nel verso del comico, resta da trarne l’ultima conseguenza, che è quanto ci interessa maggiormente in questa sede. La conseguenza è che ad Eubulo doveva essere nota quella interpretazione etimologica che faceva derivare l’ἀνιπτόποδες omerico da ἀνίπταμαι e gli dava il senso di «persone che volano col pensiero fra le cose celesti che conoscono le cose degli dei»: questo è il significato cui egli ha cambiato di valore nel senso di «persone che vivono con la testa per aria, tra le nuvole». Siccome Eubulo visse nel IV secolo a.C.,16 avremmo qui testimonianza dell’esistenza in epoca pre-alessandrina — prima del grosso lavoro esegetico compiuto dai filologi del Museo — di una glossa omerica che spiegava la rara e problematica parola ἀνιπτόποδες: una testimonianza cioè dei

 14 L.S.J. s.v.: «dwelling in air (mock heroie)»; Edmonds l. c.: « skydwellers»; Webster 1953, 53: «dwellers in the open». 15 Μετεωϱολόγος è usato con connotazione negativa da Eurip. fr. 913, 2 N2 (cfr. Arrighetti 1964, fr. 38 I, 50) e con ironia satirica da Plat. Cratyl. 401 b. Nelle Nuvole di Aristofane il riferimento alle ‘cose celesti’ costituisce un tema frequente dell’arguta ed ironica caricatura di Socrate: troviamo μετεωϱοφέναϰας (v. 333), μετεωϱοσοφιστῶν (v. 360); inoltre sch. Pac. 92: μετεωϱολέσχας (fr. 386 K.) ἔφη τοὺς φιλοσόφους (cfr. Plat. Rep. 489 c; Hesych. μετεωϱολέσχαι· φλυαϱοῦντες τὰ πεϱὶ οὐϱανὸν μετέωϱα). Per questa tematica in Aristofane v. K.J. Dover, comm. ad Nub. 333: non è improbabile che la suggestione di tali motivi comici aristofaneschi sia stata operante anche nel nostro passo di Eubulo. 16 I termini cronologici precisi sono discussi: per la datazione cfr. Kaibel 1907; Norwood 1931, 40; Webster 1953, partic. Chronological Table, 239 sgg. V. anche Suess 1966.

Una glossa omerica pre-alessandrina?  

primi elementi di ermeneutica omerica, progenitori della filologia professionale. Di questa glossa pre-alessandrina è rimasta testimonianza diretta solo negli scolî di tradizione bizantina: se era nota a Eubulo, che la sfruttò nel IV sec., essa nacque in epoca ancora classica, fu recepita nei lavori (commentari, raccolte lessicografiche) dei filologi di Alessandria e da questi si trasmise — attraverso tutti i passaggi che tale cammino conobbe — ai corpora di scoli messi insieme e redatti nell’età bizantina.17 Questo singolo caso di (probabile) glossa pre-alessandrina si affianca agli altri che ormai da tempo gli studiosi (a cominciare da Wilamowitz e Wackernagel)18 hanno rintracciato analizzando in modo simile al nostro passi di autori dell’epoca classica (come per es. Eschilo) dove compaiono parole omeriche usate in un significato che è testimone di una «interpretazione» (per la quale in genere si trova corrispondenza negli scolî). A noi importa in particolare mettere l’accento sui problemi che si pongono relativamente agli antecedenti e alla preistoria della filologia, al rapporto fra tali antecedenti e le opere criticoletterarie degli esegeti alessandrini, alle tracce che di quelli rimangono nei lacerti di critica alessandrina giunti fino a noi. Vale a dire, l’epoca assai antica cui possono risalire a volte elementi di esegesi e di ermeneutica testuale, che noi troviamo testimoniati direttamente solo in testi molto più tardi (come gli scolî e gli etimologici) e derivati dai lavori dell’epoca ellenistica. Per il testo di Omero, si può dire che il problema ha acquisito, almeno nei suoi contorni generali, una veste abbastanza netta: non c’è dubbio che prima dell’epoca tolemaica esistettero sforzi di interpretazione e di ‘traduzione’ di Omero, dettati dall’esigenza di capire e di spiegare un autore e un linguaggio diventati in molti punti difficili e problematici. La chiameremmo una rudimentale (ma quanto?) e pragmatica glossografia: non sappiamo se legata all’insegnamento scolastico soltanto o se addirittura con una sua autonomia, e dignità di circolazione e un pubblico di fruitori. Le prove indirette, fornite dall’uso-interpretazione di cui sopra di parole omeriche in poeti classici, potrebbero lasciare qualche dubbio, se non fossero ottimamente corroborate da testimonianze esplicite: fra queste basterà qui ricordare il famoso passo dei Daitalēs di Aristofane (fr. 222 K.), nel quale un personaggio chiede ad un altro di spiegargli le Ὁμήϱου γλώττας. Non c’è bisogno

 17 Per un’altra traccia di filologia omerica pre-alessandrina in Eubulo cfr. Montanari 1976a (= questo volume, ch. 26). 18 Cfr. Wilamowitz 1886, 111 sg.; Wackernagel 1935, I, 728 sgg.; Latte 1925 = 1968, partic. 641 sgg.

  Una glossa omerica pre-alessandrina? che ripetiamo cose già ottimamente dette:19 una qualche forma di glossografia omerica esisteva prima dell’opera dei filologi del Museo, ed ogni nuovo caso rilevato — come il nostro del fr. di Eubulo — andrà inserito in questo quadro per cercare di far luce sempre meglio sulle sue caratteristiche. Acquisito questo risultato generale, ci si delineano di fronte grossi problemi ancora aperti e che qui possiamo solo accennare. Innanzi tutto quello dell’utilizzazione e del tipo di utilizzazione che i filologi alessandrini fecero di questo materiale precedente, il modo in cui esso fu recepito e reimpiegato. Connessa è la questione relativa a quegli anonimi personaggi citati negli scoli genericamente come γλωσσογϱάφοι: ermeneuti pre-alessandrini (maestri di scuola?) o glossografi ‘minori’ di età ellenistica oppure anche nome generico per qualcosa, di ancor più indeterminato?20 Il tutto richiede una globale riconsiderazione, una raccolta e un riesame delle testimonianze, una ricerca più larga. Per quanto riguarda poi i cosiddetti Scholia D, ci soffermiamo un momento per sottolineare un problema che si riaggancia alla questione delle glosse pre-alessandrine. Il corpus bizantino di questi scolî è costituito, come è noto, da un certo numero di ἱστοϱίαι e da tutto un insieme di spiegazioni di singole parole o espressioni costituenti una sorta di vera e propria ‘traduzione’ passo passo del testo omerico. Fu Wilamowitz21 il primo a sostenere che affonda le sue radici in epoca molto antica l’elemento glossografico di quelli che per un errore umanistico furono detti Scholia Didymi e ritenuti il corpus scoliografico omerico di origine più recente. Questa opinione è oggi comunemente accettata e si è precisata. Da ultimo Henrichs, nel suo esemplare lavoro sugli Scholia Minora omerici, scrive giustamente che queste Worterklärungen sono testimoniate fin dal V sec. a.C. e costituiscono, accanto alla interpretazione allegorica, la più antica forma della philologische Homererklärung;22 e giova riportare anche le chiare e sintetiche parole di Erbse a proposito di questi scolî: «... reliquias eorum interpretamentorum continent, quae pueri Athenienses Homeri intellegendi causa inde a quinto a. Chr. n. saeculo in schola discebant. Hoc e ratione concludimus, qua poetae Attici et Alexandrini verta rariora Homeri interpretabantur [il corsivo è mio]. Nec non complures talium explanationum cum interpretamentis glossographorum a  19 Oltre alla bibliogr. cit. alla n. prec., cfr. Erbse 1953, 170 sg.; Pfeiffer 1955, 72 (= 1960, 154); Pfeiffer 1973, 55, 59, 146 sg., 230 (e in generale la prima parte, dedicata alla «Preistoria della Filologia Greca»); infine Henrichs 1971, 99 sgg., con tutta la bibliografia. 20 Cfr. Lehrs 1882, 36 sgg.; Latte 1925 = 1968, 641; Erbse 1953; Pfeiffer 1973, 147; Henrichs 1971, 100 e n. 10; Ludwich 1884–1885, II, 118; Tolkiehn 1925, 2434. 21 Wilamowitz 1888, 142–147. 22 Henrichs 1971, 99 sg.: in questo lavoro si trova delineata con nitida sintesi la problematica cui accenniamo.

Una glossa omerica pre-alessandrina?  

grammaticis Alexandrinis usurpatis congruunt. Iam Aristarchus haud paucas emendavit vel supplevit. Etiam nonnullis papyris satis antiquis prorsus evincitur fundamentum illorum scholiorum ante aetatem Hellenisticam iactum esse».23 Grazie infatti ai ritrovamenti papiracei — da Henrichs ottimamente editi e interpretati — la storia degli Scholia Minora (intendendo con questa dicitura il nocciolo di Worterklärungen del corpus bizantino denominato Scholia D, unitamente ai suoi predecessori restituiti da papiri) si può ora seguire per testimonianza diretta fino al I sec. a.C. Le testimonianze indirette, del tipo di quella che abbiamo ipoteticamente ricostruito, ci aiutano a gettare lo sguardo alle spalle dell’epoca tolemaica.

 23 Erbse 1969–1988, I, Praefatio p. XI.

 Aristarchus on Odyssey II 136–7: A Case Study in Ancient Homeric Philology Odyss. II 132–137: κακὸν δέ με πόλλ’ ἀποτίνειν Ἰκαρίῳ αἴ κ’ αὐτὸς ἑκὼν ἀπὸ μητέρα πέμψω. ἐκ γὰρ τοῦ πατρὸς κακὰ πείσομαι, ἄλλα δὲ δαίμων 135 δώσει, ἐπεὶ μήτηρ στυγερὰς ἀρήσετ’ ἐρινῦς οἴκου ἀπερχομένη· νέμεσις δέ μοι ἐξ ἀνθρώπων ἔσσεται· ὣς οὐ τοῦτον ἐγώ ποτε μῦθον ἐνίψω.

Since antiquity, these verses have raised several problems of a philological and exegetical nature for readers of Homer, and the Homeric scholia attest to ancient scholars’ attempts to resolve these problems.1 Aristarchus proposed the athetesis of v. 137: in addition to the brief note found in the scholion on that verse (see n. 2 below), the second part of the composite scholion MV on v. 134 presents a fuller version of Aristarchus’ argument: τοῦ πατρὸς] τὸ δὲ τοῦ πατρὸς οὐ περὶ Τυνδάρεω, ἀλλὰ περὶ Ὀδυσσέως· οὐ γὰρ ἀπεγνώκει αὐτόν, ἐπειδή φησιν «ὀσσόμενος πατέρ’ ἐσθλὸν ἐνὶ φρεσίν» (α 115). — ἄλλως τε κατὰ Καλλίμαχον «χαλεπὴ μῆνις ἐπιχθονίων». διὸ Ἀρίσταρχος ἀθετεῖ τὸν «ἔσσεται, ὣς οὐ τοῦτον ἐγώ ποτε μῦθον ἐνίψω»· περισσὸς γάρ ἐστι πρὸς ταύτην τὴν ἀπόδοσιν.2

First of all, it is important to note that in this scholion two different annotations on distinct verses have been awkwardly combined under one lemma; this is clear from the fact that, from ἄλλως τε onwards, the text of the scholion no longer discusses v. 134 but rather vv. 136–137. The two parts therefore must be analyzed separately, and it is the second part which presents greater interpretative difficulties. Aristarchus’ reasoning for the deletion of v. 137 (presented in the last sentence: περισσὸς γάρ ἐστι) is preceded by a phrase attributed to Callimachus and is introduced by the conjunction διὸ (therefore).3 However, the  1 In general, I cite the text of the Odyssean scholia from Dindorf 1855; for the scholia on Odyss. I 1–309, however, one must use instead the text of Ludwich 1888–1890. The Iliadic scholia are cited from the edition of Erbse 1969–1988, I–V. 2 Cf. schol. HM on v. 137: ἔσσεται] ἀθετεῖται μὲν ὑπὸ Ἀριστάρχου, στικτέον δὲ ὅμως μετὰ τὸ ἔσσεται, ἵνα τὸ ὣς κέηται ἀντὶ τοῦ οὕτως. 3 The use of διὸ is typical of the technical language of the scholia when they present arguments that have been adopted in order to support an interpretation or a textual intervention; examples of the usage can be found on practically any page of the scholia. “The words διὸ https://doi.org/10.1515/9783110772371-028

Aristarchus on Odyssey II 136–7  

intended connection between the citation from Callimachus and the deletion proposed by Aristarchus is not at all clear. Indeed, what could it possibly mean to say, “According to Callimachus ‘χαλεπὴ μῆνις ἐπιχθονίων’, therefore Aristarchus deletes v. 137, since it is superfluous with regard to this statement”?4 The difficulty was perceived by Victor Bérard:5 not being able to explain the connection between the citation from Callimachus and the deletion of the verse, he proposed that the text of the scholion had suffered corruption, and that this corruption had obscured a variant reading which referred to Callistratus or Antimachus in such a way that what followed would have been intelligible. The text of the scholion as he proposed correcting it — including an error in the punctuation (for δέ· θεῶν, one must understand δὲ θεῶν, see n. 6 below) — reads as follows:6 ἄλλως τ’ ἡ κατὰ Καλλί vel potius μαχον χαλεπὴ μῆνις, διὸ Ἀρίσταρχος ἀθετεῖ τὸν ἔσσεται· ὡς οὐ τοῦτον ... περισσὸς γάρ ἐστι πρὸς ταύτην τὴν ἀπόδοσιν cf. Ε 178: ἱρῶν μηνίσας· χαλεπὴ δὲ θεοῦ ἔπι μῆνις.

 ἀθετεῖται make it clear that an argument for the athetesis was offered in the schol.”: van der Valk 1963–64, I, 538. 4 On Aristarchus’ athetesis of Odyss. II 137, see Carnuth 1869, ad loc.; Carnuth 1875, ad loc.; A. Ludwich 1884–1885, I, 522; Lehrs 1882, 338; Roemer 1912, 185 f.; van der Valk 1949, 222. These works limit themselves to a discussion of the notion of περισσόν as the motive for Aristarch’s athetesis; they do not address the question raised by the citation from Callimachus. Van der Valk (loc. cit.) explains: “Aristarchus athetized... β 137 owing to a misunderstanding of the text”; and he comments: “Aristarchus wrongly explained πατρός (β 134) as being said of Odysseus”. I do not understand why the interpretation of πατρός as referring to Odysseus (leaving aside the question of whether it is correct or not) necessarily requires the athetesis of v. 137: the connection between the two issues escapes me. 5 Extensive discussion can be found in Bérard 1920, 202 ff. Bérard subsequently employed the results of this article in the edition of the Odyssey which he completed for “Les Belles Lettres”: Bérard 1933, vol. I, 37. 6 I reproduce the text of the scholion found in Bérard’s edition of the Odyssey (Bérard 1933), where the scholion appears in the apparatus under the section Corr. Bérard went so far as to print in the text of the poem itself the variant reading he had devised and, accordingly, to eliminate the verse athetized by Aristarchus (recording it only in the apparatus). His text reads: οἴκου ἀπερχομένη· . | ὑμέτερος δ’ εἰ μὲν θυμὸς νεμεσίζεται αὐτῶν, κτλ. It is particularly odd to see that in the apparatus the “variant reading” of v. 136 presents a raised dot after χαλεπὴ δέ, whereas in the text of the poem it is written correctly, without the raised dot (as I have reproduced it in this footnote). One might suppose the raised dot to be a simple typographical error, but then one notes that there is also the difference of accent upon the particle δέ...

  Aristarchus on Odyssey II 136–7 Given that the phrase χαλεπὴ μῆνις ἐπιχθονίων does not fit within a hexameter, Bérard recalled the Homeric verse Ε 178 and thereby ‘excogitated’ the variant reading χαλεπὴ δὲ θεῶν ἔπι μῆνις, which would have been found at this point in the ekdosis of Callistratus or in that of Antimachus; in Bérard’s reconstruction, Aristarchus would have accepted this text and, since in this form v. 136 would already contain the verb, he would have proposed the deletion of the following verse, insofar as it would now contain a superfluous addition to Telemachus’ speech. However, Bérard’s interpretation and his textual interventions do not convince, resulting as they do from an approach that presents too many weak links to be at all persuasive. First of all: the athetesis of v. 137 does leave a nominal sentence in the preceding verse, but this does not in any way require us to believe that a variant reading was mentioned in the scholion, and still less that said variant reading must have been of such a sort as to restore the supposedly missing verb. Even from a brief survey, sporadic and limited in scope (which I relegate to the appendix below, since at this point it would take the argument too far afield), one can see how frequently the athetesis of a verse judged to be περισσός7 leaves a sentence with an elided verb in the text of the poem; and in these cases, one does not find that ancient critics resorted to variant readings to restore the verb, but rather that they were concerned to clarify for readers the correct manner to interpret the text and to supply the missing verb mentally, either by appealing to the notion of ἀπὸ κοινοῦ, where possible, or to that of (προσ)ὑπακούειν (i.e., to understand an implied verb). The present case is analogous to several others of this sort (listed below in the appendix): a verse is judged superfluous or prolix and is therefore deleted; in the preceding verse one is left with a nominal sentence, in which the implied verb must simply be understood (in this case, it is clearly ἔσται). Therefore, we should assume that in our scholion part of Aristarchus’ argument is missing, presumably because it was lost during the course of the various reductions and epitomizations that affected the mass of ancient critical and exegetical materials. Bérard, for his part, was probably spurred to “find” a variant reading because of the way he interpreted the word ἄλλως. Now, it is true that this word is a ‘technical’ term used in the scholia to introduce an alternative explanation or a textual variant (examples can be found on practically any page of the scholia).  7 For examples of athetesis proposed by Aristarchus which are based upon the concept of περισσόν and derive from his peculiar sensitivity to related factors of style and esthetics, cf. Lehrs 1882, 338 f.; Roemer 1912, 173 ff.; Ludwich 1914, 698 ff.; van der Valk 1949, 220 ff.; van der Valk 1963–1964, II, 84 ff., 450 ff., 465 ff.

Aristarchus on Odyssey II 136–7  

But in that case, with regard to the present scholion, one would be compelled to correct not only the proper name (as Bérard did), but also the τε that follows ἄλλως. It is simpler to assume that it is precisely the connective value of the expression ἄλλως τε — which is well attested in the language of the scholia8 — that should serve as the principal clue for correctly interpreting the scholion, since it functions as the link between the two parts of the scholion as it is found in the form that has come down to us. Finally, there do not seem to be valid reasons for believing that the phrase χαλεπὴ μῆνις ἐπιχθονίων is not a real citation from Callimachus and that, instead, it hides a textual variant — a textual variant, moreover, that (in Bérard’s reconstruction) would have been corrupted both in its reading and in the name of the author to whom it is attributed. Rudolf Pfeiffer, for his part, had no doubts about the matter, and he included this fragment in his edition of Callimachus,9 just as Schneider had done before him.10 We may also add several further considerations. The word ἐπιχθόνιοι raises no difficulties: even if it is not attested elsewhere in what we possess of Callimachus,11 it is used quite normally as a substantive in poetry from Homer onwards.12 The phrase χαλεπὴ μῆνις ἐπιχθονίων forms a pentameter, lacking only the first foot and the arsis of the second: ‒ ⏑ ⏑ ‒ χαλεπὴ μῆνις ἐπιχθονίων

Now, the adjective χαλεπός occurs thirteen other times in Callimachus (not counting the present fragment); and of these thirteen passages, each time that the adjective occurs in a pentameter (six passages total), it is always found in the same metrical sedes, namely, just before the central caesura,13 precisely as it is found in the present fragment. We are dealing, therefore, with a typical and  8 Even though the technical use of ἄλλως discussed above is extremely common, this does not mean that the word can only appear in the scholia with that specific meaning. The current lack of adequate tools makes it difficult to find a large number of loci similes, but nonetheless the following may be cited: schol. AT on Il. I 1; schol. bT and AD on Il. VIII 284; schol. on Odyss. I 330; II 48; III 71; among others. 9 Pfeiffer 1949, fr. 637. Cf. Trypanis 1958, fr. 637. For examples of the confusion of the names Callistratus and Callimachus, see Pfeiffer on Callim. fr. 452. 10 Schneider 1873, fr. 262. 11 The word μῆνις is also found only here in Callimachus: cf. Pfeiffer 1949, Index Vocabulorum, s. vv. 12 Il. XXIV 220; Odyss. XVII 115, XXIV 197; Hymn. XV 2. Cf. Apoll. Rh. II 250, IV 805 and 1306. 13 E. 30.4; E. 46.10; frr. 75.35, 75.49, 85.13, 178.20. The other instances are in hexameters: one may note, however, that at Hymn. VI 66 (hexameter) χαλεπός is found in the corresponding metrical position, namely, just before the penthemimeral caesura.

  Aristarchus on Odyssey II 136–7 straightforward case in which the poet’s ear prefers a specific metrical slot for a given word. These observations are confirmation of a fact that ought not to have been doubted in the first place: the scholiast has cited a verse of Callimachus. We must, therefore, attempt to understand the motivation for this citation — that is, its function within the condensed argumentation presented in the scholion. Pfeiffer maintains that the scholiast cites the passage as a parallel for Telemachus’ words concerning his father’s anger (v. 134);14 but, if we accept this interpretation, it is still impossible to understand how the parallel would be relevant to the deletion of v. 137. In v. 136, however, Telemachus refers to the indignation and the condemnation that he would incur in the eyes of men if he were to treat his mother unjustly: νέμεσις δέ μοι ἐξ ἀνθρώπων | ἔσσεται. A parallel between ἄνθρωποι and ἐπιχθόνιοι is much more credible (the two words are synonyms, after all) than one between ἐπιχθόνιοι (a generic term) and πατήρ (referring to a specific individual). Therefore, we can with good reason suppose that the Callimachean phrase, which is written de hominum ira gravi (Pfeiffer, loc. cit., fn. 14 below), has been cited as a parallel for νέμεσις δέ μοι ἐξ ἀνθρώπων | ἔσσεται. As we have seen, Aristarchus’ proposed deletion of v. 137 results in a nominal sentence in v. 136. Instead of searching for implausible textual variants that would restore the verb, perhaps we ought instead to note that even the fragment from Callimachus itself consists of an elliptical phrase lacking a verb. Now, one might object that the absence of the verb could be due to the fragment’s being extracted from its context, or to a loss that occurred in the course of textual transmission; but given that the nominal sentence has a specific function of its own, we ought to suppose that the citation from Callimachus forms a complete independent clause and has been cited both for its content and for its elliptical form.15 Moreover, the Callimachean expression is clearly gnomic in character, and this certainly does not clash with the above assumption. Indeed, it is precisely this fact which seems to be the key for understanding why Callimachus appears in this scholion at all: the Callimachean fragment, in all likelihood, was cited as a parallel not only for the similarity of its content, but also for the stylistic similarity it displays with regard to the Homeric phrase νέμεσις δέ μοι ἐξ ἀνθρώπων. At this point, the conclusion seems obvious: Aristarchus judged that

 14 Pfeiffer 1949, 428: At Call., nisi valde erravit scholiasta fragmentum ad Telemachi verba de patris ira afferens, de hominum ira gravi loquitur. Cf. Schneider 1873, 495. 15 Bergk, Anth. Lyr.2, 162, proposed that we read χαλεπὴ μῆνις ἔπι χθονίων, but this correction is unmotivated and it changes the meaning of the Callimachean passage. It was rejected by both Schneider and Pfeiffer; Bérard does not even mention it.

Aristarchus on Odyssey II 136–7  

v. 137 was superfluous, and he proposed its athetesis; this intervention resulted in a nominal sentence in the preceding verse; the philologist then cited a parallel — both semantic and stylistic — from the text of Callimachus, in which one finds a similar concept expressed in a nominal sentence.16 It must be remembered that, in the scholia, the word περισσόν marks something which ancient scholars considered non-essential or pointlessly prolix in the context of the poetic narrative: deletions that stem from this criterion, therefore, reveal an esthetic judgment, a stylistic sensibility that preferred a text which was pithy and concise. It is not surprising, then, that such atheteses frequently result in elliptical phrases that lack a verb, which were not at all felt to be awkward or bothersome; to the contrary, such elliptical phrases were felt to be entirely suitable for a style marked by brevity and concision (see for example, in the appendix, schol. A on Il. IX 44, where it is remarked how the narrative becomes ἐμφαντικώτερος when the superfluous verse has been deleted). In these cases, as has already been remarked, the ancient philologist’s only further concern was to elucidate the passage by explaining how the nominal sentence must be understood, i.e., how to supply the missing verb mentally. We must therefore conclude that, as the critical and philological material contained in the scholia was transcribed and epitomized over the course of the centuries, one element of the argumentation contained in the present scholion was lost, namely, the indication of the verb that ought to be understood in the resulting nominal sentence of v. 136 (for this argumentative sequence in the scholia, see the examples that have been collected and discussed in the appendix). At this point, we can reconstruct with precision Aristarchus’ argument regarding vv. 136–137 in each of its elements, step-by-step. He considered v. 137 to be superfluous (περισσός) within Telemachus’ speech, and for that reason he proposed that it be deleted (διὸ ἀθετεῖται); this requires that one understand mentally (ex. gr. δεῖ προσυπακούειν) the verb ἔσται in the phrase νέμεσις δέ μοι

 16 Anyone who has even limited familiarity with the scholia will not be at all surprised to see a parallel cited from a poet of the Classical or Hellenistic period (in addition, of course, to the vast majority of parallels cited internally from within the text of the Homeric poems themselves): it is an extremely common practice, and it seems pointless to spend ink in accumulating long lists of comparanda. To give just one example in its entirely, see schol. HMQR on Odyss. IV 1: Ζηνόδοτος δὲ γράφει καιετάεσσαν ἀντὶ τοῦ καλαμινθώδη. δοκεῖ δὲ καὶ Καλλίμαχος ἐντετυχηκέναι τῇ γραφῇ δι’ ὧν φησιν «ἵππους καιετάεντος ἀπ’ Εὐρώταο κομίσσαι» (p. 169, 4 Dind.). In this case the passage from Callimachus appears to be cited precisely in order to support a textual intervention, in a manner analogous to that which we have attempted to reconstruct in the scholion discussed in the present article. For other examples, cf. schol. on Odyss. XIV 199 and XVII 599.

  Aristarchus on Odyssey II 136–7 ἐξ ἀνθρώπων within the preceding verse (v. 136); a parallel is found in Callimachus, which reads χαλεπὴ μῆνις ἐπιχθονίων. The scholion, in the form in which it has come down to us, presents this argument not only in reduced form, but also with a certain confusion as regards the order of its various elements. There remains, lastly, a slight uncertainty with regard to the use of κατὰ + acc. to introduce the citation; one is tempted to suggest the correction ἄλλως τε καὶ Καλλίμαχος (cf. e.g. schol. V on Odyss. XVII 599; schol. V on Odyss. XXII 299; etc.). * Another interesting problem is presented by the first part of the scholion we are discussing (schol. MV on v. 134). Since this part has not been analyzed above, I reproduce it again, for convenience: τοῦ πατρὸς] τὸ δὲ τοῦ πατρὸς οὐ περὶ Τυνδάρεω, ἀλλὰ περὶ Ὀδυσσέως· οὐ γὰρ ἀπεγνώκει αὐτὸν ἐπειδή φησιν «ὀσσόμενος πατέρ’ ἐσθλὸν ἐνὶ φρεσίν» (α 115).

The same issue is discussed in almost the same way in schol. BEHQ on the same verse (v. 134): ἐκ γὰρ τοῦ πατρὸς κακὰ πείσομαι] εἰ ἐπανέλθοι ὁ Ὀδυσσεύς. οὐ γὰρ ἀπεγνώκει αὐτὸν παντελῶς. τινὲς δέ, ἐκ γὰρ τοῦ πατρός, τοῦ αὐτῆς πατρός· καὶ στίζουσι τῷ Ἰκαρίῳ.

The ancient philologists were troubled by πατρός in v. 134: to whom exactly does Telemachus refer when he says ἐκ γὰρ τοῦ πατρὸς κακὰ πείσομαι? In schol. BEHQ we find two possible interpretations, without a stated preference for one over the other: the first interpretation understands Telemachus to be referring to his own father, Odysseus (εἰ ἐπανέλθοι ὁ Ὀδυσσεύς. οὐ γὰρ ἀπεγνώκει αὐτὸν παντελῶς); the second interpretation assumes that he is referring rather to Icarius, the father of Penelope (τινὲς δέ, ἐκ γὰρ τοῦ πατρός, τοῦ αὐτῆς πατρός). In schol. MV, on the other hand, we find a clearly stated preference for Odysseus, which is expressed in a language that recalls that of the other scholion (οὐ γὰρ ἀπεγνώκει αὐτόν); and we also find a reference to α 115 in support of this interpretation. It is clear, then, that MV preserves more of the source material from which both of the scholia must ultimately derive: obviously, it grants greater weight and space to the interpretation that it accepts, whereas BEHQ is more balanced precisely because it does not state an explicit preference for one interpretation over the other. The text of MV, however, presents an obvious problem in the phrase τὸ δὲ τοῦ πατρὸς οὐ περὶ Τυνδάρεω, ἀλλὰ περὶ Ὀδυσσέως. There is no doubt that the

Aristarchus on Odyssey II 136–7  

interpretative issue at stake is whether πατρὸς refers to Icarius or to Odysseus:17 so what in the world is Tyndareus doing here? It is clear that the text is corrupt; but what kind of corruption should we assume has taken place? Bérard18 thought that an original, correct text presented ΤΟΥ ΙΚΑΡΙΟΥ and that this could have been mistakenly read as ΤΥΝΔΑΡΕΩ; such an error might, possibly, have been encouraged by the fact that Icarius and Tyndareus were brothers and appeared in some of the same mythical narratives, and as a result they are sometimes mentioned together in the ancient commentaries (cf. e.g. schol. HMQ on Odyss. II 12). In that case, a simple solution would be to emend the text of the scholion: οὐ περὶ (τοῦ) Ἰκαρίου, ἀλλὰ περὶ Ὀδυσσέως. Such a solution is fair enough, in and of itself, but it does leave a certain bitter taste in the mouth: the mass of ancient critical and exegetical materials (it is certainly not a textbook case of the confusion of letters); and the supporting argument has to assume a scholiast who was, on the one hand, very well-informed regarding mythography, and yet, on the other hand, so inattentive as to confuse the names of two brothers, in a context in which one of the two is not relevant at all and, indeed, the whole question turns upon the identity of the “father” mentioned by Telemachus. To repeat: Bérard’s explanation is the simplest, and it is certainly possible; it definitely ought not to be discarded outright. But perhaps it is not the only explanation, and it is enticing to consider another one. Indeed, it is quite possible that the error in the scholion could be explained in a somewhat different manner. If we consider the long process of epitomization, of which the scholia contained in the medieval manuscripts are the last stage, it is not difficult to believe that behind the present scholion there stands a richer and more ample source. A mythographical question immediately calls to mind the common and well-known mythographical historiae that are to be found in the Homeric scholia. Let us therefore direct our attention to certain pieces of information present in Eustathius and in other Homeric scholia, from which perhaps we may derive some useful hints. The question of Penelope’s possible return to her father’s house in view of a second marriage appears at Odyss. I 275–276. I reproduce below a portion of Eustathius’ commentary (1417.23 ff.): Ἰστέον δὲ καὶ ὅτι ὁ ἀνωτέρω ῥηθεὶς τῆς Πηνελόπης πατὴρ Ἰκάριος, ἀδελφὸς ἦν Τυνδάρεω τοῦ Λάκωνος, τοῦ πατρὸς τῆς Ἑλένης. οἳ τῆς οἰκείας ἐκπεσόντες ὑπὸ Ἱπποκόωντος, ἦλθον  17 See also Eustathius ad loc., 1438.33 ff.: ποῖα δὲ πολλὰ ἀποτίσει; ἐκ τοῦ πατρὸς κακά, φησι, πείσομαι, ὅ ἐστιν ἐκ τοῦ Ὀδυσσέως εἰ τυχὸν ἐπανέλθοι. οὕτω γάρ τινες καὶ οὐκ ἐπὶ τοῦ Ἰκαρίου νοοῦσι τὸ πατρός. 18 In the article cited above (n. 5), pp. 204 ff.

  Aristarchus on Odyssey II 136–7 εἰς Θέστιον Πλευρωνίων ἄρχοντα. καὶ τὴν πέραν Ἀχελῴου συγκατεκτήσαντο. ὕστερον δέ, Τυνδάρεως μὲν τὴν Θεστίου θυγατέρα Λήδαν λαβών, ἐπανῆλθεν. ἀφ’ ἧς ἐγένοντο αὐτῷ Ἑλένη καὶ οἱ Διόσκουροι. Ἰκάριος δὲ ἐπέμεινε μέρος φασὶν ἔχων τῆς Ἀκαρνανίας. ἔνθα καὶ τεκνωσάμενος Πηνελόπην ἐκ Πολυκάστης, τῷ Ὀδυσσεῖ πρὸς γάμον ἐξέδοτο.

It is important to note that the father of Penelope is mentioned (without being named explicitly) for the first time in the poem precisely here, at α 276: a very apt moment, therefore, to insert in the commentary a mythographical historia concerning the character who has just been introduced in the poem. In the scholia, however, nothing is found resembling the historia in Eustathius19: did it once exist in ancient commentaries, and Eustathius derived his account from a source that had suffered fewer losses than that which has come down to us? Or perhaps the learned Byzantine scholar has simply added his own material, which he derived from a completely different source altogether? The very same question reoccurs at Odyss. II 50 ff. and XV 14 ff. Icarius and Tyndareus appear together in schol. HMQ on II 52: ... δῆλον ἐκ τούτων ὅτι ὁ Ἰκάριος Ἰθακήσιος ἦν ... Ἰκάριος γὰρ καὶ Τυνδάρεως ἀδελφοὶ καὶ περιείχετο τῇ κατάρᾳ ἡ Πηνελόπη, ὡς θυγάτηρ ἐκείνου τοῦ Ἰκαρίου ...

and also in schol. H on XV 16: ὅτι Ἰθακήσιος ὁ Ἰκάριος, οὐ Σπαρτιάτης, ἀδελφὸς Τυνδάρεω. ἀδελφοὶ τῆς Πηνελόπης δύο, Σῆμος καὶ Αὐλήτης· ὁ δὲ Ἰκάριος ἐκ Μεσσήνης ἦν ...

The presence of this sort of annotation within the Odyssean scholia (to which Bérard himself refers) leads one to suspect that something similar may have existed in the source material behind the scholion under discussion. Was it a fairly extensive historia, or perhaps simply a brief statement, itself reduced from a fuller discussion in the course of successive prunings? Naturally it is impossible to say; but if one accepts this hypothesis (which I submit for consideration fully aware that it can only be tentative) the error contained in the text of the scholion cannot be a mechanical substitution, pure and simple; rather it will be the lingering trace of an earlier redaction, which was richer and more ample than that which we read today. An involuntary omission, a moment of distraction, an awkward epitome, the final fruit of a long process of compression and reduction of the ancient critical and exegetical material — whatever form the  19 One finds merely the tiny annotation πατρός] τοῦ Ἰκαρίου (which however does not appear in Dindorf’s edition; it is to be found in the specimen volume of Ludwich’s edition, cited above, in n. 1).

Aristarchus on Odyssey II 136–7  

loss may have taken, of what the text once contained it has left only an apparent substitution, due to which we find Tyndareus in the place that belongs to Icarius. * We have examined the two parts of a scholion dedicated to a passage from the Odyssey, addressing the textual problems and the interpretative difficulties that each part presents in the form in which it has come down to us. In the first case (Aristarchus’ argument regarding vv. II 136–137), I believe that the account presented here offers the most plausible explanation of what we find in the scholion and how to understand it; in the second case (the substitution of Tyndareus for Icarius), the hypothesis presented here, if not definitive, deserves to be taken into consideration. From these investigations into specific and local problems, we can draw the following general conclusions. It happens every so often that the brief annotations found in the scholia, cut into pieces and whittled down over time, present us with enigmatic statements or obvious errors. In such cases, the most fruitful approach is not necessarily that of emending the text in order to restore a hypothetical or even obvious correctness; indeed, such emendations may obscure the true nature of the problem and fail to do justice to the concrete facts of the textual tradition. Sometimes a more productive explanation can be found, instead, by identifying in these problematic passages the last traces of the process of transformation, reorganization, and impoverishment that has affected the philological material of the scholia during the long history of their transmission. Each scholion that we read today in the medieval manuscript is the result of a process that began over a millenium before our text was written; every comment represents the end of an itinerary that took numerous turns and passed through various stages, each of which has left its own peculiar mark. The enormous mass of erudition, which the ancient philologists and scholars employed in the service of great works of literature, followed a tortuous course and was channeled over time into numerous different streams, until it reached the form we have: scattered scraps, sometimes textually corrupt, often enigmatic and obscure. This process entails a certain continuity, of course; but it equally entails a profound distance, a distance measurable in centuries of cultural history, during the course of which the interpretative and philological material of these texts has been frequently reconsidered, reordered, condensed, dispersed, epitomized, or otherwise transformed. Every small glance, therefore, that with painstaking care manages to penetrate beyond the state of the text as it has been preserved for us and allows us to

  Aristarchus on Odyssey II 136–7 reconstruct even a small part of the history of these transformations, is of great importance. One precious and significant form of testimony, of course, is provided by papyrus finds, when these restore to us excerpta from commentaries or fragments of scholarly works of various types, which constitute one of the numerous intermediate phases in this long process.20 But sometimes an enigmatic passage in the scholia, or even a blatant textual error, if considered in the proper light and within the proper perspective, can help us to open a window onto the tradition and thereby to understand better how the text we are reading came to take its present form.

Appendix I offer here a brief collection of comparanda for the foregoing discussion. These comparanda are not meant to be exhaustive; they are the provisional byproducts of the research that has informed the elucidation of the second part of the scholion MV on Odyss. II 134 (see above), where Aristarchus’ athetesis of v. 137 is discussed. To begin, a rather similar case is found at schol. A on Il. IX 416: ἔσσεται, : ἀθετεῖται, ὅτι νομίσας τις κρέμασθαι τὸν λόγον προσέθηκεν αὐτόν· καὶ γὰρ κατὰ τὸ περισσὸν ἐπιλέγεται «οὐδέ κέ ὦκα». δεῖ οὖν κοινὸν λαβεῖν τὸ ἔσται ἀπὸ τοῦ προκειμένου τοῦ «ἄφθιτον ἔσται».

Here too the issue at stake concerns the deletion of a verse judged to be περισσός, as a result of which one is left with a nominal sentence in the preceding verse. There is one difference, however, namely that, in this case, one can have recourse to an ἀπὸ κοινοῦ construction, taking advantage of the verb ἔσται which is found two verses previously. The concept of ἀπὸ κοινοῦ (κοινὸν λαμβάνειν, κοινὸν εἶναι) can be properly applied, naturally, in those cases in which the verb that is missing in the nominal sentence does not need to be supplied ex nihilo but rather is present in the text within another sentence or clause. The following scholia may be cited as further examples. Schol. A on Il. I 443: παῖδά τε σοὶ ἀγέμεν, : ὅτι τὸ ἀγέμεν κοινόν, ὥστε γίνεσθαι τὸν ἑξῆς περισσόν· διὸ ἀθετεῖται.

 20 G. Arrighetti has recently reexamined this subject in its entirety, and he has offered a better and more efficient framework for understanding it than has been previously available. See Arrighetti 1977a; 1977b.

Aristarchus on Odyssey II 136–7  

Schol. A on Il. IX 43–44: νῆες δέ τοι ἄγχι θαλάσσης: ὅτι ἀπὸ κοινοῦ τοῦ «πάρ τοι ὁδός» τὸ πάρεισί σοι· ὅπερ οὐ συννέντες τινὲς προστεθήκασι τὸν ἑξῆς. — 44 ἑστᾶσ’, : ἀθετεῖται, ὅτι περισσός ἐστι καὶ μὴ προσκειμένου αὐτοῦ ἐμφαντικώτερος ὁ λόγος γίνεται· ἐφορμοῦσιν αἱ νῆες πορευσόμεναι. (In this case, after the athetesis of v. 44, πάρ (sc. πάρεστι) from v. 43 must be applied ἀπὸ κοινοῦ to νῆες as well, with a slight (mental) accomodation of the original singular verb to a plural subject.) Schol. T on Il. XIV 158: : περισσόν· δύναται γὰρ ἀπὸ κοινοῦ τὸ «ἔγνω» (Ξ 154). Schol. A on Il. XXI 479: ἀλλὰ χολωσαμένη : ὅτι κοινὸν δεῖ δέξασθαι τὸ «προσέφη» (Φ 478)· ἀλλὰ χολωσαμένη Διὸς αἰδοίη παράκοιτις προσέφη. οὕτως οὖν κἀκεῖ «παῖδά τέ σοι ἀγέμεν Φοίβῳ θ’ ἱερὴν ἑκατόμβην» (Α 443)· ἀπὸ κοινοῦ γὰρ τὸ ἀγέμεν, καὶ περιττὸς ὁ ἑξῆς· «ῥέξαι ὑπὲρ Δαναῶν» (Α 444) (cf. schol. bT).

(On the application of the σχῆμα ἀπὸ κοινοῦ in cases of athetesis διὰ τὸ περισσόν within Aristarchus’ Homeric criticism, cf. Roemer, Aristarchs Athetesen, cit., pp. 196–200.) Let us now examine several cases in which the scholia speak explicitly of “understanding” a verb in a given sentence, that is to say, of adding mentally a verb that is not present at any point in the text. A useful parallel for the Odyssean scholion that we have discussed above may be found at schol. A on Il. VII 353: ἔλπομαι ἐκτελέεσθαι, : ἀθετεῖται, ὅτι ἀγνοήσας τις ὅτι ὑπακοῦσαι δεῖ τῷ «οὔ νύ τι κέρδιον ἡμῖν» (H 352) τὸ ἔσται, ὡς ἐλλείποντος τοῦ λόγου προσανεπλήρωσεν.

The athetesis of v. 353 leaves a nominal sentence in the preceding verse; therefore, in that sentence the verb ἔσται must be understood. We may compare also: Schol. A on Il. XXI 570: ἔμμεναι· αὐτὰρ οἱ : ἀθετεῖται, ὅτι ὡς ἐλλείποντος τοῦ λόγου ἐνέταξέ τις αὐτόν. δεῖ δὲ τῷ «θνητὸν δὲ ἕ φασ’ ἄνθρωποι» (Φ 569) προσυπακούειν τὸ εἶναι (cf. schol. T). Schol. A. on Il. XXIV 45: γίνεται, ἥ τ’ ἄνδρας : ἀθετεῖται, ὅτι ἐκ τῶν Ἡσιόδου (Op. 318) μετενήνεκται ὑπό τινος νομίσαντος ἐλλείπειν τὸν λόγον ... δεῖ δὲ ἔξωθεν προσυπακοῦσαι τὸ ἐστίν, ἵν’ ᾗ οὐδέ οἱ αἰδώς ἐστι (Ω 44)· καὶ πλήρης ὁ λόγος. (Note the specification ἔξωθεν.)

In these three scholia, although it is not explicitly stated, the criterion of περισσόν is clearly being applied: in each scholion an argument is developed which holds

  Aristarchus on Odyssey II 136–7 that certain verses are to be deleted, and that, in fact, they were added to the poem by someone who had thought the text was incomplete and therefore went on to complete it himself. The “superfluousness” of the verses with respect to the genuinely Homeric text, therefore, is simply considered obvious, and it is clear that these atheteses are also based on the criterion of περισσόν. Indeed, it is more than probable that, here too, the process of impoverishment discussed above is responsible for the loss of certain parts of the original argumentation — parts which, here at least, can easily be recovered. The three scholia just discussed all present atheteses that result in the need to understand a form of the verb εἰμί. For the sake of comparison, we can now add other cases in which the elliptical syntax is not due to an athetesis; instead, in these cases the scholia address difficult passages already present in the text itself — passages which, in order to be understood correctly, demand that the reader supply mentally something that is not stated by the poet. In some scholia, the concept of ἀπὸ κοινοῦ is explicitly contrasted with that of (προσ)ὑπακούειν: Schol. bT on Ιl. III 327: ἵπποι ἀερσίποδες : τὸ ἔκειντο οὐκ ἔστι κοινὸν ἐπὶ τῶν ἵππων, ἀλλὰ προσυπακούομεν τὸ ἑστήκεσαν, ὡς ἐπὶ τοῦ ... (Κ 407). Schol. AbT on Il. VIII 507: σῖτόν τ’ ἐκ μεγάρων: οὐκ ἔστιν ἐπὶ τοῦ σίτου κοινὸν τὸ «ἄξεσθε» (Θ 505)· οὐδὲν γὰρ τῶν ἀψύχων ἄγεται, φέρεται δὲ μᾶλλον. προσυπακουστέον οὖν τὸ φέρετε. Schol. bT on Il. X 407: ποῦ δέ οἱ ἔντεα κεῖται : οὐκ ἔστι κοινὸν ἐπὶ τῶν ἵππων τὸ κεῖται· δεῖ γὰρ † προσυπακουστέον † τὸ ἑστᾶσιν (T). — δεῖ τὸ ἑστᾶσι τῷ ἵπποι προσυπακούειν· οὐ γὰρ κοινὸν τὸ κεῖται (b). Schol. Q on Odyss. IX 166: ἐλεύσσομεν] ἄνω τὸ ἐλεύσσομεν ἐπὶ μὲν τῆς γαίας καὶ τοῦ καπνοῦ κοινόν· ἐπὶ δὲ τοῦ φθόγγου ὑπακουστέον τὸ ἠκούομεν, ὡς τὸ «καρπαλίμως οἶνον δὲ μελίφρονα οἰνίζεσθε, σῖτον δ’ ἐκ μεγάρων» (Θ 506–507). ἐπὶ τοῦ σίτου δεῖ ὑπακούειν τὸ φέρετε.

The same terminology is used even when that which must be understood is not a verb. Schol. A on Il. XV 155: ὅτι τὸ ἐναντίον ὑπακοῦσαι δεῖ, ἀλλ’ ἀπεδέξατο. Schol. bT on Il. V 341: καὶ μὴν πολλὰ τῶν ζῴων οὐ σῖτον ἔδουσιν, οὐ πίνουσιν οἶνον καὶ οὔτε ἄναιμα οὔτε ἀθάνατά εἰσιν. δεῖ τοίνυν προσυπακούειν τῷ οὐ σῖτον, ἀλλ’ ἀμβροσίαν· οὐ πίνουσιν οἶνον, ἀλλὰ νέκταρ. Schol. T on Il. X 428: προσυπακουστέον δὲ τὸ μέρος, οἷον «πρὸς μὲν τὸ μέρος τῆς θαλάσσης».

Aristarchus on Odyssey II 136–7  

I believe that this limited review of select passages from the scholia ought to have presented, if not an exhaustive survey, at least a sufficiently clear framework for the problem addressed above. Nevertheless, I would like to mention one more passage, namely, schol. Α on Il. II 681–685: νῦν αὖτοὺς ὅσσοι τὸ Πελασγικὸν Ἄργος ἔναιον : μακρόθεν ὑπακούεται τὸ «ἔσπετε» (Β 484) ἢ τὸ «ἐρέω» (B 493).

In this case, even though the verb that must be understood is in fact present in the text, it is so far distant that it is difficult to apply the concept of ἀπὸ κοινοῦ, and the scholiast, therefore, must resort to the concept of ὑπακούειν; at the same time, however, the scholiast felt the need to add the specification found in the adverb μακρόθεν (compare, above, schol. A on Il XXIV 45, where we find instead ἔξωθεν προσυπακοῦσαι). If I am not mistaken, then, the examples collected here are enough for our purpose: they present sufficient grounds for believing that, in the second part of schol. MV on Odyss. II 134, as a result of the process of epitomization and compression that affected the critical and philological material in the scholia, part of Aristarchus’ argument regarding the athetesis of v. 137 was lost — namely, the part in which the concept of (προσ)ὑπακούειν was introduced, and, thereby, it was explained that, once v. 137 has been deleted, one must understand the verb ἔσται in the clause νέμεσις δέ μοι ἐξ ἀνθρώπων in v. 136. From the scholia that we have examined here, it seems clear that, among the cases of athetesis, those which result in a nominal sentence requiring a verb in the future tense are the rarest of all. C. Guiraud, in his study La phrase nominale en grec d’Homère à Euripide, Paris 1962 (esp. pp. 323 ff.) has shown that in poetic language nominal sentences which require one to understand a verb in the future or in a past tense — keeping in mind that this observation applies only to the verb εἰμί — are far less common than those which require one to understand a verb in the present. In the examples above, we have seen only two examples requiring the future tense: Il. IX 416 (where the difficulty is less severe and indeed is resolved by appealing to ἀπὸ κοινοῦ and taking the verb ἔσται from two verses prior, that is to say, by taking advantage of a form already present in the text elsewhere), and Il. VII 353, which is in every way similar to the case discussed in the present article and, in fact, has explicit recourse to the concept of ὑπακούειν.

 On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401 The hypothesis to Od. β restored by P.Oxy. 3160 + P.Strasb. 1401 was the subject of a recent article of mine, which, due to a series of unfortunate circumstances, proved to be outdated already at the time of its publication, and to a degree that is not insignificant.1 It seems worthwhile therefore to return to the subject and to reconsider what I had written, in order to eliminate what is no longer valid, to reexamine the parts that do remain valid, and to expand the discussion by adding a few remarks. In that article, in fact, I had not taken into consideration a study by W. Luppe, who had recognized that the two papyrus fragments (which I had considered distinct) not only belong to the same scroll, but even supplement each other directly.2 This observation, admittedly, was only possible if one had been able to examine at least the photographs of the fragments; due to the collocation of the traces of ll. II 6–10 of the editio princeps of P.Oxy 3160, along with a few imprecisions in the transcription of the editio princeps of P.Strasb. 1401, there was no reason to suppose, on the basis of the published texts alone, that the second fragment might supplement precisely the lacuna present in the first.3 But we now know, thanks to Luppe’s observations, that we are dealing with parts of only one document, rather than two, and what is more, we can now read the hypothesis to β in its entirety. As a result, whereas my earlier analysis of the overall structure of the text of the hypothesis to β remains substantially valid (and therefore I reproduce it here more or less unmodified, adding only an analysis of the text of the hypothesis to γ, which I had not previously considered), I must now reject the other considerations I had made based on the mistaken conviction that these were two completely distinct documents and that, therefore, we had discovered a unique case in which the same hypothesis was attested twice among extant papyri. For the present, we must instead uphold the statement that, in the few cases in which it is possible to make a comparison, whenever different papyrus fragments present hypotheseis to the same book of Homer, the hypotheseis themselves are always different; in other words, at the moment we do not have any cases of identical hypotheseis to the same book of Homer attested twice on papyrus. The validity of this assertion is naturally  1 Montanari 1980. 2 Luppe 1977. 3 Ed. pr. of P. Oxy 3160: M.W. Haslam, The Oxyrhynchus Papyri, vol. XLIV, London 1976, 52–57; ed. pr. of P. Strasb. gr. 1401: J. Schwartz, Papyrus et tradition manuscrite, ZPE 4, 1969, 175–176. https://doi.org/10.1515/9783110772371-029

On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401  

limited, due to the small number of cases in which comparison is even possible; currently there are no counterexamples, but one might appear at any given moment, thanks to a new find.4 Taken as one item, the document P.Oxy. 3160 + P.Strasb. 1401 forms part of a bookroll that contained hypotheseis and scholia minora on the Odyssey. The extant portions preserve the remains of four columns of text that contained hypotheseis and scholia minora related to the first three books of the poem: a few scraps of words and a handful of glosses remain from the section dedicated to the latter part of book α, and it seems reasonable to suppose that two columns have been lost which treated the earlier part of that book; substantial sections have been preserved of the material dedicated to book β, namely, the entire hypothesis and a large portion of the glosses; as for book γ, we possess only the first half or so of the hypothesis. We can therefore deduce that this was a volume containing, for each book of the Odyssey, a hypothesis, followed by a fairly sparse set of glossographical annotations; it began with α and continued, presumably, to the end of the poem, given that a mere six columns cover the first two books in their entirety and the beginning of the third. I do not wish to discuss here the glossographical component (i.e., the scholia minora); I want to discuss instead the text of the hypotheseis. That of α has been completely lost, but we possess the hypothesis to β in its entirety, and at least half of the hypothesis to γ. Let us examine the hypothesis to β, which occupies col. II, ll. 7–25. I present the text according to the reconstruction proposed by Luppe, without indicating the correction of purely graphic mistakes:

10

15

20

ἦμος δ’ ἠ[ρ]ι̣[γ]έ̣ν̣εια φάνη̣ [ῥοδ]οδ̣άκτυλο̣ς Ἠώ̣ς — ἅμα ἡ̣[μέρ]α̣ Τηλέμαχος, τοὺς Ἰθακ̣[ησίο]υ̣ς ε]ι̣ ς ἐκκλησίαν̣ συναγαγώ[ν,] κ̣ελ]εύει ἐξιένα[ι] ἐκ τοῦ οἴκου α[ὐ]τ̣ οῦ. μ̣ὴ πεισθέντω̣ν δὲ ἀποπλε̣ ῖ̣ ε̣ ἰ̣ ς Πύλον καὶ εἰς Λακεδαίμονα καὶ πορευθεὶς πεύθεται περὶ τοῦ πατρὸς καὶ εἰ μὲν ζῶντα αὐτὸν πύθοιτο, ἀνέξασθαι ἄλλον ἐνιαυτόν, εἰ δὲ ἀποθανότα, κενοτάφιον αὐτῷ κατασκευάσαι καὶ τὴν μητέρα ἀποπέμψαι εἰς τὸν τοῦ πατρὸς αὐτῆς Ἰκαρίου οἶκον ὅπως αὐτὴν ἐκδώσει ᾧ

 4 For the Homeric hypotheseis cfr. Henrichs 1973, 23 ff.

  On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401

βούλεται. διασκωφθεὶς δὲ ὑπὸ τῶν μνηστήρων ἐπὶ τούτοις συμπραττούσης Ἀθηνᾶς ναῦ καὶ ἐρέτας συναγαγὼν ἀ̣ποπλε[ῖ.

l. 11: The mention of the suitors is necessary both for the sense of the sentence itself as well as for the referent implied in the following genitive absolute; compare the corresponding sentences in the hypotheseis found in the MSS, reproduced below. In contrast to Luppe, I prefer to supplement the text explicitly by adding τοὺς μνηστῆρας, on the assumption that we are dealing here with a simple case of mechanical omission. ll. 12 ff.: As we will see in greater detail below, the hypothesis here rather faithfully reproduces Telemachus’ speech at vv. 208ff, in which he expresses his intention to journey to Pylos and Sparta in search of news of his father and then to act on the basis of what he will have found out; it is very surprising, then, that the hypothesis seems to report as facts that have occurred (l. 12 ἀποπλεῖ, l. 14 πεύθεται) what is merely announced by Telemachus as a plan (ll. 214 ff.) — a plan which, moreover, he expresses in future-tense verbs and which will effectively take place in the following books of the poem. The difficulty is made still greater when one realizes that it is not clear what governs the infinitives in ll. 16 and 18 f. This textual problem has been analyzed well by Luppe: he maintains that the text is not damaged, but rather that it is simply characterized by poor syntactic construction on the part of the original writer; he rejects, therefore, the temptation to intervene with extensive emendation, and instead supposes that from κελεύει in l. 10 f. one must understand an implied verb of speaking (λέγει ὅτι): “And since they do not agree, (he says that) he is leaving by ship to go to Pylos and to Sparta and, once he arrives there, he asks for news of his father ...” At the moment, I do not know of any other solution better than this one. We can now analyze the text of the hypothesis taken as a whole. Its structure is simple: syntactically it consists of three periods, which very clearly articulate the three sections of the summary. The first part (A, ll. 8–11) states that at daybreak Telemachus summons the assembly of the Ithacans and he orders the suitors to leave his home. The second part (B, ll. 12–22) is the most developed, reproducing quite faithfully the content of Telemachus’ speech at vv. 208–223, which has evidently been identified as a crucial moment in the development of the plot: in those verses Telemachus, having realized that the suitors are not at all willing to leave, announces his plan to travel to Pylos and to Sparta in search of news of his father, saying that if he receives positive news, he will wait another year, whereas if learns that his father is dead, he will construct a cenotaph

On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401  

for him and he will consent to a second marriage for his mother. The third and final part (C, ll. 22–25) relates how, after being mocked by the suitors, Telemachus (with Athena’s help) prepares his ships and assembles a crew for his voyage. Let us now compare this hypothesis restored to us via papyrus with the two hypotheseis to β that appear in the MS tradition of the Odyssey. I reproduce here the texts as found in Dindorf’s edition (hypoth. I: p. 71, 19–22; hypoth. II: p. 72, 2–6), but I divide each text, for ease of comparison, into three sections, corresponding to the three parts by means of which, as we have seen, the structure of the hypothesis on papyrus was articulated. hypoth. I A)

συναγαγὼν ἐκκλησίαν Τηλέμαχος παραγγέλλει τοῖς μνηστῆρσιν ἐξιέναι τῆς οἰκίας τοῦ Ὀδυσσέως.

C)

καὶ λαβὼν παρὰ μὲν Εὐρυκλείας τὰ πρὸς τὴν ἀποδημίαν ἐπιτήδεια, παρὰ δὲ τῆς Ἀθηνᾶς ἑταίρους τε καὶ ναῦν, εἰς πλοῦν ἀνάγεται ἡλίου δύναντος.

hypoth. II A)

ἅμα ἕω Τηλέμαχος συναγαγὼν εἰς ἐκκλησίαν τοὺς Ἰθακησίους κελεύει τοὺς μνηστῆρας ἀπαλλάττεσθαι τῆς οἰκίας.

B)

καὶ ναῦν αἰτήσας παρ’ αὐτῶν, ὅπως εἰς Πύλον καὶ Σπάρτην πορευθῇ, ἀποτυγχάνει.

C)

παρὰ δὲ Νοήμονος λαβὼν καὶ ἐφόδια παρὰ Εὐρυκλείας τῆς τρόφου αὐτοῦ λάθρα τῆς μητρὸς ἀποπλεῖ.

As one can see, the hypothesis on papyrus has more points in common with hypoth. II, even if parallels with hypoth. I are not entirely lacking. These convergences with hypoth. I, however, are concentrated in the first part (A), which in any case is also the section in which the two manuscript hypotheseis themselves are the most similar; in what follows, however, we can notice phrasing similar to that of the papyrus only in hypoth. II, namely, in the verbs πορευθῇ and ἀποπλεῖ (for which cf. l. 14 and ll. 12 and 25, respectively). As for the structure of each text, with reference to the three parts into which the papyrus hypothesis is articulated, we can remark as follows: in the first part (A), there is a certain homogeneity among all three summaries; in the third (C), the manuscript hypotheseis contain greater scope or at least greater richness of detail, and in any case they both prove to be quite different from the papyrus text in substance. It is instead the central part (B) that seems to warrant the most attentive examination: in hypoth. I this part is lacking entirely; in hypoth. II it is represented by the

  On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401

sentence καὶ ναῦν... ἀποτυγχάνει, which is much more concise and much less detailed than what we read on the papyrus, where this section is quantitatively the most extensive part of the summary and qualitatively the one most thoroughly and explicitly developed. It seems, therefore, that these comparisons fully confirm, even strengthen, the conclusions at which we had arrived by means of an internal analysis of the papyrus text: namely, that the central part (B) of the hypothesis to Odyssey β restored to us by P.Oxy. 3160 + P.Strasb. 1401 forms its most distinctive and characteristic element. This section, in fact, is not only more ample and more detailed than the rest; it is also exceptional in its high degree of adherence to the original text (as I already mentioned above) in the reproduction of Telemachus’ speech of vv. 208–223. Thus, a relatively large portion of the hypothesis covers a relatively small portion of the original poem; indeed this section of the hypothesis turns out to be, more or less, a paraphrase of the original (even with something more than the original), whereas the other two sections condense hundreds of verses into a few lines of prose. This element of the papyrus hypothesis — the close paraphrase of a small number of verses in the central part (B) — is notable in and of itself, and it is not commonly found in other similar documents. It undoubtedly demonstrates a desire on the part of the writer to privilege and to highlight, within the overall summary of the book, the decisive moment (when Telemachus decides to make the journey) and also to foreshadow the potential outcome of that journey; in other words, the writer has tried to emphasize what is, structurally, the most significant moment within the Telemachy as a whole. However, there is another interesting aspect of this section of the hypothesis which, it seems to me, renders it even more particular and noteworthy. For in this section we can see what might be termed a certain ‘hermeneutic intent’, specifically in the lines that read: καὶ τὴν μητέρα ἀποπέμψαι εἰς τὸν τοῦ πατρὸς αὐτῆς Ἰκαρίου οἶκον ὅπως αὐτὴν ἐκδώσει ᾧ βούλεται (ll. 19–22). These words correspond to Telemachus’ statement, in the original text, καὶ ἀνέρι μητέρα δώσω (v. 223), with which he concludes the speech that is represented in the central part of the hypothesis. Now, it is clear that in this part the writer has chosen to expand upon the original poetic text and to explain clearly the custom by which the woman must be sent back to her father, since it is the father’s task to arrange a subsequent marriage. In the poem, this practice is referenced not only at α 275 ff., in the speech in which Athena urges Telemachus to make the journey in search of news of his father (and vv. 280 ff. of this speech will be repeated at β 212 ff.), but also in two other moments of β: these are, first, vv. 50– 54, where the expression in v. 54, (Icarius) δοίη δ’ ᾧ κ’ ἐθέλοι (sc. Penelope), is an excellent parallel for the phrasing ἐκδώσει ᾧ βούλεται found in the papyrus

On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401  

hypothesis (ll. 21 f.); and second, vv. 130 ff., where we find Telemachus using the words κακὸν δέ με πόλλ’ ἀποτίνειν | Ἰκαρίῳ, αἴ κ’ αὐτὸς ἑκὼν ἀπὸ μητέρα πέμψω — which words, again, quite readily call to mind τὴν μητέρα ἀποπέμψαι εἰς τὸν τοῦ πατρὸς αὐτῆς Ἰκαρίου οἶκον (ll. 19 ff. in the hypothesis). Regarding this detail, then, there can be little doubt that, after closely paraphrasing the preceding verses, the author of the hypothesis decided to “interpret” the brachylogy present in the poetic text at v. 223 (καὶ ἀνέρι μητέρα δώσω) by referring to the various hints disseminated elsewhere in the very verses of the Odyssey, and thus, in this case, he expands significantly upon the specific verse that he is summarizing. We can now move on to consider the hypothesis to γ, partially transmitted by the same papyrus. Again I present the text in Luppe’s reconstruction; naturally, the supplements with which he filled the various lacunae are to be considered conjectural; but given that what remains of the original document allows us to follow the drift of the argument, we can be fairly sure that Luppe’s text must be close to the truth, perhaps erring only in a few small details. In any case, my analysis is concerned with the overall development of the speech, and to that end Luppe’s reconstruction is fully adequate. The hypothesis to γ is found in col. IV, ll. 4–19, which is followed by a margin, and it would have continued in the next column, which is now lost. 5

10

15

Ἠέ[λι]ος δ’ ἀνόρ[ουσε λιπὼν περικαλλέα λ̣ ί̣ μ̣[νην. ἅμα ἡ̣μέρᾳ Τηλ[έμαχος καὶ Ἀθηνᾶ κατάγο̣ ν̣ τα̣ι̣ [εἰς Πύλον. ἐκεῖ δὲ] κ̣αταλαμβάν̣ [ουσι τούς τ(ε) ἄλλους π̣[ά]ντα̣ς̣ κ̣α̣ὶ̣ Ν̣[έστορα τῷ Ποσειδῶνι θύοντας καὶ̣ [ξενίζονται παρ’ αὐτῷ. πυν[θανομένου δὲ Τηλ[εμ]άχου ἃ ε[ἰδείη περὶ (τοῦ) Ὀδυσσέως, αὐ[τὸς μὲν οὐδὲν εἰδέναι φησὶ Ν[έστωρ, προτίθησ̣ ι δὲ αὐτῷ εἰς Λ̣[ακεδαίμονα πρὸς Μενέλαο[ν πορευθῆναι εἴτε ἐπὶ νεὼς [εἴτε πεζῇ, καὶ τά τ(ε) ἄλλα καὶ ἅρμα̣ [αὐτῷ παρέξειν ὑπισχνεῖται, ε̣ι̣α̣[

What remains of this hypothesis to the third book of the Odyssey is clearly articulated into two sections. The first section (A), ll. 6–11, relates how Telemachus and Athena/Mentor arrive at Pylos, where they find Nestor and the local inhabitants on the beach, preparing a ritual sacrifice to Poseidon, and how they

  On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401

are invited to partake in the banquet. The second section (B), ll. 11 ff., ultimately interrupted by the lacuna, reports Telemachus’ request for news regarding his father, Odysseus, along with Nestor’s negative response and his exhortation to Telemachus to make his way to Sparta and visit Menelaus (to ask news from him as well), continuing the journey either by sea on his ship or else by land, in which latter case Nestor offers to provide Telemachus with the necessary equipment. At this point γ’s summary is interrupted. It would have continued in the next column, but the rest has been swallowed up by the lacuna, and it is not worth the effort to spend time on conjectures as to what it contained; nevertheless, judging from the point at which the summary has arrived, it must have immediately passed to a report of the preparations for the night, of Telemachus’ being entertained by Nestor in the palace, and of Athena’s departure in the form of an eagle, which is recognized by Nestor. Let us compare now what remains of this papyrus hypothesis with the corresponding parts of the hypotheseis found in the MS tradition, dividing the latter into sections as well, in order to facilitate the comparison. I cite the text of Dindorf’s edition (hypoth. I: p. 118, 3–8; hypoth. II: p. 118, 9–14). hypoth. I A)

Τηλέμαχος εἰς Πύλον καταχθεὶς ἅμα τῇ Ἀθηνᾷ ἐν Μέντορος μορφῇ καταλαμβάνει τοὺς Πυλίους θυσίαν ταύρων ἐπιτελοῦντας τῷ Ποσειδῶνι,

B)

καί τι περὶ τοῦ πατρὸς αὐτῷ πυθομένῳ ἐκτίθεταί τινα τῶν Ἰλιακῶν διηγημάτων.

C)

μετὰ τοῦτο ἡ μὲν Ἀθηνᾶ ἐν ὀρνέου μορφῇ κτλ.

hypoth. II A)

Τηλέμαχον ἐλθόντα σὺν Ἀθηνᾷ ξενίζει Νέστωρ,

B)

καὶ διηγεῖται αὐτῷ τὰ συμβεβηκότα τοῖς Ἕλλησι καὶ τὸν ἐκ Τροίας ἀπόπλουν. πυθόμενος δὲ τὰ περὶ τοὺς μνηστῆρας

C)

καὶ γνωρίσας τὴν Ἀθηνᾶν ἀπιοῦσαν κτλ.

The three hypotheseis all contain first sections (A) that, despite some differences in extent and in quantity of facts reported, are essentially equivalent and do not present any distinct elements worthy of note: each one substantially summarizes vv. 1–66 of Od. γ; hypoth. I appears to be more similar to the papyrus hypothesis in terms of both its language and its greater richness of details, whereas hypoth. II is more concise. It is more interesting to compare the second sections

On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401  

(B), in which each of the three summaries takes a different turn. In all three cases, section B covers more or less vv. 67–337, which in the original poem cover: first, Telemachus’ exposition of the reason for his voyage and his request for news regarding Odysseus’ fate; then, Nestor’s first, prolix response, along with his accounts of what occurred after the fall of Troy; and finally the dialogue that follows. Of these roughly three hundred verses, a large part is occupied by Nestor’s reports (on the departure from Troy, on the return voyages of the other heroes, and in particular on the fate of Agamemnon), and indeed these reports are mentioned in both of the manuscript hypotheseis. In the poem, Telemachus also speaks of the situation in his home in Ithaca and he tells Nestor about the suitors, but this is hinted at only in hypoth. II (which for precisely this reason is fuller than hypoth. I in this section). The papyrus hypothesis, however, shows no trace of either of these elements in its second section (B), and we can be practically certain that it never contained them, given that the preserved text breaks off after these points would have been covered; indeed, the papyrus hypothesis, in what remains, has already summarized vv. 323 ff., which means (as we already observed above) that it must have passed immediately afterwards to the material that belongs to section C in the manuscript hypotheseis. In the papyrus hypothesis, then, section B quite clearly consists of three elements: after reporting Telemachus’ request for news, it relates the upshot of Nestor’s long digressions (effectively: the old king of Pylos does not have any news to share regarding what has happened to Odysseus after his departure from Troy); then it reports how Nestor urges Telemachus to go to Sparta to visit Menelaus (which, incidentally, corresponds to the plan that Telemachus had already conceived and declared in the preceding book, as was highlighted explicitly in the hypothesis to β, as we saw above) and how Telemachus will have to make this voyage either by sea, on his own ship, or else by land, with a team provided by Nestor himself and with his son Peisistratus as a guide. These elements correspond to Od. γ 313–329: in other words, once again we find that the papyrus hypothesis privileges a small portion of the original text, and it extracts and highlights something that might have seemed unimportant for the overall summary, but in fact is far from unimportant, namely, Telemachus’ decision whether to go to Sparta by sea or by land. Telemachus will decide to make the journey by land (as even Athena advises him to do, vv. 368–370): and he will in fact leave his ship and his companions from Ithaca at Pylos, and he will depart for Sparta by chariot accompanied by Peisistratus; thus he will arrive at Sparta at the beginning of the following book, and he will begin his return voyage in book ο, stopping at Pylos, where he will collect his companions and his ship, take on board the seer Theoclymenus, and finally make his way to Ithaca (ο 1–300). This is,

  On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401

therefore, a detail that will later have notable consequences in the narration of Telemachus’ journey; and the papyrus hypothesis registers it carefully and intentionally. All that we have seen so far demonstrates that, in section B, the papyrus hypothesis is utterly different from the manuscript hypotheseis: the only shared element is fairly banal and consists in the report of Telemachus’ request for news (cf. hypoth. I; this element is actually omitted in hypoth. II). In sum, the papyrus hypothesis is characterized by the way it privileges and highlights Telemachus’ journey in terms of its purpose and its effective means: since Nestor knows nothing about Odysseus; Telemachus will continue to Sparta (as planned), but he will leave his ship at Pylos and he will go on land by chariot. The intention of the writer of the hypothesis is clear, and it is obtained at the cost of providing a summary that is not entirely “faithful”, since much of what is found in book γ (Nestor’s long tales, to begin with) is simply left out; evidently the writer felt that such material was secondary as regard the essential plotline of the Telemachy (thus revealing a sensibility for the bare ‘plot’ that is, naturally, somewhat elementary). Effectively, the characteristics that distinguish the papyrus hypothesis to γ can be considered in line with what we have observed in our corresponding analysis of the hypothesis to book β. There, in fact, the characteristic element — from the point of view of the structure of the summary, and leaving aside what we have remarked about its ‘hermeneutic intent’ — consisted in the significant space and emphasis that was given to Telemachus’ decision to make a journey in search of news of his father — a journey of which the hypothesis took care to report not only the purpose, but also the stages that had been planned and the possible consequences. In short, these two hypotheseis to books of the Odyssey do not seem to be anodyne summaries with the sole intent of registering the content of each book: in each hypothesis we have analyzed, instead, it is quite clear that the writer was concerned, here and there, to clarify certain points in the text; and it is also quite clear that his decisions regarding what and how to summarize were based on the desire to delineate the essential plotline of the Telemachy. In other words — without intending to exaggerate the value of his work, but also without wishing to diminish it — we can assert that the writer of the hypothesis performed his scholastic task with a certain amount of elementary, but nonetheless attentive, concern for depicting the overall structure of the plot of the poem. In conclusion, our analysis of the hypotheseis to β and γ which have been restored to us by P. Oxy. 3160 + P. Strasb. 1401 has shed light on two elements,

On the Hypotheseis to Odyssey β and γ in P.Oxy. 3160 + P.Strasb. Gr. 1401  

which we may call ‘hermeneutic intent’ and ‘concern for plot structure’. Although it is obvious that such interpretative strategies will be found only to a very small degree and at a very elementary level in texts of this nature, nonetheless it is probable that these categories may prove useful for the study of analogous texts in the future.

 The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing The Alexandrian-born grammarian Ptolemaeus, known as Pindarion (also referred to as ὁ Πινδαρίων), the son of Oroandes (and hence also referred to as ὁ τοῦ Ὀροάνδρου), is one of the many grammarians described as pupils of Aristarchus’, μαθητὴς Ἀριστάρχου: the biographical source is Suda, π 3034. Therefore, Ptolemaeus was presumably active in the second half of the 2nd cent. BC, and quotes in the Homeric scholia show that his work was used by Didymus. The scant evidence pertaining to his works essentially suggests that he was active in two fields: on the one hand the philological analysis and exegesis of poetry, on the other the more specifically grammatical field.1 The same individual has been identified in a quote by Apollonius Dyscolus, De con. p. 241, 14 Schn. (GG II 1), which mentions a Πτολεμαῖος ὁ ἀναλογητικός:2 the grammarian designated by this epithet has been identified as Ptolemaeus Pindarion, chiefly on the basis of what is attributed to him in the most extensive testimony we have concerning his grammatical theory, namely Sextus Empiricus, Adv. math. I (Adv. gramm.), 202 ff. A champion of analogy, Ptolemaeus  1 Bibliography on Ptolemaeus Pindarion: Boeckh 1821, 654–655; A. Gräfenhan 1843–1850, II, 43; Beccard 1850, 64; La Roche 1866, 72; Blau 1883, 17 ff.; Lehrs 1882, 27; Ludwich 1884–1885, I, 50; Susemihl 1892, II, 155 ff.; Christ/Schmid/Stählin 1920, 268, 441, 444; Tolkiehn 1925, col. 2442; Dahlmann 1932, 27, 60; Dihle 1962; Mariotti 1966, 82 ff. No collection of Ptolemaeus Pindarion’s fragments has been published: most quotes come from Homeric scholia by Didymus and Herodian (a list in Susemihl 1892, n. 47; Mariotti 1966, n. 26), but also other sources — as we shall see — provide various fragments. The attribution to “Ptolemaeus” in the sch. to Bacchylides fr. 20A, 19 = P.Oxy. 1361 fr. 5, I 13 remains doubtful: cf. ed. princ. ad loc..; Körte 1918, 124 f.; Snell-Maehler ad loc. and p. XVI, p. 132 (test 10). Material by Ptolemaeus Pindarion has been used by the author of the commentary on Φ of P.Oxy. 221 (on which see Erbse, 1969– 1988, I, p. LIX; V, 78 ff.), as unequivocally emerges from a comparison between col. XVII 10–14 and sch. ex. ad Φ 356 (cf. Erbse, 1969–1988, V, p. 113, test. ad loc.): it is likely, therefore, that the material in col. VII 2 ff., sch. ad Φ 163 (also on the basis of sch. h ad Φ 163: see Erbse, test. ad loc.; already Blau 1883, 17 n. 4) to date back to our grammarian, and in particular to the monograph on Asteropeus mentioned by Suida; this idea was first put forward by Wilamowitz, “Hermes” 35, 1900, 356 = Kl. Schr. IV, 143: cf. Erbse, 1969–1988, V, 90 ff. 2 On the stress in αὐτάρ: see Schneider, comm. ad loc., 243 ff.; sch. A (Herod.) ad K 292: ἧνιν: Πτολεμαῖος ὁ τοῦ Ὀροάνδου ἀνέγνω ἧνιν ὡς “μῆνιν” (A 1 al.)· καὶ τοῦτο ὀφείλει εἶναι τό ἀνάλογον. Ptolemaeus Pindarion is also quoted by Apollonius Dyscolus in De pronom. 79, 25 Schn. (GG II 1) with regard to the text of Odyss. δ 244, where he read αὑτὸν μὲν (instead of αὑτόν μιν): see Schneider, comm. ad loc., p. 106; cf. Ludwich, Allen, Von Der Mühll, app. cr. ad Odyss. δ 244; Sch. Odyss. Ibid.; Eust. 1494, 38. https://doi.org/10.1515/9783110772371-030

The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing  

Pindarion followed his teacher Aristarchus in this area too, in some way taking part — as we shall see in detail later on — in that polemic with the anomalist followers of Crates, which had apparently been triggered by the divergences between the two leading grammarians. In the Sextus Empiricus passage — which is what I intend to focus on here — mention of Ptolemaeus Pindarion is made in an extensive section devoted to grammatical theory that contrasts the anomalist συνήθεια of the Stoic-Pergamon tradition to the ἀναλογία championed by the Alexandrian school. The issue revolved around the definition of the correct form of the Greek language and the rules by which it could be attained: on the one hand, some paradigmatic rules were seen to provide the criterion by which to infer the overall system of this language through the procedure of comparing like to like; on the other hand, anomalies deriving from usage and open to constant increase on the basis of the concrete contribution provided by speakers’ free use of the language were regarded as acceptable. In Sextus Empiricus’ work, Ch. 10, 176 ff. discusses precisely the criterion of Ἑλληνισμός, the method to establish what constitutes correct and pure Greek:3 a standard is needed to determine what can be accepted as good Greek and what must instead be rejected as bad Greek. Sextus Empiricus sets out by distinguishing two types of Ἑλληνισμός (whose aim is τὸ σαφῶς ἅμα καὶ ἀκριβῶς παραστῆσαι τὰ νοηθέντα τῶν πραγμάτων, 176): one based on the γραμματικὴ ἀναλογία, the other on the observation of the κοινὴ συνήθεια; the former is described as ἄχρηστος, the latter as εὔχρηστος (177). The following part of the argument suggests that the κριτήριον τοῦ τί ἑλληνικὸν καὶ τί ἀνελλήνιστον lies in the observation of speakers’ common usage, the κοινὴ συνήθεια (181). Analogy is discarded as an essentially abstract rule, devoid of any genuine foundation, and of no practical usefulness. In § 196 ff. the reasoning goes as follows: even those who proceed according to the method of analogy must also resort to common usage and set out from it, since it is the source of the testimonies and examples on the basis of which the rules to be used for the analogical procedure are established. If it is claimed that a word must have a certain form by analogy with other words, the latter are always drawn from usage; the fundamental criterion adopted in such cases can ultimately be traced back to this usage. Analogy means the comparing of like with like, but this is based on usage, which thus provides the matter and foundations for analogy, ultimately making the latter superfluous. Therefore, rejecting common usage as the criterion for Ἑλληνισμός de facto means also rejecting analogy, which without this criterion rests on noth 3 Concerning this section of Sextus Empiricus’ Adv. gramm. and his testimony on the analogy/ anomaly controversy, see Mette 1952, 30 ff. and F 64 e; Dahlmann 1932, 65 ff.

  The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing ing at all: ἡ δὲ ἀναλογία οὐκ ἰσχυροποιεῖται, εἰ μὴ συνήθειαν ἐχοι τὴν βεβαιοῦσαν (201).4 The argument ultimately leads to the paradoxical conclusion that rejecting usage on the basis of analogy is tantamount to rejecting usage on the basis of usage itself and, consequently, to considering the same thing at once trustworthy and untrustworthy (202 in.). At this point the argument continues as follows (202–208: quoted from J. Mau’s edition): (202) … ἐκτὸς εἰ μή τι φήσουσι μὴ τὴν αὐτὴν συνήθειαν ἐκβάλλειν ἅμα καὶ προσίεσθαι, ἀλλ᾽ ἄλλην μὲν ἐκβάλλειν ἄλλην δὲ προσίεσθαι. ὅπερ καὶ λέγουσιν οἱ ἀπὸ Πινδαρίωνος. ἀναλογία, φασίν, ὁμολογουμένως ἐκ τῆς συνηθείας ὁρμᾶται· ἔστι γὰρ ὁμοίου τε καὶ ἀνομοίου θεωρία, (203) τὸ δὲ ὅμοιον καὶ ἀνόμοιον ἐκ τῆς δεδοκιμασμένης λαμβάνεται συνηθείας, δεδοκιμασμένη δὲ καὶ ἀρχαιοτάτη ἐστὶν ἡ Ὁμήρου ποίησις· ποίημα γὰρ οὐδὲν πρεσβύτερον ἧκεν εἰς ἡμᾶς τῆς ἐκείνου ποιήσεως· διαλεξόμεθα ἄρα (204) τῇ Ὁμήρου κατακολουθοῦντες συνηθείᾳ. ἀλλὰ πρῶτον μὲν οὐχ ὑπὸ πάντων ὁμολογεῖται ποιητὴς ἀρχαιότατος εἶναι Ὅμηρος· ἔνιοι γὰρ Ἡσίοδον προήκειν τοῖς χρόνοις λέγουσιν, Λίνον τε καὶ Ὀρφέα καὶ Μουσαῖον καὶ ἄλλους παμπληθεῖς. οὐ μὴν ἀλλὰ καὶ πιθανόν ἐστι γεγονέναι μέν τινας πρὸ αὐτοῦ καὶ κατ᾽ αὐτὸν ποιητάς, ἐπεὶ καὶ αὐτός πού φησι (α 351)· τὴν γὰρ ἀοιδὴν μᾶλλον ἐπικλείουσ᾽ ἄνθρωποι ἥτις ἀκουόντεσσι νεωτάτη ἀμφιπέληται, τούτους δὲ ὑπὸ τῆς περὶ αὐτὸν λαμπρότητος ἐπεσκοτῆσθαι. (205) καὶ εἰ ἀρχαιότατος δὲ ὁμολογοῖτο τυγχάνειν Ὅμηρος, οὐδὲν εἴρηται ὑπὸ τοῦ Πινδαρίωνος ἱκνούμενον. ὥσπερ γὰρ προηποροῦμεν περὶ τοῦ πότερόν τε τῇ συνηθείᾳ ἢ τῇ ἀναλογίᾳ χρηστέον, οὕτω καὶ νῦν διαπορήσομεν πότερον τῇ συνηθείᾳ ἢ τῇ ἀναλογίᾳ, καὶ εἰ τῇ συνηθείᾳ, ἆρα τῇ καθ’ Ὅμηρον ἢ τῇ τῶν ἄλλων ἀνθρώπων· πρὸς ὅπερ (206) οὐδὲν εἴρηται. εἶτα κἀκείνην μάλιστα δεῖ τὴν συνήθειαν μεταδιώκειν ᾗ προσχρώμενοι οὐ γελασθησόμεθα· τῇ δὲ Ὁμηρικῇ κατακολουθοῦντες οὐ χωρὶς γέλωτος ἑλληνιοῦμεν, «μάρτυροι» λέγοντες καὶ «σπάρτα λέλυνται» καὶ ἄλλα τούτων ἀτοπώτερα. τοίνυν οὐδ᾽ οὗτός ἐστιν ὁ λόγος ὑγιής, μετὰ καὶ τοῦ συγκεχωρῆσθαι τὸ κατασκευαζόμενον ὑφ’ ἡμῶν, τουτέστι τὸ μὴ χρῆσθαι ἀναλογίᾳ. (207) τί γὰρ διήνεγκεν εἴτ’ ἐπὶ τὴν τῶν πολλῶν εἴτ᾽ ἐπὶ τὴν Ὁμήρου συνήθειαν ἐλθεῖν; ὡς γὰρ ἐπὶ τῆς τῶν πολλῶν τηρήσεώς ἐστι χρεία ἀλλ᾽ οὐ τεχνικῆς ἀναλογίας, οὕτω καὶ ἐπὶ τῆς Ὁμήρου· τηρήσαντες γὰρ αὐτοὶ πῶς εἴωθε λέγειν, οὕτω καὶ διαλεξόμεθα. (208) τὸ δὲ ὅλον, ὡς αὐτὸς Ὅμηρος οὐκ ἀναλογίᾳ προσεχρήσατο ἀλλὰ τῇ τῶν κατ᾽ αὐτὸν ἀνθρώπων συνηθείᾳ κατηκολούθησεν, οὕτω καὶ ἡμεῖς οὐκ ἀναλογίας πάντως ἑξόμεθα βεβαιωτὴν ἐχούση‹ς› Ὅμηρον, ἀλλὰ τὴν συνήθειαν τῶν καθ᾽ αὑτοὺς ἀνθρώπων παραπλασόμεθα. (202) … Unless, indeed, they shall declare that they do not accept and reject simultaneously the same common usage, but reject one sort and accept another sort. And that is what is said by the School of Pindarion. Analogy, they say, confessedly sets out from common usage; (203) for it is the consideration of the like and the unlike, and the like and the unlike

 4 On §§ 196–208, see Mette 1952, 24 ff. For the concept that analogy is based on usage, cf. Varro, De ling. lat. IX 2–3: consuetudo et analogia coniunctiores sunt inter se quam iei credunt, quod est nata ex quadam consuetudine analogia ...; ibid. IX 63: ... analogiae fundamentum esse … naturam et usum; ibid. IX 70: in quibus usus aut natura non subsit, ibi non esse analogiam.

The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing  

are derived from approved common usage, and the poetry of Homer is that which is approved and most ancient; for no poem older than his poetry has come down to us. Thus we shall converse following the usage of Homer. (204) But, in the first place, it is not agreed by all that Homer is the most ancient poet; for some say that Hesiod preceded him in time, Linos, too, and Orpheus and Musaeus and a host of others. And, in fact, it is probable that there were some poets before Homer and in his time (since he himself says somewhere – Surely that song above all by men is most loudly applauded Which to their listening ears as the newest of songs resoundeth), and that these poets were eclipsed by his own brilliance. (205) And even if it should be agreed that Homer is the most ancient, what Pindarion has asserted is not convincing. For just as we questioned before whether one should adopt common usage or analogy, so, too, now we shall be in doubt whether to adopt common usage or analogy; and if usage, is it to be that of Homer or that of all other men; and as to this Pindarion has said nothing. (206) Further, we should follow that usage above all the adoption of which will not bring ridicule upon us; but if we follow Homer’s, our Greek speech will not escape ridicule, when we say μάρτυροι (for μάρτυρες) and σπάρτα λέλυνται (for σπάρτα λέλυται), and other things still more absurd. Neither, then, is this argument sound, besides the fact that our contention is conceded, namely, that analogy should not be used. (207) For what’s the odds whether we have recourse to the usage of the many or to that of Homer? For just as there is need of observation, but not of technical analogy, in the case of the usage of the many, so there is also in the case of that of Homer; for when we have observed how he is wont to speak, we ourselves too will converse in the same way. (208) To sum up, just as Homer himself made no use of analogy but followed the usage of the men of his age, so too will we by no means cleave to an analogous form which has the authority of Homer, but will conform our speech to the usage of the men of our own age.5

This passage from Sextus Empiricus, §§ 196–208, would appear to preserve a rather clear testimony of a stage in the polemic between analogists and anomalists (the champions of common usage) by offering a few exchanges in this debate, so to speak. The anomalists argue: analogy too is based on usage, which must therefore be regarded as the fundamental criterion underlying the whole system; so the rule of analogy is only a useless addition, and we might as well always follow συνήθεια. This argument, which traces the whole system back to usage, is quite stringent and difficult to refute in its pithiness: analogy is shown to lack any normative foundation apart from usage and is traced back to the latter; hence, its very raison d’être is called into question, all the more so because the analogists themselves acknowledge that usage is anomalous and changeable, and thus cannot provide stable criteria. The response is bound to start from this polemical argument: Ptolemaeus Pindarion sets out by accepting the self-evident claim that analogy is based on commonly accepted usage. How 5 Transl. Bury 1949.

  The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing ever, he counters that Homer’s poetry, on account of the universal appreciation it enjoys, represents a usage that is accepted by all and moreover is the most ancient; consequently, we may be following συνήθεια, but it is Homer’s συνήθεια, the one codified in his poems — the linguistic heritage they encapsulate and to which they lend authority. The usage taken as a starting point, then, is provided by a specific text, whose authoritativeness is commonly accepted and which, moreover, represents the first known attestation of the linguistic system (ποίημα γὰρ οὐδὲν πρεσβύτερον ἧκεν εἰς ἡμᾶς τῆς ἐκείνου ποιήσεως, 203). So although Ptolemaeus keeps within the coordinates laid out by the argument he is countering, through his response he reasserts the existence of a precise and well-established basis for the language, quite different from speakers’ fluctuating usage and not subject to the latter’s changeability. Through this linguistic foundation, it is possible to establish the rule necessary to the analogical τέχνη: an analogy βεβαιωτὴν ἔχουσα Ὅμηρον (208). The key point is identified as the need to define what συνήθεια is, i.e. what usage we are referring to: Ptolemaeus Pindarion’s response is played out precisely on this issue insofar as, while accepting the indisputable starting assumption of the anomalists’ reasoning, he de facto disproves their argument. The strategy in this reply is to accept the use of συνήθεια as unavoidable, but to replace the κοινὴ συνήθεια with the Ὁμηρικὴ συνήθεια — the changeable usage of the community of speakers with the codified usage of an undisputed authority, Homer. It seems as though in his argument Ptolemaeus Pindarion is drawing upon an idea developed by his teacher, Aristarchus. This emerges from what Apollonius Dyscolus tells us in De pronom. p. 71, 20 ff. Schn. (GG II 1): Aristarchus criticised the use of composite plural forms of the third-person singular reflexive pronoun ἑαυτούς and ἑαυτῶν, in his view unduly derived from the singular ἑαυτοῦ: for in the case of first- and second-person plural reflexive pronouns we do not have composite forms, but rather separate ones (ἡμῶν αὐτῶν, ὑμῶν αὐτῶν). To prove the soundness of his argument, Aristarchus invoked Homer — παρ’ ᾧ τὰ τοῦ Ἑλληνισμοῦ ἠκρίβωται — who always uses the non-composite forms σφᾶς αὐτούς and σφῶν αὐτῶν.6 Both the kind of reasoning (for the third-person pronoun an analogy with first- and second-person pronouns applies: see Homer’s usage) and the explicit definition of Homer as he παρ’ᾧ τὰ τοῦ Ἑλληνισμοῦ ἠκρίβωται show that the idea of taking the Ὁμηρικὴ συνήθεια, the usage wit 6 See Lehrs 1882, 353. Apollonius Dyscolus explicitly claims to share Aristarchus’ opinion (cf. 72, 4 ff.) and, in agreement with what I have been arguing here, concludes his argument with the following words: ἐξετάσαντι τῆς ἐν τοῖς μέρεσι τοῦ λόγου, ὡς ἡ Ὁμηρικὴ ποίησις μᾶλλον τῶν ἄλλων ἠνύσθη (72, 16).

The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing  

nessed by the Homeric poems, as the starting point and basis of Ἑλλη-νισμός — and thus as the foundation of the analogical ratio — derives from Aristarchus’ insights: Ptolemaeus Pindarion, one of his faithful pupils, evidently developed these insights and deployed them in the debate with the anomalists. It is worth dwelling on this point to advance a few considerations. In the Homeric scholia we frequently find references to the use of the Ὁμηρικὴ συνήθεια with regard to linguistic usage in its various aspects — from vocabulary to morphology and grammar, from inflection to syntax. In scholia by Didymus, Aristonichus, and Herodian we find expressions such as Ὁμηρικὴ συνήθεια, τὸ σύνηθες τῷ ποιητῇ, συνήθως τῷ ποιητῇ, συνήθως χρῆται Ὅμηρος, and other similar ones, in passages in which a comparison with other Homeric verses is used to support the philologist’s stance and his choice.7 In other words, a reference to Homer’s common usage, as witnessed by his poems, is used to support a method of internal criticism that in all likelihood can be traced back to Aristarchus and to the principle of explaining Homer through Homer that is no doubt crucial to and typical of his exegetical method.8 This principle was applied through a constant and close comparison between similar elements within Homeric poetry, which was evidently perceived to constitute a unitary and coherent system, this is: τὸ Ὁμηρικόν as a single whole that can be circumscribed and distinguished from what is set in contrast to it. Clearly, this is a principle of analogical criticism: even when it is applied with discernment, without excessive attempts to normalise the text by eliminating unique elements that instead

 7 See e.g. Herodian ad A 277 (Ἀρίσταρχος ἀποφαίνεται ὡς κατὰ τὴν Ὁμηρικὴν συνήθειαν ...), B 498, B 585, B 783, Δ 27, Ω 318; Aristonichus ad A 41, 291, B 135, Γ 220, Γ 459, M 159, X 110; Didymus ad Γ 18, P 134–6, etc.; see also P.Oxy. 1086, I 18 ff., ad B 763; sometimes the expression χρῆσις Ὁμηρική vel sim. is used in a similar sense: Herodian ad ad A 129; Didymus ad B 266, etc. (other testimonies could be found with the help of Baar 1961). See also Ludwich 1884–1885, II, 128. It is worth recalling that the term συνήθεια is also used in the scholia to mean ordinary Greek, i.e. κοινὴ συνήθεια (by contrast to Ὁμηρικὴ συνήθεια): see e.g. Herodian ad B 190, B 504, K 38. Moreover, συνήθεια is used to refer to the Vulgata in relation to the punctuation of the text in Nikanor’s scholia: see Wackernagel 1876, 433. 8 See Sandys 1920, 132; Heitsch 1960, 21; Erbse 1961, 225; and Montanari 1974, 119 ff. Pfeiffer 1968, 225 ff. has disputed the claim that the famous maxim Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν explicitly comes from Aristarchus. This sparked a debate, with the involvement of: Wilson, 1971, 172; Lee 1975 (dating it to the 5th cent. BC); Wilson, 1976 (a response to the previous article); Schäublin 1977 (according to whom the maxim certainly expresses Aristarchus’ exegetical method, but is not bound to have originated with him, and was at least inspired by rhetorical theories). Regardless of who was the first to formulate the famous axiom, the fact remains that it encapsulates a primary aspect of the Alexandrian philological-exegetical method — widely witnessed by the scholia — and in a way constitutes a theoretical formulation of it.

  The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing ought to be retained, it still entails the fundamental notion of a single, consistent system allowing us to form an idea of what is proper to and typical of Homer’s poetry. Out of all the various components of “the Homeric” (τὸ Ὁμηρικόν), linguistic usage (Ὁμηρικὴ συνήθεια) in its various aspects is certainly an element of primary importance. It is only natural, therefore, that it is the object of constant attention and careful examination. It is hardly surprising that Aristarchus staunchly opposed those who argued that Homer was only the author of the Iliad, and not of the Odyssey as well: for Aristarchus, this position, adopted by the so-called Chorizontes critics, basically meant that his Ὁμηρικόν was not one, but divided, thereby blurring its distinctive traits; it meant that his rich and broad comparisons, so fruitful on the philological and exegetical levels, only partially applied, and were fallacious whenever they crossed the limits of one of the two epic poems, thereby casting a shadow of doubt on these comparisons as a whole; finally, it partly undermined the foundations of Aristarchus’ overall vision of the most archaic period of Greek literature and of his “history of epic”, according to which the distinction between Homer as the author of the Iliad and Odyssey and the Cyclical poets also had a chronological, literary-historical, and historical-linguistic value. This polemic was inevitable, and it seems quite plausible that Aristarchus felt the matter to require a full commitment on his part. The Homeric scholia are replete with internal references to the Homeric poems and reflect an obstinate search for connections between the Iliad and the Odyssey in terms of content and thought, mores and customs, linguistic and stylistic features, and so on: all in all, an imposing collection of observations designed to outline and confirm the unitary quality of τὸ Ὁμηρικόν: an indivisible whole bound to stand as the counterpart of the Chorizontes’ thesis, which threatened to undermine the very foundations of Aristarchus’ literary-historical conceptions and methodological framework.9 The concept of Ὁμηρικὴ συνήθεια in Aristarchus’ analysis thus belongs to a critical method founded on the principle of internal analogy applied to Homeric poetry, a principle inextricably linked to the notion of the unity and consistency of the system within which this method is adopted. On the one hand, we have the idea of Ὁμηρικὴ συνήθεια, conceived of and interpreted as a unitary and coherent linguistic-grammatical system; on the other, we have the idea that

 9 The Chorizontes’ fragments have been collected by Kohl 1917; see also Susemihl 1892, II, 149 ff.; Kohl 1921; Schmid 1929, I 1, 83 ff., 89 ff.; Christ/Schmid/Stählin 1920, 268 ff.; Pfeiffer 1968, 213, 230 n. 7 + addenda p. 290, Rüter 1969; Lesky 1971; Fraser 1972, I, 467; Maffei 1976 (on which see N.J. Richardson in CR 28, 1978, 340 ff.); Montanari 1976a (= this volume, ch. 26), with further bibliography.

The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing  

Homer is the author παρ’ ὧ τὰ τοῦ Ἑλληνισμοῦ ἠκρίβωται. These are the two Aristarchean ideas underlying the argument attributed to Ptolemaeus Pindarion in his polemic about grammatical theory (although it is difficult to determine precisely which notions were developed by the teacher and which by his pupil). At this point we can return to the Sextus Empiricus passage from which we set out. The presentation of Ptolemaeus Pindarion’s opinion (202–203) is followed by the anomalists’ responses (204–208). First of all, they object that not everyone agrees that Homer is the earliest poet (204) — we shall come back to this shortly. Then the anomalists put forward a series of arguments (205–208) essentially designed to undermine Ptolemaeus Pindarion’s position by stating that he is not at all offering a solution to the problem, but simply shifting it onto another level; that Homeric usage leads to absurdities in everyday speech; and hence that analogy continues to be an added element which is unnecessary and must therefore be rejected.10 In conclusion, getting to the core of the matter, in Sextus’ text we read that any analogy based on Homer’s authority must resolutely be rejected, in order to adopt the anomalous common usage of contemporary speakers. The first objection (204) is rather significant for my purposes, and I shall dwell on it for a moment. Disputing the claim that Homer is the earliest poet, and indeed arguing  10 It has long been acknowledged that Sextus Empiricus, Adv. gramm. 176–235, and Varro, De ling. lat. VIII 26–43 (+ IX 40–48) can essentially be traced back to the same Greek source, an anomalist grammarian active before Varro (of course), but no earlier than the first decades of the 1st cent. BC. See Dahlmann 1932, 57–70, esp. 68; Dahlmann 1940, 4, 105 ff., 116 ff.; Mette 1952, 30 ff., esp. 33, 39, 44: whereas Dahlmann (who chiefly refers to Varro) is inclined to identify this source with Crates himself, Mette suggests it may be a pupil of Crates’ who wrote after the latter’s lifetime but faithfully borrowed his basic views, combining them with scepticalempirical methods of argument. Naturally, the information about Ptolemaeus Pindarion is missing in Varro, given its specific reference to Greek: scholars assume either that Sextus Empiricus added this information to source or that Varro removed it because it was not applicable to Latin; the latter hypothesis seems far more likely, given the close contextual connection to be found in §§ 196–208, examined above. Therefore, at least as far as Sextus Empiricus 196–208 is concerned, the direct source cannot be Crates, who for chronological reasons could not have known — and sought to respond to — the ideas of Ptolemaeus Pindarion and his followers (οἱ ἀπὸ Πινδαρίωνος, 202). If Ptolemaeus Pindarion was a pupil of Aristarchus’, who died around 144 BC and was a rough contemporary of Crates’ (according to a tradition we have no solid reasons to dispute: Suida, κ 2342 = Aeschylus of Miletus; see Pfeiffer 1968, 238), and if the source of Sextus Empiricus 196–208 is an anomalist grammarian who responded to Ptolemaeus Pindarion’s refutation of an anomalist attack on Aristarchean analogy, then the grammarian in question must be dated to the years between the late 2nd and early 1st cent. BC — precisely the period in which Ptolemaeus Pindarion and Varro were active (the two were probably one or at most two generations apart, since Varro was born in 116 BC).

  The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing that it is far more likely that there were already poets before him, means not only recalling a popular opinion and endorsing it, but making an effort to undermine one of the key elements of Ptolemaeus Pindarion’s argument. Indeed, the choice of Homer as συνήθεια, to be taken as a starting point and source of material, is not an arbitrary one, first of all because of the widespread acceptance his poetry enjoys, but also because his poems make up the most ancient texts whose authoritativeness is commonly accepted: their reputation as the “first” codified texts in Greek — texts that somehow mark the beginning and origin of the development of this language — is an essential element corroborating the right to draw upon Homer’s works as the foundational texts of Ἑλληνισμός. It has been objected that there were poets even before Homer: this argument is rejected, as Ptolemaeus Pindarion does not claim that Homer is the most ancient of all poets, but that the Homeric poems are the most ancient among those known to us, as no earlier poem has reached us. This objection immediately loses its validity in relation to Hesiod, whom the Alexandrians are known to have considered posterior to Homer;11 but it equally fails to apply to Linus, Orpheus, and Musaeus: the grammarian can hardly have been unaware that these poets were commonly regarded as earlier than Homer; but evidently none of the poems attributed to them and considered genuine was known to the Alexandrians. What Ptolemaeus Pindarion claims, therefore, is that the Iliad and Odyssey make up the oldest Greek literary texts known to us, the first codification of Ἑλληνισμός to have reached us: while there were certainly some poets before Homer (for he was commonly assumed to have had some predecessors), their works have not been preserved. Let us now examine the scholium to Dionysius Thrax, p. 490,7 ff. Hilg. (GG I 2). Φασί τινες ὅτι ἕως τῶν Τρωϊκῶν οὐκ ἐγινώσκετο γράμματα· καὶ δῆλον ἐκ τοῦ μὴ σωθῆναι ποίημά τι τῶν κατὰ τοὺς Ὁμηρικοὺς χρόνους, καὶ ταῦτα αὐτοῦ τοῦ Ὁμήρου εἰσάγοντος ποιητάς, τόν τε Φήμιον καὶ Δημόδοκον, ἱστορουμένου δὲ καὶ Ὀρφέως προγεγενῆσθαι καὶ Μουσαίου καὶ Λίνου· ἀλλ’ ὅμως πλὴν ὀνόματος οὐδὲν πλέον εἰς τὰ μετὰ ταῦτα διασωθῆναι συμβέβηκε πρὸ τῆς Ὁμήρου ποιήσεως, μηδὲ πρεσβύτερον τῆς Ἰλιάδος καὶ Ὀδυσσείας σώζεσθαι ποίημα. Ἀλλ’ ἐρεῖ τις, πῶς ἐπιγράμματα σώζεται νομιζόμενα πρεσβύτερα; Τούτων δὲ τὰ μὲν ἐψευσμένους τοὺς χρόνους, τὰ δὲ νεωτέρων τινῶν ὁμωνυμίᾳ τῶν παλαιῶν τὰς ἐπιγραφὰς ἔχουσι. Καὶ οἱ ἥρωες δὲ ἀγράμματοί τινες ἦσαν, καὶ σημείοις καὶ συμβόλοις πρὸς ἀλλήλους ἐν τῇ κατὰ τὸν βίον ἀναστροφῇ χρώμενοι ἐδήλουν ἀλλήλοις ἃ ἤθελον· τὸ γοῦν γράψας ἐν πίνακι πτυκτῷ διαζωγράφησίν τινα καὶ διατύπωσιν δηλοῖ ὧν ὁ γράφων ἐβούλετο ἐνδείξασθαι. Ἀλλ’ ἄτοπόν ἐστι τοὺς ἀρίστους τοῦ ἀνθρωπείου γένους ἀπαιδεύτους καὶ ἀγραμμάτους λέγειν. Ἀλλά φαμεν ὡς οὐκ ἔστιν ἀπαιδευσίας τεκμήριον τὸ ἀγράμματον· ὁ γοῦν θεῖος Πλάτων ...  11 See Jacoby 1902, 118; Jacoby 1904, 152; Severyns 1928, 39; Rzach 1912; Pfeiffer 1968, 164.

The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing  

Some say that up until the Trojan events the letters of the alphabet were unknown: this clearly emerges from the fact that no poem from the Homeric times has been preserved, despite the fact that Homer himself portrays poets such as Phemius and Demodocus, and that Orpheus, Musaeus, and Linus are said to have lived before him; and yet as far as what came before Homer’s poetry, only the names have been preserved and no poem older than the Iliad and the Odyssey has survived. But someone will say: why have epigrams been preserved that are deemed older? In the case of some of these the dating is wrong, while in the case of others their attribution to ancient authors is due to homonymy with more recent authors. Heroes were illiterate and in their everyday lives they would mutually convey their desires by using signs and conventional symbols: hence, the expression γράψας ἐν πίνακι πτυκτῷ (Z 169) describes a drawing and image of what “the writer” sought to convey. “But it is absurd to say that the best representatives of the human race were uncultured and illiterate.” We answer that being illiterate is not a sign of unculturedness: for example, the divine Plato …

Let us consider the salient points that emerge from this scholium. No poem from the age in which the Trojan events took place and, more generally, no poem earlier than the Iliad and the Odyssey, has been preserved, despite the fact that before Homer there lived poets such as Orpheus, Musaeus, and Linus, and that within the Trojan saga itself Homer mentions Phemius and Demodocus. This has occurred because the letters of the alphabet — i.e. writing — were unknown at the time; hence, poets’ works could not be preserved: the Homeric poems are the most ancient works to have been preserved, and what some people regard as anterior to them is actually posterior; in the famous verses Z 168–169, Homer is not actually referring to the use of alphabetic writing; the heroes were illiterate, which does not mean that they were ignorant. In this scholium, belonging to the range of material brought together in commenting § 6, Περὶ στοιχείου, of the Ars grammatica, traces of a polemic clearly emerge: the main opinion, which serves as a foundation, had evidently been subjected to some objections, which the scholiast records along with their refutations. This excerptum preserves the markers of a sound erudite source. The fact that the fundamental opinion upheld here coincides with what we read in Sextus Empiricus, as do some of the expressions used, suggests that we are dealing with material derived, probably indirectly, from Ptolemaeus Pindarion. In addition to Phemius and Demodocus — Homeric characters who are regarded as historic aoidoi — as poets anterior to Homer the text mentions Orpheus, Musaeus, and Linus, three traditional names also occurring in Sextus Empiricus: no mention is made of Hesiod, of course, who — as noted above — the Alexandrians considered to be posterior to Homer. Instead, the age before Homer — which is to say in the age he describes, that of the Trojan epic — there was no alphabet; it follows that the Homeric heroes were illiterate, so the famous σήματα λυγρά of the Bellerophon episode (Z 168 ff.) should not be interpreted as letters of the alphabet, but rather

  The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing as symbols and drawings, which is to say as an ideographic representation rather than as an alphabetic form of writing. This element leads us to the ancient interpretations of Z 168 ff., Z 176–178, H 175, and H 187, as witnessed by a group of scholia already connected to the above-quoted scholium to Dionysius Thrax, and which it is worth newly considering here by bringing together the range of testimonies found in the scholia to the Iliad. We essentially find two interpretations: the first occurs in the following scholia by Aristonichus, which naturally reflect Aristarchus’ position: Sch. ad

Z 169:

Sch. ad Sch. ad Sch. ad

Z 178: H 175: H 187:

γράψας ἐν πίνακι πτυκτῷ: ὅτι ἔμφασίς ἐστι τοῦ τῆς λέξεως γράμμασι χρῆσθαι. οὐ δεῖ δὲ τοῦτο δέξασθαι, ἀλλ’ ἔστι γράψαι τὸ ξέσαι· οἷον οὖν ἐγχαράξας εἴδωλα, δι’ ὧν ἔδει γνῶναι τὸν πενθερὸν τοῦ Προίτου. σῆμα κακόν: ὅτι σημεῖα λέγει, οὐ γράμματα· εἴδωλα ἄρα ἐνέγραψεν. ὅτι σημείοις χρῶνται, οὐ γράμμασιν. ὅς μιν ἐπιγράψας: ὅτι οὐ γράμμασι τῆς λέξεως, ἀλλ’ ἐγχαράξας σημεῖα· εἰ γὰρ κοινῶς ᾔδεσαν γράμματα, ἔδει τὸν κήρυκα ἀναγνῶναι καὶ τοὺς ἄλλους, οἷς ἐπεδείκνυτο ὁ κλῆρος.

Exactly the opposite opinion is expressed in a group of exegetical scholia to the same Iliad passages: Sch. ad

Sch. ad Sch. ad Sch. ad

Z 168:

σήματα λυγρά: γράμματα· «οἱ δὲ κλῆρον ἐσημήνατο» (Η 175), «ὅς μιν ἐπιγράψας» (Η 187)· ἄτοπον γὰρ τοὺς πᾶσαν τέχνην εὑρόντας οὐκ εἰδέναι γράμματα. Z 168–9: σήματα μὲν τὰ γράμματα, πίνακα δὲ τὸ λεγόμενον πινακίδιον. Z 176: σῆμα: σῆμα τὸ ἐπίταγμα, «σημαίνων ἐπέτελλεν» (Φ 445)· ἢ τὸ πτυκτὸν τὸ ἔχον τὰ σήματα, ὅ ἐστι τὰ γράμματα. H 175: ἐσημήναντο: γράμμασιν. καὶ πῶς οὐ γινώσκει ὁ κῆρυξ; ἐθνικὰ γὰρ ἦν (cf. Aristonichus ad H 187; sch. ex. ad H 185).

The connection between these testimonies and the above-quoted scholium to Dionysius Thrax is evident, as is especially the agreement between the latter’s opinion about σήματα and Aristarchus’ (transmitted by Aristonichus’ excerpta). Moreover, we find a significant correspondence between sch. ex. ad Z 168, … ἂτοπον γὰρ τοὺς πᾶσαν τέχνην εὑρόντας οὐκ εἰδέναι γράμματα, and the sentence in the scholium to Dionysius Thrax, ἄλλ ἄτοπόν ἐστι τοὺς ἀρίστους τοῦ ἀνθρωπείου γένους ἀπαιδεύτους καὶ ἀγραμμάτους λέγειν: the second objection recorded and refuted by the scholium to Dionysius Thrax agrees with the Iliad sch. ex., which confirms that the scholium to Dionysius Thrax provides reliable information about the philological polemics that emerged concerning these

The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing  

issues, while supporting the hypothesis that this material can be traced back to the Aristarchean Ptolemaeus Pindarion. The debate on the interpretation of the Iliad’s σήματα, connected to the idea that the Homeric poems are the most ancient works to have been preserved, is also clearly mentioned by Flavius Josephus in C. Apion. I 11–12, a passage that is usually compared with these scholia: (11) … περὶ τῶν ἐπὶ Τροίαν τοσούτοις ἔτεσι στρατευσάντων ὕστερον πολλὴ γέγονεν ἀπορία τε καὶ ζήτησις, εἰ γράμμασιν ἐχρῶντο, καὶ τἀληθὲς ἐπικρατεῖ μᾶλλον περὶ τοῦ τὴν νῦν οὖσαν τῶν γραμμάτων (12) χρῆσιν ἐκείνους ἀγνοεῖν. ὅλως δὲ παρὰ τοῖς Ἕλλησιν οὐδὲν ὁμολογούμενον εὑρίσκεται γράμμα τῆς Ὁμήρου ποιήσεως πρεσβύτερον, οὗτος δὲ καὶ τῶν Τρωϊκῶν ὕστερος φαίνεται γενόμενος, καί φασιν οὐδὲ τοῦτον ἐν γράμμασι τὴν αὑτοῦ ποίησιν καταλιπεῖν, ἀλλὰ διαμνημονευομένην ἐκ τῶν ᾀσμάτων ὕστερον συντεθῆναι καὶ διὰ τοῦτο πολλὰς ἐν αὐτῇ σχεῖν τὰς διαφωνίας. (11) … it is a highly controversial and disputed question whether even those who took part in the Trojan campaign so many years later made use of letters, and the true and prevalent view is rather that they were ignorant of the present-day mode of writing. Throughout the whole range of Greek literature no undisputed work is found more ancient than the poetry of Homer. His date, however, is clearly later than the Trojan war; and even he, they say, did not leave his poems in writing. At first transmitted by memory, the scattered songs were not united until later; to which circumstance the numerous inconsistencies of the work are attributable.12

This famous passage takes up the question of the Homeric heroes’ knowledge of writing in the poems, which is connected to the opinion according to which the Iliad and the Odyssey are the most ancient works of Greek literature to have survived, since previously writing was unknown: this is precisely the connection found in the above-quoted scholium to Dionysius Thrax.13 Besides, the πολλὴ

 12 Josephus. The Life. Against Apion. Translated by H. St. J. Thackeray. Loeb Classical Library 186. Cambridge, MA: Harvard University Press, 1926. 13 On this range of testimonies, see (among others) Schmidt 1920, 56 ff.; Jeffery 1967; Maffei 1976, 29 ff.; Heubeck 1979, 1266 ff. The bibliography on the issue of writing in the Homeric poems is vast, of course: here I will resist the temptation to touch upon modern discussions of issues ranging from Homer’s familiarity with writing and the written recording of the Homeric poems to Peisistratos’ redaction and question of whether this was known to the Alexandrians or not (see Merkelbach 1969, 239 ff.). This discussion has often found a starting point in the above-mentioned sources, ever since, precisely on the basis of the Flavius Josephus passage and of Aelianus, V.H. 13, 14, Perizonius, in his Animadversiones historicae (Amsterdam 1684, ch. VI in.), spoke of the Homeric poems as the earliest literary text to have been transmitted: a text initially committed to memory and then written down and rearranged in Athens in Peisistratos’ day. Perizonius did not know the scholia of Ven. A, but these were certainly familiar to

  The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing ἀπορία τε καὶ ζήτησις of which Flavius Josephus speaks has left evident traces in the Iliad scholia we have examined, according to a clear divergence of opinion: whereas Aristarchus did not believe that any trace of alphabetic writing was to be found in Homer, the corpus of exegetical scholia bears witness to the opinion that writing already existed in the age described by Homer and hence long before his own lifetime — which is why the poet could know writing very well. The latter position falls within popular traditions which assigned the invention of the alphabet to Cadmus and the Phoenicians, to the Egyptians, to the Pelasgians (the Greeks’ ancestors), to mythical figures such as Palamedes, Prometheus, and so on.14 I shall not dwell on all these well-known traditions, but only recall that the idea of an ancient origin of the alphabet and of writing, stretching back to a time long before Homer, was by far the most common and widespread. This idea inspired quite a few traditions about texts predating Homer and authors whom the latter had reportedly drawn upon.15 Finally, it is worth recalling another element. In the exegetical scholia we often find the idea of “Homer as the source of all knowledge”:16 it seems quite natural that this representation of the poet as someone from whom we can learn a bit of everything would include knowledge of something as important as the technique of writing. Indeed, in sch. ex. ad Z 168 (whose connection with the scholium to Dionysius Thrax was noted above), the Homeric heroes are described as those πᾶσαν τέχνην εὑρόντας, which agrees with the idea that Homer bears witness to all forms of knowledge. The response to the argument made in this scholium, namely that illiteracy does not at all undermine the image of the Homeric heroes as highly worthy figures, only occurs in the scholium to Dionysius Thrax. However, as we have seen, the idea that the origins of writing stretched back to remote

 F.A. Wolf (cf. Erbse 1979), who a century later, in 1795, published his famous Prolegomena. In this work he argued — partly drawing upon the above-mentioned sources — that the Homeric poems were originally composed orally by illiterate aiodoi and only subsequently committed to writing. The debate extends into modern times and, from Milman Parry onwards, entails the vast issue of the oral origin of the Homeric poems, naturally also touching upon the question of literacy. This is an extensive debate which, as already noted, is only partially relevant to the scope of the present enquiry. 14 See e.g. Jeffery 1967; Pfohl 1968, 21. 15 In addition to the well-known tales about Dictys Cretensis and Dares Phrygius, see (among others) Suda, s.v. Palamedes and s.v. Korinnos; Sch. Dion. Thr. 185, 8 ff. Hilg.; Aelianus, V.H. 14, 21; Ptolemaeus Chennus apud Phot., Biblioth. 190, III, 66 Henry, etc. As mentioned above, the Greeks generally believed that Homer had had some predecessors: see Allen 1924, 130 ff. 16 See van der Valk 1963–1964, I, 465 ff.

The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing  

times or even mythical characters — albeit according to different traditions — was by far the most common and widespread among the Greeks. By contrast to these semi-mythical traditions, what has been described as “a sceptical view of such extravagances”17 is witnessed by Aristonicus’ scholia, the Josephus passage, and the scholia to Dion. Thr. p. 490,7 ff. Hilg. examined in the previous pages. In all likelihood, this position can be traced back to Aristarchus and his school. There is a fair chance that the issue of the alphabet and its invention were discussed, for instance, by one of the very first pupils of the great Alexandrian teacher, namely Apollodorus of Athens, in a section of his lengthy commentary on the Catalogue of Ships.18 According to Aristarchus and his followers (which leads us back to Ptolemaeus Pindarion, who was also a direct pupil of Aristarchus), the invention of the alphabet and the use of writing are later than the world depicted by Homer, and the passages Iliad Z 168 ff. and H 175 ff. do not at all bear witness to the use of writing. In all likelihood, the history of written Greek poetry begins with Homer, and the Iliad and the Odyssey were the first poems to be committed to writing. This explains why they were preserved, whereas none of the previous poems has survived (the Josephus passage further records the opinion according to which the poems were first transmitted orally and only written down after Homer’s lifetime). Another element might be brought into play here. In the scholia to Dionysius Thrax p. 192, 8 ff. Hilg., Ptolemaeus Pindarion (the name stems from a slight correction, which seems certain: see appar. ad loc.) is credited with the peculiar theory according to which the alphabet was invented by a certain Stoichos (evidently a speaking name), a native-born Athenian. This idea of the Athenian origins of the alphabet is bound to recall that of the Athenian origins of Homer, which would appear to have been endorsed by Aristarchus and Dionysius Thrax, according to two Lives of Homer, p. 244, 13 and p. 247, 8 Allen.19 Therefore, there may be some connection between the Athenian origins of Homer, the theory of the Athenian origins of the alphabet endorsed by Ptolemaeus Pindarion, and the idea that the Homeric poems were the first to have been committed to writing and hence the first to have been preserved. This element is far from certain; however, as an hypothesis it is perfectly consistent with the context reconstructed above.

 17 Jeffery 1967, 161. 18 See Jeffery 1967, 158, on the basis of Apollodorus, FgrHist 244 F 165 = Sch. Dion. Thr. 183, 5 ff. Hilg. 19 See Pfeiffer 1968, 228, 267.

  The Grammarian Ptolemaeus Pindarion, the Homeric Poems, and Writing Within the range of opinions of Aristarchus and his pupils just outlined, the history of the alphabet and of the technology of writing appears to be closely intertwined with the earliest history of Greek literature: evidently, tracing the history of the primae origines of Greek literature entailed the study of the methods that had enabled it to emerge from obscurity and to be preserved and transmitted into later centuries. Ptolemaeus Pindarion must have taken up this debate and contributed to it by closely connecting the question of the theory of language and grammar with the beginnings of literary history and the invention of alphabetic writing, thereby creating a new, broader synthesis. To conclude by returning to the points made at the beginning of this essay, we might say that Ptolemaeus Pindarion’s overall argument can be reconstructed as follows: since there was no writing before Homer, the Homeric poems are the earliest example of Greek literature to have reached us and the first known codification of the Greek language; they enjoy the widest acceptance and undisputed authoritativeness; therefore, the linguistic usages they attest to, the Ὁμηρικὴ συνήθεια, may be taken as the starting point and point of reference for the analogical system of rules governing correct Greek, Ἑλληνισμός. As we have seen, this view was formulated as a response to a polemical attack by the anomalists on Aristarchean grammatical theories — an attack intended to show the uselessness and unreliability of the analogical system of rules. In the light of this, within the field of grammar theory, Ptolemaeus Pindarion emerges not as the representative of a conciliatory theory,20 but rather as someone who defended his teacher Aristarchus’ views by somehow reformulating the theoretical foundations of the analogical procedure, yet always on the basis of Aristarchean principles.

 20 As was instead suggested by Christ-Schmid-Stählin 1920, 268.

 Ρ.Οxy. 574 verso riconsiderato: frammento di hypotheseis dell’Iliade Il P.Oxy. 574 è un piccolo frammento di rotolo di cm. 13 x 3,8, che fu reso noto piuttosto sommariamente nella sezione “Descriptions” del vol. III degli Oxyrh. Pap. (London 1903, p. 279). Il recto presenta resti di una lista di pagamenti di tasse ed altro. Invece sul verso (Pack2 1193), assegnato al II sec., si trova, secondo l’ed. pr., “an extract from a narrative of the preservation of Eurypylos by Patroclos (cf. Il. xi. 575 sqq.)”. Si tratta di cinque righi di scrittura non formale: mancano del tutto spiriti e accenti; lo iota muto è omesso; per il numero degli errori grafici in piccola estensione, il testo può dirsi non accurato: una correzione interlineare della stessa mano compare al r. 2. La striscia di spazio bianco che costituisce il margine superiore mostra che sono conservati i primi righi di una colonna. Vi si legge quanto segue: 1 3 5

margine Πάτροϰλον μαϑηϲόμενον τίϲ εἴη· ὁ δὲ ἐπιγνoὺϲ ϰαὶ ἀπολυόμενοϲ τῆϲ παρατάξεωϲ, Εὐρυπύλῳ ϲυναντᾷ, ὃν ϰαὶ ὣϲ ἐϰ τῆϲ πληγῆϲ ἔχοντα ἀπαγαγὼν εἰϲ τὴν ἐϰείνου ϲϰηνὴν ἰᾶται. τῆϲ Μ¯ –––––

1. nella col. prec.: cfr. infra. 1-2. Dapprima aveva scritto επι|ϰνωϲ, poi la stessa mano è intervenuta a correggere il ϰ con un γ posto nell’interlineo, ripristinando in parte la grafia esatta. 3. La lettura ΚΑΙΩϹ è sicura. Merkelbach preferisce la congettura ϰαϰῶϲ. 3. εχοντοϲ pap. 4. ειαται pap.

Pur nella sua esiguità, il nostro testo può forse essere caratterizzato con maggiore precisione. Il riferimento al canto XI dell’Iliade non può ovviamente essere messo in dubbio, ma confronti illuminanti si trovano precisamente con le hypotheseis di tale canto tramandate nei codici omerici. Come è noto, per ogni canto iliadico sono normalmente tramandate due hypotheseis, sensibilmente differenti per quanto riguarda lo stile e l’estensione: una I più breve e concisa, una II più lunga e diffusa e discorsiva.1 Il canto XI dell’Iliade presenta una situazione abbastanza abnorme per quanto riguarda la I hypoth.; di essa si trovano  1 Cfr. Henrichs 1973, 23 sgg. https://doi.org/10.1515/9783110772371-031

  P.Oxy. 574 verso riconsiderato infatti nei codici ben tre redazioni, che coincidono fino ad un certo punto e poi divergono notevolmente nella seconda parte (una di esse risulta da una chiaramente consapevole fusione della I con la II). E questo oltre al fatto, da non trascurare nel sottolineare la particolarità del caso, che essa sfugge in maniera sensibile a quella tendenza alla brevità e concisione che è propria della I hypoth. Riportiamo i testi per le parti che ci servono qui.2 Ia: ... Ἀχιλλεὺϲ διαπέμπεται Πάτροϰλον †δεηϑῆναι βοήϑειαν αὐτοῖϲ ἐπιτρέψαι. ἐντεῦϑεν ὁ Πάτροϰλοϲ ϲυναντᾷ Εὐρυπύλῳ τετρωμένῳ ϰαὶ ἰᾶται αὐτόν. Ib: ... Ἀχιλλεὺϲ διαπέμπεται Πάτροϰλον μαϑηϲόμενον τίϲ εἴη ὁ τετρωμένοϲ. ὁ δὲ ἀφιϰόμενοϲ ...... ἐπανερχόμενοϲ δὲ, περιτυχὼν Εὐρυπύλῳ, ϰομίϲαϲ εἰϲ τὰϲ ναῦϲ ϑεραπεύει. Ic: ... Ἀχιλλεὺϲ διαπέμπεται Πάτροκλον μαϑηϲόμενον τίϲ εἴη ὁ τετρωμένοϲ. ἀφικομένῳ δὲ αὐτῷ - - - - - - ϑεραπεύει ( = II hypoth.) II: ... Ἀχιλλεύϲ ... πέμπει Πάτροϰλον μαϑηϲόμενον τίϲ εἴη ὁ τετρωμένοϲ. ἀφιϰομένῳ δὲ αὐτῷ ...... ἐπανερχόμενοϲ δὲ ὁ Πάτροκλοϲ πρὸϲ Ἀχιλλέα, περιτυχὼν Εὐρυπύλῳ τετρωμένῳ βαστάζει τε αὐτὸν ϰαὶ ἐπὶ τὰϲ ναῦϲ ϰομίϲαϲ ϑεραπεύει. Per tutta la parte precedente a quanto abbiamo riportato, le tre redazioni della hypoth. I coincidono perfettamente: da un certo punto in poi, come si vede, divergono in modo notevole. L’hypoth. Ia, contenuta nei mss. degli Sch. D iliadici3 (e in altri), non è stata finora edita e se ne da notizia qui per la prima volta;4 l’hypoth. Ib si trova in [Eustazio],5 Barnes,6 Bekker;7 l’hypoth. Ic è stata rinvenuta per ora solo nel ms. Vat. gr. 30 (sec. XIV, cont. Iliade con scoli e hypoth.): tutta la seconda parte coincide esattamente con quella corrispondente

 2 Per il testo di tutte le hypotheseis di Iliad. XI mi servo di Vitarelli 1981, un lavoro dal quale la conoscenza del testo delle hypotheseis di Iliade XI risulta, come appare evidente, molto arricchito. 3 Per la tradizione di questi scoli cfr. ora Montanari 1979, 4 sgg. 4 Notizia tratta dal lavoro di M.C. Vitarelli cit. alla n. 2. Nell’ed. pr. degli Sch. D iliadici, dovuta a J. Lascaris, Roma 1517, è riportata solo l’hypoth. II (cfr. infra). 5 Il nome di Eustazio è messo tra parentesi quadre perchè, com’è noto, le hypotheseis dei canti omerici non appartengono al testo originale dell’autore, ma si trovano nelle edizioni a stampa che ripetono l’ed. pr. di N. Maiorano, Roma 1542–50: cfr. van der Valk 1971–1987, Praefatio p. CXLI (nella nuova edizione, basata sull’autografo eustaziano, le hypotheseis naturalmente non compaiono). 6 Barnes 1711. 7 Bekker 1827.

P.Oxy. 574 verso riconsiderato  

dell’hypoth. II,8 che si trova nei mss. degli Sch. D iliadici (e in altri) e in Lascaris,9 Barnes, Bekker. Le corrispondenze che possiamo facilmente rilevare indicano chiaramente che in P.Oxy. 574 abbiamo conservata la parte finale di una hypothesis del canto XI dell’Iliade. Una conferma decisiva sul carattere del frammento viene da quanto si legge al r. 5: l’indicazione τῆϲ M si riferiva evidentemente all’hypothesis del canto successivo,10 secondo una maniera ben documentata, ad esempio nel P.Mich. inv. 1315, che è il più ampio e significativo brano di una raccolta di hypotheseis iliadiche conservatoci su papiro.11 È difficile dubitare, a questo punto, che P.Oxy. 574 conservi un magro resto di una raccolta di hypotheseis dell’Iliade. E’ naturalmente impossibile dire di quale estensione fosse originariamente la raccolta di cui è rimasta solo la parte finale del canto XI seguita dal titolo del canto XII. Un piccolo problema può essere rappresentato dal fatto che, mentre fra il r. 4 e il r. 5 l’interlineo è uguale ai precedenti, dopo il titolo del r. 5 sono conservati 8 mm di spazio che appare senz’altro bianco: si potrebbe anche sospettare che la copiatura della raccolta sia stata interrotta per una qualsiasi ragione e che l’hypothesis del canto XII poi non sia stata effettivamente trascritta, ma in effetti nulla vieta che il testo continuasse anche dopo un interlineo un po’ più ampio. L’hypothesis del canto XI occupava quindi almeno una parte anche della colonna precedente, la quale doveva terminare con qualcosa come Ἀχιλλεὺϲ πέμπει oppure διαπέμπεται (cfr. il testo delle hypotheseis riportate sopra), che doveva costituire la reggente dell’acc. del r. 1 della colonna conservata. P.Oxy. 574 va dunque ad aggiungersi agli altri frammenti papiracei che restutuiscono parti di raccolte di hypotheseis omeriche. È infatti assodato con certezza, grazie alla continuità rilevabile fra la fine di una hypothesis ed il titolo della successiva, che si trattava precisamente di una raccolta di hypotheseis da sole e non di hypotheseis seguite dagli Scholia minora (come accade per esempio in P.Oxy. 3159 e in P.Oxy. 3160 + P.Strasb. gr. 1401). Oltre al già citato P.Mich. inv. 1315 (Iliad. K-P), frammenti di raccolte di hypotheseis omeriche si  8 Risulta dunque dalla fusione della prima parte dell’hypoth. I con la seconda parte dell’ hypoth. II: forse chi ha creato questa redazione aveva sottomano l’hypoth. la ed ha rilevato la corruttela, per cui ha deciso di cambiare esemplare, producendo una contaminazione: cfr. M.C. Vitarelli, op. cit. sopra alla n. 2. 9 Cfr. sopra n. 4. 10 Nell’ed. pr. del papiro per il r. 5 si dava l’errata lettura τῆϲ η: un controllo eseguito su una buona fotografia avuta dalla Bodleian Library di Oxford non lascia dubbi sulla correttezza della lettura τῆϲ M. 11 Edito da Henrichs 1973, 23 sgg.

  P.Oxy. 574 verso riconsiderato trovano ancora in P.Antin. 69 (Iliad. Χ–Ψ e Odyss. β–γ); P.Ryl. 23 (Odyss. γ e ζ); P.Laur. 53 (Od. ε–ζ). Come risulta da questo elenco, l’unico altro testimone papiraceo che contiene le hypotheseis agli stessi canti iliadici è il P.Mich. inv. 1315: in esso dell’hypothesis del canto XI sono conservati solo pochi resti, sufficienti tuttavia per dire che si trattava di una redazione del tutto differente da quella del nostro frammento. Le hypotheseis omeriche restituite dai papiri sono di solito, per estensione e per stile, più vicine alle I hypoth. dei codici, più brevi e stringate: non di rado anzi quelle papiracee risultano, rispetto a quelle, ancora più concise e stringate, ridotte veramente all’essenziale del contenuto del canto.12 L’hypothesis di Iliad. XI di P.Oxy. 574 non sfuggiva certo a questa tendenza: per quanto ci è dato vedere, per esempio, siamo certi che in essa mancava completamente un resoconto dell’incontro fra Patroclo e Nestore, sul quale si soffermano le hypotheseis dei codici. Dal punto di vista della storia della tradizione e del rapporto fra le hypotheseis restituite dai papiri e quelle dei codici, è interessante mettere in luce le particolari coincidenze che legano P.Oxy. 574 con l’hypoth. Ia. Con quest’ultima, il frammento papiraceo condivide le lezioni particolari ϲυναντᾷ (r. 3) e ἰᾶται13 (r.4), alle quali si oppongono rispettivamente περιτυχών e ϑεραπεύει delle altre hypotheseis. Purtroppo, il testo dell’hypoth. Ia contiene certamente una corruttela: questo particolare legame sembra confortare l’opinione che nell’ hypoth. Ia ci sia una lacuna, nella quale possono essere cadute almeno le parole μαϑηϲόμενον τίϲ εἴη (ὁ τετρωμένοϲ), più qualcos’altro che spieghi quanto segue.14 Un’ultima osservazione riguarda la frase ἀπαγαγὼν εἰϲ τὴν ἐϰείνου ϲϰηνήν del r. 4. Sembra di poter sorprendere nell’hypothesis un desiderio di chiarire e rendere esplicito con ἐϰείνου che la tenda di cui si parla è quella di Euripilo e non quella di Patroclo. La precisazione trova il suo senso nel fatto che il testo di Iliad. XI 842 sgg. non è esplicito e può lasciare un’ambiguità, che si risolve soltanto grazie ad un confronto con XV 392, dove invece si dice a chiare lettere che Patroclo si trova dentro la tenda di Euripilo.15

 12 Cfr. Henrichs 1973. 13 Che dovrebbe essere preso direttamente dal testo omerico, cfr. Iliad. XII 2. 14 Che la corruttela testuale dell’hypoth. Ia nasconda una lacuna è un convincente risultato del lavoro di M.C. Vitarelli, cit. sopra alla n. 2. 15 Di intento “interpretativo” nelle hypotheseis omeriche (che parrebbero a prima vista una pura esposizione del contenuto dei canti senza nessun altro elemento) ho parlato in Montanari 1981, 273 sgg. (= questo volume, ch. 29).

 The Homerica on Papyrus: Notes Toward a Classification of Genres It is a well-known fact that Homer is, by far, the most abundantly attested author in extant papyrus fragments. In Pack2 (published in 1965), we find 681 Homeric papyri, whereas the next most well-represented authors, Euripides and Demosthenes, follow with about 80 items each. In the Spring of 1983, during a seminar held at the University of Pisa, Paul Mertens (who is currently working on a third edition of Pack’s inventory) was able to count no less than 1004 Homeric papyri, as opposed to 111 for Euripides and 105 for Demosthenes. The difference, then, has grown from around 600 to around 900 documents — a truly extraordinary amount, which handily separates Homer from the next most commonly attested authors, and which forms a concrete and unequivocable proof of the Homeric poems’ extraordinary reception and of their deep penetration into school teaching and into culture in all its forms, even into the very fabric of society, in a manner that has no parallels elsewhere in all of the ancient world. Of this massive amount of Homeric papyri, in addition to the fragments that contain parts of the poems themselves, a considerable number of documents belong to the category of Homerica: exegetical texts dedicated to interpretation of various kinds and at various levels; commentaries; aides for reading and aides produced while reading; scholarship of different sorts on Homer or related to Homer. These texts form a kind of vast and variegated halo, surrounding the Homeric poems and more or less directly deriving from them; in quantity they almost constitute a “literature” all of their own. Of the 681 Homeric papyri present in Pack2, 605 are fragments of the poetry, whereas 76 are Homerica; and if we consider once again the numbers made available by Martens, we find that of the 1004 items currently inventoried, 864 contain poetry and 140 are Homerica. In other words, the fragments of poetry, in passing from 605 to 864 items, have increased by 42%, whereas the Homerica, passing from 76 to 140, have increased by 84% — an increase twice as large as that seen for the poetry, in terms of percentages. And it is precisely on the panorama of Homerica that I would like to focus my attention here. For the panorama of Homerica represents a problem, insofar as it is extremely broad, composite, and heterogeneous; its boundaries and its component parts have not yet been traced or clarified overall,

https://doi.org/10.1515/9783110772371-032

  The Homerica on Papyrus and as a result one feels the need for ever greater effort to be made in introducing organizational and classificatory criteria.1 As a starting point, we may look to the papyrus texts that H. Erbse included in his monumental edition of the Scholia maiora to the Iliad. Among these texts we find philological and exegetical material of various kinds, from the explanation of realia to the analysis of strictly literary aspects of the poetry, from the examination of textual variants and other text-critical problems (such as punctuation) to the discussion of difficult or controversial points of interpretation, not to mention the treatment of linguistic and grammatical issues. These texts bear the mark (more or less faded, depending on the type of text and its distance from the original) of the great philological work of the Hellenistic and Roman periods, and they are the texts that we may term hypomnemata or commentaries.2 Representative examples of this genre include the following fragments: (for the Iliad) P. Oxy. 1086, P. Oxy. 1087, P. Oxy. 221, P. Oxy. 2397 (all included in Erbse), P. Trieste (ed. S. Daris in Studi Stella, Trieste 1975, 463ff), P.W.U. inv. 217 (ed. Z.M. Packman, BASP 10, 1973, 53 ff.); (for the Odyssey) P. Flor. (ed. V. Bartoletti, ASNP 35, 1966, 1 ff.), P. Yale inv. 551 (ed. G.M. Parassoglou, Hellenika 28, 1975, 60 ff.; cf. W. Luppe, Würzb. Jahrb. N. F. 2, 1976, 99 ff.), P. Alex. 198 (cf. A. Carlini in Papiri lett. greci, Pisa 1978, 89 ff.), P. Fayum 312 (= Pack2 1213: unpublished, currently being studied). Texts of this genre consist of a series of lemmata (marked in various ways) deriving from the text of Homer, which are followed by philological and/or exegetical comments (of the kind indicated above in broad outline); not infrequently one finds references to the famous critical signs (σημεῖα) of ancient philology. As far as both their structure and their content are concerned, these hypomnemata or commentaries have a privileged relationship with a specific type of scholia to Homer found in the medieval tradition, namely the scholiastic material that derives from the VMK, the socalled Four-man Commentary, as well as from the Scholia exegetica (cf. the edition of Erbse mentioned above).3  1 Important contributions in this direction include, above all, Henrichs 1971, 100–102; Renner 1979, 311 ff. 2 My discussion here is limited to Homeric philology; that said, for general discussions of the problems regarding the structure and the definition of hypomnemata, cf. Turner 1980, 112 ff.; Arrighetti 1977a, esp. 49 ff.; Arrighetti 1977b; del Fabbro 1979 (in this last item, however, at least as far as Homer is concerned, genuine hypomnemata are lumped together with paraphrases, Scholia minora, and collections of ἱστορίαι [on which categories, see further below], which renders the definitions imprecise and the discussion not entirely clear). 3 The process by which the most ancient hypomnemata were transformed into the scholia transmitted in the margins of medieval manuscripts was long and complex, and it saw various

The Homerica on Papyrus  

Another category, easily recognizable, is represented by texts which we may call alphabetic lexica. In this case, there is a fundamental difficulty, which resides in the need to keep the specifically Homeric lexica distinct from more wide-ranging lexica which, nonetheless, contain a great deal of Homeric material. (It should not be surprising that even in a general lexicon the Homeric words tend to represent the most numerous lemmata.) Obviously, there is an element of chance inherent in the discovery of small fragments containing a limited number of lemmata, and this will inevitably leave a certain shadow of doubt around these items, which it is very hard to clarify definitively. Nonetheless, I am not at all convinced that we should include among the Homerica a lexicon like P. Heidelb. Siegmann 200, which also contains non-Homeric lemmata; Pack2 lists it among the Homerica as n. 1220 with the description “Glosses, partly Homeric” (and in this he is in agreement with Mette, Rev. Phil. 29, 1955, pap. Z2), but in my opinion it seems better to maintain that, in cases like this one, we are dealing not with a specialized Homeric lexicon (a lexicon dedicated to one author), but rather with more an extensive lexicon in which the Homeric lemmata simply happen to be in the majority. (Similar observations can be made regarding P. Mich. inv. 9 = Pack2 1228, which A. Henrichs, ZPE 13, 1973, 30 has justly described using the more hesitant term “Episches Glossar”.) Among the Homeric alphabetic lexica, the most emblematic items are those found on the papyrus texts attributed to Apollonius Sophista, which are now extant in a considerable quantity and are noteworthy for their relationship with the medieval tradition of that author’s Homer Lexicon.4 Apart from this, scholars have already noted the connection (in the form of shared material) between the glosses in the Homeric alphabetic lexica and those present in the collections of Scholia minora (see below), not to mention the connections that both of these categories of texts have with the D-Scholia and the Lexeis Homerikai.5 Let us focus our attention for a moment on the latter text, an alphabetic Homeric lexicon transmitted in medieval manuscripts and extant in its entirety; it is entitled Λέξεις Ὁμηρικαὶ (κατὰ στοιχεῖον),6 and it presents glossographic material in

 phases, above all involving epitomization and fusion of materials, which began very early. On these problems, cf. Wilson 1967, 244 ff.; Wilson 1984; Arrighetti, works cited above; Montanari 1979b (= this volume, ch. 28). 4 Cf. Henrichs/Müller 1976; Renner 1979; Haslam 1982, 33 ff.; and in general, cf. Naoumides 1969. 5 Cf. Calderini 1921, esp. 324–326; Henrichs 1971, 97 ff., on Apollonius Sophista, 110ff. 6 The edition begun by De Marco 1946, came to a halt at the letter ε and was never brought to completion. Moreover, in addition to the two manuscripts used by De Marco, Urbin. gr. 157 (U)

  The Homerica on Papyrus alphabetical order deriving from Homer that, in the tradition in which it has arrived to us, partly coincides with, but partly diverges from the corpus of the D-Scholia. This characteristic has led at least one scholar to define it as “quasi alteram recensionem”7 of the glossographic component of the D-Scholia. Now, it is clear that the Homeric alphabetic lexica attested on papyrus constitute, in their form, ancient antecedents to this type of Homeric alphabetic lexicon preserved in medieval manuscripts, in the sense that, horizontally, the papyrus lexica have the same relation to the Scholia minora on papyrus (see below) that the Lexeis Homerikai have with the glosses in the D-Scholia; within each pair, one type of text is distinguished, above all, at the level of form by the fact that it follows alphabetic order (i.e. a paradigmatic order), whereas the other is distinguished by the fact that it follows the order of the text being interpreted (i.e. a syntagmatic order), even though the two pairs allow (as we have mentioned above) a high degree of osmosis of shared material; it scarcely needs mentioning that the formal differences between the two types of each pair derive from (and bear witness to) different uses and different exigencies of consultation. I shall return to these and other questions regarding the relationships between various categories of texts in the final part of this contribution. Quite different from the last category, within this panorama of Homerica, is that represented by the Homeric anthologies. Not very numerous, these texts have recently been analyzed and described by G. Nachtergael, who in the course of his work was able to understand and interpret a few fragments that had hitherto remained rather mysterious.8 In these anthologies one finds select excerpts of Homer connected by sections written in prose which summarize the omitted passages; these texts might also be considered, conversely, summaries that contain substantial groups of verses cited literally; in short, they are textual productions with an intrinsically double character resulting from the simultaneous presence of these two elements. The conceptual validity of the category and its usefulness have been demonstrated, if nothing else, by the fact that the category permits us to understand and to describe new finds, as happened recently in T. Renner’s work on P. Mich. inv. 4832c.9 The three categories that I have outlined and discussed thus far seem to me to be significant. That said, the principal goal of my current research into the

 and Selest. 105 (S), one must now also add a third, Bodl. Ms. gr. class. f. 114 (R). On these manuscripts, cf. also Naoumides 1975, 14 f. 7 De Marco 1946, vi. 8 Nachtergael 1971. 9 Renner 1979.

The Homerica on Papyrus  

Homerica on papyrus is that of identifying, as precisely as possible, the categories of papyrus documents that constitute the ancient antecedents of the medieval corpus of the Scholia D in Iliadem, in order to draw up an inventory of the papyrus texts that ought to be included in a complete edition of said scholia. (In doing so, I will be following the method employed by Erbse for his edition of the Scholia maiora, in which he included papyrus fragments of Iliadic hypomnemata, on which see above). As is well known, the corpus of the D-Scholia consists for the most part of two fundamental components: first, the glosses, that is to say, the explanations of single words or expressions within the Homeric texts, which amount to a large body of paraphrastic Worterklärungen that follow the text divided into lemmata (in this they differ from the continuous paraphrases of the Byzantine era); second, a series of more ample comments, composed substantially of ἱστορίαι and ζητήματα. In addition to these two major components, there is also a relatively small number of purely grammatical and exegetical annotations; and the whole corpus is characterized by the regular inclusion of the hypotheseis, normally two for each book, of which the first is shorter and more concise, the second longer and more discursive.10 This brief description should be a sufficient basis for a discussion of the papyrus antecedents to the D-Scholia, to which I now turn. We may begin with the mythographic ἱστορίαι, a type of exegetical text that is by now well-characterized and is attested in a considerable number of papyrus fragments.11 These texts contain a series of Homeric lemmata followed by the respective “comment” explaining some mythological topic, and from an analysis of the extant fragments it appears that there are effectively two characteristics that are always found in texts of this genre: 1) the presence of the Homeric lemmata (marked out in some manner, occasionally with the ekthesis) listed in an order that follows the succession of the verses of the Iliad or of the Odyssey — a fact which assures the close connection between these collections and the poetic text of Homer, and indicates that their primary and principal purpose was to provide “commentary” on Homer, even though they may well have assumed an autonomous role later; 2) the focus dedicated exclusively to mythographic material, designed to explain mythological references found in the text

 10 On the Scholia D in Iliadem, cf. Erbse, 1969–1988, vol. I, Praefatio p. xi; van der Valk 1963– 1964, vol. I, 202 ff.; Schmidt 1976, 2 f.; Montanari 1979, 4 ff. (further bibliography can be found in each of these works). For the manuscript tradition, cf. also Schimberg 1892; de Marco 1932, 371 ff.; de Marco 1941. On the relationships between the D-Scholia and texts found on papyrus, cf. esp. Henrichs 1971, 97 ff., which includes an extensive bibliography. 11 Cf. most recently Montanari 1985a (= this volume, ch. 20).

  The Homerica on Papyrus of Homer and arranged in a series of (more or less) brief narrationes, which are independent of each other and are distinguished by various means (such as the paragraphos, or beginning on a new line, among others). A considerable number of ἱστορίαι are followed by an attribution indicating the author of the mythographic material, in the form ἡ ἱστορία παρὰ τῷ δεῖνα (vel sim.), and this third characteristic — an extremely important aspect of these texts, and a clue to the origin and the history of these collections — can be considered distinctive of this genre of Homeric scholarship: although it is not constantly present (as the other two characteristics are), it is fairly common, and in any case when it is absent from the texts in the form in which they are known to us, this absence can hardly be attributed to anything other than the gradual impoverishment of this material in the course of the tradition. Regarding the mythographic content of these ἱστορίαι, it is important to recognize that it not infrequently goes beyond what would be strictly necessary in order to elucidate the Homeric text; it assumes a partially independent role, which reveals the presence of a scholarly or antiquarian interest in mythology that is pursued for its own sake. This partially independent role, however, is comparable to the accumulation of antiquarian and grammatical (or linguistic) material that is sometimes found in the scholia, and it cannot be understood as a characteristic that diminishes the close connection between these documents and the Homeric text; it merely bears witness to the existence of scholarly and antiquarian activity that is fully aware of its own specific value as a collection and as a reservoir of scholarly material. On the other hand, insofar as they are generally trustworthy, the subscriptions, with the attributions to specific authors, also testify that this material has its origin in a scholarly context of some seriousness,12 even if it is true that reductions, modifications, and other kinds of impoverishment have transformed the material in such a way that the form in which it has come to us is noticeably different from that of its original source. And it is precisely with reference to these subscriptions that one can make several interesting observations about the relationship between the papyrus ἱστορίαι and those found in the scholia (both the D-Scholia to the Iliad as well as the corresponding materials for the Odyssey). First of all, one notices a striking amount of agreement, namely in the many cases in which the subscription found in a papyrus fragment corresponds to one found in the respective scholion; in contrast, it is interesting to note the case of the ἱστορία to Α 263, for  12 On the problem of the origins of the corpus of ἱστορίαι (the so-called Mythographus Homericus), cf. Schwartz 1881; Panzer 1892; Lünstedt 1961; van der Valk 1963–1964, vol. I, 303 ff.; Arrighetti 1968, 79 ff.; Arrighetti 1977b, 51 ff.; Arrighetti 1977a, 52 ff.; Montanari 1979, 14 ff.

The Homerica on Papyrus  

which the subscription is lacking both in P. Oxy. 418 and in the corresponding D-Scholion. And yet there are some divergences to be found, as well: P. Oxy. 3003 omits the subscription of the ἱστορία to Ν 302, which instead is found in the D-Scholion; on the other hand, P. Hamb. III 199 preserves the subscription on Α 38 Τενέδοιο, whereas the D-Scholion has nothing.13 We thus find before us a panorama that contains every possible variation: agreement in presence or absence of an attribution; greater richness either in the papyrus document or in the medieval scholion. The differences must derive from the vicissitudes of the transmitted scholarly material, with the processes of transformation and impoverishment it endured along the way. It is important to note, however, that when the citation of the source is preserved both in the papyrus ἱστορία and in the corresponding scholion, the two citations are always identical: that is to say, thus far there have been found no cases of the same ἱστορία attributed to two different sources, one named in the papyrus and another named in the scholion. This is, naturally, both a strong sign of the reliability of the preserved subscriptions and also an important link joining the collections of ἱστορίαι restored to us by the papyri and the ἱστορίαι present in the scholia. As for the content of the ἱστορίαι, on the other hand, it is true that we do not always find an exact correspondence in the two traditions — though a general agreement in the basic outlines of the discussion, determined in part by reference to the Homeric text, is of course the rule — and sometimes the narrations appear in redactions that are clearly divergent, such that a close comparison reveals a complicated and varied web of similarities and differences; nonetheless, there can be no doubt that the collections of mythographic ἱστορίαι dedicated to Homer restored to us by the papyri constitute the ancient antecedent to the analogous collection present in the corpus of the scholia. It is by now well established that the connections are very solid, if nothing else on the basis of the three formal characteristics that have been delineated above, which prove valid both for the papyri and for the medieval scholia. The next category to take into consideration is that constituted by the hypotheseis, which in the papyri appear in various forms but are nonetheless readily classifiable. In the first place, however, we must separate from the proper hypotheseis, as rigorously and as precisely as possible, two different kinds of documents: namely, works in which one finds more or less free rhetorical reelaborations of  13 I thank the editor B. Kramer for authorization to include this fact regarding P. Hamb. III 199. All the facts that I present here regarding the text of the D-Scholia have been checked against the manuscripts, in order to ensure the highest degree of reliability possible.

  The Homerica on Papyrus Homeric themes; and works which consist of summaries, more or less extensive, of specific groups of verses (scholastic or rhetorical exercises, with varying degrees of sophistication). In these two kinds of documents the relationship with the poetic text is much closer and much more precise, and therefore the resulting prose text is incomparably more extensive and detailed than even the fullest hypothesis (for now I leave out the question of the literal paraphrases, for which see below). In these cases one might speak of reelaborations, summaries, or something along those lines; but given that the specific natures of the texts vary to a high degree, perhaps it is best not to search for a single common denomination but rather to be as precise as possible in the description of each case. One can imagine a gradated spectrum of vicinity to, or distance from, an exact correspondence with the poetic text; such a spectrum would have at its extremes, on the one hand, the concise summary of content offered by a hypothesis, and on the other, a literal paraphrase; in the middle there would be a wide range of possible reworkings of the Homeric text, each with its own specific and different characteristics. But to return to the hypotheseis proper: these are quite distinct and identifiable, and in the papyrus fragments they are generally found in two types of texts or ‘arrangements’.14 The first is the simple collection of hypotheseis, which contains the summaries of the content of the Homeric books one after the other; each hypothesis is identified by a title indicating the book of the poem by means of a letter of the alphabet, either by itself (as in P. Antin. 69) or else preceded by the feminine definite article in the genitive case, thus: (sc. ὑπόθεσις) τῆς A (sc. ῥαψῳδίας) (as in P. Mich. 1315, P. Oxy. 574, P. Laur. III 53). Each form of title appears also in the collections of Scholia minora: the first, e.g., in P. Amst. I 6 (ed. P.J. Sijpesteijn, ZPE 6, 1970, 130f.), the second, e.g., in P. Berol. 11634v (G. Pöthke, Forsch. u. Bericht. 8, 1967, 105 ff.) and in P. Colon. 2381 (Henrichs, ZPE 8, 1971, 3ff.). In general, the title is followed by the first verse of the book, and then the summary of the book’s content, which is sometimes introduced by the formula περιέχει ὡς (though this has up to now been found only in P. Ryl. 23).15 At least one fragment, P. Antin. 69, which contains the hypotheseis for X–Ω and β–γ, proves that a single document sometimes gathered together the hypotheseis for both of the Homeric poems; whereas P. Mich. 1315, at least judging from

 14 On the hypotheseis found on papyrus, cf. Henrichs 1973, 23 ff.; Montanari 1981, 101 ff. (= this volume, ch. 29); Montanari 1982 (= this volume, ch. 31). The overall picture of the Homeric hypotheseis restored to us via papyrus fragments was planned as part of my book, Montanari 1995a. 15 On this aspect, cf. also Luppe 1980.

The Homerica on Papyrus  

the subscriptio found to the right of col. II, seems to have contained only the hypotheseis for books Α–Ρ.16 Of this subtype overall, barring errors, five fragments are known up to this point (all mentioned above) and all of them date to the second and third centuries CE. The second arrangement differs from the first in its overall form: in this case we are dealing with collections of Scholia minora, entirely alike other examples of this genre (see below), which normally follow the order of the Homeric text; in these collections the glosses of each book are preceded by the hypotheseis of the same book. We have three examples of this type of text, two from the third century CE (P. Oxy. 3159, P. Oxy. 3160 + P. Strasb. gr. 1401)17 and one dated to the third or fourth century CE (P. Achmîm 2).18 The hypotheseis that appear in this arrangement are in and of themselves exactly identical to those found in the collections mentioned above: they have the same elements and the same structure. This union of hypotheseis + Scholia minora is, however, extremely interesting if one considers the history of the tradition, because it constitutes the exact antecedent of the close union between hypotheseis and D-Scholia which is found in the manuscript tradition:19 in fact, as I mentioned above, the hypotheseis in the manuscripts are very closely connected to this corpus of scholia, which is in part the heir and continuator of the Scholia minora; it is therefore significant that in antiquity we find the combination hypotheseis + Scholia minora, whereas hitherto no fragment of a hypomnema (or any other product of Homeric exegesis) has been found which is accompanied by hypotheseis. In addition to these two arrangements, however, we can list a third, for which I believe that I have found plausible identifying factors.20 These texts are what we might call “isolated hypotheseis”, papyrus documents in which, in all probability, has been recorded a single hypothesis; thus the single hypothesis appears isolated, an outlier with respect to the collections like those described above. Such texts belong in all likelihood to a cultural context of a fairly low level (perhaps that of elementary school), having as they do an episodic and desultory character, which reveals a reading of the Homeric texts that is not complete or even necessarily wide-ranging, but rather partial and limited, attesting to what we might call a sporadic attention for one or the other of the Homeric poems; in

 16 An edition of this document is presented by Henrichs 1973. 17 On P. Oxy. 3160 + P. Strasb. gr. 1401, cf. Luppe 1977; Montanari 1981 (= this volume, ch. 29). 18 The editio princeps is that of U. Wilcken in Sitz. Berlin. Akad. Wiss. 1887, 816 ff.; cf. Wilamowitz 1888; Ludwich 1902, 2 f.; a new edition has been produced by Collart 1931, 43 ff. 19 Cf. Henrichs 1973. 20 Cf. Montanari 1982a.

  The Homerica on Papyrus this, then, these texts clearly differ from the collections described above, which are sistematic and organic in character. And yet, if the analysis I have made elsewhere is valid, we must maintain that, despite their obvious differences from the other two types of arrangements, these documents have a close textual relationship with the manuscript hypotheseis, and they have clearly analogical relationships with the other papyrus hypotheseis, to the extent that we can claim for them too the character of hypotheseis, and we therefore include them in this category, even if we must admit that they have a different, indeed quite unique, character. At this point we must address the most complex category, or at least the richest one in terms of number of finds, namely, the Scholia minora. In connection with the Scholia minora one is immediately faced with the problem of the paraphrases; for in the first place, it is necessary to establish a rigorous, clear, and definitive distinction (both conceptual and terminological) which, although purely conventional to a certain extent, allows for a categorical distinction, commonly accepted and therefore immediately understandable.21 In the past, the terminology employed by scholars — and by papyrologists in particular — has not been entirely consistent, and indeed it is still not entirely consistent today: for instance, both in earlier writings and in the present day, one finds terms like “school exercise”, “paraphrase”, or “glossary” used, along with “Scholia minora”, to indicate one and the same kind of document. Therefore, I take the term Scholia minora to refer to a specific arrangement, which presents the Homeric text divided into a series of lemmata, which are separated from each other and consist of single words or expressions (properly chosen), after which there follows a Worterklärung (which may involve multiple glosses: e.g. μῆνιν: χόλον, ὀργήν, θυμόν) and sometimes in addition one or more grammatical or exegetical comments. Lemmata and interpretations are divided into two parallel, facing columns, or else presented as continuous text, in which each lemma need not begin on a new line. These documents can in no way be considered copies of the Iliad or the Odyssey; for even if the poetic text is reproduced in its entirety for certain verses, this does not happen for long stretches, much less in a systematic and complete fashion, albeit for a single book. Less frequently, however, it does happen that we find single Worterklärungen (glosses) written in a desultory manner in the interlinear space, or in the margin, of copies of the Homeric poems; such texts fall within our definition of Scholia minora, and when producing an autonomous edition, the lemmata should be extracted from the poetic text below and printed within angle brackets, that is to  21 Some discussion may be found in Henrichs 1971, 100–102.

The Homerica on Papyrus  

say, with the same convention adopted for lemmata that are supplemented in the editions of scholia proper. These two arrangements, of which the first is by far the more common, have two elements in common: they follow the order of the Homeric text — i.e., they have what we may call a syntagmatic structure, and thus they differ from the lexica, which have an alphabetic/paradigmatic structure (see above); and they present the interpretative material split up as isolated Worterklärungen, and in this way they differ from the paraphrases. Indeed, the term paraphrase should be employed to indicate a continuous prose rendering of the content of the Homeric poetry. The decisive characteristic of this kind of document is its formal independence from the Homeric text, i.e., its ability to be read as an autonomous text with its own grammatical and syntactic integrity. A paraphrase thus is a mode of interpretation that must necessarily proceed in a systematic fashion without making choices and splitting its individual components into single words or expressions and without introducing multiple or alternative explanations. It does, however, adhere closely to the original, even to the point of an extreme literalness; indeed, the precise, one-toone correspondence with the original is what distinguishes paraphrase from summary (see above). The textual arrangement of paraphrases can vary: sometimes the Homeric text is absent, and the papyrus fragment presents only the prose version;22 other times we find the regular alternation of one line of poetic text followed by one line of paraphrase, in exact correspondence.23 This type of document is, formally speaking, the closest antecedent to the Byzantine paraphrases, such as the one attributed to Michael Psellos (edited by Bekker), the one by Manuel Moschopoulos, the one by Theodorus Gaza, and so on;24 on the other hand, it is obvious that the documents belonging to the class of Scholia minora constitute the ancient antecedent of the substantial patrimony of glosses contained in the corpus of the D-Scholia. In the future, then, it would be advisable to maintain the terminological and conceptual distinction that we have outlined, both when producing editions of new fragments and when drawing up inventories of extant

 22 For example, Bodl. Gr. Inscription 3019 (ed. Parsons 1970); P. Erl. 5; P. Köln IV 180; Bodl. Gr. Inscription 3017 (ed. Hombert/Préaux 1951), which contains the paraphrase of Δ 349–363 on the recto and Sch. min. to Δ 364–373 on the verso. 23 For example, PSI 1176, a substantial fragment which alternates text/paraphrase for Β 617– 670. 24 On the Homeric paraphrases, cf. esp. Ludwich 1884–1885, II, 483 ff.; along with Lehrs 1873, 54 ff.; Parsons 1970, 135 ff.; Henrichs 1971, 101 with n. 14; Giangrande 1962, 152 ff. On the paraphrase by Moschopoulos, cf. Grandolini 1981; Melandri 1981; 1983.

  The Homerica on Papyrus documents; if nothing else, this will help to assure greater clarity and ease of reference in scholarship. The Scholia minora, the paraphrases, and the lexica all have in common the fact that, for the most part, they consist of glosses. Though it is perhaps already be clear from the previous discussion, it seems important to specify that we employ the term glosses to indicate the raw material, as it were, of literal interpretation and linguistic or lexical decipherment of the Homeric text; whereas we define lexica, Scholia minora, and paraphrases as, precisely, the various arrangements or forms in which that material is presented. Therefore, the terminological distinctions in this field depend more upon form of presentation than on the type of content. One thus finds, on the one hand, agreement and osmosis of interpretative material among the various categories of text, and, on the other hand, differences (both quantitative and qualitative) not only between documents of different categories but also between documents belonging to the same category. This is proof that the Homeric glosses we currently possess (in their various forms) are the remnants of a large mass of glossographic material, variable and fluctuating, that derives from both scholastic and scholarly contexts and was subsequently funneled into various forms, as needed, in different moments. The interpretative materials transmitted in the medieval manuscripts are perfectly analogous to those found in the ancient documents: we have already highlighted how, formally, the Lexeis Homerikai correspond to the papyrus lexica, the glosses found in the D-Scholia correspond to the Scholia minora, and the Byzantine paraphrases correspond to the paraphrases found on papyrus. Moreover, just as for the Scholia minora, the ancient lexica and the ancient paraphrases, one finds agreement and osmosis of interpretative material, and at the same time quantitative and qualitative differences between the D-Scholia, the Lexeis Homerikai, and the Byzantine paraphrases. Naturally, above and beyond these synchronic relationships, there are also diachronic relationships, as evidenced in the fact that interpretative materials of the ancient documents appear broadly distributed throughout the medieval ones. Of course, one must not forget that many interpretative comments can be found only in a single document (out of all the various categories here discussed), without any echo in any of the others, as is only reasonable given the unfixed and fluctuating nature of the glossographic material dedicated to Homer and the long course of its transmission. We are thus faced with a situation in which every single text of the categories discussed above, as a carrier of glossographic content, may have potential links with any of the others, whether directly, or indirectly by means of intermediate texts.

The Homerica on Papyrus  

I have already mentioned above that the ultimate purpose of my review of the Homerica on papyrus is that of compiling an inventory of the papyrus texts that must be included in my edition of the Scholia D in Iliadem. To be somewhat fuller and more precise, my goal has been to define the use that should be made of those papyrus documents in said edition. Now that the categories have been distinguished and defined, the task that remains is to compile and publish, for each category, an inventory that may serve as a reference point for scholars. The papyrus hypotheseis need to be included in the edition insofar as they represent the antecedents of the hypotheseis found in the manuscripts of the D-Scholia; an inventory of the texts belonging to this category has already been compiled by myself and will be made available in my book, Studi di filologia omerica antica. II, which ought to be published by the end of 1985 (see above, fn. 14). The mythographical ἱστορίαι dedicated to the Homeric poems and restored to us by papyrus fragments must be included insofar as they constitute the specific antecedent of the analogous corpus of ἱστορίαι that forms one of the major components of the D-Scholia; a complete inventory (up to Spring 1983) has been compiled by myself and is currently in press in the Atti del XVII Congresso Intern. di Papirologia (see above, n. 11). As for the glosses — another fundamental component of the D-Scholia — we have just seen how, considered all together (that is to say, as a type of interpretative material rather than as a form of presentation of said material), these have as their ancient antecedents not only the Scholia minora but also the lexica and the paraphrases. That said, I think it is necessary to resist the temptation of including the complete text of all three of these categories of papyrus documents; rather, it seems best to include, in the upcoming edition of the D-Scholia, only the texts of the Scholia minora on papyrus (according to the description and delineation of that category presented above); the lexica and the paraphrases found on papyrus will find adequate representation if they are used as Testimonia.25 A complete inventory of the papyrus fragments containing Homeric Scholia minora has been compiled by Lucia Raffaelli.26 With these three inventories, we ought to have a full list of all the papyrus texts that should be included within an edition of the Scholia D in Iliadem (barring errors or omissions, naturally, along with any new finds in the meantime). In publishing  25 The same goes for the Lexeis Homerikai and for the Byzantine paraphrases: in the edition of the corpus of the D-Scholia to the Iliad, these works will be cited in the apparatus of Testimonia. That said, the Lexeis Homerikai deserve to be edited and published separately, in a standalone volume which can be consulted with clear references to the D-Scholia for easy comparison. The publication of a corpus of Byzantine paraphrases of the Homeric poems would also represent a useful contribution to our studies. 26 Raffaelli 1984; see now Lundon, this vol. 320 n. 5.

  The Homerica on Papyrus these inventories in advance and independently of the edition of the D-Scholia, I hope both to render a useful service to scholars in the short term, and also to stimulate other researchers to notify me of any new finds and to point out anything else that may be useful for the forthcoming edition.

 Ancient Homeric Philology as Attested in the Papyri Over the past few decades, numerous contributions have greatly enriched the study of literary papyri and our understanding of Homeric philology in the ancient world. As a result, it seems timely to offer a survey and an appraisal of the extant fragmentary papyri which contain texts dedicated to the interpretation of the Homeric poems and other scholarly research on Homer. This will also present the opportunity for additional analyses and considerations that are firmly based on the latest research and should be of great interest. In the first place, it is not insignificant simply to note that, today, we actually find ourselves in a position even to make a survey and an appraisal of this kind: the study of ancient Homeric philology, as it is attested in the papyri, now has at its disposal enough texts, and texts of good enough quality, that we can be confident in our assessments — that is to say, we can at least be confident that we shall not be proven wrong quite as quickly, and as easily, as was once the case. This state of affairs holds true in particular for our field, the study of ancient scholarship on Homer; for it does not seem that the same can be said of many other related subject areas. Indeed, the Homerist — at least, one who is also concerned with the history of Homeric exegesis — can feel some comfort in this fact. Accordingly, therefore, he should now accept the challenge of moving beyond the limits imposed by the convenient protections of a methodological prudence invoked with facile refrains. I am, of course, fully aware that my confidence is open to criticism in some respects, and I believe that calls for caution are entirely legitimate, even necessary. That said, we should not allow excessive methodological discretion to prevent us from interpreting the texts we have at our disposal. Our task, now, is precisely this: to interpret and mobilize past research, in order to illuminate history where it seems hidden and obscure. The fifth volume of Erbse’s edition of the scholia on the Iliad appeared in 1977 (the sixth and the seventh volumes appeared in 1983 and 1988 and contain the indices). At that moment, Erbse — the worthy editor of these difficult texts, which represent the remnants of ancient erudition and are priceless documents of the history of philology — was able to list a small group of papyrus fragments (fifteen in total) which contained, in one form or another, exegetical material that could be considered the ancient prototype of the sort of content found in the scholiographic corpora which are identified by the acronym VMK (the socalled Viermännerkommentar or Four-man commentary: Didymus, Aristonicus, Nicanor, Herodian) and by the term Scholia exegetica. Each of those fifteen

https://doi.org/10.1515/9783110772371-033

  Ancient Homeric Philology as Attested in the Papyri items belonged to one of two types: the majority were fragments of hypomnemata, whereas a few took the form of marginal scholia to copies of the Iliad, as in the celebrated case of the Hawara Homer. Since then, at least up until the end of 1985, the developments in the field are as follows (barring errors or misinformation on my part, both of which are, as is well known, all too easy in the complex world of papyrology). In the first place, in the course of his work Erbse himself eliminated one of the fragments he had originally listed: PMich. 3688 = pap. XIV E. This item had already been tagged in the inventory with a cautious question-mark, and was later attributed instead to a commentary on Callimachus (cf. Erbse, vol. V, 509). We should now also add that another fragment listed by Erbse ought to be excluded from his inventory, since it belongs to a different category of texts. This is PLit.Lond. 142 = pap. VII E., an item which, on the basis of a study by Pfeiffer, Erbse had already listed with reservations, due to the difficulty of defining its content (cf. Erbse, vol. II, 392 f.) and which has now been identified as another fragment of the so-called B Homericus: this identification can be considered certain, thanks to the work of Annette Harder, who recognized that the item in question formed part of a larger unpublished fragment contained in the Oxyrhynchus Papyri collection (the entire text will appear in vol. 55 or 56), with remnants of ἱστορίαι relating to Books VII, VIII, and IX of the Iliad. To make up for these losses, we can now register two acquisitions, which I presume Erbse will mention in his Addenda. These are: first, a papyrus now in Trieste, published by S. Daris in the volume Studi in onore di L. A. Stella, Trieste 1975, 463 ff., containing portions of a commentary on Book IV; second, a papyrus now in Washington, PW.U. inv. 217, published by Z.M. Packman in BASP 10 (1973), 53 ff., which contains scraps of a commentary on Book IX. Thus the total number of securely-identified fragments of Iliadic scholarship amounts to fifteen (and perhaps will rise to sixteen, if we should also include here the hitherto unpublished PBerol. 9960: see below). Among these fragments there are some celebrated and highly authoritative specimens: from the already-mentioned marginal scholia contained in the Hawara Homer (its comments are very few in quantity, but they are of great interest insofar as they constitute a kind of apparatus criticus, as it were, registering many variant readings, especially ones attributed to Aristarchus), to the much more extensive portions of hypomnemata contained, for instance, in POxy. 1086, POxy. 1087, POxy. 221, etc. As far as the Odyssey is concerned, the principal inventories are less refined and less developed, given that the critical study of the Odyssean scholia remains very far behind compared to the study of the Iliadic scholia. Nonetheless, to my knowledge we currently possess six fragments of commentaries on the

Ancient Homeric Philology as Attested in the Papyri  

Odyssey; among these, however, the character of POxy. 2888 is, in my opinion, still uncertain and its text might constitute the remnants of a σύγγραμμα for this assessment, see Haslam/Montanari 1983, 114. Included in the above reckoning is the unpublished POxy. item that Haslam has announced will be made available in Oxyrhynchus Papyri vol. LIII; Haslam, moreover, is currently preparing a general study of the Odyssey-commentaries found on papyrus). To these six we must add PLit.Lond. 30, which presents marginal scholia on Odyssey Book III (cf. Kenyon 1894). Despite these comparatively smaller numbers, it is the Odyssey, rather than the Iliad, which has benefited most from recent papyrus finds, one of which represents by far the most interesting new item in our field: I mean of course the famous papyrus PLille inv. 83 + 134 + 93 b + 93 a + 114 t + 114 o + 87, which presents “commented extracts” (to use the definition of the text’s first editor) of Books XVI and XVII; this papyrus, as everyone knows by now, dates to the 3rd century BCE, and is therefore by far the most ancient commentary on Homer currently known to us, almost certainly having been written (just to give an idea!) before the birth of Aristarchus (see now the edition of Meillier 229 ff.). As has been remarked above, fragments of this kind — containing scholarship of a philological or exegetical nature — clearly constitute the ancient antecedents of the scholia that have been defined by Erbse as “Scholia grammatica vel maiora,” which, as he writes, “e commentariis aetatis Hellenisticae orta sunt” (vol. I, Praefatio, p. xi), that it to say, precisely those scholia that, as mentioned above, have been identified as excerpts of VMK, along with the corpus of the socalled exegetical scholia. Therefore, the place of these fragments in the history of ancient Homeric philology is clear and well-defined; the connections and the cultural continuities that link them with the corpora of scholia that took shape in late antiquity and in the Byzantine period are not generally difficult to imagine, whatever may be said of specific exegetical problems. Our reasoning applies to the Iliad, but it is clearly applicable to the Odyssey as well (and this will be even more clear once a reliable modern edition of the Odyssean scholia has been produced, which will distinguish the different materials contained in said scholia and will make these distinctions usable in scholarly practice, as has been done for the Iliad). Now, a few years ago I undertook a reexamination of the Homerica contained in the papyri, with the general goal of clarifying for myself, as far as possible, what kind of material they present and their characteristics in terms of form and content. But I also had a specific goal, more concrete and delimited: namely, to identify as precisely as possible the categories of scholarly texts discussing Homer that are found in papyri and must form the antecedents of the corpus known as Scholia D in Iliadem (D-Scholia, for short). As is well known

  Ancient Homeric Philology as Attested in the Papyri even to non-specialists, this is the major class of Iliadic (but also Odyssean) scholia that was excluded from Erbse’s edition and therefore still awaits a modern critical edition that will supplant the editio princeps of Janus Lascaris, published in Rome in 1517 (cf. Erbse, vol. I, Praefatio, pp. xi and lxvii; Montanari 1979, 4 ff., with bibliography). The overall picture of the antecedents to the D-Scholia found in the papyri has, I believe, now been clarified, both in terms of quality and quantity. The edition that I am planning will consist of five volumes, of which the first will be dedicated entirely to texts found on papyrus. The decision to isolate the papyrus texts and group them in a separate volume was made on the basis of three factors: first, the large number of such fragments, which amounts to a corpus of considerable size; second, the differing natures of these various fragments (we are dealing effectively with three different categories of texts found on papyrus, each of which is noticeably distinct from the others, as we shall see below); third, the greater ease of presentation and consultation which will result, especially given the need for an effective system of cross-references that will indicate the relationships between these texts and the corpus of scholia. Thus, the entire first volume will be dedicated, in structure and content, to the papyrus texts which represent the antecedents of the Scholia D in Iliadem. These antecedents to the D-Scholia can be grouped, as mentioned above, into three categories, as follows: first, fragments of Scholia minora (these are sometimes described by editors as “glossaries” or by other similar denominations; I have adopted the terminology that I outlined in my article, Montanari 1984 [= this volume, ch. 32]); second, Hypotheseis; and third, the so-called Mythographus Homericus. In what follows we shall briefly describe each category of text.

Scholia minora In a series of articles (Henrichs 1971, 1971b, 1971c, 1973), A. Henrichs clarified, in large part, the place that this type of scholia occupies in the history of ancient scholarship. As a result, my colleague Lucia M. Raffaelli was able to publish in 1984 an inventory of these texts (Raffaelli 1984, with bibliography), which now constitutes an essential reference point for a general (and, one hopes, exhaustive) view of the corpus of Scholia minora on the Iliad and the Odyssey. This inventory lists 59 papyrus fragments for the Iliad and 10 for the Odyssey (in these numbers are included — barring errors or omissions — those fragments of Homer that present interlinear or marginal glosses; such fragments represent, in any case, a very small percentage of the total). I would like to make some brief remarks regarding the texts related to the Iliad. Of the 59 papyrus fragments,

Ancient Homeric Philology as Attested in the Papyri  

three are tagged with a question-mark to indicate their doubtful status: these are no.s 029, 032, 052, of which no. 032 is a papyrus from Vienna that received mention at the end of the nineteenth century but is no longer to be found and should be considered lost; whereas no. 029 = PBerol. 9960 is an unpublished fragment found in East Berlin, which at the present state of research (still in progress) seems to contain not Scholia minora but rather some other kind of commentary (it may turn out to be a tiny remnant of a hypomnema, in which case it will increase the number of extant fragments of that kind of text, mentioned above). This means that, of this subtype, the actual number of texts that can be consulted — according to Raffaelli’s inventory — turns out to be 57. But at this point it is important to add that, in the past two years, we have seen a considerable increase in such texts (considerable at least compared to the numbers we are used to in our field): in total, eight new pieces have been made available; of these, two constitute parts of fragments that were already known, whereas the other six appear to be (at least as far as we can now ascertain) completely new. Therefore, the number of items rises to a total of 63. Such a number is, in my opinion, really quite significant, and it places this category of texts in the first rank among the remains of ancient philology and scholarship, as far as the overall distribution is concerned, — let us not forget, in a temporal arc that ranges from the first to the seventh century CE.

Hypotheseis It seems unnecessary to dwell at length on what is well-known by now: (1) within the manuscript tradition that has come down to us, the hypotheseis are closely connected to the D-Scholia, but not to the other corpora of scholia; (2) in the papyri, hypotheseis are found in three distinct sorts of context, namely, (i) in collections dedicated specifically to the hypotheseis of all the books of one or both of the Homeric poems, (ii) added by way of prefaces to the Scholia minora to the individual books, and (iii) as isolated hypotheseis copied occasionally (cf. Henrichs 1973, 23ff.; Montanari 1981 [= this volume, ch. 29]; ZPE 48, 1982, 89ff. [= this volume, ch. 31]; Montanari 1982a; Montanari 1984 [= this volume, ch. 32]; Luppe 1977; Luppe 1980). If my reckoning is correct and complete, we have at our disposal eight fragments of hypotheseis for the Iliad and four for the Odyssey (in this total I count PAntin. 69 twice, since it — and, I add in passing, it alone — presents both Homeric poems together). It should be further noted that in two cases regarding the Iliad (PAchmîm 2 = 003 Raffaelli; POxy. 3159 = 041 Raffaelli) and in one case regarding the Odyssey (POxy. 3160 + PStrasb. 1401 = 003 Raffaelli) we are dealing with texts that present the Scholia minora preceded by the

  Ancient Homeric Philology as Attested in the Papyri respective hypotheseis; these texts, therefore, have been also included in the inventory of Scholia minora. The chronological range of the extant hypotheseis is from the second to the fourth century CE.

Mythographus Homericus The inventory of the fragments of MH, which I compiled recently (Montanari 1985a [= this volume, ch. 20]), needs to be updated in order to include the hitherto unpublished POxy. fragment (already mentioned above) which Annette Harder has identified as forming part of the same text present in PLit.Lond. 142. This is an exciting development, not only because it adds to our knowledge, but also because it clarifies an uncertainty that was present in the inventory as previously drawn up, by allowing us to remove one of the two question-marks that appeared therein. Thus we now possess no fewer than six securely-identified fragments for the Iliad, against only one for the Odyssey (PSI 1173 — which is, however, the most substantial in size); the chronological arc of these fragments ranges, roughly, from the first to the fifth century CE. The MH texts are perhaps the most interesting of all those discussed here; certainly they are the most problematic. This is not the place, however, to discuss the complex questions regarding the origin, the character, and the original form of the Mythographus Homericus, or the issues relating to the trustworthiness (or not) of the subscriptions to the ἱστορίαι that, in the present state, constitute the MH, not only in the papyrus fragments but also in the corpus of the D-Scholia. Nonetheless, we can at least mention that in the papyri, which are attested across several centuries of the Imperial period, the MH appears not, as one might expect, in the form of a mythographical textbook, but rather in the form of a hypomnema, following the order of the Homeric text, to which it is connected precisely and organically, thanks to the constant use of lemmas: this formal element, in and of itself, leads one to suppose that the earliest phase of MH is to be found in the circles of the most authoritative scholarly and philological research on Homer (cf. Montanari 1984, 130 ff., with bibliography [= this volume, ch. 32], and Montanari 1985a [= this volume, ch. 20]; Luppe 1984; Luppe 1985; Janko 1986, 51 ff.); only later, after numerous subsequent phases and modifications, did the MH texts arrive in the hands of the individual who compiled the D-Scholia and was therefore responsible for the fusion of MH with the bulk of the paraphrastic-glossographic materials. There are of course other kinds of materials containing ancient scholarship on Homer, but they do not fall within the remit of this contribution and need not be mentioned. Therefore, I would like to end this survey by remarking that

Ancient Homeric Philology as Attested in the Papyri  

three updated inventories (dedicated, precisely, to the three categories of texts discussed above) will be included in my Studi di filologia omerica antica II, which should be published by the end of 1988. Now, given that the texts in these inventories that contain material related to the Iliad will be included in the first volume of my projected edition of Scholia D in Iliadem (mentioned above), it seems highly appropriate to make use of this occasion — the Congresso di Papirologia — to ask my friends and colleagues in papyrology to call my attention to any gaps or omissions they have noticed, or even other unpublished fragments that ought to be included (if such exist). This sort of collaboration is essential, in order that complex corpora of this kind do not prove lacking or obsolete at the very moment of their release into the world. There can be no doubt, in my opinion, that we are dealing with a very significant quantity of texts. When they will be published all together in one place, they will certainly offer a great amount of material for analysis, for further research, and for discussions and investigations of various kinds. But it is also already possible, at the present state of research, to make several interesting observations, not only if we keep within the boundaries indicated in the foregoing discussion, but also if we broaden our perspective to include, for instance, lexica (whether general or exclusively Homeric) or continuous paraphrases of the Homeric texts (in which case we are faced with the daunting quantity of Byzantine paraphrase, an important sub-field in which much work remains to be done and many critical editions are needed). I would like now to present a few of such observations, more in the form of problems than solutions, and in a manner that is, admittedly, somewhat rhapsodic. If we consider them all together, the papyrus fragments containing Scholia minora, Hypotheseis, and Mythographus Homericus amount to a total of nearly 80 items for the Iliad (compared to around fifteen for the Odyssey), and they span a chronological arc that ranges from the first to the seventh century CE. It will require more in-depth reflection and analysis to determine whether these items will provide a sufficient basis for ascertaining when the corpus of the Scholia D in Iliadem came to be formed (analogous considerations will hold true for the scholia in Odysseam); and the same can be said, above all, for our understanding of the quality of the materials employed by whoever compiled the D-Scholia; nonetheless, the items thus far at our disposal will certainly provide essential facts for consideration. A separate issue is presented by Mythographus Homericus, in terms of the date of its compilation, its features, and the possibility of identifying the stages of epitomization and/or addition that have affected the original text, not to mention the transformations that the text underwent before it came to be used by the compiler of the D-Scholia.

  Ancient Homeric Philology as Attested in the Papyri One element of great interest in these texts, especially for its continuous presence across time, is that represented by lexical glosses. As a practice, we know that the glossing of words or phrases existed already in the fifth century BCE, and it is to be found in abundance not only in the Homeric D-Scholia but also in many other scholarly texts, such as, of course, lexica and in the papyri its diffusion is so wide, and above all so prevalent, that it cannot help but strike one as the bridge that stretches from the earliest forerunners of Homeric criticism, who lived before the birth of philology as a scientific and professional field, to the time when manuscripts circulated in the Byzantine era. To be more precise: whereas we find the practice of isolated glossing of individual words — as attested in the papyrus fragments of the Scholia minora — starting in the first century CE, traces of similar glossographic activity can also be found (as has already been remarked by de Marco 1946, xxx ff.) in important hypomnemata, such as POxy. 1086 and POxy. 1087, which date back to the first century BCE, not to mention the large fragment of a hypomnema which was found in POxy. 221, which is instead from the second century CE. Therefore, the mixture of glossographic or paraphrastic material with the critical and exegetical material typical of the hypomnemata, although not very common, is nonetheless securely attested both prior to, and contemporaneous with, the more extensive glossographic presence of the fragments of the Scholia minora as such. This fact should not be overlooked: it cannot but remind us that paraphrase is also found in the texts of Alexandrian exegesis, even in Aristarchus; as a result, the oftrepeated idea that there is a direct line of connection from the didactic practices of the schoolroom (at a very basic level, even) to the Scholia minora must be treated with great caution, even scepticism. Indeed, I would go so far as to maintain that the tradition of the Scholia minora contains elements that derive directly from scholarly lexicography at its highest levels and can be connected to the contexts in which said lexicography was produced. And it is precisely with this in mind that we should study with great care one text in particular, due to its extreme importance, namely the very old commentary on the Odyssey restored to us by the Lille papyrus of the third century BCE (mentioned above), which combines both critical exegesis and paraphrase. As I have already remarked, this commentary was almost certainly produced before the birth of Aristarchus, when Aristophanes of Byzantium was active — that is to say, very early on in the period that saw the greatest productivity in the writing of important hypomnemata. My space here is limited, so for now I will simply remark that in the future it will be essential to analyze this commentary very closely, and precisely with regard to the combination of exegesis and paraphrase, since

Ancient Homeric Philology as Attested in the Papyri  

it will certainly offer interesting material that will help to illuminate the history of ancient Homeric philology and its various interpretative strategies. We can say with confidence, then, that the practice of glossing words in Homeric texts, and thus, closely related, the glossographic element contained in the scholia, are both ancient. The same can be said of the ideological and interpretative problems presented by the mythological content of the Homeric poems; and the Mythographus Homericus cannot be considered separately from this tradition. Therefore, the materials present in the corpus of the D-Scholia are directly related to the two major areas of inquiry regarding Homer in antiquity, and indeed, to the two most ancient forms of reading and explaining the Homeric texts — forms of reading and exegesis that predate the methodologies developed by Hellenistic philologists, but later continued to be employed and developed alongside the work of those scholars. Then, after the corpus of the D-Scholia was compiled in the late antique or proto-medieval period, these materials were widely available in manuscripts during the medieval era and subsequently in the Renaissance. Only later did they pass into relative obscurity, when Villoison discovered the A- and B-scholia, which provided the substance behind Wolf’s celebrated Prolegomena. Today, however, the D-Scholia are at the center of renewed scholarly interest, which is of course nothing other than the most recent chapter of the long history of critical exegesis of the Homeric poems; and this renewed interest demands a new edition, an edition that can be called truly critical. Addendum: By way of a brief update, I note that the commentary on the Odyssey, which I mentioned above as being unpublished, has now been published: M.W. Haslam, Oxyrhynchus Papyri, vol. LIII, London 1986, nr. 3710, 89 ff. (the fragment presents a substantial portion of a commentary on Odyssey Book XX). Also regarding the Odyssey, on the problematic item POxy. 2888, see now Gangutia 1986. Additionally, it is worth noting that the fragment of a commentary on Iliad IX presented by PW.U. inv. 217 has been reexamined by Wouters 1976; Wouters proposes a new dating for the fragment, arguing that it goes back to the second century BCE; see also Haslam 1985.

 Note sulla tradizione manoscritta degli scholia D in Iliadem: Un caso di errore da archetipo Da diversi anni (come è stato annunciato) sto lavorando a una nuova edizione degli Scholia D all’Iliade, che si desidera ormai da molto tempo (non essendo mai stata sostituita l’editio princeps pubblicata da J. Lascaris a Roma nel 1517) e che deve completare l’insieme degli Scholia Vetera all’Iliade, dopo che H. Erbse ha edito gli scoli derivati da VMK (cioè il Commentario dei Quattro: Didimo, Aristonico, Nicanore, Erodiano) e i cosiddetti Scholia Exegetica.1 L’edizione completa degli Scholia D in Iliadem prevede un insieme di cinque volumi: nel vol. I troveranno posto i precedenti papiracei (Hypotheseis, Scholia minora, Mythographus Homericus)2 del corpus degli scoli trasmesso dai manoscritti bizantini; i voll. II–IV recheranno il testo degli Scholia D con gli indici; il vol. V conterrà le Lexeis Homerikai con i loro indici.3 Di questa raccolta scoliografica — esclusi i testi papiracei, ai quali dedicò solo sporadiche attenzioni — cominciò a occuparsi molti anni fa Vittorio de Marco, che riprese poi il lavoro in età molto avanzata e non potè arrivare al suo compimento. De Marco ha lasciato un testo provvisorio fino al canto XVII, basato sulla collazione dei quattro manoscritti CHRV + l’ed. pr. di Lascaris. Per sua volontà, tali carte sono state consegnate a me, che avevo indipendentemente iniziato l’edizione degli Scholia D e, venuto a conoscenza del suo ritorno di interesse, avevo preso contatto una decina d’anni prima della sua morte e avevo avuto periodicamente con lui simpatiche e fruttuose conversazioni.4 Il lavoro per un’edizione critica moderna fu intrapreso verso la fine del secolo scorso da A. Schimberg, che fece approfondite indagini sulla tradizione  1 Erbse 1969–1988: per le classi degli scoli e i problemi ecdotici si veda la Praefatio nel vol. I e Erbse 1960. Cfr. anche Schmidt 1976, 1 sgg. 2 Cfr. Montanari 1979; 1984 (= questo volume, cap. 32); 1985a (= questo volume, cap. 20); 1988b (= questo volume, cap. 33). Per i frammenti papiracei con Scholia minora cfr. anche Henrichs 1971; 1971b; 1971c; 1973; Raffaelli 1984: a questa essenziale lista di riferimento devono già essere aggiunti otto testi per l’Iliade e uno per l’Odissea (è in preparazione una nuova edizione aggiornata); cfr. anche quanto pubblicato in The Oxyrhynchus Papyri, LVI, Londra 1989, 17 sgg., e infine Montanari 1993a. 3 Cfr. le opp. citt. alla nota precedente e Gnomon LVII, 1985, 763; sugli Scholia D vedi anche van der Valk 1963–1964, I 202–413, con la recensione di H. Erbse in Gnomon XXXVI, 1964, 553–555. 4 Cfr. Geymonat 1989. https://doi.org/10.1515/9783110772371-034

Note sulla tradizione manoscritta degli scholia D in Iliadem  

manoscritta ma arrivò a produrre soltanto un piccolo specimen limitato ai primi cinquanta versi dell’Iliade.5 Come dicevo, il lavoro fu ripreso qualche decennio dopo (presumibilmente negli anni Venti di questo secolo) da Vittorio de Marco, che portò importanti correzioni e miglioramenti ai risultati di Schimberg e modificò sensibilmente lo stemma della tradizione manoscritta.6 Inoltre de Marco ebbe il merito di valorizzare il rapporto con l’elemento glossografico degli scholia D di un lessico omerico alfabetico, che reca nei manoscritti il titolo Λέξεις Ὁμηριϰαὶ ϰατὰ στοιχεῖον: pensò di iniziare la sua edizione con queste Lexeis Homerikai e ne pubblicò un primo fascicolo.7 Purtroppo de Marco abbandonò poi il lavoro e non videro mai la luce né il resto delle Lexeis né l’edizione degli Scholia D: come dicevo sopra, egli ritornò a occuparsene in tarda età verso la fine degli anni Settanta, ma il lavoro rimase lontano dalla conclusione.8 Per diversi anni ho dedicato principalmente le mie forze a mettere insieme il materiale dei testi papiracei per il Vol. I, che ha richiesto moltissimi sforzi e ha comportato non poche difficoltà di vario genere (selezione dei testi e definizione del piano di quelli da includere, reperimento delle fotografie, collazione degli originali, aggiornamento per l’apparizione di nuovi frammenti e cosí via) e adesso non è lontano dal suo compimento.9 Nel frattempo, comunque, è proseguito anche il lavoro sulla tradizione manoscritta e la costituzione del testo del corpus scoliastico vero e proprio e delle Lexeis Homerikai. Allo stato attuale degli studi non è piú possibile considerare separatamente questi tre elementi (vale a dire i precedenti restituiti dai papiri, gli Scholia D e le Lexeis Homerikai), per cui l’edizione deve prevederli tutti, corredati da un congruo apparato di testimonia e loci similes e dagli opportuni rimandi incrociati secondo il piano definito sopra. I manoscritti principali che recano gli Scholia D in Iliadem sono quattro, cioè C H R V , nei quali si trova il corpus di scoli autonomo, vale a dire senza il testo del poema.10 Bisogna poi includere il celebre codice Ven. A, nel quale sono confluiti,

 5 Schimberg 1890; 1891–1892; 1897. 6 de Marco 1932; 1941. 7 de Marco 1946. Ai due manoscritti U S bisogna aggiungere O = Bodl. gr. class, f 114. 8 Alcuni frutti del lavoro sono stati pubblicati da de Marco nei due articoli: de Marco 1975; 1984. 9 Cfr. opp. citt. sopra alla nota 2: si può prevedere che sia pronto per la stampa in due o tre anni circa. 10 Cfr. Schimberg e de Marco, opp. citt. sopra alle note 5 e 6; Henrichs 1971, 107 sgg.; Montanari 1979; Cavallo 1989. Un frammento di Scholia D autonomi, precisamente da A 10 a B 266, si trova anche nel manoscritto Crypt. Z.α.XXV, sul quale cfr. Wilson 1976: è un apografo di V, scritto dalla stessa mano, per cui è eliminato dallo stemma.

  Note sulla tradizione manoscritta degli scholia D in Iliadem come è ben noto, anche estratti di scoli della classe D.11 A questi si è ultimamente aggiunto un frammento restituito dal codice miscellaneo Paris. Suppl. Gr. 679 (P), che nei fogli 1–20, probabilmente del sec. X, restituisce gli scoli da Π 140 a Σ 593 (con qualche lacuna).12 Facciamo seguire le sigle: Manoscritti A C H P R V

Venetus Graec. 822 (olim Marc. gr. 454), sec. X Rom. Bibl. Naz. gr. 6 + Matrit. 4626 (71 Ir.), sec. IX ex./X in. Vat. gr. 2193, sec. XII Paris. Suppl. Gr. 679, ff. 1–20, sec. X Vat. gr. 32, sec. XII Vat. gr. 33, sec. XI ex.

Edizioni La

J. Lascaris, Sch. D in Iliadem, editio princeps, Romae 1517

Barnes Heyne Bekker

J. Barnes, Homeri Ilias et Odyssea cum scholiis, Cantabrigiae 1711 Homeri Ilias, cur. C.G. Heyne, acced. Scholia Minora, Oxonii 1821 Scholia in Homeri Iliadem, ex rec. I. Bekkeri, Berolini 1825

Lo stemma della tradizione manoscritta, a cui è pervenuto de Marco nei suoi lavori preparatori, comprende i quattro codici C H R V , il codice Ven. A e l’ed. pr. La (la posizione di P, scoperto in seguito, deve ancora essere definita). Per quanto ho potuto vedere a fino a questo punto del lavoro, tale stemma rimane valido ed è il seguente:

 11 Cfr. Erbse 1969–1988, vol. I, Praefatio, pp. XLV e LXXIV sg.; de Marco 1932, 403 sgg. 12 Cavallo 1989, 611, data questi fogli al sec. X. Il codice riunisce pezzi di epoche e contenuti differenti; inoltre i ff. 1–20 sono in successione disordinata: f. 1 r contiene l’incipit di Σ, la parte iniziale conservata è al f. 11 r con lo sch. Π 140, la parte finale è al f. 20 v; qualche passo è del tutto illeggibile. Per i fogli con scoli esiodei (risalenti al sec. XII) il ms. è stato utilizzato in Di Gregorio 1975: vd. IX.

Note sulla tradizione manoscritta degli scholia D in Iliadem  

In questa nota ci occupiamo ora di uno scolio che offre un esempio di estremo interesse per la tradizione manoscritta. Il verso in questione è Iliade N 225, il lemma è ἀνδύεται. Posidone, assunte le sembianze di Toante, capo degli Etoli, rivolge una esortazione a combattere a Idomeneo, che cosí risponde (vv. 222–225): ὦ Θόαν, οὔ τις ἀνὴρ νῦν γ᾿ αἴτιος, ὅσσον ἔγωγε γιγνώσϰω· πάντες γὰρ ἐπιστάμεθα πτολεμίζειν. οὔτε τινὰ δέος ἴσχει ἀϰήριον οὔτε τις ὄϰνῳ εἴϰων ἀνδύεται πόλεμον ϰαϰόν. Toante, nessuno ora è colpevole, almeno per quanto io sappia: tutti infatti sappiamo combattere. Nessuno trattiene un vile terrore e nessuno all’ignavia cedendo si ritira dalla guerra funesta.

Per il lemma ἀνδύεται al v. 225 i manoscritti degli Scholia D recano una glossa, che si presenta come segue in C H R V e in La:13 C Η R V La

ἀ ν δ ύ ε τ α ι : ἀναδύεται, ἀναφύεται ἀ ν δ ύ ε τ α ι : ἀναδύεται, ἀναφύεται ἀ ν α δ ύ ε τ α ι : ἀναφύεται ἀ ν α δ ύ ε τ α ι : ἀναφύεται ἀ ν δ ύ ε τ α ι : ἀναδύεται

 13 La glossa manca in A; il frammento P comincia più avanti.

  Note sulla tradizione manoscritta degli scholia D in Iliadem A quanto pare, la prima preoccupazione è stata quella di ripristinare la forma normale ἀναδύεται14 rispetto al poetico ἀνδύεται con apocope della preposizione:15 in conseguenza di questo, i manoscritti R e V (dunque v) hanno preso ἀναδύεται come lemma (e in verità ἀναδύεται è anche lezione erronea che si trova in diversi manoscritti dell’Iliade).16 La vera «traduzione» della parola poetica è data con ἀναφύεται, che compare in tutti e quattro i testimoni C H R V , ma non dà certo un senso accettabile. Il significato di ἀναδύομαι è senza dubbio quello di «indietreggiare, ritirarsi» e dunque «fuggire» dalla guerra, ritrarsi dalla battaglia. Che ἀναφύεται non dia un senso accettabile deve essersi accorto facilmente anche Lascaris, che lo ha omesso, limitandosi a una poco significativa glossa ἀνδύεται: ἀναδύεται: in questa forma la glossa è stata ripresa tale e quale da Barnes e da Heyne (non è presente in Bekker). Che cosa si nasconda sotto il misterioso ἀναφύεται ce lo rivela un frammento di papiro, cioè PRyl 536, che contiene Scholia minora a Iliade N 198–562 (con lacune)17 e che a recto col. I 18 presenta la seguente glossa: ἀ]ν̣ δύεται ἀν[αδύετα]ι̣ ϰαὶ φεύγει

Per ragioni di spazio, giunto alla fine del rigo, lo scriba ha completato la glossa scrivendo γει sopra φευ. L’integrazione non pone problemi: anche qui la glossa offre prima la forma usuale senza apocope e poi la «traduzione». L’uso di (ἀνα)φεύγω per glossare ἀναδύομαι è confermato da paralleli lessicografici: Esichio α 4287 ἀ ν α δ ῦ ν α ι : ἐϰϰλῖναι. ἀναφυγεῖν ϰτλ. Esichio α 4288 ἀ ν α δ ύ ο μ α ι : φεύγω. ἀνανεύω. ὑποχωρῶ Συναγωγὴ λέξ. χρ. 82. 25 ἀ ν α δ υ ό μ ε ν ο ς : φεύγων, ἀναχωρῶν, ἀναβαλλόμενος.18 Fozio α 1464 ἀ ν α δ ῦ ν α ι : ἐϰϰλῖναι. ἢ ἀναφυγεῖν ϰτλ.  14 Come si trova per es. nella Parafrasi di Bekker e in Eustazio 928, 37. 15 Per questi fenomeni vd. Chantraine 1973–1981, I, 87. ἀνδύεται è la lezione accettata da Aristarco, ma esisteva anche ἀδδύεται con assimilazione (che, secondo gli apparati di Ludwich e Allen, non compare nei manoscritti): cfr. sch. N 225 a Erbse, con i riferimenti. Su questo verso cfr. anche Wackernagel 1935, vol. I, 140 sg. 16 Cfr. gli apparati di Ludwich e Allen ad loc. 17 Catalogue of the Greek and Latin Papyri in the John Rylands Library, Manchester. Vol. III, Theological and Literary Texts (Nos. 457–551), ed. C.H. Roberts, Manchester, University Press, 1938, 176–178. PRyl 536 = Pack2 1195 è incluso nella lista di Raffaelli 1984, 153, Iliade 051; cfr. Montanari 1993a. 18 Σb: dal codice Coislin. 345 ed. L. Bachmann, AG I, 1–422.

Note sulla tradizione manoscritta degli scholia D in Iliadem  

Fozio α 1466 ἀ ν α δ υ ό μ ε ν ο ς : φεύγων, ἀναβαλλόμενος. Su. α 1865 ἀ ν α δ υ ό μ ε ν ο ς : φεύγων, ἀναβαλλόμενος. Mi pare, dunque, che ci siano prove sufficienti per dire che il corrotto ἀναφύεται negli Scholia D deve essere corretto con una forma di ἀναφεύγω. La prima idea che viene sarebbe di emendare in ἀναφεύγεται, ma la forma media quanto meno crea dei problemi, perché non è usata. La lezione giusta, suggerita da PRyl 536, potrebbe dunque essere ἀναφεύγει, solo all’apparenza piú lontano di ἀναφεύγεται dal tradito ἀναφύεται, ma forse addirittura piú vicino (anche paleograficamente), e in ogni caso risultante da un errore di lettura facilmente immaginabile. Abbiamo comunque un bell’esempio di sicuro errore congiuntivo, risalente all’archetipo. La dipendenza dei nostri manoscritti dall’archetipo comune ipotizzato da Schimberg e de Marco, già ben sostenuta da altri errori congiuntivi, trova un’ulteriore conferma. Nell’edizione il comportamento corretto sarà quello di mettere tra cruces ἀναφύεται e indicare in apparato la possibile correzione con ἀναφεύγει ο ἀναφεύγεται.

 Aristarchus in the D-Scholia to the Iliad For several years now, I have been at work on a new edition of the Iliadic D-Scholia,1 which has represented a long-standing desideratum in the field; in effect, the editio princeps published by J. Lascaris in Rome in the year 1517 has never yet been truly surpassed or replaced. The new edition of the Scholia D in Iliadem will consist of five volumes: Vol. I will contain the ancient scholarly antecedents that are found in papyrus fragments (Hypotheseis, Scholia minora, Mythographus Homericus); Voll. II–IV will contain the corpus of the D-Scholia that has been transmitted to us in the Byzantine manuscripts, along with the relative indices; and Vol. V will contain the Lexeis Homerikai with their indices.2 Many years ago, Vittorio de Marco began to dedicate his scholarly attention to this same collection of scholia (not including the papyrus texts, which he studied only sporadically); after a long hiatus, he took up the task again in his later years and was not able to complete it. De Marco left behind a provisional critical text of the D-Scholia up to Book XVII, which was based on the collation of the four manuscripts C H R V along with Lascaris’ editio princeps. In his will he left his papers to me, since I had independently begun my own edition of the D-Scholia, and, when I found out that he had returned to the subject, I had reached out to him about ten years before he passed away, and we had numerous friendly and fruitful conversations.3 It was A. Schimberg who, towards the end of the nineteenth century, began the work of producing a modern critical edition of these scholia. Schimberg carried out detailed research into the manuscript tradition, but he only managed

 1 See the announcement in Gnomon 57 (1985), 763. See also Montanari 1979, along with my subsequent contributions: Montanari 1984 (= this volume, ch. 32); 1985a (= this volume, ch. 20); 1988b (= this volume, ch. 33); 1994b (= this volume, ch. 34) (all of these items have been updated and included in Montanari 1995a). Lastly, see also my article dedicated to the problems presented by the text known as the Mythographus Homericus: Montanari 1995 (= this volume, ch. 22). 2 The corpus of the D-Scholia completes the publication of the Scholia Vetera on the Iliad, of which H. Erbse has already published the scholia deriving from Didymus, Aristonicus, Nicanor, and Herodian (that is, the scholia deriving from the so-called Viermännerkommentar [Fourman-commentary] = VMK) and the so-called Scholia Exegetica, in Erbse 1969–1988; for discussion of the various classes of scholia and related text-critical problems, see Erbse’s Praefatio in vol. I of his edition along with his monograph Erbse 1960. See also: van der Valk 1963–64, on the D-Scholia esp. vol. I, 202–413, along with Erbse’s review of that work in Gnomon 36 (1964), 553–555; Schmidt 1976, 1 ff. 3 See Geymonat 1989. https://doi.org/10.1515/9783110772371-035

Aristarchus in the D-Scholia to the Iliad  

to complete a small specimen of a definitive text, limited to the first fifty lines of the Iliad.4 As I mentioned above, the work was then taken up by Vittorio de Marco a few decades later (presumably in the 1920s). De Marco was responsible for important corrections and improvements upon the results obtained by Schimberg, and he revised the stemma of the manuscript tradition to a significant degree.5 Furthermore, it was de Marco who recognized the importance of the connection between the glossographical content of the D-Scholia and the alphabetic Homeric lexicon that bears the title Λέξεις Ὁμηρικαὶ κατὰ στοιχεῖον at the head of its manuscript copies. De Marco decided that he would begin his edition precisely with these Lexeis Homerikai, and he published a first fascicle of a critical text of that material.6 Unfortunately, however, he subsequently abandoned the project, and neither the rest of the Lexeis nor his projected edition of the D-Scholia ever saw the light of day. As I mentioned above, he returned to the work in his later years, towards the end of the 1970s, but the edition always remained far from being completed.7 For my part, for several years my principal attention was dedicated to identifying and systematizing the papyrus texts for Vol. I. This was intense work, and it entailed numerous difficulties of various kinds, such as: the selection of the texts to include; the definition of the organizational plan of the volume; the acquisition of photographs; the collation of the original texts; not to mention the periodic updates and modifications made necessary due to the appearance of new fragments, and so on. The volume, in any case, is now not far from completion.8 In the meantime, I have extended my research into the medieval manuscript tradition and I have continued the work of establishing a critical text of the corpus of the scholia and of the Lexeis Homerikai. Given our current understanding of ancient scholarship and its subsequent forms, it is no longer possible to consider these three elements separately from one another (the three elements being of course the antecedents that have been restored to us via papyrus

 4 Schimberg, 1890; 1891–1892; 1897. 5 De Marco 1932; 1941. 6 De Marco 1946. To the two manuscripts employed by de Marco one must now add Bodl. gr. class. f 114 = O. 7 De Marco published some of the results of his research in two articles: De Marco 1975; 1984. 8 See the works cited in footnote 1 above, with bibliography. On the Scholia minora found in the papyri, see also Henrichs 1971; 1971b; 1971c; 1973; Raffaelli 1984 (to this inventory, essential as a point of reference, a few items must already be added for both the Iliad and the Odyssey; an updated edition of the inventory is therefore currently in preparation). See also the material published in The Oxyrhynchus Papyri, LVI, London 1989, 17 ff. Lastly, see Montanari 1993a (now also included in Montanari 1995a).

  Aristarchus in the D-Scholia to the Iliad fragments, the D-Scholia themselves, and the Lexeis Homerikai); for this reason the edition must comprise all three elements, each equipped with an adequate apparatus of testimonia and loci similes and with the appropriate crossreferences, in accordance with the structure outlined above. In anticipation of the new edition (which will certainly occupy me for several more years), I present here a small selection of items: specifically, the text of those D-scholia in which Aristarchus is mentioned. The texts are without apparatus of testimonia and apparatus of variant readings, and they are in a form that must be considered provisional (even if I am fairly confident that any differences between the texts presented here and those of the definitive edition will be few and insignificant). Following Erbse’s practice, I list the sigla of the relevant manuscripts at the end of each scholion (or, when appropriate, at the end of each portion of the scholion they preserve). In addition to the medieval manuscripts, I also list (for now, at least) the editio princeps of J. Lascaris.9 A T C H P R V

Venetus Graec. 822 (olim Marc. gr. 454), Xth cent. Townleianus, Brit. Mus. Burney 88, a.D. 1014 aut 1059 Rom. Bibl. Naz. gr. 6 + Matrit. 4626 (71 Ir.), IXex/Xin cent. Vat. gr. 2193, XIIth cent. Paris. Suppl. Gr. 679, ff. 1–20, Xth cent. Vat. gr. 32, XIIth cent. Vat. gr. 33, XIex cent.

La J. Lascaris, editio princeps, Romae 1517 A 5 Δ ι ὸ ς δ ὲ τ ε λ ε ί ε τ ο β ο υ λ ή : Διὸς βουλὴν, οἱ μὲν τὴν εἱμαρμένην ἀπέδοσαν. ἄλλοι δὲ ἐξεδέξαντο, δρῦν ἱερὰν μαντικὴν τοῦ Διὸς ἐν Δωδωναίῳ ὄρει τῆς Θεσπρωτίας, ὡς αὐτὸς Ὅμηρος λέγει ἐν Ὀδυσσείᾳ· «τὸν δ’ ἐς Δωδώνην φάτο βήμεναι, ὄφρα θεοῖο / ἐκ δρυὸς ὑψικόμοιο Διὸς βουλὴν ἐπακούσοι» (ξ 327–328). ἄλλοι δὲ ἀπὸ ἱστορίας τινὸς εἶπον εἰρηκέναι τὸν Ὅμηρον· φασὶ γὰρ τὴν γῆν βαρουμένην ὑπὸ ἀνθρώπων πολυπληθείας, μηδεμιᾶς ἀνθρώπων οὔσης εὐσεβείας, αἰτῆσαι τὸν Δία κουφισθῆναι τοῦ ἄχθους· τὸν δὲ Δία πρῶτον μὲν εὐθὺς ποιῆσαι τὸν Θηβαϊκὸν πόλεμον, δι’ οὗ πολλοὺς πάνυ ἀπώλεσεν· ὕστερον δὲ πάλιν, συμβούλῳ τῷ Μώμῳ χρησάμενος, ἣν Διὸς βουλὴν Ὅμηρός φησιν. ἐπειδὴ οἷός τε ἦν κεραυνοῖς ἢ κατακλυσμοῖς ἅπαντας διαφθείρειν. ὅπερ τοῦ Μώμου κωλύσαντος, ὑποθεμέ-

 9 The stemma codicum can be found in several of the works cited above in footnotes 1 and 5. For the presence of the D-Scholia in the MSS A and T, cf. Erbse 1969–1988, vol. I, Praefatio, LXXIVf. The MS P = Paris. Suppl. Gr. 679, ff. 1–20, probably to be dated to the 10th century, presents the scholia from Π 140 to Σ 593 (with a few gaps); its position in the stemma codicum remains to be determined (cf. Montanari 1994b [= this volume, ch. 34]), but, in any case, it does not contain any of the scholia that interest us here.

Aristarchus in the D-Scholia to the Iliad  

νου δὲ αὐτῷ γνώμας δύο, τὴν Θέτιδος θνητογαμίαν καὶ θυγατρὸς καλῆς γένναν, ἐξ ὧν ἀμφοτέρων πόλεμος Ἕλλησί τε καὶ Βαρβάροις ἐγένετο, ἀφ᾽ οὗ συνέβη κουφισθῆναι τὴν γῆν πολλῶν ἀναιρεθέντων. ἡ δὲ ἱστορία παρὰ Στασίνῳ τῷ τὰ Κύπρια πεποιηκότι, εἰπόντι οὕτως· «ἦν ὅτε μυρία φῦλα κατὰ χθόνα †πλαζόμενα† / βαθυστέρνου πλάτος αἴης· / Ζεὺς δὲ ἰδὼν ἐλέησε καὶ ἐν πυκιναῖς πραπίδεσσι / †σύνθετο κουφίσαι παμβώτορα γαίης ἀνθρώπων†, / ῥιπίσσας πολέμου μεγάλην ἔριν Ἰλιακοῖο, / ὄφρα κενώσειεν θανάτῳ βάρος. οἱ δ’ ἐνὶ Τροίῃ / ἥρωες κτείνοντο, Διὸς δ᾽ ἐτελείετο βουλή» (Cypr. fr. 1). καὶ τὰ μὲν παρὰ τοῖς νεωτέροις ἱστορούμενα περὶ τῆς τοῦ Διὸς βουλῆς ἐστὶ τάδε. ἡμεῖς δέ φαμεν κατὰ τὴν Ἀριστάρχειον καὶ Ἀριστοφάνους δόξαν τῆς Θέτιδος εἶναι βουλὴν, ἣν ἐν τοῖς ἑξῆς φησι λιτανεύουσαν τὸν Δία ἐκδικῆσαι τὴν τοῦ παιδὸς ἀτιμίαν, καθάπερ καὶ τὰ κεφάλαια ἐν τῷ προοιμίῳ κεῖται τῆς ποιήσεως. ACHVLa A 591 ἀ π ὸ β η λ ο ῦ : ἀπὸ τοῦ οὐρανοῦ. CHRVLa ἀπὸ γὰρ τοῦ οὐρανοῦ ἐρρίφη. RVLa ἀπὸ γὰρ τοῦ βεβηκέναι τοὺς θεοὺς ἐπ’ αὐτῷ καὶ τοὺς ἀστέρας κεῖθι βαίνειν εἴρηται. τῷ δὲ τόνῳ καθάπερ οἱ Ἀριστάρχειοι βηλὸν ὡς χωλὸν προφερόμενοι, ἀποδιδόντες τὸν τῶν θεῶν βαθμόν. ἕτεροι δὲ βηλὸν εἶπον τὸν ἀνωτάτω πάγον καὶ περιέχοντα τὸν πάντα ἀέρα· ἄλλοι δὲ τὴν περίοδον τοῦ αἰθέρος καὶ τῶν ἄστρων. κοινῶς δὲ σημαίνει ὁ βηλὸς τὸν οὐδὸν τῆς θύρας, ὃν καλοῦσι φλιάν· ὁ γὰρ ποιητὴς ὑποτίθεται τὸν οὐρανὸν πύλας ἔχοντα. εἴρηται δὲ βηλὸς ἀπὸ τοῦ βαίνεσθαι, ὡς καὶ ὁδὸς ἀπὸ τοῦ διοδεύεσθαι. ACHVLa καὶ ὁ Πανύασις (fr. 23 Matth.) δὲ τὰ πέδιλα †βιολα† λέγει. CH B 316 π τ έ ρ υ γ ο ς : παροξυτόνως. καὶ ὁ μὲν κανὼν θέλει προπαροξυτόνως ὡς ὄρτυγος· ἀλλ’ ἐπειδὴ οὕτως δοκεῖ †στίζειν† τῷ Ἀριστάρχῳ, πειθόμεθα αὐτῷ ὡς πάνυ ἀρίστῳ γραμματικῷ. ATCHRVLa B 494 B ο ι ω τ ῶ ν μ ὲ ν Π η ν έ λ ε ω ς κ α ὶ Λ ή ϊ τ ο ς ἦ ρ χ ο ν : ἄρξασθαι φασί τινες ἀπὸ Βοιωτῶν τὸν Ὅμηρον τοῦ καταλόγου εἰς κεχαρισμένον τῶν Μουσῶν ὧν ἐπεκαλέσατο· αὐτόθι γὰρ ἐν Βοιωτίᾳ εἶναι τὸν Ἑλικῶνα, τὸ ὄρος ὅπου σύνηθες αὐταῖς διατρίβειν. ἀντιλέγουσι δέ τινες τούτοις, φάσκοντες οὐ τῆς Βοιωτίας ἀλλὰ τῆς Φωκίδος εἶναι τὸν Ἑλικῶνα· μὴ οὖν ἀπὸ Βοιωτῶν τὴν ἀρχὴν διὰ τοῦτο γεγενῆσθαι. τινὲς δὲ ἀνθ’ ὧν οὗτοι μόνοι πέντε εἶχον ἡγεμόνας. ἄλλοι δέ φασιν οὐ διὰ τοῦτο, ἀλλ’ ἐπεὶ πολλὰς πόλεις ἔχει ἡ Βοιωτία, διὰ τοῦτο ἐντεῦθεν ἄρξασθαι αὐτὸν τοῦ καταλόγου. οὐκ ἔστιν δὲ οὐδὲ τοῦτο ὑγιές· οὐ γὰρ ἀπὸ πλήθους πόλεων ὁ κατάλογος παραμεμέτρηται· ἤρξατο γὰρ ἂν μᾶλλον ἀπὸ Κρητῶν ἐχόντων ἑκατὸν πόλεις. βέλτιον δὲ λέγειν αὐτὸν ἀπὸ Βοιωτῶν ἦρχθαι ἐπειδήπερ ἐν Αὐλίδι πόλει τῆς Βοιωτίας συνήχθη ἅπαν τὸ πλῆθος τῶν ἐπὶ τὴν Ἴλιον μελλόντων στρατεύειν· ἔχει δὲ αὕτη ἡ πόλις λιμένας, τοὺς μὲν βλέποντας ἐπὶ τὴν Ἴλιον, τοὺς δὲ καταγωγοὺς ἀπὸ τῆς Ἑλλάδος ἀπάσης, δι’ ἣν αἰτίαν συνηθροίσθησαν αὐτόθι καὶ ἐντεῦθεν ἀπέπλευσαν. ὁ δὲ Ἀρίσταρχός φησι κατὰ ἐπιφορὰν αὐτὸν ἀπὸ Βοιωτῶν τὴν ἀρχὴν πεποιῆσθαι· εἰ γὰρ καὶ ἀπ’ ἄλλου ἔθνους ἤρξατο, ἐζητοῦμεν ἂν τὴν αἰτίαν τῆς ἀρχῆς. ACHVLa E 385 τ λ ῆ μ ὲ ν Ἄ ρ η ς : Ἀρίσταρχος ἀξιοῖ τὰ φραζόμενα ὑπὸ τοῦ ποιητοῦ μυθικώτερον ἐκδέχεσθαι κατὰ τὴν ποιητικὴν ἐξουσίαν, μηδὲν ἔξω τῶν φραζομένων ὑπὸ τοῦ ποιητοῦ περιεργαζομένους. Ὦτος οὖν καὶ Ἐφιάλτης γόνῳ μὲν ἦσαν Ποσειδῶνος, ἐπίκλησιν δὲ Ἀλωέως. οὗτοι ὑπερφυεῖς κατὰ μέγεθος γενόμενοι — ἄλλοι δὲ, πιθανώτερον βουλόμενοι ἐξηγεῖσθαι τὸν Ὅμηρον, τοὺς Ἀλωϊάδας τούτους φασίν γενέσθαι βασιλεῖς τῆς Ἑλλάδος— οἱ δέ φασιν Ἄρη περὶ τῆς ἱστορίας τῶν ἄστρων καὶ περὶ τῶν μετεώρων διαλέγεσθαι τὸν

  Aristarchus in the D-Scholia to the Iliad Ὅμηρον — βέλτιον δὲ ὅτι φιλοσοφεῖν βούλεται διὰ τῆσδε ῥαψῳδίας ὡς καὶ δι’ ὅλης τῆς ποιήσεως — ἵνα μὴ ὑπὸ τῶν πολεμίων ληφθέντες κακῶς πείσονται. CHRVLa10 E 397 ἐ ν Π ύ λ ῳ ἐ ν ν ε κ ύ ε σ σ ι : ὁ μὲν Ἀρίσταρχος τὸν πύλον ἀκούει κατὰ μεταπλασμὸν ἀντὶ τοῦ, πύλην, ὡς ἀπὸ τοῦ πολύαρνι πολυάρνῳ· καὶ τὸ ‛ἐν νεκύεσσι’ κυρίως ἀκούει, ἔνθα αἱ τῶν τελευτηκώτων ἀναχωροῦσι ψυχαί. CHRVLa καὶ τὸν Ἀϊδωνέα [τρωθῆναι] ὑπὸ Ἡρακλέους εἰκὸς περὶ τὴν τοῦ Λερβέρου ἀναγωγὴν τετοξεῦσθαι, ἐμποδὼν γιγνόμενον αὐτῷ καὶ κωλύοντα, ἐπειδὴ καὶ ἐν ἑτέροις τῆς ἱστορίας ταύτης μέμνηται ὁ ποιητὴς λέγων· «εὗτε μιν εἰς Ἀΐδαο πυλάρταο προὔπεμψεν / ἐξ Ἐρέβευς ἄξοντα κυνὰ στυγεροῦ Ἀΐδαο» (Θ 367– 368). Οἱ δὲ νεώτεροι Ὁμήρου ποιηταὶ τὸ ‛ἐν Πύλῳ’ κυρίως περὶ Πύλον τὸν Μεσσηνιακὸν ἤκουσαν τοῦτο πεπονθέναι αὐτόν, καθ᾽ ὅν χρόνον Ἡρακλῆς ἐπόρθησε Πύλον καὶ προσεμυθεύσαντο ἴδιον διήγημα, ὅπερ ἀνωτέρω εἴπομεν ἐπὶ τῆς Ἥρας ἐν τῇ πρὸ ταύτης ἱστορίᾳ. C E 422 Κ ύ π ρ ι ς : ἡ Ἀφροδίτη. C τὸ ἐπίθετον Ἀφροδίτης, ὅ οὐκ ἐνόησαν οἱ πρὸ ἡμῶν, τί σημαίνει. συμπλανηθέντες τῷ Ἡσιόδῳ ἔδοξαν, ὅτι Κύπρις λέγεται, ὥς φησιν Ἡσίοδος, Κυπρογένεια διότι γεννᾶται «περικλύστῳ ἐνὶ Κύπρῳ» (Th. 199), ὥσπερ καὶ τὴν φιλομειδῆ, ὅτι «μηδέων ἐξεφαάνθη» (Th. 200). Ὅμηρος δὲ οὐκ εἶπεν . ἀλλὰ τὴν μειδιάματα φιλοῦσαν, οἷον ἱλαρὰν, διὰ τὴν ἐγκειμένην αὐτῇ δύναμιν ἀπὸ τῆς συνουσίας. HRVLa ὥσπερ οὗν τὸ πῦρ Ἥφαιστον λέγει (Β 426) —— Πυθοῖ ἐνὶ πετρηέσσῃ (I 404). HVLa οὐδ᾽ ἐπεὶ φησιν· «ἵκετο — καὶ γὰρ εἰ σπανίως, Ἑλικώνιον τὸν Ποσειδῶνα εἴρηκεν (Υ 404) ἀπὸ Ἑλικῶνος, ὡς ὁ Ἀρίσταρχος βούλεται, ἐπεὶ ἡ Βοιωτία ὅλη ἱερὰ Ποσειδῶνος (οὐ γὰρ ἀρέσκει ἀπὸ Ἑλίκης, ἐπεί φησιν·«οἱ δέ τοι εἰς Ἑλίκην τε καὶ Αἰγὰς δῶρ’ ἀνάγουσιν» (Θ 203)· Ἑλικήϊον γὰρ ἂν εἶπε, συγχωροῦντος τοῦ μέτρου), δύναται δὲ Ἑλικώνιος λέγεσθαι διὰ τὸ ἕλικας καὶ περιφερεῖς εἶναι τὰς δίνας τῆς θαλάσσης — ἀπὸ τῆς Κύπρου τὸ ἐπίθετον ἐξ ἰδίου προσώπου. λέγων·«ὁ δὲ Κύπριν ἐπῴχετο νηλέϊ χαλκῷ» (Ε 330). VLa11 I 395 ἀ ν ’ Ἑ λ λ ά δ α τ ε Φ θ ί η ν τ ε : Ἀρίσταρχος παρ’ Ὁμήρῳ Θεσσαλίαν μόνην τὴν Ἑλλάδα φησὶν εἶναι καὶ Ἕλληνας τοὺς Θετταλοὺς, παρακολουθῶν Θουκυδίδῃ (Ι 3, 3) καὶ ἀθετεῖ τὸν στίχον τοῦτον «ἐγχείῃ δ᾽ ἐκέκαστο Πανέλληνας καὶ Ἀχαιούς» (Β 530) καὶ ἐκείνῳ δὲ τόπῳ πάλιν ἀκούει «φεῦγον ἔπειτ’ ἀπάνευθε δι’ Ἑλλάδος εὐρυχόροιο» (Ι 478). ACHRVLa K 282 μ έ γ α ἔ ρ γ ο ν : μεγάλην πρᾶξιν. CHRVLa Ἀρίσταρχος δέ φησι τὸ φονεῦσαι Ἕκτορα. Ail CHRVLa K 351–354 : Ἀρίσταρχος οὕτως ἐξηγήσατο· ὅσον, φησὶν, ὑφ’ ἓν καὶ ὁμοῦ ὑπό τινος ἀφεθέντος καὶ ἀπολυθέντος

 10 It did not seem necessary to reproduce here the entirety of this long scholion, whose content has frequently been the subject of discussion: cf. Pfeiffer 1973, 351 ff.; Meijering 1987, 65; Porter 1992, 70. 11 In this case too it did not seem necessary to reproduce the entirety of the scholion, which contains a lengthy discussion regarding the epithets of the Homeric divinities (for which reason some have suspected that it is derived from the work Περὶ θεῶν by Apollodorus: cf. FGrHist 244 F 353).

Aristarchus in the D-Scholia to the Iliad  

ζεύγους ἡμιόνων καὶ ζεύγους ἄλλου βοῶν φθάσωσι καὶ προλάβωσι τοὺς βοῦς αἱ ἡμίονοι (ταχύτεραι γάρ εἰσι τῶν βοῶν), τοσοῦτον, φήσιν, ἐάσαντες διάστημα παρελθεῖν τὸν Δόλωνα, μεταστραφέντες ἐδίωξαν. TRVLa O 741 o ὐ μ ε ι λ ι χ ί η π ο λ έ μ ο ι ο : Ἀρίσταρχος κατὰ δοτικήν. ἵν’ ᾖ ὁ λόγος· φῶς ἡμῖν ἐν ταῖς χερσίν ἐστιν, οὐκ ἐν τῇ μειλιχίᾳ τοῦ πολέμου. ὁ δὲ Θρᾷξ (fr. 17 Linke) φησὶ οὐκ ἔστι προσήνεια τοῦ πολέμου, ἀλλ’ εἰρήνης. VLa Y 53 θ έ ω ν ἐ π ὶ Κ α λ λ ι κ ο λ ώ ν ῃ : ὅ ἐστι τρέχων. Ἀρίσταρχος ἀνέγνω περισπωμένως, τὸ χωρίον θεῶν λέγων. ἡ δὲ Καλλικολώνη ἐστὶν ὑψηλὸς τόπος. HRVLa Υ 357 τ ό σ σ ο υ ς δ ’ ἀ ν θ ρ ώ π ο υ ς : ὁ Ἀσκαλωνίτης (p. 59 = 193 Baege) προπερισπᾷ κατὰ δεῖξιν, ὡς «τοσσῆς δ’ ὑσμίνης» (Y 359). ὁ δὲ Ἀρίσταρχος κατὰ παρολκὴν ἐδέξατο τόν ‘δέ’. RVLa Φ 319 χ ε ρ ά δ ο ς : τῆς ὑπὸ ποταμῶν ἀθροιζομένης ψάμμου καὶ ξύλων συγκομιδῆς. CHRVLa Ἀρίσταρχος δέ φησι, χεράδας καλεῖσθαι τοὺς ποταμοὺς, καὶ τοὺς ἐν αὐτοῖς λίθους· εἶναι δὲ τὸν νοῦν τοιοῦτον· πολλῇ ἰλύi οὐκ ὀλίγον πλῆθος τῶν ποταμίων λίθων ἐπισωρεύσας χώσω, χεράδος οὖν, λιθάδος. ACHRVLa Φ 323 Τ Υ Μ Β Ο Χ Ο Η Σ : χωστοῦ τάφου· τύμβος γὰρ ὁ τάφος. CHRVLa τὸ πλῆρες τυμβοχοῆσαι. HRVLa οὕτως Ἀρίσταρχος καὶ Ἀσκαλωνίτης (p. 60 = 194 Baege) καὶ οἱ πλείους· ἄμεινον γὰρ ἐν τῷ ῥήματι τὸ πρᾶγμα λαμβάνειν ἢ ἐν τῷ ὀνόματι. HVLa X 491 ὑ π ε μ μ ή μ υ κ ε : καταμέμυκε, κατεστύγνακε, κάτω βλέπει. ὁ δὲ Ἀρίσταρχος ἐπὶ τοῦ κατανένευκεν ἐκδέχεται. οἱ δὲ, ἀντὶ τοῦ εἰς ἀνάμνησιν ἔρχεται τοῦ πατρός. ACHRVLa

 The Wrong Verses of Homer and Ancient Philology In studies on the history of Alexandrian philology, topics related to the metre and prosody of the dactylic hexameter are rarely discussed or investigated, despite the fact that epic-hexametric poetry is the field in which HellenisticRoman grammarians were most active and for which we have the richest testimonies. What survives of the copious ancient exegesis of Homer, however, includes a significant range of testimonies about prosodic and metrical issues: at the end of the 19th century, this material from the Homeric scholia and Eustathius was brought together into a collection, without any claim to exhaustiveness or detailed scrutiny.1 Since then, however, to the best of my knowledge this topic has never been addressed and studied, in order to set it within the historical framework of the history of scholarship. We can begin our analysis with verse Χ 379 and the related scholium by Aristonicus. Χ 379 ἐπεὶ δὴ τόνδ’ ἄνδρα θεοὶ δαμάσασθαι ἔδωκαν Aristonicus in sch. Χ 379a: ὅτι τὴν αἰτίαν προτέταχεν, “ἐπειδὴ τόνδ’ ἄνδρα θεοί (Χ 379)”. ἔδει δὲ οὕτως· “εἰ δ’ ἄγετ’, ἀμφὶ πόλιν σὺν τεύχεσι πειρηθῶμεν, / ὄφρα κ’ ἔτι γνῶμεν Τρώων νόον, / ἢ καταλείψουσι πόλιν ἄκρην τοῦδε πεσόντος, / ἦε μένειν μεμάασιν, / ἐπειδὴ τόνδ’ ἄνδρα θεοὶ δαμάσασθαι ἔδωκαν” (Χ 381–4, 379). ὁ δὲ Διονύσιος (fr. *13 Linke) διστάζει, μὴ πρὸς τὴν ἀπόλειψιν τοῦ χρόνου· παραβάλλει γὰρ τὸ “ἐπειδὴ νῆάς τε καὶ Ἑλλήσποντον ἵκοντο” (Ψ 2) καὶ “ἐπίτονος βέβλητο” (μ 423)· τὰ γὰρ τοιαῦτα ἐσημειοῦντο πρὸς κρίσιν ποιημάτων, ὅτι σπανίως Ὅμηρος κακομέτρους ποιεῖ.2

Aristonicus’ scholium first of all raises the issue of the order in which the poet expresses his thoughts in vv. 379–384. It notes that in v. 379 (scil. + 380) he anticipates the motif of the exhortation which Achilles will address to his companions in vv. 381–384. Subsequently (ὁ δὲ Διονύσιος etc.) the scholium dwells on a metrical issue pertaining precisely to v. 379, which is what specifically interests us here. As he usually does, Aristonicus explains an Aristarchean σημεῖον, and more precisely a diple (ὅτι, scil. ἡ διπλὴ ὅτι), which is also preserved — as is often the case — in the well-known codex Ven. A. The scholium presents a number of difficulties: not least the fact — attested elsewhere, as we

 1 Rauscher 1886; Grossmann 1887. 2 Friedlaender 1853, 323 f. https://doi.org/10.1515/9783110772371-036

The Wrong Verses of Homer and Ancient Philology  

shall see — that it combines two separate annotations, a stylistic annotation and a metrical-prosodic one.3 According to the first part of the commentary, the natural and ‘normal’ order of the poet’s thoughts would feature first an exhortation and then the reasons for it, which is to say: ‘come, let us make trial in arms about the city … seeing the gods have vouchsafed us to slay this man’, and not ‘seeing the gods have vouchsafed us to slay this man … come, let us make trial in arms about the city.’ At first sight, this observation might seem to pertain to the issue of the different forms of ἀν(τ)εστραμμένη περίοδος, which Nicanor discusses in relation to matters of punctuation, for the usual purpose of bringing out the underlying structure of the poetic phrasing as clearly as possible and elucidating its meaning.4 In this case, however, an anomaly was clearly detected by Friedlaender, and it is worth quoting his words in the book on Nicanor in full:5 Etenim qui ordo in rerum natura obtinet, ut causa antecedat effectum, eum etiam in elocutione naturalem esse censuit Nicanor; unde si sententiae primariae postponitur ea quae a conjunctione causali (ὅτι, ἐπεί) incipit, hoc genus ad inversas periodos retulit. Miramur igitur Aristonicum quum Aristarchus diple notavisset Χ 379 … hanc notam ideo positam esse sibi persuasit ὅτι τὴν αἰτίαν προτέταχεν, ἐπειδὴ τόνδ’ ἄνδρα θεοί. ἔδει δὲ οὕτως· εἰ δ’ ἄγετ’, ἀμφὶ πόλιν — πειρηθῶμεν — ἐπειδὴ τόνδ’ ἄνδρα θεοὶ δαμάσασθαι ἔδωκαν. Hunc si audis Aristarchus non modo a Nicanor dissensit, ex cujus sententia hic locus recto ordine enuntiatus est, sed etiam ab Apollonio (conj. 505,22 – 506,5). Quod ut fieri potuerit: certe cur hoc tam usitatum dicendi genus diple notaret nulla causa erat. Unde propius ad verum accessisse Dionysium puto, qui διστάζει, μὴ πρὸς τὴν ἀπόλειψιν τοῦ χρόνου (in ἐπεί) sc. ᾖ ἡ διπλῆ.6 Aristonicus autem ni fallor hoc sententiarum genus cum eo confudit, in quo membra γάρ habentia praecedunt sententiam primariam, quod Aristarchus attendisse scimus (Lehrs p. 9) neque Nicanorm neglexisse videmus B805.7

In other words, the scholium’s first explanation for Aristarchus’ diple raises numerous doubts. In fact, it is wholly unconvincing: this case does not at all fit with the type of ἀν(τ)εστραμμέναι περίοδοι and it is truly difficult to see “cur hoc tam usitatum dicendi genus diple notaret”. Indeed, Friedlaender deems far more convincing the explanation provided by Dionysius, who is quoted immediately afterwards, and who has been identified as Dionysius Thrax.8 Before  3 See Erbse 1969–87, testimonia ad loc.; cf. Linke 1977, fr. *13, with a commentary. 4 Friedlaender 1850, esp. 71 ff.; Blank 1983, esp. 51; Sluiter 1990, 68 and n. 117. 5 Friedlaender 1850, 73 f. 6 Cf. infra n. 9. 7 See also Friedlaender 1853, 324: “Hanc notam non credo ex Aristarchi mente illustratam esse ab Aristonicus, qua de re dixi in proll. ad Nicanorem…” 8 See Linke 1977, fr. *13.

  The Wrong Verses of Homer and Ancient Philology analysing this second part of the scholium, though, we must dwell on the verb διστάζει that introduces it. Διστάζω is hardly one of the more frequent verbs in the technical terminology of ancient philology. In Homeric scholiography, I have found 27 occurrences of διστάζω and three of ἐπιδιστάζω: in 21 cases διστάζω (like ἐπιδιστάζω in all three of its occurrences) is used to describe the behaviour or attitude of a character who ‘hesitates, is uncertain, finds himself in doubt’; in three cases the verb has the poet as its subject; in four, we instead find an explicitly philological use of διστάζω: out of these four occurrences, only one comes from the Iliad, i.e. from our sch. Χ 379a, while three are from Odyssey. Sch. HPQ ε 337: οὐκ ἐφέρετο ἐν τοῖς πλείοσι. Ἀρίσταρχος δὲ περὶ μὲν τῆς ἀθετήσεως διστάζει κτλ. (Carnuth, Arist. 1869, p. 60; Ludwich, AHT 1884–5, I p. 557). Sch. HQ ζ 244–45: ἄμφω μὲν ἀθετεῖ Ἀρίσταρχος, διστάζει δὲ περὶ τοῦ πρώτου, ἐπεὶ καὶ Ἀλκμὰν κτλ. (Carnuth, Arist. 1869, p. 68; Ludwich, AHT 1884–5, I p. 562). Sch. P η 311: τοὺς ἓξ (311–16) Ἀρίσταρχος διστάζει Ὁμήρου εἶναι. εἰ δὲ καὶ Ὁμηρικοί, εἰκότως αὐτοὺς περιαιρεθῆναί φησι. πῶς γὰρ κτλ. (Carnuth, Arist. 1869, p. 76; Ludwich, AHT 1884–5, I p. 568).

In these three scholia, the verb διστάζω has Aristarchus as its subject, and it is used to express doubts about the wording of the text, with regard to certain instances of athetesis. More specifically still, in the first two cases, Aristarchus ‘is uncertain, doubtful’ whether to athetise or not; in the third, he ‘doubts, wonders’ whether the six verses in question are by Homer. In sch. Χ 379a, according to Friedlaender’s interpretation (shared by Erbse),9 Dionysius (Thrax) “wonders whether (the diple) may not (have been attached to the verse) in relation to the lack of (a) tense”, scil. at the beginning of the verse: for in Χ 379 the first foot (ἐπεί) inevitably has a iambic structure (˘ ¯), which is to say that it lacks a mora.10 What we have here is the well-known issue of ἀκέφαλοι hexameters, for which the scholium offers two perfectly fitting parallels: Ψ 2, beginning with ˘ ¯, and μ 423, beginning with ˘ ˘ ˘. It thus seems as though we are dealing with a passage in which the interpretation of the Aristarchean diple raised certain doubts, prompting further reflections. While this is not a particularly common situation, certain parallels for it may be found: evidently, in some cases Aristarchus’ reasons and comments were not entirely available or accessible. Particularly reminiscent of our case, Χ 379a, are some scholia in which we again find Aristonicus first offering

 9 Erbse 1969–1987, app. ad loc.: “χρόνου, sc. τὸ σημεῖον ᾖ”. 10 See Richardson 1993, ad loc.

The Wrong Verses of Homer and Ancient Philology  

an explanation for an Aristarchean diple and then reporting a different one from Dionysius Thrax. The clearest example is Aristarchus’ scholium Ο 86a: Ἀρίσταρχος σημειοῦται ὅτι ... ὁ δὲ Θρᾷξ Διονύσιος (fr. 16 Linke) παρακεῖσθαί φησι τὸ σημεῖον ὅτι κτλ. I should stress that this testimony, with the complete indication ὁ δὲ Θρᾷξ Διονύσιος, confirms that the Διονύσιος quoted in the group of similar scholia we are considering is precisely Dionysius Thrax. The same situation is found in sch. Ρ 125a, Aristonicus: ὅτι ἔοικε ὁ Ἀρίσταρχος ... ὁ δὲ Διονύσιος (fr. *12 Linke) τὸ σημεῖόν φησιν, ὅτι κτλ. Other similar cases are: sch. Μ 301a, Aristonicus: ὅτι ... καὶ ὅτι ... ὁ δὲ Διονύσιος (fr. *38 Linke) ὅτι κτλ.; sch. Π 810a, Aristonicus: ὅτι ... ὁ δὲ Διονύσιος (fr. *40 Linke) οὐκ ἐπὶ μελέτης, ἀλλ’ ὅτι κτλ.; sch. Τ 49a, Aristonicus: ὅτι ... ὁ δὲ Διονύσιος (fr. *43 Linke) τὸ σημεῖόν φησιν ὅτι κτλ. It should further be added that in sch. Ρ 24a Aristonicus only reports Dionysius Thrax’s interpretation: τὸ σημεῖον Διονύσιος (fr. *41 Linke) τὸν Ὑπερήνορά φησι κτλ.11 Aristonicus therefore drew upon Dionysius Thrax’s works on Homeric philology and reported his opinions, which sometimes differed from his own12 (it is also quite likely that our scholia only record some of the cases in which this occurred). Dionysius Thrax, a direct pupil of Aristarchus’, more or less occasionally provided, presumably in a Homeric commentary of his, explanations and interpretations of Aristarchean semeia. This evidently led him to reason about his teacher’s choices and views. Dionysius Thrax’s information about Aristarchus’ opinions has often been preserved by Aristonicus, as is the case in sch. Χ 379a. Here, as we have seen, Aristonicus reports that “Dionysius wonders whether (the diple) may not (have been attached to the verse) in relation to the lack of (a) tense”, scil. at the beginning of the verse, and invokes Ψ 2 and μ 423 for the sake of comparison. He then concludes: τὰ γὰρ τοιαῦτα ἐσημειοῦντο πρὸς κρίσιν ποιημάτων, ὅτι σπανίως Ὅμηρος κακομέτρους ποιεῖ. This metrical observation is taken up in sch. Χ 379b (a scholium from class D, also present in manuscript T): σημειῶδες ὅτι καὶ (σημειωτέον ὅτι D) ἀκεφάλοις χρῆται ὁ ποιητής, ὥστε οὐ δεῖ προπερισπᾶν (περισπᾶν D) τὸ “φίλε κασίγνητε” (Φ 308), where we also find one of the few occurrences in Homeric scholia of the term ἀκέφαλος, which does not occur in the Aristonicus scholium. Which was the true meaning of the Aristarchean diple in Χ 379 — that suggested by Aristonicus or that supported by Dionysius Thrax? As we have seen, it  11 See also the scholia to O 571: the comparison with Aristonicus in sch. Ο 571c suggests that sch. ex. Ο 571b also contains material which can be traced back to Aristonicus (Dionysius Thrax fr. *11 Linke). 12 Valk 1963–1964, I, 561–562, 565, 590; Linke, comm. on the quoted frr.

  The Wrong Verses of Homer and Ancient Philology is possible that Dionysius Thrax is providing the true interpretation of the diple, since Aristonicus’ interpretation raises quite a few puzzles.13 However, a further remark must be made with regard to those cases in which different explanations are provided for the same mark: it cannot be ruled out that a diple might have more than one meaning, i.e. that it might flag not just one but two arguments worth noting and commenting on in a verse. For example, in sch. Ω 17a Aristonicus himself provides two reasons: ὅτι ... καὶ ὅτι κτλ.; cf. the aforementioned sch. Μ 301a, Aristonicus: ὅτι ... καὶ ὅτι ... ὁ δὲ Διονύσιος κτλ. Therefore, two different, yet not contradictory, explanations of a diple do not necessarily rule each other out, since the marked verse could easily offer two opportunities for commentary. In addition, it is worth recalling certain insightful remarks by Lehrs:14 in relation to some Homeric passages Aristarchus may have failed to notice a phenomenon that he would usually detect and discuss; or he may also have consciously overlooked certain occurrences of a phenomenon already extensively attested and discussed; or, finally, an Aristarchean semeion for a verse worth noting and commenting may sometimes have been lost. In such cases, the commentator’s pupils would add and integrate marks and comments over the verses, according to the guidelines provided by their teacher. This, in a way, would easily explain the addition by a pupil, such as Dionysius Thrax, of observations in line with Aristarchus’ ideas and annotations. Aristarchus might only have associated the diple in Χ 379 with the problem indicated by Aristonicus (namely, that of the order of the poet’s thoughts), or with both this and the metrical problem discussed by Dionysius Thrax, or again with the metrical problem alone. Be that as it may, even assuming that Aristarchus said nothing about the acephaly of verse Χ 379 and that only Dionysius Thrax traced the diple back to this metrical problem, or even that he personally added the mark for this reason, sch. Χ 379a retains all its value as a testimony: for Dionysius expressly states that verses of this sort were noted and annotated, meaning — in this specific case — verses lacking a mora in the first foot, i.e. acephalous ones. He states that observations of this sort were made quite regularly: τὰ γὰρ τοιαῦτα ἐσημειοῦντο πρὸς κρίσιν ποιημάτων, ὅτι σπανίως Ὅμηρος κακομέτρους ποιεῖ.15 It is an established fact, then, that Aristarchus and his pupils noted and commented on metrical anomalies of this sort.

 13 This is far from being a general rule: Valk 1963–4, I, 561–562, has examined certain instances of divergence between Aristonicus and Dionysius Thrax. 14 See Lehrs 1882, 13 f. 15 Lehrs 1882, 14, had doubts about ἐσημειοῦντο: “Aut scripsit ἐσημειοῦτο aut Aristarchi aequales discipulos vel contubernales comprehendit”; cf. Erbse 1969, app. ad loc.

The Wrong Verses of Homer and Ancient Philology  

Verses described as ἀκέφαλοι (i.e. missing a mora in the first foot), as λαγαροί (i.e. missing a mora in the middle section of the verse, often in the fourth foot), or as μείουροι or μύουροι (i.e. missing a mora in the last foot) are sometimes flagged in Homeric scholia and other sources. This amounts to the simple recording of anomalous phenomena in relation to the metrical structure of certain hexameters, for which no explanation or theoretical interpretation is provided.16 Such matters must not have elicited a great deal of interest among epitomisers and the compilers of scholiographical corpora, given that observations of this sort are not particularly common in surviving scholia, although — as we shall see — it is quite plausible that they were originally far more numerous, given the frequency of such phenomena. Let us consider the two verses compared by Dionysius Thrax in sch. Χ 379: in the case of Ψ 2, ἐπεὶ δὴ νῆάς τε καὶ Ἑλλήσποντον ἵκοντο, we only find a mention in sch. D: ἀκέφαλον καλεῖται τοῦτο τὸ μέτρον· πέπονθε γὰρ κατ’ ἀρχὰς ὁ στίχος, which led Erbse to note ad loc.: “fort. erat sch. de versu acephalo, sc. de prosodia vocum ἐπεὶ δή initio versus positarum”; in the case of μ 423, ἐπίτονος βέβλητο, βοὸς ῥινοῖο τετευχώς, no mention at all of the metrical problem is found in surviving scholia.17 A valuable testimony is provided by sch. ex. to verse Μ 208: Μ 208 Τρῶες δ᾽ ἐρρίγησαν ὅπως ἴδον αἰόλον ὄφιν Sch. ex. Μ 208c: μείουρος ὁ στίχος. ἀκέφαλος “ἐπειδὴ νῆάς τε” (Ψ 2), λαγαρὸς “τῶν αὖθ’ ἡγείσθην Ἀσκληπιοῦ δύο παῖδε” (Β 731). οἱ δὲ “ὄφιν” φασίν· ἐμφαντικώτερον δὲ ἐχρήσατο τῇ τοῦ στίχου συνθέσει, καίτοι γε ἐγχωροῦν εἰπεῖν “ὅπως ὄφιν αἰόλον εἶδον” κτλ. Sch. D: πέπονθε δὲ ἐπὶ τέλους ὁ στίχος, ἔχων τὸν ἐκ δύο βραχειῶν πυρρίχιον. καὶ καλεῖται μείουρος.

The last foot consists of ὄφιν, so a mora is missing in the arsis: this is the only occurrence of an explicit definition of μείουρος in the Homeric scholia, to which one example of ἀκέφαλος and one of λαγαρός are added. The sch. ex. further informs us that someone attempted to explain this case by positing the form ὄφιν, with a kind of doubling of the aspirated consonant into a voiceless consonant + an aspirated consonant: the reply is that the anomaly appears to be consciously designed to achieve a stylistic effect, all the more so since Homer could easily have avoided it simply by adopting a different word order. Despite

 16 See Erbse 1969–1988, testimonia ad Μ 208c, Χ 379a, Ψ 2. See also Leeuwen 1918, 95 f.; Chantraine 1973–1981, I, 103 ff.; Korzeniewski 1968, 23 and 75. Chantraine, 105, speaks of “libertés métriques, plutôt que de l’influence du vers sur la langue: il faut les distinguer des allongements métriques proprement dits.” 17 Concerning Eustathius, cf. Grossmann 1887, 36 ff.

  The Wrong Verses of Homer and Ancient Philology the possibility of invoking the supposedly lengthening quality of the aspiration, which would make the first syllable in ὄφιν long as far as the prosody is concerned,18 apparently the predominant view was that the στίχος μείουρος was acceptable as it was. Aristarchus’ comment on verse Ε 203, preserved by Herodian, strikes me as most interesting: Ε 203 ἀνδρῶν εἰλομένων, εἰωθότες ἔδμεναι ἅδην Herodian in sch. Ε 203a: ἄδδην vel ἅδην: Νικίας (fr. 8 B.) διὰ δύο δδ γράφει διὰ τὸ μέτρον, ὁμοίως τῷ “κύον ἀδδεές” (Θ 423. Φ 481), καὶ ψιλοῖ. Ἀρίσταρχος δὲ δι’ ἑνὸς δ καὶ βραχέως καὶ δασέως· τὸ γὰρ α πρὸ τοῦ δ δεκτικὸν μὲν ψιλοῦται, ὡς τὸ ἀδολέσχης ἀδρανής, μὴ οὕτως δὲ ἔχον δασύνεται· “ἅδε δ’ Ἕκτορι” (Μ 80), “ἅδος τέ μιν ἵκετο θυμόν” (Λ 88) κτλ. — Ἀρίσταρχος “ἅδην”, ἄλλοι δὲ “ἄδδην” διὰ δύο δδ.

The ‘normal’ prosody of ἅδην is no doubt ˘ ¯ and the verse is μείουρος, i.e. it lacks a mora in the last foot, just like Μ 208: some philologists had sought to remedy the metrical anomaly by assuming a lengthening through the doubling of the consonant, but Aristarchus opposes this hypothesis on the basis of linguistic arguments and accepts ἅδην, βραχέως καὶ δασέως. This simply means that Aristarchus here preferred to accept the verse as μείουρος, rather than taking into account a prosody/spelling which he evidently regarded as unmotivated. The grammarian Nicias adduced verses Θ 423 and Φ 481 as parallels for ἅδδην: in both these verses the word ἀδεές occupies the fourth foot, with the metrical structure ˘ ˘ ˘ – in other words, these are στίχοι λαγαροί, in which a mora is missing in the fourth foot. The scholia say nothing about Φ 481, whereas with regard to Θ 423, in sch. Θ 423a, Didymus states: τὸ δὲ “ἀδεές” δι’ ἑνὸς δ ὁ Ἀρίσταρχος. Here too, then, Aristarchus preferred to accept the metrical anomaly of the λαγαρός verse (although this term is not mentioned), rather than introduce an exceptional prosody/spelling. As we have seen, sch. Μ 208 adduces Ψ 2 as example of an ἀκέφαλος verse, and Β 731 τῶν αὖθ᾽ ἡγείσθην Ἀσκληπιοῦ δύο παῖδε as an example of a λαγαρός verse. In relation to this last verse, our scholia offer a note by Herodian, sch. B 731:  18 Cf. sch. B Heph., Ench. 291, 1 ff. Consbr.: φασὶ δέ τινες καὶ περὶ τούτων ὅτι ὁ ποιητής, εὐφωνίας μᾶλλον ἢ μέτρου φροντίζων, ἔστιν ὅπου τῆς τοῦ μέτρου ἀκριβείας ὑπερορᾷ, καὶ τοῦτο δῆλον ἐκ τοῦ ‘Τρῶες δ᾽ ἐρρίγησαν ὅπως ἴδον αἰόλον ὄφιν’ (Μ 208), ἠδύνατο γὰρ εἰπεῖν ‘Τρῶες δ᾽ ἐρρίγησαν ὅπως ὄφιν αἰόλον εἶδον’· ἢ ὅτι ἐνταῦθα ὁ ποιητής ᾔσθετο τῆς τοῦ δασέος φ ἐκφωνήσεως πλέον τι ἐχούσης διὰ τὴν σφοδρότητα τοῦ πνεύματος, ὡς καὶ Ἡλιοδώρῳ δοκεῖ τῇ δασείᾳ πλέον τι νέμειν. τὸ αὐτό φασι καὶ περὶ τοῦ ‘Ζεφυρείην’ (η 119). The issue has been addressed a number of times: see Ebelinge 1885, s.v.; Chantraine 1973–1981, I, 104; Korzeniewski 1968, 23. The same prosody occurs in Hypponax, fr. 39.6 Degani.

The Wrong Verses of Homer and Ancient Philology  

οὕτως ἐκτατέον διὰ τὸ μέτρον τὸ “Ἀσκληπιοῦ”, which must therefore be scanned ¯ ¯ ¯ ¯ , by considering the syllable -πι- long, for otherwise a mora would be missing in the fourth foot, -σκληπι-. As a parallel, Erbse refers to verse Β 518 υἷες Ἰφίτου μεγαθύμου Ναυβολίδαο, where Herodian comments, sch. B 518: τὴν μέσην συλλαβὴν ἐκτείνομεν τοῦ Ἰφίτου διὰ τὸ μέτρον. ἐν ἑτέροις γοῦν φυλάσσομεν τὸν χρόνον· “Ἴφιτος Εὐρυτίδης” (φ 14). For other parallels and similar cases, see Erbse’s testimonia on these two passages.19 Let us briefly dwell instead on two Odyssey verses and the relevant scholia. κ 60 βῆν εἰς Αἰόλου κλυτὰ δώματα: τὸν δ᾽ ἐκίχανον sch. HQ. ὁ στίχος σφηκώδης. σφηκῶδες δέ ἐστι τὸ ἐλλεῖπον ἐν μέσῳ τοῦ στίχου χρόνου, ὡς ἐνταῦθα. χρῄζει γὰρ ὁ δεύτερος ποὺς χρόνου. τὸ γὰρ αιο τροχαῖός ἐστιν. ἀλλὰ τὸ ο μονόχρονον ὡς δίχρονον λαμβάνομεν. — sch. B. Vind. ὁ στίχος λαγαρός ἐστιν.

From a terminological perspective, the scholium presents σφηκώδης as a synonym of λαγαρός. The metrical problem occurs in the second foot, the trochee which, being formed by the first two syllables of Ἀιόλου, therefore lacks a mora: an identical case is provided by κ 36 δῶρα παρ᾽ Αἰόλου μεγαλήτορος Ἱπποτάδαο, but for this verse no scholia or comments by Eustathius are available. The solution suggested for this anomaly seems rather rash, insofar as it devises a wholly unjustified prosody: ἀλλὰ τὸ ο μονόχρονον ὡς δίχρονον λαμβάνομεν.20 In some modern editions, even though the manuscript tradition unanimously transmits Ἀιόλου, the editors have chosen to restore in both verses the ancient genitive form Ἀιόλοο, ¯ ˘ ˘ ¯, with a lengthening of the last vowel through the two consonants that follow it.21 But while this might certainly be a suitable historical explanation for the anomaly, it is hardly the one supposed by ancient philologists.

 19 As regards vv. Α 362 and Ζ 443, metrical observations have only been preserved by Eustathius: cf. Erbse 1969–88, testimonia ad locc. 20 Cf. Eustathius 1647.22 ff. ὅτι ἐν τῷ “βῆν εἰς Αἰόλου κλυτὰ δώματα” λαγαρότης ἐστὶν, ὡς τοῦ Αἰόλου ἀντὶ μακρᾶς ἔχοντος τὴν παραλήγουσαν καὶ σφιγγομένου ἀπὸ δακτύλου εἰς τροχαϊσμὸν ἀπεναντίας τῷ προκοιλίῳ [cf. infra n. 32] ... θεραπεία δὲ τοῦ τοιούτου μετρικοῦ πάθους ἴσως μὲν καὶ τὸ εἶναι κύριον ὄνομα τὸν Αἴολον καὶ διὰ τοῦτο ὑποπίπτειν ἐξανάγκης ἔστιν ὅτε ἀδιαφορίᾳ. μάλιστα δὲ ἡ ὀξεῖα, δυναμένη ἐκτείνειν, ὡς ἀλλαχοῦ ἐῤῥέθη, οὐ μόνον βραχὺ φωνῆεν ᾧ ἐπίκειται ὡς ἐν τῷ “Αἰόλον ὄφιν” (Μ 208), ἀλλὰ καὶ τὸ πρὸ αὐτῆς καὶ μετ’ αὐτήν κτλ. 21 This is what von der Mühll 1952 and Heubeck 1983 have printed; cf. comm. ad κ 36, with further bibliography.

  The Wrong Verses of Homer and Ancient Philology γ 230 Τηλέμαχε, ποῖόν σε ἔπος φύγεν ἕρκος ὀδόντων sch. HM. Τηλέμαχε] οὗτος ὁ στίχος λαγαρός ἐστι· διὸ Ζηνόδοτος ἴσως μετέγραφε “Τηλέμαχ’ ὑψαγόρη μέγα νήπιε, ποῖον ἔειπες;”.22

The arsis of the second dactyl is represented by the last syllable of Τηλέμαχε. The scholium here does not offer any prosodic solution for the anomaly, but reports a variant by Zenodotus that is presented — albeit tentatively — as being related to this problem (διὸ Ζηνόδοτος ἴσως μετέγραφε): indeed, the text proposed by Zenodotus is free of metrical flaws. Some manuscripts present the variant Τηλέμαχος, the nominative instead of the vocative: another way, it would seem, to eliminate the λαγαρότης of the verse.23 A different possibility is suggested by two concurrent considerations: on the one hand, the vocative was regarded as a περίοδος αὐτοτελής, which thus had to be delimited by a punctuation mark to indicate a pause; on the other, this pause was believed to lengthen the previous or subsequent syllable. Such observations are clearly attested in relation to Nicanor, and Eustathius’ commentary suggests that they were also applied to γ 230, where nothing of the sort is preserved in our scholia.24 Therefore, two courses of action were identified to remedy the metrical anomaly: the first was to change the text in order to eliminate the problem; the second was to explain the lengthening of the short syllable in the arsis by invoking the lengthening power of a punctuation pause. The former, Zenodotean solution was probably rejected by the Aristarchean school, which instead would have found the latter solution perfectly consistent with its own approach. It should be added that this latter solution must not necessarily be seen as standing in contrast with an acknowledgement of the λαγαρότης of the verse. The argument may have been more sophisticated and critically refined: it may have been argued that the verse was λαγαρός in the sense that the missing mora had been replaced by a pause; but this was not tantamount to assuming that the verse was  22 Carnuth 1869, p. 32; Ludwich 1884–1885, I p. 530. 23 See the discussion in West f. 1981, comm. ad loc.; Carnuth 1869, 32, believed that “Zenodotus v. 230 mutavit propter metrum”. 24 Eustathius 1464.62 ff.: Τὸ δὲ “Τηλέμαχε ποῖόν σε ἔπος φύγεν”, λαγαρότητα ἔπαθε κατὰ τοὺς παλαιοὺς, διὰ τὴν λήγουσαν τοῦ Τηλέμαχε ... ὅμοιον δὲ καὶ ἐν τοῖς ἑξῆς τὸ “βῆν εἰς Αἰόλου κλυτὰ δώματα” (κ 36). καὶ ἐκεῖ γὰρ ἡ παραλήγουσα τοῦ Αἰόλου λαγαρὸν τὸν στίχον ποιεῖ ... ἐνταῦθα μέντοι θεραπεύει τὴν λήγουσαν τῆς κλητικῆς τοῦ Τηλεμάχου, οὐ μόνον ἡ στιγμὴ καὶ τὸ ἐπαγόμενον π, ἀλλὰ καὶ ἡ περισπωμένη τοῦ “ποῖόν σε ἔπος φύγεν”. ὡς ἡ δύναμις ἔχει τῆς κοινῆς συλλαβῆς. καθ’ ἣν δύναμιν, καὶ ἡ στιγμὴ χαρίζεταί τι χρόνου τῷ πρὸ αὐτῆς βραχεῖ ἢ βραχυνομένῳ φωνήεντι, καὶ τὰ ἐν ἀρχῇ λέξεων ἄφωνα ὡς πεφυκότα διπλοῦσθαι κτλ. Cf. Carnuth 1875, 33 f.; Friedlaender 1850, 36 f. and 121 f.; see esp. Nicanor to Τ 189: ... ἵνα διὰ τῆς σιωπῆς τοῦ χρόνου τὸ μέτρον σώζηται.

The Wrong Verses of Homer and Ancient Philology  

perfectly identical to a ‘normal’, completely ‘corrected’ verse, devoid of any noteworthy metrical features. The selection of examples we have examined is rather small and could easily be expanded. However, it offers enough elements for us to continue our enquiry. It is clear that the metrical anomalies of certain Homeric hexameters were an object of observation and study by Alexandrian philologists: the testimony from Dionysius Thrax — to which we shall be returning — certainly reveals that Aristarchus commented on such phenomena. We have seen how the problems posed by the scanning of certain verses could give rise to different reactions. The explanation developed often involved an unusual, if not wholly exceptional, prosody that could be more or less justified. Ad κ 60 it is stated that a μονόχρονον vowel should be scanned as δίχρονον, and we do not know whether in Antiquity any justification for this was provided; ad Μ 208 a prosody of ὄφιν is proposed that makes the first syllable long. We may add an example that explicitly quotes Aristarchus: ad Α 36, Herodian states: Ἀπόλλωνι: τὸ α ἐκτείνει Ἀρίσταρχος διὰ τὸ μέτρον. Here a solution to the possible acephaly of the verse, if the first two syllables of Ἀπόλλωνι are scanned ˘ ¯, is found precisely in the use of a special prosody, namely the metrical lengthening of the short syllable in the arsis. The problem of the epic lengthening of short syllables — which entails different instances of the adaptation of words to the metre, and which modern scholars have interpreted in a variety of ways25 — was well-known to ancient philologists: it falls within the varied phenomenology of what was regarded as the κοινὴ συλλαβή. It would be worth conducting an in-depth study of the different cases provided especially by the Homeric scholia and Eustathius.26 But let us take a fresh look at some of the cases we have already discussed, in order to deepen our enquiry. Ad Β 518 Herodian notes: τὴν μέσην συλλαβὴν ἐκτείνομεν τοῦ Ἰφίτου διὰ τὸ μέτρον. ἐν ἑτέροις γοῦν φυλάσσομεν τὸν χρόνον· “Ἴφιτος Εὐρυτίδης” (φ 14). The particular prosody of the noun Ἰφίτου in this verse is noted and compared with the other cases, in which the ‘normal’ prosody with the second short syllable is instead to be found. The name Ἰφίτου occurs another four times in the Homeric poems, namely ad Ρ 306 and φ 14, 22, and 37, always with a short second syllable; only ad Β 518 is the second syllable scanned long διὰ τὸ μέτρον: the criterion of internal comparison offers a norm based on frequency; the exception is noted yet not condemned; and only ad Β  25 Cf. Chantraine 1973–1981, I, 94 ff.; Korzeniewski 1968, 23. This field has been investigated in modern studies on epic diction, but their perspective falls outside the scope of our enquiry. 26 The first collections of such material were provided by Rauscher 1886 and Grossmann 1887, but the knowledge and editions available today enable us to move well beyond them.

  The Wrong Verses of Homer and Ancient Philology 518 is a prosodic observation recorded, whereas in relation to the other verses the scholia say nothing about their ‘normal’ prosody. A similar case is represented by Herodian’s scholium to Β 731: οὕτως ἐκτατέον διὰ τὸ μέτρον τὸ Ἀσκληπιοῦ. The prosodic peculiarity here lies in the fact that the last syllable -πι- must be considered long: in the parallels provided (Ἀσκληπιοῦ Δ 194, Λ 518; Ἀσκληπιάδῃ Δ 204, Λ 613; Ἀσκληπιάδην Ξ 2), the syllable -πι- is always short and the scholia do not note anything in relation to these passages. Here too, therefore, we find that an observation has only been preserved for a prosody that is ‘irregular’ owing to metrical requirements: perhaps, the ‘regular’ parallels were listed when commenting on the exceptional case.27 This would fall within the specific framework outlined by the methodological principles of Alexandrian philology: the principle of analogical criticism on the level of the grammar and vocabulary, usages and customs, realia, and so on, led to the definition of what was specifically Homeric, τὸ Ὁμηρικόν; yet, in doing so, it ran the risk of over-normalising things, to the point of rejecting original and genuine peculiarities. Aware of this danger, Aristarchus noted the presence in the Homeric poems of many words that only appear once and yet must be preserved; in other words, he acknowledged the possibility of exceptions and tempered the risks of over-normalisation through the concept of hapax legomenon: this critic was confident he could distinguish what was an acceptable exception from what had to be rejected as non-Homeric.28 The case examined above shows how this method was clearly adopted and applied even in the metrical-prosodic field. With regard to the acephalous verse from which we set out, Χ 379, Dionysius Thrax concludes (sch. Χ 379a, Aristonicus): ... σπανίως Ὅμηρος κακομέτρους ποιεῖ; and sch. D(T) Χ 379b states: σημειῶδες ὅτι καὶ ἀκεφάλοις χρῆται ὁ ποιητής, ὥστε οὐ δεῖ προπερισπᾶν τὸ “φίλε κασίγνητε” (Φ 308). In other words: sometimes, albeit rarely, Homer also presents some acephalous verses such as this one, i.e. verses lacking a mora in the first foot; these metrical anomalies are rare yet admissible phenomena, and we must take account of their existence in Homeric poetry. Sch. D(T) also preserves an additional element: ὥστε οὐ δεῖ προπερισπᾶν τὸ “φίλε κασίγνητε” (Φ 308): ad Φ 308 φίλε κασίγνητε, σθένος ἀνέρος ἀμφότεροί περ, there is nothing in the scholia about the prosody and accent of φίλε, and Erbse remarks:

 27 Cf. also Erbse 1969–88, testimonia ad sch. Β 731. At verse Β 316 Herodian notes the prosody of ἀμφιαχυῖαν (σημειωτέον). 28 Pfeiffer 1968, 227, 229.

The Wrong Verses of Homer and Ancient Philology  

nullum signum ante versum in A; tamen dubitaveris, num sch. Aristonici exstiterit, in quo versum a sillaba brevi incipientem notaverit, ad Χ 379 (Ariston.) Cf. D ad Χ 379. Etiam Herodianus hanc rem tetigisse potest, vide test. ad Ε 359 collecta.

Ad Ε 359 φίλε κασίγνητε κόμισαί τέ με δός τέ μοι ἵππους again we find nothing in the scholia, and Erbse notes: Eust. 556, 33: τὸ “φίλε” ἐκτείνει καὶ ἐνταῦθα τὸ τῆς ἀρχούσης δίχρονον, ὡς καὶ ἐν τῇ Γ (sic, leg. Δ, scil. Δ 155) ῥαψῳδίᾳ. Φησὶ γὰρ “φίλε κασίγνητε, κόμισαί τέ με, δὸς δέ μοι ἵππους” (Ε 359). Διὸ καὶ ὁ Ἀσκαλωνίτης, φασί, περιέσπα τὸ φι. Καὶ ὅρα τὴν ἀδιαφορίαν καὶ νῦν τῶν διχρόνων, εἴγε καὶ τὸ φίλος ποτὲ μὲν ἐκτείνει, ποτὲ δὲ συστέλλει τὴν ἄρχουσαν ... fort. partim ex hyp. Iliad. (sch. Herodiani), ad Δ 151, Φ 308 ecc.

It seems, then, that the issue concerns the three occurrences of φίλε κασίγνητε at the beginning of a verse, which are evidently compared with the normal prosody of φίλος, with a short first syllable, and perhaps also — and especially — with the frequent occurrence of ὦ φίλ(ε) or ὦ φίλοι (the latter followed by a vowel) at the beginning of a verse and always ¯ ˘ ˘. The three occurrences of φίλε κασίγνητε at the beginning of a verse are the aforementioned Φ 308 and Ε 359 and moreover Δ 155, where again the scholia provide no observations, but Eustathius preserves traces of the problem.29 The issue is whether to refer to the frequent and varied issue of the κοινὴ συλλαβή by positing a lengthening in the arsis, or whether to accept the στίχος ἀκέφαλος beginning with ˘ ˘ ˘. The opinion of sch. D(T) Χ 379b, σημειῶδες ὅτι καὶ ἀκεφάλοις χρῆται ὁ ποιητής, ὥστε οὐ δεῖ προπερισπᾶν τὸ “φίλε κασίγνητε” (Φ 308), is clearly that there is no need to posit the use of a circumflex accent and hence the lengthening of the first syllable of φίλε, since an acephalous verse is an acceptable possibility in Homeric metre: given that there undoubtedly are some acephalous verses, it is unnecessary to come up with peculiar prosodies when the metre of a verse can be explained by resorting to this concept. Unlike in the two cases previously examined,30 we here find the idea that an anomalous and exceptional prosody (in this case, the presence of so many occurrences of φίλος with a short first syllable) is less acceptable than the metrical anomaly, i.e. an acephalous verse.

 29 Cf. Eust. 458, 13 ff.: Ἰστέον δέ, ὅτι τὸ “φίλε” ἐκτείνει ἐνταῦθα τὴν παραλήγουσαν, οὐχ᾽ ὥστε μὴν καὶ περισπαθῆναι αὐτήν, ἀλλὰ κατὰ λόγον κοινῆς συλλαβῆς διὰ τὴν ὀξεῖαν, ὡς καὶ ἐν ἄλλοις γίνεται. Erbse 1969–87, ad loc.: “fort. aderat nota Herodiani de accentu vocis φίλε, ad Ε 359”. Cf. Eust. 527, 5 ff., ad E 100. 30 These were two proper names (cf. Eustathius, n. 20 above), whose prosody was probably regarded as more uncertain or less fixed. However, this is not a very meaningful observation, since there are examples involving common names.

  The Wrong Verses of Homer and Ancient Philology However, there is another important aspect worth noting. In the case of γ 230, we have seen how, in relation to the metrical anomaly of the λαγαρός verse, and in addition to the possibility of explaining the lengthening of the short syllable in the arsis by invoking the lengthening power of a punctuation pause, the possibility was also considered of changing the text to eliminate the problem. Among those resorting to the latter option — it seems (for the scholium suggest an element of doubt) — was Zenodotus, who changed the verse completely. If this is true, then the ancient pioneer’s attitude was to refuse to tolerate ‘erroneous’ verses, to the point of preferring heavy emendation — and it would not be surprising to find that the level of critical refinement was not yet that of Aristarchus’ day. Let us consider a few more testimonies, starting from Ν 172 and Ζ 34, with Aristonicus’ scholia. Ν 172 ναῖε δὲ Πήδαιον πρὶν ἐλθεῖν υἷας Ἀχαιῶν Aristonicus in sch. Ν 172: ὅτι Ζηνόδοτος γράφει “ὃς νάε Πήδαιον”, ἵνα κατάλληλον τὸν λόγον κατὰ συναφὴν ποιήσῃ. ἀγνοεῖ δὲ ὅτι Ὅμηρος διακόπτει τὰς φράσεις, ἵνα μὴ μακροπερίοδος γένηται. ἄλλως τε καὶ κακόμετρον τὸ ἔπος ποιεῖ. Ζ 34 ναῖε δὲ Σατνιόεντος ἐϋρρείταο παρ᾽ ὄχθας Aristonicus in sch. Ζ 34: ὅτι Ζηνόδοτος γράφει “ὃς ναῖε Σατνιόεντος”, ὁ Ἔλατος. κακόφωνον δὲ γίνεται.

It is debated whether Zenodotus ad Ν 172 wrote ὃς νάε or ὃς ναῖε, as ad Ζ 34. Leaving aside both this and the issue raised by Aristonicus for Ν 172 with regard to the μακροπερίοδος,31 what interests us here is the criticism that Aristarchus directed against Zenodotus (Ven. A preserves the diple periestigmene), accusing him of making the verse κακόμετρος through a reading that scanned the first foot ¯ ¯ ˘, i.e. by giving it an extra syllable.32 In sch. Ν 172 Aristonicus provides the only other occurrence of κακόμετρος in the Homeric scholia, in addition to the one already discussed at the beginning, namely Aristonicus in sch. Χ 379a; one wonders whether in sch. Ζ 34 κακόφωνον may not be a scribal error for κακόμετρον. So while accepting the possibility of κακόμετροι verses, Aristarchus here rejected a reading adopted by Zenodotus because it implied an unacceptable metrical pattern. Here are a few other examples preserved by Aristonicus:  31 Cf. Friedlaender 1853, 117 and 216; Erbse 1969–88, testimonia ad locc. 32 If Zenodotus had synizesis in mind, Aristarchus either did not realise it, or else regarded it as an impossible or unacceptable solution. Cases of this sort were defined as προκέφαλοι, within a classification where the options were: προκέφαλος / ἀκέφαλος, προκοίλιος / λαγαρός, δολιχόουρος / μείουρος. See sch. Β Heph., Ench. 289, 4 Consbr.; Erbse 1969–88, testimonia ad Ψ 266; cf. Erbse 1969–88, testimonia ad Μ 208c, Χ 379a, Ψ 2.

The Wrong Verses of Homer and Ancient Philology  

Aristonicus in sch. Β 520 καὶ Πανοπῆα: ὅτι Ζηνόδοτος γράφει “καὶ Πανοπτέων”, ἄμετρον ποιῶν τὸν στίχον καὶ οὐ καθ’ Ὅμηρον σχηματίζων τὸ ὄνομα κτλ. Aristonicus in sch. Β 634 ἠδ’ οἳ Σάμον ἀμφενέμοντο: ὅτι Ζηνόδοτος γράφει “ἠδ’ οἳ Σάμην ”, καὶ ἄμετρον ποιῶν καὶ ἀγνοῶν ὅτι κατὰ τὸ ἀρσενικὸν ἐνίοτε ἐκφέρεται τὰ εἰς η λήγοντα κτλ. Aristonicus in sch. Β 658 Ἡρακληείῃ: ὅτι Ζηνόδοτος γράφει “Ἡρακλεΐῃ” καὶ ἄμετρον ποιῶν καὶ οὐχ Ὁμηρικὸν τὸ σχῆμα κτλ. Aristonicus in sch. Ε 638a ὅτι ... Ζηνόδοτος δὲ καὶ ἀμέτρως γράφει “Ἡρακλεΐην”. Aristonicus in sch. Σ 222b ὄπα χάλκεον: ἀμέτρως ὁ Ζηνόδοτος “ὄπα χαλκέην”, οὐ συνεὶς ὅτι παραπλήσιόν ἐστι τὸ σχῆμα τῷ “κλυτὸς Ἱπποδάμεια” (Β 742) κτλ. Aristonicus in sch. Υ 484a Πείρεω υἱόν: ὅτι Ζηνόδοτος γράφει “Πειρέως υἱόν”, ἄμετρον ποιῶν τὸν στίχον καὶ παράλογον κτλ.

Whereas ad γ 230 we have seen Zenodotus change the text, apparently to eliminate a metrical anomaly which Aristarchus presumably explained and accepted, in the above cases we find Zenodotus proposing a text that falls short of the metrical criteria for hexameters and which — for this and other reasons — is therefore rejected by Aristarchus. Aristarchus’ polemical remarks on Zenodotus are well-known and concern linguistic usage, style, and Homeric realia: prosodic-metrical elements would also appear to play a role within this framework. It seems as though Aristarchus would often reject his predecessor’s choices also on account of metrical flaws: presumably, he sometimes believed that Zenodotus lacked the means to understand an anomaly which needed to be preserved, and, therefore, had changed the text without any real need; that he had not yet adequately studied the hexameter, and had thus proposed ways of reconstructing the text that needed to be rejected as ἄμετροι. This topic must be addressed within the framework of Alexandrian philology, in which it played a far from negligible role: the examples we have analysed outline a research path which must be developed as broadly as possible, on the basis of a detailed examination of the available material. Significant progress was made in Antiquity also in the study of the prosody and metre of hexameters: by the time of Aristarchus and his school, the expertise acquired was such that sophisticated analyses could be conducted by criticising and overcoming previous uncertainties in terms of methods and outcomes, so as to promote an in-depth understanding of epic diction and a rigorous and methodologically conscious construction of the text. Let us sum up our enquiry so far. In relation to certain problems posed by the metre of some Homeric verses, we have seen how ancient philologists — as is well known — often resorted to different arguments that led them, in specific cases, to posit a peculiar prosody that could eliminate the possible flaw in a verse’s metrical structure: in such cases, metrical normalisation is imposed to

  The Wrong Verses of Homer and Ancient Philology the detriment of linguist-prosodic regularity. However, the criterion of metrical normalisation, which privileges verse structure over linguistic norms, and sometimes requires peculiar prosodies, is not an absolute criterion: the tendency may have been to accept ‘wrong’ verses in the smallest possible number of cases, but hexameters with a ‘flaw’ in their metrical pattern were sometimes accepted as exceptions that tempered extreme, over-mechanistic regularisation. We have seen how metrical anomalies were sometimes preferred over linguistic prosodies which were regarded as even less acceptable. We have also noted how this perfectly fits with the methodological framework of Alexandrian philology and how it signifies a high degree of critical refinement and of faith in the means of analysis and evaluation adopted: critics were confident they could determine when it was preferable to accept a linguistic exception with a peculiar prosody, or a metrical anomaly with a ‘wrong’ hexametric pattern; they were confident that they could find a balance between the norm and exceptions in relation to linguistic prosody and hexametric patterns. At this point, we can return one last time to the scholium from which we set out and, more precisely, from Dionysius Thrax’s statement: τὰ γὰρ τοιαῦτα ἐσημειοῦντο πρὸς κρίσιν ποιημάτων, ὅτι σπανίως Ὅμηρος κακομέτρους ποιεῖ. Aristarchus’ pupil can argue that poor or ‘wrong’ verses in Homer are indeed to be found, albeit rarely. We have already sufficiently clarified this concept, which must have become entrenched in critics’ awareness through the work of Aristarchus and his school.33 However, this development also involved another important step: verses with metrical problems were an object of observation and study πρὸς κρίσιν ποιημάτων. We should not forget that in the section of the Techne grammatikè that is unquestionably genuine,34 namely § 1, Dionysius Thrax concludes his definition of the six parts of grammar (understood as ἐμπειρία τῶν παρὰ ποιηταῖς τε καὶ συγγραφεῦσιν ὡς ἐπὶ τὸ πολὺ λεγομένων) with the following words: ἕκτον κρίσις ποιημάτων, ὃ δὴ κάλλιστόν ἐστι πάντων τῶν ἐν τῇ τέχνῃ. The κρίσις ποιημάτων is therefore the culmination of γραμματική, its point of arrival and loftiest outcome. The scholia on this passage from the Techne discuss the issue carefully and in depth: far from consisting in a mere value judgement, it amounts to the acquisition of technical competence with regard to all the elements pertaining to literary composition, narrative content, historical facts, metre, style, vocabulary, compositional structure, and  33 An echo of this is also to be found in the following statement from sch. Β Heph. Ench., 291, 1 Consbr.: ὁ ποιητής, εὐφωνίας μᾶλλον ἢ μέτρου φροντίζων, ἔστιν ὅπου τῆς τοῦ μέτρου ἀκριβείας ὑπερορᾷ. 34 Cf. Schenkeveld 1994.

The Wrong Verses of Homer and Ancient Philology  

characters. This competence allows the grammarian to assess (κρίνειν) whether part of a work, or an entire work, is genuine or spurious; to confidently master the specific features of each author so as to discern between what is authentic and what is false; and to decide where each genre excels.35 It was with Aristarchus and his pupils that philology reached so high a level of development and such in-depth expertise and methods. Certain metrical anomalies found in Homer are rare yet admissible phenomena; they are exceptions, not signs of spuriousness: sometimes, they can be explained on the basis of more or less complex linguistic and grammatical arguments; other times, they are accepted as they are found. The philologist is confident he possesses the tools to judge whether a certain construction of the text can be accepted with a peculiar metrical structure, or whether it must be interpreted in a different way or even rejected. Along with all other issues, problems pertaining to the prosody and meter of epic hexameters had been investigated in a sufficiently broad and in-depth way to allow Alexandrian philologists to record them πρὸς κρίσιν ποιημάτων.

 35 Lallot 1989, 80 f.

 Antichi commenti a Omero. Alcune riflessioni 1. Quando e come si incominciò a commentare Omero? La risposta a questa domanda dipende molto da quello che si intende per «commentare» e per «commento». Sembra assodato che i rapsodi siano stati i primi a fornire occasionalmente spiegazioni di vario tipo a proposito dei versi che recitavano: quando il testo risultava oscuro, quando il pubblico lo richiedeva, forse quando semplicemente si coglieva l’occasione per dire qualcosa ritenuto interessante. Sappiamo che, a partire almeno dal VI secolo a.C., ci si impegnava a interpretare il contenuto mitico dell’epica, per lo più con lo strumento dell’allegoresi, e a spiegare il significato letterale del testo, «traducendo» in greco corrente parole ed espressioni di una lingua poetica talmente lontana dall’uso da risultare spesso incomprensibile. Il primo tipo di esegesi pare fosse caratteristico di ambienti filosofici, il secondo nasceva prevalentemente nell’ambito dell’istruzione: ma la scarsità della documentazione sconsiglia di assolutizzare e generalizzare. La spiegazione del significato letterale del testo (essenzialmente per ragioni di «estraneità» linguistica) e l’interpretazione del contenuto mitico (per i suoi presupposti culturali e di pensiero) emergono come i due generi di intervento esegetico nati più anticamente: nel corso della storia, saranno quelli di più lunga durata e di più costante presenza. Se per parlare di «commento» abbiamo bisogno di pensare a un lavoro sistematico e continuo, che segua regolarmente il testo nel suo ordine sintagmatico, allora la documentazione in nostro possesso indica chiaramente l’età ellenistica come il periodo in cui si produssero i primi commenti a Omero. Anzi, dobbiamo dire che probabilmente solo con Aristarco fu scritto un vero e proprio commentario, hypomnema, di grande portata all’Iliade e all’Odissea, con l’aggiunta di un cospicuo numero di saggi monografici, i cui contenuti si intersecavano e si scambiavano con quelli dell’hypomnema. In seguito, dal II secolo a.C. in poi, nella seconda parte dell’età ellenistica e per tutto l’arco dell’età imperiale, tardoantica e bizantina, la produzione di commenti omerici fu enorme e raccolse intorno ai versi del più venerato poeta del mondo antico una messe grandiosa di materiali esegetici. Che gli infiniti materiali di esegesi omerica prodotti nell’antichità non fossero consegnati soltanto alla forma del commentario è ben noto e anche del tutto naturale: però i commentari ne erano senza dubbio i depositari di elezione e contemplavano nella loro forma fluida e libera una varietà di contenuti praticamente senza limiti, dalle varianti testuali ai problemi di grammatica, dalle questioni antiquarie all’interpretazione del significato del testo, dall’esegesi allegorica e filosofica alla spiegazione letterale del

https://doi.org/10.1515/9783110772371-037

Antichi commenti a Omero. Alcune riflessioni  

dettato poetico, dagli aspetti storico-letterari a quelli propri del mito e della religione. Di fronte a questo imponente corredo esegetico, viene da chiedersi che cosa e in che modo, di tutta l’enorme messe di materiali dell’esegesi omerica antica, sia stato e sia utilizzato nell’interpretazione omerica dell’età moderna. In estrema sintesi, diciamo che i materiali esegetici antichi si possono dividere grosso modo in due grandi categorie: una è costituita da ciò che riguarda in senso stretto la storia e la costituzione del testo; l’altra, di grande ampiezza e varietà, comprende tutto ciò che, con definizione molto generica, possiamo chiamare «storia della ricezione». Il primo tipo di informazioni, che per lo più fu raccolto e discusso in età ellenistica e nei primi secoli dell’età imperiale, sembra essere stato utilizzato più o meno costantemente dai moderni studiosi e commentatori di Omero (anche se in quantità e con approfondimenti senza dubbio diversi), almeno dopo la pubblicazione degli scoli veneti A e B da parte di Villoison (Venezia 1788) e dei Prolegomena ad Homerum di F.A. Wolf (Halle 1795):1 su questo torneremo più avanti. 2. Le possibilità di utilizzare criticamente negli studi moderni i prodotti dell’antica esegesi omerica hanno fatto passi avanti notevoli, grazie alle nuove edizioni che sono state portate a compimento per l’Iliade e agli studi sulla storia della filologia antica. L’acquisizione più importante resta senz’altro l’opera di H. Erbse, che ha costituito un progresso decisivo per la scoliografia all’Iliade e fornisce un modello per quella all’Odissea, che non possiede ancora un simile strumento. Con una panoramica riassuntiva, possiamo dire che nella scoliografia omerica si distinguono le seguenti classi: 1) gli scoli derivati dal «Commentario dei Quattro» (in tedesco Viermännerkommentar, sigla VMK: Didimo e Aristonico, attivi in età augustea; Nicanore ed Erodiano, vissuti nel II secolo d.C.); 2) i cosiddetti Scholia exegetica; 3) quelli comunemente chiamati Scholia D. A questi corpora propriamente scoliografici2 si aggiungono nei manoscritti più o meno abbondanti excerpta di altro genere, tratti per lo più da opere di natura lessico 1 Per i riferimenti completi delle opere citate, si veda la bibliografia in fondo al volume (= Montanari 1998). 2 Alle tre classi suddette si aggiunge una classe indicata con la lettera h, che risulta composta da elementi derivati dalle altre tre e dunque non necessita di essere edita autonomamente: la sua funzione è quella di controllo e integrazione per la costituzione del testo delle altre classi, laddove fornisca lezioni utili o conservi materiali altrimenti perduti (cioè emendi o supplisca), il che accade sporadicamente. In questo modo la utilizza Erbse: cfr. Erbse 1969–1988, vol. I, Praefatio, pp. LVI–LVIII e LXXV–LXXVI; in modo analogo sarà utilizzata anche per l’edizione degli Scholia D (cfr. infra).

  Antichi commenti a Omero. Alcune riflessioni grafica, dagli scritti di esegesi omerica di Porfirio, dalle Allegorie Omeriche di un Eraclito vissuto probabilmente all’inizio dell’età imperiale. Un cospicuo numero di papiri, infine, ci fa conoscere frammenti ormai considerevoli dei predecessori antichi di quanto confluito poi nella scoliografìa e nella lessicografia. Il quadro è ormai ben definito sulla base del materiale relativo all’Iliade: nelle sue linee generali, dobbiamo pensarlo applicabile anche all’Odissea, benché gli scoli odissiaci attendano ancora di essere indagati a fondo per una nuova edizione critica e comunque siano in generale tramandati in modo peggiore e in quantità assai meno ricca. L’edizione di Erbse comprende essenzialmente gli scoli derivati da VMK e gli Scholia exegetica all’Iliade, presentati con ben precise distinzioni e indicazioni di appartenenza (i nomi abbreviati dei quattro grammatici di VMK oppure ex. per gli Scholia exegetica) e corredati di un ricco apparato di passi paralleli e riferimenti bibliografici: è dunque uno strumento di pronta utilità per chi si dedica a commentare l’Iliade, che può scegliere l’uso selettivo che desidera fare degli abbondanti materiali a sua disposizione in una forma criticamente elaborata. Il completamento dell’edizione curata da M. van der Valk del commento all’Iliade di Eustazio offre agli studiosi un altro testo in cui sono confluiti moltissimi dati dell’interpretazione omerica antica. Resta invece relativamente problematica (soprattutto per chi non sia specialista di questo genere di testi) la condizione in cui si possono utilizzare i resti delle Questioni omeriche di Porfirio e in generale le ricche fonti grammaticali e lessicografiche, tenendo conto anche del fatto che fra scoliografìa e lessicografia c’è normalmente una copiosa osmosi di materiali.3 Per quanto riguarda le condizioni critiche in cui gli studiosi possono utilizzare il materiale esegetico antico sull’Odissea, abbiamo già detto che manca una nuova edizione critica degli scoli e del commento di Eustazio: il modello del lavoro svolto per l’Iliade offre binari ben tracciati e le condizioni appaiono del tutto mature per affrontare questo problema. Nel quadro della scoliografia omerica c’è una carenza vistosa e da tempo lamentata, che riguarda gli Scholia D, dei quali manca un’edizione critica moderna sia per l’Iliade che per l’Odissea. Questa classe di scoli è ben caratterizzata dalle tre componenti che la costituiscono: una grande quantità di spiegazioni glossografico-parafrastiche di singole parole o espressioni; l’insieme di ἱστορίαι mitografiche raggruppate sotto la definizione di Mythographus Homericus; un certo numero di zetemata, cioè proposizioni di problemi esegetici con la loro  3 Un’edizione critica dell’Etymologicum genuinum è la carenza principale, ma non certo l’unica, in questo settore. Importante è il compimento dell’edizione degli Epimerismi omerici di A. Dyck in due volumi, Berlin/New York, 1983–1995.

Antichi commenti a Omero. Alcune riflessioni  

soluzione. Erbse trascurò volutamente (salvo casi sporadici e limitatissimi)4 la classe D, di cui il filologo italiano Vittorio de Marco stava preparando una nuova edizione (per l’Iliade soltanto), che non vide la luce. Come credo sia noto, da tempo ho ripreso il lavoro di de Marco in vista dell’auspicata edizione critica degli Scholia D all’Iliade, con lo scopo di completare l’insieme degli scoli iliadici aggiungendo quanto tralasciato da Erbse.5 Dei corrispondenti Scholia D all’Odissea si dovrà occupare chi porrà mano all’edizione critica della scoliografia odissiaca nel suo complesso: anche questa classe appare per l’Odissea meno ricca e importante che per l’Iliade.6 Allo stato attuale, gli Scholia D devono comunque essere utilizzati con le opportune cautele, perché sono stati fatti solo desultori e limitati progressi dopo le editiones principes cinquecentesche: quelli all’Iliade furono pubblicati per la prima volta da Giano Lascaris a Roma nel 1517; pochi anni dopo Francesco Torresano d’Asola pubblicò a Venezia un’edizione aldina in due tomi: nel primo, del 1521, riprese gli scoli iliadici dall’edizione lascaridea, nel secondo, del 1528, stampò per la prima volta quelli all’Odissea. Dopo le edizioni principi, la fortuna degli Scholia D come «commento» omerico è stata cospicua e di notevole interesse storico-culturale: insieme ai commentari di Eustazio, pubblicati a Roma negli anni 1542–1550 da Niccolò Maiorano, questi scoli furono costantemente usati come strumento di interpretazione omerica nel corso dei secoli dal XVI al XVIII. L’uso della scoliografia antica nella ricerca moderna è una storia solo in parte indagata: molte cose restano da scoprire e interpretare, anche solo seguendo la storia del testo stampato e della sua utilizzazione. Già nel corso del XVI secolo, gli Scholia D furono più volte ristampati in numerose edizioni omeriche, sia dell’Iliade che dell’Odissea, per lo più assieme al testo di Omero, indicati usualmente come «Interpres Vetustissimus», «Didymus» oppure Σχόλια ψευδεπίγραφα Διδύμου. Risale a Francesco Torresano e all’edizione aldina del 1528 di questi scoli all’Odissea la connessione pseudoepigrafa con il grande erudito di età augustea,7 ben presto rivelatasi del tutto infondata: anche se Didimo non c’entra nulla, il nome è all’origine della loro attuale sigla D, da intendersi dunque come del tutto convenzionale. Alcune tappe successive di questa fortuna sono ben note: una è l’edizione di Cornelius  4 Di cui si dà spiegazione nella Praefatio, pp. LXXIV s. 5 Per uno status quaestionis si possono vedere Montanari 1979; Montanari 1995 (soprattutto pp. 147 ss. [= questo volume, cap. 22]). 6 Nell’edizione disponibile degli scoli all’Odissea, cioè quella di Dindorf 1855, la classe degli Scholia D odissiaci è quella indicata con la sigla V (che sta per «Vulgata», scil. «Scholia», un vecchio modo di indicare questi scoli). 7 Solo qui compare un Διδύμου, che non c’è nell’edizione del 1521 degli scoli iliadici, dove si parla solo di Σχόλια παλαιά.

  Antichi commenti a Omero. Alcune riflessioni Schrevel, pubblicata a Amsterdam e a Leida nel 1656, nella quale il testo omerico è accompagnato dagli Scholia D; un’altra, a un livello ben differente e di ben maggiore significato, è l’importante edizione di Joshua Barnes, pubblicata a Cambridge nel 1711, che nel commento fece un largo uso degli Scholia D e di Eustazio: il materiale dei commenti antichi comincia a essere utilizzato non più da solo, bensì «citato» all’interno di un discorso del commentatore redatto in latino. Al rapporto diretto fra il commento antico e il testo omerico si comincia ad aggiungere l’intermediario di un discorso del commentatore moderno, che ingloba i materiali antichi nella propria argomentazione anziché — per così dire — lasciarli soli accanto al testo poetico. Una vicenda particolare di questo discorso merita qui un poco di spazio. Nel corso dei secoli XVII e XVIII, a Oxford e a Cambridge si affermò l’idea di stampare il testo dell’Iliade con gli Scholia D come commento. Ho fatto qualche ricerca (che penso di approfondire e ampliare quando possibile) e al momento mi consta che la più antica di queste edizioni sia quella del 1665: Ὁμήρου Ἰλιὰς ϰαὶ εἰς αὐτὴν σχόλια ψευδεπίγραφα Διδύμου, Oxonii 1665.8 Ma ne conosco almeno altre due del XVII secolo (datate 1676 e 1695), almeno due del XVIII secolo (datate 1780 e 1792) e ancora nella prima parte del XIX secolo tre edizioni del 1802, 1811, 1817:9 sono pubblicate a Oxford e tutte hanno gli Scholia D in fondo alla pagina (nel frontespizio, alla formula riportata sopra si alterna: σὺν τοῖς σχολίοις ψευδεπιγράφοις Διδύμου). Una tradizione analoga doveva esserci a Cambridge: conosco un’edizione dell’Iliade del 1679 e una del 1689, entrambe presentate con gli «Scholia quae vulgo appellantur Didymi»; l’edizione di Barnes uscirà pochi anni dopo. Con ogni probabilità esistettero varie altre edizioni omeriche di questo tipo, sia oxoniensi che cantabrigiensi, che sarà interessante censire con maggiore ampiezza e precisione: ma già così abbiamo una serie notevole, sufficiente al momento per il nostro discorso. Ancor più interessante è quello che sembra il punto di arrivo di questa tradizione inglese, sul versante oxoniense. Accanto alla grande edizione dell’Iliade con commento e vari apparati eruditi, uscita in nove volumi a Lipsia nel 1802,10 C.G. Heyne pubblicò a Lipsia nel 1804 una editio minor in due volumi, «cum brevi annotatione», probabilmente per uso della scuola. Questa editio minor dell’Iliade di Heyne fu ristampata molte volte, in Germania e in Inghilterra: in

 8 Erbse 1969–1988, vol. I, Praefatio, p. LXVI; cfr. Allen 1931, vol. I, Prolegomena, p. 259. 9 Alcune di queste edizioni contemplano Iliade e Odissea; per altre conosco solo il volume iliadico. 10 Nel 1822, dopo la morte di Heyne, uscì un ultimo volume, il IX, con gli indici a cura di A.G. Gräfenhahn.

Antichi commenti a Omero. Alcune riflessioni  

due occasioni a mia conoscenza, cioè nel 1821 e nel 1834, fu edita a Oxford con l’aggiunta degli Scholia D, per cui la pagina si presenta divisa in tre settori, il testo iliadico seguito prima da un commento in latino e poi in fondo da un secondo «commento» costituito dagli scoli.11 Un Monitum non firmato (a p. viii, dopo la Praefatio) nell’edizione del 1821 dice: «Huic novae editioni [...] accesserunt Scholia minora, quae Didymi nomine aliquando perperam appellantur, ad fidem editionis Romanae ab interpolationibus liberata [...] Oxon. d. 21 Octob. 1821». Nell’edizione del 1834 il Monitum è riprodotto, con una brevissima aggiunta in fondo: «Nihil in hac repetita editione mutavimus [...] Oxon. d. 30 Octob. 1834». Il testo sembra dunque indicare che quelle del 1821 e del 1834 siano state le due sole volte in cui l’editio minor dell’Iliade di Heyne sia stata pubblicata con l’aggiunta degli scoli come secondo «commento». Poiché Heyne morì nel 1812, non può essere stato egli stesso l’autore dell’aggiunta nel 1821 degli Scholia D alla brevis annotatio, idea che sembra collegarsi strettamente alla tradizione inglese di cui abbiamo detto. Evidentemente gli Scholia D, grazie agli elementi di cui sono costituiti, fornivano un aiuto considerevole per due aspetti basilari nella lettura di Omero: l’interpretazione letterale del testo e la spiegazione dei riferimenti e contenuti mitologici. Del resto, questi due elementi furono i più costantemente e abbondantemente presenti nella storia dell’esegesi omerica: con la lettera del dettato poetico e con i problemi del mito essa aveva cominciato a cimentarsi dalle sue origini e continuerà a farlo senza interruzione, fino alle parafrasi bizantine e le traduzioni umanistiche da una parte e agli sforzi allegoristici per conquistare il contenuto mitico omerico al pensiero neoplatonico e cristiano dall’altra. Per tornare al punto di vista dell’omeristica attuale, abbiamo richiamato la cautela che deve adottare chi voglia utilizzare gli Scholia D. La lunga serie di riprese e ristampe dal Cinquecento all’inizio dell’Ottocento ha spesso trattato il testo di questi scoli con una certa libertà e senza più fare alcun ricorso alle fonti manoscritte. È relativamente facile che uno studioso si trovi tra le mani una copia della suddetta editio minor dell’Iliade di Heyne (non rarissima nelle biblioteche) e la utilizzi per consultare gli Scholia D all’Iliade: ma anch’essa va utilizzata con prudenza. Benché si dica «ad fìdem editionis Romanae», non è stato ripristinato del tutto coscienziosamente il testo degli Scholia D all’Iliade di Lascaris: rispetto all’editio princeps ci sono differenze, che non derivano da collazione di

 11 Ὁμήρου Ἰλιάς. Homeri Ilias. Cum brevi annotatione, curante C.G. Heyne, accedunt Scholia Minora passim emendata, Oxonii, 1821; 1834 («Scholia minora» è un vecchio modo con cui erano indicati gli Scholia D, cfr. anche sopra, nota 6).

  Antichi commenti a Omero. Alcune riflessioni manoscritti, ma sono state introdotte in seguito per via congetturale.12 Naturalmente, le carenze dell’editio princeps sono facilmente immaginabili, compresa la mancanza di parti di scoli assenti nel codice usato da Lascaris e tramandate da altri:13 tuttavia un confronto con essa è d’obbligo, per sapere quale testo si abbia di fronte, dal momento che non c’è a disposizione un testo costituito criticamente e bisogna poter valutare gli interventi prodottisi durante la storia delle edizioni a stampa. Queste rapide notizie danno un’idea della fortuna degli Scholia D, che furono frequentemente usati come «commento» omerico, insieme ai testi di Eustazio, fino agli inizi del XIX secolo. Tale fortuna cominciò a declinare rapidamente e vistosamente tra la fine del Settecento e gli inizi dell’Ottocento, soprattutto a causa della scoperta degli scoli veneti A e B pubblicati da Villoison nel 1788 e del riconoscimento del loro valore storico-documentario. Gli Scholia D furono in seguito piuttosto trascurati, fino a quando, in tempi più recenti, il loro valore cominciò ad essere riconosciuto e sottolineato: il primo a indicare il desideratum di un’edizione critica fu Wilamowitz, poi l’interesse divenne sempre più evidente, grazie anche al quadro delineato dall’opera di Erbse, come abbiamo visto. Dopo l’edizione di Villoison, gli omeristi si dedicarono ai nuovi testi scoliografici dei codici veneti, la cui importanza (sfuggita a Heyne) fu colta pienamente da F.A. Wolf, che rivolse per primo l’attenzione in modo particolare alle informazioni sulla storia del testo omerico e sulle sue varianti antiche, che essi comprendono in modo assai copioso. Iniziata così, dopo Wolf e con il XIX secolo, quella che si usa chiamare l’epoca filologica negli studi omerici e nella storia della questione omerica, le notizie sulle varianti antiche e la storia del testo furono considerate e si affermarono come uno dei contenuti più importanti della scoliografia, che fu utilizzata soprattutto per tutto quello che ci dice sull’assetto testuale nel mondo antico. Dopo Villoison e Wolf, come ho anticipato all’inizio, queste informazioni trovarono normalmente posto negli apparati critici e furono normalmente discusse con maggiore o minore abbondanza nei commenti all’Iliade e all’Odissea. Mi pare si possa dire che questo tipo di notizie fornite dai materiali dell’esegesi antica, cioè quelle sulle variae lectiones e la storia del testo, sia oggi comunemente utilizzato dall’omeristica moderna e che in un commento scientifico ci si aspetti senz’altro di trovarlo esaminato, almeno in quantità considerevole.

 12 Ricordare anche il «passim emendata» del frontespizio; cfr. la nota precedente. 13 Progressi nello studio della tradizione manoscritta sono stati fatti da A. Schimberg e V. de Marco; cfr. nota 5.

Antichi commenti a Omero. Alcune riflessioni  

3. Un discorso molto più articolato e complesso richiede la seconda delle due grandi categorie di materiali esegetici indicate all’inizio, cioè tutto il ricchissimo e differenziato insieme di notizie genericamente definibile «storia della ricezione» di Omero, fornito da tutte le classi di scoli (e anche da altri testi): chiunque ne abbia avuto esperienza, anche solo sporadica, capisce bene a quale multiforme quantità di notizie si faccia riferimento, nelle quali trovano posto i contenuti più disparati. Di questo insieme di dati, negli studi moderni si fa un uso molto variabile, che voglio provare a definire in modo molto sintetico, anche correndo il rischio di un eccesso di semplificazione e quindi di far torto a qualcosa di significativo. Un impiego cospicuo di queste informazioni si trova, naturalmente, negli studi sulla storia della filologia e della grammatica nell’antichità: qui l’oggetto di interesse principale e del tutto prevalente è appunto la storia specifica di queste discipline, piuttosto che la storia dei modi in cui il testo omerico è stato letto e assimilato, la storia delle sue interpretazioni e della sua fortuna. Abbiamo così una grande quantità di studi sull’esegesi omerica antica volti a far emergere e analizzare, in corpore Homeri, i problemi e i risultati della filologia e della grammatica nell’antichità, per ricostruirne la storia. L’interesse di questo lavoro sta principalmente nel far luce sui metodi e gli orientamenti culturali degli interpreti antichi e le notizie che chiama in causa sono sfruttate con questo orientamento: tutto ciò ha ben poco interesse diretto e preciso per l’interpretazione omerica in se stessa e di conseguenza questo genere di ricerche offre risultati meno interessanti e utili per un commento omerico attuale, come siamo abituati a concepirlo comunemente. Per esemplificare, utilizzo un caso personale. Qualche tempo fa ho scritto un articolo dedicato alle notizie, ricavabili dagli scoli, sulla rilevazione di anomalie metriche e prosodiche nel testo omerico da parte dei filologi antichi e sui loro atteggiamenti, le loro reazioni intellettuali e le loro interpretazioni a questo proposito, secondo le loro dottrine.14 È chiaro che i risultati di quel tipo di ricerche hanno interesse per la storia della filologia e delle dottrine metriche, di cui fanno emergere e analizzano un problema particolare, e anche per la storia della ricezione e della lettura dotta di Omero, di cui potranno eventualmente contribuire a spiegare qualche fenomeno. Tuttavia, credo che nessun autore di un commento moderno inserirebbe simili ricerche nelle sue note, per quanto dossograficamente ricche egli desideri che siano.

 14 I versi «sbagliati» di Omero e la filologia antica, in Fantuzzi/Pretagostini 1995, vol. I, 265– 287.

  Antichi commenti a Omero. Alcune riflessioni Riflettendo su questo, vediamo emergere con chiarezza un dato di fatto, di cui bisogna avere piena consapevolezza. Le osservazioni e le notizie esegetiche reperibili nei commenti antichi, più che per l’autore e il testo che intenderebbero spiegare, sono utili e interessanti per quello che ci dicono su loro stesse, cioè sulla cultura di cui sono espressione: per questo è importante interpretare questi materiali, caratterizzarli, contestualizzarli, quando possibile anche datarli, almeno a grandi linee. Il caso dell’allegoresi è di totale evidenza: lo studio delle notizie antiche sulle interpretazioni allegoriche di Omero sarà sempre di grande interesse per la storia del pensiero, ma solo sporadicamente e in casi particolari offrirà elementi per capire Omero di per sé e in senso proprio. Non è vero tuttavia che nulla della filologia antica venga chiamato in causa e utilizzato dal punto di vista dell’esegesi omerica moderna, dunque proprio per interpretare Omero e non per le ragioni di un’altra disciplina. Si tratta naturalmente di una presenza desultoria e assai selettiva, che, almeno in grande maggioranza, mi sembra sostanzialmente condizionata da un confronto con l’interpretazione moderna del testo omerico. Il punto estremo è quando i dati della filologia e dell’esegesi antica sono addotti e discussi con il principio (per dire brutalmente) di «giusto» o «sbagliato», insomma per vedere e argomentare quanto gli antichi filologi si siano avvicinati all’interpretazione «corretta» di Omero e quanto ne siano rimasti lontani. Ma anche senza arrivare a questa valutazione poco storica, resta comunque operante con una certa costanza un criterio di selezione basato strettamente su ciò che appare utile o inutile dal punto di vista dello studioso moderno, al fine precipuo di interpretare Omero: laddove c’è un problema filologico-esegetico che appare importante o comunque dubbio, dibattuto, irrisolto, quale che ne sia la ragione, allora il dato dell’erudizione antica può diventare interessante e significativo. È probabilmente una constatazione scontata, nel senso che non potrebbe essere altrimenti: interessano ovviamente i dati di storia del testo (come dicevamo), possono talvolta servire notizie riguardanti un fatto linguistico, un elemento del mito, un dato antiquario. Sembra insomma che l’omeristica moderna tenda in sostanza a interessarsi dell’omeristica antica quando può trovarvi un aiuto, un supporto o quanto meno un termine di confronto per l’esegesi attuale, cosa ovviamente legittima e doverosa, ma se ne curi poco o nulla per ricostruire gli elementi di una storia della «lettura» di Omero nel mondo antico e dei vari modi in cui la «ricezione» dell’epica greca arcaica abbia avuto influsso sul pensiero e sulla civiltà: tenda quindi a ritenere i materiali antichi importanti o non importanti su queste basi, selezionando di conseguenza cosa includere nelle proprie dossografie e di cosa tener conto. Per questo, le varie considerazioni offerte dagli scoli ai versi del proemio dell’Iliade saranno abbondantemente citate a proposito della variante

Antichi commenti a Omero. Alcune riflessioni  

δαῖτα / πᾶσι del v. 5, dell’atetesi di Zenodoto dei vv. 4–5, del significato e della sintassi di ἐξ οὗ del v. 6, ma non certo ricordate per un’osservazione come quella che vedremo verso la fine di questo discorso. Vengono così del tutto trascurate e lasciate da parte molte notizie e disquisizioni, che non danno niente di significativo per l’interpretazione «corretta» e «valida» e non aiutano nella ricerca di capire Omero in sé, ma offrono molti elementi sulla storia della «ricezione» di Omero e dunque sulla storia della cultura dei suoi lettori e interpreti. Per chiarire meglio, cerco di fornire qualche esempio, oltre a richiamare quanto già accennato a proposito di osservazioni antiche su fenomeni prosodico-metrici e delle varie forme di allegoresi. 4. È ben noto come Platone e Aristotele15 abbiano istituito un preciso rapporto «genetico» fra Omero e la poesia teatrale: per Platone, Omero «sembra sia stato primo maestro e guida di tutti questi bravi poeti tragici» (Repubblica 595c); Aristotele delinea la derivazione da Omero delle forme drammatiche: la tragedia come genere serio corrisponde a Iliade e Odissea, la commedia corrisponde al comico Margite. Alla prospettiva del rapporto fra Omero e le forme poetiche del teatro greco antico si ricollegano numerose osservazioni presenti negli scoli, che non è diffìcile esemplificare (il tema rientra fra l’altro nella ben nota e diffusa idea di Omero iniziatore e origine delle conoscenze, prótos heuretés di molti aspetti della civiltà greca: un filone che offre una grande quantità di argomenti interessanti). a) Compiutasi la famosa lite all’inizio dell’Iliade, imbarcata Criseide per rinviarla al padre, Agamennone manda due araldi (Taltibio ed Euribate) alla tenda di Achille, per dare compimento alla sua minaccia e portar via Briseide. Gli araldi trovano Achille seduto presso la tenda e la nave: intimoriti e rispettosi «si fermarono, non gli dicevano nulla e non parlavano» (libro I, v. 332). Achille naturalmente capisce, saluta i due che, dopo un breve discorso dell’eroe, se ne vanno con la fanciulla. Commenta lo scolio ex. a A 332b: «Dire qualcosa infatti sarebbe stato irritante, prestare omaggio sarebbe stato servile. Omero per primo introdusse nella tragedia i personaggi muti (πρόσωπα ϰωφά)». Dunque, l’osservazione riferita dallo scoliasta porta a ritrovare in Omero l’inventore del ϰωφὸν πρόσωπον come personaggio teatrale. b) La prova dell’esercito nel II libro dell’Iliade, conseguente al fallace sogno di Agamennone, sta per portare a una fuga in massa degli Achei dall’assedio di Troia: «E allora contro il destino sarebbe avvenuto il ritorno degli Achei, / se Era non si fosse rivolta ad Atena con queste parole» (vv. 155–156). Atena scende  15 Rispettivamente Repubblica III 392e ss., 595c e Poetica 1448a 20 ss., 1460a 5 ss. 1448b 34 ss.

  Antichi commenti a Omero. Alcune riflessioni dall’Olimpo e convince Odisseo ad arrestare la fuga, che infatti è scongiurata. Lo scolio ex. a B 156 commenta: «Il poeta spinge le “peripezie” (τὰς περιπετείας) fino a un punto tale, che solo il dio può cambiarne il corso. Per primo Omero insegnò ai tragediografi l’uso dei marchingegni teatrali (μηχανάς)». Qui dunque si attribuisce a Omero l’introduzione del deus ex machina e simili trovate. c) Si ricollega bene a questo esempio un caso, che riguarda il IV libro dell’Odissea.16 Telemaco arriva a Sparta da Menelao e questi, secondo l’uso, offre agli ospiti cibo e bevanda prima di chiedere loro chi siano. La situazione si era presentata identica nel III libro, a Pilo presso Nestore. Fino al termine del banchetto di accoglienza le due sequenze sono perfettamente parallele: l’identità degli ospiti resta ignota, le presentazioni sono rinviate a dopo il pranzo. Poi la scena a Pilo (libro III) si svolge in maniera semplice: Nestore chiede agli ospiti chi siano e Telemaco risponde di essere il figlio di Odisseo. A Sparta invece il riconoscimento degli ospiti da parte di Menelao avviene in modo inatteso, malgrado la normale preparazione dei vv. 60–62: «in seguito poi, / quando avrete pranzato, chiederemo chi siete / fra gli uomini». Invece di una semplice domanda, accade qualcosa di imprevisto: Telemaco esprime sottovoce a Pisistrato la sua ammirazione per la reggia, Menelao se ne accorge e prende spunto per un discorso sulla ricchezza e le amarezze che spesso le si accompagnano, discorso nel quale menziona con dolore Odisseo e le sue pene. Il ricordo del padre muove Telemaco al pianto e questo offre un primo indizio sulla sua identità, però ancora in modo implicito, non dichiarato (vv. 116–119: «lo riconobbe Menelao / e allora rimase incerto nella mente e nell’animo / se lasciare che egli stesso rammentasse il padre / o per primo interrogarlo e chiedergli ogni cosa»). Menelao è incerto se lasciare che il giovane parli del padre oppure interrogarlo direttamente, ma l’attesa che accada una di queste due cose viene ancora frustrata a causa dell’arrivo di Elena, che chiede al marito notizie sugli ospiti stranieri e ravvisa la somiglianza del giovane con il figlio di Odisseo (vv. 155 ss.). Menelao conferma questa impressione e a questo punto, finalmente, per bocca di Pisistrato si svela apertamente l’identità dell’ospite: sì, questi è effettivamente il figlio di Odisseo (vv. 155 ss.). La scena del riconoscimento di Telemaco, dunque, ha uno svolgimento narrativo complesso e articolato, una struttura scandita in particolare da due momenti in cui l’attesa del riconoscimento viene frustrata e il riconoscimento di Telemaco è rinviato da un fatto imprevisto, che complica lo sviluppo della sequenza. Tutto questo in contrasto con l’analoga scena del libro precedente, a Pilo presso Nestore. La differenza fu notata e

 16 L’ho analizzato tempo fa in: Montanari 1979c, 171–174, ora riveduto in Montanari 1995a, 21–25.

Antichi commenti a Omero. Alcune riflessioni  

commentata dagli antichi esegeti e delle loro osservazioni è rimasta traccia negli scoli. Sch. MQ a Odissea IV 69: [scil. il poeta] ha dato occasione a Menelao per un riconoscimento insolito. La forma comune di riconoscimento infatti è che uno chieda e l’altro risponda di essere il figlio di Odisseo, come è accaduto presso Nestore [libro III]. Questo riconoscimento invece include una «peripezia», cioè che, scoppiato in lacrime il giovane per il ricordo di Odisseo, a Menelao nacque il sospetto della verità. Questo stato di sconvolgimento del giovane offre l’opportunità a Menelao di parlare. Sch. Odissea IV 113: la stessa reazione [si trova] anche presso i Feaci da parte di Odisseo (VIII, 522): Odisseo infatti piange ascoltando il racconto delle proprie imprese, Telemaco invece al ricordo del padre. L’atteggiamento di Telemaco, dice lo scolio, costituisce il punto di partenza per un riconoscimento insolito (ἀναγνωρισμὸς ϰαινός), diverso dal modo ovvio e comune per cui uno chiede «chi sei» e l’altro risponde «sono il figlio di Odisseo», come è avvenuto presso Nestore. Qui invece la scena del riconoscimento contiene una «peripezia» (περιπέτεια), si svolge attraverso una «peripezia»: quindi, riconoscimento semplice a Pilo presso Nestore (libro III), riconoscimento con «peripezia» nella scena corrispondente a Sparta presso Menelao (libro IV). Non è difficile notare che la terminologia e gli strumenti concettuali di analisi sono quelli della Poetica di Aristotele: riconoscimento e «peripezia» come elementi costitutivi fondamentali del racconto sono codificati in passi famosi, nei capp. 10, 11, 24. Forse possiamo richiamare, in particolare per quanto stiamo vedendo qui, Poetica 24.2 (1459b 12 ss.), dove si dice che la struttura dell’Odissea è complessa perché è piena di riconoscimenti: infatti a 10.2 (1452a 14 ss.) Aristotele aveva definito «complessa» (πεπλεγμένη) l’azione in cui il cambiamento di situazione avviene con un riconoscimento o una «peripezia» o entrambi. Ma ancor più precisamente ci interessa Poetica 11.2 (1452a 32 s.), dove si dice che il riconoscimento più bello si ha quando esso si produce attraverso una «peripezia». Perfettamente consono al nostro scolio è pure quanto si trova nel cap. 16, dedicato a analizzare i vari tipi di riconoscimento: a 16.2 si parla del celebre riconoscimento di Odisseo da parte di Euriclea grazie alla cicatrice e si conclude ribadendo che i riconoscimenti migliori sono quelli basati su una «peripezia», come quello di Odisseo in quell’episodio. Mi pare chiaro che negli scoli citati si trovino utilizzati per l’Odissea strumenti di analisi del racconto e una terminologia tipicamente aristotelici e da Aristotele applicati alla trama sia dell’epica che della tragedia. d) Superata la possibile sedizione e conclusa l’assemblea del II libro dell’Iliade, dopo i sacrifici di rito, si chiama a raccolta l’esercito per la battaglia:

  Antichi commenti a Omero. Alcune riflessioni prima del celebre catalogo delle forze in campo, una splendida serie di similitudini cattura la possente immagine dell’armata, in mezzo alla quale spicca il condottiero Agamennone, «negli occhi e nel capo simile a Zeus fulminatore, / simile ad Ares nella cintura, nel petto a Posidone» (vv. 478–479). Lo scolio ex. a B 478–9a commenta: «I pittori perseguono il vero, i tragediografi il superiore (più nobile), i commediografi l’inferiore, cose che si trovano tutte in Omero: la commedia come nell’episodio di Tersite, la pittura come per esempio nel verso I 209: “Automedonte teneva fermo, Achille glorioso tagliava”, mentre in questa descrizione di Agamennone l’immagine è plasmata più bella e più sontuosa del reale». Qui l’idea della derivazione omerica delle forme teatrali è esplicitamente operante, ma vale la pena di aggiungere un’osservazione. Se rimane costante, ovviamente, il rapporto della tragedia con l’epica, la derivazione omerica della commedia si arricchisce, in aggiunta al rapporto con il Margite già istituito da Aristotele nella Poetica (come abbiamo già detto), grazie alla messa in luce di elementi «comici» rintracciabili all’interno degli stessi poemi epici: l’episodio di Tersite è naturalmente l’esempio preferito, ma una vena comica nei poemi omerici è rintracciabile in altri punti, sui quali varrebbe la pena di vedere se ci sono osservazioni antiche (senza dimenticare il caso della Batrachomyomachia e le notizie di altre fonti erudite su opere comiche anticamente attribuite a Omero).17 e) Avevo promesso di concludere l’esemplificazione con un’osservazione relativa al proemio dell’Iliade. Gli scoli si soffermano molto sulla prima parola del poema, la menis, «l’ira funesta del Pelide Achille», quella parola chiave la cui fama nella storia della letteratura è forse insuperata. Attira una certa attenzione lo scolio 1a, rr. 9 ss.: «Omero iniziò dalla menis, poiché essa fu presupposto e principio degli avvenimenti. D’altra parte, egli ha inventato il proemio tragico per le tragedie: infatti il racconto delle sventure ci rende attenti e partecipi e, come il miglior medico, prima mette in luce i mali dell’anima e poi applica la cura». Una scarna osservazione sulla costruzione dell’Iliade fondata sul confronto con la tragedia mostra anche qui operante l’idea del parallelo fra Omero e le forme teatrali classiche. L’analisi dell’insieme di scoli al proemio dell’Iliade, in verità, credo possa fornire diversi altri spunti di interesse, ma rimandiamo l’analisi più ampia a un’altra occasione. 5. Qual è il senso di portare all’attenzione dati di questo genere, che assai di rado hanno qualche utilità per capire Omero e che, ancor più di quanto abbiamo esemplificato sopra (par. 3), nessuno studioso moderno inserirebbe mai nelle sue annotazioni di commento, per quanto ricche e dotte voglia farle? In primo luogo, è ovvio che non ho nessuna intenzione di sostenere che questi  17 Su questi aspetti cfr. Montanari 1988c.

Antichi commenti a Omero. Alcune riflessioni  

materiali dell’esegesi antica debbano essere oggetto di studio per capire Omero e nemmeno che debbano avere spazio in un commento, nel quale tutto quanto è storia della «ricezione», certamente enorme, non può che trovare un impiego limitato e opportunamente selezionato. Emerge con sempre maggiore chiarezza che per la maggior parte le osservazioni reperibili nei commenti antichi servono a spiegare se stesse, cioè la cultura da cui sono prodotte, più che il testo cui sono dedicate: saranno dunque di valore ridotto e sporadico per capire Omero, ma sempre di grande interesse per la storia del pensiero e della cultura. È proprio con questo orientamento e da questo punto di vista che il loro significato e la loro utilità possono e devono essere valorizzati e sfruttati molto di più di quanto sia mai stato fatto, per capire le «letture di Omero» che hanno influenzato i secoli e i fenomeni culturali. Per capire e interpretare l’Omero di un poeta ellenistico18 o di un poeta latino, tutti quei materiali servono in molti modi e offrono elementi importanti: mi sembra che valga la pena di porsi il problema se tenerne conto in qualche modo oppure di non utilizzarli per niente, lasciandoli dormire nelle pieghe di edizioni consultate solo dagli eruditi.

 18 Cfr. Montanari 1995b, in particolare 52 ss. (= questo volume, cap. 28).

 Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis In un articolo recente ho preso in esame alcuni problemi relativi alla forma e alle caratteristiche dell’ekdosis alessandrina e fra questi, in particolare, l’annosa e dibattuta questione che riguarda il numero e la sequenza delle edizioni e dei commentari di Aristarco dedicati a Omero. Un’attenta e approfondita riconsiderazione di diversi aspetti del problema, compresa una nuova analisi di due casi ben noti agli studiosi di storia della filologia antica e frequentemente studiati, vale a dire Il. 10, 397–99 e 19, 365–68 con le notizie fornite dagli scoli relativi, mi hanno portato alla convinzione che le cose siano andate in un modo, che qui riassumerò senza ripetere i ragionamenti e le motivazioni, che si trovano nell’articolo citato.1 Per entrambi i suddetti passi omerici, gli scoli ci informano a proposito di un ripensamento e cambiamento di opinione da parte di Aristarco rispetto alla sua prima scelta testuale. È una situazione che trova abbondanti paralleli nella documentazione scoliografica: tuttavia in quei due casi possiamo dire, malgrado le incertezze che permangono su alcuni dettagli, che gli scoli in aggiunta ci informano sul fatto che i filologi posteriori cercavano notizie sui termini di tale ripensamento e sulla “situazione testuale” determinatasi, che risultavano loro non chiari e incerti, consultando opere di Ammonio, allievo e diretto successore di Aristarco “nella scuola”. Per conciliare l’apparente contraddizione fra i due titoli di opere di Ammonio Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως, citato da Didimo in sch. Il. 10, 397–399a Erbse, e Περὶ τῆς ἐπεκδοθείσης διορθώσεως, citato da Didimo in sch. Il. 19, 365–368a1 Erbse, si è immaginato che effettivamente Aristarco stesso avesse fatto personalmente una sola ἔκδοσις, come presupposto dal primo titolo, e che la ἐπεκδοθεῖσα διόρθωσις, menzionata nel secondo, fosse stata fatta in seguito, dopo la morte del maestro, nella cerchia dei suoi primi e più anziani allievi, magari proprio da Ammonio.2 Tuttavia, non ci si può nascondere il fatto che una nuova ἔκδοσις ‘postuma’, curata da allievi utilizzando il materiale aristarcheo, sarebbe stata  1 Montanari 1998a, con le indicazioni bibliografiche essenziali (= questo volume, cap. 23). 2 Pfeiffer 1973, 338: “... la ‘recensione pubblicata di nuovo’ deve essere stato un testo riveduto, curato non da Aristarco stesso, ma da un discepolo come Ammonio sulla base del materiale lasciato dal maestro”. Cfr. van Thiel 1996, p. III n. 1: “Er erscheint danach sicher, dass Aristarchs beide ‘Ausgaben’ Randapparate neben dem Homertext waren, die eine von ihm selbst publiziert, die andere, sein Arbeitstext, postum von einem Schüler”, con rimando al suo articolo van Thiel 1992. https://doi.org/10.1515/9783110772371-038

Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis  

comunque una seconda ἔκδοσις di Aristarco: mi pare che questo fatto avrebbe reso ben difficile scrivere (proprio da parte di Ammonio) Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως. La considerazione del carattere dell’ἔκδοσις alessandrina e l’analisi di quei due problemi testuali, in rapporto ai quali le opere di Ammonio sono citate, mi ha portato a formulare su questo punto specifico un’ipotesi un poco diversa, che credo possa farci capire meglio cosa sia accaduto e che cosa fosse effettivamente la ‘seconda’ ἔκδοσις o ἐπεκδοθεῖσα διόρθωσις di Aristarco. Il fatto che non ci siano state più edizioni della διόρθωσις aristarchea e che la διόρθωσις medesima sia stata riedita (ἐπεκδοθεῖσα), mi pare si possa concretamente conciliare per davvero soltanto supponendo che la ἐκδοθεῖσα διόρθωσις e la ἐπεκδοθεῖσα διόρθωσις fossero in un certo senso (fisicamente) la stessa cosa, però modificata, cioè riveduta, corretta e ulteriormente ‘lavorata’. Ripercorriamo in sintesi l’insieme delia vicenda. Aristarco cominciò scrivendo ὑπομνήματα basati sul testo omerico del suo predecessore Aristofane di Bisanzio, che sono citati esplicitamente da Didimo nello sch. Il. 2, 133a: ἐν τοῖς κατ᾿ Ἀριστοφάνην ὑπομνήμασιν Ἀριστάρχου Ἴλιον ἐγέγραπτο, οὐκ Ἰλίου (anche qui dunque un caso di cambiamento di opinione). Poi si dedicò all’imponente fatica della sua διόρθωσις e produsse la propria ἔκδοσις, vale a dire la copia di Omero da lui scelta come testo-base su cui lavorare e recante l’annotazione di tutti i suoi interventi testuali (segni critici e lezioni), volti a rappresentare il suo testo omerico (ἐκδοθεῖσα διόρθωσις).3 A questo punto scrisse nuovi ὑπομνήματα, basati sulla propria ἔκδοσις e sentiti/definiti (da lui stesso o da altri, e per ragioni ben comprensibili) come più approfonditi e accurati (Didimo in sch.Il. 2, lllb: ἔv τινι τῶν ἠκριβωμένων ὑπομνημάτων).4 Fino a questo punto, mi sento di accettare questa ricostruzione, che è quella approvata da Pfeiffer: mi differenzio invece a proposito dellʼultima tappa della vicenda, vale a dire l’interpretazione del significato di ἐπεκδοθεῖσα διόρθωσις, che per Pfeiffer (e altri) sarebbe stata curata “non da Aristarco stesso, ma da un discepolo come Ammonio sulla base del materiale lasciato dal maestro”.5 È un dato acquisito e accettato negli studi sulla storia della filologia antica che la ricerca e la riflessione di Aristarco su Omero non trovarono durante la

 3 Ho argomentato nell’art. cit. alla n. 1 come ritengo che fosse fatta materialmente l’ekdosis. 4 La testimonianza dello sch. Il. 7.130a1 a questo proposito deve essere usata con cautela: cfr. Pfeiffer, 338–339; Erbse, ad toc. 5 Pfeiffer 1973, 334–340; cfr. anche sopra, n. 2. Pfeiffer rigetta l’idea di Erbse 1959, cioè che Aristarco avesse fatto un solo commentario una sola volta e questo hypomnema si identificasse con l’ekdosis, che dunque non sarebbe stata un testo omerico autonomo.

  Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis vita una fine assoluta, una conclusione definitiva: oltre che a riflettere e studiare, Aristarco continuò a insegnare e Omero dovette rimanere sempre tra le sue mani. Gli scoli, infatti, offrono abbondante testimonianza sul fatto che egli abbia continuato a lavorare alla sua διόρθωσις omerica, parlandoci spesso di scelte testuali ed esegetiche diverse sia fra le due ἐκδόσεις che fra edizioni, commentari e scritti di carattere monografico (συγγράμματα), di ripensamenti e cambiamenti di opinione del filologo di fronte al suo autore più grande, di una ricerca in continua evoluzione.6 Il problema della seconda ἔκδοσις ο ἐπεκδοθεῖσα διόρθωσις si identifica in sostanza con quello di capire come e dove potessero essere conservate e annotate le osservazioni sul testo omerico fatte da Aristarco dopo che egli aveva prodotto la sua (prima) ἔκδοσις, quale uso sia stato fatto di questo materiale e da parte di chi: fu Aristarco stesso a riversarlo in una seconda ἔκδοσις separata oppure furono allievi a utilizzarlo per una postuma nuova ἔκδοσις? La conclusione, a cui sono pervenuto nel lavoro citato all’inizio, è che una seconda ἔκδοσις separata e distinta, chiunque e in qualunque modo l’avesse fatta, difficilmente giustificherebbe da parte proprio di Ammonio un titolo come Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως e ancor meno permetterebbe di capire situazioni di dubbio sulle lezioni aristarchee, come quelle discusse negli scoli a Il. 10, 397–99 e 19, 365–68 (e probabilmente anche altre di diverso genere).7 Casi simili, in combinazione con le caratteristiche dell’ἔκδοσις alessandrina, mi hanno convinto che la soluzione più probabile sia quella già enunciata, e cioè che la ἐκδοθεῖσα διόρθωσις e la ἐπεκδοθεῖσα διόρθωσις fossero in un certo senso (fisicamente) la stessa cosa, però modificata, cioè riveduta e corretta con nuovi interventi. Mi pare del tutto probabile, in sostanza, che Aristarco non abbia lasciato da parte la sua prima ἔκδοσις omerica per usare un altro supporto (un nuovo esemplare omerico da usare come nuova base di lavoro?) su cui scrivere i suoi ripensamenti e le sue diverse scelte testuali: mi pare probabile invece che abbia ripreso in mano la sua ἔκδοσις e su quella stessa copia abbia continuato a scrivere e annotare ancora ripensamenti, riflessioni e nuove scelte testuali (una pratica che a nessuno studioso può risultare strana). La copia di Omero, che era stata la ἐκδοθεῖσα διόρθωσις, divenne così in seguito la ἐπεκδοθεῖσα διόρθωσις. Tuttavia, poiché non fece ancora nuovi

 6 Ludwich, 1884–1885, I, 27, diceva che la “Aristarch’s Homerkritik ... eine allmählich werdende war”, un’espressione molto efficace: cfr. sotto e n. 9. 7 Il problema delle incertezze di Didimo nel ricostruire le lezioni aristarchee è ben vivo negli studi di storia della filologia: cfr. Ludwich, 1884–1885, I, 38 sgg.; Pfeiffer 1973, 339; Montanari 1998a, 21 (= questo volume, cap. 23).

Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis  

ὑπομνήματα, Aristarco non ebbe l’occasione di redigere per esteso i motivi dei suoi ripensamenti e di fissare per iscritto il punto a cui essi si erano fermati: non è diffìcile immaginare che qualche volta le sue riflessioni si siano interrotte a uno stadio che egli stesso riteneva non definitivo, ancora da studiare. Questo esemplare, peraltro, potè ben essere chiamato ἐπεκδοθεῖσα διόρθωσις, “recensione pubblicata di nuovo”, perché effettivamente fu messo e rimase a disposizione dei dotti e dei numerosi scolari (nel Museo), forse quando Aristarco fuggì da Alessandria in occasione della crisi politica del 145/4 oppure alla sua morte. L’ipotesi di una sola copia di Omero recante due successivi strati di interventi aristarchei è del tutto idonea e sufficiente a spiegare il fatto che nella tradizione posteriore si sia parlato comunemente di αἱ Ἀριστάρχου o αἱ Ἀριστάρχειοι, di ἡ ἑτέρα (anche ἡ προτέρα, ἡ δευτέρα, ἡ χαριεστέρα)8 τῶν Ἀριστάρχου ο τῶν Ἀρισταρχείων, come pure a motivare l’uso dell’avverbio διχῶς per indicare una oscillazione di Aristarco fra due lezioni.9 Il modo di esprimersi della tradizione erudita e scolastica trova un’ottima giustificazione nel fatto che, in un certo senso, si poteva effettivamente parlare di due assetti diversi del testo, anche se materialmente consegnati alla stessa copia di Omero: il che peraltro giustificava l’affermazione che non esistessero πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως. In genere, la prima e la seconda versione dovevano essere abbastanza leggibili e distinguibili su quella tormentata copia, così da permettere alla successiva tradizione filologica di conoscerle entrambe e di renderne conto, constatando quando il filologo era rimasto della stessa opinione e quando invece le sue valutazioni erano state diverse nel tempo e si poteva registrare una oscillazione di Aristarco fra due scelte testuali differenti. D’altra parte, nuovi interventi materialmente vergati su una copia già annotata (che non poteva essere altro che in forma di una serie di volumina, presumibilmente con spazi comunque non troppo abbondanti) spiegano bene anche possibili situazioni di incertezza di vario genere sul testo di Aristarco e sull’evoluzione delle sue idee: casi in cui, per una qualsiasi ragione, poteva non risultare agevole distinguere chiaramente quali fossero i due assetti testuali indicati prima e dopo dal maestro, quale il suo ultimo pensiero e se egli avesse davvero raggiunto una decisione definitiva. Ammonio aveva tutti i suoi motivi per spiegare come la ἔκδοσις

 8 In scoli di Didimo: cfr. sch. Il. 16, 613 (ἡ δευτέρα). 18, 182c (ἡ προτέρα), 2, 579 e Od. 4, 727 (ἡ χαριεστέρα). È interessante l’osservazione di Didimo in sch. 18, 182a, μήποτε δὲ ἀμείνων ἡ προτέρα, con la successiva motivazione: cfr. lo sch. Il. 2, 133a, dove Didimo dice καὶ μήποτε ἄμεινον ἔχει, a proposito della lezione adottata da Aristarco nei primi commentari. 9 Tutte queste testimonianze risalgono per lo più a Didimo, Περὶ τῆς Ἀρισταρχείου διορθώσεως, e sono state analizzate in Ludwich 1884–1885; cfr. Erbse 1959 e Pfeifter 1973, 334–340.

  Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis fosse in realtà una sola, che a un certo punto però era stata “riedita” con cambiamenti. È probabile che i titoli Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως e Περὶ τῆς ἐπεκδοθείσης διορθώσεως si riferiscano ο a due parti di una stessa opera o comunque a due opere strettamente collegate, il cui scopo doveva essere quello di spiegare come fossero andate le cose e soprattutto dare chiarimenti sui casi dubbi e problematici: passi omerici in cui la decisione finale del maestro non risultava sicura o perché non era stata annotata con totale chiarezza o perché erano intervenute difficoltà (danni) materiali o anche perché la sua riflessione era stata interrotta dalle circostanze. Come dicevo, l’analisi dei due casi di Il. 10, 397–99 e 19, 365–68, a proposito dei quali si è conservata negli scoli la citazione da parte di Didimo delle due opere di Ammonio, ha offerto una preziosa testimonianza per ricostruire questa vicenda. Nella tradizione scoliastico-erudita, con le sue brachilogiche e stringate annotazioni, il più delle volte l’informazione su un cambiamento d’idea da parte di Aristarco resta limitata al fatto che si forniscono due scelte differenti per lo stesso luogo, introdotte con la terminologia appena ricordata. Assai più di rado invece (probabilmente per la riduzione del materiale esegetico nella formazione dei corpora scoliografici e le scelte degli epitomatori) è rimasta traccia di una discussione fra i filologi posteriori su quale fosse la situazione alla quale Aristarco si era fermato, quali i termini delle sue scelte e dei suoi ripensamenti. Nei due casi ricordati sopra, la situazione ‘finale’ non risultava chiara, per cui è naturale che nella tradizione fìlologico-esegetica si discutesse e si cercasse di capire come stessero le cose a proposito delle opinioni del maestro sul testo di quei passi omerici. Giunti a questo punto, si può utilmente proseguire la ricerca esaminando altri casi in cui abbiamo una notizia, che ci parla di cambiamenti da parte di Aristarco a proposito della lezione da scegliere. Ci occupiamo ora di due passi, per cui negli scoli viene chiamata in causa la testimonianza di Dionisio Trace. Essi risultano particolarmente interessanti, perché appartengono allo stesso ambiente nel quale si collocano le notizie ricordate sopra, che coinvolgevano Ammonio: vale a dire la cerchia dei primi scolari e immediati successori di Aristarco, una fase nella quale i problemi evocati dovevano essere vivi e oggetto di discussione. Cominciamo dalla notizia presentata da Didimo in sch. Il. 18, 207a: Didimo in sch. 207a: ὡς δ᾿ ὅτε καπνός ἰὼν : οἱ περὶ Διονύσιον τὸν Θρᾷκα φασὶν Ἀρίσταρχον πρῶτον ταύτῃ χρώμενον τῇ γραφῇ μεταθέσθαι καὶ γράψαι “ὡς δʼ ὅτε πῦρ ἐπὶ πόντον ἀριπρεπὲς αἰθερʼ ἵκηται”. ἐμφατικῶς τὸ ἐν πολέμῳ πῦρ ἐπιτεθὲν τῷ Ἀχιλλεῖ παρέβαλε τῷ ἐν πολεμουμένῃ ἁπτομένῳ. А Didimo in sch. 207b: Ἀρίσταρχος “ὡς δʼ ὅτε πῦρ ἐπὶ πόντον ἀριπρεπὲς αἰθέρ’ ἵκηται” καὶ γὰρ ἄτοπόν φησι πῦρ εἰκάζεσθαι καπνῷ. Т

Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis  

Dionisio Trace qui ci fa sapere che Aristarco in un primo tempo aveva accettato la lezione ὡς δʼ ὅτε καπνὸς ἰὼν ἐξ ἄστεος αἰθέρʼ ἵκηται, presupposta dallo scolio come lemma (Ἀρίσταρχον πρῶτον ταύτῃ χρώμενον τῇ γραφῇ): in seguito poi avrebbe cambiato idea (μεταθέσθαι καὶ γράψαι) per adottare la lezione ὡς δʼ ὅτε πῦρ ἐπὶ πόντον ἀριπρεπὲς αἰθέρʼ ἵκηται.10 L’espressione μεταθέσθαι καὶ γράψαι non contrasta certo con il quadro descritto sopra, non implica necessariamente una copia diversa e separata, si adatta all’idea di un cambiamento apportato scrivendo la nuova lezione laddove prima era stata accolta la precedente: per essere più precisi, possiamo pensare alla seconda versione del verso scritta in margine accanto alla prima oppure alla parte differente (καπνὸς ἰὼν ἐξ ἄστεος / πῦρ ἐπὶ πόντον ἀριπρεπές) scritta nell’interlinea. È forse probabile che la nuova proposta aristarchea fosse questa volta una sua congettura, piuttosto che una lezione rinvenuta in una copia collazionata: ma non è facile avere certezze su queste valutazioni.11 Tutta la nostra tradizione manoscritta reca concordemente ὡς δ’ ὅτε καπνὸς ἰὼν ἐξ ἄστεος αἰθέρ’ ἵκηται, accettato senz’altro dagli editori moderni, mentre la seconda lezione di Aristarco è testimoniata solo da Didimo negli scoli. Negli scoli è conservata anche la motivazione del ripensamento di Aristarco. Mentre infuria la lotta per il corpo di Patroclo, Era manda Iris da Achille per esortarlo a mostrarsi sopra il fossato allo scopo di spaventare i Troiani e aiutare gli Achei a impadronirsi del cadavere. Poiché Achille è momentaneamente disarmato, interviene Atena a renderlo terribile agli occhi del nemico: lo copre con l’egida sulle spalle, diffonde intorno alla sua testa una nube dorata e fa sprigionare da lui una fiamma splendente (φλόγα παμφανόωσαν, v. 206). Il motivo del fuoco luminoso è amplificato dalla successiva similitudine, che inizia al v. 207 (ὡς δ’ ὅτε...: “come quando...”) e termina al v. 214 (ὣς ἀπ’ Ἀχιλλῆος κεφαλῆς σέλας αἰθέρ’ ἵκανε: “così dalla testa di Achille il bagliore saliva al cie 10 Cfr. Ludwich 1884–1885, II, 92–94; Erbse, ad loc., Linke 1977, fr. 19 e comm. p. 46 (superato Schmidt 1852, fr. 23, 376); Edwards 1991, 170–171: in questi luoghi si trova la bibliografìa precedente. Pasquali 1952, 238 n. 5 (sulla scia di Wilamowitz) riteneva che la seconda lezione di Aristarco non fosse intesa come una vera e propria ‘lezione’, bensì come una semplice proposta exempli causa (“Aristarco, poiché l’altro termine della comparazione è lo splendore della testa di Achille, non può se non aver desiderato che Omero si fosse espresso così”), cosa che non posso accettare; Pasquali era scettico sul fatto che si potesse dimostrare che qualche lezione fosse senz’altro congettura di Aristarco: vedi subito dopo e la n. seg. 11 In Montanari 1998a (= questo volume, cap. 23) ho argomentato la mia convinzione, secondo cui la διόρθωσις comportava sia il lavoro di emendamento congetturale sia la scelta fra varianti testuali rinvenute grazie alla collazione di copie diverse: decidere caso per caso di cosa si trattasse è sicuramente assai arduo, ma non mi sento di ammettere che nel lavoro testuale degli alessandrini ci fossero solo congetture (come è stato sostenuto più volte di recente) oppure solo lezioni testimoniate. Come è noto, si tratta di un tema ancora oggetto di copiose discussioni.

  Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis lo”): essa descrive una città isolana assediata, dalla quale si alzano, visibili da lontano, durante il giorno colonne di fumo e durante la notte bagliori di fuoco, mezzi con i quali gli assediati chiedono aiuto ai vicini. Nell’attacco della similitudine, appunto al v. 207, Aristarco trovava strano che il fuoco sprigionantesi da Achille venisse paragonato al fumo: sch. 207b καὶ γὰρ ἄτοπόν φησι πῦρ εἰκάζεσθαι καπνῷ; e riteneva più efficace che la similitudine si avviasse con il fuoco innalzantesi dalla città assediata: sch. 207a ἐμφατικῶς τὸ ἐν πολέμῳ πῦρ ἐπιτεθὲν τῷ Ἀχιλλεῖ παρέβαλε τῷ ἐν πολεμουμένῃ ἁπτομένῳ. Da qui il suo ripensamento e la proposta di scrivere al v. 207: ὡς δ᾽ ὅτε πῦρ ἐπὶ πόντον ἀριπρεπὲς αἰθέρ᾽ ἵκηται. Se queste poche righe di motivazione fossero andate perdute, eliminate da un epitomatore troppo desideroso di abbreviare, avremmo potuto avere, come accade nella maggior parte dei casi, soltanto una stringata informazione del tipo: ἐν τῆ ἑτέρᾳ τῶν Ἀριστάρχου “ὡς δ᾽ ὅτε πῦρ ἐπὶ πόντον ἀριπρεπὲς αἰθέρ᾽ ἵκηται”. Anche la versione ridotta del materiale di Didimo conservata nel manoscritto Т (sch. 207b), benché conservi un piccolo brano della motivazione aristarchea, non ci informerebbe né sul fatto che si trattava di una seconda proposta né sul fatto che se ne discutevano le ragioni. Ne dobbiamo concludere che, se il materiale didimeo di VMK questa volta non si fosse conservato in forma più ampia, probabilmente parleremmo di una semplice variante o al massimo di una seconda lezione aristarchea, ma non sapremmo nulla a proposito della prima scelta e del successivo cambiamento di opinione di Aristarco e neppure del fatto che di questo si parlava fra gli immediati allievi, evidentemente esaminandone le ragioni e l’opportunità. Interessante nella sua problematicità è il caso per cui Didimo cita ancora Dionisio Trace nello sch. Il. 9, 464b: ἀμφὶς ἐόντες: Διονύσιος ΑТ ὁ Θρᾷξ ἐν τῷ Πρὸς Κράτητα διὰ τῆς ἱπποδρομίας Α φησὶ γεγραμμένου “ἀντιόωντες” μεταθεῖναι τὸν Ἀρίσταρχον ἀμφὶς ἐόντες. ΑТ12

Secondo Didimo, dunque, nel trattato Πρὸς Κράτητα Dionisio Trace riferiva che Aristarco aveva cambiato la precedente lezione ἀντιόωντες (γεγραμμένου: mentre prima era scritto ἀντιόωντες), sostituendo al suo posto (μεταθεῖναι) la lezione ἀμφὶς ἐόντες.13 Si nota il parallelo nell’uso del verbo μετατίθημι a sottolineare il  12 Come si vede, anche qui il manoscritto A reca una versione dello scolio più ricca rispetto al manoscritto Т, che ha eliminato l’epiteto e il riferimento all’opera di Dionisio.  Cfr. Erbse, ad loc.; Linke 1977, fr. 15 e comm. pp. 43–44 (superato Schmidt 1852, fr. 5, 371): in questi luoghi si trova la bibliografìa precedente, cui aggiungere Ludwich 1884–1885, II, 93– 94; Helck 1905, 80–81.

Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis  

cambiamento testuale: μεταθεῖναι qui, μεταθέσθαι nel caso precedente. C’è ancora (a mio avviso) un serio problema nell’interpretare cosa esattamente significhi l’espressione διὰ τῆς ἱπποδρομίας, ma la cosa non è di importanza decisiva per il nostro discorso in questo momento e possiamo lasciarla da parte, rinviandone l’esame ad altra sede:14 resta il fatto che lo scolio ci informa che di questa scelta testuale aristarchea e delle sue motivazioni parlava Dionisio Trace nella sua monografìa Contro Cratete. Inoltre lo sch. 464c (Aim), forse ancora risalente a Didimo (?), riporta un’altra variante, senza indicarne la paternità: ἐv ἄλλῳ “ἐγγὺς ἐόντες”. Nella tradizione manoscritta ha prevalso nettamente ἀμφὶς ἐόντες, normalmente accolto dagli editori moderni: in alcuni codici si trova ἐγγὺς ἐόντες; nessun testimone (a quanto pare) conserva ἀντιόωντες. Non c’è dubbio che lo scolio didimeo, così come è conservato, sia molto brachilogico e che del discorso di Dionisio Trace sia rimasto un estratto piuttosto criptico, secondo cui in quel verso omerico prima si trovava scritto ἀντιόωντες e poi Aristarco aveva cambiato (per congettura o per collazione) in ἀμφὶς ἐόντες. Risulta così molto difficile ricostruire quale fosse esattamente la situazione cui faceva riferimento Dionisio Trace e che egli aveva voluto descrivere. Ci sono due possibilità: o Dionisio Trace diceva che Aristarco aveva scritto ἀμφὶς ἐόντες al posto della lezione ἀντιόωντες che trovava scritta nel suo esemplare (γεγραμμένου), cioè semplicemente che ἀμφὶς ἐόντες era la lezione di Aristarco, l'unica da lui scelta, senza cambiamenti di opinione; oppure voleva dire che dapprima Aristarco aveva lasciato e dunque accettato la lezione ἀντιόωντες che gli offriva la sua copia, ma poi aveva avuto un vero e proprio ripensamento rispetto alla prima idea e aveva deciso di cambiare in ἀμφὶς ἐόντες. Nel primo caso la precisazione γεγραμμένου “ἀντιόωντες” μεταθεῖναι suona abbastanza curiosa e non usuale, ma d’altra parte non abbiamo nessun motivo per ritenere che le cose siano andate nel secondo modo e dobbiamo restare in dubbio. Certo infine Aristarco voleva ἀμφὶς ἐόντες: ma questa era stata subito la sua scelta testuale, rimasta costante, oppure era stato un intervento successivo, un ripensamento che l’aveva portato a sostituire (per congettura o per collazione) la lezione del suo testo base, che in un primo tempo era stata accettata?

 14 Discussione del problema nella bibliogr. cit. alla n. prec. Sembra che si intenda che Aristarco sostenesse ἀμφὶς ἐόντες sulla base di un confronto con uno o più versi appartenenti all’episodio della corsa dei carri durante i giochi funebri per Patroclo nel libro ΧΧΠΙ (Erbse, ad loc., indica Il. 23, 330 vel 393, sulla scia di Ludwich, 1884–1885, I, p. 306; altri indicano Il. 23, 643, cfr. Valk 1963–1964, II, 86 e n. 15): ma i termini del confronto aristarcheo mi sembrano ancora poco chiari e la questione da approfondire.

 Discussioni su «Οlimpo» nella filologia omerica antica e il grammatico Leogoras Il significato del nome «Olympos» nell’uso omerico è una vexata quaestio ben nota agli studiosi per le sue numerose comparse nei materiali riguardanti l’esegesi di Omero nel mondo antico. Per riassumere in un modo forse un po’ sbrigativo e convenzionale, essa si basa sulla differenza e opposizione fra un approccio esegetico di tipo filologico-grammaticale e uno di tipo filosofico e allegorizzante. Nella prospettiva filosofico-allegoristica, Olimpo è parte di una cosmologia e indica i luoghi celesti come sede degli dei; nell’interpretazione filologica, Olimpo è il monte della Macedonia, luogo geografico terrestre. La scoliografìa conserva non poche testimonianze, che lasciano facilmente intravvedere i termini di un dibattito esegetico di età alessandrina. Lehrs nel suo libro su Aristarco e Gudeman nell’articolo sui segni critici hanno fatto emergere e trattato la questione, sottolineando come non pochi scoli omerici contrassegnati con una diplè e risalenti ad Aristonico (e dunque a materiale aristarcheo) trattino specificamente di questo problema.1 Più recentemente sulla questione sono tornati ancora, per ragioni e in modi diversi, M. Schmidt e M.S. Funghi (ricordando fra l’altro come la discussione sul significato del nome «Olympos» in rapporto a «ouranos» si trovi già nel commento orfico del papiro di Derveni).2 Gli interventi di Aristarco erano chiaramente volti a respingere l’identificazione di Olimpo con il cielo e comunque la sua localizzazione nei luoghi celesti. Egli osservava che «Olympos» è usato con il significato di «cielo» in poeti posteriori a Omero e restava coerente al suo principio filologico e metodologico di distinguere l’uso omerico da quello posteriore. Gli Sch. D a Iliade 1.18 e 1.353 affermano in modo esplicito che, in Omero, Olimpo è il monte della Macedonia, sacro agli dei e loro abitazione, mentre ϰατὰ ἀλληγορίαν / ἀλληγοριϰῶς Olimpo è il cielo, sede degli dei, riportando dunque tale identificazione a una forma di spiegazione allegorica.3 Alla stessa problematica appartengono anche le divergenze sull’interpretazione della parola βηλός in Iliade 1.591 (cfr. Iliade 15.23): Aristarco riteneva che il termine indicasse la soglia della dimora  1 Lehrs 18823, cap. III I; Gudeman 1922, 1919. 2 Schmidt 1976, 85–87; Funghi 1983: qui (18 s.) anche un accenno alle osservazioni sull’identificazione di Olympos nel papiro di Derveni. 3 Sch. D A 18 Ὀλύμπια δώματ᾽ ἔχοντες: οἱ τὸν Ὄλυμπον ϰατοιϰοῦντες. Ὄλυμπος δὲ ϰατὰ μὲν Ὅμηρον ὄρος τῆς Μαϰεδονίας μέγιστον, ἱερὸν τῶν ϑεῶν· ϰατὰ δὲ ἀλληγορίαν Ὄλυμπός ἐστιν ὁ οὐρανός, παρὰ τὸ ὁλολαμπὴς εἶναι. Sch. D A 353 Ὀλύμπιος: ὁ τὸν Ὄλυμπον ϰατοιϰῶν Ζεύς. ὁ δὲ Ὄλυμπος ὄρος Μαϰεδονίας, ἱερὸν οἰϰητήριον τῶν ϑεῶν· ἀλληγοριϰῶς δὲ ὁ οὐρανός. https://doi.org/10.1515/9783110772371-039

Discussioni su «Οlimpo» nella filologia omerica antica e il grammatico Leogoras  

degli dei sul monte Olimpo, Cratete identificava in βηλός la parte più alta (esterna) del cielo, la circonferenza-involucro del cosmo.4 Dunque, due linee distinte e contrapposte: quella alessandrina aristarchea per cui «Olympos» è il monte della Macedonia, quella filosofìco-allegorica per cui «Olympos» si identifica con il cielo; alla base, il problema dell’identificazione della dimora o delle dimore degli dei.5 A questa problematica vanno ricollegati6 due frammenti papiracei, che presentano interessanti e difficoltosi testi di esegesi omerica: si tratta di un papiro di Bruxelles e di un papiro di Milano. Il papiro Brux. Inv. E. 7162, databile al Ι sec. d.C., fu pubblicato per la prima volta nel 1937 da M. Hombert e C. Préaux,7 che lo definirono semplicemente come «Extraits homériques relatifs aux lieux célestes». Nel repertorio del Pack (nr. 1224) è indicato come «on Homeric cosmology (or philosophy)», ma A. Körte lo classifìcò fra quelli aventi un carattere fìlologico-grammaticale e riteneva che esso non presentasse alcun elemento dell’interpretazione omerica di tipo filosofico.8 Il papiro fu studiato anche da F. Della Corte, che credeva di avere di fronte un frammento di un trattato cosmologico e vi riscontrava soltanto questo tipo di interesse.9 Per contro, nel suo libro sugli scoli omerici M. Schmidt avanza l’idea che lo si possa guardare come un resto di una monografìa aristarchea («ein Rest einer aristarchischen Monographie»), forse intendendo non tanto Aristarco stesso quanto qualcuno della sua scuola, che scriveva muovendosi su linee di contenuto aristarchee.10 Infine M.S. Funghi, pur con la dovuta cautela, ha sostenuto che il testo, per quanto si può capire, appare più vicino a interessi filologico-grammaticali che a tematiche fìlosofico-cosmologiche.11 Di fatto, le uniche parti ricostruibili con sicurezza del testo molto frammentario sono le citazioni omeriche, che compaiono nel seguente ordine: Iliade 1.497–99, 5.749– 50 (= 8.393–4), 19.128–9, 24.97, 24.104, 1.496, 5.748–49, 19.128–9, dove sembra di vedere la ripresa da una sequenza che va dal I al XXIV e poi ricomincia, come se gli stessi versi — rilevanti per il problema — venissero riconsiderati una seconda  4 Cfr. Montanari 1988, 27–28 e 36–38, con tutti i riferimenti. 5 Per l’analisi della dizione omerica a questo proposito cfr. Merritt Sale 1984, 1–28, con i riferimenti bibliografici. 6 Lo ha fatto egregiamente Funghi 1983, che ricorda anche (18 s.) il precedente del papiro di Derveni. 7 Hombert/Préaux 1937, con una riproduzione fotografica. 8 Körte 1941, 139–140. 9 Della Corte 1939. 10 Schmidt 1976. 11 Funghi 1983 con una ripresentazione del testo in trascrizione diplomatica.

  Discussioni su «Οlimpo» nella filologia omerica antica e il grammatico Leogoras volta. Il resto è del tutto ipotetico e comunque piuttosto misero: si nota la presenza di paragraphoi e di spazi bianchi per segnalare inizio e fine dei versi omerici citati, ma la parte esegetica sfugge per ora a una caratterizzazione precisa a causa delle estese lacune. Rinuncio a riprodurre il testo, per il quale non ho progressi significativi rispetto alla revisione della Funghi. Il papiro Med. Inv. 71.82, databile al I sec. a.C., fu pubblicato nel 1972 da S. Daris con il titolo di «commentario».12 Una definizione in verità fondata su elementi piuttosto deboli: non solo il papiro è molto frammentario, ma pone anche serie difficoltà di lettura e il poco rimasto non mostra le caratteristiche che ci si aspetterebbe per un hypomnema o simili. La parte che permette di leggere un poco di testo concerne l’aggettivazione (omerica) di «Olympos». È possibile che l’autore prenda in considerazione e discuta i versi rilevanti nel corso dell’Iliade e che in questo punto stia mettendo in luce la differenza fra Ὄλυμπος e οὐρανός sulla base degli epiteti che li accompagnano. Anche in questo frammento compaiono paragraphoi e spazi bianchi (non menzionati dal primo editore). E anche per questo frammento M.S. Funghi ritiene che il testo riveli interessi filologici centrati sull’uso linguistico di Omero, suggerendo che essi possano avere la loro origine nel quadro degli studi su temi omerici in una prospettiva aristarchea e forse nella corrente della tradizione aristarchea.13 Il testo è consultabile nell’editio princeps citata. Entrambi questi frammenti possono essere ancora studiati, facendo forse qualche passo avanti, malgrado le loro condizioni veramente misere: le lacune resteranno comunque oggetto di semplici ipotesi, a meno che qualche nuovo frustulo degli stessi papiri non intervenga a cambiare la situazione. In ogni caso, mi pare che del loro contenuto si possa capire quanto basta perché venga alla mente la problematica testimonianza concernente il grammatico greco Leogoras (Λεωγόρας) di Siracusa, vissuto con ogni probabilità nell’arco del II sec. a.C.14 La testimonianza più significativa su di lui si trova nell’excerptum grammaticale conosciuto come «Grammaticus Parisinus» o «Anecdotum Parisinum» e afferma che Leogoras fu il primo a usare la diplè (il segno critico tipico della prassi filologico-esegetica di Aristarco) con lo scopo di segnalare i versi omerici rilevanti per distinguere «Olympos» dal cielo. Leogoras, si dice nell’excerptum, notava proprio che il nome «Olympos» era usato solo per il

 12 Daris 1972 (con riproduzione). 13 Funghi 1983. 14 Gudeman 1925; Usener 1865; Montanari 1999.

Discussioni su «Οlimpo» nella filologia omerica antica e il grammatico Leogoras  

monte e mai per indicare il cielo: di conseguenza, nell’uso omerico Ὄλυμπος e οὐρανός non devono essere considerati come intercambiabili e sinonimi.15 Lehrs non dava credito a questa testimonianza e se ne liberava piuttosto sbrigativamente,16 mentre Usener credeva che Leogoras fosse un grammatico più anziano di Aristarco e più giovane di Aristofane di Bisanzio, che avesse operato nell’ambiente alessandrino e avesse dato un contributo personale di rilievo allo sviluppo del sistema dei segni critici.17 Gudeman invece sosteneva che Leogoras fosse in realtà un discepolo di Aristarco e fosse stato l’autore di una specifica ricerca sull’uso del nome «Olympos» in Omero: a questo tema egli avrebbe dedicato un syngramma, da cui sarebbero nate le varie testimonianze sull’argomento nella scoliografia omerica.18 È difficile non mettere in relazione l’esistenza dei materiali sull’argomento negli scoli omerici, la testimonianza dell’anonimo excerptum sul grammatico Leogoras di Siracusa e i due problematici frammenti papiracei, che al problema nel suo insieme conferiscono per così dire maggiore corposità.19 Naturalmente tutto rimane aperto al dubbio. Le opere testimoniate nei due frammenti papiracei possono o devono essere messe in relazione con l’opera di Leogoras? E in quale modo: rappresentano in qualche modo excerpta della sua opera oppure di scritti che hanno utilizzato la sua opera? La ricerca di Leogoras sul problema di Olimpo aveva assunto la forma di un syngramma nel quale veniva trattato e discusso il materiale selezionato a questo scopo e che sarebbe confluito nella linea aristarcheo-alessandrina della filologia omerica antica? Leogoras ha avuto effettivamente un ruolo nello sviluppo del sistema dei segni critici e in particolare per l’uso della diplè?

 15 Nauck 1867, 280; Gudeman 1922, 1919: «diplen aperistikton primus Leogoras Syracusanus apposuit Homericis versibus ad separationem Olympi a coelo, proprie Olympum ab eo pro monte positum adnotans, nusquam pro coelo, quod saepe οὐρανὸν εὐρὺν dicat et μαϰρὸν Ὄλυμπον neque e contrario epitheta permutet. Ponebat autem tam ad montis signfìcationes quam ad caeli». 16 Lehrs 1882, 332 nota 240: «Fuisse iam ante eum [scil. Aristarchum] aliquem diples usum id potest verum esse: sed illi testimonio nulla fides. Fuisse aliquam ante Aristarchum editionem Homericam, et quidem iam diple usam, ignoti cuiusdam Leogorae, cuis editionis nullum praeterea vestigium, certe non admodum probabile»; nel cap. dedicato alla questione di Olimpo (cit. sopra nota 1) Leogoras non è menzionato. 17 Usener 1865. 18 Gudeman 1922, 1919, nota. 19 Lo scetticismo di Lehrs (cfr. sopra nota 16) sembra basarsi sull’idea di un’edizione di Leogoras, senza prendere in considerazione la possibilità di un syngramma.

 I percorsi della glossa: Traduzioni e tradizioni omeriche dall’antichità alla cultura bizantina È molto noto e citato il fr. 233 CGF dei Δαιταλῆς di Aristofane (la sua prima commedia, rappresentata nel 427 a.C.),1 nel quale si trova una delle piú antiche testimonianze dell’interpretazione omerica antica e in particolare della pratica della glossografia nell’ambito del sistema educativo ateniense del V secolo. Il frammento è restituito da Galeno nel suo glossario ippocratico2 (e da Polluce II 109 per il solo v. 1). Ci sono problemi di ricostruzione puntuale del testo aristofaneo, ma il senso generale appare acquisito. La commedia metteva in scena un padre con due figli, uno ‘cattivo’, che in qualche modo risultava seguace della cattiva educazione moderna (sofistica), e uno ‘buono’, piú vicino all’educazione tradizionale approvata dal padre: il contrasto padre-figlio sarebbe stato uno dei motivi portanti della commedia, incentrata sul tema e sul problema della formazione dei giovani. A quanto pare, sono conservate due battute di un dialogo fra il padre e il figlio ‘cattivo’, alla presenza del figlio ‘buono’. Il testo qui riprodotto è quello dell’edizione di A.C. Cassio.3 πρὸς ταύταις δ᾿ αὖ λέξον Ὁμήρου γλώττας· τί καλοῦσι κόρυμβα. ⏑ ⏑ ⎼ ⏑ ⏑ ⎼ ⏑ ⏑ ⎼ ⏑ ⏑ ⎼ τί καλοῦσ᾿ ἀμενηνὰ κάρηνα. ὁ μὲν οὖν σός, ἐμὸς δ᾿ οὗτος ἀδελϕὸς ϕρασάτω τί καλοῦσιν ἰδύους. ............. †τί ποτέ έστι τὸ εὖ ποτεῖν†. Padre Oltre a queste, dimmi poi le glosse omeriche: che vuol dire κόρυμβα? ............................................. che vuol dire ἀμενηνὰ κάρηνα Figlio 1 Adesso questo tuo figlio e mio fratello dica un po’ che vuol dire ἰδῦοι?

Nella battuta dei primi due versi, il padre chiede al figlio “cattivo” di spiegare e interpretare talune glosse omeriche: qualunque cosa sia perduta nella parte iniziale del v. 2, c’era evidentemente un incalzare di domande di questo tipo. Se i versi 3–4 seguivano immediatamente (come è dato per sicuro nelle ricostruzioni degli editori), il giovane non sapeva rispondere (o in ogni caso non rispondeva) e, a quanto pare, in un certo senso sfidava il fratello (il figlio ‘buono’) a interpretare glosse soloniane o comunque di linguaggio giuridico, almeno per  1 Cfr. Pfeiffer 1973, 59 e 147; Cassio 1977, 21–36, 75–77. 2 Linguarum seu dictionum exoletarum Hippocratis explicatio XIX 66, 1 K. 3 Cassio (cit. alla nota 1), fr. 28, con il commento. Kassel–Austin in CGF danno all’inizio πρὸς ταῦτας e il v. 4 è ricostruito così: ⏑ ⏑ ⎼ ⏑ ⏑ ⎼ ⏑ ⏑ ⎼ ⏑ ⏑ ⎼ ⏑ ⏑ ⎼ τί ποτ᾿ ἐστὶν ὀπύειν; https://doi.org/10.1515/9783110772371-040

I percorsi della glossa  

il caso di ἰδῦοι,4 mentre il v. 4 rimane abbastanza problematico.5 Il contesto si chiarisce presupponendo che il padre voglia saggiare le conoscenze tradizionali di base del figlio, che invece andava dietro ai moderni sofisti. L’educazione prevedeva basilarmente una certa conoscenza di Omero e la capacità di capire la sua lingua, in molti punti diventata difficile ed estranea al greco corrente già nel V secolo e sicuramente anche prima. Le ‘parole difficili’ di Omero menzionate nel frammento sono κόρυμβα e ἀμενηνὰ κάρηνα. La prassi glossografica e parafrastica divenne una presenza talmente comune e diffusa nel corso della civiltà greca, che è quasi impossibile tenere presente tutte le modalità di comparsa di questa interpretazione letterale della lingua epica, dalle tracce riscontrabili in poeti e scrittori fin dall’età classica ai numerosissimi testi di questo genere nell’età alessandrina e imperiale, fino alla scoliografia e alle parafrasi disseminate nei manoscritti di età bizantina. La parola κόρυμβα6 si trova una sola volta in Omero, in Iliade IX 241. Odisseo parla ad Achille e gli descrive la situazione negativa degli Achei in sua assenza: Ettore impazza e con l’aurora στεῦται γὰρ νηῶν ἀποκόψειν ἄκρα κόρυμβα αὐτάς τ’ ἐμπρήσειν μαλεροῦ πυρός. minaccia di strappar via alle navi gli ornamenti piú alti e di incendiarle con fuoco violento.

Gli ἄκρα κόρυμβα sono gli aplustri, gli ornamenti in cima alla poppa della nave, ambito trofeo di guerra. Nella scoliografia omerica troviamo puntualmente per questo verso uno sch. D, che ‘traduce’: ἄκρα κόρυμβα· τὰ τῶν νεῶν ἀκροστόλια.7

La glossa ritorna in diverse altre fonti. Esichio α 2614: ἄκρα κόρυμβα· τὰ ἀκροστόλια τῶν νεῶν.

 4 Parola delle leggi di Solone secondo Elio Dionisio ι 4 (citato da Eustazio 1158, 20; cfr. Fozio, Lex. ι 36): per la precisione Elio Dionisio la attribuisce a Dracone e a Solone. 5 Il verbo ὀπύειν è frutto di una ricostruzione di Dobree e Kaibel, accettata da Kassel–Austin, vedi n. 3; Kock e Meineke escludevano queste parole dal frammento di Aristofane. La possibilità che si tratti di parola soloniana si basa su Esichio β 466, ma cfr. Cassio e Kassel–Austin ad loc. 6 Da κόρυμβος, plurale κόρυμβοι e κόρυμβα. 7 Il testo è così nell’ed. di Erbse, che lo prende dal solo ms. A: ma tutti i mss. degli sch. D hanno lo scolio nella stessa forma.

  I percorsi della glossa Cfr. anche κ 3700: κόρυμβα· τὰ ὑπ᾿ ἐνίων ἄϕλαστα (vedi sotto), τὰ ἀκροστόλια, τὰ ἄκρα τῶν πρυμνῶν. τὰ ἀπεξυσμένα πρὸς κόσμον τῶν νεῶν ἄκρα, καὶ ἐπικεκαμμένα, ἅ ἐστι τὴν πρύμναν καὶ κατὰ τὴν πρῶραν. Synagogè 55, 21 Ba; Fozio, Lex. α 834; Suida α 953: ἄκρα κόρυμβα· τὰ ἀκροστόλια τῶν νεῶν τὰ ἐξέχοντα κατὰ πρύμναν ἢ πρῷραν. Cfr. anche Synagogè 282.2 Ba, Fozio, Lex. κ 990, Suida κ 2124: κόρυμβοι· κλῶνες ἢ ἄκρα (ἄκροι Su.). Eustazio 750, 34 (ad Il. IX 241): κόρυμβα δέ εἰσιν ἄκρα καθ᾿ Ὅμηρον, κατὰ δὲ ἑτέρους ἀκροστόλια καὶ οἷον κορυϕαὶ νεῶν. Καὶ ἔοικεν οὕτω λέγεσθαι τὰ ἐξέχοντα καὶ μάλιστα ἐν ὄρεσιν.

Mentre l’uso del plurale maschile κόρυμβοι viene interpretato e glossato variamente con riferimento a cose alte, che stanno in cima o alla fine di qualcosa (monti, acropoli, rami e cosí via, cfr. Esichio κ 3701: πάντα τὰ μετέωρα καὶ εἰς ὕψος ἀνατείνοντα κορύμβους λέγουσιν),8 il neutro plurale κόρυμβα sembra avere assunto tendenzialmente (anche se non in assoluto, cfr. per es. Eustazio) una certa specializzazione per indicare gli aplustri delle navi, certamente in dipendenza dallo hapax omerico di Iliade IX 241, compreso un ricorrente collegamento sintagmatico e semantico con ἄκρα. In Eschilo, Persiani 410 s., si legge che una nave greca dà inizio allo scontro κἀποθραύει πάντα Φοινίσσης νεώς/ κόρυμβ(α), “e strappa via tutti gli aplustri alla nave Fenicia”.9 La ripresa dell’omerico Iliade IX 241 νηῶν ἀποκόψειν ἄκρα κόρυμβα è evidente e marcata non tanto dall’equivalenza prosodica fra πάντα e ἄκρα (considerando che il significato di κόρυμβα di fatto può inglobare quello di ἄκρα), ma dal riferimento specifico alla nave10 e dall’uso di ἀποθραύω in parallelo semantico con ἀποκόπτω. Nella glossa di Esichio κ 3700 abbiamo visto comparire la parola ἄϕλαστα come possibile alternativa sinonimica di κόρυμβα. Anche ἄϕλαστον è uno hapax omerico: in Iliade XV 717, Ettore incalza le navi degli Achei e afferra con le mani l’aplustre della nave di Protesilao: ἄϕλαστον μετὰ χερσὶν ἔχων. Sch. D ἄϕλαστον: τὸ λεγόμενον ἀκροστόλιον τῆς νεὼς πρύμνης (cfr. Αp. Soph. 49, 6 ἄϕλαστον: τὸ ἀκροστόλιον).

 8 Esichio κ 3701: κόρυμβον· τὴν ἀκρόπολιν, ἐπειδὴ ἐϕ᾿ ὕψους ἐστί. τὴν ἐπὶ τῆς κεϕαλῆς τοῦ ὄρους ὕλην. καὶ καθόλου πάντα τὰ μετέωρα, καὶ εἰς ὕψος ἀνατείνοντα κορύμβους λέγουσιν, ὅθεν κόρυς, κύρβις, κυρβασία. 9 Lo sch. ad loc. glossa κόρυμβα con ἄκρα. 10 Cfr. anche Eur., Ifig. Aul. 258: ἀμϕὶ ναῶν κόρυμβα.

I percorsi della glossa  

Sch. ex. ἄϕλαστον: οὐ τὸ ἀκροστόλιον, ἀλλὰ τὸ ἐπὶ τῆς πρύμνης εἰς ὕψος τεταμένον ἐκ κανονίων πλατέων κτλ. “ἄϕλαστον: non è lo ἀκροστόλιον, bensí quell’oggetto sulla poppa allungato verso l’alto e fatto di tavolette piatte...” (segue una dettagliata spiegazione su fattura e posizione dell’oggetto, che tralascio); cfr. Eustazio 1039, 36 (ad Il. XV 717). Synagogè 172.40 Ba; Fozio, Lex. a 3369; ecc.: ἄϕλαστα: τὰ ἀκροστόλια Ἀπολλόδωρος (FGrHist 244 F 240). ὁ δὲ Δίδυμος (p. 79 et 181 Schmidt) τὸ ἐπὶ τῆς πρύμνης ἀνατεταμένον εἰς ὕψος ἐκ κανονίων πλατέων ἐπικεκαμμένων. Etymologicum Genuinum α 380, 6 ἄϕλαστον: ἔστιν οὖν ἀκροστόλιον τὸ ἄϕλαστον (Ap. Rh. I 1089) τὸ κατὰ τὴν πρῷραν, ἀλλ᾿ Ὅμηρος ἐπὶ τῆς πρύμνης αὐτὸ παραδίδωσι λέγων (O 717): ἄϕλαστον μετὰ †χεῖραν† ἔχων. Etymologicum Magnum 53, 20 ἄϕλαστον: ἔστιν οὖν ἀκροστόλιον, τὸ ἄϕλαστον τὸ κατὰ τὴν πρῷραν. Ἀλλ᾿ Ὅμηρος ἐπὶ τῆς πρύμνης αὐτὸ παραδίδωσι, λέγων ἄϕλαστον μετὰ χερσὶν ἔχων (O 717). Etymologicum Magnum 177, 46 ἄϕλαστα: ... Διαϕέρει δὲ τῶν κορύμβων. Ἄϕλαστα μὲν λέγονται τὰ πρυμνήσια· κόρυμβα δέ, τὰ πρωρήσια.

Anche senza riportare qui tutti i numerosi materiali esegetici e glossografici e commentarli puntualmente, possiamo osservare come la Synagogè ci informi del fatto che ci sia stata una divergenza di interpretazione fra Apollodoro e Didimo a proposito di ἄϕλαστον. Didimo obiettava all’interpretazione di Apollodoro, forse nell’hypomnema all’Iliade, da cui può aver preso origine il materiale erudito che si è disseminato nella successiva tradizione esegetica e lessicografica.11 Il punto di partenza per la discussione era stato l’interpretazione dei due hapax omerici κόρυμβα in Iliade IX 241 e ἄϕλαστον in Iliade XV 717. Ma intorno al significato di queste due parole (considerate in parte sinonimi, in parte non esattamente equivalenti) e all’identificazione degli oggetti precisi che esse designavano, collocati a poppa o a prua della nave, c’era senz’altro anche una discussione precedente ai due eruditi appena citati: la troviamo testimoniata in poeti alessandrini, come vedremo subito. È possibile che, nella commedia di Aristofane, assieme a κόρυμβα si menzionasse anche ἄϕλαστον? Forse nella parte perduta del v. 2 del frammento o nel prosieguo della scena? Mi sembra del tutto plausibile che nel comico si riflettessero i problemi esegetici posti da entrambi gli hapax omerici: già i rapsodi dovevano soffermarsi sull’interpretazione di parole e passi particolarmente  11 Così Erbse in apparato ad sch. Il. XV 717a, con altri materiali; cfr. Rengakos 1994, 104–105, 153–154, 156, 167; 1994a, 113.

  I percorsi della glossa difficili e forse anche per questo caso possiamo immaginare con un certo azzardo di risalire cosí indietro. Certo è che il problema rimase e lo possiamo inseguire fino ai repertori eruditi bizantini, passando per testi di varie epoche intermedie: Aristofane, i poeti ellenistici, la filologia alessandrina. Due riflessi poetici dotti di tali discussioni linguistiche e semantiche li troviamo in Apollonio Rodio e in Licofrone: autori ben anteriori a Didimo (età augustea) e ad Apollodoro (allievo di Aristarco e vissuto nel II sec. a.C.), che dunque ci fanno risalire un po’ piú indietro nella cronologia della storia esegetica. Anche queste occorrenze sono state ben analizzate, in relazione al significato preciso attribuito dagli antichi e dai moderni ai due termini.12 Apollonio Rodio II 601: nella morsa fra le Simplegadi, la nave Argo passa con l’aiuto di Atena e allora le rupi, scontrandosi l’una con l’altra, ἀϕλάστοιο παρέθρισαν ἄκρα κόρυμβα tagliarono via le punte estreme dell’aplustre.

Il raro verbo παρέθρισαν è evidentemente chiamato a variare l’omerico ἀποκόπτω, che già Eschilo aveva variato con ἀποθραύω, entrambi mantendo l’immagine del tagliare e strappare via. Ma ancora piú interessante risulta l’espressione ἀϕλάστοιο ἄκρα κόρυμβα:13 Apollonio (che a I 1089 usa il solo ἀϕλάστοιο e a III 543 il solo ἀϕλάστῳ, sempre nella stessa posizione del verso) qui ha utilizzato insieme κόρυμβα e ἄϕλαστον, usando il primo termine per indicare la parte estrema (sottolineata da ἄκρα) dell’oggetto che indica complessivamente con il secondo. Licofrone, Alexandra 295: Cassandra, dopo aver dato un’immagine estremamente negativa di Achille, sta immaginando una disordinata e rovinosa fuga dell’esercito acheo. ἀλλ᾿ ὡς μέλισσαι συμπεϕυρμένοι καπνῷ καὶ λιγνύος ῥιπαῖσι καὶ γρυνῶν βολαῖς, ἄϕλαστα καὶ κόρυμβα καὶ κλῃδῶν θρόνους πυκνοὶ κυβιστητῆρες ἐξ ἑδωλίων πηδῶντες αἱμάξουσιν ὀθνείαν κόνιν. Ma come api sconvolte dal fumo e dalle ondate di caligine e dai colpi dei legni ardenti, sugli ἄϕλαστα καὶ κόρυμβα e sui sedili dei rematori

 12 Cfr. Rengakos 1994 and 1994a, con tutti i riferimenti. 13 Sch. ad loc.: τὰ ἀκροστόλια ἀπέκλασαν αἱ πέτραι.

I percorsi della glossa  

come una massa di saltimbanchi dai loro posti balzando macchieranno di sangue la polvere straniera.

Le parole che ci interessano sono cosí interpretate in due recenti traduzioni di Licofrone: ἄϕλαστα καὶ κόρυμβα

per le poppe, le prore (Paduano)14 sugli aplustri, sui rostri (Gigante Lanzara)15

Vediamo anche gli scoli a Licofrone. Sch al v. 295 ἄϕλαστα δὲ καὶ κόρυμβα: ἄϕλαστα καλεῖται τὰ ἄκρα τῆς νεώς ... καὶ κόρυμβα δὲ τὸ αὐτὸ δηλοῖ καὶ ἔστιν ἐκ παραλλήλου τὸ αὐτό: ἢ διαϕέρουσιν ὅτι τὰ μὲν ἄϕλαστα πρυμνήσια, τὰ δὲ κόρυμβα πρωρήσια. Cfr. sopra: Etymologicum Magnum 177,46 ἄϕλαστα: ... Διαϕέρει δὲ τῶν κορύμβων. Ἄϕλαστα μὲν λέγονται τὰ πρυμνήσια· κόρυμβα δέ, τὰ πρωρήσια.

A quanto pare, dunque, i due termini sono qui accostati con diversi significati ed indicano la poppa e la prua della nave (a meno che non si tratti di un voluto gioco ambiguo o di una endiadi), seguendo e sfruttando una possibilità offerta dalla tradizione esegetica. Sicuramente, in ogni caso, un gioco dotto sui problemi semantici posti da questi termini e una testimonianza di interpretatio omerica in Apollonio Rodio e in Licofrone.16 Inseguendo i materiali glossografici legati a queste parole omeriche, un problema abbastanza interessante ci propone il lessico noto come Lexeis Homerikai e tramandato in tre manoscritti (due dell’XI sec. e uno un poco piú tardo).17 Lex. Hom. α 288 ἀκροκόρυμβα: τὰ ἄκρα τῶν νεῶν (OU: om. S)

Il lemma ἀκροκόρυμβα sembra una neoformazione, o diciamo pure un errore, sulla base di ἄκρα κόρυμβα in Iliade IX 241, che sarà dunque il passo di riferimento: come peraltro abbiamo visto sopra in Esichio α 2614 e altri lessici. Però al v. 178 del Carminis de viribus herbarum fragmentum di autore anonimo, datato su basi metriche forse al III sec. d.C. e comunque anteriore a Nonno,18 troviamo usato l’aggettivo ἀκροκόρυμβος:

 14 Fusillo/Hurst/Paduano 1991, 77. 15 Gigante Lanzara 2000, 85. 16 Cfr. Rengakos 1994; 1994a; 2002; Fantuzzi 1988, 65–85; 2001, partic. 178. 17 Cfr. de Marco 1946; Montanari 1979 e 1995 (= questo volume, cap. 22); Cavallo 1989. 18 Testo in Heitsch 1961–1964, II 23, con la bibliografia: cfr. Wellmann 1894, 2327.

  I percorsi della glossa καὶ κόπον ἐξελάσαι δύναται τριϕθεῖσα μετ᾿ ὄξους ἡ βοτάνη καταχειομένη πόδας ἀκροκορύμβους. e può eliminare il dolore la pianta tritata con aceto e versata sulla punta dei piedi.

Dobbiamo dunque ritenere, mi sembra, che l’aggettivo ἀκροκόρυμβος esistesse almeno dall’età imperiale, evidentemente come coniazione poetica di imitazione omerica. Possibile dunque che l’errore di Lex. Hom. α 288 non sia della tradizione manoscritta (e dunque da emendare), bensí un vero lapsus originale dell’estensore del lessico. È venuto il momento di rientrare da queste peregrinazioni un po’ desultorie. Tutto comincia, per noi, con un frammento di Aristofane: ma forse — lo abbiamo detto — la storia dell’interpretazione problematica di queste parole era iniziata prima, forse si cimentavano con tali questioni semantiche già i rapsodi. Poi abbiamo attraversato secoli fino a vari testi bizantini (ci sono anche le parafrasi omeriche, che non abbiamo citato, ma appartengono al dossier), passando per scoliasti, poeti dotti e testi eruditi di vario genere. Quali sono i percorsi della glossa? Essi corrono lungo le strade del sistema nervoso di una civiltà, lungo i gangli e i neuroni che portano gli impulsi attraverso i quali una cultura si esprime e fissa i propri contenuti antropologici e i mezzi per comunicarli. I percorsi della glossa si aggirano nella circolazione sanguigna della paideia, che attraverso i grandi vasi dell’espressione e la serie infinita dei capillari delle parole porta il nutrimento della cultura, nei suoi contenuti e nelle forme per esprimerli. Come una sonda armata di filologia, lo studioso deve insinuarsi nel corpo della civiltà antica, cercando e inseguendo i percorsi intrecciati, multiformi, imprevedibili che ne fanno emergere il sangue e i nervi.

 La filologia omerica antica e la storia del testo omerico È ben noto che la nascita dei famosi Prolegomena di F.A. Wolf fu strettamente legata alla scoperta e pubblicazione degli Scholia A e B all’Iliade da parte di Villoison, con tutte le informazioni che essi fornirono a proposito della storia del testo omerico e dell’attività dei filologi alessandrini.1 Dunque sono state intimamente e profondamente connessi la nascita della questione omerica in senso moderno (il problema della genesi e della formazione dei poemi secondo le prospettive impostate da Wolf) e lo svilupparsi delle ricerche riguardanti la filologia alessandrina come capitolo importante sia della storia del testo omerico che in generale della storia del pensiero antico. Forse il solo altro momento dell’omeristica moderna in cui si può percepire una interconnessione altrettanto sorprendente e una relazione così stretta fra questi due aspetti è proprio il periodo attuale, con l’intrecciarsi della discussione a proposito sia dei metodi e dei principi operativi della filologia antica sia del processo genetico che ha portato alla forma conservata dell’epica greca arcaica. In uno studio recente Martin Schmidt osserva:2 “In the second half of the 20th century research in the field of ancient criticism of Homer has advanced in a way that could not have been foreseen”, riconducendo giustamente la nuova stagione e i mutamenti di prospettiva all’opera di H. Erbse e di M. van der Valk, che “in different ways [corsivo mio], made a totally new situation [...] So many aspects of ancient Homeric philology are far better known today than some decades ago”. Schmidt sottolinea tre principali settori della ricerca in corso: “1) The always on-going debate on the value and the basis of the textual criticism of the Alexandrian grammarians. 2) Works on the development of ancient grammar in the context of Homeric philology [...] 3) [...] works on literary criticism and the ancient theories about it, with inclusion of Homer and ancient explanation of him. So the literary and rhetorical theories of the scholia [...] have been collected and classified”. I progressi della seconda metà del XX secolo proseguono in modo assai rilevante in questa parte iniziale del XXI secolo. Il primo punto isolato da Schmidt è quello che ci riguarda in questo momento. Negli ultimi anni, la discussione intorno al metodo adottato dagli Alessandrini nello studiare il testo e fare una ‘edizione’ di Omero è stata particolarmente vivace a proposito della vera natura  1 Villoison 1788; Wolf 1795; cfr. Canfora 1999; Nagy 1999; Rossi 1999. 2 Schmidt 2002, 159–161. https://doi.org/10.1515/9783110772371-041

  La filologia omerica antica e la storia del testo omerico delle lezioni attribuite dalla documentazione erudita a grammatici di primo piano come Zenodoto, Aristofane di Bisanzio e Aristarco: congetture ope ingenii e basate solo su criteri soggettivi, varianti derivate dalla collazione di copie e dunque risultato di una scelta, oppure una mistione e compresenza di entrambe le procedure? Questo punto rimane controverso e in verità si tratta del nodo centrale e assolutamente cruciale per tutte le questioni connesse. Mi sono occupato di questa problematica in alcune occasioni, ricordando e sottolineando in primo luogo il fatto che si tratta di un problema di metodo e di principi, non di quantità né di qualità dei risultati. Io credo che gli Alessandrini conseguirono la consapevolezza del fatto che un testo letterario è caratterizzato da una storia della trasmissione, durante la quale può avere subito alterazioni di vario genere e in vari punti, per cui il testo corretto deve essere recuperato e restaurato o per congettura oppure scegliendo la lezione migliore fra quelle offerte da una tradizione testuale non univoca. L’idea del riconoscimento di un danno prodottosi nel corso del tempo, ponendo il problema del testo autentico e della necessità di allestire metodi e strumenti intellettuali e culturali per ripararlo, rivela che si è stabilita l'unità organica fra interpretazione dei testi e critica testuale e che una conquista del pensiero è stata acquisita (indipendentemente dal tipo di risultati raggiunti).3 Negli ultimi anni il dibattito si è ulteriormente arricchito anche in seguito alle reazioni, in senso positivo o negativo, suscitate dalle opinioni sostenute da M.L. West,4 che segue (anche se per lo più tacitamente) una linea di svalutazione della filologia alessandrina risalente a M. van der Valk e ripresa più recentemente da H. van Thiel, ma criticata decisamente da M. Schmidt. R. Führer, J.-F. Nardelli, A. Rengakos e G. Nagy (oltre che dal sottoscritto).5 A proposito della pratica di collazionare manoscritti, A. Rengakos ha offerto un quadro sintetico che vale la pena riprendere.6 L’evidenza è molto più cospicua

 3 Montanari 1998a (= questo volume, cap. 23) e 2000a (= questo volume, cap. 38); Rengakos 2002a e 2002b; Montanari 2002a (= questo volume, cap. 47) e 2002b (= questo volume, cap. 24). 4 M.L. West 1998–2000; 2001a; e 2001b (e West 2002); alla sua edizione dell’Iliade sono state dedicate ampie recensioni critiche da Nagy 2000 e Nardelli 2001b, ai quali risponde M.L. West 2001b; cfr. anche Janko 2000. 5 Si vedano le opere citate nella bibliografia; sugli interventi di van Thiel 1992 e 1997 cfr. Montanari 1998a; Schmidt 1997 e 2002. Führer–Schmidt 2001; del lungo lavoro di Nardelli 2001a vedi soprattutto pp. 57–70. 6 Rengakos 2002b, con tutti i riferimenti. Rengakos presenta anche valide osservazioni sulle caratteristiche del testo usato come base da Zenodoto per il suo lavoro di diorthosis, in particolare sul fatto che si trattasse di un testo rapsodico ionico, sulla sua ‘eccentricità’ e sulla sua

La filologia omerica antica e la storia del testo omerico  

di quanto si tenda a presentarla. Essa comprende una serie di fattori, che si sostengono a vicenda e che non è facile eliminare in blocco: a) il confronto di manoscritti è una pratica della produzione libraria, è ben conosciuta e non ha nulla di eccezionale;7 b) collazione di copie e introduzione di varianti sono testimoniate in papiri tolemaici fin dal III sec. a. C.;8 c) nei poeti ellenistici si riscontrano prassi compositive che rinviano a variae lectiones più antiche dei maggiori filologi alessandrini e da questi ultimi citate, evidentemente trovandole nelle copie di Omero che consultavano;9 d) alcune testimonianze scoliastiche appaiono inequivocabili nel mostrare il confronto di più copie almeno da parte di Aristarco e non è una buona pratica trascurarle o non interpretarle;10 e) i filologi alessandrini fecero uso anche della tradizione indiretta, citando poeti postomerici come argomento per una lezione omerica: è pensabile che trascurassero la tradizione diretta? Su questa linea ha continuato la sua ormai tradizionale argomentazione G. Nagy. Nel più recente intervento, alle sue già diffuse argomentazioni egli aggiunge un particolare impegno nell’analizzare le testimonianze fornite dalla scoliografia omerica, con un esame del loro exact wording in serrato contraddittorio con le opinioni espresse (per vero dire, talvolta in modo piuttosto sbrigativo) da M.L. West.11 Il risultato è ancora un sostegno molto forte, con numerosi passi quali elementi di prova, all’idea che i maggiori filologi alessandrini effettivamente collazionassero copie di Omero, che Aristarco abbia usato abbondantemente questo metodo per il suo lavoro sul testo omerico e non sia stato Didimo a introdurlo. Mi pare difficile rinunciare all’idea di fondo che ne consegue, quella che è di fatto l’idea tradizionale da Wolf in poi:12 la filologia degli Alessandrini, almeno nel caso di Omero, comprendeva insieme sia l’intervento congetturale sia la selezione fra varianti provenienti da collazione di manoscritti, che era praticata già da Zenodoto e certo in misura considerevole da Aristofane e Aristarco. Il

 ‘brevità’: osservazioni che mi inducono a ripensare le posizioni che esprimevo in Montanari 2002, 123, 133–134 (= questo volume, cap. 21). 7 Cfr. Montanari 1998a (= questo volume, cap. 23) e 2002a, 121–123 (= questo volume, cap. 47). 8 Cfr. Haslam 1997, 63–69; S. West 1967. 9 Cfr. Rengakos 1993; 2002a, con il giusto riconoscimento al valore dell’opera di Nickau 1977. 10 Cfr. Rengakos 2002b; Nagy 2003, infra. 11 Nagy 2003. 12 Cfr. M.L. West 2001a, 36: “It is lime to challenge this assumption, inherited from Wolf, that collation of different copies was a normal and essential part of what Aristarchus and his predecessors did”. La reazione è stata ispirata soprattutto dalle idee di M. van der Valk, a partire dalla seconda metà degli anni Sessanta (van der Valk 1963–64).

  La filologia omerica antica e la storia del testo omerico fatto che spesso non si possa dire se una lezione è una congettura o una variante (cioè che non sappiamo quale sia il procedimento adottato in ogni singolo caso e per quali ragioni) è un difetto della nostra informazione, che non mette in discussione la sostanza del metodo e le sue implicazioni intellettuali e culturali. Non è stato Didimo il primo a confrontare fra loro manoscritti e precedenti edizioni di Omero, a raccogliere ed accostare varianti testuali, come sostiene West: Didimo è stato un grande erudito, che ha portato alla sua massima espansione un metodo già consolidato, ma egli dipende essenzialmente da Aristarco per le posizioni che esprime e per molti dei materiali che adduce (ai quali aggiunge, con il proprio gusto di raccoglitore, la considerazione dei lavori dei numerosi grammatici postaristarchei, che arricchiscono il quadro e il dibattito della filologia dell’ultima età alessandrina fino alla sua epoca). A me sembra che questo dibattito abbia un aspetto di artificialità, che bisogna focalizzare e superare. Una osservazione attenta rivela che nessuno degli studiosi recentemente coinvolti nel dibattito nega totalmente che i filologi alessandrini utilizzassero la collazione di esemplari (copie anonime oppure edizioni con una indicazione di provenienza o di paternità),13 nessuno si pone in una posizione così radicale. R. Janko non esclude affatto che gli Alessandrini procedessero talvolta sulla base di ‘manuscript evidence’ da loro consultata.14 Anche quello che è attualmente il più radicale esponente di questa posizione, M.L. West, malgrado la sua idea che Didimo sia stato il primo a collazionare diverse copie per raccogliere e valutare varianti testuali,15 a proposito dello sch. Il 9.222,

 13 Mi riferisco alle edizioni delle città o a quelle riferibili a una precisa persona (Antimaco, Riano, Zenodoto, Aristofane di Bisanzio, eccetera). 14 Cfr. per esempio Janko 1992, 22–23: “It was the task of Zenodotus of Ephesus to extract order from the chaos of MSS collected in the Museum. He evidently concluded that the then very numerous longer MSS were inferior: S. West, in her study of the extant Ptolemaic papyri, rightly agrees [...] His caution was salutary, given the abundance of interpolated texts; he certainly had MS authority for some omissions”; per Aristarco, Janko 1992, 27 e 2002, 659: “This [scil. the seemingly unequivocal statement in schol. A to Il. 9.222] certainly implies that Aristarchus did check manuscripts for variant readings”. Mi sembra interessante che Janko 1998a, 206 (in fondo alla seconda colonna), ricordando van der Valk come efficace sostenitore del fatto che gli Alessandrini introducevano congetture nelle loro ‘edizioni’, sente il bisogno di precisare: “He [scil. van der Valk] did not of course believe that all Alexandrian readings are conjectures”. Questo significa ritenere che anche van der Valk non fosse così radicale da negare del tutto l’esistenza di lezioni provenienti da “manuscript evidence”. 15 Questo è l’assunto di base nella sua visione della storia del testo in età alessandrina, ma in M.L. West 2001a, 36 si legge una sfumatura non marginale: “In fact the first scholars known to have cited manuscript authority for various readings are Aristarchus’ contemporaries Callistratus and Crates. Didymus is the first known to have compiled anything in the nature of a ‘critical

La filologia omerica antica e la storia del testo omerico  

deve fare una concessione che suona decisiva: “Didymus knew that Aristarchus consulted more than one text, because he cited different scholars’ readings on different occasions, but it is just his own assumption that Aristarchus systematically checked ‘many’ copies before discussing any reading”.16 Un sottile equivoco deve essere eliminato.17 Quando si parla di confrontare ‘altre’ o ‘varie’ copie di Omero a disposizione e utilizzabili, non bisogna aver l’aria di suggerire che si pensi a centinaia e centinaia di esemplari, attribuendo all’idea del confronto una esagerazione che diventa subito inverosimile e implausibile. Questa ambiguità viene fuori, forse in modo indesiderato, in scritti di autori che appartengono a entrambi gli orientamenti: diverse copie vuol dire alcune copie, non vuol dire centinaia. Si può parlare di confronto fra copie solo quando si superano certi numeri oppure basta confrontarne ‘alcune’? Si ritiene di poter affermare che gli Alessandrini facessero collazioni solo se si pensa a decine e decine di copie confrontate con implacabile sistematicità, oppure basta la consultazione abbastanza frequente di alcuni esemplari, soprattutto in passi problematici? Chi ritiene che gli Alessandrini non facessero collazioni, pensa a paradossali grammatici (in verità piuttosto strani) che tenacemente rifiutassero di guardare altre copie di Omero casualmente incontrate sul loro cammino, onde non vedere assolutamente possibili differenze testuali, oppure intende che lo facessero molto sporadicamente, solo qualche volta, solo per pochissime copie? Se il problema viene affrontato in modo rigido e come una opposizione radicale, si rischia di scivolare nel ridicolo: non è una buona strada. E mi pare che, affinando l’analisi di quanto scrivono gli studiosi, alla fine non se ne trovi neppure uno che si senta di spingere il suo scetticismo fino all’estrema conseguenza, cioè negare totalmente che gli Alessandrini confrontassero copie e rilevassero varianti. Non era certo ignoto il fatto che i testi di Omero non fossero tutti identici, che c’erano differenze fra i diversi esemplari. Fino a un certo punto (e certo anche in seguito, per chi non apparteneva alla élite colta) probabilmente questo non era considerato né eccessivamente fastidioso né particolarmente importante o comunque non tale da richiedere un intervento forte e  apparatus’”. Dunque due contemporanei di Aristarco: il suo rivale in serrati dibattiti (Cratete di Mallo) e un allievo di Aristofane di Bisanzio (Callistrato); cfr. M.L. West 2001a, 37 e Montanari 2001a, 134. 16 West 2001a, 37 n. 19; la testimonianza di questo scolio è da ritenersi inequivocabile e irrefutabile (cfr. Janko 2002, 659; Rengakos 2002b; Nagy 2003, 489–491), ma anche altri sono stati addotti ed analizzati ed è inutile ripetere qui cose già dette e facilmente reperibili nella bibliografia citata. A quanto pare, West non riesce a mantenere saldamente la sua posizione radicale. 17 Cfr. Montanari 2002a, 134 (= questo volume, cap. 47).

  La filologia omerica antica e la storia del testo omerico organizzato: poi qualcuno prese coscienza del fatto che si trattava di un problema molto serio e che bisognava cominciare ad attrezzarsi per affrontarlo in modo efficace.18 Vediamo ora di rivolgere l’attenzione al problema delle congetture, che sono ope ingenii, arbitrarie, basate su principi soggettivi, talora ridicole, comunque sbagliate e quindi da escludere dal testo omerico originale, corretto, genuino: con questi toni e colori piuttosto ‘caricati’ si presenta talvolta un procedimento che non dovrebbe creare un enorme scandalo nella sensibilità del filologo. Credo sia stato sufficientemente provato che i filologi alessandrini non esitavano molto a introdurre nel testo omerico lezioni costituite da loro congetture: anche su questo punto risulterebbe assai difficile sostenere una posizione che negasse del tutto la presenza di lezioni introdotte ope ingenii e senza alcuna evidenza documentaria. Se consideriamo esclusa questa possibilità radicale, l’analisi ci offre una serie di connotati interessanti. Janko sostiene che gli Alessandrini introdussero congetture proprie oppure di loro predecessori e già penetrate in precedenza nel testo omerico. “The problem, in my view, is that the Alexandrian (or Didymus, on West’s theory) tended to misjudge which type of manuscripts was actually better — those in which rhapsodes and scholars had emended away the ‘problems’ which the logic of carping critics like Zoilus and Megaclides of Athens had detected in Homer’s text, or those in which these problematic readings remained unaltered. As A. Rengakos observed,19 variants known in Hellenistic times ‘may well have been the result of earlier conjectures (from the 5th or 4th century B.C.) simply adopted by the Alexandrians’. The poet Aratus is said to have asked Timon of Phlius how he could obtain a reliable copy of Homer; Timon replied ‘if you come across old copies, and not those which have now been corrected’. We could hardly hope for better evidence that emendation had become endemic by Zenodotus’ time”20 “We can hardly know whether such conjectures are his own [scil. di Zenodoto], or derive from the fourth-century lytikoi who proposed solutions to problems raised by Zoilus and other ‘floggers of Homer’; what count is that they are conjectures, and nearly all bad”.21 Se gli Alessandrini, oltre alle proprie, accoglievano anche congetture di loro predecessori (la pratica di emendare era ormai ‘endemica’), da dove potevano

 18 In termini di giusto e sbagliato, di testo originale/autentico e testo corrotto/falso: esattamente come vogliono gli intendimenti di Janko e West nei confronti del testo omerico. 19 Rengakos 2002a, 155 n. 22, ma cfr. anche la successiva n. 23. 20 Janko 2002, 659. 21 Janko 1992, 24.

La filologia omerica antica e la storia del testo omerico  

conoscerle se non collazionando copie di Omero a loro disposizione? Questa teoria, dunque, presuppone la pratica del confronto fra esemplari (che Janko infatti ammette, come abbiamo visto). Ma allora non si può sfuggire a un passo ulteriore: quando e come gli Alessandrini potevano sapere che una lezione che trovavano in un esemplare omerico era una vecchia congettura? Dal punto di vista di chi collaziona un manoscritto e trova una lezione, questa lezione egli tratta di fatto come una variante rispetto a un’altra, a meno che non sappia che si tratta di una emendatio attribuibile a qualcuno di precedente. Anche ammettendo che tutte o quasi tutte le lezioni scelte dai filologi alessandrini fossero davvero emendamenti, potevano essi riconoscere vecchie congetture altrui oppure dobbiamo accettare che le trattassero alla stregua di variae lectiones di una tradizione non univoca22 e che imponeva una scelta? Siamo portati a riconoscere come del tutto plausibile e probabile, anzi difficilmente negabile, che il metodo filologico degli Alessandrini comportasse sia il ricorso all’emendamento congetturale ope ingenii che la scelta fra varianti documentate grazie al confronto di copie (nelle quali si potevano trovare anche vecchie congetture). Ripetiamo di nuovo: un problema di metodo, di princìpi e di atteggiamento intellettuale, non di quantità delle applicazioni o di qualità dei risultati. Vorrei ritenere sgombrato il campo dalla questione di quante copie bisogna utilizzare e quante volte bisogna consultarle perché si possa parlare di pratica della collazione di esemplari nel lavoro filologico. Certo il problema della qualità dei risultati appare un poco più pesante e forse fastidioso. Janko dice che le lezioni di Zenodoto sono “nearly all bad”;23 West trova che la lezione di Aristarco “modo melior videtur, plerumque peior”: in nota cita quattro casi di lezione migliore (una gratificata di un certissime) e commenta “de bona traditione agitur, non de coniecturis”.24 Se Zenodoto non è proprio sempre bad e può contare su un nearly, Aristarco qualche volta ce l’ha proprio fatta a scegliere la lezione giusta: dobbiamo intendere che ha capito quale era il testo omerico autentico fra quelli consultati.25 Mi sento scivolare di nuovo verso un problema quantitativo che non trovo molto attraente. Da Richard Bentley in poi, quante devono essere le congetture e le scelte giuste fra varianti per avere la patente di filologo o addirittura di buon

 22 E non è questa una difficoltà talvolta anche per gli agguerritissimi filologi moderni? 23 Janko 1992, 24. 24 M.L. West 1998–2000, vol. 1, Praefatio VII. 25 Sull’operato dei filologi alessandrini e sulla qualità dei loro risultati esistono beninteso valutazioni di altro genere.

  La filologia omerica antica e la storia del testo omerico filologo? E quante le interpretazioni accolte come indiscutibili dalla comunità scientifica? Ma cercare di recuperare e stabilire il testo corretto per via di congettura e scelta fra lezioni diverse non è esattamente il lavoro e il compito del filologo? Janko indica con estrema chiarezza una distinzione che dobbiamo seguire attentamente nello studiare la filologia alessandrina. Da una parte, egli dice, sta il problema dell’origine delle lezioni che essi propongono e quindi dei loro procedimenti di lavoro, del metodo e dei principi; dall’altra parte “my own concern, as a Homerist, has always been whether such readings are authentic”;26 una dichiarazione che penso rappresenti esattamente anche West.27 I filologi moderni possono essere severi quanto vogliono nel valutare le scelte testuali e le opinioni esegetiche degli antichi, considerandole accettabili o inaccettabili dal loro punto di vista filologico, linguistico ed esegetico: ma non possono né sopprimere né trascurare il fatto che gli Alessandrini emendavano e collazionavano copie di Omero allo scopo di ottenere il testo autentico, un metodo che combina l’interpretazione supportata dalle competenze possibili, la consapevolezza della storia della trasmissione e la critica testuale. Questa distinzione è essenziale e deve essere tenuta ben salda, cosa che purtroppo non sempre succede. Il valore e il significato storico-culturale della filologia alessandrina non devono dipendere dalle opinioni dell’omerista moderno a proposito della genesi dei poemi omerici e della costituzione del testo. E così torniamo alla stretta relazione che si percepisce fra queste due problematiche nell’omeristica attuale. Da una parte, abbiamo l’idea (Janko, West) che il testo omerico sia stato fissato molto presto, all’atto stesso della sua composizione (scritto direttamente o dettato), e abbia subito in seguito alterazioni di vario genere, dapprima  26 Janko 2002, 661: “As a scholar of Hellenistic poetry, Rengakos’ proper concern has been whether such readings were pre-Hellenistic, and his work has shown decisively that they sometimes were. My own concern, as a Homerist, has always been whether such readings are authentic, as when I wrote ‘We can hardly know whether such conjectures are his own [scil, di Zenodoto], or derive from the fourth-century lytikoì who proposed solutions to problems raised by Zoilus and other ‘floggers of Homer’; what counts is that they are conjectures, and nearly all bad’ [cfr. supra e n. 21]. I was then too pessimistic about the limits to our knowledge of preHellenistic scholarship; but the evidence, whether it derives from the analysis of Homeric diction, historical linguistics, oral composition, archaeological discovery, Linear B, or the principles of textual criticism, still demonstrates that most such readings derive from manuscripts [corsivo mio] which have been ‘improved’ by scholarly and rhapsodic interpreters, like those against which Timon is said to have wamed Aratus”; cfr. Montanari 1998b (= this volume, ch. 37). 27 Cfr. West 2001b.

La filologia omerica antica e la storia del testo omerico  

soprattutto da parte di rapsodi che lo memorizzavano imperfettamente oppure che si sentivano relativamente liberi di cambiare singoli punti, o da altri per altre ragioni. Con tale modello, la filologia omerica può contare su un punto di riferimento teorico ben preciso: il suo scopo è quello di ricostruire e recuperare quel testo, le alterazioni successive furono comunque congetturali e arbitrarie, si diffusero nelle varie copie e determinarono le divergenze nella tradizione manoscritta. Questa situazione si presentò agli Alessandrini, che però difficilmente riuscivano a recuperare il testo giusto, quello originale della prima fissazione: le lezioni da loro scelte sono in genere da respingere perché adottano alterazioni (congetture) operate da loro stessi oppure da altri in precedenza.28 A questa si contrappone l’idea (rappresentata soprattutto da Nagy) che sostiene per la genesi dei poemi un modello evolutivo, secondo il quale il testo omerico si sarebbe fissato molto più tardi e non c’erano testi scritti dei poemi almeno fino al VI secolo. Sintomo e riflesso di questa prolungata fase di trasmissione orale sarebbero proprio l’incertezza del testo e le sue fluttuazioni, testimoniate dal gran numero di varianti conservatesi nelle varie copie fino alla stabilizzazione avvenuta con Aristarco. Gli Alessandrini si trovarono di fronte questa varietà, l’hanno conservata e ne hanno discusso: facevano scelte in base alle loro idee, ma in realtà le varianti a loro disposizione riflettevano l’instabilità della fase orale di un testo, per il quale è teoricamente e praticamente impossibile scegliere l’unica lezione corretta (Nagy parla di “multitext”: ogni variante è ‘potenzialmente autentica’ una volta entrata nel sistema di una poesia intesa come performance). Secondo il modello di Janko e West, le varianti rapsodiche appartengono alla storia della ‘ricezione’ e della lettura di Omero nel corso dei secoli che seguirono alla sua nascita, sono differenze prodottesi successivamente alla formazione di quello che è l’unico testo autentico, che per definizione coincide con quello fissato alla sua origine. Secondo il modello di Nagy, le varianti rapsodiche appartengono alla storia della ‘formazione’ dei poemi, per una parte la storia formativa del testo e la storia della sua ricezione coincidono e non si possono separare: per qualche secolo non esisteva un testo, bensì un multitesto con variabili equipollenti, tutte sullo stesso piano di valore e autenticità. La particolarità del problema omerico e della sua genesi orale sono così spinte all’estrema conseguenza, che contempla un’opera esistente in una forma priva di un testo autentico bensì variabile nella performance, collocandosi sul lato opposto rispetto al razionalismo di West e Janko, secondo cui l’unica distinzione che ha senso è  28 Diversa la prassi editoriale di van Thiel, che sceglie il testo della vulgata dei codici, mostrando una decisa sfiducia sulla possibilità di risalire in modo plausibile a un testo ‘originale’.

  La filologia omerica antica e la storia del testo omerico quella fra testo autentico (nato in un momento preciso e unico) e scostamenti da esso (errori), e ci sono conoscenze e competenze sufficienti per impegnarsi a recuperarlo. Nella loro differente visione della genesi dei poemi, queste due correnti di pensiero concepiscono il testo omerico (e quindi anche il modo di interpretarne molti aspetti) in modo nettamente diverso e si contrappongono per quanto riguarda il modo in cui le varianti omeriche si sarebbero prodotte storicamente. Questo porta (a mio avviso impropriamente, quanto meno senza necessità) a una diversa valutazione della filologia alessandrina: positiva nella prospettiva di Nagy, negativa in quella di West e Janko. Per semplificare: nel primo caso, le lezioni alessandrine sono utili e importanti perché documentano l’instabilità di un testo che ha nella lunga fluidità il suo principio d’identificazione; nel secondo caso, sono solo alterazioni e allontanamenti dal testo originario che i filologi alessandrini, pur mirando proprio a questo, non sono stati capaci di riconoscere e di scartare. Per West e Janko i filologi alessandrini volevano (come loro) ricostituire il testo omerico genuino ma non sono stati capaci di farlo e hanno per lo più sbagliato metodi, scelte e prospettive; per Nagy essi non hanno ottenuto il risultato perseguito (il testo di Omero) non per loro debolezza, ma perché questo è in essenza impossibile, e tuttavia, grazie al loro metodo e al loro lavoro, hanno conservato testimonianza preziosa di questa caratteristica peculiare. Un’opinione sulla filologia alessandrina dipendente dalle idee sulla genesi dei poemi non mi sembra necessaria e nemmeno particolarmente utile. Credo invece che sia meglio scorrelare i due aspetti, distinguendo il metodo e l’oggetto. I filologi alessandrini hanno imparato e insegnato che: un testo letterario è caratterizzato da una storia della trasmissione che ha prodotto varie alterazioni; il testo corretto deve essere recuperato e restaurato o per congettura oppure scegliendo fra le varianti di una tradizione testuale non univoca; bisogna dotarsi di metodi, strumenti intellettuali e conoscenze culturali per operare in questo senso. Quale che sia il valore e l’origine delle lezioni prese in considerazione, la filologia alessandrina resta una tappa importante nella storia del pensiero. La connessione fra le notizie sulla filologia alessandrina e il problema della genesi dei poemi omerici risale a F.A. Wolf, come abbiamo ricordato all’inizio. Su questo vale la pena riprendere alcune parole di R. Pfeiffer: “F.A. Wolf acknowledged the ‘insigne meritum Villoisoni’ when he made the first attempt at a history of the Homeric text [...] One should always keep in mind his startingpoint from the wealth of the new material in the Venetian codex and the new spirit of bold historical inquiry, even if one sees him taking the wrong way in individual arguments and conclusions (corsivo mio) [...] F.A. Wolf, starting from

La filologia omerica antica e la storia del testo omerico  

the newly discovered Venetian Scholia, tried to give proofs for the new historical research step by step, in contrast to the vague generalities of the Homeric enthusiasts; he at least paved the way for the analytical efforts of the following generations of scholars who were eager to unveil the mysteries of epic stratification”.29 Sulla scia di Wolf, l’interesse e la fiducia nelle informazioni fornite dai resti della filologia alessandrina giocarono un ruolo nell’atmosfera intellettuale del pluralismo analitico ottocentesco, a caccia delle stratificazioni dell’epica greca arcaica. Il neounitarismo (più o meno fondamentalista), che si affermò nel Novecento, prestò spesso una scarsa attenzione alla filologia antica su Omero, poco utile e significativa per i suoi intendimenti, e gli studi in questo campo non fiorirono molto in quel periodo. Nella seconda metà del XX secolo (a partire dall’opera di Erbse e van der Valk) la ricerca cominciò a riprendere in considerazione questo settore su basi e con orientamenti molto diversi, e da allora i progressi continuano sempre più copiosi. Abbiamo visto come gli atteggiamenti nei confronti della filologia alessandrina si siano intrecciati in forme diverse con le teorie più recenti sulla genesi dei poemi: potremmo dire che attualmente essa trova favore da parte di teorie oraliste cui offre indizi di fluidità del testo conservando testimonianza della diacronia della sua formazione, mentre trova sfavore presso quel tipo di neounitarismo (oralista o anti-oralista) che punta la sua attenzione sulla formazione e fissazione precoce di un testo unico dei poemi, svalutandone le successive modifiche rapsodiche. Forse il quadro è eccessivamente semplificato e certamente non rende giustizia alle diverse posizioni e alle singole idee degli omeristi, che potranno precisare moltissimi punti e criticare le inopportune imprecisioni. Serve solo per concludere quella che ritengo sia stata la linea essenziale dei ragionamenti svolti fin qui. Dobbiamo fare ogni sforzo per tenere distinta la valutazione del significato e dell’importanza storico-culturale della filologia alessandrina nella storia del pensiero dal ruolo che intendiamo assegnare ai filologi alessandrini nelle nostre idee sul problema della genesi dei poemi omerici.

 29 Pfeiffer 1968, 214, 231.

 The Fragments of Hellenistic Scholarship The area of Greek literary culture that embraces the history of philology in antiquity is a highly interesting field for inquiry into the nature of a fragment, raising questions concerning the most privileged and richest sources of a given fragment, the form such sources take and the manner in which they present philological, grammatical and lexicographical fragments. Reflection on this subject also opens up interesting lines of investigation into how fragments themselves can convey information and what kind of information is effectively transmitted. I believe that the arguments I put forward in this paper will help to understand that the concept of fragment developed in this framework has a number of remarkable aspects, which can make an illuminating contribution to various levels of knowledge when facing the most widespread and diverse utilizations of fragments of poetic, historical, philosophical or rhetorical works. Over the course of history, the sphere of philology and ancient erudition has been a powerful creator of fragments and has proved a veritable goldmine in the quest for fragments of lost authors and indirect tradition of authors who have come down to us. It hardly needs be underlined that we are concerned here with fragments of ancient scholarship and not with fragments transmitted by ancient scholarship. Despite this, problems connected with the conception of authorship, the role of secondary sources and the figure of the reporter of the fragments are of paramount importance: the fragments of ancient scholarship are themselves transmitted by intermediaries, whose impact and significance cannot simply be relegated to a brief mention in parenthesis. What is a fragment in the history of ancient philology? What is a fragment of Aristarchus, undoubtedly the ancient philologist about whose work we know the most? For the sake of clarity, I will at first limit myself to examples drawn from the sphere of Alexandrian philology in the Hellenistic age. I have chosen the case of Aristarchus and Homeric philology for two excellent reasons, first, it provides a framework in which the richest possible documentation is available, and second, it also happens to be the area with which I am best acquainted. Information concerning statements by Aristarchus is frequently found in lexicographical collections (above all the Etymologica), or in miscellaneous works such as that of Athenaeus, but the richest source is surely represented by the scholia: in the most typical case, an Aristarchus fragment is composed of a  English translation by Rachel Barritt, revised by Glenn Most.

https://doi.org/10.1515/9783110772371-042

The Fragments of Hellenistic Scholarship  

scholium, not infrequently associated with a critical sign or what remains of the latter.1 In these cases the scholium mentions a textual choice and / or an exegetical statement by Aristarchus, discussing the motivations and reasons that dictated the particular choice and frequently adducing arguments that contrast with the stance taken by other philologists. This typical case is what could be termed the optimum situation and it is well represented in A. Ludwich’s collection of Aristarchus’ fragments of Homeric textual criticism transmitted by Didymus.2 In some cases, however, the sources simply contain Aristarchus’ textual choice, with or without a tiny remnant of a comment, for instance a single word on the dialectal or grammatical characteristics of Aristarchus’ reading, but devoid of any of the background critical arguments. Let us take a look at a few scholia to the Iliad (quoted according to Erbse’s edition).3 B 707 ὁ δ᾿ ἄρα πρότερος· Ἀρίσταρχ(ος) ἁ δ᾿ ἄ μ α π ρ ό τ ( ε ρ ο ς ) . (the Hawara Iliad = Pap. I Erbse, in mg.) οὕτως αἱ Ἀριστάρχου ἄ μ α , οὐ διὰ τοῦ ρ̄ ἄρα. (Did. in sch. Aint ad B 707) N 383 ⟨εἷλκε·⟩ οὕτως Ἀρίσταρχος ἕλ κ ε. (Did. in sch. Airn ad N 383 a) N 384 ⟨ἦλθεν ἀμύντωρ·⟩ Ἀρίσταρχος ἦ λ θ ᾿ ἐ π α μ ύ ν τ ω ρ. (Did. in sch. Aim ad N 384 b1) — τ ι ν ὲ ς μ ὲ ν ἦ λ θ ε ν ἀ μ ύ ν τ ω ρ γράφουσιν, Ἀ ρ ί σ τ α ρ χ ο ς δ ὲ ⟨ ἧ λ θ ᾿ ⟩ ἐ π α μ ύ ν τ ω ρ . (Did. in sch. Τ ad Ν 384 b2) Ν 399 ⟨ αὐτὰρ ὁ γ᾿ ἀσθμαίνων·⟩ Ἀρίσταρχος χωρὶς τοῦ γ̄, α ὐ τ ὰ ρ ὁ ἀ σ θ μ α ί ν ω ν . (Did. in sch. Aim ad N 399 a1) B 682 Τρηχ(ε)ῖν᾿ ἐνέμοντο· Ἀρίσταρχ(ος) Ἰακῶς Τ ρ η χ ε ῖ ν α ν έ μ ο ν τ ο . (the Hawara Iliad = Pap. I Erbse, in mg.) Ἰακῶς τὸ ἐ ν έ μ ο ν τ ο Ἀ ρ ί σ τ α ρ χ ο ς . (Did. in sch. Aint ad B 682) E 560 ⟨ἐοικότες:⟩ δυικῶς ἐ ο ι κ ό τ ε Ἀ ρ ί σ τ α ρ χ ο ς . (Did. in sch. Til ad E 560 a) H 458 ὅσον τ᾿ ἐπικίδναται {ἠώς}· οὕτως Ἀρίσταρχος, ὅ σ η ν τ ᾽ ἐ π ι κ ί δ ν α τ α ι θηλυκῶς, Ζηνοδότ⟨ε⟩ιος δὲ ἡ “ὅσον τ᾿ ἐπικίδναται”. (Did. in sch. A ad H 458) O 94 ⟨ἐκείνου·⟩ Ἀρίσταρχος “κείνου” Ἰακῶς. (Did. in sch. Aim ad O 94 a)

These examples will suffice to let us state very clearly the following question: what is the Aristarchus fragment in these cases? What is it that provides infor-

 1 Typically a ὅτι, scil. τὸ σημεῖον ὅτι, introducing the scholium. 2 Ludwich 1884–1885. 3 Erbse 1969–1988.

  The Fragments of Hellenistic Scholarship mation regarding his thought? There can be but one answer: the Aristarchus fragment consists of the Homeric word in the form the philologist selected. In other words, all the information concerning Aristarchus’ views derivable from these fragments actually comes from the words of the text that forms the object of his philological criticism, according to the reading he considered correct. Such an assertion may seem obvious, above all for those who are familiar with the field of the history of philology in antiquity, yet I feel it does embody a concept particularly worthy of note in this field, which has theoretical relevance for research in this area as well. For at times the fragment that acts as a source of information on the thought of an ancient philologist is given exclusively by a word of the author studied by the philologist, i.e. by the reading he selected or the conjecture he felt it necessary to introduce in a given passage of his author. The Aristarchus fragment is the Homeric variant and the rest must be deduced: the Aristarchus fragment exists and provides information even in the absence of words by Aristarchus himself. This situation is due to the way this material arose (activity of the Alexandrian philologists) and ultimately came down to us (transmission of the scholarly collections). The system of critical signs was perfected, as is known, by Aristophanes of Byzantium and Aristarchus, and was widely used in their philological practice.4 Let us take a few more examples from Homer, for whom we have the most abundant documentation. We will consider here a few examples from Iliad 2, with the comment by Erbse in app. B :

diple ante versum in A, fort. exstabat nota Aristonici de v. Ἄργος, ad Δ  d.

B :

diple periestigm. ante versum in A, fort. exstabat sch. Aristonici, cf. sch. δ  κτλ.

B :

diple ante versum in A, fort. exstabat sch. Aristonici, ad B  (Ariston.).

B :

diple ante versum in A, fort. excidit sch. Aristonici, cf. Eust. . κτλ.

B –:

obeli ante – in A, ante – (!) in pap. I [scil. the Hawara Iliad], fort. erat sch. Aristonici, vide ad B –.

In the above cases, all that remains of the Alexandrian exegesis to the lines in question is the critical sign in a manuscript of the poem.5 As can be seen from Erbse’s comments, one often finds at least one parallel to other scholarly material, which allows a reasonable hypothesis to be made as to why the critical sign

 4 Pfeiffer 1968; Gudeman 1922; Turner 19802, 100 ff.; Montanari 1979, 43 ff.; McNamee 1992 (with bibliography). 5 Wismeyer 1884–1885.

The Fragments of Hellenistic Scholarship  

was introduced. One can thus make a good guess as to the nature of the exegetical problem under discussion. Nevertheless, conclusions of this kind are the fruit of scholarly deductions, in the total absence of any verbal material connected to the sign itself. Given the lack of a scholiastic annotation directly connected to the critical sign, which is effectively ‘orphaned’, what we actually have is a situation in which the fragment proper becomes identified exclusively with the sēmeion. Each of the cases mentioned above lends itself to interesting considerations, but now I will restrict the scope of observations to the obeli that appear in Ven. A at B 874–5. Erbse compares these lines to B 860–1, also marked by obeli both in Ven. A and in the Hawara Iliad (and there he says: “860–1 obeli ante versus in A et in pap. I; vide ad B 874–5”). The Aristonicus scholium A at B 860–1 claims that B 860–1 are spurious, basing its argument on an internal criticism of a clearly Aristarchean slant: since B 874–5 are equivalent in situation and meaning to B 860–1 (and 860 + 861a = 874 + 875a) and likewise present the obeli, it can be deduced that Aristarchus had, for the sake of coherence, obelized 874–5 as well as 860–1; if this is the case, then one may surmise that Aristonicus repeated at 874–5 the motivation already given at 860– 1, unless, that is, he thought that the first time, at 860–1, was sufficient and there was no need to repeat it at 874–5. This is certainly a very plausible deduction concerning B 874–5: but what we actually have for B 874–5 are only two obeli remaining in Ven. A and in the Hawara Iliad, where, in any case, they are mistakenly moved one line forward (athetesis of 875–6 would not be a workable hypothesis). The critical sign as the one and only source of information concerning Alexandrian scholarship is sometimes to be found isolated not only in the great codex Ven. A of the Iliad, but also in different kind of papyrus fragments: 1) papyrus fragments of the Homeric text, which have both critical signs and scholia (ranging from fairly rich to extremely sporadic) in the margin of the text, but in some cases have a critical sign without any correspondence in the scholia of the papyrus itself or in scholia of other sources or in commentary material of various origin; 2) papyrus fragments containing just the Homeric text (no scholia), with critical signs without any correspondence in the whole corpus of the preserved scholia or in commentary material of various origin. As far as the first case is concerned, consider three more examples from the Iliad with Erbse’s comment in app. B 746: diple periestigm. ante versum in pap. I [scil. the Hawara Iliad]; incertum, an sch. Aristonici exstiterit. Ψ 657: asteriscus ante versum in pap. XIII [scil. P. Lit. Lond. 27]; qui quid sibi velit nescio. Ω 115: diple ante ν. in pap. XIII [scil. P. Lit. Lond. 27], fort. error scribae.

  The Fragments of Hellenistic Scholarship Both, the Hawara Iliad (2nd century AD) and P.Lit. Lond. 27 (1st century BC) are papyrus fragments of the Homeric text with critical signs and scholia in the margin. B 746 is particularly emblematic and deserves comment: the diplē periestigmenē is the critical sign specifically used by Aristarchus against Zenodotus, to indicate the places where he disagreed with choices made by his forerunner, as is documented innumerable times in the material available to us. The assumption that this is a case of an Aristarchus fragment preserved in the Hawara Iliad and reduced to the critical sign alone is therefore fully warranted: we know the type of content embodied therein (a polemic against Zenodotus), but not the specific content (the reasons for the polemic and the arguments adduced). The asterisk and the diplē of P.Lit. Lond. 27 show no link with the exegetic material and it does not appear feasible to put forward valid hypotheses: what we have, therefore, are two fragments of Alexandrian philology admitting of probable attribution to Aristarchus or his school but presenting a content that, at best, remains highly problematic.6 The second case is well represented by a certain number of papyrus fragments containing just the Homeric text with critical signs. In most instances, the signs are found to have a correspondence in the scholia or are justified by exegetic material from another source, but there are times when this does not occur and the semeion remains completely isolated. Let us examine a few cases. In P.Köln 37 (late 1st century AD) line Ω 92 is marked by a diplē, absent in Ven. A and lacking justification in the scholia. Erbse comments: “diple pura ante versum in pap. Colon. inv. nr. 55 (nondum edita) [scil. P. Köln 37]; quae diple quid sibi velit, non video”; B. Kramer, in the P. Köln edition, hypothesizes an annotation referring to the meaning of ἅλιον, as is found in the sch. ex. to the same line,7 but the meaning of the diplē remains obscure. In PSI 113 (5th century AD) line A 471 is marked by a wavy horizontal sign ~ of unknown meaning, line A 472 is marked perhaps by an asterisk or some similar shaped sign,8 line A 473 by an obelos, line A 474 has no sign in the margins, and line A 475 is marked by an obelos. The situation is radically different in Ven. A, which has nothing at A 471, presents the diplē at lines A 472 and 473, the obelos at A 474, and no sign at A 475: the scholia fully justify the sēmeia of Ven. A.

 6 Edition of the papyrus: Kenyon 1893; cf. W. Leaf, JPh 21, 1893, 17–24. On the two signs cf. also Martinazzoli 1953, 62–63. 7 Kramer/Hübner 1976, 87. The parallel with Herodian’s sch. Δ 26 a, indicated by Erbse in app. ad Ω 92, offers a weak support. 8 Edition of the papyrus: E. Pistelli, in Papiri della Società Italiana, II, Firenze 1913, 1–3; cf. McNamee 1992, 12 n. 12 and Table 1 p. 29.

The Fragments of Hellenistic Scholarship  

In PSI 113 there is certainly something amiss, as may happen when a copyist transcribes the signs that he finds in his model (in all likelihood understanding them only dimly or not at all): the obelos at line 475 is the result of an error, which may consist of a mere displacement from line 474 or a mistaken transcription for a paragraphos (as suspected by Allen, in app. ad loc.); again, the obelos at line 473 may be a mere displacement from line 474 or a mistaken transcription for a diplē (which is present in Ven. A); the supposed asterisk at line 472 may be a mistaken transcription for a diplē (which we find in Ven. A) or a genuine asterisk or another sign of unknown meaning; as to the sign ~ at line 471, its meaning remains shrouded in mystery. But this accumulation of errors (which affects all the critical signs preserved in the papyrus) gives rise to a suspicion: should the fragments of philology constituted by the semeia of PSI 113 simply be written off as crude errors by the scribe, or might one conjecture that, in part at least, they merely remain uninterpreted?9 The isolated critical sign, shorn of the support of any exegetical material linked to it, often constitutes a highly problematic fragment, from which it may be difficult to extract information in the absence of accompanying critical comment that can lend credibility to an interpretation. Furthermore, not only does this type of fragment give rise to problems of interpretation, but it also frequently occasions attribution problems: indeed, this is a special case of the problem of attribution of a fragment. A considerable number of such fragments is represented by Aristarchean critical signs but it is not always possible to make a definite attribution of a critical sign, for at times it may be difficult to establish to which philologist it can be traced back. Remaining within the realm of Homeric exegesis, the critical signs accompanied by Aristonicus and Didymus scholia can be attributed with a fair amount of certainty to Aristarchus, and the same holds true for the diplē periestigmenē, even when it occurs without any explanatory comment (as we saw above). In many cases, however, the conundrum has to be left unresolved. Finally, one also finds instances in which a sign in the  9 The two cases of P.Lit. Lond. 27 are mentioned by Erbse because the papyrus also contains marginal scholia (albeit extremely sporadic: only two have been preserved, in col. 22, at Ψ 842 or 845 and at Ψ 850): this papyrus is therefore considered in his edition of the Iliad scholia, but Erbse by no means provides a systematic account of critical signs encountered in Iliad papyri; he dwells (as we have seen) on the case of P.Köln 37 at Ω 92, but not on the sēmeia of PSI 113. The critical signs appearing on Iliad papyri arc reported systematically in the critical apparatus of the editio maior by T.W. Allen (Allen 1931), but naturally much has been published since then. In actual fact there is a lack of systematic collection and in depth analysis of the critical signs appearing in the margins of Homeric papyri, above all in connection with the exegetic material of ancient Homeric scholarship.

  The Fragments of Hellenistic Scholarship margin could be interpreted as a critical semeion, but there is no certainty that this is the correct reading. Such a situation may arise for horizontal lines possibly interpretable as obeli or paragraphoi, for angular signs conceivably interpretable as diplai, but also for a large number of signs of different kinds (dots, slashes or slanting lines and a plethora of other signs: one such example was shown above). At this point, the question as to whether the sign preserved in the papyrus is a genuine fragment of ancient philology or merely an occasional reading sign devoid of any true critical-philological content is a matter of conjecture.10 It may be interesting at this point to extend the range of issues under consideration here. Mention can thus be made of the accents and breathings that appear sporadically in the papyri.11 Likewise one can include the various lectional signs: different means used to indicate the separation of words or to indicate that at a given point there is no word boundary, or to highlight the difference between different parts of the text (lemma and comment, citations, turntaking in dialogue, and so forth); or again, consider the punctuation marks used to separate units within a text: examination of this aspect allows discussion of the work of a grammarian such as Nicanor on the Homeric text. The papyrus presented by A. Hanson12 offers a rather special case (which she rightly emphasizes) of use of the blank space to articulate units of thought: note that the blank space is clearly based on critical interpretation of the text. In effect — and this is no paradox — this constitutes a limit case of a fragment of ancient philology composed only of the blank space, i.e. literally of the absence of any pen-mark whatsoever. It is not my aim to delineate a complete typology, nor do I propose to offer a wealth of examples, but I believe I have clarified the concept to which I wished to draw attention: an isolated critical sign (that is, a sign lacking a link to critical remarks) represents a fragment of ancient philology that has come down to us in a truly exceptional form, a non-verbal form, as it were. Yet this peculiar form by no means precludes the emergence of exactly the same array of problems concerning correctness of transmission, interpretation, attribution and historical setting that arise for the type of fragment more commonly forming the focus of scholarly discussion. That the fragment should be reduced to the bare word of the author who forms the object of the commentary, indeed sometimes to the mere critical sign  10 A large amount of material is gathered together in McNamee 1992. 11 Breathings for ex. in connection with cases in which so-called interaspiration was being discussed: cf. Montanari 1988 (with bibliography). 12 Hanson 1997.

The Fragments of Hellenistic Scholarship  

of any type (non-verbal fragment), is undoubtedly one of the strangest situations encountered in the field of the history of philology, although far from rare in the material available to us. In both cases, it has to be said that the fragment exists and offers its information or its problems despite the utter absence of even a single word by the author of the fragment. This can be just as true in the case of an Aristarchean Homeric variant as in that of an Aristarchean sēmeion: not a single word by Aristarchus has survived, yet we have considerable information on his thought. Such a situation occurs when the motivations and documentations on which the textual choice was made, or the exegetical stance adopted, have been completely lost, whatever the causes of this loss may have been. When nothing is preserved to accompany the reading or the sēmeion, the reasons for the statement and its supporting critical arguments can only be deduced, whenever this proves possible: clearly, the degree of uncertainty involved is a consequence of the fact that deduction from a mere textual choice or even from the critical sign alone is a very special case indeed of deduction of information. Often the problem stems from the drastic selection that has occurred over time in the exegetic material, not infrequently resulting in the loss of all or virtually all that accompanied and justified a textual choice or an exegetic position. But at times the problem was already perceived in antiquity and here one encounters a most remarkable situation. The case of Zenodotus, who apparently wrote no commentaries, has been well analyzed, highlighting the difficulties encountered in trying to shed light on Zenodotus’ motivations for deciding on a particular text constitution and its interpretation. As early as the time of Aristarchus the problem was already making itself felt very clearly. Interpreting and commenting on the Homeric text sometimes provoked soulsearching doubts as to which reading Zenodotus had in fact selected and on what foundations and lines of reasoning he had constructed his argument. Such was the lack of certainty that veritable attempts were made to reconstruct Zenodotus’ opinion as regards various passages of the Homeric text. Thus in cases like these, our knowledge of what we extract as fragments of Zenodotus comes from a somewhat unusual type of mediation.13 We will return to this point at the end. * * * We will take for granted the basic distinction between direct and indirect tradition in the transmission of ancient literary texts. Chance discovery of papyri may lead, if the identification is successful, to recovery of pieces of varying length from a lost work and these pieces, of course, fully belong to the direct  13 On this problem cf. Nickau 1972 and 1977; Montanari 1988.

  The Fragments of Hellenistic Scholarship tradition. However, there remains the fact that a large part of the fragmentary literature derives exclusively from the indirect tradition. Within the indirect tradition, the underlying problem is constantly that of separating the fragment, i.e. of distinguishing what is to be attributed to the author cited from what belongs to the intermediary or intermediaries of the citation. Perhaps I may briefly recall that in the case of prose, where no assistance can be expected from metrical considerations or from interaction between metre and linguistic-stylistic elements, the problems can be extremely complex and tricky to unravel. Linguistic criteria (for example in the case of clear-cut dialectal differences) and stylistic touches, considerations based on knowledge of the content and thought (but from what sources?), other specific and contingent data: all these are frequently not sufficient to provide a fully reliable guarantee that it will be possible to make a precise delimitation of the fragment and recognize the part, if any, in which it is transcribed literally. What are the specific conditions of the products of Hellenistic philology, when examined from this aspect? First of all, it should be called to mind that the direct tradition is very scanty. One can mention Aristophanes of Byzantium, of whom we have, in extract form, parts of his lexicographical work and of the epitome of Aristotle’s Historia animalium, and Dionysius Thrax, with the problematic question of the Technē grammatikē, of which §§ 1–4 are in any case authentic.14 One can then raise the further problem of the fragments of hypomnēmata that have come down to us from papyri and have the subscription preserved. If one takes a look at the Pack2 repertory of literary papyri, a very curious phenomenon emerges: it is difficult to explain why the large fragment of commentary on Demosthenes preserved in P.Berol. 9780 (BKT I), with the subscription Διδύμου περὶ Δημοσθένους κη Φιλιππικῶν γ̄, is listed under the name of Didymus author of the commentary (Pack2 339), while the small fragment preserved in P.Amh. 12 with the subscription Ἀριστάρχου Ἡροδότου ᾱ ὑπόμνημα is classified under the name of the author forming the object of the comment, i.e. Herodotus (Pack2 483). Furthermore, the skimpy remnant P.Mil.Vogl. 1.19, bearing only the subscription Ἀπολλοδώρ[ου] Ἀθηναίου γραμματικο[ῦ] ζητήματα γραμματικ[ὰ] εἰς τὴν Ξ τῆς Ἰλιάδο[ς, is listed under Homer (Pack2 1197), even though the only information remaining is that Apollodorus of Athens had written some Zetemata on Iliad 14. Under what heading, for instance, would one list P.Oxy. 2536, which has the subscription Θέω[νος] τοῦ Ἀρτεμιδώρου Πινδάρου Πυθιονικῶν ὑπόμνημα: where will it be listed in the new edition of the repertory, under the name of the author being commented on, i.e.  14 Cf. recently Schenkeveld 1994 (with bibliography).

The Fragments of Hellenistic Scholarship  

Pindar, or under that of the author of the comment, i.e. Theon? It does not appear to me that there is any specific difference among these works by Didymus, Aristarchus, Apollodorus and Theon, and one is left with the sneaking suspicion that the guiding principle is actually the size of the preserved part, which in the case of the work of Didymus is fairly extensive.15 In contrast, the small Apollodorus frustulum seems to be clearer: the only information it provides is that the volumen from which it derives was a work by Apollodorus on Iliad 14, of which only the tiny final snippet with the title is preserved. In an edition of the fragments of Apollodorus of Athens, it must definitely be included in the direct tradition.16 Another aspect that remains to be considered is the fact that in the cases cited here, what we are dealing with are epitomes and not the original work. Doubt has been cast on this assertion only as far as the Didymus fragment is concerned, though I would argue that this distinction is not acceptable. Indeed, it is known that the fragments of hypomnēmata preserved in papyri generally offer works that are already epitomized. In sum, there is no real reason to make distinctions. The papyrus fragments of hypomnēmata or syngrammata bearing the subscription with the name of the author should all be considered in the same manner: for these are first and foremost part of the direct tradition of those philologists who were the authors of the commentaries, which means that remnants of exemplars of works by Didymus, Aristarchus and so forth have been preserved directly (even if they do not appear in their original form, but rather in epitomized form). In turn, such works include passages from the author who forms the object of the commentary and quotations from other authors as well, and act as part of the indirect tradition of these authors. The papyrus fragments of Philodemus’ work Περὶ τοῦ καθ’ Ὅμηρον ἀγαθοῦ βασιλέως belong of course to the direct tradition of the work of Philodemus and they would still be classed as direct tradition of Philodemus even if we only had a small (identified) fragment with exegetical considerations on a passage from the Iliad. At this point, however, one notes that there is one serious flaw in my line of argument, which casts doubt on the whole construction: namely, what we actually have are epitomes rather than the work in its original form. The following question can therefore be raised: if we consider Aristophanes of Byzantium’s epitome of Aristotle’s Historia animalium as a work by Aristophanes of Byzantium

 15 Pearson/Stephens 1983. 16 On this fragment cf. Pfeiffer 1968, 263; Erbse 1969–1988, vol. ΙII, 557–558.

  The Fragments of Hellenistic Scholarship (and not as part of the direct tradition of Aristotle),17 then why should the epitome by an anonymous X of Aristarchus’ Commentary to Herodotus be considered as the direct tradition of Aristarchus and not a work by the anonymous X? Is there a difference or was there a fundamental mistake in the argument that would consider the epitomes of hypomnēmata of a known author as belonging to the direct tradition of works of the philologist-author? I fear this is a question on which it is impossible to give a clear-cut and cogent decision, for it opens up the hornet’s nest of how epitomes should be regarded in the history of tradition and in text constitution. Naturally, on the theoretical level we cannot rest content with answers based merely on whether the name of the epitomator is known or not. When one attempts to be more precise, it appears that the only criterion valid both in theoretical and practical terms is that of an approach based on specific differences, distinguishing between different types of epitomes or excerpta. These range from cases in which a simple ‘cut and paste’ operation is carried out without modifications or interventions, to those in which the epitomator re-elaborates the text on which he is working and thereby introduces variations. This means that a definition can become certain only when the epitome and the epitomized work are both preserved, at least for a portion sufficiently large to allow a well-founded idea to be reached through comparison. But this is by no means the case with the exegetic products of Hellenistic philology. Consequently, my enthusiasm manifested in defining fragments of excerpta from hypomnēmata by Aristarchus or Theon as direct tradition of these authors will have to be dampened somewhat. At the very least, such an approach should not misleadingly convey the idea that we possess portions of the original work of the Alexandrian philologists. Naturally, it should be stated that we have an excerptum of a work by Aristarchus or Theon, but this should not lead to the illusion that we have literal segments of their critical assessment, i.e. their exact wording. We will now turn to the principal source of the fragments of Hellenistic philology, the collections of scholia (but a similar line of reasoning, mutatis mutandis, will also apply to the great lexicographic collections of the Byzantine age). As an example, let us take the richest and best-known case. Aristarchus’ work on  17 Another case is that of the epitome of the Aristotelian Politeiai made by Heraclides Lembus (and preserved in manuscripts together with the Varia Historia by Aelianus): Dilts 1971. Heraclides Lembus himself made an epitome of several works by Hermippus, an epitome of which P.Oxy. 1367 has preserved a fragment (with the colophon: Ἡρακλείδου τοῦ Σαραπίωνος ἐπιτομὴ τῶν Ἑρμίππου περὶ νομοθετῶν καὶ περὶ ἑπτὰ σοφῶν καὶ Πυθαγόρου): Gallo 1975, 35–39; Idem, in Corpus dei Papiri Filosofici, I 1**, Firenze 1992, 249–257: here the fragment is classified as Hermippus 1.

The Fragments of Hellenistic Scholarship  

Homer was utilized by Aristonicus to explain the Aristarchean sēmeia, and by Didymus to provide an account of Aristarchus’ text constitution. The works of Aristonicus and Didymus (in the original form or already in epitome?) were then utilized by a grammarian, who merged and re-elaborated them together with those by Nicanor and Herodian, producing the well-known VMK (Viermännerkommentar). Through some other intermediary, or possibly two or three intermediaries, an abundant quantity of this material subsequently found its way into the Ven. A manuscript (and a lesser quantity into other manuscripts of scholia to the Iliad). We read a scholium, which utilized x —> y —> z, which utilized VMK, which utilized Aristonicus and Didymus, who both utilized Aristarchus. The aim of Erbse’s edition is that of reconstructing the models from which were derived the scholia that we read in the manuscripts, and indeed he could not do otherwise. The Aristarchus fragment that we read in a scholium is thus basically a ‘citation’, already re-elaborated a number of times in successive utilizations. The fragments of Aristarchus transmitted by the scholia should therefore properly be viewed as indirect tradition, cited by other authors. At best, they could be considered as deriving from the work of Aristonicus or the unknown grammarian of VMK, but in actual fact we are not even in possession of these, and Aristarchus’ original work is much further away than would be assumed to be the case with, say, a citation of Pindar in Plato, since in the latter case we have the work of the first citer directly preserved. The great quantity of materials that have come down to us from Hellenistic philology is recovered from erudite sources coming under the following typology: corpora of scholia and lexicographic collections of the Byzantine tradition, or fragments of hypomnēmata and syngrammata preserved by papyri; naturally one also finds citations of the Hellenistic philologists in other kinds of works (for example miscellanea such as the Deipnosophistai of Athenaeus), but this does not change the situation greatly. For whatever the situation, we are dealing with an indirect tradition that has passed through a number of stages, and in such cases the exact wording of the philologist is generally fairly distant from what we read. If there exist cases in which we can claim to have been able to identify the ipsissima verba of a Hellenistic grammarian, such cases are few and far between and inevitably problematic; and the same is true, as we have seen, of the little we have that bears a closer resemblance to a direct tradition of their works, the papyrus fragments with an undisputed subscription. The philologists of the Hellenistic age are certainly not the only ancient authors whose original works have not come down to us and regarding whom we cannot therefore be certain whether and to what extent we have words that can be argued to have been preserved literally and therefore to be the ipsissima

  The Fragments of Hellenistic Scholarship verba of the lost author: similar problems frequently occur for philosophers and historians as well, for instance, but there is also a broader area that should be considered separately. This is the area of literature that may be termed technical-scientific or professional, which ranges over fields such as medicine, geography, mathematics, astronomy, mechanics, military treatises and so forth, but which also extends to the field of rhetoric, grammar and philology. In these spheres the formation and application of a technical language plays a particularly crucial role, closely connected with the definition of conceptual tools. In the field of philology, the situation outlined here concerning the transmission of materials over time strongly affects our knowledge of the technical language and therefore also of the precise conceptual tools of Hellenistic philology. Analysis of the available material, which is not as scanty as is sometimes thought, reveals the extensive presence of a set of expressions and technical terms used in philological criticism as well as in grammar and rhetoric. We do therefore have a genuine thesaurus of technical language. But the most serious problem lies in the attribution of the terms to a specific philologist and therefore also in the establishment of a precise chronological series. More simply stated, we can readily highlight a technical language of philology from the material available, but we generally have difficulty in stating exactly to whom it belongs and sometimes also exactly what it means, because we have lost the original works of the Hellenistic philologists and, in addition, their fragments normally do not preserve the ipsissima verba.18 One extremely famous case is that of the σαφηνίζειν of Aristarchus. Pfeiffer carefully reconsidered the celebrated maxim Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν, which was traditionally attributed to him. Pfeiffer argued that this epigraphic and icastic statement undoubtedly expresses a fundamental principle of Hellenistic philological-exegetical practice and is therefore well suited to summarize its procedures and intentions: however the formulation itself, or shall we say the abstraction of the methodological principle, is — Pfeiffer contended — not due to Aristarchus but rather to Porphyry himself who cites it without indicating an author.19 Pfeiffer’s position stirred up considerable debate, leading to statements of diverging views by a number of scholars.20 I believe that a substantial new contribution to the documentation can be made by consideration of a Pindaric scholium, which does not appear to have received due attention so far.

 18 Together with these reflections it is useful to read the remarks of Humphreys 1997. 19 Pfeiffer 1968, 225–227. 20 Wilson 1971; Lee 1975; reply by Wilson 1976; Schäublin, 1977; Porter 1992, 76 ff.

The Fragments of Hellenistic Scholarship  

Sch. Pind. Οl. 2. 152 α. φ ω ν ᾶ ν τ α σ υ ν ε τ ο ῖ σ ι ν · ἀκουστὰ τοῖς συνετοῖς. β. ἀντὶ τοῦ φανερά, σαφῆ καὶ ἐκτιθέμενα. γ. ὁ δὲ Ἀρίσταρχος οὕτω· διάδηλά φησιν ό Πίνδαρος τοῖς συνετοῖς τῶν ἀνθρώπων εἶναι, εἰς δὲ τὸ κοινὸν ἀγόμενα ἑρμηνέως χρῄζειν τοῦ σαφηνίζοντας αὐτά, ὡς οὐ πᾶσι καταδήλως φράζων· ὥστε τοῖς μὲν σοφοῖς σοφὰ διαλέγεσθαι καὶ μὴ ἔκθεσμα, τοῖς δὲ ἰδιώταις μὴ κατάδηλα γίνεσθαι. δ. τινὲς δὲ ὅτι πολλὰ ἔχων λέγειν, εἰς τὸ πάντα εἰπεῖν καὶ διατίθεσθαι προσδεῖται ἑρμηνέων· ὡς Ὅμηρος· ‘πληθὺν δ᾿ οὐκ ἂν ἔγωγε μυθήσομαι’ (Β 488).

When I encountered this evidence, I was carried away by the first flush of enthusiasm and concluded that the question was resolved: Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν was genuinely a maxim by Aristarchus. However I subsequently became convinced that greater caution should be exercised: can we really be sure that the scholiastic formulation ὁ δὲ Ἀρίσταρχος οὕτω introduces Aristarchus’ exact wording? Overjoyed with optimisim, we answer in the affirmative: but in this case, what does this scholium actually tell us? It tells us that Aristarchus utilized the verb σαφηνίζειν in a technical sense, but it gives no conclusive proof for the complete maxim Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν presented by Porphyry without attribution. As a matter of fact, the overall documentation, which I have analyzed elsewhere,21 does seem to allow us to conclude fairly plausibly that the maxim in question was a methodological principle genuinely stated by Aristarchus. Yet in this case too, despite the excellent evidence from the Pindaric scholium and other less explicit material, we cannot go beyond the confines of probability, which is rather strong for σαφηνίζειν only, and somewhat less convincing for the entire maxim. Another case, briefly: χαρακτήρ used as a term to define an author’s style appears in scholia in which Aristarchus’ opinions are outlined (as for ex. Aristonicus in sch. A ad Il. 24.614–617), “although I suppose that this does not necessarily reproduce Aristarchus’ own terminology”.22 The question is of importance precisely because of the history of the technical terminology and its meaning. Here too the situation would appear to be quite promising, inasmuch as we have one or more Aristonicus scholia and therefore material that is certainly Aristarchean. However we cannot even be sure whether the literal formulation is due to Aristonicus, let alone to Aristarchus. * * * The issues surrounding the fragments and the proper definition of what is a fragment of Alexandrian philology in the Hellenistic age are unquestionably  21 In my article “Aristarco e il σαφηνίζειν,” in preparation. 22 Richardson 1992, 359.

  The Fragments of Hellenistic Scholarship highly complex and I hope I have been able to provide a few significant examples of some of the interesting points raised by this kind of inquiry. Let me now make a few final remarks. Many of the problems stem from two specific facts: first, at times the earliest philologists did not produce extended commentaries, so that it soon became difficult to grasp their arguments; second, the original works of those who did write more extensively (such as Aristarchus and Didymus) were very soon subjected to epitomization and replaced in current use by more compact versions, which served the useful purpose of gathering together miscellaneous materials intended for a refined educated public or for advanced school use. Now, even allowing for all the necessary distinctions, it appears to me that some of the problems dealt with here in this perspective began to crop up from earliest antiquity. For what else is Aristonicus’ work on Aristarchus’ critical signs, or that of Didymus on Aristarchus’ textual readings, if not in some sense a collection of fragments of the great philologist's work on the Homeric poems, presumably put together not much more than a century after his death? These works focused primarily on Aristarchus and neglected many of the observations of Aristophanes of Byzantium, presumably because the latter was frequently in agreement with Aristarchus, but preserved many of Zenodotus because Aristarchus’ citations from the latter author bore a highly polemical thrust: there is good reason to suggest that such a work in effect was the very first collection, which thereafter exerted overwhelming influence over the selection of the fragments that have come down to us. A grammarian of no great fame, Ptolemy Epithetes, a contemporary of Aristarchus, wrote a work dedicated to gathering together the readings chosen by Zenodotus: indeed it was defined as ὁ ... τὰς Ζηνοδότου γραφὰς ἐκτιθέμενος, in all likelihood (as can be gathered from the few fragments that remain) because precise knowledge of Zenodotus’ work was being lost, and it was therefore becoming particularly difficult to follow the critical arguments he had put forward.23 That we are not too far from a collection of Zenodotean fragments on Homeric text constitution is a very compelling idea. If these considerations are correct, then we have here something that differs both from the groups of fragments of authors present in works by grammarians, metricists and the like used for purposes of exegesis or exemplification (therefore occasional and lacking any systematic criteria), and also from the miscellaneous works and theme-based florilegia and from the doxographic sylloges as well. Conceivably, we have the very first examples of a collection of fragments that were intended to reconstruct as faithfully as possible a work that no longer existed or was on the verge of being completely lost.  23 Cf. Montanari 1988.

 Die Berliner Wachstafeln P. 10508 – 10512: Scholia minora und grammatikalische Passagen (zusammen mit Davide Muratore und Fabian Reiter) Unter den Wachstafeln der Berliner Papyrussammlung ragt eine Gruppe von vier zu einem Kodex gehörigen Tafeln heraus, die von singulärer Bedeutung sind:1 Die Tafeln, deren Herkunft unbekannt ist, bilden das umfangreichste der wenigen bislang bekannten Beispiele für die Überlieferung von Homer-Scholien auf Holz- oder Wachstafeln.2 Im Jahre 1907 wurden sie von Ulrich von Wilamowitz-Moellendorf und Wilhelm Schubart kurz beschrieben, die bereits erkannten, daß die Tafeln Worterklärungen zu Passagen aus dem 11., 13., 14. und 15. Gesang der Ilias enthielten, und die Schrift in das 2. Jh. n.Chr. datierten.3 Aristide Calderini (1921) publizierte dann aufgrund von Transkriptionen Schubarts die lesbaren Textpassagen auf je einer Seite der Tafeln P. 10508, 10509 und 10511, während die am stärksten abgeriebenen Tafeln P. 10510 und 10512 unbearbeitet geblieben sind. Nach der partiellen Erstedition durch Calderini haben die Tafeln zwar Eingang in die Kataloge schulischer und scholastischer Texte von Giorgio Zalateo, Lucia Μ. Raffaelli und Raffaella Cribiore gefunden, sind ansonsten aber nicht näher erforscht worden.4 Im Rahmen eines vom Deutschen Akademischen Austauschdienst (DAAD) und dem Ateneo Italo-Tedesco geförderten Gemeinschaftsprojektes (VigoniProjekt) zwischen Ägyptischem Museum und Papyrussammlung (ÄMP) Berlin und der Universität Genua arbeiten wir an einer vollständigen Edition der Tafeln,

 1 P. 10508; 10509; 10510 und 10511 + 10512. Die Maße der Tafeln sind 15,5 x 11,8 cm. Der Holzrand hat eine Breite von ungefähr 0,9 cm. Die Schriftfelder sind 13,7 x 10 cm groß. 2 Vgl. bisher die Berliner Holztafel P. 11636 (3.–4. Jh. n.Chr.), publiziert von Plaumann (1913), neuediert von Raffaelli 1990: Il. 5, 265–286, sowie die jüngst von Guido Bastianini und John Lundon publizierte Wachstafel P.Bingen 8 (1. Hälfte 2. Jh. n.Chr.): Il. 1, 405–436, außerdem die von Hombert/Préaux 1951 publizierte Holztafel Oxford Ashm. Mus. Bodl. Gr. Inscr. 3017 (2.–3. Jh. n.Chr.) mit Paraphrasen zu Il. 4, 349–363 und Glossar zu Il. 4, 364–371. Listen aller bisher bekannten Wachstafeln bieten Brashear/Hoogendijk 1990 sowie Cauderlier 1992; vgl. für unsere Tafeln Nr. 259–262. Vgl. jetzt auch Worp 2012, 12–13, Nr. 32–35. 3 BKT V.l.S. 6. 4 Zalateo 1961, Nr. 283–286; Raffaelli 1984, Nr. 53, 54, 52 und 48; Cribiore 1996, Nr. 326–329. https://doi.org/10.1515/9783110772371-043

  Die Berliner Wachstafeln P. 10508 – 10512 die in einem Band der Berliner Klassikertexte (BKT) publiziert werden soll.5 Im folgenden präsentieren wir erste Resultate der Kooperation.

Inhalt und Aufbau der Tafeln Tafel P. 10508 enthält auf beiden Seiten Homer-Scholien: Auf dem bisher unpublizierten Rekto, d.h. der Seite, auf der die Löcher sich auf der linken Seite befinden, sind einzelne Erklärungen zu Wörtern aus dem 10., 11. und 13. Gesang der Ilias verzeichnet. Die Lemmata und Scholien sind hier in großzügiger Weise in fünf verschiedenen Blöcken geschrieben, die durch horizontale Striche und Freiräume voneinander getrennt sind. Auf dem Verso dagegen finden sich in systematischer Weise Scholien zu einer zusammenhängenden Partie von Il. 14, 227–521 sowie zu einem einzelnen Vers aus dem 13. Gesang (612). Der Aufbau der Scholien ist schlicht und verändert sich nicht: Auf das homerische Wort folgt direkt eine «Übersetzung» in prosaische Sprache, von der durch einen kurzen Schrägstrich das nächste Lemma abgetrennt ist, z.B. in Z. 15: πλῆτο προcεπέλαcεν / 14, 438. Akzente, Spiritus und weitere diakritische Zeichen finden sich weder hier noch auf den anderen Tafeln. Nur eine weitere Tafel ist nach Zusammensetzung der beiden Fragmente P. 10511 und 10512 vollständig. Hier sind auf der Versoseite (Löcher rechts) in dicht aufeinander folgenden Zeilen homerische Wörter aus dem Bereich von Il. 11, 19– 263 mit ihren Erklärungen geschrieben. Auf dem Rekto dagegen, dessen Oberfläche zu großen Teilen abgerieben und daher unleserlich ist, befinden sich keinerlei Scholien. Vielmehr scheint die in Z. 1–2 von uns entzifferte Passage τ̣ α̣c̣ c̣ ό̣ μεν̣ α̣ δέ̣ μετ̣ ά̣ τ̣ ῶ̣ν̣ φωνηέντ̣ ω̣ν̣ | φων̣ ή̣ν̣ ἀ̣π̣ο̣τ̣ ελεῖ einem Satz aus dem grammatikalischen Werk (τέχνη γραμματική) des Dionysios Thrax zu entsprechen (1, 1, 11, 2–3 Uhlig): cύμφωνα δὲ †λέγονται, ὅτι αὐτὰ μὲν καθ᾿ ἑαυτὰ φωνὴν οὐκ ἔχει, cυνταccόμενα δὲ μετὰ τῶν φωνηέντων φωνὴν ἀποτελεῖ. Da die erste Zeile der Tafel mitten im Satz einsetzt, muß es sich um die Fortsetzung von einer anderen Tafel handeln. Ob wir ein Originalzitat des Dionysios oder nur eine gleichlautende Stelle aus einem anderen grammatikalischen Werk vor uns haben, ist bisher unklar. Die Identifizierung der Stelle ist von hohem Interesse, da

 5 Außer den genannten Institutionen danken wir der Holzrestauratorin des ÄMP Margarethe Pohl für eine Sicherungsrestaurierung der fragilen Originale, der Photographin des ÄMP Sandra Steiß für hochwertige Streiflichtaufhahmen der Stücke.

Die Berliner Wachstafeln P. 10508 – 10512  

die einzigen direkten Zeugnisse für Dionsysios Thrax P.Hal. inv. 55a und PSI I 18 erst aus dem 5. Jh. n.Chr. stammen.6 Allerdings bietet auch P.Osl. II 13 (= Wouters 1979, Nr. 9; 2. Jh. n.Chr.), die Versoseite eines Papyrus, der auf dem Rekto (P.Osl. 12) Homer-Scholien enthält, grammatikalische Passagen, die zu einem guten Teil dem unter Dionysios’ Namen überlieferten Text entsprechen; diese befassen sich unter anderem gerade mit der auf der Wachstafel nachweisbaren Thematik.7 Die Zusammenstellung von grammatikalischer Abhandlung und Scholia Minora begegnet auch in der genannten Parallele P.Bingen 8, die vor den Scholien Passagen aus einer grammatikalischen Abhandlung enthält, die angesichts der Definition der Konjugationsklassen mit Dionysios Thrax verwandt zu sein scheint, sowie in PSI I 18 und 19 (Dionysios Thrax und Ilias-Glossar), die zu ein und demselben Kodex gehören. Die Tafeln P. 10509 und 10510 sind beide nur zur Hälfte erhalten. Auf P. 10510 finden sich auf der einen Seite Scholien zu Il. 10, 599–799. Die andere Seite enthält auf dem Kopf stehend einen zusammenhängenden Text. Einige von uns entzifferte Stichwörter wie cύμφωνα und φωνήεντα, δίχρονα und δίcημα scheinen dafür zu sprechen, daß wir hier einen grammatikalischen Traktat vor uns haben. Die Drehung um 180 Grad könnte ein Indiz dafür sein, daß die beiden Beschriftungen auf Vorder- und Rückseite nicht in unmittelbarem Zusammenhang standen. Der Wechsel der Schriftrichtung erschwert die Bestimmung von Rekto und Verso, denn beide Seiten haben die Löcher für die Bindung auf der rechten Seite. Die Analogie zu Tafel P. 10511 + 10512, wo auf dem Verso die Scholien und auf dem Rekto der grammatikalische Text stehen, und zu P. 10508, wo ebenfalls auf dem Verso die systematischen Scholien verzeichnet sind, legt jedoch die Vermutung nahe, daß in gleicher Weise auch P. 10510 auf dem Verso die Scholien trägt, auf dem Rekto kopfstehend dazu Passagen einer grammatikalischen Abhandlung. Auch P. 10509 ist auf den beiden Seiten in umgekehrter Schriftrichtung beschrieben, auf der einen Seite mit Scholien zu Il. 15, 17–191, auf der anderen möglicherweise wiederum mit einem grammatikalischen Text, doch die wenigen auf der abgeriebenen Oberfläche bisher von uns entzifferten Wörter χρονικήν und πνευματικά sind in dieser Hinsicht keineswegs eindeutig. Wiederum ist angesichts des Befundes der anderen Tafeln zu vermuten, daß die Scholien auf die Versoseite geschrieben waren, der andere Text kopfstehend auf das Rekto.  6 Vgl. Bastianini/Lundon, Einleitung zu P.Bingen 8 mit Anm. 9. 7 Vgl. die Definition von Konsonanten und Vokalen und deren Unterteilungen in P.Osl. 13, iv, 17 – v, 31.

  Die Berliner Wachstafeln P. 10508 – 10512

Abfolge der Tafeln Die Frage nach der Abfolge der Tafeln und der verschiedenen Beschriftungen ist nicht leicht zu beantworten.8 Immerhin spricht die identische Struktur der Scholientexte auf den Versoseiten der Tafeln dafür, daß die vier Tafeln, wahrscheinlich zusammen mit weiteren, ein einheitliches Schreibheft bildeten. Da die Tafeln von verschiedenen Händen beschrieben sind, besteht keine Notwendigkeit für die Annahme, daß sämtliche Texte zu demselben Zusammenhang gehörten. Vielmehr könnten sie aus verschiedenen Stadien der Beschriftungsgeschichte stammen. Möglich scheint etwa das folgende Szenario der Abfolge der Beschriftungen: a) Zunächst könnten die Rektoseiten des Wachstafelheftes für unzusammenhängende Homer-Scholien zu Wörtern verschiedener Gesänge genutzt worden sein, wovon sich nur noch das Rekto von P. 10508 erhalten hätte. b) Darauf mag der Besitzer die Versoseiten für die systematische Niederschrift kontinuierlicher Scholien entlang des Homertextes verwendet haben, von denen die vier Versoseiten erhalten blieben. Ob die Reihenfolge der Aufzeichnungen der Abfolge der Gesänge entsprach, und wie viele Tafel fehlen, läßt sich nicht erschließen. c) Mehrere Rektoseiten könnten zu einem späteren Zeitpunkt ausgewischt worden sein, um Passagen aus einem oder mehreren grammatikalischen Werken aufzunehmen. Mindestens zwei Tafelvorderseiten wurden vor der Neubeschriftung auf den Kopf gedreht. Die angeführten Parallelen P.Bingen 8, P.Osl. 12+13 und PSI 1 18+19 mit Erklärungen des Homertextes in naher Umgebung von grammatikalschen Passagen gibt zu berücksichtigen, daß beide Themen durchaus von einem und demselben Benutzer des Wachstafelbuches behandelt worden sein könnten. Als sicher kann die vermutete Reihenfolge der Beschriftungsstadien nicht gelten. Die weitere Entzifferung mag jedoch zu einer Präzisierung oder Modifizierung der Hypothese verhelfen.

 8 Einkerbungen des Randes (vgl. Cauderlier 1992, 68), die die Abfolge der Tafeln des Kodex verdeutlichen würden, wie es bei dem Berliner Kodex P. 14000 (= SB III 6215–6218) der Fall ist, sind nicht zu erkennen.

Die Berliner Wachstafeln P. 10508 – 10512  

Die Scholien im Licht der bisherigen Zeugnisse Die Berliner Wachstafeln stammen wahrscheinlich aus dem 2. Jh. n.Chr. und gehören damit in den Zeitraum, aus dem uns der größte Teil an Zeugnissen für Scholia Minora zur Ilias überliefert ist. Die Statistik von Lucia Raffaelli, die auf den von den jeweiligen Editoren vorgeschlagenen Datierungen beruht, erkannte unter den 54 damals bekannten Zeugnissen 35, die dem 2.–3. Jh. n.Chr. zuzuordnen waren (64%).9 Die seitdem publizierten Papyri haben dieses Bild nicht wesentlich geändert: 24 Texte sind von ihren Herausgebern auf das 2. Jh., 9 auf das 2./3. Jh. und 15 auf das 3. Jh. datiert worden; insgesamt sind dies 47 von den 73 bisher bekannt gewordenen Zeugnissen, d.h. etwa 66%. Bemerkenswert an den Berliner Tafeln ist abgesehen vom Material die Tatsache, daß sie Scholien zu Partien späterer Gesänge des Epos bieten, für die bisher nur wenige Zeugnisse an Scholien vorliegen.10 In den dreißig Jahren seit Raffaellis Studie haben sich diesbezüglich die Proportionen allerdings in einigen Punkten geändert. So besitzen wir etwa heute 18 Zeugnisse an Scholien zu Gesang 2 (= 23% aller Papyri mit Ilias-Scholien) gegenüber 7 bekannten im Jahre 1984 (= 12,5% der Gesamtsumme), und wir haben zwei Zeugnisse für Gesang 19 sowie eines für Gesang 22, für die 1984 kein einziges überliefert war. Weiterhin ohne Scholienpapyri verbleiben die Gesänge 12, 18 und 23–24. Bezüglich der von den Tafeln abgedeckten Ilias-Passagen kennen wir nur einen einzigen Papyrus mit Scholien zu Gesang 10 (P.Mil.Vogl. III 119), einen mit Scholien zu Gesang 13 (P.Ryl. III 536), zwei weitere für Gesang 15 (P.Berol. inv. 13230 und P.Vindob. inv. G 39940v) und drei weitere für Gesang 11 (P.Alex. inv. 28759, P.Med. inv. 72.13 und P.Amh. II 19). Für Gesang 14 ist unsere Tafel P. 10509 das einzige bekannte Zeugnis. Die Textpartien der genannten Zeugnisse überlappen sich allerdings nicht mit den Passagen der Berliner Tafeln.

Häufigkeit der Scholien Bezüglich der Frequenz der Scholien macht Raffaelli insbesondere auf drei Tendenzen aufmerksam: erstens, daß die Frequenz der erklärten Lemmata sich mit der Abfolge der Gesänge allmählich verringere; zweitens, daß die Dichte an Scholien, Hinweis auf eine höhere Sorgfalt bei der exegetischen Arbeit, zu Beginn

 9 Vgl. Raffaelli 1984, 155–157. 10 Vgl. Raffaelli 1984, 158.

  Die Berliner Wachstafeln P. 10508 – 10512 der Gesänge am höchsten sei; und drittens, daß die Frequenz der Glossen in den jüngeren Zeugnissen höher sei als in den älteren.11 Auf den Berliner Wachstafeln ist die Situation folgendermaßen: P. 10508 enthält auf dem Verso — vom noch unpublizierten Rekto, welches wie oben gesagt verstreute Scholien zu den Gesängen 10, 11 und 13 enthält, sehen wir momentan ab — Scholien zur zweiten Hälfte von Il. 14, 227–521. Ungefähr 300 Versen sind 32 Scholien gewidmet, zu denen sich ein Scholion zu 13, 621 gesellt. Durchschnittlich ergibt sich also eine Frequenz von gut einem Scholion pro 10 Verse. Tatsächlich aber verdichten sich die Scholien in drei Blöcken, nämlich in den 45 Versen von 227–271 13 Scholien (= 3 Scholien / 10 Verse), in den 40 Versen von 333–372 6 Scholien (= 1,5 Scholien / 10 Verse) und in den 109 Versen von 413–521 14 Scholien (1,3 Scholien / 10 Verse). Zwei größere Textpassagen, die Verse 272–332 und 373–412, bleiben ohne Erklärungen. Auf P. 10509 sind auf dem Verso mindestens 20 Scholien zu Il. 15, 17–191 (2,3 Scholien / 10 Verse) lesbar, auch hier allerdings in ungleichmäßiger Verteilung: In Z. 2–6 finden sich mindestens 7 Scholien für die 11 Verse von 17–27 (6,4 Scholien / 10 Verse), danach läßt die Frequenz stark nach (einzelne Scholien zu den Versen 42, 84, 91, 106, 132, 137, 140, 153, 180, 185 und 191).12 Auf P. 10510v haben wir bisher knapp 30 Scholien für Il. 13, 599–799 identifiziert (ca. 1,5 Scholien / 10 Verse), die relativ gleichmäßig auf die Partie verteilt sind.13 In den ersten 15 Zeilen von P. 10511+10512v (der Rest der Tafel ist nicht sicher lesbar) sind etwa 27 Scholien zu den 116 Versen von Il. 11, 136–251 identifizierbar (= 2,3 Scholien / 10 Verse). Auch hier sind zwei Blöcke zu erkennen, nämlich die Verse 136–183 mit 19 Scholien (4,2 / 10 Verse) und die Verse 253–261 mit 8 Scholien (9 / 10 Verse), zwischen denen eine größere Passage (184–252) unkommentiert bleibt. Das Auftreten mehr oder weniger ausgedehnter Passagen ohne Scholien hat Parallelen sowohl in der papyrologischen Überlieferung als auch in der mittelalterlichen Überlieferung der D-Scholien. In P. von Scherling G 99 finden sich 12 Scholien zu Il. 9, 454–468. Die Verse 458–461 bleiben ohne Scholien. Es sind die berühmten Verse, in denen Phoenix über die Möglichkeit nachdenkt, den eigenen

 11 Vgl. für diesen Punkt bereits Calderini 1921, 314. 12 Eine ähnliche Situation, wenn auch mit höherer Scholienfrequenz, begegnet etwa in P.Ryl. III 537 (4. Jh. n.Chr.), der auf dem Rekto 9 Scholien für Il. 5, 5–11 enthält (13 Scholien / 10 Verse), und auf dem Verso 8 Scholien für die Verse 37–53 (4,7 Scholien / 10 Verse). 13 Scholien zu den Versen 599, 612, 620, 622, 625, 630, 634, 649, 657, 669, 685, 686, 703, 715, 725, 736, 747, 754, 782, 793, 797 und 799.

Die Berliner Wachstafeln P. 10508 – 10512  

Vater zu ermorden, die von Aristarch eliminiert wurden und in der mittelalterlichen Tradition des Epos nicht enthalten sind.14 In P.Oxy. XLIV 3159 (3. Jh. n.Chr.) folgen auf eine Zusammenfassung des 7. Gesangs in Prosa in der zweiten Kolumne 17 oder 18 Scholien zu den Versen 4– 80 (?) desselben Gesangs. Es fehlen gänzlich Scholien zu den Versen 10–25 — einer Passage, die auch in der mittelalterlichen Tradition geringe Erläuterung erfahren hat — und zu den Versen 65–79 (65–69 sind auch in den D-Scholien ohne Glossierung).15 In P.Haun. I 3 (3. Jh. n.Chr.) dagegen, der 17 Scholien zu den Versen 1–48 des 5. Gesangs bewahrt (3,6 / 10 Verse), fehlen gänzlich Scholien zu den Versen 26–38, ohne daß diese Lücke einer entsprechenden in der mittelalterlichen Scholienüberlieferung entspräche. Auch den Passagen ohne Scholien in den Wachstafeln entsprechen keine Lücken an Scholien in der mittelalterlichen Tradition.

Reihenfolge der Scholien Bezüglich der Abfolge der Scholien gibt es auf den Wachstafeln einige Vertauschungen zu verzeichnen. Auf P. 10508 und P. 10511 + 10512 finden sich je drei Irrtümer dieser Art: a) Σ14, 335 πεφράδοι steht versehentlich hinter dem zu 14, 340 ἐπεί νύ τοι. b) Σ14, 414 ἐξερίπῃ steht zwischen Σ14, 413 cτρόμβον und Σ14, 413 περὶ δ᾿ ἔδραμε. c) Vor dem letzten Scholion zu 14, 521 ἐπιπέρθεcθαι ist ein Scholion zu 13, 612 πελέκκω̣ eingefügt. d) Σ11, 155 ἐν ἀξύλω̣ und Σ11, 156 εἰλυφόων stehen vor dem zu 11, 137 ἀμείλικτον. e) Σ11, 165 cφεδανόν (-νων Taf.) findet sich nach Σ11, 172 φοβέοντο. f) In P. 10511 + 10512v, 12–15 scheint die Reihenfolge gestört: Σ11, 26 (?) ὀρωρέχατο (-έχετο Taf.), glossiert mit ὄρoυcαν; 11, 252 [νύξε] | ἔτρωcε oder 11, 261 [ἀπέκοψε] | ἔτρωcε; 11, 248 ἀριδείκετοc λογιώτατοc; 11, 257 ὄπατρον πρὸc ὁμόπατρον; 11, 251 cτῆ δ᾿ εὐράξ· ἔc̣ τ̣ [η δ᾿ ἐκ πλα]γίων; 11, 263 (?) ἔδυν{ε}· ἔδυ{να}cα̣ν̣ . In Ρ. 10509v und 10510v sind derartige Irrtümer nicht nachweisbar. Auch für diese Art von Vertauschungen sind Parallelen in der papyrologischen Überlieferung zu finden, wenngleich die Fehlerquote auf den Wachstafeln etwas  14 Vgl. Apthorp 1998. 15 Für 10–25: ein Scholion zu Vers 12, zwei zu Vers 15, jeweils eines zu den Versen 22 und 24.

  Die Berliner Wachstafeln P. 10508 – 10512 höher als in der Mehrheit der anderen Zeugnisse zu sein scheint. Immerhin gibt es mit P.Turner 13 (P.Strasb. inv. gr. 39, 40, 41) und P.Köln inv. 2281 auch Papyri mit leicht erhöhter Rate an irrtümlichen Versetzungen von Lemmata und Scholien.

Charakteristika der Scholien a) Die Scholien auf den Berliner Wachstafeln bestehen meist aus dem homerischen Wort und einer einzigen Glosse. Allerdings lassen sich auch einige Ausnahmen anführen. Einige Lemmata bestehen aus mehreren zusammengehörigen Wörtern: P. 10508r, 11–11a (Σ11, 416) γν]α̣μπτῇc γένυccιν· ἐν ταῖc κεκαμμένα[ιc] | cιαγόcι; 14 (11, 424) κατὰ] π̣ρ̣ό̣ τ̣ μηc̣ ιν· {/} μεταξὺ τοῦ ὀμφάλου ; Ρ. 10508v, 1–2 (14, 232 und 235) ἔν̣ τ̣ ᾿ ἄρα οἱ φῦ ιρί· ἐνε̣ φυ δὴ̣ | τῇ χειρὶ αὐτοῦ / ἐγὼ δέ κέ τοι ἰδέ `ω´ · ἐγὼ δ᾿ ἄν cοι εἰδείην χάριν ; 8– 9 (14, 333 und 340) πῶc κ᾿ ἔο̣ [ι]· πῶc ἂν εἴη / ἐπ̣ί νύ τοι ἐπιδή cοι; 17 (14, 484) τ̣ ὼ καί δ̣ι̣ὸ̣ κ̣α̣ί̣ ; Ρ. 10510v, 3–4 (13, 612) ἀμφὶ πελέκῳ· περὶ cτειλειῷ ; 11–12 (13, 703) ἐν νειῷ ἐπὶ ἀρο[ύρᾳ]; 15 (13, 728) [περιίδμεναι ἄλλων· π]εριcότερον εἰδέναι τῶν ἄλλων ; Ρ. 10511+10512v, 2 (11, 155) ἐν ἀξύλῳ· ἐν πολυξύλῳ ; 4 (11, 153) χαλκῶι δηό̣ ωντεc· τῶι χ[α]|λκῶι διακόπτο[ντεc]; 5–6 (11, 160) [κείν᾿ ὄ]χ̣ε̣α̣· [κενὰ] ἅρματα ; 14–15 (11, 251) στῆ δ᾿ | εὐράξ· ἔc̣ τ̣ [η δ᾿ ἐκ πλα]γίων. b) In einigen Fällen ist ein homerischer Terminus mit einer Umschreibung wiedergegeben. P. 10508v, 16–17 (Σ14, 463) λικριφ{ρ}ίc· ἐπὶ τ̣ ὰ̣ πλέ|τια (l. πλάγια); 19 (14, 499) κώδει̣ αν μήκωνοc κεφαλήν ; Ρ. 10509, 5–6 (15, 27) ἀτ]ρύγετον τὴν πολὴν θάλαccαν ; Ρ. 10510v, 9–10 (13, 685) ἑλκεχί[τωνεc· βαθὺν τὸν] χιτῶνα ἔχονταc ; 20–21 (13, 793) ἀμοιβοί· ἐκ διαχῆc. c) Selten sind die Fälle doppelter Glossierung.16 P. 10508r, 1 (Σ10, 94) [ἀλαλύκτημαι τε]θ̣ ορύβημαι πλανῶ[μαι]. Auch die D-Scholien haben hier eine zweifache Erklärung, die nur teilweise mit unserer übereinstimmt: ἀλαλύκτημαι· τεθορύβημαι, ἠπόρημαι. Immerhin stimmt die auf der Wachstafel überlieferte Glosse mit dem Beginn der Glosse von ἀλαλύκτημαι im Lexikon des Apollonios Sophistes (p. 22, 32) überein.17 P. 10508v, 21–22 (Σ14, 521) ἐπιπέρθε̣ c̣ θ̣ αι  16 In P.Bingen 8 (P.Mil.Vogl. inv. 8), der anderen Gruppe von Wachstafeln mit Scholia Minora zu Il. 1, 405–436, gibt es nur einen Fall doppelter Glossierung eines homerischen Terminus. Seite C 11–13: αὐτοῦ αὐτόθι, ἐν αὐτῷ τῷ τόπῳ. Die Herausgeber erwägen, daß «si affaccia perciὸ la possibilità che l’estensore abbia compilato le interpretazioni attingendo a più fonti oppure che abbia operato una selezione da un testo più ricco». 17 Vgl. z.B. Suda α 1061 (= L. Zon. p. 134, 8); Ἀλάλημαι: τεθορύβημαι, πεπλάνημαι. καὶ Ἀλαλύκτημαι ὁμοίως.

Die Berliner Wachstafeln P. 10508 – 10512  

ἐπακολουθῆcαι, ἐπιδιῶξαι. Zu dem Lemma, welches nicht dem überlieferten Homertext entspricht, der vielmehr ἐπιcπέcθαι hat, vgl. unten. Die Glosse scheint derjenigen bei Apollonios Sophistes (p. 74, 6) und beinahe derjenigen in Σex. 14, 521b, ἐπισπέσθαι ἐπιδιῶξαι, ἐπακολουθῆcαι τῇ τῶν ποδῶν ὁρμῇ, zu entsprechen. Die D-Scholien erklären den Ausdruck mit ἐπακολουθῆcαι und in Teilen der Tradition auch mit τρῶcαι. Die Glosse ἐπακολουθῆcαι ist die einzige, die auch in den Lexeis Homerikai ε 701 De Marco (= van Thiel) und in Hesych ε 5210 erscheint. d) In drei Fällen weicht das Lemma von dem überlieferten Homertext ab. In P. 10508v, 4 (Σ14, 463) hat die Tafel ὑπιcχοίηc ὑπίcχεc für ἐπιcχοίηc ἐπίcχεc. Die Lesart, die hier durch die Glosse ὑπίεχεc sichergestellt ist, ist bisher weder in den Homerkodizes noch in der indirekten Überlieferung bezeugt. In P. 10508v, 5 (Σ14, 245 [oder 248]) begegnet die einfache Form εὐνήcαιμι für das Kompositum κατευνήcαιμι des homerischen Textes. Es ist unklar, ob es sich um eine beabsichtigte Simplifizierung handelt; sie findet sich auch im Homerlexikon des Apollonios Sophistes (p. 79, 8 εὐνήcαιμι κοιμήcαιμι ‘οὐδὲ κατευνήcαιμ’, ὅτε μὴ αὐτόc γε Κρονίων᾿). Dagegen ist κατευνήcαιμι weder in den von Erbse herausgegebenen Scholien, noch in den Scholia Minora, noch in den Lexeis Homerikai Objekt von Erklärungen. In P. 10508v, 21 (Σ14, 521) hat die Tafel ἐπιπέρθε̣ c̣ θ̣ αι, der überlieferte Text lautet ἐπιcπέcθαι. Das Kompositum ἐπιπέρθεcθαι ist bisher nicht bezeugt (die einfache Form πέρθεcθαι immerhin in Eustath. 961, 34). An eine Variante mag man nicht denken, zumal die Bedeutung nicht paßt, aber auch ein schlichter Fehler schiene nicht leicht zu erklären. Die homerischen Textpassagen auf den Wachstafeln sind relativ korrekt, doch es finden sich eine Reihe von Abweichungen, die sich folgendermaßen ordnen lassen: a) Vokalismus: P. 10508v, 4 ἰλαπινάζω̣ν̣ ; 6 δμήτιρα, δαμάζευcα ;18 8 ἄγρ{ι}ει ; 13–14 π̣ε̣[ρ]ιεδ{ε}ινήθη{ι} ; 16–17 πλέτια Taf., l. Πλάγια ;19 Ρ. 10509v, 7 πημανει κακο ; 16 ὑπεκλ{ε}ῖναι ; 10510v, 5 απήν, 6–7 cχήcεcθαι für cχήcεcθε ; Ρ. 10511–10512v, 4 δηϊω̣ντεc ; 10 π{ε}ίδακαc.

 18 Für -ευ- anstelle von -ου- (δαμάζουcα) vgl. Gignac 1976, 216. 19 Für ε anstelle von α in akzentuierter Silbe vgl. Gignac 1976, 281.

  Die Berliner Wachstafeln P. 10508 – 10512 b) Konsonantismus:20 α) irrtümliche Konsonantendopplung: Ρ. 10508v, 7–8 τελ{λ}έccαι τέλ{λ}ειν ; β) Konsonantenkürzung: Ρ. 10509v, 5 πολήν ; 16 ὑπεκλ{ε}ῖναι ; 10510v, 3 πελέκω̣ ; 10510v, 15 περιcότερον ; γ) Vereinfachung von Konsonantenverbindungen: Ρ. 10508v, 12 c̣ τ̣ ρ̣ό̣ βον ; δ) Dissimilation von Nasalen: Ρ. 10508v, 12 [ῥό]νβον ; Ρ. 10509v, 13 cυ̣νλήμψεται. Interessant ist auch die Schreibung λικριφ{ρ}ίc in P. 10508v, 16, wo die Liquida irrtümlich in einer zusätzlichen Silbe eingefügt ist.21 c) Haplographie/Silbenauslassung: Ρ. 10508v, 1 ιρί ; P. 10510v, 20 διαχῆc ; eine überzählige Silbe dagegen anscheinend in P. 10511–10512v, 15 ἔδυ-{να}cαν. d) Fehler bzw. Auslassungen der Endung: P. 10508r, 10 περιεξυεμένοι ; 11 γν]α̣μπτῇc.

Die Berliner Wachstafeln und die mittelalterliche Scholienüberlieferung Wenn sich die glossierten Partien der Berliner Wachstafeln, wie oben ausgeführt, auch nicht mit anderen papyrologischen Zeugnissen überlappen, so sind die Übereinstimmungen mit der mittelalterlichen Scholienüberlieferung dagegen zahlreich. Insbesondere mit den D-Scholien, die zuweilen zusätzliche Erklärungen enthalten, gibt es enge Entsprechungen:22 – P. 10508v, 1 (Σ14, 227) σεύατο (σεύατο) ὥρμηcε ; 3 (240) θρῆνυν ὑπο̣π̣όδι̣ ο̣ν̣ ; 4–5 (241) / ἰλαπινάζω̣ν̣ · ε̣ ὐ̣ω̣χ̣[ο]ύ|μενοc ; 5–6 (249) τεῇ· c{ε}η (τεῇ· τῇ cῇ) ; 9 (335) π̣εφράδ̣οι φράcει̣ εν (+ δηλώcειε, εἴποι) ; 10–11 (360) παρ|ήπαφε παρηπάτηcε (+ παρέπειcεν) ; 12 (413) c̣ τ̣ ρ̣όβον [ῥό]ν̣ βον ; 12–13 (414) ἐ̣ ξερίπῃ· ἐκπέcῃ ; 15–16 (457) cκηπτόμενον̣ ἐπερειδόμενον̣ (+ cκηριπτόμενον) ; 18–19 (497) ἀπήραξεν ἀπέκο|ψ̣ε̣ν (+ ἀπετίναξεν) ; 19 (499) κώδει̣ αν· μήκωνοc κεφαλήν. – Ρ. 10509v, 7 (Σ15, 42) πημανει· κακο [vgl. Synag.] (+βλάπτει) ; 12 (132) ἀναπλήcαc ἀναπληρώcα ; 13 (137) μά̣ρψεν cυ̣νλήμψεται (+ἐπὶ τιμωρίᾳ) ;  20 Gignac 1976, 154 ff. mit weiterer Literatur auf 154 Anm. 4; für die Verdoppelung von λ vgl. insbes. 155–156; für die Kürzung 155, mit Beispielen für Formen von πολύc, wenn auch später; für die Kürzung von κκ vgl. 160; für die von cc 158–159. 21 Für die Einfügung des ρ vgl. Gignac 1976, 108: in allen Beispielen bis auf eines (ποcάρκειc für ποcάκιc) findet sich der Fehler vor (cπρόρου, καθάρπερ, φλυραρήcω) oder nach (φρέατροc, γραφρούcηc) einer Silbe mit regulärem ρ. 22 Das zusätzliche Material in den D-Scholien ist mit «+... » gekennzeichnet. Varianten in D sind in Klammern verzeichnet.

Die Berliner Wachstafeln P. 10508 – 10512  

–

–

–

15 (153) θυόεν τεθυμιαμένον (θυόεν νέφος· τὸ εὐῶδεc καὶ τεθυμιαμένον) ; 16 (180) ὑ]πεξαλ̣ έ̣ α̣c̣ θαι· ὑπεκλεῖναι (+ φυγεῖν) ; 16–17 (185) ὑπέρ|[οπλον ὑπερ]ήφανον. Ρ. 10510v, 4 (Σ13, 620) λείψετέ θην· καταλείψετε δή ; 7 (634) ἀτάcθαλον ἄδικον ; 9 (669) θωή]ν· ζημίαν (+ τὴν) ; 9–10 (685) ἑλκεχί|[τωνεc βαθὺν τὸν] χιτῶνα ἔχονταc (ἑλκεχίτωνεc βαθεῖc χιτῶναc ἔχοντεc, ἄζωcτοι) ; 15 (728) [περιίδμεναι ἄλλων· π]εριc-ότερον εἰδέναι τῶν ἄλλων (περιίδμεναι· περιccῶc εἰδέναι). Ρ. 10511+10512v, 1 (Σ 11, 136) προcαυδήτην· προcε̣ φώ̣[νου]ν (+προcεφθέγγοντο̖ ) ; 2 (156) εἰλυφόω̣ν̣ · εἰλῶν (+ταράccων) ; 6 (172) {ἐ}φοβοντο· ἔφευ|γο{υ̣}ν (φοβέοντο· ἐφοβοῦντο, ἔφευγον) ; 13–14 (257) ὄπατρον] | πρὸc ὁμόπατρον (ohne πρόc). Ρ. 10511+10512v, 13 (Σ 11, 248) ἀριδείκετοc· μ̣ε̣γάλ̣ ω̣[c - - scheint übereinzustimmen mit den Lexeis Homerikai α 738 van Thiel (= 733 De Marco) ἀριδείκετοc μεγάλωc ἔνδοξοc, vgl. außerdem Hsch. α 7199 ἄγαν ἔνδοξον AS ἢ μεγάλωc (S) ἀcπαcτόν, φανερόν (S). Die D-Scholien erklären dagegen folgendermaßen: πάνυ ἔνδοξοc, εὐπρεπέcτατοc.

In anderen Fällen findet sich eine exakte Übereinstimmung bzw. die nächstliegende Erklärung im Lexikon des Apollonios Sophistes oder in anderen lexikographischen Quellen: – P. 10508v, 5 (Σ14, 245 oder 248) εὐνήcαιμι κοιμήcαιμι, vgl. oben. – P. 10509v, 10–11 (Σ15, 128) ἠλεέ μ̣ά̣[τα]|ιε, vgl. Ap. Soph. p. 83, 19 (aber auch ΣVOd. 2, 243 ἠλεὲ] ἠλίθιε καὶ ἀνόητε τὰc φρέναc, μάταιε und die weiteren im Apparat der Edition von F. Pontani zitierten Parallelen).23 – P. 10510v, 8–9 (Σ13, 657) ἀνέcαντεc· ἀνακαθίcαντεc. Die Glosse für den Terminus lautet in ΣbΤ 13, 657a (III p. 525 Erbse) ... τὸ δὲ ἀνέcαντεc ἀντὶ τοῦ καθίcαντεc. τινὲc δὲ ἀναθέντεc· νεκρὸc γάρ ἐcτιν, während die D-Scholien an der Stelle ἀναθέντεc, ἀναβιβάcαντεc und die Lexeis Homerikai α 576 van Thiel (α 524 De Marco) ἀναπείcαντεc haben. Eine identische Glosse wie auf der Wachstafel findet sich in Hesych (α 4962), in der Synagoge (α 602), im Lexikon des Photios (α 1873), in der Suda (α 2317) und außerdem im Kommentar des Eustathios zu unserem Passus (952, 59–60 καὶ ὡc οἱ τοιοῦτοι ἑταῖροι αὐτὸν ἐc δίφρον ἀνέcαντεc ἤγουν ἀνακαθίcαντεc καὶ ἀναθέντεc, ἀπῆγον εἰc Τροίαν ἀχνύμενοι). – In anderen Fällen dagegen gibt es keine Übereinstimmung mit den bekannten Scholien, sondern die Wachstafeln bieten eigene Erklärungen, wenn 23 Pontani 2007, 316.

  Die Berliner Wachstafeln P. 10508 – 10512

–

–

–

–

–

–

gleich die gedeuteten Termini oftmals auch in den D-Scholien, den Lexeis Homerikai oder der lexikographischen Tradition behandelt sind: P. 10508v, 6 (Σ14, 259) δμήτιρα· δαμάζευcα. Δμήτειρα wird in den Scholia vetera und den D-Scholien mit δαμάcτρια wiedergegeben, ebenso in ΣGenev., Hsch. δ 2068, Etym. Magnum p. 281, 9; Lexicon Casinense δ 56 †δητεῖρα δαμάcτρια; Suda (δ 42) δαμάτειρα· ἡ δαμάcτρια (vgl. daneben auch ΣDΙl. 14, 259a), aber auch mit ἄρχουcα (D, ΣGenev.) und δαμαcτική (Hsch. δ 2067). Für eine auf den Stamm von δαμάζω zurückgehende Paraphrase vgl. auch Eustath. 981, 12 und 991, 37. P. 10508v, 6–7 (Σ14, 261) ἀ̣πο̣ θ̣ [ύμ]ια· δυcάρεcτα. Die üblichsten Glossen sind ἀηδῆ καὶ οὐκ ἀρεcτά (ΣD, codd. YQX van Thiel), μὴ καταθύμια, λυπερά, ἀπαρέcκοντα / προcάντη τῇ ψυχῇ, ἐχθρά. Die Glosse der Wachstafel ist bisher nicht als solche zu ἀποθύμια bezeugt, erscheint allerdings als Erklärung für andere Termini (δυcπέμφελοc, δύcτροποc etc.). Für P. 10508v, 13–14 (Σ14, 413) πε̣ ρ̣ὶ δ᾿ ἔδρ̣α̣με· π̣ε̣[ρ]ιεδ{ε}ι̣ νήθη gibt es keine Parallelen. In den D-Scholien ist περὶ δ᾿ ἔδραμε glossiert durch περιῆλθεν δέ, πέριξ. Im übrigen scheint die Erklärung poetisch vgl. z.B. ΣDΙl. 17, 680 δινήθην. ἐδινοῦντο. ὅ ἐcτι περιεcτρέφοντο; ΣDΙl. 22, 165 περιδινηθήτην. ἐκύκλωcαν, περιῆλθον; auch in den wenigen lexikographischen Belegen findet sich δινέω als zu erklärendes Lemma, nicht als Element der Erklärung. P. 10509v, 2 (Σ15, 17) ἐπαύρηαι· ἐπονάcθηc. Die D-Scholien haben an der Stelle ἐπαύρῃαι· ἐπιψαύcῃ, ἀπολαύcῃc, die Lexeis Homerikai ε 677 van Thiel (= De Marco) ἐπαύρῃ· ἐπιτύχῃ (bezogen auf 11, 391). P. 10510v, 6–7 (Σ13, 630) cχήcεcθαι (l. cχήcεcθε)· κωλ[υθήcεcθε. Der Terminus wird in den D-Scholien mit ἀφέζεcθε (v.l. φεύζεcθε) erklärt, in den Scholien von Meliteniotas des Genavensis gr. 44 mit ἐφέζεcθε und bei Apollonios Sophistes mit ἐπιcχήcεcθε. Die Erklärung der Wachstafel ist nichtsdestotrotz gebräuchlich in Bezug auf andere Stellen, insbesondere in der in die D-Scholien eingegangenen Tradition, vgl. ΣDIl. 11, 819 cχήcωcιν· ἐπίσχωσι, κωλύcωcιν; 12, 4 σχήσειν ἀνθέξειν, κωλύcειν; 12, 107 cχήcεcθαι· cχεθῆναι, κωλυθῆναι ; 13, 151 cχήcουcι· ἐφέξουcι, κωλύcουcι; 17, 182 cχήcω· κωλύcω, außerdem Eustathios, der 1920, 15 in Bezug auf Od. 22, 70 οὐ γὰρ cχήcω· ἀνὴρ ὅδε χεῖραc schreibt; τὸ δὲ cχήcει ἐνταῦθα ἀντὶ τοῦ ἐφέξει, κωλύcει, vgl. auch Hsch. c 3001 σχήσειν· ἀνθέξειν, κωλύcειν, παύcειν (auch α 8601 ἀφέξει· ἀποcτήcει, ἢ ἀποκωλύcει ἢ ἀποcχήcει). Ρ. 10510v, 18 (Σ 13, 754) νιφόεντι: ὑψηλῷ. Das Wort wird weder in den DScholien noch in den Lexeis Homerikai erklärt. Die Scholien bei Erbse erläutern nur, daß das Wort «schneereich» verwendet werde, weil es sich um Berge handele, die Schnee trügen. In ΣΗes. Theog. 953 wird das Wort mit χιονώδει,

Literaturverzeichnis  

–

λαμπρῷ erklärt (χιονώδει auch in ΣΝic. Ther. 440a, während es an einer Stelle in ΣΝic. Al. 252b mit λευκῷ, οἷα τῷ ὀπῷ τῆc cύκηc erklärt wird; vgl. Phot. Lex. p. 301, 6 νιφόεccα cελήνη· λευκή; Hsch. ν 599 νιφόεcca Ἑλένη: ἀντὶ τοῦ λευκή). Ρ. 10511+10512v, 3 (Σ11, 137) ἀμείλικτον wird nur auf unserer Wachstafel mit ἄκοcμον erklärt; vgl. ΣDΙl. 11, 137 ἀηδή, cκληράν; ΣDΙl. 21, 98 ἀπηνῆ, cκληράν ; Lexeis Homerikai α 442 van Thiel (= 431 De Marco) πικρόν, οὐ προcηνῆ; Ap. Soph. 25, 11 πικράν ; Hsch. α 3546 und 3548; Synag. α 373–374 usw.

Die Aufstellungen zeigen, daß die Scholien auf den Wachstafeln teilweise identisch mit bereits überlieferten Scholien sind, teilweise auf homerische Lemmata bezogen sind, die das Interesse der antiken Erklärer auf sich gezogen haben, aber andere Deutungen bieten und teilweise auch Termini erklären, die anderweitig nicht Gegenstand der Exegese waren. Dieser Befund ähnelt dem Bild, welches sich anhand der papyrologischen Tradition der Scholia Minora zeigt.

 Homeric Papyrology: Themes, Problems, and Perspectives Homeric papyrology is nothing more and nothing less than a chapter in Homeric studies as a whole, as a broad and varied field of research. This is such an obvious claim that it immediately compels us to move on to consider and analyse, in concrete terms, what the place and role of papyrology is within the overall framework of Homeric studies, what problems it significantly contributes to through the provision of data and sources, and what issues it crucially affects. Since we are dealing here with many specific aspects, which cut across several areas of enquiry, I should note that I will not be discussing all of them. Indeed, certain aspects I will only briefly mention, while many others will remain in the background: for convenience’s sake, I might say that I do not have enough room to consider everything, but for the sake of honesty I must say that with regard to several topics I am not competent enough to treat them at length. As is natural, I shall chiefly be focusing on the Iliad and the Odyssey: I will ease my conscience by mentioning right from the start that the substantial dossier we are about to explore also includes the Hymns,1 Lives, and Certamen, on which — I fear — I will not have time to dwell. First of all, it is all too evident that Homeric papyrology provides important contributions for the textual history of the Iliad and the Odyssey, and for the history of their interpretation. We will therefore be discussing data concerning, on the one hand, Homer’s text as such and, on the other, the various kinds of exegetical material pertaining to it. We will find that, unsurprisingly, these two sets of issues often intertwine — something that is neither peculiar to Homer (for the same thing can be said about other authors) nor to data drawn from papyri (for the same applies to the whole range of codices and scholia, to mention just the most obvious example). What is peculiar to the testimonies provided by papyri, instead, is everything that concerns the material aspects related to the history of books in the ancient world, i.e. before the Byzantine period. Of course, this concerns not only Homer, but many other great literary figures as well: indeed, it concerns every fragment of a copy of a work by an author, whether he be known or unknown. Some of the issues that come into play here are the number of books per scroll, scrolls featuring more than one book, the relation between the division into  1 A papyrus with the Hymn to Dionysus that is of great interest for the arrangement of the text was published by Hurst 1994 and later as P.Gen. III 118. https://doi.org/10.1515/9783110772371-044

Homeric Papyrology: Themes, Problems, and Perspectives  

books and the Homeric poem’s distribution across the scrolls that make up a complete copy, the switch from scrolls to codices and the forms in which both of these occur, the arrangement of editions produced when codices were in use into one or more volumes, along with various other issues pertaining to the material aspect of the Homeric poems as books. While I am keenly interested in studies of this sort, I am not an expert in the field, so I will not touch upon this problem, which others are investigating with genuine competence. Alongside papyri with the Homeric text, scholars speak of a broad set of written material collectively defined as ‘Homerica’. We might try to provide a ‘negative’ definition by stating that it includes anything that is not a self-standing specimen of the Homeric text, but is somehow related to Homer. Evidently, this is a mare magnum to which scholars often attempt to bring some order, with uneven and — above all — never exhaustive results. This time too I will try to provide as much orientation as possible, while being aware that I am bound to leave out a few pieces of the picture, given how broad and complex it is. One broad area in Homeric papyrology is school texts. These provide material and information pertaining to two spheres that are important for our knowledge of the ancient world. On the one hand, they rank among the sources used to reconstruct the socio-historical context, as they throw light on a sphere that is crucial for any ancient or modern society, namely that of education and training, starting from the most elementary level; they thus help reveal the cultural level of the various strata of the population. On the other hand, school papyri in a way also fall within the field of the history of ancient Homeric scholarship and exegesis, despite their mostly simple and elementary character. Indeed, this does not at all rule out forms of juxtaposition and osmosis with respect to texts of a higher level, such as treatises or commentaries; it is precisely for this reason that scholars frequently harbour doubts and uncertainties as to whether a certain fragment is a school text or not. Consider paraphrase, a device typical of schools that nonetheless can also occur in texts belonging to different genres, as a useful instrument in exegetical reasoning and the process of interpretation. One case that is highly revealing in this respect is that of Homeric glossaries. Sometimes these are rashly labelled as purely scholastic texts (and indeed they often are, for example when they occur on tablets from school notebooks, such as those in Berlin),2 but they frequently present correspondences with exegetical texts of a different nature. It is important to distinguish between glossaries and paraphrases: the latter are characterised by a continuous and ‘rhetorically’  2 Calderini 1921; Montanari/Muratore/Reiter 2012 (= this volume, ch. 43).

  Homeric Papyrology: Themes, Problems, and Perspectives developed discourse, by contrast to the fragmenting of the text into lemmas made up of individual words and expressions which we find in scholia. In my view, when we are dealing with a series of poetic words followed by an interpretamentum, we should avoid speaking of ‘paraphrase’, even when we have a sequence of words or expressions covering the whole poetic text and showing signs of continuity (such as elisions) in the succession of lemmas: for the main thing is that lemmata and interpretamenta alternate. A paraphrase will instead take the form either of a continuous text or of an interlinear one, which is to say a text made up of lines interposed between the Homeric verses, as in the 1st-century BC PSI XII 1276. Glossaries (often called scholia minora, as we shall see) are the most widespread Homeric papyri after the textual fragments of the Iliad and the Odyssey. This leads us to two topics with respect to which Homeric papyrology — while not unique or different from that of other authors — no doubt represents the most striking, or even most surprising, example. The two topics are reference lists and what we may described as the ‘known unpublished text’. Without going too far back in time, we find that already the index which Lucia Raffaelli published in 1984 listed sixty witnesses for the Iliad and ten for the Odyssey. Two lists available online have been updated to the year 2010: that of the Scholia minora in Homerum website within the “Aristarchus” portal of Genoa University (edited by D. Muratore and the present writer),3 which lists 81 papyri for the Iliad and eleven for the Odyssey;4 and J. Lundon’s list, The Scholia minora in Homerum: An Alphabetical List, within the portal of Cologne University,5 which — in addition to a most useful list of lexeis from papyri of this sort — provides at the beginning of the Bibliography what is de facto a list of published papyri, 81 for the Iliad and 17 for the Odyssey. A survey carried out on this occasion shows that the two lists do not perfectly overlap (even in the case of the Iliad, although they give the same number, 81; for the Odyssey, Pontani (2005) lists nine papyri with Sch. Min. and six with marginalia).6 There are various reasons for this, such as: the two different ways in which these lists consider fragments that have marginal and interlinear annotations of a glossographic sort, and which can therefore be traced back to the kind of material belonging to glossaries; the divergent ways in which they list witnesses stemming from the combination of fragments published separately (i.e. by assigning each only one  3 http://www.aristarchus.unige.it/scholia/papiriList.php. 4 The numbering is that of Raffaelli 1984, with the addition of decimal places. 5 http://ifa.phil-fak.uni-koeln.de/6348.html. 6 Pontani 2005, 111–126.

Homeric Papyrology: Themes, Problems, and Perspectives  

number or two); whether or not they take into account certain papyri which are known yet remain unpublished (Lundon obviously cannot take account of these for his lexeis, since he does not have the text on hand, whereas these are present in the Montanari-Muratore list). While, on the one hand, there is a need to come up with a list that is as reliable as possible and based on well-defined, unambiguous criteria, the fact remains that in roughly 25 years the number of fragments of this kind has increased by over one third. I should further add that, in at least one large papyrus collection I am familiar with, several yet unpublished fragments of Homeric glossaries have been identified, to which others from other collections should no doubt be added (not to mention all the fragments that have yet to be identified). This makes the available figure an underestimation, and hence any numerical statistics based on the published material must be regarded as very provisional. As we shall see, this is a recurrent situation, particularly when it comes to fragments of the Homeric text itself. Alongside glossaries, we have Homeric alphabetic lexica, which comprise a similar, if not identical, kind of material, namely the combination of a Greek word with an interpretamentum. The basic difference is that glossaries present a syntagmatic arrangement that follows the order of the Homeric text, whereas lexica present a paradigmatic arrangement that follows an alphabetical order. One example is Apollonius the Sophist’s lexicon: indeed, the fragments of Homeric alphabetical lexica can often be traced back to this work, which has reached us in epitome form. This category does not include the Homeric material found in general lexica, which are on the same level as the Homeric quotes featured in commentaries on other authors, which I will be discussing shortly. The issue of glossaries leads us to hypotheseis, which in papyri occur in different forms that can nonetheless easily be grouped into three kinds of arrangement: 1) hypotheseis placed before the glossary of a Homeric book (P.Achmim 2 and P.Oxy. XLIV 3159 for the Iliad; P.Oxy. XLIV 3160 + P.Strasb. inv. 1401 and P.Oxy. LVI 3833 for the Odyssey); 2) a self-standing collection of all the hypotheseis of one of the Homeric poems or both (to the best of my knowledge, for now only P.Ant. II 69 presents the Iliad and Odyssey together); 3) isolated hypotheseis, which in all likelihood can be interpreted as occasional transcriptions.7 Hypotheseis are present — in similar or different forms from those we find in papyri — in many medieval codices of the Homeric  7 Montanari 1984 (= this volume, ch. 32) and 1988b (= this volume, ch. 33); Pontani 2005, 115– 117 for the Odyssey.

  Homeric Papyrology: Themes, Problems, and Perspectives text, with or without scholia. More particularly, they are regularly present in the codices of Scholia D, the corpus of scholia constituting a rich repository for a widespread ancient glossographic element which is chiefly (albeit not exclusively) found in glossaries, just as in certain papyri we find hypotheseis inserted before glossaries. Based on a degree of proximity established a posteriori, so to speak, which is to say a degree of proximity found in the scholiographic corpora of medieval codices but which so far has not been found in papyri (at any rate, as far as I am aware), I will now say something about the so-called Mythographus Homericus (MH). Its material has flowed into the corpus of Scholia D, probably via a lateantique copy used by the individual responsible for the archetype of these scholia. The collection of fragments of the MH, published a few years ago,8 includes eight undisputed fragments for the Iliad, but only one for the Odyssey (which is actually the most extensive one of all, PSI X 1173):9 these cover a time period roughly ranging from the 1st to the 5th century AD. I cannot delve here into the complex issues of the origin, character, and original form of the MH, and of the reliability of the subscriptions of its historiai, not just in the papyri but also in the corpus of Scholia D. However, I will at least note that in the papyri, which span several centuries of the Imperial Age, the MH does not take the form of a mythographical handbook, but rather that of a hypomnema — albeit a specialist one, so to speak — which follows the order of the Homeric text, to which it is closely and organically connected thanks to the constant presence of lemmas. Through various passages and changes, the MH must have reached the compiler of D, who then merged it with the glossographic-paraphrastic element, as already noted. One major phenomenon to be taken into account is the presence of Homeric quotes across a highly diverse range of works. Homer is by far the most oftquoted author in ancient Greek literature, which is significant both for the history of his reception over the centuries and for the history of the Homeric text as such. One obvious example is that of a papyrus fragment — say, Plato or Aeschines — containing a Homeric quote that raises certain problems because the text of the quote is different from (or possibly even because it is the same as) the text of the quote preserved in the works of Plato or Aeschines transmitted by medieval codices. In such cases, the testimony from the fragment, compared to

 8 Montanari 1995 (= this volume, ch. 22). 9 Montanari 2002 (= this volume, ch. 21); Pontani 2005, 126–130.

Homeric Papyrology: Themes, Problems, and Perspectives  

the so-called vulgate that has more or less been followed by modern editions,10 concerns the earliest known phases in the history of the text (at a stage often coinciding with that of the Ptolemaic papyri),11 and will confirm or diverge from what is known about the Homeric passage in question from the rest of the textual tradition of Plato or Aeschines. Clearly, a papyrologist editing any fragment with Homeric quotations will consider this aspect to be an accessory factor of lesser importance for him; but for the Homeric philologist it constitutes a piece in the puzzle of the history of the transmission and formation of the text that rightfully — if secondarily — belongs to Homeric papyrology. Yet — and this is probably the most interesting case — we may also be dealing with fragments of texts that have not otherwise been transmitted, works of the most varied nature that feature quotations of Homeric verses with or without any accompanying interpretation. This is truly a terribly varied galaxy that de facto is difficult to analyse. In D. Sutton’s catalogue, Homer and the Papyri, published on the website of the Washington Center of Hellenic Studies under the supervision of G. Nagy and other scholars,12 we find a list of numbers preceded by the letter w (witness) that M. West has taken up for the Iliad,13 and which constitutes a generous attempt to provide a point of reference. Under the present conditions, however, it strikes me as a very problematic and ill-defined tool (curiously enough, for example, this list includes ostrakon PSI VIII 1000, which has been explicitly traced back to the MH; while it is quite right to consider Homeric quotations occurring in commentaries on other authors, only a systematic survey of

 10 Haslam 1997, 63: “Our earliest Homeric manuscripts, those of the 3rd cent. B.C., are characterized by their startling degree of difference from the text that prevailed later, sometimes known as the ‘vulgate’ ... the very term ‘vulgate’ is a misnomer. It designates no particular version of the text; there is no vulgate of Homer as there is a vulgate of the Bible ... that is a construct which may never had any existence in the real world”; p. 84: “Just how the vulgate came to be the vulgate, is an unsettled question. As to the part played by the Alexandrian scholarship, diametrically opposed views have been arrived at ... The number and sequence of verses in our manuscripts ... conform with nearly perfect exactitude to the verses recognized by Aristarchus (verses which he athetized included) ... At the same time, the proportion of Alexandrian scholars’ readings in the vulgate is very low.” Certainly, the modern concept of vulgate does not coincide with what is called the koine edition in the ancient source, which is very difficult for us to define (Haslam 1997, 71). 11 Haslam 1997, 74–79. 12 Address http://chs.harvard.edu. 13 West 2001a, 129–138, esp. 136–138.

  Homeric Papyrology: Themes, Problems, and Perspectives exegetical material on papyri can provide any certainty).14 The possibility which the online Mertens-Pack database offers to search for quotations by author is no doubt useful here: for Homer it gives 122 quotations15 and it would be interesting to compare this with the aforementioned w list by Sutton-Nagy-West. Likewise, we should bear in mind that, while our discussion about quotations has been limited to literary texts and the works of known or unknown authors, Homeric quotations and references are certainly also to be found in non-literary texts, which is to say in the vast and varied range of sources that are commonly and most generically called “documentary papyri”. Just how much and what Homer is to be found in documentary papyri (private letters, petitions, lists, administrative documents of all sorts) constitutes — at least, to the best of my knowledge — a yet unexplored field. I would expect it, though, to provide some important evidence on Homer’s pervasive presence across the various levels and contexts of ancient society: evidence to be interpreted and analysed together with everything that the history of book production and the circulation of copies can tell us about a specific socio-historical and cultural problem in the broad sense of the term. Another most interesting sector that may be briefly mentioned in all of its intriguing complexity is that which concerns the history of the Greek language in relation to the history of Homeric diction and the Homeric text, along with the transformation of its linguistic form over the course of the centuries-old tradition that led to a profusion of medieval manuscripts, and then up to modern editions. It is rather disquieting to wonder what “Homer” we are reading in one of the volumes we are holding in our hands, as compared to what “Homer” was prior to his becoming fixed in the written copies and editions of the Hellenistic Age. If I am not mistaken, language history and hence the development of formulaic diction are not among the most commonly explored topics among papyrologists; yet, Homeric papyri sometimes provide highly interesting data that throws light precisely on such questions. And perhaps in most cases scholars of Homeric philology (nostra culpa) fail to grasp the meaning of a variant offered by a papyrus, possibly an isolated variant that is swiftly dismissed but which bears witness to different phonology or morphology from that represented in all or most codices (I would venture to say: in our so-called vulgate). Note,

 14 This is one of the aims of the series Commentaria et Lexica Graeca in Papyris reperta (CLGP), edited by G. Bastianini/M. Haslam/H. Maehler/F. Montanari/C. Römer, several volumes of which have been published so far. 15 At the time of this article (2011), the website http://web.philo.ulg.ac.be/cedopal/en/ gives “état mai 2011’ as the date of the last update.

Homeric Papyrology: Themes, Problems, and Perspectives  

however, that “different” does not necessarily mean more ancient or more recent in absolute terms, within the history of the Homeric language, compared to the more widespread and commonly accepted reading: the historical-linguistic analysis of specific variants can yield information of the stages of development of formulaic diction, or even on the development of the meaning of an expression. This is certainly a field in which there will still be much to do and investigate, as new Homeric papyri are published; so let me state here that it would be a serious mistake to overlook them because they are banal or too numerous. Finally, let us turn to the issues mentioned at the beginning, namely the relationship between Homeric papyrology, on the one hand, and the textual history of the Iliad and Odyssey (already mentioned in relation to linguistic history questions) and the history of their exegesis in Antiquity on the other. Lists are a recurrent and significant element in Homeric papyrology, and they are most useful, given the exceptional number of finds. As far as the Iliad is concerned, the data has been clearly summarised by M. West in his book on the textual history of the Iliad,16 which accompanies his Teubner edition of the epic poem;17 so in what follows I will only provide a succinct overview. While A. Ludwich listed 33 papyri in his 1902–1907 edition of Iliad and T.W. Allen 122 in his editio maior of 1931, the subsequent lists drawn by Collart, Mette, Pack, and Sutton — down to the summer of 1990 — reached a total number of 703.18 For his 1998–2000 edition, West used 1544 papyri, a number that has increased even further, to 1569, in his most recent list, published in the 2001 volume, where he has added nos. 1545–1569, all of which were published in the year 2000.19 The real number of papyri, however, is even higher, since at one stage in the development of this list, instead of continuing with the linear numbering, Mette chose to insert into the list numbers in the form of letters and decimals: I have counted at least 50, so we may sum the situation up by saying that at the beginning of the third millennium, there were over 1,620 known Iliad papyri; in the meantime, new ones have certainly been published. This list has undergone a sharp increase particularly owing to the fact that West has undertaken an enormous labour, by collating with the original texts something like 840 unpublished papyri from the Oxonian P.Oxy. collection. While he has used these  16 West 2001a, 86–138. 17 West 1998 and 2000. 18 West 2001a, 86; an updated version of Sutton’s list is available from the website mentioned in n. 12 above. 19 West 2001a, 87: as no. 1545 he adds an inscription from Rome, published in 1938 but previously not included in the lists; as no. 1569, P.NY II 1; as nos. 1546–1568, the papyri published in Manfredi 2000 (see West 2001a, 87 n. 9).

  Homeric Papyrology: Themes, Problems, and Perspectives for his edition, they have not really been published, so in practice they cannot be used in relation to all aspects of Homeric papyrology.20 Let us consider these purely numerical data. In one decade, the number of Homeric papyri listed (i.e. papyri with Homer’s text) has more than doubled, from 703 in 1990 to over 1,620 in 2001.21 We should also bear in mind that West has limited his efforts — so to speak — to those P.Oxy. which have been identified, and which he has access to in Oxford. However, this raises a few questions: first of all, how much identified and unpublished material is there in other collections? And how much lies in the boxes of papyri yet to be restored and identified? But the problem, obviously, is not just quantitative; we must ask ourselves to what extent such conspicuous increases can change our ideas and assessments of the history of the Homeric text and of other aspects of Homeric papyrology. West himself, in the first chapter of his book about the history of the text, which is devoted to papyri, warns us not to set our expectations too high: “The purpose of this chapter is not to assess the contribution of the papyri to our knowledge of the tradition, but to perform the more basic public service of providing a more or less up-to-date list of those that have been published hitherto and of the unpublished ones cited in my edition.”22 Therefore, West has used the many new P.Oxy. for the text and apparatus of his edition, basing his choices on his philological expertise and editorial principles, but the mass of new data has never been analysed as a whole, not even in relation to the history of the text, not to mention all the other papyrological aspects (indeed, West speaks of “unpublished ones cited in my edition”). I find it rather disturbing that these “unpublished” papyri account for well over half the total; but even more troubling is the fact that nothing of the sort has yet been done for the Odyssey.23 I should note that this is not merely a problem of lists — of more or less upto-date and specialist ones: to specific lists we can also add more general catalogues, like the online Leuven Database of Ancient Books (LDAB) and MP3.24 The fact is that, through the remarkable effort he made for his edition of the Iliad, West has highlighted and lent concreteness to a problem that previously had  20 In the MP3 list, they are correctly labelled as “inédit”. 21 These numbers may vary slightly, for different reasons. Haslam 1997, 60: “Homer manuscripts [scil. for the Iliad and Odyssey] now number well over a thousand.” 22 West 2001a, 86. 23 Mertens 1985 provides an updated version of the list that can be drawn from Pack2, and which can now be integrated through MP3 (see below); H. van Thiel decided to use a selection of published papyri for his edition of the Odyssey (van Thiel 1991). 24 Cf. n. 15.

Homeric Papyrology: Themes, Problems, and Perspectives  

arguably been underestimated or neglected, a problem especially affecting Homeric papyrology, for quantitative reasons. The extension of the list of Iliad papyri to include a large number of unpublished yet known ones — even leaving aside possible ‘black holes’ in the form of papyri that have been discovered but remained wholly unknown, as they have yet to be identified — and the possibility of carrying out a similar operation for the Odyssey (probably a smaller operation in terms of the overall number of papyri involved, yet a comparable one in terms of percentages) raise a significant problem: how and to what extent is the picture of the earliest stages in the history of the Homeric text likely to change, from all points of view, through this unprecedented increase in the number of testimonies and in the amount of data to be studied and analysed? And when, if ever, will they be adequately studied? What is emerging, it seems, is a rather peculiar situation for Homeric papyrology, which suggests the two following points: it is wrong to assume that Homeric fragments are of little interest simply because there are so many of them, and hence to neglect them; rather, it would be worth publishing the Homeric papyri in large blocks, in order to make what I have called ‘unpublished yet known’ material available as soon as possible — and this applies not just to the Iliad, but also to the Odyssey. We are talking about issues of the utmost relevance to our understanding of what Homer we are reading in the modern editions we commonly use: the most ancient phases in the history of the Homeric text, the fixing of the numerus versuum roughly starting from the second half of the 2nd cent. BC, essentially through the influence of Aristarchian Alexandrian philology, the varying circulation of Alexandrian readings, and the formation of the so-called vulgate. In my view, at the moment the best overview on topic of “Homeric papyri and the transmission of the text” remains the one provided by M. Haslam in 1997.25 As anticipated, the questions related to data pertaining to the Homeric text as such are closely connected to those related to the different kinds of exegetical material directly concerning it (therefore with the exclusion of the quotations of Homeric verses in exegetical texts dealing with other authors). This subgroup of ‘Homerica’ includes some kinds of texts I have already discussed, namely alphabetical lexica, glossaries, hypotheseis, and the MH. Finally, this leads us to commentaries and treatises, or hypomnemata and syngrammata according to the ancient terminology: it is not always easy to distinguish between the two, unless it is possible to identify or reconstruct a certain or at any rate likely sequence of lemmas following the syntagmatic order of

 25 Haslam 1997, with further bibliography.

  Homeric Papyrology: Themes, Problems, and Perspectives the Homeric text.26 To these we should add marginal annotations with a philological-exegetical content, which often present textual variants and quote ancient grammarians: these are annotations like those we find in the Hawara Iliad, and are therefore different from the purely glossographic, marginal or interlinear ones associated with the kind of material that makes up glossaries. As far as marginalia of the glossographic sort are concerned, they differ from glossary not because of their content (which is of the same kind), but rather because of the form in which they have been transmitted (paratexts framing the commented text vs. material transmitted independently of the text). Sometimes the distinction between different kinds of exegetical texts may prove problematic, but in principle it is important not to overlook these differences and to pay attention to the terminology.27 For the sake of completeness, to hypomnemata and marginal annotations we should add critical marks (semeia), which are connected to commentaries and fully rank among philological-exegetical marginalia, even though I cannot dwell on them here. Before saying some words about works of this kind, we must once again return to the problem of lists. In the aforementioned Sutton-Nagy catalogue, Homer and the Papyri, we also find a list of items preceded by the letter h (homerica), which has been taken up by M. West.28 This list, which in West reaches number h 142, would appear to encompass all kinds of ‘Homerica’: commentaries, treatises, glossaries, alphabetical lexica, paraphrases, hypotheseis, the MH, school exercises, and more. The same thing holds for MP3, which presents an overall list of ‘Homerica’ (after the papyri with the Homeric text), ranging from no. 1157.000 to no. 1232.030 and arranged according to the Homeric text, with the method of adding three-figure decimal numbers. Attempts at simplification are quite understandable, but I believe that a single list is not a good idea and that it is worth making the effort to distinguish between the various kinds of exegetical genres, and to draw up a specific list for each — doing everything we can to develop coherent and precise definitions (although it is only natural that certain things will escape us or remain uncertain). In the two lists just mentioned we find quite a few errors, particularly when it comes to the typological definition of the texts, as was bound to occur with such a varied and complex

 26 In certain cases, it may be difficult to distinguish between a glossary and a hypomnema, for instance because of a fragment’s brevity: what glossaries and hypomnemata have in common is the fact that their exegetical material follows the order of the commented text but is transmitted independently of it. 27 Cf. n. 34. 28 West 2001a, 129–136.

Homeric Papyrology: Themes, Problems, and Perspectives  

range of material. Other catalogues available online can be used, such as the aforementioned LDAB29 and the Catalogue of Paraliterary Papyri (CPP).30 Besides, we already have some excellent, or at any rate good, foundations to build specialist lists31 for glossaries (or Scholia minora: these two expressions are synonymous, as they describe the same kind of content, but they should not both be used in the same list),32 paraphrases, hypotheseis, the MH, alphabetical lexica, school exercises, anthologies, hypomnemata, and syngrammata. I firmly believe that it would be useful to proceed in this direction in order to find our way through rather murky waters.33 One essential aspect concerns terminology, to which we ought to pay more attention: it would be most useful to adopt a shared one with normative value.34 To the best of my knowledge, we have no list of papyrus fragments of Homeric syngrammata, and indeed I believe that a list of this sort is one of the most problematic to establish, starting from available lists of ‘Homerica’.35 What is less difficult, albeit often challenging given the state and size of the items, is to draw up a list of hypomnemata to the Iliad and the Odyssey. Erbse’s one for the Iliad dates from 1988,36 Pontani’s for the Odyssey from 2005:37 in 2011, however,

 29 http://www.trismegistos.org/ldab/. 30 http://cpp.arts.kuleuven.be/. 31 I made an initial attempt in this direction in Montanari 1984 (= this volume, ch. 32): the problem ought to be newly addressed with better defined criteria and more advanced tools (e.g. electronic lists that can be updated). 32 Cf. n. 34. 33 For example, one wonders why h 43, a fragment of Julius Africanus’ Kestoi with a quote occurs in the h(omerica) list rather than in the (w)itness list — cf. above. 34 For example, the term ‘scholia’ is used not only in relation to hypomnemata, but also for the Mythographus Homericus, as well as in the definition Scholia minora: first of all, it would be appropriate to reserve ‘scholium’ for marginal annotations in the strict sense — but this is contradicted precisely by the common use of the expression Scholia minora to describe what simply ought to be called ‘glossaries’, at least in the case of papyrus witnesses. Things would be much clearer if we decided once and for all — as clearly outlined and argued by Montana 2011 — to reserve the term ‘scholium’ for those commentary notes collected in medieval corpora, seen as parts of an overall editorial project and hence as conceptually different — at least, generally speaking — from the marginal annotations in papyri. Therefore, we should only use ‘glossaries’ for papyri and do away with the definition Scholia minora, which after all is neither particularly helpful nor clear. This is also an exercise in self-criticism for me, since I use the expression Scholia minora myself, even for the above-mentioned website — see n. 3. 35 Pontani 2005, 130–133 discusses three cases pertaining to the Odyssey: P.Oxy. LXV 4453 (MP3 1207.21), P.Oxy. XXXIX 2888 (MP3 1209.3), P. Med. inv. 210 (MP3 1210.2). 36 In the edition of the Scholia Vetera, with the Addenda et Corrigenda of volume VII, 1988. Erbse does not consider syngrammata, such as P. Brux. inv. E 7162 or P. Med. Inv. 71.82 (MP3,

  Homeric Papyrology: Themes, Problems, and Perspectives John Lundon published an article presenting what is currently the most up-todate reference list, accompanied by a reliable and compelling discussion of certain items and the criteria adopted.38 Here we have a total of 21 papyri, 14 for the Iliad and 7 for the Odyssey. Further confirming the fact that things tend to change quite rapidly in this field — and possibly also confirming the fact that recent years have witnessed a rising interest in texts of this kind — Lundon’s list (which I believe was actually updated in 2008 or 2009)39 is already in need of an update in relation to a couple of points, as far as the Iliad is concerned: one is P.Oxy. LXXVI 5095, which we shall shortly be discussing; the other, I should imagine, we will discover here through Luciano Canfora and Rosario Pintaudi’s presentation of a new Homeric commentary from the Laurentian Papyiri. Lundon published his list before he could take account of P.Oxy. 5095, a preview of which I offered some time ago.40 This consists of fragments from a commentary on Il. XII and XV, belonging to a papyrus codex written in a fine Alexandrian majuscule that can be dated to the 5th/6th cent.: in all likelihood it is a hypomnema to the whole Iliad. This witness is particularly significant for a number of reasons. It is the latest Homeric commentary known to us and the first to have been preserved in a papyrus codex, which is rather surprising, considering Homer’s popularity and the number of finds pertaining to him. It is a very high-quality specimen, certainly a valuable item intended for the book market, carefully written and conveying a high-level critical-exegetical content, which circulated in Late Antiquity. It is the only known Homeric commentary displaying a distinctive and practically one-way relation with the corpus of Scholia exegetica, thereby providing some important evidence regarding the history of the formation of this corpus.41

 respectively 1224 and 1227, with bibliography). It is worth mentioning the case of P. Gen., Bibl. Nicole inv. 72, the first edition of which — produced by J. Nicole in 1893 — remains practically unknown: the new edition by Trachsel/Schubert 1999 shows that it is probably a discussion of the topography of Troy in relation to the setting of the duel between Achilles and Hector (no lemmas are identified); hence, it falls within the syngrammata category (bibl. in MP3 1204). 37 Pontani 2005, 130–136. 38 Lundon 2011. 39 See Lundon 2011, n. 94. 40 Montanari 2009 (= this volume, ch. 24); published in P.Oxy. LXXVI (2011), 177–192. 41 This is a very different relation from that we find in the case of certain parts of the commentary on Il. XXI in P.Oxy. II 221: in P.Oxy. LXXVI 5095 the complete lack of quotes from grammarians, variants, and parallels clearly distinguishes this papyrus from P.Oxy. 221, in which all these elements are instead noticeably present and material that can be traced back to the sch. ex. is combined with conspicuous contributions from the Alexandrian-Aristarchean tradition.

Homeric Papyrology: Themes, Problems, and Perspectives  

I will choose only one case to illustrate an idea mentioned at the beginning, namely the fact that exegetical witnesses can provide significant data for the history of the Homeric text as such. The commentary on Il. XIX 351 transmitted by the 2nd-century P.Oxy. LXV 4452 proves particularly striking as regards precisely the lemma, namely v. 351, which is broadened by the insertion of another four verses between its two hemistichs: a textbook case of a longer text being completely replaced by a more concise one in our manuscript tradition, in accordance with the Aristarchean taste for a relatively less wordy text, which has determined the dominant numerus versuum in the tradition. This hypomnema fragment gives us four previously completely unknown verses (unknown, in this case, even in the scholiography, which sometimes instead provides information on verses that have disappeared from the direct tradition); not only that, but it also informs us about the source from which these verses are drawn: one of the kata poleis editions, more specifically the Marseilles one, the Massaliotike (ekdosis) mentioned repeatedly (at least thirty times) in scholia by Didymus on account of textual variants that are sometimes accepted, but more often rejected, by Aristarchus.42 The quotes from the Massaliotike which we find in the scholia usually concern textual variants pertaining to verses included in our manuscript tradition: one case that concerns the numerus versuum is Od. I 97– 98, two verses which Aristarchus expunged, but this is certainly a rare, if not isolated, example. Here we have four verses which were present in the Massaliotike (only in the Massaliotike?), but which are nowhere to be found in our manuscript tradition, evidently because they were not included in the Alexandrian/Aristarchean numerus versuum: in a hypomnema that fully reflects the Alexandrian tradition and which survives in a specimen from the 2nd century AD — by which time in the papyri of the Homeric text the additional verses had long disappeared — some verses are presented as a lemma (i.e. not simply quoted in the commentary); hence verses which must not at all have belonged to the Alexandrian/Aristarchean numerus versuum (unless we are to posit a really unlikely twist of fate), but which rather occurred in one of the kata poleis editions, are taken into consideration. What are we to make of all this? Should we expect to find in hypomnemata presented as lemmas and commented on not just verses athetised by Aristarchus (which is not at all strange), but also verses that were not at all included in the numerus versuum approved by him, which is to say verses that were not at all accepted as part of the Homeric text, even with the kind of benefit of the doubt represented by athetisation? Evidently, this is at

 42 Haslam 1997, 69–70.

  Homeric Papyrology: Themes, Problems, and Perspectives least sometimes indeed the case, which strikes me as something particularly noteworthy and interesting. In moving towards a conclusion, I would like to mention at least two other issues that especially concern fragments of hypomnemata, syngrammata, and philological-exegetical marginal scholia, albeit not just Homeric ones, but all texts of this kind: these issues, then, do not pertain to Homeric papyrology alone, although they are most frequently associated with it. One useful aspect of such texts is that they increase the data available to us about the history of Alexandrian philology, particularly through valuable quotes from grammarians whose works are largely lost, the recording of variants, information on ancient ekdoseis, and the preservation of philological semeia (which do not occur exclusively in the margins of specimens of the Homeric text, but sometimes accompany the lemmas of a hypomnema, as in P.Oxy. VIII 1086). While this factual element is so well-known that it is almost superfluous to recall it, what is less common is the problem concerning the exact wording of ancient philology. Hellenistic-age grammarian-philologists are hardly the only ancient authors whose original works have not reached us and with regard to whom it is difficult to tell with any certainty whether, and to what extent, what we have are words and sentences preserved verbatim — i.e. the lost author’s ipsissima verba. This problem is especially relevant in areas such as the one I have been discussing, in which the formation, application, and development of a technical language are particularly important, and are closely intertwined with the definition of conceptual instruments.43 Are the papyrus fragments of hypomnemata of any help to us in this respect? They can certainly be of some help, and it will be worth examining the fragments of such works also from this perspective, when it is possible to do so. I will conclude without drawing any real conclusions, but only noting that what I have done here is mostly offer a list of open questions and things to do. In short, this is the list of problems, and it is therein that the beauty of research lies.

 43 Concerning these issues, see Montanari 1997 (= this volume, ch. 42).

 Dialect Glossai in the Scholia to Homer The study of the dialect glossai mentioned in the Homeric scholia can be viewed, in the first place, as forming one part of a larger (and well-known) subject, namely, ancient scholarship in dialectology. It is also, however, closely related to the copious lexicographic production of the ancients, within which we find that specific dialect forms and other observations on dialect are abundantly represented. For the Alexandrian lexicographers, the concept of dialektos has an unequivocally geographical meaning: thus, as a scholarly term, dialektos indicates the local manner of speech in a distinct and specific area,1 and these various local forms of speech are often referred to in studies of literary works, in primis poetic ones, as we can see in the case of the Homeric scholia. We know of a collection entitled Ἐθνικαὶ λέξεις (two certain fragments are found in Galen, one possible fragment in Athenaeus), which is attributed to Zenodotus, even if it is not entirely clear whether this is Zenodotus of Ephesus or another Zenodotus; nor is it clear, if our Zenodotus is in fact the one from Ephesus, whether or not this work should be considered distinct from his Γλῶσσαι. As K. Nickau writes: «Gehörte das Werk, wie unsere Vermutung lautet, dem Z(enodotos) von Ephesos und erklärte er sowohl Dialektwörter als auch homerische Glossen, so läßt sich seine Stellung in der Geschichte der Griechischen Lexikographie wie folgt verstehen: Philetas hatte die dunklen Wörter... gesammelt, vielleicht auch, um den Schmuck des seltenen Ausdrucks (im Sinne von Aristot. poet. 22, 1459 a 9; vgl. Latte, Kl. Schriften 654) den Dichtern verfügbar zu machen, und Z(enodotos) mag ihm darin gefolgt sein. Ebenso wie Philetas erklärte Z(enodotos) Dialektwörter und Homerglossen, doch den Ephesier führten seine Homerstudien wohl noch stärker zur Deutung schwer verständlicher Wörter in Ilias und Odyssee».2 To this one may add the useful synthesis offered by R. Pfeiffer: “There was hardly any following age in which the Greek mind was not attracted by the problem of explaining λέξεις. Their origin and their changes, the differentiation of kindred words, the comparison between Greek dialects or between Greek and foreign words were discussed by the Sophists, by Democritus, and by the great Attic philosophers. In the new era the poets revived these studies with fervour, not only Philitas and Simias, but also Callimachus and Apollonius ... Whatever had been undertaken piecemeal here and there in the course of time, was all now united into one great enterprise, the

 1 Cf. Cassio 1993; 1993b, 79 ff.; 1997; Tosi 1994, esp. 209; Cassio 2008, 3–8. 2 Nickau 1972, 40–42 (citation from col. 42). https://doi.org/10.1515/9783110772371-045

  Dialect Glossai in the Scholia to Homer Λέξεις of Aristophanes. A collection of γλῶσσαι was usually limited to obsolete and obscure terms; but under the neutral title Λέξεις every word which was peculiar in form or significance and therefore in need of explanation could be listed, whether it was out of date or still in use”.3 Pfeiffer then mentions two aspects of Aristophanes’ work which he considers fundamental: first, his attention to the historical development of language, with regard to which he made a chronological distinction between ancient and modern usage (with the related idea that the latter may have its origins in a specific geographic area [p. 315]); second, his innovative interest in dialects and in dialect terms, which is clear from the fact that his Λέξεις contained a section dedicated to Ἀττικαὶ λέξεις and another dedicated to Λακωνικαὶ γλῶσσαι. (Aristophanes’ teacher, Dionysius Iambus, had written a work entitled Περὶ διαλέκτων, and Sosibius Lacon one entitled On Spartan Cults, which included a collection of Laconian glossai.)4 In the context of Hellenistic and Imperial-era studies in philology and grammar, research in dialectology was commonly practiced. It developed especially in connection with the attempt to establish the linguistic correctness and the dialectal facies of the authors whose text was to be established and interpreted. An important figure in this field is the grammarian Demetrius Ixion, a pupil of Aristarchus who later distanced himself from his teacher. Demetrius Ixion undertook studies into Attic (Ἀττικαὶ λέξεις) and into the dialect of Alexandria (Περὶ τῆς Ἀλεξανδρέων διαλέκτου); his work thus demonstrates, on the one hand, an interest in a dialect which it was impossible to ignore, given the literature it had produced, and, on the other hand, a growing interest in contemporary spoken language.5 In addition to various treatises and collections dedicated to individual dialects (among which there is the curious Ἰταλικαὶ γλῶσσαι of Diodorus of Tarsus, of the first century BCE), it is worth mentioning a work by the scholar Tryphon entitled Περὶ τῶν παρ’ Ὁμήρῳ διαλέκτων καὶ Σιμονίδῃ καὶ Πινδάρῳ καὶ Ἀλκμᾶνι καὶ τοῖς ἄλλοις λυρικοῖς; in this work, it is

 3 Pfeiffer 1968, 197 f. = 311 f. 4 Pfeiffer 1968, 201 f. = 316 f. In addition to an awareness of the diachronic evolution of language, Pfeiffer sees in Aristophanes a related interest in the spoken language of his time. This issue has been discussed subsequently by Callanan 1987, 75–82 and Ax 1990, 13–15: Callanan strongly rejects the idea that an interest in the diachronic evolution of language can be seen in Aristophanes, but in Ax’s view such a position is excessive and not borne out by the evidence (among which Ax makes use of the testimony of the Homeric scholia). On the whole question, see most recently Pagani 2011, 37 (with n. 81) and 48 (with n. 121). 5 The interest in contemporary spoken language is perhaps due to the influence of Aristophanes; cf. n. 4 above. On Demetrius Ixion, see Ascheri 2009 and 2010, which contain useful summaries and bibliography. On the dialect of Alexandria, see Fournet 2009.

Dialect Glossai in the Scholia to Homer  

notable that we find attention dedicated not only to Homer but also to the lyric poets, whose texts — given their linguistic richness and variety — posed problems which were at times extremely difficult to resolve.6 This important strand of research (which, for reasons of space, I can only mention here in brief outline) branched over time into at least three directions: 1) lexicography, which has been mentioned above; 2) linguistic-grammatical research, which, after the epoch-making systematic treatises of Apollonius Dyscolus and Herodian, extends all the way to dialect studies of a much later date, such as the famous Περὶ διαλέκτων of Gregory of Corinth (X–XI cent.); 3) philology and the interpretation of texts, which is the primary subject area that I wish to examine in the present contribution. In the history of the interpretation of poetry in the ancient world, Aristotle’s Poetics played an essential role. In it one finds reflections upon the entire preceding period, and the foundations are definitively laid out for the philological and hermeneutic activity of the following centuries. I begin by citing several well-known passages from this work: 1)

2)

3) 4)

5)

21.2 (1457 b 1–7). ἅπαν δὲ ὄνομά ἐστιν ἢ κύριον ἢ γλῶττα ἢ μεταφορὰ ἢ κόσμος ἢ πεποιημένον ἢ ἐπεκτεταμένον ἢ ὑφῃρημένον ἢ ἐξηλλαγμένον. λέγω δὲ κύριον μὲν ᾧ χρῶνται ἕκαστοι, γλῶτταν δὲ ᾧ ἕτεροι· ὥστε φανερὸν ὅτι καὶ γλῶτταν καὶ κύριον εἶναι δυνατὸν τὸ αὐτό, μὴ τοῖς αὐτοῖς δέ· τὸ γὰρ σίγυνον Κυπρίοις μὲν κύριον, ἡμῖν δὲ γλῶττα. 22.1 (1458 a 21–23). σεμνὴ δὲ καὶ ἐξαλλάττουσα τὸ ἰδιωτικὸν ἡ τοῖς ξενικοῖς κεχρημένη· ξενικὸν δὲ λέγω γλῶτταν καὶ μεταφορὰν καὶ ἐπέκτασιν καὶ πᾶν τὸ παρὰ τὸ κύριον. ἀλλ’ ἄν τις ἅπαντα τοιαῦτα ποιήσῃ, ἢ αἴνιγμα ἔσται ἢ βαρβαρισμός. 22.7 (1459 a 9–11). τῶν δ’ ὀνομάτων τὰ μὲν διπλᾶ μάλιστα ἁρμόττει τοῖς διθυράμβοις, αἱ δὲ γλῶτται τοῖς ἡρωικοῖς, αἱ δὲ μεταφοραὶ τοῖς ἰαμβείοις. 24.5 (1459 b 32–37). τὸ γὰρ ἡρωικὸν στασιμώτατον καὶ ὀγκωδέστατον τῶν μέτρων ἐστίν (διὸ καὶ γλώττας καὶ μεταφορὰς δέχεται μάλιστα· περιττὴ γὰρ καὶ ἡ διηγηματικὴ μίμησις τῶν ἄλλων. 25.6 (1461 a 10–15). τὰ δὲ πρὸς τὴν λέξιν ὁρῶντα δεῖ διαλύειν, οἷον γλώττῃ τὸ “οὐρῆας μὲν πρῶτον” (Il. 1.50)· ἴσως γὰρ οὐ τοὺς ἡμιόνους λέγει ἀλλὰ τοὺς φύλακας· καὶ τὸν Δόλωνα, “ὅς ῥ’ ἦ τοι εἶδος μὲν ἔην κακός” (Il. 10.316), οὐ τὸ σῶμα ἀσύμμετρον ἀλλὰ τὸ πρόσωπον αἰσχρόν, τὸ γὰρ εὐειδὲς οἱ Κρῆτες τὸ εὐπρόσωπον καλοῦσι· καὶ τὸ “ζωρότερον δὲ κέραιε” (Il. 9.203) οὐ τὸ ἄκρατον ὡς οἰνόφλυξιν ἀλλὰ τὸ θᾶττον.

At the end of chapter 19 of his Poetics, Aristotle introduces his discussion of the term lexis (1456 b 9), which is then developed further starting in chapter 20 (1456 b 20 ff.). In chapter 21 (passage 1), we have the general definition of glossa as a linguistic/lexical concept: it designates an onoma employed by “others” as

 6 Cf. works cited in n. 1 above.

  Dialect Glossai in the Scholia to Homer opposed to the onoma commonly used by all, that is to say, an onoma that differs from ordinary usage. In passage 2, in line with his theoretical framework, Aristotle includes glossa in an overall category that he designates as ξενικόν, i.e., “foreign” with respect to that which is ordinary (παρὰ τὸ κύριον).7 This latter designation is highly generalizing and comprises numerous types of “alterity,” which may be distinguished principally in terms of time (a word that is obsolete with respect to the moment in which it is being employed) or in terms of dialectal geography (a word that derives from a geographical area that is different from the area in which it happens to be employed at a given moment). Later (in passage 5) it will emerge that the term glossa creates some obscurity in the discussion and therefore needs to be explained further. There follows the important distinction whereby the same lexical item may be both glossa and ordinary, but not for the same people, that is to say, not in the same time period and in the same place. The example Aristotle offers is dialectal and geographical rather than chronological: he cites the word σίγυνον, which belongs to the dialect of Cyprus (with the meaning “spear,” as we find already in Herodotus V 9.3);8 thus in Cyprus this word is linguistically κύριον, whereas it is a glossa “for us,” i.e., among speakers of Attic. In chapter 22 Aristotle is concerned to define the distinctive qualities of poetic language, which (as he states) must be elevated and must differ from ordinary language, thanks to the use of linguistic forms that are foreign to the latter (cf. above, along with passage 2). Any and all expressions that are foreign to everyday speech may contribute to the creation of a style that is distinct from the ordinary and the common; at the same time, however, it is important not to make excessive use of glossai, metaphors, or other kinds of linguistic ornamentation, since these may constitute an obstacle to comprehension — ordinary speech, on the other hand, is the foundation of clarity in expression: cf. 22.2, 1458 a 34 and 1458 b 2–5 — and they may even lead to ridiculousness (22.4, 1458 b 13–14). Aristotle invites the reader to consider epic as the example par excellence of the appropriate usage of these qualities of language (1458 b 15–18), and further on (passages 3 and 4) he asserts that the use of glossai is particularly suitable for epic: in passage 3 he speaks only of glossai, whereas in passage 4 he speaks of both glossai and metaphors (the latter of which he had previously reserved for iambic poetry, in passage 3); in any case, the specific connection  7 Lucas 1972, 208, ad loc.; Barabino 2010. 8 Herodotus V 9.3: σιγύννας δ᾿ ὦν καλέουσι Λίγυες οἱ ἄνω ὑπὲρ Μασσαλίης οἰκέοντες τοὺς καπήλους, Κύπριοι δὲ τὰ δόρατα. This glossa is found abundantly in later lexica and in other scholarly works.

Dialect Glossai in the Scholia to Homer  

between glossai and epic (τὸ ἡρωικόν) is both consistent and emphasized, even if it is obviously in no way exclusive, given that glossai are employed to a great extent in the tragic style as well, as is clear from the examples Aristotle adduces at 22.4–5 (1458 b 11–1459 a 9). Chapter 25 illustrates a set of exegetical criteria that may be employed in order to interpret difficulties encountered in poetic language. At the beginning of the chapter, Aristotle states the general rule that, since the objects of mimesis are communicated by means of language, it is necessary to take into consideration glossai, metaphors, and “many [other] transformations of language” (πολλὰ πάθη τῆς λέξεως). He then develops this argument further (passage 5) by presenting three Homeric passages where an interpretative problem may be resolved (in his view, at least) by assuming that a problematic word is being used in an obsolete or in a dialectical sense: in this way, one can eliminate illogicalities, inconsistencies, or other inconvenient aspects of the text. Now, it is not my purpose here to examine the validity of the interpretative problems Aristotle presents or of the solutions that he offers.9 Instead, for my present purposes, it is important simply to note that, in the second example, the problem’s lysis involves having recourse to the notion of dialect. In Il. 10.316 we read that Dolon εἶδος ἔην κακὸς ἀλλὰ ποδώκης,10 and for Aristotle the apparent contradiction can be resolved by interpreting the first adjective as meaning not that Dolon had a bodily deformation (which would contradict the fact of his being a swift runner), but rather that he had an unattractive face: according to Aristotle, we are dealing with a dialect usage deriving from Crete, since in that dialect εὐειδές means εὐπρόσωπον. In his discussion of λέξις, therefore, Aristotle establishes a definition of glossa, which he understands to be a stylistic element typical of poetic language and particularly typical of epic (and tragedy as well: cf. chapter 25.4–5, 1458 b 11–1459 a 3).11 Aristotle also discusses the question of glossai in Book 3 of the Rhetoric (chapters 1–3), where he includes explicit references to the Poetics (1404 b 7, 1404 b 28, 1405 a 5). In the course of his discussion of the difference between the style of poetry and the style of prose, Aristotle asserts that those literary embellishments that are obtained by means of linguistic forms different from ordinary language are not, in general, suitable for prose, but rather are

 9 See (ad loc.) the commentaries of Bywater 1909; Gudeman 1934; Lucas 1972; and Gallavotti 1974. 10 The last part of the verse (ἀλλὰ ποδώκης) is not present in the text of the Poetics as it has come down to us. 11 Pfeiffer 1968, 12, 75–79 = 55, 142–147.

  Dialect Glossai in the Scholia to Homer typical of poetry: the effect of estrangement (διὸ δεῖ ποιεῖν ξένην τὴν διάλεκτον, 1404 b 10) is common and indeed appropriate in poetry, but much less so in prose — especially given that an essential quality of prose style is clarity, and clarity is obtained by employing what is linguistically κύριον.12 Among the stylistic ornaments found in elevated language (λέξις κεκοσμημένη, as opposed to λέξις ταπεινή), Aristotle explicitly cites glossai (the term is found at 1404 b 28, 1406 a 6, 1406 b 2). Aristotle does point out, however, that there has been an evolution in prose style over time: “The first prose style was poetic, for instance that of Gorgias, and even today the majority of uneducated people think that the best orators are the ones of this sort, whereas in fact such an idea is mistaken, since the style of prose is different from that of poetry”; and even in the language of tragedy, one finds a growing tendency to use ordinary language, both in terms of meter (in the shift from the trochaic tetrameter to the iambic trimeter, which is closer to spoken language) and also in terms of diction. With regard to the latter, Aristotle states that the tragic poets “have even eliminated those words that are foreign to ordinary language (ὅσα παρὰ τὴν διάλεκτον, i.e. glossai),13 which earlier tragic poets employed to embellish their works and which epic poets continue to employ in composing their hexameters” (1404 a 25–35). Thus, alongside the element of “archaism” that may be present in the use of glossai, we find once again the close connection between glossai and hexameter epic poetry (which is now viewed as running counter to the current trend), and this connection is reiterated at 1406 b 2, where Aristotle affirms again that glossai are particularly suitable for use by epic poets (οἱ ἐποποιοί). Aristotle, of course, has not invented ex nihilo the problem of how to interpret difficult words found in poetic texts, nor has he coined the term glossa for such words. Rudolf Pfeiffer has already discussed in detail the major antecedents to Aristotle’s treatment of these issues; as a result, it seems necessary at present only to mention these antecedents in brief outline. First of all, I believe that Pfeiffer is entirely correct when he asserts, first, that the rhapsodes themselves, in the course of their recitations, must have already faced the necessity of offering rudimentary interpretative comments, for instance explanations of difficult words — many of which were without question utterly foreign to the ordinary language of their audiences; and second, that we can see traces of this

 12 Cf. also Topica VI 2, 140 a 2–6: πᾶν γὰρ ἀσαφὲς τὸ μὴ εἰωθός. 13 Slightly later, Aristotle remarks that Euripides was the one who paved the way for the use of words from everyday language in poetic compositions (1404 b 24–25).

Dialect Glossai in the Scholia to Homer  

early form of exegesis in the texts of the archaic and classical poets.14 Pfeiffer also highlights the significance of the Sophists, with their interest in language and their studies dedicated to specific linguistic phenomena. Among the Sophists, Prodicus stands out for his reflections on linguistic problems and in particular for his treatment of synonyms and semantic distinctions, which (according to Pfeiffer) led him “to an awareness of the different usage in different parts of the country” — to the extent that he even claimed that Pittacus was unable to correctly distinguish words, since he was from Lesbos and had been educated in a foreign idiom (Plat., Prot. 341 c). Pfeiffer also notes that, “in his Cratylus, Plato seems to be reproducing similar Sophistic discussions from a source of the fifth century about ξενικὰ ὀνόματα, when he refers to Aeolic or Doric words as ‘alien’, that is, different from the familiar Attic form”.15 In this general sketch — which we are drawing in brief — one should also mention Democritus’ work Περὶ Ὁμήρου ἢ ὀρθοεπείης καὶ γλωσσέων (the title of which, however, may perhaps be doubtful).16 Finally, there is the well-known fragment (233 PCG) from Aristophanes’ Δαιταλῆς (a comedy performed in 427 BCE),17 in which we find one of the earliest testimonies to Homeric exegesis in general and, in particular, to the practice of explaining difficult words in the context of education in fifth century Athens. Although the fragment presents some textual problems, its overall sense seems both clear and generally accepted by scholars: a young man is asked (apparently by his father) to explain and to translate certain Homeric glossai (λέξον Ὁμήρου γλώττας); there follows a request to explain a word that perhaps comes from a work of Solon, and then another that comes from an unknown work.18 We are dealing, then, with two Homeric glossai and two glossai deriving from other texts; each is presented as problematic due to its being archaic and obsolete.

 14 Pfeiffer 1968, 4–5, 12, 79 = 45–46, 55, 147. On the earliest traces of the interpretation of problematic words in the poets themselves, cf. also Montanari 1976, 208–210, with bibliography (= this volume, ch. 27). 15 Pfeiffer 1968, 37–43 = 1973, 91–99 (the citations are from p. 41 = p. 96, where Pfeiffer also points out that Herodotus (I 142.3–4) distinguishes between several variants of Ionic spoken in different cities (see Asheri 1988, 348 ad loc.) 16 Pfeiffer 1968, 42–43 and 79 = 1973, 97–99 and 147. 17 Cf. Pfeiffer 1968, 14–15 and 79 = 1973, 59 and 147; Cassio 1977, 75–77. 18 The word in question is ἰδῦοι or ἰδυῖοι (cf. Cassio ad loc.), which comes from the Laws of Solon according to Aelius Dionysius ι 4 Erbse (cited by Eustathius 1158,20). To be exact, Aelius Dionysius attributes the word to both Draco and Solon, whereas Photius (Lex. ι 36) attributes it only to Solon (fr. 41 b Ruschenbusch); clearly it is a word that derives from legal language. Cf. Montanari 2003 (= this volume, ch. 40).

  Dialect Glossai in the Scholia to Homer In short, we have sufficient evidence to believe that Aristotle has taken a pre-existing practice — which was already very widespread, both within an educational setting and also within broader cultural contexts of various kinds, ranging from the activity of the rhapsodes to the linguistic reflections of the philosophers — and he has codified this practice as a method of exegesis and as a specific criterion for stylistic analysis. However, it is important to emphasize that this exegetical method is based (as was inevitably the case) upon the observation of an essential and distinguishing characteristic of the language of Homeric epic (and of course, although perhaps to varying degrees, of the other “high” genres of poetry, especially tragedy.) Indeed, given that, as a general rule, the observation of specific phenomena is the foundation upon which stands the legitimacy of any given method of exegesis, one may invoke the glossa in the context of the interpretation of poetry precisely because the glossa is itself a recognized element of poetic language and style. Let us now try to follow the thread of the glossa, as foreign element due to dialectal reasons (without forgetting, however, that Aristotle took pains to emphasize that the same lexical item may be both glossa and ordinary, but not for the same individuals, that is to say, not in the same time period and in the same place). Among the examples of glossai that Aristotle adduces in the Poetics, two are clearly of a dialectal nature: 1) the word σίγυνον, which belongs to the Cypriot dialect, and therefore in Cyprus it is linguistically κύριον whereas it is a glossa for the Athenians; 2) εἶδος κακός (in Il. 10.316), which means that Dolon had an unattractive face, rather than a bodily deformation, and belongs to the dialect of Crete (where εὐειδές means εὐπρόσωπον) and is therefore a glossa in Homer. To these we may add the compound word formed of three elements (Ἑρμοκαϊκόξανθος), which Aristotle cites as typical of the dialect of Marseille (chap. 21, 1457 a 35); of course, compound formations are in and of themselves a central element of the high style in poetic language, and in this case we may speak of a compound that is also a glossa for reasons of dialect. If we now turn, finally, to the Homeric scholia, as a first step we may cast a quick glance at the lemma Dialecti within Index V: Res potiores, imprimis grammatica et rhetorica in Erbse’s edition of the Scholia in Iliadem (vol. VII, 84–88). Here we find a huge quantity of items listed, and we are immediately disabused of any ingenuous desire to analyze each and every one of them (especially if one were to add examples from the Odyssey to those cited for the Iliad!). If we then seek out a point of departure within this vast sea of material, a good place to begin might be to verify whether or not, within the corpus of Homeric scholia we currently possess, we find any mention of dialect glossai attributed to the two geographic areas Aristotle cites in the Poetics, namely Cyprus and Crete. As a

Dialect Glossai in the Scholia to Homer  

starting-point, this is admittedly somewhat arbitrary, but the results seem, in my opinion, rather interesting. Let me begin, then, by mentioning that I have found no less than 13 glossai attributed to the dialect of Cyprus, all of them within the Iliad. 1) 2)

3) 4) 5)

6) 7) 8) 9) 10)

11)

12) 13)

5.387 χαλκέῳ δ’ ἐν κεράμῳ: χαλκῷ ἀγγείῳ, πίθῳ, ἢ δεσμῳτηρίῳ. οἱ γὰρ Κύπριοι τὸ δεσμωτήριον κέραμον καλοῦσι. D 12.29 οὕτως Ἀμαθούσιοι; cf. Eust. 890.62 τὸ δὲ «φιτρῶν καὶ λάων» Ἀμαθουσίων γλώσσης ἐστίν, ὥς φασιν οἱ παλαιοί. Κύπριοι δὲ οἱ Ἀμαθούσιοι. (Erbse test. ad loc.). 14.479 ἰόμωροι: ἰοῖς μόρον ἐπάγοντες, ἢ περὶ ἰοὺς κακοπαθοῦντες, ἢ ἰοὺς ὀξεῖς ἔχοντες· μωρὸν γὰρ τὸ ὀξὺ Κύπριοι· ὅθεν καὶ “ὑλακόμωροι” (ξ 29. π 4). 15.263 ἀκοστήσας: Κύπριοι ἀκοστὰς τὰς κριθάς. 17.51 κόμαι Χαρίτεσσιν ὁμοῖαι: ἀντὶ τοῦ ταῖς τῶν Χαρίτων κόμαις ἶσαι. ἀπίθανον δέ ἐστιν. ἀμόρφωτος γὰρ ἡ κόμη, εἰ μὴ οὕτως. Μακεδόνες δὲ καὶ Κύπριοι Χάριτας λέγουσι τὰς συνεστραμμένας καὶ οὔλας μυρσίνας, ἃς φαμὲν στεφανίτιδας. D 21.262 χώρῳ ἔνι προαλεῖ: Παρμενίων ὁ Βυζάντιος· “Κύπριοι καὶ Ἀρκάδες καὶ Λακεδαιμόνιοι προαλὲς κάταντες.” 21.329 πνίξειε. Κυπρίων ἡ λέξις. 21.455 Eust. 1246.26 … ἄλλοι δὲ «ἀπολουσέμεν», ἤτοι κολοβώσειν· λοῦσον γάρ, φασί, παρὰ Κυπρίοις τὸ κολοβόν (Erbse, test. ad loc.). 22.441 … δηλοῖ δὲ κατὰ Κυπρίους τὸ ποικίλλειν, ἀφ’ οὗ καὶ παστός. — ἔπασσε: πάσσειν Κύπριοι τὸ ποικίλλειν, ἀφ’ οὗ καὶ ὁ παστός.19 23.1 πτόλιν: Κυπρίων τῶν ἐν Σαλαμῖνι ἡ λέξις· cf. Eust. 842.63 ὁ δὲ πτόλεμος Κυπρίων καὶ Ἀττικῶν λέξις καθ’ Ἡρακλείδην ἐστί, καθὰ καὶ ἡ πτόλις (Erbse test. ad loc.). 23.29 τάφον … Κύπριοι δὲ τάφον τὸν φόνον· “πολέων τάφῳ ἀνδρῶν ἀντεβόλησα” (λ 416, ω 87), ὅ ἐστι φόνῳ. — οἱ Κύπριοι καὶ τὸν φόνον τάφον καλοῦσι. D 24.323 … Σικελοὶ δὲ τὴν αὔλειαν θύραν. Κύπριοι δὲ παστάδα ἀμφίθυρον. 24.753: ἀμιχθαλόεσσαν: κατὰ Κυπρίους εὐδαίμονα.

 19 In the same verse we find the word θρόνα, for which the scholion to Theocritus 2, 59–62 (citing Cleitarchus) mentions various different meanings found among the Θεσσαλοί, the Κύπριοι, and the Αἰτωλοί (cf. Erbse, 1969–1988, testimonium ad loc.).

  Dialect Glossai in the Scholia to Homer We cannot examine every single instance in detail; a few observations on certain individual scholia should suffice. In no. 1 above, the possible Cypriot glossa adds an important nuance to the meaning of the passage: the extraordinary exploit whereby the two Aloadae, Ephialtes and Otos, emprisoned Ares within a huge bronze vase becomes even more charged with significance since that the word used by the poet to indicate the vase, was also used in Cyprus to mean “prison” (δεσμωτήριον). In no. 2 we can observe how the reference to the dialect of Cyprus is rendered even more precise by the specification that the glossa belongs to the spoken language of the city of Amathous. In no. 3 the compound ἰόμωρος is explained as containing ἰός ‘arrow’ as its first element; but for the second element, we find that the scholiast adduces support for the possible Cypriot glossa (μωρόν = ὀξύ) by means of two internal citations within the genre of epic poetry, which refer to an adjective used twice in the Odyssey. In no. 4 the meaning of the verb ἀκοστάω, used of horses that feed at the trough, is explained with reference to ἀκοστή, which according to the scholiast is a Cypriot word meaning ‘barley’. Elsewhere, however, we find that the very same word is described as Thessalian: in sch. 6.506–8 we read, οἱ δὲ παρὰ Θεσσαλοῖς ἀκοστὰς τὰς κριθάς. And in 6.506 b we find instead: κυρίως (NB) πᾶσαι αἱ τροφαὶ ἀκοσταὶ καλοῦνται. In no. 8 a Cypriot glossa is invoked to explain the variant reading ἀπολουσέμεν, which we find cited only in Eustathius (with a parallel in Hesychius α 6465); in the manuscript tradition of the epics and in the scholia, this variant is otherwise completely absent. In no. 9 the fact that πάσσειν = ποικίλλειν in Cypriot usage would create a sort of figura etymologica (ποικίλα ποικίλλειν) in the expression ἐν δὲ θρόνα ποικίλ’ ἔπασσε. This is not necessary to understand the passage, but it contributes to the elevation of the style to a certain degree. In no. 13 the idea that, in the Cypriot dialect, ἀμιχθαλόεσσαν could mean εὐδαίμονα is adduced as a possible explanation of the epithet of uncertain meaning found in Λῆμνον ἀμιχθαλόεσσαν (cf. LfrE s.v.). In four of the passages cited above, other dialects are invoked alongside that of Cyprus: Macedonian (no. 5), Arcadian and Lacedaemonian (no. 6), Attic (no. 10), and Sicilian (no. 12). Below we will have reason to discuss Attic at greater length; for now I offer a few brief considerations regarding the other dialects mentioned. Macedonian. The word ἄττα is described as Macedonian in the scholion to Od. 16.31 and as Thessalian in the D-Scholion to Il. 9.607 (cf. LfrE s.v.). In the

Dialect Glossai in the Scholia to Homer  

scholion Il. 20.404c, a Macedonian word with /a/ vocalism is mentioned along with a group of words that are found παρὰ Δωριεῦσι. Arcadian. In the scholion Il. 13.390a, for the Homeric adjective βλωθρή a range of different meanings are supplied by adducing the usages of various dialects (cf. LfrE s.v.): Arcadian, Boeotian, Magnesian (is this the Magnesia in Thessaly, or in Caria, or in Lydia?), along with the dialect of the Dryopes (an Ionic population: cf. Herodotus I 146.1, etc.), the dialect of the Tyrrhenians, and the dialect of Carystus (in Euboea). In the scholion to Il. 14.385, the meaning that the word ἄορ possessed among the Arcadians and the Aetolians is distinguished from its meaning in the Homeric text. Lacedaemonian. Another glossa attributed to the dialect of the Λακεδαιμόνιοι is found in the scholion Il. 16.184b τὰ μετέωρα δὲ τῶν οἰκημάτων ὦα Λακεδαιμόνιοι καλοῦσιν, ἃ νῦν ὑπερῷά φαμεν. The dialect of the Λάκωνες is invoked to explain glossai at Il. 6.43 (στόμα τὸ πρόσωπον. οὕτως Λάκωνες), 16.221 (along with the Μυτιληναῖοι), and 23.83c (Λάκωνες τὸ θάπτειν τιθήμεναι φασίν). Sicilian. I have not found any other references to the Sicilian dialect. An assertion found in the scholion to Od. 20.383 might have linguistic implications as well: ἐς Σικελοὺς: ἐγινώσκετο ἄρα τὰ κατὰ τοὺς Σικελούς ... διαφέρουσι δὲ Σικελιῶται Σικελῶν· οἱ μὲν γὰρ Ἕλληνες, οἱ δὲ βάρβαροι. At this point we may consider the Cretan dialect. We find five glossai that are described as belonging to this dialect, four within the Iliad and one in the Odyssey; there is, however, no trace within the scholia of the glossa in Il. 10.316 that Aristotle discusses in the Poetics (cf. above, passage 5, 1461 a 10–15, and see Erbse ad loc.). 1)

Il. 1.39. Σμινθεῦ: Ὦ Σμίνθιε. ἔστι δὲ ἐπίθετον τοῦ Ἀπόλλωνος. Σμίνθος γὰρ τόπος τῆς Τρωάδος, ἐν ᾧ ἱερὸν Ἀπόλλωνος Σμινθίου, ἀπὸ αἰτίας τοιᾶσδε· ἐν Χρύσῃ πόλει τῆς Μυσίας Κρῖνίς τις ἦν ἱερεὺς τοῦ κεῖθι Ἀπόλλωνος. τούτῳ δὲ ὀργισθεὶς ὁ θεὸς, ἔπεμψεν αὐτοῦ τοῖς ἀγροῖς μύας, οἵτινες τοὺς καρποὺς ἐλυμαίνοντο. βουληθεὶς δέ ποτε ὁ θεὸς αὐτῷ καταλλαγῆναι, πρὸς Ὄρδην τὸν ἀρχιβούκολον αὐτοῦ παρεγένετο· παρ’ ᾧ ξενισθεὶς ὁ θεὸς ὑπέσχετο κακῶν ἀπαλλάξειν, καὶ δὴ παραχρῆμα τοξεύσας τοὺς μῦς διέφθειρεν. ἀπαλλασσόμενος οὖν ἐνετείλατο τὴν ἐπιφάνειαν αὐτοῦ δηλῶσαι τῷ Κρίνιδι· οὗ γενομένου ὁ Κρῖνις ἱερὸν ἱδρύσατο τῷ θεῷ, Σμινθέα αὐτὸν προσαγορεύσας, ἐπειδήπερ, κατὰ τὴν ἐγχώριον αὐτῶν διάλεκτον οἱ μύες σμίνθοι καλοῦνται. ἡ ἱστορία παρὰ Πολέμωνι (fr. 31 Preller). ἄλλοι δὲ οὕτως εἶπον, ὅτι Κρῆτες ἀποικίαν στέλλοντες χρησμὸν ἔλαβον παρὰ τοῦ Ἀπόλλωνος ὅπου αὐτοῖς ἐναντιωθῶσιν γηγενεῖς — ἔλεγεν δὲ περὶ μυῶν — ἐκεῖ κτίσαι τὴν πόλιν. οἱ δὲ ἀπέλυσαν τοὺς ἀποίκους. ἐλθόντες δὲ εἰς τὸν Ἑλλήσποντον καὶ τῆς νυκτὸς ἐπιγενομένης, μύες ἔκοψαν αὐτῶν τοὺς τελαμῶνας τῶν ὅπλων. πρωῒ δὲ ἀναστάντες καὶ θεασάμενοι τοῦτο, σοφισάμενοί τε καθ’ ἑαυτοὺς, ἔκτισαν ἐκεῖ πόλιν, ἥντινα ἐκάλεσαν Σμινθίαν· οἱ γὰρ Κρῆτες τοὺς μύας σμίνθους καλοῦσιν. ἐκ τούτου καὶ Ἀπόλλων Σμίνθιος ἐκλήθη διὰ τὸ ὑπερασπίζειν αὐτῆς. D (ed. H. van Thiel).

  Dialect Glossai in the Scholia to Homer 2) 3) 4) 5)

Il. 8.448 Eust. 722,60 καὶ ἔστιν κατ᾽ αὐτὸν τὸ “οὐ μέντον” Ἀργείων καὶ Κρητῶν γλώσσης. Il. 12.77 πρυλέες: οὕτω Γορτύνιοι. Il. 13.600 θεράπων: δοῦλος ὁπλοφόρος κατὰ Κρῆτας. Od. 3.444 Περσεὺς δ᾽ ἀμνίον εἶχε: τὸ αἱματοδόχον ἀγγεῖον. Καὶ γίνεται ἀπὸ τοῦ αἷμα αἱμνίον καὶ ἀμνίον … ἔστι δὲ τῶν ἅπαξ εἰρημένων ἡ λέξις. ἄλλοι δὲ μικρὸν μαχαιρίδιον, ὅ καὶ σφάγιον καλοῦσιν οἱ Ἀττικοί. — ἀμνίον ὡς πηνίον. Κρῆτες δὲ αἱμνίον αὐτό φασι.

In no. 1 we find that the word σμίνθοι is described as commonly meaning μύες in the local dialects of Crete and Mysia. In the Homeric text it constitutes a dialect glossa, knowledge of which is essential in order to understand the meaning of the epithet that the priest Cryses uses when he addresses Apollo. It is interesting to note that, even though two different mythological narratives are cited (one for Crete and one for the Mysian city), nonetheless the epithet for Apollo is explained with reference to the noun meaning ‘mouse’ in each of the two dialects. Regarding no. 2, Eustathius (cf. Erbse, test. ad loc.) is the only source for the variant μέντον (which goes back to Heracleides of Miletus) for the reading μέν θην transmitted in the manuscripts; μέντον would appear to be an expression deriving from the dialect of Argos and from that of Crete. In no. 3 a Cretan glossa is identified and is assigned specifically to the dialect of Gortyn (cf. Eust. 893,32; Erbse test. ad loc.). In no. 5, for the word ἀμνίον a first possible meaning is presented, namely “vase that collects the blood” of victims (αἱματοδόχον ἀγγεῖον), and for this meaning the scholiast cites the form αἱμνίον (from αἷμα), which is ultimately characterized as Cretan. Alternatively (ἄλλοι δέ) a second possible meaning is mentioned, according to which ἀμνίον would indicate instead a small knife (μικρὸν μαχαιρίδιον), which will presumably have been used to slit the throats of sacrificial victims; there follows a reference to the Attic dialect, with the statement that this ἀμνίον (= μαχαιρίδιον) is the same item that Attic-speakers call σφάγιον (which latter term must be understood either as an adjective, “that which kills, that which sacrifices, fatal,” modifying an understood μαχαιρίδιον, or else as a substantive, “killer, sacrificer”). In short, the two possible interpretative solutions would appear to be as follows: ἀμνίον (Cretan αἱμνίον) = αἱματοδόχον ἀγγεῖον, or else ἀμνίον (Cretan αἱμνίον) = μικρὸν μαχαιρίδιον, in Attic σφάγιον. In any case, it is clear that this scholion does not discuss an Attic glossa in the Homeric text; rather, it cites the Attic equivalent of one of the possible meanings possessed by the Homeric word, which is, if anything, very close to a Cretan lexical item. In other words, Attic is merely employed as language of reference in order to ensure that the meaning of the term is understood. This survey is perhaps already too long for the present context; its purpose, in any case, is merely to offer a set of examples, which at this point seems

Dialect Glossai in the Scholia to Homer  

substantial enough (however limited it may be). We have seen that ancient scholars in their study of Homer referred to numerous dialects belonging to a wide geographic range; these dialects are more often identified by the name of the people who spoke them, less often by the name of the place where they were spoken: thus we find Κύπριοι for the Cypriot dialect, Κρῆτες (Cretan), Ἀττικοί/ Ἀθηναῖοι/Ἀτθίς (Attic), Μακεδόνες (Macedonian), Ἀρκάδες (Arcadian), Λακεδαιμόνιοι (Lacedaemonian), Σικελοί (Sicilian), Θεσσαλοί (Thessalian), Δωριεῖς (Doric), Βοιωτοί (Boeotian), along with references like “of Magnesia” (the one in Thessaly, or Caria, or Lydia?), “of the Dryopes” (a population with Ionic origins, cf. above), “of the Tyrrhenians”, “of Carystus” (in Euboea), “of Mytilene”, “of Mysia”. Overall — taking into account the indices to critical editions as well as my own searches using the digital TLG — we find denominations of varying taxonomic breadth, from those referring to the larger dialect groups (Αἰολεῖς/ Αἰολικός/Αἰολίς, Δωριεῖς/Δωρικός/Δώριος, Ἴωνες20/Ἰακός/Ἰωνικός), to those referring to regional areas (these are in fact the most numerous: Attica, Thessaly, Boeotia, Cyprus, Crete, Macedonia, Aetolia, etc.), to those that specify individual cities (like Mytilene, Gortyn in Crete, Amathous in Cyprus). I must admit that, even though for some time now I have had an underlying sense that there was a vast and rich range of references to the ancient Greek dialects within the Homeric scholia, and even though I knew in advance that this contribution would not be able to cover each and every case, still, when I finally began the concrete task of examining the material in detail, what I found proved to be even more surprising than I could have imagined. For, above and beyond that which we can analyze directly, we must recall two important facts: 1) that which remains within the collections of scholia we currently possess represents only a small part of the enormous quantity of philological scholarship produced by the ancients; 2) the data that I have assembled here could easily be extended and enriched by culling items from other important sources, such as the vast sea of lexicography, or Eustathius (whom I have cited above, but only sporadically), or various other studies dedicated to the Homeric texts. In short, there exists a massive quantity and variety of materials related to dialectology invoked by ancient criticism of Homer, and we thus find ourselves faced with a phenomenon of extraordinary interest and significance. Here I shall limit myself to sketching a framework within which future researches might proceed in order to better understand the various possible implications — at the level of geography, mythology, and anthropology (local customs and practices) — that may be latent in the many references to dialects  20 Ἰάονες in sch. Il. 13.685.

  Dialect Glossai in the Scholia to Homer found within ancient Homeric scholarship. As a first example, one may recall the case of Il. 1.39 and the two different mythological narratives regarding Apollo, one set in Mysia and the other set on Crete, which are presented in an attempt to explain the god’s epithet by an appeal to a shared dialect term for mice (σμίνθοι = μύες): Apollonius Sophista (143.9) informs us that Aristarchus believed the epithet to be related to Sminthe, in the Troad (where a cult of Apollo Smintheus did exist, though in fact this cult was found even outside the Troad and indeed not far from Mysia) and that Apion invoked a religious procession on Rhodes called the Smintheia, which celebrated Apollo’s protection of the community against an infestation of mice. I leave aside the question (one might say the temptation) to analyze this set of data from the point of view of Greek dialectology as it is currently understood within the modern fields of linguistics and philology. This type of research, in any case, extends well beyond my own personal range of expertise, and as a result I do not think it prudent to guess what the possible results of such inquiries might look like. At the beginning of this essay, I offered a highly schematic overview of the various elements which make up the rich panorama of ancient studies in dialectology, to which the present material can be linked. For the moment, however, I would like to return to the question of poetic style and poetic language. For ancient scholars and critics, the analysis of literary language (be it that of a specific author or of an entire genre) is anchored to a fundamental principle: every author writes in a language that is characterized overall, and in essence, by the cultural and geographic context in which that author lived and acted; within a given author’s language, however, there may be elements, more or less common, that derive from other dialects and play the role of glossai in the manner we have described above, following Aristotle’s analysis. To wit: a glossa is a stylistic embellishment insofar as it introduces an element of “estrangement”; an individual word is a glossa when it is inserted in the context of a different base dialect, whereas it is not a glossa when it is employed within the habitual context of its dialect of origin.21 The passages we have discussed above demonstrate that the identification and the explanation of dialect glossai may aid in the interpretation of a text or clarify some aspect of the text that is not immediately, or not entirely, comprehensible. Thus, the developments in dialectology over the course of time, and the increasing wealth of materials available to scholars within treatises and  21 On these concepts, see most recently Cassio 1993a; 1993b; 1997, along with the summaries in Cassio 2008 (especially chapter 3, 70 ff., and the contribution by C. Vessella, 392–395).

Dialect Glossai in the Scholia to Homer  

collections of various kinds, provide modern philologists with an ever more powerful and ever more productive tool.22 However, whereas the methodological basis for the concept of glossa and for the search of the glossa as a hermeneutical tool now seems to be clear, it must be admitted that the concrete applications of this concept/tool, as found in ancient scholarship, often leave one somewhat perplexed; as a result, it seems to me that we must consider this phenomenon from another point of view as well. It is not my intention to re-examine in detail the three examples of Homeric glossai that Aristotle adduces in chapter 25 of his Poetics, since the facts regarding these are generally well-known. I would like to recall one case only, the one in which Aristotle mentions dialect as a determining factor: namely, Il. 10.316, where Dolon is described as εἶδος κακὸς ἀλλὰ ποδώκης. It seems legitimate to ask whether it is really necessary to have recourse, as Aristotle does, to a Cretan glossa (εὐειδές = εὐπρόσωπον) in order to explain how the expression is not self-contradictory insofar as, by this reading, εἶδος κακός would not indicate bodily deformity but simply ugliness, which does not contrast in any way with the fact of being a good sprinter. In a similar vein, among the other cases discussed above, let us re-examine that of Il. 17.51, in which, à propos of Euphorbus (who has been killed by Menelaus) we find the following verse: αἵματί οἱ δεύοντο κόμαι Χαρίτεσσιν ὁμοῖαι. The D-Scholion comments: ἀντὶ τοῦ ταῖς τῶν Χαρίτων κόμαις ἶσαι. ἀπίθανον δέ ἐστιν ... Μακεδόνες δὲ καὶ Κύπριοι Χάριτας λέγουσι τὰς συνεστραμμένας καὶ οὔλας μυρσίνας, ἃς φαμὲν στεφανίτιδας. Here too one may legitimately ask whether the problem (and thus the corresponding recourse to a dialect glossa for the solution) has not simply been “invented”; indeed, the first explanation of Χαρίτεσσιν ὁμοῖαι offered in the scholion does not seem to present any real difficulty.

 22 As in other cases, this development within Alexandrian philology must have been prompted by an Aristotelian/Peripatetic influence. There were numerous factors that encouraged research into the Greek dialects and their respective differences: for instance, the interest in the customs and practices of different areas of Greece (seen, for instance, in the collection of Politeiai, which abounded in such information); the study of language, both in and of itself and also as employed in poetry (Rhetoric and Poetics); and the researches — conducted already by the first generations of the peripatetics — into the lives and the works of individual poets, which pointed out the dialect differences. This is one of several aspects that should be studied further regarding the relationship between Aristotle and the Peripatetic school, on the one hand, and Hellenistic philology and scholarship, on the other: cf. Montanari 1994a; 2000b (= this volume, ch. 55); 2001; and 2008 (= this volume, ch. 54); Schironi 2009; summary in Montanari 2012 (= this volume, ch. 56).

  Dialect Glossai in the Scholia to Homer Now, we have seen above that the interpretation of the epithet Smintheus for Apollo in Il. 1.39 is significantly enriched by the ancient scholars’ reference to considerations of dialect, along with the narration of the relevant mythological events in which Apollo was the protagonist. In other cases, however, one must admit that the invocation of a dialect term in order to explain a passage does not seem at all necessary to the understanding of the text. Perhaps it may sound somewhat harsh, but there does at times seem to be within the exegetical tradition a desire to “invent” interpretative difficulties, to “discover” a glossa within the poetic text at any cost, with the sole purpose of presenting facts of dialectology merely to have the chance to explain them, or else to enrich the scholarly tradition itself by adding to the range of possible interpretations of a given passage (in this regard one thinks of those cases in which we find a series of possible meanings, each corresponding to a different dialect, for example in the scholion Il. 13.390a, cited above with reference to the Arcadian dialect). In short, although the reference to a given dialect does sometimes appear to be concretely useful for understanding an obscure passage in which a glossa is present, at other times we seem to find this practice employed with the intent of adding a nuance that, in one way or another, grants greater value to the meaning of the text, or else with the sole intent of offering an information that is essentially of scholarly interest only. Thus, it seems that we can observe in the exegetical tradition a desire to pursue additional details that prove to be of interest to the interpreter alone; in other words, it seems, at times, that we are witnessing a kind of “hunt for glossai”, carried out simply in order to have the chance to explain them by adducing a dialect item which possesses certain additional nuances of meaning or permits the mention of certain antiquarian details. Somewhat differently, one might even go so far as to say that we can observe at work, in these cases, the specific desire of revealing a Homeric text that is richly, and uniquely, characterized by the simultaneous presence of lexical items deriving from a wide range of dialects. In this way, the Aristotelian method of identifying and analyzing glossai in order to explain the text — an exegetical tool offered to scholars and critics on the basis of a specific theory of language and style — has been repurposed for a new use, much broader than that for which it was originally intended, with the result that the linguistic treasure-trove found in the text of Homer is enriched by a considerable number of “discoveries” — “discoveries”, however, that are rather debatable as far as Homeric language is concerned (even if this does not mean — and here we encounter yet another aspect of the problem — that they are wrong or unreliable if considered simply as evidence for dialect usage in and of itself). In short, there was a notion that any given author’s poetic language consisted of a mix of

Dialect Glossai in the Scholia to Homer  

elements deriving from different dialects; but whereas for other authors this was understood to mean the occasional use of foreign dialect terms within a well-characterized base dialect, in the (unique and peculiar) case of Homer this idea seems to have been understood, ultimately, as meaning that the Homeric text was composed of elements deriving from all the known dialects of Greece, items of which could then be discovered within the poems with stubborn persistence, and at any cost. Such an idea finds expression in a passage of ps.Plutarch’s Περὶ Ὁμήρου B 8: Λέξει δὲ ποικίλῃ κεκχρημένος τοὺς ἀπὸ πάσης διαλέκτου τῶν Ἑλληνίδων χαρακτῆρας ἐγκατέμιξεν, ἐξ ὧν δῆλός ἐστι πᾶσαν τὴν Ἑλλάδα ἐπελθὼν καὶ πᾶν ἔθνος. There follows an exemplification of this concept with instances of the Homeric usage of Doric, Aeolic, and Ionic, concluding with the affermation (chap. 12) μάλιστα δὲ τῇ Ἀτθίδι διαλέκτῳ κέχρηται, and then repeating (chap. 14) ὅλως μὲν οὖν τὰς πάντων τῶν Ἑλλήνων φωνὰς ἀθροίζων ποικίλον ἀπεργάζεται τὸν λόγον καὶ χρῆται ποτὲ μὲν ταῖς ξέναις. It was Aristarchus who had maintained that Homer was of Athenian origin and that the language of Homer was an archaic form of Ionic, spoken in Attica in ancient times, before it spread into the areas later known as Ionia; Homer then enriched this base language (according to Aristarchus) by making use of dialect forms belonging to other areas of Greece, which he would have encountered in his travels. By this interpretation, the specific case of Homer can be made to fit within the general model applied to other authors: a base dialect, well-characterized within a specific cultural and geographic context (in which the poet himself lived), is enriched by the addition of foreign elements from other dialects that serve as glossai. Such an interpretation of Homer’s language is, one may add, coherent with Aristarchus’ overall philological and historical methodology as it appears in his discussion of other aspects of Homer: for Aristarchus it is not legitimate to attribute to Homer (by means of allegoresis, for instance) outlooks or ideologies that are foreign to the culture in which he lived or chronologically posterior to it, nor is it legitimate to attribute to him any knowledge of any kind that has been demonstrated or otherwise discovered in subsequent eras. It is this same kind of rigorous historical-philological analysis that prevents Aristarchus from believing that Homer’s language is simply a mixture of all the dialects of the Greek world; for such an idea posits an implausible linguistic congeries, inexplicable in historical terms and therefore unacceptable. The Alexandrian and Aristarchean view, then, rejected interpretations that resulted in anachronism and chronological incoherence; and yet, as is well known, other critical and ideological trends arose that directly opposed this view and maintained, instead, that Homer was the originator of any of a number of philosophical and religious ideas, and that his text was the source and the

  Dialect Glossai in the Scholia to Homer reservoir of all forms of knowledge. This latter view is well represented among the scholia in the corpus of the so-called scholia exegetica, while it is completely absent from the scholia that derive from the genuinely Aristarchean tradition of the so-called VMK (Viermännerkommentar or Four-Man Commentary: Didymus, Aristonicus, Nicanor, Herodian). The scholia mentioning dialect glossai that I have examined thus far are all scholia exegetica, or else D-Scholia or passages of Eustathius which probably derive from scholia exegetica that have since been lost;23 none of them, in fact, are scholia deriving from VMK. There is, however, another witness to the ancient Homeric scholarship that discusses dialects, which differs from the above passages and must be added to the picture sketched thus far. In his contribution Nünlist 2012, René has highlighted two scholia of Aristonicus which present us with interesting information regarding dialect usage and which it seems worthwhile to examine further here. Ariston. ad 16.856b {ψυχὴ δ’} ἐκ ῥεθέων: ὅτι πάντα τὰ μέλη ῥέθη Ὅμηρος προσαγορεύει. οἱ δὲ Αἰολεῖς μόνον τὸ πρόσωπον. Ariston. ad 17.112a παχνοῦται, ἀέκων δέ τ’ : ὅτι τὴν κατὰ ἀγρὸν ἔπαυλιν μέσσαυλον. οἱ δὲ Ἀττικοὶ τὴν μέσην θύραν τῆς αὐλῆς, τὴν διορίζουσαν τήν τε γυναικωνῖτιν καὶ τὸν ἀνδρῶνα.

Regarding 16.856, the scholion states that in Homer the word ῥέθος can refer to any of the μέλη of the body, whereas for the Aeolians it indicates only the πρόσωπον. On 17.112, the scholion discusses the word μέσαυλον and the different meanings this word possesses respectively in Homer and in Attic usage (sc. chronologically posterior). In these two cases, then, it is clear that the observation regarding dialect usage is not presented in order to help interpret some aspect of the Homeric text, but rather as an additional consideration of a purely linguistic or lexicographic nature — i.e., a purely scholarly comment, with no exegetical function vis-à-vis the text of Homer itself. Observations of this type may fall within the Aristarchean framework of Homeric study (discussed above); that is to say, they may contribute to the thesis that Homer’s language was an archaic form of Ionic spoken long ago in Attica and therefore different both from (e.g.) Aeolic (the first example) and from contemporary Attic (the second example). Such an interpretation is perhaps supported by the fact that the scholia of Aristonicus are generally Aristarchean in methodology and scholarly outlook. Moving forward, it will be important to follow up and expand the research into this

 23 Fenoglio 2009.

Dialect Glossai in the Scholia to Homer  

question, in order to determine whether there are other scholia of a similar kind that may adduce further evidence to support this idea. To sum up our findings, we may conclude by saying that the extant remnants of ancient Homeric criticism discussing dialect are particularly rich; moreover, what we now possess must represent only a fraction of the total that was produced, and of what we possess I have only examined a small portion in these pages. With respect to this phenomenon, in my opinion there are, at least at present, two possible lines of interpretation. The Homeric criticism discussing dialect found in the scholia may represent: 1) An additional component of the non-Aristarchean tradition of interpreting Homer, for which Homer was the source and reservoir of all forms of knowledge. According to this view, one can find everything in Homer, every fact from every place and every time — even as far as language and dialects are concerned. Such a view serves the interests of the interpreter, here as in many other realms of knowledge. 2) Traces of interpretative activity that continue the Aristarchean view. This tradition aimed to identify the base dialect of Homeric language (in part by highlighting specific differences from other forms of Greek), which was thought to be an archaic form of the Ionic dialect spoken in Attica. Interpreters in this tradition held that this base dialect was enriched by the occasional use of various elements deriving from other dialects and recognizable as glossai. If I had to express my own opinion, at the present state of our knowledge, it would be as follows: the first hypothesis above probably applies to all those scholia that have their origin in the scholia exegetica, whereas the second probably applies instead to the scholia that derive from a source with its origins in Aristarchaen and Alexandrian scholarship (i.e., the VMK scholia, such as the two of Aristonicus cited above). I do not think we can exclude the possibility that the two different strands originally played different roles during different moments of activity within ancient Homeric scholarship, but that their eventual products have been mixed together in the composite tradition we now possess, which constitutes a mass of miscellaneous material in which osmosis of content is hardly infrequent. What I can say with certainty is that our research is far from being complete; I shall be content if in these pages I have been able to suggest some useful guidelines and to identify the path that future research should follow.

 Penelope at the Symposium: Od. 1, 328–335 and Dicaearchus Penelope’s first entrance in the Odyssey is one worthy of the queen of Ithaca. In Book 1, while the suitors hold a banquet in the palace of Odysseus, from her rooms on the upper floors Penelope hears the song of the bard Phemius and decides to go down to the hall, accompanied by two maids: τοῦ δ᾿ ὑπερωϊόθεν φρεσὶ σύνθετο θέσπιν ἀοιδὴν κούρη Ἰκαρίοιο, περίφρων Πηνελόπεια· κλίμακα δ᾿ ὑψηλὴν κατεβήσετο οἷο δόμοιο, οὐκ οἴη, ἅμα τῇ γε καὶ ἀμφίπολοι δύ᾿ ἕποντο. ἡ δ᾿ ὅτε δὴ μνηστῆρας ἀφίκετο δῖα γυναικῶν, στῆ ῥα παρὰ σταθμὸν τέγεος πύκα ποιητοῖο, ἄντα παρειάων σχομένη λιπαρὰ κρήδεμνα· ἀμφίπολος δ᾿ ἄρα οἱ κεδνὴ ἑκάτερθε παρέστη.

330

335

From the upper rooms she understood his divine song, the daughter of Icarius, wise Penelope: she went down the high stair from her room, not alone, for, together with her, two maids followed. And she, dazzling among women, when she reached the suitors stood by a pillar of the strongly-built roof holding in front of her cheeks shimmering veils: and a loyal maid stood by her on each side.

The schol. Od. 1, 332 preserves, most probably via an excerpt derived from Porphyry, an intriguing discussion by Dicaearchus on this passage of the Odyssey.1 Schol. Od. α 332 ἡ δ’ ὅτε δὴ μνηστῆρας κτλ.] αἰτιᾶται ἐκ τῶν ἐπῶν τούτων Δικαίαρχος τὴν παρ’ Ὁμήρῳ Πηνελόπην· “ἡ δ᾿ ὅτε δὴ μνηστῆρας ἀφίκετο δῖα γυναικῶν, / στῆ ῥα παρὰ σταθμὸν τέγεος πύκα ποιητοῖο, / ἄντα παρειάων” καὶ τὰ ἑξῆς. οὐδαμῶς γὰρ εὔτακτον εἶναί φησι τὴν Πηνελόπην, πρῶτον μὲν ὅτι πρὸς μεθύοντας αὔτη παραγίνεται νεανίσκους, ἔπειτα ὅτι τῷ κρηδέμνῳ τὰ κάλλιστα μέρη τοῦ προσώπου καλύψασα τοὺς ὀφθαλμοὺς μόνους ἀπολέλοιπε θεωρεῖσθαι. περίεργος γὰρ ἡ τοιαύτη σχηματοποιΐα καὶ προσποίητος, ἥ τε παράστασις τῶν θεραπαινίδων ἑκάτερθεν εἰς τὸ κατ’ ἐξοχὴν φαίνεσθαι καλὴν οὐκ ἀνεπιτήδευτον ἐπιδείκνυσι. φαμὲν οὖν ὅτι τὸ καθόλου ἔθος ἀγνοεῖν ἔοικεν ὁ Δικαίαρχος. σύνηθες

 1 Dicaearchus, fr. 95 Mirhady 2001 (= 92 Wehrli). https://doi.org/10.1515/9783110772371-046

Penelope at the Symposium  

γὰρ παρὰ τοῖς ἀρχαίοις τὰς ἐλευθέρας γυναῖκας εἰς τὰ τῶν ἀνδρῶν εἰσιέναι συμπόσια. μαρτυρία δὲ τούτων κτλ.2 And when she reached the suitors etc.] On the basis of these words, Dicaearchus rebuked the Penelope of Homer: “And she, dazzling among women, when she reached the suitors / stood by a pillar of the strongly-built roof / holding in front of her cheeks” and what follows. For he says that Penelope did not behave well at all, firstly because she has come into the presence of drunken young men, then because with her veil she covers the most beautiful part of her face and leaves only her eyes visible. A behaviour like this is excessive and affected and, further, the presence of the maids on each side of her, in order to appear beautiful by her excellence, shows that all this was done intentionally. We may say that in reality Dicaearchus does not seem to know the customs as a whole. For among the ancients it was the custom for free women to go to banquets with the men. The evidence for this is etc.

The critical observation of Dicaearchus on the behaviour of Penelope is rebutted by the claim that he has not taken account of the specific usages and customs of the period and of the context, and in particular of the fact that among the ancients it was usual for free women to take part in the banquets of the men. The next part of the long scholion adduces as proofs (μαρτυρία δέ ...) the parallels in the Odyssey of Arete and Helen, women who are free and even royal, who are present at banquets, respectively in the palace of Alcinous among the Phaeacians (Od. 7 and 8) and in the palace of Menelaus in Sparta (Od. 4). Then, citing the words of Nausicaa in Od. 6, 286–288, the scholion notes that spending time in male company and gatherings would be inappropriate for unmarried girls but not for married women.3 Penelope has thus done nothing inappropriate by intervening personally during the banquet in order to interrupt the song of Phemius on the troubled returns of the Greeks from Troy, all the more so since the ancients regarded it as proper for a free woman, even the daughter of a king, to be personally engaged in various tasks and activities of everyday life. Consequently, her presence is not out of place, but rather shows wisdom and prudence (sophrosyne), reminding the suitors of the odiousness of their courting her while the memory of her husband is unforgettable to her.4 In the scholion there follow observations on the behaviour of the suitors and then a long inquiry on the kredemnon and on the gesture of covering the face by various Homeric characters, above all when weeping. It concludes with two notes on the  2 P 171, 61–73 Pontani 2007. 3 Nausicaa said, when speaking to Odysseus: “I too would fault another who behaved like that, / one who without the consent of her father and mother met with men before going to her public wedding” (Od. 6, 286–288). 4 P 172, 87–90 Pontani 2007.

  Penelope at the Symposium maids who accompany Penelope: firstly, this too is a normal habit among the ancients and hence not a malicious display; further, they are wise and faithful maids, not two of the treacherous and insolent ones mentioned in Od. 22, 424, which helps cast light on the attitude and behaviour of Penelope herself.5 Thus we have a long excerpt6 that discusses the passage in which Penelope makes her entrance into the poem and, beginning from an observation by Dicaearchus on this appearance on the scene, it comments on it by debating in various ways the “customs of the ancients” (archaioi, palaioi) in women’s social behaviour. The topic is treated in a broadly ethical way on the historical level, based on observation of appropriate and inappropriate social behaviours, but the “documentation,” so to speak, adopted in the discussion is entirely drawn from Homer (Odyssey and Iliad), with the quotation of various passages to support the argument prompted by the lines in Book 1 and in connection with the scene of the entrance of the wife of Odysseus into the poem. Evidently it was a stimulating passage, on which Dicaearchus had offered at least two reflections: Penelope had not behaved well by joining drunken men right in the middle of a symposium; and the gesture of covering with a veil part of her face while leaving her eyes uncovered and the fact of having two maids at her side denote intentionally affected and pretentious behaviour, aiming to make herself stand out. It is natural to think that his attention was drawn by the rather distinctive theatricality of this entrance, which “presents” a primary character such as the wife of Odysseus. Probably the discussion turned on the way the poet had realized this important narrative passage, using it as a peg for disquisitions of an ethical character on usages and customs of the Homeric period. It is not surprising to find this type of theme in Dicaearchus, who devoted important researches to the cultural history of Greece and wrote a Βίος Ἑλλάδος. Our fragment is transmitted without a work-title but, since poetic works, and especially archaic ones, form an important and essential source for cultural and anthropological history, it seems clear that the line between this type of investigation and exegesis of the poetry must have often and intentionally been rather blurred, indeed that it was not a genuine boundary at all.7 However, it seems to me crucial to note how Dicaearchus’ critical observations are met with objections based in a precise and specific way on what is found within the Homeric poems. To put it simply, it is argued against Dicaearchus that the behaviour of Penelope should not be criticized on an ethical level  5 P 173, 22–27 Pontani 2007. 6 It takes up almost three printed pages: 171, 61–173, 27 Pontani 2007. 7 Mirhady 2001 places it in the section “Contests, Literary Criticism” (fr. 95).

Penelope at the Symposium  

because it corresponds to the usages of the time, as is shown well by analysis of similar situations in the Homeric poems. In sum, Dicaearchus is mistaken because he does not take account of how the scene is entirely consistent with that which is specifically Homeric, τὸ Ὁμηρικόν, as was taught by the AlexandrianAristarchean method. The scholion insists on the fact that the observations concern the ancients, introducing a notion of historicity in moral concepts that runs in a similar direction. All the elements present suggest that this may be an observation born in a Peripatetic setting and developed by Dicaearchus, with the discussion continuing around it in the Alexandrian sphere, when the interpreters of Homer attentively applied the method of explaining Homer primarily from Homer himself, deploying the well-honed armoury of a battle-hardened Homeric philology also upon earlier observations that were weak precisely on the basis of their method. Some years ago I took an interest in the case of Il. 2, 409, a line criticized for its content by the Peripatetic Demetrius of Phalerum,8 who regarded it as inappropriate (the issue concerned the rules of invitation to symposia, and so likewise a problem of public ethics). This line subsequently became the object of discussion by the Alexandrian philologists: some of them proposed to delete it, whereas Aristarchus preserved it. I will not repeat the argument set out elsewhere,9 but it is likely that here too a Peripatetic zetema on ethical content found a continuation in a genuinely philological discussion in the more grammatical climes of Alexandria. Comparing the two cases, it is certainly striking that later sources interested in Homeric exegesis have preserved observations by two intellectuals of the early Peripatos (Demetrius of Phalerum and Dicaearchus) on matching themes, not only and not primarily for the topic of usage and customs linked to the symposium, but more generally in relation to the behavioural ethics of the archaic society attested in the Homeric poems. From our point of view, it is even more interesting to note how such considerations were not left aside and passed over thereafter, but clearly stimulated developments and discussions in the sphere of later Homeric exegesis, even that with a thoroughly philological scope. A similar point could be made in another case that I have already treated: the famous scene of the serpent turned to stone in Il. 2, 319, recalled by Odysseus as part of the prophecy of the seer Calchas at the departure of the fleet from Aulis: a serpent with a red back devours a brood of eight sparrow chicks and their mother and is then turned to stone. This portentous appearance was inter 8 Fr. 143 SOD (= 190 Wehrli; omittit Jacoby, FGrHist 228). 9 Montanari 2000b (= this volume, ch. 55); 2001.

  Penelope at the Symposium preted on the basis of the number of victims: nine years of war would pass before in the tenth year the Achaeans would win out over besieged Troy. A twofold, delicate philological problem, already among the ancient philologists, concerned a variant to line 318, connected to whether the following line 319 was present in the text or not, i.e. a problem of athetesis. In this case too I will not repeat the argument made elsewhere.10 An excerpt of Porphyry11 informs us that Aristotle had taken an interest in the interpretation of this passage of the Iliad, carefully discussing various problematic aspects linked to the nature of the prodigy and the meaning of Calchas’ prophecy: this is fr. 145 Rose = 369 Gigon, evidently from the Aporemata Homerica.12 We thus have an interesting case in which a documented Aristotelian aporia focused attention on the difficulties of a passage for which a likewise documented philological discussion by the Alexandrian grammarians shows a typical divergence between Zenodotus and Aristarchus, not only in the choice of reading but also in the preservation or athetesis of an “incriminated” line. To draw conclusions, we can confirm that, even if only in fragmentary form, documentation survives for the fact that Aristotle himself or his students at the Peripatos discussed significant Homeric passages, devoting zetemata to them on very varied topics, and that for these same passages the sources show clearly that the particular questions raised in their comments subsequently became the subject of further discussions and deeper treatments by the Alexandrian grammarians, sometimes with identifiable developments also of a specifically textual character. The analysis of these cases, of the relevant passages and of the topics involved brings to light very clearly how the deep relation between Aristotle and the Peripatetic school, on the one hand, and Hellenistic philology, on the other, was realized namely in the interpretation and study of poetic texts. What has concerned us here is an example, and others too have been highlighted and studied,13 with a view to compiling a broad and careful collection that would redraw the picture of these relations, at least for the revealing elements that have been preserved.

 10 Montanari 2008 (= this volume, ch. 54). 11 Transmitted in the manuscript Ven. Marc. 821 = *B: cf. Erbse 1969–1988, I, p. xvii; text in: Dindorf 1877, 115–117; Schrader 1880, xxx. 12 Hintenlang 1961, 137–141. 13 As well as Montanari 2000b (= this volume, ch. 55); 2001; 2008 (= this volume, ch. 54); cf. also Schironi 2009: synthesis in Montanari 2012 (= this volume, ch. 56).

 Alexandrian Homeric Philology The Form of the Ekdosis and the Variae Lectiones 0. A few years ago I addressed several problems linked to positions modern scholars may have taken up about the way Alexandrian philologists worked on the Homeric text.1 I emphasize that the discussion focused specifically on the history of the Homeric text; its extension to other authors must therefore be carefully analyzed case by case. I believe it is useful to revisit these issues after the intervening debate, if for no other reason than to delineate (as I see it, of course) what can be held as reconstructed with a good level of probability and what still remains controversial. But I also believe greater emphasis needs to be placed on some aspects which in my view have not been given due regard. I will outline the relevant points.2 1) It is crucial to determine the material form of the ekdosis of a grammarian and how it was concretely built up. A number of aspects are connected with what the ekdosis was like in concrete terms: a) a possible explanation of the terminology used in the scholia that have come down to us for the atheteseis and line omissions carried out by Zenodotus; b) how the critical work on the Homeric text by the various philologists was preserved and became known, in particular the pre-Aristarchean criticism — namely, that performed before it became general practice to write extensive hypomnemata in which a number of subjects could be dwelt on in some depth; c) a possible solution to the problem concerning the number, form and succession of Aristarchus’ Homeric ekdoseis and hypomnemata.  English translation by Rachel Costa.  1 Montanari 1998 (= this volume, ch. 23), 2000 (= this volume, ch. 38) and 2002b (= this volume, ch. 58). In the following notes the references are limited: full references can be found in the bibliography and in the cited articles. 2 E.J. Kenney, in his review in CR n.s. 50 (2000) 118–120, of the volume Editing Texts — Texte Edieren (ed. by G.W. Most, Aporemata 2, Göttingen 1998), did not venture into a discussion of my article (Montanari 1998a) and dismisses it in two lines: “Franco Montanari, ‘Zenodotus, Aristarchus and the Ekdosis of Homer’, sketches a speculative maze which had better be left to specialists to thread” (p. 119). The poor man is evidently surprised to find debate on Alexandrian philology in a volume of Editing Texts (instead of being left to ‘specialists’: specialists in what?), and has clearly grasped little or nothing of the problems we deal with (again) above and had difficulty in following the line of reasoning, which I have to confess is actually rather complex. A quite different assessment is given by a scholar such as S.R. Slings, BMCR 1999.05.27. https://doi.org/10.1515/9783110772371-047

  Alexandrian Homeric Philology 2) The real nature of the readings attributed by the erudite tradition to Alexandrian grammarians such as Zenodotus, Aristophanes of Byzantium and Aristarchus: conjectures ope ingenii and based only on subjective criteria, variants deriving from collation of copies and thus the result of a selection, or a mixture of both proceedings. As far as point 1) is concerned, I feel that fairly widespread agreement has been reached, building up a picture I will describe below. On the aspects linked to point 1), namely a, b and c above, I think I have contributed to clarifying certain points and I will try to summarize what I believe can be reconstructed with a fair degree of reliability. Point 2) remains controversial and has recently been the focus of intense debate. Indeed this is the most crucial and central node of the work of the Alexandrian philologists and has a bearing on their role in the intellectual and cultural history of our civilization. The solution is easier if we firmly ground our approach on the following key element: it is a problem of principles and method, not of quantity or quality of the results. 1. Let us begin with the question of the material form of the Alexandrian ekdosis and the way in which it was created. Physically, i.e. from the point of view of the written manuscript, what was the ekdosis of Homer, how did a grammarian proceed and what exactly did he do if he wanted to produce one? That the result of his work was a completely new exemplar, containing his own text, i.e. the entire work re-written according to the way he believed was appropriate and right, is not a widely held view today (among other things, it is biased by the modern concept of critical edition) and to my mind it is implausible. More generally accepted is the view (which I share) that he would work on an already existing copy, carefully chosen from among those he had available, and this copy would then become the base-text for his own alterations and comments: he would thus write only what he deemed necessary, in the margins and in the blank space between the columns and above the lines. In this case, one can easily imagine that the critical signs would be placed beside the lines in this copy: Zenodotus began3 with the obelos for the proposal of athetesis, and then the system was gradually enriched and became notably differentiated. Doubts remain concerning the extent to which it was supplied not only with the semeia and what amounted to veritable differences in reading as compared to the base-text, but also with brief and sporadic explanatory annotations and comments.  3 It is possible that there existed pre-Alexandrian forms of σημείωσις: cf. Cassio 2002, 127–128.

Alexandrian Homeric Philology  

Today it is widely agreed (and I would go as far as to say that it can be considered as established)4 that the Homeric ekdosis of an Alexandrian philologist was a copy chosen as the base-text, which he gradually annotated as he proceeded with his studies and his work of diorthosis. It was therefore essentially a unique exemplar (things could be copied from it, but it could not be reproduced in the same form); in addition, it must have been perfectly identifiable given that it bore the name of the scholar who had ‘personalized’ it with his work. This manner of producing the ekdosis was adopted by Zenodotus and remained general practice among later grammarians. A philologist would choose, according to his own preferences, an exemplar he judged to be suitable as a basis for his work. Whenever (for some reason or other) he did not approve of the text he found in this exemplar, he would jot down his preferred reading in the available space by or above the line involved. His Homeric text resulted from the set represented by the base-text plus the various indications of his changes contained in the ‘surrounding material’. He would set the required semeia by the side of the lines and it is highly likely that he also added concise explanatory and exegetic notes. We must imagine all this as the outcome of years of study, which repeatedly prompted a number of interventions on one and the same exemplar: garnished and disfigured by the work of diorthosis, it was materially the Homeric ekdosis of that philologist, his own personal copy containing the entire fruit of his reflections and investigations, an ekdosis inasmuch as it was precisely ἐκδοϑεῖσα, that is to say available to learned men, pupils, poets and intellectuals. Until the practice came into use of writing commentaries, which allowed greater space, the working exemplar with the semeia, the different readings and the notes in the margins (in addition, naturally, to other separate works such as lexicographic collections and monographic studies, or recollections preserved in the environment and the ‘school’, cf. below) constituted the essential source providing knowledge on the opinions of a given philologist about the Homeric text. Such must have been the case for Zenodotus, Aristophanes and in general the pre-Aristarchean grammarians. With Aristarchus, the ekdosis certainly remained in use, consisting of the working exemplar with its ‘surrounding material’, and it was accompanied by the hypomnema, which became the ideal recipient for developing arguments on a wide range of philological-exegetical themes.

 4 I argued in favour of this view in the works cited in n. 1, where the numerous precedents are also indicated; cf. Haslam 1997; West 2001a and 2002 for Zenodotus.

  Alexandrian Homeric Philology These considerations hold in general for the material form of the Alexandrian ekdosis and the different generations of philologists. Let us now deal more specifically with Zenodotus and Aristarchus. Zenodotus, we have said, worked on an exemplar that was available to him and which he deliberately chose for this specific purpose. However he had more than a few reservations about it, concerning both the numerus versuum and also a certain quantity of readings. He had doubts about the authenticity of some lines and adopted a sign indicating his suggestion that the line should be expunged, the obelos, which he marked in besides the lines: this was the ἀϑετεῖν, the ἀϑέτησις. But it has always been more difficult to determine how he proceeded with lines he believed should most certainly be deleted from the text as definitely spurious. Normally, such lines would have been present in his basetext (as were those for which he proposed the ἀϑέτησις by means of the obelos). For these cases of deletion, in the scholia one finds the expressions οὐ γϱάφειν (the most frequent), οὐκ εἶναι, οὐ φέϱεσϑαι:5 reconstructing exactly and concretely what the difference was between the operations indicated by the different terminology has always been problematic. I think one may gain a good idea of what was meant by noting that the literary papyri bear witness to the use of different modes of deleting something present in a text: a horizontal or slanting stroke could be marked through or above the words or letters to be deleted, or they could be distinguished by dots or by strokes placed above or beneath, or alternatively be closed within a sort of round brackets used in pairs.6 The deletions were usually carried out, when necessary, by the διοϱϑωτής, whose task in the writing workshop was to re-read and correct, often in comparison to the model. The most noble scriptorial practice of manuscript production thus offered Zenodotus various means to perform deletions and corrections, which he could then perfectly well transfer to his own particular διόϱϑωσις, the διόϱϑωσις of the philologist rather than that of the one who corrects a scribe’s work.7 One should not overlook — in fact, it should be highlighted — that the methods and practices of book production advocated operations such as comparison between exemplars (the copy as compared with the model) and other practical resources for correcting, deleting, adding or indicating certain points concerning the literary text. These resources were applied to the techniques adopted by the grammarian:

 5 Cf. Nickau 1977, 1–30; 1972, 29 ff. 6 Turner/Parsons 1987, 15–16, with a reference to examples in the plates; cf. Turner 1977, 93 and Pl. VIII (= 1980, 112–113 and plate VIII). 7 Cf. Ludwich 1883–1884, II 134; Nickau 1977, 10 ff.

Alexandrian Homeric Philology  

this is probably why the procedure did not seem particularly unusual. But it did imply an innovative principle, of which more below. Thus the copy used by Zenodotus presented not only lines marked with the obelos but also lines that were clearly intended to be deleted from the base-text, by one or other of the graphic methods mentioned above. In addition, the grammarian could equally well jot down something like οὐ γϱάφειν or another equivalent term;8 alternatively, he could rely purely on the deletion sign without verbal annotations, in which case the terminology for line deletion may have been noted down by those who followed his teachings at the Museum, or it may have been created, modified or extended by the subsequent tradition (this is conceivable above all for οὐκ εἶναι and οὐ φέϱεσϑαι).9 I would argue this is the most plausible explanation for the much-debated terminology used in the scholia for Zenodotus’ text alterations of athetesis and line omission, including the problem of the material difference in the manuscripts between athetised and deleted lines. To recapitulate, I believe that the copy used as the base-text by Zenodotus, once completed, contained all the alterations intended by the grammarian: the athetesis by means of the obeloi, the deletions by means of the elimination signs, the readings he believed to be correct, at times also concise exegetic remarks. It is difficult to determine how long the exemplar of the ekdosis then remained available and accessible. I think that what has been preserved of the arguments and comments on the Homeric text of the first διοϱϑωτής derives essentially from three sources: the memory of his lessons at the Museum, aided in various ways by notes; a posteriori reconstructions, aided by a brief annotation (if and when this existed) or simply deduced from the text alterations; materials available in other works (such as the Γλῶσσαι and so forth). One can thus readily understand the causes of the well-known uncertainty, already testified in the Aristarchean age, regarding not only Zenodotus’ arguments but at times also the actual readings he intended: if his exemplar was lost, it is quite plausible to imagine partial and imprecise reconstructions, or conflicting information, a tradition that was not fully reliable; but even if his exemplar remained extant for a fairly long period of time, it is not implausible to imagine that some points

 8 The verbs πεϱιγϱάφω and διαγϱάφω are technical terms for ‘deleting’ using the material means mentioned: cf. Turner and Parsons, p. 16; some examples have remained in the scholia: for πεϱιγϱάφω cf. Nickau 1977, 10–12 and 29. 9 In the articles cited in n. 1 I discussed some possible objections to this reconstruction, which I will not repeat here.

  Alexandrian Homeric Philology did not appear completely clear, that certain annotations were not easy to interpret, or that discordant traditional information was sometimes in circulation.10 Martin West has recently put forward an idea that I feel is worth examining in some depth.11 The peculiarity of Zenodotus’ text (West argues) cannot be entirely attributed to eccentricity in his judgments and opinions: it must at least to some extent reflect a peculiarity of the tradition on which he depended and based his work. The claim is that he worked on a rhapsodic exemplar produced in an Ionic context probably in the 4th century. This exemplar thus reflected a different traditional line as compared to that which subsequently became more widespread, which was predominantly of Attic origin. It is quite plausible that the Zenodotus exemplar came from Ephesus, his native city, and that he brought it with him to Alexandria. In other words, according to this hypothesis Zenodotus worked on a Homeric text characterized by idiosyncratic elements that are reflected in the tendencies of the text choices attributed to him: thus his Homeric text, resulting from the combination of the base-text plus the alterations jotted down in the margins, was therefore necessarily directly influenced by these features. Such a view is supported both by the (widely shared) conception of this edition as ‘an old exemplar’ with alterations and notes, and also by the Ionic tendency of the Zenodotean text: further, this line of reasoning can be supplemented with the idea that Zenodotus worked on a rhapsodic exemplar stemming from the 4th century Ionic environment and perhaps originally deriving from Ephesus. This is a very interesting and plausible hypothesis, which has a number of undeniable advantages and provides a better explanation of certain issues. Not only does it provide a historically more concrete motivation for some characteristics of the Zenodotean text and “the erratic character that it evidently did have”,12 but it also provides additional support for the idea of the ekdosis as delineated above. This idea will also prove seminal, as we will see, for discussion of the problem of the origin of the readings. The external and material form of the Homeric ekdosis of Aristophanes of Byzantium and Aristarchus must have been very similar to that of Zenodotus. But Zenodotus’ choice of base-text seemed highly debatable and was open to criticism, which is why Aristophanes and Aristarchus chose exemplars with  10 On these much-debated problems cf. Pfeiffer 1968, 108 ff. (= 1973, 185 ff.); Nickau 1972, 28– 31 and Nickau 1977, chap. I (he believes that Aristarchus no longer had Zenodotus’ original and is also sceptical about the existence of copies of the latter); Montanari 1988, 83 ff. and 98–105 (in particular on the work of Ptolemaios Epithetes); Rengakos 1993, 12 ff.; van Thiel 1997, 20; West 2001a, chap. I.2–3. 11 West 1998–2000, vol. I, p. VII; West 2001a, 33–45 and 2002; approved by Janko 2002. 12 West 2001a, 39.

Alexandrian Homeric Philology  

noticeably different characteristics.13 Consequently, a line of tradition predominantly of Attic origin gradually spread, partly by virtue of the base-text of working copies used by grammarians who worked at a later period than Zenodotus. The latter base-text proved decisive above all as regards the numerus versuum, whereas the readings suggested by the individual grammarians generally did not become standard in the vulgata. The plus-verses present in the Zenodotean text were not his own interpolations but were instead typical of exemplars that were current in his day:14 they disappeared because the work of Aristarchus led to general recognition of a text that had a very similar number of lines to our vulgata. It is significant that Aristophanes did not go as far as to carry out the drastic act of line deletion: in other words (following the hypothesized reconstruction), it is significant that he abandoned the use of performing material deletions on his own copy by resorting to the graphic techniques mentioned above. The obelos became the prime tool for expressing a cautious doubt on parts of the text. The οὐ γϱάφειν disappeared, leaving only the ἀϑετεῖν.15 Aristarchus followed exactly the same procedure. This explains why many of the lines Zenodotus had decided to eliminate from the Homeric text once and for all — but which were present in the copies chosen by later grammarians — were preserved in the numerus versuum that became the generally accepted tradition after the Aristarchean age and thus remained in our vulgata.16 The abandonment of the drastic practice of material deletion highlights the increasing sense of caution that had developed in the meantime, and accounts for the fact that many of the lines ‘deleted’ by Zenodotus were in effect no longer deleted17 and were thus not obliterated from the tradition. The work of Aristarchus marked the beginning of the period in which Alexandrian philological production included the drafting of extensive hypomnemata. This greatly facilitated and enriched the communication and preservation of the arguments and motivations put forward by the grammarians, so that the material which has come down to us from this tradition is much more substantial.  13 Most recently, West 2001a, 36: “Clearly Aristophanes and Aristarchus were not dependent on Zenodotus’ text but followed another source or sources much more similar to the vulgate”; cf. West 2002, 138. 14 West 2001a, 40. 15 Or else, if genuine deletions were still carried out, they were of such minor relevance that all knowledge of them was lost. 16 Haslam 1997, 85; West 1998–2000, vol. I, p. VII; Führer/Schmidt 2001, 7. 17 It is often stated, instead, that they were ‘recovered’ or ‘reintroduced’: this would have involved far more complicated operations.

  Alexandrian Homeric Philology Yet the ekdosis as an annotated working copy by no means went out of use, as clearly testified by the information on the Aristarchean editions. However, the possibility of dwelling at length in the hypomnema on arguments pertaining to text criticism and exegesis constituted an important resource. In practice, the need to write on the copy chosen as the base-text was no longer so strongly felt, especially as regards philological-exegetic arguments. Previously, before the rise of separate hypomnemata, there had been a greater need to write on the working exemplar,18 but with Aristarchus the critical signs placed next to the lines (and this system, enriched quite early, soon after Zenodotus, became particularly elaborate)19 as well as the variants and the readings to be adopted must have been present in the margins and interlinear spaces, while the philologicalexegetic treatment was mostly developed in the commentary. Therefore, to recapitulate my view and render it more explicit,20 starting from Aristarchus there was an increase in the presence of critical signs on the working base-text and a parallel decrease in use of brief explanatory notes because there was space for that in the commentary. What surely remained were the variants to be chosen, which were discussed more extensively in the hypomnema (where, presumably, the readings were also presented). The number of Aristarchus’ ekdoseis and their philological and chronological relation to the commentary or commentaries is still a subject of dispute. I will not go over the entire background here, nor report the treatment already given elsewhere: I will take these for granted and will restrict myself to summarizing the results, in order to set them within the framework that is being delineated.21 On the one hand we have the frequent inequivocable references to the plural for Aristarchus’ editorial work on Homer: αἱ Ἀϱιστάϱχου (scil. ἐκδόσεις or διοϱϑώσεις), ἡ ἑτέϱα τῶν Ἀϱιστάϱχου and similar. On the other, the titles of two works by the grammarian Ammonius, the direct successor of Aristarchus: Πεϱὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀϱισταϱχείου διοϱϑώσεως and Πεϱὶ τῆς ἐπεκδοϑείσης διοϱϑώσεως, the former in apparent contradiction with the latter and with the scholiastic citations that indicate two editions. Nevertheless, the solutions proposed go as far as to hypothesize a ἐπεκδοϑεῖσα διόϱϑωσις carried

 18 It is therefore plausible that Zenodotus wrote relatively more on his exemplar than did Aristarchus on his own copy. 19 They could also be repeated in the hypomnema beside the lemmas, as was the case for instance in P. Oxy. 1086 (pap. II Erbse). 20 Cf. Nagy 2000, n. 9; Führer/Schmidt 2001, 7 n. 32. 21 Montanari 1998a (= this volume, ch. 23) and 2000a (= this volume, ch. 38); cf. West 2001a, 61–67.

Alexandrian Homeric Philology  

out by Aristarchus’ immediate pupils, probably by his successor Ammonius, who in any case was familiar with it. As regards the hypomnemata, I think it is difficult to deny that Aristarchus made two successive versions: a first commentary based on the ekdosis by Aristophanes of Byzantium is explicitly cited in sch. B 133 a: ἐν τοῖς κατ᾿ Ἀϱιστοφάνην ὑπομνήμασιν Ἀϱιστάϱχου. In contrast to this stands the citation of perfected (ἠκϱιβωμένα) hypomnemata in sch. B 111 b: it is perfectly plausible to assume that Aristarchus produced a second version of the hypomnemata in which he took into account the progress achieved over time by his work as a Homeric philologist.22 We have two Homeric passages on which Aristarchus is known to have changed his mind in comparison to his first text choice, as reported in the scholia to K 397–399 and T 365–368. Such a situation has many parallels in the scholiographic documentation, describing Aristarchean second thoughts and changes of heart.23 However, in the case of these two Homeric passages, it can confidently be stated (despite uncertainties about details) that later philologists were searching for information on the reasons and circumstances for his change of mind and on the text situation that had ensued: since they were far from certain, they consulted the cited works by Ammonius. In the attempt to explain this situation and reconcile the apparent contradiction between the two titles of the successor in the school, it has been suggested that Aristarchus himself personally composed only one ekdosis, as suggested by the first title, and that the ἐπεκδοϑεῖσα διόϱϑωσις mentioned in the second one was actually composed later, after the master’s death, in the circle of his first and most senior pupils, possibly even by Ammonius or else by other followers. However, I believe that a slightly different hypothesis can lead to a better understanding of what really happened. The paradox that there were no more editions of the Aristarchean διόϱϑωσις and that this work itself was issued again (ἐπεκδοϑεῖσα) can only be resolved, in my view, by supposing that the ἐκδοϑεῖσα διόϱϑωσις and the ἐπεκδοϑεῖσα διόϱϑωσις were in a sense truly (physically) the same thing, but modified, revised, corrected and further worked on. So let us briefly trace out the entire story. Aristarchus began by writing hypomnemata based on Aristophanes’ text; then he devoted himself to his διόϱϑωσις and produced his own ekdosis; at this point he then wrote new hypomnemata, based on his own ekdosis, presenting what he  22 Allow me to take this opportunity to refer to Montanari 1997, 285–286 (= this volume, ch. 42), on the question of the famous Ὅμηϱον ἐξ Ὁμήϱου σαφηνίζειν. 23 Another two examples are examined in Montanari 2000a (= this volume, ch. 38).

  Alexandrian Homeric Philology felt was a more careful and refined analysis. But the research and reflection of a philologist like Aristarchus on a text like Homer were unlikely to come to a definitive conclusion, for not only did Aristarchus continue to meditate and study, but he also continued to teach, and Homer was constantly in his hands. And so he would resume work on his διόϱϑωσις, pick up his ekdosis again and jot down his second thoughts and new ideas on the text: thus the very copy, that had been the ἐκδοϑεῖσα διόϱϑωσις, then became the ἐπεκδοϑεῖσα διόϱϑωσις. And since he wrote no further new hypomnemata, he could not write at length on the reasons for changing his mind and explain what stage his thought had reached, and so on some points there is considerable uncertainty as to the final stage of his thought.24 Thus the assumption that there existed a single exemplar of Homer, bearing successive layers of alterations and jottings, provides a perfect explanation for the fact that the tradition commonly spoke of αἱ Ἀϱιστάϱχου, ἡ ἑτέϱα τῶν Ἀϱιστάϱχου, διχῶς Ἀϱίσταϱχος (cfr. particularly Didymus in sch. B 517 a) and so forth. In a certain sense, one could indeed speak of two editions, i.e. one and the same exemplar displayed a dual set-up: in the majority of cases the first and second version must both have been legible and distinguishable, thereby permitting the subsequent philological tradition to be familiar with and discuss them both. But if new alterations were made on an already densely annotated copy, it is hardly surprising that uncertainties may have arisen, where for some reason (at times possibly due to material conditions) it may not have been clear which of the two types of text set-up represented the master’s final position, which was his ultimate decision and whether indeed he had actually come to any definitive conclusion. Ammonius certainly had perfectly valid reasons for explaining that there was effectively one and only one ekdosis, that at a certain point it had been ‘re-issued’ with changes: in fact it is quite likely that the titles Πεϱὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀϱισταϱχείου διοϱϑώσεως and Πεϱὶ τῆς ἐπεκδοϑείσης διοϱϑώσεως referred either to two parts of one and the same work or to two very closely linked works, which were designed to explain what had really happened and above all to clarify the doubtful points. In some passages the master’s final decision was not clear: because it had not been annotated properly, or because there was physical damage at that particular point, or perhaps because his reflections had been interrupted by unforeseen

 24 This solution of the problem ekdosis / ekdoseis of Aristarchus is approved by S.R. Slings, loc. cit. in n. 2. I am unsure whether it was perceived by West 2001a, 61–67.

Alexandrian Homeric Philology  

circumstances. In any case, Didymus evidently felt that not everything was perfectly clear, because he consulted Ammonius on doubtful points.25 2. We will now turn to the problem of the true nature of the readings attributed by the erudite tradition to the Alexandrian grammarians such as Zenodotus, Aristophanes of Byzantium and Aristarchus: were they subjective conjectures, ope ingenii and based only on arbitrary criteria, or were they variants deriving from a documentary source, i.e. from the collation of copies and thus resulting from a deliberate choice? Was there a practice of comparing a variety of exemplars of the Homeric text to spot the differences? This is one of the most disputed issues in ancient philology, and it is indeed of crucial importance, not only for the history of the Homeric text in antiquity, but also for an understanding and assessment of the role played by the philologists of the Hellenistic age in our cultural and intellectual history. My own position, already expressed and argued elsewhere, can be summarized in the conviction that the Alexandrian philologists’ production of the ekdosis of a literary work involved both conjectural emendations on the text that had come down to them, and also choice among text variants they had discovered through the collation of different copies. This overall work on the text was referred to by the term διόϱϑωσις, and this was the procedure to which they adhered from Zenodotus onward. Therefore I do not feel it is appropriate to use absolute and exclusive categories — that is, either to assert that all the readings espoused by the Alexandrian grammarians were merely arbitrary conjectures devoid of a documentary source, or alternatively to claim that they were all readings deriving from exemplars that had been collated and thus resulted from a choice among variants. I think it is far more likely that their work involved both conjectural criticism and also choice among variants deriving from collation.26 Naturally, with the evidence available to us today it will be difficult to distinguish case by case whether a reading represented a personal and subjective  25 Pfeiffer 1968, 217 (= 1973, 339): “Whether Didymus was able to work on copies of these original διοϱϑώσεις and ὑπομνήματα of Aristarchus and of his monographs, the συγγϱάμματα, is an insoluble problem”; cf. West 2001a, 61–67. 26 That the philological work of the Alexandrians involved both arbitrary conjecture and the use and collation of copies is an opinion espoused by Pfeiffer, and it had already clearly been expressed with regard to Zenodotus, cf. 1968, 110 (= 1973, 188–189): “It is not improbable that Zenodotus, examining manuscripts in the library, selected one text of Homer, which seemed to him to be superior to any other one, as his main guide; its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Διόϱϑωσις can be the term for either kind of correction. It is hard to imagine any other way.”

  Alexandrian Homeric Philology conjecture or whether it rested on a documentary source, unless we resort to hypotheses and deductions that may not always be reliable. But this is our own problem in interpreting individual cases: it is not a problem regarding the modus operandi and the method of the Alexandrian philologists. The two levels must not be superimposed, and the fact that we lack definitive criteria to consistently distinguish what is a conjecture and what is a variant by no means implies that one of the two categories is in abeyance. Such an approach to work on the text (which does not mean, and does not involve, an anachronistic assimilation of the ancient grammarian to the modern philologist) was extensively applied by Zenodotus. His ekdosis (in the meaning defined above) was its first systematic result, and it was materially visible and consultable in erudite Alexandrian circles. It was a decisive stage in the history of the Homeric text and an intellectually important acquisition. Surely it is inconceivable that, once this method had been inaugurated, it was not further pursued, worked out in greater depth and expanded to include further applications with the progress of post-Zenodotean philology. The idea that the Alexandrian philologists offered only arbitrary conjectures, without any documentary basis and without comparison among copies, has had a number of supporters, starting above all from the positions of M. van der Valk, whose line of interpretation was also adopted (naturally with individually differentiated stances) by H. van Thiel and most recently by M.L. West, to whom we will return later.27 Arguments against this tendency, which leads recta via to a (quite unfair) underestimation of the value of the work performed by the Alexandrians, have recently been put forward by M. Haslam, M. Schmidt, G. Nagy, J.-F. Nardelli and A. Rengakos.28 Rengakos has recently dwelt on this problem again, presenting very precise and cogent arguments concerning the testimony offered by the poets of early Hellenism, i.e. of the Zenodotean age. First and foremost, there is no doubt that these poets revealed knowledge of different pre-existing Homeric readings. But one can go further: “Do Hellenistic poets offer cases which prove beyond doubt  27 Discussion of the positions of H. van Thiel in Montanari 1998a, 4–6 (= this volume, ch. 23). Sharp criticism of van der Valk’s ideas has been made in a number of papers: for ex. Rengakos 1993, 38–48; 2002, 146–148; Haslam 1997, 70 n. 31: “… he does not concern himself with the transmission. In categorizing readings he operates with an opposition between ‘original, old readings’ and ‘only subjective conjectures’ … a schematization that is surely too simple to cope successfully with the complex vicissitudes of the Homeric text.” 28 Haslam 1997; Schmidt 1997; Führer/Schmidt 2001, 6–7; Nagy 2000 (and elsewhere) and Nardelli 2001a (partic. pp. 52–70) in direct opposition to West’s theories; Rengakos 2002 (cf. also Rengakos 2001).

Alexandrian Homeric Philology  

that they made use of different Homeric manuscripts? In other words, do their works display Bindefehler which point to the older Homeric tradition? The answer is clearly ‘yes’.”29 I believe the cases discussed in his article support this ‘yes’: indeed we may confidently maintain that some of the Homeric variants testified in the lines of the philologist-poets of the Zenodotean age derived from the consultation of various manuscripts. A very clear-cut stance has been assumed recently by M. West,30 who claims that Zenodotus and the other Alexandrian philologists known as authors of ekdoseis of Homer, “Aristarchus and his predecessors”, did not carry out any collation of manuscripts or any choice among variants from documentary sources: “In fact the first scholars known to have cited manuscript authority for variant readings are Aristarchus’ contemporaries Callistratus and Crates. Didymus is the first known to have compiled anything in the nature of a ‘critical apparatus’. It is entirely unjustified to project his methods back onto Aristarchus or Zenodotus, or to assume that all the various copies available to Didymus in the time of Augustus were already part of the library’s holdings in the early third century.”31 But it is simply begging the question to claim that Didymus’s method is projected back onto Aristarchus or Zenodotus, for in actual fact there is no evidence Didymus was the first to apply this method rather than having inherited it from his predecessors, unless it be simply the sheer conviction that Zenodotus, Aristophanes and Aristarchus did not apply it at all.32 In West’s view, Zenodotus’ readings are held to be either conjectural emendations or peculiarities of his base-text, but certainly do not result from any form of comparison among copies. Zenodotus is claimed to have lacked the concept of variants, his only critical concern being the existence and identification of spurious lines: “The one kind of textual criticism we know Zenodotus practised was not concerned with choices between variant readings but with the identification of spurious lines and passages. The one feature of his text that marked it out as a critical διόϱϑωσις was the presence of obeloi in the margins

 29 Rengakos 2002, 149. 30 West 2001a, 33–85 and 2002. 31 West 2001a, 36; 2002, 140. 32 West’s Didymean hypothesis is rejected by Nagy 2000, Nardelli 2001a (pp. 61–64), Janko 2002. The evidence of Didymus in sch. Il. 9.222 b is rightly underlined by Nagy and Janko, and cannot be dismissed out of hand as does West (2001a, 37 n. 19): in that scholium Didymus does indeed state that Aristarchus found that reading in some exemplars.

  Alexandrian Homeric Philology (and perhaps brackets enclosing longer passages) to signal the critic’s suspicion that certain verses were un-Homeric.”33 But this, obviously, means that with Zenodotus the problem had arisen of how to discriminate the authentic from the spurious: for whereas an entire line held to be spurious can be eliminated from the text (when the meaning and syntax make this feasible), a part of the line (a word or an expression) cannot be removed without replacing it with something else. Once a critical approach towards the way in which the text presents itself has been acquired, it is inconceivable for there to be a theoretical and essential separation which discriminates between verse athetesis and single word alteration, such that the philologist’s concern focused only on athetesis of whole lines and not on shorter text alterations:34 in either case, the problem at hand for the philologist resides in the opposition between authentic / correct vs. spurious / damaged and in seeking to identify the proper text. Furthermore, by addressing the issue of the authentic text and how to devise the critical-methodological tools to obtain it, Zenodotus achieved a major breakthrough: it was a crucial intellectual step, which he succeeded in taking (prepared by the Aristotelian tradition),35 thereby sparking the progress that characterized the Alexandrian age. This is the first point to be stressed, and it is one that must never be overlooked in the historicalcultural evaluation of the Alexandrian philology phenomenon. The next step forward is to determine which tools were invented and brought into action to fulfil this aim. Conjecture alone, based on subjective criteria, or also the search for documentation by comparing manuscripts and making a choice among variants, i.e. a mixture of both procedures? Thus the question is whether only the first was adopted, or whether the second was used as well. As I stated at the beginning, this is a problem of principle and method, not of quantity or quality of the results. But West warns against a travesty of the situation: “The misapprehension, which goes back at least to the time of Wolf, is that Zenodotus, Aristophanes, and Aristarchus were all editors in the modern sense, who wanted to establish a good text of Homer and who approached the task as a modern editor does, by collecting manuscripts and comparing their

 33 West 2001a, 38; 2002, 140. The idea of the possible use of signs for material and graphic deletion placed on the base-copy was clearly put forward in Montanari 1998, 6; West does not seem to be concerned with the distinction between ἀϑετεῖν and οὐ γϱάφειν in Zenodotus. 34 All the more so since expunction at times involves variants in the part which remains: cf. Montanari 1998a, 7 n. 17 (= this volume, ch. 23), on the subject of B 156–168 (sch. B 156–169). 35 On this problem, cf. Richardson 1992; Montanari 1994a (partic. Richardson 1994); 2000b (= this volume, ch. 55); 2001 (with the bibliography).

Alexandrian Homeric Philology  

readings.”36 Now, if Zenodotus had at least “one kind of textual criticism”, what is likely to have been his aim in carrying out emendations on the Homeric text? Are we thus to believe that Zenodotus had a conscious premeditated idea of ‘modelling’ Homer according to his own taste, i.e. “I’m going to set about reworking Homer and I’m going to make it the way I think it ought to be”? This possibility is by no means easy to accept, but actually this is the only real alternative to the view that “he wanted to establish a good text of Homer”, which is the natural goal of anyone who starts working on and correcting a text. Our debate centers on the methods and criteria we believe we can recognize in the work of the Alexandrian philologists: for in effect their aim cannot but have been “to establish a good text of Homer”, whatever the value of the result according to modern Homerists. I fear that the misunderstandings arise from the fact that there is no clear definition of the guidelines for our judgment on the work of the Alexandrian philologists. By adopting our own point of view concerning the ‘competence’ on which they base their opinions and arguments, so that it can be ascertained whether and when they are right or wrong in comparison with the ‘truth’ according to modern scientific philology, we risk producing unfounded and pointless judgments. Naturally, evaluation of the quality of their choices is the proper perspective for the interpreter (and editor) of Homer. In contrast, maintaining conscious awareness of historical distance and taking care not to overlay our criteria on their behaviour is the proper perspective for the historian of philology as a cultural and intellectual phenomenon and of the ‘reading’ of Homer in ancient civilization. Perhaps it is hard to conceive that Zenodotus’ aim (however incoherent and unsophisticated) was precisely “to establish a good text of Homer” because the testimony that has come down to us indicates that his text was far from good — in fact it was dreadful, and incoherent seen through the filter of the requirements and knowledge of present-day classical philology.37 As even as regards the successors of Zenodotus, even the great Aristarchus, we can hardly claim always to agree with their text choices. The viewpoint from which a Homeric scholar approaches his task is the need to judge whether the text Zenodotus, Aristophanes, or Aristarchus decided on is the one to print in a modern critical edition, whether their interpretations should be espoused as valid in a scientific commentary. The viewpoint from which a historian of ancient philology starts out is that of seeking to understand their methods, arguments, principles, knowledge — in a word, their historical and intellectual position.  36 West 2001a, 34; 2002, 138. 37 Which, of course, always produces excellent and fully coherent texts: or does it?

  Alexandrian Homeric Philology The tendency to scoff at the opinions of the Alexandrian philologists in terms of modern Homeric studies should by no means translate into the tendency to discredit their historical significance, which needs to be correctly positioned and contextualized. It is mistaken to blur the distinction between the two planes. R. Janko seems to render this concept explicit rather more clearly.38 His position on the methods of the Alexandrian philologists is not extremistically unilateral: he believes that the majority of their readings are indeed arbitrary conjectures (due to the Alexandrians themselves or possibly more ancient), but he does not go as far as to deny the recourse to manuscripts and comparison among copies as part of their ekdosis work.39 “On the one hand”, he argues, “there stands the problem of the origin of their proposed text choices (subjective emendation, comparison among copies) and therefore of their working procedures; on the other”, he points out, “my own concern, as a Homerist, has always been whether such readings are authentic”.40 Perfectly clear: modern philologists can to some extent be severe regarding the opinions of the Alexandrians, considering them to be fairly acceptable or fairly unacceptable from their own point of view (Janko is very negative), but they cannot downplay the fact that the ancient Alexandrians emended and compared exemplars to correct the Homeric text, a method that combined interpretation of the text with awareness of the history of the tradition. There is one point on which the question of the method adopted by the Alexandrians directly intersects the Homeric question, that is to say the birth of the current form of the poems. This requires thorough and careful clarification, even at the cost of simplification. Thus on the one hand, the idea shared by  38 Janko 2002; cfr. Montanari 1998b (= this volume, ch. 37). An extremely apt remark, perfectly applicable to Alexandrian philology as well, is offered by Cassio 2002, 132, on the issue of preAlexandrian criticism: “The earliest scholarly approach to the Homeric text is totally foreign to us … we do right to think along very different lines, but we should never forget that it was the commonest approach to the Homeric text in the times of Socrates, Plato, and Aristotle. As a consequence, we ought to be wary of looking at it with a superior smile, and ought to try to understand its motives in more depth instead”. 39 For Zenodotus, Janko 1992, 23: “His caution was salutary, given the abundance of interpolated texts; he certainly had MS authority for some omissions”; for Aristarchus, Janko 1992, 27, and Janko 2002: “This [i.e. sch. Il. 9.222, cf. above n. 32] certainly implies that Aristarchus did check manuscripts for variant readings.” 40 Janko 2002, where he continues by citing his own formulation in Janko 1992, 24: “We can hardly know whether such conjectures are his [i.e. Zenodotus’] own, or derive from the fourthcentury λυτικοί who proposed solutions to problems raised by Zoïlus and other ‘floggers of Homer’; what counts is that they are conjectures, and nearly all bad.”

Alexandrian Homeric Philology  

Janko, Haslam41 and West42 is, fundamentally, that the Homeric text was written down very early, at the actual moment of its composition and probably written down under dictation. This text subsequently underwent alterations of various types, especially by rhapsodes who either memorized it imperfectly or felt fairly free to change certain individual features, or for other reasons. Given this framework, Homeric philology can rely on precise guidelines on the theoretical plane: its aim is to reconstruct and recover that specific text, since the subsequent alterations were in any case conjectural and arbitrary, and had gradually found their way into the various copies, leading to divergences within the manuscript tradition. This was the situation the Alexandrians also found themselves facing, and in West’s interpretation they proceeded to choose a copy considered to be good and then sought to correct it sporadically through conjectures, according to their own criteria, but without bothering to make comparisons with other copies. According to Janko, in addition to choice of the exemplar and their conjectural alterations, sometimes they did make comparisons with other copies and opted for a different text set-up, but in the course of their comparisons they frequently came across rhapsodic alterations and normally felt to determine which was the right text, the original text that was first written down: and so, Janko contends, they made bad choices. Since this is the task of the modern Homeric philologist, the latter must first of all judge which is the authentic reading among those offered by the direct tradition and the erudite materials of the Alexandrian philologists. In opposition to this stands the idea, espoused today above all by G. Nagy,43 of an ‘evolutionary model’ for the genesis of the poems: in this interpretation, the Homeric text was written down far later and there were no written texts of the poems until at least the 6th century. In this view, it is in fact the very uncertainties of the text, as well as its fluctuations testified in the myriad of variants preserved in the various copies up to stabilization, which took place after Aristarchus, that represent the symptoms and the outcome of this prolonged phase of oral transmission. The Alexandrians were confronted with this variety and passed a number of judgments, making decisions on the basis of their ideas: but the truth is that many if not all of the variants that were available to them reflected the instability of a text that was actually still in the oral stage, whereby it was theoretically impossible to choose the one and only correct reading.

 41 Haslam 1997, partic. 80–81. 42 West 2001a and previous writings that can be found cited there. 43 See the works cited in the bibliography.

  Alexandrian Homeric Philology In their different conception of the genesis and writing down of the Homeric poems, these two streams of thought stand in opposition on the issue of how the Homeric variants came into being over historical time, on their meaning in the history of transmission and their evaluation with respect to a possible or impossible original text. This is the crux of the Homeric problem. The problem of the history of philology is: firstly the Alexandrians, starting from Zenodotus, perceived the problem of a non-univocal text characterized by differences among copies, and they evidently felt that such a situation should be corrected in some manner, because it revealed that damage had occurred. What were their next moves, what tools did they devise? They chose a working copy (and this implies they must at least have compared a few, considering them in the light of some approach or other and expressing certain preferences) and then they conducted their work of diorthosis, making conjectures and comparing the copies they had available (among which they had already performed an initial selection, and it would be strange indeed if the copies they had inspected were then determinedly set aside with the expressed intention of never consulting them again). As they gradually proceeded with their work, they noted down the differences in number of lines and in certain words, no doubt venturing some opinion in this regard. If they often did not make the right choice, failing to identify things that had been arbitrarily introduced by rhapsodes or linguistically unacceptable forms, or if they had inadequate awareness of the concepts of historical grammar or of the archaic epic style or of the oral tradition, this in no way affects the fact that they established a method, which was subsequently to develop into ‘le magnifiche sorti e progressive’ of our sophisticated philology. I would like to add another historical observation. Rarely have the Homeric question, that is to say the problem of the genesis and formation of the poems, and the history of Alexandrian philology as a fundamental chapter of the history of the Homeric text, been so intimately and intricately intertwined. Perhaps the only other moment of modern Homeristics in which an equally striking interconnection can be perceived is the moment of the discovery of scholia A and B by Villoison and the influence they exerted on Wolf’s Prolegomena. However, the existence of such very close relations and the enormous interest they arouse must never be allowed to lead to a confusion of the two planes delineated above, for this could result in unfounded judgments. West argues that Zenodotus used as his base-text a rhapsodic exemplar, probably dating from the 4th century, written in an Ionic context (perhaps deriving from Ephesus and brought with him to Alexandria). Such a text therefore contained text characteristics that were decidedly unusual as compared to the more widespread and commonly used paradosis from an Attic background.

Alexandrian Homeric Philology  

This — as I stated earlier — is a good idea, whose usefulness can also be seen in the perspective we are examining. Are we to believe that Zenodotus’ copy was the only existing exemplar of Homer in Alexandria, or is it quite likely that there were others around, with different text characteristics? Should we conceive of a paradoxical Zenodotus who, despite his taste and his concern for the Homeric text, made every effort not to look at other copies he may have come across, not to note the points at which they departed from his own copy and not to ask himself any questions about those differences? It seems to me far more likely that he realized the differences with respect to the Ionic rhapsodic copy he had, both as regards the number of lines and certain individual readings, and that he decided to note them down and express his opinions.44 There is a subtle ambiguity that needs to be eliminated: when speaking of ‘other’ or ‘various’ copies of Homer that were available and utilizable, one should not be misled into thinking of hundreds and hundreds of exemplars, transforming the idea of a comparison into an exaggerated undertaking that immediately becomes totally implausible. Various copies means a few copies, it does not mean hundreds. More specifically: can one speak of comparison among copies only when a certain number is reached, or is it sufficient to compare a few? A further comment by West is germane to this point: “Consider what we know of Aristarchus’ methods, for which we have plenty of material in the scholia. Of course he had the text of his teacher Aristophanes before him. He also kept an eye on [my italics] that of Zenodotus, and took up critical positions against it. But the arguments he used, as reported by Aristonicus and Didymus, were always based on the internal evidence of contextual coherence or general Homeric usage. Not once does he appeal to the authority of manuscripts.”45 So Aristarchus compared his working copy with that of Aristophanes and that of Zenodotus (although “kept an eye on” is insidiously reductive). This implies he made a certain small comparison of copies, but that he took great care not to let his eye stray onto any further copies: according to this picture, he did study and interpret Homer, but he painstakingly avoided consulting any other exemplar than his own, that of Zenodotus and that of Aristophanes, although these alone already presented him with not inconsiderable divergences. Frankly, this seems to me like yet another paradoxical portrayal.

 44 West 1998–2000, vol. I, p. VII, states that the exemplar used by Zenodotus “certe rhapsodi alicuius Ionici opus fuit, non philologi”: this is quite right, and it is on this rhapsodic copy that he carried out his philological work, which was something different and new. 45 West 2001a, 37: the case of Il. 9.222 clearly contradicts this, cf. above n. 32.

  Alexandrian Homeric Philology Earlier, we mentioned the practice of manuscript production, which commonly involved the operation of comparison among exemplars and the execution of corrections. I think that this well-known fact has not been given due consideration when discussing the work of the Alexandrian philologists. And yet we know of various Homeric manuscripts of the Ptolemaic age that reveal this activity. “Several of these early manuscripts give evidence of having been collated with another exemplar”, states M. Haslam,46 who adduces extremely interesting 3rd century B.C. examples. The question that arises quite spontaneously is the following: is it realistic to think that a consolidated practice which was applied in the production of copies, involving comparison among exemplars and corrections, would not have struck the philologist as being a good model for conducting his work? Is it believable that, if the diorthotes of a scriptorium compared copies and corrected them as part of his work as a craftsman, such a procedure was not adopted by this new type of diorthotes who, in his capacity as an erudite scholar and an intellectual, was working on the Homeric text? In conclusion, it is worth pointing out once more that we are dealing with a problem of principles and methods, not of the quantity or quality of the results.47 All that was needed in order for there to be a decisive step forward in intellectual achievement was the very fact of addressing the problem, even if only partially, erratically and incoherently: a literary text had a history of its transmission, during which it could become distorted at various points; the correct text could then be restored by conjecture or by choosing the best reading among those offered by a divergent tradition. The idea of the recognition of damage and of a way to repair it reveals that the organic unity between interpretation and textual criticism has become established. Much progress still remained to be made, Wolfian scientific philology was still in the distant future: no anachronism, but the historical evaluation that a nodal step had been taken from Zenodotus to Aristarchus.

 46 Haslam 1997, 65. 47 Montanari 1998a, 2 (= this volume, ch. 23); Rengakos 2002, 146.



Part III: Other Authors

 Ancient Scholarship on Hesiod 1. It can hardly be doubted that the most ancient forms of exegesis of Hesiod’s text developed in parallel and in association with the Homeric poems. They undoubtedly arose in the context of recitation by rhapsodes and involved explanation of the meanings of individual words or expressions,1 allegorical interpretations of the mythic contents and probably also some elements concerning the person and lives of the poets themselves, many of which may have been quite fanciful. Plato, Ion 531a, rests on the presupposition that the rhapsodes, who clearly had different levels of skills, did not merely recite the poets but also interpreted them: mention is made of Homer, Hesiod and Archilochus. Isocrates, Panath. 18, speaks scornfully of the sophistai who believe they are allknowing, and who recite the lines and wax eloquent on the poets, in particular Homer and Hesiod, offering a potpourri of the finest arguments put forward by others.2 The works of the philosophers and orators abound in quotations from the great poets of the most authoritative traditional paideia, who were adduced and interpreted to buttress a wealth of subjects and were taken as maîtres à penser with whom a rhapsode could engage in dialogue. Among these great poets, Homer enjoyed first place, closely followed by Hesiod. During the period of time extending from the archaic rhapsodes to Aristotle and his first group of pupils, a wide variety of problems concerning the texts of the poets were raised, discussed and handed down over the generations, and the same issues would later be taken up again at least in part and addressed in different perspectives in the Homeric and Hesiodic scholarship of the Hellenistic age, using different methods, different tools, and with different sensibilities. Our documentation is gradually building up more solid evidence of critical activity focusing specifically on the personality and works of Hesiod. We know of a Πεϱὶ τῆς ποιήσεως Ὁμήϱου ϰαὶ Ἡσιόδου by Hecataeus of Abdera (who lived between the 4th and 3rd century BC)3 and of a Πεϱὶ Ὁμήϱου ϰαὶ Ἡσιόδου by Antidorus of Cyme (unknown chronology, possibly 3rd century BC).4 But as

 English translation by Rachel Barritt Costa.  1 Like those represented in the well-known fr. 233 PCG of the Daitales of Aristophanes: cf. Pfeiffer 1968, 15 and 79; Cassio 1977, 21–36 and 75–77; Montanari 2003 (= this volume, ch. 40). 2 Pfeiffer 1968, 3–15; West 1978, 63. 3 Suid. ε 359 = FGrHist 264 T 1; West 1978, 63; Most 2006, T 131. 4 Pfeiffer 1968, 157–158; West 1978, 63; Most 2006, T 133: the evidence on Antidorus is in sch. Dion. Thr., GG I/iii p. 448, 6–7 Hilgard. https://doi.org/10.1515/9783110772371-048

  Ancient Scholarship on Hesiod early as the mid-4th century BC, the sophist Alcidamas certainly used biographical material concerning Homer and Hesiod in his Mouseion,5 which had some form of continuation — through how many and what kind of stages is unknown — in the small work known as Certamen Homeri et Hesiodi. The Certamen has come down to us in a 2nd century AD version bearing (in the only extant manuscript) the rather elaborate title Πεϱὶ Ὁμήϱου ϰαὶ Ἡσιόδου ϰαὶ τοῦ γένους ϰαὶ ἀγῶνος αὐτῶν. However, a fragment with minor differences is found in a papyrus dating from the 3rd century BC (P.Lit.Lond. 191), which already takes us several centuries further back in time, while another papyrus (P. Michigan 2754, from the 2nd–3rd century AD) preserves the final part with the subscription Ἀλϰι]δάμαντος Πεϱὶ Ὁμήϱου: this takes us back to Alcidamas and thus to the 4th century BC.6 The well-known method of making every possible effort to derive biographical data from the works of the poets, which was consolidated and practised in the peripatetic context, found helpful supporting elements in Hesiod, who was rather given to talking about himself. Such an attitude contrasted with the anonymity that characterized the Homeric aoidos (and which by no means thwarted the idea of inventing a “biography of Homer”, with its inevitable variants and alternatives).7 Alcidamas, or whoever created the story, certainly drew on the lines where Hesiod (Works and Days, ll. 650–660) tells of how he once went to Chalcis in Euboea to the funeral of king Amphidamas, in order to take part in the poetic competition held in honor of the deceased, and of how he emerged victorious. Hesiod makes no mention of adversaries in the contest, and the idea that one of his antagonists on that very occasion, or even his specific antagonist, was in fact Homer — who was defeated — may stem from a later date or may be a subsequent addition. On the other hand, some lines attributed to Hesiod (fr. dub. 357 M.–W., from sch. Pind. Nem. 2.1c [III 31,10 Dr.]) allude to an occasion when he himself and Homer supposedly met and sang in Delos, albeit not necessarily in a competition: this is clearly a different occasion from that of the Certamen, but it does bear some similarity, if only in the presumed contemporaneity of the two poets. The episode of the Certamen, which does not appear in the Vitae Homeri that have been preserved, may have been a total invention of Alcidamas himself or it may rest on a more ancient nucleus, formed of stories that circulated on the  5 Fr. 5–7 Avezzù 1982, 38–51, with comm. (84–90) and bibliography. 6 Overview with bibliography in West 1978, 32 and 319 (comm. to 650–652) and West 2003, 297–300. 7 Cf. West 1999; Graziosi 2002.

Ancient Scholarship on Hesiod  

life of the two great poets and which sometimes included the attempt at fabricating an occasion for a competition between them, drawing on some autobiographical lines of Hesiod. These stories were subsequently channelled into a rhetorical and biographical-erudite tradition on the two most ancient and authoritative figures of Greek poetry.8 Yet one can hardly fail to link the problem of the Certamen and its origins to the information concerning ancient debate on the authenticity of the very lines involved. The source is Plutarch, who wrote a lost commentary on the Works and Days in four books. The fragment that concerns us is transmitted via Proclus in sch. Op. 650–662 (pp. 205–206 Pertusi): it informs us that, according to Plutarch, everything about the journey to Chalcis, Amphidamas and the poetic competition was interpolated, ἐμβεβλῆσϑαι:9 πάντα οὖν ταῦτα ληϱώδη. As West comments: “In a manuscript containing scholia of Pertusi’s class c [p. 206, 1–2], the sentence ἀϑετοῦνται δέϰα στίχοι διὰ τὸ νεώτεϱον τῆς ἱστοϱίας sits rudely in the middle of Proclus. It looks like a fragment of sch. vet. If so, the athetesis was probably Alexandrian. The ten lines affected were presumably 651–660”.10 One may imagine an ancient debate on the episode and on the way it unfolded: were there perhaps some who identified Homer as Hesiod’s adversary and some who argued instead that the whole thing had been invented and the lines were not authentic? Had such a suspicion given rise during the Alexandrian age to a proposal of athetesis, which Plutarch endorsed in his commentary? 2. As regards the peripatetic environment, we have a number of significant pieces of information. They can be considered as forming part of the well known interest in the personalities of the poets and their works as developed by Aristotle and his school, where on the one hand the traditions that had been handed down were collected, and on the other efforts were made to infer data from analysis of the works. In addition to the analogous books on Homer, Archilochus, Euripides and Choerilus, Aristotle himself also dedicated one to Ἀποϱήματα Ἡσιόδου11 (but Hesiod is not mentioned in the Poetics). Heraclides  8 On the subject see Most 2006, xi–xxv, and the discussion by Nagy 2009. 9 The verb is used by Plutarch with exactly this meaning in Theseus 20 τοῦτο γὰϱ τὸ ἔπος ἐϰ τῶν Ἡσιόδου (fr. 298 M.–W.) Πεισίστϱατον ἐξελεῖν φησιν Ἡϱέας ὁ Μεγαϱεύς, ὥσπεϱ αὖ πάλιν ἐμβαλεῖν εἰς τὴν Ὁμήϱου νέϰυιαν τὸ Θησέα Πειϱίϑοόν τε, ϑεῶν ἀϱιδείϰετα τέϰνα (scil. Od. 11.631). 10 Cf. West 1978, 319, comm. to 650–652. 11 Cited in the Vita Hesychii of Aristotle, Gigon p. 28: cf. Jacoby 1930, 45; Moraux 1951, 114 ff.; Düring 1957, 87; Schwartz 1960, 610; Pfeiffer 1968, 145 n. 2; West 1978, 63 n. 4; Most 2006, lxvii and T 128.

  Ancient Scholarship on Hesiod of Pontus wrote a treatise in two books Πεϱὶ τῆς Ὁμήϱου ϰαὶ Ἡσιόδου ἡλιϰίας: this we know from Diogenes Laertius (5.87, 5.92), who also maintained that Chamaeleon claimed that Heraclides had plagiarized a work of his with similar content.12 In a Vita Homeri we find confirmation that Heraclides was interested in investigating the chronology of the two poets and contended that Homer was more ancient than Hesiod.13 Furthermore, the famous fragment on the Palinodia relates that Chamaeleon had made a note stating that Stesichorus was critical not only of Homer but also of Hesiod on the question of Helen, thereby revealing that with regard to this question he took into consideration both of the two poets and their works in parallel.14 Another peripatetic, Praxiphanes, claimed to have found a copy of the Works and Days that lacked the proem (ll. 1–10) and which began ex abrupto with the reference to the two Erides (l. 11).15 Ὅτι δὲ τὸ πϱοοίμιόν τινες διέγϱαψαν ὥσπεϱ ἄλλοι τε ϰαὶ Ἀϱίσταϱχος ὀβελίζων τοὺς στίχους ϰαὶ Πϱαξιφάνης, ὁ τοῦ Θεοφϱάστου μαϑητής … οὗτος μέντοι ϰαὶ ἐντυχεῖν φησιν ἀπϱοοιμιάστῳ τῷ βιβλίῳ ϰαὶ ἀϱχομένῳ χωϱὶς τῆς ἐπιϰλήσεως τῶν Μουσῶν, ἐντεῦϑεν (l. 11) οὐϰ ἄϱα μοῦνον ἔην ἐϱίδων γένος· ϰαὶ γάϱ τοῦτο πϱέπον ἦν, ὡς ἔοιϰεν, ἀνδϱὶ γϱάφειν ἄνευ σϰηνῆς ποιητιϰῆς ἐγχειϱοῦντι ϰαὶ πϱὸ ϑυϱῶν ὄγϰον οὐϰ16 ἐπιδειϰνυμένῳ πεϱιττόν. ἐπάγεται δέ τινας ϰαὶ τὸ μὴ ἀπὸ τοῦ

 12 Cf. Pfeiffer 1968, 145 n. 2; West 1978, 63 n. 4; Most 2006, T 129–130, 264–265. Heraclides of Pontus, fr. 22.24, fr. 176 and 177 Wehrli, with comm. 122–123; fr. 1.87, 92 and fr. 105 Schütrumpf; cf. Gottschalk 1980, 2, 4, 136–137. Chamaeleon, fr. 46 Wehrli = 47 Giordano; cf. Corradi 2007; on Chamaeleon and ancient biography, cf. Arrighetti 2007 and 2008, with references to the previous bibliography. New edition of Chamaeleon: A. Martano 2012, was announced (with distribution of a proekdosis) at the Congress “Filosofi della scuola di Aristotele. Cameleonte e Prassifane”, Rome, Istituto Svizzero, 5–7 September 2007, to be published in the forthcoming Proceedings, in the framework of the Theophrastus Project (http://www.ucl.ac.uk/GrandLat/ research/research–projects/theophrastus_extras): in the proekdosis see fr. 16, with comm. pp. 142–143. 13 Vita Romana (VI), 252.34–35 Allen = 52.29–30 Montanari 1979 (Heraclides Pont. fr. 177 Wehrli = fr. 105 Schütrumpf): Ἡϱαϰλείδης μὲν οὖν αὐτὸν (scil. Ὅμηϱον) ἀποδείϰνυσι πϱεσβύτεϱον Ἡσιόδου. 14 P.Oxy. 2506, fr. 26, col. I 1–14: Chamaeleon fr. 29 Wehrli 1969 = 29 Giordano 1990; cf. Montanari 1989 (commented edition of the fragments of Chamaeleon on papyrus), 29 4 T with bibliography. 15 Prolegomena in Opera et Dies Ac, Pertusi 1955, 2.7–20 (cf. Test. et adn. cr., p. 1, the sch. of I. Tzetzes in Op. 1: see below, n. 19) = Praxiphanes fr. 22 a Wehrli = fr. 5 Brink 1946 (Aristarchus fr. 5 Waeschke [1874]); Most 2006, T 9 and 138, 194–195; cf. Jacoby 1930, 45; Aly 1954, 1780; Blum 1977, 109 (= 1991, 52); a new edition of Praxiphanes, edited by E. Matelli, was announced on the occasion cited above in n. 13, with distribution of a proekdosis. 16 Suppl. C.M. Mazzucchi apud Matelli, proekdosis (n. 15), fr. 28.

Ancient Scholarship on Hesiod  

Ἑλιϰῶνος λαβεῖν, Βοιωτὸν ὄντα, τὰς Μούσας, ὥσπεϱ ἐν τῇ Θεογονίᾳ (l. 1) πεποίηϰεν, ἀλλ᾿ ἀπὸ τῆς Πιεϱίας, εἰς τὸ μὴ πϱοσίεσϑαι τὸ πϱοοίμιον. Some have crossed out the proem, as for example Aristarchus among others, who obelizes the verses, and Theophrastus’ student Praxiphanes … This latter says that he encountered a copy without the proem, which lacked the invocation to the Muses and began with So there was not just one birth of Strifes after all. (transl. G. Most) This, it would appear, was befitting to a man who sets out to write without resorting to poetic fiction and who makes no conceited show of superfluous affectation before embarking on the task. A further element inducing some not to accept the proem is the fact that the writer, as a Boeotian, does not assume that the Muses come from Helicon, as he has done in the Theogony, but rather from the Pieria region.

This remark is supported by a passage in Pausanias (9.31.4–5), relating (and I do not think there is any reason to cast doubt on this testimony)17 that the Boeotians living around Helicon had an ancient tradition which held that the Works and Days was the only work composed by Hesiod and that it began without the proem; he adds that they also showed him a copy of the poem engraved on lead and placed by a spring. A certain degree of fluctuation in the proemial parts is credible, given that the hymn to the gods which preceded (the reciting of) a poem in hexameters could easily be felt to be something independent and thus be treated separately (the Homeric Hymns are or derive from a collection of such proems, in the short or also in the more extensive and elaborate forms). To me it seems plausible that this aspect of ancient composition and recitation constituted, so to speak, a sort of ancient traditional basis, reflected in the differences among copies, for the philological problem that arose in the Alexandrian age, concerning the authenticity of the proems.18 The same piece of evidence also informs us that Aristarchus athetized this proem, marking it with the obeloi, which means he was working on a copy that did indeed have the proem. It is difficult to ascertain who may have been the source of the observation on the discrepancy compared to the proem of the Theogony with regard to the provenance of the Muses, but it certainly fits well with the philology of the Alexandrians (and there is nothing to rule out the possibility that it may have been more ancient and subsequently taken up again).  17 This was wrongly done by Wehrli 1969a, 114, in contrast to Preller 1842, 24–25, and Aly 1954, 1780; cf. Most 2006, T 42, 188–189; cf. also Matelli, proekdosis (above n. 16), fr. 28, comm. 152–153. 18 On the ancient athetesis of the Hesiodic proems and the related problems, cf. Leo 1894, 14– 16 = 354–357; West 1966, 150; Pfeiffer 1968, 220, 241; West 1978, 136–137; Broggiato 2001, xxiii and Crates F 78, with comm. 239; Pucci 2007, 23–24; Cingano 2009, 102 f. In general on the hymn-proems, cf. also Janko 1981; Lamberton 1988, 44–48; Càssola 1975, lix–lxi.

  Ancient Scholarship on Hesiod Moreover, as far as we know, Aristarchus only athetized the proem of the Works and Days, unlike Crates of Mallos (as we will see shortly), who expunged both of them: and the above evidence seems to imply that the proem of the Theogony was considered authentic because the Muses are located in the author’s homeland. The terminology clearly bears a technical character, which is extremely suggestive of a connection with the Alexandrian grammarians: both διαγϱάφω and ὀβελίζω19 express very specific concepts and tools, respectively “delete”20 and “mark with the obelos to indicate athetesis”, and this means that what we have here is a perfect case of “translation” of a problem springing from a concrete difference among circulating copies into a language, vision and instrumental apparatus of a philological-grammatical nature. In the past few years, the widely debated problem of the documentary foundation of the activity of the Alexandrian philologists, i.e. whether they worked on the basis of comparison among exemplars or by pure autoschediastic conjecture, has been enriched with the reactions to the opinions put forward by M.L. West,21 who follows a line that tends to discredit Alexandrian philology. Originally attributable to M. van der Valk, this line was taken up again by H. van Thiel and R. Janko, but has been strongly criticized by M. Schmidt, R. Führer, J.-F. Nardelli, A. Rengakos and G. Nagy, as well as by the present writer.22 The case of the proem of the Works and Days takes on remarkable significance in this regard, for we have evidence both that there existed copies without the proem (as also described by Pausanias) and, in addition, that grammarians such as Aristarchus and Crates of Mallos athetized the proem itself while working on copies which did have it (Aristarchus marked it with the obeloi). One could hardly hope for a better example of a philological problem of editorial choice, which is paralleled by an ascertained divergence among different exemplars of the work.  19 In the sch. of Tzetzes, cit. supra n. 16, a further simplification can be read: ἰστέον ὅτι Ἀϱίσταϱχος ϰαὶ ἕτεϱοι ὀβελίζουσι τὸ πϱοοίμιον ϰαὶ Πϱαξιφάνης ϰτλ. 20 The verbs πεϱιγϱάφω and διαγϱάφω are technical terms for materially deleting matter from a written exemplar: cf. Turner/Parsons 1987, 16; some examples have remained in the scholia: for πεϱιγϱάφω cf. Nickau 1977, 10–12 and 29; Dickey 2007, 174 and 254. On the use of deletion as well as athetesis in philological practice and bookmaking, cf. Montanari 1998a (= this volume, ch. 23); 2002 (= this volume, ch. 21); and 2009 (= this volume, ch. 24). 21 West 1998–2000; West 2001a-b (and 2002); extensive critical reviews of his edition of the Iliad have been written by Nagy 2000 and Nardelli 2001, with subsequent responses by West 2001b; cf. also Janko 2000. 22 Cf. van Thiel 1992 and 1997; Schmidt 1997 and 2002; Führer/Schmidt 2001; of the long work by Nardelli 2001–2002, see above all I 52–70; Montanari 1998a; 2002; and 2009a (= this volume, ch. 7); an excellent overview, which shows how the evidence on the practice of collating is far greater than is usually suggested, can be found in Rengakos 2002, with all the references.

Ancient Scholarship on Hesiod  

Of the four known fragments of Crates concerning Hesiod,23 one speaks specifically of the proems of the Works and Days and the Theogony, and states that Crates expunged both of them.24 The citation can be found in an anonymous Life of Dionysios Periegetes,25 in a passage which asserts that the proem must be linked to the context that follows it and must be suited to the specific content; it should not be so generic as to be adaptable to any work indiscriminately. Homer presents a proem (here the reference is evidently both to the Iliad and the Odyssey) whose content is drawn from the actual poem itself and hints at its essence, while the proems of the Works and Days and the Theogony of Hesiod could easily be set at the opening of various different poems. Therefore Crates was right to athetize them (τὸ δὲ τῶν Ἔϱγων ϰαὶ Ἡμεϱῶν Ἡσιόδου ϰαὶ τῆς Θεογονίας πάσης ἔστι πϱοτάξαι ποιήσεως· διὸ ϰαὶ ὁ Κϱάτης αὐτὰ ϰατὰ λόγον ἠϑέτει, “that [sc. the proem] of Hesiod’s Works and Days and of the Theogony is a prelude for his poetry as a whole; hence Crates too athetized them, reasonably”). Going by what the source says, then, the reason for this athetesis was precisely the fact that the Hesiodic proems, in contrast to those of Homer, did not seem to be indissolubly linked to the work itself; rather, their content would have enabled them to have an independent existence as compositions, or at least to be or not to be present without impairing the whole poem. This is an important observation, and it needs to be considered very carefully. One possible objection to this point might be that the last line of the proem (l. 10) of the Works and Days contains a precise reference to the “truth” the poet is about to proclaim to Perses, the brother whom Hesiod then explicitly addresses several times in connection with the feud over the inheritance and also with the modes of behavior giving rise to the reflections on ethical issues expressed in the poem. However, the fact remains that only this line is linked to the rest of the poem, whereas the previous lines (ll. 1–9) constitute a hymn to Zeus that can effectively be read independently, and the two components can be kept clearly distinct (l. 10 has no cogent syntactic links with the prior and following sentences, and one can thus imagine it may have been devised as a text bridge). Besides, the autobiographical passage inserted into the Hymn to the Muses, which constitutes the proem of the Theogony, likewise has nothing that explicitly connects it to the contents of the sacred story that follows in the poem, just as the “autobiographical” passage of the Homeric Hymn to Apollo (ll. 166–173) has nothing that genuinely  23 F 78–81 Broggiato 2001: cf. xxiii, 78–80, 239–241. 24 F 78 Broggiato 2001. 25 Transmitted without indication of authorship by the codex Chisianus R.IV20: edition in Kassel 1985, 72, 56–60.

  Ancient Scholarship on Hesiod links it to the Iliad or the Odyssey, even though this hymn could easily have preceded a recitation as a “proem” by Homer to a work “of his own”. Thus Crates and Aristarchus, with their philological approach, addressed a problem whose roots dated back to a much earlier time and which had left its imprint on copies still circulating at the time of Pausanias. In this context, it is worth recalling the evidence of the so-called Excerptum de vetere Iliade,26 according to which Crates27 and Nicanor28 knew an ἀϱχαία Ἰλιάς whose proem was: Μούσας ἀείδω ϰαὶ Ἀπόλλωνα ϰλυτότοξον. It strikes me as unlikely that there would have been only one proemial line: it is far more likely that the brief erudite excerptum preserves only the incipit of a proem, the length of which is today indeterminable, consisting of a hymn to the gods of poetry (the Muses) and to the god whose rage sets in motion the Iliad (Apollo).29 Immediately afterwards comes a citation of Aristoxenus as evidence of yet another version of the Iliadic proem: ἔσπετε νῦν μοι Μοῦσαι, Ὀλύμπια δώματ᾿ ἔχουσαι, / ὅππως δὴ μῆνίς τε χόλος ϑ᾿ ἕλε Πηλεΐωνα / Λητοῦς τ᾿ ἀγλαὸν υἱόν· ὁ γὰϱ βασιλῆι χολωϑείς. It thus becomes clear at once that Aristoxenus’ version, which begins with a line identical to Il. 2.484 (beginning of the Catalogue of Ships), in some sense reproduces the Iliadic proem that had become accepted and had been handed down by tradition, but it “condenses” the content into three lines as compared to nine, and of these three lines the first is one of pure invocation, in contrast to the strictly thematic and organic manner in which the Iliadic proem proceeds from its very first word (as had been noted even by the ancients). Therefore this is a different organic proem, specifically linked to the poem of the Iliad, whereas the version given by Crates and Nicanor may well have been a proem of a hymnic character and therefore distinct from the content of the poem itself (unless the only line still extant is simply the remains of another alternative form of organic proem). In other words, as far as the proem of the Iliad is concerned one begins to perceive the traces of a rather unstable tradition, in which the difference between an organic proem and an independent proem can be discerned, as described above. Observation of these characteristics must have led to misgivings in reflections on archaic epic poetry, resulting in explicitly philological debate, as we

 26 Contained at the beginning of the codex Bibl. Naz. Roma gr. 6: latest ed. in Montanari 1979, 56; cf. Erbse 1969–1988, vol. I, 3 app. 27 F 1 Broggiato 2001, with comm. 140–141. 28 Cf. Montana 2009; unfortunately this piece of evidence is generally neglected in the overview of the grammarian. 29 West 1978, 137: “Crates knew of a version which began with a hymn to the Muses and Apollo”.

Ancient Scholarship on Hesiod  

have seen. One could say that the contrast of “authentic vs. spurious” for the epic proems is the translation into philological terms (Aristarchus and Crates) of the difference between “organic proem vs. independent proem” in the poetic practice of composition and recitation (hence the evidence on the existence of exemplars of the Works and Days without proem: Praxiphanes and Pausanias). For Homer, tradition became consolidated around organic proems, which sidelined or at least relegated to the background other proemial forms and allowed no doubts as to their authenticity thanks to their link with the authentic poems.30 For Hesiod, tradition became consolidated around considerably more independent proems, which led to doubts about their authorship, as testified among the scholars of the Hellenistic age. But it should be clear that this issue by no means concerns the real authorial authenticity of the proems in the sense of modern philology: it may very well be the case that an independent proem is actually authentic while an organic one is spurious.31 For Hesiod, the question of the authenticity of the proems did arise and the evidence on Crates’ motivations clarifies the underlying reason: in the consolidated text, the Homeric proems were organically linked to the authentic poems (and only fragments of erudite memory conserved the trace of other proems for an “ancient Iliad”), while the Hesiodic proems had an independent character and may have been quite distinct from the poem they were linked to.32 It seems to me quite natural to think that the consolidation of a proem with “organic” characteristics for the Iliad and the Odyssey represented an important stage (perhaps even the final stage) of the process through which the overall text became fixed, i.e. that it was an aspect of the history of the formation of the poems as they have been preserved and handed down to us. If the organic proem is linked to the fixed and consolidated text, the independent proem, on the other hand, can be adapted and is suggestive of a more fluid situation, which does not involve a necessary connection and allows freedom to choose a hymn

 30 The discussions on the meaning of the Διὸς βουλή in Il. 1.5 perhaps retain a reflex of this problem, at least with regard to the fact that Aristarchus and Aristophanes were supporters of the close interconnection with the content of the Iliad, while others gave an interpretation that referred to the broader context of the sacred story: the concept of explaining Homer with Homer as in the case of the proem accentuates the aspect of its organic cohesion and close connection to the whole work. Cf. Kirk 1985, ad loc.; Latacz 2002, ad loc.; Pagès 2005. 31 “Modern scholars with few exceptions have accepted it as Hesiod’s own preface, and they are certainly right”: West 1978, 137, with regard to the proem of the Works and Days; cf. Arrighetti 1998, 380–381; personally, I am convinced of the Hesiodic authenticity of the two proems and their authorial link with the poems. 32 See bibliogr. above, n. 18.

  Ancient Scholarship on Hesiod to the divinity with a proemial function taken from within the repertory, and then to associate with it one or another narrative passage. Following this line of reasoning, we might go so far as to say that Hesiod, in a sense, reflects a less stabilized situation of his works as compared to the way the Homeric poems have come down to us. It is a fact that a precise cut-off of a limited narrative segment in the great overarching mythic cycle constitutes for the Iliad and the Odyssey — and this is underlined very clearly as early as by Aristotle in the Poetics — a very precise element of identification and individuation, which involves a specific proem for each poem and also binds the proem closely to the poem in question. For Hesiod, it may seem highly surprising that the novelty and above all the unusual personalization of the content of the Works and Days (unthinkable in the framework of epic narrative of the “Homeric” type) did not necessarily involve the presence of an individual and particular organic proem: but it was precisely this feature that must have appeared striking to those among the ancients who concluded that the non organic proem handed down with the text of the poem, and connected to it only by l. 10, must be spurious. 3. We know that in antiquity there was much debate on the authenticity of the works handed down under the name of Hesiod. The Aspis (Shield of Heracles) formed the object of in-depth investigation, and the first part of hypothesis A allowed for considerable insight into this controversy.33 τῆς Ἀσπίδος ἡ ἀϱχὴ ἐν τῷ τετάϱτῳ Καταλόγῳ φέϱεται μέχϱι στίχων ν᾿ ϰαὶ σ᾿ (Hes. fr. 195 M.–W, vd. app.), διὸ ϰαὶ ὑπώπτευϰεν Ἀϱιστοφάνης (fr. 406 Slater) ὡς οὐϰ οὖσαν αὐτὴν Ἡσιόδου, ἀλλ᾿ ἑτέϱου τινὸς τὴν Ὁμηϱιϰὴν ἀσπίδα μιμήσασϑαι πϱοαιϱουμένου. Μεγαϰλείδης34 ὁ Ἀϑηναῖος γνήσιον μὲν οἶδε τὸ ποίημα, ἄλλως δὲ ἐπιτιμᾷ τῷ Ἡσιόδῳ· ἄλογον γάϱ φησι ποιεῖν ὅπλα Ἥφαιστον τοῖς τῆς μητϱὸς ἐχϑϱοῖς. Ἀπολλώνιος δὲ ὁ Ῥόδιος ἐν τῷ τϱίτῳ φησὶν αὐτοῦ εἶναι ἔϰ τε τοῦ χαϱαϰτῆϱος ϰαὶ ἐϰ τοῦ πάλιν τὸν Ἰόλαον ἐν τῷ Καταλόγῳ εὑϱίσϰειν ἡνιοχοῦντα Ἡϱαϰλεῖ (Hes. fr. 230 M.–W.; Apoll. Rh. fr. XXI Michaelis). ϰαὶ Στησίχοϱος (fr. 92 Page PMG, vd. app.) δέ φησιν Ἡσιόδου εἶναι τὸ ποίημα. The beginning of the Shield is transmitted in Book 4 of the Catalogue up to line 56. For this reason, Aristophanes suspected that it did not belong to Hesiod but to someone else who had chosen to imitate the Homeric “Shield”. Megacleides of Athens considered the poem to be genuine but censured Hesiod: for he said it was illogical that Hephaestus should  33 Pp. 86–87 in the OCT edition of F. Solmsen; cf. Russo 19652, 29–31, 36–37, text 67–68; Arrighetti 1998, 479 and 482, text 230–231; Most 2006, T 52. 34 μεγαϰλῆς codd., corr. Schweighäuser: the correction, supported by the comparison with the cited passage of Athenaeus, is normally accepted (but Jacoby [1930] 45, speaks of “Megacles Atheniensis”); the same may also hold for sch. Il. 5.640 (cf. Erbse ad loc.); cf. Arrighetti 1998, 482; Most 2006, T 132 = T 52.

Ancient Scholarship on Hesiod  

make weapons for his mother’s enemies. Apollonius Rhodius says in Book 3 that it is his (i.e. Hesiod’s), because of the style and because he finds Iolaus elsewhere in the Catalogue driving his chariot for Heracles. And Stesichorus says that the poem is Hesiod’s (transl. G. Most).

In its account of the different opinions on the question, this erudite excerptum seems to date back to Stesichorus, but Page suspects that the name of the last sentence is corrupted.35 In effect, the formulation that attributes to Stesichorus an explicit declaration on Hesiodic authorship seems rather implausible, but perhaps the question should be considered from a different perspective.36 Athenaeus [12.512e–f (= 12.5–6 Kaibel)] tells us that the peripatetic Megaclides, who was active in the second half of the 4th century BC (a contemporary of Alcidamas) and known for a Πεϱὶ Ὁμήϱου,37 investigated the figure of Heracles and compared the way he was represented by various different poets. The passage in Athenaeus begins with Μεγαϰλείδης ἐπιτιμᾷ τοῖς μεϑ᾿ Ὅμηϱον ϰαὶ Ἡσίοδον ποιηταῖς ϰτλ. and goes on with considerations about the way the hero is characterized and about his attire, concluding that Stesichorus was the first to describe Heracles with a bow, club and lion skin: ϰαὶ ταῦτα πλάσαι πϱῶτον Στησίχοϱον τὸν Ἱμεϱαῖον. Overall, then, in the hypothesis and in Athenaeus we have two remnants of a work in which Megaclides enquired into the variations of the representation of Heracles by poets and criticizes certain aspects of the authors’ descriptions: according to our hypothesis he found absurd that Hesiod in the Aspis (which he considered authentic) says that Hephaestus constructs weapons for one who is the enemy of his mother Hera, whereas Athenaeus informs us on some considerations referring to the character and attire of Heracles in poets of a later date than Homer and Hesiod and explicitly mentions Stesichorus, albeit without the title of the work. What seems to me most probable is that Stesichorus, in a work featuring Heracles (for instance the Cycnus), made some reference to the Aspis, possibly in order to differentiate himself, but in some sense implying its Hesiodic authorship. The formulation of the hypothesis would thus be of a later date and may have been a form of simplification, suitable to be included in a doxography that sprang from a background of philological discussions on authenticity in the Alexandrian age. Apollonius Rhodius argued in favor of authenticity on the basis of considerations of style and content, while Aristophanes seems rather to invoke elements of the Hesiodic manuscript tradition, i.e. the presence of ll. 1–56 in book IV of  35 Cf. PMG cit., with bibliography. 36 Cf. Russo 1965, 29–31, 36–37. 37 Montanari CPF 1992 (Megaclides 66 1T); Pagani 2006.

  Ancient Scholarship on Hesiod the Catalogue of Women, in order to express a doubt on the authorship of the Shield of Heracles attached to these lines. Aristophanes must have devoted considerable attention to authenticity questions, because we know that he also denied Hesiodic authorship of the Precepts of Chiron. The observations by Apollonius and Aristophanes imply that both of them believed the Catalogue to be authentic, as was also the case for Crates:38 we will return to this point later (§ 5). We have no information on the position of Aristarchus with regard to the Aspis, but we can certainly say that this short poem remained among the texts that were worked on by the grammarians, although this did not dispel the doubts on its authenticity. A Ὑπόμνημα Ἀσπίδος Ἡσιόδου, a title implying that a specific stance had been taken, was written by Epaphroditus (1st century AD).39 The sch. ad Sc. 415 mentions an intervention by a certain Σέλευκος, probably the grammarian Seleucus (called Homericus), who was likewise from the early imperial age:40 this too may presuppose the attribution to Hesiod. The anonymous author of the On the Sublime expresses a doubtful attitude, introducing the citation of Sc. 267 with the words εἴ γε Ἡσιόδου ϰαὶ τὴν Ἀσπίδα ϑετέον, while Pausanias, in the list of Hesiod’s works in 9.31.5, mentions the Xείϱωνος Ὑποϑῆϰαι, but not the Aspis. Finally, in the sch. Dion. Thr., GG I/iii p. 124, 6–7 Hilgard, the small poem is explicitly called pseudoepigraphic: ἑτέϱου γάϱ ἐστιν, ἐπιγϱαφῇ δὲ ϰαὶ ὀνόματι ἐχϱήσατο τῆς Ἡσιόδου, ἵνα τῇ ἀξιοπιστίᾳ τοῦ ποιητοῦ ἀξία ϰϱιϑῇ ἀναγνώσεως. In short, the question dated back at least to the peripatetic Megaclides, and perhaps even earlier, and it remained an object of debate in the sphere of philology of the Hellenistic-Roman age. Yet despite the doubts raised, the Aspis remained solidly included in the Hesiodic corpus, as clearly testified by some papyri of the imperial age and by the later codices.41 Particularly significant is P.Oxy. 4652, a fragment of a parchment codex, which preserves remnants of a

 38 Pfeiffer 1968, 177–178; Martin 2005, 155, 167, 171–174; Cingano 2009, 104–107. The source for the Χείϱωνος Ὑποϑῆϰαι is Quint. 1.1.15 (fr. 407 Slater 1986), cf. Merkelbach/West 1967, 143– 145; Most 2006, T 69 (cf. T 70, and lxii, where pro “Aristarchus” leg. “Aristophanes”); D’Alessio 2005, 232; this work is likewise listed among those which, according to Pausanias (9.31.4–5, cited in § 2 above), the Boeotians of the Helicon rejected, believing that only the Works and Days was authentic. 39 Braswell/Billerbeck 2007, 40 and 319–326, fr. 54–55; Martano 2004a. 40 See Russo 1965, 36 and comm. ad loc.; Müller 1891; cf. Razzetti 2002; see § 5 below. 41 Russo 1965, 37.

Ancient Scholarship on Hesiod  

glossary of the Aspis circulating in the 5th century AD:42 this is unequivocal evidence that it continued to enjoy a favorable reception in late antiquity. 4. We have evidence of a Homeric allegoresis starting from the 6th century BC, and it is natural to think it must have developed in parallel for Hesiod as well, obviously applied to the “inviting” content of the Theogony, although we have specific and precise information in this regard only from the Stoic Zeno of Citium (born around 333 BC). The material has recently been studied in depth by K. Algra and here we can briefly outline his results:43 “Zeno of Citium provided physical interpretations of elements of Greek myth as they can be found in Hesiod’s Theogonia … His purpose was not so much to defend the (consistency of the) old poet, nor to provide a sophisticated and enlightened version of traditional religion, but rather to ‘appropriate’ Hesiodic myth and to use it in support of his own Stoic cosmology and cosmogony” (p. 562). The earliest specific mention of Hesiod’s Theogony in Latin literature can be found in Cicero, Nat. deor. 1.36 = Zeno SVF I 167: Zeno autem … cum vero Hesiodi Theogoniam id est originem deorum interpretatur, tollit omnino usitatas perceptasque cognitiones deorum; neque enim Iovem neque Iunonem neque Vestam neque quemquam qui ita appelletur in deorum habet numero, sed rebus inanimis atque mutis per quandam significationem haec docet tributa nomina. Zeno’s view … again in his interpretation of Hesiod’s Theogony (or Origin of the Gods) he does away with the customary and received ideas of the gods altogether, for he does not reckon either Jupiter, Juno or Vesta as gods, or any being that bears a personal name, but teaches that these names have been assigned allegorically to dumb and lifeless things” (transl. H. Rackham).

“What it indicates, is merely that Zeno interpreted (elements of) the Theogony of Hesiod and that he did so by claiming that, if we pay attention to what the names of the gods actually mean (per quandam significationem), these traditional names can be shown to refer to elements of nature — i.e. what Cicero here for the sake of polemics has labeled res inanimae atque mutae. It does certainly not straightforwardly imply that Zeno presented these allegorical or etymological

 42 Edition by D. Obbink in P.Oxy. LXVIII, 2003, 69–75; cf. Martano 2004b; Stroppa 2008. A. Martano has devoted various studies to the tradition of ancient exegesis on the Aspis: see the bibliography. 43 Algra 2001.

  Ancient Scholarship on Hesiod interpretations in the specific context of ‘ein fortlaufender Kommentar zur Theogonie Hesiods’”.44 Algra analyzes seven fragments in which the name of Zeno is in some way associated, either directly or indirectly, with a passage or a figure of Hesiod’s Theogony, namely SVF I 100, 103, 104, 105, 118, 121, 276 (but it should be borne in mind that the texts present in the SVF collection may be misleading and should be checked against more recent and more reliable editions). The themes addressed are: the Titans and their names (fr. 100), the conception of Chaos and primordial entities (fr. 103, 104, 105),45 the Cyclopes (fr. 118), Helios (fr. 121), the spherical shape of the earth (fr. 276). His enquiry leads to the conclusion that only in a few cases there is an explicit and precise reference to Hesiod’s text (SVF I 104 and 105, to which we will return shortly, also adding SVF I 235, which Algra does not take into consideration, probably because it concerns the Works and Days): “In all other cases some connection with Hesiod’s text can be made, and it is just plausible to assume that Zeno himself indeed made that connection as well. After all in Zeno’s days Hesiod’s Theogony still provided the most authoritative version of mythical cosmogony. Nevertheless … in a number of cases … Zeno’s comments appear to go beyond what is in Hesiod’s text and to focus rather on the traditional world picture, the traditional names and the traditional cosmogony of which Hesiod presents us a poetical version. In this respect it is useful to adduce a general observation made by Anthony Long about the practice of what is usually called ‘allegorical interpretation’ in the Stoic school. Long rightly points out that ‘allegorical interpretation’ strictly speaking involves laying bare the hidden intentions of a particular literary author, in this case Hesiod. But this, apparently, is not why Zeno turned to Hesiod. What he interprets is not Hesiod’s poem — qua creation of Hesiod — but the hidden meaning of the myths, or of the mythical names, he represents”.46 Actually, we should maintain a careful distinction between the idea of allegorical interpretation of the mythic content of a literary work and the allegorical interpretation of the myth as such. This is a delicately problematic point, but the two levels should not be confused, nor, much less, should they be identified, however intertwined or indeed overlaid they may come to be in actual practice, because poetry is certainly the major source of myth. And, as is said above, Hesiod’s Theogony provided the most authoritative version of mythical cosmogony.  44 Algra 2001, 564–565; the sentence cited at the end is from Steinmetz 1986, 21. 45 Zeno believed that the original Chaos was actually water, of which the sediment consolidated in mud was regarded as having given rise to the earth (Gaia). 46 Algra 2001, 576–577; the reference is to Long 1992, cf. Steinmetz 1986, 29.

Ancient Scholarship on Hesiod  

The title Εἰς Ἡσιόδου Θεογονίαν, which can be read in SVF I, p. 71, as a work of Zeno’s, is a pure invention by H. von Arnim, which unfortunately has at times been taken up again and occasionally accepted. The careful examination by K. Algra has now put things back in the right perspective. “But if he did not write such a commentary, where then should we suppose him to have discussed these elements of traditional myth? … The fact that all Zeno’s recorded interpretations of myth dealt with cosmological or cosmogonical phenomena, would rather seem to suggest that they originally figured in one of Zeno’s physical works … But what reason do we have to suppose that the references to Hesiodic myth were made in the same context? Here we may adduce another interesting piece of evidence, to be found in Diogenes Laertius X, 27 … Diogenes, who here professes to report the opinion of Carneades, tells us that both Zeno and Chrysippus, as well as Aristotle, but unlike Epicurus, interspersed their own texts with marturia … Zeno’s philosophical works contained many marturia, references to poets or to traditional or popular views. The preserved etymological explanations of names of mythical divinities all appear to derive from contexts dealing with cosmogony or cosmology. By contrast, the hypothesis that these etymological explanations stem from an original line-by-line commentary on Hesiod’s Theogony, lacks support”.47 I believe that this conclusion can be shared. Furthermore, the utilization and interpretation of passages from authoritative poets belongs to a rich and well established tradition in the sphere of philosophy. It is also worth noting that observations and considerations of relevance for the text itself, including observations of an exegetic nature, could perfectly well find their place in contexts of this type. To our knowledge, Plato and Aristotle represent the first testimony concerning the problem of the lines immediately following the proem, Th. 116–120, which constitute the opening part of the Hesiodic cosmogonic exposition. ἤτοι μὲν πϱώτιστα Χάος γένετ᾿· αὐτὰϱ ἔπειτα Γαῖ᾿ εὐϱύστεϱνος, πάντων ἕδος ἀσφαλὲς αἰεὶ ἀϑανάτων οἳ ἔχουσι ϰάϱη νιφόεντος Ὀλύμπου, Τάϱταϱά τ᾿ ἠεϱόεντα μυχῷ χϑονὸς εὐϱυοδείης, ἠδ᾿ Ἔϱος, ὅς ϰάλλιστος ἐν ἀϑανάτοισι ϑεοῖσι.

116

120

In truth, first of all Chasm came to be, and then broad-breasted Earth, the ever immovable seat of all the immortals who possess snowy Olympus’ peak and murky Tartarus in the depths of the broad-pathed earth, and Eros, who is the most beautiful among the immortal gods (transl. G. Most).

 47 Algra 2001, 578–580.

  Ancient Scholarship on Hesiod In Plato, Symp. 178b, at the beginning of the development of his argument Phaedrus maintains that Eros is worshipped because he is the most ancient among the gods, as proven by the fact that no-one mentions a genealogy in which he appears. Phaedrus then cites Hesiod’s lines on the beginning of the history of the cosmos. ἀλλ᾿ Ἡσίοδος πϱῶτον μὲν Χάος φησὶ γενέσϑαι, αὐτὰϱ ἔπειτα Γαῖ᾿ εὐϱύστεϱνος, πάντων ἕδος ἀσφαλὲς αἰεὶ ἠδ᾿ Ἔϱος … Ἡσιόδῳ δὲ ϰαὶ Ἀϰουσίλεως σύμφησιν μετὰ τὸ Χάος δύο τούτω γενέσϑαι, Γῆν τε ϰαὶ Ἔϱωτα.48 As Hesiod says: First Chaos came, and then broad-bosomed Earth, the everlasting seat of all that is, and Love … and Acusilaus agrees with Hesiod that after Chaos, the Earth and Love, these two, came into being (transl. B. Jowett).

The context guarantees that Plato specifically considered the passage from the Theogony as testimony that the original Chaos was followed first by Earth and then by Eros, as the third entity: if the presence or absence of l. 118 is virtually irrelevant for the content of the passage, the absence of l. 119 is crucial for the vision of the primordial phases because it eliminates the Tartarus in the succession of the most ancient beings. The same holds true for Aristotle, Metaph. 984b27. ὑποπτεύσειε δ᾿ ἄν τις Ἡσίοδον πϱῶτον ζητῆσαι τὸ τοιοῦτον, ϰἂν εἴ τις ἄλλος ἔϱωτα ἢ ἐπιϑυμίαν ἐν τοῖς οὖσιν ἔϑηϰεν ὡς ἀϱχήν, οἷον ϰαὶ Παϱμενίδης· ϰαὶ γὰϱ οὗτος ϰατασϰευάζων τὴν τοῦ παντὸς γένεσιν “πϱώτιστον μέν” φησιν “ἔϱωτα ϑεῶν μητίσατο πάντων”, Ἡσίοδος δὲ πάντων μὲν πϱώτιστα Χάος γένετ᾿, αὐτὰϱ ἔπειτα Γαῖ᾿ εὐϱύστεϱνος … ἠδ᾿ Ἔϱος, ὃς πάντεσσι μεταπϱέπει ἀϑανάτοισιν, ὡς δέον ἐν τοῖς οὖσιν ὑπάϱχειν τιν᾿ αἰτίαν ἥτις ϰινήσει ϰαὶ συνάξει τὰ πϱάγματα.

 48 The text is that of Burnet OCT. The textual problem involving the sentence Ἡσιόδῳ δὲ ϰαὶ Ἀϰουσίλεως σύμφησιν does not affect the problem we are dealing with.

Ancient Scholarship on Hesiod  

One might suspect that Hesiod was the first to look for such a thing — or some one else who put love or desire among existing things as a principle, as Parmenides, too, does; for he, in constructing the genesis of the universe, says: “Love first of all the Gods she planned.” And Hesiod says: First of all things was Chaos made, and then broad-breasted Earth … and Love, ’mid all the gods pre-eminent, which implies that among existing things there must be from the first a cause which will move things and bring them together (transl. W.D. Ross).

Despite the lack of textual precision, evidently due to citing by heart, here too one finds confirmation of the succession Chaos–Earth–Eros, without the Tartarus (and without the formulaic l. 118).49 If Plato and Aristotle reflect a Hesiodic text without ll. 118–119 — since it is unlikely that they both suffered memory loss at exactly the same point, and even more unlikely that Aristotle depended on Plato, given the differences — it is difficult to say exactly what this really signifies: original text or text with lacuna? In other words, are ll. 118–119 a later interpolation as compared to the original text or were Plato and Aristotle reading a mutilated text from which these lines had disappeared? Can one go so far as to imagine that Plato and Aristotle had a text of the Theogony containing ll. 118–11950 and that they deliberately eliminated these lines because they regarded them as incorrect, placed in the wrong position, or possibly even non authentic? Moreover, the different weight of the two lines cannot be disregarded: what is truly important is actually only the presence or absence of l. 119. Is it conceivable that among the copies of the Theogony in circulation there were some with and some without this or these line(s) and that a philosopher considered the version that was more in harmony with his own ideas on cosmology and the origins of the world to be the valid version? If what we are dealing with is a genuine fluctuation in the text, involving the presence or absence of Tartarus at that particular point, then it would be interesting to try to place it within the history of the Hesiodic work and its composition and transmission.51 The use of authoritative figures from the world of thought, like the prestigious poets of the tradition, is a widely recognized characteristic of philosophical discourse: both Plato and Aristotle have extensive recourse to this manner of presenting their arguments. But what is particularly interesting is to note how  49 Aristotle also cites the passage from Hesiod in Phys. 208b30, but he gives only ll. 116–117. 50 Which are present in all the manuscripts, including P.Achmîm 3 of the 4th–5th century AD: cf. West 1966, ad loc. 51 Cf. the comm. to these lines in West 1966.

  Ancient Scholarship on Hesiod the debate focusing on a certain passage subsequently became transformed and thereby gave rise to a genuinely philological problem, taking on the distinctive features of the era of the grammarians and the corresponding technical language. Thus with regard to the passage from the Theogony which we have discussed here, in sch. Th. 117b it is stated that Πλάτων … ἀϑετεῖ τοὺς στίχους (scil. ll. 118–119): the debate on the content of the initial lines of the Hesiodic cosmology, taken up again in the commentaries and the exegetic treatises of the Hellenistic age, produced an approach to the problem in terms of athetesis, that is to say a deliberate and well motivated intervention on a text that did contain those lines, which were evidently regarded as questionable. Perhaps Zeno too was able to read the Hesiodic lines on the origin of the cosmos without the Tartarus, that is to say without l. 119: the evidence is uncertain but the Stoic philosopher may have directed attention to the problem again along the lines of Plato and Aristotle, in the framework of treatises of a cosmological nature.52 Similarly, for the Works and Days the scholia have likewise preserved evidence on Zeno, in connection with ll. 293–295 (SVF I 235): οὗτος μὲν πανάϱιστος, ὃς αὐτῷ πάντα νοήσει φϱασσάμενος τά ϰ᾿ ἔπειτα ϰαὶ ἐς τέλος ᾖσιν ἀμείνω, ἐσϑλὸς δ᾿ αὖ ϰἀϰεῖνος ὃς εὖ εἰπόντι πίϑηται

293 295

 52 The sources are sch. Hes. Th. 117a–b and sch. Apoll. Rh. 1.496–498b: cf. Algra 2001, 565– 569. In sch. Hes. Th. 117a–b (SVF 105), on the basis of the edition by Di Gregorio 1975, 24–25, it seems fairly clear that reference is being made specifically to Zeno’s conception of Chaos and to the absence of ll. 118–119 in Plato, Symp. 178b, without any apparent connection between the two things. In sch. Apoll. Rh. 1.496–498b (SVF 104) there may be a reference to the fact that even Zeno considered Eros to be the third primordial entity, without the Tartarus of l. 119 (nothing can be said about l. 118), but it remains doubtful whether the citation of Zeno, which started with regard to Chaos, continues with the inclusion of the sentence on Eros: ϰαὶ Ζήνων δὲ τὸ παϱ᾿ Ἡσιόδῳ (Th. 116 f.) χάος ὕδωϱ εἶναί φησι … τϱίτον δὲ ἔϱωτα γεγονέναι ϰαϑ᾿ Ἡσίοδον (Th. 120). Algra 2001, 569: “Given the fact that the scholia were recycled and abbreviated a number of times, it is not impossible that we are dealing with a garbled version of what originally was a reference to a proposal to athetize line 119 on Tartarus, on the ground that both Plato and Zeno ignored Tartarus as a separate factor in their exegesis … Anyway, given the problems involved in the constitution of the text, and given the fact that there is no clear reason to assume that the scholia here report a suggestion made on Zeno’s part, the sensible way to proceed would seem to be to limit the testimony on Zeno to the claim about the earth coming from the water”. I would reject the idea put forward by West 1966, comm. to 118–119, and West 1967, 5 n. 1, according to whom Zeno athetized l. 117 (but he uses the old edition of the scholia, without Di Gregorio 1975).

Ancient Scholarship on Hesiod  

The man who thinks of everything by himself, considering what will be better, later and in the end — this man is the best of all. That man is fine too, the one who is persuaded by someone who speaks well (transl. G. Most). Sch. Hes. Op. 293–297 οὗτος μὲν πανάϱιστος: Ζήνων μὲν ὁ Στωικὸς (SVF I 235) ἐνήλλαττε τοὺς στίχους, λέγων ‘οὗτος μὲν πανάριστος ὃς εὖ εἰπόντι πίθηται· ἐσθλὸς δ’ αὖ κἀκεῖνος ὃς αὑτῷ πάντα νοήσῃ·’ τῇ εὐπειθείᾳ τὰ πρωτεῖα διδούς, τῇ φρονήσει δὲ τὰ δευτερεῖα

This piece of information is confirmed in Diogenes Laertius 7.25–26: τούς ϑ᾿ Ἡσιόδου στίχους μεταγϱάφειν οὕτω· “ϰεῖνος μὲν πανάϱιστος ὅς εὖ εἰπόντι πίϑηται,/ ἐσϑλὸς δ᾿ αὖ ϰἀϰεῖνος ὃς αὐτὸς πάντα νοήσῃ”. ϰϱείττονα γὰϱ εἶναι τὸν ἀϰοῦσαι ϰαλῶς δυνάμενον τὸ λεγόμενον ϰαὶ χϱῆσϑαι αὐτῷ τοῦ δι᾿ αὑτοῦ τὸ πᾶν συννοήσαντος· τῷ μὲν γὰϱ εἶναι μόνον τὸ συνεῖναι, τῷ δ᾿ εὖ πεισϑέντι πϱοσεῖναι ϰαὶ τὴν πϱᾶξιν. It is said, moreover, that he corrected Hesiod’s lines thus: … The reason he gave for this was that the man capable of giving a proper hearing to what is said and profiting by it was superior to him who discovers everything himself. For the one had merely a right apprehension, the other in obeying good counsel superadded conduct (transl. R.D. Hicks).

I do not think one can imagine a discussion by Zeno concerning a textual problem in the strict sense. The observation clearly derives from a context involving a treatment of ethical themes53 and it offers a good illustration of the possibility of inserting the citation of a poet into a philosophical discourse, exploiting it for the purpose of developing the overall argument, e.g. by contending that the poet’s words would provide a better meaning and a better teaching if they were interpreted in some particular manner rather than another, or perhaps if they were modified and “corrected” according to a given approach. It is interesting to note that Diogenes speaks of στίχους μεταγϱάφειν, introducing a verb that had certainly taken on a “technical” philological value, so that he seems to be suggesting a veritable editorial intervention by Zeno,54 which seems rather implausible. Thus once again we have a “translation” into philological-grammatical terms of a particular point of a philosophical investigation, in which the passage

 53 The fragment is placed under the title Vitae agendae praecepta in SVF I, 56: under the same fr. 235 are registered also two references found in Themistius; cf. West 1978, 50–51 and comm. ad loc. 54 A typical illustration of this is the use made by Aristonicus in sch. Il. 2.681a Erbse; other cases can easily be found, for instance by utilizing Erbse’s index of the sermo grammaticus s.v. μεταγϱάφειν; cf. Dickey 2007, 206 and 246.

  Ancient Scholarship on Hesiod as well as its interpretation were called into play55 (as we saw above for terms such as ἐμβάλλω, διαγϱάφω, ὀβελίζω, ἀϑετέω). 5. “In the field of epic poetry Homer as the author of Iliad and Odyssey and Hesiod as the author of Theogony and Erga always occupied the first places which Aristotle and his school, followed by Zenodotus and his pupils, had assigned to them”. Thus wrote R. Pfeiffer, in the context of a survey of the selective lists of authors who formed the object of philological treatment in the culminating period of Alexandrian philology, the era of Aristophanes of Byzantium and Aristarchus. As Pfeiffer pointed out, there was a major thematic line of investigation that dated back to Aristotle and his school and which continued with the first generations of Alexandrian grammarians.56 “Aristotle and his school in the course of their literary researches had occasionally dealt with Hesiodic problems: the relative date of Hesiod and Homer, the authenticity of some of his poems, antiquarian questions. But we know of no edition before that of Zenodotus, who was the first to have the collected manuscripts at his disposal, nor can we tell whether his edition contained other Hesiodic texts besides the Theogony. Homer was without a rival; next to him Hesiod was a favourite with the great poets of the first half of the third century, especially with Aratus and Callimachus”.57 In effect, it is with Zenodotus that we find the first testimony to refer precisely to an ekdosis of Hesiod’s works.58 The sch. Th. 5 b2 cites τὰ Ζηνοδότεια, scil. ἀντίγϱαφα, with regard to a textual variant.59 Πεϱμησοῖο: οἱ μὲν ποταμοὶ τῆς Βοιωτίας οὗτοί εἰσι… ἐν δὲ ταῖς Ζηνοδοτείοις γϱάφεται Τεϱμησοῖο [ϰαϰῶς· ὁ γὰϱ Τεϱμησὸς ὄϱος ἐστὶ ϰαὶ οὐ ποταμός].60

 55 Some Homeric cases in Montanari 2000a (= this volume, ch. 38); 2008 (= this volume, ch. 54). 56 Pfeiffer 1968, 204–205. 57 Pfeiffer 1968, 117, cf. 177: “The scholar poets of the third century were remarkably fond of Hesiod … and their interest stimulated the activity of the grammarians”. 58 I explain what I mean by Alexandrian ekdosis in Montanari 1998a (= this volume, ch. 23); 2000a (= this volume, ch. 5); 2002a (= this volume, ch. 13); 2004 (= this volume, ch. 41); and 2009 (= this volume, ch. 24). 59 Cf. Jacoby 1930, 46; West 1966, 153; Pfeiffer 1968, 117; West 1978, 64; Most 2006, T 134; on Zenodotus, cf. also Martin 2005, 171–174. In reality ἐν δὲ ταῖς Ζενοδοτείοις would presuppose διορθώσεσι or ἐκδόσεσι, but the manuscript offers ἐν δὲ τοῖς Ζ., which has been defended by Schroeder 2009, in my view rightly and with good arguments: therefore the substantive understood would indeed be ἀντιγράφοις (see above).

Ancient Scholarship on Hesiod  

The reading proposed by Zenodotus can be found in several manuscripts: both forms are attested, but Πεϱμ– is predominant and accepted by modern editors.61 The citation of the name Zenodotus in sch. Th. 116 c1 raises a different kind of problems. Χάος γένετ᾿: … οἱ δὲ τὸ ὕδωϱ λέγουσιν, ἄλλοι δὲ τὸν ἀέϱα … ἄλλως. χάος λέγει τὸν ϰεχυμένον ἀέϱα· ϰαὶ γὰϱ Ζηνόδοτος (οὕτως) φησίν. Βαϰχυλίδης δὲ χάος τὸν ἀέϱα ὠνόμαζε, λέγων πεϱὶ τοῦ ἀετοῦ· νωμᾶται δ᾿ ἐν ἀτϱυγέτῳ χάει (5.26–27 S.–M.).62

After the proem, the account of the cosmogony starts at l. 116 with the primordial Chaos, which was at the origins of the world: this fundamental theme could not have been disregarded by philosophers and the scholia to this line mention a certain number of them.63 Zeno had argued that the primordial Chaos was water:64 here this opinion is quoted anonymously (οἱ δὲ τὸ ὕδωϱ λέγουσιν),65 but Zeno is then explicitly cited in the sch. 117a (SVF I 105). Others believed that Chaos was air, and Zenodotus in particular identified Chaos with “τὸν ϰεχυμένον ἀέϱα”, i.e. “the mist shed around”, according to the translation by Pfeiffer,  60 The sch. is cited according to Di Gregorio 1975, 4; on the expunged sentence ϰαϰῶς— ποταμός: “additamentum videtur indocti hominis qui Termessum a Teumesso non distinxit”. The same sch. preserves a citation of Crates, which is probably spurious: cf. Di Gregorio ad loc.; Broggiato 2001, lxvi; West 1966, ad loc. 61 Cf. West 1966, 153, ad loc., who points out that the initial Τ- is Attic, in contrast with Πwhich is Boeotian. The scholium says that Zenodotus read Τερμη(σ)σοῖο rather than Περμη(σ)σοῖο in Th. 5. The explanation of the variant could be simply graphic or could derive from different copies (both readings are attesed in the manuscripts), but could also go back to a more complex problem. If the initial consonant was a labiovelar, then Π- would fit Aeolic and Boeotian, whereas Τ- would fit Ionic and Attic. In the latter case Τερμη(σ)σοῖο could be an Ionicism owed to the copy that Zenodotus would have brought with him to Alexandria. Schroeder 2009 advances this possibility together with West’s idea that the Zenodotean copy of the Homeric poems would have had an Ionic patina, being an Ionicising rhapsodic copy from the home city of Zenodotus himself, Ionian Ephesus. For this Schroeder also compares the neoor hyper-ionic ἑωυτῇ at Th. 126, attested by a papyrus and by an indirect quotation, which West sets in the text in place of the much more widely attested Attic form ἑαυτῇ. 62 On the tradition of the text of this sch., see Di Gregorio 1975, ad loc.: the mention of Zenodotus is found in a part of the sch. within double square brackets, which the editor uses to indicate portions of text that are preserved by some mss. but which “ab archetypo aliena videntur” (p. xxxi). 63 Pp. 22–25 Di Gregorio; cf. West 1966, 192–193, ad loc. 64 SVF I 103, 104, 105: cf. Algra 2001, 566–569. 65 He is probably also the figure referred to in sch. Th. 116 c2: χάος παϱὰ τὸ χεῖσϑαι, cf. SVF I 103 (Ps.-Probus in Verg., Buc. VI 31, p. 344, 14–15 Thilo–Hagen): Zeno Citieus sic interpretatur, aquam χάος appellatam ἀπὸ τοῦ χέεσϑαι.

  Ancient Scholarship on Hesiod who seems to have no doubt that the scholar cited here is precisely Zenodotus of Ephesus (suggesting that the fragment could derive from his glossary).66 In contrast K. Nickau hesitates between Zenodotus of Ephesus and Zenodotus of Alexandria (2nd–1st century BC), to whom the Suda attributes a work Εἰς τὴν Ἡσιόδου Θεογονίαν, perhaps a commentary;67 on the other hand, the topic also suggests that one should not exclude Zenodotus of Mallos (2nd–1st century BC), probably to be identified with the Ζηνόδοτος ὁ Κϱατήτειος cited in some Homeric scholia and also with the Ζηνόδοτος ὁ Στωϊϰός mentioned in Diogenes Laertius 7.30 as a pupil of Diogenes of Babylon.68 If Zenodotus of Ephesus put forward an opinion of his own with regard to the Chaos that appears at the beginning of the Theogony, then we have to come to the conclusion that the Hesiodic exegesis by the grammarian who was the first “editor” of Hesiod encroached on a field which was already being seriously addressed by philosophers, among whom the grammarian’s contemporary Zeno. But of course this remains doubtful. The scanty remains of the work of Apollonius Rhodius on Hesiod certainly do not authorize the hypothesis of a commentary, but they suffice to state that he dealt with this poet far from sporadically.69 We saw earlier (§ 3 above) that Apollonius expressed himself in favour of the Hesiodic authenticity of the Aspis in the third book (ἐν τῷ τϱίτῳ) of a treatise the general subject of which is unknown: one can imagine it may perhaps have been a text on Hesiodic problems that addressed questions of authenticity and other issues, with scattered observations on significant passages, of which we have a couple of fragments. We already noted (§ 3 above) that the observations by Apollonius and Aristophanes on the Aspis imply that they both regarded the Catalogue as authentic. This opinion was shared by Crates, as we will see further on. A sch. vet. to the end of the Works and Days informs us that Apollonius regarded as spurious the so-called Ornithomanteia (of which we do not know the extension), i.e. the part following the existing l. 828, the line that concludes the poem as it has come down to us.

 66 Pfeiffer 1968, 117; Nickau 1972, 22, 20 and 40, 34. 67 Cf. Nickau 1972, 20–23, esp. 22, 20; this should be the person cited in P.Oxy. 1241 within a group of grammarians who flourished under Ptolemy IX (116–80 BC): cf. Nickau 1972, 21, 42; Pfeiffer 1968, 254; West 1978, 64 n. 3; Most 2006, T 140, labels the fr. “Zenodotus Alex.” 68 Cf. Nickau 1972, 45–47 and 49. 69 On Apollonius and Hesiod cf. West 1966, 50–51 and 162; Pfeiffer 1968, 144; West 1978, 64 and 364; West 1985, 127; Most 2006, T 80 and 135 and p. lxii.

Ancient Scholarship on Hesiod  

Sch. 828a … τούτοις δὲ ἐπάγουσί τινες τὴν Ὀϱνιϑομαντείαν (***), ἅτινα Ἀπολλώνιος ὁ Ῥόδιος (fr. p. 42 Michaelis) ἀϑετεῖ.70 At his point some people add the Bird Omens, which Apollonius Rhodius marks as spurious (transl. G. Most).

Perhaps the athetesis of Apollonius, or at least the athetesis he upheld, exerted some influence in later times, since the Ornithomanteia has not come down to us, unlike the Aspis. It is not clear exactly what problem Apollonius raised concerning l. 26 of the Theogony: ποιμένες ἄγϱαυλοι, ϰάϰ᾿ ἐλέγχεα, γαστέϱες οἶον. Sch. 26b … Ἀπολλώνιος μὲν ὁ Ῥόδιος (fr. 19 Michaelis) λείπειν τὸν πϱῶτον στίχον φησίν· οὐ λείπει δέ, ἀλλ᾿ ἔστι· ποιμένες ἄγϱαυλοί ἐστε, ϰάϰ᾿ ἐλέγχεα, γαστέϱες μόνον.

What is meant is not the suspicion of a lacuna after the first line of the Muses’ words:71 rather, I believe, Apollonius felt that the meaning of the sentence was not complete, and that something was missing (whether he was striving to hypothesize something is of course unknown).72 The anonymous respondent argues that nothing is missing in the meaning, but that the line should simply be interpreted as a nominal phrase, with the verb ἐστε understood. We have somewhat more information on Aristophanes of Byzantium.73 As we have seen (§ 3 above), he too addressed questions of the authenticity of works attributed to Hesiod. Pfeiffer hypothesizes that he discussed the matter “in his supplement to the Pinakes of Callimachus” and this is quite plausible.74 We do not know what formed the basis for his belief that the Precepts of Chiron was not a work by Hesiod, whereas his doubts concerning the Aspis sprang from a careful examination of exemplars of the Catalogue of Women, where in book IV he found ll. 1–56 of the Aspis, i.e. the Alcmena–Ehoie (the mother of Heracles), separately and without the description of Heracles’ shield, which in the Aspis is attached to these lines as the major portion of the poem.75 This argument clearly

 70 Text of Pertusi 1955, 259: on the lacuna cf. Test. et adn. cr., p. 227; cf. Jacoby 1930, 48; West 1978, 364; Most 2006, T 80 and p. lxii; Cingano 2009, 103 f. 71 As appears to be assumed by Pfeiffer 1968, 144. 72 West 1966, 162, ad loc.; Di Gregorio 1975, 7, ad loc. 73 Cf. Jacoby 1930, 46; West 1966, 50–51; Pfeiffer 1968, 177–178; West 1978, 64–65; Aristophanes fr. 405–407 Slater 1986; Most 2006, T 52, 69, 136; the name also appears in sch. Th. 126, a passage that gave rise to knotty problems: cf. Di Gregorio ad loc., Slater fr. sp. 439. 74 Pfeiffer 1968, 178; cf. West 1978, 64. 75 Fr. 406–407 Slater 1986; West 1978, 64, suspects that he also rejected the Wedding of Ceyx.

  Ancient Scholarship on Hesiod presupposes a conviction that the Catalogue was by Hesiod. In any case, barring errors, the only ancient evidence against the Hesiodic authorship of this work can be found in the story, related by Pausanias,76 according to which the Boeotians living around the Helicon believed that only the Works and Days was authentic, and in a passage from Aelian, where the reference to a content of the Catalogue is accompanied by an expression of doubt as to its attribution.77 For Aristophanes, we owe to sch. Th. 68a an important piece of information. αἳ τότ᾿ ἴσαν: ἐπεσημήνατο ταῦτα ὁ Ἀϱιστοφάνης (fr. 405 Slater). νῦν πεϱὶ ἀνόδου αὐτῶν φησι τῆς εἰς τὸν Ὄλυμπον· ὁ πϱότεϱος γὰϱ ἦν λόγος αὐτῷ πεϱὶ τῆς ἐν τόπῳ αὐτῶν χοϱείας.

We cannot tell what critical semeion Aristophanes put at this line and whether its motivation was that indicated in the subsequent part of the scholion, but this is nonetheless sufficient to state that “he must have followed Zenodotus in editing Hesiod”, or at least the Theogony.78 We will come back to the question of the Hesiodic semeia shortly. It is no doubt purely by chance that we have no specific information on the opinions of Aristarchus with regard to the problems of authenticity that had long been a bone of contention.79 As we have already seen (§ 2 above), we only know that he (also on a documentary basis) athetized the proem of the Works and Days, but we have no information concerning the proem of the Theogony. As regards the philological activity of Aristarchus concerning Hesiod, we have rather more information. In addition to that concerning the proem of the Works and Days, we have another seven certain (plus two conjectural) citations for textual or exegetic problems. Sch. Th. 114–115 tells us that the grammarian

 76 Paus. 9.31.4–5, cited above, § 2 n. 38. 77 Ael., V. H. 12.36 = fr. 183 M.–W.; concerning the ancient scholarship on the Catalogue cf. also Schwartz 1960, 609–618; Arrighetti 1998, 445–447; Hirschberger 2004, 42–43. 78 Pfeiffer 1968, 177; cf. West 1978, 64–65. 79 As far as the Catalogue is concerned, it is likely that Aristarchus also considered it authentic, following his predecessors Apollonius Rhodius and Aristophanes. In the sch. Il. 23.683b Hesiod is defined as more recent than Homer and a reference to the Catalogue is attributed to him, indicating that the source must have believed the Catalogue to be authentic; however, there is no reliable evidence to say whether this source was Aristarchus himself (West 1985, 127 and n. 3, p. 135), because the relative chronology of the two poets was treated by various authors who deemed Hesiod to be more recent, as can be found in the Lives (and elsewhere: see T 5–9 Most 2006, 164–167). In sch. Th. 338 and Op. 94a it is stated that Hesiod is more recent than Homer and this may effectively be a “doctrine characteristic of Aristarchus” (West 1978, 65 and nn. 6–7), though this remains on the plane of hypothesis.

Ancient Scholarship on Hesiod  

Seleucus (1st century AD)80 athetized the two lines, whereas Aristarchus athetized only the second;81 Aristarchus also expunged Op. 210–211 (sch. Op. 207–212, with motivation)82 and Op. 740–741 (sch. Op. 740a);83 in sch. Th. 991 his name can be restored conjecturally with regard to a variant.84 Exegetic interventions are preserved in sch. Th. 76 (πϱοφεϱεστάτη ἀντὶ τοῦ πϱεσβυτάτη), sch. Th. 138 (οὐ ϰαλῶς τοῦ πϱεσβυτέϱου Οὐϱανοῦ ϑαλεϱοῦ εἰϱημένου), sch. Th. 253 (a problem involving the names of the Nereids); according to a proposed correction, the text of sch. Th. 379 could be regarded as citing him in connection with the adjectivization of the wind Zephyrus.85 The EM (539.22) and the EGud (346.46 Sturz) preserve an observation by Aristarchus on the spelling and accentuation of the name of the Titan Κϱ(ε)ῖος, mentioned in Th. 134.86 In sch. Op. 97 we find a response by Aristarchus against the grammarian Comanus of Naucratis on the problem of the Elpis remaining in the pithos.87 From these data we can reach the conclusion that Aristarchus, continuing the work of his predecessors and maintaining a lively interest in the other great ancient epic poet alongside Homer, made an ekdosis of the Theogony and the Works and Days and in all probability also wrote a commentary.88 His studies on archaic Greek epic poetry led him, in the wake of his predecessors, to draw a distinction between the Hesiodic and the Homeric style, and to use this distinction as a criterion for evaluating the genuinely Homeric origin of lines with Ἡσιόδειος χαϱακτήϱ. A famous case is that of the list of Nereids in Il. 18.39–49 (cf. Did. in sch. ad loc.), where in any case the observation and the proposed athetesis already dated back to Zenodotus, but see also the sch. ad Il. 24.25–30 (the passage that mentions the judgement of Paris) and the sch. ad Il. 24.614–617a. It should also be recalled that Apollonius Rhodius adduced precisely the χαϱακτήϱ as a reason for the authenticity of the Aspis.89 Recently C.M. Schroeder has found that the sch. of

 80 Fr. 27 Müller 1891; cf. above, § 3 end and n. 40, for the Aspis; cf. below, § 6 and n. 102. 81 Cf. West 1966, ad loc. 82 Cf. West 1978, ad loc. and p. 65. 83 Pertusi’s integration ϰαὶ τὸν ἑξῆς is necessary: l. 740 cannot be athetized alone. 84 Cf. Di Gregorio 1975, ad loc. 85 Cf. Di Gregorio 1975, ad loc. 86 Cf. West 1966, 202; see also Algra 2001, 570–572. 87 It is likely that the fragment comes from the polemical Aristarchean work Πϱὸς Κομανόν, of which a few fragments of Homeric criticism have been preserved: cf. West 1978, 65; Verdenius 1985, ad loc.; Dyck 1988, fr. 16, pp. 251–254; Montanari a, 1999. 88 Cf. Jacoby 1930, 46; Pfeiffer 1968, 220; West 1966, 50–51; 1978, 65 (see bibliography in n. 2) expresses some doubt: for the issue on Aristonicus see below, n. 94; Most 2006, T 49 and 137. 89 See above in § 3; cf. Schwartz 1960, 43–46.

  Ancient Scholarship on Hesiod Aristonicus to Iliad 10.431 a Erbse restores the citation of a hitherto unknown monograph of Aristarchus: Περὶ ἡλικίας Ἡσιόδου. Of it we know only the title: it would seem to be biographical in character and could have engaged in the debates on the dates of Homer and Hesiod and on which of the two was the earlier.90 As had been the case with Homer, the progress in the editorial activity of the Alexandrian scholars concerning Hesiod was likewise characterized by the use of critical signs. We have no information on Zenodotus, but in actual fact there is no reason why in his ekdosis of Hesiod he would not have used the obelos he also used for Homer. For Aristophanes of Byzantium a semeion is presupposed by sch. Th. 68a, as we saw above; for some interventions by Aristarchus concerning the Theogony and the Works and Days one finds in the scholia the verb ὀβελίζω. In the sch. vet. critical signs, without indication of a grammarian’s name, are implied by sch. Th. 233 (τὸ δὲ σημεῖον παϱάχειται), 484 (τὸ σημεῖον ὅτι), 927 (διὸ ϰαὶ τὸ σημεῖον); sch. Op. 276b (τὸ σημεῖον ὅτι), 649a (σημειοῦται ὁ στίχος οὗτος).91 All this must have been transmitted through the work of the grammarian Aristonicus (cited for a reading in sch. Th. 178) entitled Πεϱὶ τῶν σημείων τῶν ἐν τῇ Θεογονίᾳ Ἡσιόδου92 and therefore it must go back to Aristarchus. There is no evidence for a similar treatise on the Works and Days, but the assumption that it must have existed seems to me plausible. In the Etymologicon of Orion (96.28–29 Sturz) we find the citation Ἀϱίσταϱχος ἐν τοῖς σημείοις Ἡσιόδου: this is a most curious way of citing a commentary, but the expression becomes meaningful if one imagines there may have been a mistake in the author’s name for Ἀϱιστόνιϰος (wrong interpretation of an abbreviation?). A remarkable piece of evidence of this tradition is P.Oxy. 3224, a small fragment of an exemplar of the Works and Days of the 2nd century AD, which presents a lefthand margin that is extremely rich in semeia: in just a few lines we have the obelos, the diple, the asteriskos and probably the chi.93

 90 Schroeder 2007. 91 West 1978, 65 n. 4 mentions another four scholia beginning with τὸ σημεῖον ὅτι, which are present in only one ms. but absent in the edition by Di Gregorio 1975. 92 Suid. s.v. Aristonikos (α 3924); cf. Cohn 1895, 966; Pfeiffer 1968, 220 and n. 3; West 1978, 65; Montanari 1996, 1120. 93 Edition by D. Obbink in P.Oxy. XLV, 1977, 51–52. An overview of Hesiodic exegesis preserved in the papyri can be found in Stroppa 2008a: while maintaining the due caution called for by the chance nature of the discoveries, we can certainly say that the remains of exegetic material on Hesiod preserved in the papyri are rather scanty, especially if compared to Homer; none of the preserved fragments can be attributed with certainty to a commentary and P.Oxy. 3224 is probably the most significant piece of evidence.

Ancient Scholarship on Hesiod  

The work by Aristonicus on the Hesiodic critical signs of Aristarchus94 charts a path that is exactly parallel to the well-known history of Homeric critical signs, although the documentation is far less rich, and naturally for Hesiod we have neither a Ven. A codex nor a corpus of scholia comparable to that existing for Homer.95 For the other great erudite figure of the Augustan age, namely Didymus, we know even less than we do for Aristonicus. We may perhaps, for Didymus too, go as far as to hypothesize that in the case of Hesiod he played a role and a function parallel to his known role for Homer, acting as the scholar who conserved the philological heritage of the Hellenistic age and transmitted the scholarship of Aristarchus, but the two scanty shreds of exegetic remains that we have allow us little more than a plausible hypothesis. The remains in question are sch. Th. 126 and sch. Op. 304b.96 A subject touched on earlier (§ 2 above) is the Hesiodic exegesis of Crates of Mallos, whom we mentioned in connection with the athetesis of the proems of the Theogony and of the Works and Days. We have another three fragments of Crates concerning Hesiod, which deal with exegetic and textual problems. There is no certainty that Crates did indeed write commentaries on Hesiod, but the insight offered by his remarks and the specific problem of text criticism on Op. 530 seem to point in this direction. But what is certain is that in turning his attention to Hesiod, Crates was working in the wake of a Stoic tradition that was interested in themes of a cosmological and religious nature,97 which in the exegetic material that has come down to us is well testified by the citations of Zeno (cf. § 4 above) and of Chrysippus. Sch. Th. 126 preserves an observation on a cosmological theme, in which Crates is said to have found in Hesiod the concept that the earth and the sky are spherical. It is stated in Th. 126–127 that Earth generated Uranos equal to herself so that he would cover her completely, but Crates said that the earth cannot be “equal” (l. 126: ἶσον) to the sky: ἀλλ᾿ ὁ Κϱάτης ἀποϱεῖ· εἰ γὰϱ ἶσον, πῶς δύναται ϰαλύπτειν? Such a difficulty can be comprehended only in terms of a spherical conception, which would require that the sky covering the earth is more extend 94 Cf. Jacoby 1930, 46–47; Pfeiffer 1968, 220 n. 3; Most 2006, T 142. I am surprised to find the statement in West 1978, 65, that “If Aristarchus had left his own commentary, there would have been no need for Aristonicus”: this hardly strikes me as a reason for doubting the existence of an Aristarchean commentary, as shown by the parallel Homeric example (cf. above, n. 89). 95 On the corpus of the Hesiodic scholia, see the recent overview by Dickey 2007, 40–42. 96 Schmidt 1854, 299–300, cf. p. 12; West 1966, 51 and 69; 1978, 66 and 233 (comm. to v. 304); Most 2006, T 143. 97 Broggiato 2001, xxiii, F 78–81, pp. 78–80, comm. pp. 239–241; cf. Jacoby 1930, 46; Pfeiffer 1968, 241; Most 2006, T 139.

  Ancient Scholarship on Hesiod ed. In the subsequent part of the scholion the solution of the difficulty, evidently put forward by Crates himself, is to say that “equal” should be understood in the sense of the spherical shape and not in terms of extension.98 In Th. 142–143 it is stated that the Cyclopes were similar to the gods, except for the fact of having only one eye. The scholion to l. 142 says that Crates noted a contradiction with the section of the Catalogue which relates that the Cyclopes were killed by Apollo, and therefore they could not be equal to the gods, i.e. immortal (fr. 52 M.–W.). Crates untangled this contradiction by replacing l. 142 with another line, which claimed that the Cyclopes were human, but that they were brought up by the gods. The predominant opinion is that the replacement line is the work of Crates himself and does not derive from a different version of the Theogony.99 This line of reasoning presupposed that Crates, like Apollonius Rhodius and Aristophanes before him, believed the Catalogue to be by Hesiod. Sch. Op. 529–531 (from Proclus) and the EGen AB = EM 594.46 state that in l. 530, in the description of winter, instead of μυλιόωντες “gritting their teeth”, Crates actually wrote (γϱάφει: a conjecture?) μαλϰιόωντες “stiff with cold” (from μαλϰιάω, normally μαλϰίω).100 6. Other sporadic evidence of Hesiodic exegesis testifies to a constant interest at various levels, of which only few fragments have survived. Of Dionysius Thrax, a pupil of Aristarchus, we have a gloss on the word φεϱέοιϰος in Op. 571, that has come down to us in the scholion of Proclus,101 but information as to the type of work from which this derived is totally lacking; for Demetrius Ixion (today dated in all probability to the 2nd century BC) the Suda preserves the title Εἰς Ἡσίοδον ἐξήγησις.102 We have already encountered Seleucus (1st century AD), one of the grammarians most frequently cited in the Hesiodic scholia (sch. Th. 114–115, 160, 270, 573; sch. Op. 96a, 150b, 549; sch. Sc. 415) for athetesis and textual variants:103 it would appear that he was concerned with the Theogony, the Works and Days and the Shield, perhaps producing editions and commentaries; he may possibly have played an important role in the transmission of Alexandrian exegetic materials. Mention has also already been made of the hypomnema to the Aspis by his contemporary Epaphroditus (1st century BC).104  98 F 79 Broggiato 2001, with comm. 99 F 80 Broggiato 2001, with comm. and bibliographic references; cf. West 1966, ad loc. 100 F 81 Broggiato 2001, with comm.; cf. West 1978, ad loc. 101 Sch. Op. 571–577, p. 186, 5–6 Pertusi: cf. Pfeiffer 1968, 267; West 1978, 66; Pagani 2008. 102 Ascheri 2009; Most 2006, T 141. 103 §§ 3 and 5 above: cf. West 1978, 66–67; fr. in Müller 1891; Most 2006, T 144. 104 Cf. above, § 3 and n. 40; Most 2006, T 145. For the imperial age, cf. Jacoby 1930, 47.

Ancient Scholarship on Hesiod  

Other 1st century AD grammarians cited in the scholia vetera are Tryphon and Habron (both in sch. Th. 389). Plutarch devoted much attention to Hesiod and wrote a major commentary on the Works and Days, which certainly preserved information on the previous Hesiodic studies. Proclus used it and a number of fragments have thereby been preserved in the scholia.105 During the imperial age, Hesiod’s style was the object of interest and analysis by a number of rhetoricians and literary critics.106 As we have seen, problems of authenticity and selection within a fairly wide-ranging and considerably variegated corpus were frequently discussed during the Alexandrian age, following a long path strongly similar to that which for Homer, as far as we know, dates back at least to Herodotus (who in 2.117 adduced content-based arguments to claim that the Cypria is not by Homer), and which led to restricting the authentically Homeric works to the Iliad, the Odyssey and the Margites. Of the two major works forming the object of discussion, the Aspis has been preserved in the mss., while the Catalogue is lost, even though the numerous papyri testify to its fortune during the imperial age, in addition to the triad consisting of the Theogony, the Works and Days and the Shield.107 We will conclude with a brief remark on the scholia vetera that have come down to us. The roots of the corpus very probably can be traced back to a commentary (or possibly more than one, combined at a later period) of the early imperial age, given that no citations of grammarians beyond the time of Tryphon and Habron (1st century AD) are found. The commentary was a repository of the many streams of the scholarship of the Hellenistic age (above all, but not exclusively, Alexandrian) and of the philosophical interpretations deriving chiefly from the Stoics. As usual and in accordance with general practice, the commentary was epitomized, abridged, altered in the manuscripts in which it was transmitted in the form of marginal annotations; moreover the resulting corpus itself was then further enriched over time with new materials of later origin, namely paraphrases and other contributions by Byzantine erudite scholars. Since Hesiodic scholarship began fairly early, at least with Aristotle, in the scholia vetera one finds fragments of respectable antiquity, citations of the grammarians touched on in this paper, references to philosophers such as Zeno

 105 Cf. § 1 above; Maes 1939; West 1978, 67–68; Most 2006, lxix and T 147; on the commentaries of the Byzantine age (Proclus, John Tzetzes, Planudes, Moschopoulos), see West 1978, 68–75. Proclus: Marzillo 2009; Moschopoulos: Grandolini 1991. 106 See Hunter 2009. 107 On the problems of the Hesiodic corpus, see Cingano 2009.

  Ancient Scholarship on Hesiod and Chrysippus.108 This fragmentary evidence allows us a glimpse of the rich array of critical-exegetical achievements, the intellectual products of which have for the most part been lost: it is the rise and development of this heritage that we have sought to bring to light in these pages.

 108 The essentials in West 1966, 69–70; 1978, 67–68; Dickey 2007, 40.

 Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201 Il quadro degli hypomnemata, in particolare pindarici, restituiti dai papiri si è recentemente arricchito di un interessante reperto, P.Oxy. LXXIX 5201, un volumen datato «tentatively» dagli editori al Ia/Ip, dunque uno hypomnema fra i più antichi. Si conservano due colonne di testo, mutile in diversi punti, di un commentario alla Olimpica 1.1 Le righe si possono ricostruire grazie ai lemmi pindarici, presenti con abbondanza: contano da 24 a 33 lettere e quindi sono abbastanza larghe, cosa piuttosto frequente in questo genere di testi (se ne conoscono anche di ben più larghe). Il commentario è scritto sul recto («along the fibres») e secondo gli editori si tratta di «a fragment … apparently cut down to the present size and shape for a letter to be written on the back, upside down in relation to the text on the front».2 Si tratta dunque di un rotolo originariamente usato proprio per il commentario pindarico e in seguito riutilizzato per altro, in questo caso una lettera (inedita, databile al I/IIp).3 Giusto per menzionare solo qualche caso fra i più noti e significativi: il celebre hypomnema di Teone alle Pitiche di Pindaro (P.Oxy. XXXI 2536, del IIp) è scritto sul recto e il verso è rimasto bianco, dunque il rotolo non è stato riutilizzato ed è stato conservato solo per il commentario pindarico; il contrario accade per il commentario alle Pitiche di P.Berol. inv. 13419 del III/IVp, che è stato scritto riutilizzando il verso di un documento, e per il commentario agli Epinici di Bacchilide, P.Oxy. XXIII 2367 del I/IIp, anch’esso scritto sul verso di un documento. Il cosiddetto «Commento di Ammonio» a Iliade XXI, P.Oxy. II 221 del IIp, è scritto sul verso di un trattato di metrica (P.Oxy. II 220); lo hypomnema a Iliade II, P.Oxy. VIII 1086 del Ia, è scritto sul recto di un rotolo, che è stato poi riutilizzato per ricette mediche sul verso; lo hypomnema a Iliade VII, P.Oxy. VIII 1087 del Ip, è scritto sul recto di un rotolo che non è stato riutilizzato perche il verso è rimasto bianco. Il famoso hypomnema di Didimo alle Filippiche di Demostene, P.Berol. inv. 9780 del IIp, è scritto sul recto di un volumen, che è stato riutilizzato nel verso per copiare un trattato di etica del filosofo stoico Ierocle.

 Grazie a Fausto Montana per le sue preziose osservazioni.  1 W.S. Barrett and D. Obbink, P.Oxy. 5201. Commentary on Pindar, Olympian 1, in: The Oxyrhynchus Papyri, Vol. LXXIX, London 2014, 116–124. 2 Vd. p. 116. 3 Barrett/Obbink 2014, 117. https://doi.org/10.1515/9783110772371-049

  Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201 Credo che varrebbe la pena fare un’indagine sistematica sugli hypomnemata papiracei conservati in volumina (indipendentemente dal fatto che si tratti di una epitome oppure no) da questo punto di vista, cioè se sono stati scritti sul recto o sul verso di un rotolo, nel primo caso se il verso è rimasto bianco e dunque non utilizzato oppure se il rotolo è stato riutilizzato per altro, se la riutilizzazione è stata del rotolo intero oppure se esso è stato smembrato e adattato in pezzi a diversi usi. Tenendo conto delle epoche di appartenenza e di altri aspetti librari, compreso il fatto che un frammento bianco sul verso non garantisce del tutto che il rotolo non sia stato riutilizzato in qualche altra parte, credo comunque che un’analisi dei dati relativi a questo aspetto nei suoi diversi elementi consentirebbe un tentativo di indagine storico-culturale e forse anche socioculturale con risultati di un certo interesse.4 In P.Oxy. LXXIX 5201 abbiamo resti di un commentario alla Olimpica 1, composta per la vittoria di Ierone di Siracusa nella corsa con il cavallo montato nel 476a. Una fonte antica ci dice che fu Aristofane di Bisanzio a collocare la Olimpica 1 nella privilegiata posizione di apertura della raccolta delle Olimpiche e di conseguenza di tutti gli Epinici,5 che divenne in seguito canonica: una scelta significativa, che viola la consueta gerarchia delle discipline sportive, secondo la quale era privilegiata la corsa dei carri, a favore di quello che Luciano definisce «il più bello di tutti i canti».6 È possibile che la scelta sia dovuta principalmente al fatto che l’ode racconta il mito di Pelope, riconosciuto aition dell’agone olimpico qui cantato, unitamente all’importanza del personaggio celebrato, ma appare plausibile anche una motivazione estetica legata all’eccellenza poetica dell’ode. Inoltre, l’ode si pone come un vero manifesto di poetica pindarica, che affianca l’eccellenza del poeta a quella del celebrato. In ogni caso, sembra difficile respingere una motivazione precisa e specifica, assolutamente non casuale, dell’importanza conferita alla Olimpica 1. Qualche ragionamento sul commentario può essere fatto, anche tenendo conto dei suggerimenti ricostruttivi degli editori stampati a testo, che appaiono del tutto plausibili. Le due colonne parzialmente conservate contenevano il commento ai vv. (17) 19–39 (45?), cioè a una trentina di versi dell’ode. Il commento alla parte iniziale fino ai versi 16/17 doveva occupare una o al massimo due colonne precedenti. Gli editori stimano circa 70/71 colonne per commentare tutte le Olimpiche, più eventuale materiale introduttivo (non certo ma plausibile,  4 Osservazioni sul tema in Lama 1991. 5 Arist. Byz., fr. 381 Slater: Vita Thomana p. 7, 14 sgg. Drachmann e P.Oxy. XXVI 2438, 35–39; sul problema cfr. Negri 2004, 16–34. 6 Luciano, Gallus 7.

Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201  

dato che siamo all’inizio almeno delle Olimpiche, se non dell’insieme degli Epinici), per un totale che avrebbe potuto arrivare fino a 75 colonne in un rotolo di circa 6 metri.7 Mi sento di osservare che si tratta di un’impresa editoriale (per autore e opera) e libraria (per il manufatto) certamente importante, per cui acquista rilevanza il fatto che sia stata distrutta e tagliuzzata per una lettera privata e presumibilmente per altro: possiamo pensare a un cambio di proprietà, a un oggetto passato da una persona interessata a questo genere di libri (uno studioso?) a un’altra che invece lo ha usato come semplice carta da riciclo, oppure semplicemente a un guasto materiale del rotolo? Il commentario mostra che i lemmi contenevano il testo dell’ode per intero, come rilevano gli editori:8 «The lemmata … are written out in full, as prose (i.e. without colometry) and in consecutive entries, so as to reproduce (at any rate in the preserved portion) the entire text of the poem without omission». Per visualizzare con chiarezza questo fatto, riporto il testo dell’ode per la sezione interessata. Il commento alla parte iniziale, dall’incipit fino alla fine della pericope terminante al v. 17 con τράπεζαν, doveva occupare (come dicevamo) una o due colonne prima delle due rimaste. Sono evidenziati in grassetto i lemmi conservati, nei quali alle parti materialmente visibili sul papiro sono state integrate in corsivo le parti cadute ma la cui presenza è garantita dal commento. 

ἄριϲτον µὲν ὕδωρ, ὁ δὲ χρυϲὸϲ αἰθόµενον πῦρ ..............



ἄνδρεϲ ἀµφὶ θαµὰ τράπεζαν. ἀλλὰ Δωρίαν ἀπὸ φόρµιγγα παϲϲάλου λάµβαν᾽, εἴ τί τοι Πίϲαϲ τε καὶ Φερενίκου χάριϲ



νόον ὑπὸ γλυκυτάταιϲ ἔθηκε φροντίϲιν, ὅτε παρ᾽ Ἀλφεῶν ϲύτο, δέµαϲ

I –

ἀκέντητον ἐν δρόµοιϲι παρέχων, κράτει δὲ προϲέµειξε δεϲπόταν,

I –

Ϲυρακόϲιον ἱπποχάρµαν βαϲιλῆα. λάµπει δέ οἱ κλέοϲ

I –

ἐν εὐάνορι Λυδοῦ Πέλοποϲ ἀποικίᾳ∙ 25

30

τοῦ µεγαϲθενὴϲ ἐράϲϲατο γαιάοχοϲ Ποϲειδᾶν, ἐπεί νιν καθαροῦ λέβητοϲ ἔξελε Κλωθώ, ἐλέφαντι φαίδιµον ὦµον κεκαδµένον.

I – I –

ἦ θαυµατὰ πολλά, καί πού τι καὶ βροτῶν φάτιϲ ὑπὲρ τὸν ἀλαθῆ λόγον

lac. I/II

δεδαιδαλµένοι ψεύδεϲι ποικίλοιϲ ἐξαπατῶντι µῦθοι.

lac. I/II

 7 Barrett/Obbink 2014, 116 e 122–123. 8 Barrett/Obbink 2014, 117.

  Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201 Χάριϲ δ᾽, ἅπερ ἅπαντα τεύχει τὰ µείλιχα θνατοῖϲ, ἐπιφέροιϲα τιµὰν καὶ ἄπιϲτον ἐµήϲατο πιϲτὸν

II 

ἔµµεναι τὸ πολλάκιϲ∙ ἁµέραι δ᾽ ἐπίλοιποι

II 

µάρτυρεϲ ϲοφώτατοι. 35

ἔϲτι δ᾽ ἀνδρὶ φάµεν ἐοικὸϲ ἀµφὶ δαιµόνων καλά∙ µείων γὰρ αἰτία.

II –/

υἱὲ Ταντάλου, ϲὲ δ᾽, ἀντία προτέρων, φθέγξοµαι,

II –

ὁπότ᾽ ἐκάλεϲε πατὴρ τὸν εὐνοµώτατον ἐϲ ἔρανον φίλαν τε Ϲίπυλον, 40

ἀµοιβαῖα θεοῖϲι δεῖπνα παρέχων, τότ᾽ Ἀγλαοτρίαιναν ἁρπάϲαι.

II –

20 l. Ἀλφεῷ

I lemmi sono formati da pericopi del testo, tagliate in modo certamente non casuale bensì legato al senso, per cui sono di lunghezza molto disuguale, da 2 a 28 parole. Come dicevamo, importante è il fatto che le pericopi siano assolutamente consecutive, senza alcuna omissione, così da riprodurre l’intero testo dell’ode: questo fatto è constatabile nella parte conservata ed è del tutto plausibile che fosse mantenuto per tutto il commentario. Questo fatto mi induce a riprendere una nozione sulla quale ho richiamato l’attenzione in alcune occasioni passate, quella di “edizione commentata”, le cui caratteristiche essenziali sono le seguenti: il testo oggetto di commento è diviso in pericopi e riportato per intero, così che il fruitore possa leggere l’opera senza bisogno di ricorrere a un altro supporto librario; il discorso esegetico è intercalato alle pericopi di testo e caratterizzato da una parafrasi, alla quale sono occasionalmente aggiunti elementi di contenuto esplicativo, senza alcuna sistematicità e in forma stringata. Nel lavoro più recente su questo tema scrivevo: «In verità, non vorrei che questi due esempi [scil. P.Fay. 3 dei Topica di Aristotele e il P.Lille degli Aitia di Callimaco] inducessero a connettere per forza il modello della ‘edizione commentata’ alle fasi iniziali del percorso esegetico su un autore … Più prudentemente, mi accontenterei di ritenere identificata una tipologia libraria ed esegetica … un prodotto evidentemente diverso sia dall’hypomnema [aggiungerei stricto sensu] che dalla nuda parafrasi interlineare e dal glossario (anche continuo)».9 Confrontata con quanto possiamo intendere come uno hypomnema in senso proprio e stretto, la ‘edizione commentata’ è dunque un po’ diversa (azzarderei: per la  9 Avezzù/Scattolin 2007, con i riferimenti alla bibliografia precedente; i passi riportati sono alle pp. 13 e 14.

Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201  

forma e per gli scopi, quindi anche per il pubblico?), e spero che prima o poi ci si decida ad adottare una terminologia più precisa nella definizione dei testi dell’esegesi antica. Gli editori rilevano la particolarità del prodotto nel quadro dell’esegesi pindarica conosciuta:10 «No other commentary on Pindar reproduces the text to this extent in its lemmata or explanations». Credo sia abbastanza naturale connettere l’allestimento di una ‘edizione commentata’ almeno per la Olimpica 1 (o anche per tutte le Olimpiche di Pindaro, dato che potevano essere contenute in un solo rotolo di circa 6 metri, vedi supra) a fattori come l’importanza e l’eccellenza dell’ode (o della raccolta intera), la difficoltà di comprensione, oppure anche alla committenza, che avrà pure giocato un ruolo nella produzione di questo tipo di opere. Vediamo ora più da vicino il contenuto delle parti esegetiche, cominciando con il commento alla prima pericope della sezione conservata: ἀλλὰ Δωρίαν (v. 17) — ἐν δρόµοιϲι παρέχων (v. 21), che occupa col. I 3–15. ἀλλ᾽ ἀπ̣[ὸ τ]οῦ | [πα]ϲϲάλου λάµβανε τὴν Δ[ω]ρ̣ικὴν | [κιθ]άρα · Δωρικὴ ὅτι δωρ[ιϲτ]ὶ γράφει | [Αἰο]λ̣εὺϲ ὤν· [εἴ τ]ί ϲ̣ο̣ι ἡ τ̣[οῦ Φερε]νίκου | [χάρι]ϲ ἐκ τῆϲ Π[ίϲηϲ  γ]λ̣υκυ|[τάτ]αιϲ φροντ̣[ίϲι ]υ̣  | ]α̣ν̣ Ἀλφειο[ ] | ] Φερενικ[ ][] | ] χάριϲ ἡ νικ[ Ἀ]λ̣φε̣[ι|ὸϲ π]οταµ(ὸϲ) ἐν Ἤλιδ[ι  ]ων Ἦλιϲ | [] Ὀλυµπια· Φ[ερένικ]οϲ δὲ ἔδραµεν | [ὁ ἵπ]π(οϲ) ἀκέντριϲ̣τοϲ̣· ἔµφαϲιϲ γενναιό|[τητ](οϲ).

Vediamo poi il commento alla terza pericope pindarica della sezione conservata: λάµπει δέ οἱ (v. 23) — ἀποικίᾳ (v. 24), che occupa col. I 25–29. λά]µπε̣ι δὲ̣ [τοῦ Ἱέρον]ο̣[ϲ] τὸ κλέοϲ ἐν | [τῇ] τ̣οῦ Λυδο[ῦ Πέλοπ(οϲ) εὐάν]δρωι ἀποι|[κία]ι· τὴν Ἧλιν̣ [λέγ(ει), ἣν ὁ] Πέ̣λοψ πρὸ προι|[κὸ]ϲ̣ ἐκτήϲατο [λαβὼ]ν̣ ἀπ᾽ Οἰνοµάου | [πα]τ̣ρὸϲ τὴν Ἰππο[δά]µειαν.

Sono le due pericopi il cui commentario è meglio ricostruibile. Il commento procede chiaramente con una sistematica e puntuale parafrasi del dettato pindarico, intervallata da brevi note esplicative, che qui abbiamo evidenziato con la sottolineatura. La parafrasi segue il testo poetico nel modo più aderente possibile, talvolta conservandone singole parole poetiche, che vengono mantenute nella parafrasi anche quando possano risultare bisognose di una spiegazione. È il caso per esempio, nella prima pericope, dell’espressione Πίϲαϲ τε καὶ Φερενίκου χάριϲ, che indica la gloria prodotta dalla vittoria conquistata da Ferenico a Olimpia. Poco più avanti, infatti, l’identificazione geografica dell’Alfeo,

 10 Barrett/Obbink 2014, 117.

  Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201 dell’Elide e di Olimpia, è preceduta da una spiegazione di χάριϲ, parola che a quanto pare (la ricostruzione è plausibile) non era stata ‘tradotta’ nella parafrasi: la spiegazione è purtroppo mutilata, ma in χάριϲ ἡ νικ[ si scorge comunque l’idea che χάριϲ fosse connessa all’idea di vittoria. In questo senso gli editori suggeriscono nel commento χάριϲ ἡ νικ[ητήριοϲ ἡδονή, sulla base dello sch. ad loc. χάριϲ δὲ ἡ ἐπὶ τῇ νίκῃ ἡδονή. Dunque il commentatore prima parafrasa tenendosi strettissimo al testo (ἡ τοῦ Φερενίκου χάριϲ ἐκ τῆϲ Πίϲηϲ), per poi spiegare cosa vuole dire qui χάριϲ e aggiungere elementi di identificazione geograficotopografica del luogo della vittoria (Alfeo, Pisa, Elide, Olimpia). Nel successivo brano di parafrasi il pindarico δέµαϲ ἀκέντητον è ‘tradotto’ con ὁ ἵπποϲ ἀκέντριϲτοϲ, cui si aggiunge ἔµφαϲιϲ γενναιότητοϲ: sintentica notazione con la quale il commentatore rileva che l’aggettivo secondo cui Ferenico si slancia «senza sprone» verso la vittoria enfatizza la speciale nobiltà e il particolare valore del famoso «puledro dalla fulva criniera veloce come la tempesta», come lo definisce Bacchilide, 5.38–40, epinicio dedicato alla stessa occasione della Olimpica 1. Il medesimo procedimento esegetico si riscontra bene per la terza pericope di testo pindarico, molto breve: la parafrasi è seguita da una nota intesa a spiegare cosa intende il poeta per ἀποικία. Si tratta dell’Elide, appunto la regione di Olimpia, che Pelope ottenne avendo preso in sposa Ippodamia dal padre Enomao: un riferimento sintetico alla ben nota vicenda della gara vinta da Pelope contro Enomao, mito eziologico degli agoni olimpici. Questo modo di procedere caratterizza tutto il commentario. La stessa commistione di parafrasi con occasionali aggiunte esegetiche si riscontra per es. nella col. I per la seconda pericope, vv. 22–23, anche se i righi di commento sono meno ricostruibili (per es. ai rr. 19–23, vedi infra) e ancora nella col. II, commento ai vv. 30–32 (vedi ἀλήθειαν r. 8; per [πα]ραδοχῇ vedi più avanti); commento al v. 35 (r. 19), al v. 36 (rr. 24–25), ai vv. 37–39 (rr. 31 sgg.). Qualche volta troviamo soltanto la parafrasi: per i vv. 25–26 nella col. I, rr. 31–32; per il v. 36, prima parte, nella col. II, r. 20. Gli editori, dopo aver descritto il procedimento consistente in parafrasi con occasionali inserimenti di note esegetiche «of varying relevance», aggiungono:11 «More learned material makes a brief appearance: at ii 33–5 we find a reference and comparison to Pindar’s ὑπορχήµατα; at i 19–23 there is a brief historical note;12 at i 14 a rhetorical term (ἔµφαϲιϲ) is invoked in the explanation». Al «more learned material» qui elencato aggiungerei che W.S. Barrett, autore di una prima edizione di questo papiro, aveva ipotizzato la presenza di due citazioni da  11 Barrett/Obbink 2014, 117 12 Il contenuto è problematico, vedi p. 122, comm. ad loc.

Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201  

tragedie non identificate a col. II, rr. 7–8 e 12–13: frammenti recepiti da R. Kannicht in TrGF II, adespota 453a e 453b.13 Un confronto interno all’opera pindarica non è banale. W.B. Henry14 congettura un riferimento del commentatore ‘in avanti’, entro il libro delle Olimpiche, a col. I, r. 27 (ipotizzato riferimento a Ol. 9.6–10: [ὡϲ ἐρεῖ (scil. ὁ Πίνδαροϲ)] e a col. II, rr. 12–13 (ipotizzato riferimento a Ol. 10.53–55):15 oltre al fatto che questi due sono del tutto ipotetici, certamente il riferimento a un diverso genere poetico (gli Iporchemi) è più significativo rispetto a un riferimento ad altri epinici della stessa raccolta, ed è sicuro. Credo che sarebbe utile un’analisi sistematica dei riferimenti interni alla sua opera nei commentari a un autore per far luce su caratteristiche e interessi degli antichi esegeti. Abbiamo già detto sopra del «rhetorical term» ἔµφαϲιϲ a col. I, r. 14. Aggiungerei che a col. II, rr. 5–6, il termine [πα]ραδοχῇ (del tutto plausibilmente integrato) rimanda alla stessa sfera, come notano gli stessi editori:16 «Potentially a term of textual or literary criticism, either in the sense of ‘tradition’ (including oral or written textual paradosis), or of ‘acceptance’, i.e. of a mythological tradition … The latter sense has obvious relevance in the context of the present passage, especially in reference to the obbligation (or not) of the poet to preserve the tradition». Non c’è bisogno di sottolineare quanto sia importante il tema di allinearsi alla oppure discostarsi dalla tradizione mitica in un’ode come la Olimpica 1, con i problemi connessi alle modifiche apportate dal poeta al mito di Tantalo e Pelope: sembra che si facesse riferimento a questo nel commento ai vv. 36–39, come ipotizzano gli editori, possibilmente con un confronto con gli Iporchemi, forse per una diversa presentazione della vicenda.17 Da queste brevi osservazioni vorrei trarre una considerazione di ordine generale. Malgrado il carattere ipotetico di molte delle indicazioni estratte da un testo frammentario e conservato molto parzialmente, mi sembra si possa dire che bisogna guardarsi da un’abitudine purtroppo abbastanza diffusa e radicata, cioè quella di caratterizzare in modo riduttivo e giudicare con sufficienza il valore di testimonianza storico-culturale di un prodotto esegetico che non contiene materiale erudito ‘di alto livello’ secondo i nostri standard di giudizio. È facile trovare esempi di questo atteggiamento scorrendo le prime edizioni o anche le riprese di testi di questo genere. Si cercano le varianti testuali, la vasta

 13 Barrett/Obbink 2014, 118 alla fine dell’introduzione e p. 123, comm. ad locc. 14 Presso Barrett/Obbink 2014, 118 alla fine dell’introduzione. 15 Presso Barrett/Obbink 2014, 122 e 123, ad loc. 16 Barrett/Obbink 2014, 123, ad loc. 17 Barrett/Obbink 2014, 124, comm. ai rr. 23–25 e ai rr. 33–35 della col. II.

  Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201 erudizione e la dotta ermeneutica dei grandi filologi, soprattutto di età ellenistica, le osservazioni linguistiche, grammaticali, etimologiche, e se questi contenuti non compaiono in modo evidente e con sufficiente abbondanza, allora il prodotto esegetico è di basso valore, di assai ridotta importanza, di scarso significato. Per di più, questa prospettiva rischia di indurre a trascurare elementi non abbastanza appariscenti, ma presenti e tali da rendere il quadro esegetico più variegato di quanto appaia a prima vista.18 Ho cercato di mostrare che nel nostro caso non è esatto parlare di una semplice parafrasi, magari senza pretese e di scarso significato: mettere in evidenza e valorizzare gli altri aspetti del prodotto esegetico ci ha aiutato a capirlo un po’ meglio e soprattutto in modo più completo. Mi pare poi opportuno aggiungere che non ritengo affatto giusto dal punto di vista storico-culturale svalutare del tutto, magari con malcelata sufficienza, la parafrasi come procedimento esegetico (come se ‘tradurre’ e spiegare Pindaro sia così semplice e banale…): ricordiamo che, all’interno di un dato sistema culturale, la parafrasi è il più antico e basilare modo di traduzione, e quindi di interpretazione, che si usa definire ‘traduzione intralinguale’.19 Concludo affermando che anche a questo proposito credo sia necessario cambiare orientamento e atteggiamento metodologico, come da tempo invoco per gli studi sulla filologia ed erudizione antica nel suo complesso. Agli antichi filologi non dobbiamo chiedere principalmente o addirittura solamente quello che interessa a noi e quello che per noi è importante e significativo. Gli antichi filologi e le loro opere non sono importanti per quello che ci dicono sugli autori che commentano e interpretano (non è frequente in verità che ci diano informazioni importanti su questo piano, e non è produttivo andare a cercarle con il microscopio), bensì per quello che ci dicono su se stessi, sulle loro idee, i loro metodi, la loro cultura. Non sono interessanti per il numero di volte che hanno scelto la lezione o l’interpretazione giusta e corretta secondo noi e le nostre opinioni, bensì per quello che hanno detto, osservato, interpretato, indipendentemente se sia giusto o sbagliato dal nostro punto di vista. Gli antichi eruditi e filologi sono importanti anche o forse soprattutto per quello che ci dicono su se stessi (anche se sbagliano in modo che a noi può risultare persino ridicolo) oltre che per quello che ci dicono sull’autore che commentano e interpretano (che in verità non è così spesso decisivo per lo studioso moderno). Lo stesso principio

 18 Un esempio emblematico: in CLGP I 1.4, Berlin/Boston 20122, p. 277, il commento a Bacchilide P.Oxy. XXIII 2367 è definito da H. Maehler come un prodotto senza pretese, spiegazioni di parole e semplice parafrasi di un livello molto modesto, ma a Fr. 7, 2 compare l’avverbio ἀλληγορικῶϲ… 19 Montanari 1991; 1995c; 2003 (= questo volume, cap. 40).

Osservazioni sul Commentario a Pindaro P.Oxy. LXXIX 5201  

vale per i prodotti esegetici che troviamo nei papiri, studiamo ed editiamo, e spero di avere espresso con chiarezza cosa voglio dire. Non ha importanza se questi testi contengono informazioni di alto livello secondo i nostri standard e i nostri punti di vista, ha importanza quello che hanno scelto di fare, qualunque cosa sia, perché ci dà informazioni utili a comprendere un capitolo di storia culturale del mondo antico per quello che era, per quello che riteneva interessante, per quello che chiedeva e leggeva, e in tutti i suoi vari aspetti, siano esse di ‘alto’ oppure di ‘basso’ livello secondo la nostra filologia.20

 20 Riflessioni in questa direzione: Montanari 2015 (= questo volume, cap. 10); Montanari 2015a (= questo volume, cap. 9); Montana 2015, 60–70; Montanari 2016 (= questo volume, cap. 12).

 Citazioni di Pindaro negli scoli omerici Le citazioni di autori nella scoliografia antica sono un fenomeno abbondante e frequente, sul cui significato torneremo in conclusione. Gli scoli omerici lo documentano ampiamente, e si può osservare inoltre che i frammenti di commentari omerici conservati su papiro si incaricano di darci un’idea di quanto più numerose fossero le citazioni in fasi più antiche della formazione dei prodotti esegetici e di quanto materiale di questo tipo sia andato perso nella trasmissione dei materiali filologico-eruditi. Negli scoli omerici fra gli autori di rilievo Pindaro è citato con una certa frequenza per motivi molto diversi, che cercheremo di esemplificare utilizzando in particolare gli scoli all’Iliade. Naturalmente si tratta appunto di una esemplificazione, che ha semplicemente lo scopo di mostrare qualche tema che viene in causa e peraltro prospetta una ricerca eventualmente da proseguire: in primo luogo si può completare lo studio dei frammenti pindarici negli scoli omerici; poi si può continuare allargando il campo ad altri corpora scoliografici e in seguito anche ad altri testi eruditi, come i lessici e gli etimologici, e così via. Cominciamo con un caso molto semplice. In Iliade 8.365–369 si fa riferimento alla spedizione di Eracle agli Inferi per catturare il cane di Ade e lo sch. ex. ad 8.367–368 informa: οἶδε δὲ τὸν κύνα καὶ τὴν φύσιν αὐτοῦ. Πίνδαρος (fr. 249b, p. 75 Sn.–M.) δὲ ἑκατόν, Ἡσίοδος (cf. Theog. 312) δὲ πεντήκοντα ἔχειν αὐτὸν κεφαλάς φασιν.

Null’altro di più, almeno per quanto è rimasto, di un particolare erudito relativo al numero delle teste del cane di Ade, che — come notato poco più avanti nello sch. di Aristonico a Il. 8.368 — solo più tardi avrà il nome di Cerbero (a partire da Esiodo, Theog. 311). Disaccordi fra gli autori antichi su particolari di questo genere sono abbastanza normali e spesso vengono notati. Vediamo ora un caso in cui Pindaro è ricordato per una variante del mito: anche questa è una tipologia di osservazione erudita che certamente non sorprende nei testi della filologia antica, ma è fonte di dati ed elementi non di rado significativi e interessanti. In Il. 10.435 la prima menzione di Reso nel poema è commentata come segue: Sch. ex. ad Il. 10.435 ἐν δέ σφιν Ῥῆσος: Ῥῆσος Στρυμόνος τοῦ ποταμοῦ τῆς Θρᾴκης υἱὸς καὶ Εὐτέρπης Μούσης. ἱστορεῖ δὲ Πίνδαρος (fr. 262 Sn.–M.) ὅτι μίαν ἡμέραν πολεμήσας πρὸς Ἕλληνας μέγιστα αὐτοῖς ἐνεδείξατο κακά, κατὰ δὲ πρόνοιαν Ἥρας καὶ Ἀθηνᾶς ἀναστάντες οἱ περὶ Διομήδεα ἀναιροῦσιν αὐτόν.

https://doi.org/10.1515/9783110772371-050

Citazioni di Pindaro negli scoli omerici  

Sch. D ad Il. 10.435 ἐν δέ σφι Ῥῆσος βασιλεύς: Ῥῆσος γένει μὲν Θρᾷξ ἦν, υἱὸς Στρυμόνος, τοῦ αὐτόθι ποταμοῦ, καὶ Εὐτέρπης, μιᾶς τῶν Μουσῶν. διάφορος δὲ τῶν καθ᾿ αὑτὸν γενόμενος ἐν πολεμικοῖς ἔργοις ἐπῆλθε τοῖς Ἕλλησιν, ὅπως Τρωσὶ συμμαχήσῃ, καὶ συμβαλὼν πολλοὺς τῶν Ἑλλήνων ἀπέκτεινεν. δείσασα δὲ Ἥρα περὶ τῶν Ἑλλήνων, Ἀθηνᾶν ἐπὶ τὴν τούτου διαφθορὰν πέμπει. κατελθοῦσα δὲ ἡ θεὸς Ὀδυσσέα τε καὶ Διομήδην ἐπὶ τὴν κατασκοπὴν ἐποίησεν προελθεῖν. ἐπιστάντες δὲ ἐκεῖνοι κοιμωμένῳ Ῥήσῳ αὐτόν τε καὶ τοὺς ἑταίρους κτείνουσιν, ὡς ἱστορεῖ Πίνδαρος (fr. 262 Sn.– M.). ἔνιοι δὲ λέγουσιν κτλ.

Qui Pindaro è chiamato in causa per una variante mitica nella vicenda di Reso: in sintesi, Reso arriva a Troia, compie gesta importanti che danneggiano cospicuamente gli Achei, per cui Era e Atena reagiscono suscitando la spedizione di Odisseo e Diomede, che conduce all’uccisione di Reso e dei Traci. Poi solo nello sch. D si trova un’altra versione ancora, introdotta da ἔνιοι δὲ λέγουσιν e del tutto adespota, della quale non c’è traccia nello sch. ex., che dunque sembra avere perduto una parte del materiale erudito conservato dall’altro scolio, più ricco. Lo sch. D presenta la versione pindarica in una forma un po’ più estesa e con qualche particolare in più: l’edizione teubneriana Maehler 1989 riporta come frammento pindarico solo il primo scolio, ma questo è molto imprudente, perché non abbiamo alcuna certezza su quale delle due versioni corrisponda meglio al testo perduto di Pindaro.1 Un problema da lasciare agli studiosi di Pindaro, ma certo non privo di significato. Ancora uno sch. ex. omerico cita un verso della Pitica 3 di Pindaro in parallelo con Il. 13.20. Il critico antico sostiene che Pindaro qui si sarebbe ‘ispirato’ all’Iliade e aggiunge una serie di considerazioni basate sul confronto fra i due passi, per chiudere con un problema testuale pindarico.2 Posidone si slancia giù dall’Olimpo per andare in soccorso degli Achei e τρὶς μὲν ὀρέξατ᾿ ἰών, τὸ δὲ τέτρατον ἵκετο τέκμωρ. (Posidone) fece tre passi, al quarto raggiunse la meta. sch. ex. ad Il. 13.20 τρὶς μὲν ὀρέξατ᾿ ‹ἰών›: ὑπερφυῶς τοσαύτην θάλασσαν καὶ ἔθνη τρισὶ βήμασι παρῆλθεν. ὑπερβαλέσθαι δὲ τοῦτο θελήσας Πίνδαρος εἰς ὑπόνοιαν ἧκεν ψεύδους· «βάματι δ᾿ ἐν πρώτῳ παῖδ᾿ ἐκ νεκροῦ / ἅρπασε» (Pyth. 3.43–44).

Pind. Pyth. 3.43–44 βάματι δ᾿ ἐν πρώτῳ κιχὼν παῖδ᾿ ἐκ νεκροῦ / ἅρπασε

 1 Cfr. Hainsworth 1993, 151: egli attribuisce entrambi gli scoli a Aristonico, un errore palese. 2 Del problema testuale e metrico ho trattato in Montanari 1993b, 461–464 (= questo volume, cap. 50).

  Citazioni di Pindaro negli scoli omerici (Apollo) raggiunto al primo passo il bimbo lo rapì dal cadavere sch. Pyth. 3.75 βάματι δ᾿ ἐν πρώτῳ: ἐὰν μὲν γράφηται ἐν πρώτῳ, ὡς νῦν γέγραπται, ἔσται ἑνὶ βήματι ὁ Ἀπόλλων ἀπὸ τῆς Πυθῶνος ἐληλυθὼς εἰς τὴν Θεσσαλίαν· ἐὰν δέ, ὡς ἔν τισι, βάματι δ᾿ ἐν τριτάτῳ, ἔσται παραφράζων τὸ «τρὶς μὲν ὀρέξατ᾿ ἰών» (Il. 13.20). ἔστι δὲ σεμνότερον καὶ πρέπον τῷ θεῷ τὸ ἅμα τῷ πρώτῳ βήματι ῥύσασθαι τὸν παῖδα, ἔστι δὲ καὶ πρὸς τὴν ἀντίστροφον σύμφωνον. ὁ μέντοι Ἀρίσταρχος γράφει τριτάτῳ.

Il passo di Pindaro si riferisce alla nascita di Asclepio da Coronide, che portava in seno il seme di Apollo e quello dell’amore clandestino con il mortale Ischi. Coronide fu punita per la sua colpa, ma Apollo volle salvare il suo figlio e, raggiungendolo con un solo passo, lo rapì dal cadavere di Coronide che stava bruciando sul rogo. La base del discorso critico sembra chiaramente un confronto fra il verso omerico e il verso pindarico, fondato sul confronto fra la possanza di Posidone da una parte e di Apollo dall’altra: le due divinità si caratterizzano per la sovrumana capacità di percorrere un inusitato tratto di cammino in tre o un solo passo. Lo scolio omerico, dopo avere rilevato che superare una simile distanza in tre passi da parte di Posidone è ὑπερφυῶς, introduce una considerazione piuttosto sorprendente: Pindaro, volendo superare l’immagine presentata da Omero e parlando per Apollo di un solo passo per raggiungere e salvare il figlio, εἰς ὑπόνοιαν ἧκεν ψεύδους: «cadde nel sospetto di falso, di menzogna» (?). Vuol dire che, ispirandosi a Omero ma parlando di un solo passo anziché di tre, Pindaro ha per così dire esagerato, attribuendo ad Apollo una cosa così inverosimile da ingenerare il sospetto che non sia vero? Lo scolio pindarico, dal canto suo, prima sostiene che è σεμνότερον καὶ πρέπον τῷ θεῷ superare la distanza in un solo passo, ma poi presenta la possibilità che nel passo pindarico non si dovesse leggere ἐν πρώτῳ bensì ἐν τριτάτῳ, lezione peraltro scelta da Aristarco (verosimilmente per imitazione omerica). C’era evidentemente una discussione sul rapporto fra i due passi poetici e in particolare su quello pindarico, per il quale da una parte si può sostenere che un solo passo è più efficace per descrivere la sovrumana possanza del dio, dall’altra che questo è eccessivo e quindi sospetto di falsità, per cui meglio leggere ἐν τριτάτω, dunque in tre passi come fa Posidone in Omero.3 Più complesso e problematico è il caso di cui ci occupiamo ora, basato anch’esso su un parallelo fra Omero e Pindaro. Nel catalogo delle navi del II canto dell’Iliade, nel passo relativo al contingente dei Rodiesi, si trova un riferimento all’insediamento a Rodi dei discendenti di Eracle, guidati dall’Eraclide Tlepolemo, esiliato da Argo in seguito all’assassinio di Licimnio. Alla fine del passo si legge (vv. 667–670):  3 Cfr. Gentili/Bernadini/Cingano/Giannini 1995, comm. ad loc. (p. 414).

Citazioni di Pindaro negli scoli omerici  

αὐτὰρ ὁ γ᾿ ἐς Ῥόδον ἷξεν ἀλώμενος ἄλγεα πάσχων· τριχθὰ δὲ ᾤκηθεν καταφυλαδόν, ἠδὲ φίληθεν ἐκ Διός, ὅς τε θεοῖσι καὶ ἀνθρώποισιν ἀνάσσει, καί σφιν θεσπέσιον πλοῦτον κατέχευε Κρονίων.

667

670

(Tlepolemo) Giunse a Rodi peregrinando e soffrendo dolori 667 e lì si stanziarono in tre sedi, divisi per tribù, cari a Zeus che regna su uomini e dèi: su di loro immensa ricchezza versò il Cronide. 670

Il passo pose alcuni problemi alla filologia antica (come si legge negli scoli) e diversi altri ne pone agli studiosi moderni (la colonizzazione argiva di Rodi, la divisione in tre sedi),4 ma qui ci interessa l’osservazione conservata dallo sch. di Aristonico al v. 670, con un rimando a Pindaro.5 Ariston. ad Il. 2.670 καί σφιν θεσπέσιον πλοῦτον ‹κατέχευε Κρονίων›: ὅτι Πίνδαρος (Ol. 7.34) κυρίως δέδεκται χρυσὸν ὗσαι τὸν Δία, Ὁμήρου μεταφορᾷ κεχρημένου διὰ τοῦ κατέχευε πρὸς ἔμφασιν τοῦ πλούτου.

Lo scolio appartiene a un tipo ben noto e viene dall’opera di Aristonico sui σημεῖα di Aristarco: ὅτι, scil. τὸ σημεῖον ὅτι, qui certamente una diplè, che segnalava un luogo sul quale Aristarco aveva qualche commento da fare. Possiamo dunque dire con fiducia che il contenuto è aristarcheo. Lo stesso tema, in modo un po’ diverso, è trattato nello sch. D ad Il. 2.670: καί σφιν θεσπέσιον πλοῦτον: καὶ αὐτοῖς πολὺν καὶ ἀθρόον κατέπεμψε πλοῦτον ὁ Ζεύς, καὶ πάνυ πλουσίους αὐτοὺς ἐποίησεν. ἱστορεῖται γὰρ ὅτι τοῖς Ῥοδίοις χρυσὸν ἔβρεξεν ὁ Ζεύς, ἐπεὶ πρῶτοι γεννωμένῃ τῇ Ἀθηνᾷ ἔθυσαν.

In modo un po’ diverso e anche, pur nella stringatezza dell’espressione, un po’ più ricco: da dove venga l’aggiunta del riferimento a Atena si capisce immediatamente leggendo il passo pindarico, che era citato nello sch. di Aristonico e qui è alluso con ἱστορεῖται (le citazioni esplicite e precise dei passi chiamati a confronto sono uno dei primi elementi che vanno perduti nella trasmissione delle opere dell’erudizione antica). Il passo della Olimpica 7 chiamato a confronto sono i vv. 32–37, con la ripresa nei vv. 49–51:

 4 Le tre città di Rodi — Lindo, Ialiso e Camiro — sono citate al v. 656. 5 Erbse 1969–1988, vol. I, 321: in apparato sono indicati i paralleli.

  Citazioni di Pindaro negli scoli omerici τῷ μὲν ὁ χρυσοκόμας εὐώδεος ἐξ ἀδύτου ναῶν πλόον 32 εἶπε Λερναίας ἀπ᾿ ἀκτᾶς εὐθὺν ἐς ἀμφιθάλασσον νομόν, ἔνθα ποτὲ βρέχε θεῶν βασιλεὺς ὁ μέγας χρυσέαις νιφάδεσσι πόλιν, ἁνίχ᾿ Ἁφαίστου τέχναισιν 35 χαλκελάτῳ πελέκει πατέρος Ἀθαναία κορυφὰν κατ᾿ ἄκραν ἀνορούσαισ᾿ ἀλάλαξεν ὑπερμάκει βοᾷ. … κείνοις ὁ μὲν ξανθὰν ἀγαγὼν νεφέλαν {Ζεύς} πολὺν ὗσε χρυσόν· αὐτὰ δέ σφισιν ὤπασε τέχναν 50 πᾶσαν ἐπιχθονίων Γλαυκῶπις ἀριστοπόνοις χερσὶ κρατεῖν. A lui (scil. Tlepolemo) il dio dall’aurea chioma dal recesso odoroso prescrisse 32 la giusta rotta delle navi dal lido di Lerna al pascolo cinto dal mare, dove un giorno il grande re degli dèi faceva piovere sulla città fiocchi di neve dorata quando per arte di Efesto 35 sotto il colpo della scure di bronzo Atena balzata fuori dalla testa del padre lanciò l’urlo fortissimo di guerra. … Addensata su di essi una nuvola bionda piovve molto oro e la stessa Glaucopide concesse loro 50 di superare con mani industriose in ogni arte i mortali.

Pindaro sta raccontando anche lui la vicenda della colonizzazione argiva di Rodi da parte di Tlepolemo che, in seguito all’assassinio di Licimnio, per consiglio dell’oracolo di Delfi se ne va da Argo (il lido di Lerna) per insediarsi a Rodi (il pascolo cinto dal mare): è la stessa vicenda cui si riferisce il passo omerico da cui siamo partiti. Il parallelo con Pindaro citato dallo scolio omerico trova riscontro negli scoli pindarici a questi versi. Sch. Pind. Ol. 7.63 = 34 a. βρέχε θεῶν βασιλεύς: τοῦτο ἐκ τοῦ Ὁμηρικοῦ στίχου τοῦ ἀθετουμένου μετενήνεκται (Il. 2.670)· «καί σφιν θεσπέσιον πλοῦτον κατέχευε Κρονίων». ὅτι ἄρα νεφέλην ἐπιστήσας τοῖς Ῥοδίοις χρυσὸν ἔχευε. ἢ ὅτι πολὺν χύδην πλοῦτον αὐτοῖς κατεσκεύασεν, ὁμοίως τῷ (Od. 15.426)· «κούρη δ᾿ εἴμ᾿ Ἀρύβαντος ἐγὼ ῥυδὸν ἀφνειοῖο». b. βρέχε θεῶν βασιλεὺς ὁ μέγας: τοῦτο ἐκ τῶν Ὁμήρου ὁρμηθεὶς λέγει, ὅτι Ζεὺς χρυσὸν ἔβρεξε τοῖς Ῥοδίοις διὰ τὸ τὸν ποιητὴν εἰρηκέναι· (Il. 2.670) «καί σφιν θεσπέσιον πλοῦτον κατέχευε Κρονίων»· οὐκ εἰρηκότος Ὁμήρου ὅτι ἔβρεξε χρυσόν, ἀλλὰ περιεποίησε χύδην αὐτοῖς τὸν πλοῦτον.

Sgombriamo intanto il campo da un problema che tocca solo tangenzialmente il nostro discorso. Il riferimento a un verso omerico espunto, che leggiamo

Citazioni di Pindaro negli scoli omerici  

all’inizio dello sch. a, trova riscontro in due scoli di Aristonico a Il. 2.668 e 669, dove sembra riguardare il v. 669 (ἀθετεῖται), mentre qui a quanto pare si riferisce al v. 670. La questione è dubbia e le motivazioni non sono chiare: al momento possiamo lasciarla da parte.6 Si tratta comunque di un elemento che, per quanto incerto, va nel senso di una conferma del fatto che questi scoli a Omero e a Pindaro sono collegati e derivano da una fonte comune, cioè Aristarco.7 La base del ragionamento e delle osservazioni del critico antico è senza dubbio l’evidenziazione e l’analisi del parallelo fra il passo di Iliade 2 e quello della Olimpica 7 di Pindaro. I due luoghi parlano della stessa vicenda, cioè l’arrivo dell’Eraclide Tlepolemo a Rodi da Argo e il suo stanziarsi sull’isola con i suoi seguaci, caratterizzato da un elemento importante: la grande ricchezza che Zeus concesse ai Rodiesi (Omero), sotto forma di vera e propria pioggia di neve d’oro (Pindaro). La prima cosa che ha attirato la mia attenzione sono state le parole Ὁμήρου μεταφορᾷ κεχρημένου dello sch. iliadico, che racchiudono un’interessante notazione di carattere stilistico-espressivo: Aristarco (dando per acquisito che sia stato lui) osservava che Pindaro intende in senso proprio che Zeus fece piovere neve d’oro sui Rodiesi, mentre Omero usa una metafora con l’espressione θεσπέσιον πλοῦτον κατέχευε, cioè «riversò» su di loro una ricchezza immensa, miracolosa, per enfatizzare l’idea di ricchezza: πρὸς ἔμφασιν τοῦ πλούτου: a μεταφορᾷ si contrappone κυρίως. In Pindaro leggiamo infatti che Zeus βρέχε ... χρυσέαις νιφάδεσσι πόλιν (v. 34), faceva piovere sulla città fiocchi di neve d’oro, o bagnava la città con fiocchi di neve dorata, e poi κείνοις ὁ μὲν ξανθὰν ἀγαγὼν νεφέλαν {Ζεύς} πολὺν ὗσε χρυσόν (v. 50), addensata su di essi una nuvola bionda fece piovere molto oro. Al «versare molta ricchezza» si contrappone «far piovere oro, neve d’oro», con una identificazione concreta e materiale del fenomeno specifico utilizzato da Zeus per concedere grande ricchezza ai Rodiesi e un’enfasi sul carattere divino dell’episodio. Anche gli sch. pindarici chiariscono la differenza: nello sch. a abbiamo ὅτι ἄρα νεφέλην ἐπιστήσας τοῖς Ῥοδίοις χρυσὸν ἔχευε. ἢ ὅτι πολὺν χύδην πλοῦτον αὐτοῖς κατεσκεύασεν, cioè «stesa una nube versò oro su di loro» oppure «fornì loro ricchezza in abbondanza». E anche lo sch. b, riconosciuta l’ispirazione omerica di Pindaro, rileva che Omero non parla propriamente di far piovere specificamente oro bensì di procurare molta ricchezza. Credo non ci siano dubbi che questi scoli omerici e pindarici risalgano in ultima analisi alla stessa fonte e che questa fonte sia la filologia alessandrina, probabilmente Aristarco stesso.  6 Cfr. però Lehrs 1882, 186 «hoc videmus non Aristarcheum esse»: giudizio limitato al particolare della motivazione dell’atetesi. 7 Erbse 1969–1988, vol. I, 320.

  Citazioni di Pindaro negli scoli omerici La storia di Tlepolemo ha chiaramente il suo fondamento nel fatto di spiegare le origini della prosperità di Rodi e dei Rodiesi dopo la colonizzazione argiva. Lo ha ben visto Strabone 14.2.20. ἔοικε δὲ καὶ ὁ ποιητὴς μαρτυρεῖν τὴν ἐκ παλαιοῦ παροῦσαν τοῖς Ῥοδίοις εὐδαιμονίαν εὐθὺς ἀπὸ τῆς πρώτης κτίσεως τῶν τριῶν πόλεων· τριχθὰ δὲ ᾤκηθεν καταφυλαδόν, ἠδ᾿ ἐφίληθεν ἐκ Διός, ὅστε θεοῖσι καὶ ἀνθρώποισιν ἀνάσσει, καί σφιν θεσπέσιον πλοῦτον κατέχευε Κρονίων (Il. 2.668–670). οἱ δ᾿ εἰς μῦθον ἀνήγαγον τὸ ἔπος καὶ χρυσὸν ὑσθῆναί φασιν ἐν τῇ νήσῳ κατὰ τὴν Ἀθηνᾶς γένεσιν ἐκ τῆς κεφαλῆς τοῦ Διός, ὡς εἴρηκε Πίνδαρος. A quanto pare, anche il poeta (scil. Omero) testimonia la prosperità dei Rodiesi fin dai tempi antichi, subito a partire dalla prima fondazione delle tre città: e lì si stanziarono in tre sedi, tribù per tribù, cari a Zeus che regna su uomini e dèi: su di loro immensa ricchezza versò il Cronide. Altri invece riconducono questi versi a un mito e dicono che dell’oro piovve nell’isola al momento della nascita di Atena dalla testa di Zeus, come ha detto Pindaro.

Strabone mette in luce il fatto che la vicenda spiega e motiva la prosperità di Rodi, riconducendola al favore di Zeus: e per questo basta il passo omerico. Poi però aggiunge la notazione di una differenza significativa fra il racconto omerico e quello pindarico, quella che abbiamo visto rilevare solo nello sch. D ad Il. 2.670: la pioggia di neve d’oro su Rodi, che ne realizzò la prosperità, fu inviata da Zeus in occasione della celebre nascita di Atena dalla testa di Zeus. Che Strabone dipenda dalla filologia alessandrina mi pare del tutto probabile: nel suo resoconto però è omesso un ulteriore elemento, che troviamo appunto nell’appena ricordato sch. D ad Il. 2.670, cioè che i Rodiesi «meritarono» il favore di Zeus ἐπεὶ πρῶτοι γεννωμένῃ τῇ Ἀθηνᾷ ἔθυσαν, cioè perché furono i primi a fare un sacrificio a Atena quando nacque. È un altro segmento del mito: Elio ricorda agli Eliadi che Atena si sarebbe stabilita nel luogo in cui le sarebbe stato offerto il primo sacrificio; gli Eliadi si preparano a farlo alla nascita della dea, ma al momento buono dimenticano il fuoco e fanno quindi un sacrificio ἄπυρος: malgrado questo, Zeus regala loro una pioggia d’oro e Atena ogni arte (Pind. Ol. 7.39–49). Il passo omerico di Il. 2 e quello pindarico della Ol. 7 sono stati molto studiati per i diversi problemi che pongono e per i loro significati.8 La nostra prospettiva

 8 La bibliografia è molto ricca: cfr. Cantilena 1987 e 1990, Gentili/Catenacci/Giannini/Lomiento 2013, con i riferimenti essenziali.

Citazioni di Pindaro negli scoli omerici  

in questa sede è del tutto diversa. Cercando di mettere insieme una serie di elementi, trasmessi da fonti diverse ma riconducibili — mi sembra — ad un discorso unitario, abbiamo ricostruito un ragionamento critico articolato e complesso, che ha preso le mosse dall’osservazione del fatto che Pindaro — secondo il critico antico — si è ispirato a Omero per il suo racconto su Tlepolemo e la colonizzazione di Rodi come eziologia della prosperità dell’isola e del favore di Zeus e di Atena: racconto che ha modificato e adattato da par suo secondo la sua poetica (il ruolo primario del divino) e in particolare alla specifica occasione della celebrazione di un vincitore rodiese (l’Olimpica 7 è dedicata a Diagora di Rodi). A noi non interessa qui il possibile riferimento ai Dori nel passo iliadico né se la colonizzazione di Tlepolemo sia di fatto pre-dorica, oppure il senso miticoreligioso e antropologico della pioggia dorata di Pindaro, il momento esatto in cui essa è avvenuta, se si possa mettere in dubbio che Pindaro si sia ispirato a Omero oppure se i due racconti derivino da tradizioni diverse o altri temi ancora a cui gli studiosi hanno dedicato le loro attenzioni. Quello che ci interessa è il ruolo della filologia antica (in questo caso quella alessandrina, verosimilmente Aristarco stesso) e quello che una testimonianza di questo genere ci dice su di essa. Ritengo che la differenza di prospettiva risulti ancor più significativa ed evidente proprio perché stiamo parlando di versi che hanno sollevato molto impegno e molte interpretazioni di per sé, cioè per i contenuti specifici dei loro stessi autori. Per la filologia antica è stato essenziale rilevare e sottolineare il rapporto fra i due passi e la dipendenza di Pindaro da Omero, e analizzarne le differenze, sia dal punto di vista stilistico (μεταφορᾷ vs. κυρίως) sia da quello del contenuto e della trattazione del mito. Il resto è un problema degli studi moderni: chiariamo bene che non c’è una gerarchia di importanza fra le due prospettive e che sono importanti due aspetti: 1) tenerli distinti, 2) non subordinare il secondo al primo, liquidando la critica filologica antica come irrilevante o sbagliata perché non risolve i problemi della ricerca moderna. Ho richiamato altrove questi concetti per me fondamentali ma non ancora sufficientemente interiorizzati e ‘digeriti’ dagli studiosi.9 Mentre l’utilità diretta della filologia antica per l’esegesi moderna è relativamente sporadica e in fondo casuale (ma certo non per questo negligibile e da lasciare da parte a cuor leggero), permanente, costante e duraturo per il valore della filologia antica è quello che essa ci dice su se stessa e sulla propria storia.10 L’erudizione antica non può  9 Cfr. Montanari 2016 (= questo volume, cap. 12) e 2018 (= questo volume, cap. 3); Perrone 2010. 10 La storia di una disciplina è essenziale per capire l’autoconsapevolezza e il progresso della medesima e per questo deve essere comunque tenuta nella massima considerazione anche nelle sue fasi più antiche, addirittura embrionali: lo ha ben sottolineato Pfeiffer 1968,

  Citazioni di Pindaro negli scoli omerici certo essere sminuita, considerandola principalmente come testimone di frammenti o di notizie su opere perdute e come fonte di informazioni per l’interpretazione delle opere oggetto di commento, ma i suoi prodotti sono almeno altrettanto importanti per quello che ci dicono a proposito di se stessi, dunque della storia della disciplina. La filologia e l’erudizione antica sono ormai accettate come uno degli aspetti indispensabili del panorama storicoculturale del mondo antico. Per questo è di primaria importanza che i testi filologico-eruditi siano interrogati per quello che ci dicono su se stessi, adottando il principio basilare che non ha importanza se quanto ci dicono è giusto o sbagliato, se le loro interpretazioni sono buone o cattive dal nostro punto di vista e secondo i nostri metodi moderni. Che i grammatici antichi a proposito di Omero o Eschilo scelgano una lezione o una interpretazione sbagliata e da respingere secondo i nostri metodi e i nostri principi è del tutto secondario rispetto alla comprensione dei metodi e dei fondamenti culturali che determinano le loro scelte e i loro orientamenti. Difendere la ‘utilità’ dei testi dell’erudizione antica non significa affatto trovare che qualche volta hanno ragione e danno interpretazioni valide, significa capire cosa dicono e perché. Oggi la ricerca deve muoversi in questa prospettiva storico-culturale, nella quale rientrano la consapevolezza e il riconoscimento dell’importanza che riveste la presenza di citazioni di autori antichi nei materiali esegetico-eruditi che si sono conservati. Queste citazioni non vanno più considerate soltanto per le informazioni che ci danno su opere perdute o sull’eventuale interpretazione di un passo problematico. Bisogna di conseguenza acquisire e promuovere una generale percezione del fatto che le citazioni d’autore, i passi paralleli addotti dagli eruditi, devono essere visti e studiati come un aspetto caratteristico e importante nel quadro complessivo della filologia ed erudizione antica nel suo farsi, nel suo sviluppo e nel suo significato storico-culturale. La presenza di innumerevoli citazioni d’autore e la loro funzione nei testi di carattere esegetico-erudito devono essere studiate più a fondo e con molto maggiore attenzione rispetto alla visione ‘ancillare’ del passato: è un fenomeno di grande rilievo, di grande utilità per valutare il significato storico-culturale dalla filologia antica.

 VII = 1973, 31: «The history of classical scholarship, therefore, is classical scholarship in the making. And a book reconstructing its history under this aspect can claim to be regarded as an integral part of scholarship itself»; cfr. Montana 2012, IX).

 L’esegesi antica di Eschilo da Aristotele a Didimo Nel contesto della trattazione sulle hypotheseis della poesia drammatica realizzate da Aristofane di Bisanzio (la cui importanza era stata sottolineata già da U. von Wilamowitz),1 R. Pfeiffer scrive: “Recently discovered hypotheses have provided important new evidence for Aeschylus”.2 Poco prima egli aveva dichiarato che dell’opera di Aristofane sul testo di Eschilo “there is still no evidence”.3 In primo luogo è degno di nota che R. Pfeiffer dica “still”: questa piccola parola rivela da una parte la prudenza dello studioso esperto, che sa bene come un nuovo ritrovamento possa intervenire a sconvolgere alcuni contorni facilmente dati per acquisiti, dall’altra la convinzione che Aristofane di Bisanzio abbia davvero lavorato anche sul testo di Eschilo, benché le testimonianze non siano ancora emerse in modo significativo ed esplicito. In effetti oggi, quaranta anni dopo la History of Classical Scholarship di R. Pfeiffer,4 non abbiamo nuovi testimoni da aggiungere a quelli su cui egli poteva contare5 e dobbiamo rimanere attestati sulla dichiarazione che non ci sono ‘ancora’ prove specifiche e dirette di un lavoro di Aristofane di Bisanzio proprio sul testo di Eschilo, come ce ne sono invece per la sua critica testuale relativa a Euripide, dato che negli scoli si trova almeno qualche riferimento a varianti e segni critici;6 mentre per quanto riguarda Sofocle alcune testimonianze su papiro sono oggetto di discussione: in quattro luoghi di P.Oxy. 1174 (Ichneutae), in uno di P.Oxy. 1805 (Trachiniae) e in uno di P.Oxy. 2452 (Theseus?) è stata ipotizzata la presenza del suo nome in annotazioni marginali, ma lo scioglimento delle abbreviazioni rimane dubbio.7  1 von Wilamowitz-Moellendorf 1906, 146. 2 Pfeiffer 1968, 194. 3 Pfeiffer 1968, 192. Per Aristofane cf. Wartelle 1971, 143–161; von Wilamowitz-Moellendorf 1906, 145 scriveva: “Daß Aristophanes für die Tragiker dieselbe Bedeutung hat wie für die Lyriker ist nicht überliefert. Dennoch ist es ganz unzweifelhaft”. 4 Pfeiffer 1968. 5 L’incremento più importante risaliva al 1952, con un piccolo addendum nel 1954: Lobel 1952 e 1954. 6 Cf. schol. ad Eur. Or. 713, 1038, 1287; Hipp. 171; Tr. 47. 7 Pfeiffer 1968, 192 è piuttosto fiducioso che si tratti di Aristofane. In P.Oxy. 1174, Ichneutae, i passi in questione sono col. III 20 = v. 79, col. VI 5 e 8 = vv. 143 e 146, col. IX 6 = v. 221: cf. Hunt 1912, ad locc.; Maltese 1982, ad locc.; Diggle 1998, ad locc.; Radt 1999 [= TrGF 4], ad locc.; Razzetti 2002a, 337–345; McNamee 2007, 366–370, nr. 1473, ad locc. — Per P.Oxy. 1805, Trachiniae v. 744, cf. Grenfell/Hunt 1922, ad loc.; Lloyd-Jones/Wilson 1990, ad loc.; McNamee 2007, 364, nr. 1471, ad loc. —Per P.Oxy. 2452 (Theseus?), fr. 2 r. 16, cf. Turner 1962, ad loc.; TrGF 4 F 730e r. https://doi.org/10.1515/9783110772371-051

  L’esegesi antica di Eschilo da Aristotele a Didimo Vedremo tuttavia che sia per la tragedia in generale che per Eschilo in particolare possiamo fiduciosamente attribuire ad Aristofane un lavoro approfondito e già erede di una tradizione abbastanza rigogliosa. Sappiamo che i filologi alessandrini dedicarono più precocemente abbondanti cure alla commedia antica, ma abbiamo sufficienti indizi per dire che con Aristofane di Bisanzio anche la tragedia si era ormai decisamente collocata fra i loro principali oggetti di interesse e di studio. E se la documentazione non è cospicua (almeno fino ad ora, per seguire la prudenza di R. Pfeiffer), questo non obbliga a concluderne che si trattò solo di un lavoro sporadico e limitato. La selezione tradizionale intervenuta nella trasmissione del materiale di carattere esegetico-erudito prodotto nell’antichità ci ha consegnato una situazione troppo chiaramente squilibrata perché questo fattore non eserciti un peso determinante nell’immagine che risulterebbe tenendo conto solo di un aspetto quantitativo. La differenza fra i corpora scoliografici conservati è un dato evidente: mentre gli scoli alle commedie di Aristofane costituiscono uno dei più ricchi e importanti che abbiamo, assai meno consistenti sono quelli dei tre tragici, e fra questi il meno cospicuo è proprio quello ad Eschilo.8 La maggior ricchezza dei corpora scoliografici a Aristofane e Euripide, rispetto a quelli a Eschilo e Sofocle, è uno dei risultati della maggiore fortuna dei primi due autori fino all’età bizantina. Lo stesso vale per la quantità di copie delle opere degli autori di teatro che sono venute fuori dalle sabbie dell’Egitto. Anche accontentandoci di un dato molto elementare e non sottoposto ad analisi raffinate, possiamo vedere che una semplice lettura del repertorio in rete9 dei papiri pubblicati ci dà un totale dei frammenti papiracei conosciuti che va nello stesso senso: Eschilo 32, Sofocle 32, Euripide 162, Aristofane 57. Per la filologia antica su Eschilo la casualità dei ritrovamenti papiracei ha fornito una documentazione molto particolare, quella delle già citate hypotheseis, su cui giustamente R. Pfeiffer attirava l’attenzione nel passo ricordato all’inizio. Il progetto Commentaria et Lexica Graeca in Papyris reperta (CLGP) ha lo scopo di raccogliere le testimonianze di filologia ed erudizione antica restituite dai papiri, sottoponendole a un riesame approfondito e proponendole in nuove edizioni critiche, corredate di introduzioni e commenti. I due volumi finora usciti ci permettono di confrontare al più avanzato stadio di elaborazione

 16; McNamee 2007, 370, nr. 1479, ad loc. — Il nome di Aristofane compare una volta negli schol. sofoclei: ad Aj. 746 è citato per un proverbio incluso nella sua raccolta, fr. 359, Slater 1986. 8 Si veda il recente quadro in Dickey 2007, 18–42; Herington 1972; Smith 1976 e Smith 1982; Morocho Gayo 1989. 9 [= Mertens–Pack3].

L’esegesi antica di Eschilo da Aristotele a Didimo  

critica il materiale disponibile per Aristofane e per Eschilo10 e la differenza risulta cospicua anche a prima vista: 31 sono le testimonianze sicure per Aristofane (più 4 dubbie), 9 quelle sicure per Eschilo (più 2 dubbie). A un’osservazione più approfondita il divario appare ancor più significativo. Distribuiti nell’arco dei secoli dal I al VI d.C., per Aristofane abbiamo 6 frammenti di hypomnema e 18 copie con marginalia (16 sicure e 2 dubbie), cui si aggiungono 11 voci di lessico riferibili al poeta comico (9 sicure e 2 dubbie). Per Eschilo nessun frammento di hypomnema è conservato, abbiamo 4 frammenti sicuri di copie con note marginali (cui possiamo affiancare quattro frammenti con segni critici),11 un frammento di rotolo con hypotheseis e note marginali alle Aetn(ae)ae e una raccolta di hypotheseis a varie tragedie,12 nessuna voce di lessico sicura (una dubbia).13 Benché il materiale che riguarda Euripide e Sofocle non abbia ancora avuto la revisione critica e la riconsiderazione sistematica prevista dal progetto CLGP,14 possiamo addurre i dati complessivi disponibili allo stato attuale, sia pure limitandoci in questa sede a un confronto puramente quantitativo. Come per Eschilo, anche per Sofocle neppure un frammento di hypomnema è conservato, mentre abbiamo 7 esemplari con note marginali (e quattro con segni critici e varianti), tre papiri con hypotheseis (uno dei quali probabilmente comprendeva anche un bios)15 e due voci di lessico riconducibili a sue opere. Per Euripide abbiamo due hypomnemata, 20 frammenti con note marginali, 6 voci di lessico e ben 19 papiri con hypotheseis. Nell’introduzione alla sezione del CLGP dedicata ad Aristofane, F. Montana traccia un ben fondato e informato panorama complessivo dei resti papiracei di erudizione aristofanea e osserva che “la documentazione di note marginali e commentari è più che raddoppiata rispetto a quella disponibile appena tre decenni or sono”.16 Dunque per Aristofane negli ultimi decenni le cose sono andate ben diversamente da quanto accaduto per Eschilo, per il quale da oltre mezzo secolo non si registrano incrementi di documentazione, come abbiamo già detto.

 10 Eschilo in Bastianini/Haslam/Maehler/Montanari/Römer 2004 [= CLGP I 1.1]; Aristofane in Bastianini/Haslam/Maehler/Montanari/Römer 2006 [= CLGP I 1.4.]. 11 Cf. CLGP I 1.1, Aeschylus 15 e n. 15. 12 Il termine hypothesis è usato con una certa convenzionalità, come vedremo esaminando concretamente i testi. 13 CLGP I 1.1, 71–72, Aeschylus 11. 14 L’esperienza mostra che lo studio approfondito di questo genere di testimoni può portare a modifiche anche significative del quadro di partenza. 15 P.Vindob. G 29779: Luppe 1985a, con bibliografia; van Rossum-Steenbeek 1998, 34–35. 16 CLGP I 1.4, 3–12: 3: il riferimento è a McNamee 1977, che censiva 11 testimoni (3 hypomnemata e 8 copie con marginalia).

  L’esegesi antica di Eschilo da Aristotele a Didimo L’attenzione del mondo della ricerca per i pezzi recanti materiali esegeticoeruditi è molto cresciuta negli ultimi decenni, parallelamente con l’acquisizione di una più matura consapevolezza dell’importanza storico-culturale di questo aspetto della civiltà letteraria e della vita intellettuale nel mondo antico. Una simile considerazione indurrebbe a pensare che nelle maggiori collezioni papirologiche difficilmente frammenti riguardanti un autore importante e attraente come Eschilo possano essere lasciati da parte a favore di altri pezzi ritenuti di maggiore interesse. In base a questo ragionamento, dobbiamo affievolire la speranza di ritrovare novità importanti e rassegnarci per Eschilo a non arricchire in modo significativo la situazione che abbiamo oggi davanti? Mi pare possibile che le cose stiano effettivamente così almeno nel medio-breve periodo, in attesa che in un futuro più lontano il tesoro delle numerose scatole di pezzi non identificati cominci a essere disvelato e studiato: ma questo riguarderà le prossime generazioni, noi ora dobbiamo sforzarci di ricavare il più possibile da quello che abbiamo. Il quadro risultante dalle osservazioni prevalentemente quantitative sulla diffusione di copie delle opere dei tre tragici più Aristofane invita a tracciare una differenza abbastanza visibile fra Aristofane e Euripide come autori di maggior successo e diffusione rispetto a Eschilo e Sofocle. Ciò corrisponde in effetti alle aspettative suscitate da quanto sappiamo in generale sulla fortuna di questi autori a partire dall’età ellenistica e imperiale, ma un altro tipo di testimonianza può indurre ad aggiungere qualche considerazione ulteriore e ad arricchire il quadro. Mi riferisco ora proprio al variegato materiale che si usa etichettare come hypothesis e ai vari problemi che esso suscita. Prenderemo le mosse dall’opera sulle hypotheseis dei drammi di Euripide e Sofocle, attribuita al peripatetico Dicearco sulla base di una testimonianza di Sesto Empirico (Adv. Math. 3.3), di cui parleremo fra poco. Diversi frammenti papiracei (datati dal I al III sec. d.C.) contenenti hypotheseis dei drammi di questi due autori tragici, molto più numerosi per Euripide che per Sofocle, sono stati ritrovati e via via pubblicati da alcuni decenni. L’opinione che essi conservino resti dell’opera di Dicearco, formulata da C. Gallavotti, è stata poi ripresa e sostenuta da M.W. Haslam.17 Una diversa opinione ha avanzato J.S. Rusten: egli accetta l’identificazione dei frammenti conservati come parti dell’opera conosciuta in antico con il titolo Ὑποθέσεις τῶν Εὐριπίδου καὶ Σοφοκλέους μύθων, ma ne contesta la paternità dicearchea e pensa invece a una falsa attribuzione a Dicearco di una circolante raccolta di hypotheseis dei drammi di Euripide e Sofocle, ritiene cioè che si tratti in effetti di uno pseudo-Dicearco del I o II sec.  17 Gallavotti 1933, 188, sulla base di PSI 1286; Haslam 1975.

L’esegesi antica di Eschilo da Aristotele a Didimo  

d.C.18 Invece W. Luppe ritiene che l’opera citata da Sesto Empirico risalga effettivamente a Dicearco e che i frammenti restituiti dai papiri appartengano proprio ad essa.19 Il comportamento di D. Mirhady nell’edizione dei frammenti di Dicearco20 implica la convinzione che l’opera citata da Sesto Empirico sia autentica (egli dunque respinge, a mio parere giustamente, l’idea che si tratti di uno pseudoDicearco): la testimonianza di Sesto è accolta come fr. 112 e sotto lo stesso titolo sono raccolti anche i frr. 113–115, come segue (non riportiamo i testi del fr. 115).21 Argumenta (scil. Hypotheseis) 112 Sextus Empiricus, Adversus Mathematicos 3.322 πολλαχῶς μὲν καὶ ἄλλως ὑπόθεσις προσαγορεύεται, τὰ νῦν δὲ ἀπαρκέσει τριχῶς λέγεσθαι. καθ᾿ ἕνα μὲν τρόπον ἡ δραματικὴ περιπέτεια, καθὸ καὶ τραγικὴν καὶ κωμικὴν ὑπόθεσιν εἶναι λέγομεν καὶ Δικαιάρχου τινὰς ὑπόθεσεις τῶν Εὐριπίδου καὶ Σοφοκλέους μύθων, οὐκ ἄλλο τι καλοῦντες ὑπόθεσιν ἢ τὴν τοῦ δράματος περιπέτειαν. 113 Hypothesis Sophoclis Aiacis23 Δικαίαρχος δὲ Αἴαντος θάνατον ἐπιγράφει. ἐν δὲ ταῖς Διδασκαλίαις ψιλῶς Αἴας ἀναγέγραπται. 114 Hypothesis Rhesi 1.2624 τοῦτο τὸ δρᾶμα ἔνιοι νόθον ὑπενόησαν ὡς οὐκ ὂν Εὐριπίδου· τὸν γὰρ Σοφόκλειον μᾶλλον ὑποφαίνει χαρακτῆρα. ἐν μέντοι ταῖς Διδασκαλίαις ὡς γνήσιον ἀναγέγραπται. καὶ ἡ περὶ τὰ μετάρσια δὲ ἐν αὐτῷ πολυπραγμοσύνη τὸν Εὐριπίδην ὁμολογεῖ. πρόλογοι δὲ διττοί φέρονται. ὁ γοῦν Δικαίαρχος ἐκτιθεὶς τὴν ὑπόθεσιν τοῦ ‘Ρήσου γράφει κατά λέξιν οὕτως < Ῥῆσος, οὗ ἀρχή· > “νῦν εὐσέληνον φέγγος ἡ διφρήλατος” καὶ * * * καὶ ἐν ἐνίοις δὲ τῶν ἀντιγράφων ἕτερός τις φέρεται πρόλογος, πεζὸς πάνυ καὶ οὐ πρέπων Εὐριπίδῃ κτλ. 115A Hypothesis Euripidis Alcestis25 115B Papyrus Oxyrhinchus 245726

 18 Rusten 1982; cf. Kassel, 1985a, 53–59. 19 Sintesi con riferimenti e bibliografia in Luppe/Montanari 1992 e van Rossum-Steenbeek 1998, 1–52; cf. anche Pfeiffer 1968, 193–194; Bagordo 1998, 24–26; Luppe 2001; Luppe 2002. 20 Mirhady 2001. 21 Fr. 112 = Wehrli 1967, 78 = Bagordo 1998, F 8 = TrGF 4, T 153 = TrGF 5, T 212. — Fr. 113 = Wehrli, 79 = Bagordo 1998, F 9. — Fr. 114 = Wehrli 1967, 81 = Bagordo 1998, F 11 = TrGF 2, F*8 l = TrGF 5, F 60 ia (qui sopra è riportato il testo di D. Mirhady, ma si veda anche TrGF 5, F 60 ia), cf. Luppe 1990. — Sul fr. 115 cf. infra n. 36. 22 Mau 1961, 107, 11–17. 23 Pearson 1946, vol. 1, 11–13. 24 Zanetto 1993, 4, 3–12. 25 Diggle 1987, 33, 3–14.

  L’esegesi antica di Eschilo da Aristotele a Didimo I frr. 113 e 114 sono costituiti da passi delle hypotheseis dell’Aiace di Sofocle e del Reso (di cui si difende la genuinità euripidea), dove Dicearco è citato in relazione con le Didascalie di Aristotele, nel primo caso per una divergenza, nel secondo per una concordanza.27 In verità, mentre il fr. 114, riguardante il Reso, appare sicuramente riferibile alle Hypotheseis (cf. infra), qualche dubbio rimane invece per il fr. 113, dove si dice che Dicearco dava come titolo Αἴαντος θάνατον, mentre nelle Didascalie si trovava semplicemente Aἴας. Anche in una citazione conservata nella seconda hypothesis dell’Edipo re di Sofocle si tratta di un problema di titolo, questa volta appunto dell’Edipo re di Sofocle: è il fr. 101, che D. Mirhady raccoglie sotto una diversa rubrica rispetto al fr. 113. 101 Hypothesis secunda Sophoclis Oedipodis Tyranni28 ὁ Τύραννος Oἰδίπους ἐπὶ διακρίσει θατέρου ἐπιγέγραπται. Χαριέντως δὲ Τύραννον ἅπαντες αὐτὸν ἐπιγράφουσιν ὡς ἐξέχοντα πάσης τῆς Σοφοκλέους ποιήσεως, καίπερ ἡττηθέντα ὑπὸ Φιλοκλέους, ὥς φησι Δικαίαρχος. εἰσὶ δὲ καὶ Πρότερον, οὐ Τύραννον αὐτὸν ἐπιγράφοντες, διὰ τοὺς χρόνους τῶν διδασκαλιῶν καὶ διὰ τὰ πράγματα.29

Qui possiamo intanto osservare che il riconoscimento dell’eccellenza dell’Edipo re riprende certamente l’idea espressa a questo proposito nella Poetica di Aristotele. Dicearco aveva scritto anche un trattato sugli agoni dionisiaci, nel quale si trovavano senz’altro notizie sulle tragedie rappresentate.30 Mi pare possibile che le notizie dei frr. 101 e 113 provengano dalla stessa opera, che si tratti delle Hypotheseis oppure del Περὶ Διονυσιακῶν ἀγώνων, e che quindi possano essere collocate insieme sotto l’una o l’altra rubrica:31 peraltro, non è certo da escludere che fra i due scritti ci fossero sovrapposizioni e osmosi di informazioni e materiali e che un dato potesse trovarsi in entrambi o anche altrove.32 Al trattato sugli agoni dionisiaci può risalire la notizia, riportata dalla Vita di Eschilo, secondo cui Dicearco sosteneva che il terzo attore sarebbe stato introdotto da

 26 Luppe 1982, 14. 27 Gigon 1987, fr. 419 e fr. 428. 28 Pearson 1946, 109, 20–24. 29 Fr. 101 = Wehrli 1967, 80 = Bagordo 1998, F 10 = TrGF 4, T 39. 30 Fr. 99 = Wehrli 1967, 75 = Bagordo 1998, F 3; cf. Luppe 1990, 331–332. 31 In Bagordo 1998 sono poste di seguito come F 9 e F 10. 32 Per esempio, la hypothesis della Medea di Euripide (Diggle 1987, 89, 25–27 = TrGF 5, T 85) cita Dicearco (Mirhady 2001, fr. 62 = Wehrli 1967, 63) per la notizia secondo cui Euripide avrebbe composto la tragedia rielaborando un’opera di Neofrone e la notizia è attribuita all’opera Ἑλλάδος βίος (anche qui in consonanza con Aristotele, Gigon 1987, fr. 774).

L’esegesi antica di Eschilo da Aristotele a Didimo  

Sofocle, e non da Eschilo come altri dicevano.33 A differenza di quanto sappiamo del suo lavoro su Sofocle ed Euripide, mi pare che questa sia la sola notizia che abbiamo di Dicearco a proposito di Eschilo. In ogni caso, è inequivocabile quanto si legge nella hypothesis del Reso (fr. 114 riportato sopra), dopo le notizie sul problema dell’autenticità con la citazione dalle Didascalie, dove si dava il dramma per genuinamente euripideo: ὁ γοῦν Δικαίαρχος ἐκτιθεὶς τὴν ὑπόθεσιν τοῦ Ῥήσου γράφει κατὰ λέξιν οὕτως κτλ. La combinazione di questa testimonianza con quella di Sesto Empirico mi pare risolutiva: esisteva di sicuro un’opera di Dicearco dedicata alle hypotheseis di Sofocle ed Euripide, un’opera che evidentemente proseguiva e ampliava il lavoro delle Didascalie aristoteliche, dal momento che alle basilari informazioni essenzialmente storico-teatrali ‘aggiungeva’ quanto meno (la prudenza è d’obbligo, ne sappiamo troppo poco) l’importante dato identificativo costituito dall’incipit e l’esposizione del mythos della tragedia, elemento essenziale dell’opera teatrale secondo i principi del maestro nella Poetica,34 e con ogni probabilità anche altre notizie e osservazioni. Per quanto riguarda i numerosi frammenti papiracei che sono stati ricondotti alle hypotheseis di Dicearco, non possiamo occuparci a fondo del problema in questa sede. Osserviamo solo che è abbastanza verisimile che di opere erudite di questo tipo si producessero epitomi e selezioni, suscettibili di una più agevole consultazione e di una più ampia circolazione: un buon parallelo si può indicare nelle epitomi di Eraclide Lembo delle Politeiai aristoteliche e di opere biografiche di Ermippo (quest’ultima conservata parzialmente in P.Oxy. 1367 del II sec. d.C.) e di Satiro.35 Non è necessario che tutti i frammenti conosciuti e riconducibili a questa tipologia di hypotheseis risalgano alla stessa opera, che si tratti proprio dello scritto originale di Dicearco oppure di un’epitome o altra compilazione da essa derivata, cosa che mi pare più probabile.36 Per il momento, aggiungiamo

 33 Mirhady 2001, fr. 100 = Wehrli 1967, 76 = TrGF 4, T 98 = Bagordo 1998, F 4. (cf. TrGF 3, 36, 56–59; Herington 1972, 62, 5–8). 34 Sul mythos torneremo più avanti; sul tema è importante Meijering 1987, 99–132. 35 Dilts 1971; Gallo 1992; per Satiro cf. Schorn 2004, 18. 36 Cf. van Rossum-Steenbeek 1998, 1–32, partic. 2–3. La scelta editoriale di Mirhady 2001 a questo proposito lascia qualche dubbio. Dei diversi papiri con hypotheseis di tragedie, egli include nella sua edizione di Dicearco come frammento autentico soltanto P.Oxy. 2457, numerato come fr. 115 B in parallelo con la hypothesis dell’Alcesti conservata nei codici euripidei (e nello schol. ad Plato Symp. 179 b), numerata come fr. 115 A (= Wehrli 1967, 82 = Bagordo 1998, F 12: vd. sopra p. 429): in quest’ultima Δικαιάρχου è un’aggiunta congetturale di Triclinio, che D. Mirhady non mette fra parentesi uncinate, come invece fa J. Diggle 1987. Il testo papiraceo è pubblicato in una redazione molto integrata e ricostruita, mi pare con un eccesso di fiducia: si

  L’esegesi antica di Eschilo da Aristotele a Didimo solo che anche in questo caso rileviamo come i testimoni riguardanti Euripide siano nettamente più numerosi di quelli riguardanti Sofocle, in armonia con gli altri dati quantitativi che abbiamo visto sopra.37 La testimonianza di Sesto Empirico su Dicearco è rivelatrice non solo per il titolo che conserva, ma anche per il contesto nel quale è inserita, vale a dire un esame dei diversi significati del termine hypothesis, il primo dei quali comporta un discorso di chiara matrice aristotelica, cui appartiene anche l’uso del ben connotato termine περιπέτεια. Riconsideriamo il passo di Adv. Math. 3.3. Alla base del discorso sta l’idea dell’importanza primaria, indicata da Aristotele nella Poetica, del mythos nella costruzione della tragedia (la coerente σύστασις τῶν πραγμάτων), vale a dire del lavoro fondamentale per cui il poeta rielabora il contenuto tradizionale della vicenda in base ai criteri di necessità e verosimiglianza e gli conferisce un significato universale. Il riferimento all’opera di Dicearco si proietta dunque anche sullo sfondo della riflessione teorica di Aristotele, per così dire in parallelo al rapporto già evidenziato con la raccolta di materiali delle Didascalie: diciamo che lo studio e l’esposizione di Dicearco dei mythoi delle tragedie si ricollegava al pensiero e all’opera del maestro sui due piani, quello pratico delle Didascalie e quello teorico della Poetica, secondo una ben nota connessione organica, tipicamente aristotelico-peripatetica, fra documentazione storica e riflessione filosofica. Senza nessuna pretesa di compiere una ricognizione sistematica e completa, possiamo dire che ci sono sufficienti testimonianze sul fatto che i ben noti interessi di Aristotele e della sua scuola per le opere e gli autori della poesia,38 accanto alla copiosa messe di studi dedicati all’epica e alla lirica (oltre che alla  confronti l’opposto comportamento di J. Diggle all’inizio del suo testo dell’Alcesti, dove i resti di P.Oxy. 2457 sono riprodotti senza integrazioni. Se è vero che il papiro offre una redazione più ampia di quella dei codici, si è indotti a pensare che l’editore intenda quest’ultimo come il Dicearco autentico rispetto a quello della hypothesis dei codici, e P.Oxy. 2457 come l’unico autentico fra i numerosi frammenti papiracei con hypotheseis di questo tipo. Diverso il comportamento di Bagordo 1998, 121–122, che aggiunge anche P.Oxy. 3013, hypothesis del Tereo di Sofocle, come fr. dubbio *F 13. Come trattare editorialmente i frammenti papiracei di hypotheseis di Sofocle e Euripide, se e quali distinzioni fare fra i diversi testimoni, è un problema difficile e delicato, che può produrre conseguenze pesanti in chi utilizza tali edizioni senza un’approfondita conoscenza del problema. Esso sarà affrontato nella relativa sezione del CLGP. 37 Salvo errori, i frammenti euripidei finora conosciuti sono 19, quelli sofoclei 2 o 3. Nei relativi volumi del progetto CLGP sarà contenuto un riesame completo di questi materiali per Euripide e Sofocle, con nuove edizioni commentate. 38 Cf. Richardson 1992 e Richardson 1994; Blum 1977, 27–109; Montanari 2000b (= questo volume, cap. 55); 2001; efficace sintesi con utilissima raccolta dei frammenti in Bagordo 1998. A questo tema ho dedicato Montanari 2012 (= questo volume, cap. 56).

L’esegesi antica di Eschilo da Aristotele a Didimo  

musica), comprendevano anche diversi trattati su argomenti relativi al teatro sia comico che tragico. In questo quadro dunque, nel periodo fra Aristotele e le prime generazioni dei suoi allievi, si collocano i più antichi studi e le prime raccolte di repertori eruditi dedicati al teatro greco, nei quali anche Eschilo doveva avere la sua parte. Nel catalogo delle opere di Aristotele compare anche un Περὶ τραγῳδιῶν, di cui abbiamo solo il titolo,39 come pure dello scritto Sui tre tragici e del Περὶ τῶν παρ᾿ Εὐριπίδῃ καὶ Σοφοκλεῖ di Eraclide Pontico;40 invece all’opera Sui poeti tragici di Aristosseno sono congetturalmente attribuiti tre frammenti oltre all’unico sicuro.41 Risulta che Eraclide Pontico parlasse delle vicissitudini di Eschilo in relazione alle sue presunte rivelazioni di segreti dei misteri e raccontasse di come il tribunale lo avrebbe assolto a causa della sua partecipazione alla battaglia di Maratona.42 La fonte introduce la citazione come segue: λέγει δὲ περὶ Αἰσχύλου καὶ Ἡρακλείδης ὁ Ποντικὸς ἐν τῷ πρώτῳ περὶ Ὁμήρου. Abbiamo qui un utile richiamo alla, peraltro evidente, possibilità che di Eschilo si parlasse anche in trattati dedicati ad autori di altri generi poetici e non solo ad autori e tematiche specificamente teatrali. Mi pare giusto ricordare come questa tradizione di carattere aneddotico-biografico avesse già un precedente interessante nelle Epidemie di Ione di Chio, dove si trovavano resoconti di episodi accaduti a vari personaggi, tra cui anche Sofocle ed Eschilo, a proposito del quale Ione certamente menzionava la partecipazione ai fatti di Salamina e con tutta probabilità un episodio accaduto durante un incontro di pugilato ai Giochi Istmici.43

 39 Diog. Laert. 5. 26: Gigon 1987, 24 nr. 137 e 548 nr. 136; raccolta di testimonianze sul tema in Bagordo 1998, 91–94. 40 Rispettivamente Wehrli 1969, fr. 179 (titolo in Diog. Laert. 5. 88) e fr. 180 (titolo in Diog. Laert. 5. 87) = Bagordo 1998, F 1 e F 2. Un Περὶ Εὐριπίδου καὶ Σοφοκλέους è attribuito anche a Duride: FGrHist 76 F 28 = Bagordo 1998, F 2 = TrGF 4, T 150. 41 Wehrli 1967a, fr. 113 + frr. 114–116 (= Bagordo 1998, F 11–14). 42 Wehrli 1969, fr. 170 = TrGF 3, T 93 b (la fonte è un anonimo commento all’Etica Nicomachea di Aristotele). Nuova edizione di Eraclide Pontico: Heraclides of Pontus, Edited by E. Schütrumpf, P. Stork, J. van Ophuijsen, and S. Prince, RUSCH XIV, New Brunswick (USA) and London (UK), 2008. L’episodio rientra nel topos biografico dell’avversione degli Ateniesi nei confronti degli intellettuali: cf. Schorn 2004, 58 e 284. 43 Edizione dei frammenti di Ione in Leurini 2000, 63–72 Epidemie: frr. 100–112; cf. Radt 1985, 102, che rimanda a T 14 (= Leurini 2000, f 101), a T 149 “probabilissime” (= Leurini 2000, f *108), e a T 111, 112, 114, a proposito dei quali haud male coniecerunt viri docti (= Leurini 2000, f **133, f **135, f **134); sulle Epidemie di Ione (considerata un’opera pionieristica e addirittura l’invenzione di un genere) cf. West 1985a, 75 sg.; Leurini, 1991; Leurini 2006, 35–40 con bibliografia precedente; Pelling 2007, 75–109, cf. anche introd. e passim nel volume; Jouanna 2007,

  L’esegesi antica di Eschilo da Aristotele a Didimo Un’altra celebre notizia deve essere inclusa nel dossier che stiamo costruendo: quella relativa alla copia ‘ufficiale’ dei tre grandi tragici fatta fare da Licurgo ad Atene. R. Pfeiffer ricorda il fatto con queste parole: “At about the same time (after 334 B.C.) as Aristotle was compiling the records of performances of plays from the Athenian archives, his friend and fellow student Lycurgus, who was in charge of the public finances from 338 to 326 B.C., had an official copy made of the works of the three great tragedians; this was deposited in the public archives and the actors were compelled by law to keep to this authorized text”.44 Riprenderemo in seguito questa notizia per aggiungere che a quanto pare, circa un secolo più tardi, questa copia pervenne alla biblioteca di Alessandria (cf. p. 413). Per il momento, registriamo anche questo elemento nel quadro del lavoro che, a partire dalle Didascalie di Aristotele, l’ambiente del Peripato dedicò al fenomeno teatrale, con una particolare attenzione a quelli che si erano imposti come i tre autori di maggiore rilievo. J. Jouanna ha ben sottolineato come lo scopo di questo intervento risulti chiaramente non tanto quello di conservare l’esistenza dei testi, bensì di conservarne la forma autentica e corretta, preservandola dalle corruzioni soprattutto portate dalle riprese teatrali e dagli attori nelle rappresentazioni.45 Si tratta di un fatto che merita di essere sottolineato in tutto il suo valore, anche nella cornice degli influssi peripatetici sugli orientamenti e le attitudini intellettuali dei filologi alessandrini. Accanto a diversi scritti su argomenti legati al teatro, Cameleonte compose anche un Περὶ Αἰσχύλου, di cui rimane qualche frammento riportato da Ateneo, con citazione di alcuni versi di Eschilo e certamente con un prevalente interesse di tipo biografico-aneddotico (al quale è riconducibile anche la suddetta notizia di Eraclide Pontico a proposito dei misteri), sviluppato attraverso il ben noto metodo di interpretazione in chiave biografistica di passi delle opere del poeta.46 Cameleonte sosteneva che Eschilo componesse in stato di ubriachezza e che fosse stato lui, e non Euripide, il primo a introdurre sulla scena personaggi ubriachi. Inoltre, a proposito della coreografia, utilizzando testimonianze di Aristofane (perché, dichiara, nei comici si trovano notizie attendibili sugli autori tragici), Cameleonte dice che Eschilo fu il primo a creare schemi di danza per i  35–38. Sintesi e raccolta delle testimonianze sul periodo prearistotelico in Bagordo 1998, 14 (per Ione). 44 Pfeiffer 1968, 82 e 192; Aeschylus: TrGF 3, T 145 = Sophocles: TrGF 4, T 156; Wartelle 1971, 101–124. 45 Jouanna 2007, 524–526. 46 Giordano 1990; cf. Arrighetti 1987, 161–190; Montanari 1989 (“Chamaeleon”) (ediz. commentata dei frammenti su papiro); Arrighetti 1994, 230–234; 240–243; Bagordo 1998, 26 sg. e 112–116; D. Mirhady 2012; nuova edizione Martano 2012; riguardano Eschilo i frr. 42–44).

L’esegesi antica di Eschilo da Aristotele a Didimo  

cori, senza ricorrere a maestri, e in genere assumeva di persona l’intera direzione artistica.47 Come per molti autori, a quanto pare anche per Eschilo gli interessi per la biografia e la personalità del poeta diedero luogo, essenzialmente in ambiente peripatetico, a forme di interpretazione di passi delle sue opere in senso biografico e alla ricerca (o anche invenzione) di aneddoti capaci di dipingere icasticamente tratti del carattere e della vita del personaggio, (ri)costruendone un’immagine. Un discorso che vale anche per l’opera di Satiro, del quale possiamo leggere ampie porzioni della Vita di Euripide grazie a un fortunato ritrovamento papiraceo, P.Oxy. 1176, che ci ha restituito parti del sesto libro della sua raccolta di biografie, quello dedicato appunto ai tre tragici, come si legge nel titolo conservato: Σατύρου Bίων ἀναγαφῆς ς΄ Αἰσχύλου Σοφοκλέους Εὐριπίδου.48 I frammenti rimasti riguardano la vita di Euripide, mentre non abbiamo nulla a proposito di Eschilo: il metodo ampiamente applicato è appunto quello della ricostruzione biografica attraverso l’individuazione di passi delle opere suscettibili di una interpretazione e di uno sfruttamento a questo scopo.49 Il lavoro tipicamente peripatetico della ricostruzione delle biografie degli autori fu dunque un importante ambito entro il quale si sviluppò una ricca attività di esegesi dei testi mirata a questo scopo. In seguito il filone biografico si sviluppò variamente e il prodotto più cospicuo che abbiamo è l’ampio bios adespoto pervenuto in un manipolo di codici:50 ma in questa sede non ci occuperemo ulteriormente di questo aspetto della ricerca antica su Eschilo. Aggiungiamo soltanto un altro scampolo di erudizione eschilea, che si può avvicinare a questa tematica. Si tratta del Κατάλογον τῶν Αἰσχύλου δραμάτων, conservato in tre codici, che presenta un elenco delle opere di Eschilo in ordine alfabetico. È difficile dire che provenienza abbia e come lo si possa collocare cronologicamente: R. Pfeiffer lo riconduce in ultima analisi alla tradizione che risale ai Pinakes di Callimaco e alla loro continuazione nell’opera biografica di Ermippo di Smirne peripatetico-callimacheo, ritenendo

 47 Cf. Wehrli 1969a, frr. 39–42; Giordano 1990, frr. 39–42; TrGF 3, T 103, T 117ab, F 309–311; Bagordo 1998, F 5–9. 48 TrGF 4, T 148 (om. TrGF 3). Nuova edizione in Schorn 2004, con introduzione, commento e bibliografia, cf. 15–18 sulla struttura dell’opera. 49 Cf. Schorn 2004, 56–63 (6. Satyros als Peripatetiker), sulla ricostruzione biografica anche in rapporto con Eraclide Pontico e Cameleonte: cf. 194 sg. sul tema dell’ubriachezza già presente in Cameleonte, 284 sul processo dei misteri già trattato da Eraclide Pontico (cf. sopra e n. 34); cf. anche Bagordo 1998, 54 sg.; 162–164. 50 Edizione recente in TrGF 3, T 1; ad esso si affianca la breve voce biografica in Suda αι 357 = TrGF 3, T 2.

  L’esegesi antica di Eschilo da Aristotele a Didimo che esso fosse posto in appendice a una biografia del poeta e conservi l’ordinamento più seguito nelle liste antiche.51 Il Περὶ Αἰσχύλου di Cameleonte è l’unica opera specificamente dedicata a Eschilo che abbiamo incontrato finora (a parte la porzione eschilea di Satiro),52 ma ora un’altra deve essere chiamata in causa. La hypothesis dei Persiani trasmessa nei codici si apre con la notizia secondo cui Eschilo avrebbe ripreso e imitato le Fenicie di Frinico, fatta risalire a una fonte di nome Glauco:53 Γλαῦκος ἐν τοῖς περὶ Αἰσχύλου μύθων ἐκ τῶν Φοινισσῶν Φρυνίχου φησὶ τοὺς Πέρσας παραπεποιῆσθαι. ἐκτίθησι καὶ τὴν ἀρχὴν τοῦ δράματον ταύτην· “τάδ᾿ ἐστὶ Περσῶν τῶν πάλαι βεβηκότων”. Πλὴν ἐκεῖ εὐνοῦχός ἐστιν ἀγγέλλων ἐν ἀρχῇ τὴν Ξέρξου ἧτταν στορνύς τε θρόνους τινὰς τοῖς τῆς ἀρχῆς παρέδροις· ἐνταῦθα δὲ προλογίζει χορὸς πρεσβυτῶν.

Osserviamo che l’excerptum non si limita all’informazione sulla dipendenza di Eschilo da Frinico, ma aggiunge la citazione del verso iniziale di Frinico e alcuni elementi di confronto fra i due drammaturghi per quanto riguarda l’avvio della tragedia: l’esile frammento ci lascia dunque intravvedere che la fonte doveva essere un trattato di rispettabile livello erudito. Si è discusso su chi sia questo Glauco, autore di un Περὶ Αἰσχύλου μύθων, e si è imposta una vulgata secondo cui si tratterebbe di Glauco di Reggio, attivo fra V e IV sec. a.C. e noto per i suoi interessi di carattere storico-letterario, mentre per ragioni tematiche non è sembrato entrare in gioco Glauco di Samo, di cui si hanno solo poche notizie relative a studi di carattere grammaticale, in particolare sulla dottrina degli accenti.54 È uno di quei casi in cui una identificazione, avanzata dapprima come possibile o tutt’al più probabile, diventa tacitamente acquisita, mentre forse ci sono buoni motivi quanto meno per lasciarla allo stato di dubbio o forse per considerare altre possibilità. Su Glauco di Samo sappiamo assai poco, il suo inquadramento cronologico e culturale è problematico: si è parlato di uno studioso di grammatica anteriore ad Aristofane di Bisanzio oppure

 51 TrGF 3, T 78 (testo con indicazione dei mss. e riferimenti bibliografici); Murray 1957, 375; Page 1972, 335; cf. Pfeiffer 1968, 128–130, con altri esempi di cataloghi. 52 Fra le opere di Teofrasto in Diog. Laert. 5. 50 compare il titolo Πρὸς Aἰσχύλον α΄, ma pare assodato che non si tratti del poeta tragico: Radt, 1985, 102; Fortenbaugh 1992, I, 288, nr. 42. 53 Page 1972, 2 = West 1998, 3; TrGF 3, T 86 (cf. p. 101 sub Q); TrGF 1, Phrynicus T 5 + F 8. 54 Cf. Lanata 1963, 278 sg., fr. 7, con la bibliografia (che si basa sostanzialmente su Hiller, 1886, 428 sg.); cf. Bagordo 1998, 14 sg. e 137 sg.; Fornaro 1996–2003; G. Ucciardello, s.v. “Glaucus [1] (di Reggio)”; s.v. “Glaucus [2] (di Samo)”, LGGA, con la bibliografia.

L’esegesi antica di Eschilo da Aristotele a Didimo  

di un erudito peripatetico vissuto fra IV e III sec. a.C.,55 due collocazioni che peraltro non appaiono in decisivo contrasto. Per quanto riguarda i suoi interessi, solo quelli sulla dottrina degli accenti sono noti, ma bisogna ovviamente guardarsi dal vizio di dare per scontato che un autore, del quale abbiamo scarsissime notizie, si sia occupato per tutta la vita soltanto dell’argomento di cui è rimasta traccia. Mi pare da tenere presente la considerazione che il titolo Περὶ Αἰσχύλου μύθων non sembra rientrare agevolmente nella tipologia di opere di carattere generalmente e genericamente mitografico, cui è stato accostato.56 La focalizzazione su uno specifico autore e sulle trame delle sue tragedie (cioè, come dicevamo, sui μῦθοι da lui rielaborati nella sua peculiare σύστασις τῶν πραγμάτων) evoca piuttosto l’atmosfera di pensiero e di ricerca di Aristotele e del Peripato, con gli interessi precisamente diretti alle personalità degli autori e alle loro opere: in altre parole, più che di interesse mitografico dovremmo parlare di un’attenzione per il poeta Eschilo e per i suoi drammi, e precisamente per i suoi μῦθοι, come suona il titolo Ὑποθέσεις τῶν Εὐριπίδου καὶ Σοφοκλέους μύθων di Dicearco, con le connessioni e implicazioni aristoteliche di cui abbiamo detto. Non voglio avanzare identificazioni spericolate, non ci sono indizi sufficienti: è possibile anche che si tratti di un altro Glauco ancora rispetto ai due chiamati in causa.57 Ma certo il tipo di opera suggerito dal titolo e il contenuto del frammento (con le sue puntuali osservazioni in parallelo sul modo di iniziare la tragedia) inviterebbero a pensare a qualcuno collocabile entro la cornice in cui stanno anche Aristotele e i suoi scolari di cui abbiamo detto, dunque nel quadro dell’erudizione peripatetica. Purtroppo rimane solo il titolo del Περὶ Σοφοκλέους μύθων di Filocoro di Atene, ma possiamo ugualmente permetterci qualche considerazione. “Philochoros is the first scholar among the Attidographers”: così F. Jacoby riassumeva “in a short formula” i tratti di una personalità dagli interessi culturali molteplici, anche se quelli di carattere letterario hanno subito un naufragio pressoché totale e sono in genere dimenticati.58 Oltre a quello menzionato, gli sono attribuiti

 55 Cf. Fornaro 1996–2003, col. 1094; Ucciardello, art.[2] cit.; non mi sembrano privi di fondamento gli argomenti di Hanschke 1914, 113; 119–123; 127, per collocarlo fra le prime generazioni di Peripatetici. 56 Hiller 1886 e Lanata 1963, menzionano i Tραγῳδούμενα di Asclepiade di Tragilo (FGrHist 12). 57 Altri si trovano menzionati per es. in Fornaro 1996–2003. Cf. le relative voci in LGGA: i problemi di omonimia sono comuni per autori di cui si conserva pochissimo. 58 FGrHist 328, Commentary, Introd. 327 sgg. (la citazione è a 327, rr. 35 sg.).

  L’esegesi antica di Eschilo da Aristotele a Didimo trattati Περὶ Εὐριπίδου, Περὶ Ἀλκμᾶνος, Περὶ τραγῳδιῶν.59 Contemporaneo di Cameleonte e di Prassifane, Filocoro ha certamente conosciuto non solo la lezione di Aristotele, ma anche quella di Dicearco ed Eraclide Pontico, più vecchi di lui di almeno una generazione. Collocare questi aspetti della sua attività nell’ambito e nell’atmosfera influenzata dalle ricerche peripatetiche appare del tutto naturale e degno di nota, anche se non capita facilmente di vederlo sottolineato. Non sembrerà azzardato, a questo punto, ricostruire un percorso intellettuale che trova i suoi fondamenti di pensiero nello studio del fenomeno teatrale della Poetica di Aristotele e nella messe di ricerche erudite e di raccolte di materiali prodotte già dal maestro e in seguito per generazioni da allievi diretti della scuola o da uomini di cultura ispirati dall’influenza peripatetica. L’interesse per il mythos come elemento essenziale dell’opera drammatica veniva naturale in base a quanto il maestro aveva insegnato: si trattava di analizzare non solo la materia narrativa mitica utilizzata come serbatoio di storie, ma anche e, anzi, soprattutto cosa ne aveva fatto il singolo autore nella singola tragedia, come aveva costruito la sua trama per realizzare il fine proprio dell’opera teatrale. Probabilmente Dicearco (di non molti anni più giovane di Aristotele) fu il primo a dedicarsi a questa analisi con ampiezza60 e certo fu seguito da altri, fra i quali Filocoro, che studiò Sofocle, e quel Glauco che si occupò di Eschilo. Purtroppo la tradizione (compresa quella dei papiri recanti brandelli di hypotheseis di Euripide e Sofocle) ha molto immiserito (anzi, per lo più perduto) quanto sappiamo di questo lavoro e dei suoi contenuti, ma qualche indizio rimane di un’opera ben diversa da un banale riassuntino: si consideri per questo il contesto della citazione di Sesto Empirico, il contenuto del frammento di Dicearco sul prologo del Reso e del frammento dell’incerto Glauco. È usuale distinguere le hypotheseis tragiche grosso modo in due tipologie: quelle “narrative”, dette anche di tipo “dicearcheo”, e quelle “erudite”, dette anche di tipo “aristofaneo”.61 Un riesame dell’intera questione esula dai limiti di

 59 I primi due citati in Suda φ 441 = FGrHist 328 T 1, il terzo in schol. ad Eur. Hec. 3 = FGrHist 328 F 90; Bagordo 1998, 33, F 1 e 2, ritiene che, negli scritti di Filocoro riguardanti il teatro, gli interessi antiquari e mitografici prevalessero su quelli filologici e storico-letterari, ma questa idea può essere influenzata dalla selezione dei frammenti pervenuti. Peraltro, il rapporto fra erudizione, filologia e antiquaria dovrebbe essere ripensato in termini assai meno schematici: cf. Montana 2002; 2009; Montanari 2009a (= questo volume, cap. 7). A riflessioni di grande interesse storico-culturale sul tema porta la lettura di Momigliano 1950. 60 Ricordiamo che Sesto Empirico dice Δικαιάρχου τινὰς ὑποθέσεις κτλ.: è possibile che la formulazione si riferisca alla non esaustività del lavoro. 61 Sostanzialmente anche le hypotheseis della commedia possono ricondursi al medesimo schema tipologico: cf. per es. Zimmermann 1998; van Rossum-Steenbeek 1998, 1–52, distingue

L’esegesi antica di Eschilo da Aristotele a Didimo  

questo intervento, ma vorrei suggerire di considerare il materiale trasmesso dai codici e dai papiri da una prospettiva un po’ diversa, secondo un’angolazione che individua una linea tradizionale sostanzialmente unitaria da Aristotele ad Aristofane di Bisanzio, attraverso i Pinakes di Callimaco.62 Alcuni degli elementi che portano in questa direzione sono già emersi e ora proviamo a riprenderli in sintesi. Nelle ricerche di Aristotele e della sua cerchia, lo studio della produzione teatrale si affiancò con abbondanza di risultati a quello degli altri generi poetici (più arcaici), vale a dire l’epica e i vari generi che noi raggruppiamo sotto l’etichetta di “lirica”. Aristotele stesso realizzò con le Didascalie la prima raccolta sistematica di notizie sulla storia del teatro greco fino ai suoi tempi (autori, opere, rappresentazioni). D’altra parte, la riflessione teorica della Poetica analizzò autorevolmente l’essenza dell’opera drammaturgica e mise in luce l’importanza di alcuni elementi costitutivi, fra cui in primo luogo lo specifico mythos (σύστασις τῶν πραγμάτων) che individua ogni dramma di un particolare autore. Queste indicazioni furono raccolte come orientamenti di studio e di ricerca in primo luogo da Dicearco e poi anche da altri, che si dedicarono ad analizzare i mythoi delle tragedie dei maggiori autori e produssero un importante ampliamento e complemento di notizie rispetto all’opera aristotelica: ricordiamo i due casi in cui Dicearco è citato in relazione con le Didascalie di Aristotele nelle hypotheseis dell’Aiace di Sofocle (per divergenza) e del Reso (per concordanza) e l’analisi del rapporto Frinico-Eschilo nel frammento del problematico Glauco. Alle ricerche e ai materiali prodotti in ambiente peripatetico si aggiunse l’immenso lavoro dei Pinakes di Callimaco, che dedicò agli autori teatrali uno speciale pinax ordinato cronologicamente, evidentemente utilizzando le Didascalie aristoteliche e quanto di utile si trovava a disposizione.63 Tutto ciò fu messo a frutto da Aristofane di Bisanzio, come sottolineava già R. Pfeiffer: “The making of the hypotheses is thus typical of the interrelation between the Peripatetic tradition and Alexandrian scholarship”:64 ne risultò un’opera dotta ed erudita, che raccoglieva una copiosa messe di materiali e li rielaborava secondo gli ormai maturi orientamenti della filologia alessandrina, senza che questo

 “narrative hypotheses”, “learned h.”, “descriptive h.” e “Menandrean h.”, ma cf. p. 48. Si possono considerare a parte le hypotheseis omeriche, anche se di fatto esse sono riconducibili al tipo “narrativo”; un discorso analogo vale anche per le diegeseis callimachee: Ibid., 53–84. 62 Suggerita già da Pfeiffer 1968, 192 sg.: cf. l’accenno di Zimmermann 1998, 819. 63 Pfeiffer 1968, 127–134. 64 Ibid., 193.

  L’esegesi antica di Eschilo da Aristotele a Didimo decretasse rapidamente e totalmente la perdita delle altre raccolte e delle altre opere. Dobbiamo tenere ben presente il fatto che a noi sono pervenuti pressoché soltanto materiali filtrati attraverso epitomi, selezioni, fusioni, citazioni e adattamenti di vario genere, che hanno pesantemente alterato il materiale trasmesso. Questo vale ovviamente anche per tutto quanto possiamo e usiamo catalogare sotto la non univoca etichetta di hypothesis e deve indurre a prudenza nel valutare le caratteristiche e le attribuzioni dei testi pervenuti attraverso i papiri e i codici. La conclusione del discorso sviluppato fino ad ora è che per Eschilo, come per molti altri autori di poesia del periodo arcaico e classico, le prime tappe di un’attività critica ed esegetica si trovano nel contesto dell’ambiente peripatetico, che fin da Aristotele sviluppò interessi profondi e molteplici per le personalità degli autori e le opere dei grandi poeti della paideia greca a partire dall’epica e da Omero: raccolte di materiali eruditi, ricerche biografiche, studio dei testi, riflessione teorica inaugurarono un’attitudine intellettuale nei confronti dei monumenti della poesia e della letteratura del passato, che segnò una svolta fondamentale nella storia culturale antica e costituì il fermento e l’insegnamento essenziale per la filologia alessandrina. Non si può dubitare che Eschilo abbia avuto la sua parte e il suo spazio in tutto questo, anche se la tradizione successiva non lo ha privilegiato nella conservazione di materiali, con la significativa eccezione di cui già si è detto e di cui ora parleremo un poco più a fondo. Vediamo dunque ora il testo delle hypotheseis eschilee restituito dai papiri, da cui ha preso avvio il nostro discorso. La recente edizione in Commentaria et Lexica Graeca in Papyris reperta rende superfluo esporre qui tutti i dettagli dell’esame approfondito condotto in quella sede, dalla quale riprendiamo i testi (senza gli apparati) e alla quale rimandiamo per le notizie di carattere papirologico, il commento puntuale e i riferimenti bibliografici. P.Oxy. 2257,65 del II sec. d.C., offre notizie sulle perdute Etnee di Eschilo nel contesto (purtroppo conservato molto parzialmente) di una trattazione meritevole di particolare attenzione per il tipo di erudizione che contiene. Del testo tragico rimangono solo pochissime lettere in alcuni dei frammenti, che presentano tutti annotazioni marginali vergate da una mano corsiva, diversa da quella che ha scritto il testo poetico. Nel fr. 1, scritta dalla stessa mano corsiva, si legge la parte conclusiva di un testo in prosa, che si può convenzionalmente definire una hypothesis, anche se sono evidenti le peculiarità che la distinguono dai testi usualmente indicati con questo termine: abbiamo in effetti, almeno a quanto si legge nella parte rimasta, una sorta di breve trattazione su aspetti rilevanti  65 CLGP I 1.1, Aeschylus 1, con la bibliografia.

L’esegesi antica di Eschilo da Aristotele a Didimo  

dell’azione drammatica della tragedia oggetto di studio. Il testo di questa hypothesis è contenuto in un ritaglio di papiro, incollato sulla superficie del rotolo e posto forse nell’agraphon iniziale.66 Il motivo di questo stato di cose non è individuabile con certezza: si può forse supporre che il rotolo in quel punto presentasse un difetto di fabbricazione o un danneggiamento successivo, per cui la hypothesis sia stata realizzata a parte e poi incollata nella posizione voluta. Poiché la mano che ha scritto la hypothesis e i marginalia è la stessa, si tratta evidentemente di un complessivo e coerente lavoro erudito, costituito da una trattazione tematica e da frequenti annotazioni puntuali: dunque un esemplare dotato di un apparato esegetico ricco di materiali diversi (forse l’edizione di un solo dramma?). Fr. 1

5

10

——— ]…[…].φ̣[ ].ρ̣…ωναφ̣[…]……[ ]..[..].τατ(ων) ..[..].α̣…π̣α (ἑπτὰ) ].[.].παυτ[..]..π̣ρα…. ]..[…].μ̣η̣( )..[.]..ετ̣ ( ).τε Αἰcχύ̣(λ- ) ].. [εἰc] Ἀ̣θήνα̣c ἐκ̣ Δελ̣φῶν μ(ε)τ(α) βι]β̣άζο̣ ν̣ [τ(αι) ].[… ]….. ὁ Tρωΐλ̣ο̣ c C̣ ο̣ φοκλ(έουc) κ(αὶ)] ο̣ ἱ Ἀχιλλ(έωc) ἐρα[c]τ̣ (αί). κ(ατὰ) μ(ὲν) γ(ὰρ) τὸ πρῶτον μέ̣ ρ̣[oc α̣ὐ̣τοῦ ἡ cκην̣ ὴ ὑ(πό)κει̣ τ̣(αι) Αἴτνη, κ(ατὰ) δ(ὲ) τὸ δεύτ(ερον) Ξουθία, κ(α)τ(ὰ) δ(ὲ) τ̣ ὸ̣ τρίτο̣ ν πάλιν̣ Αἴτνη, εἶτ᾿ ἀ πὸ ταύτηc εἰ[c Λε]ο̣ ν̣ τίνουc μ(ε)τ(α)βάλλει κ(αὶ) γί(νεται) ἡ cκηνὴ Λεον[τ(ίνων) χῶ(ροc)], μ(ε)τ(ὰ) δ᾿ αὐτὸν Cυρακοῦccαι κ(αὶ) τὰ λοιπὰ .[ ±8 ].ηι δ(ια)περαίνετ(αι) ὅc (ἐστι) τόπ̣(οc) .[ ] >———

... sette... di Eschilo si trasferiscono da Delfi ad Atene... il Troilo di Sofocle e Gli amanti di Achille. Nella prima parte, infatti, la scena è Etna, nella seconda Xuthia, nella terza ancora Etna, poi si sposta da quest’ultima località a Leontini e la scena diventa [il territorio] dei Leontinesi, e dopo questo Siracusa, e il resto si conclude a..., che è un luogo...

“Whether this text is really part of a hyp. (locus actionis) or rather a sort of commentary (cf. γάρ in l. 8) comparing the Aetnaeae to other plays in which changes of scene are found cannot be concluded”.67 Purtroppo la perdita di una parte  66 Ma non si può escludere una collocazione alla fine del rotolo: cf. P.Oxy. IV.663, tavola in BICS 29 (1982), pl. 7; Bastianini 1996, 28. 67 Van Rossum-Steenbeek 1998, 36.

  L’esegesi antica di Eschilo da Aristotele a Didimo probabilmente cospicua del testo ne rende ancor più difficile l’interpretazione generale. Si potrebbe forse parlare di una breve trattazione (ma non sappiamo quanto e cosa venisse prima) volta a mettere in luce talune peculiarità dell’azione della tragedia. Vi si legge infatti la descrizione di numerosi cambiamenti di luogo dell’azione (ai quali fa riscontro una divisione in μέρη), che costituisce un problema anche dal punto di vista tecnico. È possibile che questo si spieghi come una sorta di espansione analitica e di approfondimento rispetto alla semplice indicazione del luogo, che si trova comunemente nelle hypotheseis. Del resto le Etnee, dal punto di vista dei cambi di luogo dell’azione, costituiscono un caso così peculiare,68 che è comprensibile si sia sentito il bisogno di corredarne un esemplare con qualche spiegazione in proposito. A P.Oxy. 2256,69 del II/III sec. d.C., appartengono ben 89 frammenti, per i quali E. Lobel ipotizzava la possibilità che provenissero da un corpus di tragedie eschilee in più rotoli.70 I frr. 1–5 contengono materiali che riguardano diverse tragedie: bisogna precisare che i frr. 1–4 presentano informazioni di carattere “didascalico” sui drammi, mentre il fr. 5 proviene con ogni probabilità da un riassunto (secondo E. Lobel del Filottete di Eschilo). I frr. 1 e 2 riguardano la trilogia tebana, di cui viene confermata la sequenza dei drammi data dall’hypothesis del codice M (Laur. 32.9, del X sec.); il fr. 3 testimonia la corretta cronologia delle Supplici, posteriori al 468 a.C. Se da una parte il contenuto appare dunque ricco e impegnativo dal punto di vista documentario, dall’altra il testo di Eschilo presenta correzioni diverse ed è corredato da varie note marginali (fra cui forse anche varianti testuali): si trattava dunque di un esemplare oggetto di non poche cure testuali ed esegetiche. Il livello critico-erudito di questi frammenti fa pensare a resti di materiali risalenti alla tradizione (di cui si è detto) che va da Aristotele e Peripatetici ad Aristofane di Bisanzio, arrivati poi anche a manoscritti di età bizantina (come appunto il codice M).

 68 Vengono in mente soltanto le Eumenidi di Eschilo e l’Aiace di Sofocle, dove però il cambio è uno solo. Sul problema cf. Fraenkel 1954; Kakridis 1958, I. 141–153; Ferrari 1968; Lesky 1972, 152; Taplin 1977, 416 sg. 69 CLGP I 1.1, Aeschylus 3, con la bibliografia. 70 Lobel 1952, 29. Per vero dire, egli sottolineava all’inizio che solo per un frammento la paternità eschilea è assicurata dalla presenza di una citazione antica; poi più avanti scriveva: “I can offer no opinion how many Aeschylean plays are represented in the following fragments. I have not succeeded in identifying even one”.

L’esegesi antica di Eschilo da Aristotele a Didimo  

Fr. 1

——— ] ζων Λά̣[ϊοc >– ] [ ] [ ] τὰ π[ρ]όcω̣[πα· 5 ] Λάϊ[οc ——— 1 ὁ προλογί] | ζων E. Lobel

Fr. 2

5

Αἰcχύλο]ῡ̱ ] ἐπὶ ἄρχoντoc Θεαγ]ε̣ νίδoυ ὀλ[υ]μ̣πιάδoc [οη̅ ἔτει] α̅ ἐνίκα Αἰcχύλ]οc Λαΐωι Οίδ̣[ί]ποδι Ἑπτ̣ ὰ̣ ἐπὶ Θήβαc Cφιγγὶ cατυ(ρικῆι)· ] δεύτερoc Ἀ̣ριcτίαc ταῖc τοῦ πα τρòc Πρατίνο]υ τραγωιδ[ί]αιc· τρί[τ]ο̣ c̣ [Πο]λυ φράcμων] Λυκoυ̣ργε̣ [ίαι] τ̣ [ετρ]αλογίαι̣ · ] [ ] [

... di Eschilo. [Sotto l’arconte] Theagenides nel primo [anno della settantottesima] olimpiade, [vinceva Eschilo] con Laio, Edipo, Sette a Tebe [e il dramma satiresco Sfinge]. Per secondo Aristias con le tragedie del padre [Pratinas]. Per terzo Polyphrasmon con la tetralogia Lycurgeia.

I frr. 1 e 2 sono riferibili entrambi alla trilogia tebana: il primo conteneva probabilmente l’hypothesis del Laio, che il fr. 2 conferma essere stata la prima tragedia della trilogia, seguita dall’Edipo e dai Sette a Tebe. Il testo del fr. 2 è ben confrontabile con la hypothesis dei Sette a Tebe conservata dal solo ms. M: ἐδιδάχθη ἐπὶ Θεαγένους Ὀλυμπιάδι οη᾿. ἐνίκα Λαΐῳ, Oἰδίποδι, Ἑπτὰ ἐπὶ Θήβας, Σφιγγὶ σατυρικῇ. β΄ Ἀριστίων Περσεῖ, Ταντάλῳ, Παλαισταῖς σατυρικοῖς τοῖς Πρατίνου πατρός. γ΄ Πολυφράσμων Λυκoυργείᾳ τετραλoγίᾳ.71

Le differenze fra i due testi e la possibilità di integrarsi reciprocamente fanno ben capire come nessuno dei due riproduca la dotta e filologica fonte originale, che possiamo indicare nella hypothesis di Aristofane di Bisanzio: entrambi presentano evidentemente una loro versione ridotta e manipolata, passata attraverso un  71 Cf. Smith 1976, Scholia Aesch. Sept., 1, rr. 4–7; Morocho Gayo 1969, 1, 5–10; TrGF 3, T 58 ab; West 1998, 61. Il testo dato sopra corrisponde al ms. M (τοῦ add. Wilamowitz), non è il testo ricostruito nelle edizioni utilizzando sia M che il papiro.

  L’esegesi antica di Eschilo da Aristotele a Didimo certo numero di tappe di trasmissione. Nello stesso codice, alla hypothesis delle Eumenidi è premessa la titolazione Ἀριοτοφ(άνους) γρ(αμματικοῦ) ἡ ὑπόθεσις, omessa dagli altri testimoni.72 Fr. 3

5

ἐπὶ α̣.[ ὀλυμπιάδοc .. ἔτει .̅ ἐνίκα [Aἰ]cχύλο[c Ἱκέτιcι, Aἰγυπτίοιc, Δαν̣ [α]ΐ̣ cι, Ἀμυ[μώνηι cατυ( ) δεύτ[ε]ρ̣[ο]c Cοφοκλῆ[c, τρίτοc Mέcατοc ⟦N.[.].[ ⟧ ⟦Bάκχαιc, Kωφοῖc̣ [cατυ( )⟧ Ποι]μέcιν, Κύκ.[ ] cατυ() 1 ἄρ[χοντοc vel Ἀρ[χεδημίδου: alii alia

sotto [l’arconte(?)... nell’anno... dell’olimpiade.] vinceva Eschilo con [Supplici, Egizi], Danaidi, il [dramma satiresco] Amimone. Secondo fu Sofocle, [terzo] Mesato con [N..., ..., Baccanti e il [dramma satiresco] Kophoi], Pastori, Cic[ ], [...] e il dramma satiresco [ ].

Prima che fosse pubblicato questo papiro, la maggior parte dei critici riteneva che le Supplici fossero un’opera giovanile di Eschilo: da qui risultava invece (come alcuni avevano già sostenuto) che la trilogia delle Supplici doveva essere datata tra le opere più tarde. Il frammento contiene la hypothesis che ricorda la vittoria di Eschilo con la trilogia della quale facevano parte le Danaidi (unico titolo leggibile nel papiro), contro Sofocle classificatosi secondo, mentre probabilmente un certo Mesato fu il terzo.73 Poiché è noto che Sofocle partecipò per la prima volta a un concorso tragico e contestualmente ottenne la sua prima vittoria nel 468 e poiché qui si fa riferimento a un secondo posto di Sofocle, si tratta evidentemente di una occasione successiva: quindi le Supplici di Eschilo furono presentate almeno dopo il 468 e non sono la sua più antica tragedia pervenuta, essendo posteriori ai Persiani e ai Sette a Tebe del 467.74 Il confronto con il fr. 2, in cui si vede un titolo subito prima dell’indicazione cronologica relativa alla trilogia tebana, deve far ritenere che il fr. 3 contenga l’inizio della hypothesis, probabilmente preceduto da un titolo come nel fr. 2 e seguito dalle informazioni sulla prima tragedia della trilogia, cioè proprio le Supplici. Dopo il nome di Mesato al r. 5 c’era evidentemente un errore nei titoli

 72 Smith 1978, Scholia Aesch. Ag. ecc., 42, 2; West 1998, 341. 73 Su Mesato vedi Yorke, 1954; Stoessl 1970. 74 Questi dati sono stati oggetto di ampie discussioni, ma allo stato attuale della documentazione credo che si possa considerarli acquisiti: cf. da ultimo Jouanna 2007, 91 sg. e 102 sg.

L’esegesi antica di Eschilo da Aristotele a Didimo  

riportati, per cui lo scriba ha posto una parentesi rotonda come segno di espunzione, così come ha fatto all’inizio del r. 6: poiché le parentesi appaiono poste in scribendo, è del tutto verosimile che egli copiasse un esemplare che già recava la stessa correzione e lo abbia riprodotto meccanicamente. Se la ricostruzione è giusta, è con il titolo Ποι]μέcιν al r. 7 che comincia la trilogia di Mesato. Che quella precedente possa essere attribuita a Sofocle è solo un’ipotesi, il cui unico appiglio è dato dal fatto che non si conoscono altre opere teatrali dal titolo Κωφοί se non un dramma satiresco di Sofocle.75 Fr. 4

5

ἡ μὲν] cκηνὴ τοῦ δρά̣ματο]c ὑπόκειται ἐν̣ ] ὁ δὲ χο(ρὸc) cυνέcτηκεν] ἐ̣ κ πολιτῶν γε̣ ]ν. ὁ προλογί(ζων) ———

Fr. 5 (a) τ-

5

10

πίδ-

].ἀδυνα ].λημφθῆ(ναι) επε]μπον ]ι̣ .[..].[ ]αρ Εὐρι Νεο]π̣τολεμο( ) ] Φ̣ιλοκτή(τ ) Ὀδυc]cεύc ] ] ]ω̣ οὗ — — —

(b)

5

— — — ]. ]τον ].αυ ]ενο ]. ]. — — —

fr. 4: la scena del dramma è a… il coro è costituito da cittadini… colui che pronuncia il prologo è... fr. 5. impossibile… essere preso… mandavano… Euripide… Nettolemo, Filottete, Odisseo...

Pur nella loro frammentarietà, questi papiri sono testimoni di due prodotti dei quali si percepisce molto bene il particolare livello filologico-erudito. Gli esemplari pervenuti sono del II e del II–III sec. d.C., ma credo non si possa dubitare che l’erudizione di cui sono testimoni risalga in ultima analisi agli esiti più maturi  75 Radt 1985, 326–328.

  L’esegesi antica di Eschilo da Aristotele a Didimo di quella linea di ricerche e indagini su autori e opere del teatro, che da Aristotele ad Aristofane di Bisanzio (e Aristarco, come vedremo più avanti) aveva consolidato un’autorevole tradizione di studi e una cospicua raccolta di materiali eruditi. Questo considerevole tesoro fu consegnato ai secoli successivi e attraversò un lungo percorso di rielaborazioni varie, che ne fece arrivare i resti fino ai codici di epoca bizantina, in rapporto con le diverse fortune dei materiali stessi e degli autori implicati. Le fonti da cui derivano le opere di P.Oxy. 2256 e 2257 risalgono certo alla filologia alessandrina e ci mostrano in modo inequivocabile come anche Eschilo avesse avuto la sua parte di cure e attenzioni durante il percorso di sviluppo che abbiamo tracciato, anche se in seguito la selezione dei materiali conservati lo mise più in ombra. Senza questi due reperti il nostro panorama sarebbe assai più sbiadito e deformato, il che deve far riflettere ancora una volta sulla casualità da cui dipende il quadro delle nostre informazioni.76 Aggiungiamo ancora una considerazione non secondaria. I papiri di Eschilo con note marginali e hypotheseis riguardano in gran parte tragedie perdute: questo rende difficile fare confronti con la tradizione scoliografica, ma conferisce a questi testimoni un ulteriore significato. A quanto abbiamo già visto, possiamo aggiungere tre pezzi con note marginali: P.Oxy. 2255, che riguarda il Glaucus Marinus;77 PSI 1211, appartenente ai Myrmidones;78 P.Oxy. 2164, che appartiene alle Xantriae.79 Infine, in un gruppo di frammenti del II sec. d.C., R. Cantarella riconobbe brandelli di un “progetto editoriale”: doveva trattarsi di una copia di opere di Eschilo (difficile comunque ipotizzare l’opera completa), costituita forse da una ventina di rotoli e comprendente diversi drammi perduti, oltre a due tragedie conservate, cioè l’Agamennone e i Sette a Tebe; in punti problematici troviamo che sono state scritte alcune annotazioni marginali.80 Se da una parte lamentiamo una certa povertà di testimonianze sull’esegesi antica di Eschilo, vale a dire un naufragio ancora più grave di quello che ha colpito altri autori, dall’altra il fatto che si trovino materiali esegetici non banali anche relativi a diverse tragedie perdute indica quanto della sua produzione  76 Van Rossum-Steenbeek 1998, 34–36, considera solo tre esempi di “learned hypotheses” su papiro: queste due di Eschilo e quella di Sofocle in P.Vindob. G 29779: cf. sopra e n. 15. 77 CLGP I 1.1, Aeschylus 2. 78 CLGP I 1.1, Aeschylus 4. 79 CLGP I 1.1, Aeschylus 6. 80 Tutti i riferimenti in CLGP I 1.1, pp. 14–15. Si tratta di PSI 1208–1210, P.Oxy. 2159–2162, 2163 + PSI 1472, P.Oxy. 2164, 2178, 2179, P.Oxy. 2245–2255: cf. Cantarella 1941, 328–329; Cantarella 1942, 75–79; Cantarella 1948, 133–141; Lobel 1952, 1; Wartelle 1971, 316–317. Cavallo 1986, 108, a ragione sostiene che non necessariamente vi dovevano rientrare le altre cinque tragedie pervenute per tradizione medievale: di diverso avviso Cantarella, ll. cc.; Wartelle 1971, 230, 335.

L’esegesi antica di Eschilo da Aristotele a Didimo  

non solo fosse ancora disponibile nei primi secoli dell’età imperiale, ma anche oggetto di un lavoro di un certo, talvolta alto, livello culturale, come fanno pensare opere quali le due hypotheseis. Per questo aspetto della fortuna e della storia del testo, i papiri offrono un’informazione molto significativa: nei primi secoli dell’età imperiale, alcuni drammi di Eschilo in seguito scomparsi non solo erano conosciuti e copiati in nuovi esemplari, ma ad essi erano anche dedicate cure esegetico-erudite per le quali esistevano evidentemente dei lettori (ribadisco che il caso del volume con le Etnee è di notevole significato).81 La scelta che più tardi ha determinato la selezione delle tragedie giunte fino a noi e la ridotta “triade bizantina” erano ancora molto lontane. A tutto questo si aggiungono alcuni altri dati offerti dai papiri di Eschilo, che vanno ricordati a questo punto. Il primo è la presenza di un certo numero di segni critici, che devono risalire all’attività dei filologi alessandrini.82 L’altro è il fatto che nei papiri si trova rispecchiata la colometria dei testi lirici, che solitamente viene riferita all’opera di Aristofane di Bisanzio (o comunque ricondotta fondamentalmente alla filologia alessandrina) e che in seguito si è ampiamente diffusa e imposta nella tradizione testuale. Un caso molto significativo è dato dalla disposizione dei versi di P.Oxy. 2333, che contiene resti dei Sette a Tebe con note marginali.83 Vale la pena spendere qualche parola anche sul P.Oxy. 220, dell’inizio del II sec. d.C., contenente un trattato di metrica.84 L’opera non è attribuibile, ma alcuni elementi fanno pensare che sia stata composta fra il I sec. a.C. e il I sec. d.C. da un autore di un certo livello culturale, forse un poeta.85 Eschilo è citato due volte. Nella col. V, rr. 1–8, l’autore afferma di essersi accorto in un secondo momento che un metro (impossibile capire quale), di cui credeva di essere l’inventore, in realtà era già stato utilizzato da Eschilo e prima di lui da Alcmane

 81 Per altre informazioni su questo tema cf. CLGP I 1.1, pp. 13–17. 82 Degni di menzione sono il segno χ in P.Oxy. 2160, fr. 2, col. II, a sinistra di r. 24; ancora il χ in P.Oxy. 2250(a), a sinistra di r. 1; la diplé in P.Oxy. 2249, a sinistra di r. 10; la diplé periestigmene in P.Oxy. 2163, fr. 1, a sinistra di r. 4. 83 CLGP I, Aeschylus 5. Cf. Fleming 1975; Hammerstaedt 1989; Irigoin 1994, 81; Bravi 1996. In questo quadro di riferimento, come ricorda Haslam 1978, 34, non sembra verosimile che Aristofane abbia prodotto dal nulla la colometria di tutta la poesia in metri “lirici” (compresa quella dei testi teatrali), lavorando solo su testi scritti come prosa: deve avere avuto dei precedenti e avere utilizzato esemplari che presentavano già una qualche evidenziazione grafica della struttura metrica. 84 CLGP I, Aeschylus 8, 61–65. 85 Cf. Grenfell/Hunt 1899, 43; Leo 1899, 506.

  L’esegesi antica di Eschilo da Aristotele a Didimo e Simonide.86 Nella col. XI, rr. 1–6, l’autore riporta due versi distinti come esempio del parteneo, cioè un dimetro ionico a minore catalettico (trattato anche nella successiva col. XII, dove ne viene presentato lo schema). In base al testo del r. 4 (πάλιν), si può affermare che le citazioni di Eschilo erano due: non è sicuro che una di esse si trovi al r. 2,87 mentre la citazione al r. 6 è esplicitamente tratta da Eschilo, da un dramma con protagonista Prometeo.88 Non è strano trovare la citazione di Eschilo per illustrare un metro ionico, dal momento che fra i tre tragici maggiori egli è quello che ne fa più largo uso.89 Il frammento ci documenta l’interesse per le strutture metriche da lui utilizzate, che porta anche a ricerche e confronti con altri autori. Un insieme di indizi, dunque, che può effettivamente far intravvedere un’attività esegetica ben più ricca rispetto agli scarsi resti conservati. Le tracce di erudizione eschilea, che emergono dai reperti papiracei appartenenti all’arco cronologico I–III sec. d.C., affondano evidentemente le loro radici nel lavoro compiuto in età ellenistica, a sua volta cresciuto sulle fertili basi delle ricerche peripatetiche. Altri indizi cospirano nello stesso senso e ad essi ci dedicheremo nelle pagine che seguono, ultima parte di questo lavoro. Per l’età alessandrina,90 la prima notizia di un lavoro sul testo degli autori teatrali appartiene ai tempi di Zenodoto, ma non lo riguarda direttamente. Si discute ancora sulla notizia di Tzetzes91 relativa alla diorthosis dei tragici e dei comici, affidata rispettivamente ad Alessandro Etolo e a Licofrone, poeti-filologi contemporanei di Zenodoto (che, per quanto ne sappiamo, non si occupò della poesia drammatica) e attivi in Alessandria: il primo si sarebbe occupato della tragedia, il secondo della commedia. Non trovo argomenti per respingere l’informazione o per sforzarsi di non dare un valore ‘tecnico’ al verbo διορθοῦν, con il quale nella fonte è qualificato il loro lavoro (come pure quello di Zenodoto). È del tutto verosimile, invece, che il loro lavoro sia stato superato e di conseguenza offuscato dai progressi compiuti in seguito e che per questo non si abbiano riscontri a tale notizia e non siano rimaste tracce dei loro interventi.92  86 Cf. PMG, Alcm. fr. 161(c) = PMGF, T B 13 (xiii); PMG, Sim. fr. 652 (iv). 87 TrGF 3, F 476a. 88 TrGF 3, F 188a. 89 Cf. West 1982, 124. 90 Rimandiamo ancora alla buona sintesi con raccolta delle testimonianze che si trova in Bagordo 1998. 91 Tz. Proll.Com. XIa I 1 e XIa II 1 (Koster 1960, I 1 A 22 e I 1 A 31): TrGF 3, T 147= TrGF 4, T 158 a–b. 92 Cf. Pfeiffer 1968, 105–107; Wartelle 1971, 136–138; Lloyd-Jones 1994, 377–379; Bagordo 1998, 35 sg., 150; Magnelli 1999, 10–12, 86 sg. (Alex. Aet. T 7); Jouanna 2007, 526.

L’esegesi antica di Eschilo da Aristotele a Didimo  

Ovvio che, occupandosi di tragedia, anche Eschilo fosse preso in considerazione al pari degli altri due tragici maggiori. R. Pfeiffer ricorda che Tolomeo III (247–221) si procurò la copia ‘ufficiale’ di Eschilo, Sofocle ed Euripide fatta fare da Licurgo ad Atene (di cui abbiamo già detto sopra a proposito dell’ambiente aristotelico) e che questa rimase nella biblioteca di Alessandria.93 Egli ha certamente ragione nel dire: “we should not overestimate its critical value”, però “it may have been of some use in the Alexandrian library” mi pare un po’ riduttivo.94 Si trattava di un esemplare innegabilmente prestigioso, altrimenti la sua storia non avrebbe spiegazione: secondo il racconto di Galeno, il re la ottenne dagli Ateniesi per farsene fare una copia lussuosa, ma poi preferì pagare la cauzione pur di tenersi l’originale.95 La cosa accadde mentre Eratostene era bibliotecario, per cui il suo successore Aristofane ebbe senz’altro a disposizione questo testo dei tre grandi tragici, di cui sarebbe difficile negare il valore simbolico e quindi di stimolo nell’ambiente alessandrino. Certo non sarà stato per questo che gli studi sul teatro presero slancio e approfondimento proprio con Aristofane, piuttosto l’interesse per le opere teatrali avrà seguito alla richiesta: ma la concomitanza storico-cronologica deve aver conferito un qualche peso anche all’episodio della copia ateniese, dato che fra l’altro Tolomeo non si sarà privato del parere e del consiglio degli eruditi della sua corte. Se per gli studi su Eschilo di Aristofane di Bisanzio abbiamo una testimonianza concreta solo nelle hypotheseis e abbiamo iniziato questo lavoro con l’affermazione (di R. Pfeiffer) secondo cui le testimonianze di un suo lavoro specifico sul testo non sono ancora emerse in modo esplicito, giunti a questo punto mi pare si possa dire che tutto quanto abbiamo osservato sulle hypotheseis, sulla storia che le ha precedute e il contesto che presuppongono, fa capire che esse non potevano assolutamente prescindere da un serio lavoro sul testo stesso delle opere drammatiche da vari punti di vista. Un papiro ercolanese (P.Herc. 1012) restituisce brani di un’opera attribuita a Demetrio Lacone, il cui contenuto è stato definito “aporie testuali ed esegetiche in Epicuro”.96 Alla colonna XXII, rr. 1–6, si trova un riferimento ad Aristofane di Bisanzio in connessione con una citazione da Eschilo:97

 93 Pfeiffer 1968, 82; 192; Jouanna 2007, 524–526. 94 Pfeiffer 1968, 82. 95 Gal. in Hippocr. Epid. 3. 2. 4: Wenkebach 1936, 79,23–80,6 = TrGF 3, T 146 = TrGF 4, T 157; cf. Jouanna 2007, 526 e n. 15. 96 Puglia 1988. 97 Riportiamo anche questo frammento dalla recente edizione in CLGP, alla quale rimando per commento e bibliografia. CLGP I 1.1, Aeschylus 7, pp. 59 sg.

  L’esegesi antica di Eschilo da Aristotele a Didimo

5

……]c ἀλ[λὰ] καὶ Ἀριcτοφ[άνηc ὁ γ]ραμματικòc εὗρε̣ [ν πα]ρ’ Aἰc̣ [χύ]λ̣ωι τοῦτ᾿ ἐν τη̣[ ±8 ν]εότικτα δ᾿ ὑπὸ μ̣η̣ρο[ῦ ………]cι κορω ν̣ [………..᾿

1 suppl. Puglia 1988 3–4 ἐν | τῆ̣[ι Cεμέληι· Puglia 1988: “Multa quoque alia feminarum nomina supplere possis, si τη[ι erat; an Τη[λέφῳ?” Radt 1985 5 suppl. Puglia 1988

Mentre il titolo della tragedia è perduto in lacuna e rimane incerto, il nome di Eschilo è in buona parte conservato, per cui possiamo considerare certa l’attribuzione a lui del verso in questione, del quale si ricostruisce bene la frase ν]εότικτα δ᾿ ὑπὸ μ̣η̣ρο[ῦ, “neonati sotto la coscia”.98 Altrettanto sicura deve essere considerata la menzione di Aristofane “grammatico” e sembra del tutto probabile che la citazione provenga dalle Lexeis.99 Tuttavia non è possibile sapere quale fosse il termine che il grammatico εὗρεν παρ᾿ Aἰσχύλῳ e che aveva trovato degno di nota: piuttosto che il comune μηρόν, buoni candidati sembrano essere νεότικτος del r. 4 oppure forse κορώνη, ricostruibile subito dopo,100 ma la cosa resta in dubbio. Questa testimonianza si affianca ai pochi casi citati da altre fonti (principalmente Eustazio e opere lessicografiche): un piccolo gruzzolo in base al quale risulta documentata la presenza di parole di Eschilo nella raccolta lessicografica di Aristofane.101 In seguito, l’attività lessicografica non cessò di svilupparsi nell’ambito del lavoro dei grammatici: fra le numerose opere lessicali di Didimo, al teatro erano

 98 Che si trattasse della Semele è opinione di Puglia 1980, 33, e Puglia 1988, 158 sg., accolta da Amarante 1998, ma non da Radt 1985, che colloca il frammento tra quelli di drammi non identificati (F 317a). 99 Manca nell’edizione Slater 1986; altre citazioni da Eschilo nei frr. 21, 161, 203, 337, 338; cf. anche Tosi 1994, 157. 100 A favore di quest’ultimo Puglia 1988, 220–222, e Amarante 1998, 147, adducono il fatto che nella colonna seguente è riportato un passo di Apollonio Empirico contenente la parola κορώνη: poteva dunque esserci una discussione sui significati di questo termine, ma la struttura e l’interpretazione del discorso non sono chiari e lo stesso collegamento con la citazione di Aristofane di Bisanzio non può dirsi sicuro. 101 Si trovano raccolti in Nauck 1848 (cf. index iii, 302, s.v. Aeschylus) e poi Slater 1986, index 3, 235, s.v. Aeschylus. Merita di essere segnalata la possibilità che nel frammento di lessico P.Berol. inv. 9965, databile fra III e II sec. a.C. (dunque cronologicamente assai vicino ad Aristofane), si trovi glossata una parola di Eschilo: cf. Ucciardello 2006, 43.

L’esegesi antica di Eschilo da Aristotele a Didimo  

dedicate una Λέξις κωμική e una Λέξις τραγική.102 In età imperiale le sillogi di lexeis tendevano sempre più a raccogliere i frutti del lavoro dei secoli precedenti: quella di Panfilo alessandrino (I sec. d.C.) sta degnamente accanto all’opera di Didimo come summa delle conoscenze accumulatesi. Panfilo fu epitomato da Giulio Vestino (II sec. d.C.), che poco più tardi fu utilizzato, assieme ad altre fonti, da Diogeniano di Eraclea (II sec.), a sua volta poi base principale per il lessico di Esichio di Alessandria (V–VI sec.). A Diogeniano è stato attribuito un lessico dedicato specificamente a Eschilo, ma la notizia non è sicura.103 Non si trovano citazioni dei maggiori filologi alessandrini negli scholia vetera a Eschilo che abbiamo, ma uno scolio a Teocrito104 ci offre una preziosa informazione a proposito di un commentario al dramma satiresco Licurgo, appartenente alla tetralogia Lycurgeia.105 Theoc. 10.18: μάντις τοι τὰν νύκτα χροϊξεῖται καλαμαία. Schol. e. Ἀρίσταρχος ἐν ὑπομνήσει Λυκούργου Αἰσχύλου φησὶ τὴν ἀκρίδα ταύτην, εἴ τινι ἐμβλέψειε ζώῳ, τούτῳ κακόν τι γίνεσθαι. f. ἔστι δὲ χλωρὰ καὶ περιμήκεις τοὺς ἐμπροσθίους πόδας ἔχουσα καὶ λεπτοὺς καὶ συνεχῶς αὐτοὺς κινοῦσα· οὐ γὰρ ὡς ἡ κοσκινόμαντις.

I numerosi scoli al verso discutono il senso generale dell’immagine e la precisa interpretazione di μάντις καλαμαία (la cavalletta o la mantide, cf. schol. f). Quello che ci interessa è lo schol. e, nel quale il manoscritto migliore reca Ἀρίστοφάνης invece di Ἀρίσταρχος e omette le parole seguenti ἐν – φησί: scrive cioè Ἀριστοφάνης τὴν ἀκρίδα ταύτην κτλ. Nella sua edizione C. Wendel preferisce “Aristarco” e la redazione più completa, approvato da R. Pfeiffer.106 Mi pare difficile dubitare che Ἀρίσταρχος sia la lezione corretta: è del tutto plausibile che Aristarco, e non Aristofane, abbia composto un vero e proprio commentario (tre manoscritti su sette recano ὑπομνήματι al posto di ὑπομνήσει) a una o più opere di Eschilo,107 magari proprio prendendo le mosse dalle ricerche e dal lavoro sul testo già fatti in precedenza. Un’altra testimonianza conferma che Aristarco si occupò di Eschilo. La troviamo nello schol. ad Ar. Ra. 1124.

 102 Schmidt 1854, 11 sg.; 15–111, per la lexis tragica, 82–111: Eschilo è citato al fr. 6; cf. Pfeiffer 1968, 278; Tosi 1989 e 2006. 103 L’attribuzione risale a Schmidt 1862, 4, xc: cf. Cohn 1905, 782, 48–55; Wartelle 1971, 345 n. 1; Radt 1985, 102 e F 286; Bagordo 1998, 68. 104 Schol. vet. ad Theocr. 10. 18e, p. 229 Wendel. 105 Radt 1985, 54, T 67–69 e 234–236, F 124–126; cf. West 1990, 26–50. 106 Pfeiffer 1968, 222 sg.; Wartelle 1971, 163–167. 107 Per l’ipotesi che il commentario riguardasse tutta la tetralogia cf. Radt 1985, 234 sg.

  L’esegesi antica di Eschilo da Aristotele a Didimo Τετραλογίαν φέρουσι τὴν Ὀρέστειαν αἱ διδασκαλίαι, Ἀγαμέμνονα, Χοηφόρους, Εὐμενίδας, Πρωτέα σατυρικόν. Ἀρίσταρχος καὶ Ἀπολλώνιος τριλογίαν λέγουσι, χωρὶς τῶν σατυρικῶν.

La presenza del Proteo come dramma satiresco dell’Orestea è confermata dalla hypothesis dell’Agamennone conservata nei codici.108 A quanto pare, Aristarco109 concepiva l’Orestea non tanto come una tetralogia unitaria e organica, bensì piuttosto come una trilogia, alla quale si accodava il Proteo come qualcosa di distinto, a differenza dalle Didascalie aristoteliche,110 cui evidentemente risale (attraverso Aristofane di Bisanzio) l’informazione presente anche nella hypothesis. Insieme ad Aristarco è menzionato un grammatico Apollonio, noto come commentatore di Aristofane e citato soprattutto negli schol. alle Rane: che sia stato seguace o scolaro di Aristarco e che si possa collocare nel II sec. a.C. è solo un’ipotesi plausibile.111 Osserviamo, per curiosità, una certa analogia fra le due testimonianze aristarchee relative a Eschilo. Entrambe riguardano i due soli casi eschilei per i quali è tramandata la definizione di τετραλογία, vale a dire la Lycurgeia112 e la Orestea, e in entrambi i casi l’attenzione verte in particolare sul dramma satiresco (il Licurgo nel primo caso, il Proteo nel secondo). Mi pare possibile che si tratti di un indizio relativo a riflessioni sia sulla terminologia che sulle problematiche legate alla nozione di tetralogia e trilogia nell’evoluzione della tragedia antica.113 Non troviamo testimonianze su Eschilo per i grammatici postaristarchei, a parte probabilmente lo sfuggente Apollonio di cui si è detto e forse qualche altra  108 Cf. Smith 1976, Scholia Aesch. Ag. ecc., 1, 20–22; TrGF 3, T 65 (per il Proteo cf. 331 sgg.). 109 Cf. Wartelle 1971, 165 sg.; Pfeiffer 1968 non menziona questa testimonianza. Uno scolio ai Sette a Tebe merita forse una piccola osservazione. A proposito del verso 79: μεθεῖται στρατὸς στρατόπεδον λιπών, lo schol. 79a menziona il segno χ come segue: τὸ δὲ χ πρὸς τὴν διαφορὰν τοῦ στρατοῦ καὶ τοῦ στρατοπέδου· στρατόπεδον γὰρ καλεῖται τὸ ἐνδιαίτημα τοῦ στρατοῦ. Segno critico e tipo di annotazione (una precisazione di carattere linguistico) fanno pensare a un commentario e forse il tema può richiamare alla mente l’autore del trattato omerico Περὶ τοῦ ναυστάθμου. Una pura ipotesi, senza impegno. 110 Cf. Gigon 1987, fr. 418, che però utilizza solo la hypothesis dell’Agamennone dei codici e non questo scolio aristofaneo: comportamento strano perché il riferimento alle Didascalie si trova solo qui e manca nella hypothesis. 111 Cf. Wartelle 1971, 141 sg., 165, 171; Montanari 1996a, 880; S. Perrone, s.v. “Apollonius” [1] (RE n. 77), in LGGA. Problemi di identificazione e assegnazione dei frammenti sono dovuti, come di norma, ai numerosi omonimi: in LGGA sono censiti 13 personaggi di nome “Apollonio”. 112 Così definita in schol. ad Ar. Thesm. 135 = TrGF 3, T 68. 113 Sulla questione cf. Pickard-Cambridge 1968, 80–81: “The name τετραλογία … is not known to have been applied to tragedy before the time of the Alexandrian scholars Aristarchus and Apollonius, who also were apparently the first to use τριλογία in this application” (p. 80).

L’esegesi antica di Eschilo da Aristotele a Didimo  

labilissima traccia.114 Didimo si occupò molto del teatro comico e tragico:115 la Λέξις κωμική e la Λέξις τραγική sono già state menzionate; inoltre abbiamo abbastanza informazioni sui suoi hypomnemata a Sofocle e Euripide, ma nessuna notizia certa che scrisse anche commentari a opere di Eschilo.116 Vediamo gli indizi che si possono chiamare in causa. In Eur. Phoen. 751 Eteocle dice che farebbe perdere troppo tempo dire il nome di ciascuno dei guerrieri che sta per schierare alle porte della città, e nello schol. è citato Didimo per un rimando ai Sette a Tebe:117 Schol. 751 ὄνομα δ᾿ ἑκάστου: πεφύλακται τὰς ὀνομασίας αὐτῶν εἰπεῖν, ὥς φησι Δίδυμος, διὰ τὸ ὑπὸ Αἰσχύλου εἰρῆσθαι ἐν τοῖς Ἑπτὰ ἐπὶ Θήβας.

Nello schol. Ar. Ra. 704, è citata un’affermazione di Didimo secondo cui l’espressione aristofanea è ripresa da Eschilo, ma nello schol. stesso l’osservazione è contestata dicendo che si tratta invece di una ripresa da Archiloco.118 Schol. 704 κυμάτων ἐν ἀγκάλαις: Δίδυμός φησι παρὰ τῷ Αἰσχύλῳ.119 ἔστι δὲ ὄντως παρὰ Ἀρχιλόχῳ ψυχὰς ἔχοντες κυμάτων ἐν ἀγκάλαις.120

Nello schol. Ar. Ra. 965 Didimo è citato per notizie a proposito di Formisio, allievo di Eschilo.121 Schol. 965 Φορμίσιος: Δίδυμός φησιν ὅτι Φορμίσιος δραστικὸς ἦν καὶ τὴν κόμην τρέφων καὶ φοβερὸς δοκῶν εἶναι. διὸ καὶ Αἰσχύλου μαθητὴν αὐτόν φησιν εἶναι. βαθὺς δὲ ἦν καὶ καθίει τὸν πώγωνα. κωμῳδεῖται δὲ καὶ εἰς δωροδοκίαν.

Nello schol. Ar. Ra. 1028 Didimo spiega che Dioniso vuol dire che nei Persiani di Eschilo Dario è morto (compare il suo fantasma), non che si mette in scena la morte di Dario. Quanto segue fa pensare a discussioni sulle due rappresentazioni

 114 Tenacemente cercate da Wartelle 1971, 165–184. Anche per Callistrato, allievo di Aristofane e contemporaneo di Aristarco, si è sospettata l’esistenza di esegesi eschilea, ma le basi sono troppo fragili: cf. Gudeman 1919, 1742 sg.; Barth 1984, 19. 115 Schmidt 1854, 12 e 240 sgg.; cf. Pfeiffer 1968, 276–277; per Eschilo: Wartelle 1971, 185–195. 116 Pfeiffer 1968, 277; Schmidt 1854, 12, 245, 249 bis, 250; cf. anche Wartelle 1971, 187–190, Morocho Gayo 1989, 14–17. 117 Schmidt 1854, 245, fr. 14. 118 Ibid., 249, fr. 12. 119 TrGF 3, F 462. 120 West 1980, Archiloc. fr. 213. 121 Schmidt 1854, 249, fr. 14.

  L’esegesi antica di Eschilo da Aristotele a Didimo della tragedia, che forse coinvolgevano anche questo punto: ma la cosa rimane oscura.122 Schol. 1028 ἐχάρην γοῦν ἡνίκ᾿ †ἤκουσα περὶ† Δαρείου τεθνεῶτος: ... Δίδυμος, ὅτι οὐ περιέχουσι θάνατον Δαρείου οἱ Πέρσαι τὸ δρᾶμα. διὸ τινὲς διττὰς καθέσεις τουτέστι διδασκαλίας τῶν Περcῶν φασι, καὶ τὴν μίαν μὴ φέρεσθαι.

È evidente che nessuno di questi indizi ha valore probante, anche aggiungendo la considerazione di come siano particolarmente numerose le citazioni di Eschilo negli scoli aristofanei, nei quali confluì notoriamente molto materiale didimeo. Tuttavia mi pare difficile che, in tutto il suo poderoso lavoro esegetico, Didimo abbia trascurato Eschilo, pur avendo dedicato commenti a tutti i più importanti autori del teatro comico e tragico: non credo quindi del tutto azzardato attribuirgli anche un’opera esegetica su Eschilo, della quale purtroppo non è rimasto pressoché nulla. L’ipotesi che il nocciolo più antico degli scoli a Eschilo conservati risalga al commentario di Didimo è senza dubbio plausibile, ma al momento non ci sono prove:123 la tradizione esegetica può avere preso altre strade, anche se risulta difficile pensare che in tutto quanto conservato nella pur scarna scoliografia eschilea non ci sia nulla di origine didimea o predidimea. Ai corpora scoliografici possiamo dedicare ancora una breve considerazione. Tenendo presente non soltanto quelli agli autori di teatro, ma almeno anche quelli agli altri maggiori poeti (come Omero, Esiodo, Pindaro), in essi troviamo una rilevante presenza di citazioni da Eschilo, utilizzate come paralleli, testimonianze di carattere linguistico o antiquario, materiali di commento. Sarebbe interessante fare una ricerca sistematica su queste citazioni (magari coinvolgendo anche altre opere erudite, come i grandi etimologici), con lo scopo specifico di vedere se e in che misura esse possano offrire indizi di interessi esegetici specificamente eschilei, cioè se le modalità con cui Eschilo è utilizzato per commentare non rivelino in qualche modo elementi di filologia eschilea in senso proprio. Ma il tema esula dai confini che ci siamo dati in questa sede.124 Non ci sono vere e proprie conclusioni da trarre sulla base dei discorsi fatto fin qui e dell’analisi delle scarse testimonianze disponibili sui primi secoli della filologia eschilea nel mondo antico. Il percorso che abbiamo cercato di rintracciare ha visto dapprima il consolidarsi di interessi per gli autori e le opere del  122 Ibid., 250, fr. 17. 123 Cf. Dickey 2007, 36, con bibliografia. 124 Faccio riferimento al tipo di ricerca utilizzato per la filologia callimachea in Montanari 1995b (= questo volume, cap. 62); 2002b, 64 sgg. (= questo volume, cap. 58). Per un paio di esempi eschilei, mi limito qui a rimandare a Jouanna 2001 e Easterling 2006.

L’esegesi antica di Eschilo da Aristotele a Didimo  

teatro nell’ambiente peripatetico, con la produzione di cospicui lavori di ricerca e di raccolta di materiali. Nel contesto della filologia alessandrina, dopo i primi passi in età zenodotea e il grande regesto callimacheo, un’importante svolta si ebbe quando Aristofane di Bisanzio mise a frutto tutti i precedenti che aveva a disposizione: lo studio lessicografico, le dotte hypotheseis, l’analisi colometrica delle parti liriche sono tre aspetti che mostrano un lavoro di grande impegno anche sui testi eschilei, che pose solide basi per successivi commentari prodotti da Aristarco in poi, fino a Didimo. Così riassunto, questo percorso segue le comuni linee e i canoni del formarsi e consolidarsi di una tradizione criticoesegetica relativa ai principali autori e alle maggiori opere della letteratura greca pre-aristotelica. Non è possibile dubitare che Eschilo sia stato uno dei poeti oggetto di interesse e di studio al pari degli altri grandi: il fatto che le testimonianze per lui siano così scarse e talvolta davvero misere, tanto da obbligare a spremere ogni frustulo e anche ad avventurarsi in ipotesi sostenute soprattutto dal parallelo con altri autori (teatrali), è evidentemente conseguenza di una minore fortuna in epoca imperiale, tardoantica e bizantina, per cui una larga parte del materiale esegetico antico andò perduto col tempo, lasciandoci una situazione compromessa e un’immagine deformata.125 È noto peraltro che Eschilo godette invece di particolare fortuna e di particolari onori ufficiali in Atene nel V sec., tanto che fu decretato il diritto di “avere il coro” per chi volesse rimettere in scena le sue tragedie. Fra la rappresentazione delle Rane di Aristofane e il provvedimento di Licurgo (di cui si è detto sopra) si consolidò in modo inequivocabile il canone che consacrò la triade tragica, ben presente nella visione aristotelica e peripatetica, ma d’altro canto si affermò anche l’idea — evidentemente destinata ad accentuarsi — di un Eschilo arcaico, campione solenne di un modo e di un gusto un po’ lontano e sempre più estraneo rispetto alla contemporaneità.126 L’autorevolissimo Sofocle era il più apprezzato da Aristotele127 e in seguito Euripide divenne il più conosciuto e il più diffuso: abbiamo visto come, almeno a quanto ne sappiamo, questi due risultino ben più presenti di Eschilo nelle opere dei peripatetici, compresa la Poetica di Aristotele, dove Eschilo è il meno citato. Si può comprendere che Eschilo sia rimasto relativamente più in ombra negli sviluppi storico-culturali postaristotelici fino all’età bizantina. Questo spiega la scarsità di materiale rimasto, ma non credo abbia significato un minore interesse e un ridotto impegno di  125 Per Eschilo la tradizione erudita di età bizantina non è particolarmente abbondante e rivela gli inevitabili immiserimenti: cf. per es. Tosi 1994, 195 sg.; 1989 e 2006. 126 Cf. le considerazioni e la documentazione in Zimmermann, 2006. 127 Abbiamo visto sopra (p. 430) come Dicearco riprendesse l’idea dell’eccellenza dell’Edipo re.

  L’esegesi antica di Eschilo da Aristotele a Didimo studio negli ambienti di intellettuali, critici e filologici. Mi pare che lo sforzo di documentazione e di analisi fatto in questo lavoro debba portarci a ritenere del tutto plausibile, anche se in parte ipotetica, la ricostruzione di una storia dei primi secoli di critica e filologia su Eschilo, da Aristotele a Didimo. Ancor più che per molti altri autori, per Eschilo dobbiamo ragionare pensando a una realtà antica assai più ricca di quella che vediamo oggi, anche se in seguito la sua fortuna fu meno ampia.128

Discussion M. Griffith: Thank you for your very informative and up-to-date account of what we know about the continuation of Aristotelian modes of literary/dramatic analysis down to the time of Didymus, especially as these relate to the ancient study and interpretation of Aeschylus’ plays. You have provided a convincing case, in my opinion, for the persistence of Aristotelian criteria of literary/dramatic criticism in the Peripatetic tradition represented by Dicaearchus, Heraclides, Chamaeleon, and Glaucus (incidentally, I am persuaded that you have correctly identified the author of the remark in hypoth. Aesch. Pers. about Phrynichus etc., as being Glaucus of Samos, not the earlier Glaucus of Rhegium), as manifested by references to their concern for issues of mythos, peripeteia, etc. (and one might add the high esteem accorded to enargeia as a prime poetic-stylistic goal, in many critics of this period and in the scholia to the tragedians — cf., e.g., R. Meijering, Literary and rhetorical theories in Greek scholia, [Groningen 1987]). I wonder whether you might refer also to the papyrus P. Herc. 1581, fr. I (M.L. Nardelli, CErc 8 [1978], 96–103, and mentioned e.g. in R. Janko, Aristotle on Comedy, [London 1984], 142), which seems to employ the notion of katharsis in tragedy in what appear to be very much the terms envisaged by Aristotle in Poetics, 6. The author and date of that papyrus text are unknown; but it seems to be another piece of evidence that the Poetics and Aristotle’s “theory of tragedy” were in fact known and quite fully incorporated into literary studies in the Hellenistic and/or early Roman imperial eras — a fact which used to be flatly denied by modern scholars until recently. But then how are we best to account for the combining of such criteria with the interest in biography which shows up in several of the same critics whose work on the dramatists you describe in your paper, when Aristotle himself shows no  128 Cf. Wartelle 1971, 169 sg.; per il periodo posteriore a Didimo, ibid., 197 sgg.; cf. anche in questo volume il saggio di A.J. Podlecki (Podlecki 2009).

L’esegesi antica di Eschilo da Aristotele a Didimo  

interest at all in the bios of a tragic poet, only in his technê and the constituent parts of his poetic text? (We find the same combination of analysis of style with biography, of course, in the rhetorical tradition, notably in the Preface of Dionysius of Halicarnassus’ de Oratoribus Veteribus, and in his Lysias. I’m not sure what Dionysius of Halicarnassus’ relationship is to the Aristotelian critical tradition.) F. Montanari: Sono sempre più convinto dell’importanza di Aristotele e della tradizione peripatetica negli sviluppi della filologia e dell’erudizione di età ellenistica, del ruolo decisivo che essi giocarono come fermento ispiratore: è un tema al quale hanno posto attenzione diversi studiosi negli ultimi decenni ed al quale anch’io ho dedicato alcuni lavori. Nel quadro rientra anche la Poetica di Aristotele, che a mio avviso era senza dubbio conosciuta: il fatto che non la si trovi citata nei corpora scoliografici pervenuti non è un argomento, perché la Poetica fornisce fondamenti teorici e filosofici alla concezione e alla comprensione della techne poietikè, quindi la si può trovare rispecchiata in concetti e terminologia piuttosto che addotta per interpretazioni specifiche e puntuali. Come ho cercato di argomentare sopra, nel pensiero e nell’attività di Aristotele e del Peripato le ricerche di documentazione storica ed erudita si collocano accanto ai fondamenti teorico-filosofici: i due aspetti procedono in parallelo e si integrano (si pensi alle ricerche sulle Politeiai accanto alla Politica). Questa impostazione, a mio avviso, spiega anche la nascita e la crescita di interesse per la biografia degli autori e per altri dati di documentazione (per il teatro, le Didascaliai e anche i mythoi) in parallelo con la riflessione teorica sulla techne poietikè e il suo statuto filosofico. Sulla tematica “aristotelico-peripatetica” in questa sede è particolarmente opportuno menzionare i già citati Entretiens del 1994, La philologie grecque à l’époque hellénistique et romaine, e quanto si trova nella mia nota 29 (da cui risalire ad altra bibliografia). Non posso dire nulla di preciso a proposito di P. Herc. 1581, non avendolo studiato specificamente, ma certo mi pare un testimone da prendere in seria considerazione, partendo da quanto dicono l’editrice e R. Janko nei luoghi da Lei citati: grazie per averlo ricordato. M. Griffith: Perhaps this next question will take you outside the bounds of your paper — but in case you feel like commenting, I would be very grateful for your opinion on the following. The sands of Egypt have yielded a remarkable number of papyrus finds containing (pieces of) Greek satyr-drama. Several are Aeschylean (admittedly, many of them from the same hand and apparently part of a single collection: PSI 1209, P.Oxy 2161, 2245, 2256?), but we also have in separate papyri quite large amounts of Sophoclean satyr-drama, and even a number of definite satyric-fragments by Achaeus and Ion and Adespota — i.e. authors who

  L’esegesi antica di Eschilo da Aristotele a Didimo were not included among the Big Three. How do you think we can best account for this apparent continuing popularity and interest in satyr-dramas, when Aristotle seems to show so little interest in them in the Poetics (or elsewhere in his surviving works, for that matter) except as a stage in the early development of tragedy? Certainly, satyr-drama continued to be widely composed and performed in the City Dionysia, and increasingly abroad as well, during the 4th C. and beyond. But copying and studying texts of these plays in Egypt through into the 2nd C. AD — I’m not sure how best to account for that. I wonder too if this issue is related to the issue you raise about the uncertainty as to whether Peripatetic and Alexandrian critics usually thought in terms of a tragedian producing a trilogy or a tetralogy (i.e. including satyr-play)? F. Montanari: Per quanto riguarda il dramma satiresco, in primo luogo non bisogna mai dimenticare il forte aspetto di casualità dei ritrovamenti (e inoltre della decifrazione e pubblicazione dei pezzi conservati nelle collezioni), che rappresenta, per così dire, un fattore di rischio non piccolo quando si fanno considerazioni quantitative e statistiche, che pure offrono elementi utili alle ricostruzioni storiche (io stesso ho utilizzato questo tipo di dati nel mio intervento). Detto questo, in effetti la documentazione papiracea del dramma satiresco appare considerevole, anche se in verità non mi sembra tanto squilibrata numericamente rispetto ai frammenti delle tragedie. Non svaluterei l’attenzione di Aristotele nella Poetica, anche se in forma sintetica, e non dimenticherei che il secondo libro sul comico (alla cui esistenza io credo) poteva ben riguardare anche questo genere teatrale. Mi pare plausibile che il dramma satiresco abbia continuato a godere di popolarità presso il pubblico e che quindi si sia continuato a produrne esemplari anche in età imperiale. In effetti, le poche testimonianze che abbiamo a proposito della definizione di “tetralogia” o “trilogia” fanno pensare a una discussione anche intorno al dramma satiresco: in fondo, l’unica cosa sicura per Aristarco è che commentò un dramma satiresco, il Licurgo (cf. sopra pp. 717–718). A. Podlecki: Aiskhylos seems to have been rather poorly served by Peripatetic and Alexandrian scholarship. (The shadowy Glaukos, who is mentioned in the hypothesis to Persai, cannot be identified.) Khamaileon of Herakleia Pontika was the exception. From his Περὶ Αἰσχύλου Athenaios drew a miscellany of information. He discussed Aiskhylos’s innovations in choreography (TrGF 3, T 103) and preserved some of his satyric verses (Ff 309–11). The story of Sophokles’s dismissive comment that his predecessor composed τὰ δέοντα ..., ἀλλ᾿ οὐκ εἰδώς γε (TrGF 3, T 117 b) — that is, allegedly while drunk — should

L’esegesi antica di Eschilo da Aristotele a Didimo  

probably be labeled fiction rather than biography. Otherwise, there are only passing references: T 93b from Herakleides Pontikos’s Περὶ Αἰσχύλου and T 113a from Περὶ ἡδονῆς by Theophrastos or Khamaileon. Of the Alexandrians the name of Aristophanes Byzantinus occurs just once (CLGP I, Aeschylus 7 = TrGF 3, F 317a, presumably from the Lexeis), although R. Pfeiffer thought he might also be the source of some of the material in P.Oxy. 2256 and 2257 (CLGP I, Aeschylus 3 and 1). Aristarkhos wrote a hypomnema on Lykourgos (Radt 1985, 234 sq.). From Didymos, apparently nothing. The point of my question is this. Ancient scholarship evinced much more interest in Sophokles; Pfeiffer 1968, 223 posits a commentary by Aristarkhos and remarks that he was Didymos’s “favourite tragedian” (277). It is therefore a little surprising to find him so poorly represented in the commentaries preserved in the papyri. What do you think accounts for that? Had Sophokles already, like his great but somewhat impenetrable forerunner, become ‘difficult’ and therefore less popular (than, say, Euripides) among even serious later readers of the classical canon? Or is this just an accident of preservation? F. Montanari: A dire il vero, non credo che Eschilo sia stato davvero “rather poorly served by Peripatetic and Alexandrian scholarship”: penso piuttosto (cf. la parte conclusiva del mio intervento) che per Eschilo sia andato perduto molto più materiale che per gli altri due a causa della sua molto minore fortuna successiva, non solo rispetto a Euripide (e Aristofane), ma anche rispetto a Sofocle. La scarsa presenza di Sofocle nei papiri e nei resti della antica scholarship, credo si possa in effetti attribuire a una sua minore popolarità rispetto a Euripide (e Aristofane), ma anche in questo caso dobbiamo mantenere un saggio margine di dubbio a causa di quello che potrebbe rivelarsi prima o poi, almeno in parte, “an accident of preservation”. A. Podlecki: Regarding Ion of Khios. I would be very hesitant to assign to the Ἐπιδημία any material of a serious literary-critical nature. It seems unlikely, for example, that Sophokles’s description of his stylistic evolution could have found a place there (TrGF 4, T 100, pace U. von Wilamowitz). What a typical passage looked like can be seen from TrGF 4, T 75: Sophokles comes on to the pretty wine-pourer; there follows light repartee peppered with learned quotes from Phrynikhos and Simonides; Sophokles’s stratagem to kiss the boy over the cup, followed by his report of Perikles’s playful put-down, all capped by Ion’s own sour depreciation of Sophokles’s meager skills at ta politika. The only attested direct contact between Ion and Aiskhylos (TrGF 3, T 149a, from Plut. De prof, in virt. 79), with Aiskhylos’s trite comment that the shouting spectators at

  L’esegesi antica di Eschilo da Aristotele a Didimo the Isthmian games showed less askêsis than the boxer who had just been pummeled, smacks of the kind of airy fictionalizing that has cast the ancient biographical tradition into (not always deserved) disrepute. F. Montanari: A proposito delle Epidemie di Ione di Chio, non ho parlato di “material of a serious literary-critical nature”, bensì di un precedente interessante per la tradizione di carattere aneddotico-biografico (cf. i riferimenti a p. 699 n. 43). L’elemento aneddotico (anche “airy fictionalizing”, come Lei dice) nella biografia antica è stato oggetto negli ultimi tempi di attente e approfondite analisi e valutazioni, che ne hanno messo in luce la funzione e il significato, togliendogli una reputazione negativa in effetti spesso immeritata. G. Avezzù: La ricostruzione, per quanto in parte indiziaria, è comunque affascinante. Sull’hypothesis II all’Edipo re mi pare di intravedere (a) che la forma ὁ Τύραννος Oἰδίπους ... ἐπιγέγραπται presupponga un adattamento sintattico piuttosto consistente, che ha probabilmente inserito Oἰδίπους dove originariamente stava un anaforico — dunque il giro di frase e, soprattutto, la trattazione, dovevano essere considerevolmente più ampi; beninteso, la lingua di questo genere di assemblaggi è sempre di un tipo molto particolare e sarebbe insensato assumere come riferimento un oratore del canone attico; (b) potrebbe anche essere di Dicearco solo l’informazione della sconfitta di Sofocle ad opera di Filocle, e non l’intera e alquanto stravagante considerazione che feliciter eum Regem appellant, quia etc. In ogni caso direi che si tratta di un documento molto interessante, ricco di informazioni e testimone di punti di vista diversi, appartenessero o meno a una sola trattazione più ampia. F. Montanari: Sul primo punto, mi sentirei di dire che la fonte della hypothesis doveva certamente contenere una trattazione più ampia: è del tutto normale nella tradizione di questo genere di testi, che hanno subito progressivi depauperamenti. Sul secondo punto, credo che provenga da Dicearco anche la parte precedente al dato sulla sconfitta di Sofocle: del resto, come dicevo, il riferimento all’eccellenza dell’Edipo Re risente dell’opinione espressa da Aristotele nella Poetica. J. Jouanna: Votre communication reconstitue avec la plus grande précision ce que les témoignages nous apprennent sur les différents aspects de l’exégèse antique depuis les travaux de l’école aristotélicienne jusqu’à Didyme au premier siècle avant J.C. en passant par l’érudition alexandrine, avec une utilisation des papyrus tout à fait précise.

L’esegesi antica di Eschilo da Aristotele a Didimo  

Après les remarques de Guido Avezzù sur l’hypothesis de l’Oedipe Roi, concernant la double appellation de cette tragedie, je voudrais remarquer deux choses. 1) l’expression χαριέντως n’a rien d’étonnant car τύραννον n’a pas un sens péjoratif: le titre d’Oedipe Roi est bien trouvé parce que la pièce surpasse toutes les autres. Je pense, comme vous, que l’ensemble de cette phrase est attribuée à Dicéarque et non pas seulement la fin (καίπερ ἡττηθέντα ὑπὸ Φιλοκλέους). 2) La justification du second titre Le premier Oedipe est remarquable. Elle est double: cet Oedipe est Le premier par rapport à Oedipe à Colone pour deux raisons, la chronologie de la représentation (διὰ τοὺς χρόνους τῶν διδασκαλιῶν), la chronologie du mythe (διὰ τὰ πράγματα). La première raison suppose une connaissance des didascalies qui remontent à Aristote. L’auteur de l’hypothesis devait connaître la date de la représentation de l’Oedipe Roi, ce qui malheureusement ne nous a pas été transmis. F. Montanari: Grazie per l’apprezzamento e per le Sue osservazioni, che aggiungono elementi interessanti e con le quali sono del tutto d’accordo. R. Parker: I wonder if I could draw you out on one of the passages which you quote, Schol. ad Ar. Ra. 1124. It draws a puzzling distinction between the didaskaliai, which listed the satyric Proteus along with the three plays of the Oresteia to make a tetralogy, whereas “Aristarchos and Apollonios speak of a trilogy, apart from the satyr plays”. What could this distinction mean? Did Aristarchos and Apollonios separate Proteus off because they saw no thematic connection between it and the other plays of the trilogy? Or is the scholion drawing a false distinction? Modern scholars often speak off the Oresteia as a trilogy without wishing to deny its place within a tetralogy, and it is conceivable that the scholion has made too much of similar language in Aristarchos and Apollonios. F. Montanari: Sono molto riluttante a pensare che Aristarco in un’opera filologica potesse usare “trilogia” o “tetralogia” senza troppo badare alla precisione e al significato specifico dei termini, tanto più se, come pare, l’uso critico consolidato di essi risale proprio agli Alessandrini (cf. sopra, n. 113; è possibile che la terminologia avesse anche una funzione classificatoria). Mi pare più probabile che una considerazione e una lettura strettamente filologiche e letterarie, vale a dire una fruizione ormai ben lontana dall’originaria prassi che li vedeva legati nella successione rappresentativa in teatro, portassero a percepire fortemente la differenza e la distinzione fra la trilogia tragica e il dramma satiresco e a rendere per così dire naturale esaminarli, studiarli e trattarli separatemente (cf. anche sopra la parte finale della risposta a M. Griffith, p. 724).

  L’esegesi antica di Eschilo da Aristotele a Didimo J. Jouanna: À propos de Ion de Chios: quelles sont exactement les indications que nous avons sur Eschyle dans les Ἐπιδημίαι de ce contemporain de Sophocle? Je rappelle que Ion de Chios avait eu l’idée singulière de faire des ‘reportages’ sur les grandes personnalités qui avaient visité son île natale. Sa description du banquet où Sophocle, de passage lors de l’expédition contre Samos, était la vedette est bien connue. Qu’en est-il exactement pour Eschyle? F. Montanari: Il solo frammento su Eschilo citato con nome e titolo dell’opera Ἐπιδημίαι di Ione è il TrGF 3, T 14 = Leurini, f 101, che menziona la partecipazione del poeta ai fatti di Salamina; è del tutto probabile che provenga dalla stessa opera il TrGF 3, T 149 = Leurini, f *108, che racconta un episodio accaduto durante un incontro di pugilato ai Giochi Istmici, al quale Ione ed Eschilo avrebbero assistito assieme; altri tre riferimenti ad aneddoti su Eschilo sono di dubbia attribuzione, ma su questi S. Radt dice haud male coniecerunt viri docti (cf. p. 699 n. 43). J. Jouanna: Une dernière question sur les scholies d’Eschyle. Est-il possible de déterminer, malgré l’absence d’indication sur les sources, si elles renferment du matériel remontant à l’époque hellénistique ou même aristotélicienne? F. Montanari: Purtroppo gli scoli ad Eschilo sono poveri e di conseguenza danno scarse informazioni anche sui materiali che contengono e sulla loro provenienza: quando un corpus di scoli è molto depauperato, dice poco anche sulla sua stratificazione, sulla storia della sua formazione, sugli eruditi e filologi di cui ha inglobato i materiali. Poiché abbiamo indizi consistenti che i Peripatetici e gli Alessandrini si occuparono di Eschilo come degli altri due, è assolutamente improbabile, per non dire impossibile, che nulla del loro lavoro sia rimasto nella pur impoverita tradizione esegetica giunta fino a noi. Il nostro problema è che il patrimonio della filologia eschilea antica si è particolarmente immiserito, a causa delle vicende della fortuna dell’autore di cui ho detto nella parte finale del mio intervento e in questa discussione: questo dato di fatto ci pone di fronte a particolari difficoltà e ci obbliga anche a formulare ipotesi fondate sui pochi indizi conservati. F. Macintosh: You have very clearly and helpfully defined three strands of ancient scholarly criticism: the biographical, the Aristotelean “formalist” and what one might term the “stage-craft” approaches. To what extent, do you think, does this third tradition relate to the performance tradition proper?

L’esegesi antica di Eschilo da Aristotele a Didimo  

F. Montanari: In primo luogo, grazie per aver così ben focalizzato le tre linee di interesse degli antichi sull’opera di Eschilo, che sono emerse dalla mia ricerca. Aggiungerei soltanto che la definizione di “aristotelico” si adatta, di fatto, anche all’interesse biografico e a quello relativo alla rappresentazione teatrale (vedi le Didascaliai), a testimoniare di quanto Aristotele e il Peripato abbiano influito sugli sviluppi della filologia e dell’erudizione in età ellenistica (come abbiamo appena ribadito all’inizio di questa discussione). Mi pare di poter dire che, a partire almeno da Aristotele, ci sia stato un evidente interesse per l’arte scenica e la rappresentazione, oltre che per il testo delle opere e le vicende biografiche degli autori: anche per questo aspetto posso ricordare la consonanza fra le considerazioni che si trovano nella Poetica (per esempio, il ruolo della ὄψις) e il lavoro di documentazione storico-erudita che sta alla base delle Didascaliai. Come abbiamo visto, questo tema interessò anche i filologi alessandrini, così che l’eredità peripatetica fu raccolta in primo luogo nella produzione delle hypotheseis, da Aristofane di Bisanzio fino ai diversi tipi di testimoni papiracei. In tempi recenti, è stata compiuta qualche ricerca (anche dal nostro collega G. Avezzù) dedicata alle osservazioni di carattere drammaturgico reperibili nei corpora scoliografici agli autori di teatro: Meijering 1987; cf. Grisolia 2001; Avezzù 2007; Turra 2006, partic. 153, nr. 4. Credo che su questa strada si possa ancora proseguire. Comunque, datare con precisione questo tipo di informazioni sarà sempre molto difficile, per non dire impossibile. Ad esempio, nei papiri accade che taluni marginalia si possano interpretare come derivati da una precedente parepigraphe (stage direction), non più capita come tale in esemplari che non erano più copioni per una rappresentazione (un caso nel citato CLGP I 1.4, 136 e n. 6; cf. Turner/Parsons 1987, 13). Più problematico dunque mi sembra riuscire a stabilire quanto delle osservazioni di questo genere riscontrabili nei materiali eruditi antichi si riferisca alla prima messa in scena, o comunque alle più antiche, e si sia in qualche modo conservato per tradizione, e quanto invece dipenda da riprese successive nel corso del tempo, e dunque riguardi la storia del teatro, e quanto sia basato essenzialmente sull’interpretazione del testo e dunque sia piuttosto specifico della (storia della) filologia.

 Alcuni passi di Eustazio e il Prometeo di Eschilo La nuova edizione del commentario all’Iliade di Eustazio, che sta uscendo per le cure di M. van der Valk,1 intraprende la meritoria opera di colmare una delle più sensibili lacune negli studi sulla byzantine scholarship. Si era fermi sostanzialmente all’editio princeps dei due commentari omerici procurata da Niccolò Maiorano a Roma, in tre volumi, negli anni 1542–1550,2 e più volte ristampata: l’avere a disposizione un’edizione critica moderna (fondata ora sull’autografo stesso dell’autore, il ms. Laur. 59, 2 e 3) svolge l’essenziale funzione di stimolare ad approfondire lo studio e l’esegesi delle pagine eustaziane, e c’è da augurarsi che, dopo il commentario all’Iliade, venga riedito anche quello all’Odissea. Van der Valk correda la sua edizione del testo di un primo orientativo apparato di testimonia, per forza di cose non molto ampio e limitato ai rimandi essenziali e più evidenti. Vale dunque la pena, mi pare, di render noto un piccolo contributo, che va nella direzione di scavare e mettere in luce il più possibile i numerosi e diversi elementi della cultura di Eustazio, che trasudano in continuazione nel suo dettato e lo infarciscono di una gran quantità di citazioni, riprese, reminiscenze, allusioni provenienti dai più svariati autori, nel composito gioco del riuso di un’ampia dottrina letteraria ed erudita. Nel Proemio del commentario all’Iliade (p. 2, 24 sgg.), si legge il seguente passo: ... τὰ χϱήσιμα κατὰ ἀϰολουϑίαν εὐσυνϑέτως ἐκλέγονται, οὐχ ὥστε μέντοι τὰ πάντων ἐνταῦϑα εἶναι τῶν πονησαμένων εἰς τὸν ποιητήν (τοῦτο γὰϱ ϰαὶ μόχϑος μάταιος ϰαὶ πεϱιττὸς ϰαὶ οὐδὲ ῥᾷον ἀνύσιμος), ἀλλ᾽ ὥστε τὸν γινώσϰειν ἐϑέλοντα εὑϱίσϰειν ϰατὰ τόπον εὐτάϰτως τὰ μὴ παϱέλϰοντα... «... le cose utili vengono scelte opportunamente secondo l’ordine, non tuttavia in modo tale che qui ci sia quanto hanno detto tutti coloro che si sono occupati del poeta (questo infatti sarebbe un’impresa vana e superflua e non facile a compiersi), bensì in modo che colui che vuole conoscerle trovi ad ogni luogo ordinatamente le cose che servono...»

Eustazio spiega che nel suo commentario non intende accumulare tutta l’ingente mole di materiale critico-esegetico prodotto dai suoi predecessori che si sono occupati di Omero, bensì scegliere da essa e presentare ai fruitori del suo

 1 Van der Valk 1971–1987. 2 Sull’ediz. di N. Maiorano cfr. van der Valk 1971–1987, vol. I, Praefatio, XXXI sgg. https://doi.org/10.1515/9783110772371-052

Alcuni passi di Eustazio e il Prometeo di Eschilo  

lavoro soltanto ciò che risulta utile e non superfluo. Ci interessano in questo passo le parole che abbiamo messo in corsivo, nelle quali è precisamente riconoscibile l’espressione che si trova in Eschilo, Prometeo 383: è la frase con cui Prometeo taglia corto e respinge i prudenti consigli di Oceano: μόχϑον πεϱισσὸν ϰουφόνουν τ᾽εὐηϑίαν.

Il parallelo con l’espressione eustaziana è evidente. Ma si arricchisce ulteriormente, e si spiega per intero, se si chiama a confronto anche lo scolio al verso eschileo. Si legge infatti in sch. rec. ad Aesch., Prom. 383 (p. 218, 16 Dind.): μόχϑον πεϱισσὸν] πόνον μάταιον ὁϱῶ.

A costruire l’espressione di Eustazio concorrono quindi, giustapposte, le parole di Eschilo e la glossa esplicativa, che viene inserita fra il sostantivo e l’aggettivo del sintagma eschileo. Più o meno analogo è quanto si riscontra in un luogo dell’esegesi di Eustazio al canone per la festività di Pentecoste attribuito a Giovanni Damasceno: comm. ad ode V, trop. 2, v. 3 πάνταϱχος (p. 297 Mai);3 ὅτι δὲ ϰαὶ δίχα τοῦ τ ἐν τοῖς τοιούτοις ἡ αὐτὴ ἔννοια σῴζεται ὁποία ἦν ἂν ϰαὶ μετὰ τοῦ τ, δῆλον δίχα πολλῶν ἄλλων ϰαὶ ἐϰ τοῦ πανωλεϑϱία ϰαὶ τοῦ «πανόπτην ϰύϰλον ἡλίου ϰαλῶ» (Aesch., Prom. 91), ταὐτὸν δ᾽ εἰπεῖν ‘τὸν πάντ᾽ ἐφοϱῶντα’ ϰαὶ, ὡς ἂν φαίη τις, ‘παντεπόπτην’. «Nelle parole cosiffatte, anche senza il τ si mantiene lo stesso significato che hanno nella forma con il τ: ciò risulta chiaro, a parte molte altre cose, anche dalla parola πανωλεϑϱία e dall’espressione πανόπτην ϰύϰλον ἡλίου ϰαλῶ (Aesch., Prom. 91), dove per dire ‘che tutto vede’ uno potrebbe anche, se volesse, dire παντεπόπτην».

Le parole eustaziane che abbiamo messo in corsivo nel testo coincidono esattamente con una delle glosse di πανόπτην che si leggono nello sch. rec. ad Aesch., Prom. 91 (p. 182, 20 Dind.): πανόπτην] τὸν πάντ᾽ ἐφοϱῶντα, τὸν πάντα βλέποντα, ϰτλ.

 3 L’editio princeps della Exegesis in canonem iambicum de festivitate Spiritus Sancti (attribuito a Giovanni Damasceno) di Eustazio fu pubblicata da Mai 1841, 161–383: una nuova edizione stanno preparando S. Ronchey e P. Cesaretti, ai quali devo la segnalazione del passo qui trattato.

  Alcuni passi di Eustazio e il Prometeo di Eschilo Anche se in una forma un po’ diversa dal caso precedente, ritroviamo ancora Eustazio che cita un verso del Prometeo ed accompagna alla parola eschilea un elemento esegetico di essa, che ritroviamo testimoniato nel corpus degli scolî. Nel commento ad Iliad. K 394 (ϑοὴν διὰ νύϰτα μέλαιναν) troviamo (p. 814, 22 sgg.): ... ὁ φιλόμηϱος Σοφοϰλῆς τὸ ϑοὴ μεταλαβών “αἰόλα νύξ” (Trach. 94) λέγει, οὐ μόνον ϰατὰ τὸ ἐν ἄστϱοις ποιϰίλον αὐτῆς, ϰαϑ᾽ ὃ καὶ “ποιϰιλείμων” (Prom. 24) παϱ᾽ Αἰσχύλῳ λέγεται, ἀλλὰ ϰαὶ ϰατὰ τὰς ῥηϑείσας Ὁμηϱιϰὰς ἐννοίας. «... l’omerizzante Sofocle, sostituendo la parola ϑοή, dice αἰόλα νύξ (Trach. 94), non soltanto per il suo carattere variegato dovuto alle stelle, concetto per il quale in Eschilo si trova ποιϰιλείμων (Prom. 24), ma anche per i suddetti significati omerici».

Eustazio interpreta l’espressione sofoclea αἰόλα νύξ come imitazione dell’omerico ϑοὴ νύξ: l’aggettivo di Sofocle sarebbe usato per il suo duplice significato di «veloce» (che rimanda ad Omero) e di «variegato» (in riferimento al manto di stelle del cielo notturno).4 Per quest’ultimo, Eustazio chiama ad ulteriore confronto col passo sofocleo l’aggettivo di Aesch., Prom. 24: ἡ ποιϰιλείμων νὺξ ἀποϰϱύψει φάος.5

Se in questo passo di Eustazio abbiamo l’esplicita citazione di un locus parallelus, l’uso dell’aggettivo eschileo (a quanto ne sappiamo un hapax) assume un carattere allusivo nella seguente frase del Proemio del commento a Pindaro, 2 (p. 286, 9 sgg. Drachmann): ἡ δὲ λυϱιϰὴ διαχείϱισις ... οὐ ποιϰίλως ἐγϰϱοαίνουσα, ϰαϑὰ τϱέχοι ἂν Ὅμηϱος ϰαὶ οἱ ϰατ᾽ αὐτόν, ἀλλ᾽ ὁδὸν ἁπλουστέϱαν ἑλομένη τινὰ ϰαὶ ὀλιγοσχιδῆ, ϰαὶ οὐδὲ ὡϱαϊζομένη πανηγυϱιϰῶς οἷς ποιϰιλείμων πϱόεισι ϰαὶ τανύπεπλος, ἄλλως δέ... «La lirica... non si diffonde variegatamente, come procede Omero e quelli simili a lui, ma ha scelto una strada più semplice e con poche ramificazioni; e non si abbellisce pomposamente con elementi grazie ai quali inceda con un manto trapunto ed un lungo peplo, invece...»

Qui l’epiteto da Eschilo attribuito alla notte viene trasferito, negativamente, al genere lirico, con una metaforizzazione di secondo grado (la lirica non è come la

 4 Mi pare certo che così vadano interpretate le parole di Eustazio: cfr. l’intero contesto da 814, 15 sgg.; diversamente van der Valk 1971–1987. 5 Per l’interpretazione dell’agg. ποιϰιλείμων cfr. anche Sch. ad Prom. 24 (p. 171, 18 sg. Dind.).

Alcuni passi di Eustazio e il Prometeo di Eschilo  

notte che è come rivestita di un manto trapunto = la lirica non ha un manto trapunto): non più quindi, come si diceva, una citazione di Eschilo, bensì un reimpiego allusivo dell’aggettivo eschileo. È ben noto che della cosidetta triade bizantina delle tragedie di Eschilo (Prometeo, Persiani, Sette a Tebe), il Prometeo è quella più ampiamente ed abbondantemente presente nella cultura degli uomini di Bisanzio, per i quali era questa le tragedia eschilea più letta. Eustazio obbedisce perfettamente a questa regola, ed anche il suo rapporto con il materiale esegetico testimoniato negli scoli eschilei è già stato rilevato:6 in questo quadro si inserisce il piccolo contributo che abbiamo portato. Nei passi esaminati sopra abbiamo visto tre diversi modi di reimpiego da parte di Eustazio di elementi della sua cultura classica nella tessitura del suo scrivere: la ripresa di un’espressione arricchita ed accresciuta da un elemento esegetico, la vera e propria citazione esplicita, il riuso allusivo.

 6 Cfr. van der Valk 1971–1987, I, Praefatio, LXXXVII; vol. II, Praefatio, XLVII.

 Gli storici greci e la filologia di età ellenistico-romana A Emilio Gabba, grande uomo e grande maestro

Nell’affrontare il tema della presenza degli storici greci nella filologia ed erudizione di età ellenistico-romana potevo scegliere la comoda strada di fare una panoramica di quello che conosciamo, limitandomi ai dati e alle informazioni disponibili: avrei compiuto uno sforzo documentario forse un po’ faticoso, ma certamente più tranquillo rispetto alla scelta di arrischiare un’interpretazione, come invece ho fatto. Il tema che ci accingiamo ad affrontare si muove in un’area di intersezione e di sovrapposizione fra filologia e storiografia antiche, che è più ampia di quanto forse si possa ritenere a prima vista. In tempi recenti è stato sottolineato come la linea di confine fra storici e grammatici-eruditi non sia così netta, e soprattutto non lo fosse nel mondo antico, tanto è vero che molti autori che compaiono in un Lessico dei Grammatici Greci Antichi (LGGA) 11 si trovano anche nei FGrHist di Jacoby, o viceversa a seconda dei punti di vista, e che nella continuazione in corso dei FGrHist la sovrapposizione si presenta con la stessa o con crescente abbondanza.2 Questo comporta che, nell’enorme quadro della storiografia e dell’erudizione antica, la figura dello storico-filologo è più frequente di quanto siamo abituati a pensare nei termini delle tradizionali linee di confine disciplinari.3 A parte questo aspetto (che va tenuto presente nel lavoro e nell’organizzazione della ricerca), due modalità di intersezione possono essere individuate: 1) i filologi antichi utilizzarono abbondantemente i testi storiografici a scopi esegetici, dunque come fonte di notizie nell’attività di spiegazione e commento delle grandi opere della poesia e di altri generi della prosa;4 2) i filologi antichi, a partire dall’età ellenistica, si interessarono della prosa storiografica in quanto tale e in quanto genere letterario, dunque considerata per se stessa e come oggetto di cure filologico-esegetiche, messe in opera con prassi e orientamenti

 Ringrazio Fausto Montana per le sue preziose osservazioni.  1 Consultabile in rete: http://www.aristarchus.unige.it/lgga/index.php. 2 Cfr. da ultimo Montana 2009, con la bibliografia precedente. 3 Cfr. Montana 2007; 2009. 4 Per fare un esempio, questo concetto può spiegare la particolare fortuna dell’Athenaion Politeia di Aristotele: cfr. Montanari 1993c (= Ancient Authors, ch. 23); 2006 (= Ancient Authors, ch. 24), cfr. anche Montana 2009. https://doi.org/10.1515/9783110772371-053

Gli storici greci e la filologia di età ellenistico-romana  

simili a quelli usati per i poeti e altro, con citazioni e riferimenti vari.5 È su quest’ultimo piano che si muove il nostro discorso, anche se alla fine faremo un cenno anche alla prima modalità. È naturale che la testimonianza da cui partire sia quella dei papiri, cercando di trarre da essi il massimo di informazioni possibile: vedremo a questo proposito che le testimonianze papiracee di un interesse filologico-erudito per Erodoto sono quantitativamente piuttosto esigue (ma di primaria importanza), mentre Tucidide dà maggiori soddisfazioni, e che il confronto fra i due suscita considerazioni di un certo interesse. L’altro tipo di fonte a cui dobbiamo rivolgerci è la letteratura erudita più tarda, cioè quella prodotta nell’epoca bizantina, che in questo caso va utilizzata evidentemente dal nostro punto di vista, cioè come possibile serbatoio di notizie sulle epoche precedenti, lasciando agli studiosi dei periodi successivi le considerazioni di loro competenza sul fenomeno letterario e storico-culturale in età tardo-antica e bizantina. I corpora scoliografici pervenuti di solito sono stati costituiti nei primi secoli dell’età bizantina, come esito ultimo di un processo sviluppatosi grosso modo fra la tarda antichità e la rinascenza bizantina. Possiamo dire in modo assai generico che nel corso dell’età imperiale, con dislocazioni cronologiche ed evolutive differenti e non lineari, si produssero opere esegetiche di carattere miscellaneo, basate sulla fusione di fonti che erano direttamente o indirettamente i vari grandi hypomnemata, i syngrammata, le raccolte lessicografiche e altre opere erudite di età ellenistica o dei primi secoli dell’età romana; a loro volta, queste sillogi e compilazioni costituirono sostanzialmente e prevalentemente le fonti da cui furono realizzate le raccolte di scoli che troviamo nei codici (spesso con l’aggiunta di interventi di dotti bizantini). Il valore dei vari corpora scoliografici conosciuti è molto diseguale in dipendenza da due fattori: le fonti utilizzate dai compilatori e la qualità e quantità di ciò che è stato selezionato e si è conservato. Partiamo dunque dai papiri, se non altro per ragioni cronologiche del tipo di materiale. E comincerò col ripetere un’osservazione già fatta più volte, un’osservazione che definirei pesante nella sua ovvietà, per cui deve indurci a riflettere sulle posizioni che prendiamo e le conclusioni che tiriamo. Se non avessimo il frammento P.Amh. 2.12 (MP3 483) non sapremmo che Aristarco di Samotracia si è occupato specificamente di Erodoto, e non certo in modo sporadico e

 5 Montana 2009, 169: «Nei casi considerati di esegesi a Erodoto e Tucidide, in effetti, ci imbattiamo in un rovesciamento di ruoli fra storia e poesia: ciò cui altrove si ricorre come explicans (la fonte storiografica, in quanto sorgente d’informazione) funge qui da explicandum (in quanto interessante in sé come opera letteraria) e viceversa (la fonte poetica è addotta in funzione esplicativa)».

  Gli storici greci e la filologia di età ellenistico-romana occasionale bensì così a fondo e sistematicamente da scrivere un vero e proprio hypomnema, almeno al primo libro. Il papiro reca la sottoscrizione Ἀριστάρχου | Ἡροδότου | ᾱ | ὑπόμνημα. Come chiariremo alla fine del nostro discorso, se ignorassimo l’esistenza di questa opera del principe dei grammatici-filologi alessandrini, ciò avrebbe serie conseguenze sull’idea che possiamo farci degli sviluppi della filologia in età ellenistica: in sostanza, sbaglieremmo non di poco la valutazione e la comprensione del fenomeno storico-culturale in sé e nella sua portata generale, e in particolare del ruolo della filologia alessandrina nella storia della fortuna della storiografia greca. Questo piccolo papiro (un frammento di cm 22,7 x 16,5 con resti di due colonne di testo, databile al III sec. d.C.) fu reso noto nel 1901 grazie alla editio princeps di Grenfell e Hunt6 e la loro presentazione offre un esempio da manuale del cambiamento di prospettiva intercorso nell’arco del secolo scorso a proposito della considerazione, utilizzazione e valutazione delle opere della letteratura erudita. La breve introduzione si conclude come segue: «The chief points of interest in the fragment are the proof of a hitherto unknown variant ἅμιπποι for ἄνιπποι in chap. 215 and a new quotation from the Ποιμένες of Sophocles». Per Grenfell e Hunt, dunque, il commentario è interessante per due fattori: 1) quello che ci dice intorno al testo commentato, cioè Erodoto: varianti testuali e magari qualche volta elementi esegetici di particolare rilievo; 2) la scoperta di nuovi frammenti di opere perdute. Sul commento in sé, una sola notazione: «Unless the papyrus gives only a series of excerpts from Aristarchus’ commentary, which is not very likely, that work must have been extremely brief». A rigore, possiamo dire solo che Aristarco scrisse un hypomnema sistematico al libro I di Erodoto, evitando di pensare automaticamente a un lavoro sull’intera opera erodotea, che resta soltanto possibile o forse probabile. Mi sembra tuttavia plausibile il recentissimo risultato di Ivan Matijašić, che in una voce di Stefano di Bisanzio (St. Byz. 466, 12–13 Meineke) ha rinvenuto un frammento di Aristarco di contenuto morfologico-grammaticale a proposito della forma del genitivo del toponimo egiziano Momemphis, citato in Erodoto II 163, 2 e II 169, 1. Questo induce a pensare a un commentario aristarcheo quanto meno anche al II libro.7

 6 Grenfell/Hunt 1901, 3–4 e Plate III; cfr. Viljoen 1915, no. V, 17–22; Paap 1948, no. 10, 37–40; il testo del papiro è riprodotto in apparato nell’edizione teubneriana di Erodoto di Rosén 1987– 1997, I, 135–136; cfr. Pfeiffer 1973, 348–349; Dickey 2007, 54; Montana 2009, 166–167; S. West 2011, 77–81. Piuttosto stravagante la posizione di Hemmerdinger 1981, 154 sgg. 7 Matijašić 2013, 217–220.

Gli storici greci e la filologia di età ellenistico-romana  

Il commento di Aristarco è ritenuto breve e molto desultorio, il che sottintende di un valore piuttosto limitato, per il fatto che nella col. II, abbastanza ben conservata, sono coperti ben 22 capitoli del primo libro e dal cap. 194 si passa al cap. 215, saltando del tutto i capp. 195–214. Nulla, ovviamente, sulla storia della filologia alessandrina in quanto fenomeno storico-culturale, pur avendo di fronte la testimonianza di un’opera in precedenza del tutto sconosciuta e nientemeno che di Aristarco: nel 1901 era ancora molto lontana la History of Classical Scholarship di R. Pfeiffer, che vide la luce nel 1968. Da allora il frammento è stato più volte oggetto di attenzioni, sia da parte di studiosi ed editori del testo di Erodoto, sia da parte di chi rivolge i suoi interessi in particolare appunto alla History of Classical Scholarship: la letteratura erudita non ha più un valore solo o prevalentemente ancillare, per gli altri testi che chiama in causa (commentati o citati), ma viene vista e studiata come un insieme di opere e di generi letterari che testimoniano e identificano un fenomeno culturale e intellettuale di primaria importanza nel mondo antico greco e latino. E noi oggi ci muoviamo su questo registro culturale. In tempi recentissimi, un contributo importante (per certi aspetti decisivo) su questo papiro è stato pubblicato da Fausto Montana, che riconsidera la bibliografia precedente e ottiene risultati nuovi soprattutto per la prima e difficoltosissima colonna.8 Senza bisogno dunque di dilungarmi sulle tappe della ricerca sul nostro hypomnema erodoteo, dirò solo in modo riassuntivo che considero acquisiti i seguenti punti. 1) Si tratta in effetti di excerpta dell’opera originale di Aristarco,9 che era quindi considerevolmente più ricca. Nel fatto che sia pervenuta una versione epitomata di qualcosa come quattro secoli posteriore alla sua morte, poi, non c’è nulla di strano, anzi al contrario diversi paralleli fanno pensare che il fenomeno rientri in una prassi piuttosto normale: se P.Amh. 12 è del III sec. d.C., per es. P.Oxy. 1086 è della seconda metà del I sec. a.C. (quindi solo un secolo circa

 8 Montana 2012a con tutti i riferimenti bibliografici, a cui aggiungere il più recente Matijašić 2013. 9 Grenfell/Hunt 1901 definivano la paternità aristarchea «almost certain» (p. 3); Haslam 1994, 45, n. 169, si dichiara scettico, mentre altrove (The Oxyrhynchus Papyri, vol. LXV, London 1998, nr. 4455) sembra piuttosto convinto (cfr. p. 55 e p. 57, nota a Col. I 12); nessun dubbio per Montana 2009, 166–167 e 2012a, S. West 2011, Matijašić 2013. In effetti, non vedo davvero quali motivi possano esserci per dubitare dell’esplicita subscriptio del papiro, soprattutto una volta assodato che non si tratta dell’opera originale completa bensì di excerpta, il che dovrebbe placare le ansie di chi vorrebbe qualcosa, per così dire, all’altezza del principe dei filologi antichi, cioè di come egli se lo immagina, magari dimenticando che non abbiamo nessuna opera originale di Aristarco.

  Gli storici greci e la filologia di età ellenistico-romana dopo la morte di Aristarco) e contiene un commentario omerico di contenuto aristarcheo sicuramente già epitomato; P.Oxy. 221, del II sec. d.C., contiene un commentario omerico di notevole ricchezza, frutto dell’epitome di più fonti confluite in una sintesi prodotta probabilmente in età adrianea.10 Altri esempi si possono aggiungere senza difficoltà: i grandi commentari di età ellenistica furono ben presto oggetto di epitomazioni e di fusioni in opere di natura miscellanea, diffuse a partire dai primi secoli dell’età imperiale. Per Erodoto (come per Tucidide e Senofonte) possiamo aggiungere il fatto che abbiamo notizie, se pur indirette e frammentarie, di numerosi eruditi ed esegeti che si occuparono della sua opera in età imperiale, soprattutto nel periodo della Seconda Sofistica (torneremo su questo punto):11 un contesto nel quale si inserisce bene anche la circolazione di un’epitome dello hypomnema di Aristarco, oltre che dell’altro commentario erodoteo su papiro, di cui parleremo fra poco. 2) Per quanto riguarda il carattere dell’opera, non si può parlare di un commento di contenuto specificamente storiografico, bensì, come ha ben sintetizzato Montana, «di tipo linguistico-letterario di marca erudita», che per spiegare il testo erodoteo adduce paralleli poetici (Omero implicitamente, Sofocle citato esplicitamente) per fatti sia di contenuto sia di forma: una «Sachphilologie funzionale all’interpretazione letteraria» e attenta ai contenuti del testo e ai Realien.12 Si constata insomma che il modello del commentario alle opere poetiche riesce ad adattarsi bene a opere di generi diversi e con contenuti diversi. Per esemplificare, tra la fine della col. I e l’inizio della col. II abbiamo una spiegazione di Erodoto I 194, 2 διφθέρας στεγαστρίδας, pelli usate dagli Armeni o dai Babilonesi per rivestire le imbarcazioni a scopo di trasporto commerciale;13 più avanti il lemma presenta la variante ἅμιπποι per ἄνιπποι a I 215, 1 e il commento comprende un riferimento al modo di usare il carro in combattimento da parte degli eroi omerici; segue la spiegazione della parola σάγαρις ancora in I 215, 1 (pl. ionico σαγάρις nel testo), intesa come un’ascia scita, che sarebbe usata anche dalle Amazzoni; chiude un commento all’affermazione di Erodoto in I 215, 2 (fine del capitolo) secondo cui i Massageti non usano né ferro né argento, nel quale viene addotto un verso dei Ποιμένες di Sofocle (fr. 500 Radt); segue la subscriptio (nulla sull’ultimo capitolo del libro, il cap. 216). Inoltre, il possibile frammento relativo al libro II, come abbiamo visto, riguarda un problema linguisticogrammaticale a proposito della forma del genitivo di un toponimo.  10 Pfeiffer 1973, 342 e nt. 55, 367 e nt. 30. 11 Cfr. Matijašić 2013, 220 e nt. 17. 12 Montana 2012a, 72 e nt. 5. 13 Montana 2012a, 75.

Gli storici greci e la filologia di età ellenistico-romana  

Conosciamo un solo altro frammento papiraceo di commento a Erodoto, questa volta al libro V: si tratta di P.Oxy. 4455, anch’esso del III sec. d.C. e pubblicato nel 1998 da M.W. Haslam.14 Malgrado la vicinanza cronologica e le somiglianze di carattere grafico, pare assodato che non si tratti di un altro pezzo dello stesso rotolo di P.Amh. 12 (per cui non può assolutamente essere utilizzato come indizio a proposito dell’estensione dell’opera aristarchea). Sono conservate due colonne di testo, con lemmi erodotei abbastanza estesi, che appartengono a V 52–55. Caratterizzare il contenuto dello hypomnema non risulta agevole, se non altro perché una porzione considerevole del testo conservato è costituita da lemmi e la ricostruzione delle parti di commento è per lo più assai ipotetica o addirittura impossibile. Ci sono sicuramente elementi di esegesi di tipo lessicale (col. II 6–7, 12–14, 15–16), con possibili contatti con la ricca tradizione lessicografica, nelle cui opere di carattere generale Erodoto è spesso citato: per restare fra i papiri, tre citazioni erodotee si trovano per esempio nel lessico alfabetico P.Oxy. 2087, del II sec. d.C., di cui è rimasto un frammento di termini in alpha, ma non poche occorrenze troviamo nel grande bacino della lessicografia posteriore fino alla tarda età bizantina (su questo torneremo più avanti). Compare tuttavia in questo commentario un tipo di contenuto su cui vale la pena di soffermarsi. Sembra sicuro che si parli della distanza fra certi luoghi e si rilevino incoerenze interne nell’indicazione della lunghezza di tali distanze. Inoltre, se è giusta la ricostruzione di Haslam dei rr. 2–11 della col. I (la cui leggibilità è estremamente difficoltosa), il commentatore noterebbe una discrepanza fra Erodoto e Senofonte a proposito ancora della lunghezza di un tratto della famosa strada reale per Susa. Questo secondo caso, per vero dire, è lungi dall’essere privo di dubbi: la ricostruzione è molto ingegnosa e Haslam stesso la propone solo come possibile. Che nella parte conservata ci sia una particolare attenzione a una simile tematica è un dato da trattare con prudenza, evitando generalizzazioni spericolate: in altre parole, non siamo assolutamente autorizzati a dire che questo tipo di interesse caratterizzasse sistematicamente il commentario, e di conseguenza il suo autore, anche se rimane motivo di interesse che nella parte conservata questo tipo di argomenti sia visibile. Insomma, nello hypomnema ricorrono osservazioni sul lessico e discussioni riconducibili al dibattito specialistico dei geografi (come Strabone) circa il modo di confrontarsi con le dissonanze (diaphoniai, cfr. διάφωνα in col. I 18) delle fonti in merito a misurazioni terrestri e, più in generale, con l’affidabilità (pistis, apistia, axiopistos) delle fonti storiche e geografiche.15 In verità, non sorprendono certo le osservazioni  14 Haslam, P.Oxy. LXV, cit. (cfr. supra, nt. 9). 15 Montana 2009a; 2009, 167–168.

  Gli storici greci e la filologia di età ellenistico-romana legate a problemi di coerenza interna delle opere, se si pensa ovviamente al ben noto principio della filologia ellenistica secondo cui un autore va spiegato anzitutto con se stesso (che è un principio fondato appunto sull’analogia e la coerenza interna), come pure quelle legate a discrepanze con altri autori su un certo argomento. Interessante, se vogliamo, è che in questo caso un problema tipicamente storiografico presente nel testo commentato conduca a un confronto con un’altra opera storica (Erodoto con Senofonte: sempre che la ricostruzione sia valida, non dimentichiamolo), il che peraltro sarebbe portato naturalmente dal tipo di problema che si decide di commentare (la lunghezza di un tratto della strada reale per Susa). Un simile atteggiamento filologico-esegetico richiede un adattamento a vari livelli rispetto ai contenuti che il testo propone, ed è anche metodologicamente corretto, per non dire ovvio, che il commento dipenda dal tipo di testo che commenta: è il tipo stesso di esegesi che richiede di corrispondere alle sollecitazioni del testo, oltre che agli interessi dell’esegeta. Se per alcuni contenuti risulta naturale e opportuno il confronto con testi poetici, come in P.Amh. 12 abbiamo visto che Aristarco trovava utile addurre confronti con Omero e Sofocle, in casi come quelli della misurazione di distanze, per l’anonimo commentatore erodoteo di P.Oxy. 4455, viene stimolata piuttosto la ricerca di confronti interni all’opera commentata e/o con testi dello stesso tipo e genere. Guardiamoci assolutamente dal dire in senso assoluto che in un caso l’erudito esegeta predilige il rimando ai poeti e nell’altro i confronti interni o con altri storici, perché questa sarebbe davvero una generalizzazione indebita. Quello che possiamo dire è che anche P.Oxy. 4455 mostra un tipo di commento improntato a quella Sachphilologie funzionale ai contenuti del testo, di cui si è detto per il commentario aristarcheo: rientra dunque nella medesima visione filologico-esegetica, in relazione alla quale non si può comunque parlare di un commento di carattere specificamente storiografico e improntato a questioni di metodo. Un ultimo punto a proposito di P.Oxy. 4455 è che vi troviamo citato Didimo (col. I 12–14) per il significato puntuale di una parola di Erodoto 52, 2, precisamente le πύλαι del fiume Halys. È corretto osservare, con Haslam, che questo non implica necessariamente che Didimo abbia scritto un commento a Erodoto, anche se il fatto non desterebbe una grande sorpresa, considerato anche il precedente di Aristarco.16 Secondo Montana, la spiegazione didimea potrebbe non avere un contenuto strettamente glossografico (Erodoto dice πύλαι nel senso di γεφύραι, come ipotizza Haslam), ma spiegare l’espressione brachilogica di Erodoto chiarendo che le πύλαι in effetti davano direttamente sul fiume: la cosa è  16 Haslam, P.Oxy. LXV, cit., 57.

Gli storici greci e la filologia di età ellenistico-romana  

tutt’altro che sicura, ma se si trattasse di una esegesi di Sachphilologie, ne risulterebbe un poco rafforzata l’ipotesi che la nota provenga non da un lessico di Didimo bensì da un hypomnema.17 In ogni caso, l’autore del commentario è postdidimeo e quindi si colloca fra il I e il III sec. d.C. (data del papiro). È molto difficile mettere a fuoco il problema se si tratti anche in questo caso di una selezione, quindi di excerpta dallo hypomnema di un singolo autore oppure di una fusione miscellanea da una o più opere esegetiche: certo si osserva che la copertura del testo è tale da non comportare salti, abbiamo cioè annotazioni a tutti i paragrafi della porzione di testo interessata, il che farebbe propendere per un’opera miscellanea dei primi secoli dell’età imperiale (quando il commentario aristarcheo era evidentemente ancora letto, se non altro nella forma epitomata di P.Amh. 12). Un altro paio di testimonianze possono essere chiamate in causa a proposito dell’interesse di Didimo per Erodoto. Appare sicura quella che si trova in una voce di Esichio (Μυκερίνα, μ 1824, la cui fonte è Diogeniano di Eraclea, II sec. d.C.), secondo la quale pare che Didimo identificasse nel nome “Mykerina” la città egiziana di Sais (e non Memphis come altri) sulla base appunto di un passo di Erodoto II 129–130: il frammento è attribuito congetturalmente all’opera Lexis komikè.18 Pur ricordando ancora una volta come sia pericoloso e azzardato tirare troppe conseguenze da testimonianze così esigue, non possiamo non osservare che qui compare l’erudita Sachphilologie funzionale alla spiegazione di un testo (qui, essendo di Erodoto il parallelo utilizzato a scopo esegetico, il testo di riferimento poteva non essere Erodoto stesso, bensì per esempio un’opera comica). Soggetto a gravi problemi di attribuzione è invece quanto si legge in un passo del grammatico latino Prisciano (VI sec. d.C.) a proposito di un dato lessicale: un termine per il quale Didimo, nella sua opera sull’analogia presso i Romani (uno dei trattati di contenuto grammaticale di cui abbiamo notizia),19 distingueva l’uso

 17 Montana 2009a, 253–254. 18 Schmidt 1854, 35: Hesych., s.v. Μυκερίνα, μ 1824 (nome che troviamo solo in Esichio come città egiziana, in Erodiano e Teognosto solo come «nome di città»; s.v. Μουκερίναι in Esichio, μ 1729) = Memphis, ma per Didimo = Sais (solo = Sais in 1729), in base al legame di Sais con Micerino testimoniato da Erodoto (II, 129–130). La voce di Esichio è attribuita da Schmidt alla Lexis komikè di Didimo, sulla base del fatto che nell’epistola a Eulogio Esichio cita questa opera lessicografica di Didimo come una delle sue fonti (Latte I, p. l incipit, cfr. Braswell 2013, 90–92), anche se mi pare più probabile che Didimo fosse già stato impiegato da Diogeniano piuttosto che aggiunto da Esichio. 19 Schmidt 1854, 345–349; Braswell 2013, 90–92.

  Gli storici greci e la filologia di età ellenistico-romana ionico e l’uso attico citando Erodoto.20 Tuttavia non solo questo singolo passo (e alcuni altri di Prisciano), ma anche l’opera stessa è da alcuni attribuita a Claudio Didimo, vissuto una o due generazioni più tardi del Calcentero; per non dire della non abbandonata ipotesi che i due siano la stessa persona.21 Per completare il quadro dell’esegesi erodotea testimoniata dai papiri, aggiungiamo che conosciamo tre testimoni, sempre della prima età imperiale, nei quali si trovano annotazioni interlineari o marginali, che si collocano nei primi tre secoli d.C.: P.Oxy. 19 (sec. II–III d.C.), P.Oxy. 1092 (sec. II d.C. ex.)22 e P.Oxy. 1619 (sec. I–II d.C.), che contiene un paio di note di carattere esegetico-parafrastico.23 Possiamo a questo punto cercare di ricostruire la seguente (magra) “storia” di quello che sappiamo del lavoro dei grammatici di età alessandrina su Erodoto. Nei frammenti superstiti di Aristofane di Bisanzio si trovano una citazione di Tucidide e due di Senofonte, ma nessuna di Erodoto, cosa probabilmente dovuta al caso e alla scarsità di quanto conservato.24 Aristarco scrisse un hypomnema erodoteo, quanto meno al libro I. Che nel suo solco Didimo abbia scritto un commentario è un’ipotesi non infondata, ma per la quale non ci sono prove decisive, anche se possiamo dire che studiò Erodoto, ne ricavò materiali di carattere linguistico-lessicale e dialettologico e lo utilizzò come testo di confronto a scopi esegetici (vedi sopra Esichio μ 1824). In età postdidimea, cioè nei primi secoli dell’età imperiale, circolava una epitome del commentario aristarcheo e furono prodotti nuovi commentari, per quanto ne sappiamo (P.Oxy. 4455) con un orientamento esegetico-erudito simile per tenore al precedente alessandrino testimoniato per Aristarco. Come dicevamo, sappiamo che nei primi secoli dell’età imperiale non pochi eruditi ed esegeti si occuparono degli storici e anche di Erodoto: Alessandro Cotieo (I–II sec.);25 Minucio Pacato Ireneo (I sec.), se è corretta l’identificazione con lui di un Ireneo citato per un commentario a Erodoto;26 un Apollonio menzionato in Etym.Magn. 552, 5 e 722, 22, che potrebbe

 20 Schmidt 1854, 348–349; Braswell 2013, 91–92 (Priscian., De figuris numerorum, III 411.9 – 412.2 Hertz–Keil = 17.14 – 18.4 Passalacqua 1987). 21 Status quaestionis in Braswell 2013, 90–92: vd. Montana 2006a e 2013; Montanari 1997a. 22 S. West 2011, 81. 23 Gli scoli di P.Oxy. 1619 si trovano nell’edizione di Rosén 1987–1997, ad III 54, 2 e III 60, 3, e in Asheri 1990, 199–203 (con traduzione italiana). 24 Tucidide nel fr. 33 Slater, Senofonte nei frr. 16 e 17 Slater; Pfeiffer 1973, 311, per le Lexeis di Aristofane afferma «vi sono riferimenti ad Erodoto, Tucidide, Senofonte, Isocrate, Demostene»: in realtà Erodoto non risulta dai frammenti conosciuti, ma certo è assai probabile che ci fosse. 25 Vd. Montana 2005. 26 Vd. Regali 2007.

Gli storici greci e la filologia di età ellenistico-romana  

essere l’autore del lessico omerico (fine del I sec.);27 degli hypomnemata a Erodoto, Tucidide e Senofonte sono attribuiti a Erone di Atene, figlio di Kotys, nel quale del tutto ipoteticamente si può vedere un retore del I sec.;28 e altri ancora.29 Rivolgiamoci ora a un altro tipo di fonte, la letteratura erudita conservata nei codici di età bizantina: raccolte scoliografiche, raccolte lessicografiche, trattati vari di contenuto erudito e grammaticale. La scoliografia per le opere in prosa è complessivamente assai meno ricca di quella per i maggiori poeti conservati attraverso il filtro dell’età bizantina: si rispecchia in questo un dato storico reale, che riguarda le modalità di sviluppo e la portata della filologia nella sua fioritura di età ellenistico-imperiale. Per quanto concerne gli storici, il corpus scoliografico più considerevole è quello a Tucidide ed è stato oggetto di una certa attenzione da parte degli studiosi; esiste una manciata di scoli a Erodoto, per lo più molto brevi,30 e si conosce anche uno sparuto manipolo di scoli a Senofonte, Strabone e Plutarco:31 tutto questo attende ancora di essere indagato in modo approfondito.32 Accanto alla scoliografia, e in stretta relazione per le frequenti osmosi di materiali, si colloca la lessicografia, che affonda parimenti le sue radici in età alessandrina e imperiale. Uno dei frutti del lavoro dei grammatici erano materiali costituiti da glossai, cioè parole “difficili” trovate in autori classici e legate a una interpretazione puntuale, e da lexeis, cioè parole che per una ragione qualunque fornivano motivo o aggancio per la trattazione di un argomento  27 Cfr. 1895a, 140; Jacoby 1913, 514. 28 Kroll 1912. 29 Un quadro in Jacoby 1913, 513–515, e nel già citato Matijašić 2013, 220 e nt. 17. Sintesi sulla filologia erodotea antica in Rosén 1962, 211, 217–234; Dickey 2007, 53–54. 30 Non abbiamo una raccolta e un’edizione sistematica degli scoli a Erodoto; quanto edito si trova in: Stein 1871, II: scholia pp. 431–440, Λέξεις ’Ηροδότου 441–482 (Latte/Erbse 1965, XII sg. e 191–230); questo materiale è riportato e utilizzato nell’apparato dell’edizione Teubner di Rosén (1987–1997), cfr. Praef., p. XLIX; le lexeis sono edite anche in Rosén 1962, 217–231; scholia e lexeis sono riprodotti con traduzione italiana in fondo ai volumi dell’edizione di Erodoto della Fondazione Valla, a cura di vari autori (cfr. Asheri 1988, 240); una recente acquisizione in Cantore 2012. Gli scoli a Tucidide sono ben più ricchi di quelli erodotei, ma non c’è ancora una buona edizione né uno studio approfondito: si utilizza ancora l’edizione teubneriana Hude 1927; cfr. Luschnat 1954; Tosi 1983; Dickey 2007, 55. Una nuova edizione preparava da tempo lo scomparso A. Kleinlogel, con progressi cospicui anche per la storia della formazione del corpus: K. Alpers, dal Nachlass di Kleinlogel e con l’aiuto di I.C. Cunningham, si sta occupando di completare l’edizione degli Scholia Graeca in Thucydidem (pubblicazione prevista per fine 2014 come vol. XV della Samml. Gr. Lat. Gramm.). 31 Senofonte: Dindorf 1855a, 381–396. Strabone: Diller, 1954, 29–50. Plutarco: Manfredini 1975 e 1979. 32 Dickey 2007, 53–55, 70–71.

  Gli storici greci e la filologia di età ellenistico-romana linguistico, lessicale, filologico o antiquario d’ogni genere, e che fungevano da Stichwort o lemma per raccolte dapprima preferibilmente tematiche e in seguito disposte sempre più rigorosamente in ordine alfabetico. Le imponenti Lexeis di Aristofane di Bisanzio, articolate in diverse sezioni tematiche, costituirono un modello per metodo, vastità di conoscenze in ogni campo e ricchezza dei materiali. Dagli allievi di Aristofane, da Aristarco e dai suoi successori fino a Didimo, e poi in tutta l’età imperiale, tardoantica e bizantina, l’attività lessicografica non cessò di svilupparsi e la produzione fu enorme: lessici dialettologici, sinonimici, etimologici, di singoli autori o generi letterari, onomastici, compilazioni su ambiti particolari. In questo enorme fiume trova posto anche una corrente rappresentata dalle parole degli storici. Come accennavamo, i resti di letteratura erudita erodotea che si sono conservati nei codici medievali sono piuttosto poveri. Abbiamo un piccolo gruppo di scoli e due raccolte di lexeis, che furono pubblicati da H. Stein nel 1871, in appendice alla sua edizione di Erodoto, e ripresi in seguito.33 Delle due raccolte di lexeis, una è disposta secondo l’ordine del testo e divisa per libri, l’altra in ordine alfabetico.34 Quella che segue il testo sembra la più antica e potrebbe includere materiali derivanti da un commentario, dato che in qualche caso rivela uno scopo esegetico puntuale sul passo interessato: per esempio una lexis a III 156 discute il significato della forma verbale καταπροΐξεται e cita due passi di Archiloco e gli epimerismi di Erodiano. La raccolta alfabetica comprende in gran parte gli stessi materiali, con qualche modifica per aggiunta o omissione. Quanto agli scoli conservati, la loro quantità, come di norma, va diminuendo con il procedere dell’opera, fino a ridursi a poche righe per gli ultimi libri. Essi appartengono certamente a una redazione abbastanza tarda, nella quale sono entrati anche elementi cristiani, ma è assai probabile che includano resti di origine più antica. Pur nell’esiguità del testo, troviamo citato un bel gruppo di autori: Omero, Archiloco, Dionisio di Mileto, Democrito, Aristofane, Iseo, Erodiano. Questo testimonia in favore di un’origine da buona erudizione antica, di età alessandrina o prima età imperiale. Un caso interessante a questo proposito è quello della parola σάγαρις, che è commentata — come abbiamo visto — nel frammento di hypomnema aristarcheo P.Amh. 12 come lemma di I 215, in due scoli a I 215 e IV 5 e nelle lexeis al libro I, dunque per I 215:35 probabilmente per

 33 Stein 1871. 34 Nelle edizioni di Rosén 1987–1997 e Asheri 1988 le due redazioni sono fuse secondo i libri erodotei, mentre in Stein erano separate. È possibile o forse anche probabile che nei mss. ci siano altri materiali. 35 Vd. p. 225, r. 70 Rosén 1987–1997 = p. 157, r. 76 Asheri 1988.

Gli storici greci e la filologia di età ellenistico-romana  

effetto di epitomazioni differenti, non c’è coincidenza né dei due scoli fra loro né di uno dei due scoli con il papiro, tranne che per il significato di σάγαρις glossato πέλεκυς, che è comune al papiro, allo sch. I 215 e alle lexeis. L’elemento esegetico di carattere in ampio senso linguistico ha una presenza preponderante: pressoché unica nelle lexeis, maggioritaria negli scoli. Esso affonda certamente le sue radici nell’erudizione grammaticale di età alessandrina e i papiri, come abbiamo visto, ce ne attestano la presenza (del resto per nulla sorprendente) almeno da Aristarco all’età imperiale: rientra in una linea tradizionale di studi di tipo linguistico-lessicale-dialettologico, che riguardava un po’ tutti i dialetti letterari e che interessò pressoché costantemente i filologi (come strumento esegetico risaliva alla concezione della glossa come elemento del linguaggio poetico espressa da Aristotele nella Poetica e nella Retorica), per arrivare alla trattatistica dialettologica di età bizantina (pensiamo al ricco trattato sui dialetti di Gregorio di Corinto, XI–XII sec., oppure specificamente per lo Ionico al Περὶ Ἰάδος di Manuele Moscopulo, XIII–XIV in. sec.). Nell’ambito della dialettologia antica, è naturale che Erodoto giocasse un ruolo di primo piano nello studio dello Ionico (peraltro il dialetto in cui furono scritte le prime opere in prosa di cui abbiamo notizia dalle fonti antiche), sul piano sia morfologico sia lessicale: Dionigi di Alicarnasso offre una prima classificazione laddove (Ad Pomp. 3, 16) indica in Erodoto il canone migliore per lo Ionico e in Tucidide il migliore per l’Attico. Si tratta di un fenomeno ben noto, che ricordo soltanto per inquadrarvi la cospicua presenza dell’elemento linguistico nei magri resti della filologia erodotea antica. Un secondo elemento riscontrato nei papiri è quello che abbiamo chiamato Sachphilologie funzionale ai contenuti e all’interpretazione di un testo. Mi pare interessante e significativo rilevare che una simile compresenza si ritrova nello sparuto manipolo di scoli erodotei, e fa qualche comparsa persino nelle lexeis, come nel caso citato sopra della lexis a III 156. Solo per fare qualche esempio: a I 71 uno scolio discute il significato di ἀναξυρίς = sandalo di cuoio; a I 85 si spiega il valore di εὐεστώ citando Democrito; a I 125 si ragguaglia sul nome dei Persiani con un excursus mitografico; e ci sono altri casi. La presenza di due elementi così caratteristici della filologia erudita di età ellenistico-romana non è certo sorprendente e non costituisce un dato particolare della filologia erodotea antica: abbiamo già osservato che esso corrisponde alla tipologia comune del commentario erudito alessandrino. Esso ci permette tuttavia due considerazioni, che non mi sembrano inutili. La prima è in positivo, e come tale deve essere ritenuta valida benché si basi su testimonianze abbastanza esigue: la filologia erodotea antica rientra nel mainstream della tradizione esegetico-erudita che dai maggiori filologi alessandrini arriva alle sintesi erudite

  Gli storici greci e la filologia di età ellenistico-romana di età imperiale e tardoantica, e infine bizantina. La seconda è in negativo, e come tale è da prendere con molta prudenza in quanto soggetta a possibili smentite: riprendendo quanto dicevamo all’inizio a proposito del commento aristarcheo, constatiamo che per Erodoto non abbiamo quello che potremmo definire un commento di carattere specificamente storico e storiografico, cioè rivolto alle problematiche specifiche riguardanti il metodo dello storico nei suoi vari aspetti. Ripeto che il quadro potrebbe mutare in seguito a nuovi ritrovamenti, che aggiungano altri elementi: in questo caso, il primo punto (esegesi linguistico-lessicale + Sachphilologie) rimarrebbe valido anche se non più unico, la seconda considerazione (assenza di discussioni sul metodo) potrebbe rimanere valida o essere smentita a seconda delle caratteristiche di una eventuale nuova testimonianza. Occupiamoci ora di Tucidide. Per quanto riguarda i papiri, il materiale è nettamente più ricco che per Erodoto: abbiamo due grossi pezzi di hypomnemata, ai quali si è di recente aggiunto un frammento probabilmente di hypomnema ma assai esiguo; conosciamo poi dieci frammenti del testo con note marginali di varia portata (anche solo semeia), un paio di voci in lessici generali (P.Oxy. 2087 e P.Oxy. 1803), un breve Bios inserito in una raccolta di biografie di uomini illustri (P.Oxy. 1800 + 2081) e citazioni in trattati di vario genere. Gli hypomnemata sono: 1) P.Vindob. G 29247 (MP3 1535), sec. III ex., commento a Tucidide I 1–9: Gerstinger 1925; 2) P.Oxy. 853 (MP3 1536), sec. II ex., commento a Tucidide II 1–45: Grenfell/ Hunt 1908, 107–149; 3) P.Berol. 21313 (MP3 1522.04), sec. I ex. – II in., può restituire un minuscolo frammento di commento a Tucidide III 83, 4. Purtroppo si ricostruisce solo una porzione di testo tucidideo, che pare essere il lemma, seguito da ὅτι e non dalla continuazione di Tucidide: è questo il principale indizio in favore del commentario invece che della copia dell’opera tucididea. Se si tratta di un commentario, è il più antico conosciuto.36 I tre testimoni sono tutti datati ai primi tre secoli dell’età imperiale. Il papiro di Vienna appartiene a un codice di grande formato, di cui rimangono parti di un bifoglio: la porzione di testo conservata è considerevole (molto maggiore dei due reperti erodotei) e copre i capitoli 1–9 del I libro, vale a dire il proemio e la prima parte della Archaiologìa, ma presenta serie difficoltà di ricostruzione e di  36 Edizione di P.Berol. 21313: Parsons/Pellé 2012; il recto del frustulo era stato anticipato da L. Canfora in QS LXXIV (2011), 97–98.

Gli storici greci e la filologia di età ellenistico-romana  

interpretazione, e bisogna dire che l’interesse suscitato finora è stato molto ridotto se rapportato all’importanza del testimone. Più che mai necessaria è una nuova edizione, fondata su un approfondito lavoro sull’originale e corredata di un adeguato commento. Sulla base di quanto si trova nell’editio princeps e in alcuni interventi successivi,37 possiamo per il momento affermare quanto segue. Il commentario deve essere stato prodotto nella seconda metà o forse verso la fine del II sec., perché cita come autore noto Elio Aristide, morto dopo il 187. Esso appartiene senz’altro a un ambiente atticista: nel f. 2a, rr. 20–28, si trovano osservazioni di carattere stilistico-retorico sul proemio tucidideo, in connessione con Dionigi di Alicarnasso (il trattato Su Tucidide e la Seconda Lettera a Ammeo sulle particolarità dello stile di Tucidide) e con la citazione di Demostene e di Elio Aristide, il che corrisponde alla scelta dei tre autori indicati come i migliori modelli della lexis attikè, appunto Tucidide, Demostene e Elio Aristide. Tuttavia, osserva Luschnat,38 resterebbe deluso chi per questo motivo si aspettasse un’abbondanza di osservazioni linguistico-lessicali riconducibili al tipo dei lessici atticisti: il commento si sofferma piuttosto e in modo preponderante, nello stile degli zetemata, su problemi vari del contenuto e introduce osservazioni storico-culturali, come quella sulla pirateria esercitata dai Greci antichi e quella sulla ginnastica presso i Greci e i barbari, oppure mitografiche, come le notizie sulla genealogia di Euristeo. Interessante è anche il fatto che con l’espressione ἃ ἄλλοι ἄλλως l’autore alluda con ogni evidenza a diversi commenti di altri esegeti a lui noti. Nel f. 1a il commento riguarda Tucidide I 1, 2, la celebre definizione della guerra come κίνησις μεγίστη: «il più grande sconvolgimento prodottosi fra i Greci e in una parte del mondo non greco, insomma per la gran parte dell’umanità». L’alternanza fra lemmi e parti esegetiche non è del tutto chiara allo stato attuale, ma certo il commento a questo passo occupa buona parte della pagina ed è molto più ricco degli scoli tramandati.39 Dobbiamo dunque ancora una volta chiamare in causa il concetto di Sachphilologie funzionale ai contenuti e all’interpretazione di un testo, accanto all’elemento di carattere linguistico, che in questo commentario (almeno allo stato attuale della conoscenza di un frammento per il quale sembra abbastanza chiaro che molti progressi siano possibili) assume in prevalenza l’aspetto dell’analisi stilisticoretorica (principalmente nel passo citato sul proemio) rispetto a quello più squisitamente lessicale.

 37 Luschnat 1954, 29–31; Montana 2009, 168; Stroppa 2008, 52–53; 2009, 301 e 315. 38 Luschnat 1954. 39 Sch. Thuc. p. 2, 4–9 Hude (1927); cfr. Stroppa 2008, 53; 2009, 315.

  Gli storici greci e la filologia di età ellenistico-romana L’altro frammento di hypomnema tucidideo è P.Oxy. 853, un rotolo del tardo II sec. d.C., di cui si conservano parti di ben 19 colonne (oltre a piccoli frammenti non collocati): anche qui dunque possiamo leggere una considerevole porzione di testo, che copre Tucidide II 1–45 (con parti perdute) ed è redatto con chiara evidenza nella consueta forma che alterna lemma + commento.40 L’autore cita Dionigi di Alicarnasso, che costituisce il terminus post quem, mentre gli ultimi decenni del II sec. d.C. (data del papiro) costituiscono il terminus ante quem. Una particolarità da sempre messa in evidenza è la lunghissima nota che risponde puntualmente alle critiche di Dionigi di Alicarnasso all’esposizione storiografica di Tucidide: sarebbe tuttavia un errore metodologico fare automaticamente di questo aspetto, certamente di grande importanza, la caratterizzazione unica, generale e sistematica del commentario. Vediamo qualche dato. La col. I comincia esattamente con l’incipit del libro II: la prima nota riguarda proprio la frase iniziale, occupa due righe, è molto mutila ma certo dice qualcosa sull’avverbio ἐνθένδε; la seconda occupa una riga e mezza, riguarda l’espressione di poco successiva παρ᾿ ἀλλήλους e ne rileva l’uso da parte dell’autore per il più usuale πρὸς ἀλλήλους; segue come lemma la frase finale di II 1, 1 (un paragrafo che occupa in tutto cinque righe di edizione moderna e dal quale sono presi ben tre lemmi): γέγραπται δὲ ἑξῆς ὡς ἕκαστα ἐγίγνετο κατὰ θέρος καὶ χειμῶνα, «Il racconto riporta i fatti nell’ordine in cui accaddero, per estati e inverni». È la celebre enunciazione del criterio cronologico scelto da Tucidide, alla quale il commentatore dedica un’annotazione che va dalla r. 5 della col. I fino alla r. 9 della col. IV, per un totale di 109 righe. Questa trattazione risponde esplicitamente a Dionigi di Alicarnasso su tre punti («Dionigi di Alicarnasso nel trattato Su Tucidide critica Tucidide su non molti argomenti e si sofferma soprattutto su tre punti»): l’esposizione per estati e inverni; la conseguente frantumazione del racconto, per cui si passa da un episodio a un altro senza completare il precedente; il fatto che Tucidide non parte dalla causa che egli stesso indica come quella vera, ma espone le cause indicate dagli altri. Alla riga IV 10, terminato questo tema, il commento riprende con un lemma di II 2, 1 (cioè subito dopo la fine del paragrafo precedente) e prosegue fino a II 45, 2, con annotazioni la cui estensione va da due a dieci righe, superate solo in tre casi da note di 14 o 15 righe. Gli argomenti sono vari e contemplano notazioni di carattere linguistico-lessicale e grammaticale insieme a variegate spiegazioni di problemi e difficoltà del testo e note su questioni storiche e geografiche: abbiamo insomma il medesimo insieme di elementi già trovato nei commentari  40 Grenfell/Hunt 1908, 107–149; cfr. Luschnat 1954, 25–29; L. Canfora, Teorie e tecnica della storiografia classica, Roma/Bari 1974; Montana 2009, 168.

Gli storici greci e la filologia di età ellenistico-romana  

esaminati finora. P.Oxy. 853 tuttavia è un caso interessante per illustrare quanto vari potevano essere i contenuti degli hypomnemata e quanta libertà il commentatore si potesse prendere nello sviluppare un tema o un altro a seconda dei suoi gusti e interessi, ma anche per mettere in guardia dal trarre conclusioni affrettate quando consideriamo un frammento, anche se non piccolissimo. Se avessimo soltanto un paio delle prime colonne (magari con la prima mutila all’inizio), penseremmo senz’altro a un trattato sulla composizione storiografica oppure a un opuscolo polemico nei confronti di Tucidide e della sua esposizione, mentre si tratta proprio di un tipico esempio di commentario con i suoi lemmi e le sue ben differenziate note esegetiche, che tra l’altro per i suoi scopi esegetici cita poeti come Callimaco, Euripide, Omero, Pindaro. Se invece le prime colonne fossero perdute e non avessimo la lunga annotazione che va da col. I 5 fino a col. IV 9, potremmo pensare semplicemente a un commentario soltanto con interessi simili ai precedenti che abbiamo visto e senza lo sviluppo particolare del tema della struttura espositiva tucididea. In questo caso invece lo hypomnema, oltre alla tipologia di contenuti esegetici già riscontrati, rivela anche un interesse incentrato sulle modalità di esposizione della storia, il racconto e l’economia della materia, che al momento non ha paralleli. Torneremo su questo aspetto. Il fatto che questo ci sia testimoniato per Tucidide e non per Erodoto sarà un caso, dovuto anche alla scarsità dei testimoni disponibili, oppure è un fenomeno che va interpretato altrimenti? E il fatto che i due soli frammenti di commentario tucidideo che possediamo, essendo di età imperiale, si riallaccino entrambi, anche se per ragioni diverse (una linguistico-stilistica, l’altra sulle modalità del racconto storiografico), alla trattazione di Dionigi di Alicarnasso sarà ancora una volta uno scherzo della sorte oppure indicherà un’influenza forte e condizionante? Sono domande sulle quali è difficile prendere posizione, anche perché il quadro può cambiare e si rischia di incorrere nel caso di un piccolo papiro che fa cadere una grande ricostruzione storico-culturale (come dicevamo all’inizio a proposito del commento di Aristarco a Erodoto). Azzarderei qualcosa su uno di questi punti, quello che riguarda l’aspetto linguistico-stilistico. Il peso dell’Atticismo nella cultura di età imperiale fa pensare che effettivamente la strada indicata da Dionigi di Alicarnasso sia stata assai influente, considerato che l’Atticismo fu in primo luogo un problema di scelta di modelli e Tucidide giocò un ruolo di primo piano a questo proposito (assieme a Demostene, ai quali si aggiunse poi Elio Aristide).41 Nel suo trattato Su Tucidide, Dionigi dice rapidamente qualcosa sugli storici precedenti al suo autore (capitolo 5): Erodoto è liquidato in un brevissimo paragrafo (5, 5), nel  41 Per Dionigi vd. De Jonge 2011.

  Gli storici greci e la filologia di età ellenistico-romana quale si riconosce il suo progresso rispetto ai predecessori nella scelta dell’argomento e si sottolinea il valore di avere deciso di raccontare una grande quantità di avvenimenti d’Europa e d’Asia; infine, due parole sullo stile: «Per quanto riguarda lo stile, Erodoto gli ha conferito le qualità (ἀρεταί) che erano state trascurate dagli storici precedenti (oἱ πρὸ αὐτοῦ συγγραφεῖς)». Mentre lo Ionico di Erodoto restava oggetto di interesse per eruditi e studiosi di lessico e dialettologia, Tucidide rientrava nel vivo del dibattito atticista sui modelli di lingua e stile. E inoltre, aggiungiamo, magari in connessione più o meno stretta con questo aspetto stilistico-retorico, veniva stimolata anche una discussione sulle modalità della sua esposizione storiografica e sulla sua organizzazione del racconto, che il commentario P.Oxy. 853 ci testimonia particolarmente vivo nei primi secoli dell’età imperiale anche nell’ambito dell’esegesi erudita. Del resto, che in età imperiale la storiografia fosse ricondotta sotto il grande ombrello della retorica non è certo una novità.42 Tutto questo si colloca nei primi tre secoli dell’età imperiale, e abbiamo già accennato sopra alle non poche tracce di svariati interessi da parte degli eruditi per le opere degli storici maggiori in questo stesso periodo, che è quello della Seconda Sofistica e dell’Atticismo.43 È il caso che ci ha negato per Tucidide una testimonianza sull’esegesi di età alessandrina, come quella del frammento aristarcheo per Erodoto, oppure, come pensava Wilamowitz, la filologia alessandrina non si occupò di esegesi tucididea e l’interesse per Tucidide si sviluppò solo più tardi nell’ambito dell’Atticismo? Per vero dire, più o meno contemporaneamente H. Usener sosteneva che Dionigi di Alicarnasso conoscesse scritti grammaticali alessandrini su Tucidide,44 il che del resto sembra provato da quanto Dionigi stesso afferma proprio in Su Tucidide 51, 1 e 55, 2. In 51, 1 leggiamo: «È facile contare [scil. sono poche] le persone in grado di comprendere tutto quanto di Tucidide, e anche costoro non capiscono certe cose senza esegesi» (χωρὶς ἐξηγήσεως);45 in 55, 2: «Non bisogna ammirare né imitare le espressioni enigmatiche e difficili da capire e bisognose di spiegazioni grammaticali» (γραμματικῶν ἐξηγήσεων δεομένας).46 Che si tratti o meno di veri e propri commentari, non mi pare dubitabile che si parli di un lavoro esegetico su Tucidide.47 Certo è difficile dire se questi commenti tucididei fossero contemporanei a Dio-

 42 Luschnat 1954, 20, nt. 2: «la storiografia fu totalmente sequestrata dalla retorica». 43 Vd. supra, e ntt. 25–29; cfr. Matijašić 2013, 220 e nt. 17. 44 Cfr. Luschnat 1954, 23; Tosi 1983, 8–9; Montana 2009, 169–170; De Jonge 2011, 451–456. 45 G. Aujac traduce «sans le secours d’un commentaire grammatical», 1991, 118. 46 Aujac 1991, 125 traduce «nécessitant des explications grammaticales». 47 Cfr. De Jonge 2011, 451–456.

Gli storici greci e la filologia di età ellenistico-romana  

nigi o antecedenti e di quanto: per lo meno, però, siamo già risaliti all’età augustea, che vuol dire l’età di Didimo, anche se ancora una volta dobbiamo ammettere che non ci sono prove certe che Didimo abbia prodotto un commento a Tucidide, benché sia ben testimoniato che egli scrisse una Vita di Tucidide, citata tre volte nella Vita di Tucidide di Marcellino.48 Abbiamo in verità una testimonianza presso Ammonio (sec. I–II d.C.), De adfinium vocabulorum differentia, 451 Nickau, dove Didimo è citato a proposito della diffenza fra συμμαχεῖν e ἐπιμαχεῖν, per la quale viene addotto un passo di Tucidide (I 44, 1),49 ma non si tratta certo di una prova conclusiva. Pfeiffer dice: «Non potrebbe far molta meraviglia che Aristarco abbia anche scritto il primo commentario a Tucidide [scil. come scrisse il primo commentario a Erodoto], perché non si può mettere più a lungo in dubbio che Didimo poté servirsi di una precedente opera filologica alessandrina»:50 sembra dunque dare per scontata l’esistenza del commentario di Didimo e ipotizzare con ottima probabilità quello di Aristarco.51 Purtroppo mi pare che non venga utilizzata una buona testimonianza offerta dallo scolio di Aristonico a Od. III, 71:52 vi si discute di un riferimento alla pirateria in Od. III 71–74 (versi che ricorrono uguali in IX 253–255) e Aristarco è esplicitamente citato per aver addotto la testimonianza di Tucidide I 5, 1, sul fatto che i Greci antichi non ritenevano turpe praticare la pirateria (καθάπτεται δὲ καὶ Θουκυδίδου Ἀρίσταρχος λέγοντος κτλ.). Su questo tema si sofferma, come abbiamo notato sopra, anche il commentario P.Vindob. 29247 proprio a proposito di Tucidide I 5, 1 in alcune righe purtroppo mutilate (f. 2a, rr. 16–20): si tratta comunque di un buon parallelo allo scolio odissiaco e sembra del tutto plausibile pensare che la nota del commentario papiraceo avesse alle spalle l’osservazione di Aristarco. E neppure viene chiamato in causa a questo proposito uno scolio a Tucidide che, pur nel testo probabilmente immiserito che leggiamo nell’edizione di Hude, a margine di III 84 parla chiaramente di parti obelizzate (τὰ ὠβελισμένα) che «nessuno degli interpreti riteneva fossero di Tucidide».53 Con questo, mi sembra difficile sfuggire all’idea che ci sia dietro un commentario piuttosto antico, probabilmente di età ellenistica (Aristarco stesso?): certamente è un’ipotesi, ma direi che prende abbastanza corpo.  48 Schmidt 1854, 321–333; Braswell 2013, 86–88. 49 Schmidt 1854, 333–334; Braswell 2013, 88. 50 Pfeiffer 1973, 349–350. 51 Cfr. Luschnat 1954, 23; Tosi 1983, 8–9; Montana 2009, 169–170. 52 Pontani 2010, 29–30; Luschnat 1954, 30, nt. 1, menziona questo scolio in parallelo con il passo di P. Vindob. citato supra (f. 2a, rr. 16–20). 53 Sch. Thuc. p. 216, 10–11 Hude: τὰ ὠβελισμένα οὐδενὶ τῶν ἐξηγητῶν ἔδοξε Θουκυδίδου εἶναι.

  Gli storici greci e la filologia di età ellenistico-romana Infine, anche un frammento di Aristofane di Bisanzio, conservatoci da Eustazio, deve essere preso in considerazione: Aristofane discuteva dei significati del verbo ἀμύνεσθαι, adducendo passi di Alcmane e di Tucidide.54 Abbiamo già ricordato che in Aristofane di Bisanzio non troviamo citazioni di Erodoto, però è vero che, oltre a Tucidide, troviamo due citazioni di Senofonte (anche se non sicure al cento per cento). Si può essere più o meno fiduciosi su tutti questi indizi, che in effetti non sono prove conclusive, soprattutto se presi separatamente, ma non bisogna trascurare il loro insieme né dimenticare che troppo abbiamo perduto per non valorizzare anche piccole tracce dell’attività esegetica degli Alessandrini su autori come Erodoto e Tucidide, tracce che peraltro si radunano intorno a una prova vera e importante come lo hypomnema di Aristarco a Erodoto.55 Su queste basi si è affermata l’idea di un’origine antica, anche alessandrina, del materiale esegetico tucidideo confluito negli scoli, nei quali dunque si potranno cercare materiali utili da questo punto di vista. Solo due parole per Senofonte, per non trascurare il terzo membro della triade canonica. I papiri ci offrono ben pochi materiali: alcuni testimoni recano annotazioni marginali o segni critici (uno in particolare della Ciropedia del II sec. d.C.)56 e troviamo Senofonte citato in un paio di lessici alfabetici generali. Dickey e Gudeman ritengono che la popolarità delle sue opere e le tracce, per quanto scarse, facciano pensare che ci siano stati prodotti esegetici, che purtroppo si sono perduti.57 Anche a me pare probabile, ma dovremo aspettare che qualche papiro spunti fuori a illuminarci; oppure aspettare che si studino almeno decentemente quegli scoli senofontei di cui Dindorf diede un’idea nel 1855 con quelli della Anabasi: presto venne fuori altro, ma siamo ben lontani da una edizione e da un’indagine soddisfacenti.58 Forse a questo punto — volendo mirare a un quadro completo — dovrei parlare degli storici frammentari, ma non lo farò per un motivo assai semplice: solo una ricerca molto lunga e impegnativa (certamente di anni) potrebbe portare a qualche risultato attendibile, almeno dal punto di vista implicato dal mio tema in questa sede, vale a dire lo studio degli storici nell’ambito della filologia ed erudizione di età ellenistico-romana (che riguarda anche la poderosa questione di quanto i filologi di età ellenistica conoscessero degli storici per noi perduti e  54 Fr. 33 Slater, già ricordato supra, nt. 24. 55 Cfr. Luschnat 1954.; Tosi 1983, 8–9; Montana 2009, 169–170; De Jonge 2011, 451–456. 56 P.Vindob. G 26010 + 29283 + 29782 = MP3 1551: Cyropaedia con marginalia; 26010 ed. K. Wessely, in MPER VI (1897), 81–97; 29283 + 29782 ed. H. Bannert/H. Harrauer, in WS, NF XIV (1980), 29–37 (con riedizione di 26010, coll. X–XI). 57 Gudeman 1921, 692–693; Dickey 2007, 55. 58 Cfr. Dickey 2007, loc. cit.

Gli storici greci e la filologia di età ellenistico-romana  

come li conoscessero). Questo comporta un’analisi sistematica delle citazioni dei vari storici nella letteratura erudita, che 1) sono una enormità, 2) possono anche essere di seconda mano e/o “di repertorio”, 3) possono o meno implicare elementi di esegesi sul passo che viene citato. Una ricerca molto elementare con il TLG elettronico, senza alcuna elaborazione dei risultati, mostra che nell’enorme insieme costituito da quanto conservato di raccolte scoliografiche, lessici, etimologici e opere grammaticali, le citazioni di storici sono abbondanti, anche tenendo conto del fatto che spesso si tratta di occorrenze in realtà duplicate (a causa del sistema automatico della ricerca) e che alcune sono rappresentate non da citazioni bensì da voci biografiche, soprattutto della Suda. Oltre alle numerose citazioni di Tucidide, Erodoto, Senofonte (questo è l’ordine di frequenza), abbiamo anche storici non conservati come (anche qui in ordine di frequenza) Teopompo, Ellanico, Eforo, Timeo, e si potrebbe proseguire per altri dei numerosi storici frammentari. Più limitato è il numero di tali citazioni nei papiri di opere esegetico-erudite: per esempio, accade che uno storico sia chiamato in causa per commentare un altro autore, per lo più un poeta (così come Omero è citato per commentare Erodoto e Tucidide, può accadere che Tucidide — come abbiamo visto per il caso della pirateria arcaica — sia chiamato in causa per commentare Omero), ma troviamo gli storici frequentemente citati (come già detto) anche nei lessici generalisti, nei trattati o altro (come accade nei frammenti su papiro). Tutto questo rientra nella prima della due modalità individuate all’inizio, ed è chiaro che riguarda tutti gli storici e non solo quelli frammentari. Riprendiamo ancora una volta quanto emerso fin qui. La compresenza di elementi di osservazione di tipo linguistico-lessicale e di esegesi funzionale all’interpretazione puntuale dei contenuti (riassumibile come Sprach- e Sachphilologie) è documentata fin dallo hypomnema di Aristarco e permane nei secoli successivi nei prodotti riconducibili all’esegesi erudita di età ellenistica e imperiale. Ci sono buoni indizi che il dato linguistico-lessicale relativo agli storici fosse oggetto di studio già da parte di Aristofane di Bisanzio, trovando spazio nella grande raccolta delle Lexeis; poi lo abbiamo riconosciuto nei pur magri resti del commento aristarcheo e lo troviamo in Didimo, sia per Erodoto che per Tucidide. Sembra delinearsi con chiarezza il fatto che per lo studio della lingua e dello stile di Tucidide (come pure di Demostene) siano stati fondamentali gli interessi di questo tipo degli eruditi di età ellenistica anteriori a Didimo, che fu il grande bacino collettore di età augustea dell’attività erudita dei secoli precedenti. Sul fertile terreno dell’osservazione degli aspetti di carattere linguistico-lessicale, in una linea di cui si riesce a vedere il percorso nei mutamenti storico-culturali, l’Atticismo sviluppò i suoi studi sulla lingua degli autori che lo interessavano

  Gli storici greci e la filologia di età ellenistico-romana maggiormente per la problematica dei modelli di lessico e di stile, tra cui — come abbiamo visto — c’era Tucidide, mentre per un’altra strada gli studi di grammatica e dialettologia si interessavano non poco anche a Erodoto per il suo Ionico. A un certo punto si sviluppò il dibattito sul metodo storiografico, inteso come studio delle modalità di racconto della storia, della composizione della storiografia come genere letterario e dello stile degli storici: questo accadde verosimilmente per lo più in trattati come quelli di Dionigi di Alicarnasso, ma anche talvolta all’interno di un commentario del tipo consolidato in età ellenistica e peraltro caratterizzato essenzialmente dalle questioni di lingua e lessico e dall’interpretazione puntuale dei variegati contenuti del testo (lasciando momentaneamente da parte la questione della ekdosis, che non è necessariamente presupposta dall’esistenza di un hypomnema). È molto interessante cogliere un momento di congiunzione, fortunatamente conservato nella casualità dei ritrovamenti, di questi due approcci nel commentario tucidideo P.Oxy. 853, dove la trattazione sulla composizione storiografica, in risposta all’opinione di Dionigi di Alicarnasso, viene esposta in una lunga e (almeno per noi) anomala annotazione a un lemma, dunque esattamente nello stile ipomnematico che caratterizza l’intero prodotto, per il resto dotato dei contenuti tipici della sua tradizione di riferimento, che abbiamo più volte ricordato. Riassumendo in estrema sintesi, gli storici — almeno i maggiori — furono trattati dai filologi alessandrini con un approccio simile a quello degli altri autori, vale a dire quello della Sprach- e Sachphilologie. A partire dall’età imperiale — almeno per quanto ne sappiamo finora — due altri fattori fecero la loro comparsa o irruzione ed esercitarono la loro influenza (e la nostra documentazione indica per Dionigi di Alicarnasso un ruolo chiave per entrambi): da una parte l’Atticismo, con i suoi problemi dei modelli di lessico e stile, che riguardò anche gli storici; dall’altra le questioni specifiche della prosa storiografica e del genere letterario “storiografia”, quanto a problemi di composizione e di organizzazione retorica e stilistica. Le riflessioni sul metodo storico in quanto tale (di cosa fare storia, come accertare la realtà dei fatti, le fonti, i documenti, i testimoni, il problema dell’individuazione delle cause degli avvenimenti e del loro concaternarsi come problema fattuale e non solo espositivo) restarono proprie degli storici, mentre le problematiche relative alle scelte compositive, la strutturazione narrativa, lo stile dell’opera storiografica divennero terreno della retorica, che accolse nel suo grande alveo anche la storiografia. Nel nostro caso, il criterio della scansione stagionale o l’ordine in cui vengono esposte le cause degli avvenimenti in Tucidide sono criticati da Dionigi non perché pregiudicano la veridicità della ricostruzione storica, ma perché nuocciono alla chiarezza dell’esposizione: e su questo stesso piano gli risponde l’autore del commentario

Gli storici greci e la filologia di età ellenistico-romana  

P.Oxy. 853.59 Del resto, in età imperiale e tardoantica, nell’ambito dell’educazione, Tucidide è visto come uno degli autori fondamentali della formazione retorica, e in questo rientra anche l’analisi e la discussione sulla sua composizione narrativa. Allo stesso modo, delle questioni della lingua non si occuparono più solo i grammatici per i loro studi, ma anche (di nuovo) i retori per i loro scopi: che Dionigi di Alicarnasso, storico e retore, abbia svolto un ruolo chiave pare dunque plausibile e forse persino naturale. Tutto questo, sotto il grande ombrello della retorica, si aggiunse a una tradizione esegetica di stampo filologicogrammaticale alessandrino che continuava riccamente: ha ragione Pfeiffer a ritenere notevole che un esemplare di commentario aristarcheo sia ancora copiato con il suo titolo quattro secoli dopo Aristarco, nel III sec. d.C.,60 ma adesso abbiamo anche un altro hypomnema a Erodoto di tipo simile, che circolava anch’esso nel III sec., e anche i cospicui pezzi di hypomnemata tucididei dei sec. II e III; e in seguito avremo ancora i commentari di età tardo-antica e poi i corpora di scoli formatisi in età bizantina: a long lasting history. Infine, un passo indietro cronologicamente per un’ultima osservazione, assai rapida. Nello sviluppo della filologia alessandrina vennero prima i poeti: per Pfeiffer perché la filologia fu creazione della poesia, ma probabilmente anche per altri fattori, per esempio il fatto che i poeti erano sentiti come i principali depositari della paideia ellenica, quella che Aristotele aveva identificato come la cultura ed educazione consegnata dal passato come patrimonio imprescindibile, e per la quale in età alessandrina ci si pose il problema anche di conservarne e interpretarne i testi fondamentali, a partire da Omero. Si cominciò quindi con i poeti (così fecero i pionieri Filita e Zenodoto), ma presto si cominciò a interessarsi anche dei prosatori, gli storici e gli oratori (Demostene in testa, ma anche Isocrate, Eschine e altri), e anche gli scienziati (in primis la medicina con Ippocrate). Come credo di avere mostrato qualche tempo fa, iniziando con qualche interesse riscontrabile in Aristarco, si passò infine ai “contemporanei”, cioè ai grandi poeti ellenistici, su molti dei quali, forse su tutti i principali, il grammatico Teone, contemporaneo di Didimo, risulta avere lavorato e scritto commentari.61 Con l’età augustea, dunque, la filologia di età ellenistica ha raggiunto un panorama di attività che copre praticamente ogni genere di autori e opere (e così possiamo avere una vaga idea di quanto abbiamo perso): interpretare la lingua e i testi con le armi della grammatica e della filologia era ormai una conquista che non avremmo mai più perduto.62  59 Cfr. Montana 2009, 168. 60 Pfeiffer 1973, 349. 61 Su questo cfr. Montanari 1995b (= questo volume, cap. 62); 2002b (= questo volume, cap. 58). 62 Montanari 2011; 2015a (= questo volume, cap. 9).

 Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319 In ricordo del συμϕιλολογϵῖν

Dopo la poco esaltante prova del morale delle truppe voluta da Agamennone e conclusa con le percosse al rivoltoso Tersite, una volta ristabilite le condizioni della normale assemblea dell’esercito, Odisseo rivolge agli Achei un lungo discorso (vv. 278–332), nel quale li esorta a perseverare nell’impresa, malgrado siano ormai trascorsi ben nove anni di guerra. Egli ricorda che si deve ancora verificare la veridicità della profezia pronunciata dall’indovino Calcante alla partenza della flotta da Aulide. Allora gli Achei offrivano un sacrificio propiziatorio accanto a una fonte sotto un platano, quando apparve un fatto portentoso: improvvisamente un serpente dal dorso scarlatto balzò da sotto l’altare, si gettò verso l’albero e divorò una covata di otto piccoli passeri più la madre che svolazzava intorno straziata; terminato lo scempio, il serpente fu trasformato in pietra sotto lo sguardo attonito dei presenti. Calcante interpretò il prodigio sulla base del numero delle vittime: nove anni di guerra dovranno passare prima che al decimo gli Achei abbiano ragione di Troia assediata. Odisseo conclude con una nota ottimista, che esorta ad attendere ancora un poco il realizzarsi della predizione. I versi in cui si racconta la parte finale dell’evento prodigioso accaduto alla partenza avevano suscitato discussioni già nell’antichità. αὐτὰρ ἐπϵὶ κατὰ τέκνα ϕάγε στρουθοῖο καὶ αὐτήν, τὸν μὲν ἀρίζηλον θῆκεν θεὸς ὅς περ ἔϕηνε· λᾶαν γάρ μιν ἔθηκε Κρόνου πάις ἀγκυλομήτεω· ἡμεῖς δ᾿ ἑσταότες θαυμάζομεν οἷον ἐτύχθη.

317

320

Quando poi (il serpente) ebbe mangiato i piccoli passeri e la madre, meraviglioso lo rese il dio che l’aveva fatto apparire: pietra lo fece diventare il figlio di Crono dai pensieri tortuosi; e noi restammo attoniti per lo stupore di quanto accaduto.

Il primo problema riguarda la variante ἀρίζηλον / ἀίζηλον al v. 318: il serpente viene definito “meraviglioso, straordinario” (così il testo e la traduzione sopra) oppure “invisibile”; il secondo problema riguarda la presenza o meno del seguente v. 319. Le due questioni sono connesse. Con ἀίζηλον il verso 318 dice che la bestia venne fatta sparire dal dio che l’aveva fatta apparire (con un effetto di antitesi con il verbo ἔϕηνε): il senso sarebbe che il serpente scompare in qualche

https://doi.org/10.1515/9783110772371-054

Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319  

modo alla vista, come prima era detto che esso era apparso sbucando da sotto l’altare (v. 310), e non che fu mutato in pietra come si dice al successivo v. 319. Con ἀρίζηλον il v. 318 dice che il dio fece una prodigiosa meraviglia della bestia che aveva fatto apparire, vale a dire che la trasformò in pietra (essendo Κρόνου πάϊς ἀγκυλομήτεω la specificazione del precedente θεός). Dunque, la scelta testuale al v. 318 è condizionante: con ἀίζηλον il v. 319 è da espungere, con ἀρίζηλον il v. 319 offre l’opportuna spiegazione del prodigio metamorfico e soprattutto dell’aggettivo che vi fa riferimento, secondo una modalità ben nota per la dizione epica arcaica. Notizie sulle opinioni dei filologi alessandrini ci vengono fornite da due scoli di Aristonico, sch. Il. B 318 e B 319, qui citati dall’edizione di Erbse. Sch. B 318 {τὸν μὲν} ἀρίζηλον: ὅτι Ζηνόδοτος γράϕει “ἀρίδηλον” καὶ τὸν ἐχόμενον προσέθηκεν τὸ γὰρ “ἀρίδηλον” ἄγαν ἐμφανές, ὅπερ ἀπίθανον· ὃ γὰρ ἐὰν πλάσῃ, τοῦτο ἀναιρεῖ. λέγει μέντοι γε ὅτι ὁ ϕήνας αὐτὸν θεὸς καὶ ἄδηλον ἐποίησεν. Sch. B 319 a.1 λᾶαν γάρ μιν ἔθηκε Κρόνου παῖς ἀγκυλομήτεω: ἀθετεῖται· πιθανώτερον γὰρ αὐτὸν καθάπαξ πεποιηκέναι ἀϕανῆ τὸν καὶ ϕήναντα θεόν. Sch. B 319 a .2 ἀθετεῖται. προείρηται δὲ ἡ αἰτία.

Lo sch. B 318 informa che Zenodoto al v. 318 leggeva ἀρίδηλον (= ἀρίζηλον) e quindi προσέθηκεν il successivo v. 319;1 Aristonico riferisce con chiara preferenza l’opinione di Aristarco, che leggeva ἀίζηλον (= ἀίδηλον) e atetizzava il v. 319. La questione è ben conosciuta e commentata, per cui non è necessario qui diffondersi ulteriormente.2 Sono chiari i termini della divergenza fra Zenodoto e Aristarco sull’assetto del testo e la conseguente interpretazione: sottolineiamo soltanto che per Zenodoto nel testo c’era la trasformazione del serpente in pietra, mentre per Aristarco l’episodio aveva un carattere meno prodigioso e il serpente spariva come era comparso, per volontà divina. La profezia di Calcante, secondo quanto riferisce Odisseo, aveva interpretato il fatto che il serpente aveva divorato otto passerotti più la madre, ma non faceva parola della trasformazione in pietra: l’opinione di Aristarco teneva dunque conto anche della coerenza interna.  1 È bene chiarire che il verbo προστίθημι non significa affatto che Zenodoto abbia aggiunto un verso di sua creazione; un parallelo per l’uso del verbo in riferimento a Zenodoto si trova nello sch. R 456 (Didimo), a proposito del verso “addizionale” 456a. 2 Si veda la discussione in Valk, 1963–1964, ii, 146–148, con paralleli e bibliografia; cfr. Erbse 1969–1988, ad loc.; Chantraine 1973–1981, i, 169; LfgrE, ss. vv.; West 1998–2000, app. ad loc.; Kirk 1985, ad loc.; Mirto 1997, ad loc.; Latacz 2003, ad loc.

  Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319 Un excerptum di Porfirio3 ci informa che Aristotele si era occupato dell’interpretazione di questo passo iliadico. Si tratta del frammento 145 Rose = 369 Gigon, evidentemente dagli Aporemata Homerikà,4 che inizia con queste parole: περὶ τούτων τῶν ἐπῶν ἠπόρησεν ὁ Ἀριστοτέλης τοιαῦτα· διὰ τί ὁ Κάλχας, εἰ μὲν οὐδὲν ἦν τέρας τὸ γινόμενον, ἐξηγεῖται ὡς τέρας; τί γὰρ ἄτοπον ὑπὸ ὄφεως στρουθοὺς κατέδεσθαι ἢ τούτους ὀκτὼ εἶναι; περὶ δὲ τοῦ λίθον γενέσθαι οὐδέν λέγει, ὃ ἦν μέγα; A proposito di questi versi Aristotele ha esaminato i seguenti problemi. Per quale ragione Calcante, se quanto accaduto non era un prodigio, lo interpreta come un prodigio? In effetti, cosa c’è di strano nel fatto che dei passeri siano mangiati da un serpente oppure che siano otto di numero? Invece Calcante non dice nulla sul fatto che il serpente sia diventato pietra, che era una cosa importante.

A quanto pare nel seguito (c’è certamente una corruzione a questo punto) Aristotele si interrogava sul possibile rapporto fra la partenza dell’esercito greco da Troia per la fine della guerra e il numero dei passeri, dato che la madre era la nona vittima divorata dal serpente, mentre Troia sarebbe stata presa nel decimo anno. Si prosegue poi riferendo che Aristotele sosteneva che la pietrificazione del serpente non si riferisce al ritorno da Troia e che per questo il poeta non ne ha fatto parlare Calcante (φησὶν οὖν μὴ εἰς τὸν νόστον εἰρῆσθαι τὰ περὶ τῆς ἀπολιθώσεως τοῦ δράκοντος· διὸ οὐδ᾿ ἐποίησε λέγοντα): in primo luogo non tutti i Greci sarebbero stati privati del ritorno; poi sarebbe stato ridicolo che Odisseo in qualche modo invitasse alla partenza gli uomini ai quali stava mostrando i motivi per restare. Il significato della pietrificazione poteva invece essere la lentezza, che già c’era stata e non incuteva più paura: erano già trascorsi nove anni interi, che corrispondono al numero dei passeri uccisi, e iniziava il decimo (Dindorf p. 115, rr. 1 – 18 = Gigon 527 II 44 – 528 I 19). Introdotta da οἱ δὲ τὰ περὶ οἰωνιστικῆς τῆς καθ᾿ Ὅμηρον γράψαντες, nell’excerptum porfiriano si trova poi una lunga sezione, che contiene minuziosi e variegati ragionamenti intorno ai valori simbolici da attribuire ai passeri dell’episodio (Dindorf p. 115, 18 – 116, 15 = Gigon 528 I 19 – 528 II 17), dopo la quale ricompare il nome di Aristotele e si ribadisce la sua idea a proposito della pietrificazione del serpente.

 3 Trasmesso dal codice Ven. Marc. 821 = *B: cfr. Erbse 1969–1988, i, p. XVII; testo in: Dindorf 1877, 115–117; Schrader 1880, xxx. 4 Cfr. Hintenlang 1961, 137–141.

Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319  

ἡ δὲ τοῦ δράκοντος ἀπολίθωσις κατὰ μὲν Ἀριστοτέλην τὴν βραδυτῆτα ἐδήλου καὶ τὸ σκληρὸν τοῦ πολέμου, κατὰ δὲ ἄλλους τὴν τῆς πόλεως ἐρήμωσιν καὶ τοῦ μὲν ζωτικοῦ παντὸς κένωσιν τῶν δὲ λίθων καὶ οἰκοδομημάτων κατάλυσιν, τῶν Τρώων ἁλόντων Ἑλλήνων ἀποπλευσάντων. Secondo Aristotele la pietrificazione del serpente indicava la lentezza e la durezza della guerra; secondo altri invece lo spopolamento della città e la totale sparizione della vita, la distruzione delle pietre e degli edifici, dopo la cattura dei Troiani e la partenza dei Greci.

Seguono ancora altre considerazioni su possibili significati simbolici dei vari elementi dell’episodio, che a quanto pare non hanno più nulla a che fare con Aristotele, almeno in modo esplicitamente riconoscibile. Per riassumere, da questa fonte veniamo a sapere che Aristotele aveva affrontato e discusso le difficoltà poste dal passo del II canto dell’Iliade nel quale Odisseo riferisce l’episodio accaduto in Aulide alla partenza della spedizione a Troia con la connessa profezia di Calcante. Lo svolgersi dell’episodio e l’interpretazione dell’indovino gli erano risultati problematici per diversi motivi: non apparivano immediatamente perspicui né la corrispondenza del numero dei passeri divorati con gli anni di guerra né il significato della pietrificazione del serpente, e c’era bisogno di fornire spiegazioni. Sul primo punto non sappiamo con certezza cosa egli pensasse esattamente, probabilmente che il numero nove si riferiva agli anni completi trascorsi; sul secondo invece è chiaramente detto che la trasformazione in pietra significava secondo lui la lentezza esasperante della guerra e non il decimo e conclusivo anno della medesima, e dunque non era in relazione con il ritorno dell’esercito acheo. La considerazione aporetica indicata come punto di partenza del ragionamento appare legata alla coerenza complessiva del passo. Aristotele si chiedeva come mai Calcante avesse interpretato come un prodigio il fatto che dei passeri fossero stati mangiati da un serpente oppure che essi fossero in numero di otto più la madre, mentre non aveva detto nulla sul fatto che il serpente divoratore fosse stato trasformato in pietra, cioè sulla sola cosa veramente prodigiosa dell’intero episodio. Non si può negare, in effetti, che il testo ponga questa difficoltà: se non come una questione di puro razionalismo, certo come un problema di coerenza della rappresentazione poetica. Perché mai il poeta doveva introdurre il racconto di un episodio nodale nella vicenda in un forma in cui un fatto portentoso veniva interpretato dal celebre indovino in un modo così curioso e parziale? Perché Calcante si occupa dei passeri divorati e non del serpente pietrificato, che è sicuramente un prodigio tale da richiedere una spiegazione (sulla quale si affaticavano gli esegeti, compreso Aristotele, come abbiamo visto)? La soluzione di Aristotele era: la pietrificazione non riguarda la partenza da Troia

  Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319 e per questo Calcante non la prende in considerazione nella sua interpretazione profetica, che riguarda la fine della guerra e il tempo del ritorno in patria.5 Mi sembra assai difficile non vedere la connessione fra le questioni sollevate dai ragionamenti di Aristotele e la documentazione sulla divergenza filologica fra Zenodoto e Aristarco, che è centrata esattamente sul v. 319, l’unico che parli della soprannaturale metamorfosi. Nella fonte (Porfirio) che ci trasmette il materiale aristotelico non abbiamo alcuna traccia relativa a varianti testuali, come quelle documentate per il v. 318 e collegate alla conservazione o meno nel testo del v. 319, di cui abbiamo parlato sopra. Tuttavia, è ben presente il problema della coerenza fra la presenza del v. 319 nel testo e la successiva interpretazione dell’indovino che non ne tiene conto e trascura la pietrificazione del serpente: Aristotele era sorpreso e discuteva su questo, oltre che sui significati simbolici dell’episodio, fornendo le sue opinioni esegetiche. Abbiamo un interessante caso in cui una documentata aporia aristotelica focalizzava l’attenzione sulle difficoltà di un passo, per il quale un’altrettanto documentata discussione filologica dei grammatici alessandrini mostra una tipica divergenza fra Zenodoto e Aristarco non solo sulla scelta di una lezione, ma anche sulla conservazione o l’atetesi di un verso “incriminato”. La situazione è simile a quella di Iliade II 409, di cui mi sono occupato qualche tempo fa: un verso criticato per il suo contenuto da Demetrio Falereo, che lo riteneva inopportuno, e in seguito oggetto di discussione per i filologi alessandrini, alcuni dei quali proponevano di espungerlo, mentre Aristarco lo conservava.6 Se pure in forma frammentaria, ci è pervenuta documentazione sul fatto che Aristotele stesso oppure suoi scolari peripatetici avevano discusso passi omerici significativi, dedicando ad essi zetemata su svariati argomenti: passi per i quali le fonti mostrano con evidenza come le puntuali questioni sollevate nei loro commenti siano state oggetto in seguito di ulteriori discussioni e approfondimenti da parte dei grammatici alessandrini, talvolta con riconoscibili sviluppi filologicoesegetici anche di carattere testuale. L’analisi di questi casi, dei passi rilevanti e delle tematiche coinvolte, porta alla luce una modalità ben chiara in cui si è concretato, proprio sull’interpretazione e lo studio dei testi poetici, il rapporto profondo tra Aristotele e scuola peripatetica e la filologia di età ellenistica. Quello di

 5 Un’analisi del frammento aristotelico in Hintenlang 1961, che parla di «sachlich-rationale Charakter der aristotelischen Lösung» (p. 139). 6 Cfr. Montanari 2000b (= questo volume, cap. 55); 2001.

Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319  

cui ci siamo occupati qui è un esempio, altri sono stati evidenziati e studiati,7 in vista di una raccolta ampia e attenta, che ricomponga il quadro, almeno per gli elementi rivelatori che si sono conservati.

 7 Cfr. Montanari 2000b; 2001; 2012 (= questo volume, cap. 56).

 Demetrius of Phalerum on Literature Relatively little is known of Demetrius of Phalerum’s writings on literature, reflecting his interest in literary works and the personalities of poets and writers,1 even less than is known concerning his production in other fields of learning, such as philosophy, historical-political studies, oratory or rhetoric. Diogenes Laertius states that he outstripped all the Peripatetics of his day in quantity of books and number of lines, for he was highly educated and rich in experience in a multitude of fields: he composed historical and political works, essays on poets and treatises on rhetoric, made collections of public and diplomatic speeches, compilations of Aesop’s fables, and much else.2 In addition to the generic information concerning Demetrius of Phalerum’s writings on poets (τὰ περὶ ποιητῶν), Diogenes Laertius also provides us with a few titles. On Homer we have a Περὶ Ἰλιάδος in two books, a Περὶ Ὀδυσσείας in three books, a Ὁμηρικός in one book.3 We are also told of a Περὶ Ἀντιφάνους in one book, on the subject of Antiphanes, an eminent poet of the Middle Comedy who was active in the first half of the fourth century.4 It is interesting to note that this book concerned an author who was fairly close in time to Demetrius of Phalerum (Antiphanes may have been perhaps a couple of generations older, or thereabouts), but unfortunately no fragment at all has come down to us.5 As far  1 Tertullian, Apolog. 18, defines him as “grammaticorum tunc probatissimus” (62 SOD = fr. 188 Wehrli 1968 = 228 T 10 FGrH); “grammaticus” is the definition given also by Marius Victorinus, Ars Gramm. I 4 = 147 = fr. 196 W. 2 DL 5.80–81 = 1.58–109 = fr. 74 (+189) W = 228 T 1 FGrH. 3 143–6 = frr. 190–93 W: see below. 4 1.102 = fr. 194 W = Antiphanes T 5 PCG. A syngramma was devoted to Antiphanes also by Dorotheus of Ascalon, a grammarian of the first imperial age: Athen. 14.662f = Antiphanes T 7 PCG; for Aristophanes of Byzantium v. fr. 369 Slater = Antiphanes T 6 PCG. 5 Wehrli 1968, 86, hypothesizes: “dann stammt von D. vielleicht ein Teil des Materials bei Suda s.v. und im Anonymus De comoedia 13 p. 9 Kaibel.” I venture to propose a very hazardous suggestion, not to be taken seriously. Plutarch (Demosth. 9.3–4 = 135A = fr. 163 W) and Photius (Bibl. 493a41 = 135B = fr. 164 W) report that Demetrius of Phalerum says that once Demosthenes, as if under inspiration, swore the famous metrical oath “by land, by springs, by rivers, by floods” and so he caused an uproar in the assembly and that the episode was ridiculed by the comic poets. Now, ps.-Plutarch (Vitae dec. orat. 845b) tells that the comic poets Antiphanes and Timocles made fun of Demosthenes in their comedies and then quotes the same oath, adding that in this way Demosthenes caused an uproar in the assembly. No explicit connection is stated, but perhaps it is possible to imagine that the point regarding Demetrius of Phalerum about Demosthenes could be traced back to his book about Antiphanes and originated from a comment or an analysis of a comedy of Antiphanes dealing with Demosthenes. https://doi.org/10.1515/9783110772371-055

Demetrius of Phalerum on Literature  

as the poets are concerned, nothing else remains: Homeric criticism represents the only setting from which one can hope to derive some results. We will return to this later. One difficulty consists of attribution problems, due to the fact that our sources sometimes simply quote the name “Demetrios” without specification, in a situation in which homonymy is far from rare. In the history of the editions of Demetrius of Phalerum fragments, this type of uncertainty has given rise to a number of oscillations, some of which are quite curious. The old Ostermann collection6 was rather generous and included many citations which were later attributed to other figures bearing the name “Demetrios.” Jacoby, on the other hand, openly declared he would not include all the fragments allowed by Ostermann,7 although the exclusion criterion he adopted is not altogether clear. In FGrH 228 Komm. he states:8 “nicht aufgenommen sind außer den dürftigen überresten von reden (o. p. 643, 44; 645, 19) und den Χρεῖαι (o. p. 644, 26) eine reihe von zitaten in den Homerscholien (Il. Z 414; Ξ 221; N 5; Π 411; Eust. Il. O 680; Schol. Od. σ 17; Athen. V 4 p. 177 EF), die apophthegmata…” The same considerations are found in Wehrli, in the introductory comment to the fragments of “philological” works (85): “Demetrios allein bedeutet in Zitaten besonders für grammatikalische Fragen wahrscheinlich den Aristarcheer D. Ixion (Ostermann frr. 58–62; cf. Martini 2835, 20;9 Bayer 140:10 es handelt sich um die Scholien zu Homer Z 414; N 5; Ξ 221; Π 411; Eustath. zu O 680; schol. σ 17).” However, in this list Wehrli stops at sch. Od. σ 17, since in his edition he naturally does not omit the certain fragment found in Athen. V 177 ef, as we will see shortly (actually 143 = fr. 190 Wehrli). Examining this series, one finds that in sch. Il. 6.414, 13.5, 14.221, 16.411 and Od. 18.17 there is a citation of a “Demetrios” without specification: these fragments have been attributed to the grammarian Demetrius Ixion (whose name recurs in another twenty or so cases in Homer Scholia, where he is cited also as Δημήτριος ὁ Ἰξίων or only ὁ Ἰξίων).11 In contrast, the citation of “Demetrios” in

 6 Ostermann 1857. 7 FGrH 228, II B Komm., 641: “Ostermann … mit bisher vollständigster sammlung der fragmente, die hier nicht alle aufgenommen sind (s.u. p. 648, 32).” 8 II B Komm., 648, 32ff. 9 Martini 1901. 10 Bayer 1942. 11 The cases in question are respectively fr. 7 = 48, 14, 15, 18, 25 in Staesche 1883 = 170, 172, 171, 173, 175; one can also add sch. Il. 15.194, quoting “Demetrios” = fr. 17 Staesche. In Scholia and other erudite sources “Demetrios” without specification is likely to be Demetrius Ixion, of

  Demetrius of Phalerum on Literature Eustathius’ commentary ad Il. 15.680 (1037.57) = 174 certainly belongs to the grammarian Demetrius Gonypesós, as confirmed by the parallel sch. ex. Il. 15.683–84, in which he is cited with his full name (as in the other two extant fragments of Demetrius Gonypesós).12 Therefore as far as these six fragments are concerned, Jacoby’s omission (in contrast to Ostermann’s inclusion) would appear to have been guided by uncertainty in the attribution and perhaps even by awareness of the different authorship possibilities. Consequently, it comes as something of a surprise to note the omission in Jacoby’s collection of the fragment preserved by Athenaeus book V, 177 ef, which is certainly a Demetrius of Phalerum fragment, accompanied as it is by the complete citation of the name Δημήτριος ὁ Φαληρεύς (143 = fr. 190 W, see below). If one surmises that Jacoby omitted it on account of the type of its “philological” content, then one may wonder why he included sch. Od. 3.267 + Tzetzes Prol. ad Lycophr. 4.5ff. (FGrH 228 F 32a–b = 144, 146 = frr. 191–192 W) and Stobaeus 3.5.43 (FGrH 228 F 33 = 145 = fr. 193 W), to which we will turn shortly: the claim that the latter two exemplify contents that are partly of a historical-cultural nature, in addition to observations on literary interpretation, is not a valid explanation, since a similar claim could easily be made regarding the above-mentioned Athenaeus fragment. In actual fact, these three (or four) certain fragments, i.e. with the complete name Δημήτριος ὁ Φαληρεύς, form the entire stock of Homeric interpretation of Demetrius on which considerations can be made. This group of fragments will be dealt with in depth later. Another problem is that of sch. AD ad Il. 6.35, in which a “Demetrios” is cited. Jacoby includes this scholium in the Zweifelhaftes und Unechtes section of Demetrius of Phalerum fragments (FGrH 228 F 50 = 157), and he comments telegraphically: “zitate D.s in den Homer-scholien o. p. 648, 34.” In other words, he refers back to none other than the place seen above, in which he lists the citations from Homer Scholia he had omitted in his collection: but this one, also bearing only the name “Demetrios,” is indeed present, albeit in the Zweifelhaftes und Unechtes section. The same situation is found in Wehrli’s edition, where the fragment is included in the section headed Unbestimmbares, Zweifelhaftes, Unechtes, numbered 207= 157, and the comment says nothing at all about its authorship. This is a Homeric scholium of the D class, one of the well-known mythographical ἱστορίαι of the so-called Mythographus Homericus,13  course, particularly concerning grammatical and philological questions: cf. also Wehrli 1968, quoted above. 12 Cf. M. van der Valk ad Eust. l.c.; Erbse ad sch. Il. cit. 13 Montanari 1995 (= this volume, ch. 22), with bibliography.

Demetrius of Phalerum on Literature  

which tells of the conquest of the city of Pedasos by Achilles. The subscription says ἡ ἱστορία παρὰ Δημητρίῳ καὶ Ἡσιόδῳ: the ‘Demetrios’ cited here has been identified elsewhere with Demetrius of Scepsis and therefore the scholium has been included in the collection of fragments of the latter.14 Statements in favour of Demetrius of Scepsis can also be found by Merkelbach-West in the edition of Hesiod fragments (fr. 214). In addition, Erbse (who did not include the D-Scholia in his edition of the Scholia to the Iliad) reports the two attributions without committing himself in any direction. But perhaps something a little more definite can be said concerning this fragment. We do not yet have a modern critical edition of the Scholia D in Iliadem and for these Scholia the text of the 1517 editio princeps by Janus Lascaris is used, while the Dindorf edition of the Scholia A in Iliadem can be used for the D-Scholia extracts contained in the codex Ven. A. As mentioned above, Erbse did not include the D class in his edition and merely gave an indication whenever a D-Scholium happened to be present in codex A (as is the case here, and so it is labelled as a sch. AD). It is known, perhaps, that I have long been working on an edition of the Scholia D in Iliadem. I have found that in the most authoritative manuscript of these scholia, the codex C partly preserved in Rome and partly in Madrid (dated between the end of the ninth and the beginning of the tenth century),15 the subscription of the ἱστορία in question is different from that given in the AHRV codices. ἡ ἱστορία παρὰ Δημητρίῳ καὶ Ἡσιόδῳ ἱστορεῖ Δημήτριος καὶ Ἡσίοδος ἱστορεῖ Δημήτριος ὁ †ἀσκητης†

ARV Η C

I would argue that the blatant corruption of C masks the correct original subscription ἱστορεῖ Δημήτριος ὁ Σκήψιος. In addition I also suspect that the name of Hesiod could be the fruit of a conjecture16 intended to amend the corrupted ἀσκητης, but this is perhaps casting my net too far. A good parallel for the Δημήτριος ὁ Σκήψιος solution in sch. Il. 6.35 is offered by the subscription of the D-Scholium to Il. 20.3 in the manuscripts: ἡ ἱστορία παρὰ Δημητρίῳ τῷ Σκηψίῳ, confirmed by a parchment fragment, P. Schubart 21, of the fifth century A.D. (the alternation between the formula ἡ ἱστορία παρὰ + dative or ἱστορεῖ + nominative

 14 Gaede 1880, fr. 32. 15 Cf. Montanari 1979, passim; 1995a, partic. 147 ff. 16 Doubts on the Hesiodic contents of the sch. also in Merkelbach-West, apparatus to the fr. 214.

  Demetrius of Phalerum on Literature does not constitute a problem).17 If these considerations are acceptable, then sch. AD ad Il. 6.35 has to be regarded as indeed containing a remnant of Demetrius of Scepsis and definitely not of Demetrius of Phalerum. Quite different is the case of the citation in Stephanus of Byzantium s.v. Κορώπη, where it would appear that a hypomnema to Nicander is attributed precisely to Δημήτριος ὁ Φαληρεύς. Wehrli includes the fragment in his section Unbestimmbares, Zweifelhaftes, Unechtes, listing it as Nr. 208 = 158 and specifying in the comment that a hypomnema by Demetrius of Phalerum to Nicander “ist chronologisch unmöglich; mit Recht nimmt man an, daß der Phalereer mit Demetrios Chloros verwechselt wird”.18 Jacoby, on the other hand, omits the fragment altogether: a somewhat bizarre editorial choice, as the presence of the full name would certainly suggest it should be included among the spuria with an ad hoc explanation (see Jacoby himself for 228 F 51 = 65 and 52 = 66). That the reference of Stephanus of Byzantium is to Demetrios Chloros (cited elsewhere in the Nicander Scholia) is in my view virtually certain, and in any case there can hardly be any doubt that Demetrius of Phalerum is to be ruled out here: but, I believe, the presence of the full name needs to be mentioned and discussed in a collection of Demetrius of Phalerum fragments. This is an isolated example in which the authenticity of a fragment that has come down to us in a text presenting the citation of the full name of Δημήτριος ὁ Φαληρεύς has to be rejected. Problems often arise elsewhere from the fact that, as mentioned earlier, one sometimes finds citation of a “Demetrios” without specification. This problem is compounded by the existence of several homonyms, especially when dealing with scholiographic, lexicographic and other erudite sources. Before going further into this matter, I would like to make a remark concerning what we have considered so far. Discussion has centered on a fairly limited number of examples, but has concretely confirmed what I believe to be a fairly widespread opinion, namely that we sorely lack a reliable edition of the fragments of Demetrius of Phalerum and that it is extremely desirable to have a new edition based on serious re-examination of the evidence. I am far from convinced that the editions currently available — I mean, editions of various

 17 Cf. Montanari 1995 (= this volume, ch. 22). To my knowledge, Il. 6.35 and 20.3 contain the only two citations of a “Demetrios” of the Scholia D in Iliadem as they have been edited (Lascaris) and in both cases the citation is Δημήτριος ὁ Σκήψιος. Another two can be derived from the manuscripts, with the bare “Demetrios,” that is sch. D Il. 5.906 and 6.23: the first may be Demetrius Ixion according to Erbse ad loc., the second could be Demetrius of Scepsis. 18 Cf. Susemihl, 1892, II.20; Kroll 1936; Guhl 1969, 4.

Demetrius of Phalerum on Literature  

figures named “Demetrios” — have provided the best or the most plausible solution in all the cases in which the sources exhibit the citation of a “Demetrios” without any specification.19 It should not be overlooked that new editions, above all of scholiographic and lexicographic works, may well lead to notable progress in this field. However, this is a problem that goes beyond the scope of my paper. The observations made so far have been concerned with criticism of poetry: we will investigate Homeric criticism in greater depth, but it is appropriate at this point to offer a few words on prose writers first. The Demetrius fragments concerning prose works are normally set in the framework of rhetorical or rhetorical-philosophical production. However when treatises on rhetoric make direct use of authors’ works, with detailed close stylistic analysis, it is clear that such treatises are in some sense grounded on interpretation of the texts. This means that they have a close relationship with hermeneutic and literary criticism, and it is certainly true that there are naturally overlapping areas and shared materials between these disciplines. Let us examine one case regarding Plato. Dionysius of Halicarnassus, in de Demosth. 5, severely criticizes Plato’s use of the high style. When Plato indulges in sophisticated stylistic refinements and fine flourishes of phrase, Dionysius argues, he reveals a number of flaws. Dionysius lists a number of criticisms and concludes by claiming that Plato makes excessive and clumsy use of allegory, unpleasant figures of speech and Gorgianic expressions introduced in an inappropriate and puerile manner, to the point of appearing as some kind of possessed mystic, as maintained by Demetrius of Phalerum and many others as well (133 = fr. 170 W = 228 F 11b FGrH). 1. ἡ δὲ δὴ Πλατωνικὴ διάλεκτος βούλεται μὲν εἶναι καὶ αὐτὴ μῖγμα ἑκατέρων τῶν χαρακτήρων, τοῦ δὲ ὑψηλοῦ καὶ ἰσχνοῦ, καθάπερ εἴρηταί μοι πρότερον, πέφυκε δ᾿ οὐχ ὁμοίως πρὸς ἀμφοτέρους τοὺς χαρακτῆρας εὐτυχής. 2. ὅταν μὲν οὖν τὴν ἰσχνὴν καὶ ἀφελῆ καὶ ἀποίητον ἐπιτηδεύῃ φράσιν, ἐκτόπως ἡδεῖά ἐστι καὶ φιλάνθρωπος· καθαρὰ γὰρ ἀποχρώντως γίνεται καὶ διαυγής … 4. … ὅταν δὲ εἰς τὴν περιττολογίαν καὶ τὸ καλλιεπεῖν, ὃ πολλάκις εἴωθε ποιεῖν, ἄμετρον ὁρμὴν λάβῃ, πολλῷ χείρων ἑαυτῆς γίνεται … 6. ἀλληγορίας τε περιβάλλεται πολλὰς 〈καὶ μακράς〉, οὔτε μέτρον ἐχούσας οὔτε καιρόν. σχήμασί τε ποιητικοῖς ἐσχάτην προσβάλλουσιν ἀηδίαν καὶ μάλιστα τοῖς Γοργιείοις ἀκαίρως καὶ μειρακιωδῶς ἐναβρύνεται. καὶ ῾πολὺς ὁ τελετὴς ἐστὶν ἐν τοῖς τοιούτοις παρ᾿ αὐτῷ᾽, ὡς καὶ Δημήτριος ὁ Φαληρεὺς εἴρηκέ που καὶ ἄλλοι συχνοί· οὐ γὰρ ἐμὸς ὁ μῦθος.

 19 For ex. quotations of “Demetrios” in Aristophanes Scholia, which are attributed to Demetrius of Scepsis.

  Demetrius of Phalerum on Literature Attacked for his biting criticism of Plato, Dionysius then replied in self-defense with the Letter to Pompey Geminos, in which he also endeavoured to correct his position: in this second work Dionysius quotes his own de Demosth. passage containing criticisms of Plato’s style together with the citation of Demetrius of Phalerum. Shortly prior to this self-quotation, Dionysius (Epist. ad Pomp. 1.16 = 133 = fr. 195 W = 228 F 11a FGrH) defends himself by arguing that many others before him had voiced criticism of Plato’s ideas and mode of expression, starting with his most important disciple Aristotle and subsequently Cephisodorus, Theopompus, Zoilus, Hippodamas, Demetrius (of Phalerum) and many others, all of them not motivated by envy or mean-spiritedness, but simply in search of the truth. καὶ γὰρ τὰ δόγματα διέβαλον αὐτοῦ τινες καὶ τοὺς λόγους ἐμέμψατο πρῶτον μὲν ὁ γνησιώτατος αὐτοῦ μαθητὴς Ἀριστοτέλης, ἔπειτα οἱ περὶ Κηφισόδωρόν τε καὶ Θεόπομπον καὶ Ζωΐλον καὶ Ἱπποδάμαντα καὶ Δημήτριον καὶ ἄλλοι συχνοί, οὐ διὰ φθόνον ἢ διὰ φιλαπεχθημοσύνην κωμῳδοῦντες ἀλλὰ τὴν ἀλήθειαν ἐξετάζοντες.

Commenting on the first passage (de Demosth. 5.6 = 133 = fr. 170), Wehrli then refers to the second (Epist. ad Pomp. 1.16 = 133 = fr. 195) defining it as “inhaltliche Kritik,” perhaps on account of the sentence ἀλλὰ τὴν ἀλήθειαν ἐξετάζοντες. In my opinion, a reading of the entire passage of the Letter to Pompey 1–2 together with de Demosth. 5 shows the two passages to be so closely interconnected that it was definitely not advisable to divide them and consider them separately, classifying the first (fr. 170) under the heading Rhetorisches and the second (fr. 195) under the heading Über Platon. Jacoby achieves better results by grouping them as F 11 a–b = 133. Demetrius of Phalerum is adduced by Dionysius of Halicarnassus as a precedent for criticism of Plato20 and it is highly plausible to suggest that Dionysius of Halicarnassus may also have used Demetrius of Phalerum as a source for some of his own arguments. The catalogue of Diogenes Laertius mentions no work by Demetrius of Phalerum on Plato and it may perhaps be said that hypothesizing a specific work Περὶ Πλάτωνος or some such title is unnecessary. Where Demetrius of Phalerum is cited in isolation (i.e. in de Demosth. 5, reproduced in Letter to Pompey 2 = 133 = fr. 170), the content of the citation is stylistic-rhetorical, while in the other case (Epist. ad Pomp. 1.16 = 133 = fr. 195) his name is included within a group of figures jointly described as critics of Plato’s doctrines and dialogues. If Demetrius of Phalerum criticized Plato also or predominantly on the  20 Cf. Wehrli on fr. 170 for parallels of criticism of Plato’s use of the high style in Peripatetic circles, from Aristotle to Dicaearchus (see fr. 42 W, from DL 3.38 [Wehrli 1967]).

Demetrius of Phalerum on Literature  

grounds of style, this may have been encompassed within a work on rhetoric extending to much broader horizons, even in the Περὶ ῥητορικῆς. In any case, this is exactly what we could regard as fragments of “literary criticism” concerning prose writers. It will have become clear from what we have seen so far that the theme of “Demetrius of Phalerum on Poetry” seems in fact to be restricted to “Demetrius of Phalerum on Homer.” It is indeed the case that Homeric criticism is the area in which we have a small number of pieces of definite evidence, which allow us to achieve results. As I pointed out earlier, Diogenes Laertius provides us with three titles concerning Homeric issues: Περὶ Ἰλιάδος in two books, Περὶ Ὀδυσσείας in three books, Ὁμηρικός in one book. While the first two fit perfectly into the well-known typology of peripatetic Περὶ-Literature or Problemata-Literatur, the third could have been a speech21 or, perhaps more likely, a dialogue, although this by no means excludes the possibility that it may have contained exegetic reflections on passages from the Iliad or the Odyssey. However the evidence that has come down to us never contains any indication of the work from which the fragments came. Let us begin by examining the fragment in Athenaeus, Deipn. V 177 e–f, which concerns Iliad 2.409 (143 = fr. 190 W; omittit Jacoby, FGrH 228). It occurs in the context of a discussion on the symposia and above all on who should be invited to attend. Athenaeus says that Homer also gives indications as to which persons do not require any invitation but instead come to a feast on their own initiative. An example is given from the chieftain’s dinner arranged by Agamemnon after testing the army’s morale in Iliad book 2: Homer says (2.404–7) that Agamemnon invited the Achean princes, and he mentions Nestor, Idomeneus, the two Aiaxes, Diomedes, Odysseus. He then adds (2.408) that Menelaus came αὐτόματος, of his own accord. Athenaeus comments: “For it is plain that neither a brother, nor parents, nor wife need be invited, nor anyone else whom one holds in equal esteem with these: otherwise it would be cold and unfriendly. And yet some authorities have added a verse which further explains the reason: “for he knew in his heart that his brother was troubled” [v. 409] — as though it were necessary to tell the reason why a brother might come to dinner of his own accord…”22 (177 c). After a prolonged explanation of the absurdity of adding 1.409 to introduce a non-existent motivation, Athenaeus continues with

 21 Cf. Wehrli, comm., 85: “eine rhetorische Deklamation wie Dion von Prusa oratio LIII, denn solche Reden über Homer sind alte Uebung.” 22 Transl. by C. Burton Gulick (Loeb 1928).

  Demetrius of Phalerum on Literature an illustration of the opinions of the grammarian Athenocles of Cyzicus and those of Demetrius of Phalerum: Ἀθηνοκλῆς δ᾿ ὁ Κυζικηνὸς μᾶλλον Ἀριστάρχου κατακούων τῶν Ὁμηρικῶν ἐπῶν εὐπαιδευτότερον ἡμῖν φησι τοῦτον Ὅμηρον καταλιπεῖν, ὅσῳ τῆς ἀνάγκης ὁ Μενέλεως οἰκειοτέρως εἶχεν. Δημήτριος δ᾿ ὁ Φαληρεὺς ἐπαρίστερον τὴν τοῦ στίχου παράληψιν ἐπειπὼν καὶ τῆς ποιήσεως ἀλλοτρίαν, τὸν ᾔδεε γὰρ κατὰ θυμὸν ἀδελφεὸν ὡς ἐπονεῖτο (Β 409) μικρολογίαν ἐμβάλλειν τοῖς ἤθεσιν. “οἶμαι γάρ, φησίν, ἕκαστον τῶν χαριέντων ἀνθρώπων ἔχειν καὶ οἰκεῖον καὶ φίλον πρὸς ὃν ἂν ἔλθοι θυσίας οὔσης τὸν καλοῦντα μὴ περιμείνας.” Πλάτων δ᾿ ἐν τῷ Συμποσίῳ περὶ τῶν αὐτῶν λέγει οὕτως (p. 174 b) κτλ. Athenocles of Cyzicus, with a better understanding of the Homeric poetry than Aristarchus, explains to us with greater refinement that Homer passed Menelaus over without mention because he was more closely related in kinship to Agamemnon. And Demetrius of Phalerum declared that the inclusion of the verse “for he knew in his heart that his brother was troubled” is awkward and foreign to the poet’s style, and imputes meanness to the characters. “For,” says he, “I think that every man of refinement has someone, either relative or friend, to whom he can go when a feast is on without waiting for an invitation.” And Plato, in The Symposium, has this to say on the same subject (p. 174b) etc.23

Athenaeus then goes on to discuss the passage from the Symposium where Plato quotes the proverb according to which the ἀγαθοί go αὐτόματοι to the feasts held by the ἀγαθοί: Plato also mentions the passage from Iliad 2.407ff. with a disquisition as to whether or not it was suitable for Homer to present Menelaus as going αὐτόματος to Agamemnon’s dinner, claiming that Menelaus ranked below Agamemnon. However in Plato there is no reference to l. 409 and no mention in the terms discussed above of the attendant problems it could raise. Therefore Demetrius of Phalerum is the most ancient authority we have on this particular question. Basically, then, what we are dealing with is a problem of internal coherence of the Homeric passage. For if it is right and proper, according to the established rules of personal relations, for Menelaus to go to Agamemnon’s dinner of his own accord, i.e. αὐτόματος, as l. 408 would have it, then the motivation supplied in l. 409 is quite out of place: it “is awkward and foreign to the poet’s style, and imputes meanness to the characters.” Introducing the discussion on the discrepancy of l. 409, Athenaeus asserts that some authorities have added the verse (τινὲς στίχον προσέγραψαν), i.e. that

 23 Ibid.

Demetrius of Phalerum on Literature  

the line was an interpolation24 which had evidently been added without accepting or without understanding the supposed profound value of the declaration that Menelaus went αὐτόματος to Agamemnon’s dinner. That l. 409 was not authentic was clearly also the opinion held by Demetrius of Phalerum, who argued that acceptance of the line would have been foreign to the poet’s style (τῆς ποιήσεως ἀλλοτρίαν). However we cannot definitely state that he was the first to maintain that l. 409 is spurious, but only that to our knowledge he is the most ancient author to have expressed this opinion.25 The suggestion of deleting the line was certainly a subject of discussion among Alexandrian philologists. Like Demetrius of Phalerum, Athenocles of Cyzicus (a grammarian of the Alexandrian age, possibly anterior to Aristarchus or at the most contemporary, against whom the Aristarchean pupil Ammonios polemically wrote a Πρὸς Ἀθηνοκλέα)26 was favorable to athetizing the line. It would also appear that Aristarchus held the opposite view and did not athetize the line (allusion in the words: μᾶλλον Ἀριστάρχου κατακούων τῶν Ὁμηρικῶν ἐπῶν): perhaps a trace of this has remained in the scholia to this passage, where the question is discussed.27 Moreover, the source used by Athenaeus throughout the discussion on l. 409 shared the view that l. 409 was not authentic since it was incoherent with l. 408 and was therefore to be deleted: this source, openly antiaristarchean, is thought to have been the grammarian Seleukos, of the first imperial age, known for his frequently critical attitude towards Aristarchus and the Aristarchean tradition.28 From these considerations, then, it can be concluded that this specific problem of internal consistency of the Homeric passage was indeed a subject of discussion among Alexandrian philologists: Aristarchus (and pupils and followers as well?) interpreted the text in such a manner as to save l. 409, while others (like

 24 Cf. Ludwich 1884–1885, I, 216–17; van der Valk 1963–1964, II, 499; Erbse ad sch. Il. 2.405–9; discussed by Kirk 1985, I, 157–58 (409 possible oral or rhapsodic expansion of 408?). 25 Ludwich, l.c.: “Von wem die Athetese des genannten Verses ausging, sagt er [scil. Athenaeus] nicht.” 26 Cohn 1896, 2049; Ludwich 1884–85, I, 49, 51; Müller 1891, 11–14. 27 Cf. Erbse ad sch. Il. 2.408 and 409. 28 On the sources of Athenaeus V 1–20, cf. Müller 1891, 11–14; Düring 1941, 90–105: the main source of the section dealing with convivial customs is Herodicus of Seleuceia, called ὁ Κρατήτειος, a grammarian of the 2nd c. B.C., a follower of Crates of Mallos and an opponent of Aristarchus and his school; the part concerning Iliad 2.404–9 and particularly the discussion of l. 409 is traced back precisely to Seleukos, of the first imperial age, antiaristarchean as well: it was inserted by some unknown grammarian into Herodicus material. Cf. also Erbse ad sch. Il. 2.408 and 409.

  Demetrius of Phalerum on Literature Demetrius of Phalerum, Athenocles of Cyzicus, Athenaeus’ source [Seleukos?]) emphasized the inconsistency and opted in favor of athetizing it. The most ancient authority known to have noticed these contradictory elements and to have believed the line to be spurious was, we said, Demetrius of Phalerum. It seems likely that we have the remains of a zétema in genuine peripatetic style, perhaps for example on the Homeric representation of the symposia, on rules and customs in symposia or something of this nature, in which interpretation of the text also led naturally to debate on an issue of coherence. But even if it is clear that Demetrius of Phalerum went as far as to argue explicitly that l. 2.409 was not “authentic,” it is difficult to ascribe to him the further step of proposing an “athetesis” in the strict sense, i.e. in the true Alexandrian sense of the term. However a brief chronological description needs to be given at this point. Demetrius fled to Alexandria after 297 (i.e. after the death of Cassander) and he was certainly still alive in 283 when Ptolemy II came to the throne: it is plausible that in this period he devoted himself particularly to literature. The akmè of Zenodotus is placed under Ptolemy I, i.e. before 283, even though he completed his work under Ptolemy II. Therefore there was a period in which Demetrius and Zenodotus were both active in Alexandria, Demetrius being the older of the two (born perhaps around 345) and contributing authoritatively to the development of the city’s cultural institutions. Zenodotus, on the other hand, who would have been younger (born perhaps around 330–325), was a preceptor of the royal family and was later appointed as its first librarian.29 In the light of these observations, it is perhaps not altogether unrealistic to suggest that Demetrius of Phalerum’s Homeric criticism stimulated debate and in-depth inquiry in Alexandrian intellectual circles during the initial period of Hellenistic philology. Perhaps we may perceive the glimmers here of a first phase in which traditional peripatetic discussions on the works of poets — mainly Homer — provided stimuli and materials that would then be reworked with a variety of different methods, at the dawn of scientific philology. Perhaps this is a single but good example. Perhaps during this period Demetrius of Phalerum’s observation on the passage of Iliad 2 may actually have given rise to a proposal of a Zenodotean obelós. But at this point I have overreached myself in the direction of speculation and a word of caution against the development of hypotheses is in order now. Stobaeus (3.5.43) preserves a quotation from Hermippus (taken from a work entitled Συναγωγὴ τῶν καλῶς ἀναφωνηθέντων ἐξ Ὁμήρου), in which it is said that Demetrius of Phalerum claimed that line Od. 23.296 had been composed by  29 Pfeiffer 1968, 99 ff.; Fraser 1972, 114, 314 f., 689 f.; Blum 1991, 99–107; Richardson 1994, 13 f.

Demetrius of Phalerum on Literature  

Homer εἰς σωφροσύνην (Δημήτριος ὁ Φαληρεὺς εἰς σωφροσύνην ἔλεγε ταῦτα ποιεῖν: 145 = fr. 193 W; FGrHist 228 F 33).30 We know from Diogenes Laertius that Hermippus wrote on Demetrius of Phalerum (a biography, perhaps?)31 and it is possible that he lifted Demetrius of Phalerum’s observation on Od. 23.296 to include it in a collection of καλῶς ἀναφωνηθέντα taken from Homer. An analogy with the previous case can be recognized: thus if the previous fragment concerned correct behavior in the symposia, here we are dealing with reflections and precepts of a philosophical-moral nature found in Homer, or rather with comments on Homeric passages from a moral point of view.32 But what is the significance of the observation by Demetrius? Od. 23.296 is the line that ratifies the re-union of Odysseus and Penelope, the very end of Odysseus’ wanderings and his return to his family: therefore the σωφροσύνη should reside, I would argue, in Odysseus’ preference for returning home and accepting the love of his legitimate spouse, rejecting the temptations and enticements that lured him during his journeys, as well as in Penelope’s patient wait for her legitimate husband’s return as against all the overtures and gifts from her suitors. All this education to the σωφροσύνη of the adventures of the Odyssey is condensed and ratified in the moment in which husband and wife, happy and full of desire, finally return to their original bed, in line Od. 23.296. Sch. Od. 23.296 ἀσπάσιοι λέκτροιο] ἀσπαστῶς καὶ ἐπιθυμητικῶς ὑπεμνήσθησαν τοῦ πάλαι τῆς συνουσίας νόμου. Ἀριστοφάνης δὲ καὶ Ἀρίσταρχος πέρας τῆς Ὀδυσσείας τοῦτο ποιοῦνται – ἀσπάσιοι λέκτροιο παλαιοῦ θεσμὸν ἵκοντο] τοῦτο τέλος τῆς Ὀδυσσείας φησὶν Ἀρίσταρχος καὶ Ἀριστοφάνης.

The scholiastic claim that Aristophanes of Byzantium and Aristarchus had indicated the τέλος or πέρας of the Odyssey as residing in Od. 23.296 has given rise to a debate in modern criticism concerning the meaning to be attributed to this observation. Essentially, two basic possibilities can be outlined: either the two great Alexandrian grammarians intended to state that the authentic Odyssey actually ended at that line and that all following material was therefore spurious, or they intended to state that the τέλος of the action in the Aristotelian sense was contained at this point, i.e. that Od. 23.296 represented the culminating point of the meaning, the final solution of the essential line of the whole

 30 Hermippus fr. 92 W (Wehrli 1974). 31 DL 5.78 = Demetrius of Phalerum 1.34–44 = fr. 69 W = Hermippus fr. 58 W. 32 There is a parallel in Dicaearchus fr. 92 W (from sch. Od. 1.332), concerning the first appearance of Penelope among the suitors in Od. 1.332 ff.

  Demetrius of Phalerum on Literature poem.33 I do not wish to investigate this question here: suffice it to say that there is good likelihood that Demetrius of Phalerum’s focus on Od. 23.296 represented an early stand on the significance of this point. The actual verse itself does not seem to offer profound teaching as regards σωφροσύνη, but it acquires much greater depth of meaning if it is seen as the culminating point of the adventures of Odysseus and Penelope in the sense described above. It is impossible to assess exactly what Demetrius thought in philological terms, nor can it be ascertained whether he truly held an opinion as to whether some parts were authentic and others spurious (as happened to be the case with Il. 2.409).34 But his attention undeniably focused closely on that particular point of the Odyssey 23.296, and he can hardly have failed to be aware of the types of issues raised here concerning the meaning of the passage in relation to the meaning of the whole poem, regardless of any question of authenticity or otherwise of the subsequent parts. Aristotle had given an exemplary synopsis of the poem’s main plot in his Poetics (ch. 17, 1455 b 16–23), indicating that it reached its goal with the suitors’ death and the rest consisted of episodes. Thus in Aristotle’s conception the τέλος of the Odyssey was achieved with Odysseus’ vengeance and his successful reappropriation of his home and his kingdom. I would suggest it is not unlikely that discussions on the main plot of the Odyssey and on the culminating point of a well-rounded narrative continued within peripatetic circles subsequently to Aristotle’s statement, with Demetrius of Phalerum putting forward not exactly the same view, but a slightly different position as compared to that held by Aristotle: and such a position may have received attention in Alexandria and undergone further development among the grammarians. Perhaps we can add an observation in this regard, which will serve to underline this significant, though not drastic, difference of opinion. Aristotle’s viewpoint rests on a reading of the Odyssey that was more linked to a heroic conception, so that the essential themes were thought to be Odysseus’ experience of wars and political affairs, his participation in the expedition to Troy and his subsequent return, his prolonged absence from home and his lengthy wanderings, the danger to his household and kingdom caused by internal enemies, his revenge and triumphant return. The other view holds that the real culminating point and the real

 33 Rossi 1968, 151–163; Pfeiffer 1968, 116, 175–177, 231 n. 1; Gallavotti 1969; Erbse 1972, 166– 177; Richardson 1994, 21 f.; also Montanari 1995a, partic. 21 ff. 34 Wehrli ad loc. comments: “Es handelt sich um den Vers, mit dem Aristophanes und Aristarchus die Odyssee schließen lassen; vielleicht stützt D. mit seiner moralischen Würdigung die gleiche Auffassung.”

Demetrius of Phalerum on Literature  

solution to the poem — and therefore in some sense the enshrinement of its true meaning — lies in the moment when husband and wife are reunited and in the reestablishment of marital bliss between the protagonist and his faithful wife. Such a view points to a reading oriented more towards feelings and human interest, more sensitive to intimacy and sentiment. This latter reading, which seems to reflect more profound observation of the psychology of the characters and more attention to the sentimental themes, is suggestive of a more “Menandrean” reading and is perhaps closer to Hellenistic tastes. Nevertheless, this is a difference of orientation which can already be traced in Aristotle’s Poetics, in his considerations on the difference between the Iliad and the Odyssey. The development flowed within the Aristotelian-peripatetic stream. The third fragment of Demetrius of Phalerum’s Homeric criticism is provided by the sch. Od. 3.267 (144 = fr. 191 [+146 = 192] W; FGrH 228 F 32a [+32b]). The comment concerns the famous passage in which Nestor, at Pylos, tells Telemachus about Clytemnestra, Aegisthus and the murder of Agamemnon, and more specifically the lines that tell how Agamemnon had entrusted his wife to a bard when he left for Troy. Aegisthus makes an attempt on Clytemnestra’s virtue: ἡ δ᾿ ἦ τοι τὸ πρὶν μὲν ἀναίνετο ἔργον ἀεικές, δῖα Κλυταιμνήστρη· φρεσὶ γὰρ κέχρητ᾿ ἀγαθῇσι· πὰρ δ᾿ ἄρ᾿ ἔην καὶ ἀοιδὸς ἀνήρ, ᾧ πόλλ᾿ ἐπέτελλεν Ἀτρεΐδης Τροίηνδε κιὼν εἴρυσθαι ἄκοιτιν. ἀλλ᾿ ὅτε δή μιν μοῖρα θεῶν ἐπέδησε δαμῆναι, δὴ τότε τὸν μὲν ἀοιδὸν ἄγων ἐς νῆσον ἐρήμην κάλλιπεν οἰωνοῖσιν ἕλωρ καὶ κύρμα γενέσθαι, τὴν δ᾿ ἐθέλων ἐθέλουσαν ἀνήγαγεν ὅνδε δόμονδε.

265

270

Homer does not tell us the name of the bard of l. 267 and it is only natural that speculation on this matter has been rife: a scholium reports that some believed him to have been called Chariades (or Charidemos according to one manuscript), others Demodokos, others Glaukos. This passage and Clytemnestra’s bard was a subject also addressed by Dicaearchus (fr. 93 W),35 who says that “the ancients thought that the singer was also wise, as is clear from the one left with Clytemnestra.” Here the theme is “the ancients thought that the singers were wise,” but no name is given. It seems that Aristoxenus (fr. 123 W) touched on the same theme, but I am not sure if the material immediately following the name of Aristoxenus in Strabo 1.2.3 is also to be assigned to Aristoxenus him-

 35 From Philodemus, On Music, p. 20, 32.21 Kemke.

  Demetrius of Phalerum on Literature self: here too no name of the bard is given. The excerptum of Demetrius of Phalerum in sch. Od. 3.267 is much richer and places the issue in a broader context. This is the content of the scholium: Menelaus went to Delphi together with Odysseus to question the oracle concerning the forthcoming expedition against Troy. At that time Creon was ἀγωνοθέτης of the nine-yearly Pythian games: the race was won by Demodokos Lacon, a disciple of Automedes of Mycenae, who was the first to write an epic poem on the battle of Amphytrion against the Teleboai and also wrote another on the contest between Cithaeron and Helicon, after whom the mountains of Boeotia were named. Automedes was therefore a disciple of Perimedes of Argos, who was the teacher of Automedes of Mycenae himself and also of Licymnios of Bouprasion, Sinis the Dorian, Pharidas the Lacon and Probolos the Spartan. So Menelaus dedicated to Athena Pronoia a necklace belonging to Helen. Demodokos, on the other hand, was given the task by Agamemnon, who had brought him to Mycenae, of guarding Clytemnestra. They held the bards in high esteem as masters of divine and ancient noble actions,36 and they loved the lyre above all other instruments. Even Clytemnestra manifested the high regard she had for the bard, by ordering that he not be killed but banished instead. Timolaos said that Demodokos was the brother of Phemios, who followed Penelope to Ithaca to act as her guard: therefore, only because he was forced to did he sing for the suitors. Thus the bard to whom Agamemnon had entrusted Clytemnestra was called Demodokos and he was Lacon. His presence is explained in the fragment by the fact that Menelaus is claimed to have met him at the poetry competition in Delphi and later Agamemnon is supposed to have taken him to Mycenae (the excerptum does not preserve the explanation of all the transitional steps). The high esteem in which bards were held at the time thus explains his role in this episode: one may note that the same role is attributed also to Phemios, acting as Penelope’s guard, and recall Dicaearchus’ fragment quoted above. Such was the solution Demetrius gave to the problem of identifying the bard mentioned by Homer in Od. 3.267, a solution which — to my knowledge — is unparalleled, whether this Demodokos is meant to be the same Demodokos of the Phaeacians or a different and homonymous figure.37  36 Cf. Dicaearchus’ fragment 93 W and Strabo 1.2.3 (Aristoxenus fr. 123 W [Wehrli 1967a]), quoted above: cf. Floratos 1972, 12, 43–44. 37 According to Wehrli, comm. ad 191–92, this Demodokos is the same as Demodokos of the Phaeacians, the supposition being that he met the heroes (Menelaus and Agamennon) at the Pythian games and then sang of their enterprises to the Phaeacians: cf. Gostoli 1986; as different persons are given in RE s.vv. In Pausanias 1.2.3 there is mention both of Demodokos at the court of Alcinoos and also of the poet left by Agamemnon with the latter’s wife: this poet is not

Demetrius of Phalerum on Literature  

By itself, this is just an erudite detail. However there is one aspect highlighted by this fragment that goes beyond this basic account and is worth emphasizing. Demetrius of Phalerum tells us that the bard Demodokos Lacon was a disciple of Automedes of Mycenae; now, the teacher of the latter was Perimedes of Argos, who was also the teacher of other singers, and these are listed. A teacher-pupil diadochè is therefore delineated, as follows: Perimedes ↔ Automedes ↔ Demodokos Lacon. Moreover, a number of other singers are mentioned by name and all of them, like Automedes, were disciples of Perimedes of Argos. At the end a Timolaos is quoted and in all probability the reference here is to Timolaos of Larisa,38 a pupil of Anaximenes of Lampsacus and therefore more or less a contemporary (perhaps a little younger) of Demetrius of Phalerum. It is interesting to note that Timolaos contended that Demodokos was the brother of Phemios, the bard of the royal palace at Ithaca: we must therefore deduce from this that Timolaos agreed with the identification put forward by Demetrius for Clytemnestra’s bard and added his own contribution in the form of this family relationship and the idea that Phemios too acted as a guard. I cannot say whether it was Demetrius himself who cited Timolaos or whether the second citation was added by the source of the scholiastic material, but in any case Timolaos of Larisa was plainly dealing with the same problem and adopting a similar approach. A similar content is displayed by the citation from Tzetzes, Prolegomena to Lycophron, 4, 5ff. Scheer (146 = fr. 192 W; FGrHist 228 F 32b), where it is said that Demetrius of Phalerum mentioned a few very ancient poets who had written songs to be performed with the accompaniment of musical instruments: the names are those of the Kerkyraioi (sic: probable confusions in the tradition of the material?) Automedes, Demodokos and Chaeris, the Ithacan Phemios and others. I have not carried out in-depth research into all the names mentioned here, gathering together all the available evidence on these figures. I suspect that the evidence will turn out to be very scanty. However the aspect I wish to underline here is the general interest in very ancient poets, the bards who preceded Homer, an interest which usually starts out from the bards mentioned in the Homeric poems. This is a type of information found in the Lives of Homer, where there is often talk of poets prior to Homer: relations between such poets and  given any name and therefore Pausanias either does not know or does not accept the idea of Demetrius of Phalerum; the same can be said of Strabo 1.2.3, quoted above. 38 Also cited by Eustathius at Od. 11.521 (1697.57); his Homeric activities are known: he had doubled the Iliad by adding a line to each line: cf. Ziegler 1936.

  Demetrius of Phalerum on Literature Homer are frequently suggested, identifying them as his teachers or in some sense as part of a diadochè that included him, or even inserting them into his genealogical tree. The Lives of Homer were put together at a fairly late date, but there is a strong likelihood that the information contained therein derived from an erudition that may well go back to respectable antiquity. The inventions that developed around the bards of the Odyssey, around other figures of minstrels prior to Homer together with the genealogy and teachers of the poet are quite likely to have sprung up just as gratuitously as the other elements of a biographical tradition that had little to rely on in the way of personal declarations by the author about himself, but which nevertheless exploited as far as possible the information supplied by the poetic text. Indeed, peripatetic research followed this method with remarkable achievements. At a certain point this type of information evidently found suitable channels through which it took shape more concretely and a multifaceted biographical tradition began to form. As early as in Herodotus, the claim was made that Homer and Hesiod were the most ancient known poets, while the poets who were said to have lived before them were actually of a later date. But it was Aristotle, in the Poetics (4.4, 1448 b 27ff.), who codified not only the established image of Homer as the author of the Iliad, the Odyssey and the Margites, but also the idea of Homer as the most ancient poet whose works have come down to us: prior to Homer there were many other poets but, it was said, their works are lost and for us poetry starts with Homer. This historical-literary conception became the standard view in the Alexandrian age, both as regards Homer’s “authentic” works (Iliad, Odyssey and Margites) and also as regards his priority in time and the idea that before him there lived a multitude of poets whose works have not been preserved. Thus in one way or another, investigation touched on the problem of these poets of very ancient times, poets who lived before the “first” poet of Greek culture. The effort to bring the pre-Homeric poetry into sharper focus led to the attempt to recover some supposed data by starting from the texts and adopting a number of different procedures, irrespective of the type of text and with little heed for how far-fetched the conjectures might be. In order to organize these data, scholars resorted to cultural models that were already well established in collective intellectual consciousness and operated very effectively: genealogy, an extremely ancient structure for the organization of mythical and historical time, and the diadochè, linkage through succession based on a cultural and intellectual bond rather than on family relationships. Genealogical or teacherpupil connections, or both together, clearly provided a response to the need to establish structural links both within a pre-Homeric timescape shrouded in semi-legendary mists and also between this and the emergence of a figure such

Demetrius of Phalerum on Literature  

as Homer. Aristotle’s codification must have given a decisive impulse in this direction, which in some sense was part of an embryonic need for history of literature that would later be developed along various lines through the research of the erudites in the Hellenistic age. Here too I do not wish to stray beyond the bounds of plausible speculation, but it seems to me that the information concerning Demetrius of Phalerum fits very well into this framework, that is to say, the interest in the pre-Homeric and in providing a structured picture of this obscure age. Indeed, I would argue that the significance of this fragment goes beyond the mere curiosity of an erudite hankering after an ingenious solution to a zétema.39 I have no difficulty in acknowledging that in order to extract reasonably interesting results from these meagre remains it has been necessary to indulge in deductions and speculations, which do not constitute conclusive evidence and offer neither irrefutable data nor a large number of clues. Yet I would argue that overall this investigation has not been without foundation, above and beyond this or that specific point. If my arguments concerning Demetrius of Phalerum’s criticism of Homer are granted some validity as a whole, then I think we can perceive, albeit dimly, some of the ways in which peripatetic research on works of literature (poetry) and the personalities of authors (poets) acted as a crucial impetus along the road that led to Alexandrian philology: we can perceive the elements of a link that remains of primary importance, namely the relationship between Aristotle and his school and Alexandria.40

 39 Montanari 1981a, 97–114 (= this volume, ch. 30); 1992, 13–17; Blum 1977, 47–49; Arrighetti 1987; Gostoli 1986. 40 Rossi 1976; Blum 1977, 47–49, 99–107 and passim; Montanari 1993, partic. 259–264 (= this volume, ch. 1); Arrighetti 1987; Richardson 1994.

 The Peripatos on Literature Interpretation, Use and Abuse I have been asked to make some opening remarks for this conference that forms part of the well-known and important “Theophrastus Project” and I will just try to draw an outline, offering a few examples and making some general considerations concerning “the Peripatos on Literature”. This is, I believe, a subject worthy of new and deeper investigation and reflection, because the scrappy and scanty evidence makes it rather difficult to perceive its overall significance and content. So what I have aimed to do for this occasion is to cast a retrospective glance at the series of “Theophrastus Project” conferences held starting from 1981, a task which is greatly facilitated thanks to the website1 and of course through consultation of the published books. These materials constitute an impressive collection of scientific results, gathered together in over a quarter of a century. If I have counted correctly, the conference devoted to Chamaeleon and Praxiphanes (in 2007) was the 14th meeting of the “Theophrastus Project”;2 and now the 15th meeting, on Aristoxenus, has already been held at DePauw University (Indiana, USA) on September 2009. The beginning of the project saw attention focusing on the Stoic and Peripatetic Ethics, with regard to the work of Arius Didymus (1981); thereafter, Theophrastus was the theme most widely addressed (1983, 1985, 1989, 1993 and 1999), interspersed with Cicero (1987) and Peripatetic Rhetoric after Aristotle (1991). In 1995 the conference held in Boulder (Colorado) on Demetrius of Phalerum and Dicaearchus of Messene inaugurated a line of research which is the aspect that ideally leads us to this meeting on Chamaeleon and Praxiphanes. This line can be described as the study of the intellectual figures of the Peripatos who were pupils of Aristotle and Theophrastus and who can be dated to a period of time referring roughly to the second half of the 4th and 3rd century. The focus of this research is a renewed investigation into the possibility of reconstructing their personality, their work and their interests on the basis of a new edition and through reconsideration of the extant fragments. In addition to Theophrastus, attention has already dwelt on Heraclides of Pontus, Demetrius

 English Translation by Rachel Barritt Costa.  1 http://www.ucl.ac.uk/GrandLat/research/research-projects/theophrastus_extras/conf. 2 Swiss Institute and British School at Rome, Rome, Italy, 5–7 September 2007. https://doi.org/10.1515/9783110772371-056

The Peripatos on Literature Interpretation, Use and Abuse  

of Phalerum, Dicaearchus, Eudemos of Rhodes, Strato of Lampsacus (2005, not yet published), Lyco, Hieronimus of Rhodes, Aristo of Ceos; and subsequently, on Chamaeleon, Praxiphanes, and Aristoxenus of Tarentum (see above). Over the next few years we can probably expect Clearchus of Soli, Phaenias of Eresus and perhaps others as well, such as Critolaus of Phaselis. The themes addressed by the activity of this group of figures are very numerous and range over the most disparate fields of knowledge. Personally, during the course of my studies I have never dealt with ancient philosophy stricto sensu, and I have devoted attention to the majestic figure of Aristotle only as regards a particular sphere of his investigations, namely his work on language and on man’s use of language for the creation of works of literature in a general sense, from oratory to poetic genres. Substantially, this means that I have dealt with two works that have come down to us, the Rhetoric and the Poetics, and a few lost works, such as the Homeric Problems or the On Poets. More precisely, I would say that I have dealt with Aristotle’s reflection on poetic art as such, on its history, on the personality of the authors and on the interpretation of the works. The same kind of inquiry also underlies my interest in the school of Aristotle and the Peripatetic intellectuals, who form the object of the “Theophrastus Project”. This is why I took part in the above mentioned conference held in Boulder in 1995 on Demetrius of Phalerum and Dicaearchus: my subject was “Demetrius of Phalerum on Literature”.3 I would now like to build on a few aspects of that research and try to offer a few reflections that link it to the present context. From the scrappy and scanty evidence of a few fragments concerning Homer there emerged several results that seemed noteworthy, at least from my point of view. For example, I think it is rather interesting that, as far as one can tell, Demetrius reflected on the narrative plot of the Odyssey and its telos, following in the footsteps of Aristotle (in the Poetics) but to some extent adopting positions of his own (fr. 145 SOD = 193 Wehrli): we will come back to this topic below. What I want to go back to now is the fragment in Athenaeus, Deipn. 5.177 e–f (fr. 143 SOD = 190 Wehrli), which concerns Iliad 2.409. The subject it deals with occurs in the context of a discussion on the symposia and above all on who should be invited to attend. Athenaeus says that Homer also gives indications as to which persons do not require any invitation but instead come to a feast on their own initiative. An example is given from the chieftain’s dinner arranged by Agamemnon in Iliad book 2: Homer says that Menelaus came autómatos, of his own accord (2.408). Athenaeus comments (177 c): “For it is plain that neither a  3 Montanari 2000b (= this volume, ch. 55).

  The Peripatos on Literature Interpretation, Use and Abuse brother, nor parents, nor wife need be invited, nor anyone else whom one holds in equal esteem with these: otherwise it would be cold and unfriendly. And yet some authorities have added a verse which further explains the reason: ‘for he [scil. Menelaus] knew in his heart that his brother was troubled’ [v. 409] — as though it were necessary to tell the reason why a brother might come to dinner of his own accord...”. After a prolonged explanation of the absurdity of adding l. 409 to introduce a non-existent motivation, Athenaeus continues with an illustration of the opinions of the grammarian Athenocles of Cyzicus and those of Demetrius of Phalerum: “Demetrius of Phalerum declared that the inclusion of the verse ‘for he knew in his heart that his brother was troubled [scil. v. 409]’ is awkward and foreign to the poet’s style, and imputes meanness to the characters. ‘For’, says he [scil. Demetrius], ‘I think that every man of refinement has someone, either relative or friend, to whom he can go when a feast is on without waiting for an invitation’”.4 It appears that this specific problem of internal consistency of the Homeric passage was indeed a subject of discussion among Alexandrian philologists: Aristarchus interpreted the text in such a manner as to save the line, while others underlined the inconsistency and opted in favour of athetizing it. Demetrius of Phalerum is the most ancient authority known to have noticed these contradictory elements and to have believed the line to be spurious. It is likely that we have the remains of a zétema in genuine Peripatetic style, possibly on rules and customs in symposia, in which interpretation of the Homeric text also led naturally to debate on an issue of coherence. It seems not altogether unrealistic to suggest that this kind of Homeric criticism stimulated debate in Alexandrian intellectual circles: Demetrius of Phalerum’s observation on the passage of Iliad 2 may actually have given rise to a proposal of a Zenodotean obelós. The idea of considerations on rules and customs in symposia may have had a fitting place in the context of a treatise, whatever form the treatise may have had, concerning themes of behavioral ethics in the sphere of political social life, the family, or relations among individuals. It was quite natural to make reference to authoritative texts of the paideia sanctioned by the cultural tradition, first and foremost Homer. I would suggest that, seen in this light, the small fragment of Demetrius takes on considerable significance, especially if one reflects on the various routes that might cause observations springing from  4 Transl. by Gulick 1928. Athenaeus then goes on to discuss the passage from the Symposium where Plato (Smp. 174 b) also mentions Iliad 2.407 ff., with a disquisition as to whether or not it was suitable for Homer to present Menelaus as going autómatos to Agamemnon’s dinner, claiming that Menelaus ranked below Agamemnon.

The Peripatos on Literature Interpretation, Use and Abuse  

different contexts to prompt meditation on more strictly philological problems, and thus, as it were, to migrate and be transported under the Alexandrian sky (as Demetrius himself did). At that time my attention was also attracted by a fragment of Dicaearchus and now I would like to take the opportunity to re-examine it. Sch. Od. 1.332 preserves, presumably by means of an excerptum from Porphyry, an intriguing comment by Dicaearchus on the passage from the Odyssey in which Penelope makes her first appearance (fr. 95 Mirhady = 92 Wehrli). While the banquet of the suitors is in full swing at Odysseus’s palace, from her rooms on the upper floors Penelope hears the song of the bard Phemios and goes down into the hall, accompanied by two handmaidens. ἡ δ᾿ ὅτε δὴ μνηστῆρας ἀφίκετο δῖα γυναικῶν, στῆ ῥα παρὰ σταθμὸν τέγεος πύκα ποιητοῖο, ἄντα παρειάων σχομένη λιπαρὰ κρήδεμνα· ἀμφίπολος δ᾿ ἄρα οἱ κεδνὴ ἑκάτερθε παρέστη.

332

335

When Penelope, queen among women, had reached the suitors, she stood by the pillar that held the strong-built roof, holding her shining veil across her cheeks, and a loyal maid stood on either side of her.5 Sch. Od. 1.332 a. ἡ δ᾿ ὅτε δὴ μνηστῆρας κτλ.] αἰτιᾶται ἐκ τῶν ἐπῶν τούτων Δικαίαρχος [fr. 92 Wehrli] τὴν παρ᾿ Ὁμήρῳ Πηνελόπην ... οὐδαμῶς γὰρ εὔτακτον εἶναί φησι τὴν Πηνελόπην, πρῶτα μὲν ὅτι πρὸς μεθύοντας αὕτη παραγίνεται νεανίσκους, ἔπειτα τῷ κρηδέμνῳ τὰ κάλλιστα μέρη τοῦ προσώπου καλύψασα τοὺς ὀφθαλμοὺς μόνους ἀπολέλοιπε θεωρεῖσθαι. περίεργος γὰρ ἡ τοιαύτη σχηματοποιΐα καὶ προσποίητος, ἥ τε παράστασις τῶν θεραπαινίδων ἑκάτερθεν εἰς τὸ κατ᾿ ἐξοχὴν φαίνεσθαι καλὴν οὐκ ἀνεπιτήδευτον δείκνυσι. φαμὲν οὖν ὅτι τὸ καθόλου ἔθος ἀγνοεῖν ἔοικεν ὁ Δικαίαρχος. σύνηθες γὰρ παρὰ τοῖς ἀρχαίοις τὰς ἐλευθέρας γυναῖκας εἰς τὰ τῶν ἀνδρῶν εἰσιέναι συμπόσια. μαρτυρία δὲ…6 From these words Dicaearchus found fault with the Penelope in Homer ... He says that Penelope is by no means well-behaved, first because she comes into the presence of drunken young men, then because she covers the most beautiful parts of her face with the veil, allowing only her eyes to be seen. For such posturing is superfluous and pretentious, and the accompanying of the women servants on each side for the purpose of appearing outstandingly beautiful shows it was not unintended. However we say that Dicaearchus seems to be ignorant of the general custom. For it was common among the ancients that the free women entered the men’s symposia.7

 5 Transl. by Martin Hammond. 6 Pontani 2007, 171–173. 7 Transl. by Mirhady 2001, 95–97.

  The Peripatos on Literature Interpretation, Use and Abuse Dicaearchus’s critical observation on Penelope’s behavior is rebutted by pointing out that he has not taken into account the common custom according to which it was usual among the ancients for free women to take part in men’s symposia. As proof of this (μαρτυρία δέ ..., in the continuation of the sch., omitted above), mention is made of the Odyssean parallels of Arete and Helen, who are present, respectively, during the banquets at Alcinous’s palace among the Phaeacians (Od. 7 and 8) and at Menelaus’s royal palace in Sparta (Od. 4). Then, citing words by Nausicaa in Od. 6.287–288, it is stated that being in the company of males and taking part in male gatherings is improper for young girls, but not for married women. Thus Penelope has done nothing wrong by intervening personally to interrupt Phemios’s song on the sorrowful return of the Greeks from Troy, especially since the ancients believed that it was appropriate for a free woman, even if she was the daughter of a king, to devote herself personally to various tasks and activities. Therefore, her intervention is not out of place: rather, it shows sophrosyne,8 reminding the suitors how hateful their courtship is, given that the memory of her husband cannot be erased from her mind. There follow observations on the suitors’ behavior and then a long disquisition on the kredemnon (the veil) and on the gesture of covering the face made by various Homeric characters, particularly on occasions when tears are shed. The passage closes with two notes concerning the handmaidens who accompany Penelope: firstly, this too is a normal custom for the ancients, and thus cannot be interpreted as a malicious exhibition; moreover, these are devoted and wise handmaidens, not two of the treacherous and impudent ones mentioned in Od. 22.424, a fact that helps to shed light on the attitude of Penelope herself. Thus we have a long excerptum, which draws its inspiration from the passage in which Penelope makes her appearance: then, starting from an observation by Dicaearchus, a commentary is provided, debating in various ways on the “customs of the ancients” (archaioi, palaioi) with regard to the social behavior of women. The theme is examined from an extensively ethical perspective, but the “documentation”, so to speak, is entirely derived from Homer (Odyssey and Iliad), with citation of various passages to support the arguments put forward concerning the lines from Book 1 and in connection with the appearance of Odysseus’s spouse on the scene of the poem. It is clearly a stimulating passage of the Odyssey, on the subject of which Dicaearchus had at the very least advanced two considerations: Penelope does not show proper behavior when she appears among drunken men in the very middle of a symposium; the gesture of  8 Pontani 2007, 172.87–88: ὥστε ἡ μὲν παρουσία ἐκ τούτων οὐκ ἄτοπος, ἀλλὰ καὶ σωφροσύνην ἐμφαίνουσα.

The Peripatos on Literature Interpretation, Use and Abuse  

covering only a part of her face with the veil, leaving her eyes uncovered, and the fact of having two handmaidens at her side demonstrate a deliberately precious affected form of behavior, designed to highlight her own person. It is natural to assume that the attention is drawn by this rather theatrical appearance on the scene, which presents a major character as Odysseus’s spouse: perhaps the discussion revolved precisely around the way the poet had constructed this important narrative passage, or alternatively the observation may have belonged to a thematic context of cultural history such as that of the Βίος Ἑλλάδος.9 It is certainly a coincidence that the first consideration concerns the symposia, just as is the case for the above-mentioned fragment of Demetrius of Phalerum, and one must definitely not venture into hazardous connections. Of course, it is striking that later sources interested in Homeric exegesis preserved observations by two intellectuals of the early Peripatos that were so thematically consonant: I mean, not only as regards the theme of the customs of the symposium, but more generally as regards the behavioral ethics of archaic society as testified in the Homeric poems. This is indeed a far from secondary theme of the Greek paideia, which can also be seen as linked to the well-known sphere of critical reflection on the concepts of prepon and aprepès. Equally, it is interesting to note that such considerations were by no means ignored and neglected thereafter: on the contrary, they clearly prompted developments and debate in the framework of the rich subsequent Homeric exegesis. We have already seen that this was relevant for Demetrius, but it cannot be overlooked that the treatment preserved by the scholium (from Porphyry) in reply to Dicaearchus is centered on a comparison that is fully internal to Homer, aiming to delineate the Homeric specificity, τὸ Ὁμηρικόν, as laid down by the Alexandrian-Aristarchean method. Moreover, the notion of the historical nature of the moral concepts under discussion that is introduced in the treatment goes in the same direction. At this point, I cannot disregard another connection, likewise dictated by a fragment of Demetrius of Phalerum: namely, the one concerning the telos and the narrative plot of the Odyssey, which I mentioned earlier. Stobaeus (3.5.43) preserves a quotation from Hermippus,10 according to whom Demetrius of  9 Mirhady 2001 classifies the fragment (94–97, nr. 95) under the heading of “Contests, Literary Criticism” and not under that of “Cultural History”. 10 Taken from the work Συναγωγὴ τῶν καλῶς ἀναφωνηθέντων ἐξ Ὀμήρου: Fr. 92 Wehrli = 1026 F 59 Bollansée (FGrHist 4 A 3, with comm.). We know from Diogenes Laertius that Hermippus wrote on Demetrius of Phalerum (Diogenes Laertius 5.78, Demetrius of Phalerum fr. 69 Wehrli = 1, 78 SOD, Hermippus fr. 58 Wehrli = 1026 F 75 Bollansée, with comm.) and it is possible that he lifted Demetrius of Phalerum’s observation on Od. 23.296 to include it in a collection of καλῶς ἀναφωνηθέντα taken from Homer.

  The Peripatos on Literature Interpretation, Use and Abuse Phalerum claimed that line Od. 23.296 (οἱ μὲν ἔπειτα / ἀσπάσιοι λέκτροιο παλαιοῦ θεσμὸν ἵκοντο — “and they [Odysseus and Penelepe] / happily went to the place of their original bed”) had been composed by Homer εἰς σωφροσύνην (fr. 145 SOD = 193 Wehrli: Δημήτριος ὁ Φαληρεὺς εἰς σωφροσύνην ἔλεγε ταῦτα ποιεῖν). But what is the significance of the observation by Demetrius? He asserts that Homer composed the line εἰς σωφροσύνην, which can hardly mean anything other than “for, in view of, with regard to sophrosyne” (pudicitiae causa), or “according to, in conformity with sophrosyne”. Od. 23.296 is the line that ratifies the re-union of Odysseus and Penelope, the end of Odysseus’s wanderings and his return to his family: therefore the sophrosyne should reside, I would argue, in Odysseus’s preference for returning home and accepting the love of his legitimate spouse, rejecting the temptations and enticements that lured him during his journeys, as well as in Penelope’s patient wait for her legitimate husband’s return as against all the overtures and gifts from her suitors. All this sophrosyne of the adventures of the Odyssey is condensed and ratified at the moment when husband and wife finally return to their original bed; and the sch. comments ἀσπαστῶς καὶ ἐπιθυμητικῶς ὑπεμνήσθησαν τοῦ πάλαι τῆς συνουσίας νόμου, “with joy and with desire they revived the memory of the ancient custom of being together”. One may also wonder whether, by drawing attention to the passage and speaking of sophrosyne, Demetrius intended to refer to that of the character or that of the poet himself in creating the characters and their story with this ending. In any case, neither option in any way affects the basic fact, namely that Demetrius placed emphasis on this line on account of its value in relation to the meaning and moral value of the issue. The information that Aristophanes of Byzantium and Aristarchus had indicated the telos or peras of the Odyssey as residing in Od. 23.296 has given rise to great debate in modern criticism concerning the meaning to be attributed to this observation. Not a word about this question here: suffice it to say that there is likelihood that Demetrius of Phalerum’s focus on Od. 23.296 represented an early stand on the significance of this point of the poem. The actual verse itself does not seem to offer profound teaching as regards sophrosyne, but it acquires much greater depth of meaning if it is seen in context in the sense described above. Aristotle had given an exemplary synopsis of the poem’s main plot in his Poetics (ch. 17, 1455 b 16–23), indicating that it reached its goal with the suitors’ death and the rest consisted merely of episodes. Thus in Aristotle’s conception the telos of the Odyssey was achieved with Odysseus’s vengeance and his successful reappropriation of his home and his kingdom. I would suggest it is not unlikely that discussions on the main plot of the Odyssey and on the culminating point of a well-rounded narrative continued within Peripatetic circles subsequently to

The Peripatos on Literature Interpretation, Use and Abuse  

Aristotle’s statement, with Demetrius of Phalerum putting forward not exactly the same view, but a slightly different position as compared to that held by Aristotle. Moreover, such a position may have received attention in Alexandria and undergone further discussion among the grammarians. Demetrius’ view holds that the real culminating point of the poem, and therefore in some sense the enshrinement of its true meaning, lies in the moment when husband and wife are reunited and in the re-establishment of marital bliss between the protagonist and his faithful wife. If the first fragment of Demetrius concerned correct behavior in the symposia, here we are dealing with comments on Homer involving reflections about the behavior of wedded couples and its social relevance, on the basis of the story of Odysseus and Penelope. It is not far-fetched to cast one’s mind back to Dicaearchus’s critical observations on Penelope’s public behavior at the banquet of the suitors, seen above in the comment on Od. 1.332, with the arguments which, however, led to underlining the sophrosyne of Penelope: traces of discussions centering around ethical issues, or maybe of divergent opinions concerning Homeric passages and characters, arising in a Peripatetic environment? It is likewise interesting to recall that we have a fragment, in which Aristotle described as aprepès the fact that when Odysseus charges into the fray to restrain the routed army that is retreating in disarray after Agamemnon’s testing of the morale of the troops, in Il. 2.183 he takes off his mantle (chlaina) and is left with only his chiton on. Thereby Aristotle evokes the suggestion that Achean army had been brought to a halt by their amazement at the unseemly gesture of a character such as Odysseus (fr. 143 Rose = 368 Gigon: Sch. Il. 2.183 from Porphyry). The source has preserved no fewer than another six possible interpretations (introduced by ἄλλοι δέ and then by a series of οἱ δέ): such a doxography highlights the way this detail was extensively discussed, and shows that Aristotle’s remark was by no means isolated. In this particular point of the story Odysseus focuses attention on himself with a highly theatrical gesture, which obtains a spectacular result in an exceedingly striking scene that marks a turning point in the development of the action. The parallel with the case of Penelope commented on by Dicaearchus is noteworthy: here too discussion centers on an important narrative element, underscoring the way the poet calls attention to the action and accentuates it by attributing to the character a mode of behavior that borders on affectation and is open to criticism due to its unseemliness. The debate addresses the issue of questionable forms of behavior, of seemly or unseemly attitudes displayed by the characters, but one should not overlook that this occurs in relation to places that are important for the narrative plot: this can hardly be a coincidence.

  The Peripatos on Literature Interpretation, Use and Abuse We are thus dealing with a series of ethical considerations (on behavioral rules in the symposia and elsewhere, on archaic social and family customs) which build on motifs inspired by comments concerning certain particular Homeric passages. Later exegetic debate would continue to address these issues, but also exploring other avenues of enquiry, concerning topics like those we have already seen: the figure of Penelope, Odysseus’s manner of acting, the plot and the telos of the Odyssey and so forth. Particularly interesting from this point of view are the passages whose content was an object of discussion and criticism in the Peripatetic environment, and on which we later find the Alexandrian philologists dwelling in terms of textual problems such as a possible expunction or a choice among variants. Earlier we mentioned Il. 2.409; a similar case may be Il. 2.319: Aristonicus says the line was expunged, scil. by Aristarchus, while Zenodotus retained it, and we know that the passage had been discussed by Aristotle (fr. 145 Rose = 369 Gigon), who seems to cast doubt precisely on this line.11 I feel it is helpful, at this point, to recall that a fragment of Aristotle’s Homeric Problems (Sch. Od. 12.129 = fr. 175 Rose = 398 Gigon) shows us the grand master, the founder of the Peripatos, mentioned in connection with a genuinely allegorical interpretation, in a physical-cosmological perspective, of the passage from the Odyssey that talks of the cattle of the Sun (Od. 12.127–136). Sch. Od. 12.129 says that Aristotle φυσικῶς interprets the meaning of the number of cows, seven herds composed of fifty head of cattle apiece, in reference to the days of the lunar year, which number three hundred and fifty, i.e. precisely the product of seven by fifty. It is difficult to say whether the addition present in another sch. to the same verse, according to which the identical number of sheep refers to an equal number of nights, can be traced back to the complete interpretation by Aristotle himself or whether it is only a punctilious addition by others (as the formulation would seem to suggest). I believe it is plausible that Aristotle is the source of the entire exegesis and that as far as this passage is concerned he is reporting an interpretation he knew, which must have struck him as worthy of consideration, even though he had not necessarily put it forward himself. There need be absolutely no surprise on finding that Aristotle admits the ancient practice of allegory, in this physical case, as an exegetic possibility to be taken into consideration.12 For this too belonged to a consolidated and authoritative  11 I addressed this case in Montanari 2008 (= this volume, ch. 54). 12 Tulli 1987, 51–52, with other elements in this direction; Pfeiffer 1968, 237, is too cursory: “Plato and Aristotle rejected allegorism, and so consequently did the Academy as well as the Peripatos”.

The Peripatos on Literature Interpretation, Use and Abuse  

tradition, which he contemplates with the historical attitude of the scholar who enquires into precedents and takes into account the reflections put forward before his time, as happens not only for philosophy but also for rhetoric and naturally for poetry itself in famous parts of the Poetics. Aristotle took a novel interest in what we refer to by the overall term of “literature”. In the first place, this accords with his marked and systematic interest in the history of the various disciplines: considerable attention was devoted to erudite research and antiquarian collections, with an effort of historical documentation in the spheres of thought in which Aristotle elaborated a doctrine of his own. The link between the ordered collection of opinions expressed by predecessors (from which doxography drew its origin) and theoretical reflection appears as a characterizing intellectual trait: the scientific foundation of a discipline cannot disregard conscious knowledge of its history, and this holds true for rhetoric and poetics as well. Almost the entire range of available information on the prior techne rhetorike is due to Aristotle, even if his collection of Technai is lost. But special attention should be devoted to everything that can be traced back to research on literature. The Homeric Problems and the dialogue On Poets have already been mentioned; to these can be added the treatise On Tragedies, of which we have only the title. If we cite the collection of Didascaliae and the list of Olympic and Pythic winners, their relevance for the works, respectively, of classical drama and choral lyric is unmistakable; the investigations on proverbs certainly had a relation to the poetic texts in which the paroimiai are often taken up again; in the Athenaion Politeia one notes copious utilization of the political elegy of Solon, and moreover, from a different point of view, the Constitution of Athens is likely to have had considerable relevance for the interpretation of allusions and historical references in works such as that of old comedy.13 The great collection of historical-antiquarian erudition amassed by Aristotle and his school undoubtedly had an organic connection with the sphere of history of literature, with study of the works and reconstruction of the lives of the authors. There was a convergence between literature as a source of information and literature as the object of exegesis and commentaries: the link between erudite documentation and text interpretation was profound and left a fertile training in work and method. This background was then meaningfully continued both in the work of some scholars of the Peripatetic school and also in the criticalexegetic activity of Alexandrian philology, stimulating lively intellectual ferment and an extraordinary thrust.

 13 Montanari 2006.

  The Peripatos on Literature Interpretation, Use and Abuse Together with erudite research on the works and the authors, Aristotle devoted himself to theoretical reflection and constructed a doctrine of his own on the issue of the techne poietike: as in other fields of learning, these two aspects could not remain separate. Although the loss of the materials is a serious impediment, one certainly cannot detach the theoretical discourse on epic literature and tragedy conducted in the Poetics from the shreds of evidence provided by titles and fragments which hint at study and analysis of questions of various kinds prompted by the texts. In any case even the Poetics, despite its speculative theoretical aim, clearly shows that the starting point is observation of reality, its history and its products. Reality, in this context, is the result of the activity of man in the field of poetic art, that is to say concretely the evolution of literary genres from Homer onwards. Without any need for demonstration, Aristotle presupposes straight out that poetry is a techne, that it belongs to the range of human activities governed by rationality and is subject to identifiable rules, and that it is distinguished by its own essence and by the aim that is intrinsic to it. This has been an unquestionable change of perspective. If Plato did not grant poetry a value of knowledge inasmuch as it was imitation of sensible things, for Aristotle poetry generates knowledge in its own right by virtue of being mimesis of nature, which is indeed genuine reality: not, however, in the sense that poetry imitates the accidental detail, but rather in that it imitates the universal, because there is no true knowledge unless it be that of the universal (and mimesis produces pleasure because knowledge produces pleasure). Thus, if Plato set the education springing from poetry in opposition to that from philosophy, obviously to the advantage of the latter, for Aristotle poetry belongs to philosophy because it is knowledge of the universal: and this is a decisive positive appraisal. The epistemological plane is unequivocally accompanied by the plane that concerns the end (telos) and the effect. In the Aristotelian vision, the value of the products engendered by the art of poetry depends on men’s capacities (rather than on divine inspiration) and, further, on accomplishment of the specific end, i.e. knowledge through mimesis, more than on the fact of teaching good and appropriate things and educating to the good, in the sense of not conveying harmful messages or indecent content (condicio sine qua non for poetry to be accepted). Naturally, for Aristotle, not only is the possibility of an ethical function of poetry by no means excluded: on the contrary, its ethical function is a value to be pursued and achieved in appreciating and interpreting the works of the poets. Yet the essential and specific aim and function of poetry is of a cognitive order in relation to the intellect, and of a psychological-emotive order in relation to pleasure and the passions (as in the case of tragic catharsis). I would argue that one of the most significant passages is to be found in the initial

The Peripatos on Literature Interpretation, Use and Abuse  

part of Chap. 25, where Aristotle explains that there are two possible mistaken approaches in the art of poetry, the specific mistake that concerns poetics in its own right, and the accidental mistake that concerns any other discipline, such as politics or medicine: if an impossible thing is represented, but the poem achieves its intrinsic aim, then the result is good and positive. Poetic art is a success or a failure in relation to itself and not to any other techne or discipline (including history, as stated in another famous passage from Chap. 9, and including political and moral education). Like any other sphere of human activity, poetry has a value of its own, an independent value, and it must be studied and understood with the aid of the appropriate tools. As compared to the idea that poetry should offer righteous moral messages that are valid for society and the state, and teach correct and true things to shape the individual, this novel approach embodied an intellectual change of enormous importance: investigating and explaining the world, the nature and life of men also included investigating and explaining the fundamental texts of men’s culture and the sphere of knowledge that concerns literature and poetry. Aristotle’s thought thus cast the consideration of literature in a completely different light. In the global vision of his philosophy, the cultural models and the great poetic creations in which the heritage of the Greek paideia was concentrated were a segment of the world, an important portion of the history of civilization and of men, a treasure that was to be studied and known, interpreted, understood. Thus were laid the intellectual foundations that explain not only certain interests and lines of enquiry cherished by the Peripatetics, but also the subsequent long-lasting developments of the great season of Alexandrian philology, which in this respect is the progeny of Aristotle in the most profound sense. With his reflections, his activity and his teaching, Aristotle brought about a cultural change and molded the Peripatos according to an approach which is clearly recognizable among his pupils, despite the almost total loss of their writings.14 We can now return to the subject broached in the first part of this paper, examining a few examples of exegesis and comments on some Homeric passages by Aristotle himself, as well as by Demetrius of Phalerum and Dicaearchus. The presentation of zetemata on matters of an ethical nature, on the behavior of characters, on habits and customs of various kinds certainly owed a major debt to the traditional themes concerning the end and the educational effect and the moral value of poetry, just as the resumption of allegorical exegesis  14 Cf. Richardson 1992 and 1994; useful overview with collection of the fragments concerning drama in Bagordo 1998.

  The Peripatos on Literature Interpretation, Use and Abuse had roots in a distant past and a well known continuity. The sources that have preserved these fragments clearly reveal that the well defined questions raised in the comments of these Peripatetic figures later became the object of further discussion, with recognizable philological-exegetic developments. There is a distinct perception of standing on the crest of the ridge, in the intellectual phase during which the exegetic intent is making headway and gaining ground as an independent value, also based on a retrospective vision that encompasses awareness and knowledge. Let us imagine a setting that places us in the last decades of Aristotle’s activity, accompanied by the first generation of his pupils. We have already mentioned Demetrius15 and Dicaearchus. With regard to the latter, we may briefly add that his production in the field of studies on literature included not only Homeric criticism but also writings On Alcaeus, On Dionysiac Contests, On Musical Contests, as well as a collection of tragic hypotheseis.16 Theophrastus’s interests in the fields of rhetoric and poetics are well documented, with titles such as On the Art of Rhetoric, Rhetorical Precepts, On Style, On (the Art of) Poetry, On Comedy.17 Among the very poor shreds of evidence on Phaenias of Eresus one also finds an On the Poets, and among those on Eudemos of Rhodes there is also a Περὶ λέξεως. Strato of Lampsacus was generally known as “the physicist” (ὁ φυσικός): literature and the art of poetry do not seem to play a role in his production, but the title Λύσεις ἀπορουμένων, Solutions of difficulties, may be attractive from our point of view. It is likewise interesting to note that Strato, who was the head of the Peripatos after Theophrastus, spent a period of time in Alexandria at the court of the Ptolemies,18 as did Demetrius of Phalerum. Unfortunately we are told almost nothing explicitly about their activities and the cultural influence, but this link in the chain cannot be neglected. Heraclides of Pontus, an Academic and associated with Speusippus, did not officially belong to the Peripatos, but his relations with the Aristotelian school have been ascertained and indeed explain some of his specific interests, first and foremost his well documented literary interests. With regard to Heraclides and Strato, see the proceedings of the 2005 conference of the Theophrastus Project in Grenoble (Strato: RUSCH XVI) and those of 2003 in Leeds (Heraclides: fragments edited in RUSCH XIV, discussion in RUSCH XV).

 15 Survey of his fragments on literature in Montanari 2000b (= this volume, ch. 55). 16 Frr. 1.14, 1.15, 1.17, 1.18, 89–115 Mirhady 2001, 89–115; overview of the problem of the hypotheseis in Luppe/Montanari 1992; cf. Luppe 2001. 17 Fr. 666–709 Fortenbaugh et al. 1992. 18 Cf. D.L. 5.58–60.

The Peripatos on Literature Interpretation, Use and Abuse  

Aristoxenus is known above all for his theoretical treatises on music. His musicological writings built on the results of previous thinking and became a point of reference: here too the interwoven presence of erudite documentation and theoretical reflection can clearly be perceived. The subject of music was also investigated by Heraclides of Pontus and Dicaearchus; the former wrote a treatise On Music and a collection on the most important composers, the latter on musical competitions. A moment’s reflection on lyric and drama suffices to realise the connection of music with poetry and the intensity of the link between the two fields of research. But Aristoxenus was also the author of a number of biographies, above all of philosophers (Pythagoras and his followers, Archytas, Socrates, Plato), but of scholars from other disciplines as well, given that he wrote a bios of the dithyrambographer Telestes and a work On the Tragic Poets. We have noted several times that there was considerable interest in literary works and in the personality of the authors. The focus on Aristoxenus prompts us to extend our field of interest to biography, a theme to which another fragment of Demetrius of Phalerum’s Homeric criticism can be linked. I mean the fragment provided by the Sch. Od. 3.267 (fr. 144 [+146] SOD = 191 [+192] Wehrli) and concerns the famous lines that tell how Agamemnon had entrusted his wife to a bard when he left for Troy. πὰρ δ᾿ ἄρ᾿ ἔην καὶ ἀοιδὸς ἀνήρ, ᾧ πόλλ᾿ ἐπέτελλεν Ἀτρεΐδης Τροίηνδε κιὼν εἴρυσθαι ἄκοιτιν. ἀλλ᾿ ὅτε δή μιν μοῖρα θεῶν ἐπέδησε δαμῆναι, δὴ τότε τὸν μὲν ἀοιδὸν ἄγων ἐς νῆσον ἐρήμην κάλλιπεν οἰωνοῖσιν ἕλωρ καὶ κύρμα γενέσθαι.

267

270

And there was also a bard with her (Clytemnestra), whom the son of Atreus had earnestly charged, as he left for Troy, with the guarding of his wife. But when the time came when the fate of the gods tied her to subjection, then Aigisthos took that bard to a deserted island and left him there to be the prey and food for birds.19

Homer does not tell us the name of the bard and it is only natural that speculation on this matter has been rife. This passage was a subject also addressed by Dicaearchus, who says that “the ancients thought the singer was also wise, as is clear from the one left with Clytemnestra”, but no name is given.20 The excerptum  19 Transl. by Martin Hammond. 20 Fr. 39 Mirhady = 93 Wehrli, from Phld., On Music, 32.21 Kemke. It seems that Aristoxenus also (Fr. 123 Wehrli) touched on the same theme, but I am not sure if the material immediately following the name of Aristoxenus in Strabo 1.2.3 is also to be assigned to Aristoxenus himself: here too no name of the bard is given.

  The Peripatos on Literature Interpretation, Use and Abuse of Demetrius of Phalerum places the issue in a broader context, which cannot be summarized here:21 he presents a solution to identify the anonymous character and furnishes a few elements of his biography; furthermore, he sets this character in a complex diadochè of ancient bards (which also includes Phemios, the bard of the royal palace at Ithaca). The aspect I wish to underline is the general interest in very ancient poets, who preceded Homer, an interest which usually starts out from the singers mentioned in the Homeric poems. This is a type of information found in the Lives of Homer, where there is often talk of poets prior to Homer: relations between such poets and Homer are frequently suggested, identifying them as his teachers or in some sense as part of a diadochè that included him, or even inserting them into his genealogical tree. The inventions that developed around the bards of the Odyssey, and around other figures of minstrels prior to Homer together with the genealogy and teachers of the poet, are quite likely to have sprung up just as gratuitously as the other elements of a biographical tradition that had little to rely on in the way of personal declarations by the author about himself, but which nevertheless exploited as far as possible the information supplied by the poetic text. At a certain point this type of information evidently found suitable channels through which it took shape more concretely: thus a multifaceted biographical tradition began to form.22 Indeed, Peripatetic research followed this method with remarkable achievements. Reconstructing the life of an author by starting out from an interpretation of passages from his works, discovering real or presumed biographical elements, exploiting — and overexploiting — the texts to this end, is a method which apparently had one of its most fertile applications in the activity of Chamaeleon. The lyrical poets offered rich pickings for this type of work; with others, such as the tragedians, it was certainly more difficult, and Homer no doubt severely tested their speculations. Use and abuse of literature took place copiously. This brings me to the closing remarks of my paper, and the real beginning of this conference, during which we will listen to in-depth analyses on the above-mentioned figures of Chamaeleon and Praxiphanes. In what remains of their works it can be noted that the study of literature is awarded a much more important and significant place in comparison to the earlier Peripatetics cited so far, even though they were separated by no more than just a few decades. Having metabolized the teachings of Aristotle ever more deeply, the study of literary

 21 Cf. Montanari 2000b, 406–410 (= this volume, ch. 55). 22 Montanari 1992, 14–17; Graziosi 2002.

The Peripatos on Literature Interpretation, Use and Abuse  

works and research into the authors was by now a frequently perused field of enquiry and of rewarding investigation. By the early decades of the 3rd century BCE the term grammatikὸs appears to have become a specialist term to designate a philologist who is an exegete of poetry, whereas it previously referred to a teacher at an elementary level. The ancients took note of the change in meaning and raised the question of who had been the first to receive the epithet in its new sense. There was a tradition that pointed to the Peripatetic Praxiphanes, a pupil of Theophrastus: this piece of information can be linked to the conception that regards Aristotle as the initiator of grammatikè in the sense of “the erudite study of literature” and as the ideal master and source of inspiration, with his school, of the Alexandrian grammatikoi.23 Praxiphanes was roughly a contemporary, probably just slightly older, of Zenodotus of Ephesus, the first librarian of Alexandria. The attitude of Aristotle and of the first generations of intellectuals in the Peripatos marked a genuine novelty within the Greek cultural framework between the 4th and 3rd century BCE. The surge of interest in literature sprang from the importance attributed by Aristotle to poetic-literary creation as a sphere of human activity. Furthermore, it was profoundly rooted in the attitude of great attention he displayed towards the related historical-cultural themes. The value attached to the personalities and the works representing the consolidated cultural heritage also increased the weight acquired by an intellectual who was capable of interpreting and understanding this highly important heritage. The Peripatetic school and the Alexandrian school were the two avenues opened up by that cultural vision.

 23 Pfeiffer 1968, 156–159.

 Un nuovo frammento di commentario a Callimaco Si tratta di un frammento di volumen (cm. 3 × 5,3) di provenienza sconosciuta, acquistato sul mercato antiquario del Cairo:1 il recto contiene registrazioni di conti di arourai,2 il verso resti di 14 righi di una colonna di testo letterario, per il quale il rotolo sembra essere stato riutilizzato. Non sono conservati resti di margini né di intercolumnio, per cui nulla si può dire sulle dimensioni del rotolo né sull’ampiezza della colonna. Un piccolissimo frustulo è stato accostato nell’angolo a sinistra in basso, ma l’unione dei due pezzi lascia dei problemi: cfr. le note al testo. La scrittura, di tipo corrente e senza particolarità grafiche di rilievo, è abbastanza fìtta, bilineare (scendono sotto il rigo ρ, ι e leggermente χ), vergata con scioltezza e con una certa regolarità; il modulo è piuttosto piccolo, ma i tratti sono notevolmente marcati; alcune lettere, come υ, γ, π (che ha il tratto verticale destro incurvato), presentano alla base dei trattini di coronamento; compaiono qua e là gruppi di lettere attaccate, come α e ι. Su base paleografica, si può assegnare la scrittura del nostro testo al II sec. d. C.3 La lettura è in alcuni punti incerta a causa dei danneggiamenti subiti dal papiro. Unico segno di lettura visibile, i due punti su ο iniziale al r. 7. La caratteristica esterna più evidente è la frequente presenza di spazi bianchi, particolarmente ampi e numerosi negli ultimi righi, dove nessun mezzo tecnico ha permesso di vedere tracce di scrittura (cfr. infra). L’esiguità di quanto è conservato non permette di raggiungere grandi risultati, ma si può forse arrivare a formulare una ipotesi sulla natura del testo cui il nostro frammentino apparteneva. Premettiamo comunque la trascrizione, alla quale faremo seguire il commento.

 1 Il papiro (un pezzo maggiore cui è stato aggiunto un frustulo minimo, cfr. n. ai rr. 12–14 del testo greco) è stato acquistato al Cairo una decina d’anni or sono dalla prof.ssa Edda Bresciani, che ne ha fatto dono al prof. Emilio Gabba: sono grato al prof. Gabba di avermi mostrato il frammento e di avermi concesso, con grande generosità e liberalità, di studiarlo e pubblicarlo. Devo un sentito ringraziamento a G. Arrighetti, A. Carlini ed E.G. Turner, che mi hanno fornito aiuti e consigli determinanti; a I. Crisci, che si è curato con grande perizia del restauro. 2 Cfr. Appendice. 3 Cfr. P. Med. 17 (comm. ad Antimaco di Colofone); P. Oxy. 2636 e 2637 (comm. a lirica corale). La datazione è confermata da E.G. Turner. https://doi.org/10.1515/9783110772371-057

Un nuovo frammento di commentario a Callimaco  

1

5

10

14

– – –

– – –

–

–––

]. παραπο . . [ ]γυναιϰασ .[ ].σϰατεχεασ [ ]αλοναυλειου[ ].ου ελλειπε[ ]ντομη̣ε̣ ξ̣ιναι[ ]ενεϰο̈ υδ̣ο̣.αυ[ ]σ[.]μβαλονϰα̣[ ]υ̣ [ ]λ̣ ιον [ ]χεισϑαι [ ]. . [ ]ω̣ι̣ [ ] [ ]πο . λα . . . [ –––

↓

1

–––

].παρα πο . .[ ]γυναῖϰας .[ ].ς ϰατέχεας [ ἐϰδοὺς σάμβ]αλον αὐλείου[ 5 ].ου ἐλλείπε[ ]ν τὸ μὴ̣ ἐ̣ ξιναι[ ]ενεϰ ο̈ ὐδ̣ο̣ῦ̣ αυ[ ]σ[ά]μβαλον ϰα̣[ ]υ̣ [ 10 ]λ̣ ιον [ ]χεισϑαι [ ]. . [ ]ω̣ι̣ [ ] [ 14 ]πο . λαν̣ . . [ –––

1. La prima lettera potrebbe essere η; dopo o si può leggere π oppure τι: Turner esprime una preferenza per τι. 2. Dopo lo spazio bianco, traccia di una lettera: λ, α oppure χ (Turner). 3. Della lettera iniziale resta solo un brevissimo tratto orizzontale mediano. Dopo ϰατέχεας si ha certamente uno spazio bianco, come mostra inequivocabilmente il sigma con il tratto superiore allungato (cfr. r. 2); non è certo che dopo lo spazio sia conservata una traccia di scrittura. Per questa forma di sigma ‘finale’ usato prima di spazio bianco interno al rigo, cfr. per es. P. Oxy. 2636 (comm. a lirica corale). 4. Callimaco, fr. 631 Pf. (cfr. sch. T ad Il. XVI 235). 5. La traccia iniziale potrebbe essere di un λ, di un μ o di un δ; più difficile α. 7. Dei due segni sopra ο quello a sinistra potrebbe essere il resto di un apostrofo; dell’υ finale di oὐδ̣ο̣ῦ̣ non resta quasi nulla, ma le tracce non si oppongono a tale lettura. 12–14. L’angolo sinistro in basso (. ./ω̣ι̣ /πο.) è costituito da un frammentino staccato, che è stato collocato nel restauro operato da I. Crisci, il quale ritiene certo il saldarsi dei due pezzi. Anche D. Foraboschi, che ha esaminato il testo documentario del recto (cfr. Appendice) è di questa opinione. In base a questi convergenti pareri, mettiamo da parte — pur conservando qualche perplessità — i dubbi che suscita la scrittura letteraria, nella quale sembrerebbero differenti la pressione del calamo e la forma delle lettere: può trattarsi di impressione fallace dovuta all’esiguità delle lettere testimoniate o anche a cause materiali. In ogni caso, l’accostamento lascia irrisolti alcuni dubbi di lettura: cfr. nn. segg.

  Un nuovo frammento di commentario a Callimaco –

– –

12. Turner propone ε̣ ξ̣ oppure ε̣ ι̣ : poiché la prima lettera è spaccata in due dalla frattura, egli evidentemente pensa che il tratto mediano dell’ ε sia andato perduto nell’accostamento. Non è impossibile ο̣ ι̣ . 13. La lettura accolta nel testo è di Turner. 14. In un primo tempo avevo letto πολλα (letto anche da Turner e da Crisci), ma in seguito ho dovuto revocare in dubbio il primo λ: sembra di vedere un trattino obliquo in alto a destra nella lettera, che porterebbe a leggere un χ, ma leggermente diverso da quello del r. 3. Il dubbio di lettura è accresciuto dal fatto che la lettera si trova proprio nel punto di frattura, metà da una parte e metà dall’altra (cfr. n. 12–14).

Al r. 4 ]αλον αυλειου[ è certamente riferibile al fr. 631 Pf. di Callimaco, conosciuto per tradizione indiretta dallo sch. T ad Il. XVI 235: † ἧσεν ἐϰδοὺς σάμβαλον αὐλείου. Mi pare che l’estrema rarità dell’espressione non lasci dubbi sull’ identificazione:4 non c’è dubbio che al r. 8 si debba leggere σ[ά]μβαλον. Al r. 5, dopo lo spazio bianco, si ha certamente una forma del verbo ἐλλείπω (si può facilmente proporre come integrazione ἐλλείπε[ι ο ἐλλείπε[ται): la presenza di questo verbo tipico del linguaggio scoliografico,5 gli spazi bianchi, il ricomparire della rara forma σάμβαλον a pochi righi di distanza, sono tutti elementi che convergono sull’ipotesi che si abbia a che fare con un frammento di commentario a un’opera (non sappiamo quale) di Callimaco.6 Considerando gli spazi bianchi, possiamo con verisimiglianza pensare che il lemma, di cui faceva parte il fr. 631 identificato al r. 4, proseguisse nel r. 5 e arrivasse fino allo spazio bianco prima di ελλειπε[ : la desinenza ].ου apparterrebbe dunque al testo di Callimaco. Dopo lo spazio bianco, iniziava il commento, nel quale tra l’altro veniva probabilmente spesa qualche parola per il raro ed arcaico σάμβαλον (cfr. r. 8). Il sigma con il tratto superiore allungato viene usato prima dello spazio bianco, per cui non c’è dubbio che dopo ϰατέχεας c’era un vuoto (cfr. n. al r. 3): si può dunque supporre che il lemma in questione iniziasse a questo punto del r. 3. Se abbiamo correttamente delimitato la porzione di testo callimacheo, a titolo d’ipotesi si può riproporre la possibilità, già avanzata in  4 Fra le attestazioni della parola αὔλειος registrate in LSJ s. v., nessun’altra è preceduta dalla terminazione -αλον. 5 Negli scoli ἐλλείπω viene usato con particolare frequenza in riferimento a una preposizione (la deplorevole povertà di indici dei corpora scoliografici rende difficili tali ricerche: si può usare Baar 1961; cfr. per es. sch. ad Il. A 65, 93–5, 111, etc.). Nel nostro caso, è possibile che il discorso si riferisse alla ἔλλειψις della preposizione ἐϰ davanti al genitivo nell’espressione ἐϰδοὺς σάμβαλον αὐλείου (cfr. però infra i problemi esegetici relativi al frammento). 6 E.G. Turner è d’accordo su questa identificazione.

Un nuovo frammento di commentario a Callimaco  

forma diversa da Schneider,7 che si abbia a che fare con un distico elegiaco. La terminazione in -ου sia di αὐλείου che della parola mutilata al r. 5 ci fa venire in mente, infatti, come sia frequente la rima interna fra le due parti del pentametro che si potrebbe allora ricostruire così: σάμβαλον αὐλείου [⎼ ⏑⏑ ⎼ ⏑⏑].ου. Al r. 2 troviamo γυναῖϰας seguito da uno spazio bianco, dal che si può ipotizzare che qui finisse il commento ad un lemma precedente. Per quanto riguarda ciò che era contenuto fra lo spazio bianco del r. 2 e quello del r. 3, ci troviamo di fronte a notevoli difficoltà. La seconda persona singolare ϰατέχεας non può certo appartenere al discorso di commento, ma d’altra parte non ci sentiamo neppure di attribuirla sic et simpliciter al testo di Callimaco: l’usus normale del poeta richiederebbe ϰατέχευας (cfr. fr. 260, 11; Hymn. VI 5; Ind. Vocab. s. v. χέω). Se abbiamo ricostruito bene la struttura del testo, l’esiguità dello spazio a disposizione inviterebbe a pensare ad un breve lemma, seguito da una glossa o parafrasi terminante con ϰατέχεας.8 Comunque stiano le cose, questa parola, per la forma in cui si presenta, non può essere del tutto estranea al testo callimacheo: da essa ricaviamo che nel passo di Callimaco doveva esserci certamente un verbo all’aoristo II sing., che fosse ϰατέχευας o qualcosa di significato analogo. La II pers. sing. testimoniata da ϰατέχεας, comunque, è in accordo con il sing. ἐϰδούς, trádito indirettamente: essa implica o un discorso diretto o un’apostrofe, per cui viene ad essere impossibile la correzione ἤιεν proposta da Schneider9 per il corrotto ἧσεν tràdito dallo scolio omerico: Pfeiffer avanza la possibilità che ἧσεν sia corruzione di φησίν e non sia quindi parola di Callimaco, ma dello scoliasta.10 Purtroppo il nuovo papiro non consente di risolvere il problema. Nello sch. T ad Il. XVI 235 il passo di Callimaco è riportato nel commento alla parola omerica ἀνιπτόποδες, attributo dei Selli, sacerdoti di Zeus Dodoneo.11  7 Schneider 1873, fr. 492, riteneva che il frammento appartenesse a un distico e che σάμβαλον αὐλείου costituisse il primo hemiepes del pentametro (cfr. n. 9). 8 Non impressiona la presenza di lemmi di proporzioni anche molto diverse: nei numerosi frammenti di hypomnemata a noi giunti, si va da lemmi formati da una sola parola, commentata o anche solo glossata, fino a lemmi formati da più versi. Solo per fare qualche esempio, si veda il comm. ad Il. B 751–827 del P. Oxy. 1086; il comm. ad Antimaco del P. Med. 17; il comm. a Nicandro, Ther. 377–95 del P. Oxy. 2221; etc. 9 Schneider, loc. cit., ricostruiva: [οὐδὲ] . . . . . ἤιεν ἐϰδοὺς σάμβαλον αὐλείου. 10 Pfeiffer ad fr. 631: «fort. ἧσεν non poetae vocabulum est, sed scholiastae: ϰαὶ K. φησιν». 11 Si tratta della famosa preghiera di Achille a Zeus prima dell’ingresso in battaglia di Patroclo, vv. 233–5:

  Un nuovo frammento di commentario a Callimaco Fra le varie spiegazioni di questo aggettivo fornite dallo scolio, una suona: μὴ προϊόντες τοῦ ἱεροῦ, ὡς μὴ δεῖσϑαι νίπτρων – ϰαὶ Καλλίμαχος “ἧσεν ἐϰδοὺς σάμβαλον αὐλείου.”.12 Non sappiamo se αὔλειος sia usato come sostantivo o come aggettivo:13 il processo di epitomazione e contrazione che ha portato dagli hypomnemata agli scoli potrebbe essere stato responsabile della perdita del sostantivo nella citazione e — se abbiamo correttamente delimitato il lemma — la desinenza -ου (alla fine del verso, secondo la nostra ipotesi) potrebbe essere di un sostantivo al genitivo accordato con αὐλείου (v. anche infra). Sul piano esegetico, non mi sembra che il nostro papiro porti un contributo decisivo: il significato delle parole callimachee resta estremamente dubbio. Attenendosi allo scolio che le cita, si tratterebbe di un’espressione indicante il non uscire dal tempio o comunque da un qualsiasi luogo chiuso; si dovrebbe intendere dunque «avendo messo sandalo fuori dalla porta esterna», dove «sandalo» starebbe per piede, e bisognerebbe supporre la caduta di una negazione («negatio desideratur»: Pfeiffer ad loc.), cioè «non avendo messo piede fuori dalla porta esterna», per giustificare il parallelo istituito dallo scoliasta col significato di ἀνιπτόποδες = che non escono dal tempio, sì da non aver bisogno di lavacri. Seguendo questa interpretazione, si può confrontare la formulazione dello stesso concetto nello sch. AD al medesimo verso: τινὲς δὲ αὐτοὺς διὰ τοῦτο λέγουσιν ἀνιπτόποδας, ὅτι οὐϰ ἐξίασιν ἔξω τοῦ ἱεροῦ; e nello sch. Ge: ποδονίπτροις μὴ χρώμενοι διὰ τὸ ϰαϑαρὸν ϰαὶ ἁγνὸν ϰαὶ διὰ τὸ μὴ ἐξιέναι τοῦ σηϰοῦ. Ciò porterebbe a correggere εξιναι del r. 6 in ἐξιναι, ottenendo τὸ μὴ ἐξιέναι, che dovrebbe essere parafrasi delle parole di Callimaco.14 Questa interpretazione, che ha in suo favore il fatto di fondarsi sulla testimonianza dello scolio che cita il passo, lascia tuttavia qualche perplessità per la durezza dell’espressione e anche per la necessità di supporre la caduta accidentale di una negazione così importante per il senso del parallelo (a meno che essa non si nasconda sotto il

 Ζεῦ ἄνα, Δωδωναῖε, Πελασγιϰέ, τηλόϑι ναίων, Δωδώνης μεδέων δυσχειμέρου· ἀμφὶ δὲ Σελλοὶ σοὶ ναίουσ᾿ ὑποφῆται ἀνιπτόποδες χαμαιεῦναι. Sui Selli cfr. Bürchner 1923, 1323; Bölte, 1912; Cook 1903, 179 sgg.; Cook 1925, II, 677 sgg.; Nilsson 1955, 423 sgg.; soprattutto Parke 1967, partic. 20 sgg. 12 Cfr. lo stesso concetto negli scoli A e Ge, riportati infra, dove però non compare la citazione di Callimaco. 13 Entrambi gli usi sono possibili: v. LSJ s. v.; la parola non è usata altrove da Callimaco. 14 Per il tipo di errore cfr. per es. P.Oxy. 1800 fr. 2, col. II 46: απιναι per ἀπιναι. Sarebbe possibile anche τὸ μὴ ἐξ(ε)ῖναι (προϊέναι τοῦ ἱεροῦ): mi sembra meno probabile. Le prime tre lettere (ντο), poi, potrebbero anche costituire una desinenza verbale.

Un nuovo frammento di commentario a Callimaco  

corrotto ἧσεν). Non è da escludere che il significato letterale dell’espressione fosse diverso, per esempio «avendo messo il sandalo (calzare) fuori dalla porta esterna»: ogni interpretazione dovrà comunque render conto del parallelo che lo scoliasta sembra istituire con la sua spiegazione della costumanza dei Selli ἀνιπτόποδες. I nostri sforzi esegetici sono d’altra parte sottoposti al dubbio di fondo lasciato dalla citazione dello scolio. Non è affatto detto che ci sia stata conservata una unità semantica e che essa non sia stata infirmata da qualche maldestra epitomazione del materiale di commento al passo omerico: il frammento presenta una corruzione. Il dubbio resta, secondo me, anche sul fatto che la frase di Callimaco si riferisse precisamente ai Selli e a Dodona.15 L’uso del singolare (decisivamente confermato dal nostro papiro) potrebbe fare difficoltà in riferimento ai Σελλοί: ma trova un significativo parallelo in Call. fr. 23, 3 Pf., dove è usato Σελλός al singolare;16 è vero anche che, se le γυναῖϰες menzionate al r. 2, pur nel corso del commento e non in lemma (cfr. sopra), avevano qualcosa a che fare con il contesto callimacheo, si può invocare la ben nota presenza a Dodona di un clero di sesso femminile (le Peleiades).17 Tuttavia, se il passo si riferisce ai Selli, il suo significato dovrebbe (stando sempre allo scolio omerico) essere in relazione con il costume dei piedi non lavati: tale costume trova un parallelo unicamente in Eurip., Erecht. fr. 367 N2 (= fr. 58 Austin),18 citato da Eustazio nel commento allo stesso passo omerico.19 Le altre testimonianze sui sacerdoti dodonei fanno riferimento soltanto all’uso di dormire per terra: Callimaco stesso, Hymn. IV 286, li definisce γηλεχέες, ma non fa parola del non lavare i piedi. E non sono riuscito a trovare agganci che confortino l’opinione che il frammento appartenesse ad un contesto ‘dodoneo’. Resta tutt’altro che da scartare la possibilità che il passo fosse riportato solo come parallelo concettuale, nel qual caso poteva anche essere estratto da un contesto non religioso. Purtroppo, su questo groviglio di problemi di interpretazione letterale e di contestualizzazione, il nostro papiro non offre nuovi elementi: e resta una serie di

 15 Schneider 1873, 91, nelle sue ampie ricostruzioni, attribuiva il fr. al II libro degli Aitia, in un episodio ambientato a Dodona (cfr. fr. 53 Pf.). Pfeiffer lo colloca più prudentemente fra i Fragmenta incertae sedis, pur ritenendo fondamentalmente valido il legame con Dodona. 16 ... σὺ δ᾿ὡς ἁλὸς ἦχον ἀϰούει / Σ]ελλὸς ἐνὶ Τμαρίοις οὔρεσιν Ἰϰαρίης ... Le altre testimonianze sui Selli nella bibl. cit. alla n. 11. 17 Cfr. Hdt. II 52 sgg.; Soph., Trach. 171, etc.; Kroll 1937; Kern 1903, 1261 sg.; Parke 1967, 60 sgg.; v. anche Suida s.v. Δωδώνη· πόλις ... ἐν ᾗ μαντεῖον ἦν γυναιϰῶν προφητίδων. 18 Nova Fragmenta Euripidea in papyris reperta ed. C. Austin, Berlino 1968. Questo il frammento: ἐν ἀστρώτῳ πέδῳ / εὕδουσι, πηγαῖς δ᾿ οὐχ ὑγραίνουσιν πόδας. 19 Eust. 1058, 4 a proposito dei Selli ἀνιπτόποδες: ὃ ϰαὶ ἐν τῷ τοῦ Εὐριπίδου Ἐρεχϑεῖ δηλοῦται εἰπόντος ὅτι πηγαῖς οὐχ ὑγραίνουσιν πόδας (cfr. la cit. di Clem. Alex., Strom. VI 2, 7).

  Un nuovo frammento di commentario a Callimaco dubbi. In ogni caso, se si ritiene valida l’appartenenza del passo ad un qualsivoglia contesto religioso (magari non ‘dodoneo’!), ϰατέχεας del r. 3 potrebbe riferirsi ad un atto rituale. Ma non credo valga la pena, con tanto poco a disposizione, accumulare e inseguire ipotesi che non si riesce a corroborare. Al r. 7, se è stato letto correttamente οὐδοῦ, è pressoché certo che si abbia a che fare con una citazione, trattandosi di parola poetica che non può appartenere al linguaggio scoliastico: citazione di un passo parallelo nel corso del commento al lemma dei rr. 4-5? Viene da integrare οὐδοῦ αὐ[λείου, che si può confrontare con Odyss. I 104: στῆ . . . οὐδοῦ ἐπ᾿ αὐλείου; ma non sono riuscito a identificare la citazione del nostro testo. Un’altra possibilità è che non si tratti di citazione da altro autore, ma vengano riportate o parafrasate parole di Callimaco. Tutto ciò potrebbe suggerire di leggere come δ la lettera iniziale del r. 5 e di integrare οὐ]δοῦ nel lemma: ma tale integrazione contrasta con la metrica se si considera valida la ricostruzione del pentametro proposta sopra. Se οὐδοῦ è parola di Callimaco, o non va integrata al r. 5 oppure il pentametro non finiva con la desinenza — ου che compare prima dello spazio bianco a questo rigo. Per quanto riguarda le lettere che precedono οὐδοῦ al r. 7 (ενεϰ), non sono riuscito a trovare una soluzione soddisfacente.20 I rr. 9–13 presentano ampi spazi bianchi nella parte destra e nessun mezzo tecnico (raggi infrarossi e ultra violetti) ha permesso di scorgere eventuali tracce di scrittura svanita: se si tratta di reali spazi bianchi, l’unica possibilità di spiegazione che mi viene in mente, come pura ipotesi, è di pensare a una serie di glosse. Al r. 11, ]χεισϑαι va integrato in ϰατα]χεῖσϑαι ed ha qualche relazione con ϰατέχεας del r. 3? Con il r. 14, il rigo torna ad essere scritto per intero: un nuovo lemma? Per quanto riguarda la filologia callimachea nell’antichità, la tradizione non ci dà notizia di nessun altro hypomnema a Callimaco precedente a quello scritto da Teone, figlio di Artemidoro, vissuto in epoca augustea.21 Tuttavia alcune scoperte, che non sono ancora di dominio pubblico per gli studiosi, mi permettono ora — se pure al momento soltanto per un breve cenno — di modificare in modo piuttosto consistente il quadro degli inizi della fortuna critica di Callimaco nella filologia alessandrina. Mentre stavo attendendo a questo lavoro, infatti, ho avuto comunicazione — grazie al cortese spirito di collaborazione scientifica del Prof. Claude Meillier dell’università di Lille — che nella collezione di Lille si  20 La soluzione più semplice sembra dividere ]εν ἐϰ οὐδοῦ e intendere ἐϰ come preposizione reggente οὐδοῦ: ma perché non ἐξ? È solo un errore del copista? Non va scartato ]ἕνεϰ᾿ οὐδοῦ, pensando forse a un precedente genitivo retto da ἕνεϰ(α) e inghiottito dalla lacuna. 21 Pfeiffer 1953, II, XXVII.

Un nuovo frammento di commentario a Callimaco  

trovano due frammenti papiracei inediti di commentari a Callimaco, sicuramente databili ad epoca tolemaica (approssimativamente al II sec. a. C.). P. Lille 79 (identificato da P. Parsons) restituisce una sezione di commentario in cui si individuano come lemmi alcuni versi del fr. 176 Pf. (Aitia); P. Lille 82 (identificato da C. Meillier e già oggetto di una comunicazione al Congresso dell’APLAES tenutosi a Lille il 24 maggio 1975) conserva un brano di commentario relativo ai primi versi del fr. 383 Pf. (Elegia in Victoriam Nemeam?). I due frammenti sono scritti dalla stessa mano: si può dunque supporre che provengano da un’edizione commentata di Callimaco comprendente almeno gli Aitia e gli Epinicia. I papiri presentano ampi lemmi che riportano continuativamente il testo di Callimaco, a quanto pare con intenzioni esaustive: le sezioni di commento sono «incorporate» nel testo e inframmezzate ai versi, ma non costituiscono una parte preponderante nel corpo del volumen, nel quale un peso pressoché equivalente è dato al testo poetico in sé. Il termine hypomnema — come siamo abituati ad intenderlo, ponderosa ed autonoma opera di commento di grande ricchezza ed estensione — sembra poco adatto a questi testi, per i quali converrà parlare più genericamente di edizione commentata. Si è invitati a ripensare anche allo sviluppo dell’aspetto formale e librario dei lavori di commento: si può forse ritenere che, nella fase iniziale del lavorio critico su un autore, il materiale esegetico ed erudito avesse una mole abbastanza piccola per poter andare assieme al testo commentato e che il progressivo accrescersi di tale mole abbia richiesto che l’hypomnema si presentasse come un ponderoso libro autonomo e staccato dal testo nella materiale presentazione libraria. Sono grato al Prof. Meillier ed ai suoi collaboratori di avermi fornito le notizie qui riferite ed avermi concesso di approfittare dell’occasione per dare notizia in anteprima di questi importanti testimoni: a quanto mi viene comunicato, essi saranno pubblicati — assieme agli altri papiri della stessa collezione — nel Cahier de Recherches de l’Institut de Papyrologie et d’Egyptologie de Lille (C.R.I.P.E.L.) n° IV, che uscirà nel corso del 1976. Il Callimaco commentato di età tolemaica restituito dai P. Lille 79 e 82 offre una testimonianza di estrema importanza dal punto di vista storico-culturale: esso in primo luogo smentisce il dato ricavabile dalla tradizione e ne rivela la grave lacunosità, mostrando per testimonianza diretta che Teone non fu affatto il primo a commentare la poesia di Callimaco. Nei papiri di Lille si trova certo un precedente importante e una fonte del lavoro di Teone: ciò significa che, per un prodotto così maturo della critica filologica alle opere del poeta di Cirene, bisogna ora risalire molto più indietro, addirittura all’epoca di Aristarco. Le conseguenze di questi nuovi dati, però, travalicano i limiti della filologia callimachea e investono più in generale il problema dell’esegesi critico-filologica dei grandi poeti ellenistici.

  Un nuovo frammento di commentario a Callimaco Recentemente F. Lasserre22 ha individuato, in un papiro della seconda metà del II sec. a. C., un commentario (di Apollodoro?) ad un’elegia alessandrina (di Filita?). Diventa necessario rivedere globalmente il ruolo finora assegnato a Teone nella storia della filologia antica: più o meno coscientemente, si è sempre rischiato di considerare dati obiettivi quelli che erano soltanto argomenti ex silentio, vedendo in Teone il personaggio di svolta con il quale si cominciò a pubblicare veri e propri hypomnemata ai poeti ellenistici, mentre prima ci sarebbe stato per essi soltanto qualche più o meno sporadico interesse.23 Ciò che prima dovevamo nebulosamente divinare, ora possiamo toccarlo con mano: ben prima di Teone si cominciò a scrivere commenti alla poesia ellenistica; se Teone poi diede a quest’opera di esegesi un impulso forte e probabilmente decisivo per la critica posteriore, egli si inserì comunque certamente in un contesto e in un filone di interesse già piuttosto consolidato. Sempre in tema di filologia callimachea antica, dall’insieme delle testimonianze risulta che dalla fine del I sec. a. C. (cioè dall’epoca di Teone) al II sec. d. C. vi fu una vera fioritura di interesse e molti filologi si occuparono delle opere del poeta di Cirene.24 Fra i papiri, si rileva una particolare frequenza nel II e III sec. d. C. (certo anche dovuta, in vero, alla maggiore quantità di ritrovamenti papiracei relativi a quest’epoca) di testimoni della cospicua attività critica svoltasi intorno alla poesia callimachea: dal famoso papiro milanese delle Diegheseis (P. Med. 18), agli Scholia Florentina (PSI 1094 e 1219), al P. Berol. 11521, ai P. Oxy. 2262 e 2263. In questo panorama, ben si inserisce il nostro frammento, collocabile — come si è detto — nel II sec. d. C. Non avendo certo esaurito la ricerca sui problemi che pone, ritengo tuttavia più utile far conoscere il nuovo testo e consegnare il suo contributo, seppur minimo, a studiosi esperti di questioni callimachee e meglio in grado di conseguire ulteriori risultati.

 22 Lasserre 1975. 23 Su Teone cfr. Wilamowitz 1906, 187 sgg.; Susemihl 1892, II, 215 sgg; Wendel 1934; Pfeiffer 1953, XXVII; Guhl 1969 (per Callimaco pp. 6 sg. e 29 sg.); Fraser 1972, I 474 e nn. Cfr. anche Lasserre 1975, 176. 24 Pfeiffer 1953, II, xxvi sgg. e CII sgg.

 Callimaco e la filologia Il titolo del mio intervento è consapevolmente ambiguo: Callimaco e la filologia può fare riferimento alla filologia di Callimaco e dunque alla molto indagata problematica del poeta erudito e filologo, oppure può fare riferimento alla filologia su Callimaco e dunque agli studi callimachei nel mondo antico. Non c’è dubbio che le ricerche su Callimaco poeta-filologo siano assai più diffuse e differenziate rispetto a quelle su Callimaco oggetto di filologia, specie nei secoli stessi dell’età alessandrina, dunque per quanto riguarda i primordi e i primi passi della filologia callimachea. Qui parleremo di entrambi gli aspetti, la filologia di Callimaco1 e la filologia antica su Callimaco: ma sarà la seconda ad avere lo spazio maggiore, riprendendo e proseguendo la discussione su una questione sollevata di recente, vale a dire l’epoca e le modalità di inizio nell’antichità del lavoro filologico sui poeti del primo ellenismo e in particolare su Callimaco.

Callimaco filologo e l’ekdosis omerica di Zenodoto Pfeiffer pose con prudenza il problema della possibile conoscenza da parte di Callimaco del testo omerico di Zenodoto, sulla base di testimonianze costituite da alcuni casi in cui il testo omerico presupposto dal poeta nella sua versificazione si accorda in modo particolare e univoco con quello scelto da Zenodoto.2 Data anche la vicinanza cronologica (Zenodoto visse tra il 330 e il 260 circa), la questione è di assoluta rilevanza e si capisce bene la cautela mostrata da Pfeiffer. In seguito, tuttavia, credo che l’approfondita e documentata ricerca di A. Rengakos abbia raggiunto il risultato di provare (almeno con grande probabilità) che Callimaco e Apollonio Rodio abbiano effettivamente conosciuto il testo omerico di Zenodoto e lo abbiano talvolta anche presupposto e utilizzato nel loro fare poesia, riprendendolo allusivamente in alcuni luoghi.3 È senz’altro un

 1 Su questo aspetto resta fondamentale la sintesi di Pfeiffer 1973, 207–231 (= 1968, 123–140); cfr. Blum 1977. 2 Pfeiffer 1949, ad fr. 12,6 + addenda con i rimandi (cui aggiungi fr. 633 e fr. 497 + addenda), 1973, 229–230 (= 1968, 139–140); cfr. Erbse 1953. 3 Rengakos 1993, 169: “Beide Dichter, Kallimachos und Apollonios, haben Zenodots Text nicht nur gekannt, sondern auch in ihren Dichtungen benutzt”, cfr. Montanari 1995b, 53–54 (= questo volume, cap. 62); Rengakos 2001. https://doi.org/10.1515/9783110772371-058

  Callimaco e la filologia fatto importante, che indica come l’attività filologica di un grammatico4 abbia in qualche modo influenzato la prassi poetica di due eminenti poeti-grammatici. Il discorso coinvolge la tematica relativa alla forma in cui si presentavano le edizioni alessandrine, cioè che cosa era effettivamente l’ekdosis zenodotea, sia dal punto di vista materiale e librario, sia dal punto di vista dei contenuti filologici, implicando anche il problema del metodo e delle fonti delle lezioni di Zenodoto, basate o meno sull’utilizzazione e sulla collazione di diversi esemplari. Mi sono occupato di recente di tali questioni e, senza riprendere qui gli elementi e la bibliografìa di un dibattito ultimamente piuttosto ricco, riassumo i risultati ai quali ritengo di essere pervenuto.5 Come procedeva e cosa faceva esattamente un grammatico alessandrino che voleva produrre una ekdosis di Omero? Una possibilità è che il risultato del suo lavoro fosse proprio una nuova copia continua recante il suo testo, l’intero testo riscritto come egli riteneva che fosse giusto e corretto: un’idea molto (troppo) vicina al concetto moderno di edizione. L’altra è che egli lavorasse su una copia già esistente, opportunamente scelta fra quelle a sua disposizione e tenuta come testo-base per fare i suoi interventi, dunque senza riscrivere un testo continuo e intero, ma scrivendo solo il necessario nelle parti disponibili, vale a dire margini, intercolumni e interlinei. In questo caso, si può immaginare facilmente che i segni critici venissero posti accanto ai versi in questa stessa copia: ma resta oggetto di dubbio definire se essa fosse corredata anche di annotazioni, in quale misura e con quali contenuti, insomma come e dove fossero scritte e conservate le vere e proprie lezioni volute dal filologo autore di una specifica ekdosis. Questa seconda possibilità (cioè la copia annotata) sembra prevalere negli studi più recenti e anche a me pare la più verisimile. Zenodoto scelse una copia già esistente e la annotò via via con il procedere dei suoi studi e del suo lavoro di diorthosis. Su alcuni versi aveva dubbi di autenticità e accanto ad essi tracciò un obelos, il segno inventato per la proposta di atetesi. Più difficile è sempre stato capire come andavano le cose a proposito di quei versi che egli riteneva sicuramente spuri e dunque decisamente da eliminare dal testo: versi che  4 Un grammatico che, se davvero fu anche poeta, certamente non lo fu di grande rilievo: le sole indicazioni in questo senso vengono dall’articolo della Suda (Z 74: ἐποποιὸς ϰαὶ γραμματιϰός) e dall’esistenza di tre epigrammi a lui attribuiti, ma di dubbia paternità (cfr. Gow/Page 1965, II 557–559). 5 Montanari 1998a (= questo volume, cap. 23); 2000a (con la bibliografia citata [= questo volume, cap. 38]); cfr. anche Janko 1998; Nagy 1998; Matthaios 1999; Nagy 2000; Nardelli 2001a; Führer/Schmldt 2001; solo dopo gli Entretiens ho potuto vedere West 2001a. Mi propongo di tornare fra breve sull’argomento in altra sede.

Callimaco e la filologia  

comunque dovevano essere presenti nel suo testo-base. Mi pare che una buona indicazione su questo ci possa venire dal fatto che i papiri letterari ci testimoniano bene l’uso di diversi modi per cancellare qualcosa presente in un testo: sulle parole o le lettere da eliminare si poteva tirare un tratto orizzontale o obliquo, oppure esse potevano essere contrassegnate da punti o da linee al di sopra o al di sotto, oppure ancora essere racchiuse entro una sorta di parentesi tonde in coppia.6 Le cancellazioni di solito erano operate, quando era il caso, dal διορθωτής, che nella bottega di produzione libraria aveva il compito di rileggere e correggere, spesso confrontando la copia con il modello (di fatto, un confronto di esemplari). La più nobile pratica scrittoria della produzione libraria indicava quindi a Zenodoto modi di procedere a ‘cancellazioni’, che egli poteva benissimo trasferire alla sua particolare διόρθωσις, la διόρθωσις del filologo invece che quella del correttore del lavoro di uno scriba.7 Così la copia usata dal filologo come testo-base presentava, durante e dopo il suo lavoro, sia versi contrassegnati con obelos che versi decisamente ‘cancellati’ con qualcuno dei mezzi suddetti: accanto a questi ultimi, Zenodoto poteva benissimo annotare qualcosa come οὐ γράφειν o altro termine equipollente;8 oppure affidarsi al solo segno di cancellazione senza annotazioni ‘verbali’, nel qual caso la terminologia per l’eliminazione di versi (οὐ γράφειν, οὐϰ εἶναι, οὐ φέρεσθαι)9 può essere stata riportata da chi seguì il suo insegnamento al Museo o anche essere stata creata dalla tradizione successiva per descrivere i suoi interventi. A questa ricostruzione si possono fare alcune obiezioni. La prima è: cosa succedeva quando una proposta di atetesi o ancora di più una cancellazione comportavano di necessità cambiamenti nel testo restante perché esso fosse leggibile? Risposta: non vedo difficoltà a pensare che Zenodoto potesse scrivere il testo alternativo in margine, accanto al verso interessato, prima e/o dopo la parte atetizzata o cancellata.10 Un’altra è: come mai nella tradizione scoliastica

 6 Turner/Parsons 1987, 16, con rinvio a esempi nelle tavole; cfr. Turner 1980, 93 e PI. VIII (= 1984, 113 e tav. VIII). 7 Cfr. Ludwich 1884–1885, II 134; Nickau 1977, 10 sg. 8 I verbi περιγράφω e διαγράφω sono termini tecnici per ‘cancellare’ con i mezzi materiali di cui sopra: cfr. Turner/Parsons 1987, 16; alcuni esempi sono rimasti negli scoli: per περιγράφω cfr. Nickau 1977, 10–12 e 29. 9 Cfr. Nickau 1977, 1–30; 1972, 29 sgg. 10 Il caso dei vv. B 156–168 è emblematico: cfr. schol. ad B 156–169. Se la sua copia conteneva tutto questo passo e Zenodoto voleva eliminare i vv. 157–168, allora egli doveva per forza adottare per il v. 156 un testo diverso da quello che introduce 157 sgg.: accanto al v. 156 che trovava nella sua copia, cioè εἰ μὴ Ἀθηναίην Ἥρη πρὸς μῦθον ἔειπεν, poteva benissimo scrivere nel margine il resto da lui voluto per il v. 156 in seguito alla cancellazione di 157–168, cioè εἰ μὴ

  Callimaco e la filologia per le eliminazioni di Zenodoto troviamo anche l’uso di οὐϰ εἶναι, se i versi eliminati c’erano nella sua copia, cioè erano visibilmente cancellati ma presenti? La spiegazione più semplice è che questa sia una maniera semplificata (e imprecisa) di descrivere l’intervento, probabilmente utilizzata ben più tardi e senza una precisa cognizione dei fatti: la terminologia fornita dagli scoli per gli interventi zenodotei pone problemi con qualsiasi ricostruzione, il che non può stupire troppo.11 Una terza è: come può accadere che si trovino interventi testuali per versi eliminati e dunque cancellati? Risposta: e perché no? Caso mai, si deve pensare proprio al fatto che i versi eliminati erano nel testo-base e in qualche modo vi restavano anche se appunto ‘cancellati’: di conseguenza, potevano essere oggetto di attenzione o esserlo stati in un primo tempo (soprattutto pensando a un lavoro svolto su una copia pre-esistente, che recava i versi ‘indesiderabili’ per Zenodoto); potevano anche esserci ripensamenti, l’idea della eliminazione essere adottata in un secondo momento, dopo che si era pensato a un intervento esegetico su uno dei versi poi cancellati.12 Bisogna pensare a un lavoro protrattosi nel tempo, la stessa copia utilizzata durante tutto il periodo di attività del filologo per i suoi studi omerici. Mi sembra, insomma, che questa idea spieghi in modo plausibile quello che sappiamo della filologia di Zenodoto, compreso il problema della differenza materiale ‘libraria’ fra i versi atetizzati (obelos) e i versi eliminati (οὐ γράφειν), senza comportare al momento obiezioni forti. A proposito del metodo filologico degli alessandrini, in relazione al dibattuto problema se le loro lezioni fossero soltanto congetture basate su opinioni e gusti personali oppure avessero una fonte documentaria in esemplari omerici collazionati, ripeto qui brevemente che condivido del tutto l’idea che le due posizioni estreme (solo congetture autoschediastiche oppure solo fonti documentarie collazionate) siano entrambe da respingere, essendo ben più verisimile che ci fosse nel lavoro dei filologi alessandrini sia la critica congetturale che l’uso di materiale documentario proveniente dalla collazione di copie e dalla conseguente osservazione di varianti.13 Naturalmente, con le testimonianze in  Ἀθηναίη λαοσσόος ἦλθ’ ἀπ’ Ὀλύμπου. Cfr. Ludwich 1884–1885, II 134; Nickau 1972, 30; 1977, 6 sgg. 11 Cfr. sopra e n. 9; vd. anche Montanari 1988, partic. 83 sgg. e 95 sgg. 12 Ancora il passo B 156–168 (v. n. 10) ci offre un buon esempio: lo schol. B 161a ci informa di una variante zenodotea a B 161, anche se dallo schol. B 156–169 sappiamo che Zenodoto eliminava i vv. 157–168: entrambi gli scoli risalgono a Aristonico, che evidentemente non si stupiva. Su questo problema cfr. Nickau 1977, 6 sgg. 13 Sintesi recente in Nardelli 2001a, 51–70 e in Führer/Schmidt 2001, 3–8; diversa la posizione di West 2001a, 33–45, 54–56 (cfr. sopra n. 5).

Callimaco e la filologia  

nostro possesso sarà difficile distinguere di volta in volta se una lezione è congettura personale oppure ha una base documentaria, se non ricorrendo a ipotesi e deduzioni più o meno fondate e affidabili. Questo modo di lavorare nel suo complesso (che non intende e non comporta una anacronistica assimilazione del grammatico antico al filologo moderno) fu applicato da Zenodoto e la sua ekdosis omerica ne fu il primo risultato sistematico, anche materialmente visibile e consultabile nell’ambiente dei dotti alessandrini. Una novità significativa e una acquisizione intellettualmente e culturalmente importante: si può ben capire dunque che essa abbia influenzato poeti eruditi come Callimaco e Apollonio Rodio, che sicuramente ne furono a conoscenza. Pensando alle loro riflessioni di poeti-filologi, il fatto di avere a disposizione la copia omerica annotata da Zenodoto risulta un elemento per nulla secondario del quadro culturale e un supporto certamente utile.

Gli inizi della filologia su Callimaco (e su altri poeti ellenistici) Qualche anno fa ho cercato di far emergere gli elementi per cui, a mio avviso, si può ritenere che l’attività filologica dei grammatici alessandrini sui maggiori poeti del primo periodo ellenistico non sia iniziata nel I sec. a.C. con Asclepiade di Mirlea14 e Artemidoro (il padre di Teone), per realizzarsi soltanto con i commentari di Teone in età augustea, ma avesse mosso i primi passi qualche generazione prima e avesse conosciuto un considerevole sviluppo già ai tempi di Aristarco e dei suoi immediati discepoli. Sostenevo che, se in età augustea i poeti ellenistici erano evidentemente diventati un comune terreno di lavoro filologico, tuttavia la storia della loro esegesi doveva essere cominciata già da qualche tempo, ancora ben dentro l’età ellenistica. Non si dovette aspettare fino all’epoca di Teone perché l’esperienza sul passato, cioè sui prodotti della grande produzione letteraria prearistotelica, fosse trasferita pienamente anche sulla poesia del primo periodo alessandrino, iniziando il processo per cui i suoi autori sarebbero entrati nel novero dei πραττόμενοι, cioè ‘trattati’ dai grammatici.15 Una parte consistente di quel lavoro era dedicata ad analizzare alcune citazioni di Callimaco nella scoliografia omerica e il loro impiego come materiale di commento a Omero. Avanzavo l’ipotesi che l’uso abbondante di Callimaco nella filologia omerica (sia come testimonianza per la costituzione del testo, sia come parallelo e supporto per l’interpretazione), specialmente considerando le  14 Cfr. infra e n. 61. 15 Montanari 1995b (= questo volume, cap. 62).

  Callimaco e la filologia caratteristiche talvolta ‘impegnative’ per l’analisi del passo callimacheo chiamato in causa, costituisca un valido indizio di un lavoro su Callimaco stesso come autonomo oggetto di studio: che si tratti insomma di resti (per quanto ridotti) delle prime fasi della filologia callimachea (per limitata che fosse) e di un interesse attivo per la poesia ellenistica da parte dei grammatici alessandrini anteriori al I sec. a.C. (segnatamente Aristarco e il suo entourage). Grosso modo tra la fine del III e la prima metà del II sec. a.C. (cioè fra gli ultimi tempi di Aristofane di Bisanzio e il fiorire del lavoro di Aristarco) devono essersi via via affacciati l’idea e l’orientamento per cui ci si doveva occupare in qualche modo anche dei ‘moderni’, vale a dire delle più significative figure di poeti del periodo ellenistico. Che le testimonianze dirette anteriori all’età augustea non siano abbondanti, non è davvero difficile da spiegare: in generale, i frammenti papiracei di hypomnemata o di altri prodotti eruditi databili all’età tolemaica sono complessivamente assai scarsi rispetto a quelli di età imperiale; e peraltro di per sé la produzione filologico-grammaticale sui poeti ellenistici non sarà certo diventata subito copiosa. Ma riflettendo bene, non è comunque più verisimile che l’abbondanza dei commentari a poeti ellenistici scritti da Teone in età augustea non sia un punto di partenza bensì, in un certo senso, un punto di arrivo? Punto di arrivo, naturalmente, di una prima fase della filologia sui poeti ellenistici, che ha portato da interventi critico-esegetici relativamente limitati e desultori fino a includere i ‘moderni’, come dicevamo sopra, nel quadro degli autori trattati dai grammatici. All’epoca di Teone, la filologia mostra di essersi ormai definitivamente impadronita del segmento di storia letteraria rappresentato dalla poesia dell’età ellenistica: mi sembra difficile pensare che questo fatto rappresenti una fase ancora relativamente iniziale.16 L’argomento porta inoltre ad aggiungere una considerazione sui progressi del lavoro filologico alessandrino sulle opere in prosa degli storici e degli oratori. Per Demostene, abbiamo una scoliografia povera di fronte alla testimonianza di un assai considerevole hypomnema di Didimo conservato su papiro e frammenti di altri hypomnemata. La consistenza della scoliografìa questa volta ci trae in inganno, Demostene fu riccamente commentato almeno a partire dall’età augustea: sembra diffìcile che gli hypomnemata di Didimo sugli oratori17 fossero senza precedenti e abbiano rappresentato la prima prova significativa della filologia sulla grande oratoria. Per quanto riguarda gli storici, ricordiamo che la  16 A. Harder mi ha ricordato le acute osservazioni di Cameron 1995, 229–232, a proposito dell’esegesi callimachea; cfr. infra, nn. 44, 46 e 66. Per Apollonio Rodio cfr. infra. 17 Oltre a Demostene, Eschine, Iperide, Iseo; forse anche Isocrate e Dinarco: cfr. Schmidt 1854. Per Demostene: Lossau 1964; Pearson/Stephens 1983.

Callimaco e la filologia  

scoliografia erodotea ci dice pochissimo e se non avessimo avuto il caso fortunato del piccolo frammento PAmh. 12, del III sec. d.C., con la sottoscrizione “hypomnema di Aristarco al libro I di Erodoto”,18 non sapremmo che lo stesso Aristarco scrisse un commentario apposito sullo storico. Pfeiffer si esprime con cautela sul fatto che Aristarco abbia commentato anche Tucidide, ma è fiducioso sull’esistenza di studi tucididei prima di Didimo.19 Considerando dunque da una parte i prosatori (storici e oratori) e dall’altra i poeti ellenistici, appare chiaro come nei tre secoli fra Zenodoto e l’età augustea (con personaggi come Didimo, Aristonico, Teone e altri) l’attività filologico-esegetica sia arrivata a un raggio così vasto di interessi, da coprire praticamente tutti gli aspetti principali della letteratura greca arcaica, classica ed ellenistica. L’ampliamento dell’orizzonte fino a queste dimensioni deve essere maturato fra Aristofane di Bisanzio e Aristarco, per consolidarsi, nell’arco del II sec. a.C., con quest’ultimo e i suoi discepoli. L’idea di studiare i testi poetici dei passato, di conservare e interpretare con sofisticati strumenti critici e intellettuali i monumenti di una paideia venerata e codificata nella sua autorevolezza, era evidentemente diventata, nella consapevolezza dei suoi esponenti, un metodo abbastanza sicuro ed evoluto da essere applicabile e applicato a quanto era da considerarsi letteratura importante, arte della parola. Dalla novità dell’ekdosis di Zenodoto al maturo lavoro delle edizioni e degli scritti esegetici di Aristarco, dedicarsi alla filologia, all’interpretazione e allo studio erudito dei testi letterari era entrato nello spirito del tempo e nella formazione culturale degli intellettuali. Rovesciando il punto di vista, ci sarebbe forse da stupirsi che un filologo come Aristarco non provasse interesse e non si occupasse di poeti come Callimaco o Apollonio Rodio (anche indipendentemente dalle preferenze del suo gusto).20 Alcuni indizi assai problematici vanno presi in considerazione ancora. Una notizia dell’Etymologicum Genuinum ci parla di un personaggio di nome “Edilo” che avrebbe scritto un’opera sugli Epigrammi di Callimaco: è una pura ipotesi che questo Edilo possa essere Edilo di Samo, poeta di epigrammi addirittura contemporaneo di Callimaco, e non un altrimenti ignoto grammatico di età  18 Cfr. Paap 1948, 37–40; Pfeiffer 1973, 348–350 (= 1968, 224–225). Il testo del papiro è riprodotto in apparato nell’edizione teubneriana di Rosen 1987–1997, I 135–136. 19 Pfeiffer 1973, 348–350, 419–420 (= 1968, 224–225, 277–278); Montanari 1993, 247–249, 272 (= questo volume, cap. 1); Maehler 1994, 121–124. La definizione di Dionisio Trace della grammatikè come empeiria delle opere di poeti e prosatori (Techne 1, definizione sicuramente autentica: cfr. da ultimo Schenkeveld 1994) significa che con la generazione immediatamente dopo Aristarco era normale parlare di lavoro filologico-grammaticale anche sulle opere in prosa. 20 Richard Hunter ha sottolineato il problema posto dal passo di Quint. 10, 1, 54: cfr. la discussione.

  Callimaco e la filologia posteriore, per cui possiamo solo sperare in nuovi dati che possano dirimere il dubbio.21 Una parte della voce riguardante la differenza di significato fra ὀλίγος e μιϰρός nei lessici sinonimici di Ammonio e di Erennio Filone22 presenta un irrisolto problema di costituzione e comprensione del testo, che ci può riguardare:23 Amm. 352 Nickau: Ἀριστόνιϰος ἐν ὑπομνήματι †ἐϰάλεσ᾽ ἐπὶ στοιχείου† ‘ὀλίγην νησῖδα Καλυψοῦς’ (Call. fr. 470 b Pf). Herenn. Phil. 132 p. 202 Palmieri: οὕτω Νιϰίας (loc. ign.) ἐν Ὑπομνήματι Ἑϰάλης ἐπὶ στί[χ]ου ῾–⏑⏑– ὀλίγην τε παρὲϰ νησῖδα ⏑⎼⏓᾽ (Call. fr.47O b Pf).

In primo luogo, i due lessici si differenziano per il nome che introduce la citazione, Ἀριστόνιϰος oppure οὕτω Νιϰίας, mentre concordano per ἐν ὑπομνήματι. Quanto segue, rimane piuttosto misterioso. In Ammonio, Nickau lascia fra cruces quello che danno (concordemente) i manoscritti prima del frustulo poetico. In Erennio Filone, Palmieri accetta una congettura di Valckenaer, che per primo attribuì il frammento a Callimaco e per di più introdusse la menzione dell’Ecale come opera da cui sarebbe tratta la citazione: la prima cosa è stata accettata dagli studiosi, la seconda invece respinta a favore di una possibile collocazione negli Aitia, e il testo accolto da Palmieri risulta piuttosto imprudente. Poiché la locuzione ἐν ὑπομνήματι è tramandata concordemente, non c’è motivo di dubitare che qui si parli di un commentario: ma rimane assolutamente in dubbio se esso fosse di Aristonico o di Nicia e se riguardasse Omero oppure Callimaco.24 Un commentario callimacheo di Aristonico non è altrimenti attestato; Nicia dovrebbe essere il grammatico25 citato più volte negli scoli omerici e vissuto probabilmente nel I sec. a.C.26 Per entrambi, l’esegesi omerica sarebbe la cosa  21 Etym.Gen. α 551 = Etym.M. α 960 (72,12 Gaisford) ἀλυτάρχης (vol. I 342–343 Lasserre/ Livadaras) = Call. T 45, ap. Pfeiffer 1949–1953, II p. cii: Ἡδύλος δὲ εἰς τὰ Ἐπιγράμματα Καλλιμάχου; cfr. Parsons 1977, 5. 22 Nickau 1966; Palmieri 1988. 23 Me lo ha ricordato Luigi Lehnus durante gli Entretiens. 24 Cfr. Pfeiffer 1949–1953, ad fr. 470 (sulla scia di Schneider, ipotizza che il riferimento fosse al commentario omerico di Aristonico, in cui sarebbe stato citato il verso di Callimaco); esaustive informazioni in Massimilla 1996, fr. 120, p. 160 e comm. a pp. 451–452; Benedetto 1993, 72–76. 25 Lehnus avanza la possibilità che si tratti di Nicia di Mileto, amico di Teocrito e poeta a sua volta (Geffcken 1936, 335–336; Albiani 2000), il che ci porterebbe nettamente più indietro, alla stessa epoca di Edilo. 26 Wendel 1936, 337; Damschen 2000, con bibliografìa; cfr. Massimilla 1996, 451. Temo che il Nicia citato in schol. ad Hom. Od. 23, 218 non sia questo grammatico, anche perché sembra proprio che lo scolio recante la sottoscrizione ή ἱστορία παρά Νιϰίᾳ τῷ πρώτω non c’entri nulla

Callimaco e la filologia  

più ovvia, stando alla documentazione che abbiamo: certo un riferimento a Callimaco è possibile, in relazione all’uso di ὀλίγος nel senso di μιϰρός nel verso citato. Nel caso di Aristonico, avremmo un buon parallelo cronologico per il lavoro di Teone, che confermerebbe lo spazio già ben acquisito dalla filologia callimachea; nel caso di Nicia, potremmo forse collocarci un poco prima: ma anche la cronologia di questo grammatico è dubbia. Tutto è estremamente incerto e anche questa notizia, allo stato attuale, mi sembra difficilmente utilizzabile. Infine, negli scoli a Nicandro si trova citato alcune volte un grammatico di nome Demetrio Cloro, autore di un hypomnema nicandreo: l’opinione comune è che egli sia anteriore a Didimo di almeno una generazione, ma la datazione è ipotetica e quindi non possiamo fare molto conto neppure su questo.27 La mia proposta di retrodatazione degli inizi del lavoro erudito degli alessandrini sui grandi poeti del primo ellenismo è stata discussa in un articolo recente da A. Rengakos, con grande competenza e acribía, facendo un passo avanti nella discussione di questa tematica.28 Mi pare che Rengakos condivida senz’altro (e anzi corrobori in modo importante, come vedremo subito) il risultato essenziale dell’indagine a proposito dell’interesse filologico alessandrino, in età aristarchea, per i poeti ellenistici, in primo luogo Callimaco. In conclusione tuttavia esprime un orientamento un po’ diverso nella valutazione storicoculturale del fenomeno: e su questo mi pare utile tornare e riflettere ancora e aggiungere altre considerazioni, in un dibattito il cui progresso deve molto al suo intervento. Al sondaggio da me fatto analizzando soltanto alcuni esempi callimachei significativi, Rengakos aggiunge una serie di altri casi in cui nella scoliografia omerica si trovano riferimenti di fonte aristarchea a Callimaco, Antimaco29 e Euforione (i due poeti più citati dopo Callimaco, di gran lunga il più presente), distinguendo le occasioni in cui il passo di un poeta ellenistico è addotto in relazione alla costituzione del testo omerico, da quelle in cui esso è utilizzato nel contesto di una discussione sul significato di una espressione omerica.30  con il problema dell’autenticità dei versi 218–224: su questo NP dipende direttamente da RE, che riprende un’attribuzione di J. Tolkiehn; cfr. FGrHist 60. 27 Cfr. Susemihl 1892, II 20; Kroll 1936, 262; Id., “Demetrios”, nr. 100 a, in RE Suppl.-Bd. VII (1940), 124; Montanari 1997b (con bibliografìa). 28 Rengakos 2000. 29 “For the purposes of this paper I will consider him a Hellenistic poet”: Rengakos 2000, 326; Schironi 1999. 30 Rengakos 2000, 328, ricorda anche Od. 2, 136–137, dove l’atetesi aristarchea del v. 137 lascia nel v. 136 una frase nominale per la quale viene citato il parallelo di Call. fr. 637 (Montanari 1979b = 1995a). Bisogna aggiungere il brillante risultato di Pontani 1999, che nello schol. ad Od.

  Callimaco e la filologia Qualche caso interessante egli trova anche in Aristofane di Bisanzio e nell’aristarcheo Apollodoro.31 Il risultato concreto è un incremento considerevole del gruzzolo evidenziato di osservazioni di filologi alessandrini sui poeti ellenistici. Ma cosa significa questo fenomeno? Rengakos dichiara: “My main objective is to raise again the question whether Aristarchus did take an interest in the Hellenistic poets per se or whether he used their work merely as a tool in his interpretation of Homer” (p. 326). E più avanti: “The fact that Aristarchus’ criticism in the field of language but also of Realien adheres exclusively to Homer combined with his habit to regard Hellenistic poetry quasi as a part of the indirect Homeric tradition and use it for the constitution of the Homeric text shows that he did not grant the right to these poets ‘to be as unhomeric as possible while being in Homer’s tracks’32... this means that he had apparently no sense of the individuality of Hellenistic poetry” (p. 331). Infine la conclusione: “Long before the heyday of Hellenistic studies in the 1st cent. B.C.... Hellenistic poets, from Lycophron, Callimachus and Apollonius to Euphorion and Nicander, were studied by professional Alexandrian scholars, esp. by Aristarchus, and used in their learned work on Homer. The Hellenistic poetae dotti are regarded as a sub-group of the infamous neoteroi, who in the interpretation of Homer served only as means to distinguish what is Homeric from what is post-Homeric. The works of these poets were of particular value to the Alexandrian scholars who recognized the affinity of these works to the Homeric epics and thus used them as evidence for the constitution of the Homeric text. But this one-sided attitude of the Alexandrian scholars, as is apparent in particular from Aristarchus and Apollodorus, means that they did not appreciate the special character of the Hellenistic poets. This may also have been the reason why these poets were not recognized in the 2nd cent. B.C. as an object of scholarship in their own right” (p. 334). Rengakos esprime con chiarezza e precisione la sua visione, che in sostanza è la seguente: i grammatici alessandrini, specialmente Aristarco, si interessarono effettivamente dei poeti ellenistici e li studiarono, ma soltanto in funzione strumentale per la filologia omerica e senza riconoscere loro un proprio autonomo valore come oggetto di studio erudito. Questa idea non mi pare del tutto convincente e vedo alcune debolezze negli argomenti sui quali appoggia, che cercherò di mettere in luce.  2, 50 ha acutamente reperito il primo verso degli Aitia usato come parallelo stilistico-sintattico per Omero. 31 Rengakos 2000, 333–334. 32 Riprende qui una frase di H. Herter; cfr. sotto per Apollonio Rodio.

Callimaco e la filologia  

Non c'è dubbio che il punto di partenza fosse normalmente Omero e che molta parte del lavoro filologico nascesse da oppure arrivasse a un problema di costituzione del testo o di interpretazione omerica: Omero era la base irrinunciabile dell’educazione e il banco di prova comune del lavoro dei grammatici. Ma ciò non significa certo che tutto si facesse in funzione soltanto di Omero e che ogni altro autore servisse e fosse oggetto di attenzione solo in quanto strumento e testimonianza per distinguere quanto è omerico da quanto è postomerico. Questa nozione storico-letteraria (riguardante lingua, stile, contenuti, Realien) era sicuramente l’obiettivo principale della scholarship e un cardine dei suoi ragionamenti, ma il suo risultato non doveva certo essere che solo τὸ Ὁμηριϰόν fosse interessante come oggetto autonomo di studio e che tutto il resto avesse un valore puramente ancillare. Certamente i poetae docti ellenistici erano considerati come un gruppo di neoteroi, ma sembra diffìcile sostenere che questo implicasse assolutamente che fossero indegni di interesse per se stessi. La qualifica di νεώτερος è utilizzata per gran parte dei (forse praticamente per tutti i) maggiori poeti posteriori a Omero, a cominciare da Esiodo, per proseguire con Archiloco, Pindaro, Sofocle e altri, fino agli ellenistici, tutti spesso utilizzati per confronto con passi omerici.33 Dovremmo applicare lo stesso ragionamento a tutti questi autori e vederli come oggetto di studio gregario, finalizzato solo a Omero? Mi pare invece che, se Omero doveva essere comunque il punto di partenza e/o di arrivo dell’argomentare filologico, non si possa affatto dedurne l’assenza di un interesse reale e autonomo anche per il poeta chiamato a confronto o utilizzato come parallelo, diverso e distinto in sequenza storica.34 Molti studiosi arrivano a occuparsi di un autore essendo partiti da un altro (magari sempre dallo stesso), ma qui entriamo in distinzioni sottili, che riguardano i percorsi intellettuali della ricerca e la ricostruzione della vita scientifica di ognuno. È possibile, anzi probabile, che Aristarco sia arrivato a Callimaco solo secondariamente rispetto ai suoi studi su Omero: ma certo studiò anche Callimaco, e mi pare difficile dire se lo fece con maggiore o minore passione e convinzione rispetto a uno qualunque dei poeti postomerici e prearistotelici, cui si dedicò con risultati ben noti. L'altro aspetto da chiamare in causa è quello delle fonti delle nostre informazioni. Le fonti utilizzate sono sostanzialmente e con grande prevalenza gli scoli omerici, il che significa comunque una fonte orientata in partenza su Omero.  33 Cfr. Severyns 1928, 31–61. Per fare qualche esempio significativo: per Pindaro cfr. POxy. 1086 ad Il. 2, 783 (Erbse 1969–1988, I 168): per Sofocle schol. A ad Il. 23, 679a (Erbse 1969–1988, V, 471). 34 Fatte salve naturalmente preferenze e predilezioni: fra i poeti ellenistici, l’apprezzamento per Callimaco si profila chiaramente.

  Callimaco e la filologia Inoltre, l’epitomazione e selezione del materiale esegetico pervenuto attraverso la scoliografia deve avere accentuato questo aspetto, sopprimendo o riducendo di preferenza, nei commenti omerici, i materiali meno strettamente omerici, quali discussioni più o meno ampie e approfondite su passi paralleli presi da altri autori o excursus eruditi di vario contenuto.35 Se altri corpora scoliografici non offrono attualmente materiale altrettanto interessante a proposito dell’uso dei poeti ellenistici da parte di Aristarco,36 questo può indicare diverse cose: 1) l’esegesi omerica era effettivamente la più ricca nell’antichità ed è anche la più riccamente conservata, dunque è naturale che dia più materiali anche per questo aspetto (come per molti altri); 2) gli altri poeti offrivano sicuramente meno occasioni di confronto e parallelo; 3) della filologia antica abbiamo perso moltissimo, per cui è molto facile che pochi indizi rivelino un fenomeno assai più cospicuo. Credo che valga la pena cercare altri dati e una piccola cosa faremo anche in questa occasione. L’ultima considerazione riguarda la produzione di opere esegetiche specifiche sui poeti del primo ellenismo, che siano state sicuramente scritte prima della fine del II sec. a.C. Questa è già una limitazione molto forte, come già osservato sopra, perché sono veramente pochi i frammenti conservati di opere erudite su qualunque autore (incluso Omero) che siano anteriori a questa data. Tuttavia due esempi ci sono e mi pare giusto ribadire l’importanza di questi due reperti, la cui esistenza è un pilastro fondamentale per il ragionamento che stiamo facendo.37 Ripeto qui solo le notizie essenziali. Nel 1975 F. Lasserre pubblicò l’editio princeps del PLouvre inv. 7733 verso: due anni dopo il testo fu riconsiderato a fondo da P. Parsons e poi ripresentato nel Suppl.Hell.38 Il frammento, datato al II sec. a.C., presenta sul recto un trattato di ottica39 e sul verso un componimento di sei versi (un enigma, la cui risposta è: “un’ostrica”), seguito da un commento continuo di oltre cinquanta righe (forse più o meno completo). La poesia è senz’altro ellenistica, ma il nome dell’autore

 35 Cfr. sotto l’esempio dagli scoli ad Arato. 36 Rengakos 2000, 326. 37 Da questi prendevo le mosse in Montanari 1995b, 49–52 (= questo volume, cap. 62): mi pare che Rengakos 2000, 325, non dia loro il giusto rilievo: “Apart from some new papyrus finds documenting more or less learned commentaries on Hellenistic works which had appeared only a few decades ago we have no reports that the leading Alexandrian scholars worked on Hellenistic poetry”: ma anche per Erodoto, ad esempio, non abbiamo nessuna informazione al di fuori del piccolo frustulo di papiro già ricordato (cfr. sopra e n. 18). 38 Suppl.Hell. 983–984, pp. 497–500; Lasserre 1975; Marcovich 1976; Parsons 1977a; Del Fabbro 1979, 72–75; D’Alessio 1990; Sbardella 2000, 179–184. 39 Pack2 2579; il testo del verso era segnalato in Pack2 2911 come prosa non identificata.

Callimaco e la filologia  

rimane ignoto.40 Il commentatore cita il poeta Teodorida di Siracusa, attivo nella seconda metà del III sec. a.C., e il testo è stato copiato da una mano del II sec. a.C.: dunque l’autore deve aver scritto nel tardo III o comunque entro la metà del II sec. a.C.,41 dunque entro la morte di Aristarco. Del componimento si ricostruiscono praticamente cinque versi su sei; del commento è probabile che non manchino molte righe: quelle conservate sono gravemente lacunose, ma quanto rimane è sufficiente per notare alcune caratteristiche. Data la brevità del componimento, è stato agevole per lo scriba riportare prima per intero l’opera commentata, con il suo titolo “ὄστρειον”, che viene ripetuto prima dell’inizio del commento. Benché il testo poetico sia comodamente a disposizione del lettore, il commento non rinuncia ad avere estesi lemmi (non si può dire con certezza se riportasse proprio tutto). Si affronta dapprima il problema dell’identificazione del luogo del sepolcro di Memnone e il discorso sembra essere: non Abido nella Tebaide egizia, dove non ci sono ostriche, bensì Abido nell’Ellesponto, che notoriamente ne produce. La seconda sezione (rr. 25–29) spiega ἀγροτέρη come epiteto di Artemide e dunque qui equivalente a ‘luna’; poi probabilmente collega a questo l’idea che le ostriche aumentino di dimensioni quando la luna cresce. Segue il commento al v. 6, concentrato sul significato dell’aggettivo ἀφέψαλος sulla base di quello del suo componente φέψαλος (rr. 30–39): qui si cita una commedia altrimenti ignota di Difilo, il Παραλυόμενος, di cui erano riportati probabilmente tre trimetri (fr. 59, in PCG V 86), e un frammento di Sofocle (pure altrimenti ignoto, fr. 966a Radt). L’ultima parte del commento (rr. 41 sgg.) riguardava la frase finale del componimento, dalla dieresi bucolica del v. 5 alla fine: il v. 6 compare come lemma alla r. 40. La prima questione era certamente il ricercato modo di indicare il coltello con cui si apre l’ostrica: l’amante di ‘Doso’, raro epiteto di Afrodite, è appunto Ares, cioè il ferro, cioè il coltello. Abbiamo qui la già ricordata citazione di Teodorida (fr. 743 Suppl.Hell.), con ogni probabilità un parallelo per l’epiteto ‘Doso’. Seguono parti troppo lacunose. Relativamente alle dimensioni del testo poetico, il commento appare di considerevole ampiezza: la varietà dei problemi e dei contenuti esegetici e l’utilizzazione di passi paralleli rivela un’opera di ragguardevole erudizione, scritta su una poesia certamente di non molti decenni prima. Nel 1977 C. Meillier diede l’editio princeps di alcuni frammenti papiracei conservati a Lille e contenenti versi della parte iniziale del III libro degli Aitia di Callimaco, inframmezzati da parti di commento: il testo fu studiato e riedito da

 40 Si è fatto il nome di Filita (cfr. bibliogr. alla n. 38), ma senza un reale fondamento. 41 Parsons 1977a, 12.

  Callimaco e la filologia P. Parsons e ripresentato infine nel Suppl.Hell.42 La datazione dei P. Lille ha oscillato fra la fine del III sec. a.C. e i primi decenni del II.43 Il testo callimacheo è riportato per intero e ad intervalli irregolari si alterna con righe di commento rientrate di tre lettere. Nelle piccole e lacunose parti conservate, il commento per lo più consiste in glosse o parafrasi, ma non soltanto: occasionalmente esso offre anche qualche notizia di carattere storico oppure linguistico. A ragione A. Cameron ha osservato: “... the most significant feature of the Lille scholia is a detailed prosopographical note, an accurate explanation of the dynastic fiction whereby Berenice II, daughter of Magas and Apama, was officially styled the daughter of Philadelphus and Arsinoe”:44 il riferimento è alla nota che segue il v. 2 del componimento (Suppl. Hell. 255) e può forse spingerci a vedere nei problemi prosopografici e dinastici dei regnanti un plausibile tema di interesse capace di stimolare l’esegesi callimachea più antica. Possiamo aggiungere che anche al v. 9 l’estensione del commento sembra coprire diverse righe. Quanto rimane purtroppo non è molto e forse non fa molta impressione nello stato attuale: ma c’è da chiedersi che effetto facesse nella sua interezza e quanto contenesse degli Aitia una simile edizione. Dal poco superstite, non sembra un prodotto particolarmente erudito, piuttosto un lavoro volto ad aiutare la comprensione immediata, con notizie di base utilizzabili per così dire inter legendum, intercalate a un testo che si può leggere nella sua interezza. L’epoca è più o meno la stessa del commento all’Ostrica, ma qui abbiamo un testo esegetico — a quanto pare — con minori ambizioni: in questo caso, in compenso, è lecito dire con certezza che questa “edizione commentata”45 di Callimaco fu ‘pubblicata’ in Egitto da un minimo di una a un massimo di due generazioni dopo la morte del poeta e ben prima della morte di Aristarco. Possiamo permetterci di liquidare con sufficienza questi due testimoni sicuri di interpretazione specifica di poesia ellenistica per lo meno contemporanei di

 42 Meillier 1976; Parsons 1977; Suppl.Hell. 254–265 (nuova edizione e ricostruzione, con altra bibliografìa). 43 Cfr. Meillier 1976; Parsons 1977, 4; Cavallo 1983, 53; Turner/Parsons 1987, nrr. 74 e 75, pp. 124–127. 44 Cameron 1995, 229 (cfr. nn. 16, 46, 66), e prosegue: “An abbreviated version is to be found in the commentary to F 110.45 in the sixth- or seventh-century P. Oxy. 2258 (again in the hand of the text scribe). So this particular note in a late antique manuscript goes back 700 years to a text of the Aetia written barely a generation from Callimachus’s death. For another prosopographical note, we have already considered the first-century London scholiast’s identification of the ‘tenth Muse’ as Arsinoe (Ch. VI. 1), which, if correct, must also be early”. 45 Così la definivo in Montanari 1976b, 147 (= questo volume, cap. 57); cfr. Del Fabbro 1979, 70–71.

Callimaco e la filologia  

Aristarco, ritenendoli due casi sporadici e isolati, opere scarsamente significativi? Oppure dobbiamo pensare che la loro esistenza deve influenzare il modo con cui guardiamo gli indizi di lavoro sui poeti ellenistici già in età aristarchea?46

Un’interpretazione aristarchea di Arato Arato rappresenta un caso relativamente particolare, perché le vicende della filologia sui Fenomeni dipendono in modo importante da problematiche di storia dell’astronomia, che possono avere percorsi anche distinti rispetto all’esegesi della poesia in senso stretto.47 È caratteristico il fatto che il testo di Arato sia stato oggetto di cure esegetiche sia per gli aspetti propriamente scientifici che per quelli poetici e filologici: “The commentators on A(ratus) are of two kinds, grammarians and astronomers (μαθηματιϰοί), and sometimes both are involved in comment on the same topic, e.g. in sch. 23, were the former are dismissed as being ignorant of astronomy”.48 Sch. ad Arat. Phaen. 23 (p. 68, 15–69, 6 Martin) περὶ δ’οὐρανός 〈αὐτὸν ἀγινεῖ〉·49 πολλὴ ϰαὶ διάφορος ἐνταῦθα ζήτησις περὶ τὴν γραφὴν ἐγένετο τοῖς μαθηματιϰοῖς ϰαὶ γραμματιϰοῖς. οἱ μὲν γὰρ γραμματιϰοὶ ἀγνοήσαντες εἶπον· “περιάγει ὁ οὐρανὸς τὸν ἄξονα”, ἔστι δὲ τοῦτο τῶν ἀτοπωτάτων. εἰ γὰρ ἀϰίνητον αὐτὸν ἀπεδώϰαμεν, [ϰαὶ] αὐτοῦ τοῦ Ἀράτου ἄντιϰρυς εἰπόντος (21–22) “ἀλλὰ μάλ’ αὔτως / ἄξων αἰὲν ἄρηρεν”, πῶς αὐτόν φασι περιάγεσθαι; ἀλλ’ οἱ μὲν μαθηματιϰοὶ τὸ αὐτὸν δασύνουσιν, ἵν’ ᾖ ἑαυτόν, ὁ δὲ λόγος· “περὶ δὲ τὸν ἄξονα ἄγει ϰαὶ στρέφει ὁ οὐρανὸς ἑαυτόν”. δύναται δὲ ϰαὶ οὕτως ψιλουμένου τοῦ αὐτόν, ἵν’ ᾖ ὡς πρὸς τὰ προειρημένα· “περὶ δὲ τὸν ἄξονα αὐτὸν πάντα τὰ ἄστρα ὁ οὐρανὸς ἀγινεῖ (ὡς τὸ προειρημένον) ϰαὶ περιφέρει”.

 46 Cameron 1995, 229–230: “Even if there was no full-scale commentary before Theon, we now know that annotated texts existed from a very early date. The new Lille papyrus of Aetia III, written in the late third or early second century B.C., is already equipped with notes (interspered among the lines of the text) in the hand of the text scribe. Most of them go no further than gloss and paraphrase, but that is immaterial. It is the existence rather than the nature of the commentary that matters. Many Byzantine scholia are no more substantial. We might in any case expect the notes on a contemporary text to be rather different from those on a classic”. Cameron ritiene che materiali degli Scholia Florentina risalgano a commenti più o meno contemporanei del poeta, che deve avere subito stimolato il bisogno di spiegazioni per il suo carattere estremamente allusivo. Cfr. nn. 16, 44, 46 e 66. 47 Non ho nessuna competenza sull’astronomia antica; utilizzo: Martin 1956; 1998; Kidd 1997. 48 Kidd 1997, 44. 49 Ho integrato il lemma.

  Callimaco e la filologia La discussione verte sul significato dell’emistichio περὶ δ’οὐρανός αὐτὸν ἀγινεῖ. Per i γραμματιϰοί la frase vorrebbe dire che il cielo porta in giro e fa muovere l’asse, quindi αὐτόν = l’asse. Ma questo è impossibile, perché l’asse è immobile, come del resto ha detto Arato stesso pochi versi prima. I μαθηματιϰοί intendono allora αὑτόν aspirato col valore di ἑαυτόν, interpretando che il cielo muove se stesso (ἑαυτόν) = si muove e ruota intorno all'asse. Infine, viene aggiunta un’altra possibilità, di natura puramente esegetica: lasciando αὐτόν con lo spirito dolce e pensando a un oggetto sottinteso, il significato potrebbe essere che il cielo fa muovere, scil. tutte le stelle, intorno all’asse (αὐτόν), rispettando dunque la realtà astronomica e la coerenza interna del poeta. La prima interpretazione è attribuita ai γραμματιϰοὶ ἀγνοήσαντες e viene respinta dai μαθηματιϰοί sulla base sia evidentemente di una precisa nozione scientifica, sia di un confronto interno con Arato, che non può contraddirsi in pochi versi (un argomento di coerenza interna del poeta, che potrebbe andare benissimo anche per i γραμματιϰοί). Ad essa si contrappone non tanto un intervento testuale, quanto una lettura leggermente diversa (αὑτόν invece di αὐτόν) ma decisiva per il senso della frase: un ragionamento insomma di tipo esattamente ‘grammaticale’, che viene addotto dai μαθηματιϰοί. La terza possibilità è introdotta da δύναται δὲ ϰαί e sembra una replica (senza indicazione di paternità neppure generica) all’intervento testuale dei μαθηματιϰοί, sostenendo che il significato da loro voluto è ottenibile anche soltanto su basi esclusivamente ermeneutiche, con una corretta interpretazione del passo. Per completare il quadro, aggiungiamo che esiste anche una lezione οὐρανόν, con la quale il soggetto sottinteso sarebbe l’asse: lezione che è presa in considerazione, parafrasata (περιάγει, φησίν, ὁ ἄξων τὸν οὐρανόν) e criticata nell’altro scolio al verso.50 Appare dunque una contrapposizione dichiarata fra γραμματιϰοί e μαθηματιϰοί per quanto riguarda la conoscenza dell’astronomia e il senso del passo di Arato, ma bisogna dire che sul piano del metodo non si vede una vera divergenza: si ricava piuttosto l’impressione che gli ‘scienziati’ applichino una filologia dotata di controllo serrato per i contenuti scientifici del testo, ma che non siano certo lontani da principi ermeneutici fondati sulla lingua, la grammatica, la correttezza delle lezioni testuali e la coerenza interna del poeta. Il nucleo più antico del corpus degli scoli aratei pervenuti risale, a quanto pare, a un commentario di Teone, anche se, per vero dire, non sono fugati i dubbi che si tratti davvero del Teone di età augustea.51 Molti elementi si aggiun 50 P. 69, 7–10 Martin, non riportato qui. 51 Cfr. Martin 1956; 1998, cxxvi–cxxx; Kidd 1997, 43–48; Pfeiffer, Storia 1973, 203–204 n. 90 (= 1968, 121 n. 4); su questo problema di identificazione, cfr. da ultimo Schiano 2002, partic. 136–137.

Callimaco e la filologia  

sero a questi scoli nel corso della tradizione, ma cosa c’era stato prima di Teone, o comunque prima dell’età augustea? È imprudente fare riferimento a un’opera di Eratostene, o circolante sotto il suo nome anche se non sua: non ci sono indizi che egli abbia fatto un commentario, ma forse potrebbe avere scritto un trattato di astronomia poetico-letteraria per accompagnare e aiutare la lettura di Arato, con nozioni elementari di cosmografìa e notizie sui miti relativi a trasformazioni in corpi celesti. Scrive D. Kidd: “By the second century BC the Phaenomena was already being read as if were an astronomical textbook, and the commentaries of Attalus and Hipparchus, especially the latter’s, were written to explain and correct the poem from this point of view”.52 Anche gli interessi di orientamento stoico sono ben documentati: alcune annotazioni aratee di Cratete di Mallo possono provenire dalle sue opere omeriche, mentre Zenodoto di Mallo53 è possibile che abbia scritto un commentario su Arato e quello che sappiamo della sua opera fa pensare a contenuti filologici;54 e poi Boeto di Sidone, che scrisse un commentario in almeno quattro libri.55 Negli scoli aratei il nome di Aristarco ricorre due volte e D. Kidd osserva che i suoi commenti furono probabilmente presi dalle opere su Omero.56 Questo può essere vero, ma i due casi sono abbastanza diversi fra loro da meritare una breve analisi e una riflessione. Nello schol. ai vv. 254–55 (p. 204, 2–3 Martin), nel contesto di una discussione sul significato di ἐπιγουνίς del v. 254, Aristarco è citato a proposito del significato del termine in Omero (τὸ ἄνω τοῦ γόνατος) e lo scolio adduce poi l’opinione del tutto differente di Cratete e quella del grammatico Chares. La parola ricorre in Od. 17, 225 e 18, 74: in entrambi gli scoli omerici corrispondenti si trova un parallelo per il significato aristarcheo, anche se non ricorre il nome di Aristarco (schol. ad 17, 225 τὸ ὑπεράνω τοῦ γόνατος; schol. ad 18, 74 τὸν ἐπάνω τοῦ γόνατος τόπον), mentre non vi si trova alcun parallelo per l’interpretazione di Cratete. Assolutamente probabile dunque che la fonte dell’esegesi aristarchea presente in questo scolio ad Arato sia il commento omerico di Aristarco: e lo stesso può valere anche per il frammento di Cratete, di cui non è rimasta traccia negli scoli omerici, ma si ritrova adespoto in fonti lessicografiche.57  52 Kidd 1976, 45. 53 Supposto allievo di Cratete, se lo era davvero. 54 Nickau 1972, 45–47. 55 Kidd 1997, 45–46. 56 Kidd 1997, 45 (Introd. VI b: Commentators). 57 Insieme alle altre, l’interpretazione di Cratete si ritrova, se pur anonima, in Bachmann 1828, I 228,35; Etym.M. 358, 24 Gaisford; Suda ε 2269 (dove mancano anche i nomi di Aristarco e di Chares).

  Callimaco e la filologia Nello schol. al v. 28 invece è riportata una esegesi puntuale di Aristarco al verso di Arato, sulla quale vale la pena di soffermarsi. Dopo aver parlato dell’ asse (cfr. sopra) e dei due poli, Arato comincia la descrizione delle costellazioni del nord con le due Orse o Carri, che ruotano intorno al polo nord (vv. 26–44): 26 28

Δύω δέ μιν ἀμφὶς ἔχουσαι Ἄρϰτοι ἅμα τροχόωσι· τὸ δὴ ϰαλέονται Ἅμαξαι. αἳ δ’ ἤτοι ϰεφαλὰς μὲν ἐπ’ ἰξύας αἰὲν ἔχουσιν ἀλλήλων.

Schol. ad Arat. Phaen. 28 (p. 78, 9 Martin) αἵ δή τοι (vel δ’ ἤτοι) ϰεφαλάς: ἐπὶ τὰς ἀλλήλων ἰξύας τὰς ϰεφαλὰς ἔχουσι τετραμμένας, οὐϰ ἐπὶ τὰς αὑτῶν, ὡς Ἀρίσταρχος ὁ γραμματιϰὸς ᾠήθη, τὰς ϰεφαλὰς αὐτῶν πρὸς τὴν ἰδίαν ἰξὺν ἀποστρέφεσθαι, παρὰ τὸ φαινόμενον. ὁρῶσι γὰρ εἰς τοὔμπροσθεν, ϰαὶ οὐϰ εἰσὶν ἀλλήλαις ἀνατετραμμέναι.

Secondo lο scolio, la frase αἳ δή τοι ϰεφαλὰς μὲν ἐπ᾽ ἰξύας αἰὲν ἔχουσιν ἀλλήλων significa che le due Orse hanno la testa rivolta l’una in direzione della groppa dell’altra, cioè la testa di ciascuna è allineata con il corpo dell’altra. Arato — prosegue lo scolio — non vuol dire che ciascuna Orsa guarda il proprio corpo, come credeva il grammatico Aristarco, secondo il quale le teste delle due Orse sarebbero rivolte ciascuna verso la propria groppa, il che è contrario a quanto si vede. Le Orse infatti, conclude lo scolio, guardano avanti (puntano sempre in direzioni opposte) e non sono rivolte all’indietro l’una verso l’altra.58 A quanto pare, Aristarco viene rimproverato di avere trascurato il dato fenomenico osservabile, o comunque di averlo messo in secondo piano, rispetto a un’interpretazione filologica del testo che finisce con l’essere astratta e sbagliata: insomma, di essere stato solo un γραμματιϰός e non anche un μαθηματιϰός, secondo la stessa impostazione di ragionamento che abbiamo visto sopra. Quale può essere la fonte di questa annotazione? Omero menziona soltanto l’Orsa maggiore, nominata nella descrizione dello scudo di Achille a Il. 18, 487– 489: tre versi che ricorrono uguali in Od. 5, 273–275, quando Odisseo parte dall’isola di Calipso. Lo scolio di Aristonico a Od. 5, 273 si limita a rimandare al passo dell’Iliade: ὅτι ϰαὶ ἐν Ἰλιάδι τὰ αὐτὰ περὶ τῆς Ἄρϰτου ϰαὶ τῶν ἄλλων ἀστέρων φησίν. Purtroppo però negli schol. ad Il. 18, 487–489 troviamo materiale vario, ma nessun resto di Aristonico. Al v. 487 abbiamo nel ms. A una nota di Erodiano a proposito dell’aspirazione di ἄμαξα, seguita da due schol. D: il primo parla dell’esistenza di due Orse, di cui la Minore non menzionata da Omero e scoperta da Talete; il secondo reca il breve racconto di una ἱστορία callimachea sul catasterismo della ninfa Callisto trasformata in Orsa Maggiore (fr. 632 Pf.).

 58 Martin 1998, 159–160; Kidd 1997, 183.

Callimaco e la filologia  

A 18, 488 i mss. bT offrono due schol ex., nel secondo dei quali (488 b, introdotto da ἄλλως) vengono citati tre passi di Arato, in un contesto in cui si parla dell’Orsa Minore, la Κυνόσουρα ο Κυνοσουρίς. Schol. ad Il. 18, 488 b (ex.): ἄλλως· ἤ τ’ αὐτοῦ στρέφεται ϰαί τ’ Ὠρίωνα δοϰεύει· ἅμα μὲν χαριέντως ὡς ϰυνηγόν, ἅμα δὲ ὅτι ϰαὶ τὴν αὐτὴν ϰίνησιν αὐτῷ ποιεῖται, ὥς πού φησιν ὁ Ἄρατος (Phaen. 226–7)· “ὅς ῥά τε ϰαὶ μήϰιστα διωϰόμενος περὶ ϰύϰλα / οὐδὲν ἀφαυρότερον τροχάει Κυνοσουρίδος Ἄρϰτου”. οὐϰ ὀνομάζει δὲ τὴν ἄλλην, ἐπεὶ τῶν ἐμφανεστέρων μέμνηται. τινὲς δέ †φασιν ἄϰαι†, ἵν’ ᾖ “ϰαὶ Κυνόσουρα”· αὕτη γὰρ “μειοτέρη 〈...〉 πᾶσα περιστρέφεται στροφάλιγγι” (Arat. Phaen. 43). ἀλλὰ πρῶτα μέν ἄμφω Ἅμαξαι ὀνομάζονται, ὡς ϰαὶ Ἄρατος (Phaen. 27)· “ἀτὰρ ϰαλέονται Ἅμαξαι”. ἄλλως τε ὅτε τὸ τέ ὑποτάσσει ὁ ποιητής τῷ ἄρθρῳ, οὐ σημαίνει ϰαί· “ὅν τε ϰαὶ ὑψόθ’ ἐόντα” (Ρ 676), “ὅν τε ϰύν’ Ὠρίωνος” (X 29), “ἥ τε ϰατ’ αἰγίλιπος πέτρης” (I 15), “ ἥ τ’ ἀνὰ νῶτα θέουσα” (Ν 547). οἱ δὲ τὸ “Ἄρϰτον” (Σ 487) ἀντὶ πληθυντιϰοῦ, ὡς τὸ “θήγει δέ τε λευϰόν ὀδόντα” ([Hsd.] scut. 388).

La prima parte di questo scolio cita Arato 226–227, che ricorda il movimento della Κυνοσουρίς in rapporto con quello dell’Ariete, ma conclude che Omero di questa non parla, perché menziona solo gli astri più visibili (vedi sotto). La seconda parte (τινὲς δέ...) fa riferimento a una possibile variante per il v. 488, purtroppo corrotta, che vi introdurrebbe un’allusione all’Orsa Minore, e viene citato Arato 43, vale a dire un verso della parte conclusiva della sezione sulle Orse (cfr. sopra), dove si dice che la Maggiore (Ἑλίϰη) è più visibile e la Minore (Κυνόσουρα) si muove su un’orbita più piccola. Tuttavia questa possibilità per il testo omerico viene respinta, sia sottolineando che Arato può parlare in quel punto (vv. 42–44) dell’Orsa Minore perché prima (vv. 26 sgg.) aveva menzionato esplicitamente entrambe le Orse (e viene citato il v. 27: vedi sopra), sia con una analisi dell’uso linguistico omerico. Infine (οἱ δέ ...) è menzionata anche l’idea che a 18, 487 Omero usi il singolare (Ἄρϰτον) per il plurale: una speculazione puramente esegetica per cercare di introdurre in Omero anche l’Orsa Minore.59 Attira la nostra attenzione la presenza di ben tre citazioni di Arato a fornire materiali di commento e di parallelo alla menzione omerica dell’Orsa Maggiore: due delle citazioni di Arato provengono proprio dalla sezione sulle Orse, vv. 26– 44, cui appartiene il verso per il quale è conservata la nota di Aristarco che abbiamo visto (cioè il v. 28). L’insieme suggerisce senz’altro che ci fosse un commento cospicuo ad Il. 18, 487–489, che probabilmente è confluito almeno in parte negli schol. ex. mentre non si è conservato nella tradizione di VMK. I versi omerici che parlano dell’Orsa Maggiore dovevano essere commentati con dovizia di materiali e di ragionamenti e con abbondanti richiami a quello che era il  59 Segue ancora nel ms. A lo schol. 488 c, uno schol. D che parla del movimento dell’Orsa Maggiore intorno al polo nord e del fatto che essa guarda in direzione di Orione.

  Callimaco e la filologia testo di riferimento per i problemi astronomici, appunto i Fenomeni di Arato. La conclusione che la nota di Aristarco ad Arato 28 provenisse anch’essa dal commento omerico risulta dunque plausibile e non possiamo certo considerarla, così isolata, una prova che egli abbia dedicato un qualche lavoro specifico (di qualunque genere) ai Fenomeni. Ma dobbiamo anche chiederci se questo conclude le nostre possibili riflessioni sugli interessi aristarchei per Arato. Che Aristarco abbia scritto o no qualcosa proprio su Arato, la nota superstite al v. 28 mostra che non fece unicamente un uso sussidiario e ‘passivo’ del suo poema, ma si impegnò anche (almeno talvolta) nell’esegesi puntuale di esso: cosa che peraltro poteva essere facilmente indotta o addirittura resa necessaria anche soltanto per utilizzare i Fenomeni come testo di riferimento o come parallelo. Mi pare lecito pensare che, se lo usò per commentare Omero, lo fece avendolo studiato in modo sufficiente per offrire contributi filologici di rilievo. Quanto lo fece, con quale dedizione e in quale sede (cioè in quale opera), è impossibile dirlo: la selezione dei materiali esegetici nella tradizione dei corpora scoliografici è stata troppo forte per consentirci considerazioni quantitative; il lavoro di Teone può aver costituito un filtro importante, riassumendo osservazioni precedenti e facendo perdere riferimenti espliciti ai predecessori; e senz’altro molte cose si sono perdute in seguito. Certo sarebbe ben strano se la puntuale esegesi del v. 28 fosse l’unico intervento filologico specifico sul testo di Arato prodotto da Aristarco e si fosse conservato così fortunatamente. Di fronte a un caso simile, la distinzione fra il contributo specificamente arateo e l’impiego ancillare nei confronti di Omero risulta forse, nella sostanza, troppo sottile o addirittura inutile. Aristarco ‘doveva’ utilizzare il moderno poeta astronomico per integrare e aggiornare quanto diceva il vecchio Omero a proposito di aspetti scientifici, sui quali le conoscenze avevano fatto innegabili progressi. Doveva dunque studiarlo bene e a fondo, come pure Callimaco, secondo il ragionamento già fatto sopra, e come Apollonio Rodio, di cui diremo fra poco: qualunque fosse l’impulso di partenza, l’interesse non poteva essere sporadico e passeggero; e senz’altro lo stimolo era facilmente suscettibile di sviluppi.

Un accenno su Apollonio Rodio Un breve cenno voglio dedicare anche ad Apollonio Rodio. Una prima considerazione riguarda il grammatico Chares, vissuto probabilmente fra III e II sec. a.C., del quale viene tramandato il titolo di uno scritto probabilmente sulle fonti

Callimaco e la filologia  

del poema apolloniano;60 e Asclepiade di Mirlea, a proposito del quale recentemente G.B. D’Alessio ha molto ben argomentato la possibilità che si sia occupato anche di Apollonio, il che rende quantitativamente più rilevante la sua attività sui poeti ellenistici.61 Nel quadro rientra bene l’argomento che segue, a dare consistenza a una filologia apolloniana prima di Teone. M. Fantuzzi ha recentemente pubblicato un’interessante analisi di un gruppo di scoli apolloniani, nei quali alcune scelte linguistiche del poeta sono criticate come “cattivo uso”.62 Anche se non sempre esplicitato (almeno nella forma conservata), il termine di paragone “corretto” sembra essere (almeno nella maggioranza dei casi) il testo omerico approvato e accolto da Aristarco: in questi scoli non appare considerata la possibilità che gli usi di Apollonio siano innovazioni ricercate e apprezzabili rispetto alla vulgata omerica, ma solo quella che si tratti di scorrettezze e scelte sbagliate. D’altra parte, non accade per Apollonio quello che accade talvolta per Callimaco (e altri poeti ellenistici), cioè di essere spesso addotto negli scoli omerici come testimonianza per la costituzione del testo di Omero. Nel gruppo di scoli apolloniani selezionati da Fantuzzi si sottolinea piuttosto la differenza fra Apollonio e il testo omerico corretto, non per evidenziare un’apprezzabile modernità poetica del νεώτερος, bensì allo scopo di mostrare che egli pecca in termini di esegesi filologica e per questo compie scelte linguistiche non buone né apprezzabili. Una prospettiva evidentemente aristarchea, che per di più lascia intravvedere — anche se a livello di ipotesi e senza una esplicita menzione — l’idea che tali ‘errori’ di Apollonio possano essere condizionati da orientamenti ‘zenodotei’: dal momento che anche Apollonio (come Callimaco) a quanto pare conobbe e utilizzò il testo omerico di Zenodoto, il cerchio dell’argomentazione si chiude in modo plausibile.63 Abbiamo dunque una ventina di esempi di ‘lettura’ aristarchea del poema di Apollonio Rodio in termini di scarso apprezzamento: cosa ne dobbiamo pensare, nel quadro del tema che stiamo sviluppando? Dobbiamo pensare che Aristarco (o un suo allievo) abbia studiato Apollonio solo per estrarre un gruzzolo di casi da confrontare sfavorevolmente con Omero, cioè solo per togliersi il gusto di mostrare quanto il moderno sbagliasse rispetto all’antico? Credo piuttosto che si tratti dei resti di uno studio della lingua e dello stile del poeta delle Argonautiche, rimasto nel corpus degli scoli apolloniani in seguito a processi selettivi

 60 Περὶ ἱστοριῶν τοῦ Ἀπολλωνίου in Schol. ad Apoll.Rh. 2, 1052: cfr. D’Alessio 2000, 93–95. 61 D’Alessio 2000, con le informazioni bibliografiche. 62 Fantuzzi 2000; cfr. anche Rengakos 1994 e Rengakos 2001. 63 Tanto più considerando che nella filologia alessandrina una sorta di “linea zenodotea” era rimasta viva fino all’epoca di Aristarco: cfr. Montanari 1988 (SGLG 7).

  Callimaco e la filologia svariati e pesanti. Pare certo che Aristarco amasse Callimaco più di Apollonio, ma difficilmente avrà mancato di riconoscere con interesse lo spessore di erudita filologia (specialmente omerica) che quest’ultimo aveva racchiuso in ogni verso del suo poema:64 se gli interessava principalmente prendendo spunto da Omero, lo studiava in quanto Apollonio (e la linea di separazione rimane assai sottile), magari per mostrare che scelte errate di esegesi omerica portano a usi poetici non buoni. Se questi esempi di scelte censurate sono, almeno in parte, testimonianze di lezioni zenodotee, possono essere state evidenziate per questo, come accade per Callimaco, anch’egli qualche volta censurato per una scelta zenodotea.65

Conclusione Le conclusioni sono state anticipate e adesso devo soltanto riassumerle. L’attività filologica di Aristarco, ben impiantata su una tradizione già consolidata da Zenodoto ad Aristofane di Bisanzio, si concluse intorno alla metà del II sec. a.C. e fu seguita da quella rigogliosa dei suoi discepoli. Il panorama della cultura alessandrina aveva conosciuto una stagione straordinaria e aveva acquistato la sua ricchezza e la sua piena maturità. In tale contesto culturale, essere filologi non era un’eccezione: per gli ambienti intellettuali, leggere i testi poetici da eruditi e da grammatici, esprimere opinioni e dare contributi esegetici era nell’aria e nello spirito del tempo. Sembrerebbe dunque addirittura strano che non diventasse presto oggetto di queste cure e di questi atteggiamenti anche la grande poesia del primo ellenismo, dei poeti ‘moderni’. Non mi sembra paradossale dire che quella prassi poetica intrisa di erudizione invitava essa stessa alle cure filologiche, stimolava lo sforzo e il confronto intellettuale dell’esegesi, provocava il lettore colto all’analisi degli spessori abilmente dissimulati e delle ricercate preziosità.66 Con interventi rilevanti a partire dall’età di Aristarco (quale che fosse la forma materiale in cui erano redatti), cominciò allora la storia della filologia sulla poesia ellenistica: Callimaco fu probabilmente l’autore più amato e per questo è, di quel gruppo di poeti, il più citato nella  64 Cfr. da ultimo Rengakos 2001. 65 Cfr. per esempio schol. ad Il. 12, 34 + schol. ad Apoll.Rh. 1, 1309: Rengakos 1993, 61–62, 82; 1994, 61, 155; Montanari 1995b, 57–58 (= questo volume, cap. 62). 66 Cfr. Cameron 1995, 230: “The moment the Aetia began to circulate outside Alexandria this need [di spiegazioni varie] will have become pressing. The work as a whole is highly allusive, and it is not hard to imagine very early copies being equipped with at least a skeleton of explanatory notes”.

Callimaco e la filologia  

letteratura erudita antica. Grande poeta filologo, fu ben presto grande poeta oggetto di filologia.

Discussion Th. Fuhrer: Il me semble très intéressant de concevoir le texte des poèmes homériques, dès l’ekdosis de Zénodote, comme ‘copie de travail’: ayant cette forme de ‘brouillon élaboré’ (si vous me permettez d’utiliser cet oxymoron), il est aussi une documentation des apories de la philologie homérique. Si donc des poètes comme Callimaque et Apollonius utilisaient cette copie, cela explique peut-être pourquoi, dans leur poèmes, on trouve si nombreuses allusions et discussions au sujet du texte homérique: le texte était quelque chose de ‘flou’ pour eux. Puisque le texte des poèmes homériques était sujet de tant de discussions et débats philologiques, les commentaires sur ce texte ne peuvent forcément pas avoir eu le même caractère que les commentaires sur les textes des poètes contemporains ou peu antérieurs. Alors non seulement le commentaire sur Callimaque que nous a préservé le papyrus de Lille, mais aussi toute forme de travail d’un savant comme Aristarque sur le même poète devait être autre chose que le commentaire aristarchéen sur le texte d’Homère. F. Montanari: Il caso di Omero è diverso da quello di tutti gli altri poeti, è un caso unico per il grande accumulo di erudizione che c’è stato nel corso del tempo, per la quantità d’interesse suscitato a vari livelli (da quello scolastico elementare fino a quello filosofico e teologico) e anche per le fluttuazioni testuali (qualunque sia l’interpretazione e la visione storica che se ne vuole dare, e anche il peso che si vuole attribuire al fenomeno: è un problema ancora molto dibattuto). È vero dunque che anche i prodotti eruditi relativi a Omero sono per certi aspetti differenti da tutti gli altri. Inoltre, ci si può facilmente immaginare che le annotazioni erudite su un testo ‘moderno’ fossero diverse da quelle relative a un testo del periodo arcaico e classico. L. Lehnus: Trovo molto suggestiva l’idea che Teone in età augustea non sia un punto di partenza bensì, in un certo senso, un punto di arrivo. In che misura l’esistenza precoce di commentari o comunque di interessi esegetici può aver influenzato la conservazione o la perdita di opere di poeti ellenistici? Per esempio: sarei indotto a credere che l’Ibis non sia stata commentata e quindi andò perduta (forse già in età ellenistica).

  Callimaco e la filologia F. Montanari: Pensando a Teone come possibile punto di arrivo di una tradizione esegetica già formata, se pur solo da qualche generazione, devo naturalmente concepirlo anche come nuovo punto di partenza di una stagione nella quale i poeti ellenistici diventano più normalmente oggetto di cure esegetiche, fino a entrare a un certo punto, come dicevo sopra, nel novero dei πραττόμενοι, cioè i normalmente ‘trattati’ dai grammatici. Penso che l’attività esegetica possa aver avuto un ruolo nella conservazione e perdita di opere: eviterei di pensare a una relazione meccanica e necessaria, anche perché siamo costretti a ragionare e silentio (con la possibilità che un nuovo frammento di papiro ci smentisca): per esempio, a quanto pare gli Aitia furono presto oggetto di cure, ma non si conservarono. A.S. Hollis: An example (perhaps not noticed before) of learned interpretation of Aratus, Phaenomena, reflected in Virgil, Georgics. In this case the issue is purely philological, no technicalities of astronomy are involved. In Georg. 1.387 et studio incassum videas gestire lavandi, why does Virgil write incassum, “in vain”? Perhaps he shows himself aware of an interpretation of αὔτως (αὔτως) in Aratus, Phaen. 945 γαστέρι τύπτουσαι αὔτως εἰλυμένον ὔδωρ. One possible meaning of αὔτως is “in vain’’ (LSJ I.2). Kidd, however, takes αὔτως (rough breathing, as in the manuscripts) with the words which follow, translating (p. 143) “the water enclosed as it is” (see further his note on Phaen. 21). Controversy may have extended to the breathing, since (according to LSJ) some grammarians distinguished between αὕτως (likewise) and αὔτως (in vain). Chronologically, it seems possible that Virgil read a commentary on Aratus by Theon, son of Artemidorus. F. Montanari: Grazie, il caso è senz’altro interessante e l’idea mi sembra del tutto plausibile. R. Hunter: How does the famous problem of Quintilian’s report about Aristophanes Byz. and Aristarchus and poets of suum tempus relate to what you have been saying (Inst. 10,1,54: Apollonius in ordinem a grammaticis datum non venit, quia Aristarchus atque Aristophanes neminem sui temporis in numerum redegerunt)?. Is it a question of ‘genre’, an idea that may have been associated only with ‘older’ poets, or of different levels of philological interest which can be discerned in the different types of material which survive? F. Montanari: Credo che il passo di Quintiliano si riferisca alle famose (e anche abbastanza problematiche...) “liste canoniche”, che effettivamente riguardavano i

Callimaco e la filologia  

poeti ‘antichi’: mi pare del tutto naturale che, ai tempi di Aristofane e Aristarco, i canoni non comprendessero i poeti ‘moderni’, che ancora non erano normalmente e abbondantemente πραττόμενοι. A mio avviso, si deve pensare (come abbiamo già detto) che il lavoro filologico sui poeti alessandrini comportasse effettivamente nelle prime fasi materiali e modalità diverse rispetto a quello sugli autori da Omero a Aristofane, dunque anche un livello diverso di interessi filologici (si veda per esempio quanto detto nel testo a proposito delle spiegazioni di carattere prosopografico). Comunque, Quintiliano offre una testimonianza da tenere presente nella valutazione del problema e dei suoi connotati storici. P.J. Parsons: Why do we have no reference to commentaries on Alexandrian poets, e.g. by Aristarchus? F. Montanari: Colgo l’occasione per precisare che non voglio sostenere per forza che ci fossero veri e propri hypomnemata sui poeti ellenistici già ai tempi di Aristarco: questo è possibile, ma è difficile dire se e quanto fossero estesi ed impegnativi; gli unici esempi su cui possiamo ragionare sono i due frammenti papiracei di cui abbiamo parlato. Credo fermamente che esistessero già interessi filologico-esegetici diretti e specifici sui ‘moderni’, ma non voglio impegnarmi molto nel supporre quale forma potessero avere (brevi trattazioni tematiche? annotazioni marginali?). Non possiamo considerare troppo significativo il fatto che non abbiamo testimonianze e riferimenti specifici, considerando la scarsezza del materiale pervenuto: ricordiamoci (cfr. nel testo, n. 18) che nessuna testimonianza ci parla di commenti a Erodoto, ma PAmh. 12 ci ha restituito la prova che Aristarco aveva scritto un vero e proprio hypomnema su questo autore. R. Hunter: Can we be sure that Aristarchus actually ‘discussed’ (somewhere) the meaning of Aratus, Phaen. 28–29? Is it possible that he just quoted and/or paraphrased the verses without any sign that the meaning was problematic? Aratus was the acknowledged text on the stars and (if my memory is correct) Hipparchus sees problems not with the interpretation of Aratus’ text but with his facts. F. Montanari: Non posso certo negare che il mio discorso si basi su indizi e non si può certo avere una certezza assoluta; d’altra parte, credo di averlo presentato con le dovute cautele. Tuttavia, mi pare che il riferimento agli interventi di γραμματιϰοί e μαθηματιϰοί vada tenuto presente come testimonianza di discussioni su problemi sia filologici che scientifici. Dovunque fosse trattato il caso che abbiamo visto, si tratta comunque di un contributo esegetico specificamente

  Callimaco e la filologia arateo: l’interesse proprio per Arato mi sembra innegabile, qualunque fosse stato il punto di partenza o il primo impulso. P.J. Parsons: Are these two papyri (Oyster and PLille) special cases? 1) Oyster: a riddle, therefore needs explanation. Riddles (here in epigram-form) are sympotic material and their explanation could be part of the learned symposium? 2) PLille: was it designed as patriotic reading for schools (cfr. epigrams on Ptolemaic achievements in the school texts PDidot and PGuéraud-Jouguet)? Was it just this poem, not the whole of book III (not probable)? Its explanatory material is basically elementary (and glosses become less frequent as the text progresses?): should we distinguish between schoolmasters’ explication de texte and the serious work of philologists like Aristarchus? F. Montanari: Certo, è sempre possibile che la casualità dei ritrovamenti papiracei ci faccia lo scherzo di proporci ‘casi speciali’: almeno altrettanto possibile è che ci offra esempi limitati di qualcosa che esisteva con maggiore abbondanza. Il fatto è che questi testimoni esistono e che la loro esistenza appare consonante con alcuni indizi che ho cercato di far emergere (cfr. anche Cameron 1995, 44, 46 e 66). D’altra parte, non è certo mia intenzione sostenere che nel II sec. a.C. ci fosse una produzione esegetica sui poeti ellenistici abbondante quanto quella sui poeti arcaici e classici e con le stesse caratteristiche. Certamente c’è differenza fra la explication de texte in forma parafrastica per uso esclusivamente scolastico e il lavoro filologico-esegetico di più alto livello: tuttavia non credo giusto tracciare una linea di separazione totale e addirittura di contrapposizione, la parafrasi del testo è uno strumento utilizzato anche dal filologo più agguerrito. Inoltre, in PLille non ci sono soltanto glosse e parafrasi, pur nel poco che si è conservato del commento. È importante sottolineare, come ho detto nel testo, che una nota ‘lunga’ tratta di problemi prosopografici dei Tolomei.

 Satiro in P.Oxy. 2506 (fr. 26 II)? Nella sezione cui appartiene il fr. 26 col. II, l’autore del commentario o trattato su poesia lirica restituito da P.Oxy. 25061 sta illustrando con esempi i debiti dei poeti tragici nei confronti di Stesicoro (cfr. fr. 217 PMG):2 parla prima di Eschilo nelle Coefore (II 7–14), poi di Euripide nell’Oreste (II 14–24), infine ancora di Euripide nell’Ifigenia in Aulide (II 25–27). È su questi righi che ci soffermiamo, così dati da Page nell’ed. pr. Εὐριπίδ]η̣ϲ δὲ ϰαὶ τὴν Ἰφ[ιγένειαν ἐ]π̣οίηϲε γαμουμέ̣ [νην Ἀχιλλεῖ ± 4]... Ϲ̣α̣τ̣ [.]ρ̣. [ –––––

Le tracce al r.27 (dopo il quale la colonna si interrompe) fanno venire in mente un buon candidato, cioè Satiro, che per esempio nella sezione della Vita di Euripide (parzialmente conservata in P.Oxy. 1176) dedicata alla techne del tragediografo3 poteva bene aver parlato anche dei debiti di quest’ultimo nei confronti di poeti anteriori come Stesicoro. Ho potuto controllare il papiro nell’Ashmolean Museum di Oxford, assieme agli amici R.Coles e P.Parsons, e con l’aiuto del microscopio binoculare. Il c letto da Page è sicuro; segue una traccia in alto che va bene come parte di α; allo stato attuale, non vedo traccia della lettera seguente. Al probabile α segue una lacuna di un paio di lettere, dopo la quale una piccola curva alta è con tutta probabilità di un ρ, come trascritto da Page; e poi c’è un’altra curva alta (sono conservate le estremità superiori delle lettere), che può essere la parte superiore di un o. Potremmo dire di aver recuperato con buona probabilità Ϲα̣[τυ]ρ̣ο̣ [, o anche Ϲατ̣[υ]ρ̣ο̣ [ se conserviamo credito al τ letto da Page.

 1 Page 1963; cfr. le discussioni di Davison 1966; Pfeiffer 1973, 345 (= 1968, 222); Arrighetti 1977a, 47sg. Page era molto cauto nella definizione dell’opera, che intitolava “Comment” (invece del più compromettente “Commentary” che si legge nella prefazione di Turner/Skeat, p.V), mentre Pfeiffer ha preferito vedervi resti di un trattato del tipo περί τοῦ δεῖνα; giustamente dubbioso Arrighetti, l.c. 2 Cfr. Page 1963, 37. 3 In quello che abbiamo di P.Oxy. 1176, la sezione sulla techne è conservata solo in minima parte: cfr. Arrighetti 1964, 26sg. (seguendo l’ed. pr. di A.S. Hunt, P.Oxy. vol. IX, London 1912, p. 124; Fr. Leo, NGG 1912, p. 286 = Kl.Schr. II, p. 379; vedi anche H. Gerstinger, WS 38, 1916, p. 55sg.). Sulla personalità di Satiro vedi l’introd. di Arrighetti con la bibliogr.; più recentemente S. West 1974; Arrighetti 1977a; Tronson 1984. https://doi.org/10.1515/9783110772371-059

  Satiro in P.Oxy. 2506 (fr. 26 II)? Purtroppo, non si può progredire molto oltre questo punto. Le ovvie possibilità che si presentano sono: ὥϲ φη]ϲ̣ι̣ (ν̣ ) Ϲ. , ὡϲ λέγ]ε̣ ι Ϲ. oppure ὡϲ ϰ]α̣ὶ̣ Ϲ. (scil. φηϲι, ἀνέγραψε, al.) o simili. Ma le tracce prima di c non appaiono conciliabili con nessuna di esse. Ciò che si vede è un minimo resto di una lettera, seguito da due puntini in alto con una minuscola traccia (rivelata solo dal microscopio) centrale a mezza altezza: la lettera con cui meglio si concilierebbe questo insieme è η. Naturalmente, ciò è ben lungi dall’essere sicuro: accettando per buona la lettura di η, possiamo pensare alla fine di una parola legata alla frase precedente e dopo la quale mettere un punto fermo, avendo quindi al r.27 Ἀχιλλεῖ ± 4].η̣. Ϲά̣[τυ]ρ̣ο̣ [ϲ δὲ ϰτλ. oppure τῶι Ἀχιλλεῖ ±2].η̣. Ϲά̣[τυ]ρ̣ο̣ [ϲ δὲ ϰτλ. Un’altra strada sarebbe quella di pensare al dativo dell’articolo femminile, con il nome di Satiro al genitivo seguito da un sostantivo, pur considerando la difficoltà per cui di solito nel papiro lo iota muto è ascritto, mentre qui sarebbe omesso. Per esempio: ὡϲ ἐν] τ̣ ῆ̣ Ϲα̣[τύ]ρ̣ο̣ [υ (βίων) ἀναγραφῇ oppure ϲυγγραφῇ, o addirittura ἐπιτομῇ.4 È evidente che tutto questo è più o meno pura speculazione, essendo troppo poco e incerto quello che abbiamo. Sarà dunque meglio accontentarsi di aver recuperato con buona probabilità il nome di Satiro, anche se non si riesce a stabilire il contesto preciso in cui era inserito: beninteso, tale ’recupero’ va preso come un suggerimento, che può chiedere conforto al ben rilevato aspetto del lavoro di Satiro consistente nell’assiduo uso delle fonti letterarie. Quando, esaminando il papiro e controllando l’originale, ho pensato alla lettura proposta sopra, non sapevo ancora che il nome di Satiro era già stato collegato con P.Oxy. 2506, ma in modo ben diverso. Discutendo la difficoltà di caratterizzare il tipo di opera che vi è contenuto e le caute incertezze di Page nell’ed. pr., J. Davison avanzò un’idea singolare:5 per lui, “we have what seems to be a continuous discussion by the author of the work, in which quotations are interspersed and discussed for their importance as throwing light on the biography of the authors concerned”, per cui egli si sente di suggerire che “if we had the whole text of the work or even a reasonably consecutive text of a substantial part of it, we should find that its purpose was indeed biographical, and that it formed part of a series of Lives of the Poets, analogous to that life of Euripides of which substantial fragments were published in 1912 ad P.Oxy. 1176”, vale a dire la Vita di Euripide di Satiro, che faceva parte di un’opera comprendente anche

 4 In P.Oxy. 1176 (fr. 39 XXIII, p. 80 Arrighetti) l’opera di Satiro reca il titolo Ϲατύρου βίων ἀναγραφή: discussione su di esso in Arrighetti 1977a, 30sg.; cfr. Gudeman 1921a, 233. Nel caso di ἐπιτομή, si dovrà pensare a quella ben nota di Eraclide Lembo: cfr. Diogene Laerzio IX 26 ϰαϑά φηϲιν Ἡραϰλείδηϲ ἐν τῇ Ϲατύρου ἐπιτομῇ (= FHG III, ρ.169). 5 Davison 1966, 98sgg. (cfr. anche Davison 1965; Davison 1968, 222).

Satiro in P.Oxy. 2506 (fr. 26 II)?  

quelle di Eschilo e Sofocle (vd. colofone, fr. 39 XXIII, p. 80 Arrighetti); infine “it is thus tempting to suggest that No. 2506 may contain parts of another book of the same work, known to Diogenes Laertius and Athenaeus as Ϲατύρου Βίοι”. Per sostenere la sua idea (che insomma si costruisce un po’ su se stessa), Davison deve per ovvie ragioni cronologiche negare che in P.Oxy. 2506 sia citato Aristarco (proposto come integrazione principalmente a fr. 6 a, 6: cfr. fr. 79,7); ma soprattutto affermare che non è decisivo il fatto che non si trovi traccia di quella forma dialogica che l’opera biografica di Satiro ha rivelato in P.Oxy. 1176 come sua caratteristica saliente, con la giustificazione che le parti leggibili con una certa continuità di P.Oxy. 2506 sarebbero troppo brevi per escludere con certezza tale forma dialogica. In sostanza, tale ipotesi pare nata soltanto dalla difficoltà di definire il tipo di opera contenuta in P.Oxy. 2506, ma le sue forzature mi paiono evidenti: non ha argomenti a favore e ne ha di non trascurabili contro. È infatti una proposta che non ha avuto seguito, e anch’io ritengo che debba essere lasciata cadere. Non c’è invece nessun argomento contrario, ed anzi una certa verisimiglianza confortata dalle tracce, riguardo alla possibilità che l’autore di un’opera filologico-erudita di qualunque genere (hypomnema, syngramma, miscellanea?), che citava autorità come Aristotele, Dicearco, Cameleonte, Aristarco, possa occasionalmente avere citato anche Satiro.

 Un papiro di Eschine con correzioni (P.Oxy. 2404): Considerazioni sull’ekdosis alessandrina È un dato di fatto che i nuovi esemplari di opere letterarie prodotti negli scriptoria potevano essere oggetto di rilettura e correzione grazie a un nuovo confronto con l’antigrafo e talvolta anche sulla base di una collazione con altre copie. Nel ricercare e indagare le prime testimonianze di questo fenomeno “librario” siamo condizionati dal fatto che i più antichi papiri greci a noi pervenuti sono pochissimi e risalgono alla seconda metà del IV sec. a.C. Uno di essi è il celebre frammento dei Persiani di Timoteo, P.Berol. inv. 9875. Alla col. IV, v. 133, dopo avere scritto βορεαιαραισον|ται, lo scriba ha inserito un δ supra lineam per ripristinare la corretta lezione βορέᾳ διαραίσον|ται. Alla col. V, v. 196, dopo avere scritto πλουτουοδε, lo scriba ha inserito uno ι supra lineam per ripristinare la corretta lezione πλούτου oἱ δέ. Due interventi di correzione (διόρθωσις) realizzati in scribendo oppure in seguito a una rilettura. Nel celebre Papiro di Derveni collocato verso la fine del IV secolo a.C., a col. XXI 11 la prima mano corregge ]ρμοναδε con l’inserimento di uno iota supra lineam per ripristinare il corretto Ἁ]ρμονία δέ, mentre a col. VI 5 non è stata corretta l’omissione di una lettera in τοιδε a fine rigo per τοῖς δέ. Si tratta di correzioni occasionali, realizzate dallo scriba probabilmente quando se ne accorgeva in scribendo, piuttosto che di un lavoro sistematico.1 Questi interventi di correzione di errori materiali non sono certo un fenomeno particolarmente vistoso, ma si tratta delle più antiche testimonianze direttamente visibili della preoccupazione di realizzare un testo corretto in ambito librario. Al III sec. a.C. appartiene il papiro milanese con gli epigrammi di Posidippo (P.Mil. Vogl. 309), un testimone assai rilevante dal nostro punto di vista, sia per l’epoca alta cui appartiene, sia per la quantità degli interventi sul testo che presenta. La maggior parte delle correzioni è opera dello stesso scriba, ma in seguito altre due mani sono intervenute con ulteriori emendamenti, e le differenze di comportamento sono da tenere presenti. «Il medesimo scriba è responsabile della maggioranza delle correzioni ... Gli interventi sono tutti assai limitati (in genere coinvolgono una sola lettera e mai più di tre) e sono tutti diretti ad emendare banali errori di stesura, cioè fraintendimenti, sostituzioni accidentali ed omissioni. La maggior parte di essi è stata manifestamente eseguita in scribendo;

 1 Turner/Parsons 1987, 92; testo in Kouremenos/Parassoglou/Tsantsanoglou 2006. https://doi.org/10.1515/9783110772371-060

Un papiro di Eschine con correzioni (P.Oxy. 2404)  

e tutti quanti sono stati compiuti con studiata accuratezza ... infatti, se gli errori emendati fossero apparsi evidenti, avrebbero dato immediatamente l’impressione che il testo fosse scritto senza la necessaria precisione ed avrebbero svilito il lavoro di chi lo aveva scritto. Dopo quella del copista, altre due mani hanno inserito emendamenti nel rotolo, operando entrambe in pochi punti e senza avere la precauzione di occultare i loro interventi, come invece cercava di fare il copista ... Si potrebbe pensare che questa [scil. la seconda mano, m. 2] sia la mano di un revisore del centro di copia in cui fu realizzato il rotolo. L’ipotesi non è del tutto esclusa; ma non è nemmeno sostenuta da indicazioni certe. Anzi, il fatto che le correzioni di m. 2 siano tutte concentrate in due sole colonne consecutive, induce a ritenere che chi le apportò fosse un lettore particolarmente interessato a quella sezione dello scritto, più che un revisore obbligato a controllare tutto il rotolo. Fu sicuramente un lettore la terza persona che intervenne sul testo [scil. m. 3]... i suoi interventi sono concentrati tutti nella col. XI. Lì egli segnalò una variante di lettura per la l. 30, annotandola nel margine superiore».2 A col. XI 30 si legge κεντρακαιεξ. [ e nel margine superiore m. 3 ha scritto καικεντρα (le ultime tre lettere non si leggono nella fotografia, ma si possono vedere nell’originale).3 È del tutto probabile che si tratti di una correzione oppure di una vera e propria variante in connessione con il κέντρα καὶ del testo: forse si indicava una possibile inversione καὶ κέντρα, ma la perdita della restante parte del verso rende difficile capire bene la situazione.4 I papiri ci testimoniano bene l’uso di diverse modalità per migliorare un esemplare librario, vale a dire correggere la copia di un testo e aggiungere elementi utili alla sua comprensione: si poteva aggiungere, togliere o modificare lettere o parole omesse o scritte per errore, cancellare quello che si riteneva sbagliato per sostituirlo con ciò che si riteneva corretto, scrivendolo sopra la riga, nei margini e negli intercolunni (talvolta con segnali di richiamo), oppure anche al posto delle parole precedenti, una volta eliminate. Per cancellare qualcosa si poteva tirare un tratto orizzontale o obliquo (anche molto vistoso) sulle lettere o le parole da eliminare, oppure esse potevano venire contrassegnate da punti o da linee al di sopra o al di sotto, oppure ancora essere racchiuse entro una sorta di parentesi tonde in coppia, oppure essere materialmente erase o

 2 Bastianini/Gallazzi 2001, 15. 3 Bastianini/Gallazzi 2001, 76–77. 4 Cfr. la discussione in Bastianini/Gallazzi 2001, 199–200; Austin/Bastianini 2002, 96; Ferrari 2005, 204; Lapini 2007, 280–281, con la bibliografia precedente sull’epigramma.

  Un papiro di Eschine con correzioni (P.Oxy. 2404) lavate via con una spugna.5 La documentazione per le fasi più antiche non è abbondante, come è naturale perché scarsi sono i reperti in generale, ma con il proseguire del tempo essa diventa cospicua ed estremamente significativa: l’esemplificazione può essere facilmente addotta e arricchita,6 più avanti esamineremo un caso significativo. Scrivono Turner e Parsons: «One of the questions the palaeographer should ask about any literary manuscript is whether it has been adequately compared against its antigraph (the exemplar from which it was copied), a task which, in a publishing house, was the duty of the diorthotès, corrector, or whether it has been collated with a second exemplar a procedure often carried out by private individuals to secure a reliable text». “But several of our surviving papyrus manuscripts, and especially those which are beautifully written, contain such serious un-noted errors that it is clear their ‘proof-reading’ was of a summary, superficial kind, if done at all ... Those ancient themselves who set store by having a dependable copy (persons like Strabo and Galen) were aware of this weakness and adopted a routine to counter it: they themselves (or their secretaries) checked the copy to be used against another exemplar. If, therefore, the text had been checked against its first exemplar, and was later collated with a second, it may well bear the marks of this double checking”.7 In alcuni lavori degli ultimi anni mi sono occupato di un problema centrale per la storia della filologia e della cultura antica: la forma dell’ekdosis alessandrina che nel periodo da Zenodoto ad Aristarco si impone come risultato tipico dell’opera dei filologi, accanto allo hypomnema, al syngramma, alla raccolta di lexeis e altri prodotti esegetico-eruditi. Riprendo qui solo alcuni punti essenziali.8 Problemi decisivi si sono posti a partire dall’interpretazione del modo di lavorare di Zenodoto, il primo dei maggiori filologi alessandrini. Condivido del tutto l’idea esplicitamente sostenuta da Pfeiffer e Nickau, secondo cui è difficile immaginare un procedimento diverso dal lavoro di correzione eseguito su una copia già esistente: una copia opportunamente scelta fra quelle a disposizione, utilizzata come testo-base sul quale operare gli interventi voluti e via via annotata dal grammatico con il procedere dei suoi studi e del suo lavoro di diorthosis.9  5 Turner/Parsons 1987, 15–16, con rinvio a esempi nelle tavole; cfr. Turner 1980, 93 e PI. VIII (= 112–113 e tav. VIII); Bastianini 2001; Montanari 2009a, 146–150 (= questo volume, cap. 7). 6 Alcuni esempi in Montanari 2009a, 146–150. 7 Turner/Parsons 1987, 15–16, e Turner 1980, 93 (= 1984, 112–113). 8 Montanari 1998a (= questo volume, cap. 23); 2002a (= questo volume, cap. 47); 2004 (= questo volume, cap. 41); 2009a. 9 Pfeiffer 1968, 110 (= 1973, 188–189): «It is not improbable that Zenodotus, examining manuscripts in the library, selected one text of Homer, which seemed to him to be superior to any

Un papiro di Eschine con correzioni (P.Oxy. 2404)  

Un simile modo di produzione dell’ekdosis fu adottato da Zenodoto e rimase in uso da parte dei grammatici posteriori. Un filologo sceglieva, secondo le proprie preferenze, un esemplare che gli risultava adatto e valido come base di lavoro: quando il testo non incontrava la sua approvazione, in corrispondenza del luogo interessato egli correggeva, toglieva o aggiungeva, annotava la lezione preferita negli spazi liberi o nell’interlinea, all’occorrenza metteva in margine gli opportuni semeia. Egli operava queste correzioni sia sulla base del confronto con altri esemplari sia ope ingenii. Il “suo” testo risultava dunque dall’insieme rappresentato dal testo-base più le indicazioni di modifica contenute in tutto il “contorno” paratestuale e semeiotico aggiunto da lui. Dobbiamo immaginare tutto questo come esito di anni di studio, che producevano via via svariati interventi sullo stesso esemplare: istoriato e martoriato dal lavoro di diorthosis (fatto di interventi ben visibili e non “mascherati” per esigenze librarie estetiche e commerciali), questo esemplare era concretamente l’ekdosis omerica di quel filologo, la sua copia personale, recante il suo nome per identificazione e contenente il frutto delle sue riflessioni e indagini, ekdosis in quanto appunto ἐκδοθεῖσα, cioè a disposizione di dotti, scolari, poeti e intellettuali. Mi risulta difficile trovare argomenti contro questa idea di fondo, secondo cui il lavoro per realizzare una ekdosis di carattere filologico si basava di fatto, sul piano operativo materiale e librario, su modalità di lavoro sicuramente non nuove e certamente per nulla stravaganti nel mondo intellettuale e colto, ormai abituato ad essere una “civiltà del libro” da almeno un paio di secoli. Nei lavori ricordati ho cercato dunque di sottolineare e valorizzare l’importanza centrale dell’aspetto librario nel formarsi di una prassi filologica e l’opportunità di chiamare in causa con il dovuto peso quanto sappiamo delle pratiche in uso e consolidate per la revisione e correzione dei testi. La migliore pratica di produzione libraria prevedeva operazioni di confronto fra copie e interventi di correzione, diorthosis realizzati da un diorthotès di professione o anche occasionale, che aveva buone risorse tecniche per cancellare, aggiungere, sostituire, segnalare  other one, as his main guide; its deficiencies he may have corrected from better readings in other manuscripts as well as by his own conjectures. Διόρθωσις can be the term for either kind of correction. It is hard to imagine any other way»; Nickau 1972, 30–31: «Dann ist zu fragen ob Z(enodotos) nicht einen durch Recensio ermittelten Homertext zugrundelegte (der jedoch nicht seinen Vorstellungen von der genuinen Form der Epen entsprach), diesen mit Obeloi versah und zu ihm Textvorschläge sowie deren Begründung mitteilte. Z(enodotos) selbst wie auch seine Hörer machten sich entsprechende Notizen, die, wären sie von Z(enodotos) schriftlich veröffentlicht worden, ‘Hypomnemata’ hätten heißen können. Aber die Zeit der schriftlich publizierten Homer-Kommentare begann erst mit Aristarchos. So würden sich auch die späteren Unsicherheiten in der Berichterstattung über Z(enodotos’) Ausgabe erklären».

  Un papiro di Eschine con correzioni (P.Oxy. 2404) vari aspetti e caratteri del testo allo scopo di migliorarlo e renderlo più affidabile. I metodi e la tecnologia libraria offrirono una base per così dire tecnicoartigianale, che trovò applicazione nel lavoro del grammatico, secondo uno sviluppo che non doveva apparire a prima vista particolarmente strano o stravagante, ma implicava un principio innovativo straordinario e consapevole: con Zenodoto, ricordato come il primo diorthotès di Omero, la diorthosis del correttore dello scriptorium divenne la diorthosis del grammatico, preoccupazioni e interventi di carattere specificamente librario ed estetico-commerciale si mutarono in preoccupazioni e interventi di carattere critico e filologico-grammaticale.10 Possiamo pensare di concettualizzare il passaggio riflettendo nei termini seguenti. Lo scopo che guidava il lavoro del correttore in uno scriptorium era quello di produrre una copia corretta da vendere a un compratore-committente, vale a dire un esemplare realizzato nel modo migliore possibile sul piano artigianale. L’idea di fondo del grammatico nel correggere il suo testo era invece quella di trovare la forma giusta dell’opera stessa di cui si stava occupando: lavorava su una copia con lo scopo e l’intenzione di realizzare l’esemplare modello, nel quale si potesse trovare quella che era secondo lui la forma esatta dell’opera letteraria, compresa l’indicazione quanto possibile codificata di dubbi e aporie testuali, un aspetto che certo non apparteneva al sistema mentale e all’orizzonte operativo dello scriptorium e dei suoi artigiani. Così alla cancellazione drastica (operazione tipica del correttore artigianale della produzione libraria) si affiancò dapprima (grazie a Zenodoto) il segno del dubbio filologico, l’obelos, che segnò il cambiamento intellettuale decisivo nel senso di pensare all’opera in quanto tale invece che alla singola copia. Poi il progresso della disciplina andò sempre più nella direzione di incrementare e sviluppare i segnali di intervento critico, di riflessione e di indagine, non tanto su una singola copia con intento produttivo artigianale, quanto sul recupero della volontà autentica dell’autore, dunque anche sull’interpretazione corretta del testo oggetto di studio. La diorthosis del grammatico non aveva di mira il singolo esemplare, pensava invece all’opera in se stessa: alcuni strumenti pratici collaudati dalla produzione libraria certamente risultarono comodi e furono utilizzati, ma con finalità e atteggiamenti intellettuali di altro genere. Abbiamo già sottolineato la difficoltà di reperire testimonianze dirette per le fasi più antiche, vale a dire il periodo iniziale della filologia alessandrina. Per quanto riguarda gli aspetti tecnico-artigianali della produzione libraria, che sono rilevanti per il nostro discorso, siamo dunque obbligati a servirci prevalentemente di documentazione più tarda, essendo troppo scarsi i papiri di età zenodotea in  10 Cfr. Nickau 1977, 10–11; Montanari, opp. citt. sopra n. 8.

Un papiro di Eschine con correzioni (P.Oxy. 2404)  

nostro possesso. Non credo tuttavia ci possano essere dubbi che stiamo parlando di aspetti che appartengono in generale alla civiltà del libro greco, per cui l’abbondante documentazione di età imperiale vale senz’altro a far luce su fenomeni presenti anche nei secoli precedenti, per i quali le evidenze scarseggiano. Entro questo quadro di riferimento e da questo punto di vista, vale la pena riconsiderare un testimone di grande interesse e molto studiato per diverse ragioni: P.Oxy. 2404 + P.Laur. inv. ΙII/278, un frammento di rotolo papiraceo recante parte dei §§ 51–53 (P.Oxy. 2404) e dei §§ 162–163 (P.Laur. III/278) dell’orazione Contro Ctesifonte di Eschine.11 Esso appartiene a un gruppo di sei rotoli, che per la peculiarità della scrittura sono stati attribuiti a uno stesso scriba, collocabile alla fine del II o all’inizio del III sec. d.C.,12 e che sono accomunati dal fatto di essere esemplari di pregio (scrittura calligrafica, ampi margini, materiale di alta qualità). Tre di essi recano segni evidenti di revisione e di correzioni di vario tipo (realizzate da mani diverse per i singoli rotoli): sono PSI 1090, P.Oxy. 2373 e appunto P.Oxy. 2404 + P.Laur. III/278, che è il testimone più ricco di interventi, rivelatori di un lavoro sistematico.13 P.Laur. ΙII/278 restituisce una sola colonna mutila da tre lati e piuttosto danneggiata per abrasioni e distacco di fibre. Delle due colonne di P.Oxy. 2404, la prima conserva i margini superiore e inferiore e quindi assicura il numero di 28 righe per colonna, ma è mutila sia a sinistra che a destra (rimane grosso modo una metà del rigo), mentre la seconda è conservata quasi per intero (manca solo un rigo all’inizio, ma gli altri sono completi) e consente una valutazione ben fondata. Analizzeremo dunque in particolare questa colonna. Osserviamo che la revisione non si è limitata a riparare piccole sviste con interventi graficamente il più possibile “mascherati” per non svilire il prodotto dal punto di vista estetico (cfr. sopra quanto osservato a proposito del papiro di Posidippo), bensì ha prodotto anche emendamenti ben visibili e fruibili, per quanto scritti con cura e realizzati senza sciatteria, il cui scopo appare quello di migliorare il testo in se stesso e costituirlo in modo che sia letto secondo gli intendimenti del correttore o dei correttori. Lo scriba si è limitato a dotare il testo di paragraphoi, mentre una punteggiatura piuttosto abbondante è stata aggiunta da una mano successiva, almeno

 11 Ed. pr. di P.Oxy. 2404: Turner 1957, cfr. Turner 1980, PI. VIII e p. 212 (= Turner 1984, Tav. VIII e p. 224); ed. pr. di P.Laur. inv. III/278: Messeri Savorelli/Pintaudi 1997, 172–174; cfr. Neri 2003, 511–514; Esposito 2004, 3–4; Colomo 2008, passim. 12 Cfr. da ultimo Colomo 2008, con la storia della questione e la bibliografia precedente (per la datazione vedi p. 14); cfr. anche Neri 2003, 499–517. 13 Colomo 2008, 6–7.

  Un papiro di Eschine con correzioni (P.Oxy. 2404) a quanto pare.14 La maggior parte è costituita da punti in alto, collocati un po’ più in alto delle lettere, che erano già scritte, curando che non fossero sopra una lettera bensì nello stretto spazio fra quella finale della parola precedente e quella iniziale della parola seguente. Si vede anche un punto in basso a col. I 17. Il sistema si può descrivere come segue: il punto in alto in combinazione con la paragraphos segna la fine di un periodo, il punto in alto semplice distingue i cola del periodo, il punto in basso corrisponde a una pausa più debole.15 Se prendiamo la seconda colonna, che comprende la porzione di testo (abbastanza piccola: le colonne sono piuttosto strette e non alte) da § 52.7 τριάκοντα a § 53.7 ὥστε (ediz. Dilts 1997, 212–213), vi troviamo sei punti in alto, che articolano correttamente tutti i cola del passo (rr. II 5, 8, 12, 15, 16, 22 = rr. 33, 36, 40, 43, 44, 50 nella numerazione continua dell’ed. pr.); due di questi punti inoltre sono in combinazione con la paragraphos (rr. 8 = 36 e 22 = 50) e segnano precisamente le due fini di periodo che si incontrano in queste righe (per intenderci diremo che corrispondono ai due punti fermi delle edizioni moderne). Nella prima colonna, dove la mancanza del margine sinistro impedisce di sapere se c’erano paragraphoi e i righi sono mutili, si vedono due punti in alto ai rr. 14 e 22 e il già citato punto in basso al r. 17. Un lavoro dunque quanto meno apprezzabile per impegno e risultato nell’evidenziare l’articolazione sintattica e retorica del testo, e il pensiero corre al ruolo dell’interpunzione nell’esegesi filologica alessandrina (anche senza evocare il complicato e idiosincratico sistema escogitato da Nicanore, ma piuttosto magari quello più semplice e diffuso delle tre stigmai di Dionisio Trace).16 A partire dall’ed. pr. di P.Oxy. 2404, tutti gli interventi di revisione e correzione sono stati attribuiti a un’unica seconda mano, ma in uno studio recente D. Colomo ritiene di individuare tre mani successive a quella dello scriba.17 Non mi addentro in questo problema, per il quale non ritengo di avere sufficienti competenze e che richiederebbe anche un’ulteriore autopsia dell’originale. Mi limiterò a una descrizione degli interventi che si riscontrano in particolare nella col. II di P.Oxy. 2404. Al r. 6 (= 34 nella numerazione continua) la prima mano scrive ἐν τοῖς διονυσίοις e il revisore ripristina il corretto ἐν διονύσου mettendo quattro punti al di sopra delle lettere del τοις da eliminare e semplicemente scrivendo ου sopra

 14 Cfr. Turner 1957, 130; Turner 1980, 212 (= 1984, 224); Colomo 2008, 15–16. Sui segni di punteggiatura nei papiri cfr. Turner/Parsons 1987, 9–10; Turner 1980, 92–93 (= 1984, 111–112). 15 Colomo 2008, 15–16. 16 Cfr. Colomo 2008, 15–22; Montana 2009 con bibliografìa; Dion. Tr., Gramm § 4. 17 Colomo 2008, 24–27, cfr. Neri 2003, 511–514.

Un papiro di Eschine con correzioni (P.Oxy. 2404)  

ιοις, senza cancellare materialmente queste ultime lettere ma facendo ben attenzione a collocare ου esattamente sopra ιο, per evitare possibili fraintendimenti (tipo σιου) e indicare bene la corretta lettura σου, evidentemente contando sul fatto che il residuale ις non facesse problemi. Al r. 21 (= 49) λείαν è stato corretto in λίαν cancellando ε con due trattini obliqui sopra la lettera; al r. II 26 (= 54) la prima mano scrive ὅπως e il revisore ripristina il corretto οὕτως cancellando l’erroneo π con un tratto obliquo sopra e scrivendo υτ esattamente sopra π.18 Più complessa e assai più intrigante la situazione ai rr. II 16–20 (44–48 nella numerazione continua dell’ed. pr.). Questo il testo del passo nei codici e nelle edizioni moderne (Blass 1908,19 Adams 1919, Leone 1977, Dilts 1997): ἀλλ᾿ ἐκεῖνο φοβούμενος, μή μοι παρ’ ὑμῶν ἀπαντήσῃ τὸ δοκεῖν ἀληθῆ μὲν λέγειν, ἀρχαῖα δὲ καὶ λίαν ὁμολογούμενα.

Prima un piccolo particolare. I codici medievali recano concordi δοκεῖν μὲν ἀληθῆ λέγειν, ma C.G. Cobet aveva proposto la trasposizione δοκεῖν ἀληθῆ μὲν λέγειν, giustamente accolta da Blass 190820 e nelle edizioni successive.21 Il nostro papiro conferma la trasposizione: la lezione corretta ἀληθῆ μέν è nel testo base e non viene toccata dal correttore. Ma per il resto la situazione testuale del passo è abbastanza complicata: né il testo base né quello ottenuto dal revisore corrispondono a quello della tradizione medievale. Sembra che il testo base fosse: ἀλλ᾿ ἐκεῖνο φοβούμενος, μή μοι παρ᾿ ὑμῶν ἀπαντήσῃ τι τοιοῦτον καὶ τὸ δοκεῖν ἀληθῆ μὲν λέγειν κτλ.

Turner commenta: «This first reading appears to mean ‘lest I be greeted on your part by some such thing as the impression of telling the truth but...’, as if τι τοιοῦτον οἷον or ὡς καί stood in the Greek».22 In seguito un τις è stato aggiunto

 18 Secondo Neri 2003, 512, le due cancellazioni ai rr. 21 e 26 sarebbero dovute alla prima mano, quella dello scriba, mentre il correttore avrebbe poi aggiunto le lettere giuste al r. 26; inoltre, a col. I 28 lo scriba avrebbe cercato di correggere un erroneo ε con un risultato pasticciato, che avrebbe indotto il correttore ad aggiungere il corretto omicron nell’interlineo; cfr. Colomo 2008, 26. 19 Nessuna variazione in Blass/Schindel 1978. 20 Confermato in Blass/Schindel 1978, XXIX. 21 Solo nell’ediz. Martin/de Budé 1928 è rimasto μὲν ἀληθῆ. 22 Cfr. Turner 1957, 132, ad loc.; secondo Colomo 2008, 24, qui «è stato operato un intervento di correzione attraverso la collazione di un altro esemplare diverso dall’antigrafo, ad opera di due mani diverse, la terza e la quarta».

  Un papiro di Eschine con correzioni (P.Oxy. 2404) supra lineam, disponendolo con attenzione fra μή e μοι (r. 16 = 44); τι τοιοῦτον (r. 18 = 46) è stato eliminato con due tratti orizzontali di penna (uno tracciato sopra le due parole, ora sbiadito, e un altro ben visibile supra lineam) e accanto a destra (nell’intercolumnio) in sostituzione è stato scritto θορ.[, comunemente integrato θόρυ[βος. Nel rigo successivo (r. 19 = 47) è stato lasciato il καί, mentre τὸ δοκεῖν è stato cancellato con un tratto orizzontale sulle lettere (doppio su οδο, probabilmente un primo tratto più breve, poi ripassato con il tratto lungo sull’intera sequenza) ed è stato aggiunto δόξω, scritto più in piccolo ma non nel margine (come prima θόρυ[βος) bensì in uno spazio rimasto bianco nella parte finale del rigo e nel quale le quattro lettere di modulo ridotto entrano a prezzo solo di una minima sporgenza a destra (vedi sotto). Il testo che risulta dall’insieme degli interventi è seguente: ἀλλ᾿ ἐκεῖνο φοβούμενος, μή τις μοι παρ᾿ ὑμῶν ἀπαντήσῃ [τι τοιοῦτον] θόρυβος καὶ [τὸ δοκεῖν] δόξω ἀληθῆ μὲν λέγειν κτλ.

La cosa più semplice è pensare che il primo fosse il testo base scritto dal copista23 e il secondo, risultante dall’insieme delle correzioni, quello voluto dal revisore o dai revisori. Rimane un dubbio a proposito della correzione al r. 19 (= 47): perché il copista aveva lasciato uno spazio bianco dopo τὸ δοκεῖν alla fine del r. 19 = 47, spazio che è stato utilizzato per aggiungere δόξω? È vero che c’è qualche irregolarità nell’incolonnamento a destra, ma qui lo spazio lasciato vuoto è un po’ troppo (corrisponde a circa tre lettere del modulo normale). Neri ritiene possibile che lo scriba stesso abbia lasciato lo spazio e abbia poi realizzato la correzione (ma se questo è avvenuto in scribendo, perché scrivere con un modulo più piccolo e compresso?) oppure che lo abbia deliberatamente lasciato per il correttore, che lo avrebbe riempito in un secondo momento.24 Colomo invece pensa che δόξω sia da attribuire ad una quarta mano, diversa da quella che ha scritto τις interlineare e θόρυ[βος in margine.25 In ogni caso, nessuno dei due assetti testuali corrisponde a quello tramandato dai codici. Quando pubblicò il papiro, Turner scriveva a proposito di questo passo: “at a place where the text had not even been suspected, there are two alternative readings, both superior to that of the manuscript tradition, no doubt both deriving from different ancient editions”.26 Contro questa opinione si è  23 Con un errore/omissione dopo τοιοῦτον, come ipotizzato da Turner, loc. cit. alla n. 22? Ma cfr. Merkelbach 1959. 24 Neri 2003, 512. 25 Colomo 2008, 25. 26 Turner 1950, 130.

Un papiro di Eschine con correzioni (P.Oxy. 2404)  

pronunciato Merkelbach, che ha difeso invece la lezione dei codici, considerando inferiori entrambe quelle del papiro e cercando di spiegarne la genesi di fatto congetturale.27 Questa opinione è stata evidentemente condivisa dagli editori di Eschine, dato che in nessuna edizione posteriore alla pubblicazione del papiro è stata messa in discussione la lezione dei codici (si vedano Leone 1977, Dilts 1997 ad loc.).28 Tuttavia, il valore in sé della(e) lezione(i) di P.Oxy. 2404, la sua posizione e quella in generale dei papiri nella tradizione manoscritta di Eschine non è il problema che ci interessa in questo momento.29 Le tecniche di cancellazione utilizzate nel papiro sono i punti al di sopra della lettera, il tratto obliquo (singolo o doppio) sopra la singola lettera e, per sequenze più lunghe, il tratto di penna nell’interlineo al di sopra delle oppure sulle lettere da eliminare (o combinati insieme).30 Abbiamo inoltre rilevato il diffuso metodo di correzione che consiste nel semplice scrivere le lettere giuste al di sopra di quelle sbagliate, che risultano eliminate per così dire automaticamente e senza bisogno di indicazioni materiali. Sempre a scopo correttivo si trova l’aggiunta di parole nell’interlineo o in margine. Insomma, un frammento che pare una vera antologia delle tecniche e dei metodi disponibili per emendare e migliorare un testo. Il fatto importante per il nostro discorso è che questo esemplare dell’orazione di Eschine sia stato sottoposto a una revisione approfondita e sistematica, volta a individuarne la struttura distinguendo cola e periodi, a correggere errori di copiatura in modo chiaro e fruibile dal lettore, a emendare il testo dove fosse ritenuto insoddisfacente. L’opinione avanzata da Turner nella ed. pr., che il lavoro sia stato effettuato per mezzo della collazione con un secondo esemplare, appare del tutto plausibile, oserei dire da accettare senz’altro.31 L’analisi e lo studio di testimoni del tipo di P.Oxy. 2404 mi pare una via maestra per fare luce, entro il quadro evocato sopra, su alcuni aspetti tecnici

 27 Merkelbach 1959; sembra meno convinto della sua prima idea Turner 1980, 212 (= 1984, 224); cfr. Colomo 2008, 30. 28 In Blass/Schindel 1978, XXIX, per questo passo in P.Oxy. 2004 stranamente si menziona solo la lezione ἀληθῆ μέν e si tace del resto. 29 Sul problema, oltre alla introduzione di Dilts 1997, cfr. Monaco 2000 (per il nostro papiro pp. 38 e 52, dove però si parla solo della lezione ἀληθῆ μέν); Colomo 2008, 30 sgg. 30 Neri 2003, 512; Colomo 2008, 24–25. 31 Turner 1957, 130: «The second hand not only revised the text for errors but collated its readings with an exemplar different from that from which it was copied», cfr. Colomo 2008, 24 (cit. sopra n. 22); che tutto sia avvenuto solo per congettura mi sembra altamente inverosimile, malgrado Merkelbach 1959. Sulle problematiche connesse cfr. innanzi tutto Turner 1980, 92–93 (= 111–112).

  Un papiro di Eschine con correzioni (P.Oxy. 2404) della filologia antica, soprattutto quelli riguardanti i modi concreti di lavorare su un esemplare di un’opera quando si volesse operare con lo scopo di produrre una ekdosis, compresa la dibattuta questione del confronto fra copie. Voglio di nuovo precisare chiaramente che adduco questo tipo di testimoni e le loro caratteristiche non a causa dell’uso e della destinazione erudita che essi possono rivelare32 o perché addirittura pensi che si possa trattare di una ekdosis in senso proprio, bensì per tornare a sottolineare l’importanza del modello artigianale delle tecniche librarie nel formarsi di una prassi filologica intesa a migliorare ed emendare testi ritenuti insoddisfacenti per i guasti sofferti. Nel trattamento dei testi, fra il punto di vista librario e quello squistamente filologico esisteva una ben evidente zona di sovrapposizione, quella per così dire materiale: insieme a tecniche e metodi ben collaudati, l’attività del grammatico ha ereditato anche elementi terminologici essenziali, se è vero che la filologia si definì diorthosis e il filologo diorthotès, come accadeva per l’opera e la persona che nello scriptorium o privatamente provvedeva a correggere le nuove copie confrontandole con l’antigrafo e talvolta con altri esemplari.

 32 Aspetto ben evidenziato negli studi: cfr. Neri 2003 e Colomo 2008.

 Iperide e Cratino in due Glosse dell’Αntiatticista L’indicazione di fonte Ὑπερείδης ὑπὲρ Κρατίνου in due glosse dell’Antiatticista ha creato dei problemi, che forse vale la pena riconsiderare. 77.27 Bk. Ἀ κ μ ή: ἀντὶ τοῦ ἔτι. Ὑπερείδης (fr. 116 J.) ὑπὲρ Κρατίνου. 100.32 Bk. Κ ά θ ῃ: ἀντὶ τοῦ κάθησαι. Ὑπερείδης (fr. 115 J.) ὑπὲρ Κρατίνου.

La glossa κάθῃ di 100.32 ha un preciso parallelo in Oros A 57 Alpers (da Zonaras 1168–1169),1 una glossa assai più ricca, che conserva per esteso (con altro materiale) le citazioni di un passo di Iperide (fr. 115 Jensen) e di un passo di Cratino (fr. 309 PCG), nei quali è usata la forma κάθῃ. Per la citazione di Iperide, la glossa di Oros dà solo il nome Ὑπερίδης, senza aggiungere il titolo dell’opera (come del resto accade anche per Cratino e per tutte le altre fonti citate nella glossa).2 Il problema è dunque se l’indicazione di fonte dell’Antiatticista conservi la citazione soltanto di Iperide con il titolo dell’opera, un altrimenti ignoto Ὑπὲρ Κρατίνου omesso nella glossa di Oros, oppure avesse in qualche modo i nomi di due autori, Iperide e Cratino. Il parallelo fornito da Oros invita a preferire la seconda possibilità. Per vero dire, W. Luppe propose di leggere Ὑπερείδης καὶ Κρατῖνος senza utilizzare la glossa di Oros, ma unicamente sulla base del fatto che il fr. 34 PCG (Hesych. β 969 + Suid. β 459) fa ritenere che la forma κάθῃ si trovasse in Cratino:3 naturalmente, Oros A 57 (= Zonaras 1168–1169), che presenta insieme citazioni di Iperide e Cratino per la forma κάθῃ, sembra confermare la correzione di Luppe. Tuttavia lo stesso Alpers, nell’edizione di Oros, ha contrapposto due argomenti: in primo luogo, il parallelo con Oros–Zonaras vale solo per la glossa di 100.32 κάθῃ e non per la glossa di 77.27 ἀκμήν, dove si trova la stessa formula Ὑπερείδης ὑπὲρ Κρατίνου; in secondo luogo, nell’Antiatticista non si usa il καὶ per unire due fonti bensì comunemente una giustapposizione asindetica. La prima obiezione obbliga naturalmente a correggere Ὑπερείδης καὶ Κρατῖνος anche a 77.27 (così Luppe, che confronta il fr. 408 PCG, su cui torneremo), la seconda rimane senza repliche: l’insieme delle due porta giustamente Alpers a ritenere ancora

 1 Alpers 1981, 178–179, cfr. p. 108 e n. 37. 2 Nell’ordine: Aristofane, Cratino, Aristofane, Iperide, Ferecrate, Archippo, Cratino, Eschine, Platone. 3 Luppe 1967, 406. https://doi.org/10.1515/9783110772371-061

  Iperide e Cratino in due Glosse dell’Antiatticista dubbia la congettura di Luppe, malgrado gli elementi a suo favore;4 e anche Kassel–Austin ritengono l’intervento troppo audace.5 Un’altra possibilità è stata proposta più recentemente da R. Tosi, sulla base della considerazione che il parallelo di Antiatticista 100.32 κάθῃ con Oros A 57 (= Zonaras 1168–1169) è troppo significativo per non essere tenuto in alcuna considerazione. Tosi propone di pensare che l’Antiatticista a 100.32 avesse originariamente la citazione dell’opera di Iperide (spesso nelle pur brevi glosse dell’Antiatticista al nome dell’autore segue il titolo dell’opera) seguita dal nome di Cratino, dunque: Ὑπερείδης ὑπὲρ < . . . >, Κρατῖνος, «e che la corruzione successiva fosse facilitata proprio dall’esistenza di un’orazione di Iperide Ὑπὲρ Κρατίνου»; addirittura, a 100.32 avrebbe potuto esserci Ὑπερείδης ὑπὲρ Κρατίνου, Κρατῖνος, poi ridottosi per aplografia. Di conseguenza Ὑπερείδης ὑπὲρ Κρατίνου potrebbe essere conservato a 77.27. La corruzione sarebbe limitata a 100.32, dove l’Antiatticista citava l’orazione di Iperide e Cratino (cfr. Oros– Zonaras), mentre a 77.27 il testo potrebbe essere sano e l’Antiatticista avrebbe citato solo l’orazione di Iperide ὑπὲρ Κρατίνου.6 Al momento, questa mi pareva la soluzione più valida: ma forse si può avanzarne un’altra che — lo premetto senz’altro — non voglio presentare come migliore, bensì soltanto come un’altra possibilità da prendere in considerazione. Diciamo che la posta in gioco è l’esistenza o meno dell’orazione di Iperide ὑπὲρ Κρατίνου (su cui non abbiamo altre notizie) e, dall’altra parte, che tipo di testimonianza su Cratino si può avere dalle due glosse dell’Antiatticista. Anch’io ritengo che il parallelo di Antiatticista 100.32 con Oros A 57 non possa essere trascurato; ma obblighi a pensare che qualche problema testuale deve essersi prodotto almeno in Antiatticista 100.32. L’idea che entrambi i passi dell’Antiatticista siano corrotti e si siano corrotti nello stesso modo ha certo una intrinseca debolezza e postulare un solo errore in una sola delle due glosse è senz’altro più economico. Tuttavia forse la debolezza è minore se si pensa a una corruzione facile, tale da potersi produrre anche indipendentemente, magari favorita la seconda volta dal fatto di essersi già prodotta la prima a distanza non grande. Considerando come sia frequente l’uso di compendi nei manoscritti di opere lessicografiche, scoliastiche e simili, e come spesso capiti che proprio i nomi degli autori siano compendiati, forse si potrebbe supporre che il nome di Iperide fosse scritto abbreviato e per semplice dittografia sia nato il secondo  4 Alpers 1981, 108 n. 37. 5 PCG IV p. 138, ad Cratin. fr. 34; non si dice nulla ad fr. 309, dove sarebbe utile avere almeno un rimando al fr. 34; cfr. ad fr. 408 (su cui v. sotto). 6 Tosi 1994, 165–166.

Iperide e Cratino in due Glosse dell’Antiatticista  

ὑπὲρ: ovviamente da questo sarà conseguito ὑπὲρ Κρατίνου (ancor più facile se anche il nome Κρατῖνος era abbreviato, per lo meno nella desinenza). In questo caso, la lezione originale sarebbe stata sia a 77.27 che a 100.32 la più semplice che ci si aspetterebbe e del tutto conforme all’uso dell’Antiatticista, vale a dire Ὑπερείδης, Κρατῖνος. Rimane un problema. Per 100.32 κάθῃ troviamo in Oros A 57 un frammento che ci testimonia esplicitamente l’uso di questa forma da parte di Cratino, uso che trova conferma anche nel fr. 34 PCG (sul quale si era basato Luppe): questo dato ha un peso considerevole e sarebbe abbastanza confortante avere qualcosa di analogo per 77.27 ἀκμήν. Come dicevamo sopra, Luppe invocava il fr. 408 PCG, così presentato da Kassel–Austin: Lex. cod. Vat. 12 (s. XV) ap. Reitz. Ind. lect. Rost. 1892/93 p. 4 ἀ κ μ ή ν· ἐπίρρημα. Αἰσχύλος (fr. 339a R.) καὶ Μένανδρος (fr. 715a Koe.–Th.) ... Κρατῖνος μέντοι τῷ (τὸ cod.) ἀκμὴν ἐπιρρηματικῶς μέν, ἀντὶ τοῦ ἀκμαίως ἐχρήσατο Antiatt. p. 77,26 [sic!, leg. 77,27] Ὑπερείδης ὑπὲρ Κρατίνου in Ὑπερείδης καὶ Κρατῖνος mutabat Luppe (vid. ad fr. 34), quamquam illic de ἀκμὴν ἀντὶ τοῦ ἔτι agitur, cf. Phryn. ecl. 93, Moer. p. 191,23 Bk. (U. Pohle, D. Sprache d. Redners Hyp. in ihren Bez. z. Koine, 1928, p. 47 sq.)

La differenza sottolineata da K.–A. riposa sul fatto che questo frammento testimonia l’uso in Cratino di ἀκμήν nel senso di ἀκμαίως, mentre nella glossa dell’Antiatticista si parla di ἀκμήν nel senso di ἔτι: usare questo frammento come sostegno alla correzione Ὑπερείδης, Κρατῖνος in Antiatt. 77.27 equivarrebbe dunque ad attribuire a Cratino l’uso di ἀκμὴν ἀντὶ τοῦ ἔτι invece che di ἀκμὴν ἀντὶ τοῦ ἀκμαίως. Non c’è dubbio che l’obiezione sussista, con una precisa puntualità che non può essere negata: ma forse si può provare a considerare le cose con una prospettiva diversa e più ampia. Alla glossa ἀκμήν del Lex. cod. Vat. 12 si affiancano diversi paralleli lessicografici, più tardi di quelli addotti da Kassel–Austin ma utili ad altro scopo. I due paralleli più antichi sono offerti da Frinico, Ecl. 93 Fischer, e da Moeris, p. 73 Pierson–Koch (= 79 Pierson). Oltre a queste testimonianze, tuttavia, bisogna prendere in considerazione almeno: Esichio α 2446, 2447, 2449, ε 6608; Σb 54.24 Bachm.; Fozio α 769 Theod.; Suid. α 904 e 906; e infine anche: Et. Gen. α 337, Et. Sym. α 403, Et. M. α 664 Lass.–Liv. (= 49,13 G.). In questi materiali lessicografici è il sostantivo ἀκμή ad avere maggiore spazio nella trattazione accanto a ἀκμήν, e si ha la chiara sensazione di avere di fronte i resti di una trattazione lessicale articolata che comprendeva osservazioni combinate su ἀκμή e ἀκμήν. Tutto ciò significa che l’avverbio ἀκμήν, con ogni probabilità in connessione con il sostantivo ἀκμή, almeno dall’età dei lessici

  Iperide e Cratino in due Glosse dell’Antiatticista atticisti di Frinico, di Moeris e dell’Antiatticista, era stato oggetto di indagini lessicali e semantiche confluite in una glossa, generatrice poi di una filiazione lessicografica che non risulterà inconsueta a chi abbia esperienza in questo campo. Se consideriamo la triade costituita da Antiatticista, Frinico e Moeris, dobbiamo ricordare alcune cose: sappiamo che l’Antiatticista è stato una delle fonti di Frinico ma in una versione più ampia di quella pervenutaci, che è sicuramente un’epitome; Frinico, Moeris e il Philetairos hanno avuto in comune una fonte, che non possiamo individuare.7 Mi pare possibile ricavarne che ci sono buone probabilità che la glossa dell’Antiatticista come oggi la leggiamo (nell’epitome pervenutaci) sia l’abbreviazione di una glossa più ampia, nella quale si davano ben più materiali intorno all’uso e ai significati di ἀκμή e ἀκμήν (possiamo pensare, se vogliamo, al tipo di quella che Oros ci ha conservato per κάθῃ), materiali che per altre strade della lessicografia sembrano essersi conservati in qualche modo e chissà con quali modificazioni. Allora, il fatto che nell’epitomata glossa della nostra versione dell’Antiatticista rimanga solo notizia del significato di ἀκμὴν ἀντὶ τοῦ ἔτι non dovrà essere ritenuto un impedimento a pensare che in origine l’Antiatticista, considerando diversi significati e usi di ἀκμήν, citasse Iperide e Cratino anche a 77.27 come a 100.32. Purtroppo, per 77.27 ci manca un parallelo chiarificatore come quello di Oros A 57 per 100.32: ma se il contrasto fra il fr. 34 PCG di Cratino e Antiatt. 77.27 è solo apparente, frutto di un’epitomazione violenta e forse maldestra, allora nulla osta che anche a 77.27 la lezione corretta fosse Ὑπερείδης, Κρατῖνος. Così avremmo definitivamente eliminato la supposta orazione di Iperide ὑπὲρ Κρατίνου e, per di più, non sapremmo bene cosa Antiatt. 77.27 ci dice a proposito di Cratino: o l’epitomazione è stata tale che questa glossa non ci dice praticamente nulla su di lui, cioè è solo possibile che semplicemente confermi l’uso di ἀκμὴν ἀντὶ τοῦ ἀκμαίως già testimoniato dal fr. 34 PCG, oppure possiamo azzardarci a pensare che Antiatt. 77.27 ci dica che Cratino usava ἀκμὴν anche ἀντὶ τοῦ ἔτι, oltre che ἀντὶ τοῦ ἀκμαίως. Ma a questo punto mi pare di aver azzardato anche troppo.

 7 Su questi problemi cfr. Latte 1915; Fischer 1974, 39–44.

 Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries” There is no doubt that the poets of the Archaic and Classical periods were not only the first to be studied, but also the most constantly and abundantly edited and commented on by the philologists of the Alexandrian and Imperial periods, because they represent the true, essential and indispensable basis of the traditional paideia. The works of the great authors of the past were held to be treasures of civilisation which characterised Greek cultural identity — a prized heritage that had to be understood and preserved through suitable means and methods. In the entire period of Alexandrian philology, starting from Zenodotus, the grammarians were assiduously and deeply involved in the study of Homer and Hesiod, of all the various genres of archaic lyric poetry, and of tragic and comic theatre. Later — certainly in the age of Aristarchus, if not before — prose writers also made their appearance in the work of the grammarians.1 However, a number of facts show that grammarians and scholars began to work on “contemporary” authors too very early on, even in the Hellenistic Age: by the Augustan period the poets of the Hellenistic Age already represented a common field of philological and exegetical work. Thus the scholia to Theocritus, Apollonius of Rhodes, Aratus, Nicander, and even Callimachus, whilst more scarce, are the outcome of an exegetical history which began certainly later than the Homeric or Pindaric, but not much later, and in any case still well within the Hellenistic period. It did not take long before the expertise gained from the study of works of the past, namely the great Pre-Aristotelian literary tradition, would come to serve as a model, and be transferred to “contemporary” matters, namely to the poetry of the early Alexandrian period. Up until not too long ago, the principal facts which we had at our disposal concerned Theon’s activity during the Augustan age: beyond Homer, Sophocles, and Pindar, we have evidence that he wrote commentaries on Theocritus, Callimachus, Apollonius of Rhodes, Lycophron, and Nicander, thus making for a nexus of activity in which the Hellenistic poets (and a good number of them, not just one or two) play a very important role.2 So there emerged a tendency to suppose that only beginning in the Augustan period did the great poets of the 3rd century BC become an object of careful philological study, entering into the

 1 An overview with bibliography: Pfeiffer 1968; Montanari 1993 (= this volume, ch. 1); 1994a. 2 Guhl 1969; Montanari 1993, 279–280. https://doi.org/10.1515/9783110772371-062

  Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries” number of πραττόμενοι authors — which is to say, of authors “dealt with” by the grammarians. I might observe to begin with that certain facts here were overlooked. The first is constituted by the edition of the Bucolic poets (Theocritus, Moschus and Bion) prepared by the grammarian Artemidorus, father of Theon: I find it difficult to doubt that this work was a basis for Theon’s own exegetical work, and that it indicates a taste for Hellenistic poetry which was sufficiently pronounced to produce so exquisitely Alexandrian a result as an edition, followed a couple of years later by a substantial commentary.3 The second fact regards the activity of Asklepiades of Myrlea: some of his critical comments on passages by Theocritus have been cited, on the basis of which one might imagine at least some limited philological activity on a selection of poems, if one hesitates to go so far as to suppose the existence of an actual commentary — too confident a conclusion, perhaps.4 While these considerations concern Theocritus and bucolic poetry, we can add at least one more. We have a fragment from Artemidorus himself which indubitably concerns an element of Callimachean exegesis, and more specifically, a lexis by the Aitia, passed down in the Etymologicum Genuinum:5 it thus appears that he in some way foreshadowed the more detailed work of his son Theon even with regard to Callimachus. Finally, we might recall a few other points, however uncertain these might be. In the scholia to Nicander, a grammarian by the name of Demetrius Chlorus is sometimes cited, the author of a Nicandrean hypomnema: the common view is that he lived at least a generation before Didymus, but the dating is hypothetical and so we cannot take much account of his activity.6 An entry in the Etymologicum Genuinum speaks of a work (a commentary?) by one “Edilus” on Callimachus’ Epigrams: it is pure conjecture that this Edilus might be Edilus of Samus, a poet of epigrams and even a contemporary of Callimachus’, rather than an otherwise

 3 On Artemidorus and the beginnings of Theocritean philology, see Wendel 1921; Maehler, 1994, 97–99; Blumenthal 1934, 2022; Wentzel 1985. 4 See the bibliography cited in the previous note; see also Wentzel 1986; Müller 1903; Adler 1914; Pfeiffer 1968, 272–273; Montanari 1993, 277–278 (= this volume, ch. 1); Slater 1972; Rispoli 1988, 170–204. 5 Etym. Gen. α 1175 = Etym. M. α 1795 (142,43 G.) ἀρισκυδής (vol. II, 198–199 Lasserre-Livadaras) = Callimachus fr. 55, v. Pfeiffer ad loc. + addenda in vol. II p. 113; see Wentzel 1985, 1332. 6 Cf. Susemihl 1982, II, 20; Kroll 1936.

Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries”  

unknown grammarian of a later age; and so we can only hope for the emergence of new information which might settle our doubts on this score.7 Leaving this aside, we have in any case noted a number of solid facts which would shift grammarians’ active interest in Hellenistic poetry at least a generation back with respect to Theon: that does not seem much, but in the general portrait that we will attempt to paint, this point will be seen to acquire greater significance. The situation has changed considerably over the past twenty years, above all thanks to the publication of two papyri of essential importance in this context. In 1975, F. Lasserre published the editio princeps of P. Louvre inv. 7733 verso; two years later, the text was thoroughly reviewed by P. Parsons and then presented anew in the Suppl. Hell.8 The fragment, dated to the 2nd century, presents on its recto a treatise on optics9 and on the verso a six-verse epigram (a puzzle, the answer to which is “an oyster”), followed by a continuous (and perhaps more or less complete) commentary of over fifty lines. The epigram is clearly Alexandrian, but the name of the author remains unknown.10 The commentator quotes Theodorides, active in the second half of the 3rd century BC, and the text itself was copied in the 2nd century BC: therefore, the author must have been writing in the late 3rd century or at the beginning of the 2nd century BC.11 Approximately five of the epigram’s six verses have been reconstructed, and it is unlikely that the commentary is missing many lines; the surviving ones contain major lacunae, but what remains permits the inference of a number of letters. Given the brevity of the composition, it was easy for the scribe first to reproduce the full work which is the subject of the commentary, along with its title “ὄστρειον”, which is repeated before the beginning of the commentary. Although the poetic text is conveniently placed at the reader’s disposal, the commentary does not for this reason forego lengthy lemmas: vv. 1, 4, 5, and 6 of the epigram (respectively l. 9, l. 24, l. 29, l. 40) undoubtedly occur as lemmas, but it is possible that the composition was not wholly reproduced by the exegete in the lemmas (the presence of vv. 2–3 remains dubious, although it seems clear  7 Etym. Gen. α 551 = Etym. M. α 960 (72,12 G.) ἀλυτάρχης (vol. I, 342–343 Lasserre-Livadaras) = Callimachus T 45 Pfeiffer (vol. II p. cii): Ἡδύλος δὲ εἰς τὰ Ἐπιγράμματα Καλλιμάχου; see Parsons 1977, 5. 8 Lasserre 1975; Marcovich 1976; Parsons 1977a; Suppl. Hell. 983–984, pp. 497–500; Del Fabbro 1979, 72–75. 9 Pack2 2579; the text of the verso was flagged in Pack2 2911 as unidentified prose. 10 The name of Philitas has been broached here (cf. bibliogr. no. 8), but without any real basis. 11 According to Parsons 1977a, 12.

  Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries” that the first section of the commentary, ll. 9–23, concerns vv. 1–3 in their entirety: it is possible that the lemma was constituted only by v. 1). The problem addressed at the beginning is that of identifying the place where Memnon’s sepulchre lies: not Abydos in the Egyptian Thebaid, where there are no oysters, but Abydos in the Hellespont, which is famous for their production. The second section (ll. 25–29) explains ἀγροτέρη as an epithet of Artemis, and hence here equivalent to “moon”; it then seems to draw a connection between this and the idea that oysters increase in size when the moon waxes. There follows the commentary on v. 5, which focuses on the meaning of the adjective ἀφέψαλος, on the basis of its component φέψαλος (ll. 30–39): here a comedy is quoted, otherwise unknown, by Diphilus, the “Παραλυόμενος”; three trimeters of this comedy were probably included (fr. 59 PCG), along with a fragment of Sophocles (also otherwise unknown, fr. 966A Radt). The last part of the commentary (ll. 41 et seq.) concerned the last phrase of the epigram, from the bucolic diaeresis of v. 5 to the end: v. 6 appears as a lemma to l. 40. The first question certainly concerned the sophisticated way of indicating the knife with which to open the oyster: the lover of Dōsṓ, a rare epithet of Aphrodite, is precisely Ares, that is, the iron, that is, the knife. Here appears the aforementioned citation from Theodorides (fr. 743 Suppl. Hell.), likely a parallel for the epithet Dōsṓ. The subsequent parts are too incomplete to make much of them. The commentary can be called brief in absolute terms, but relative to the dimension of the composition itself it appears noticeably long: the variety of the problems it addresses and of its exegetical contents, along with the use of parallel passages, defines the work as one of noteworthy erudition, written — as already noted — in the late 3rd or early 2nd century BC, about an epigram which certainly dates back to not too many decades before. In 1977, C. Meillier produced the editio princeps of some papyrus fragments from Lille, containing verses from the first part of the third book of Callimachus’ Aitia, the Victoria Berenices, interspersed with a commentary: shortly after, the text was published anew with a commentary by P. Parsons and ultimately presented in the Suppl Hell.12 The dating of P. Lille was in no way definite and even now oscillates between the end of the 3rd century and the first decades of the 2nd century BC.13 The Callimachus text is included in its entirety, and is interspersed at irregular intervals with lines of commentary, indented by three letters,  12 Meillier 1976; Parsons 1977; Suppl. Hell. 254–265 (new edition and reconstruction, with new bibliography). 13 See Meillier 1976; Parsons 1977, 4; Cavallo 1983, 53; Turner/Parsons 1987, nos. 74 and 75, pp. 124–127.

Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries”  

producing an immediate visual effect. The commentary mostly consists of glosses or paraphrases, and therefore amounts to a simple “translation” of the poetic text into contemporary Greek; occasionally, it also offers some information of historical or even linguistic nature, necessary for the comprehension of the text. Therefore, it is not a particularly erudite work: it is rather aimed at aiding comprehension in the best possible way, with basic information that seems to be geared towards inter legendum use, so to speak, interspersed in a text that can be read in its entirety. It dates back to a period almost identical to that of the commentary on the oyster epigram, but here we have an evidently less ambitious exegetical text; on the other hand, in this case we can confidently state that this “commented edition”14 of Callimachus was “published” at a minimum of one and a maximum of two generations after the death of the poet. In the commentary on the anonymous oyster epigram in the Louvre papyrus, and in the commentary on Callimachus in the Lille papyrus, we have two examples of the exegesis of Hellenistic poetry that date back to at least the first half of the 2nd century BC., if not shortly before. Therefore, they fall at the peak of Aristarchus’ career (since he died around 144 BC), perhaps towards the very end of the life of Aristophanes of Byzantium, at least a century earlier than the work of Artemidorus and of Asklepiades of Myrlea, recalled above. These fresh testimonies prompt further reflection and lead us to consider certain other elements which we have at our disposal from a different point of view and with a different orientation. Some years ago I studied a problematic Odyssey scholium, which contains a misunderstanding of a Callimachus quote. The scholium concerns Od. 2.136–7 and provides information about one of Aristarchus’ many attempts at athetesis based on the concept of περισσόν. Aristarchus believed v. 137 to be περισσός, and he proposed the athetesis (διὸ ἀθετεῖ); since this left v. 136 with a nominal phrase, which required the verb to be inferred, he presented a parallel which was both semantic and stylistic by quoting a passage from Callimachus (fr. 637) that expressed a similar concept through a nominal phrase. Aristarchus used a Callimachean parallel to support his own idea in the constitution of the Homeric text.15 Along with the large number of references within the Homeric poems, the use of parallel passages taken from other poets is notoriously frequent in the

 14 This is how I defined it Montanari 1976b, 147 (= this volume, ch. 57); cf. Del Fabbro 1979, 70–71. 15 Montanari 1979b, reprinted in a revised version in Montanari 1995a, 27–40. In the scholia to these verses, we find material from Didymus (Ludwich 1884–1885, I, 522), Aristonicus (Carnuth 1869, 22) and Nicanor (Carnuth 1875, 27–28).

  Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries” scholia, and it is one of the most impressive phenomena that can be traced back to Alexandrian erudition. Reflecting on this, I began to ask myself about the presence and the use of the works of the Alexandrian poets in scholia to the archaic and Classical poets, and I began with the examination of some cases in which Callimachus is cited in Homeric scholia. Homeric philology enjoys a privileged position, because it boasts the richest documentation that we have, and very often allows us to chronologically arrange the material in a satisfying way. The HMQR scholium ad Od. 4.1, dating back to Didymus,16 cites a reading by Zenodotus (Ζηνόδοτος δὲ γράφει ‘καιετάεσσαν’ ἀντὶ τοῦ καλαμινθώδη) and then comments: δοκεῖ δὲ καὶ Καλλίμαχος ἐντετυχηκέναι τῇ γραφῇ δι’ ὧν φησιν «ἵππους καιετάεντος ἀπ’ Εὐρώταο κομίσσαι» (fr. 639; cf. Erbse ad Il. 2.581). This case recalls a problem that Pfeiffer focused on in several detailed commentaries in his edition of Callimachus, and then revisited in a dense passage of his History of Classical Scholarship: namely, the existence of evidence of the possibility that Callimachus was familiar with Zenodotus’ Homeric text. This evidence is provided by a certain number of cases in which the Homeric text presupposed by Callimachus fits, in a unique way, with that chosen by Zenodotus.17 Given their chronological proximity, this fact could indeed be quite significant, and Pfeiffer’s cautiousness is understandable. He comments on the case of Il. 18.592 — where Zenodotus read Ἀριήδῃ for Ἀριάδνῃ, and Callimachus, alluding to the Homeric passage, used the form Ἀριήδης (fr. 67.13) — with these words: “This certainly is a most remarkable coincidence; but as Zenodotus constituted his text on earlier manuscripts that he found reliable, the same sources may have been accessible to Callimachus. Although it is possible or even very probable that he followed Zenodotus, the coincidence in this [scil. Il. 18.592] and about ten similar cases is not conclusive proof.” However, it seems to me that a detailed and well-documented study by A. Rengakos has succeeded in proving, at least with great probability, that Callimachus and Apollonius of Rhodes knew and used Zenodotus’ Homeric text. This is certainly important, because the specific problem involves the intricate question of the form of the Alexandrian editions — namely, what Zenodotus’ ekdosis looked like in physical terms — and also the question of whether or not he used different  16 Ludwich 1884–1885, I, 535: cf. Il. 2.581, which has a diplè periestigmene in Ven. A, v. Erbse ad loc. and Ludwich 1884–1885, I, 224. 17 Pfeiffer ad Callim. fr. 12.6 + addenda with references (along with fr. 633 and fr. 497 + addenda); Pfeiffer 1973, 229–230; cf. Erbse 1953, 179. A. Rengakos has recently returned to this subject in an excellent book (see below): Rengakos 1993, 49–87 for Apollonius of Rhodes and Callimachus in relation to Zenodotus’ Homeric text, and particularly pp. 79–87 for Zenodous and Callimachus (for the case of Od. 4.1 see pp. 20–21, 29, 85–86).

Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries”  

manuscripts and collated them.18 Whatever position one wishes to take here, it remains the case that the scholia sometimes inform us that the ancient philologists noticed and commented on coincidences between passages in Callimacus and in Zenodotus’ Homeric readings. So let us attempt to reason further on this simple and basic fact, whatever the final consequence may be of the fact that Callimacus knew and used precisely Zenodotus’ Homeric text. In short, we shall strive to apply this information from a different point of view. At first sight, the scholium to Il. 16.234a seems surprising: δυσχειμέρου: οἱ περὶ Ζηνόδοτον ‘πολυπίδακος’, διὰ τὸ Καλλιμάχου «κρηνέων τ’ Εὐρώπῃ μισγομένων ἑκατόν» (fr. 630). The material in all likelihood comes from Didymus, though the scholium itself belongs rather to the class of the scholia exegetica.19 Taken literally, this scholium seems to say that οἱ περὶ Ζηνόδοτον chose the reading πολυπίδακος on the basis of a passage in Callimachus (διὰ τὸ Καλλιμάχου): should we take this to refer to philologists who were “followers/pupils of Zenodotus” and who, in constructing a new Homeric text after Zenodotus and Callimachus, took account of the Callimachus parallel for the layout of this verse (meaning that they used the Callimachus passage as a parallel to uphold their idea of the Homeric text, much as Aristarchus does ad Od. 2.136–7)? This is, if I am correct, the opinion held by Erbse, according to whom the scholium “sagt m.E. aus, daß Schüler Zenodots ihre Konjektur mit dem kallimachischen Vers zu stützen suchten”.20 Bergk provided another explanation (one which was accepted by Pfeiffer): according to him, “Zenodotean grammarians” used the Callimachus verse to prove that πολυπίδακος was Zenodotus’ reading; which is to say that Zenodotean grammarians of later age, determined to reliably reconstruct

 18 Pfeiffer’s words are found in 1973, 139; Erbse’s reservations in 1953, 179; for Il. 18.592 cf. Rengakos 1993, 42–43, 85. On this specific subject, Rengakos concludes: “Beide Dichter, Kallimachos und Apollonios, haben Zenodots Text nich nur gekannt, sondern auch in ihren Dichtungen benutzt”: 1993, 170. In the same book, Rengakos offers some information and interesting ideas regarding the philological method of the Alexandrians, in relation to the much-debated problem of whether their readings were only conjectural or had some documented source in collated Homeric copies: I fully share the view that the two extreme positions (namely, that these readings were only conjectures or only from collated and documented sources) should both be rejected, as it is more likely that the Alexandrians both employed conjectural critiques and made use of the documentary material constituted by the collation of copies. Cf. also Rengakos 1994; Rengakos 1994a. 19 Erbse labels it “ex. (Did.?)”; cf. Ludwich 1884–1885, I, 405; van der Valk 1963–1964, II, 64, 233 n. 673. 20 Erbse 1953, 179.

  Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries” Zenodotus’ textual choices,21 called upon Callimachus as their witness. Therefore, πολυπίδακος was the reading favoured by Zenodotus, and not by his later “followers/scholars”. I hold this to be the correct view, perhaps with one small correction. The use of the periphrasis οἱ περὶ + acc. nominis proprii as a simple indication of the person has been proven,22 so it is likely that, in the scholium, οἱ περὶ Ζηνόδοτον simply means Zenodotus: we would then be dealing simply with a situation in which the source of the scholium, namely Didymus himself or one of his sources, claimed that Zenodotus’ reading was πολυπίδακος, calling on the above-cited verse from Callimachus as his witness. In any case, whoever the responsible party might be, Callimachus serves as a cache of clues for reconstructing Zenodotus’ Homeric text. This therefore agrees with the meaning that clearly emerged from the quoted scholium Od. 4.1: δοκεῖ δὲ καὶ Καλλίμαχος ἐντετυχηκέναι τῇ γραφῇ: it would seem that the two examples are similar precisely in the fact that both use Callimachus to maintain and corroborate information about one of Zenodotus’ Homeric readings. It is extremely likely, as we have seen, that both derive from Didymus’ work on the construction of Aristarchus’ text, and offer new elements that fit nicely within an already established framework. Aristarchus and his school often argued with Zenodotus’ textual choices, but sometimes they were not aware of the motives and the reasoning of their ancient predecessor; sometimes they were not even certain of what precisely Zenodotus’ reading of a given passage was, because they lacked adequate documentation. The problem of how much it was possible to know in a reliable way about Zenodotus’ Homeric philology has already been investigated,23 and perhaps the most relevant element here is that, in Aristarchus’ day, there was a grammarian who referred to Zenodotus’ teaching, Ptolemy Epithetes. A few fragments have been preserved from one of his works explicitly dedicated to reconstructing and collecting τὰς Ζηνοδότου γραφάς: in this work, he happened to maintain that Zenodotus’ real reading was not that attributed to him by Aristarchus and his followers, but another. The more instructive case is certainly found in the scholium to Il. 14.37b, but the scholium to Il. 2.111b is also very informative; in it, Didymus’ discussion concerns —

 21 Bergk 1915, 293: “potuerunt Zenodotei ad magistri lectionem probandam uti Callimachi versu, ut ostenderent hunc quoque Zenodoti lectionem secutum esse”; cf. Pfeiffer ad fr. 630, 426 (“de textu Zenodoti apud Callimachum…”); cf. Rengakos 1993, 42–43, 83–84. On the Zenodotan grammarians involved in reconstructing Zenodotus’ text in a later period, see below. 22 See Radt 1980; 1988; 1988a. 23 Nickau 1972; Nickau 1977; Montanari 1988, particularly the part on Ptolemy Epithetes.

Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries”  

among other things — the issue of what exactly Zenodotus’ true reading was: these two scholia respectively coincide with F3 and F1 of Ptolemy Epithetes.24 Given the existence of a serious problem in the reconstruction of Zenodotus’ opinion, it was evidently believed that even Callimachus — as a poet — on account of his imitations and allusions, could be used for this purpose, and at least some of the coincidences with Zenodotus cited in the scholia seem to be adduced precisely as witnesses to support one of Zenodotus’ readings. It is clearly not necessary for us to decide now if these can be considered valid evidence of the fact that Callimachus knew and used precisely Zenodotus’ Homeric texts, with all the above implications. Here, due importance must be given to the fact the the ancient philologists considered the Callimachean text useful for deriving clues regarding Zenodotus’ Homeric philology; this certainly occurred in the Aristarchan period, as we have just seen:25 a declaration like δοκεῖ δὲ καὶ Καλλίμαχος κτλ. (cf. above) is truly significant. Whoever engaged in such exquisitely philological reasoning must have studied the language and the poetic style of Callimachus deeply enough to value his Homeric zelos with sufficient subtlety, recognising its erudite orientations and sophisticated choices in terms of the history of the construction of the Homeric text. In the scholium ad Il. 11.62a, regarding the word οὔλιος, Aristonicus alerts us to the variant αὔλιος (ὅτι τινὲς γράφουσιν ‘αὔλιος’), which Callimachus supposedly accepted and imitated in fr. 177.6 (= 259.6 SH): if αὔλιος was accepted by Callimachus, it is possible that Zenodotus too was among the τινές, but we cannot exclude the possibility of a manuscript source common to both of them, as we recalled above whilst reporting Pfeiffer’s words. Aristonicus presents an analysis of the context and of the meaning of the two teachings, which concludes by preferring οὔλιος and rejecting αὔλιος: this material connected to a critical mark: scil. ἡ διπλῆ ὅτι, is undoubtedly Aristarchan and obviously involves, in the examination of the two variants, an authentic analysis also of the hypothetical Callimachean reprise in the third book of the Aitia.26 Another interesting case is presented by Aristonicus’ scholium ad Il. 12.34: here too we find the explanation of an Aristarchean sign, in all probability a diplè periestigmene, usually reserved for polemics against Zenodotus. Il. 12.34: ὣς ἄρ’ ἔμελλον ὄπισθεν: ὅτι Ζηνόδοτος γράφει ‘ὣς ἤμελλον ὄπισθε’. ἔστι δὲ

 24 Cf. Montanari 1988, Ptolemaios Epithetes, 75–112. 25 At least so far as we know: it is entirely possible that Aristophanes of Byzantium already had difficulties with the reconstruction of Zenodotus’ opinions and choices. 26 This case involves both Callimachus and Apollonius of Rhodes (4.1629): see Rengakos 1993, 133–134; 1994, 61, 155.

  Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries” βάρβαρον· τὰ γὰρ ἀπὸ συμφώνου ἀρχόμενα ἐπὶ τοῦ παρεληλυθότος χρόνου ἀπὸ τοῦ ε ἄρχεται. Let us compare this with the scholium to Apollonius of Rhodes 1.1309: Ap. Rh. 1.1309: καὶ τὰ μὲν ὧς ἤμελλε μετὰ χρόνον ἐκτελέεσθαι: Καλλιμάχου ὁ στίχος (Aitia fr 12. 6). κοινὸν δὲ ἁμάρτημα πάντων τῶν μεθ’ Ὅμηρον ποιητῶν, τὰ ἀπὸ συμφώνου ἀρχόμενα ῥήματα κατὰ τὸν ἐνεστῶτα χρόνον διὰ τοῦ η ἐκφέρειν ἐπὶ τοῦ παρατατικοῦ. The Apollonius of Rhodes scholium observes that the verse from Apollonius is Callimachus’, which is to say that it is taken from Callimachus. Indeed, it has been identified in fr. 12.6 of the Aitia. The two Hellenistic poets share the use of the forms ἠμελ-, instead of ἐμελ- as in Homer: this linguistic comparison supports the rejection of Zenodotus’ reading, which would have ἤμελλον in the Iliad, but is dismissed with the words ἔστι δὲ βάρβαρον. In this case, there is a recognisable difference in the language, whereby the use of Callimachus does not support a Homeric reading (indeed, it highlights post-Homeric poets’ divergence).27 The comparison between Homeric and Callimachean linguistic usage occurs repeatedly: a good parallel to the case just noted can be found for example in the Aristonicus’ scholium ad Il. 3.371b, where Callimachus is quoted, Aitia fr. 43.53. But it is worth dwelling a little bit on Il. 9.219: in scholium 219b Aristonicus (again introduced by ὅτι to explain a diplè of Aristarchus’) compared the Homeric use of θῦσαι with that of Timotheus (7 = 783 PMG) and of Philoxenus of Cythera (10 = 823 PMG), poets who lived at the turn of the 4th century and who, in contrast to Homer, behaved ὁμοίως τῇ ἡμετέρᾳ συνηθείᾳ. Scholium 219d, which belongs to the class of the sch. ex., again discusses the use of θῦσαι and also quotes Callimachus, Aitia fr. 5. A single sch. ex. certainly provides no guarantee that the material dates back to Aristarchus (as in the case of the scholia by Didymus and Aristonicus); but it is certainly plausible — indeed, it is entirely probable in my opinion — that here the scholia, in the form in which they have come down to us, provide a fragmentary record of a richer and more complex discussion, in which Aristarchus examined an element of the Homeric linguistic usage, comparing it with that of prior poets, all the way up to Callimachus. This would be a typical case of Aristarchus’ application of his historical and philological method, aimed at characterising and distinguishing — on the basis of extensive and in-depth knowledge — Homeric linguistic and stylistic usage from that of later poets. In this last case, by making use also of a sch. ex., we have somewhat strayed from the rule that we first imposed on ourselves: namely, to use only scholiastic  27 On this case, see Rengakos 1993, 61–62, 82; 1994, 61, 155.

Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries”  

sources, so as to guarantee that the examined material dates back at least to Aristarchus and thus to a date prior to the mid-2nd century BC (hence, the excerpta of Didymus and Aristonicus). But this exception to the rule is negligible if the hypothesis is valid (as I believe it is) that the discourse in its entirety in any case dates back to Aristarchus. Without harbouring any intention of betraying our initial prudence, I will nonetheless permit myself another exception by introducing a subject which I find quite interesting, although for the moment we will not follow it in its possible developments. In Il. 14.37, the scholia broadly discuss the reading ὀψείοντες, but something else of note for our purposes can be found only in Apollonius the Sophist: Ap. S. 125.32: ὀψείοντες ὀπτικῶς ἔχοντες. ὁ δὲ τύπος τῆς λέξεως Ἀττικός· κλαυσείοντες γὰρ λέγουσιν ἀντὶ τοῦ κλαυστικῶς ἔχοντες. καὶ Καλλίμαχος «οἵ κεν βρωσείοντες ἐμὸν παριῶσιν ἄροτρον» (Aitia fr. 24.17) ἀντὶ τοῦ βρωτικῶς ἔχοντες. Here a Callimachean passage is adduced as a parallel even for the Homeric usage, identified as “Attic”. The rich scholia to verse Il. 14.37 inform us that Aristarchus preferred the reading ὀψείοντες (we have already cited this case above, as it is perhaps the most revealing of the uncertainty that existed in the Aristarchan period as regards Zenodotus’ Homeric text);28 for this reason, there are grounds to suspect that the whole discourse, with its observation regarding the Attic character of the language used and its citation of Callimachus, dates back to Aristarchus. There is a certain similarity between this and the first example that we have seen, Od. 2.136–7, where a Callimachean parallel furnishes Aristarchus with a stylistic parallel to support his textual choice. Here the quotation of Callimachus is followed by a brief explanation of the same: βρωσείοντες ... ἀντὶ τοῦ βρωτικῶς ἔχοντες. If it could be established that this, too, formed part of the original discourse, we would have what we might define as evidence of exegesis on Callimachus inserted into a discourse by Aristarchus. But this is all too hypothetical, and we of course cannot take it as evidence in support of our hypothesis. Indeed, a brief examination of the citations of Callimachus in the Homeric scholia immediately show us that there are some cases in which the citation itself is accompanied by some other element, which we can define as evidence of exegesis. This matter could use considerably more research than it has received, and it would be well to set out on a very general investigation, one which enlarges the field generally to quotations from the Alexandrian poets in the scholia to the archaic and Classical poets, so as to research ways in which the possible evidence of exegesis connected to a passage  28 Ptolemaios Epithetes F 3, cf. nn. 23 and 24.

  Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries” by an Alexandrian poet might be traced back with certainty, or at least with a high degree of probability, to a grammarian prior to the Augustan period. If such research could produce significant results, this aspect too would acquire a certain weight so far as our study is concerned. Although the examples are limited, they paint a sufficiently varied picture: we have seen how Callimachean usage was set in contrast to the Homeric one, with a certain historical sensibility, and how Callimachus could be invoked as possible evidence in support of a particular textual Homeric choice or as a linguistic or stylistic parallel for the Homeric text. This last point reflects the Alexandrian philologists’ acknowledgement of the Homeric zelos of the Hellenistic poets. In the scholium to Od. 14.199, for example, it is said that Callimachus used the feminine genitive plural Κολωνάων (fr. 300.1) ζηλώσας the Homeric Κρητάων (which can be found in Od. 14.199). Unfortunately this scholium does not give us any insight into who produced this observation. But this issue, too, deserves to be studied with greater breadth, in the search for observations about the Homeric zelos of the Hellenistic poets which can be traced back to pre-Augustan philologists. The scholiographic material which we have examined here is far from extensive, but I think a conclusion can be drawn from it, one which other investigations might be able to confirm. We have seen various cases in which Aristarchus quoted Callimachus commenting on Homer, and various reasons for which the grammarian used these Callimachean passages: are we to believe that such differentiated and refined usage could be undertaken without prior in-depth study of Callimachus’s language and poetry? Is it possible to call on Callimachus to furnish a Homeric parallel — thus explicitly recognising his imitation of Homer — and to use this to reconstruct Zenodotus’ reading of a given passage, without first having submitted Callimachus, too, to a sufficiently detailed criticalphilological study? I think not: I believe that what we have seen leads us to suppose, altogether seriously, that Callimachus was used in this way, not only because he was known, but also because he was an object of independent study and interpretation. The present investigation has offered me but a glimpse of the matter, which has nonetheless persuaded me that broad and rigorous research into the citations, not only of Callimachus, but also of Theocritus, Aratus, Apollonius of Rhodes, Lycophron, Euphorion and other Hellenistic poets, and maybe even of Menander and of the authors of the New Comedy, in scholia to the archaic and Classical poets would give us an extremely interesting picture of the problem that we are investigating. Essentially, I think we are in need of a change of perspective: the use of the Hellenistic poets to comment on

Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries”  

and interpret the poets of the archaic and Classical periods can be seen not only as one of the erudite instruments deployed by the great Alexandrian philologists for the exegesis of the archaic and Classical poets, but also as a concrete sign of a philological study of the Hellenistic poets which was already fairly advanced, and one which should be investigated in this light. Our brief examination has clearly led us to Aristarchus, which is to say, to the period of the two papyrus commentaries on the Hellenistic poets we have discussed. If the use of Callimachus by Aristarchus in his Homeric exegesis constitutes valid evidence of a study also of Callimachus himself, then we are better able to understand that active interest in Hellenistic poetry on the part of the Alexandrian grammarians prior to the Augustan period from which we set out. Roughly between the end of the 3rd and the first half of the 2nd century BC — which is to say between Aristophanes of Byzantium and Aristarchus — the idea that it was necessary to deal also with “contemporary” poets — that is, the great poetry of the Hellenistic period — must have gradually become consolidated. It is not difficult to explain why there are few direct testimonies prior to the Augustan age: papyrus fragments of hypomnemata, or of other scholarly works, from the Ptolemaic period are on the whole much scarcer than ones from the Imperial age; moreover, philological-grammatical production on the Hellenistic poets surely did not become copious all at once. However, on second thought, is it in truth not more likely that the abundance of the commentaries on the Hellenistic poets written by Theon in the Augustus period represents, not a starting point, but an end point? In that epoch, philology showed that it had definitely come to master that segment of literary history represented by the poetry of the Hellenistic age. Before closing, I wish to enlarge this portrait even further, by adding one more consideration. Even the progress of Alexandrian philology with respect to prose writers is very significant: as we know, the historians came first, and then the orators were certainly studied. Our scholiography for Demosthenes is quite poor, whereas we have the testimony of a very important hypomnema by Didymus, preserved on papyrus, and fragments of other hypomnemata. This certainly means that the dearth of scholia deceives us — that Demosthenes was richly commented on, at least starting from the Augustan period. And so we have to ask ourselves: is it possible that Didymus’ hypomnemata on Demosthenes and other orators29 were entirely without precedent, that they were the first exegetical works on an orator? So far as the historians are concerned, we must remember that the scholia to Herodotus tell us very little, and if we did not have the fortunate case of  29 Schmidt 1854. For Demosthenes: Lossau 1964; Pearson/Stephens 1983.

  Alexandrian Philologists and Alexandrian Poets: Philology on “Contemporaries” P.Amherst 12, we would not know that Aristarchus himself wrote a hypomnema on Herodotus: Pfeiffer is cautious when speaking of the possibility that Aristarchus may also have commented on Thucydides, but he is confident about the existence of Thucydidean studies before Didymus.30 Considering on the one hand the prose writers (historians and orators) and on the other the Hellenistic poets, we are compelled to say that in the period of time between Zenodotus and the Augustan period (Didymus, Aristonicus, and Theon), philological activity — leaving aside the question of editions, commentaries and lexicons for each work by each author — touched upon a truly vast range of interests: one might say to practically all the principal aspects of archaic, Classical, and Hellenistic literature. The evidence all points to the fact that this broadening of the horizon occurred in the period between Aristophanes of Byzantium and Aristarchus, and was consolidated with the latter.31 The idea of studying the poetic texts of the past, and preserving and understanding, through the most advanced critical instruments, the cornerstones of a paideia which had been venerated and codified in its authoritativeness (including the revolutionary aspect of producing new editions that might take into account and repair any damage caused by textual transmission) — this idea, in its full maturity, evidently became, in the scholarly awareness of its greatest representatives, a method so sure and definite, as to be continually applicable to literature as a whole.

 30 Pfeiffer 1968, 224–225, 277–278; Montanari 1993, 247–249, 272 (= this volume, ch. 1); Maehler 1994, 121–124. The definition of Dionysius Thrax of the grammatikè as empeiria of the works of poets and prose writers (a definition which is surely authentic: cf. the latest work by Schenkeveld 1994) means that from the generation immediately following Aristarchus it was normal to talk of philological-grammatical studies even on prose works. 31 Naturally, I mean something different from the classificatory work of Callimachus in the Pinakes, or the materials occasionally inserted in the collection of Lexeis up until Aristophanes of Byzantium.

Bibliography Abbamonte, G. 2004. ‘Tipologie esegetiche nei commenti di Alessandro di Afrodisia: la parafrasi’, in: G.Abbamonte/F. Conti Bizzarro/L. Spina (eds.), L’ultima parola. L’analisi dei testi: teorie e pratiche nell’antichità greca e latina, Napoli, 19–34. Aceti, C. 2008. ‘Sarpedone fra mito e poesia’, in: C. Aceti/D. Leuzzi/L. Pagani (eds.), Eroi nell’Iliade. Personaggi e strutture narrative, Roma, 224–262. Adams, Ch.D. 1919. The Speeches of Aeschines, Cambridge/London. Adler, A. 1914. ‘Die Kommentare des Asklepiades von Myrlea’, Hermes 49, 39–46. Adler, A. 1928–1938. Suidae Lexicon, 5 vols., Leipzig. Albiani, M.G. 2000. ‘Nikias Nr. 4’, DNP VIII, 914. Albini, U. 1991. Nel nome di Dioniso. Vita teatrale nell’Atene classica, Milano. Algra, K. 2001. ‘Comments or commentary? Zeno of Citium and Hesiod’s Theogonia’, Mnemosyne 54, 562–581. Allen, T.W. 1924. Homer. The Origins and the Transmission, Oxford. Allen, T.W. 1931. Homeri Ilias, Oxford. Alpers, K. 1981. Das attizistische Lexikon des Oros, SGLG 4, Berlin/New York. Alpers, K. 1990. ‘Griechische Lexikographie in Antike und Mittelalter’, in: H. Koch/A. Krup-Ebert/ M. Eigen (eds.), Welt der Information. Wissen und Wissensvermittlung in Geschichte und Gegenwart, Stuttgart, 14–38. Aly, W. 1954. ‘Praxiphanes’, RE XXII, 1769–1784. Amarante, F. 1998. ‘Eschilo nei papiri ercolanesi’, CErc 28, 133–150. Ambaglio, D. 1980. ‘L’opera storiografica di Ellanico di Lesbo’, in: Ricerche di storiografia antica II, Pisa, 9–192. Andrei, O. 1984. A. Claudius Charax di Pergamo: interessi antiquari e antichità cittadine nell’ età degli Antonini, Bologna. Apthorp, M.J. 1980. The Manuscript Evidence for Interpolation in Homer, Heidelberg. Apthorp, M.J. 1998. ‘Double News from Antinoopolis on Phoenix’s Parricidal Thoughts (Iliad 9, 458–461)’, ZPE 122, 182–188. Arata, L./Bastianini, G./Montanari, F. 2004. ‘Aeschylus 3. P.Oxy. XX 2256. Hypotheseis di varie tragedie’, in: CLGP I 1.4, Berlin/Boston, 34–51. Argenio, R. 1964. ‘Il comico Eubulo’, RSC 12, 125–143. Arrighetti, G. 1964. Satiro, Vita di Euripide, Pisa. Arrighetti, G. 1968. ‘Il POxy 1611: alcuni problemi di erudizione antica’, SCO 17, 76–98. Arrighetti, G. 1977a. ‘Fra erudizione e biografia’, SCO 26, 13–67. Arrighetti, G. 1977b. ‘Hypomnemata e scholia: alcuni problemi’, MPhL 2, 49–67. Arrighetti, G. 1987. Poeti, eruditi e biografi. Momenti della riflessione del Greci sulla letteratura, Biblioteca di Studi Antichi 52, Pisa. Arrighetti, G. 1994. ‘Riflessione sulla letteratura e biografia presso i Greci’, in: Montanari 1994a, 211–250. Arrighetti, G. 1998. Esiodo, Opere. Testi introdotti, tradotti e commentati, Torino. Arrighetti, G. 2007. ‘Anekdote und Biographie’, in: M. Erler/S. Schorn (eds.), Die griechische Biographie in hellenistischer Zeit, Berlin/New York, 78–100. Arrighetti, G. 2008. ‘Cameleonte peripatetico e gli studi sulla biografia greca’, in: Arduini et al. (eds.), Studi offerti ad Alessandro Perutelli, Roma, vol. I, 63–69.

https://doi.org/10.1515/9783110772371-063

  Bibliography Arrighetti, G./Calvani, G./Montanari, F. 1991. Concordantia et indices in scholia Pindarica vetera, Hildesheim. Ascheri, P. 2009. ‘Demetrius [14] Ixion’, in LGGA. Ascheri, P. 2010. ‘Demetrio Issione “dialettologo”: l’attico e il dialetto degli Alessandrini’, in: F. Montana (ed.), Aner Polytropos. Ricerche di filologia greca’, Roma, 125–152. Asheri, D. 1988. Erodoto, Le storie, I: La Lidia e la Persia, Milano. Asheri, D. 1990. Erodoto, Le storie, III: La Persia, Milano. Aujac, G. 1977. ‘Eratosthène premier éditeur de textes scientifiques ?’, Pallas 25, 3–24. Aujac, G. 1991. Denys D’Halicarnasse, Thucydide. Seconde lettre à Ammée, Paris. Aujac, G./Lasserre, F. 1969. Strabon, Geographies, t. I 1, Paris. Austin, C./Bastianini, G. 2002. Posidippi Pellaei quae supersunt omnia, Milano. Avezzù, G. 1982. Alcidamante, Orazioni e frammenti, Roma. Avezzù, G. 2007. ‘Esegesi e annotazioni drammaturgiche negli scholia vetera sofoclei’, in: R. Pretagostini/E. Dettori (eds.), La cultura letteraria ellenistica, Roma, 223–232. Avezzù, G./Scattolin, P. (eds.) 2006. I classici greci e i loro commentatori. Dai papiri ai marginalia rinascimentali. Memorie della Accademia Roveretana degli Agiati, s. II vol. X, Rovereto. Ax, W. 1982. ‘Aristarch und die “Grammatik”’, Glotta 60, 96–109. Ax, W. 1986. Laut, Stimme und Sprache. Studien zu drei Grundbegriffen der antiken Sprachtheorie, Göttingen. Ax, W. 2000. ‘Aristophanes von Byzanz als Analogist: zu Fragment 374 Slater (= Varro, de Lingua latina 9, 12)’, in: F. Grewing (ed.), Lexis und Logos. Studien zur antiken Grammatik und Rhetorik, Stuttgart, 116–127 (= orig. in: Glotta 68, 1990, 4–18). Baar, J. 1952. Untersuchungen zur Terminologie der Iliasscholien, Diss. Hamburg. Baar, J. 1961. Index zu den Ilias-Scholien, Baden-Baden. Bachmann, L. 1828. Anecdota Graeca e codd. mss. Bibl. Reg. Parisin. descr. L.B., Lipsiae. Bacigalupo, V. 2018. ‘Dicearco, Aristarco e i pronomi riflessivi’, RFIC 146, 98–128. Baege, M. 1882. De Ptolemaeo Ascalonita, Halle. Bagordo, A. 1998. Die antiken Traktate über das Drama. Mit einer Sammlung der Fragmente, Stuttgart. Barnes, J. 1711. Homeri Ilias et Odyssea, Cambridge. Barrett, W.S./Obbink, D. 2014. ‘P.Oxy. 5201. Commentary on Pindar, Olympian 1’, The Oxyrhynchus Papyri, vol. LXXIX, London, 116–124. Barth, H.-L. 1984. Die Fragmente aus den Schriften des Grammatikers Kallistratos zu Homers Ilias und Odyssee, Diss. Bonn. Bartoletti, V. 1966. ‘Nuovi papiri fiorentini’, ASNP 25, 1–25. Barwick, K. 1957. Probleme der stoischen Sprachlehre und Rhetorik, Abh. Sächs. Akad. Wiss. Phil.-hist. Kl. 49, 3, Berlin. Bastianini, G. 1996. ‘Tipologie dei rotoli e problemi di ricostruzione’, in: M. Capasso (ed.), Atti del V Seminario internazionale di papirologia, Papyrologica Lupiensa 4, 21–42. Bastianini, G. 2001. ‘Osservazioni sul “Papiro di Teramene”’, in: S. Bianchetti (ed.), POIKILMA. Studi in onore di Michele R. Cataudella, La Spezia, 81–87. Bastianini, G. 2006. ‘Considerazioni sulle diegeseis fiorentine (PSI XI 1219)’, in: G. Bastianini/ A. Casanova (eds.), Callimaco. Cent’anni di papiri, Firenze, 149–166. Bastianini, G./Gallazzi, C. 2001. Papiri dell’Università degli Studi di Milano – VIII. Posidippo di Pella. Epigrammi (P.Mil. Vogl. VIII 309), Milano.

Bibliography  

Bastianini, G./Haslam, M./Maehler, H./Montanari, F./Römer, C. (eds.) 2004. Commentaria et Lexica Graeca in Papyris reperta, vol. I 1.1: Aeschines–Alcaeus, München. Bastianini, G./Haslam, M./Maehler, H./Montanari, F./Römer, C. (eds.) 2006. Commentaria et Lexica Graeca in Papyris reperta, Pars I, vol. I 1.4: Aristophanes–Bacchylides, München. Bastianini, G./Sedley, D.N. 1995. ‘Commentarium in Platonis «Theaetetum»’, in: CPF III. Commentari, Firenze, 227–562. Bayer, E. 1942. Demetrius Phalereus der Athener, Tübinger Beiträge zur Altertumswissenschaft 36, Tübingen. Beccard, Th. 1850. De scholiis in Homeri Iliadem Venetis A, Berolini. Bekker, I. 1827. Scholia in Homeri Iliadem, vol. III, Berolini. Bekker, I. 1833. Apollonii Sophistae Lexicon Homericum, Berolini. Benedetto, G. 1993. Il sogno e l’invettiva. Momenti di storia dell’esegesi callimachea, Firenze. Bérard, V. 1920. ‘Sur les scholies et le texte de l’Odyssée’, RPh 44, 173–228. Bérard, V. 1933. L’Odyssée «Poésie homérique», Paris. Bergk, T. 1915. Poetae Lyrici Graeci, vol. II4, Lipsiae. Berndt, R. 1906. De Charete, Chaeride, Alexione grammaticis eorumque reliquis, Königsberg. Bethe, E. 1887. Quaestiones Diodoreae mythographae. Diss. Göttingen. Blank, D.L. 1982. Ancient Philosophy and Grammar. The Syntax of Apollonius Dyscolus, Chico, CA. Blank, D.L. 1983. ‘Remarks on Nicanor, the Stoics and the Ancient Theory of Punctuation’, Glotta 61, 48–67. Blank, D.L./Dyck, A.R. 1984. ‘Aristophanes of Byzantium and Problem-Solving in the Museum: Notes on a Recent Reassessment’, ZPE 56, 17–24. Blass, F. . Aeschinis Orationes, post Fr. Frankium, Lipsiae. Blass, F./Schindel, U. . Aeschinis Orationes, post Fr. Frankium, Stuttgart. Blau, A. 1883. De Aristarchi discipulis, Diss. Jena. Blum, R. 1977. Kallimachos und die Literaturverzeichnung bei den Griechen, Frankfurt (= Kallimachos. The Alexandrian Library and the Origins of Bibliography, Madison, IL, 1991). Blumenthal, A. von. 1934. ‘Theokritos Nr. 1’, RE 5 A, 2001–2025. Boeckh, A. 1821. Pindari Epiniciorum Interpretatio Latina cum commentario perpetuo. Fragmenta et indices, Leipzig. Boissonnade, J. Fr. 1832. Anecdota Graeca, vol. IV, Paris (repr. Hildesheim, Olms 1962). Bölte, F. 1912. ‘Helloi’, RE VIII 1, 194–196. Bolzoni, L. 1980. L’universo dei poemi possibili. Studi su Francesco Patrizi da Cherso, Roma. Bonanno, M.G. 1972. Studi su Cratete comico, Padova. Bouchard, E. 2016. Du Lycée au Musée: théorie poétique et critique littéraire, Paris. Boudon Millot, V./Jouanna, J. 2010. Galien. Ne pas se chagriner, Paris. Boudreaux, P. 1919. Le texte d’Aristophane et ses commentateurs, Paris. Brandis, C.A. 1961. ‘Scholia in Aristotelem’, in: O. Gigon, Aristotelis Opera ex rec. I. Bekkeri, vol. IV, Berlin. Brashear, W./Hoogendijk, F.A.J. 1990. ‘Corpus tabularum lignearum ceratarumque Aegyptiarum’, Enchoria 17, 21–54. Braswell, B.K. 2013. Didymos of Alexandria. Commentary on Pindar, Basel. Braswell, B.K./Billerbeck, M. 2007. The Grammarian Epaphroditus. Testimonia and Fragments, Bern/Berlin. Bravi, L. 1996. ‘Nota al P.Oxy. 3838’, QUCC 53, 61–65. Breithaupt, M. 1915. De Parmenisco grammatico, Leipzig/Berlin.

  Bibliography Brink, K.O. 1946. ‘Callimachus and Aristotle: an Inquiry into Callimachus’ Πρὸς Πραξιφάνην’, CQ 40, 11–26. Bröcker, L.O. 1985. ‘Die Methoden Galens in der literarischen Kritik’, RhM 150, 415–438. Broggiato, M. 2001. Cratere di Mallo, I frammenti, Pleiadi, Studi sulla letteratura antica 2, La Spezia. Buck, C.D./Petersen, W. 1945. A Reverse Index of Greek Nouns and Adjectives, Chicago. Bühler, W. 1982–1987. Zenobii Athoi proverbia, vols. I, IV, Göttingen. Buffière, F. 19892. Héraclite, Allégories d’Homère, Paris. Bürchner, L. 1923. ‘Selloi’, RE II A 2, 1323. Bury, R.G. 1949. Sextus Empiricus, Against the Professors, Loeb Classical Library 382, Cambridge, MA. Bussemaker, U.C. 1847. Scholia et Paraphrases in Nicandrum et Oppianum, Parisiis. Bywater, I. 1909. Aristotle, On the Art of Poetry, Oxford. Calderini, A. 1921. ‘Commenti «minori» al testo di Omero in documenti egiziani’, Aegyptus 2, 303–326. Callanan, C.K. 1987. Die Sprachbeschreibung bei Aristophanes von Byzanz, Hypomnemata 88, Göttingen. Cambiano, G./Canfora, L./Lanza, D. (eds.) 1992. Lo Spazio letterario della Grecia antica, vol. I: La produzione e la circolazione del testo, t. I: La polis, Roma. Cambiano, G./Canfora, L./Lanza, D. (eds.) 1993. Lo Spazio letterario della Grecia antica, vol. I: La produzione e la circolazione del testo, t. II: L’ellenismo, Roma. Cameron, A. 1995. Callimachus and His Critics, Princeton. Canfora, L. 1974. Teorie e tecnica della storiografia classica, Roma/Bari. Canfora, L. 1993. ‘La Biblioteca e il Museo’, in: Cambiano/Canfora/Lanza 1993, 11–30. Canfora, L. 1999. ‘La découverte du Venetus Marcianus A par Villoison’, in: F. Létoublon/ C. Volpilhac-Auger (eds.), Homère en France après la Querelle (1715–1900), Paris, 41–50. Cantarella, R. 1941. Eschilo, Firenze. Cantarella, R. 1942. ‘Nuovi frammenti papiracei di Eschilo’, Dioniso 9, 75–79. Cantarella, R. 1948. I nuovi frammenti eschilei di Ossirinco, Napoli. Cantarella, R. 1969. ‘Aristofane di Bisanzio, Menandro e la mimesi’, RAL 24, 189–194. Cantilena, M. 1987. ‘Due studi sulla VII Olimpica di Pindaro. I: Quando cadde la pioggia d’oro?’, Prometheus 13, 209–216. Cantilena, M. 1990. ‘Due studi sulla VII Olimpica di Pindaro. II: I significati della pioggia d’oro’, Prometheus 16, 111–135. Cantore, R. 2012. ‘I marginalia dei primi trentaquattro fogli del Laur. Plut. 70.3 (A) di Erodoto’, Bollettino dei Classici 33, 3–32. Carlini, A. 1977. ‘Filologia e Erudizione’, in: Storia e civiltà dei Greci, vol. IX, Milano, 341–360. Carlini, A. 1978. Papiri letterari greci, Pisa. Carlini, A. 1990. Review of A. Giardina (ed.), Società romana e impero tardoantico, IV. Tradizione dei classici, trasformazioni della cultura, Bari 1986, RFIC 118, 225–236. Carnuth, O. 1869. Aristonici Περὶ σημείων Ὀδυσσείας reliquiae emendatiores, Lipsiae. Carnuth, O. 1875. Nicanoris Περὶ Ὀδυσσειακῆς στιγμῆς reliquiae emendatiores, Berolini. Cassio, A.C. (ed.) 1977. Aristofane, Banchettanti, Pisa. Cassio, A.C. 1993a. ‘Alcmane, il dialetto di Cirene e la filologia alessandrina’, RFIC 121, 24–36. Cassio, A.C. 1993b. ‘Parlate locali, dialetti delle stirpi e fonti letterarie nei grammatici greci’, in: E. Crespo/José L. García Ramón/A. Striano (eds.), Dialectologica Graeca, Actas del II coloquio internacional de dialectología griega, Madrid, 73–90.

Bibliography  

Cassio, A.C. 1997. ‘Futuri dorici, dialetto di Siracusa e testo antico dei lirici greci’, AION(filol) 19, 187–214. Cassio, A.C. 2002. ‘Early editions of the Greek epics and Homeric textual criticism in the sixth and fifth centuries BC’, in: Montanari/Ascheri 2002, 105–136. Cassio, A.C. (ed.) 2008. Storia delle lingue letterarie greche, Firenze. Càssola, F. 1975. Inni omerici, Milano. Cauderlier, P. 1992. ‘Les tablettes grecques d’Egypte: inventaire’, in: E. Lalou (ed.), Les tablettes à écrire de l᾽Antiquité à l᾽époque moderne, Bibliologia 12, Turnhout, 63–96. Cavallo, G. 1967. ‘Osservazioni paleografiche sul canone e la cronologia della cosidetta “onciale romana”’, ASNP 36, 209–220. Cavallo, G. 1975. ‘Grammata Alexandrina’, JÖB 24, 23–54 (= Cavallo 2005, 175–202). Cavallo, G. 1983. Libri, scritture, scribi a Ercolano, Napoli. Cavallo, G. 1986. ‘Conservazione e perdita dei testi greci’, in: A. Giardina (ed.), Società romana e impero tardoantico, IV, Tradizione dei classici. Trasformazione della cultura, Roma, 83– 172. Cavallo, G. 1987. ‘La trasmissione dei ‘moderni’ fra antichità tarda e Medioevo bizantino’, ByzZ 80, 313–329. Cavallo, G. 1989. ‘Lo specchio omerico’, in: Mélanges de l’École Française de Rome 101–102, 609–627. Cavallo, G. 1989a. ‘Libro e cultura scritta’, in: A. Schiavone (ed.), Storia di Roma, vol. IV, Turin, 693–734. Cavallo, G. 2005. Il calamo e il papiro, Firenze. Cavallo, G. 2008. La scrittura greca e latina dei papiri. Una introduzione, Pisa/Roma. Cavallo, G./Maehler, H. 1987. Greek Bookhands of the Early Byzantine Period A.D. 300–800, (BICS Suppl. 47), London. Chambers, M.H. 1971. ‘More notes on the text of the Ath. Pol.’, TAPhA 102, 41–47. Chantraine, P. 1933. La formation des noms en Grec ancien, Paris. Chantraine, P. 1973–1981. Grammaire Homérique, 2 vols., Paris. Christ, W./Schmid, W./Stählin, O. 1920. Geschichte der griechischen Litteratur, II6, München. Christodoulou, G.A. 1977. Τὰ ἀρχαῖα σχόλια εἰς Αἴαντα τοῦ Σοφοκλέους, Αθήνα. Cingano, E. 2009. ‘The Hesiodic Corpus’, in: F. Montanari/A. Rengakos/Chr. Tsagalis (eds.), Brill’s Companion to Hesiod, Leiden/Boston, 91–130. Cohn, L. 1895. ‘Aristonikos Nr. 17’, RE II 1, 964–966. Cohn, L. 1895a. ‘Apollonios Nr. 83’, RE II 1, 140. Cohn, L. 1896. ‘Athenokles’, RE II 2, 2049. Cohn, L. 1899. ‘Χωϱίζοντες’, RE III 2, Stuttgart, 2439. Cohn, L. 1905. ‘Diogenianos [4]’, RE V 1, 782–783. Cohn, L. 1913. ‘Griechische Lexicographie’, in: K. Brugmann, Griechische Grammatik, 4. Aufl. bearb. v. A. Thumb, München, 575–616. Collart, J. 1962. ‘Analogie et anomalie’, in: Varron, Entretiens Hardt 9, Vandoeuvres-Genève, 117–140. Collart, P. 1931. ‘Épitomé d’Homère, Iliade A et glossaire A 1–21’, BIFAO 31, 43–47. Colomo, D. 2008. ‘Osservazioni sullo scriba ossirinchita dell’omega quadrangolare (Johnson A2)’, S&T 6, 3–34. Comunetti, M. 2020. ‘Le citazioni di Eschilo negli scolii a Euripide’, RFIC 148, 91–113. Consbruch, M. 1906. Hephaestionis Enchiridion, cum commentariis veteribus, Leipzig. Conte, G.B. 1974. Memoria dei poeti e sistema letterario. Catullo Virgilio Ovidio Lucano, Torino.

  Bibliography Conte, G.B. 2013. Ope ingenii. Esperienze di critica testuale, Pisa. Cook, A.B. 1903. ‘Zeus, Jupiter and the Oak’, CR 17, 174–186. Cook, A.B. 1925. Zeus. A Study in Ancient Religion, vol. II, Cambridge. Corradi, M. 2007. ‘Chamaeleon’, in: LGGA. Cribiore, R. 1996. Writing, Teachers, and Students in Graeco-Roman Egypt, Am. Stud. Pap. 36, Atlanta. Crugnola, A. 1971. Scholia in Nicandri Theriaka, Milano. Crusius, O. 1888. ‘Stesichoros und die epodische Composition in der griechischen Lyrik’, in: Commentationes Philologae quibus O. Ribbeckio ... congratulantur discipuli Lipsienses, Lipsiae 1888, 1–22. Cunningham, I.C. 1986. ‘Harpocration and the Συναγωγή’, GRBS 27, 205–221. Dahlmann, H. 1932. Varro und die Hellenistische Sprachtheorie, Berlin. Dahlmann, H. 1940. Varro. De lingua latina Buch VIII, Berlin. Dähnhardt, O. 1894. Scholia in Aeschyli Persas, Lipsiae. Dain, A. 1954. Le Philetairos attribué à Hérodien, Paris. DʼAlessio, G.B. 1990. ‘Aggiunte all’“Ostrica” (Suppl.Hell. 983 v. 3)’, ZPE 81, 299–303. DʼAlessio, G.B. 2000. ‘Le Argonautiche di Cleone Curiense’, in: R. Pretagostini (ed.), La letteratura ellenistica. Problemi e prospettive di ricerca, Quaderni dei seminari romani di cultura greca 1, Roma, 91–112. DʼAlessio, G.B. 2005. ‘The Megalai Ehoiai: a survey of the fragments’, in: R. Hunter (ed.), The Hesiodic Catalogue of Women: Constructions and Reconstructions, Cambridge, 176–216. Damschen, G. 2000. ‘Nikias Nr. 9’, DNP VIII, 915–916. Danek, G. 1998. Epos und Zitat. Studien zu den Quellen der Odyssee, Wien. Danek, G. 2002. Traditional referentiality and Homeric intertextuality, in: Montanari/Ascheri 2002, 3–20. Daris, S. 1972. ‘Papiri letterari dell'Università Cattolica di Milano’, Aegyptus 52, 89–90. Daris, S. 1974. ‘Scholia minora al libro I dell’Iliade (P.Palau Rib. inv. 147)’, Studia Papyrologica 13, 7–20. Davison, J. 1966. Review of D. Page, The Oxyrhynchus papyri, part XXIX, London 1963, JHS 85, 197–198. Davison, J. 1966. ‘Oxyrhynchus Papyrus 2506’, in: Atti dell᾽XI Congresso Internazionale di Papirologia, Milano, 96–106. Davison, J. 1968. From Archilochus to Pindar, New York. Deas, H.T. 1931. ‘The Scholia Vetera to Pindar’, HSPh 42, 1–78. De Jonge, C.C. 2011. ‘Dionysius of Halicarnassus and the Scholia on Thucydides’ Syntax’, in: Matthaios/Montanari/Rengakos 2011, 451–478. De Lacy, P. 1948. ‘Stoic Views of Poetry’, AJPh 69, 241–271. de Marco, V. 1932. ‘Sulla tradizione manoscritta degli Scholia Minora all’Iliade’, Atti della Reale Accademia Nazionale dei Lincei, Memorie della Classe di Scienze Morali, Storiche e Filologiche, Ser. 6.4, 373–407. de Marco, V. 1936. ‘Sulla tradizione manoscritta degli scolii Sofoclei’, SIFC 3, 3–44. de Marco, V. 1937. ‘De scholiis in Sophoclis tragoedias veteribus’, Atti della Reale Accademia Nazionale dei Lincei, Memorie della Classe di Scienze Morali, Storiche e Filologiche 6, 105–228. de Marco, V. 1941. ‘Da un manoscritto degli Scholia Minora all’Iliade’, Atti della Reale Accademia d’Italia, Rendiconto della Classe di Scienze Morali e Storiche, Ser. 7, Suppl. al vol. II, 125–145.

Bibliography  

de Marco, V. 1946. Scholia Minora in Homeri Iliadem. 1, Λέξεις Ὁμηρικαί codd. Urb. CLVII et Selestadiensis CVII [lege: CV], Città del Vaticano. de Marco, V. 1952. Scholia in Sophoclis Oedipum Coloneum, Romae. de Marco, V. 1975. ‘Philostephanos e Esiodo’, PP 165, 453–462. de Marco, V. 1984. ‘Frammenti di Teopompo, Teofrasto, Callimaco’, SIFC 2, 193–201. de Sousa Medeiros, W. 1961. Hipónax de Éfeso. I. Fragmentos dos Iambos, Coimbra. De Stefani, L. 1909–1920. Etymologicum Gudianum, 2 vols., α-ζειαί, Lipsiae. De Velsen, A. 1853. Tryphonis Grammatici Alexandrini Fragmenta, Berolini. Degani, E. 1977–1978. ‘Problemi di lessicografia greca’, BIFG 4, 135–146. Degani, E. 1984. Studi su Ipponatte, Bari. Degani, E. 1987. ‘Lessicografi’, in: Dizionario degli scrittori Greci e Latini, vol. II, Milano, 1169– 1189. Degani, E. 1991. Hipponactis Testimonia et Fragmenta, Stuttgart/Leipzig (1st ed. 1983). Degani, E. 1995. ‘La lessicografia’, in: AA.VV., Lo spazio letterario della Grecia antica, vol. II, Roma, 505–527. Degani, E. 1999. ‘Filologia e storia’, Eikasmos 10, 279–314. Degenhardt, C. 1909. De veterum grammaticorum scholiis, Diss. Monacensis, Lipsiae. Deichgräber, K. 1930. Die Griechische Empirikerschule. Sammlung der Fragmente und Darstellung der Lehre, Berlin. Del Fabbro, M. 1979. ‘Il commentario nella tradizione papiracea’, Studia Papyrologica 18, 69– 132. Della Corte, F. 1939. ‘Il frammento cosmologico brussellese’, RFIC 67, 36–42. Di Benedetto, V. 1958. ‘Dionisio Trace e la techne a lui attribuita’, ASNP 27, 169–210. Di Benedetto, V. 1959. ‘Dionisio Trace e la techne a lui attribuita’, ASNP 28, 87–118. Di Benedetto, V. 1973. ‘La TEXNH spuria’, ASNP 3, 797–814. Di Benedetto, V. 1990. ‘At the Origins of Greek Grammar’, Glotta 68, 19–39. Di Gregorio, L. 1975. Scholia Vetera in Hesiodi Theogoniam, Milano. Di Marco, M. 1989. Timone di Fliunte, Silli, Roma. Dickey, E. 2007. Ancient Greek Scholarship, Oxford. Dickey, E. 2015. ‘The Sources of our Knowledge of Ancient Scholarship’, in: Montanari/ Matthaios/Rengakos 2015, I 459–514. Diehl, W. 1938. Die wörtlichen Beziehungen zwischen Ilias und Odyssee, Greifswald. Diggle, J. 1987. Euripidis fabulae, vol. I, Oxonii. Diggle, J. 1998. Tragicorum Graecorum fragmenta selecta, Oxonii. Dihle, A. 1959. ‘Ptolemaios Epithetes’, RE XXIII 2, 1862–1863. Dihle, A. 1962. ‘Ptolemaios 79a’, RE Suppl. IX, 1306. Dihle, A. 1986. ‘Philosophie-Fachwissenschaft-Allgemeinbildung’, in: H. Flashar/O. Gigon (eds.), Aspects de la philosophie hellénistique, Entretiens Hardt 32, Vandoeuvres-Genève, 185–232. Diller, A. 1954. ‘The Scholia on Strabo’, Traditio 10, 29–50 (= A. Diller, Studies in Greek Manuscript Tradition, Amsterdam 1983, 23–45). Dilts, M.R. 1971. Heraclidis Lembi Excerpta Politiarum, Durham, NC. Dilts, M.R. 1992. Scholia in Aeschinem, Stutgardiae/Lipsiae. Dilts, M.R. 1997. Aeschinis Orationes, Stutgardiae/Lipsiae. Dindorf, G. 1829. Aristides, vol. III, Lipsiae. Dindorf, G. 1851. Scholia in Aeschylum, Oxonii. Dindorf, G. 1852. Scholia Graeca in Aeschinem et Isocratem, Oxonii (repr. Hildesheim 1970).

  Bibliography Dindorf, G. 1855. Scholia Graeca in Homeri Odysseam, Oxonii (repr. Amsterdam 1962). Dindorf, G. 1855a. Xenophontis Expeditio Cyri, Oxonii. Dindorf, G. 1877. Scholia Graeca in Homeri Iliadem ex codicibus aucta et emendata III, Oxonii. Dörrie, H. 1979. ‘Aristippos (3–5)’, in: Der Kleine Pauly, I 562. Drachmann, A.B. 1903–1927. Scholia vetera in Pindari carmina, 3 vols., Lipsiae. Dubischar, M. 2015. ‘Typology of Philological Writings’, in: Montanari/Matthaios/Rengakos 2015, I 545–599. Dübner, F. 1877. Scholia in Aristophanem, Parisiis. Düring, I. 1941. Herodicus the Cratetean, Kungliga Vitterhets Historie och Antikvitets Akademiens Hanglingar, vol. 51, 2, Stockholm. Düring, I. 1957. Aristotle in the Ancient Biographical Tradition, Göteborg. Dyck, A.R. 1981. ‘On Apollodorus of Cyrene’, HSPh 35, 101–106. Dyck, A.R. 1983. Epimerismi omerici, Pars Prior, Berlin/New York. Dyck, A.R. 1987. ‘The Glossographoi’, HSCPh 91, 119–160. Dyck, A.R. 1988. The Fragments of Comanus of Naucratis, SGLG 7, Berlin/New York. Easterling, P. 2006. ‘Le choeur dans la tragédie grecque d’après les commentateurs anciens’, CRAI, 1585–1597. Ebeling, H. et al. 1885. Lexicon Homericum, Lipsiae. Edwards, M.W. 1991. The Iliad. A Commentary, Volume V: books 17–20, Cambridge. Egenolff, P. 1894. ‘Zu Herodianus Technicus’, Jahrb. class. Philol. 40 (149), 337–346. Erbse, H. 1953. ‘Homerscholien und Hellenistische Glossare bei Apollonios Rhodios’, Hermes 81, 163–196. Erbse, H. 1959. ‘Über Aristarchs Iliasausgaben’, Hermes 87, 275–303. Erbse, H. 1960. Beiträge zur Überlieferung der Iliasscholien, München. Erbse, H. 1961. Geschichte der Textüberlieferung, Zürich. Erbse, H. 1964. Review of van der Valk 1963–1964 vol. I, Gnomon 36, 549–557. Erbse, H. 1969–1988. Scholia Graeca in Homeri Iliadem (Scholia Vetera), 7 vols., Berolini. Erbse, H. 1972. Beiträge zum Verständnis der Odyssee, Berlin/New York. Erbse, H. 1979. ‘F.A. Wolf e gli scoli all’Iliade’, ASNP 9, 39–58. Erbse, H. 1980. ‘Zur normativen Grammatik der Alexandriner’, Glotta 58, 236–258. Ercolani, A. 2006. Omero: introduzione allo studio dell’epica greca arcaica, Roma. Esposito, E. 2004. ‘Aeschines’, in: Commentaria et Lexica Graeca in Papyris reperta (CLGP), I 1.1, München/Leipzig, 3–11. Fantuzzi, M. 1988. Ricerche su Apollonio Rodio, Roma. Fantuzzi, M./Pretagostini, R. 1995. Struttura e storia dell’esametro greco, Roma. Fantuzzi, M. 2000. ‘An Aristarchean reading of Apollonius’ Argonautica’, SemRom. 3, 313–323. Fantuzzi, M. 2001. ‘«Homeric» Formularity in the Argonautica of Apollonius of Rhodes’, in: Th.D. Papanghelis/A. Rengakos (eds.), A Companion to Apollonius Rhodius, Leiden/Boston/ Köln, 171–192. Fantuzzi, M. 2005. ‘Euripides (?) Rhesus 56–58 and Homer Iliad 8.498–501: another possible clue to Zenodotus’ reliability’, CPh 100, 268–273. Färber, H. 1936. Die Lyrik in der Kunsttheorie der Antike, München. Fehling, D. 1989. ‘Die ursprüngliche Troja-Geschichte, oder: Interpretationen zur TrojaGeschichte’, WJA 15, 7–16. Fehling, D. 1991. Die ursprüngliche Troja-Geschichte, oder: Interpretationen zur TrojaGeschichte, Innsbruck.

Bibliography  

Fenoglio, S. 2009. ‘La riflessione sui dialetti nei Commentarii all’Odissea di Eustazio di Tessalonica’, Quaderni del Dipartimento di Filologia, Linguistica e Tradizione Classica “A. Rostagni”, n.s. 8, 2009, 239–254. Ferrari, F. 2005. ‘Per il testo di Posidippo’, MD 54, 185–212. Ferrari, L. 1968. I drammi perduti di Eschilo, Palermo. Finkelberg, M. 2000. ‘The Cypria, the Iliad and the problem of multiformity in oral and written tradition’, CP 95, 1–11. Fischer, E. 1974. Die Ekloge des Phrynichos, SGLG 1, Berlin/New York. Fleming, J.Th. 1975. ‘Ancient Evidence for the Colometry of Aeschylus’ Septem’, GRBS 16, 141– 148. Floratos, C.S. 1972. Strabon über Literatur und Poseidonios, Athens. Fornaro, S. 1996–2003. “Glaukos [7] von Rhegion”, “Glaukos [8] von Samos”, DNP, vol. IV, Stuttgart, 1093–1095. Fortenbaugh, W.W. et al. (eds.) 1992. Theophrastus of Eresus. Sources for his Life, Writings, Thought and Influence, Leiden. Fortenbaugh, W.W./Schütrumpf, E. 2000. Demetrius of Phalerum. Text, Translation, and Discussion, Rutgers Univ. St. in Class. Humanit. (RUSCH) 9, New Brunswick/London. Fortenbaugh, W.W./Schütrumpf, E. 2001. Dicaearchus of Messana. Text, Translation, and Discussion, New Brunswick/London. Fournet, J.L. 2009. Alexandrie: une communauté linguistique? Ou la question du grec alexandrin, IFAO, Le Caire. Fraenkel, E. 1954. ‘Vermutungen zum Aetna-Festspiel des Aischylos’, Eranos 52, 61–75. Fraser, P.M. 1972. Ptolemaic Alexandria, Oxford. Frede, M. 1978. ‘Principles of Stoic Grammar’, in: J.M. Rist (ed.), The Stoics, Berkeley/Los Angeles/London, 27–75. Fredrich, C./Wentzel, G. 1986. ‘Anecdota aus einer athenischen Handschrift’, Nachrichten von der Königl. Gesellschaft der Wissenschaften zu Göttingen, Philologisch-historische Klasse, 309–340. Friedlaender, L. 1850. Nicanoris Περὶ Ἰλιακῆς στιγμῆς reliquiae emendatiores, Regiomonti. Friedlaender, L. 1853. Aristonici Περì σημείων Ἰλιάδος reliquiae, Gottingae. Frye, N. 1957. Anatomy of Criticism, Princeton. Führer, R./Schmidt, M. 2001. ‘Homerus redivivus’, GGA 253, 1–32. Fuhrmann, M. 1960. Das systematische Lehrbuch, Göttingen. Fuhrmann, M. 1967. ‘Xenon (15)’, RE IX a, 2, 1540. Funghi, M.S. 1983. ‘P. Brux. Inv. E 7162 e P. Med. Inv. 71.82. Due discussioni su Olimpo’, PP 208, 11–19. Funghi, M.S. 1995. ‘P. Derveni’, in: Corpus dei Papiri Filosofici greci e latini (CPF), vol. III: Commentari, Firenze 1995, 565–585. Fusillo, M. 1993. “Apollonio Rodio’, in: Cambiano/Canfora/Lanza 1993, 107–144. Fusillo, M./Hurst, A./Paduano, G. 1991. Licofrone, Alessandra/Milano. Gabba, E. 1974. ‘Storiografia greca e imperialismo romano (III-I secolo a.C.)’, RSI 86, 625–642 (= Aspetti culturali dell’imperialismo romano, Florence, 1993, 17 ff.). Gabba, E. 1984. ‘Scienza e potere nel mondo ellenistico’, in: G. Giannantoni/M. Vegetti (eds.), La scienza ellenistica, Napoli, 13–37. Gaede, R. 1880. Demetrii Scepsii quae supersunt, Diss. Gryphiswaldiae. Gaisford, T. 1848. Etymologicum Magnum, Oxonii. Gallavotti, C. 1933. ‘Nuove hypotheseis di drammi euripidei’, RFIC 61, 177–188.

  Bibliography Gallavotti, C. 1969. ‘Tracce della Poetica di Aristotele negli scolii omerici’, Maia 21, 203–214. Gallavotti, C. 1974. Aristotele, Dell’arte poetica, Milano. Gallo, I. 1975. Frammenti biografici da papiri, I, Roma. Gallo, I. 1992. ‘Hermippus F1’, CPF I 1**, Firenze, 249–257. Gallo, I./Nicastri, L. (ed.) 1995. Biografìa e autobiografia degli antichi e dei moderni, Napoli. Gangutia, E. 1986. ‘Further Comments on P.Oxy. 2888’, Philologus 130, 187–190. Garofalo, I./Lami, A. 2012. Galeno. L’anima e il dolore. De indolentia. De propriis placitis, Milano. Gärtner, H. 1972. ‘Palamedes (n. 2)’, Der Kleine Pauly, 419. Gärtner, Th. 1998. ‘Eine Verbesserung zum Text des D-Scholions zu Hom. Il. Ψ 142’, ZPE 122, 6. Geffcken, J. 1936. ‘Nikias Nr. 24’, RE XVII 1, 335–336. Gennari Santori, C. 2018. ‘Esiodo interprete di Omero: alcuni esempi di esegesi antica’, SemRom n.s. 7, 153–174. Gentili, B. Cirene. 1990. Storia, mito, letteratura, Urbino. Gentili, B./Bernardini, P.A./Cingano, E./Giannini, P. 1995. Pindaro, Le Pitiche, Milano. Gentili, B./Catenacci, C./Giannini, P./Lomiento, L. 2013. Pindaro, Le Olimpiche, Milano. Gerstinger, H. 1925. ‘Bruchstücke eines antiken Kommentars zur Archäologie des Thukydides im Papyr. gr. Vindob. 29247’, Denkschr. Wien. Akad., 67/2, 1–20. Geymonat, M. 1974. Scholia in Nicandri Alexipharmaca, Milano. Geymonat, M. 1989. ‘Vittorio de Marco’, Gnomon 61, 188–190. Giangrande, G. 1962. ‘On the origins of the Greek romance: the birth of a literary form’, Eranos 60, 132–159. Giannantoni, G. 1990. Socratis et Socraticorum reliquiae, 4 vols., Napoli. Gigante Lanzara, V. 2000. Licofrone, Alessandra, Milano. Gignac, F.T. 1976. A Grammar of the Greek Papyri of the Roman and Byzantine Periods I. Phonology, Milano. Gigon, O. (ed.) 1987. Aristotelis opera, vol. III, Librorum deperditorum fragmenta, Berlin. Giordano, D. 21990. Chamaeleontis Heracleotae Fragmenta, Bologna. Gostoli, A. 1986. ‘La figura dell’aedo preomerico nella filologia peripatetica ed ellenistica: Demodoco tra mito e storia’, in: G. Cerri (ed.), Scrivere e recitare. Modelli di trasmissione del testo poetico nell’antichità e nel medioevo, Roma, 103–126. Gottschalk, H.B. 1980. Heraclides of Pontus, Oxford. Gow, A.S.F./Page, D.L. (eds.) 1965. The Greek Anthology Hellenistic Epigrams, 2 vols., Cambridge. Gräfenhan, A. 1843–1850. Geschichte der Klassischen Philologie im Altertum, Bonn. Grandolini, S. 1981. ‘Nota all’edizione moscopulea dell’Iliade’, GIF 33, 251–253. Grandolini, S. 1991. Manuelis Moschopouli Commentarium in Hesiodi Opera et Dies, Roma. Grauert, W.H. 1827. ‘Über die homerischen Chorizonten’, RhM 1, 199–211. Graziosi, B. 2002. Inventing Homer. The Early Reception of Epic, Cambridge. Greene, W.C. 1938. Scholia Platonica, Haverfordiae in Civitate Pennsylvaniae. Grenfell, B.P./Hunt, A.S. (eds.) 1899. The Oxyrhynchus Papyri, Part II, London. Grenfell, B.P./Hunt, A.S. (eds.) 1901. The Amherst Papyri. Part II. Classical Fragments and Documents of the Ptolemaic, Roman and Byzantine Periods, Nos. 10–201, London. Grenfell, B.P./Hunt, A.S. (eds.) 1908. The Oxyrhynchus Papyri, Part VI, London. Grenfell, B.P./Hunt, A.S. (eds.) 1922. The Oxyrhynchus Papyri, Part XV, London.

Bibliography  

Grisolia, R. 1992. ‘Critica letteraria e citazioni poetiche negli Scholia Vetera ai tragici greci’, in: A. De Vivo/L. Spina (eds.), ‘Come dice il poeta…’. Percorsi greci e latini di parole poetiche, Napoli, 43–58. Grisolia, R. 2001. Oikonomia. Struttura e tecnica drammatica negli scoli antichi ai testi drammatici, Napoli. Groningen, B.A. van. 1963. Ἔκδοσις, Mnemosyne 16, 1–17. Grossmann, H. 1887. De doctrinae metricae reliquis ab Eustathio servatis, Argentorati. Gudeman, A. 1912. ‘Hellanikos (8)’, RE VIII 1, 153–155. Gudeman, A. 1919. ‘Kallistratos (38)’, RE X, 1737–1748. Gudeman, A. 1921. ‘Scholien’, RE II A/1, 625–705. Gudeman, A. 1921a. ‘Satyros Nr. 16’, RE II A/1, 228–235. Gudeman, A. 1922. ‘Kritische Zeichen’, RE XI, 1916–1927. Gudeman, A. 1925. ‘Leogoras (3)’, RE XII, 2000. Gudeman, A. 1927. ‘Lysanias (8)’, RE XIII 2, 2510. Gudeman, A. 1934. Aristoteles, Περὶ Ποιητικῆς, Berlin/Leipzig. Guhl, C. 1969. Die Fragmente des Alexandrinischen Grammatikers Theon, Diss. Hamburg. Gulick, C.B. 1928. The Deipnosophists I–II, London/New York. Haas, W. 1977. Die Fragmente der Grammatiker Tyrannion und Diokles, SGLG 3, Berlin/New York. Hahn, R. 1967. Die Allegorie in der antiken Rhetorik, Diss. Tübingen. Hainsworth, B. 1993. The Iliad: A Commentary, vol. III: Books 9–12, Cambridge. Hammerstaedt, J. 1989. ‘3838. Aeschylus. Prometheus Vinctus 123–132’, in: M.G. Sirivianou et al. (eds.), The Oxyrhynchus Papyri, Vol. LVI, London, 69–70. Hanschke, P. 1914. De Accentuum Graecorum Nominibus, Bonn. Hanson, A.E. 1997. ‘Fragmentation and the Greek Medical Writers’, in: G.W. Most (ed.), Collecting Fragments = Fragmente Sammeln, Aporemata 1, Göttingen, 289–314. Haslam, M.W. 1975. ‘The Authenticity of Euripides, Phoenissae 1–2 and Sophocles, Electra 1’, GRBS 16, 149–174. Haslam, M.W. 1978. ‘The Versification of the New Stesichorus (P.Lille 76abc)’, GRBS 19, 29–57. Haslam, M.W. 1982. ‘A New Papyrus Text of Apollonius Sophista’, ZPE 49, 31–38. Haslam, M.W. 1985. ‘P. Wash. Univ. Inv. 217: Commentary on Iliad 19’, BASP 22, 97–100. Haslam, M.W. 1990. ‘A New Papyrus of the Mythographus Homericus’, BASP 27, 31–36. Haslam, M.W. 1994. ‘The Homer Lexicon of Apollonius Sophista I. Composition and Constituents; II. Identity and Transmission’, CPh 89, 1–45; 107–118. Haslam, M.W. 1996. ‘On P. Oxy. LXI 4096, Mythographus Homericus’, ZPE 110, 115–117. Haslam, M.W. 1997. ‘Homeric Papyri and Transmission of the Text’, in: I. Morris/B. Powell (eds.), A New Companion to Homer, Leiden/New York/Koln, 55–100. Haslam, M.W./Montanari, F. 1983. ‘Commentary on Odyssey 21, (P. Fayum 312)’, BASP 20, 113– 122 (= this volume, ch. 18). Heitsch, E. 1960. ‘Überlieferung und Deutung. Philologische Überlegungen zum Traditionsproblem’, A&A 9, 19–38. Heitsch, E. 1961–1964. Die griechischen Dichterfragmente der römischen Kaiserzeit, Göttingen. Helck, I. 1905. De Cratetis Mallotae studiis criticis quae ad Iliadem spectant, Lipsiae. Helck, I. 1914. De Cratetis Mallotae studiis criticis quae ad Odysseam spectant, Progr. Dresden. Hemmerdinger, B. 1981. Les manuscrits d’Hérodote et la critique verbale, Genova. Henrichs, A. 1971. ‘Scholia Minora zu Homer. Prolegomena’, ZPE 7, 97–149. Henrichs, A. 1971b. ‘Scholia Minora zu Homer II’, ZPE 7, 229–260.

  Bibliography Henrichs, A. 1971c. ‘Scholia Minora zu Homer III: Nr. 8: Glossar zu Od. 1. 67–69, 79–116’, ZPE 8, 1–12. Henrichs, A. 1973. ‘Scholia Minora zu Homer IV’, ZPE 12, 17–43. Henrichs, A./Müller, W. 1976. ‘Apollonios Sophistes, Homerlexicon (P. Berol. inv. 16705)’, in: Collectanea Papyrologica in Honour of H.C. Youtie, Bonn, I 27–51. Henry, M. 1986. ‘The Derveni Commentator as Literary Critic’, TAPhA 116, 149–164. Hense, O. 1912. ‘Hephaistion’, RE VIII 1, 286–309. Herington, C.J. (ed.) 1972. The Older Scholia on the Prometheus Bound, Leiden. Heubeck, A. 1979. Schrift, Archaeologia Homerica 3.10, Göttingen. Heubeck, A. et al. 1983. Omero Odissea, vol. III: Libri IX–XII, Milano. Hilgard, A. 1901. Scholia in Dionysii Thracis Artem grammaticam, Grammatici Graeci 1.3, Lipsiae. Hiller, E. 1886. ‘Die Fragmente des Glaukos von Rhegion’, RhM 41, 398–436. Hintenlang, H. 1961. Untersuchungen zu den Homer-Aporien Aristotles, Diss. Heidelberg. Hirschberger, M. 2004. Gynaikôn Katalogos und Megalai Ehoiai. Ein Kommentar zu den Fragmenten zweier hesiodischer Epen, München/Leipzig. Hoerschelmann, W. 1881. ‘Untersuchungen zur Geschichte der griechischen Metriker. Die Composition der Hephaestio-Scholien’, RhM 36, 260–301. Hofmann, Ph. 1905. Aristarchs Studien «de cultu et victu heroum» im Anschluss an Karl Lehrs, Programm des Kgl. Ludwigs-Gymnasiums in München 1904/1905, München. Holwerda, D. 1987. ‘Kallistratos, Schüler des Aristophanes von Byzanz’, Mnemosyne 40, 148. Hombert, M./Préaux, Cl. 1937. ‘P. Bruxelles Inv. E 7162: Extraits homériques relatifs aux lieux célestes’, in: Mélanges É. Boisacq, Annuaire Instit. Philol. Hist. Orient. Slav. 5, Bruxelles, 493–497. Hombert, M./Préaux, Cl. 1951. ‘Une tablette homérique de la Bibliothèque Bodléenne’, in: Παγκάρπια. Mélanges Henri Grégoire, vol. III, Bruxelles, 162–168. Householder, F.W. 1981. The Syntax of Apollonius Dyscolus, Amsterdam. Hude, C. 1927. Scholia in Thucydidem, Lipsiae (repr. New York 1973). Humphreys, S. 1997. ‘Fragments, Fetishes, and Philosophies: Towards a History of Greek Historiography after Thucydides’, in: G.W. Most, Collecting Fragments = Fragmente Sammeln, Aporemata 1, Göttingen, 207–224. Hunger, H. 1967. ‘Palimpsest-Fragmente aus Herod. Vindob. Hist. gr. 10’, JÖB 16, 1–33. Hunt, A.S. (ed.) 1912. The Oxyrhynchus Papyri, Part IX, London. Hunter, R. 2009. ‘Hesiod’s Style: Towards an Ancient Analysis’, in: F. Montanari/A. Rengakos/ Chr. Tsagalis (eds.), Brill’s Companion to Hesiod, Leiden/Boston, 253–270. Hunter, R. 2015. ‘The Rhetorical Criticism of Homer’, in: Montanari/Rengakos/Tsagalis 2015, I 673–705. Hurst, A. 1994. ‘Un nouveau papyrus du premier hymne homérique; le papyrus de Genève’, in: A. Bülow-Jacobsen (ed.), Proceedings of the 20th International Congress of Papyrologists, Copenhagen, 317–321. Ieraci Bio, A.M. 1979. ‘Il concetto di paroimía: testimonianze antiche e tardo-antiche’, Rendic. Accad. Arch. Belle Arti 54, 185–214. Irigoin, J. 1952. Histoire du texte de Pindare, Paris. Irigoin, J. 1994. ‘Les éditions de textes’, in: Montanari 1994a, 39–82. Jacob, C. 1999. ‘Du livre au texte: pour une histoire comparée des philologies’, Quaderni di storia 50, 61–93 (= C. Jacob/J. Vale, ‘From Book to Text: Towards a Comparative History of Philologies’, Diogenes 47, 1999, 4–22).

Bibliography  

Jacoby, F. 1902. Apollodors Chronik, Berlin. Jacoby, F. 1904. Das Marmor Parium, Berlin. Jacoby, F. 1913. ‘Herodotos (7)’, RE Suppl. II, 205–520. Jacoby, F. 1921. ‘Kleitophon (3)’, RE XI, 661. Jacoby, F. 1929. ‘Karystios’, RE X.2, 2254–2255. Jacoby, F. 1930. Hesiodi Carmina, Berolini. Janko, R. 1981. ‘The structure of the Homeric Hymns: A Study in Genre’, Hermes 109, 9–24. Janko, R. 1982. Homer, Hesiod and the Hymns: Diachronic Development in Epic Diction, Cambridge. Janko, R. 1986. ‘The Shield of Heracles and the Legend of Cycnus’, CQ 36, 38–59. Janko, R. 1990. ‘The Iliad and his Editors: Dictation and Redaction’, CA 9, 326–334. Janko, R. 1992. The Iliad: A Commentary, Vol. IV: Books 13–16, Cambridge. Janko, R. 1998a. Review of G. Nagy, Poetry as Performance. Homer and Beyond, Cambridge, MA, 1996, JHS 118, 206–207. Janko, R. 1998b. Review of I. Morris/B. Powell (eds.), A New Companion to Homer, Leiden/New York/Köln 1997, BMCR 1998.05.20. Janko, R. 2000. Review of Homeri Ilias, rec. M.L. West, vol. I, Stutgardiae et Lipsiae, 1998– 2000, CR 50, 1–5. Janko, R. 2002. ‘Seduta di chiusura’, in: Montanari/Ascheri 2002, 653–671. Jeffery, L.H. 1967. ‘Ἀρχαῖα γράμματα: Some Ancient Greek Views’, in: Europa. Studien zur Geschichte und Epigraphik der frühen Aegaeis. Festschrift für E. Grumach, Berlin, 152–166. Jones, H.L. 1917. The Geography of Strabo, with an English translation by H.L.J., based in part upon the unfinished version of J.R. Sitlington Sterrett, London/Cambridge, MA. Jouanna, J. 2001. ‘La lecture de Sophocle dans les scholies: remarques sur les scholies anciennes d’Ajax’, in: A. Billault/Chr. Mauduit (eds.), Lectures antiques de la tragédie grecque, Paris, 9–26. Jouanna, J. 2007. Sophocle, Paris. Kaibel, G. 1907. ‘Eubulos (14)’, RE VI 1, 877–878. Kakridis, J.Th. 1958. ‘Classical Tragedy in the Light of New Texts’, in: Acta Congressus Madvigiani Hafniae 1954, I 141–153. Kannicht, R. 2004. Tragicorum Graecorum Fragmenta, vol. V, Euripides, Göttingen [= TrGF 5]. Kassel, R. 1985. ‘Antimachos in der Vita Chisiana des Dionysios Periegetes’, in: Catalepton, Festschr. B. Wyss, Basel, 69–76. Kassel, R. 1985a. ‘Hypothesis’, in: W.J. Aerts et al. (eds.), Σχόλια. Studia ad criticam interpretationemque textuum Graecorum et ad historiam iuris Graeco-Romani pertinentia D. Holwerda oblata, Groningen, 53–59. Keaney, J.J. 1991. Harpocration, Lexeis of the Ten Orators, Amsterdam. Kennedy, G.A. (ed.) 1989. The Cambridge History of Literary Criticism, vol. I, Cambridge. Kenyon, F. 1893. ‘Brit. Nus. Pap. CXXVIII (Il. XXIII, XXIV)’, JPh 21, 296–343. Kenyon, F. 1894. ‘Two new MSS. in the British Museum’, JPh 22, 238–261. Kern, O. 1903. ‘Dodona’, RE V 1, 1257–1265. Kidd, D. 1997. Aratus. Phaenomena, Cambridge. Kindstrand, J.F. 1978. ‘The Greek Concept of Proverbs’, Eranos 76, 71–85. Kirk, G. 1985. The Iliad. A Commentary, vol. I: Books 1–4, Cambridge. Kirk, G.S. 1990. The Iliad. A Commentary, vol II: Books 5–8, Cambridge. Körte, A. 1918. ‘Bacchylides’, Hermes 53, 113–147. Körte, A. 1941. ‘Literarische Texte mit Ausschluß der christlichen’, APF 14, 103–238.

  Bibliography Kohl, J.W. 1917. De Chorizontibus, Diss. Giessen, Darmstadt. Kohl, J.W. 1921. ‘Die homerische Frage der Chorizonten’, Neue Jahrbücher für das Klassische Altertum, Geschichte und Deutsche Literatur 24, 198–214. Korzeniewski, D. 1968. Griechische Metrik, Darmstadt. Koster, W.J.W. 1960. Scholia in Aristophanem. Io. Tzetzae commentarii in Aristophanem, IV 1: Prolegomena et commentarium in Plutum, Groningen. Kotzia, P./Sotiroudis, P. 2010. ‘Galenou, Perì alypías’, Hellenica 60, 63–150. Kouremenos, Th./Parassoglou, G.M./Tsantsanoglou, K. 2006. The Derveni Papyrus, Studi e testi per il Corpus dei Papiri Filosofici 13, Firenze. Kramer, B./Hagedorn, D. 1984. Griechische Papyri der Staats- und Universitätsbibliothek Hamburg, Bonn. Kramer, B./Hagedorn, D. 1976. Kölner Papyri (P.Köln), vol. 1, Opladen. Kranz, W. 1933. Stasimon, Berlin. Kranz, W. 1958. Kosmos, Bonn. Kroll, W. 1895. ‘Heron Nr. 4’, RE VIII 1, 992. Kroll, W. 1922. ‘Krates von Mallos’, RE XI 2, 1634–1641. Kroll, W. 1936. ‘Nikandros Nr. 11’, RE XVII 1, 262. Kroll, W. 1937. ‘Peleiades’, RE XIX 1, 266–267. Kroll, W. 1940. ‘Demetrios Nr. 100a’, RE Suppl. VII, 124. Kuch, H. 1965. Philologos. Untersuchung eines Wortes von seinem ersten Auftreten in der Tradition bis zur ersten überlieferten lexikalischen Festlegung, Berlin. Kuchenmüller, G. 1928. Philetae Coi reliquiae, Berolini. Lallot, J. 1989. La grammaire de Denys le Thrace, Paris. Lama, M. 1991. ‘Aspetti di tecnica libraria ad Ossirinco: copie letterarie su rotoli documentari’, Aegyptus 71, 55–120. Lamberton, R. 1986. Homer the Theologian. Neoplatonist Allegorical Reading and the Growth of the Epic Tradition, Berkeley/Los Angeles. Lamberton, R. 1988. Hesiod, New Haven/London. Lamberton, R./Keaney, J.J. 1992. Homer’s Ancient Readers. The Hermeneutics of Greek Epic’s Earliest Exegetes, Princeton. Lamedica, A. 1991. ‘La terminologia critico-letteraria dal Papiro di Derveni ai corpora scoliografici’, in: P. Radici Colace/M. Caccamo Caltabiano (eds.), Atti del I seminario di studi sui lessici tecnici greci e latini, Messina, 83–91. Lamedica, A. 1992. ‘Il papiro di Derveni come commentario. Problemi formali’, in: A.H.S. ElMosallamy, Proceedings of the XIXth International Congress of Papyrology, Cairo, 325– 333. Lanata, G. 1963. Poetica pre-platonica. Testimonianze e frammenti, Firenze. Lanza, D. 1987. Aristotele, Poetica, Milan. La Penna, A. 1985. L’ermeneutica letteraria, in: M. Vegetti (ed.), Il sapere degli antichi, Milano, 45–62. Lapini, W. 2007. Capitoli su Posidippo, Alessandria. La Roche, J. 1866. Homerische Textkritik im Altertum, Leipzig. Lasserre, F. 1975. ‘L’élégie de l’huître (P. Louvre inv. 7733 v° inéd.)’, QUCC 19, 145–176. Lasserre, F. 1981. ‘Telephos Nr. 3’, Der Kleine Pauly, 3, 1217–1218. Lasserre, F. 1986–1987. ‘Aristoph. Byz. Fragm. Neglecta’, MCr 21–22, 303–315. Latacz, J. 2002. Homers Ilias. Gesamtkommentar, vol. I 1.2, München/Leipzig. Latacz, J. 2003. Homers Ilias. Gesamtkommentar, vol. II 2, München/Leipzig.

Bibliography  

Latte, K. 1915. ‘Zur Zeitbestimmung des Antiatticista’, Hermes 50, 373–394. Latte, K. 1925. ‘Glossographika’, Philologus 80, 136–175 (= Kleine Schriften, München 1968, 631–666). Latte, K. 1948. ‘De nonnullis papyris Oxyrhynchiis. 1. De Hipponactis Epodo’, Philologus 97, 37–47 (= Kleine Schriften 468–471). Latte, K. 1953–1966. Hesychii Alexandrini Lexicon, 2 vols., Hauniae. Latte, K./Erbse, H. 1965. Lexica Graeca Minora, Hildesheim. Lausberg, H. 1973. Handbuch der literarischen Rhetorik, München. Leaf, W. 1900–19022. The Iliad, London. Lee, G. 1975. ‘An Aristarchean Maxim’, PCPhS 21, 63–64. Leeuwen, J. van. 19182. Enchiridium dictionis epicae, Lugduni Batavorum. Lefkowitz, M.R. 1985. ‘The Pindar Scholia’, AJPh 106, 269–282. Lehnus, L. 1993. ‘Callimaco tra la polis e il regno’, in: Cambiano/Canfora/Lanza 1993, 75–106. Lehrs, K. 1848. Herodiani scripta tria emendatiora, Königsberg. Lehrs, K. 1873. Die Pindarscholien, Leipzig. Lehrs, K. 18823. De Aristarchi studiis Homericis, Lipsiae. Lenz, F. 1934. Untersuchungen zu den Aristeidesscholien, Berlin. Leo, F. 1894. ‘Hesiodea’, Index Scholarum, Göttingen, 1–22 (= F. L., Ausgewählte Kleine Schriften, Rome 1960, vol. II, 343–363). Leo, F. 1899. ‘Ein metrisches Fragment aus Oxyrhynchos’, NGG, 495–507. Leo, F. 1901. Die griechisch-römische Biographie nach ihrer literarischen Form, Leipzig. Leo, F. 1960. Ausgewählte Kleine Schriften, 2 vols., Roma. Leone, P. 1977. Eschine, Oratori Attici Minori, vol. I, Torino. Lesky, A. 1971. Geschichte der griechischen Literatur, Bern/München. Lesky, A. 19723. Die tragische Dichtung der Hellenen, Göttingen. Leszl, W. 1985. ‘Linguaggio e discorso’, in: M. Vegetti (ed.), Introduzione alle culture antiche II. Il sapere degli antichi, Torino, 13–44. Leurini, L. 1991. ‘Le Ἐπιδημίαι di Ione di Chio’, AFCL 49, 99–118. Leurini, L. 2000. Ionis Chii Testimonia et Fragmenta, Amsterdam. Leurini, L. 2006. ‘Ione di Chio 1960–2005’, Lustrum 48, 7–44. Leutsch, E.L. von/ Schneidewin, F.G. 1839–1851. Corpus Paroemiographorum Graecorum, 2 vols., Gottingae. Leutsch, E.L. von/Schneidewin, F.G. 1961. Corpus Paroemiographorum Graecorum Supplementum, Hildesheim. Linke, K. 1977. Die Fragmente des Grammatikers Dionysios Thrax, SGLG 3, Berlin/New York. Lloyd, A.C. 1971. ‘Grammar and Metaphysics in the Stoa’, in: A.A. Long (ed.), Problems in Stoicism, London, 58–74. Lloyd, G.E.R. 1966. Polarity and Analogy, Cambridge. Lloyd-Jones, H. 1994. ‘Alexander Aetolus, Aristophanes and the Life of Euripides (Alexander Aetolus fr. 7 Powell, Aristophanes fr. 676 b Kock)’, in: Storia, poesia e pensiero nel mondo antico. Studi in onore di Marcello Gigante, Napoli, 371–379. Lloyd-Jones, H./Wilson, N.G. (eds.) 1983. Supplementum Hellenisticum, Berlin/New York. Lloyd-Jones, H. (ed.) 2005. Supplementum Supplementi Hellenistici, Berlin/New York. Lloyd-Jones, H./Wilson, N.G. (eds.) 1990. Sophoclis fabulae, Oxonii. Lobel, E. et al. (eds.) 1941. The Oxyrhynchus Papyri, Part XVIII, London. Lobel, E. et al. (eds.) 1952. The Oxyrhynchus Papyri, Part XX, London. Lobel, E. et al. (eds.) 1954. The Oxyrhynchus Papyri, Part XXII, London.

  Bibliography Lomanto, V./Marinone, N. 1990. Index Grammaticus. An Index to the Latin Grammarians, Alpha-Omega LXXXI, 3 vols., Hildesheim. Long, A.A. 1992. ‘Stoic Readings of Homer’, in: Lamberton/Keaney 1992, 41–66 (repr. in A.A. Long, Stoic Studies, Cambridge 1996, 58–85). Lossau, M.J. 1964. Untersuchungen zur antiken Demosthenesexegese, Berlin/Zürich. Lucas, D.W. 1968. 19722. Aristotle, Poetics, Oxford. Ludwich, A. 1884–1885. Aristarchs Homerische Textkritik, Leipzig (repr. Hildesheim/New York, 1971). Ludwich, A. 1888–1890. Scholia in Homeri Odysseae A 1–309 auctiora et emendatiora, Königsberg (repr. Hildesheim 1966 with a preface by H. Erbse). Ludwich, A. 1902. Über die Papyrus-Commentare zu den homerischen Gedichten, Regimontii Prussorum. Ludwich, A. 1902–1907. Homeri Ilias, recensuit A.L., Leipzig. Ludwich, A. 1914. ‘Die Quellenberichte über Aristarchs Ilias-Athetesen’, RhM 69, 680–734. Ludwich, A. 1917. ‘Über die Homerischen Glossen Apions’, Philologus 74, 205–247. Ludwich, A. 1918. ‘Über die Homerischen Glossen Apions’, Philologus 75, 95–127. Lührs, D. 1992. Untersuchungen zu den Athetesen Aristarchs in der Ilias und zu ihrer Behandlung im Corpus der exegetischen Scholien, Beiträge zur Altertumswissenschaft 11, Hildesheim/Zürich/New York. Lundon, J. 2011. ‘Homeric Commentaries on Papyrus: A Survey’, in: Matthaios/Montanari/ Rengakos 2011, 159–179. Lünstedt, P. 1961. Untersuchungen zu den mythologischen Abschnitten der D-Scholien, Diss. Hamburg. Luppe, W. 1967. ‘Zu einigen Kratinosfragmenten’, Wiss. Zeitschr. Univ. Halle, XVI, 397–409. Luppe, W. 1977. ‘P. Oxy. 3160 + P. Strasb. 1401 - Bruchstücke derselben OdysseeErläuterungen’, ZPE 27, 101–106. Luppe, W. 1980. ‘Die Odyssee-Hypotheseis im P. Ryl. 23 (Nr. 1208 Pack)’, APF 27, 33–36. Luppe, W. 1980. ‘Die Hypotheseis zu Euripides’ Alkestis und Aiolos P. Oxy. 2457’, Philologus 126, 10–18. Luppe, W. 1984. ‘Zum Tennes-Mythos im ‘Mythographus Homericus’ P.Hamb. 199’, ZPE 56, 31–32. Luppe, W. 1985. ‘Achilleus bei Lykomedes in P. Berol. 13930 (Pack 1203)’, AFP 31, 5–12. Luppe, W. 1985a. ‘P.Vindob. G 29779 – ein Sophokles-Kodex’, WS 19, 89–104. Luppe, W. 1990. ‘Dikaiarchos und der “Rhesos’-Prolog”’, ZPE 84, 11–13. Luppe, W. 1992. ‘Homer-Erläuterungen zu Ξ 316–348’, ZPE 93, 163–165. Luppe, W. 1995. ‘Mythographus Homericus (P. Oxy. 4096 fr. 5 Kol. I und fr. 6, Ἀπόλλων νόμος)’, Analecta Papyrologica 7, 15–18. Luppe, W. 1996a. ‘Mythographus Homericus P. Oxy. 4096 fr. 10’, ZPE 112, 25–28. Luppe, W. 1996b. ‘Die Ikarios-Sage im Mythographus Homericus’, ZPE 112, 29–33. Luppe, W. 1996c. ‘Mythographus Homericus P. Oxy. 4096 fr. 11’, Hyperboreus 2, 189–193. Luppe, W. 1996d. ‘Neileus’ ἀποικία nach Milet. Mythographus Homericus P. Oxy. 4096 fr. 3’, Eikasmos 7, 207–210. Luppe, W. 1996e. ‘Ein neues Fragment des Mythographus Homericus zu Ψ 91–92’, Prometheus 22, 97–100. Luppe, W. 1996/1997. ‘Ein Zeugnis für die Niobe-Sage in P. Oxy. 4096’, Würzburger Jahrbücher für die Altertumswissenschaft 21, 153–159.

Bibliography  

Luppe, W. 1997. ‘Nachlese und Überlegungen zum Mythographus-Homericus-Codex P.S.I. 1173’, ZPE 116, 13–18. Luppe, W. 1997a. ‘Die Psamathe- und die Plejaden-Sage im Mythographus Homericus P. Oxy. LXI 4096 fr. 1 und fr. 2 Kol. II’, APF 43, 1–6. Luppe, W. 2001. ‘Neues aus Papyrus-Hypotheseis zu verlorenen Euripides-Dramen’, in: W.W. Fortenbaugh/E. Schütrumpf (eds.), Dicaearchus of Messana. Text, Translation and Discussion, New Brunswick, 329–341. Luppe, W. 2002. ‘Σχόλια, ὑπομνήματα und ὑποθέσεις zu griechischen Dramen auf Papyri’, in: W. Geerlings/C. Schulze (eds.), Der Kommentar in Antike und Mittelalter. Beiträge zu seiner Erforschung, Leiden, 55–77. Luppe, W./Montanari, F. 1992. ‘Hypotheseis dei drammi di Euripide e Sofocle’, in: Corpus dei Papiri Filosofici greci e latini [= CPF], Firenze, I 1**, 32–33. Luschnat, O. 1954. ‘Die Thukydidesscholien. Zu ihrer handschriftlichen Grundlage, Herkunft und Geschichte’, Philologus 98, 14–58. Luzzatto, M.T. 1988. ‘L’oratoria, la retorica e la critica letteraria dalle origini ad Ermogene’, in: F. Montanari (ed.), Da Omero agli Alessandrini. Problemi e figure della letteratura greca, Roma, 207–256. Lyle, E. 1984. ‘The Circus as Cosmos’, Latomus 43, 827–841. Maass, E. 1884. ‘Die Iliasscholien des Codex Leidensis’, Hermes 19, 264–289. Maehler, H. 1994. ‘Die Scholien der Papyri in ihrem Verhältnis zu den Scholiencorpus der Handschriften’, in: Montanari 1994, 95–127. Maes, M. 1939. Contribution à l’étude du commentaire de Plutarque aux Travaux et Jours d’Hésiode, Liège. Maffei, M. 1976. Antike Diskussionen über die Episode von Glaukos und Diomedes im VI. Buch der Ilias, Meisenheim. Magnelli, E. 1999. Alexandri Aetoli testimonia et fragmenta, Firenze. Mai, A. 1841. Spicilegium Romanum, vol. V, Romae. Maltese, E. 1982. Sofocle, Ichneutae, Firenze. Manetti, D. 2006. ‘La terminologie du livre: à propos des emploi d’hyphos et edaphos dans deux passages de Galien’, REG 119, 156–171. Manetti, D. 2012a. ‘Galeno e la ekdosis di Ippocrate: una nota testuale’, in: G. Bastianini/ W. Lapini/M. Tulli (eds.), Harmonia. Scritti di filologia classica in onore di Angelo Casanova, Firenze, 475–481. Manetti, D. 2012b. ‘Galeno ΠΕΡΙ ΑΛΥΠΙΑΣ e il difficile equilibrio dei filologi’, in: D. M. (ed.), Studi sul De indolentia di Galeno, Pisa, 9–22. Manetti, D. 2015. “Medicine and Exegesis’, in: Montanari/Matthaios/Rengakos 2015, II 1126– 1215. Manetti, D./Roselli, A. 1994. ‘Galeno commentatore di Ippocrate’, Aufstieg und Niedergang der Römischen Welt 37.2, 1529–1635. Manfredi, M. (ed.) 2000. Papiri dell’Iliade, Firenze. Manfredini, M. 1975. ‘Gli scolii a Plutarco di Areta di Cesarea’, Syculorum Gymnasium 28, 337– 350. Manfredini, M. 1979. ‘Gli scolî alle Vite di Plutarco’, JÖB 28, 83–119. Maniaci, M. 2006. ‘Issues with the mise en page of manuscripts with framing commentary. An example of several textual witnesses of the Iliad’, S&T 4, 211–297. Manuli, P. 1984. ‘Lo stile del commento. Galeno e la tradizione ippocratica’, in: G. Giannantoni/ M. Vegetti (eds.), La scienza ellenistica, Napoli, 375–394.

  Bibliography Marcovich, M. 1976. ‘P.Louvre inv. 7733v’, ZPE 23, 219–220. Mariotti, I. 1966. Aristone d’Alessandria, Bologna. Martano, A. 2002. ‘Scolii e glosse allo Scudo di Eracle dal manoscritto Ambrosiano C 222 Inf.’, Aevum 76, 151–200. Martano, A. 2004a. ‘Hieronymus ἐν ὑπομνήματι Ἀσπίδος Ἡσιόδου’, in: W.W. Fortenbaugh/ S. White (eds.), Lyco of Troas and Hieronymus of Rhodes, New Brunswick/London, 457– 474. Martano, A. 2004b. ‘Brevi note a P.Oxy. 4652’, Aegyptus 84, 129–136. Martano, A. 2005a. ‘La tradizione manoscritta dell’esegesi antica allo Scudo di Eracle esiodeo. La famiglia del Vat. gr. 1332 (sec. XIII–XV) ’, Aevum 78, 461–489. Martano, A. 2005b. ‘La tradizione manoscritta dell’esegesi antica allo Scudo di Eracle esiodeo. I codici Vat. Pal. gr. 425, Vallicell. F 16, Leid. Vulc. gr. 23, Laur. Conv. Soppr. 158 e Par. gr. 2833 (sec. XIV–XVI)’, Aevum 80, 461–489. Martano, A. 2006. ‘L’esegesi antica allo Scudo di Eracle nell’Etymologicum Genuinum e Gudianum’, in: Avezzù/Scattolin 2006, 85–120. Martano, A. 2012. ‘Chamaeleon of Heraclea Pontica: the Sources, Text and Translation’, in: A. Martano/E. Matelli/D. Mirhady (eds.), Praxiphanes of Mytilene and Chamaeleon of Heraclea. Text, Translation, and Discussion, Rutgers University Studies in Classical Humanities 18, New Brunswick/London, 157–337. Martin, Jean. . Histoire du texte des Phénomènes d’Aratos, Paris. Martin, Jean. . Scholia in Aratum vetera, Stuttgardiae. Martin, Jean. 1998. Aratos, Phénomènes. 2 vols., Paris. Martin, Josef. 1974. Antike Rhetorik, München. Martin, R. 2005. ‘Pulp Epic: The Catalogue and the Shield’, in: R.L. Hunter (ed.), The Hesiodic Catalogue of Women. Constructions and Reconstructions, Cambridge, 153–175. Martin, V./de Budé, G. 1928. Eschine, Discours, vol. II, Paris. Martinazzoli, F. 1953. Hapax legomenon, vol. I 1, Roma. Martini, E. 1901. ‘Demetrios 85’, RE IV 2, 2817–2841. Marzillo, P. 2009. Der Kommentar des Proklos zu Hesiods Werken und Tagen, Tübingen. Marzullo, B. 1968. ‘La «coppia contigua» in Esichio’, QIFG 3, 70–87 (= La «coppia contigua» nella glossografia di Esichio, in: AA.VV., Studia Classica et Orientalia Antonino Pagliaro oblata, I, Roma 1969, 85–105). Massimilla, G. 1996. Callimaco, Aitia. Libri primo e secondo, Pisa. Masson, O. 1962. Les fragments du poète Hipponax. Édition critique et commentée, Paris. Matijašić, I. 2013. ‘Aristarco ed Erodoto in Stefano di Bisanzio’, RhM CLVI, 217–220. Matranga, P. 1850. Anecdota Graeca e mss. bibliothecis Vaticana, Angelica, Barberiniana, Vallicelliana, Medicea, Vindobonensi deprompta, 2 vols., Romae. Matthaios, S. 1999. Untersuchungen zur Grammatik Aristarchs: Texte und Interpretationen zur Wortartenlehre, Hypomnemata 126, Göttingen. Matthaios, S. 2015. ‘Greek Scholarship in the Imperial Era and Late Antiquity’, in: Montanari/ Matthaios/Rengakos 2015, I 182–296. Matthaios, S./Montanari, F./Rengakos A. (eds.) 2011. Ancient Scholarship and Grammar: Archetypes, Concepts and Contexts, Trends in Classics Supplementary Volumes 8, Berlin. Mau, J. (ed.) 1961. Sexti Empirici opera, vol. III: Adversus Mathematicos, Lipsiae. Mazon, P. 1947. Homère Iliade, I. II (ch. VII–XII), Paris. McNamee, K. 1977. Marginalia and Commentaries in Greek Literary Papyri [Diss. Duke University], Durham, NC.

Bibliography  

McNamee, K. 1981. ‘Aristarchus and ‘Everyman’s’ Homer’, GRBS 22, 247–255. McNamee, K. 1981a. ‘Greek Literary Papyri Revised by Two or More Hands’, in: Proceedings of the Sixteenth International Congress of Papyrology, Chico, CA, 79–91. McNamee, K. 1981b. Abbreviations in Greek Literary Papyri and Ostraca, BASP Suppl. III, Chico, CA. McNamee, K. 1992. Sigla and Select Marginalia in Greek Literary Papyri, Bruxelles. McNamee, K. 2007. Annotations in Greek and Latin Texts from Egypt, Oakville. Meijering, R. 1985. ‘Aristophanes of Byzantium and Scholia on the Composition of the Dramatic Chorus’, in: ΣΧΟΛΙΑ. Studia … D. Holwerda oblata, Groningen, 91–102. Meijering, R. 1987. Literary and Rhetorical Theories in Greek Scholia, Groningen. Meillier, C. 1976. ‘Callimaque (P.L. 76 d, 78 a b c, 82, 84 et 111 c)’, CRIPEL 4, 261–286. Meillier, C. 1979. Callimaque et son temps. Recherches sur la carrière et la condition d’un écrivain à l’époque des premiers Lagides, Lille. Meillier, C. 1985. ‘Extraits commentés d’Homère, Odyssée, 16 et 17: P. Lille inv. 83 + 134 + 93b + 93a + 114t + 114o + 87’, in: F. Geus/F. Thill (eds.), Mélanges offerts à Jean Vercoutter, Paris, 229–238. Melandri, E. 1981. ‘Per un’edizione della Technologia di M. Moscopulo ad Hom. A - B 493’, Prometheus 7, 215–224. Melandri, E. 1983. ‘La parafrasi di M. Moscopulo ad Hom. A - B 493 e la tradizione esegetica e lessicografica dell’Iliade’, Prometheus 9, 177–192. Merkelbach, R. 1959. ‘Kritische Beiträge. 3. Ein Aeschines-Papyrus’, in: Studien zur Textgeschichte und Textkritik G. Jachmann ... gewidmet, Köln/Opladen, 162–164. Merkelbach, R. 19692. Untersuchungen zur Odyssee, München. Merkelbach, R./West, M.L. 1967. Fragmenta Hesiodea, Oxonii. Merritt Sale, W. 1984. ‘Homeric Olympus and its formulae’, AJPh 105, 1–28. Merro, G. 2015. ‘Ἰδίως φησί: singolarità del mito pindarico nell’esegesi antica’, SemRom n.s. 4, 213–234. Mertens, P. 1985. ‘Vingt années de papyrologie odysséenne’, CdÉ 60, 191–203. Messeri Savorelli, G./Pintaudi, R. 1997. ‘Frammenti di rotoli letterari laurenziani’, ZPE 115, 171– 177. Mette, H.J. 1931. De Cratete Mallota sive Pergameno, Diss. Berlin. Mette, H.J. 1936. Sphairopoiia. Untersuchungen zur Kosmologie des Krates von Pergamon, München. Mette, H.J. 1952. Parateresis. Untersuchungen zur Sprachtheorie des Krates von Pergamon, Halle/Saale. Mette, H.J. 1984. ‘Krates von Pergamon 1953-83’, Lustrum 26, 95–104. Michaelis, J. 1875. De Apollonii Rhodii fragmentis, Halis Saxonum. Micheli, G. 1979. ‘Macrocosmo / microcosmo’, in: Enciclopedia Einaudi, VIII, Torino, 637–659. Mirhady, D.C. 2001. ‘Dicaearchus of Messana: The Sources, Text and Translation’, in: Fortenbaugh/Schütrumpf 2001, 1–142. Mirhady, D. 2012. “Chamaeleon on Epic and Tragedy. Something to do with Dionysus”, in: A. Martano/E. Matelli/D. Mirhady (eds.), Praxiphanes of Mytilene and Chamaeleon of Heraclea. Text, Translation, and Discussion, Rutgers University Studies in Classical Humanities 18, New Brunswick/London, 387–409. Mirto, M.S. 1997. Omero, Iliade, trad. e introd. di Guido Paduano, comm. di M.S.M., Torino. Momigliano, A. 1950. ‘Ancient History and the Antiquarian’, Journal of the Warburg and Courtauld Institutes 13, 285–315 (= A.M., Contributo alla storia degli studi classici, Roma 1955, 67–106).

  Bibliography Monaco, M. 2000. ‘La tradizione manoscritta di Eschine tra i papiri e i codici medievali’, Aegyptus 80, 3–98. Montana, F. 1996. L’Athenaion Politeia di Aristotele negli scholia vetera ad Aristofane, Pisa. Montana, F. 2005. ‘Alexander [6] Cotiaeus’, LGGA. Montana, F. 2006. ‘Zwischen Philologie und Geschichte. Il contributo dei FGrHist all’edizione dei grammatici greci antichi’, in: C. Ampolo (ed.), Aspetti dell’opera di Felix Jacoby, Pisa, 201–226. Montana, F. 2006a. ‘Didymus [1]’, LGGA. Montana, F. 2006b. ‘Nicanor [3]’, LGGA. Montana, F. 2007. ‘Menandro (e Aristofane) ad Alessandria: qualche riflessione’, in: R. Pretagostini/E. Dettori (eds.), La cultura letteraria ellenistica. Persistenza, innovazione, trasmissione, Roma, 257–269. Montana, F. 2009. ‘Storici, filologi, storici-filologi: intersezioni nella cultura ellenistica’, in: F. Gazzano/G. Ottone/L. Santi Amantini (eds.), Ingenia Asiatica inclita per gentes fuere. Fortuna e tradizione di storici d’Asia Minore, Genova, 205–225. Montana, F. 2009a. ‘A hypomnema on Herodotus’ Bk. 5: notes to P.Oxy. LXV 4455, Col. I’, in: F. Montanari/S. Perrone (eds.), Fragments of the Past. Ancient Scholarship and Greek Papyri, TiC 1, 241–253 Montana, F. 2010. ‘The Making of Greek Scholiastic Corpora’, in: F. Montanari/L. Pagani (eds.), From Scholars to Scholia, Berlin/New York, 105–161. Montana, F. 2012. La filologia ellenistica. Lineamenti di una storia culturale, Pavia. Montana, F. 2012a. ‘Nuova luce su P.Amh. II 12, Col. I (hypomnema di Aristarco al libro I di Erodoto)’, ZPE 180, 72–76. Montana, F. 2013. ‘Didymus [2] Claudius’, LGGA. Montana, F. 2014. ‘Gli Omeri di Aristarco in fumo: Gal. indol. 13’, Prometheus 40, 262‒270. Montana, F. 2015. ‘Hellenistic Scholarship’, in: Montanari/Matthaios/Rengakos 2015, 60–183. Montana, F. 2017. ‘Dal Liceo al Museo, ultima frontiera’, RFIC 145, 443–473. Montana, F. 2018. ‘The oldest textual witness of John Tzetzes’ Exegesis of the Iliad’, in: M. Ercoles/L. Pagani/F. Pontani/G. Ucciardello (eds.), Approaches to Greek Poetry, Berlin/Boston, 107–132. Montana, F. 2019. ‘Herodotus 4: P.Amh. II 12. Aristarco, Commentario a I 171, 4?-215,2’, in: G. Bastianini et al. (eds.), CLGP – Commentaria et Lexica Graeca in Papyris reperta I 2.6, Berlin/Boston, 39–61. Montana, F. 2020. ‘Editing Anonymous Voices: The scholia vetera to the Iliad’, in: S. Boodts/ P. de Leemans/S. Schorn (eds.), Sicut dicit. Editing Ancient and Medieval Commentaries on Authoritative Texts, Turnhout. Montanari, F. 1974. ‘L’episodio eleusino delle peregrinazioni di Demetra. A proposito delle fonti di Ovidio, Fast. IV 502-62 e Metam. V 446-61’, ASNP, s. III 4, 109–137. Montanari, F. 1976. ‘Una glossa omerica pre-alessandrina? (Sch. Il. XVI 235 - Eubulo fr. 139 K.)’, Rendiconti Istituto Lombardo - Classe di Lettere 110, 202–211 (= this volume, ch. 27). Montanari, F. 1976a. ‘Omero, Eubulo, i pesci e i Chorizontes’, SCO XXV, 325–331 (= this volume, ch. 26). Montanari, F. 1976b. ‘Un nuovo frammento di commentario a Callimaco’, Athenaeum 54, 138– 150 (= this volume, ch. 57). Montanari, F. 1979. Studi di filologia omerica antica, I, Biblioteca di Studi Antichi 19, Pisa. Montanari, F. 1979a. ‘I prolegomena del codice Rom. Bibl. Naz. gr. 6 (C)’, in: Montanari 1979, 43–75.

Bibliography  

Montanari, F. 1979b. ‘Aristarco ad Odissea II 136–7. Appunti di filologia omerica antica’, MD 3, 157–170 (= this volume, ch. 28). Montanari, F. 1979c. ‘Termini e concetti della Poetica di Aristotele in uno scolio a Odissea IV 69’, SCO 29, 171–174. Montanari, F. 1980. ‘L’hypothesis di Odyss. β in P. Strasb. Gr. 1401 e in P. Oxy. XLIV 3160’, in: Miscellanea Papyrologica, Papyrologica Florentina 7, Firenze, 273–279. Montanari, F. 1981. ‘Sulle hypotheseis di Odyss. β e γ di P. Oxy. 3160 + P. Strasb. Gr. 1401’, SCO 31, 101–110 (= this volume, ch. 29). Montanari, F. 1981a. ‘Il grammatico Tolomeo Pindarione, i poemi omerici e la scrittura’, Ricerche di Filologia Classica I, Pisa, 97–114 (= this volume, ch. 30). Montanari, F. 1982. ‘P. Oxy. 574 verso riconsiderato: frammento di hypotheseis dell’Iliade’, ZPE 48, 89–92 (= this volume, ch. 31). Montanari, F. 1982a. ‘P. Pis. inv. 1 e P. Bonon. 6 in rapporto con le hypotheseis dei canti omerici’, Anagennesis 2, 273–284. Montanari, F. 1984. ‘Gli Homerica su papiro: per una distinzione di generi’, in: Ricerche di filologia classica, II, Pisa, 125–138 (= this volume, ch. 32). Montanari, F. 1985a. ‘Revisione di P. Berol. 13282. Le Historiae fabulares omeriche su papiro’, in: Atti del XVII Congresso internazionale di Papirologia, Napoli, II 229–242 (= this volume, ch. 20). Montanari, F. 1987. ‘Grammatici Greci’, in: Dizionario degli scrittori Greci e Latini, vol. II, Milano, 1093–1105. Montanari, F. 1987a. ‘Un «nuovo» papiro dell’Ecuba di Euripide (P. Tebt. 683 recto) e Epimetron sui papiri di Euripide’, RFIC 115, 24–32; 441–443. Montanari, F. 1988. I frammenti dei grammatici Agathokles, Hellanikos, Ptolemaios Epithetes. In appendice i grammatici Theophilos, Anaxagoras, Xenon, SGLG 7, Berlin/New York. Montanari, F. 1988b. ‘Filologia omerica antica nei papiri’, in: Proceedings of the XVIII Congress of Papyrology, Athens, II 337–344 (= this volume, ch. 33). Montanari, F. 1988c. ‘La burla degli dei. Antiche storie del comico e del riso’, Prefazione a La battaglia delle rane e dei topi. Batrachomyomachia, a cura di M. Fusillo, Milano, 9–17. Montanari, F. 1989. See CPF 1989, 403–418 (Ancient Authors, ch. 30). Montanari, F. 1991. ‘Tradurre dal greco in greco’, in: S. Nicosia (ed.), La traduzione dei testi classici. Teoria, prassi, storia, Napoli, 221–229. Montanari, F. 19922. Introduzione a Omero. Con un’appendice su Esiodo, Firenze (3rd. ed. Roma 2013). Montanari, F. 1992a. ‘Scoliografia e teatro greco: qualche appunto’, in: L. De Finis (ed.), Dal teatro greco al teatro rinascimentale: momenti e linee di evoluzione, Trento, 73–87 (= this volume, ch. 13). Montanari, F. 1993. ‘L’erudizione, la filologia, la grammatica’, in: Lo spazio letterario della Grecia antica, vol. I 2, Roma, 235–281 (= this volume, ch. 1). Montanari, F. 1993a. ‘Nuova edizione di P. Ryl. 536: Scholia Minora a Iliade N 198–562’, in: Mosaico. Studi in onore di Umberto Albini, Genova, 135–146. Montanari, F. 1993b. ‘Due note pindariche (Pitica 3, 43 e Nemea 7, 49)’, in: R. Pretagostini (ed.), Tradizione e innovazione nella cultura greca da Omero all’età ellenistica. Scritti in onore di B. Gentili, vol. II, Roma, 461–467 (= Ancient Authors, ch. 14). Montanari, F. 1993c. ‘L’Athenaion Politeia dai papiri alle edizioni, in: L.R. Cresci/L. Piccirilli (eds.), L’«Athenaion Politeia» di Aristotele, Genova, 1–24 (= Ancient Authors, ch. 23).

  Bibliography Montanari, F. 1993d. ‘Alessandria - Cirene – Pergamo’, in: Cambiano/Canfora/Lanza (eds.) 1993, 625–656 (= this volume, ch. 14). Montanari, F. 1994a. La philologie grecque à l’époque hellénistique et romaine, Entretiens sur l’antiquité classique 40, Vandoeuvres-Genève. Montanari, F. 1994b. ‘Note sulla tradizione manoscritta degli Scholia D in Iliadem. Un caso di errore da archetipo’, in: Storia, Poesia e Pensiero nel mondo antico. Studi in onore di Marcello Gigante. Napoli, II 475–481 (= this volume, ch. 34). Montanari, F. 1995. ‘The Mythographus Homericus’, in: Greek Literary Theory after Aristotle. A Collection of Papers in Honour of D.M. Schenkeveld, Amsterdam, 135–172 (= this volume, ch. 22). Montanari, F. 1995a. Studi di filologia omerica antica II, Biblioteca di Studi Antichi 50, Pisa. Montanari, F. 1995b. ‘Filologi alessandrini e poeti alessandrini. La filologia sui «contemporanei»’, Aevum Antiquum 8, 47–63 (= this volume, ch. 62). Montanari, F. 1995c. ‘Translating from Greek into Greek in ancient Greek culture’, in: Translation and Meaning Part 3, Proceedings of the Maastricht Session of the 2nd MaastrichtLodz Duo Colloquium, Maastricht, 441–448. Montanari, F. 1996. ‘Aristonikos (5)’, DNP I, 1120. Montanari, F. 1996a. ‘Apollonius (8)’, DNP I, 880. Montanari, F. 1997. ‘The fragments of Hellenistic scholarship’, in: G.W. Most (ed.), Collecting Fragments = Fragmente Sammeln, Aporemata , Göttingen, 273–288 (= this volume, ch. 42). Montanari, F. 1997a. ‘Didymos [1] aus Alexandria, Didymus [3] Claudius’, DNP III, 550–553. Montanari, F. 1997b. ‘Demetrios Chloros’, DNP III, 439. Montanari, F. (ed.) 1998. Omero. Gli aedi, i poemi, gli interpreti, Firenze. Montanari, F. 1998a. ‘Zenodotus, Aristarchus and the Ekdosis of Homer’, in: G.W. Most (ed.), Editing Texts = Texte edieren, Aporemata 2, Göttingen, 1–21 (= this volume, ch. 23). Montanari, F. 1998b. ‘Antichi commenti a Omero. Alcune riflessioni’, in: F. Montanari (ed.), Omero. Gli aedi, i poemi, gli interpreti, Firenze, 1–17 (= this volume, ch. 3). Montanari, F. 1999. ‘Leogoras (nr. 2)’, Der Neue Pauly 7, 53. Montanari, F. 1999a. ‘Komanos (nr. 1)’, Der Neue Pauly 6, 673. Montanari, F. 2000a. ‘Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis’, in: M. Cannatà Fera/S. Grandolini (eds.), Poesia e religione in Grecia. Studi in onore di G. Aurelio Privitera, Napoli, 479–486 (= this volume, ch. 38). Montanari, F. 2000b. ‘Demetrius of Phalerum on Literature’, in: W.W. Fortenbaugh/E. Schütrumpf (eds.), Demetrius of Phalerum. Text, Translation and Discussion (Rutgers Univ. Stud. in Class. Humanities 9), New Brunswick/London, 391–411 (= this volume, cap. 55). Montanari, F. 2001. ‘Gli studi omerici di Demetrio Falereo’, SemRom. 4, 143–157. Montanari, F. 2001a. ‘Commentari antichi su papiro. II progetto Commentaria et Lexica Graeca in Papyris reperta (CLGP)’, in: I. Andorlini/G. Bastianini/M. Manfredi/G. Menci (eds.), Atti del XXII Congresso Internazionale di Papirologia, Firenze, 969–981. Montanari, F. 2002. ‘Ancora sul Mythographus Homericus (e l’Odissea)’, in: A. Hurst/F. Létoublon (eds.), La mythologie et l’Odyssée. Hommage à Gabriel Germain, Genève, 129–144 (= this volume, ch. 21). Montanari, F. 2002a. ‘Alexandrian Homeric philology. The form of the ekdosis and the variae lectiones’, in: M. Reichel/A. Rengakos (eds.), Epea Pteroenta. Beiträge zur Homerforschung, Festschrift für Wolfgang Kullmann zum 75. Geburtstag. Stuttgart, 119–140 (= this volume, ch. 47).

Bibliography  

Montanari, F. 2002b. ‘Callimaco e la filologia’, in: L. Lehnus/F. Montanari (eds.), Callimaque, Entretiens Hardt 48, Vandoeuvres–Genève, 59–97 (= this volume, ch. 58). Montanari, F. 2003. ‘I percorsi della glossa. Traduzioni e tradizioni omeriche dall’antichità alla cultura bizantina’, in: P. Volpe Cacciatore (ed.), Erudizione scolastico-grammaticale a Bisanzio, Napoli, 81–88 (= this volume, ch. 40). Montanari, F. 2004. ‘La filologia omerica antica e la storia del testo omerico’, in: A. Bierl/ A. Schmitt/A. Willi (eds.), Antike Literatur in neuer Deutung. Festschrift für Joachim Latacz, München, 127–143 (= this volume, ch. 41). Montanari, F. . ‘Aristoteles’, in: G. Bastianini/M. Haslam/H. Maehler/F. Montanari/C. Roemer (eds.), Commentaria et Lexica Graeca in Papyris reperta (CLGP), vol. I..: Aristophanes – Bacchylides, München/Leipzig, – (= Ancient Authors, ch. ). Montanari, F. 2006a. ‘Glossario, parafrasi, ‘edizione commentata’ nei papiri’, in: Avezzù/ Scattolin 2006, 9–15 (= this volume, ch. 16). Montanari, F. 2008. ‘Aristotele, Zenodoto, Aristarco e il serpente pietrificato di Iliade II 319’, in: P. Arduini/S. Audano/A. Borghini/G. Paduano (eds.), Studi offerti ad Alessandro Perutelli, Roma, vol. II, 237–244 (= this volume, ch. 54). Montanari, F. 2009. ‘A New Papyrus Commentary on the Iliad’, TiC 1, 177–182 (= this volume, ch. 24). Montanari, F. 2009a. ‘Ekdosis alessandrina: il libro e il testo’, in: M. Sanz Morales/M. Librán Moreno (eds.), Verae Lectiones. Estudios de Crítica Textual y Edición de Textos Griegos, Cáceres/Huelva, 143–167, tavv. 1–8 pp. 385–392 (= this volume, ch. 7). Montanari, F. 2009b. ‘Il progetto Aristarchus in rete e il Lessico dei Grammatici Greci Antichi (LGGA)’, in: F. Gazzano/G. Ottone/L. Santi Amantini (eds.), Ingenia Asiatica. Fortuna e tradizione di storici d’Asia Minore, Roma, 213–222. Montanari, F. 2009c. ‘Un papiro di Eschine con correzioni (P.Oxy. 2404): considerazioni sull’ekdosis alessandrina’, AFP 55, 401–411 (= this volume, ch. 60). Montanari, F. 2011. ‘Ancient Scholarship and Classical Studies’, in: Matthaios/Montanari/ Rengakos 2011, 11–24 (= this volume, ch. 30). Montanari, F. 2011a. ‘Correcting a Copy, Editing a Text, Alexandrian Ekdosis, and Papyri’, in: F. Montanari/L. Pagani (eds.), From Scholars to Scholia: Chapters in the History of Ancient Greek Scholarship, Trends in Classics Supplementary Volumes 9, Berlin/New York, 1–15 (= this volume, ch. 8). Montanari, F. 2012. ‘The Peripatos on Literature. Interpretation, Use and Abuse’, in: A. Martano/E. Matelli/D. Mirhady (eds.), Praxiphanes of Mytilene and Chamaeleon of Heraclea. Text, Translation, And Discussion, Rutgers University Studies in Classical Humanities 18, New Brunswick/London, 339–358 (= this volume, ch. 56). Montanari, F. 2012b. ‘Aristoteles’, in: Commentaria et Lexica Graeca in Papyris reperta (CLGP), vol. I.1.4, München/Leipzig, 241–245. Montanari, F. 2012c. ‘Odissea IX 43–45, 75–92’, in: F. Reiter (ed.), Literarische Texte der Berliner Papyrussammlung. Zur Wiedereröffnung des Neuen Museums, Berlin/Boston, 26–30. Montanari, F. 2013, ‘Gli storici greci e la filologia di età ellenistico-romana’, in: Le età della trasmissione. Alessandria, Roma, Bisanzio, Roma, 1–32 (= this volume, ch. 53). Montanari, F. 2014. ‘Dal Peripato ad Alessandria’, in: M. Tulli/M. Magnani/A. Nicolosi (eds.), Philia. Dieci contributi per Gabriele Burzacchini, Bologna, 79–102. Montanari, F. 2015. ‘Ekdosis. A Product of the Ancient Scholarship’, in: Montanari/Matthaios/ Rengakos 2015, 641–672 (= this volume, ch. 10).

  Bibliography Montanari, F. 2015a. ‘From Book to Edition. Philology in Ancient Greece’, in: S. Pollock/B.A. Elman/K.K. Chang (eds.), World Philology, Cambridge, MA/London, 25–44 (= this volume, ch. 9). Montanari, F. 2015b. ‘Aristarchus’ Conjectures (once) again’, in: M. Tziatzi/M. Billerbeck/ F. Montanari/K. Tsantsanoglou (eds.), Lemmata. Beiträge zum Gedenken an Christos Theodoridis, Berlin/Boston, 119–129 (= this volume, ch. 11). Montanari, F. 2016. ‘Remarks on the citations of authors and works in ancient scholarship’, TiC 8, 73–82 (= this volume, ch. 12). Montanari, F. 2017. ‘The Idea of History of Literature: the Beginnings in Ancient Greek Culture’, in: J. Grethlein/A. Rengakos (eds.), Griechische Literaturgeschichtsschreibung. Traditionen, Probleme und Konzepte, Berlin/Boston, 153–169. Montanari, F. 2018. ‘Afterword. Ancient Scholarship today’, in: M. Ercoles/L. Pagani/F. Pontani/G. Ucciardello (eds.), Approaches to Greek Poetry. Homer, Hesiod, Pindar, and Aeschylus in Ancient Exegesis, Berlin/Boston, 345–354 (= this volume, ch. 3). Montanari, F. 2020. ‘Citazioni di Pindaro negli scoli omerici’, in: C. Meliadò/G.B. D’Alessio/ L. Lomiento/G. Ucciardello (eds.), Il potere della parola. Studi in onore di Maria Cannatà Fera, Alessandria, 279–287 (= this volume, ch. 50). Montanari, F. 2020a. ‘Gli Etymologica e gli studi sulla letteratura erudita’, in: Atti del Convegno “Accademia Nazionale dei Lincei e Filologia greco-latina”, Roma, 24–25 ottobre 2019, a cura di Guglielmo Cavallo, Bollettino dei Classici, Serie III, XLI (2020), 213–226. Montanari, F. (ed.)/Ascheri, P. (collab.), 2002. Omero tremila anni dopo, Roma (repr. as: Omero tremila anni dopo. Personaggi e strutture narrative, Roma 2009). Montanari, F./Muratore, D./Reiter, F. 2012. ‘Die Berliner Wachstafeln P. 10508–10512: Scholia Minora und grammatikalische Passagen’, in: Actes du 26e Congrès international de papyrologie, Genève, 549–557 (= this volume, ch. 43). Montanari, F./Rengakos, A./Tsagalis, Chr. (eds.) 2012. Homeric Contexts. Neoanalysis and the Interpretation of Oral Poetry, Trends in Classics Supplementary Volumes 12, Berlin/Boston. Montanari, F./Matthaios, S./Rengakos, A. 2015. Brill’s Companion to Ancient Greek Scholarship, voll. I-II, Leiden/Boston. Montanari, F./Montana, F./Muratore, D./Pagani, L. 2017. ‘Towards a new critical edition of the scholia to the Iliad: a specimen’, TiC 9, 1–21 (= this volume, ch. 8). Montiglio, S. 2005. Wandering in Ancient Greek Culture, Chicago, IL. Moraux, P. 1951. Les listes anciennes des ouvrages d’Aristote, Louvain. Moraux, P. 1973. Der Aristotelismus bei den Griechen, vol. I, Berlin (= L’aristotelismo presso i Greci, 2 vols., Milano 2000). Moraux, P. 1985. Galien de Pergame. Souvenir d’un médecin, Paris. Moraux, P. 1986. ‘Les débuts de la philologie aristotelicienne’, in: G. Cambiano (ed.), Storiografia e dossografia nella filosofia antica, Torino, 127–148. Moretti, A.F. 1993. ‘Revisione di alcuni papiri omerici editi tra i P. Mil. Vogl.’, Tyche 8, 87–97. Morocho Gayo, G. (ed.) 1989. Scholia in Aeschyli Septem adversus Thebas, León. Most, G.W. 1992. ‘Prefazione. La ricerca assidua delle origini selvagge’, in: W. Burkert, Origini selvagge. Sacrificio e mito nella Grecia arcaica, Roma/Bari, V–XII. Most, G.W. 2006–2007. Hesiod, vol. I: Theogony, Works and Days, Testimonia; vol. II: The Shield, Catalogue of Women, Other Fragments, vols. 2, Cambridge, MA/London. Mühll, P. von der. 1952. Homeri Odyssea, Basileae. Müller, A.B. 1903. De Asclepiade Myrleano, Lipsiae. Müller, C. 1861. Geographi Graeci Minores, Parisiis.

Bibliography  

Müller, E.I. von. 1891. ‘Galen als Philologe’, in: Verhandlungen der 41. Versammlung deutscher Philologen und Schulmänner, München, 80–91. Müller, M. 1891. De Seleuco Homerico, diss. Gottingae. Muratore, D. 2012. ‘Studi sulla tradizione e sul testo degli Scholia in Homeri Iliadem. I. Iliade A–B: revisione del testo degli scholia vetera (Erbse) nel Venetus A’, RFIC 140, 425–438. Muratore, D. 2014. ‘Some thoughts on the interlinear scholia in the h family of the Iliad’, in: F. Montana/A. Porro (eds.), The Birth of Scholiography. From Types to Texts, TiC 6, 54–89. Muratore, D. 2016. ‘Studi sulla tradizione e sul testo degli Scholia in Homeri Iliadem. 2. Parafrasi e glosse interlineari ai canti A–B dell’Iliade nel Venetus A’, RFIC 144, 361–401. Muratore, D. 2018a. ‘On the sources of Lascaris’ edition of D-Scholia on the Iliad’, in: M. Ercoles/L. Pagani/F. Pontani/G. Ucciardello (eds.), Approaches to Greek Poetry, Berlin/ Boston, 133–160. Muratore, D. 2018b. ‘Studi sulla tradizione e sul testo degli Scholia in Homeri Iliadem. 3. Gli Scholia D nel Marc. gr. App. IX, 5 (1336)’, RFIC 146, 239–249. Murray, G. 1957. Aeschyli septem quae supersunt tragoediae, Oxonii. Musti, D. 1989. Storia greca, Roma/Bari. Naber, S.A. 1864–1865. Photii Patriarchae lexicon, Leidae. Nachtergael, G. 1971. ‘Fragments d’anthologies homériques (P. Strasb. inv. 2374; P. Graec. Vindob. 26740; P. Hamb. II 136)’, CE 46, 344–351. Nagy, G. 1996a. Poetry as Performance. Homer and Beyond, Cambridge, MA. Nagy, G. 1996b. Homeric Questions, Austin. Nagy, G. 1997. ‘Homeric Scholia’, in: I. Morris/B. Powell (eds.), A New Companion to Homer, Leiden/New York/Köln, 101–122. Nagy, G. 1998. ‘Aristarchean questions’, BMCR, 1998.07.14. Nagy, G. 1999. ‘Les éditions alexandrines d’Homère au XVIIIe et au XIXe siècle’, in: F. Létoublon/ C. Volpilhac-Auger (eds.), Homère en France après la Querelle (1715–1900), Paris, 63–72. Nagy, G. 2000. Review of Homeri Ilias, rec. M.L. West, vol. I, Stuttgart/Leipzig, 1998, BMCR 2000.09.12. Nagy, G. 2003. Review of M.L. West, Text and Transmission of the lliad, München/Leipzig, 2001, Gnomon, 481–501. Nagy, G. 2004. Homer’s Text and Language, Urbana. Nagy, G. 2009. ‘Hesiod and the Ancient Biographical Traditions’, in: F. Montanari/A. Rengakos/ Chr. Tsagalis (eds.), Brill’s Companion to Hesiod, Leiden/Boston, 271–312. Nagy, G. 2010. Homer the Classic, Cambridge, MA. Naoumides, M. 1969. ‘The Fragments of Greek Lexicography in the Papyri’, Classical Studies presented to B. E. Perry, Urbana, 181–202. Naoumides, M. 1975. Ῥητορικαὶ Λέξεις, Αθήναι. Nardelli, J.-F. 2001a. ‘Éditer l’Iliade, I. La transmission et ses débats: perspectives critiques’. Gaia 5, 41–118 (contin. in ‘Éditer l’Iliade, II. Manuscrits et affiliations manuscrites dans l’Iliade ΧΙΙΙ–XV’, Gaia 6, 2002, 47–144). Nardelli, J.-F. 2001b. Review of Homeri Ilias, rec. M.L. West, vol. II, Munich/Leipzig, 2000, BMCR 2001.06.21. Nauck, A. 1848. Aristophanis Byzantii grammatici Alexandrini fragmenta, Halis. Nauck, A. 1867. Lexicon Vindobonense, Petropoli. Negri, M. 2004. Pindaro ad Alessandria. Le edizioni e gli editori, Brescia. Neitzel, S. 1977. Apions Γλῶσσαι Ὁμηρικαί, SGLG 3, Berlin/New York. Neri, C. 2003. Erinna. Testimonianze e frammenti, Bologna.

  Bibliography Nickau, K. 1966. Ammonii qui dicitur liber de adfinium vocabulorum differentia, Lipsiae. Nickau, K. 1972. ‘Zenodotos (3)’, RE X A, 1972, 23–45. Nickau, K. 1977. Untersuchungen zur textkritischen Methode des Zenodotos von Ephesos, Berlin/New York. Nicolai, R. 1991. ‘Zenodoto e Pseudo-Scimno (Hom. Od. 1, 3)’, RFIC 119, 181–187. Nicolai, R. 1992. La storiografia nell’educazione antica, Pisa. Nicolai, R. 1992a. La fondazione di Cirene e i Karneia cirenaici nell’inno ad Apollo di Callimaco, MD 28, 153–173. Nilsson, Μ.P. 1955. Geschichte der Griechischen Religion, I2, München. Norwood, G. 1931. Greek Comedy, London. Nünlist, R. 2009. The Ancient Critic at Work. Terms and Concepts of Literary Criticism in Greek Scholia, Cambridge. Nünlist, R. 2012. ‘Aristarch und das Lexicon des frühgriechischen Epos’, in: M. Meier-Brügger (ed.), Homer, gedeutet durch ein großes Lexicon, Berlin, 193–214. Nünlist, R. 2015. ‘Poetics and Literary Criticism in the Framework of Ancient Greek Scholarship’, in: Montanari/Matthaios/Rengakos 2015, I, 706–755. Olson, S.D. 2014. Eupolis frr. 326–497. Translation and Commentary, Heidelberg. Olson, S.D. 2017. Eupolis. Testimonia and Aiges – Demoi (frr. 1–146). Translation and Commentary, Heidelberg. Olson, S.D./Seaberg, R. 2018. Kratinos frr. 299–514. Translation and Commentary, Göttingen. Ophuijsen, J.M. van. 1987. Hephaestion. On Metre, Leiden. Ostermann, Chr. 1857. Commentatio de Demetrii Phalerei vita, rebus gestis et scriptorum reliquiis, Progr. Hersfeld 1847, Fulda. Paap, A.H.R.E. 1948. De Herodoti reliquiis in papyris et membranis Aegyptiis servatis, Lugduni Batavorum. Pagani, L. 2006. ‘Megaclides’, in: LGGA. Pagani, L. 2008. ‘Dionysius [14] Thrax’, in: LGGA. Pagani, L. 2009. Review of R. Nünlist, The Ancient Critic at Work. Terms and Concepts of Literary Criticism in Greek Scholia, RFIC 137, 201–211. Pagani, L. 2011. ‘Pioneers of Grammar. Hellenistic Scholarship and the Study of Language’, in: F. Montanari/L. Pagani (eds.), From Scholars to Scholia. Chapters in the History of Ancient Greek Scholarship, Trends in Classics Supplementary Volumes 9, Berlin/New York, 17–64. Pagani, L. 2018. ‘Interpretazioni di Omero in chiave tragica negli scolii all’Iliade’, in: F. Conte Bizzarro (ed.), ΛΕΞΙΚΟΝ ΓΡΑΜΜΑΤΙΚΗΣ. Studi di lessicografia e grammatica greca, Napoli, 192–198. Pagani, L. 2020. ‘Papyrus commentaries on the Iliad’, in: S. Boodts/P. de Leemans/S. Schorn (eds.), Sicut dicit. Editing Ancient and Medieval Commentaries on Authoritative Texts, Turnhout, 331–362. Pagani, L./Perrone, S. 2012. ‘Le ekdoseis antiche di Omero nei papiri,’ in: G. Bastianini/ A. Casanova (eds.), I papiri omerici, Firenze, 97–124. Page, D.L. 1962. Poetae Melici Graeci, Oxford. Page, D.L. 1963. The Oxyrhychus Papyri, part XXIX, London. Page, D.L. 1972. Aeschyli septem quae supersunt tragoediae, Oxonii. Pagès, J. Cebrián. 2005. ‘Análisis del escolio D a Ilíada I 5: una aproximación a la génesis del corpus’, in: A. Alvar Ezquerra (ed.), Actas del XI Congreso de la Sociedad Española de Estudios Clásicos, Madrid, 453–460. Palmieri, V. 1988. Herennius Philo, De diversis verborum significationibus, Napoli.

Bibliography  

Panzer, L. 1892. De Mythographo Homerico restituendo, Diss. Gryphiswaldiae. Papageorgiou, P.N. 1888. Scholia in Sophoclem, Lipsiae. Parassoglou, G. 1975. ‘A New Commentary on the Odyssey’, Hellenika 28, 60–65. Parke, H.W. 1967. The Oracles of Zeus. Dodona. Olympia. Ammon, Oxford. Parsons, P.J. 1970. ‘A School-Book from the Sayce Collection’, ZPE 6, 133–149. Parsons, P.J. 1977. ‘Callimachus: Victoria Berenices’, ZPE 25, 1–50. Parsons, P.J. 1977a. ‘The Oyster’, ZPE 24, 1–12. Parsons, P.J./Pellé, N. 2012. 6. ‘Commentario a Tucidide III? ’, in: F. Reiter (ed.), Literarische Texte der Berliner Papyrussammlung. Zur Wiedereröffnung des Neuen Museums, Berlin/ Boston, 31–33. Pasquali, G. 1920. Filologia e storia, Firenze (repr. 1998). Pasquali, G. 19522. Storia della tradizione e critica del testo, Firenze. Patrizi, F. 1969. Della Poetica, a cura di D. Aguzzi Barbagli, vol. I; La deca istoriale, Firenze. Pearson, A.C. (ed.) 1946. Sophoclis fabulae, Oxonii. Pearson, L./Stephens, S. (eds.) 1983. Didymi in Demosthenem commenta, Stutgardiae. Pecorella, G.B. 1962. Dionisio Trace Τέχνη γραμματική, Bologna. Pelling, C. 2007. “Ion’s Epidemiai and Plutarch’s Ion”, in: V. Jennings/A. Katsaros (eds.), The World of Ion of Chios, Leiden, 75–109. Pépin, J. 19762. Mythe et allégorie. Les origines grecques et les contestations judéo-chrétiennes, Paris. Perret, J. 1982. ‘Allégorie, hyponoiai, inspiration sur les exégèses anciennes d’Homère’, in: Mélanges É. Gareau, Ottawa, 65–73. Perrone, S. 2010. ‘Paralleli comici nell’esegesi a commedia su papiro’, in: F. Montana (ed.), Aner Polytropos. Ricerche di filologia greca antica dedicate dagli allievi a Franco Montanari, Roma, 85–103. Perrone, S. 2019. Cratete. Introduzione, traduzione e commento, Göttingen. Pertusi, A. (ed.) 1955. Scholia vetera in Hesiodi Opera et Dies, Milan. Pfeiffer, R. 1937. ‘Hesiodisches und Homerisches. Zu neuen und alten Papyri’, Philologus 92, 1–25. Pfeiffer, R. 1949–1953. Callimachus, 2 vols., Oxonii. Pfeiffer, R. 1955. ‘The Future of Studies in the Field of Hellenistic Poetry’, JHS 75, 69–73 (= Ausgewählte Schriften, München 1960, 148–158). Pfeiffer, R. 1968. History of Classical Scholarship from the Beginning to the End of the Hellenistic Age, Oxford (rev. ed. in German: München 19782). Pfeiffer, R. 1973. Storia della filologia classica, Napoli (= Pfeiffer 1968). Pfeiffer, R. 1976. History of Classical Scholarship: 1300–1850, Oxford. Phillips, T. 2013. ‘Callimachus in the Pindar Scholia’, Cambridge Classical Journal 59, 152–177. Pfohl, G. 1968. Das Alphabet, Darmstadt. Pickard-Cambridge, A. 19682. The Dramatic Festivals of Athens, Oxford. Pinborg, J. ‘Classical Antiquity: Greece’, in: Historiography of Linguistics, Current Trends in Linguistics 13.1, The Hague/Paris, 69–126. Plaumann, G. 1913. ‘P. Berol. inv. 11636: scholia minora in Homerum (E 265–286)’, Amtliche Berichte aus den königlichen Kunstsammlungen 34, 219–223. Podlecki, A.J. 1969. ‘The Peripatetics as Literary Critics’, Phoenix 23, 114–137. Podlecki, A.J. 2009. ‘Aiskhylos the Forerunner’, in: J. Jouanna/F. Montanari (eds.), Eschyle à l’aube du théâtre occidental, Genève, 319–362. Poetscher, W. 1966. ‘Zeus Naios und Dione in Dodona’, Mnemosyne 19, 113–147.

  Bibliography Polaschek, E. 1965. ‘Klaudios Ptolemaios. Das geographische Werk’, RE Suppl. X, 680–833. Poli, A. 1989. ‘Tre casi di versi omessi nella διόρθωσις di Riano da Creta’, Athenaeum 67, 287– 295. Pontani, F. 1999. ‘The First Word of Callimachus’ Aitia’, ZPE 128, 57–59. Pontani, F. 2005. Sguardi su Ulisse. La tradizione esegetica greca all’Odissea, Roma. Pontani, F. 2007. Scholia Graeca in Homert Odysseam I. Scholia ad libros α–β, Pleiadi 6.1. Roma. Pontani, F. 2010. Scholia Graeca in Homeri Odysseam II. Scholia ad libros γ–δ, Pleiadi 6.2, Roma. Porro, A. 1985. ‘Manoscritti in maiuscola alessandrina di contenuto profano. Aspetti, grafici, codicologici filologici’, Scrittura e civiltà 9, 169–215. Porson, R. 1823. Photii Lexicon e codice Galeano descripsit R.P., Lipsiae. Porter, J.J. 1992. ‘Hermeneutic Lines and Circles: Aristarchus and Crates on the Exegesis of Homer’, in: R. Lamberton/J.J. Keaney (eds.), Homer’s Ancient Readers, Princeton, 67–114. Preller, L. 1842. De Praxiphane peripatetico inter antiquissimos grammaticos nobili disputatio, diss. Dorpati 1842, Berlin. Preller, L. 1848. ‘Zu den griechischen Komikern’, Philologus 3.1–4, 520–522. Pretagostini, R. 1988. ‘La poesia ellenistica’, in: F. Montanari (ed.), Da Omero agli Alessandrini. Problemi e figure della letteratura greca, Roma, 318ff. Pucci, P. 2007. Inno alle Muse (Esiodo Teogonia 
1–115), Roma/Pisa. Puglia, E. 1980. ‘Nuove letture nei PHerc. 1012 e 1786 (Demetrii Laconis Opera incerta)’, CErc 10, 25–53. Puglia, E. (ed.) 1988. Demetrio Lacone. Aporie testuali ed esegetiche in Epicuro (PHerc. 1012), Napoli. Rabe, H. 1906. Scholia in Lucianum, Lipsiae (repr. Stuttgart 1971). Radt, S. 1980. ‘Noch einmal Aischylos, Niobe Fr. 162 N.2 (278 M.)’, ZPE 38, 47–58. Radt, S. 1985. Tragicorum Graecorum Fragmenta, vol. 3: Aeschylus, Göttingen (= TrGF 3). Radt, S. 1988. ‘Oἱ (αἱ etc.) περὶ + acc. nominis proprii bei Strabon’, ZPE 71, 35–40. Radt, S. 1988a. ‘Addendum’, ZPE 74, 108. Radt, S. (ed.) 19992. Tragicorum Graecorum Fragmenta, vol. 4: Sophocles, Göttingen (= TrGF 4). Raffaelli, L.M. 1984. ‘Repertorio dei papiri contenenti Scholia minora in Homerum’, in: F. Montanari (ed.), Ricerche di filologia classica I, Pisa, 139–177. Raffaelli, L.M. 1990. ‘P. Berol. 11636: Omero E 265–317 + Scholia minora a E 265–286’, APF 36, 5–12. Raffaelli, L.M. 1992. Aristarco interprete dei poeti lirici, Diss. Genova. Rauscher, G. 1886. De scholiis Homericis ad rem metricam pertinentibus, Argentorati. Razzetti, F. 2002. ‘Seleucus [3] Homericus’, in: LGGA. Razzetti, F. 2002a. I frammenti del grammatico Aristonico, Diss. Genova. Regali, M. 1983. Macrobio, Commento al Somnium Scipionis, Libro I, Pisa. Regali, M. 1989. Macrobio, Commento al Somnium Scipionis, Libro II, Pisa. Regali, M. 2007. ‘Minucius Pacatus Irenaeus’, LGGA. Reisch, E. 1899. ‘Chor’, RE III, 2373–2404. Reitzenstein, R. 1907. Der Anfang des Lexikons des Photios, Leipzig/Berlin. Rengakos, A. 1993. Der Homertext und die hellenistischen Dichter, Stuttgart. Rengakos, A. 1994. Apollonios Rhodios und die antike Homererklärung, München. Rengakos, A. 1994a. ‘Lykophron als Homererklärer’, ZPE 102, 111–130. Rengakos, A. 2000. ‘Aristarchus and the Hellenistic poets’, SemRom. 3, 325–335.

Bibliography  

Rengakos, A. 2001. ‘Apollonius Rhodius as a Homeric Scholar’, in: Th.D. Papanghelis/A. Rengakos (eds.), A Companion to Apollonius Rhodius, Leiden/Boston/Köln, 193–216. Rengakos, A. 2002. ‘The Hellenistic Poets as Homeric Critics’, in: Montanari/Ascheri 2002, Roma, 143–157. Rengakos, A. 2002a. Review of M.L. West, Text and Transmission of the Iliad, München/Leipzig, 2001, BMCR 2002.11.15. Rengakos, A. 2012. ‘Bemerkungen zum antiken Homertext’, in: M. Meier-Brügger (ed.), Homer, gedeutet durch ein großes Lexicon, Berlin, 239–252. Renner, T. 1979. ‘Three New Homerica on Papyrus’, HSCP 83, 331–337. Repellini, F.F. 1993. ‘Matematica, astronomia e meccanica’, in: Cambiano/Canfora/Lanza 1993, 305–344. Reynolds, L.D./Wilson, N.G. 1987. Copisti e filologi, Padova (Oxford 1968; 19913). Riad, H./Schwartz, J. 1968. ‘Deux planchettes du Musée gréco-romain d’Alexandrie’, CE 43, 114–125. Richardson, N.J. 1983. ‘Recognition Scenes in the Odyssey and Ancient Literary Criticism’, Papers of the Liverpool Latin Seminar 4, 219–235. Richardson, N.J. 1992. ‘Aristotle’s Reading of Homer and its Background’, in: R. Lamberton/ J.J. Keaney (eds.), Homer’s Ancient Readers, Princeton, 30–40. Richardson, N.J. 1993. The Iliad: A Commentary, vol. VI: Books 21–24, Cambridge. Richardson, N.J. 1994. ‘Aristotle and Hellenistic Scholarship’, in: Montanari 1994a, 7–38. Rispoli, G. 1988. Lo spazio del verisimile. Il racconto, la storia, il mito, Napoli. Rizzo, S. 1973. Il lessico filologico degli umanisti, Roma. Robert, L. 1937. Etudes Anatoliennes, Paris. Roemer, A. 1912. Aristarchs Athetesen in der Homerkritik, Leipzig/Berlin. Rose, V. 1863. Aristoteles Pseudepigraphus, Lipsiae. Roselli, A. 2010. ‘Libri e biblioteche a Roma al tempo di Galeno: la testimonianza del de indolentia,’ Galenos 4, 127–148. Roselli, A. 2012. ‘Galeno e la filologia del II secolo’, in: E. Bona/C. Levy/G. Magnaldi (eds.), Vestigia notitiai. Scritti in memoria di Michelangelo Giusta, Alessandria, 63–80. Rosén, H.B. 1962. Eine Laut- und Formenlehre der Herodotischen Sprachform, Heidelberg. Rosén, H.B. 1987–1997. Herodoti Historiae, 2 vols., Lipsiae. Rossi, L.E. 1968. ‘La fine alessandrina dell’Odissea e lο ζῆλος Ὁμηρικός di Apollonio Rodio’, RFIC 96, 151–163. Rossi, L.E. 1971. ‘I generi letterari e le loro leggi scritte e non scritte nelle letterature classiche’, BICS 18, 69–94. Rossi, L.E. 1976. ‘Umanesimo e filologia (a proposito della Storia della filologia classica di Rudolf Pfeiffer)’, RFIC 104, 98–117. Rossi, L.E. 1999. ‘La méthode philologique de Jean Baptiste Gaspard d’Ansse de Villoison et le Venetus Marcianus A’, in: F. Létoublon/C. Volpilhac-Auger (eds.), Homère en France après la Querelle (1715–1900), Paris, 51–62. Rossum-Steenbeek, M. van. 1996. ‘More on P.Oxy. LXI 4096, Mythographus Homericus’, ZPE 112, 34–36. Rossum-Steenbeek, M. van. 1998. Greek Readers’ Digest? Studies on a Selection of Subliterary Papyri, Mnemosyne Suppl. 175, Leiden/New York/Köln. Rupprecht, K. 1949. ‘Paroimia. Paroimiographoi’, in: RE XVIII, 1735–1778. Russo, C.F. [19652] 1950. Hesiodi Scutum, Firenze. Rusten, J.S. 1982. ‘Dicaearchus and the Tales from Euripides’, GRBS 23, 357–367.

  Bibliography Rüter, K. 1969. Odysseeinterpretationen, Göttingen. Rutherford, W.G. 1896–1905. Scholia Aristophanica: being such comments adscript to the text of Aristophanes as have been preserved in the Codex Ravennas, vols. 3, London. Rzach, A. 1912. ‘Hesiodos’, RE VIII 1, 1167–1240. Salomons, R.P. 1984. ‘Use and meaning of the middle point in the Hawara Homer’, in: Atti del XVII Congresso Internazionale di Papirologia, Napoli 19–26 maggio 1983, Napoli, II 249– 253. Sandys, J.E. 1920. A History of Classical Scholarship, I3, Cambridge. Sbardella, L. 2000. Filita. Testimonianze e frammenti poetici, Quaderni dei Seminari Romani di Cultura Greca 3, Roma. Scarpa, L. 1981. Macrobii Ambrosii Theodosii, Commentariorum in Somnium Scipionis libri duo, Padova. Scattolin, P. 2007. ‘Sui meccanismi delle citazioni negli scolî antichi a Sofocle ed Euripide’, in: R. Pretagostini/E. Dettori (eds.), La cultura letteraria ellenistica: persistenza, innovazione, trasmissione, Quaderni dei Seminari Romani di Cultura Greca 10, Roma, 233–245. Schäublin, C.H. 1977. ‘Homerum ex Homero’, MH 34, 221–227. Scheer, E. 1908. Lycophronis Alexandra, Vol. II: Scholia continens, Berlin (repr. ibid. 1958). Schenkeveld, D.M. 1970. ‘Aristarchus and ΟΜΗΡΟΣ ΦΙΛΟΤΕΧΝΟΣ. Some Fundamental Ideas of Aristarchus on Homer as a Poet’, Mnemosyne 23, 162–178. Schenkeveld, D.M. 1982. ‘Studies in the History of Ancient Linguistics. I. Σύνδεσμοι ὑποθετικοί and ὁ ἐὰν ἐπιζευκτικός’, Mnemosyne 35, 248–268. Schenkeveld, D.M. 1984. ‘Studies in the History of Ancient Linguistics. II. Stoic and Peripatetic Kinds of Speech Act and the Distinction of Grammatical Moods’, Mnemosyne 37, 291–353. Schenkeveld, D.M. 1990. ‘Studies in the History of Ancient Linguistics. III. The Stoic τέχνη περὶ φωνῆς’, Mnemosyne 43, 86–108. Schenkeveld, D.M. 1990a. ‘Studies in the History of Ancient Linguistics. IV. Developments in the Study of Ancient Linguistics, Mnemosyne 43, 289–306. Schenkeveld, D.M. 1994. ‘Scholarship and Grammar’, in: Montanari 1994a, 263–301. Schiano, C. 2002. ‘Teone e il Museo di Alessandria’, QS 55, 129–143. Schimberg, A. 1890. ‘Zur handschriftlichen Überlieferung der scholia Didymi: I’, Philologus 49, 421–456. Schimberg, A. 1891–1892. ‘Zur handschriftlichen Überlieferung der scholia Didymi: II–III’, Wiss. Beilage z. Progr. K. Ev. Gymnasiums zu Ratibor, Göttingen. Schimberg, A. 1892. Die handschriftliche Überlieferung der Scholia Vulgata gennant Didymi, Göttingen (reprint of: Schimberg 1890 and Schimberg 1891–1892). Schimberg, A. 1897. ‘Scholia in Homeri Iliadem vulgata (A 1–50) e codicibus aucta et emendata’, Festschrift zur hundertjährigen Jubelfeier des Königlichen Friedrich-WilhelmGymnasiums zu Berlin, Berlin, 65–79. Schironi, F. . ‘Aristarco studioso di Antimaco’, RFIC , –. Schironi, F. 2009. ‘Theory into Practice: Aristotelian Principles in Aristarchean Philology’, CPh 104, 279–316. Schmid, W. 1929. Geschichte der griechischen Literatur, München. Schmidt, M. 1852. ‘Dionys der Thraker’, Philologus 7, 370–381. Schmidt, M. 1854. Didymi Chalcenteri grammatici Alexandrini fragmenta quae supersunt omnia, Lipsiae. Schmidt, M. 1858–1868. Hesychii Alexandrini Lexicon, 5 vols., Ienae.

Bibliography  

Schmidt, M. 1976. Die Erklärungen zum Weltbild Homers und zur Kultur der Heroenzeit in den bT-Scholien zur Ilias, Zetemata 62, München. Schmidt, M. 1997. ‘Variae lectiones oder Parallelstellen: Was notierten Zenodot und Aristarch zu Homer?’, ZPE 115, 1–12. Schmidt, M. 2002. ‘The Homer of the Scholia: what is explained to the Reader”, in: MontanariAscheri 2002, 159–183. Schmidt, N. 1920. ‘Bellerophon’s Tablet and the Homeric Question in the Light of Oriental Research’, TAPhA 51, 56–70. Schmidt, R. 1839. Stoicorum grammatica, Halis. Schneider, O. 1873. Callimachea, vol. II, Lipsiae. Schorn, S. 2004. Satyros aus Kallatis. Sammlung der Fragmente mit Kommentar, Basel. Schrader, H. 1880. Porphyrii Quaestionum Homericarum ad Iliadem pertinentium reliquias, Lipsiae. Schrader, H. 1890. Porphyrii Quaestionum Homericarum ad Odysseam pertinentium reliquias, Lipsiae. Schroeder, C.M. 2007. ‘A New Monograph by Aristarchus?’, JHS 127, 138–141. Schroeder, C.M. 2009. ‘Zenodotus’ Text of Hesiod’, CQ 59, 271–274. Schulz, H. 1912. ‘Herodianus’, RE VIII 1, 959–973. Schwartz, E. 1881. ‘De Scholiis Homericis ad historiam fabularem pertinentibus’, NJPhP Suppl. 12, 405–463. Schwartz, E. 1894. ‘Apollodoros’, RE I 2, 2856–2886. Schwartz, E. 1887–1891. Scholia in Euripidem, 2 vols., Berolini. Schwartz, E. 1957. Griechische Geschichtsschreiber, Leipzig. Schwartz, J. 1960. Pseudo-Hesiodeia. Recherches sur la composition, la diffusion et la disparition ancienne d’oeuvres attribuées à Hésiode, Leiden. Schwarz, Julius 1878. De scholiis in Homeri Iliadem mythologicis capita tria. Diss. Vratislaviae. Sciarra, E. 2005. La tradizione degli scholia iliadici in Terra d’Otranto, Roma. Serrano, C. 1977. ‘Historia de la Lexicografía griega antigua y medieval’, in: Introducción a la Lexicografía Griega, Madrid, 61–106. Severyns, A. 1928. Le Cycle épique dans l’École d’Aristarque, Paris. Sforza, I. 2000. ‘Nota su un verso dei papiri tolemaici dell’Iliade’, Analecta Papyrologica 12, 25–34. Siebenborn, E. 1976. Die Lehre von der Sprachrichtigkeit und ihren Kriterien. Studien zur antiken normativen Grammatik, Amsterdam. Slater, W.J. 1972. ‘Asklepiades and historia’, GRBS 13, 317–333. Slater, W.J. 1976. ‘Aristophanes of Byzantium on the Pinakes of Callimachus’, Phoenix 30, 234– 241. Slater, W.J. 1982. ‘Aristophanes of Byzantium and Problem-Solving in the Museum’, CQ 37, 336–349. Slater, W.J. 1986. Aristophanis Byzantii fragmenta, SGLG 6, Berlin/New York. Slater, W.J. 1989. ‘Problems in Interpreting Scholia on Greek Texts’, in: J. Grant (ed.), Editing Greek and Latin Texts, New York, 37–61. Slings, S.R. 1987. ‘Commentary on poem(s) of Hipponax’, in: J.M. Bremer/A.M. van Erp Taalman Kip/S.R. Slings (eds.), Some Recently Found Greek Poems, Leiden/New York/København/ Köln, 70–94, Slings, S.R. 1999. Review of: G.W. Most (ed.), Editing Texts = Texte Edieren, Aporemata 2, Göttingen 1998, BMCR 1999.05.27.

  Bibliography Sluiter, I. 1990. Ancient Grammar in Context. Contributions to the Study of Ancient Linguistic Thought, Amsterdam. Smith, O.L. (ed.) 1976. Scholia Graeca in Aeschylum quae exstant omnia, Pars I: Scholia in Agamemnonem, Choephoros, Eumenides, Supplices continens, Lipsiae. Smith, O.L. (ed.). 1982. Scholia Graeca in Aeschylum quae exstant omnia, Pars II Fasc. 2: Scholia in Septem adversus Thebas continens, Lipsiae. Snipes, K. 1988. ‘Literary Interpretation in the Homeric Scholia: the Similes of the Iliad’, AJPh 109, 196–222. Sodano, A.R. 1964. Porphyrii in Platonis Timaeum commentariorum fragmenta, Neapoli. Sodano, A.R. 1970. Porphyrii Quaestionum Homericarum liber I, Napoli. Spoerri, W. 1975. ‘Gnome’, in: Der Kleine Pauly, II, München, 823–829. Staden, H. von 1989. Herophilus. The Art of Medicine in Early Alexandria, Cambridge. Staesche, T. 1883. De Demetrio Ixione, Diss. Halis Saxonum. Stein, H. 1871. Herodoti Historiae, vol. II, Berolini. Steinicke, K. 1957. Apollonii Sophistae Lexicon Homericum, Diss. Gottingae. Steinmetz, P. 1986. ‘Allegorische Dichtung und allegorische Deutung in der alten Stoa’, RhM 129, 18–30. Stoessl, F. 1970. ‘Mesatos’, RE, Suppl. XII, 866–867. Stork, P./von Ophuijsen, J.M./Dorandi, T. 2000. ‘Demetrius of Phalerum: The Sources, Text and Translation’, in: Fortenbaugh-Schütrumpf 2000, 1–310. Stramaglia, A. 2011. ‘Libri perduti per sempre: Galeno, De indolentia 13; 16; 17–19’, RFIC 139, 118–147. Stroppa, M. 2008. ‘Lista di codici tardoantichi contenenti hypomnemata’, Aegyptus 88, 49–69. Stroppa, M. 2008a. ‘Esegesi a Esiodo nei papyri’, in: G. Bastianini/A. Casanova (eds.), Esiodo. Cent’anni di papiri. Atti del Convegno Internazionale di Studi, Firenze, 7–8 June 2008, Florence, 83–96. Stroppa, M. 2009. ‘Some remarks regarding commentaries on codex from late antiquity’, TiC 1.2, 298–327. Sturz, F.W. 1818. Etymologicum Graecae linguae Gudianum, Lipsiae. Suess, W. 1966. ‘Der ältere Dionys als Tragiker’, RhM 109, 299–318. Susemihl, F. 1892. Die Geschichte der griechischen Litteratur in der Alexandrinerzeit, Leipzig (repr. Hildesheim 1965). Taplin, O. 1977. The Stagecraft of Aeschylus. The Dramatic Use of Exits and Entrances in Greek Tragedy, Oxford. Taylor, D.J. 1986. ‘Rethinking the History of Language Science in Classical Antiquity’, in: Historiographia Linguistica 13, 175–190. Telò, M. 2007. Eupolidis Demi, Firenze. Tessier, A. 1989. Scholia metrica vetera in Pindari carmina, Leipzig. Theodoridis, C. 1976. Die Fragmente des Grammatikers Philoxenos, SGLG 2, Berlin/New York. Theodoridis, C. 1979. ‘Vier neue Bruchstücke des Apollodoros von Athen’, RhM 122, 9–17. Theodoridis, C. 1983. Photii Patriarcae Lexicon, vol. I: Α–Δ, Berlin/New York. Thiel, H. van. 1991. Homeri Odyssea, Hildesheim/Zürich/New York. Thiel, H. van. 1996. ‘Zenodot, Aristarch und Andere’, ZPE 90, 1–32. Thiel, H. van. 1996, 20102. Homeri Ilias, Hildesheim/Zürich/New York. Thiel, H. van. 1997. ‘Der Homertext in Alexandria’, ZPE 115, 13–36. Thiel, H. van. 2005. Λέξεις Ὁμηρικαί, Proecdosis correctior 2005, online at: http://kups.ub. unikoeln.de/5253/ (first release 2002).

Bibliography  

Thiel, H. van. 2014. Scholia D in Iliadem secundum codices manu scriptos, Proecdosis aucta et correctior 2014. Adiuverunt N. Conrad et S. Matthaios, online at: http://kups.ub.uni-koeln. de/5586/ (first release 2000). Thierfelder, A. 1935. Beiträge zur Kritik und Erklärung des Apollonios Dyskolos, Leipzig. Tittmann, I.A.H. 1808. Iohannis Zonarae Lexicon, Leipzig. Tocco, A. 2019. ‘“New Music” and early Peripatetic Scholarship: the “degeneration” of Music as a Historiographical Paradigm’, Greek and Roman Musical Studies 7, 33–50. Tolkiehn, J. 1925. ‘Lexikographie’, RE XII 2, 2432–2482. Tosi, R. 1983. Scoli-fantasma tucididei, Pisa. Tosi, R. 1988. Studi sulla tradizione indiretta dei classici greci, Bologna. Tosi, R. 1989. ‘Alcuni esempi di polisemia nell’Agamennone di Eschilo: esegesi antica e filologia moderna’, Lexis 3, 3–24. Tosi, R. 1994. ‘La lessicografia e la paremiografia in età alessandrina e il loro sviluppo successivo’, in: Montanari 1994a, 143–197 + 198–209. Tosi, R. 2006. ‘Note di lessicografia eschilea’, Lexis 24, 43–51. Tosi, R. 2014. ‘The history of corpora scholiastica: a series of unfortunate events’, TiC 6, 15–23. Tosi, R. 2015. ‘Typology of Lexicographical Works’, in: F. Montanari/S. Matthaios/A. Rengakos (eds.), Brill’s Companion to Ancient Greek Scholarship, Leiden/Boston, I 622–636. Trachsel, A./Schubert, P. 1999. ‘Une description de la topographie de Troie dans un papyrus de Genève (Pack2 1204): réédition’, MH 56, 222–237. Tronson, A. 1984. ‘Satyrus the Peripatetic and the marriages of Philip II’, JHS 104, 116–126. Trypanis, C.A. 1958. Callimachus. Aetia, Iambi, Hecale and other Fragments, London. Tulli, M. 1987. ‘Il giudizio di Platone sull’esegesi allegorica’, in: Interpretazioni antiche e moderne di testi greci. Ricerche di filologia classica III, Pisa, 45–52. Turner, E.G. 1957. ‘P.Oxy. 2404. Aeschines, In Ctesiphontem 51–53’, in: P.Oxy. XXIV, London, 130–132. Turner, E.G. (ed.). 1962. The Oxyrhynchus Papyri, Part XXVII, London. Turner, E.G. 1968, 19802. Greek Papyri. An Introduction, Oxford (ed. ital. a cura di M. Manfredi, Roma 1984). Turner, E.G. 1977. The Typology of the Early Codex, Philadelphia. Turner, E.G./Parsons, P.J. 19872. Greek Manuscripts of the Ancient World, BICS Suppl. 46, London. Turra, V. 2006. ‘Sul valore di alcune categorie critiche negli scholia vetera al Filottete’, in: Avezzù/Scattolin 2006, 151–171. Ucciardello, G. 2006. ‘Esegesi linguistica, glosse ed interpretamenta tra hypomnemata e lessici. Materiali e spunti di riflessione’, in: Avezzù/Scattolin 2006, 35–83. Uhlig, G. 1883. Dionysii Thracis Ars grammatica, Grammatici graeci 1, Lipsiae. Usener, H. 1865. ‘Leogoras von Syrakus’, RhM 20, 131–133. Valente, S. 2015. ‘Typology of Grammatical Treatises’, in: Montanari/Matthaios/Rengakos 2015, I 600–621. Valk, M. van der. 1949. Textual Criticism of the Odyssey, Leiden. Valk, M. van der. 1958. ‘On Apollodori Bibliotheca’, REG 71, 100–168. Valk, M. van der. 1963–1964. Researches on the Text and Scholia of the Iliad, 2 vols., Leiden. Valk, M. van der. 1971–1987. Eustathii commentarii ad Homeri Iliadem, Leiden. Vassis, I. 1991. Die handschriftliche Überlieferung der sogenannten Psellos-Paraphrase der Ilias, Hamburg. Verdenius, W.J. 1985. A Commentary on Hesiod, Works and Days, vv. 1–382, Leiden.

  Bibliography Vergados, A. 2017. ‘Der Dichter als Leser und (Fehl-)interpret: Hesiod in den homerischen Scholien’, in: J. Grethlein/A. Rengakos (eds.), Griechische Literaturgeschichtsschreibung. Traditionen, Probleme und Konzepte, Berlin/Boston, 271–297. Vian, F. 1969. Quintus de Smyrne, La suite d’Homère, vol. III: Livres X–XIV, Paris. Vian, F. 1976. Nonnos de Panopolis, Les Dionysiaques, vol. I: Chants I–II, Paris. Viljoen, H.G. 1915. Herodoti fragmenta in papyris servata, Groningen. Villoison, Jean Baptiste Gaspard d’Ansse de. 1788. Homeri Ilias ad veteris codicis Veneti fidem recensita. Venetiis. Virgilio, B. 1993, Gli Attalidi di Pergamo: Fama, eredità, memoria, Pisa. Vitarelli, M.C. 1981. ‘Sul testo e la tradizione delle hypotheseis dell’Iliade’, in: Ricerche di Filologia Classica I. Studi di Letteratura Greca, Pisa, 125–137. Volkmann, R. 1872. Die Rhetorik der Griechen und Römer in systematischer Übersicht, Berlin. Wachsmuth, K. 1860. De Cratete Mallota, Lipsiae. Wackernagel, J. 1876. ‘Nicanor und Herodian’, RhM 31, 431–439. Wackernagel, J. 1935. Kleine Schriften, vol. I, Göttingen. Waerden, B.L. van der. 1959. ‘Klaudios Ptolemaios’, RE XXII 2, 1959, 1788–1859. Waeschke, H. 1874. De Aristarchi studiis Hesiodiis, Lipsiae. Wartelle, A. 1971. Histoire du texte d’Eschyle dans l’antiquité, Paris. Weber, R. 1889. De Dioscuridis Πεϱὶ τῶν παϱ᾿ Ὁμήϱῳ νόμων libello, Leipziger Studien z. class. Philol. 11, Lipsiae. Webster, T.B.L. 1953. Studies in Later Greek Comedy, Manchester. Wehrli, F. 1967. Dikaiarchos, Die Schule des Aristoteles 1, Basel/Stuttgart. Wehrli, F. 1967a. Aristoxenos, Die Schule des Aristoteles 2, Basel/Stuttgart. Wehrli, F. 1968. Demetrios von Phaleron, Die Schule des Aristoteles 4, Basel/Stuttgart. Wehrli, F. 1969. Herakleides Pontikos, Die Schule des Aristoteles 7, Basel/Stuttgart. Wehrli, F. 1969a. Phainias von Eresos. Chamaileon. Praxiphanes, Die Schule des Aristoteles 9, Basel/Stuttgart. Wehrli, F. 1974. Hermippos der Kallimacheer, Die Schule des Aristoteles Suppl. 1, Basel/ Stuttgart. Wehrli, F. 1983. Der Peripatos bis zum Beginn der römischen Kaiserzeit, in: H. Flashar (ed.), Die Philosophie der Antike, vol. 3: Ältere Akademie - Aristoteles - Peripatos, Basel/Stuttgart, 459–599. Wellmann, M. 1894. ‘Anonymi (6)’, RE I, 2327. Wendel, C. 1914. Scholia in Theocritum vetera, Lipsiae. Wendel, C. 1921. Überlieferung und Entstehung der Theokrit-Scholien, Abh. Akad. Wiss. Göttingen, phil.-hist. Klasse, XVII 2, Berlin. Wendel, C. 1934. ‘Theon’, RE V A 2, 2054–2059. Wendel, C. 1934a. ‘Telephos Nr.’, RE V A 1, 369–371. Wendel, C. 1935. Scholia in Apollonium Rhodium vetera, Berolini. Wendel, C. 1936. ‘Nikias Nr. 29’, RE XVII 1, 337. Wendel, C. 1936a. ‘Nikanor Nr. 27’, RE XVII 1, 274–277. Wendel, C. 1939. ‘Tryphon Nr. 25’, RE VII A 1, 726–745. Wendel, C. 1942. ‘Palamedes (n. 3)’, RE XVIII.2, 2512–2513. Wenkebach, E. 1936. Galeni in Hippocratis Epidemiarum librum III, Lipsiae/Berolini (= CMG V,10,2,1). Wentzel, G. 1985. ‘Artemidoros Nr. 31’, RE II, 1331–1332. Wentzel, G. 1986. ‘Asklepiades Nr. 28’, RE II, 1628–1631.

Bibliography  

West, M.L. 1965. ‘Tryphon De tropis’, CQ 15, 230–248. West, M.L. 1966. Hesiod, Theogony, Oxford. West, M.L. 1967. ‘Alcman and Pythagoras’, CQ n.s. 17, 1–15. West, M.L. 1978. Hesiod, Works and Days, Oxford. West, M.L. 1980. Delectus ex iambis et elegis Graecis, Oxonii. West, M.L. 1982. Greek Metre, Oxford. West, M.L. 1985. The Hesiodic Catalogue of Women, Oxford. West, M.L. 1985a. ‘Ion of Chios’, BICS 32, 71–78. West, M.L. 19892. Iambi et Elegi Graeci ante Alexandrum cantati, vol. 1. Oxonii. West, M.L. 1990. Studies in Aeschylus, Stuttgart. West, M.L. 19982. Aeschyli Tragoediae, cum incerti poetae Prometheo, Stutgardiae/Lipsiae. West, M.L. 1998–2000. Homeri Ilias, 2 vols., Stutgardiae/Lipsiae. West, M.L. 1999. ‘The invention of Homer’, CQ 49, 364–382. West, M.L. 2001a. Studies in the Text and Transmission of the Iliad, München/Leipzig. West, M.L. 2001b. ‘Response by M.L.W. on Nagy and Nardelli’, BMCR 2001.09.06. West, M.L. 2002. ‘Zenodotus’ text’, in: Montanari/Ascheri 2002, 137–142. West, M.L. 2003. Homeric Hymns, Homeric Apocrypha, Lives of Homer, Cambridge, MA/London. West, S. 1967. The Ptolemaic Papyri of Homer, Köln/Opladen. West, S. 1970. ‘Chalcenteric Negligence’, CQ 20, 288–296. West, S. 1974. ‘Satyrus: Peripatetic or Alexandrian?’, GRBS 15, 279–287. West, S. 1981. Omero, Odissea, vol. I: Libri I–IV, Milano 1981. West, S. 1988. ‘The Transmission of the Text’, in: A. Heubeck/S. West/J.B. Hainsworth, A Commentary on Homer’s Odyssey, vol. I, Oxford, 33–48 (= Omero, Odissea, vol. I, Milano 1981, xxxix–lix). West, S. 2011. ‘The Papyri of Herodotus’, in: D. Obbink/R. Rutherford (eds.), Culture in Pieces. Essays on Ancient Texts in Honour of Peter Parsons, Oxford, 69–83. White, J.W. 1912. The Verse of Greek Comedy, London. White, J.W. 1914. The Scholia on the Aves of Aristophanes, Boston/London. Wilamowitz, U. von. 1886. Isyllos von Epidauros, Berlin. Wilamowitz, U. von. 1888. ‘Zu den Homerscholien’, Hermes 23, 142–147. Wilamowitz, U. von. 31906. Einleitung in die Griechische Tragödie, Berlin. Wilamowitz, U. von. 1925. ‘Die griechische Heldensage’, Sitz. Preuβ. Akad. Wissensch. 1925, 41–62 (= Kleine Schriften V 2, Berlin 1971, 54–84). Wille, G. 1967. Musica Romana. Die Bedeutung der Musik im Leben der Römer, Amsterdam. Wilson, N.G. 1967. ‘A Chapter in the History of Scholia’, CQ 17, 244–256. Wilson, N.G. 1971. ‘An Aristarchean Maxim’, CR 31, 172. Wilson, N.G. 1971a. ‘The Vienna Dioscorides and the History of Scholia’, GRBS 12, 557–558. Wilson, N.G. 1976. ‘Aristarchus or a Sophist’, PCPhS 22, 123. Wilson, N.G. 1976b. ‘A Manuscript of the D-Scholia’, ZPE 23, 61–62. Wilson, N.G. 1977. Review of G. Zuntz, Die Aristophanes-Scholien der Papyri, CR 27, 271. Wilson, N.G. 1984. ‘Scoliasti e commentatori’, SCO 33, 83–112. Wilson, N.G. 1990. Filologi bizantini, Napoli (orig. London 1983). Wismeyer, J. 1884–1885. Die durch Scholien nicht erklärten kritischen Zeichen der Iliashandschrift Venetus A, Progr. der K. Studienanstalt Passau. Wolf, F.A. 1795. Prolegomena to Homer (transl. with introd. and notes by A. Grafton/G.W. Most/ J.E.G. Zetzel, Princeton, 1985).

  Bibliography Worp, K.A. 2012. ‘A New Survey of Greek, Coptic, Demotic and Latin Tabulae Preserved from Classical Antiquity. Version 1.0, February 2012’, Trismegistos Online Publications 6. Wouters, A. 1976. ‘A Note on the Homer Commentary P.W.U. Inv. 217’, ZPE 21, 271–273. Wouters, A.P. 1979. The Grammatical Papyri from Graeco-Roman Egypt. Contributions to the «Ars grammatica » in Antiquity, Verhandelingen van de Κoninklijke Academie voor Wetenschappen, Letteren en Schone Kunsten van België. Klasse der Letteren, Jaargang 41, Nr. 92. Wyss, B. 1936. Antimachi Colophonii reliquiae, Berolini. Yorke, E.C. 1954. ‘Mesatus Tragicus’, CQ 4, 183–184. Zalateo, G. 1961. ‘Papiri scolastici’, Aegyptus 41, 160–235. Zanetto, J. (ed.) 1993. Euripides, Rhesus, Stutgardiae/Lipsiae. Ziegler, K. 1936. ‘Timolaos (6)’, RE VI A, 1, 1275–1276. Zimmermann, B. 1998. ‘Hypothesis’, in DNP V, 819–820. Zimmermann, B. 2006. ‘Aischylos-Rezeption im 5. Jahrhundert v. Chr.’, Lexis 24, 53–62. Zumbo, A. 1987. ‘Lingua omerica e varianti testuali nella ‘recensio’ alessandrina di Riano di Creta’, in: G. Bolognesi/V. Pisani (eds.), Linguistica e filologia, Milan 1984, Brescia, 558– 570. Zumbo, A. 1988. ‘Cinque variae lectiones di Riano di Creta in Omero, Odissea VI’, in: S. Costanza (ed.), Poesia epica greca e latina, Soveria, 113–118. Zuntz, G. 19752. Die Aristophanes-Scholien der Papyri, Berlin.

General Index acephalous verse 499ff. Aeschines 112f., 127, 152, 834ff. Aeschylus 13, 35, 111, 151f., 198, 691ff., 725, 728, 733 – Agamemnon 712 – Aeth(ae)ae 693, 707f., 713 – Danaids 710 – Diktyoulkoi 110, 151 – Glaucus Marinus 712 – Laius 709 – Lycurgus 717 – Myrmidones 712 – Oedipus 709 – Persians 702, 719f., 733 – Philoctetes 708 – Prometheus 730ff. – Proteus 718, 727 – Seven Against Thebes 709, 712f., 733 – Suppliants 708, 710 – Xantriae 712 Alcaeus 35, 52 Alcidamas 644f. Alcman 35, 40, 243, 713, 752 Alexander Aetolus 31, 123, 145, 714 Alexander Cotiaeus 742 Alexandria 205ff. Alexion 41 allegoresis/allegory 10, 24f., 35, 37f., 43, 50ff., 53, 222f., 319, 407, 414, 510, 518, 532, 613, 643, 655f., 767, 788 Ammonius 98, 160ff., 349ff., 357ff., 524ff., 628ff., 771, 812 analogy 19f., 37, 82f., 441ff. Andronicus of Rhodes 245 anomaly 19f., 37, 222, 441ff., Antiatticist 845ff. Antigonus of Carystus 36 Antimachus of Colophon 26, 417f., 813 Antiphanes comicus 762 antisigma 33 antistrophe 74 Apion 18, 42, 610 Apollo 214, 287, 314f., 608, 610, 612, 684 https://doi.org/10.1515/9783110772371-064

Apollodorus of Athens 36, 39, 223, 225f., 317, 562f., 804, 814 – Chronika 39, 220 – On the Catalogue of Ships 36, 39, 453 – On the Gods 39, 317, 492 n. 11 Apollonius Dyscolus 19ff., 41, 44, 66, 440, 444, 599 Apollonius Eidographos 33 Apollonius of Rhodes 31, 42, 198, 207, 214, 540, 653, 824ff., 849ff., 858 Against Zenodotus 31 Apollonius Sophista 18, 42, 296, 319f., 461, 610 Aratus 13, 37, 40, 548, 819ff., 829f. Arcesilaus of Pitane 218 Archilochus 31, 35, 643, 645 Archimedes 211 Aristarchus 14f., 19f., 29ff., 33f., 44f., 50, 97ff., 138f., 148f., 159f., 169ff., 175ff., 184ff., 223ff., 255, 339ff., 416ff., 449f., 453f., 488ff., 555ff., 613ff., 647ff., 666ff., 685, 687, 717f., 735ff., 751, 755, 757ff., 771ff., 786, 814ff., 819ff., 853f. – diorthos(e)is/ekdos(e)is 98ff., 160ff., 341ff., 348ff., 357ff., 524ff., 626ff., 725 Aristippus of Cyrene 214 Aristonicus 12, 42, 84ff., 148f., 172, 179f., 344, 354f., 496ff., 506f., 559, 565, 567f., 614, 669, 757ff., 812, 857f. Aristoxenus 775, 793 Aristophanes 13 n. 17, 198, 262f., 692ff. Aristophanes of Byzantium 17, 20, 23, 32f., 98f., 132, 169ff., 220, 237, 348ff., 525, 563f., 598, 626f., 653f., 665f., 668, 674, 691, 705, 715f., 742, 752, 786, 828, 857 n. 25 Aristotle 19, 22f., 24ff., 45, 58f., 69f., 245ff., 521f., 599ff., 611f., 620, 658f., 700ff., 728ff., 758ff., 774f., 778, 781f., 786ff., 790f., 795 Artemidorus 850, 853 Asclepiades of Myrlea 22, 40, 223, 304, 307f., 809, 825, 850, 853 asteriskos 33, 668

  General Index Athenaeus of Naucratis 4 n. 2, 43, 554, 653, 769ff. Athenocles of Cyzicus (grammarian) 770f., 782 athetein, athetesis 30, 61, 95, 107, 130, 139 n. 60, 156ff., 170, 345ff., 354ff., 417f., 426ff., 620, 625, 634, 648f., 670, 808 Attalids 215ff., 218ff. Bacchius of Tanagra 212f. Bacchylides 35 biography 4f., 27 Bion 850 Callimachus 13, 27, 31, 42, 208, 210, 214f., 246f., 419ff., 662, 796ff., 805ff., 849ff. – Against Praxiphanes 31 – Pinakes 31, 220 Callistratus 417f., 633 Campanella, Tommaso 80f. celestial bodies 72ff., 82f. celestial movements 72ff., 81ff. Certamen Homeri et Hesiodi 644ff. Chamaeleon 646, 700ff., 794 Charax 226 Chares (grammarian) 821, 824 chi 668, 713 n. 82 chorizontes 34, 261, 402ff., 446 chorus 72ff. Chrysippus 19, 23 Claudius Aelianus 4 n. 2 Claudius Ptolemaeus 72ff., 76f. Comanus of Naucratis 667 conjectures 63ff., 100f., 118f., 121f., 133ff., 136, 163ff., 174ff., 339ff., 543ff., 622 Crates of Mallus 36f., 40, 220ff., 633, 649ff., 669f., 821 Cratinus 258ff., 845ff. critical sign(s) (semeion, semeia) 88ff., 106f., 124f., 129ff., 146, 155f., 556f., 837 Cypria 316 Cyrene 213ff. Cyrillus, glossary of 18f.

Demosthenes 15, 46, 187, 459, 762 n. 4, 861 Demetrius of Adramyttium (Ixion) 224, 273, 598, 763f. Demetrios Chloros 766, 850 Demetrius Gonypesós 764 Demetrius of Phalerum 29, 208, 619, 762ff., 782f., 785f., 791f. Demetrius of Scepsis 36, 225ff., 765f. Democritus 25, 603 Demodokos 775ff. Demon 23 diagraphō 157 n. 47, 345 n. 16 Dicaearchus 616ff., 694ff., 703f., 775f., 783ff., 787, 791f. Didymus 12, 15, 23, 41f., 72, 87, 136 n. 51, 138f., 163, 169, 176ff., 316f., 349, 351f., 354f., 445, 546ff., 563, 565, 633, 669, 716ff., 740f., 751f., 861f. Diodorus of Tarsus 598 Diogenes Laertius 5, 43, 220, 273, 646, 762, 769, 773 Diogenianus of Heraclea 18, 23 Dionysius of Halicarnassus 14, 723, 747ff., 767f. Dionysius Periegetes 13 Dionysius Thrax 15, 19ff., 39f., 180f., 223f., 448ff., 495ff., 503, 508, 571 – Against Crates 180f., 223f., 530f. – Technē grammatikē 20, 40, 570 Diogenianus 18, 717 diorthosis (recension)/diorthotes 8ff., 30f., 88, 92ff., 106ff., 117f., 123f., 128ff., 135f., 145f., 154f., 158, 339ff., 524ff., 623ff., 807, 836ff., 844 Dios boulē 315f., 651 n. 30 diplē 34, 148f., 303, 351ff., 494ff., 534f., 556ff. diplē periestigmenē 34, 558f., 857f. ekdosis 8ff., 30f., 88ff., 106ff., 118f., 121ff., 134f., 144ff., 158, 339ff., 621ff., 834ff. empeiria 21, 811 n. 19, 862 n. 30 Epaphroditus 654, 670 Epicharmus 39 epode 74

General Index  

Erasistratus 211f. Eratosthenes of Cyrene 32, 51, 215f., 308f. – Chronographiai 32 – Geographika 32 Etymologicum Magnum 19 n. 37, 199f. Eubulus 402ff., 408ff. Euphorion 276, 307, 813 Eupolis 260ff. Euripides 13, 198ff., 692ff. Eustathius 53, 423f., 502f., 614, 730ff. Galenus 11, 42, 93, 129, 133ff., 167f., 227f. Glaucus of Samos 702f., 722 glossai 17, 123, 536ff., 597ff., 599ff., 743 – Cretan 604, 607ff. – Cypriot 604ff. – dialectal 599ff. glossography 16f., 24, 29 gnomai 22 grammar 19ff., 32ff., 40f., 55 grammarian (grammatikos) 26f., 40, 107f., 118, 121, 131f., 158f., 207, 339, 440, 566, 596, 620, 627, 632, 671 Habron 671 hapax legomenon 34, 504, 538f. Harpocration 42 Hedylus (of Samos?) 850f. hellēnismos 441 Hephaestion 42 Heraclides of Pontus 645f., 699 n. 42, 792f. Heraclitus (the Allegorist) 11, 38, 51 Herennius Philo 812 Hermogenes 5f. Herodianus 12, 19ff., 41, 445, 500, 503ff. Herodotus 13f., 735ff., 738ff., 753f. Herophilus 211 Hesiod/[Hesiod] 13 n. 17, 31, 35, 37, 197, 643ff. – Aspis 652ff., 664f. – Catalogue of Women 653f., 664f. – Ornithomanteia 664f. – Precepts of Chiron 654, 665 – Theogony 647ff., 655ff.

– Works and Days 644ff., 660f., 666f. Hesychius 18, 717 Hipponax 229ff. historiae, mythographical 280ff., 296ff., 304ff., 323ff., 463ff. Homer 10ff., 24, 32, 34f., 37, 50ff., 197, 202, 266ff., 288, 613, 615, 643f., 662, 668, 671, 777f., 794 Homeric text 26, 30f., 42, 88, 106f., 123, 126, 131ff., 139ff., 145ff., 163ff., 170, 176, 183 n. 26, 309 n. 31, 339ff., 370, 414, 443ff., 450ff., 504, 509, 543ff., 561, 584ff., 621ff., 827, 837 Homēron ex Homērou saphēnizein (“explaining Homer by Homer”) 314f., 566f., 619 hyperbaton 274 Hypereides 845ff. hypomnema 8ff., 12, 34, 44, 47f., 85f., 88f., 97f., 107f., 132f., 145, 160f., 185f., 247, 341, 349f., 356ff., 360ff., 366f., 460f., 480f., 525, 563f., 591ff., 595f., 621, 627ff., 676, 739ff., 810f., 829, 836, 861f. hypothes(e)is 33, 92, 430ff., 455ff., 465ff., 477ff., 585f., 691ff., 706ff., 729 Ion of Chios 699, 725f., 728 Iosephus Flavius 451f. kakometros verse 506ff. kata poleis editions 595 kola 33 koronis 33 kritikos 36f., 208 Lacydes of Cyrene 214 lagaros verse 499ff. Leogoras of Syracuse 532ff. Leschides 226 lexeis 8, 44, 123, 585, 743f. Lexeis Homerikai 462f., 470f., 482ff., 488ff. lexicography 16ff., 33, 44, 296, 480, 599, 609 Library, Alexandrian 29, 31ff., 120, 145, 172, 207, 715

  General Index Library of Pergamum 36, 220f. Lucillus of Tarrha 23 Lycon Peripateticus 218 Lycophron 13, 31, 123, 145, 540f., 714 manuscript collation 100f., 118f., 121f., 136ff., 141f., 163ff., 544ff., 622, 837 Margites 202f., 522 Massaliotikē ekdosis 595 mathematics 211 Megaclides 548, 653f. meiouros (myouros) verse 499f. Menippus of Pergamum 226 Mimnermus 35 Moschus 850 Museum of Alexandria 29, 60, 157, 207, 345, 546, 625, 840 Mythographus Homericus 286ff., 295ff., 304ff., 323ff., 478ff., 512f., 586, 593 n. 34, 764f. Nemesion 352, 355 Neoptolemus of Parium 17 Nicander 13, 42, 226 Nicanor 12, 42, 113, 153, 495, 511, 560, 650 Nicias (grammarian) 812f. numerus versuum 94f., 96f., 132, 159, 591, 595, 624 obelos 33, 63f., 71, 88ff., 95f., 129f., 139ff., 155ff., 169f., 252, 345, 350ff., 355f., 806f., 838 Occident 77f. Olymp 532ff. Oppian 13 Orient 77f. ouk einai, ou feresthai, ou grafein 95, 156f., 158 n. 51, 159f., 624f. Palamedes of Elea 235ff. Pamphilus of Alexandria 18, 717 Panaetius of Rhodes 224f. paragraphos 33, 153, 167, 559f., 839f. paraphrase 16f., 243ff. paroemiography 4, 22ff., 27, 44 paroimiai 23

Patrizi, Fran cesco 80 Pergamum 10, 35ff., 40, 120f., 215ff., 221ff. perigraphō 157 n. 47, 345 n. 16 Peri-Literatur 86f., 769 Peripatetics/Peripatos 27ff., 45, 58ff., 69f., 144, 207ff., 218f., 611 n. 22, 619f., 645f., 700f., 703ff., 721ff., 724f., 728f., 760, 772, 780ff. Phaenias of Eresus 792 Phemios 776 Philemon 17 Philicus 26 Philitas of Cos 17, 208, 804 – Ataktoi glossai/Atakta 29 Philochorus of Athens 703f. Philodemus 563 philologos 40 Philostratus 5 Philoxenus 21f. Photius, Lexicon 18f. Pindar 12, 35, 197, 673ff., 682ff. Plato 25, 202, 519f., 658, 767 – Cratylus 17 – Protagoras 49 – Theaetetus 110f., 151 plus-verses 97, 125f. Plutarch 5, 23, 221 Polemon of Ilium 36, 237 Pollux 18 Porphyry 11, 38, 43, 406f., 566f., 620, 783 Posidippus 109f., 126f., 150 Praxiphanes 27 n. 59, 60, 646f., 651, 781, 794f. Priscianus 741f. Problemata-Literatur 86f., 769 Prodicus 25, 603 Protagoras 25 Pseudo-Apollodorus 300f. Pseudo-Zonaras 18f. Ptolemaeus of Ascalon 22, 41, 224 Ptolemaeus Epithetes 75f., 212, 856f. Ptolemaeus Pindarion 440ff., 453f.

General Index  

Rhesus 697 rhetoric 5f., 21, 42, 44, 59, 543, 566, 762, 767, 789, 792 Rhianus of Crete 31 Satyrus the Peripatetic 701f., 831ff. scholia 7ff., 12ff., 189ff., 192ff., 197ff., 256ff., 564ff., 593 n. 34, 671f. scholia D 12, 280ff., 290ff., 295ff., 304ff., 323ff., 331ff., 383ff., 461f., 475f., 482ff., 488ff., 511ff., 578ff., 586 scholia exegetica (Homeric) 12 n. 16, 363ff., 368, 383, 385ff., 511f., 594, 614f. scholia maiora (Homeric) 383ff., 460ff., 475f. scholia minora (Homeric) 299f., 414f., 431f., 462, 467ff., 476ff., 486f., 569ff., 584f., 593 n. 34 Seleucus Homericus 654 Sextus Empiricus 694ff. sigma 33, 797f. Solon 603 Sophocles 13, 35, 198, 692ff., 725, 849 – Oedipus Rex 696, 726f. – Trachiniae 691 – Ichneutae 691 – Poimenes 185, 736 – Theseus 691 Sophron, mimes 39 Sosibius Lacon 598 Sotades 211 stigmē 113, 153, 355 Strato of Lampsacus 208, 792 strophe 73f. Suda 18, 664, 670, 753 synetheia 20, 441 synetheia homerikē 445ff. syngramma 8ff., 12, 44, 85f., 145, 563f., 591ff. Telephus 38, 225 Theagenes of Rhegium 24

Theocritus 13, 27, 40, 42, 198, 210f., 849ff. Theodorides 851f. Theon of Alexandria 42, 563f., 673, 803f., 809f., 813, 820f., 827f. Theophrastus 208, 792 Thucydides 14, 746ff., 753f. Timolaos of Larisa 777 Timon of Phlius 548, 550 n. 26 Timotheus 108f., 125 Tlepolemus 684ff. Tryphon of Alexandria 21, 41, 598f., 671 Tyrannion 21f., 40 varia(e) lectio(nes) 70f., 91, 136 n. 51, 165 n. 68, 122, 133ff., 163ff., 174, 339ff., 545ff., 621ff. Viermännerkommentar (VMK) 12f., 87, 295, 352, 363f., 368, 383, 386, 460, 473f., 511, 565, 614 vulgata, Homeric 97, 125f., 132, 148ff., 159, 445 n. 7, 551 n. 28, 587f., 591, 627, 825 Xenocritus of Cos 213 Xenophon 14, 752 Zeno of Citium 655ff., 661ff. Zenobius 23 Zenodotus of Alexandria 664 Zenodotus of Ephesus 29f., 88ff., 95f., 106f., 123f., 129ff., 137ff., 145f., 169ff., 207, 339ff., 548f., 597, 624ff., 662f., 668, 757ff., 772, 805ff., 825ff., 854ff. – Glossai 30, 597 Zenodotus of Mallus 223, 664 Zeuxis 213 Zodiac 79f. Zoilus 548, 636 n. 40, 768

Index of Passages Discussed Acusilaus of Argos (FGrHist 2) F 22 312 Aeschines 3.51–53 3.162–163

91, 112–115, 127, 152, 839f. 112, 127, 152, 839f.

Aeschylus PV 24 732 383 731 Fragments (TrGF 3, Radt) T1 701 T2 701 T 14 728 T 58 709 n. 71 T 65 718 n. 108 T 67–69 717 n. 105 T 68 718 n. 112 T 78 702 T 86 702 n. 53 T 93b 699, 725 T 103 701, 724 T 113a 725 T 117ab 701, 724 T 145 700 T 146 715 n. 95 T 147 714 n. 91 T 149a 725 T 234–236 717 n. 103 F 124–126 717 n. 105 F 188a 714 F 309–311 701 F 317a 716 n. 98, 725 F 462 719 n. 119 F 476a 714 Alcman Fragments (PMG) 161(c)

714 n. 86

https://doi.org/10.1515/9783110772371-065

Αlexander Aetolus Fragments (Magnelli) T7 123 n. 10, 145 n. 6 Ammonius (Nickau) 352 812 451 751 Anonymous (Heitsch) Carminis de viribus herbarum fragmentum 178 541f. Anonymous (Spengel) περὶ σχημάτων iii 188.5–7 274 Antiatticista (Bekker) 77.27 845 100.32 845 Apollodorus (FGrHist 244) F 165 453 n. 18 F 353 492 n. 11 Αpollonius Dyscolus Constr. (CG II 1) 241.14 440 Pron. 71.20ff. 444f. 79.25 440 n. 2 Apollonius of Rhodes Arg. 1.59–64 311 1.1089 540 2.500–510 214 2.601 540 3.543 540 Αpollonius Sophista (Bekker) 33.12 409 44.8 271 74.6 577

  Index of Passages Discussed 79.8 83.19 125.32 135.23 143.9

577 579 859 273 610

Aratus 26–44 28–29 42–44 226–227 945

823 829 823 823 828

Archilochus (West) Fragments 213 719 n. 120 Aristophanes Eq. 96 259 Fragments (Kassel–Austin) 233 16 n. 31, 413, 536f., 603 Aristophanes of Byzantium Fragments (Slater) 1 237 n. 32 16–17 742 n. 24 33 742 n. 24, 752 369 762 n. 4 381 674 n. 5 405–407 665f. n. 73, 75 Aristoteles Metaph. 984 b 27 995 a 4ff. 1074 b 1ff. Poet. 1448 a 20ff. 1448 b 27ff. 1448 b 34ff. 1449 b 9ff. 1452 a 14ff. 1452 a 32f. 1455 b 16–23 1456 b 9 1456 b 20ff.

658f. 25 n. 55 25 n. 55 202 778 202 202 521 521 774, 786 599 19, 599f.

1457 a 35 1457 b 1ff. 1458 a 21–23 1458 a 34 1458 b 2–5 1458 b 11–1459 a 9 1458 b 13–18 1459 a 9–11 1459 a 17ff. 1459 b 12ff. 1459 b 32–37 1460 a 5ff. 1461 a 10ff. Rh. 1404 a 25–35 1404 b 10 1406 b 2 Top. 140 a 2–6 Fragments (Rose) 143 145 146 175

604 17 n. 34, 599 599 600 600 601 600 599 202 521 599 202 17 n. 34, 599, 607 602 602 602 602 n. 12 787 620, 758f., 788 406 25, 788

Aristoxenus (Wehrli) Fragments fr. 113–116 699 fr. 123 775 Athenaeus III 113 V 1–20 V 177e–f V 181c XII 512e–f XIV 63a

410 771 n. 28 769f., 781f. 356 n. 27 653 226 n. 65

Atilius Fortunatianus (Keil) p. 294, 28ff. 77 Bacchylides 5.38–40

678

Callimachus (Pfeiffer) Hymn to Apollo 65–68 214

Index of Passages Discussed  

Hymn to Delos 286 Fragments 23.3 55 176 177.6 383 632 637

801 801 850 n. 5 803 857 803 822 813 n. 30

Chamaeleon (Wehrli) Fragments 29 646 n. 14 39–42 700f. 46 646 n. 12 Cheroboscus Th. 167.37ff.

259

Chorizontes (Kohl) Fragments XIX 261 Crates (Broggiato) Fragments 1 650 78 649 Cratinus (Kassel–Austin) Fragments 2 261 34 845 108 258 164 259 309 845 331 258 97 259 408 847 444 259 Demetrius of Phalerum (Wehrli) Fragments 42 768 n. 20 69 773 n. 31, 785 ν. 10 74 762 n. 2 163 762 n. 5

164 170 188 189 190–193 190 191 192 193 194 195 196 207 208

762 n. 5 767f. 762 n. 1 762 n. 2 762 n. 3, 764, 763f., 769 775f., 793 775–777, 793 773, 786 762 n. 4 768 762 n. 1 764f. 766

Dicaearchus (Mirhady) Fragments 95 (= 92 Wehrli) 783 39 (= 93 Wehrli) 775f., 793 n. 20 101 696 112–115 695–697 Diogenes Laertius 2.83 5.26 5.37 5.50 5.58–60 5.58 5.83–85 5.87 5.92 7.25–26 7.34 9.26

214 699 n. 39 208 n. 11 702 n. 52 792 208 n. 12 273 646 646 661f. 220 832 n. 4

Dionysius of Halicarnasus Pomp. 3.16 745 Thuc. 5 749f. Dionysius Thrax Ars 4

153 n. 39

  Index of Passages Discussed Diphilus (Kassel–Austin) Fragments 59 817, 852 Epahroditus (Braswell/Billerbeck) Fragments 54–55 654 Ephorus (FGrHist 70) F 146 406f. Etymologicum Magnum (Gaisford) 53.20 539 72.12 812 n. 21, 851 n. 7 142.43 850 n. 5 172.36 272 177.46 539 358.24 821 ν. 57 539.22 667 552.5 742f. 579.1 260 n. 17 690.42ff. 77 722.22f. 742f. Etymologicum Orionis (Sturz) 96.28–29 668 Eubulus (Kock) Fragments 120 139

403, 411 n. 13 408–415

Eupolis (Kassel–Austin) Fragments 22 261 49 260 102.3 261 102.7 261 309 261 342 260 352 260 472 260 483 261

Euripides Phoen. 751 Fragments (Nauck) 73a 367 913.2

719 312 801 412 n. 15

Εustathius Commentarii ad Homeri Iliadem 2, 24ff. 730f. 313, 34 406 458, 13ff. 505 n. 29 722, 60 608 750, 34 538 814, 22 732 890, 62 605 952, 59–60 579 1058, 5 410 n. 6, 801 n. 19 1246, 26 605 1647, 22ff. 501 n. 20 1720, 30ff. 403 n. 6 1907.10 272 1907,16–22 273 1907,40, 42 272 1908.2 272 Commentarii ad Homeri Odysseam 1417, 23ff. 423f. 1464, 62ff. 502 n. 24 Hellanicus (FGrHist 4) F 26 a–b 283 n. 7 Heracleides Ponticus (Wehrli) Fragments 22.24 646 n. 12 170 699 171 406 176 646 n. 12, 13 177 646 n. 12, 13 179 699 180 699 Herennius Philo (Palmieri) 132 812

Index of Passages Discussed  

Hermippus (Wehrli) Fragments 58 773 n. 31, 785 n. 10 92 773, 785 n. 10

Hipponax (Degani) Fragments 23 235 n. 18 129 229–241

Hermogenes περὶ ἰδεῶν (Rabe) 305f. 274 περὶ μεθόδου δεινότητος (Rabe) 14 (p. 430.1–4) 274 Herodotus 1.142.3–4 603 1.194.2 738 1.215.1 738 2.129–130 741 5.9.3 600 n. 8

Homer Iliad A5 A 399–400 Α 471–475 B 204–207 B 317–320 Β 319 B 407ff. Β 409 B 484 B 667–670 E 1–48 E 178 E 387 H 4–80 H 75–83 Θ 365–369 Θ 423 I 222 Ι 454–468 Κ 274f. K 310–312 K 316 K 394–399 K 599–799 Λ 11–263 Λ 136–251 M 91–92 M 110–111

Hesiod Th. 116–120 126–127 142–143 311 Op. 530 571 650–660 Fragments (M.-W.) 183 214 357

666 765 644

Hesychius α 2614 α 4962 α 5212 α 7199 α 7260 ε 5210 κ 3700–1 μ 1729 μ 1824 π 3581–2

537, 541 579 409 579 271 577 538 741 n. 18 741f. 273

657f. 669 670 682 669 670 644

M 136 M 139–140 M 147 N 216–218 N 225 N 599–799 Ξ 227–521 O 17–191 O 610–614 O 618–624 O 641–661

315f., 651 n. 30 315 558f. 769 756f. 619, 756–761 770 619, 774, 781, 788 650 684 575 418 605 575 258 682 500 639 n. 45 574f. 235 352f., 356 601, 604, 610 353–355 571 570 574 364, 366, 369–372 362–364, 366, 369– 372 372f. 374 364 297f. 485f. 574 570, 574 571, 574 375–377 363 378–381

  Index of Passages Discussed Ο 717 Π 746–747 Σ 592 Υ 147–404 Φ 481 Ψ2 Ω 92 Odyssey α 97–98 α 328–335 β 136–137 β 201–218 γ 162 δ 244 ζ 287–288 κ 36 κ 190–192 κ 243 λ 233 λ 235–332 λ 292–310 λ 298 λ 305 λ 321–ξ 520 λ 519–520 μ 15 μ 127–136 μ 423 ρ 225 σ 74 τ 174 χ 424 ψ 207–220 ψ 296

538f. 402–404 854 280–294 500 499 558

[Longin.] de subl. 22.1–4

274

Lycophron Alex. 33ff. Alex. 295

283 540f.

595 616–620 416–429, 813 n. 30 276–279 272 440 n. 2 784 501 274 411 273 303, 329 303 303f. 303 330f. 303 411 788 499 821 821 404 784 267–275 773f., 785 n. 10

Macrobius Commentarii in Somnium Scipionis II 3, 5 79–81

Homeric Hymn to Apollo 166–173 649f. Lexeis Homericae (de Marco) α 288 541f. α 524 579 α 733 579 ε 701 577 Iosephus Flavius C. Ap. 1, 11–12

451

Oros (Alpers) 57 Papyri P.Achmim 2 P.Alex. inv. 198 P.Alex. inv. 28759 P.Amherst 2.12

P.Amherst 2.19 P.Amst. I 6 P.Antin. 69 P.Berol.inv. 9780 P.Berol.inv. 9872 P.Berol.inv. 9875 P.Berol.inv. 9875 P.Berol.inv. 9960 P.Berol. inv. 9965 P.Berol.inv. 10508–10512 P.Berol.inv. 11521 P.Berol.inv. 11634 P.Berol.inv. 13230 P.Berol.inv. 13282 P.Berol.inv. 13419 P.Berol.inv. 21313 P.Bingen 8 B.Bodmer 2 P.Brux.inv. E 7162 P.Colon. 2381 P.Derveni P.Derveni VI, 5

845

467, 477, 585 268, 460 573 121, 184ff., 255, 562, 735–741, 744, 811, 829, 862 573 466 466, 477, 585 562, 673 92f., 110, 151 IV, 133 90, 108, 125, 147 V, 196 90, 108, 834 474, 477 716 n. 101 569–581 804 466 573 280–294, 303, 306– 308 362 n. 6, 367, 673 746 571f., 576 n. 16 92f., 111f., 152 533f. 466 125, 147 109

Index of Passages Discussed  

P.Derveni XXI, 11 P.Fay. 3 P.Fay. 312 P.Flor. P.Gen. III 118 P.Hamb. 199

109, 834 245f. 267–275, 460 460 582 n. 1 286, 289, 307, 309 n. 31, 465 P.Haun. I 3 575 P.Heidelb.Siegmann 200 461 P.Herc. 1012 715f. P.Herc. 1581 722 P.Ilias 12 (= P.Heid.Lit. 2 (inv. 1262–1266) + P.Hib. 1.22 (Bodl.Libr. inv. Ms.Gr.Class.b3(P)/2) + P.Grenf. 2.4 (Bodl.Libr. inv. Ms.Gr.class.b.3(P)) 125f., 147ff., 150 P.Köln 37 558 P.Köln 2281 576 P.Laur. III/53 466 P.Laur. III/278 112, 127, 152, 839 P.Lille 76 246f. P.Lille 79 803 P.Lille 82 803 P.Lille 83 + 134 + 93 b + 93 a + 114 t + 114 o +87 475 P.Lond.Lit. 27 558 P.Lond.Lit.30 475 P.Lond.Lit. 142 286, 298, 474, 478 P.Lond.Lit. 191 644 P.Louvre inv. 7733 248, 816f., 850 P.Med. 17 796 n. 3 P.Med.inv. 18 804 P.Med.inv. 210 268, 593 n. 35 P.Med. inv. 71.82 534 P.Med.inv. 72.13 573 P.Mich.inv. 9 461 P.Mich.inv. 1206 361f., 367f. P.Mich.inv. 2754 644 P.Mich.inv. 1315 466 P.Mich.inv. 3688 474 P.Mich.inv. 4832c 461 P.Mil.Vogl. 1.19 562 P.Mil.Vogl. III 119 573 P.Mil.Vogl. 259 303f., 305 n. 21, 329– 338 P.Mil.Vogl. 309 91, 834

P.Odyssey 31 (= P.Sorbonne.inv. 2245 A) 126, 148 P.Osl. II 13 571 P.Oxy. 19 742 P.Oxy. 220 713f. P.Oxy. 221 440 n. 1, 465, 474, 480, 594 n. 41, 673, 738 P.Oxy. 418 286f., 289 n. 18, 307, 462 P.Oxy. 418.9–22 310–316, 325, 328 P.Oxy. 574 455–458, 466 P.Oxy. 663 707 n. 66 P.Oxy. 853 746, 749f., 754 P.Oxy. 856 362 n. 6, 367 P.Oxy. 1086 460, 474, 480, 596, 673, 737f. P.Oxy. 1087 258, 260, 460, 474, 480, 673 P.Oxy. 1092 742 P.Oxy. 1174 691 P.Oxy. 1176 701, 831f. P.Oxy. 1367 564 n. 17, 697 P.Oxy. 1611 312, 330 n. 15 P.Oxy. 1619 742 P.Oxy. 1800 746, 800 n. 14 P.Oxy. 1803 746 P.Oxy. 1805 691 P.Oxy. 2004 843 n. 28 P.Oxy. 2081 746 P.Oxy. 2087 739, 746 P.Oxy. 2159–2163 712 n. 80 P.Oxy. 2160 713 n. 82 P.Oxy. 2161 91, 110, 151, 267, 723 P.Oxy. 2163 82 P.Oxy. 2164 712 P.Oxy. 2176 229–241 P.Oxy. 2178 712 n. 80 P.Oxy. 2179 712 n. 80 P.Oxy. 2221 248, 799 n. 8 P.Oxy. 2245–2255 712 n. 80 P.Oxy. 2245 723 P.Oxy. 2249 713 n. 82 P.Oxy. 2255 712 P.Oxy. 2256 92, 111, 151f., 708– 712, 723, 725 P.Oxy. 2257 706f., 725

  Index of Passages Discussed P.Oxy. 2262 P.Oxy. 2263 P.Oxy. 2333 P.Oxy. 2367 P.Oxy. 2373 P.Oxy. 2393 P.Oxy. 2397 P.Oxy. 2404 P.Oxy. 2452 P.Oxy. 2457 P.Oxy. 2506 P.Oxy. 2536 P.Oxy. 2636, 2637 P.Oxy. 2888 P.Oxy. 3003 P.Oxy. 3013 P.Oxy. 3151.2.6 P.Oxy. 3152 P.Oxy. 3159 P.Oxy. 3160 P.Oxy. 3224 P.Oxy. 3710 P.Oxy. 3722 P.Oxy. 3830 P.Oxy. 3832 P.Oxy. 3833 P.Oxy. 4096 P.Oxy. 4452 P.Oxy. 4453 P.Oxy. 4455 P.Oxy. 4652 P.Oxy. 5095 P.Oxy. 5201 P.Ryl. 23 P.Ryl. 53 P.Ryl. 536 P.Ryl. 537 P.Schubart 21 PSI I 18, 19 PSI 113 PSI 1000 PSI 1090

804 804 713 673, 680 n. 18 839 243 460 91, 112–115, 127, 152, 834–844 691 697f. n. 36 831–833 562f., 673 796 n. 3 268, 475, 481, 593 n. 35 289, 298f., 307f., 465 698 n. 36 272 267 467, 477, 575, 585 430–439, 467, 477, 585 668 268, 481 5 n. 3 298, 303, 307–309 276–279 585 299, 303, 306, 324, 326 595 593 n. 35 739 654f. 360–382, 594 673–681 466 267 298, 486f., 573 574 n. 12 286f., 302 n. 18, 303, 307f., 765f. 571f. 558f. 297f., 325 n. 6 839

PSI 1094

804

PSI 1173

280 n. 1, 286, 289, 298f., 301, 302 n. 18, 303f., 305 n. 21, 307f., 324f., 329– 338, 478, 586 PSI 1208–1210 712 n. 80 PSI 1209 723 PSI 1211 712 PSI 1219 804 PSI 1276 244, 584 PSI 1472 712 n. 80 P.Strasb.Gr. 1401 430–439, 467, 477, 585 P.Trieste 460, 474 P.Turner 13 576 P.Univ.Milan. 309 109–110, 126f., 150 P.Vindob. G 26010 752 n. 56 P.Vindob. G 29247746, 751 P.Vindob. G 29283 752 n. 56 P.Vindob. G 29779 712 n. 76 P.Vindob. G 29782 752 n. 56 P.Vindob. G 39940v 573 P. von Scherling G 99 574 PW.U.inv. 217 460, 474, 481 P.Yale inv. 551 268, 460 Pausanias 1.2.3 9.31.4–5

776 n. 37 666

Philicus (SH) 677

27

Philoxenus of Cythera (PMG) Fragments 823 858 Phrynichus (TrGF 1) Fragments T5 702 n. 53 F8 702 n. 53 Pindar Ol. 1

673–681

Index of Passages Discussed  

7.32–37, 49–51 9.6–10 Pyth. 3.43–44 Fragments (Sn.-M.) 249b 262 Plato Alcibiades II 147b 5–10 Cratylus 401b Ion 531a Protagoras 339a ff. 341c Rep. 392 e–394c 404b 595c Symp. 174b 178b Plutarch Antonius 58.9 591 Theseus 20 Mor. 2.26f.

685f.

683f. 682 682

52 n. 3 412 n. 15 643

Porphyrius (Schrader) Quaestiones Homericae I 48–49 405 Praxiphanes (Wehrli) Fragments 10 26 f. 22 646f.

49 603

Priscianus De figuris numerorum III 411.9 –412.2 Hertz–Keil

202 404 202

Ptolemaios Epithetes (Montanari) F1 857 F3 857

782 n. 4 658

Quintilian Institutio oratoria 8.2.14f. 8.6.65 10.1.54

221 221 645 n. 9 356 n. 27

[Plutarch] Vitae decem oratorum 845b 762 n. 5 [Plutarch] Περὶ Ὁμήρου Β8

Porphyrius In Platonis Timaeum commentariorum fragmenta (Sodano) 68.72 80

613

Polemon of Ilium (Preller) Fragments A6 237f. n. 30

742 n. 20

274 274 811 n. 20, 828f.

Rhetorica ad Herennium 4.44 274 Scholia on Aeschylus (Smith) Sept. 1 709 n. 71 Sept. 79a 718 n. 109 Sept. 751 719 Ag. 1, 20–22 718 n. 108 Ag. 42, 2 710 n. 72 Prom. 24 732 n. 5 Prom. 91 731 Prom. 383 731f. Scholia on Apollonius of Rhodes 1.1309 826 n. 65, 858 2.1052 825 n. 60

  Index of Passages Discussed Scholia on Aratus (Martin) 23 819f. 28 822f. 254–255 821 Scholia on Aristophanes Ra. 704 719 Ra. 965 719 Ra. 1028 719f. Ra. 1124 717f., 727 Thesm. 135 718 n. 112 Scholia on Bacchylides fr. 20A, 19 (= P.Oxy. 1361 fr. 5, I 3) 440 n. 1 Scholia on Dionysius Thrax (Hilgard) 183.5 453 n. 18 192.8ff. 453 490.7ff. 448f., 453 124.6–7 654 Scholia on Euripides Hec. 3 704 n. 59 Hec. 640 77 Scholia on Hephaestion (Consbruch) Ench. 289.4 506 n. 32 Ench. 291.1ff. 500 n. 18, 506 n. 33 Scholia on Herodotus (Stein) 1.71 745 1.85 745 1.125 745 1.215 744 4.5 744 Scholia on Hesiod Th. 5b2 26b 68a 76 114–115 116c1 116c2 117a

662f. 665 666, 668 667 666f., 670 663 663 n. 65 663f.

117b 660 126 665 n. 73, 669 138 667 142 670 160 670 178 668 233 668 253 667 270 670 338 666 n. 79 379 667 389 671 484 668 573 670 927 668 991 667 Op. 94a 666 n. 79 96a 670 97 667 150b 670 207–212 667 276b 668 304b 669 529–531 670 549 670 571–577 670 n. 101 649a 668 650–662 (= pp. 205–206 Pertusi) 645 740–741 667 828a 664f. Sc. 415 654, 670 hypothesis A Sc. 86f. Solmsen 652f. Scholia on Homer Iliad A 1a A5 A6 A7 A 8–9 A 10 A 11 A 18 A 38

203f., 522 315f., 490f. 391 309 n. 31, 391f. 392–396 396–399 400f. 532 289, 309 n. 31, 465

Index of Passages Discussed  

Α 39 A 50 A 263 A 264 A 332b A 353 A 399–406 A 399–400 A 400 A 405–436 Α 443 A 591 B 56c B 103 B 108 Β 111b Β 133a B 145 B 156–169

B 157a B 183 B 316 B 318 B 319a1, a2 B 333a B 478–479a B 494 B 517a B 518 B 520 B 579 B 581 B 590 B 634 B 649 B 658 B 668 B 669 Β 670 Β 681–685

607, 612 309 n. 31 289, 307, 464f. 289 n. 18, 307–310, 328 203, 519 532 313f. 287, 313–315, 325, 328 288 n. 17, 313f. 576 n. 16 426 491, 532f. 258 317 556 98, 157f. n. 50, 161, 350, 629, 856f. 98, 161, 350, 525, 527 n. 8, 629 317 157f. n. 50, 203, 345f. n. 17, n. 18, 519, 634 n. 34, 807 n. 10, 808 n. 12 162 787 491 757 757, 788 260 203, 522 491 630 503f. 507 527 n. 8 556 556 507 404–406 507 687 687 257, 685, 688 429

B 682 B 707 B 731 B 746 B 874–875 B 860–861 Γ 144 Γ 155b Γ 242 Γ 327 Γ 339 Γ 371b Γ 443 Δ 17 Ε 203 Ε 385 Ε 385 E 397 E 422 E 359 E 560 E 638a E 906 Z 4a Z 4b Z 23 Z 34 Ζ 35 Z 43 Z 130 Z 168 Z 168–9 Z 169 Z 176 Z 178 Z 414 Z 506–508 H 44 H 76 H 130a H 175 H 187 Η 353 H 458 Θ 423a Θ 365–368 Θ 368

555 555 500f., 504 557f. 556f. 557 296 n. 4, 323 n. 2 224 n. 58 317 428 272 858 309 n. 31, 317 272 500 50 n. 2, 491f, 319 n. 54 492 492 505 555 505 766 n. 17 179f., 359 n. 30 138 766 n. 17 506f. 764f. 607 309 n. 31, 317 450, 452 450 450 450 450 763 606 307 260 525 n. 4 450 450 427 555 500 257, 682 682

  Index of Passages Discussed Θ 507 Θ 479 Ι 43–44 I 44 I 77b1 I 219b, d I 222b

I 241 I 395 Ι 416 I 447 I 453c I 464b Ι 607 K 282 K 292 K 299a K 351–354 Κ 397–399a, b Κ 407 Κ 428 K 431a Κ 435 Λ 62a Λ 451b Λ 513c M 34 Μ 77 M 110a1, a2 Μ 158 M 208 Μ 239 M 301a N5 N 20 N 66 N 172 N 217–218 N 217 N 218 N 301 N 302 N 289–291

428 307 427 421 259 858 138, 166 n. 75, 169 n. 84, 171 n. 88, 172 n. 89, 176ff. 537 492 426 307 n. 25, 308f., 329 n. 15 309 n. 32 180f., 359 n. 30, 530f. 606 492 440 n. 2 235 492 98, 161, 350–356, 524–528, 629 428 428 668 257, 682f. 857 238 n. 37 212 826 n. 65, 857f. 608 362f. 427 499–501, 503 78 497f. 763 683f. 309 n. 31, 317 506f. 298 298 298 308 289, 307, 465 262

N 353 N 383 N 384 Ν 390a N 399 Ν 600 Ν 657a Ξ 37b Ξ 221 Ξ 241c Ξ 385 Ξ 479 Ο 86a O 94 Ο 155 Ο 194 Ο 263 O 412b1 O 449–451 Ο 717 O 618b O 623 O 741 Π 106b Π 184b Π 221 Π 234a Π 235 Π 353b Π 411 Π 428b Π 747 Π 810a Π 856b Ρ 24a Ρ 51 P 112a Ρ 125a Ρ 463 P 613 Σ 39–49 Σ 182c Σ 207a, b Σ 222b Σ 487–489 Σ 488b

260 555 555 607 555 608 579 856, 859 763 260 607 605 497 555 428 763 ν. 11 605 261 359 n. 30 539 363 363 493 238 n. 36 607 607 855 408–415, 799f. 261 763 259 402 497 614 497 605, 611 614 497 261 527 n. 8 667 527 n. 8 180, 359 n. 30, 528– 530 507 822 823

Index of Passages Discussed  

Σ 521b1 T 49a Τ 365–368a1 T 391 Y3 Y 53 Υ 67 Y 147–404 Y 145–147 Y 215–216 Y 307 Y 357 Y 403–404 Υ 404c Y 484a Φ 163 Φ 195b Φ 262 Φ 329 Φ 319 Φ 323 Φ 356 Φ 444 Φ 479 Φ 570 X 126 X 379 Χ 441 X 491 Ψ1 Ψ2 Ψ 29 Ψ 79b Ψ 83c Ψ 346 Ψ 361a1 Ψ 599 Ψ 657 Ψ 683b Ψ 679a Ω 17a Ω 25–30 Ω 45 Ω 85a Ω 115 Ω 323

259 497 98, 161, 350f., 524– 528, 629 271 286, 765f. 308, 493 272, 319 n. 54 290–293 283, 287f., 306 292 285, 292, 308 493 285, 292f., 306 607 507 440 n. 1 356 n. 27 605 605 493 493 440 n. 1 283 n. 7 427 427 127 316 494–498, 504f. 605 493 605 499 605 223 n. 51 607 303 n. 19 259 271 557 666 n. 79 815 n. 33 498 667 427 261 557 605

Ω 614–617a Ω 617a Ω 753 Odyssey α 332 β 52 β 134 β 136–137 β 137 β 154 γ4 γ 71 γ 91 γ 230 γ 267 γ 444 δ1 δ 69 δ 113 δ 727 ε 273 ε 337 ζ 244–45 η 311 θ 29 ι 166 κ 60 λ 233 λ 321 λ 321–2 λ 326 λ 326–7 λ 519–20 λ 582 μ 69–70 μ 129 μ 199 ξ 199 ο 16 π 31 ρ 225 ρ 599 σ 17 σ 74 τ 174 υ 383

567, 667 261 n. 21 605 616–620, 773 n. 32, 783, 787 424 416–423 853 416 n. 2 78 333 751 333 502 764, 775 608 421 n. 16, 854 520f. 520f. 527 n. 8 822 496 496 496 78 428 501, 503 273 333f. 334f. 335 336 336 336f. 337f. 788 421 n. 16 860 424 606 821 421 n. 16 763 821 404f. 607

  Index of Passages Discussed φ 218 267 φ 221 267 φ 230 267 ψ 218 812 hypothesis β 430–439 hypothesis Ι β (71, 19–22 Dind.) hypothesis IΙ β (72, 2–6 Dind.) hypothesis γ hypothesis Ι γ (118, 3–8 Dind.) hypothesis IΙ γ (118, 9–14 Dind.)

433f. 433f. 435f. 436f. 436f.

Scholia on Lycophron Alex. 295 541 Scholia on Pindar (Drachmann) Ol. 2.152α 567 Ol. 7.63 257, 686f. Pyth. 3.75 684 Epim. b, p. 306, 16ff. 77–78 Epim. c, p. 307, 7ff. 77 Epim. f, p. 310f. 72–77 Scholia on Sophocles (Christodoulou) Aj. 746 692 n. 7 Scholia on Theocritus (Wendel) 2.59–62 605 n. 19 10.18 717 Scholia on Thucydides (Hude) 3.84 751 Sextus Empiricus Adversus mathematicos I 202ff. 440 III 3 694f., 698f. X 176ff. 441f. X 196–208 442–444 X 202–208 447f. Simonides (PMG) Fragments 652 (iv)

714 n. 86

Sophocles (TrGF 4, Radt) Fragments T 14 699

T 75 T 98 T 100 T 111 T 112 T 114 T 148 T 149 T 150 T 156 T 157 fr. 500 fr. 906 fr. 966a

725 697 725 699 699 699 701 699 699 700 715 n. 95 752 261 817, 852

Sosiphanes (TrGF) Fragments 92 F 6 309 n. 32 Stesichorus (PMG) Fragments 217 831 Stobaeus 3.5.43

772f., 785f.

Strabo 1.2.3 1.2.31 13.624 10.4.15 14.2.20

793 n. 20 84f. 220 406f. 688

Suda α 1061 α 2317 α 3867 α 3924 ε 2269 ζ 74 κ 2342 λ 311 π 3034 π 3035 φ 441

576 n. 17 579 226 n. 66 668 n. 92 821 n. 57 123 n. 9, 145 n. 5 447 n. 10 226 n. 67 440 157 n. 48 704 n. 59

Index of Passages Discussed  

Supplementum Hellenisticum 254–265 818 n. 42, 852 n. 12 255 818 743 817, 852 983–984 816 n. 38, 851 n. 8 Tertullian Apologeticum 18

762 n. 1

Theon (Spengel) 82.19–23

274

Theophrastus (FHS&G) Fragments 600 312 666–709 792 n. 16 Thucydides 1.5.1 2.1–45

751 748

Timon (di Marco) Fragments T I 113 F 12

213 n. 30 213 n. 30

Timotheus (PMG) 783

858

Trypho De tropis (Spengel) 3.197.20–22 274 Tzetzes Prolegomena de comoed., Prooem. I 1–12 123 n. 10, 145 n. 6 II 1–4, 33.22–39 123 n. 10, 145 n. 6

Varro De lingua latina 8.26–43

447 n. 10

Vergilius Georg. 1.387

828

Vitae Homeri (Allen) Vita II 244.13 453 Vita V 247.8 453 Vita VI 252.34–35 646 n. 13 Xenophon Hellenica 3.1.6

216

Zeno (SVF) I 71 I 104, 105 I 167 I 235

657 656, 663f. 655f. 656, 660f.

Zonaeus De figuris (Spengel) 170.14–16 274