Didymus and Graeco-Roman Learning


387 66 16MB

English Pages 174 [180] Year 2020

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Didymus and Graeco-Roman Learning

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

ISSN 0000-0000 (ONLINE) ISSN 0000-0000 (PRINT)

BULLETIN OF THE

INSTITUTE OF CLASSICAL STUDIES

BULLETIN OF THE

INSTITUTE OF CLASSICAL STUDIES

VOLUME 63 NUMBER 2 2020

DIDYMUS AND GRAECO-ROMAN LEARNING Thomas R. P. Coward and Enrico Emanuele Prodi

VOLUME 63 NUMBER 2 2020

BICS_63-2_Cover-new.indd 1

15-06-2021 11:17:07

BICS-63-2 2020

BULLETIN OF THE INSTITUTE OF CLASSICAL STUDIES

DIDYMUS AND GRAECO-ROMAN LEARNING EDITED BY THOMAS R. P. COWARD AND ENRICO EMANUELE PRODI

INSTITUTE OF CLASSICAL STUDIES SCHOOL OF ADVANCED STUDY UNIVERSITY OF LONDON

Director & Editor Professor Greg Woolf, Institute of Classical Studies, London, UK Publications Committee Dr John Creighton, Dr Zena Kamash, Professor Rosalind Thomas, Dr Peter Thonemann, Professor Phiroze Vasunia, Dr Christopher Whitton Subscriptions A subscription to Bulletin of the Institute of Classical Studies comprises 2 issues. Prices include dispatch. Annual Subscription Rate (Volume 63, 2 issues, 2020) Institutional Print and online access  UK: £332.00/$608.00/e389.00 Print and online access  outside UK: £477.00/$608.00/e389.00 Site-wide online access only  UK: £265.00/$485.00/e310.00 Site-wide online access only  outside UK: £382.00/$485.00/e310.00 Please note: UK£ rate applies to UK and Rest of World, except US and Canada (US$) and Europe (Eure). There are other subscription rates available; for a complete listing, please visit https://academic.oup.com/bics Full prepayment in the correct currency is required for all orders. Payment should be in US dollars for orders being delivered to the USA or Canada; Euros for orders being delivered within Europe (excluding the UK); GBP Sterling for orders being delivered elsewhere (i.e., not being delivered to USA, Canada, or Europe). All orders should be accompanied by full payment and sent to your nearest Oxford Journals office. Subscriptions are accepted for complete volumes only. Orders are regarded as firm, and payments are not refundable. Our prices include dispatch. Claims must be notified within four months of dispatch/order date (whichever is later). Subscriptions in the EEC may be subject to European VAT. If registered, please supply details to avoid unnecessary charges. For subscriptions that include online versions, a proportion of the subscription price may be subject to UK VAT. Subscribers in Canada, please add GST to the prices quoted. Personal rate subscriptions are only available if payment is made by personal cheque or credit card, delivery is to a private address, and is for personal use only. The current year and two previous years’ issues are available from Oxford University Press. Previous volumes can be obtained from the Periodicals Service Company, 11 Main Street, Germantown, NY 12526, USA. E-mail: [email protected]. Email: [email protected]. Tel: (518) 537-4700. Fax: (518) 537-5899. Contact information Journals Customer Service Department, Oxford Journals, Great Clarendon Street, Oxford OX2 6DP, UK. E-mail: [email protected]. Tel: þ44 (0)1865 353907. Fax: þ 44 (0)1865 353485. In the Americas, please contact: Journals Customer Service Department, Oxford Journals, 2001 Evans Road, Cary, NC 27513, USA. E-mail: jnlorders@oup. com. Tel: (800) 852-7323 (toll-free in USA/Canada) or (919) 677-0977. Fax: (919) 677-1714. In Japan, please contact: Journals Customer Service Department, Oxford Journals, 4-5-10-8F Shiba, Minato-ku, Tokyo, 108-8386, Japan. E-mail: [email protected]. Tel: (81) 3 5444 5858. Fax: (81) 3 3454 2929. Methods of payment Payment should be made: by cheque (to Oxford Journals, Cashiers Office, Great Clarendon Street, Oxford, OX2 6DP, UK); by bank transfer to Barclays Bank Plc, Oxford Office, Oxford (bank sort code 20-65-18) (UK); overseas only Swift code BARC GB22 (GB£ Sterling Account no. 70299332, IBAN GB89BARC20651870299332; US$ Dollars Account no. 66014600, IBAN GB27BARC20651866014600; EUe EURO Account no. 78923655, IBAN GB16BARC20651878923655); or by credit card (Mastercard, Visa, Switch or American Express). Postal information Bulletin of the Institute of Classical Studies (ISSN: 0076-0730) is published in June and December by Oxford University Press, UK. Bulletin of the Institute of Classical Studies is distributed by Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. Periodicals Postage Paid at Rahway, NJ and at additional entry points. POSTMASTER: Send address changes to Bulletin of the Institute of Classical Studies, c/o Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. Environmental and ethical policies Oxford Journals, a division of Oxford University Press, is committed to working with the global community to bring the highest quality research to the widest possible audience. Oxford Journals will protect the environment by implementing environmentally friendly policies and practices wherever possible. Please see https://academic.oup.com/journals/ pages/about_us/ethical_policies for further information on environmental and ethical policies. Digital object identifiers For information on dois and to resolve them, please visit www.doi.org. Permissions For information on how to request permissions to reproduce articles or information from this journal, please visit https://academic.oup.com/journals/pages/access_purchase/ rights_and_permissions. Advertising Advertising, inserts, and artwork enquiries should be addressed to Advertising and Special Sales, Oxford Journals, Oxford University Press, Great Clarendon Street, Oxford, OX2 6DP, UK. Tel: þ44 (0)1865 354767; Fax: þ44 (0)1865 353774; E-mail: [email protected]. Disclaimer Statements of fact and opinion in the articles in Bulletin of the Institute of Classical Studies are those of the respective authors and contributors and not of Bulletin of the Institute of Classical Studies or Oxford University Press. Neither Oxford University Press nor Bulletin of the Institute of Classical Studies make any representation, express or implied, in respect of the accuracy of the material in this journal and cannot accept any legal responsibility or liability for any errors or omissions that may be made. The reader should make her or his own evaluation as to the appropriateness or otherwise of any experimental technique described. Copyright (c) 2020, Institute of Classical Studies, School of Advanced Study, University of London. Indexing/Abstracting All rights reserved; no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without prior written permission of the publisher or a license permitting restricted copying issued in the UK by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1P 9HE. Instructions for Authors Full instructions for manuscript preparation and submission can be found at: https://academic.oup.com/bics/pages/General_Instructions. Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide.

Style of Rembrandt van Rijn, An Old Scholar at His Desk (detail), pen and brown ink with brown wash on laid paper, overall: 11.3 x 12.3 cm (4 7/16 x 4 13/16 in.), Widener Collection, 1942.9.676. Source photographer: https://www.nga.gov/collection/ art-object-page.1854.html. Reproduced under the Creative Commons CC0 1.0 Universal Public Domain Declaration.

BICS_63-2_Cover-new.indd 2

15-06-2021 11:17:16

TABLE OF CONTENTS Acknowledgements

iv

Enrico Emanuele Prodi and Thomas R. P. Coward

Introduction

1

Lara Pagani

Chapter One: Didymus and epic poetry

8

Enrico Emanuele Prodi

Chapter Two: Didymus and lyric

21

Thomas R. P. Coward

Chapter Three: Didymus on Attic tragedy

34

Federica Benuzzi

Chapter Four: Didymus and comedy

51

Fausto Montana

Chapter Five: Didymus and the Greek historians

62

Scott J. DiGiulio

Chapter Six: The compiler compiled: Didymus in Imperial scholarly and miscellanistic literature

79

Thomas R. P. Coward A checklist of the testimonia and fragments of Didymus and Enrico Emanuele Prodi

95

Abbreviations

121

Bibliography

122

Concordances

133

Index of passages cited

159

General index

173

BICS-63-2 2020

ACKNOWLEDGEMENTS We would like to thank our contributors for accepting our invitation to take part in this volume. We are grateful to Eleanor Dickey, Joseph Howley, Ivan Matijašić, Fausto Montana, Camillo Neri, Paolo Scattolin, Oliver Thomas, and Giuseppe Ucciardello for their comments, which have saved us from many an error. We are deeply indebted to Liz Potter, who supported this project since its inception and steered it powerfully almost all the way to conclusion; her absence from BICS will be much felt. In the closing stages, Sara Charles and Greg Woolf have been invaluable. When this project was conceived, one of us was at Ca' Foscari University, Venice, and the other was on his way there. We wish to thank that institution, and the University of Oxford, for providing us with a supportive environment and much-needed resources, especially in the difficult period of the lockdown. Thomas R. P. Coward and Enrico Emanuele Prodi

Bulletin of the Institute of Classical Studies, 2020, 63, 1–7 doi: 10.1093/bics/qbaa013 Introduction

Introduction Enrico Emanuele Prodi1 and Thomas R. P. Coward2 1

University of Oxford, Oxford, UK Ca' Foscari University, Venice, Italy

2

Didymus Chalcenterus (‘Bronze-Gut’, Χαλκέντερος) is one of the most important figures in the study of Greek literature in antiquity. He stands at the intersection of what for us are the Hellenistic and the Imperial period, a crucial node in the history of classical scholarship. His output was vast and the evidence for it is substantial. The current resurgence of interest towards ancient scholarship has seen a renewed attention in aspects of his work, too: beside several articles, the first two decades of this century have seen a monograph (and most of another) on his Demosthenes, and one on his Pindar.1 Bruce Karl Braswell’s book, first published in 2013, includes a catalogue raisonné of all of Didymus’ known works, programmatically intended as the starting point for a re-evaluation of his achievements as a scholar.2 The present volume pursues that same aim by bringing together the various components of his production and assessing his interests and methods, his place in the development of Greek and Roman learning, and his legacy in later scholarship. After this brief Introduction, the chapters investigate broad areas of his work (Ch. 1–5) and its reception (Ch. 6). We have chosen not to devote a chapter to oratory given the large body of existing scholarship on that part of Didymus’ oeuvre. The volume ends with a Checklist comprising the testimonia for his life and work and all his known fragments, with the aim of replacing the numbering of Moritz Schmidt’s outdated and unwieldy edition of 1854 while adding the several fragments which he did not know or which were revealed by papyri long after his time. Throughout the volume we refer to the Checklist by using bold figures for both fragments (1, 2, etc.) and testimonia (T1, T2, etc.). A Concordance enables the reader to refer to the editions of Schmidt, Pearson–Stephens, and Braswell. What little we know of Didymus’ life comes mostly from one entry in the Suda (δ 872 Adler, T1): Δίδυμος, Διδύμου ταριχοπώλου, γραμματικὸς Ἀριστάρχειος, Ἀλεξανδρεύς, γεγονὼς ἐπὶ Ἀντωνίου καὶ Κικέρωνος καὶ ἕως Αὐγούστου· Χαλκέντερος κληθεὶς διὰ τὴν περὶ τὰ βιβλία ἐπιμονήν· φασὶ γὰρ αὐτὸν συγγεγραφέναι ὑπὲρ τὰ τρισχίλια πεντακόσια βιβλία. Didymus. Son of Didymus, a fishmonger. A grammarian of Aristarchus’ school, from Alexandria. He was active in the time of Antony and Cicero and until that of Augustus. He was called ‘bronze-gut’ on account of his dedication to books: they say that he authored over 3,500 books.

1 Gibson 2002; Harding 2006; Braswell 2017. Other essential recent bibliography includes Luzzatto 2011; Braswell 2011; Montana 2015b: 172–78. 2 Braswell 2017: 40–103, cf. 14. C The Author(s) 2021. Published by Oxford University Press on behalf of the Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected].



1

2



Prodi and Coward

He was from Alexandria and, as far as anybody knows, he spent the rest of his life there.3 His career spanned the time from about Caesar’s death in 44 BC to sometime during the long reign of Augustus (died 4 AD 14). Another entry in the Suda asserts that he was a contemporary and frequent critic of Juba, the scholar-king of Mauretania (ι 399 Adler, T2), who died in AD 23 or 24;5 a third (α 3215 Adler, T3) says that the notorious Apion, whose dazzling career spanned Augustus’ immediate successors, was his θρεπτός (perhaps ‘pupil’ rather than ‘stepson’).6 Another pupil of his in Alexandria was one Heraclides son of Heraclides of Pontus (neither being the fourth-century philosopher, obviously), who went to Rome to denounce Didymus’ critics and remained there under Claudius and Nero (Suda η 463 Adler, T4); he may have been so devoted to his master as to call his own son after him (Suda δ 875 Adler).7 So our man must have lived into the first century AD—a long career to accommodate an enormous output (T1, T5, T6, T8–10, T14). Didymus’ unnamed near-contemporary critics in Rome (T4) are a testament to the swift spread of his works beyond Alexandria; unless he visited Rome in person, there must have been networks which he (or interested readers) could tap into for the dissemination of his scholarship. Already in the mid-Republican period, Livius Andronicus, Ennius, and Naevius drew on Hellenistic scholarship and aesthetics in their works, indicating some access to these scholarly works, whether in excerpted or complete forms.8 Cicero’s translations from Homer, the Greek tragedians, and his partial translation of the Timaeus show a familiarity with the Greek commentary tradition of these respective authors; the same is true of Virgil.9 The late Republican era provides evidence of long-distance dissemination of new works along elite contact networks. For instance, in 60 BC Cicero had sent copies of his Greek commentarii on his consulship to Atticus in Greece and to Posidonius on Rhodes, and Atticus had sent him his own commentarius, which was later published;10 later, in 44 BC, Cicero asked for a copy of Posidonius’ treatise On Duty for research purposes in his own De officiis to be sent to him, and he had also asked Athenodorus Calvus, his friend and fellow student of Posidonius, for a precis of the work.11 These pieces of evidence testify to the almost real-time dissemination that written works, including scholarly ones in Greek, could have across the Roman-dominated Mediterranean. The details of Didymus’ relationship with the two other illustrious Greek literary scholars of his time, Aristonicus and Theon, remain unclear. The Suda entry just quoted (T3) synchronizes Didymus and Theon indirectly by claiming that Apion, θρεπτός of the former, was διάδοχος, ‘successor’, of the latter. Views differ on the relative chronology of the Pindaric commentaries that both scholars are known to have composed; the lack of references to Theon in the doxographies that can be attributed to Didymus makes one suspect that Didymus’ work is earlier, but conclusive evidence is lacking.12 The same is true of Aristonicus, at least as

3 Some earlier scholars suggested that he moved to Rome (e.g. Sandys 1903: 139). While this remains a possibility, there is no evidence for it other than the (probably misattributed) works on Roman themes referenced by Ammianus and Priscian (406, 410–14). 4 Braswell 2017: 28 must be right that γεγονώς refers to Didymus’ activity as a scholar rather than to his birth; ἕως is nonsensical otherwise. 5 A reasonable inference from Str. 17.3.5: see Coward, this volume, p. 46 n. 82. 6 On Apion see now Benaissa 2015: 125–27, 129–30. 7 Schmidt 1854: 3. 8 See Bishop 2019: 27 n. 86 for bibliography. 9 Cicero: Atzert 1907; Soubiran 1972. Virgil: Schlunk 1974 (Aeneid); Farrell 2016 (Eclogues). 10 Cic. Att. 2.1 [21] = Posid. T34 Edelstein–Kidd; Nep. Att. 10.6. See Dix 2013: 220 with n. 48 for the works that Cicero borrowed from Atticus. 11 Cic. Att. 16.11.4 [420] = Posid. F41a Edelstein–Kidd with Edelstein 1989–99: II.i 185–89. Dix 2013: 220 n. 50 suggests that Cicero was asking for the copy from the library of the recently deceased Cato, of whom Athenodorus was a member of the household. On Athenodorus Calvus see PA I A497; on him and Posidonius see Cic. Fam. 3.7.5, cf. also Posid. T44, F7, 41ab Edelstein–Kidd and Cic. Att. 16.14.4. Strabo himself may have met him in Rome (c. 44 BC, the same year as Cic. Att. 16.11.4 [420] was written), which led to Strabo turning to Stoic philosophy; see BNJ 746 T4 = F5 with Dueck 2000: 10–13 and 65–69. Interestingly, Strabo complains that book dealers in both Rome and Alexandria employed bad scribes and did not check the newly produced copies against the reliable exemplars (13.1.54); cf. also Cic. QF 3.4.5, 3.5.6 with White 2009: 133, 135 and Gal. Ind. 13, Libr. Propr. praef. 1–2. 12 So Deas 1931: 34–35, 38–39, who even posits that Didymus’ commentary was the basis of Theon’s. For a different view see McNamee 2007: 95–96; Meliadò 2015.

Introduction



3

far as Pindar is concerned.13 With regard to Homer, scholarly opinion has long maintained that Didymus made use of Aristonicus’ work, despite (uncharacteristically) never citing him. In her contribution to this volume, however, Lara Pagani argues for greater caution: while it seems certain that Aristonicus was not aware of Didymus’ Homeric writings, it may be that neither of the two scholars—working one in Rome, the other in Alexandria—made use of the other’s work. Demarcating what he did and did not write is sometimes arduous. Δίδυμος was not an uncommon name: the Suda alone lists four grammarians, a philosopher, and an author of a Georgics in four books (δ 871–76 Adler); there were others, too. Confusion, accordingly, is easy. Ammianus (22.16.16, T7) ascribes to Chalcenterus […] Didymus six books of churlish and often inaccurate criticisms of Cicero, probably the same work to which Suetonius composed a response (Suda τ 895 Adler, Περὶ τῆς Κικέρωνος πολιτείας α´, ἀντιλέγει δὲ τῶι Διδύμωι). These are unlikely to be by our man, who does not seem to have busied himself with anything not written in Greek. Perhaps Ammianus is conflating the Chalcenterus with Didymus Claudius, who wrote on Roman matters (Suda δ 874 Adler),14 or perhaps with the obscure ‘Didymus the Younger’, an Alexandrian grammarian who went on to practise in Rome (Suda δ 873 Adler), or with another, unknown character. Conversely, the works ascribed by the Suda to ‘Didymus the Younger’ have sometimes been reassigned to his better-known namesake, though on no good evidence.15 He certainly wrote a great deal on a great deal of subjects.16 He compiled a Λέξις κωμική in fifty books and a Λέξις τραγική in at least twenty-eight, which were among the sources used by Hesychius for his own Lexicon; the former was also epitomated by Galen, though the epitome is also lost.17 He wrote monographs on semantic differences, incorrect usage, figurative language, ‘puzzling words’ (Περὶ ἀπορουμένης λέξεως), and morphological change (Περὶ παθῶν); Herodian later furnished this last work with a hypomnēma. Didymus was also responsible for an important work On Proverbs in thirteen books which stands at the basis of the later paroemiographical tradition, and a treatise On Lyric Poets which dealt extensively with the division of lyric genres. As a scholar of ancient Attica he authored a monograph On Solon’s Axones in response to one Asclepiades, and probably a Register of Attic Demes. He is also ascribed miscellaneous works entitled Ξένη ἱστορία, Σύμμικτα, and Συμποσιακά, the last two of which may have been one and the same. But it was as a commentator that he was most renowned. He worked on archaic and classical poets, illustrious and obscure: Homer, both Iliad and Odyssey, plus a treatise on Aristarchus’ diorthōsis of those two poems; Hesiod, at least Theogony and Works and Days; Pindar, at least Epinicians and Paeans, perhaps the Hymns, but more probably the entire corpus; Bacchylides, at least the Epinicians, quite possibly the Dithyrambs too; Alcaeus; Anacreon; perhaps Sappho, if the discussion cited by Seneca (T14) comes from a commentary on her; Sophocles; Euripides; Ion of Chios (Athenaeus attests a commentary on the Agamemnon and a further book of ἀντεξηγήσεις to that poet, whatever they were); seemingly even the more obscure Achaeus’ Games; Aristophanes; Phrynichus’ Kronos; Menander (the existence of commentaries to Cratinus and Eupolis is hypothetical). Compared to many of his predecessors he is remarkable for the attention he devoted to prose writers, especially the orators: he wrote commentaries on Demosthenes (this is probably the same as the Περὶ Δημοσθένους attested by P.Berol. inv. 9780, °281),18 Aeschines, Hyperides, Isaeus, and possibly Isocrates, Dinarchus, Lysias, and Lycurgus, all of which were among the chief sources for Harpocration’s Lexicon of the Ten Orators. He seems to have written on Herodotus and Thucydides, too.

13 14 15 16 17 18

Deas 1931: 29–30 argues that none of the (very few) citations of Aristonicus in the Pindar scholia is mediated by Didymus. So Cohn 1903: 471–72; Montana 2015b: 177. e.g. Schmidt 1854: 335 (Περὶ ὀρθογραφίας). For a critical catalogue of Didymus’ writings see Cohn 1903; Braswell 2017: 40–103. See Coker 2019. This is not the place to relitigate the long-standing questions of whether the Περὶ Δημοσθένους is really a commentary and whether P. Berol. inv. 9780 recto represents excerpts from Didymus’ work rather than the original; for what it is worth, our answers are ‘yes’ and ‘no’ respectively. See West 1970: 289–91; Arrighetti 1987: 195–204; Harding 2006: 13–20; and (conclusively) Luzzatto 2011.

4



Prodi and Coward

The figures we are given for the size of Didymus’ oeuvre are quite extraordinary: 3500 books according to Athenaeus and the Suda (Τ5, Τ1, Τ10), 4000 according to Seneca (T12). They need not be an accurate estimate, but they do indicate a reputation as a vastly prolific scholar. According to Quintilian, nobody wrote more than him (T6); comparing him to that pinnacle of Roman scholarship, Varro, Jerome remarks that he wrote more books than an ordinary person can even copy out (tantos libros composuerit quantos quiuis nostrum alienos sua manu describere non potest, T8, cf. T9). He had both admirers and detractors already in antiquity, as Scott DiGiulio’s chapter shows. Seneca chose him as the byword for pointless erudition in a tirade against the meticulous study of literature (Ep. 88.37–40, Τ14).19 He acquired the nickname ‘Bronze-Gut’ (Τ1, Τ2, Τ7–10, Τ12, Τ13) from the vast number of books which he could metaphorically digest, or perhaps (so the late Τ12) from the lengths of time he could spend immersed in his studies without even rising to eat. His other nickname, ‘Book-Forgetter’ (Βιβλιολάθας), which Athenaeus found in a poem by Demetrius of Troezen (SH 376, Τ5), is more openly satirical with its implication that Didymus’ output was so large that he used to forget what he had written. But Jerome reports that Graeci Chalcenterum miris efferunt laudibus (Epist. 33.1, T8), the Suda calls him ‘Didymus the great’ (α 3215 Adler, Τ3), and still in twelfth-century Constantinople he could be used alongside Demosthenes as a paragon of dedication to learning (Τ12). On the Roman side, Macrobius spared no superlatives to describe him: omnium grammaticorum facile eruditissimus (Sat. 5.18.9, °39), grammaticorum omnium quique sunt quique fuerunt instructissimus (5.22.10, 336). It is his extraordinary learning that Macrobius emphasizes, and Didymus’ reputation for cramming in and churning out was destined to haunt him for millennia. Up until the 1960s, especially after the publication of P.Berol. inv. 9780 recto (On Demosthenes, °281) in 1904, historians of classical scholarship regarded him as an unoriginal if ultimately competent compiler; Rudolf Pfeiffer condescended that ‘he was not totally devoid of critical judgement’ while admitting that in spite of his ‘amazing feats of learning’—‘or perhaps because of them—his reputation has never been very high’.20 Then half a century ago Stephanie West took an axe to that meagre consensus and used the Berlin papyrus to denounce Didymus’ work as ‘slapdash and ill-digested’, at times ‘extraordinarily confused and incoherent’, consisting of ‘hurried compilation rather than intelligent re-interpretation’.21 Not just a mere compiler, then, but a dim one too. More recent scholarship has been somewhat more nuanced, but damning judgements still abound.22 Not that this is altogether unfair: Didymus’ writings do include things that are irrelevant, wrong, or downright silly. The aim of this collection, however, is not to damn him on the standards of (idealized) twenty-first-century academia, but rather to examine his scholarship on its own terms and, insofar as possible, in the context of ancient scholarly practices.23 Building on earlier sources was indeed a staple of Didymus’ method. The most extensive continuous fragment of his work—the Berlin papyrus of the Περὶ Δημοσθένους—consists for almost one half of verbatim quotations. Perhaps even more than Aristonicus’ Περὶ σημείων Ἰλιάδος καὶ Ὀδυσσείας, Didymus’ monograph Περὶ τῆς Ἀριστάρχου διορθώσεως is as close as antiquity gets to a history of Homeric textual criticism, with pre-Didymean material far outnumbering (and, some might say, outweighing) Didymus’ original

19 On the interpretation of this passage see the crucial remarks by Luzzatto 2011: 8: Seneca derides Didymus not as a bad scholar, but as an excellent one; literary scholarship is intrinsically futile, and the better you are at it, the more worthy you are of the philosopher’s condemnation. 20 Pfeiffer 1968: 279, 274. Note how he dispatches Didymus in six pages, compared to Aristophanes of Byzantium’s thirty-eight (171–209) and Aristarchus’ twenty-four (210–33). An exceptional admirer is Turner 1968: 120, in whose opinion ‘Didymus’ work shines like gold’ in comparison to scholarship on Thucydides found in the papyri. 21 West 1970: 288, 292, 296. 22 See e.g. Braswell 2017: 124–26 (‘Didymos’ performance as an interpreter of poetic texts is dismal’, p. 124). Harding 2006: 2 is more sympathetic without being exculpatory. 23 Cf. Harding 2006: 32; Braswell 2017: 125.

Introduction



5

contributions.24 In the Pindar scholia we find time and again Didymus appending his own interpretation of a passage to a discussion of earlier scholars’ attempts. Quintilian’s much-quoted anecdote about him (Inst. 1.8.20, T6) is also relevant. In prescribing that the elucidation of historical references must be undertaken with judicious moderation, rather than by hoarding stories from sources of all sorts, Quintilian tells of how Didymus once argued against a certain historia only to have it cited back at him from one of his own books. Despite the consonance with the nickname Βιβλιολάθας (T5), the point being made is not primarily about Didymus’ absent-mindedness, but about his unthinking accumulation of material for accumulation’s sake: his books (Quintilian intimates) were stuffed with tralaticious information which he himself barely knew and, it turns out, did not believe to be true. There are two sides to this compilatory practice. One we may term doxography. Didymus’ approach is often to list everyone else’s answer to a given problem before providing his own (or expressing agreement with one of the others). Beside his work on the Aristarchan recension of the Iliad and the Odyssey, which put earlier scholarship front and centre, the doxographic tendency is in evidence more than once in his commentaries to Pindar (140, 158)—so much so that Henry Deas called him ‘the first great variorum editor’.25 The first of the quaestiones he is notoriously ascribed by Seneca, on ‘Homer’s fatherland’ (Ep. 88.37, 394), is just the sort of topic which must have warranted an extensive literature review.26 This approach was to prove popular: in Jerome’s later telling (Adv. Rufin. 1.16), the purpose of a commentary is to provide the reader with the opinions of other interpreters and the evidence they rest on, so that he can make up his own mind which of them is truer.27 But Didymus’ doxographic inclinations should not be pressed too hard: the assumption that his work on Demosthenes, for instance, must have been based on earlier commentaries of which not a trace survives is a mere supposition, founded on the prejudice that Didymus cannot possibly have thought anything much for himself.28 The other side—explored in depth in Fausto Montana’s chapter—is Didymus’ systematic use of sources to elucidate Realien: history, geography, society, mythology, local customs, antiquarian details. In the early nineteenth-century querelle between Wortphilologie and Sachphilologie—philology of text and philology of context, to paraphrase freely—he would have firmly advocated for the importance of the latter alongside the former.29 His wide reading and industrious note-taking came in very handy when discussing authors like Aristophanes or Demosthenes, whose works are replete with specific references which often defy understanding if approached on the basis of the text alone. The surviving columns of the Demosthenes commentary (°281) contain excerpts from Philochorus, Aristotle, Theopompus, Callisthenes, Bryon, Hermippus, Dinarchus, Philemon, Timocles, Homer, Aristophanes, Timosthenes, Demon, Sophocles, Aeschylus, Callimachus, Androtion, and the letters attributed to Philip of Macedon; add those authors whom he merely references without a verbatim quotation and a few opinions attributed to ἔνιοι, ‘some’. His scholarship on poetry made ample use of historical material, both for the sake of positive interpretation (e.g. °39, 105) and also negatively, when an earlier commentator had invented a factual-sounding explanation that did not actually correspond to attested fact (e.g. 84, 111). Not that knowing lots of facts means always using them sensibly. Critics have seized upon several such inanities in the Demosthenes commentary,30 and in at least one case Didymus’ discomfort with an otherwise unattested family name led him to amend a perfectly good Pindaric 24 See Schironi 2018: 23–25; Pagani, this volume, 19–20. 25 Deas 1931: 19, cf. 22. See also Prodi, this volume, 24–26. 26 Cf. [Plu.] De Hom. I 1–3 (citing Ephor. BNJ 70 F1 and Arist. fr. 76 Rose = On Poets fr. 65a–e Janko), II 2 (citing Pi. fr. 264 Snell– Maehler, Simon. fr. 19.1 West, Antim. fr. 130 Wyss, Nic. fr. 14 Schneider, beside the two above); multiple claims are also referenced e.g. in Cert. Hom. Hes. 2, Procl. Vit. Hom. 2 West, APl 294, Gell. 3.11.6. 27 On the passage see La Bua 2019: 168–71. 28 Refuted by Harding 2006: 31–39. For an unprejudiced suggestion of earlier precedents—arising from a very different perspective—see now Montana 2015b: 95. 29 Cf. Luzzatto 2011: 7. 30 West 1970, to name but the most conspicuous example.

6



Prodi and Coward

passage (157). As Federica Benuzzi illustrates in her chapter, Didymus’ anxiety to use his extensive knowledge of Old Comedy to trace kōmōidoumenoi across different plays sometimes blinds him to the obvious implications of the individual context (256, 259). Excerption and compilation is a well-known practice of scholarly composition in the period of the late Republic and early Empire.31 Cicero admits to a version of it (Inv. 2.4), and we have a detailed description of Pliny the Elder’s methodical excerpting in one of his nephew’s letters (Ep. 3.5). Pliny too has been disparaged as a mere compiler, yet his Natural History does more than summarize contemporary knowledge or points of view; it aims to make a vast range of information, ideas, and techniques accessible, retrievable, and usable by readers, bringing together Greek and Roman ideas, along with those of the Babylonians and Egyptians.32 At the beginning of Book 1, Pliny lists the contents of the other thirty-six books and the sources he has consulted, in which he claims (Praef. 17) to have included more than 20,000 facts culled from over 2000 works. Pliny’s role was to bring together, order, and make available what had already been collectively accomplished in several disciplines, without refraining from comments and additions of his own. Closer to home, large parts of Philodemus’ Index academicorum consist of excerpts from earlier sources (mostly named, but there are some ἔνιοι). Its main witness, P.Herc. 1021, is in fact a first draft of the work with extensive additions and corrections.33 Like the 160 volumes of electorum commentarios […] opistographos which the younger Pliny reports having inherited from his uncle, the papyrus is written on both sides, though not from its inception: Philodemus kept having more materials added on the back of the roll as his research progressed after the writing of the front. Aside from its status as a working document, it is the closest known papyrological equivalent to P.Berol. inv. 9780 recto (°281),34 just as Philodemus is arguably the closest figure to Didymus among his near contemporaries about which we know a sufficient amount. Philodemus’ On Poems, On Music (Bk. 4), and On Rhetoric are also forms of ὑπόμνημα. His critical methodology, following his tutor Zeno of Sidon, is usually in two parts. First he outlines and compiles his adversary’s views and arguments with quotations or paraphrases; then he rebuts his opponent’s views, sometimes offering his own opposing theory.35 Sometimes there is a third section which contains a summary of the opponent’s claims and his own criticisms, followed by a pithy conclusion. In other words, the same structure as we find in the Pindaric doxographies (e.g. 140). In short, Didymus’ practice was far from unique if we look at the broader picture. After all, one of the most common meanings of the word ὑπόμνημα—and of its Latin equivalent commentarius—is a set of notes.36 Relatedly, the adjective ὑπομνηματικός was sometimes used to denote an unpolished first draft that was not meant for wide circulation.37 A commentary proper—a set of materials provided as an aid for understanding a work of literature—intrinsically has an aspect of that, too; papyrologists have a point when they sometimes label commentaries as ‘paraliterary’. That a commentary should consist of excerpts collected from various places is thus in the nature of things. We may imagine Didymus working in a similar way to Pliny, reading extensively and excerpting material to be reused in his publications. In fact, this process does leave a trace when it goes wrong. At one point in the Περὶ Δημοσθένους Didymus says that he omits an extract from Anaximenes’ Histories, ‘because it is useless’; the explicit reference to τὴν ἐκλογ(ήν), ‘the extract’, is telling, as is his portrayal of his own action as an omission (παρίημι).38 Shortly afterward he introduced an extract to 31 See the discussion in Dorandi 2007: 29–46, from which the following examples are drawn. 32 Taub 2017: 72–85; see Murphy 2004 and Doody 2010 for further analysis. 33 See Cavallo 1984: 13–17; Dorandi 2007: 40–42; and Essler 2019. Text: Dorandi 1991 (pp. 109–13 on the textual transmission). Del Mastro 2012 recognized another fragment of the same roll in P.Herc. 1691 pz. 2. 34 The comparison is already in Harding 2006: 39. 35 See Delattre 2007: I xxvi–xxx. Cf. e.g. Sextus Empiricus (M. 7.1–262), who summarizes the views of philosophers by time and school, and then refutes them in the same order (7.263–68). 36 On the meanings and development of commentarius and ὑπόμνημα, see Bömer 1953. 37 Dorandi 2007: 65–81. 38 P.Berol. inv. 9780 recto col. vi.62.

Introduction



7

prove a point and then forgot to copy it in (or failed to find it in his notes?), and the omission remained uncorrected—a case of ‘I’m sure I’ve read this somewhere, I’ll look it up later’ gone awry which some of us no doubt will sympathize with.39 It has been common in modern scholarship to regard Didymus as a turning point between two different ways of studying literature: the end of ‘philology’, perhaps, and the beginning of ‘erudition’. There may be a grain of truth in this. Didymus is nothing if not erudite, and his relative lack of interest in text-critical matters —and his underwhelming record when he did deal with them—sets him apart from the great editors of previous generations. With the benefit of hindsight, there is a sense of ‘coming after’ in his meticulous compilations of earlier scholars’ opinions, but his inscribing of himself into those very compilations speaks against a conscious perception of a break. Ancient readers, at any rate, do not seem to have felt one. The Suda calls him a γραμματικὸς Ἀριστάρχειος (T1); Seneca pairs him with Aristarchus and Apion (first century AD) in his broadside against studia liberalia (T14); Ammianus has him as the most eminent of a series of literary scholars comprising Aristarchus, Herodian (second century AD), and Ammonius Saccas (third century AD) (T7).40 His works combine continuity and innovation; the broader record, especially for what concerns ‘minor’ grammarians, is too patchy to assess the exact proportion of the two in comparison to others. Scholars had been debating each other’s opinions since the study of literature first began, but Didymus is the first known practitioner of systematic, or almost systematic, doxography in this realm. The use of historical and antiquarian sources as an aid for interpretation certainly did not begin with him, but one would struggle to identify an earlier commentator who practised it so thoroughly and widely. The cross-fertilization of specialized lexicography and running commentary to individual works is attested before the time of Didymus and Theon, but it seems to have been mostly restricted to technical genres.41 The mention of Theon—the great commentator of the Hellenistic poets—highlights by contrast one aspect in which Didymus was decidedly on the conservative side: namely, an apparent lack of interest in elucidating post-classical literature. The latest author that we know he commented is the last of the classics, Menander. But this appearance is deceptive: he had clearly read Callimachus and Lycophron (to name but two) with great attention, and quoted them repeatedly in his exegesis of older authors. With his gaze firmly fixed on the existing bibliography and (apparently) not much time to spare even for the scholarship of his contemporaries, Didymus may be thought an unlikely innovator, yet on the whole his profile is quite unlike any earlier scholar that we know of. Martin West spoke of his own career as going ‘forward into the past’;42 Didymus may rather give the impression of going backward into the future. Interested in prose as well as poetry and well aware that any nook and cranny of the ancient record could be called upon to elucidate another, he was a devoted practitioner of what in a later age would be labelled Altertumswissenschaft, perhaps more so than any other ancient scholar of Greek literature. As Fausto Montana remarks in a recent reassessment, he is the culmination of Alexandria’s tradition of literary scholarship.43 We hope that the present volume will inspire further work on this very interesting predecessor of our discipline.

39 Col. viii.54. After that half–sentence (the name of the author is also missing) the scribe left a blank, as his exemplar must have done, too. It is certainly an authorial mistake, see West 1970: 293. 40 On Didymus’ dependence on Aristarchus, see e.g. the contributions of Pagani and Prodi in this volume, 13–20, 28, 30. 41 e.g. Bacchius of Tanagra, see Montana 2015b: 117 and Perilli 2017: 103–4 with refs. Aristocles of Rhodes, a contemporary of Varro and Strabo and slightly earlier than Didymus, may have written a lexicographical work on Hippocrates, an On Poetry, an On Dialects, and some sort of work on the dialogues of Plato, see Corradi 2015. 42 West 2007. 43 Montana 2015b: 172.

Bulletin of the Institute of Classical Studies, 2020, 63, 8–20 doi: 10.1093/bics/qbaa014 Original article

Chapter One Didymus and epic poetry* Lara Pagani University of Genoa, Genoa, Italy

ABSTRACT This chapter studies Didymus’ scholarly activity on epic poetry, especially on Homer, in relation to both his own exegetical efforts and his work about the textual recension of Aristarchus of Samothrace. It provides a critical survey of the fragments of Didymus’ Homeric commentaries, with an overview on the subjects they covered. As for the treatise on Aristarchus’ diorthosis, which is much more significant for the exegetical tradition on Homer, it discusses the scholia that convey his reasonings, the relationship of this work with the work of Aristonicus, the sources used by Didymus, the information Didymus gives us about Aristarchus’ works and ecdotic method, and finally the methods employed by Didymus himself in handling the work of his predecessors.

The importance of Didymus’ role in the scholarly study of epic poetry1 lies especially in his work about the textual recension (diorthōsis) of the Homeric poems by Aristarchus of Samothrace dating to more than a century before him. His own commentaries on the Iliad and the Odyssey are known to a much lesser extent and had a more limited importance for the Homeric exegetical tradition. He has nothing to do with the group of scholia that was called the scholia Didymi by Franciscus Asulanus at the beginning of the sixteenth century, in a completely unfounded attribution, and which is now conventionally identified as D scholia for the Iliad and V scholia for the Odyssey.2 As for Hesiod, we can hypothesize that Didymus wrote a commentary on the Theogony, since in the scholia he is mentioned, along with Crates and Aristophanes of Byzantium, in connection with the hermeneutic problem of the generation of Ouranos from Gaia (Σ Hes. Theog. 126 Di Gregorio, 98). On the other hand, his presence in the scholia to Works and Days (304b Pertusi, 99) concerning an improbable etymology of the adjective κόθουρος (‘without a sting’) is more likely to come from a lexicographic work than from a commentary.3 The attempts made in the nineteenth century to treat Didymus as the primary source of the scholiastic corpora to

* I wish to thank Fausto Montana for helpful remarks about my paper, as well as René Nünlist and Francesca Schironi for sharing with me their forthcoming articles. Numbers in bold refer to testimonia and fragments in the Checklist (pp. 95–120). 1 For up-to-date general overviews (with bibliography), see Braswell 2017: 46–51; Montana 2015b: 173. The scholia to the Iliad can be consulted in Erbse 1969–88 and van Thiel 2014b (D scholia); the scholia to the Odyssey are to be read in Pontani 2007–20 (up to Book 8), and in Dindorf 1855 and Ernst 2006 (D scholia) for the rest. 2 Pontani 2005: 503. An account of this class in relation to both poems is in Pontani 2005: 145–48. 3 E.g. the Λέξις κωμική, with reference to κηφήν in Ar. V. 1114, as suggested by West 1978: 66. C The Author(s) 2021. Published by Oxford University Press on behalf of the Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected].



8

Didymus and epic poetry



9

Hesiod and to ascribe to him much more material within the scholia4 were refuted for want of evidence.5 On the basis of what is preserved, he cannot be credited with an important role in that context.

1. THE COMMENTARIES ON THE ILIAD AND THE ODYSSEY Didymus’ own exegetical work on the Homeric poems has been overshadowed by his image as a collector and vehicle of the previous ecdotic tradition of these poems. We have a relatively limited knowledge of his commentaries both on the Iliad and on the Odyssey6 and no information about their interactions with the treatise on Aristarchus’ diorthōsis nor any hints from which we might hypothesize a relative chronology among these works. Didymus’ hypomnēmata to Homer are explicitly mentioned in a handful of passages; some further fragments can be assigned to them with varying degrees of certainty, given the wide scope of Didymus’ activity. The following account is based on Schmidt’s edition, with a discussion of the most problematic cases. The hypomnēma to the Iliad is quoted twice in Ammonius’ De adfinium vocabulorum differentia (300 and 513 Nickau, °72 and °71), in both cases drawing material from the second book. The first fragment discusses the distinction between λῃτουργεῖν/λειτουργεῖν, ‘to serve the state’, and λιτουργεῖν allegedly meaning ‘to say bad things’, which was possibly connected with the mention of the name Leitos at Il. 2.494.7 The second fragment deals with the difference in the use and shape of two kinds of mantles, which is paralleled anonymously and much more concisely in Σ ex. Il. 2.183b Erbse.8 Other possible fragments, concerning a variety of subjects, are:

1. mythography: Σ D Il. 1.399 van Thiel (69b), about the conspiracy of the gods against Zeus; Σ Did. vel ex. Il. 19.116a1 (83), on the identity of Sthenelus’ wife; 2. Homeric geography or ethnography: St. Byz. α 8 Billerbeck (77), where the Ἄβιοι are identified as a Thracian people (presumably connected with Il. 13.6, but the scholia provide no parallel); α 361 Billerbeck (72; cf. Σ ex. Il. 2.519b), where Ἀπολλωνία is said to be in Phocis and is qualified with the epithet Κυπάρισσος; σ 122 Billerbeck–Neumann-Hartmann (73a), where Σήσαμον is identified with an island (probably pertaining to Il. 2.853, but nothing similar is preserved in the scholia ad loc.); 3. etymologies: Σ D Il. 4.475 (75; cf. Eust. ad Il. 500.43–44 van der Valk), where Didymus is said to have claimed that every mountain can be called Ἴδη because one can ‘see most things from there’; St. Byz. α 24 Billerbeck (78), where he maintains that the toponym Καβασσός originates from Καβακησός: it is unclear what is implied by Stephanus’ claim that Didymus wrote a whole book on this topic, i.e. whether it means that this discussion took up no less than one roll of the commentary,9 or it refers to an individual work, in the form of a treatise. If the latter were true, we cannot, however, rule out that Didymus both wrote a note about the origin of the name in the Iliadic hypomnēma (a possible point of reference is Il. 13.363, though there is no parallel in the scholia) and made a more detailed survey about the city in a different and specific study; 4. explanations of words or content: Σ ex. Il. 10.428b Erbse (76) clarifies that ἀγκυλοτόξοι is said of people who use both the bow and the javelin; P.Oxy. II 221, col. 17.27–28 (ad Il. 21.363, 85), on μέλδειν, explained as properly meaning ‘to eat the limbs’ (μέρη ἔδειν); Σ D Il. 22.126 van Thiel (86), about the 4 5 6 7

Flach 1876: 112–18; Dimitrijević 1899: 142–56. West 1978: 66. Well-founded interpretations also in Montanari 2009a: 339–40; Braswell 2017: 50–51. A concise but clear survey in Cohn 1903: 449.41–67. Cf. more recently Braswell 2017: 47–50. Erbse 1969–88: I 290, in the apparatus of parallels to Σ D Il. 2.494. A different etymology is preserved in the D scholion edited by van Thiel 2014b as 2.494b. 8 Schmidt 1854: 180. 9 Cf. Cohn 1903: 449.54–59.

10



Pagani

purported background and usage of the expression ‘meet under an oak or a rock, like a young man with a girl’; Σ ex. Il. 24.640 Erbse (89), on the semantic equivalence of χόρτος with τειχίον (‘wall’); 5. textual-exegetical problems: Σ Hrd. Il. 19.90c.d Erbse (82), where Didymus defends, against Ptolemy of Ascalon, the presence of the preposition διά in the expression θεὸς διὰ πάντα τελευτᾷ, ‘the god fulfils everything’. On the other hand, the glossographic explanation of ἄφλαστα preserved in the Synagoge (α 2550 Cunningham, unde Phot. α 3369 Theodoridis: 80b), which one could relate to Il. 15.717, belongs more probably to Didymus’ Λέξις κωμική, as Schmidt realized.10 Σ Did. Il. 19.81 Erbse (81), which reports the Aristarchean interpretation of the line, seems a better fit for the treatise on Aristarchus’ diorthōsis (see below, §2). Finally, a reassessment of the text has removed the reference to a λύσις by Didymus in Porph. ad Il. 3.379–80 MacPhail (ex fr. 6 Schmidt).11 It is difficult to establish from which work of Didymus the Aristotelian report about eels (P.Oxy. II 221, col. x.11–14, ad Il. 21.203, 84) derives. Four passages explicitly attest the hypomnēma to the Odyssey: Et. Gen. (AB) s.v. Σκύρος (°91b), which quotes Book 14 (probably to be emended to 11, since the likely referent seems to be Od. 11.509)12 on the explanation of the name of the island Σκῦρος as ‘gypsum’, due to its clay-rich soil; St. Byz. α 410 Billerbeck (°92), where it is reported that Didymus maintained the existence of eight springs named Arethusa (Od. 13.408: cf. Σ ad loc.); Et. Gen. (B) s.v. περισκέπτῳ (°93a), citing Book 14 on the meaning of the word (the reference is probably to 14.6); Et. Gud. p. 138.3–9 de Stefani s.v. ἀνέγναμψαν (°94), again concerned with semantics, pertaining to Od. 14.348 (cf. Σ ad loc.). Further possible fragments are Et. Gen. (AB) α 1133 Lasserre-Livadaras s.v. Ἀργινοῦσαι (cf. °91a), which belongs together with the explanation of the name of Scyros, and Σ ex. Od. 4.797c Pontani (90a), which explains the name of Πηνελόπη by relating the story of her rescue by ducks (πηνέλοπες) when she had been thrown into the sea by Nauplius. It is difficult to ascribe to this work or another the etymology of ἀρετή preserved by Et. Gud. p. 190.25–26 de Stefani (95b), and the discussion of the pronunciation, aspirated or not, of ἀλύω in Ael. Dion. α 81 Erbse and in the lexicographical tradition (96b). Much more is probably concealed in remarks reported anonymously in the scholia. From these meagre remains it is difficult to outline an overall picture of the kind of issues addressed by Didymus and his approach when personally commenting on the poems. Nevertheless, a good variety of themes emerges, such as etymological and semantic explanations, mythological accounts, geographic and ethnographic topics, and textual and exegetical discussions. 2. ON THE DIORTHŌSIS OF ARISTARCHUS Together with Aristonicus, Didymus represents our best source of information about the philological work carried out by Aristarchus on the Homeric epics: it is telling that the Suda (δ 872 Adler, T1) calls him a γραμματικὸς Ἀριστάρχειος, even though the chronological distance rules out any direct connection between him and that renowned scholar. Didymus’ treatise aimed to outline the textual choices made by Aristarchus in his recension of Homer’s poems, whereas Aristonicus’ work contained explanations of the critical signs attached by Aristarchus to the Homeric texts. Both of them were excerpted and then incorporated into a composite and now lost commentary together with selected annotations from the works of two further authors, Aelius Herodian and Nicanor, respectively on prosody and punctuation. The existence of such a compilation, conventionally called Viermännerkommentar (VMK), is inferred from the subscriptions of the

10 Schmidt 1854: 79 (fr. 48). 11 Schmidt 1854: 180. The text really present in the MSS is much more sensible: ἀδύνατόν (not Διδύμου) φασιν εἶναι κατὰ τὸ ἐγχείρημα. Cf. Schrader 1880–82: 1, 64, and recently MacPhail 2011: 80. 12 Cf. Braswell 2017: 48 and n. 85.

Didymus and epic poetry



11

tenth-century Venetus A of the Iliad,13 which mention the sources of the exegetical material contained in the margins of this codex.14 In VMK, which is generally conceived as still having the form of a continuous commentary and was presumably compiled during Late Antiquity,15 Didymus’ material had already been mingled with that of others of different origins. To make matters worse, we are not able to read it as it was at that stage, but we have to resort to later compilations of scholia into which it was incorporated: during this process it underwent further selections and abridgements, on the one hand, and was conflated with additional materials on the other hand. As a result, the marginal apparatus of the Venetus A, although it is the witness par excellence of VMK scholia, also contains comments connected with the traditional lines of the exegetica and the D classes. Moreover, scholia stemming from VMK can also be found scattered in other manuscripts that primarily bear non-VMK scholia. Furthermore, as these remarks are often reported anonymously and each one of them may appear heavily mixed with other ones, it will become clear that the attribution of any given portion to one or another group of scholia, and hence to its possible source, is the result of complex inferences based on content, wording, and positioning in the manuscript tradition, and so is not always unambiguous. Although for the Odyssey we cannot rely on a witness like the Venetus A, it is possible, though far from certain, that a compilation similar to the VMK did exist also for the Odyssey:16 we know that the four works excerpted in order to compile the Iliadic VMK were also concerned with, or had a counterpart on, the Odyssey.17 As for Didymus, the key witness is Σ Nic. Od. 2.260d1 Pontani (68), where a variant reading of Aristarchus is reported with the words ‘as Didymus says that Aristarchus wrote’ (ὡς Δίδυμός φησι γράφειν Ἀρίσταρχον).18 It is rare that the scholia to either the Iliad or the Odyssey explicitly mention Didymus as their ultimate source, and only a few of them give the title of his treatise, in an abbreviated form, such as e.g. ἐν τοῖς διορθωτικοῖς (Σ Il. 17.607c1 Erbse, °63) or ἐν τῇ διορθώσει (Σ Il. 21.110a Erbse, °65a).19 Except in these cases, Didymus’ authorship of many other scholia is inferred as described above,20 with all the inherent uncertainties. Furthermore, the intrinsic nature of a scholiastic compilation implies that what we now read as a unitary scholion may be a conglomeration of different sources, and the identification of the boundaries between them can be a matter of discussion.

2.1. Didymus and Aristonicus Didymus and Aristonicus were therefore active in the same field in approximately the same period, but in different places – the former in Alexandria, the latter in Rome – and both worked on Aristarchus’ diorthōsis of Homer, dealing with it from different points of view. This occurrence is hardly a coincidence; rather, it is 13 Venice, Biblioteca Nazionale Marciana, gr. 454 (822). A high-resolution reproduction of the entire manuscript has been made available via the Homer Multitext project () and on the Marciana’s repository in the Biblioteca digitale italiana (); see also Dué 2009. 14 The colophon appears at the end of each book, with the exception of books 17 and 24, with minor variations or omissions. On the meaning of the initial παράκειται, see Montana 2010 and 2014a (cf. also Montana 2011: 151–52 and Pagani 2014, with discussion and bibliography on the relevance of this evidence in connection with the much debated problem of the origin of scholiography). 15 Lehrs 1882: 31–32 and Ludwich 1884–85: 1, 78–82 opted for the period following the death of Herodian; van der Valk 1963–64: I 107 and Dickey 2007: 19 date it within the fourth century; Erbse 1969–88: I xlv–xlviii and Schironi 2015: 609 support the fifth/sixth century. Van Thiel 2000: 11–12 (cf. 2014: I 27–28) has taken a unique stance by bringing the very existence of VMK into question. 16 Cf. Pontani 2005: 96–100, 149–50, with bibliography. 17 E.g., for Aristonicus, Suda α 3924 Adler (cf. Carnuth 1869); for Herodian, Σ Hrd. Il. 16.390c Erbse (cf. Lentz 1867–70: I lxxxiii, II.1 129– 65); for Nicanor, Suda ν 375 Adler (cf. Carnuth 1875). A fresh perspective on Nicanor is offered by Nünlist 2019 and Nünlist forthcoming. 18 Cf. Ludwich 1884–85: I 43 and 507–631 (frr.). Schmidt 1854: 183 mistakenly considered it to come from Didymus’ own commentaries. 19 For scholia to the Iliad, see Erbse 1969–88: VI 39, s.v. Δίδυμος, ὁ grammaticus (Index I), and 312, s.vv. διόρθωσις and διορθωτικά (Index III). The expression ἐν πρώτῳ διορθωτικῶν is also used in a scholion traced back to Herodian which quotes Didymus (Σ Hrd. Il. 24.557a Erbse, °66). For the scholia to the Odyssey, see Dindorf 1855: II 796, s.v. Δίδυμος (Index scriptorum). 20 These fragments are collected in the comprehensive edition of Schmidt 1854, as well as in Ludwich 1884–85: I 175–631.

12



Pagani

probably to be explained against the background of a specific cultural need expressed by the cultivated classes of the Roman society of that time (see below, §2.5). It is generally held that Aristonicus was slightly older than Didymus, but it is difficult to confirm that his work (or part of it) was known to and used by the latter.21 This was suggested already by Lehrs and Ludwich and recently reaffirmed by West,22 although Aristonicus’ name is never mentioned by Didymus, who, however, normally does cite his sources.23 In addition to this general objection, the specific arguments supporting this theory can also be disproved. The first among them is the fact that in Σ Ariston. | Ap.H. (?) Il. 8.535–37 (57) and Σ Did. Il. 15.86a.c Erbse (62) the excerptor explicitly declines to relate what Didymus had to say on these lines on the grounds that it coincided with Aristonicus’ comment reported just before: but this could arise from the fact that the epitomizer here simply used Aristonicus’ work before Didymus’,24 all the more so given that the opposite is also attested in Σ Did. | Ap.H. Il. 7.256–57, where the remarks omitted as superfluous are those of Aristonicus.25 A further series of pairs of scholia where Didymus and Aristonicus report the same Aristarchean textual choice, and both are unable to identify the source of the rejected reading except in a generic way, are also taken by West to demonstrate that Didymus, who usually tends instead to quote any available detail of this kind, drew on Aristonicus.26 However, it is equally plausible, or more so, to ascribe this vagueness in both Didymus and Aristonicus to a common source used by them, i.e. Aristarchus himself.27 Conversely, the best we can infer from Aristonicus’ and Didymus’ scholia to Il. 2.111 (50, cf. below, §§2.2 and 2.5) is that Aristonicus seems to have been unaware of Didymus’ discussion, since here the former relates without any caveat the version of facts that caused so much trouble to the latter.28 A comparison between the information found in the scholia respectively ascribed to Didymus and Aristonicus reveals divergent practices,29 presumably to be connected with the different aims of their works, which may tell us something about the Aristarchean sources at their disposal. Aristonicus seems to assume that there was only one Homeric text established by Aristarchus, without any reference to the two phases of diorthōsis attested by Didymus, and one series of hypomnēmata, likewise split by Didymus into two (see below, §§2.2, 2.3).30 Didymus, moreover, quotes several monographs by Aristarchus, as well as by other scholars (see below, §2.2); Aristonicus, on the other hand, refers only to the treatise On the Camp and may have known the Against Philitas, and may have had access to individual opinions of some of Aristarchus’ pupils.31 Conversely, Didymus’ scholia make no reference to critical signs, a fact that is probably explained by the nature of his work (which was not focused on this topic, unlike Aristonicus’) and perhaps as a result of the selection made during the compilation of VMK (when each work may have been used according to its specific content).32 Moreover, Didymus and Aristonicus sometimes give contrasting information about Aristarchus’ readings, reinforcing the impression that they had access to divergent evidence. The traditional idea that 21 On this issue, see Schmidt 1854: 277 (who thought, to the contrary, that Didymus was a source of Aristonicus); Lehrs 1882: 28; Ludwich 1884–85: I 51; Susemihl 1891–92: II 214 n. 379; Pfeiffer 1968: 218 (in whose opinion Aristonicus was younger than Didymus); Fraser 1972: II 689 n. 265; West 2001: 49–50; Pontani 2005: 62; Razzetti 2010: 60–61; Montana 2015b: 170; Schironi 2015: 626–27; Braswell 2017: 47; Schironi 2018: 18. 22 See previous note. 23 Cf. Schironi 2015: 626 and see below, §§2.2, 2.3. 24 Cf. Razzetti 2010: 60–61. 25 West 2001: 49 thinks nonetheless that ‘in each case Aristonicus must be regarded as the original’. 26 West 2001: 49–50. 27 Cf. Schironi 2015: 626. On the possibility that Aristarchus did not take note specifically of his sources for variant readings, cf. Schironi forthcoming and see below, §2.2. 28 West 2001: 49. Cf. Pagani 2015: 88, Schironi 2015: 626. 29 Cf. Schironi forthcoming. 30 Cf. West 2001: 65; Nagy 2004: 86; Nagy 2009: 34–37; Schironi 2015: 610–11; Schironi 2018: 15–16, 19–20, 35, with bibliography. 31 Schironi 2018: 18. 32 Nagy 2009: 35, to the contrary, considers this feature to reveal that Didymus did not have the ekdosis of Aristarchus: see next, in the text.

Didymus and epic poetry



13

Didymus, working in Alexandria, exploited a wider and more reliable set of Aristarchean materials in comparison with Aristonicus, in Rome, has been recently reaffirmed.33 On this hypothesis, Aristonicus must have limited his research to the last stage of the hypomnēmata (in addition to a very poor selection of treatises) because he probably had no access to any ekdosis, and that should have been enough for his purpose.34 On the other hand, the opposite view has also been maintained, according to which the definitive version of both Aristarchus’ diorthōsis and his hypomnēmata was present in Rome, and hence available to Aristonicus, whereas Didymus had no direct access to any of them and could count only on post-Aristarchean rearrangements.35 The latter interpretation, however, seems to be less in tune with the evidence (ambiguous and problematic though it be).

2.2. Didymus’ Sources Didymus’ work on the Homeric poems went far beyond the aim of collecting and explaining Aristarchus’ textual choices: he seems in addition to have provided them with a background of other variant readings transmitted by copies of the epics (ekdoseis)36 or endorsed by different scholars, thereby producing what has been defined as ‘a sort of apparatus criticus’.37 In the scholia that can be traced back to Didymus the following materials are mentioned:38 1. ekdoseis: two versions by Aristarchus; the so-called κατ᾿ ἄνδρα (i.e. individual texts), such as those of Antimachus, Zenodotus, Rhianus, Aristophanes of Byzantium, Callistratus, and the otherwise unknown Sosigenes and Philemon; those κατὰ πόλεις (i.e. city texts), from Marseilles, Chios, Argos, Cyprus, Sinope, Crete; and one identified as πολύστιχος (possibly characterized by the presence of plus-verses); 2. the commentaries of Aristarchus, and probably others by different scholars;39 3. treatises by Aristarchus (Πρὸς Κομανόν in Σ Il. 1.97–99, 2.798a, 24.110b1 Erbse; Πρὸς Φιλίταν in Σ Il. 1.524c, 2.111b (50) Erbse; Πρὸς τὸ Ξένωνος παράδοξον in Σ Il. 12.435a1 Erbse; Περὶ Ἰλιάδος καὶ Ὀδυσσείας in Σ Il. 9.349–50 Erbse),40 and many other authors. On the basis of this picture, a lively discussion has developed about the possibility that at least some of these sources were already available to, and consulted and recorded by, Aristarchus in his commentaries and monographs, and ‘inherited’ from there by Didymus. This issue is related to the crucial question whether Aristarchus and the first Alexandrians compared different copies in order to establish the text of a literary work, or operated merely through conjectural emendation.41 The endorsement of the former view, which is also supported by some practices of Aristarchus attested by Didymus himself (see below, §2.4), has sometimes led scholars to

33 e.g. Lehrs 1882: 16–28; Ludwich 1884–85: I 23, 41–43, 64–57; West 2001: 65; Schironi 2015; Schironi 2018: 35 ff. 34 Cf. most recently Schironi 2015 and Schironi 2018: 18–20, 35 ff. The fact that Philoxenus and Seleucus, who were active in Rome approximately in Aristonicus’ time, also wrote on Aristarchus’ sēmeia may suggest the presence of the relevant bibliography there (West 2001: 47–48; Nagy 2009: 35, from a different point of view). 35 Nagy 2009: 33–37. 36 Ekdosis can simply mean a ‘text’, while an ‘edition’ as the result of textual constitution is rather indicated by διόρθωσις: the latter could be hosted within an ekdosis: cf. West 2001: 50; Montanari 2015b, with further bibliography; for an overview of the ekdoseis quoted in Homeric scholia and papyri, see also Pagani and Perrone 2012: 97–102. 37 West 2001: 46, who, however, acknowledges that this was not very systematic. 38 Cf. West 2001: 50–85; Schironi 2018: 19–22. 39 Several times ὑμπομνήματα are mentioned without further specification, and it is not always legitimate to infer the authorship of Aristarchus, as shown by the fact that at least twice there are commentaries explicitly attributed to an unknown Diogenes (Σ Did. Il. 8.296b and 441a Erbse): cf. West 2001: 74. Moreover, Didymus’ mention of οἱ ὑπομνηματισταί (Σ Il. 16.467a1 Erbse) suggests a plurality of hypomnematic sources: Schironi 2018: 19; see on this term see Coward, this volume, 40–42. 40 West 2001: 79; Schironi 2018: 31–35. 41 For a recent status quaestionis and a critical assessment, see Montanari 2015b, with bibliography.

14



Pagani

maintain that the germinal ‘critical apparatus’ that is found in the scholia ascribed to Didymus was initiated by, and at least partly derives from, Aristarchus.42 However, the two issues do not coincide: as has recently been reaffirmed in a well-balanced argument, it is possible to support the idea that Aristarchus collated manuscript evidence in order to establish his Homeric text, without necessarily positing that Didymus picked up from him the references to the readings attested in various ekdoseis.43 One of the loci classici of this discussion is Σ Did. Il. 1.423–24 Erbse, where Didymus quotes verbatim from Aristarchus’ opinion about problems of textual arrangement and meaning, introducing it as ‘the wording of Aristarchus, from the first book of his commentary to the Iliad’ (λέξις Ἀριστάρχου ἐκ τοῦ αʹ τῆς Ἰλιάδος ὑμπομνήματος). This scholion continues as follows: οὕτως δὲ εὕρομεν καὶ ἐν τῇ Μασσαλιωτικῇ καὶ Σινωπικῇ καὶ Κυπρίᾳ καὶ Ἀντιμαχείῳ καὶ Ἀριστοφανείῳ. Καλλίστρατος δὲ ἐν τῷ Πρὸς τὰς ἀθετήσεις ὁμοίως, καὶ ὁ Σιδώνιος καὶ ὁ Ἰξίων ἐν τῷ ἕκτῳ Πρὸς τὰς ἐξηγήσεις. This is the way we found it also in the (ekdoseis) of Marseilles, Sinope, Cyprus, Antimachus, and Aristophanes. Callistratus in his Against the athetēseis (has it) likewise, and so also do (Dionysius) of Sidon and (Demetrius) Ixion in the sixth book of his Against the exēgēses. The problem is whether at least the first sentence belongs to the quotation from Aristarchus, assuming that Didymus intervenes only further on to add the information about the treatises. This view is assumed in Erbse’s edition and has been recently exploited in order to claim that the expression οὕτως […] εὕρομεν was initially an Aristarchean usage, afterwards continued by Didymus.44 However, given that this kind of wording could easily not be an author-specific trait, the opposite opinion has also been maintained, claiming that expressions such as εὕρομεν and the like are a characteristic mark of Didymus (and not Aristarchus), as suggested by other scholia where he claims to ‘have found’ a variant reading in Aristarchus’ ekdosis itself.45 When read against this backdrop, the argument that Aristarchus is described twice in Didymean scholia (Σ Did. Il. 6.4b, 9.222b1 Erbse) as ‘finding’ (εὑρών) a variant reading46 does not prove the Aristarchean origin of the expression—all the less so if we consider that, where we have the parallel formulation by Aristonicus (6.4a), the latter says περιπεσών, not εὑρών. Given the level of uncertainty inherent in our evidence, the risk of circular reasoning is high,47 and the issue must to a certain degree remain open. However, it is impossible to demonstrate that any of the lists of ekdoseis cited by name in our scholia derives from Aristarchus,48 whereas, on the contrary, an approach aimed at searching and recording various sources in detail is clearly attested for Didymus in a passage such as Σ Il. 2.111b Erbse (50, attributed to him by the subscription ταῦτα ὁ Δίδυμος), where the collection of evidence cannot be Aristarchus’ legacy, since it is intended to explain problems in the transmission of the variants preferred by Zenodotus and Aristarchus himself.49 A comparison with the scholia that explain Aristarchus’ critical signs, and which are therefore attributed to Aristonicus, suggests as much. Given the aim of his work, Aristonicus must have had at his disposal Aristarchus’ commentaries (in addition to at least one or possibly

e.g. Nagy 2004: 85–109, with further bibliography. Schironi 2018: 63–75, explicitly following in the footsteps of Lehrs 1882. Erbse 1969–88: I 119–20; Nagy 2004: 89–94. e.g. Σ Did. Il. 2.131a1, 2.517a, 4.3a, 15.469–70a1, 16.636c1: see West 2001: 70–72. Cf. Nagy 2004: 90–91. Schironi 2018: 64 is well aware of this (cf. also Nagy 2004: 94). West 2001: 50–79, who, however, goes too far in claiming that Aristarchus did not check the manuscript evidence; Montanari 2015a; Schironi 2018: 67. 49 On this passage, see Pagani 2015; Schironi 2015: 612–15; and, with a focus on the direct quotation of Aristarchus by Didymus, Schironi forthcoming.

42 43 44 45 46 47 48

Didymus and epic poetry



15

two of the monographs), if not the ekdosis;50 yet he never mentions any variant reading by the name of the edition or the scholar that supports it (with the unsurprising exception of Aristophanes and Zenodotus),51 and, rather, refers to them with a generic or collective expression such as ἔνιοι or ἔν τισι. This may well depend on Aristonicus’ personal choices in excerpting Aristarchus’ work, but this situation more likely reflects the fact that Aristarchus himself did not take note in detail of all the ekdoseis that he had consulted in order to establish his Homeric text.52 Nevertheless, the problem remains of which texts can be considered to have actually been available to Didymus. According to the picture outlined by West, this seems not to be the case for Zenodotus’ ekdosis, whose variants Didymus probably found commented upon (and often criticized) in Aristarchus’ hypomnēmata.53 As for Aristophanes’ text, the picture is ambiguous: on one side Didymus once claims to ‘have found’ a variant in Aristophanes’ ekdosis (but it is uncertain whether he is indeed the subject: see above, Σ Il. 1.423–24);54 on the other side, he cites it on several occasions through some intermediate authority,55 and we have no positive evidence that other scholars active in Alexandria in Didymus’ time had it at hand. All in all, it seems more likely that Didymus did not directly consult Aristophanes’ text but knew its readings thanks to (again) Aristarchus and/or Callistratus.56 The latter may also be the source for the readings of Rhianus,57 even if nothing excludes direct knowledge of them on Didymus’ part, and for those of Sosigenes.58 As for Antimachus’ ekdosis, it would have been very old by Didymus’ time (about four hundred years), but this does not make its survival impossible.59 On the other hand, direct access on Didymus’ part to the ekdosis of an unknown Philemon is made unlikely (though not excluded) by the fact that it is mentioned only twice in the Didymean scholia, which suggests rather that Didymus found it cited in some of his sources.60 Among the city texts, the ones from Marseilles and from Chios, which are frequently quoted, may have been available to Didymus, whereas for the Sinopean ekdosis this is a more vague hypothesis. Access to the variants of the Cretan and Cyprian texts may have been provided by Seleucus,61 who may also have been Didymus’ source for the ekdosis called polystichos.62 In Didymus’ scholia we also find collective references to groups of editions, apparently distinguished on qualitative grounds, such as αἱ κοιναί/κοινότεραι (‘the [more] common ones’), or αἱ δημώδεις (‘the popular ones’), and αἱ εἰκαιότεραι (‘the more ordinary ones’); on the other side we find αἱ χαριέστεραι/χαριέσταται (‘the more/most refined ones’). The expressions αἱ πᾶσαι/ ἅπασαι may seem quite obvious, but since in a few cases they appear contrasted with something else, according to West they do not mean ‘all the editions’, but indicate a set of particularly respectable texts which Didymus held in great regard: this explanation, which in fact equates the terms αἱ πᾶσαι/ἅπασαι with αἱ χαριέστεραι/χαριέσταται, does not seem fully cogent, and further research will perhaps shed new light on the

50 Cf. Schironi 2015 and 2018: 15–18, 73–75. See above, §2.1. 51 One of Aristarchus’ critical signs (the diplē periestigmenē) was specifically devoted to marking lines where he disagreed with Zenodotus’ text, while Aristophanes’ ekdosis was at the basis of Aristarchus’ first hypomnēma: see below. 52 Schironi 2018: 16–17, 72–73, who also conveniently distinguishes three levels of discussion: 1) whether Aristarchus collated manuscripts; 2) what role this collation played in his diorthōsis; 3) whether he listed the manuscripts he had checked (p. 64). See also Schironi forthcoming and cf. Σ Did. Il. 1.423–24 Erbse, quoted above, where the variant μετὰ δαῖτα is attributed, in Aristarchus’ own words, to ἔνιοι. 53 The fact that Didymus did not have access to Zenodotus’ text is suggested by several passages where he apparently is not able to give a certain answer on his position: e.g. Σ Did. Il. 1.97, 2.111b (50), 4.3a, 13.808a, 14.37 Erbse: cf. West 2001: 54–56. 54 Wording indicating that Aristophanes γράφει something (e.g. Σ Did. Il. 3.13b, 6. 148a, 4.137a1 Erbse, etc.), also mentioned by West in this respect, have in my view less strength in suggesting autopsy. 55 e.g. Σ Did. Il. 19.327a, 21.130–35a1 Erbse; Σ Od. 1.424a Pontani, 13.152 Dindorf. 56 West 2001: 59–60. 57 Both Callistratus and Rhianus are quoted much more frequently for the Odyssey than for the Iliad: West 2001: 56–58. 58 West 2001: 58. 59 West 2001: 52–54. 60 West 2001: 58–59. 61 Cf. Σ Did. Il. 1.381 Erbse: West 2001: 70. 62 West 2001: 67–72.

16



Pagani

issue.63 At all events, the whole account is highly speculative and not uncontroversial; it has been claimed that Didymus did indeed draw the majority of the data connected with these ekdoseis from Aristarchus.64 The situation is by no means clearer when it comes to the Aristarchean material itself. Didymus seems to quote two versions of his ekdosis, speaking of αἱ Ἀριστάρχου, αἱ Ἀριστάρχειοι, ἡ ἑτέρα/προτέρα/δευτέρα/ χαριεστέρα τῶν Ἀριστάρχου/τῶν Ἀρισταρχείων (‘one [of two]/the first/second/more refined of Aristarchus’ [ekdoseis/diorthōseis?]’), or using the adverb διχῶς (‘in two ways’) when referring to a reading of Aristarchus.65 Moreover, he mentions on the one hand commentaries written by Aristarchus according to the ekdosis of his predecessor, Aristophanes of Byzantium (ἐν τοῖς κατ᾿ Ἀριστοφάνην ὑπομνήμασιν Ἀριστάρχου, Σ Il. 2.133a Erbse), and on the other hand some ‘perfected commentaries’ (ἔν τινι τῶν ἡκριβωμένων ὑπομνημάτων, Σ Il. 2.111b Erbse, 50).66 The wording of certain Didymean scholia, such as ἐν τῇ ἑτέρᾳ τῶν Ἀριστάρχου εὕρομεν (‘we found within one of the two [ekdoseis/diorthōseis?] of Aristarchus’, Σ Il. 2.131a1 Erbse) or διχῶς ἐν ταῖς Ἀριστάρχου εὑρίσκομεν (‘we find in two ways in the [ekdoseis/diorthōseis?] of Aristarchus’, Σ Il. 2.517a Erbse), may suggest that he had direct access to the text of Aristarchus.67 Nevertheless, other examples lead us to believe that Didymus favoured the evidence provided by commentaries and treatises, by Aristarchus and his pupils, even in cases when consulting the text established by Aristarchus would have been more effective.68 E.g. in Σ Did. Il. 2.111b Erbse (50) he resorts only to the contrasting information given by commentaries and treatises, by Aristarchus himself and by other scholars, in order to ascertain Aristarchus’ (as well as Zenodotus’) reading.69 Σ Did. Il. 19.365–68 Erbse reports that, in the words of Dionysius of Sidon, Aristarchus firstly athetized the four lines and afterwards removed the obeloi,70 whereas Ammonius gave no similar information. Didymus (Σ Did. Il. 21.130–35a1 Erbse) relates that Aristarchus in his hypomnēmata recorded an athetesis by Aristophanes as well as its motivation, and adds that perhaps Aristarchus himself agreed with this choice, as he inferred e silentio from the absence of a counterargument. In all these cases, we wonder why Didymus did not simply check the situation of the text and the critical signs in Aristarchus’ edition. The traditional theory argues that this approach by Didymus depended on his awareness that he was using texts that were not Aristarchus’ originals and were therefore not fully reliable.71 But an alternative explanation may be proposed, on the basis of a different interpretation of the twofold nature of Aristarchus’ work. 63 West 2001: 51–52; cf. Schironi 2018: 66. The examples mentioned by West are limited to Σ Did. Il. 12.161b and 19.95b1, b2 Erbse, out of a total of approximately fifty occurrences of αἱ πᾶσαι/ἅπασαι within the scholiastic corpus (see Erbse 1969–88: VI 268 and 444, s.vv. πᾶς and ἅπας). Σ Did. Il. 13.485a1 Erbse should be removed from this list, since the recorded reading of Zenodotus’ text is not in competition with the one ascribed to πᾶσαι (it concerns another word of the line); Σ Il. 2.196c Erbse, which West uses to corroborate his theory, risks complicating the picture even more (one of the MSS bearing it has αἱ χαριέσταται […] ἄνευ Ζηνοδότου, the other one has αἱ πᾶσαι πλὴν τῆς Ζηνοδότου). The same holds true for Σ Did. Il. 3.18b1 Erbse, where the expression ἐν πᾶσαις ταῖς χαριεστέραις appears. 64 e.g. Nagy 2004: 86–109; Nagy 2009: 21–37. 65 Cf. West 2001: 61. On the doubt about which word is implied by the elliptical expressions mentioned, see below, §2.3. 66 Didymus’ mention of αἱ ἐξητασμέναι Ἀριστάρχου (Σ Il. 7.130a1 Erbse) may refer to the second ekdosis/diorthōsis, but the plural is strange: the correction of ἐν ταῖς ἐξητασμέναις to ἐν τοῖς ἐξητασμένοις suggested by Lehrs 1882: 21 and accepted by Pfeiffer 1968: 217 aims to identify them with the ἠκριβωμένα ὑπομνήματα of Σ Il. 2.111b Erbse (50). The emendation was rejected by Erbse 1959: 293 and 1969–88: II 251; cf. Montanari 1998: 48–81; West 2001: 74; Pagani 2015: 75 and n. 9; Schironi 2018: 35, 44–45 and n. 72. 67 Cf. West 2001: 63. The passage of Σ Did. Il. 4.3a Erbse which says ἐν […] ταῖς ἐκδόσεσι […] εὕραμεν is to be left out, since it is not certain that the ekdoseis here mentioned are those of Aristarchus: cf. Montana 2014b: 264. 68 The following examples in the text are taken from West 2001: 63–64, with the omission of Σ Did. (?) Il. 4.345–46a Erbse (which remarks that these lines were not athetized according to the commentaries, but that a doubt over their authenticity arises because they were criticized by some members of the Aristarchean school: for the interpretation of οἱ ἡμέτεροι, see Villoison 1788: xxvii, followed e.g. by Erbse 1959: 285 and West 2001: 64), due to its uncertain Didymean authorship (cf. Erbse 1959: 284–85 and 1969–88: I 510). On this problem, see also Lehrs 1882: 16–17 and 27–28; Ludwich 1884–85: I 38–41; Pfeiffer 1968: 216–17; Montanari 2000: 481; West 2001: 63–67; Montanari 2003: 36. 69 For a more detailed discussion of this passage see the bibliography quoted above, n. 49. 70 On Aristarchus’ second thoughts see Montanari 2000 and 2003; Schironi 2015: 612–24. 71 Ludwich 1884–85: I 38–39; West 2001: 66–67; Nagy 2004: 86 with n. 65; cf. Nagy 2009: 21–37, who thinks that the texts with Aristarchus’ diorthōsis which Didymus had at his disposal were in fact post-Aristarchean products. According to Pfeiffer 1968: 217 this is

Didymus and epic poetry



17

2.3. Didymus as a Witness of Aristarchus’ Homeric Works In order to do so, we have to add to the picture the problematic witness of Ammonius of Alexandria, a direct pupil of Aristarchus, about the ecdotic work of his master, as reported by Didymus himself. The latter ascribes to Ammonius two titles: Περὶ τοῦ μὴ γεγονέναι πλείονας ἐκδόσεις τῆς Ἀρισταρχείου διορθώσεως (On the Fact That There Were Not More Editions of Aristarchus’ Recension, Σ Il. 10.397–99a Erbse, 59) and Περὶ τῆς ἐπεκδοθείσης διορθώσεως (On the Re-edited Recension, Σ Il. 19.365–68a1 Erbse). The fact that Ammonius felt the need to clarify the situation concerning the editorial steps of the Aristarchean diorthōsis is admittedly a symptom indicating that some confusion had arisen about the issue.72 However, Ammonius’ titles themselves seem to be inconsistent with one another. This fact has been accounted for in various ways: by interpreting μὴ […] πλείονας ἐκδόσεις in the first title as meaning ‘not more than two’;73 by supposing that the second ekdosis had been edited by Aristarchus’ pupils (i.e. Ammonius himself) and hence was not stricto sensu ‘of Aristarchus’;74 by hypothesizing that two ekdoseis indeed existed—one being the working text, the other the published text—but were then combined together and disseminated in that form after Aristarchus’ death by a pupil of his;75 or by positing that the ἐπεκδοθεῖσα διόρθωσις represented a second step of the ecdotic work of Aristarchus but found its place in the same book as the first phase rather than in a separate one.76 This last explanation, which I believe to be the most persuasive one, also allows us to understand Didymus’ concerns and doubts in consulting an edition of this kind – one which bore a stratification of different layers of diorthōsis sometimes even in conflict with each other – and his frequent recourse to the explanations found in commentaries and treatises in order to solve questions left open by the mere collation of the edition. A consequence of this point of view is that Didymus should not have referred to two ekdoseis: this is tenable if we suppose that the implied word in expressions such as αἱ Ἀριστάρχειοι and the like (see above, §2.2) is διορθώσεις, and not ἐκδόσεις.77 The evidence provided by Didymus thus allows us to reconstruct the philological activity of Aristarchus on the Homeric poems, according to the most widely accepted outline, as follows: he first wrote commentaries on the basis of the text established by Aristophanes (οἱ κατ᾿ Ἀριστοφάνην ὑπομνήματα Ἀριστάρχου), then produced his own ekdosis, which brought with it new and refined commentaries (τὰ ἡκριβωμένα ὑπομνήματα), and finally a second step of diorthōsis, either produced by his pupils (Pfeiffer) or hosted in the same copy bearing the first layer (Montanari), in both cases without the explanatory support of a further hypomnēma.78

72 73

74 75

76

77

78

an ‘insoluble problem’, and caution is exerted also by Schironi 2015: 624–26. The presence of Aristarchus’ editions of the two Homeric poems at Rome until the end of the second century AD is attested by the recently discovered Περὶ ἀλυπίας of Galen, though we cannot be sure that they were (or were believed to be) Aristarchus’ own autographs, given the ecdotic and exegetical difficulties of this passage: discussion and bibliography in Montana 2014b. A comprehensive study of this work is now to be found in Petit 2019. It has been read as a sign of a ‘crisis of transmission’ resulting in a split within the post-Aristarchean tradition: Nagy 2009: 30–33. Lehrs 1882: 23–24; Ludwich 1884–85: I 16–17; West 2001: 62–63. On the legitimacy of implying that the contrary of πλείονες is ‘two’ instead of ‘one’, see the criticism by Erbse 1959: 276–77 (who furthermore believed that the ekdosis was in some way implicit in the commentaries, which included all the textual choices of Aristarchus), endorsed by Pfeiffer 1968: 215–17 with n. 6; counterarguments in West 2001: 62–63. Villoison 1788: xxvii; Wolf 1795 (1985): 195; Pfeiffer 1968: 215–17 with n. 6. Van Thiel 2014a: I 8–14, who moreover assumed that the annotations made by the Alexandrian scholars on their copies of literary texts were not variant readings or conjectures, but references for the purpose of comparison or alternatives written down with the aim of commenting, subsequently misunderstood by their followers. Moreover, in van Thiel’s opinion the commentaries were not produced by Aristarchus himself, but were a later production put together by his pupils (cf. also van Thiel 2014a: IV 124). Montanari 1998: 10–19; 2000: 480–86; 2002: 125–27; 2003: 31–41; 2009b: 156–59; 2015a: 657–60; 2015b: 657–60; cf. Montana 2014b: 262–65 and Pagani 2015: 74–78. Cf. the doubts raised in Schironi 2018: 38 about the expression ἐπεκδοθεῖσα, allegedly indicating a real process of edition (cf. also Nagy 2009: 21–30). Montana 2014b: 262–65 has demonstrated that it is possible and not in conflict with the evidence: there is only one passage which explicitly mentions αἱ Ἀριστάρχου ἐκδόσεις (Σ Did. Il. 2.221 Erbse), which, in isolation, may be explained as the result of a terminological fluctuation in the subsequent tradition, and two which on the contrary attest αἱ διορθώσεις (Σ Did. Il. 1.522a1, 2.192b1 Erbse), while the anonymous ἐκδόσεις cited in Σ Did. Il. 3.406a1, 4.3a, and 8.163c Erbse cannot be safely ascribed to Aristarchus. The most recent overviews, with bibliography and accounts of further different proposals, are to be found in Pagani 2015: 75–77 and Schironi 2018: 35–41 and 44–45.

18



Pagani

2.4. Didymus as a Witness of Aristarchus’ Method We also owe to Didymus crucial information about Aristarchus’ ecdotic approach. As already mentioned, some scholars have dismissed the philological work of the first Alexandrians, Aristarchus included, as based mostly or exclusively on conjectural emendation and not on manuscript evidence (see above, §2.2).79 Conversely, many others have responded by collecting a significant set of testimonies contradicting this view.80 We often have no means to distinguish whether a textual arrangement attributed to an ancient grammarian by the scholia was his own conjecture or the result of accepting a variant reading found in a copy of the work; but if the former method is nowhere clearly attested, the latter is on the contrary undeniably documented by Didymus (and also Aristonicus) with reference to Aristarchus.81 A more limited role, though not a negligible one, is played by a group of passages that show Aristarchus rejecting a textual arrangement ‘written’ by ‘some people’, ἔνιοι or τινές.82 These passages testify to Aristarchus’ knowledge of, and comparison among, different copies, but still not to a textual choice based on the manuscript evidence. The latter is instead found without any doubt in two important Didymean scholia.83 In Σ Il. 6.4b Erbse, Didymus says that Aristarchus firstly adopted the text μεσσηγὺς ποταμοῖο Σκαμάνδρου καὶ Στομαλίμνης (‘between the river Scamander and the lagoon’), but afterward, ‘having found’ the alternative μεσσηγὺς Σιμόεντος ἰδὲ Ξάνθοιο ῥοάων (‘between the Simoeis and the currents of Xanthus’), he accepted the latter (ὕστερον δὲ Ἀρίσταρχος ταύτην εὑρὼν †ἐπέκρινεν†).84 This account is paralleled by that of Aristonicus, which explains a critical sign marked by Aristarchus (presumably a diplē) by reporting that the first variant was written ἐν τοῖς ἀρχαίοις (sc. ἀντιγράφοις? = ‘in the ancient copies’)85 and hence is handed down in the hypomnēmata (sc. of Aristarchus), and that afterwards Aristarchus himself, ‘having come across’ the second one, ‘wrote’ that instead (ὕστερον δὲ περιπεσὼν ἔγραψε), evidently in a subsequent phase of his diorthōsis (see above, §2.3). These two scholia clearly outline an ecdotic practice that did take into consideration the selection of variant readings preserved by the manuscript tradition, in order to improve the text. An even more significant case is that of the Didymean scholia to Il. 9.222: the less epitomized one (222b1 Erbse, MS A) tells us that Aristarchus found the text problematic, because it says that Odysseus’ speech to Achilles begins only after the three ambassadors ‘had satisfied their desire for drink and food’ (αὐτὰρ ἐπεὶ πόσιος καὶ ἐδητύος ἐξ ἔρον ἕντο) while in fact they have already taken part in a banquet in Agamemnon’s tent immediately before leaving for their mission (Il. 9.174–78). Aristarchus’ puzzlement at this passage emerges also in Aristonicus’ scholion (222a Erbse), which however maintains only that the line has been misused (κατακέχρηται) in a way characteristic of the cyclic poets (κυκλικώτερον). Didymus instead explains that, according to Aristarchus, the poet should rather have said ‘after they had acquired again (ἂψ ἐπάσαντο) drink and food’, or ‘quickly acquired’ (αἶψ᾿ ἐπάσαντο); despite this, Aristarchus ‘did not change anything due to excessive caution, because he had found the reading transmitted in many (sc. copies)’ (ἀλλ’ ὅμως ὑπὸ περιττῆς εὐλαβείας οὐδὲν μετέθηκεν, ἐν πολλαῖς οὕτως εὑρὼν φερομένην τὴν γραφήν). The same annotation by Didymus has been excerpted in another two scholia, whose content is very instructive about the gaps in information that the compiling process may entail: 1) a scholion in MS A (222b2 Erbse), written in the margin between the text and the frame scholia (Aim), reports only that Aristarchus believed it would have been better to say ‘had acquired again’; 2) in a scholion in MS T (222b3 Erbse) we find first, misleadingly, that ‘Aristarchus writes 79 e.g. van der Valk 1963–64; van Thiel 1992 and 1997; West 2001. A recent overview, with further bibliography, in Montanari 2015b. 80 e.g. Schmidt 1997; Montanari 2002 and 2004; Nagy 2004 and 2009; Montanari 2009b and 2011; Rengakos 2012; Montanari 2015a, 2015b, 2015c, 2020. Cf. Schironi 2018: 63–75. 81 Cf. Rengakos 2012: 240–41, 247; Montanari 2015a: 123; 2015b: 660–71, pace West 2001: 37. 82 Σ Did. Il. 1.423–24 (see above, §2.2), 9.159, 9.401b (cf. Σ Ariston. Il. 9.401a), 11.142a Erbse. Two further similar cases are attested by Aristonicus: Σ Ariston. Il. 3.230a and 16.105 Erbse. See Rengakos 2012: 245. 83 See Rengakos 2012: 244–46; Nagy 2004: 87; Montanari 2015a: 120–25; Schironi 2018: 70–72. 84 The reading of the MSS (ἐπέκρινεν b, ἐπέκρινε T) is not fully satisfactory: the translation is based on Bekker’s conjecture ἐνέκρινεν. 85 The implied noun is added by Rengakos 2012: 245: an alternative seems difficult to find.

Didymus and epic poetry



19

“had acquired again”’, then the explanation that the characters involved had already eaten their fill, and only at the end the adjustment that ‘however, he has not changed the reading’ (οὐ μετέθηκε δὲ τὴν γραφήν). These crucial passages therefore represent indisputable evidence that the Aristarchean diorthōsis also took into account the manuscript tradition and was not exclusively, or even mostly, based on arbitrary interventions motivated by content or linguistic coherence or propriety.86 On the other hand, Σ Did. Il. 9.222b1–b3 sheds light also on Didymus’ approach to his own work.

2.5. Didymus’ Method Didymus is generally described as a dull compiler, a mere epigone of the great productive period of the first Alexandrians.87 This view is not entirely incorrect and finds good justification in the very choice of excerpting and collecting pre-existing material. Still, it may benefit from some adjustments of perspective. Didymus’ personal imprint can be detected both in his expression of opinions on the matter at hand, and in deliberately including or excluding in his doxographical compilation items that are explicitly declared useful or not, as the case may be. In several passages he even criticizes a textual choice by Aristarchus, using adverbial forms such as κακῶς (‘badly’, Σ Did. Il. 1.106e, 21.162a2 Erbse) or οὐ καλῶς (‘not well’, 2.355a2 Erbse), or comparing it to another reading judged χαριέστερον (‘more graceful’, 7.428a1 Erbse) or οὐκ ἄχαρις (‘not graceless’, 2.462a2, cf. a3, 3.292a1 Erbse), or such that it ἔχει λόγον (‘makes sense’, 21.162a1, cf. 2.462a1 Erbse). Sometimes he admits that a reading, despite not being the one preferred by Aristarchus, fits the Homeric character or usage (ἔχει […] τὸν Ὁμηρικὸν χαρακτῆρα […], καίπερ οὐκ οὖσα Ἀριστάρχειος, 3.18a, sim. b1 Erbse: καίτοι μὴ οὖσα Ἀριστάρχειος, ὅμως ἔχει Ὁμηρικὴν συνήθειαν), or that a competing reading is sensible too (ἔχει […] λόγον καὶ ἡ/τό […], 9.584a1 and a2, cf. a3 Erbse: δύναται […] μᾶλλον […]).88 A scholion discussed above (see §2.4) shows Didymus labelling Aristarchus’ conservative stance at Il. 9.222 as an exaggerated scruple; the curtailed versions of it in Aim and T clearly show that the characterization of Didymus as a passive vehicle of third-party positions may have been magnified by the process of transmission, in which Didymus’ own remarks could be liable to omission. Although Didymus’ opinion may appear to us inferior to Aristarchus’, this should not lead us to neglect the personal contribution of the former.89 Further examples testify to a critical awareness in the activity of compilation, which will not have been accomplished in a purely mechanical way. Faced with a complicated task such as determining Aristarchus’ and Zenodotus’ readings in Il. 2.111b (see above, §2.2), Didymus suggested a mistake in the tradition within Aristarchus’ school and felt the need to explore each available source, even taking note of dead ends: when consulting the works of Aristarchus’ pupils Dionysodorus of Troezen and Ammonius, he found that they added nothing useful to the topic, and so he decided not to quote their witness (διὸ καὶ τὰς μαρτυρίας αὐτῶν οὐκ ἐγράψαμεν) – yet he nonetheless gives a full account of his research process.90 Far from being a groundless display of erudition,91 it reveals the meticulousness of Didymus’ work. In another case (Σ Did. Il. 3.406a Erbse), after noting which was the reading of both phases of the Aristarchean diorthōsis, Didymus wonders from where the competing variant derived, since he was unable to find it attested anywhere (he presumably was aware of it because it was implied in the discussion of Aristarchus’ commentaries);92 perhaps realizing 86 See Rengakos 2012: 244–46; Nagy 2004: 87; Montanari 2015a: 120–25; Schironi 2018: 70–72, pace West 2001: 37 and n. 19 (about Σ Did. Il. 9.222 Erbse): detailed counterarguments in Montanari 2015a: 122–23. 87 e.g. Pfeiffer 1968: 275–76, who, however, acknowledges some degree of personal intervention by Didymus, but downplays it, considering the value of his opinions second-rate in comparison with Aristarchus’ legacy: see next, in the text. 88 Overview in Schironi 2018: 22. 89 Cf. above, n. 87. For general remarks on this large subject, see Montanari 2018: 350–51. 90 The remains of the treatise Περὶ Δημοσθένους (P.Berol. inv. 9780, °281) provide clear evidence of the same kind of method (esp. col. iv 59–63, vi 18–43, and 59–62): cf. Montana in this volume. 91 Cf. van Thiel 2014a: I 176. 92 Nagy 2004: 103.

20



Pagani

that the item was potentially problematic, he then states that he has resolved to cite Aristarchus’ wording of the explanatory paraphrase (προσθήσειν μοι δοκῶ καὶ τὴν Ἀριστάρχου λέξιν οὕτως ἔχουσαν).93 Didymus’ attention not only to consulting many collateral sources in addition to the primary witness represented by Aristarchus’ work, but also to providing a detailed account of the whole evidence, perfectly fits his role as preserver of variants, compiler of excerpts, and collector of the outcomes of the creative phase of Hellenistic philology, but this does not make him merely an obtuse plodder. On the contrary, Didymus’ activity was probably a response to the cultural need in early Imperial Rome to read the literary masterpieces of the Greek past through the editions and commentaries that bore the quality guarantee of Alexandrian philology.94 We can only speculate that Didymus’ works, produced in the most important and paradigmatic centre of the Hellenistic scholarship, could have been presented (or been perceived by their readership) as the highest standard of sources, in comparison to many similar others that flourished at that time, perhaps Aristonicus’ included. Be that as it may, Didymus was certainly able to ensure the survival of the knowledge of the past, not only by collecting a doxography with its results but also by following up its methods and vision at the crucial turning point between Hellenistic and Imperial scholarship.95

93 Pace Nagy 2004: 102, who suggests a significantly different translation of this part. Cf. LSJ s.v. δοκέω I.3b; BDAG s.v. δoκέω 1B. 94 Cf. Montana 2016: 109–11. 95 Montana 2012a: 81. Cf. also Schironi 2018: 22, 25, 65, who, however, accepts the traditional thesis of Didymus’ ‘sloppiness’ (and hence unreliability).

Bulletin of the Institute of Classical Studies, 2020, 63, 21–33 doi: 10.1093/bics/qbaa015 Original article

Chapter Two Didymus and lyric* Enrico Emanuele Prodi University of Oxford, Oxford, UK

ABSTRACT Didymus worked extensively on archaic lyric poetry. The greatest amount of surviving material comes from the Pindar scholia and concerns Pindar’s Epinicians, but there are fragments and testimonies of his commentaries to other authors and a treatise On Lyric Poets. This chapter reviews the evidence for Didymus’ lyric scholarship, then discusses the contents of the On Lyric Poets—whose surviving fragments are concerned with the identification of lyric genres and the etymologies of their names—and the threads that run through his Pindaric exegesis: the compilation and evaluation of earlier scholarship, the use of historiographical evidence, textual criticism, a concern for the constitution of the Pindaric corpus and the contextualization of individual poems, and strategies of literary interpretation such as recourse to recurrent Pindaric themes and the train of thought of a passage.

In his invective against philology in the Letters to Lucilius (88.37, T14) Seneca chastises Didymus for indulging in extravagant disquisitions about literary minutiae.1 Two of his four examples concern archaic Greek lyric poets: was Anacreon’s life more devoted to lust or to drunkenness? (396) Was Sappho a courtesan? (397) Hans Bernsdorff has realized that a fragment of a hypomnēma to Anacreon casts light on the true scope of the first question, and indeed sounds a warning on Seneca’s polemical distortions.2 Fr. 2 of P.Oxy. LIV 3722 (second century AD) reads as follows (103):3 ] [ ][ ] θεράπων ἔμηνεν ε* λ [ ]υτον ἀπὸ τοῦ συμποσίου Πυθόμ]α*νδρον ἀπιό* ντα οὗ νυ*(ν) [ ] προπεσών· ὁ μὲν Δίδυμος ]* * βέλτ{ε}ιό* ν φησιν ἐπὶ [ ]* ἐρωτ* ικών· τόδε μεν τὸ] μ*ὲν γὰρ οἶδα καὶ λ{ε}ίαν ]* ὅ* τι μεθυσθεὶς παρὰ [

(PMG 454?) [ [ [ [

* This chapter was written at speed, following the withdrawal of its appointed author, during a long period of library closures. I apologize for any shortcomings that result from these circumstances and I take responsibility for those that don’t. The Pindar scholia are cited from Drachmann’s edition. All translations are mine. Numbers in bold refer to testimonia and fragments in the Checklist (pp. 95–120). 1 On this passage, see Luzzatto 2011: 8; DiGiulio, this volume, 81–82. 2 Bernsdorff 2020: I 48, II 272–74. I am grateful to Prof. Bernsdorff for sharing his thoughts with us before publication. 3 CLGP I.1.2.2 Anacreon 3. For bibliography and an apparatus see ibid. and Bernsdorff 2020: II 270–75. © The Author(s) 2021. Published by Oxford University Press on behalf of the Institute of Classical Studies. All rights reserved. For permissions, please email: [email protected].



21

22



Prodi

δηὖτε Πυθόμα]ν* δρον* : υ* vπ* ετ* αρτ* … [ ]εφ .[ ]…[ ] . . [

(PMG 400?)

The fragment is difficult to supplement into continuous text, but it seems that the first-person speaker in one of Anacreon’s ditties claimed to have experienced a certain mishap: either madness (2 ἔμηνεν) or a fall (4 προπεσών), if not both. A zētēma arose as to whether this mishap was due to love (6 ἐρωτ* ικῶν) or inebriation (8 μεθυσθείς),4 and Didymus—the supplement at 4–5 imposes itself—stated that the former was better, with his trademark βέλτιον.5 The likely commentary to Anacreon cited by P.Oxy. 3722 is another piece in the already extensive jigsaw puzzle of Didymus’ scholarship on archaic Greek lyric. Two marginal notes in P.Oxy. XV 1788 fr. 4 testify that he wrote on Alcaeus (100, 101).6 He is probably mentioned in the margins of another text, P.Oxy. XXI 2299 fr. 10b, which could be by either Alcaeus or Sappho (102).7 Papyrological traces of a work on Alcman are doubtful.8 It can perhaps be divined from a scholion to Pindar that he commented on Archilochus.9 Herennius Philo and Ammonius refer to a hypomnēma to Bacchylides’ Epinicians (°175), and he may have commented on the rest of his poetry too: P.Oxy. XXIII 2368, almost certainly from a commentary to Bacchylides’ Dithyrambs, is a plausible candidate for Didymean authorship.10 The most extensive evidence to have come down to us concerns his scholarship on Pindar (104–74), preserved mostly by the scholia vetera with occasional contributions from papyri and other sources. Beside his work as a commentator, he also wrote a monograph On Lyric Poets, from which only scanty fragments survive. 1. ON LYRIC POETS The only explicit testimonies for Didymus’ On Lyric Poets are the Late Antique and Byzantine Etymologica.11 They cite it sometimes as Περὶ λυρικῶν ποιητῶν (°347a, c, d), sometimes as Περὶ ποιητῶν (°345a); the true title is probably the former.12 Material which the Etymologica attribute to this treatise turns up in Proclus’ Chrestomathy, or rather in the summary of its first two books in Photius’ Library, codex 239.13 Some of it is also found in the commentary to Aphthonius by John of Sardis, which (given their similarities) probably took it from Proclus.

4 Love and drink—both frequent ailments of the Anacreontic narrator, and de rigueur at the symposion—are sometimes combined into a single image: PMG 376.2 μεθύων ἔρωτι, 450 ἔρωτα πίνων (Bernsdorff 2020: I 48). 5 See Benelli 2011: 53–54. 6 McNamee 2007: 146–47; CLGP I.1.1 Alcaeus 4. Both notes are mutilated: the first is quite extensive and may have been paraphrastic; the second explains a critical sign, the ἄλογος (Lobel 1951b: 142; McNamee loc. cit.). P.Oxy. 1788 has sometimes been attributed to Sappho, most recently by Liberman 1999: I lxxxvii–xci, but see Lentini 2007. 7 McNamee 2007: 158; CLGP Alcaeus 18. See Lobel 1951a: 71. Lasserre 1989: 31, 92–93 n. 16 also saw a reference to Didymus in P.Oxy. XVII 2076 col. i (Sappho), but the interpretation of the traces is more uncertain. 8 P.Paris 71 (CLGP I.1.2.1 Alcman 5) coll. ii.27, iv.4 as interpreted by Haslam ap. Römer 2013: 109. 9 Σ A O. 6.154a–b περὶ δὲ τῆς σκυτάλης καὶ ἐν τοῖς Ἀρχιλόχου ὑπομνήμασιν εἴρηται. The deduction of a Didymean commentary on Archilochus works if 1) the author of this anonymous note is Didymus and 2) εἴρηται indicates self-citation. Both of these things are quite possible: Didymus uses εἴρηται time and again for self-citation, so much so that Schmidt used it to assign authorship to anonymous fragments (1854: 243). Yet neither of them can be proved, and Didymus’ commentary on Archilochus remains conjectural. 10 CLGP I.1.4 Bacchylides 4. See Snell 1961: 50*. The doxography fits Didymus’ practice (see §2.1); the generic diatribe concluded by the citation of Dionysius of Phaselis parallels 166. 11 On the Περὶ λυρικῶν ποιητῶν see Grandolini 1999. On the Etymologica see Reitzenstein 1897 with his corrections in Reitzenstein 1907; Dickey 2007: 91–92, with bibliography; Alpers 2015. 12 Cohn 1903: 469; see also Grandolini 1999: 2–4. 13 On the Chrestomathy see Severyns 1938. Disagreement persists on whether its author is the fourth-century Neoplatonic philosopher or an earlier grammarian.

Didymus and lyric



23

The few secure fragments of this work are all concerned with the characteristics of poetic genres: elegy, the paean, the difference between the hymn and the prosodion. Hymns are called so because they are durable (ὑπόμονος) and cause remembrance (ὑπόμνησις); in a narrower sense, hymns differ from prosodia in that prosodia are sung to the aulos while processing towards (προσιόντες πρός) temples or altars, and hymns are sung to the kithara while stationary (μένω again?) (°347). Paeans are sung to implore a stop to some evil, deriving as they allegedly do from παύω (348). After a sizeable doxography of folk etymologies of ἔλεγος (from ἔλεος, from εὖ λέγειν, from ἒ ἒ λέγειν…) Didymus is credited with the fascinating theory that the rhythm of the pentameter after the hexameter in the elegiac distich imitates the halting breath of the dying (°345). This last point is not present in Proclus, but the rest of Proclus’ account of elegy shares several similarities with the doxography in the Etymologica, which accordingly may come from Didymus too. The parallels between Proclus and the Etymologica have raised the prospect that the entire lyric section of the Chrestomathy (24–92, or indeed 24–99 Severyns) may depend on Didymus. A repertory of every known lyric genre, down to the most obscure, sounds just like the sort of thing Didymus might write; we know from Σ Ap.Rhod. 1.972 Wendel that he also discussed the ioulos, a little-known song for Demeter (349; cf. PMG 849 ap. Ath. 14.618d), which would fit well here. Devoting three sections specifically to daphnephorics, tripodephorics, and oschophorics suggests someone with Pindaric interests (cf. frr. 6c–f, 66, 94b–c Maehler). Details differ, but the difference may be due to the vagaries of transmission. Proclus’ account of the tripodephoria (79–86 Severyns) mentions a procession from Boeotia to Dodona, while according to Ammonius, Didymus spoke of one to the Ismenion, near Thebes (°172). However, both Didymus and Strabo—who gives largely the same account as Proclus—quote Book 2 of Ephorus as a source, suggesting that their accounts may have originally been closer than they now seem. Proclus’ story, moreover, is also found in Zenobius, who had Didymus’ On Proverbs among his chief sources.14 The strong etymological slant of the securely attributed fragments is expected, given the sources that transmit them, but the role of etymology in Didymus’ work more broadly is greater than is often realized.15 It may seem surprising that Didymus’ etymology of σκόλιον should come from the Symposiaka instead (EM p. 718 Gaisford, °338), but the possibility of duplication ought not to be discounted, especially if the Symposiaka were a ‘best of’ of literary and historical zētēmata culled from his scholarly works and aimed at a somewhat wider public. But what else was there in the On Lyric Poets beyond genres and their etymologies? The title suggests a work concerned not only with lyric, but also with the lyricists as individuals. Schmidt must have assumed as much when he assigned to the treatise Didymus’ discussion of whether Theognis hailed from continental Megara or its Sicilian namesake (Σ Pl. Leg. 630a Greene, 346). Yet the same material occurs in Harpocration (θ 6 Keaney): did Didymus rather treat this topic in one of the hypomnēmata to the Attic orators, on which we know that Harpocration drew? Perhaps the one to Isocrates, who mentions Theognis in 2.43? But, again, duplication is a possibility, and (for instance) the disquisition about Sappho’s profession caricatured by Seneca (397) would fit well in an On Lyric Poets.16 Although the evidence for this aspect is scanty, the treatise may have combined biographical material with literary history and criticism, much like Aristotle’s Περὶ ποιητῶν and the several later treatises bearing that name.17 14 The clue is in the title: ἐπιτομὴ ἐκ τῶν Ταρραίου καὶ Διδύμου παροιμιῶν, ‘epitome from the Proverbs of Tarrhaeus and Didymus’. 15 The Etymologica preserve a large number of other fragments, perhaps ultimately deriving from the Περὶ παθῶν (On Modifications, sc. of words), containing more or less fantastical etymologies of a variety of words. It is unclear how much of it is genuine Didymus and how much is misinterpreted or misattributed. Unsurprisingly, this area of Didymus’ work is understudied (who wants to read the Etymologica?), but it would reward careful investigation. On paretymology in Didymus’ scholarship on Aristophanes see Benuzzi, this volume, 56–57. 16 Schmidt 1854: 384–85 n. 3 regarded the four fragments cited by Seneca as individual works, but such specialized monographs seem alien to what we know of Didymus’ writing habits, with the possible exception of the one on Solon’s axones (XLI). Of course it is possible that the discussion of whether Sappho was a hetaira came from a commentary to her poetry (cf. 103  396 with the discussion above), or perhaps to one of the comedies in which she was a character. 17 On Aristotle’s On Poets see Janko 2011: 410–83.

24



Prodi

2. PINDAR We are on firmer footing with Didymus’ role in Pindaric scholarship.18 Tradition has it that the canonical edition of Pindar’s poetry was made by Aristophanes of Byzantium (fr. 381 Slater).19 The scholia vetera relate some emendations by Aristophanes’ older contemporary Zenodotus; whether he produced an edition is doubtful.20 The most prominent Pindaric commentator before Didymus is Aristarchus, who is referenced sixty-eight times in the scholia—the second most cited scholar after Didymus himself (seventy-four).21 The scholia include a great deal of references to other, ‘minor’ figures. Two in particular, Didymus’ contemporaries Aristonicus and Theon, occur very sparsely in the scholia, but we know from other sources that they too commented on Pindar.22 As for Didymus, he certainly commented on the Epinicians (104–71) and the Paeans (°172),23 and he is cited in two fragments of P.Oxy. XXVI 2442, both incertae sedis (frr. 39 and 97, 173–74).24 It is a reasonable conjecture that he commented on the entire Pindaric corpus, as Aristarchus probably had done.

2.1. Doxography A long-standing critical trope maintains that Didymus’ hypomnēmata were ‘the chief, if not the sole, intermediary between his predecessors and his successors’.25 This is not quite true: the scholia incorporate at least one other commentary, which was not based on Didymus but drew on Hellenistic material, including some of Didymus’ own sources.26 Yet we owe a great deal not only to Didymus’ voracious reading, but also to his openness in citing his sources.27 The picture of him as a ‘mere compiler’ is reductive and misleading, but incorporating the scholarship of others was indeed a keystone of his method, with Pindar no less than with Demosthenes (°281).28 This is equally true of information drawn from other branches of learning—history first of all—and of earlier interpretations of Pindar. The doxographic scholia which punctuate the corpus are with all likelihood lifted straight from him. Out of forty-nine scholia in which Didymus’ opinion is cited together with one or more others, thirty-seven present his opinion at the end, capping the discussion; most of 18 See Deas 1931: 19–27; Irigoin 1952: 67–74; Carnevali 1980; Braswell 2011 and 2017. The last includes a critical edition, with translation and notes, of all the non-papyrological fragments of Didymus’ Pindaric scholarship. It is an indispensable tool, but the translation is frequently incorrect and needs to be handled with care. On Hellenistic scholarship on Pindar see Deas 1931: 1–27; Irigoin 1952: 31–75. On the Pindar scholia, beside Deas 1931, see Calvani Mariotti 1987; David et al. (eds) 2009 and 2015; Braswell 2012; and especially Bitto 2012: 63–240. 19 On the ascription see Slater 1986: 145–46 (against); D’Alessio 1997: 51–55; Negri 2004: 16–27 (in favour). 20 See Irigoin 1952: 32–33; Pfeiffer 1968: 117–18 (in favour); Ferrari 1992 (against). Ruffa 2001 is also relevant; see n. 55 below. 21 Aristarchus’ work on Pindar has been rather neglected in comparison to that on Homer; beside Irigoin 1952: 51–56 see Horn 1883; Feine 1883; Vassilaki 2009. 22 See respectively Razzetti 2000; P.Oxy. XXXI 2536 and Merro 2018. 23 The Pindar scholia transmit a further four fragments of a commentary on the Paeans introduced by εἴρηται ἐν Παιᾶσιν vel sim. without an author’s name. That they refer to a commentary, rather than to the Paeans themselves, is clear (Käppel 1992), but whether the commentary is by Didymus, as has been conjectured since Boeckh 1819: xvii, remains in doubt: Braswell 2017: 261 n. 352 considers the attribution ‘very likely’ but rightly classes them as dubia (frr. 69–72 Braswell). 24 P.Oxy. 2442 comprises not one but (at least) three manuscripts, preserving fragments of Pindar’s Hymns, Paeans, and Prosodia (Lobel 1961: 31; D’Alessio 1997: 35–37, 40–45). In the absence of external evidence it is impossible to establish to which of these three books each of the fragments belongs. 25 Deas 1931: 19. 26 See Prodi forthcoming. 27 This is not to say that interpretations are always attributed to their originators in the text as we have it: τινες and ἔνιοι, ‘some’, abound even in Didymean scholia (cf. e.g. Σ BD N. 7.1a, 158). It is tempting to attribute this moniker to careless scribes (cf. 111b, where Σ A O. 5.27b splits the pars destruens and the pars construens of Didymus’ interpretation and ascribes the former to him, the latter to τινες; contrast 111a), but Didymus used it too, albeit sparingly, in P.Berol. inv. 9780 recto (°281). Another set of unspecific labels is οἱ ὑπομνηματισταί/ ὑπομνηματισάμενοι/προϋπομνηματισάμενοι, ‘commentators’/‘previous commentators’. These occur six times between them, including in conjunction with Didymus (cf. 115a–b, 146; other occurrences could be Didymean too). The meaning seems to me to be the same for all three words, ‘all the commentators to date’ (or ‘all commentators to date except the one quoted individually’), but Calvani Mariotti 1999 gives a different interpretation of οἱ προϋπομνηματισάμενοι. See also Coward, this volume, 40–42 on similar words in the Sophocles scholia. 28 Deas 1931: 22–26.

Didymus and lyric



25

the other twelve, too, are informed by Didymus’ perspective.29 He seems to have wished to preserve and disseminate earlier scholarship for the benefit of those without access to it, much as he explicitly did Aristotle’s poem for Hermias in the Περὶ Δημοσθένους (P.Berol. inv. 9780 recto col. vi.20–22). Consider the literature review on the myth of Nemean 1. The ode celebrates Hieron’s henchman Chromios of Aetna. Its mythical section (vv. 33–72) narrates Heracles’ first heroic feat, the killing of the snakes that Hera had sent against him immediately after its birth, culminating in Tiresias’ prophecy of his immortalization. The unperspicuous connection between the myth and the honorand perplexed ancient critics. Σ BDPU N. 1.49c presents four solutions to the puzzle, each refuted in turn, and finally Didymus supplies his own interpretation (144). ‘Some’, Aristarchus related, thought that Pindar had been assigned that subject, which Aristarchus himself saw as a cop-out (ἀπίθανον, ‘unpersuasive’; fr. 58 Horn = 42 Feine). Or perhaps Pindar always praises those who have inborn excellence, like Heracles?30 ‘Unpersuasive’, again: if that were the point, why only mention this first adventure and not the other, more famous ones? Chaeris (fr. 20 Berndt) suggested that Chromios had toiled at Hieron’s side from the beginning and was rewarded with horse-rearing, just as Heracles had toiled and been rewarded with immortality and the hand of Hebe in marriage.31 In this case the objection is that the episode narrated does not illustrate Heracles’ willingness to toil as much as all the labours together would have done. Chrysippus (fr. 2 Braswell) claimed that Heracles was relevant to this Nemean victory qua slayer of the Nemean lion—of which, however, there is no trace in the poem.32 Didymus’ ‘better’ (βέλτιον) explanation is that, just as Heracles’ defeat of the snakes inaugurated a lifetime of even greater exploits, so Chromios’ first victory intimated more and better victories to come, with Pindar casting himself in Tiresias’ prophetic role. Didymus’ rebuttals of Aristarchus and Chaeris are underwhelming, yet his interpretation—right or wrong though it be—belongs to a sensitive reader of Pindar. He supports it by quoting a gnomic passage from Pythian 1 (33–34) where the poet connects the favourable beginning of a journey to its successful continuation. The maxim is appropriate not only generally, but particularly when viewed in its original context: there, too, it is instrumental to a wish for further victories for the city of Aetna after the one just won by her founder, Hieron. So Didymus builds on the explicit connection between that gnomic paradigm and a wish for further victories in Pythian 1 to tease out an implicit connection between a mythical instance of that same paradigm and a similar, but unstated, wish in Nemean 1. A similar doxography concerns why Nemean 7 begins with an address to E(i)leithyia, the goddess of childbirth.33 The victor Sogenes was obviously not a newborn, so why involve her? This time we have five explanations, each duly rebutted, capped by one ascribed to ‘Aristodemus the pupil of Aristarchus’ (BNJ 383 F14), which Didymus endorses (βέλτιον; Σ BD N. 7.1a, 158). Unsurprisingly given the methods of Hellenistic Pindarists, several of these explanations consist of inventing ‘facts’ outside the text which purportedly explain the passage. Didymus dismisses each of them by pointing to the lack of documentary proof. That Sogenes was too young to compete and someone else won in his place and gifted him the victory is αὐτοσχέδιον, ‘improvised’, as well as unlikely on other grounds.34 That his father was a priest of Eileithyia is ἀμάρτυρον, ‘unattested’. Was his house perhaps in the neighbourhood of a temple of the goddess?35 οὐδὲ τοῦτο δὲ 29 30 31 32 33 34

See Prodi forthcoming. The context suggests that this second, anonymous interpretation also belongs to Aristarchus (Horn 1883: 55). On Chaeris see Calvani Mariotti 2012. On this Chrysippus see Braswell 2015 with the corrections by Vecchiato 2018a and 2018b. See Fränkel 1961: 385–94, esp. 391–93; Young 1970: 35–37. As we know from Σ BD N. 7.56a (160), this was the opinion of Aristarchus (frr. 54–56 Feine), who used it to explain in one fell swoop both the invocation to Eileithyia at the beginning and the ‘digression’ on Neoptolemus later on in the ode (Neoptolemus being also the name of the putative ‘true’ victor). There, too, Didymus dismisses Aristarchus’ supposition and quotes one by Aristodemus which is equally wrong-headed (Fränkel 1961: 385–91). 35 A similar explanation of a different passage in the ode, again based on the asserted vicinity of the victor’s house to a temple, is attributed to Callistratus by Σ BD N. 7.150a.

26



Prodi

ἱστορεῖται, ‘there is no testimony of this, either’. We may think no better of Aristodemus’ explanation that Sogenes was a late, long-wanted son and his father was especially grateful to Eileithyia for his birth; the epigram of Simonides which allegedly proved it does not, in fact, do anything of the kind.36 Yet the epigram was there, and this must have satisfied Didymus. Two things are noteworthy: the dutiful inclusion of earlier solutions to the zētēma, even though he regarded them as false, and the unwillingness to countenance extrinsic interpretations which rested on no evidentiary basis. Examples of the first are everywhere in the corpus: Didymus’ opinion occurs on its own in less than a third of the scholia that cite him. The second is of a piece with his historical interests, in the broadest possible sense of the adjective.

2.2. History The extensive remains of the Περὶ Δημοσθένους (°281) are almost entirely historical in content; the absence of the fine-grained literary interpretations required by a lyric poet create a much starker imbalance in the commentary on the orator. Yet Didymus’ reliance on documented history, both as a fuel for positive explanation and as a brake on runaway fantasy, plays an essential role in his Pindaric scholarship too.37 Historicizing readings of Pindar were not a Didymean innovation: as we have just seen, earlier Pindarists, too, felt the need to read the poetry against the context with which it is so clearly intertwined. Any innovation rather consists of a shift from inference to documentation, putting his vast reading to (sometimes) good use. Scholarship on comedy, with its long-established tradition of works on κωμῳδούμενοι, may have shown him the way. A list of Didymus’ quotations from the historians in the Pindar scholia can be found in Fausto Montana’s chapter.38 They occur in almost a third of the fragments and comprise local historians, antiquarians, and more universal historiography. The prevalence of Timaeus tallies with the Sicilian focus of many of the Epinicians. Additionally, the scholia transmit a large number of historiographical quotations without the name of the Pindarist who brought them to bear. One suspects that many of them may come from Didymus, too:39 tellingly, only twice in the scholia does a historian’s name occur near that of a Pindarist other than Didymus,40 and in neither case is it clear that the historian was cited by that Pindarist. Sometimes Didymus refers to history in order to provide background information on a victor, or on the circumstances surrounding an ode. Hieron was still Hieron of Syracuse, not Hieron of Aetna, when he won the Olympics in 476 BC (Apollodorus BNJ 244 F69: Σ HQ O. 1.35c, 104), and he was a hereditary priest of Demeter (Philistus BNJ 556 F49 and Timaeus BNJ 566 F96: Σ A O. 6.158c, 114). Karrhotos, mentioned at P. 5.26, was a military ally of the victor Arkesilas rather than his charioteer (Theotimos BNJ 470 F1: Σ BDEGQ P. 5.34, 132). Chromios of Syracuse became Chromios of Aetna when his boss Hieron did the same after ‘founding’ that city (inscr. a BDPTU N. 1, 140). Such remarks do not only provide context, but sometimes they elucidate specific parts of the text: Hieron’s priesthood justifies the phrase καθαρῷ σκάπτῳ, ‘with a pure sceptre’ (O. 6.93), and Chromios’ change of city explains why the title of the ode calls him Αἰτναῖος.41 36 Young 1970: 637. 37 The same is true of his work on comedy: see Benuzzi, this volume, 57–60. On Didymus and history generally see Montana, this volume. 38 This volume, 70–71. On Didymus’ use of history in his Pindaric scholarship see also Deas 1931: 22–23; Irigoin 1952: 71; Braswell 2011: 182–87; Braswell 2012: 13–18; Braswell 2017: 113–16. On historicizing interpretations in the Pindar scholia see Bitto 2012: 220–23; Phillips 2020 (449–50 on Didymus). 39 Cf. Irigoin 1952: 72–73. 40 Hellanicus and Aristodemus, Σ A O. 3.22a; Polemon and Aristarchus, Σ BCDEQ O. 7.95a. 41 Despite not going back to Pindar, titles are transmitted together with the text and are often commented on by the scholia: see Prodi 2020: 463–67. Indeed, the headnotes that precede every ode, containing information on the victor and his victory, are effectively scholia on the respective title (inscriptio), as Drachmann recognized.

Didymus and lyric



27

Realien and their sources, too, are often called upon to elucidate an otherwise obscure element in Pindar’s text, be it the referent of the cult epithet Orthosia at O. 3.30 (Artemis, citing Apollodorus BNJ 244 F127: Σ A O. 3.54a, 109), the location of Phaisana (allegedly in Elis not in Arcadia, citing Istros BNJ 334 F41: Σ A O. 6.55a, 112a), the identification of the ‘local games of Hera’ at P. 8.79 (in Argos: Σ BDEGQ P. 8.113c, 136) and of those ‘of deep-bosomed Earth’ at P. 9.101–02 (in Athens: Σ DEGQ P. 9.177, BDEGQ 178, 137a–b), or the exact name of the Theban games in honour of Heracles (Herakleia not Iolaeia, citing ὁ περὶ ἀγώνων ἀναγραψάμενος: Σ BDP N. 4.32, 151). An instructively more complex case involves a passage in Olympian 5 (vv. 10–14). The victor Psaumis is said to sing (ἀείδει) various landmarks around his city Camarina, including the streams with which the river Hipparis ‘waters the people and quickly attaches a high-limbed grove of sturdy halls (κολλᾷ τε σταδίων θαλάμων ταχέως ὑψίγυιον ἄλσος), bringing from helplessness to light the people of this city’.42 But what does all this mean? A scholion relates two possible answers (Σ A O. 5.27b, 111b): Ἀρίσταρχος παριέναι φησὶ τὸν Ἵππαριν τὴν πόλιν καὶ προσχωννύντα ἰλὺν ἀναπλάσσειν αὐτῇ γῆν, καθάπερ τὸν Ἀχελῷον ταῖς Ἐχινάσι νήσοις καὶ τὸν Νεῖλον τῇ Αἰγύπτῳ· κολλᾷ οὖν ἀντὶ τοῦ προσαναπλάσσει γῆν. τοῦτο δέ φησιν ὁ Δίδυμος ἀμάρτυρον εἶναι· οὐ γὰρ ἱστορεῖται περὶ τὸν Ἵππαριν καὶ τὴν Καμάριναν τοῦτο γενόμενον· οὐ γὰρ πλημμύρᾳ χρῆται ὁ ποταμὸς καταφερόμενος. βέλτιον οὖν οὕτως τινές· ἐπειδὴ ἐξ ὑπογυίου συνῴκισται ἡ Καμάρινα, καθὰ καὶ νέοικον αὐτὴν προσηγόρευσε, γειτνιᾷ δὲ Ἵππαρις αὐτῇ, κατὰ τοῦτο φάναι ἐπὶ τοῦ συνοικισμοῦ κολλᾶσθαι ὑπ’ αὐτοῦ σταδίων θαλάμων ταχέως ὑψίπυργον ἄλσος, ἐπεὶ τὰ ἐπιτήδεια πρὸς τὴν οἰκοδομὴν διὰ τοῦ κατάπλου προσεκομίζετο ὥστε ποιεῖν σταδίους θαλάμους. Aristarchus says that the Hipparis passes by the city and heaps up earth on it by depositing silt, just like the Achelous in the islands Echinades and the Nile in Egypt: ‘attaches’ stands for ‘heaps up earth onto’. But Didymus says this is unattested, for there is no record of this event concerning the Hipparis and Camarina; the river does not flood as it flows seaward. So, some (explain the passage) better in this way: since Camarina was founded recently—the reason why (Pindar) also called it ‘newly founded’— and the Hipparis is adjacent (to it), the reason why he said that upon its foundation it ‘quickly attached a high-limbed grove of sturdy halls’ is that it carried to (the city) with its downward stream the materials useful for the construction work so as to build sturdy halls. Aristarchus (fr. 17 Horn = 16 Feine) maintained that the river ‘attached’ the ‘halls’ because it flooded and deposited silt, which the locals used for brick-making, as our scholion’s twin in the Vatican recension makes clear (Σ CDEHQ O. 5.20e, 111a);43 Didymus objected because of a lack of evidence. As far as we can judge, there is also no evidence in favour of his own explanation (we know that it is his from 111a), which indeed is hard to square with the letter of the text. Yet it avoids postulating a physical phenomenon for which one would expect some tangible evidence to have existed, and instead it makes reference to a specific and welldocumented historical event, the refoundation of Camarina after its destruction by Gelon in the early years of the fifth century.44

42 For clarity’s sake I translate the text the way it was understood by ancient commentators, with the Hipparis as the subject of κολλᾷ. For more recent views see Mader 1990: 81–82, 85–86 (the subject is Psaumis); Lomiento in Gentili et al. 2013: 442–43 (the subject is the river). On the scholia to this passage see also Brunel 1971; Carnevali 1980: 12–15; Lomiento 2006; Vassilaki 2009: 139–44. 43 The scholia to O. 2–12 are preserved in two ‘recensions’: the Ambrosian (consisting of MS A, in Milan’s Biblioteca Ambrosiana) and the Vatican (found in several manuscripts and named after MS B, in the Vatican Library). See Drachmann 1913: x–xiii; Deas 1931: 50–65. 44 These events are expanded upon in Σ CDEHQ O. 5.16, A 19a, CDEHQ 19b, ACDEHQ 19c, A 19d, with quotations from Timaeus and Philistus, which accordingly may also be Didymean (for Σ 19a–c see already Irigoin 1952: 72).

28



Prodi

Didymus’ history-conscious correction of an autoschediastic inference by Aristarchus is nothing new. We already saw one example (158), and there are more. Some scholia explicitly remark his greater adherence to historical fact (τὸ ἀκριβέστερον τῆς ἱστορίας ἐκτίθεται: Σ BCEHQ O. 2.29d, 105b; ἱστορικώτερον λέγει: Σ BCDEQ O. 3.1d, 108b).45 Yet often Didymus only corrects an unattested element in Aristarchus’ interpretation while retaining its fundamentals.46 111 is itself an example: Didymus does away with the floods and the bricks, but still maintains that the subject is the river, that the indirect object is the city, and that the sentence refers to the conveyance of building materials. Another example is 108b just cited. There, the invocation to the Dioscuri that opens Olympian 3 for Theron of Akragas is explained in terms of the particular devotion to the Dioscuri not in Akragas itself, as Aristarchus had surmised (fr. 12 Horn = 11 Feine), but in Theron’s ancestral homeland, Argos. Its Ambrosian twin Σ A O. 3.1a (108a) is quite right to state that ‘Didymus rather inclines towards Aristarchus’ explanation’: less-attested devotion in Akragas is replaced with better-attested devotion in Argos, but the trajectory of the two interpretations is the same.47 Whatever the exact sense of the Suda’s claim that Didymus was a γραμματικὸς Ἀριστάρχειος (Suda δ 872 Adler, T1), his predecessor certainly cast a long shadow on his work. The very notion of reading Pindar historically develops an Aristarchan precedent, albeit on a very different foundation.48

2.3. Textual Criticism As far as the scholia go, Didymus’ forays into textual criticism were relatively rare. We know of only four conjectures by him; in a few other instances he defended the transmitted text, or a transmitted variant. Characteristically, one emendation stems from a misplaced application of historical knowledge. At N. 6.31 Pindar mentions an Aeginetan clan, the Bassidai. Since (we infer) there was no other record of this clan, and since on the other hand Pythaenetus’ Aiginetika mentioned a local hero named Boudion (BNJ 299 F 2a), Didymus suggested changing Βασσίδαισιν into Βουδίδαισιν, the hero’s putative descendants (Σ BD N. 6.53a, 157). The rest of the scholia to this passage wisely ignore the conjecture (Σ BD N. 6.53b–e, 54a). Two small emendations are more felicitous. At N. 4.59 Pindar says that Akastos tried to bring about Peleus’ death ‘with the knife of Daedalus’, τᾷ Δαιδάλου δὲ μαχαίρᾳ—a unique expression. One anonymous commentator attempted to explain the transmitted text as a metaphor meaning a deadly ruse (δόλος), such as the one Daedalus visited on Minos through the daughters of Kokalos. Didymus had other ideas (Σ BDP N. 4.95b–c, 152): Δίδυμος δέ φησι δεῖν γράφειν διὰ τοῦ ῳ· δαιδάλῳ δὲ μαχαίρᾳ δόλον ἤρτυσε τῷ Πηλεῖ, παρελόμενος αὐτοῦ κρύφα, ἵνα χωρὶς ἀμυντηρίου ἁλοὺς ὑπὸ τῶν Κενταύρων φθαρῇ. ταῦτα δὲ ἱστοροῦσι πολλοὶ μέν, ἀτὰρ δὴ καὶ Ἡσίοδος λέγων οὕτως (fr. 209 Merkelbach–West)· ἥδε δέ οἱ κατὰ θυμὸν ἀρίστη φαίνετο βουλή· αὐτὸν μὲν σχέσθαι, κρύψαι δ’ ἀδόκητα μάχαιραν καλὴν, ἥν οἱ ἔτευξε περικλυτὸς Ἀμφιγυήεις· ὡς τὴν μαστεύων οἶος κατὰ Πήλιον αἰπύ αἶψ’ ὑπὸ Κενταύροισιν ὀρεσκῴοισι δαμείη.

45 46 47 48

On ἱστορία and its cognates in the Pindar scholia see Calvani 2006; Vassilaki 2015. Deas 1931: 20–21. See Vassilaki 2009: 137–39. On the cult of the Dioscuri in Argos see Pae. 18.1–2 and Moretti 1998: 237–39. Vassilaki 2009, largely followed by Phillips 2020: 441–47, seeks to defend Aristarchus from the charge of ahistoricity levelled by Wilamowitz 1889–95: I 156 and often repeated. They are right that Aristarchus was not uninterested in history and often proffered broadly historicizing interpretations, yet a historicizing interpretation is not the same thing as an interpretation based on documented historical fact, and Aristarchus had plenty more of the former than of the latter.

Didymus and lyric



29

δαίδαλον δὲ εἶπε τὴν μάχαιραν διὰ τὸ ὑπὸ Ἡφαίστου κατεσκευάσθαι. ἐπιεικῶς δὲ τὰ Ἡφαίστου ἔργα δαίδαλά φησι (Il. 18.482)· ποίει δαίδαλα πολλὰ ἰδυίῃσι πραπίδεσσιν. But Didymus says it should be written with ῳ: ‘with an ornamented knife he laid a ruse for Peleus, taking it from him covertly so that he would be caught by the Centaurs without help and would perish’. This story is related by many, including Hesiod, who says as follows: This seemed to him the best plan in his heart: hold back, but stealthily conceal the sword, beautiful, which the famous Cripple made him, so while alone he sought it on steep Pelion the mountain-lurking Centaurs would soon fell him. (Pindar) called the knife ‘ornamented’ because it had been made by Hephaestus. He fairly calls Hephaestus’ works ‘ornamented’: much finery he made with skilful mind. The passage from the Catalogue of Women gave Didymus a fuller narrative of the episode which Pindar simply alluded to.49 There was a real weapon, which belonged to Peleus, and which Akastos had hidden in the moment of need; it was the work of Hephaestus (Ἀμφιγυήεις, v. 3), not of Daedalus; it was ‘beautiful’, καλή —that is, plausibly, ornamented, i.e. δαίδαλος. He therefore corrected Δαιδάλου to δαιδάλῳ, eliminating the oddity and aligning Pindar with Hesiod.50 The Homeric verse strengthened the emendation by confirming that Hephaestus’ works could indeed be called δαίδαλα. An epic parallel is also key to Didymus’ intervention on N. 10.62 (Σ BD N. 10.114a, 165).51 The transmitted text has Lynceus espy the Dioscuri from Mount Taygetos while he was sitting (ἥμενος) in an oak’s hollow trunk, for (γάρ) he had the keenest sight on earth. Surely not: the proof of preternatural sight is to spot someone from afar when they are hidden, and the Cypria (fr. 15 Bernabé) confirm that the oak contained the Dioscuri, not Lynceus. So Aristarchus changed ἥμενος to the accusative ἥμενον, indicating Castor, for Castor alone was apparently indicated by Apollodorus, BNJ 224 F148 (cf. Bibl. 3.1.11; fr. 65 Horn = 59 Feine). Didymus objected that the Cypria located both brothers, not only one, in the fateful tree. Accordingly, he proposed to retain the letters of the transmitted text and simply move the accent forward: ἡμένος, taken as accusative plural (ἀντὶ τοῦ ἡμένους) on the parallel of the use of ἀελλόπος, τρίπος, ἕδος as accusatives plural. The analogy does not work linguistically, but the sense is the right one: all recent editors

49 Didymus’ anonymous predecessor may also have had the Hesiodic fragment in mind, or at least the well-known fact that Peleus owned a sword made by Hephaestus: his ‘absurd explanation’ (Henry 2005: 41) presupposes an attempt to avoid the natural interpretation of Pindar’s words, viz. an actual weapon made by Daedalus. 50 Modern commentators are divided. Like Snell–Maehler (who do not even cite Didymus’ suggestion in the apparatus), Willcock 1995, Henry 2005, and Cannatà Fera 2020 retain Δαιδάλου; like Turyn, Braswell 2017: 215, 218 favours the emendation. 51 On Didymus’ use of quotations from earlier poets see Braswell 2011: 188–91 and 2017: 116–18. His use of Hellenistic poetry still awaits investigation, but Callimachus’ presence in the Pindar scholia more generally is treated by Phillips 2013. On attitudes towards mythological parallels and variants in the scholia see Bitto 2012: 117–23.

30



Prodi

print ἡμένους. Once again Didymus takes the lead from Aristarchus both in the interpretation of the passage and for the type of emendation he proposes.52 The emendation is minimal, if it is one at all (Didymus only speaks of ἀναγιγνώσκειν: ‘read’ in the literal, not text-critical sense). Similarly, Σ BD N. 4.151a (153) quotes him as saying that at N. 4.93 ‘it is also possible to read’ οἷον, ‘how’ (exclamatory, θαυμαστικῶς), rather than οἶον, ‘only’, favoured by Aristarchus (fr. 68 Horn = 51 Feine). He must be right, yet he displays his characteristic tentativeness. μήποτε, ‘perhaps’, occurs so frequently that some past scholars have used it as a fingerprint of sorts to assign anonymous comments to him.53 In four cases he is on record as defending the transmitted text, sometimes rightly (πόσις against Aristarchus’ unmetrical πόσιος at O. 2.77: Σ A O. 2.140a, 107; perhaps ναίων at P. 7.6: Σ BDEGQ P. 7.6a, 135), sometimes probably not (Ἆλιν against Aristodemus’ Ἄλτιν at O. 10.45: Σ BCDEQ O. 10.55c, 123; μόλεν against μόλον at N. 7.34: Σ BD N. 7.47, 159). When faced with a variant ἁ τραχεῖα πόλις for Ἀτρέκεια πόλιν at O. 10.13 he opined that it, too, ‘makes sense’ (ἔχειν λόγον: Σ Α O. 10.17c, 122), but apparently he stayed shy of endorsing it—rightly so, because it is unmetrical.54 So the picture of Didymus as a textual critic is a mixed one, both in method and in results. Generally he seems unkeen on altering the transmitted text, choosing instead to interpret it when possible and otherwise to amend with a light touch; yet in one case he proposed a major, unwarranted emendation when Pythaenetus sent him off track.

2.4. Corpus and Performance Didymus was also interested in the constitution of the collection. Firstly, he seems to have defended the authenticity of Olympian 5. Drachmann, followed by Braswell, prints inscr. a thus (°110): αὕτη ἡ ᾠδὴ ἐν μὲν τοῖς ἐδαφίοις οὐκ ἦν, ἐν δὲ τοῖς Διδύμου ὑπομνήμασιν ἐλέγετο Πινδάρου. This ode was not (by Pindar) in the base texts, but in Didymus’ commentaries it was said to be by Pindar. —following the Vatican recension (BDEHQ); Α has a mundane περὶ αὐτῆς τάδε in place of Πινδάρου, presumably referring to inscr. b which follows. If the Vatican text is correct, Didymus defended the ode against the charge of spuriousness made in the ἐδάφια, the ‘base text’ traditionally ascribed to Aristophanes of Byzantium.55 Elsewhere he decries that Nemean 11 was ‘shoved into’ the Epinicians (συνεῶσθαι: inscr. a BD, 166). While none of the last three Nemeans honours a Nemean victory, as ancient commentators knew (inscr. BD N. 9), the last one is unique in not being a victory ode altogether. Didymus pointed out, correctly, that it was 52 Aristarchus had similarly corrected ἐσλός to the required ἐσλούς at N. 1.24, recognizing the transmitted form as a residue of the ‘old spelling’: Σ BDPU N. 1.34b (fr. 56 Horn = 40 Feine). 53 See Benuzzi, this volume, 53 and n. 23. On Didymus’ similar use of the potential optative in the Περὶ Δημοσθένους see Gibson 2002: 30–32. 54 See Braswell 2017: 164. As he remarks, ‘in the extant fragments of Didymos’ commentary discussion of metre is entirely absent’ (p. 124). Whether this betrays ‘his ineptitude in that field’ (ibid.) I am less certain: outside of the scholia metrica there is very little metrical discussion altogether in the scholia. 55 I follow Ruffa 2001: οὐκ ἦν means not that the ode ‘was not there’ but that it ‘was marked as spurious’ (the sense the phrase has in the Homeric scholia); the ἐδάφια are thus likelier to be Aristophanes’ canonical edition than a putative earlier one by Zenodotus. Ferrari 2006 instead defends A’s text. If he is right and the scholion presents inscr. b as a verbatim quotation from Didymus, it could be used as circumstantial evidence to assign to Didymus the short headnotes, with the identification of the victor and the date of the victory, that introduce the scholia to almost every ode. Such a conclusion would be consistent with Didymus’ historical interests. Not that the quotation would have to end there: the bit that follows suit in A (inscr. c, connected by δέ), citing Polemon of Ilium’s account of the ἀπήνη (fr. 20 Capel Badino = 23 Preller) and offering an improbable etymology of that word, would be at home in Didymus’ commentary (so Capel Badino 2018: 253–54).

Didymus and lyric



31

composed for Aristagoras’ inauguration as councillor (citing vv. 9–10), not to celebrate the earlier, minor athletic victories mentioned at vv. 19–21. More perplexingly, he added, following Dionysius of Phaselis, that it should have been included in the Drinking Songs (Παροίνια) instead.56 Didymus’ interest in the occasion of a song transpires also from a remark about N. 1.6, where Pindar mentions ‘Aetnaean Zeus’. Apparently Hieron and his entourage used to celebrate victories in the crown games with epinicians performed at the festival of Aetnaean Zeus; Didymus therefore suggests that plausibly (πιθανόν) Nemean 1 too was composed for performance at that festival (Σ BDU N. 1.7b, 141). Didymus’ concern for genre tallies with that shown by the Περὶ λυρικῶν ποιητῶν, and an interest in ritual performance contexts is also apparent from his discussion of the tripodephoric procession in the commentary on the Paeans (°172) and of the daphnephoric, tripodephoric, and oschophoric songs in Proclus’ Chrestomathy, if that part of it goes back to him (see §1).57

2.5. Interpretation For all the importance of history and context, much of Didymus’ Pindaric exegesis consists of down-to-earth interpretations of individual passages. Sometimes he comments on the meaning of a word or its exact connotations. At N. 1.25 he suggests that τέχναι means not ‘skills’ but ‘guiles’ (δόλους) and μάρνασθαι not ‘fight’ but ‘act’ (ἐνεργεῖν) (Σ BDPU N. 1.36, 38, 142, 143). At N. 5.6 he remarks, contrary to an unnamed predecessor, that the ὀπώρα to which Pindar compares the young victor’s incipient beard is not to be understood as ‘fruit’ but as the season bearing that name, the latter part of the summer (Σ BD N. 5.10a, 155). Commenting on O. 9.22, μαλεραῖς ἐπιφλέγων ἀοιδαῖς, he improbably maintains that the songs are not ‘flaming’—in keeping with the image of ἐπιφλέγων, ‘setting alight’, with the Homeric use of μαλερός, and with Pindar’s frequent equation of song with fire and light—but ‘gentle’ (ἀντὶ τοῦ μαλακαῖς), drawing a parallel with I. 2.8, μαλακόφωνοι ἀοιδαί (Σ O. 9.34c, 120). We already saw Didymus use a Pindaric parallel for elucidation in 144 (§2.1). In some other cases he buttresses his interpretations with references to Pindar’s usage more generally. Commenting N. 7.61, σκοτεινὸν ἀπέχων ψόγον, ‘keeping dark censure away’, he suggests that the poet is the subject, not the object, of the negated censure: ‘I will blame no-one’, not ‘no-one will blame me’ (Σ BD N. 7.89b, 161); he adds that ‘this is not alien to Pindar, since he makes many other statements of this sort’ (τοῦτο οὐκ ἀλλότριον τοῦ Πινδάρου, ἀλλὰ γὰρ πολλὰ αὐτῷ τοιαῦτα εἴρηται). Remarkably, when it came to the occurrence of the same motif at I. 1.43–45 he gave the opposite interpretation (Σ BD I. 1.60, 169).58 Or take I. 2.12, οὐκ ἀγνῶτ᾽ ἀείδω. When arguing that ἀγνῶτ᾽ is the elision of ἀγνῶτι, ‘to someone unknowing’, not of ἄγνωτα, ‘things unknown’, Didymus refutes the opposite claim as follows (Σ BD I. 2.19a, 171): ὁ δὲ Δίδυμος μειοῦσθαί φησι τὸ ἀξίωμα τοῦ ποιητοῦ, εἰ μηδέπω τοῦ ἐπινίκου συντεταγμένου γνώριμός ἐστιν ἡ τοῦ Ξενοκράτους νίκη· ἔμπαλιν γὰρ ὁ Πίνδαρος ἀφανεῖς καὶ ἀδόξους φησὶν εἶναι τὰς νίκας, εἰς ἃς αὐτὸς μηδὲν γέγραφε. Didymus, however, says that the poet’s dignity is diminished if Xenocrates’ victory is renowned before the epinician is composed. On the contrary, Pindar says that the victories for which he has not written anything are unseen and unrenowned. Didymus is right on the general facts, for Pindar does often make fame dependent on his songs (O. 10.91– 93, N. 7.14–16, etc.), but turning this trope into a rule ignores the rhetorical flexibility of his poetry and imposes on it an ideological consistency which it does not possess. 56 Discussed by D’Alessio 1997: 54–55 and 2000; Schröder 1999: 146 and n. 2. 57 On issues of corpus and genre in the Pindar scholia more broadly see Bitto 2012: 195–98. 58 On the construal of this scholion see Prodi 2014 contra Braswell 2017.

32



Prodi

There is another interesting case in which Didymus discusses what makes, and what does not make, praise. The passage is one of the most textually disputed in the Pindaric corpus, N. 7.30–37.59 In the (probably incorrect) ancient interpretation, Pindar says that after the Trojan War Neoptolemus reached Delphi and died there τεθνακότων βοαθόων, ‘after the helpers had died’. But who were these ‘helpers’? (Σ BD N. 7.47, 159): ποίων δὲ βοηθῶν; τῶν περὶ τὸν Εὐρύπυλον, οὓς αὐτὸς ὁ Νεοπτόλεμος ἀνελὼν ἐπόρθησε τὴν Ἴλιον. οὕτω γὰρ ἴδιος ὁ πόνος ἔσται τοῦ ἥρωος. […] ἐὰν δὲ ἐπὶ τῶν περὶ τὸν Ἕκτορα, καθὼς Ἀρίσταρχός φησιν, ἀναδράμωμεν, πρῶτον μὲν μακρόθεν ἔσται τὴν πόρθησιν συνάπτων, δεύτερον δὲ ἐπὶ τὴν κοινότητα μεταβησόμεθα, δι’ ἧς τὸ ἐγκώμιον οὐκ ἔσται. Which helpers? Eurypylus and his companions: Neoptolemus himself killed them and sacked Troy. In this way the labour will be the hero’s own. […] If instead we go back to Hector and his companions, as Aristarchus argues, firstly the connection with the sack will be distant, and secondly we will switch to generality, through which there will be no praise. In order to establish the helpers’ identity, Didymus appeals a priori to the workings of ἐγκώμιον, ‘praise’. In order to be effective, praise must concentrate on something specific (ἴδιος) to the individual praised; if instead its referent is common to many, no praise of the individual ensues. Pindar means to praise Neoptolemus, ergo the ‘dead helpers’ must be something specific to him, whence their identification with Troy’s ‘helper’ Eurypylus.60 A different preoccupation with the general and the particular underlies Didymus’ criticism of Aristarchus and Chaeris in Σ BDPU N. 1.49c (144; §2.1 above). Both scholars had sought to draw a general parallel between Heracles and the victor, respectively on account of his inborn excellence and of his φιλοπονία, ‘love of labour’. Didymus’ objection to both is similar: if Pindar had wanted to draw on a permanent characteristic of Heracles’ greatness, he would have told of all the labours together, rather than select one, minor episode. The scholion to Nemean 7 just examined combines an appeal to first principles with sensitiveness to context and the train of thought of the passage: Didymus goes on to note that the mythical episode of Neoptolemus’ own death illustrates the immediately preceding gnōmē that death comes equally to those who are or are not expecting it (vv. 30–31).61 Another example of a contextually sensitive explanation addresses the list of Theaios’ victories at N. 10.25–28, where Pindar mentions a Pythian victory, three crowns won ‘at the gates of the sea’, i.e. at the Isthmus, and three ‘on the holy ground according to Adrastus’ institution’ (σεμνοῖς δαπέδοις ἐν Ἀδραστείῳ νόμῳ, 28). Adrastus was a mythical king of Sicyon, so one could easily have assumed that the games of v. 28 are the Pythia celebrated there, for which Pindar composed Nemean 9 (the ode immediately preceding this one in the canonical edition). However, Didymus realized the context is a list of ‘periodic’ victories—those obtained in the Panhellenic ‘circuit’ comprising the Olympian, Pythian, Isthmian, and Nemean games—which requires the reference to be to the games in Sicyon’s neighbour Nemea rather than the local Pythia (Σ BD N. 9.49b, 164).62 In the same scholion, Didymus also discerned correctly that vv. 29–33, immediately after our passage, are a wish for an Olympic victory, completing the περίοδος. This conceit, too, is drawn from Pindar’s usage 59 See Most 1986; Braswell 2017: 229–33; Cannatà Fera 2020: 448–49. 60 The assumption that the mythical narrative of N. 7 is ‘praise’ of Neoptolemus is also noteworthy. It tallies with the insistence on the poet’s role elsewhere in the poem: vv. 61–63 (cf. Didymus’ interpretation in 161, discussed above), 64–65, 68–69, and especially 102–04 (where the doxography in Σ BD N. 7.150a could well be Didymean). 61 This is one way of taking ἀδόκητον and δοκέοντα, the other (and more plausible) being ‘obscure’ and ‘famous’. The controversy is already in Σ BD N. 7.44a. 62 For a similar case see Σ BD I. 1.67 (170) with Braswell 2017: 253. Didymus’ attempted defence of the variant ἁ τραχεῖα at O. 10.17 (122) likewise relies on presupposing continuity of sense with the next clause (Carnevali 1980: 7).

Didymus and lyric



33

(P. 5.124, I. 1.64–67, 6.7–9, etc.). Didymus also deployed it to interpret the word προκώμιον, ‘prelude’, at N. 4.11: in his view, by using that word Pindar cast the ode as the prelude of hoped-for further epinicians, implicitly expressing a wish for further victories (Σ BDP N. 4.14a, 150). As we saw, this idea also stands behind the interpretation of the myth of Heracles in Nemean 1 as portending future victories (§2.1); in that context Didymus quoted P. 1.33–34, another parallel passage.63 In Didymus’ hypomnēmata there was room also for interpretations of a more literary bent. To name only one, he recognized that the images Pindar uses in his praise of the wrestling trainer Melesias at the end of Nemean 4 are drawn, appropriately, from wrestling (ἀπὸ τῶν παλαιόντων δὲ πάλιν ἡ μεταφορά, καὶ τροπικαὶ αἱ λέξεις ἀπὸ τῆς ἀθλήσεως) (Σ BD N. 4.153, 154; cf. Σ BD 151a, 153).64 Nor was the poet immune from criticism: in Didymus’ estimation, Apollo’s laughter at the ‘upright arrogance’ of the asses being sacrificed to him at P. 10.36 (a reference to an erection) is μετὰ τοῦ γελοίου καὶ ἄσεμνα, ‘ridiculous and unseemly’ (Σ BDEGQ P. 10.51b, 139).65 3. CONCLUSIONS Didymus is a central node in scholarship on Greek lyric: not only as a transmitter of earlier scholarship, crucial though that role was, but also as an exegete in his own right. No ancient scholar that we know of contributed so much, and of such value, to the interpretation of Pindar, whether in content or in method. His vast knowledge of historical, geographical, and antiquarian literature enabled him to build historicizing readings on a much firmer foundation than had previously been done. Glitches abound, whether caused by uncritical reliance on sources or by insufficient willingness to depart from precedent; yet the principle—finding things out rather than making them up—is fundamentally sound. His frequent caution and openness to multiple possibilities are noteworthy. As a textual critic, Didymus gave his best with small-scale emendations (an ending, an accent, a breathing), but he could be misled—again—by excessive reliance on sources or by an unwillingness to reject the transmitted text. When it comes to literary criticism, Didymus shows himself to be a careful and sensitive reader of Pindar, one who noticed patterns and recurring ideas and could retrace the poet’s thinking while keeping in mind the background constituted by pre- (and post-)Pindaric Greek poetry. He could be spectacularly wrong on occasion: Σ B O. 13.27a, BCEQ 27d (126) really is ‘truly Pythonesque’, as his most recent editor put it.66 Yet when compared with his fellow ancient Pindarists he is second to none, not even to Aristarchus. The study of Pindar has made long strides since his times, but little of it would have been possible without his work and his example.

63 64 65 66

Remarkably, no such interpretation is made explicit in the scholia to Pythian 1. Σ DEFGQ P. 1.67, 69 come closest. Σ BD N. 7.150a (which we suggested above, n. 60, as potentially Didymean) also concludes by identifying the domain of a metaphor. See Braswell 2011: 195–96. On criticisms of Pindar in the scholia see Cannatà Fera 2018. Braswell 2017: 124.

Bulletin of the Institute of Classical Studies, 2020, 63, 34–50 doi: 10.1093/bics/qbaa016 Original article

Chapter Three Didymus on Attic tragedy* Thomas R. P. Coward Ca’ Foscari University, Venice, Italy

ABSTRACT This chapter examines Didymus’ studies on Greek tragedy, in particular on Sophocles and Ion of Chios. It demonstrates Didymus’ methods of exegesis and his use of other exegetical corpora. It explains the thinking behind some of his choices and preferences, including the mistakes. Furthermore, it examines the problems of identifying Didymean material where Didymus is not named.

1. INTRODUCTION This chapter examines Didymus’ contribution to the study of Greek tragedy. His name appears nineteen times in the scholia to Euripides and in a subscriptio at the end of the Medea that seems to indicate a commentary by Didymus.1 His commentary on Sophocles is attested, and he also concerned himself with Ion of Chios, and perhaps Achaeus.2 Didymus was also the author of glossaries to both comedy (6–*38) and tragedy (°39–*49), which overlapped with the commentaries and influenced one another, and in other works he quoted dramatic authors.3 This chapter examines several examples of his works on Greek tragedy, in particular those of Sophocles and Ion of Chios, as these fragments have been understudied in comparison to those of Euripides, and also considers the overlaps between these studies and Didymus’ other works.4 It demonstrates Didymus’ methods of exegesis and explains the methodological framework behind some of his

* I wish to thank Enrico Emanuele Prodi and Paolo Scattolin for their insightful comments and bibliographical assistance with this chapter. Numbers in bold refer to testimonia and fragments in the Checklist (pp. 95–120). 1 B = Cod. Par. gr. 2713 fol. 129r and 185–203d with Mastronarde 2017: 21–23 for a summary, and Schmidt 1854: 242–46; Braswell 2017: 57 for examples. 2 See 207 on Achaeus. Didymus, while he was aware of Aeschylean material, may have composed hypomnēmata on Aeschylus (no. 14 Braswell), but such a work is inferred rather than attested. See Montanari 2009c: 419–20 and Braswell 2017: 52–53. Other possible traces of a Didymean commentary on Aeschylus are 201, 264, 266, 269, which refer to Aeschylus, but do not indicate a dedicated commentary; cf. also 305b. Hsch. ε 2679 Cunningham refers to οἱ ὑπομνηματισταί for an explanation of a hapax legomenon to A. 151 TrGF3, which Schmidt 1854: 240–41 inferred to be a commentary to Aeschylus. 3 See e.g. °4, where he analysed a word from Sophocles’ Niobe in Book 7 of his On Puzzling Words, which also could have been discussed in a Comm. in S. or Glossary of the Language of Tragedy, or both. 4 See Wartelle 1971: 185–95 (focus on Aeschylus); Zuntz 1965: 253–54; Pfeiffer 1968: 277–78; Trendelenburg 1867: 56–58 on Didymus’ works on Greek tragedy, Euripides in particular. © The Author(s) 2021. Published by Oxford University Press on behalf of the Institute of Classical Studies. All rights reserved. For permissions, please email: [email protected].



34

Didymus on Attic tragedy



35

choices and preferences, including the mistakes, and his use of other exegetical corpora. Furthermore, it examines the problems of identifying Didymean material where Didymus is not named.

2. DIDYMUS ON SOPHOCLES Didymus is mentioned nine times in the scholia to Sophocles, once in Athenaeus, and once in Harpocration in connection with Sophocles.5 Papyri assign comments to several scholars, but none to Didymus, although the compilers or authors of the editions of the papyri may have been using Didymus as an intermediary source.6 Rudolf Pfeiffer stated that Didymus’ preferred tragedian was Sophocles as there are no extant derogatory comments on Sophocles as there are for Euripides.7 Pfeiffer’s assessment is understandable, based on the available evidence and the predominant slant in the Didymean material in the scholia to Euripides, but the surviving passages do not signify that Sophocles was Didymus’ favourite tragedian; rather, they do not contain negative comments on Sophocles in comparison to Euripides. The surviving fragments cover a range of topics: spelling and textual criticism, grammar and syntax, explicatory comments on the meaning of words and passages, and antiquarian researches using other verse and prose works. The first example of Didymean exegesis is on syntax/grammar:

Od.: φρονοῦντα γάρ νιν οὐκ ἂν ἐξέστην ὄκνῳ Ath.: ἀλλ’ οὐδὲ νῦν σε μὴ παρόντ’ ἴδῃ πέλας Od.: ‘Yes, for I would not have shrunk in fear from him in his right mind.’ Ath.: ‘But now he will not even see that you are nearby’ (S. Aj. 82–83) ‘ἀλλ’ οὐδὲ νῦν σε’: Δίδυμος σημειοῦται τὴν φράσιν: ‘ἀλλ’ οὐδὲ νῦν σε μὴ παρόντ’ ἴδῃ πέλας.’ τὸ δὲ μὴ ἢ παρέλκον ἢ σύνδεσμος ἀντὶ συνδέσμου, ἀντὶ τοῦ ὡς, ἵν’ ᾖ· ὡς μὴ πλησίον αὐτοῦ παρόντα. ‘But now [he will] not [even see that] you [are nearby’]: Didymus marks as an exception the sentence ‘But even now he would not see you even though you are nearby’. The μὴ is either redundant or a conjunction used in place of another conjunction, ὡς in particular, so that it becomes ὡς μὴ πλησίον αὐτοῦ παρόντα (‘as if you are not near him’). (Σ S. Aj. 83 Christodoulou (176)) The context of this passage is that Athena is attempting to counter Odysseus’ reluctance to meet Ajax with a succession of arguments (Aj. 74–88) which fail; here she says that she will ensure Ajax will not see him (Aj. 83–85). Didymus is not proposing an emendation, rather an interpretative aid. He offers two options. First, he states that μή is redundant (παρέλκον), which makes παρόντα either a concessive or a predicative participle (i.e. either ‘even though you are close to him/even if you are close to him’ or ‘he will not see that you are close’) from Athena’s point of view. While it is syntactically fine, μή is not redundant, as it refers back to Odysseus’ οὐκ in the previous line.8 In another fragment, Didymus on Antigone (4–6) rightly notes that the 5 In this chapter, I do not include 180 (εἰ δ’ οὖν: εἰ δέ τις ἀνόητος εὑρεθῇ· Δίδυμος δέ φησι …) as Didymus’ comment has been lost. 6 P.Oxy. 1805 (ad Tr. 744) and P.Oxy. 2452 (Theseus ad fr. 730e.16–17 TrGF4) respectively attribute comments to ΑΡ and ΑΡΙ, which could refer to Aristophanes of Byzantium, Aristarchus of Samothrace, or Aristonicus; the last of these is probably the author of two variant readings attributed to ΑΡΝΙ (P.Oxy. 1174 ad Ichn. fr. 314.72, 314.146 TrGF4). The same papyrus attributes over a dozen comments and readings to ΘΕ or ΘΕΩ (presumably Aristonicus’ near-contemporary, Theon of Alexandria); while ΝΙ (frr. 220a.84.2, 314.108, 314.156 TrGF4) could be Nicander of Colophon (second century BC) or Nicanor of Alexandria (second century AD). 7 Pfeiffer 1968: 277. See Elsperger 1907: 108–27 for all of Didymus’ negative criticisms of Euripides, and Wilamowitz 1889–95: I 160-2 with nn. 79–83 for other examples. 8 Jebb 1900: 24.

36



Coward

transmitted reading ἄτης ἄτερ is the opposite of what is needed here and points out the contradictions.9 He further surmises that οὐ is superfluous, but notes that it is a feature of the tragic style. If μὴ is not redundant, the second explanation is that it is σύνδεσμος, a connecting particle or ‘combiner’, which keeps the negative. σύνδεσμος was used to signify anything from conjunctions to prepositions to interjections to noun phrases, in addition to the words that are now called particles, and it seems to have been coined to cover the words that were neither noun/adjective (ὄνομα), verb (ῥῆμα), nor adverb (ἐπίρρημα/μεσότης).10 According to Didymus, μὴ should be understood as acting in the manner of the catholic ὡς (σύνδεσμος ἀντὶ συνδέσμου, ἀντὶ τοῦ ὡς). Didymus then paraphrases the passage as ὡς μὴ πλησίον αὐτοῦ παρόντα. This makes the participle from the point of view of Ajax (the subject), for whom Odysseus is not present, and removes any hypothetical notions (i.e. ‘he will not see you because [from his point of view] you are not close’ or ‘he will not see you, thinking that you are not close’).11 ὡς, with μὴ as a negative, can sometimes take the participle to express comparison (i.e. ‘as (if)’, ‘just like’), hence the paraphrase could also mean ‘[But now he will not see you] as if you were not close to him.’12 The implication is that Athena’s aid will make Odysseus invisible as if he were not nearby, even though he actually is nearby. Didymus may have been following Odysseus’ train of thought that he preferred Ajax to be indoors.13 The transmitted line is οὐ(δὲ) […] μή + (aor.) subjunctive (ἴδῃ), which is a construction used for emphatic negative denials and prohibitions, i.e. something will definitely not be the case, and is a recognized construction of Greek dramatic idiom.14 Here Ajax will not see Odysseus, due to Athena’s intervention. Didymus’ interpretation does not recognize this construction as a tragic idiom and does not raise a problem with the subjunctive, but it raises problems with the function of μὴ. The next example concerns Didymus’ interpretation of another word in the Ajax. πᾶς δὲ στρατὸς δίπαλτος ἄν με χειρὶ φονεύοι. and the whole army, grasping spears in both hands, would strike me dead! (S. Aj. 408–09) (408a) δίπαλτος ἄν με : ἀμφοτέραις ταῖς χερσίν, οἷον περιδεξίως με φονεύοι. 15 παντὶ σθένει, ὡς Δίδυμος. ἢ ὁ στρατός με φονεύοι λαβὼν τὰ δίπαλτα δοράτια, ὡς Πῖός (F16 Hiller) φησιν. 9 178: οὐδὲν γὰρ οὔτ’ ἀλγεινὸν οὔτ’ †ἄτης ἄτερ† / οὔτ’ αἰσχρὸν οὔτ’ ἄτιμόν ἐσθ’ ὁποῖον οὐ τῶν σῶν τε κἀμῶν οὐκ ὄπωπ’ ἐγὼ κακῶν (‘No, there is nothing painful or laden with destruction or shameful or dishonouring among your sorrows and mine that I have not witnessed’) with Susanetti 2012: 157–58 for the various solutions proposed for this line, and see now Agosto 2018: 13–14, who proposes that οὔτ᾽ ἄτης ἄτερ is a ‘corruption of an initial οὖσ᾽ ἄτης ἄτερ, with οὖσ(α) being an adverbial participle of concession depending on the subject of ὄπωπ(α) in l. 6’. Cf. 197, where Didymus comments on the puzzling expression τὰς ἀνάγκας οἱ νόμοι διώρισαν (Eur. Hec. 847), and says it expresses the opposite of the meaning needed in the passage. 10 De Kreij 2016: I.2.§4. 11 See Swiggers and Wouters 2002: 102–08, especially 102 n. 7 on ‘combiner’; Edelstein and Kidd 1989–99: II.1 201–04 and De Kreij 2016: I.2, especially §§14–31 for a summary of the meanings of σύνδεσμος in ancient literary scholarship, and Dickey 2007: 260 for references to further discussions on σύνδεσμος. Σ Hom. Il. 1.512b, 2.139b, 2.463b, 3.31b Erbse would classify ὡς as μόριον. 12 See K.−G. III.2, §488 on ὡς and Tryphon F61 Velsen, an anonymous grammarian on Tryphon’s discussion of ὡς. πέλας and πλησίον are synonymous: Hsch. π 1290 Cunningham, τ 1751 Hansen–Cunningham; Syn.a π 274 Cunningham; Phot. π 542, σ 561, 897 Theodoridis; ΣD Hom. Od. 10.516b Ernst; Σ A. Th. 636m, 669g, 973m–n Smith; Σ Eur. Hec. 484 Schwartz; Σ S. Aj. 1171b Christodoulou; Et. Gud. p. 458, 12 Sturz; [Hdn.] Part. p. 105 Boissonade (πέλας, ἐπίῤῥημα, ἀντὶ τοῦ πλησίον·). Eur. Ph. 159–60 use the two together. See Wilamowitz 1889–95: III 52–3 ad Eur. HF 192 on the uses of πέλας and πλησίον. 13 S. Aj. 76, 80. 14 Cf. e.g. S. Ph. 103, 611–13, El. 1029; Ar. V. 394, Av. 461. Cf. also Plaut. Bac. 1037 and Ter. An. 205 where neque […] haud could be a translation of οὐδὲ μή. See Goodwin 1889: §295 and Appendix II; Wackernagel 1926–28: II 305 = Langslow 2009: 782; MacDowell 1971: 185; CGCG §34.9 on this construction. 15 Paolo Scattolin, based on the sequence of Σ Ar. Av. 35a Holwerda, observes per litteras that the inserted above is unlikely and could be replaced with .

Didymus on Attic tragedy



37

‘[All the army] brandishing with their hands (δίπαλτος) would strike me dead’: with both hands, that is, they would kill me using both hands. ‘with all their strength’, as Didymus says. Or the army would take up spears that are hurled with both hands and kill me, as Pius says. (Σ S. Aj. 408 Papageorgiou (177)) There are two explanations here: ἀμφοτέραις ταῖς χερσίν […] ὡς Δίδυμος is the one of Didymus, and the remainder of the passage is the explanation of Pius. Didymus said that δίπαλτος meant περιδεξίως, which is paraphrased as παντὶ σθένει (‘with all their strength’), referring back to ἀμφοτέραις ταῖς χερσίν. Intensificatory δι-, however, is unparalleled.16 The Homeric warrior was normally armed with two spears (e.g. Il. 3.18–19, 10.76, 11.43, 12.298; Od. 1.256) and there was the phrase πάλλων δ᾽ ὀξέα δοῦρε (Il. 5.495 = 6.104 = 11.212), referring to the two spears of the Homeric warrior held in both hands.17 In Greek tragedy, the adjective had passed into a figurative sense, as indicated by τριπάλτων πημάτων (‘thrice-hurled sufferings’) in Aeschylus (Th. 985).18 Euripides’ δίπαλτον ἱερὸν […] κεραυνοφαὲς πῦρ (Tr. 1102) suggests that the god has a thunderbolt in each hand, and in ὡς δ᾽ εἴδομεν δίπαλτα πολεμίων ξίφη (IT 323) the adjective here means ‘two’ or ‘twin’, rather than that Orestes and Pylades are brandishing a sword in each hand.19 δίπαλτος clearly had literal and metaphorical meanings, which were the subject of discussion, as demonstrated above.20 Didymus’ παντὶ σθένει does occur elsewhere in tragedy, and his interpretation may have derived from other exegetical corpora.21 In the opening of Aristophanes’ Birds, Euelpides remarks that ‘we, being of good standing in tribe and clan, solid citizens, with no one trying to shoo us away, have up and left our country with both feet flying’ (ἡμεῖς δὲ φυλῇ καὶ γένει τιμώμενοι, / ἀστοὶ μετ᾿ ἀστῶν, οὐ σοβοῦντος οὐδενὸς / ἀνεπτόμεθ᾿ ἐκ τῆς πατρίδος ἀμφοῖν τοῖν ποδοῖν, Av. 33–35). The scholia remark that ἀμφοῖν τοῖν ποδοῖν is to be understood as ἀμφοῖν τοῖν πτεροῖν,22 and its meaning is παντὶ σθένει.23 At the beginning of Iliad 17, Menelaus is first into the fray to fight over Patroclus’ body. The scholia remark that Menelaus’ drive to help is with all his strength (παντὶ σθένει).24 In later authors, following Thucydides (1.86.3, 5.23), σθένος was understood as a reference to armed forces and παντὶ σθένει to full military might.25 Didymus, then, understands the phrase in a figurative sense, and his interpretation may have been informed by the exegeses above. The third example concerns Didymus’ emendation of a transmitted word from a verb to a noun by changing its accentuation (πειρᾷ > πείρᾳ).26 Ὦ πάντα τολμῶν κἀπὸ παντὸς ἂν φέρων λόγου δικαίου μηχάνημα ποικίλον, 16 Lorimer 1936–37: 174; Finglass 2011: 260. Cf. ἀμφοτέραις ταῖς χερσίν here with ἀμφοῖν τοῖν ποδοῖν = ἀμφοῖν τοῖν πτεροῖν (Σ Ar. Av. 35a Holwerda). 17 Asteropaeus throws both spears simultaneously from each hand (Hom. Il. 21.162–3). Didymus (not named) may have criticized Aristarchus for choosing a certain reading in this passage (Σ Hom. Il. 21.162a1 and 162b2/a2 Erbse; cf. also Σ Hom. Il. 1.106e Erbse). Cf. also Eur. Tr. 1102. 18 τριπάλτων literally means ‘thrice brandished’, i.e. with a weapon, but Sommerstein 2008: 260–61 n. 148 thinks that here it may possibly refer to the three generations of suffering in the house of Laius (cf. A. Th. 742 ff.). The alternative reading διπάλτων in Tr (Cod. Neap. II F31, c.AD 1325) may have been influenced by tragic idiom. 19 Cf. Eur. IT 242 (δίπτυχοι νεανίαι). 20 Suda δ 1256 Adler preserves these interpretations, though without named sources. Ps.-Zonaras (s.v. δίπαλτος Tittmann (p. 509)) more or less repeats the contents of the scholia vetera, but conflates the first meaning and Didymus’ together, and then quotes the Ajax passage. 21 A. Supp. 146 and S. fr. 904 TrGF4. Cf. also Pi. P. 4.79; B. 18.49; A. Ag. 643; Eur. Hec. 1155–56. 22 This is a mixing of imagery between avian and human anatomy. Cf. 40ab for a similar line of interpretation of metaphorical imagery. 23 Σ Ar. Av. 35a Holwerda; cf. Suda ν 450 Adler; cf. also Suda α 1793 Adler. 24 ΣA Hom. Il. 17.1–2a2 Erbse. 25 Th. 1.86.3; cf. also Syn.a π 82 Cunningham; Hsch. π 407 Hansen = Phot. π 173 Theodoridis; Suda π 232 Adler. Σ Th. 5.23 Hude; Σ D. Or. 3.48, 49a Dilts says that παντὶ σθένει was understood as military and naval might; cf. also ΣbT Hom. Il. 18.274b Erbse (the source of S. fr. 904 TrGF4), where some commentators took σθένος as meaning to refer to the army. 26 L has πείρᾶ. On accentuation and spelling of words, cf. Σ Hom. Od. 11.221 Dindorf; Σ Hom. Il. 14.199a1 Erbse.

38



Coward

τί ταῦτα πειρᾷ κἀμὲ δεύτερον θέλεις ἑλεῖν ἐν οἷς μάλιστ’ ἂν ἀλγοίην ἁλούς; You who will dare anything and would extract a cunning trick from any just plea, why do you make this attempt and why do you try once more to catch me in the trap that would most pain me if you caught me? (S. OC 761–64) βαρυτόνως ἀναγνωστέον πείρᾳ, οὐ περισπωμένως· οὕτω Δίδυμος. One should read it with the accent recessed i.e. πείρᾳ, not with a circumflex on the ultima, i.e. not πειρᾷ. So [says] Didymus. (Σ S. OC 763 Xenis (183)) The passage commented upon is the rhēsis of Oedipus in response to Creon. Oedipus berates Creon for his reproaches and dishonesty (OC 761–99). The transmitted πειρᾷ is a contracted present indicative second-person singular medio-passive form of πειράομαι/πειράω, whereas Didymus’ πείρᾳ is a noun in the dative singular meaning here ‘by a stratagem’.27 The reason for Didymus’ change is unclear. Earlier scholars thought that there was an error in the transmission.28 One option is that Didymus has misunderstood a line of thought of Homeric exegesis on verbs that contract in a similar fashion, but the commented verb remained verbal, rather than changing from a verb to a noun.29 Georgios Xenis also proposes that Didymus seems to have read ταύτᾳ πείρᾳ as an alpha with Doric accentuation, which would make the line unmetrical.30 If this is the case, it makes Didymus’ misunderstanding of Homeric exegesis unlikely. It is possible that Didymus’ πείρᾳ was referring back to μηχάνημα ποικίλον (‘a cunning trick’) in the preceding line. Oedipus characterizes Creon as bold and crafty, and he refers to him as a man who ‘says harsh things softly’ (σκληρὰ μαλθακῶς λέγων, 774); his ideas are ‘good as speech [i.e. they sound good] but bad as deeds’ (λόγῳ μὲν ἐσθλά, τοῖσι δ’ ἔργοισιν κακά, 782); and his mouth is both false and well sharpened (794–95).31 However, πειρᾷ does occur elsewhere in Homer and Sophocles, and it reoccurs a few lines later without an extant comment (774).32 Earlier on, the chorus has learnt of Oedipus’ identity, which he reluctantly revealed after their persistent questioning (207–24). They are horrified to have him in their company, and they demand that he leave (226–36). Antigone implores the chorus to show pity and recognize that Oedipus’ incest and parricide were unwitting and inescapable (237–54). Oedipus then argues for his moral innocence and appeals to Athens’ reputation for compassion (258–91). The chorus admit that they feel pity and they allow Oedipus to wait for Theseus (292–309). Some critics thought that Antigone’s lines (237–57) ought to be considered spurious, as

27 See Chantraine 1958–63: I 57 on the form with Et. Gen. β 158 Lassere–Livadaras (II 454–55); cf. also Choerob. in Theod. II 175.10 (GG IV.2); and EM p. 757.50–55 Gaisford s.v. τητᾷ. 28 Hermann 1825: 169 thinks that Didymus meant τί τῇδε πειρᾷ … or τί ταῦτα; πειρᾷ κἀμὲ … and Meineke 1868: 211 suggests either οὐ βαρυτόνως ἀναγνωστέον πείρᾳ, ἀλλὰ περισπωμένως or περισπωμένως ἀναγνωστέον πειρᾷ, οὐ βαρυτόνως. The second of Meineke’s suggestions was already anticipated by Schneidewin 1853; I thank Paolo Scattolin for this piece of information. Didymus’ explanations are not very convincing in those passages where modern editors assume the readings he explains are corrupt (190) or spurious (202); nor is his idea that a vocative in Hecuba is directed to Polydorus, not a self-address (197). 29 See ΣΤ Hom. Il. 14.199a1 Erbse (ex. [Hrd.?]) with Schironi 2018: 615–16 on δαμνᾷ, which Aristarchus argued was a simpler contraction of δάμνημι (cf. Hom. Od. 14.488), rather than δαμνάω, whereas others thought it was a form of Doric accentuation from δάμνασαι and so read it as δάμνᾳ. Cf. also Σ Hom. Od. 11.221 Dindorf (II 491–92) on the same verb, where Crates of Malles (fr. 55 Broggiato) read δαμνᾷ ἐπεί κεν as δάμναται ὥς κεν, and Ptolemy of Ascalon (p. 197 Baege) proposed δάμνα(ι?), apocope of δάμναται. 30 Xenis 2018: 158. See Braswell 2017: 164 on Didymus’ metrical errors in the scholia to Pindar. 31 Cf. also S. OC 981, 986, where Creon’s mouth is impious, and his speech is slanderous. 32 Hom. Il. 21.459, 24.390, 433; S. OT 360, 399, OC 774.

Didymus on Attic tragedy



39

Antigone’s interjection breaks up the exchange between the chorus and Oedipus, although it is in fact dramatically effective.33 ‘ὦ ξένοι αἰδόφρονες’: τὸ τῆς Ἀντιγόνης πρόσωπον ὅλον καὶ τοῦ χοροῦ τὸ τετράστιχον ἀθετοῦνται· κρεῖττον γάρ φασιν εὐθέως τῷ δικαιολογικῷ χρήσασθαι τὸν Οἰδίπουν πρὸς αὐτούς. ἀλλὰ τὰ πράγματα αὐτοῖς οὐκ ἐν καιρῷ ἐστιν ἀλλ’ ἐν δυσπραγίᾳ, ὥστε ἐπαφρόδιτον εἶναι αὐτοῖς τὴν ἐλεεινολογίαν, καὶ τοῦτο τὸ πρόσωπον ἡ Ἀντιγόνη πληροῖ. ἐπεὶ μέντοι οὗτοι οὐ πείθονται, τότε δικαιολογικώτερον καὶ ὥσπερ ἀπολογούμενος ἐκφέρει τὰ ἑξῆς ὁ Οἰδίπους, ὅτι ἀκούσιά ἐστιν αὐτῷ τὰ ἁμαρτήματα. Καὶ 〈ὅρα〉 εἰ ταῦτα τῆς ὁμοίας ἔχεται δυνάμεως τοῦ ποιητοῦ. Καὶ καθόλου θαυμαστή τίς ἐστιν ἡ οἰκονομία τοῦ δράματος.34 Οὐδὲν δὲ ἐν τοῖς Διδύμου τούτων ὀβελισθὲν εὕρομεν. ‘O strangers of respectful mind’: The whole speech of Antigone and the four lines of the chorus are athetized. For they say it is better that Oedipus should straightaway address his justification to them (sc. the chorus). However, their circumstances are not opportune, but unfortunate, so that in their case a piteous appeal has its charm, and Antigone supplies this speech. Nevertheless, since these (sc. the chorus) are by no means persuaded, Oedipus then delivers his justification in what follows with the defence that his errors were involuntary. And if these lines contribute in the usual manner [i.e. what we expect from Sophocles] the expressive power of the poet. In general, the disposition of the material of the drama is extraordinary. We find that none of this section [OC 237–57] has been obelized in the works of Didymus. (Σ S. OC 237 Xenis (182)) Antigone’s monody interrupts the amoebaean between Oedipus and the chorus and yet it continues in the same metrical colour of the exchange with Antigone acting as a mediator. The scholia rightly add that it is natural and graceful that an appeal to pity (ἐλεεινολογία), which Antigone makes, should precede the father’s appeal to reason (τὸ δικαιολογικόν).35 It is stated by the anonymous critic that Didymus did not athetize this section of the play in his works (ἐν τοῖς Διδύμου).36 Due to Didymus’ methods of consulting prior editions, this comment raises the possibility that (some of) the older Alexandrian copies did not either. Elsewhere, Didymus accused actors of being responsible for the interpolation of line 380 after 356 in Euripides’ Medea.37 This interpolated line in the Medea does not occur in the manuscript tradition, which suggests either Didymus saw it in a copy or copies he was able to consult, or it had been reported by an earlier scholar.38 This passage here also suggests, if the comments on the stylistic suitability of Antigone’s speech preceding 33 The transmitted reading in LR is ὀβελισθέντων, which indicates that the verses were not obelized in Didymus’ edition. Lascaris 1518 emended to ὀβελισθὲν. The transmitted reading was emended by Erbse (1955): 86 to τῶν ὀβελισθέντων, which would indicate the opposite, that the verses had been expunged. See Havekoss (1960): 50–52, the student of Erbse, for a detailed discussion of this passage. 34 Cf. ἄφατος δέ ἐστι καθόλου ἡ οἰκονομία ἐν τῷ δράματι, ὡς οὐδενὶ ἄλλῳ σχεδόν (Hyp. IV.14–15 OC Xenis) from the hypothesis attributed to Sallustius, which suggests an overlap between that passage and 182; see Grisolia 2001 and Nünlist 2009: 24–28 on οἰκονομία. 35 Cf. 191–192, where Didymus declares Andromache’s statements to Menelaus to be contrary to the requirements of the situation and not appropriate to her as a non-Greek woman; and he finds fault with the use of a conventional motif of grief-stricken speechlessness (194). 36 Schmidt 1854: 241 thought that εὕρομεν referred to Pius, De Marco 1952: xxvi, to Sallustius, and Havekoss 1960: 52, to the compiler of the collection of scholia. Scattolin per litteras posits that all of 182 bar the last sentence (Οὐδὲν δὲ […] εὕρομεν) is by Sallustius (cf. Hyp. IV.14–15 OC Xenis in n. 34), who reports the doxography and indicated agreement with Didymus’ interpretation of the passage. ἐν τοῖς Διδύμου is a hapax, but ἐν τοῖς + scholar’s name or title of work is a common occurrence, e.g. Aristarchus started his activity with a hypomnēma based on Aristophanes’ text (Σ Hom. Il. 2.133a: ἐν τοῖς κατ’ Ἀριστοφάνην ὑπομνήμασιν Ἀριστάρχου); cf. also Σ Hom. Il. 8.37a (ἐν τῇ Ζηνοδότου [ἐκδόσει?]) and Σ Hom. Il. 19.26a (ἐν τοῖς Ζηνοδότου [ἀντιγράφοις?]). The papyrus of Sophocles’ Ichneutae (P.Oxy. 1174) testifies to a number of variant readings marked by οὕτως ἦν ἐν τῷ Θέωνος, ‘thus it was in Theon’s [copy/edition]’ (Theon, frr. 19– 35 Guhl). Theon’s ἔκδοσις encompassed all the works of Sophocles, but it remains unclear, however, whether Theon also wrote a commentary (Guhl 1969: 13). 37 188–189 with Mastronarde 2017: 22 n. 90. 38 Didymus also used the works of Callistratus for the texts of Aristophanes; see West 2001: 60–61.

40



Coward

Οὐδὲν δὲ […] εὕρομεν are derived from Didymus or another (e.g. Sallustius), his reasons or the reasons he reported for keeping the passage in his commentary on Sophocles. The last extant fragment of the commentary, preserved in Athenaeus, very likely overlaps with his Glossary of the Language of Tragedy and concerns the mostly lost Phoenix/The Dolopeians.39 Didymus glossed κύναρος ἄκανθα and surmised what Sophocles meant: ‘κινάρα’ ταύτην Σοφοκλῆς ἐν Κολχίσι (348 TrGF4) κυνάραν καλεῖ, ἐν δὲ Φοίνικι ‘κύναρος ἄκανθα πάντα πληθύει γύην’ (718 TrGF4) […] Δίδυμος δ’ ὁ γραμματικὸς ἐξηγούμενος παρὰ τῷ Σοφοκλεῖ (718 TrGF4) τὸ ‘κύναρος ἄκανθα’ ‘μήποτε, φησί, τὴν κυνόσβατον λέγει διὰ τὸ ἀκανθῶδες καὶ τραχὺ εἶναι τὸ φυτόν. καὶ γὰρ ἡ Πυθία ξυλίνην κύνα αὐτὸ εἶπεν (L83 Fontenrose), καὶ ὁ Λοκρὸς χρησμὸν λαβὼν ἐκεῖ πόλιν οἰκίζειν ὅπου ἂν ὑπὸ ξυλίνης κυνὸς δηχθῆι, καταμυχθεὶς τὴν κνήμην ὑπὸ κυνοσβάτου ἔκτισε τὴν πόλιν.’ kinara (‘artichoke’). Sophocles in the Colchian Women calls this [kinara] a ‘kunara’ but in the Phoenix says ‘the thorny artichoke [kunaros] fills every field’ […] Didymus the grammarian in his explanation of the term ‘thorny artichoke’ (κύναρος ἄκανθα) in Sophocles says: ‘perhaps he means the wild rose, since the plant is prickly and rough. The Pythia, in fact, referred to it as a “wooden dog” [a wild rose, κύνα]; and after Lokros received an oracle telling him to plant a city [Ozolian Locris] in a place where he was bitten by a wooden dog, he founded the city when his shin was scratched by a wild rose [κυνόσβατος].’ (Ath. 2.70c–d (44 = 184)) ἐξηγούμενος παρὰ τῷ Σοφοκλεῖ indicates that Didymus made these remarks in his commentary on Sophocles. It is likely that Didymus delved further into antiquarian and lexicographical discussions on the artichoke. κύναρος ἄκανθα, according to this same passage of Athenaeus, was also discussed by Hecataeus of Miletus (BNJ 1 F291–92a) and Skylax (BNJ 709 F3–4) or Polemon of Ilium (F92 Preller), who say that kunara thorns are found in the mountains by the Caspian Sea (the Elburz) and in the mountains of the Chorasmians around the river Indus. kunara/kinara appears to be the cardoon, the wild progenitor of the artichoke.40 Theophrastus’ discussion of κυνόσβατος suggests a species of rose, which may be Didymus’ source.41 Hesychius also uses Didymus’ interpretation without attestation, which suggests that Didymus also discussed κινάρα/κύναρος in his Glossary of the Language of Tragedy.42 3. SOME POTENTIAL FRAGMENTS OF DIDYMEAN COMMENTARY Pfeiffer also posited four other passages in the scholia to Sophocles’ Oedipus at Colonus to be considered as Didymean on account of references to earlier ὑπομνηματισάμενοι.43 These passages raise the issue of identifying Didymean passages although he is not named as one of the ὑπομνηματισάμενοι.44 These four passages may be proposed as Didymean on the grounds of the high quality of the surviving scholia to the Oedipus 39 S. 718 TrGF4 (ap. Ath. 2.70a): κύναρος ἄκανθα πάντα πληθύει γύην (‘The kunara thorn abounds in every portion of the land’) is the full quote. γύην could be a reference to Dolopia, Phthia, Scyros, Troy, or Eion. 40 Later Athenaeus (2.71a) states that κινάρα (‘artichoke’) rather than κυνάρα was in use in his day; cf. also Galen, de alim. facult. 2.51, who regards κυνάρα as an affectation, and Hdn. Philet. 210 Dain. 41 Thphr. HP 3.18.4  Ath. 2.70d. 42 Hsch. κ 4562 Cunningham. 43 Pfeiffer 1968: 277. ΣLrT S. OC 388, 390a1, 683–4 Xenis. A similar matter occurs in the Homeric scholia with references to Aristarchus and Zenodotus, and references to Eratosthenes and Lycophron in the scholia to Aristophanes are excerpts of Didymean commentary; here, however, there are no named scholars, just οἱ ὑπομνηματισάμενοι. 44 οἱ ὑπομνηματισάμενοι: Σ Hom. Il. 16.21b, 16.467a1 Erbse (probably means Didymus); Σ Pi. O. 6.55a (112a), P. 3.18a; Σ A. Pers. hyp. I Dindorf; ΣLrT S. OC 388, 390, 681 Xenis; Σ Eur. Andr. 32, Or. 1384 (203a) Schwartz (the latter is the only occurrence in the singular). Cf. also Σ Hom. Il. 7.452a, 14.382d1 Erbse.

Didymus on Attic tragedy



41

Coloneus, and comparisons based on Didymus’ methodologies.45 οἱ ὑπομνηματισάμενοι can occur with and without Didymus named in the same phrase, which can lead to the assumption that the phrase is anterior to Didymus.46 For example, Didymus reports that line 46 of Antigone was athetized by ὑπομνηματισταί, but it is not stated that he agreed with them, only that he reported that they did this.47 The plural may be a convention and thus conceal a reference to a single scholar: e.g. Wilamowitz, who understood that οἱ φαύλως ὑπομνηματισάμενοι at Σ Eur. Andr. 32 Schwartz referred to Didymus.48 Didymus did consult other hypomnēmata and editions by other critics.49 Didymus’ mention of οἱ ὑπομνηματισταί at Σ Hom. Il. 16.467a1 Erbse suggests a plurality of commentary sources.50 Since later teachers and commentators knew that one commentary had routinely borrowed from another, they could consider a view to have been held by the transmitters as well as the original exponent, and thus the plural was natural.51 Therefore, some instances of οἱ ὑπομνηματισάμενοι/ὑπομνηματισταί may be examples of Didymean commentary which report his summation of prior commentators, but his name has dropped out. This is suggested by examples of when his name and οἱ ὑπομνηματισάμενοι/ὑπομνηματισταί have been preserved together, or Didymean material is buried beneath the phrase.52 A passage of dialogue between Ismene and Oedipus contains two potential Didymean fragments. Ismene reports that oracles have foretold that the city (Athens or Thebes) where Oedipus is buried will have power, and she warns Oedipus that Creon will arrive soon to persuade him from staying at Athens. The first passage (ΣLrT S. OC 388 Xenis) offers a summation of the oracle section, and remarks on the lack of another literary source to support the Sophoclean version. Didymus liked to quote historical and antiquarian sources and passages of poetry.53 Here, ἐβουλόμην δὲ αὐτοὺς μαρτυρίῳ χρήσασθαι indicates a wish by the anonymous critic that the οἱ ὑπομνηματισάμενοι had done so, and may suggest Didymus or another critic’s frustration at being unable to find such a resource. In the second passage (ΣLrT S. OC 390a1 Xenis) the transmitted text of Ismene’s speech is εὐνοίας (‘goodwill’ or ‘favour’), but the scholia and Triclinian MSS have the alternative reading of εὐσοίας, which has been adopted in subsequent critical editions even though the word only occurs here in Sophocles and very likely in Alcaeus and Crinagoras of Mytilene.54 εὐσοίας was favoured by οἱ ὑπομνηματισάμενοι based on the readings preserved in the better-quality copies of the plays (ἐν τοῖς ἀναγκαιοτέροις τῶν ἀντιγράφων).55 Didymus elsewhere refers to using better-quality copies.56 The critic summarizing the remarks of οἱ ὑπομνηματισάμενοι states the meaning of εὐσοία, and cites a parallel from

45 Since Didymus is not named in these passages, they are not included in the Checklist, per the principles stated. 46 Didymus and ὑπομνηματισάμενοι: 112a, 146, 179. 47 179. S. Ant. 45–46: τὸν γοῦν ἐμόν, καὶ τὸν σόν, ἢν σὺ μὴ θέλῃς, / ἀδελφόν· οὐ γὰρ δὴ προδοῦσ’ ἁλώσομαι (‘I certainly intend to bury my brother—and yours, if you will not. I will not be caught betraying him’). While the two-line utterance breaks the flow of stichomythia, which is perhaps why the ancient critics suspected it (cf. 182); the two-line reproach by Antigone to Ismene works for dramatic effect and emphasizes Antigone’s isolation and sense of duty; see Lloyd-Jones and Wilson 1990: 116–17 and Griffith 1999: 130–31. 48 Wilamowitz 1889–95: I 160 n. 179. 49 See West 2001: 75–83 for examples from Homeric scholarship. 50 Schironi 2018: 19. 51 Mastronarde 2017: 10 n. 37. 52 Another instance is where a title (known as a work by Didymus) survives, but not the author, e.g. *49 ap. Σ S. Tr. 1159 Xenis: ‘ἐμοὶ γὰρ ἦν πρόσφαντον’: προσμεμαντευμένον, προσειρημένον ὑπὸ τοῦ Δίος· οὕτω δὲ διὰ τοῦ σ ἐν τῆι Τραγικῇ λέξει. (‘“For there was a prophecy for me”: having been prophesized, having been spoken to him by Zeus. Thus, with a sigma, in the Tragic Vocabulary’). This could be from Didymus’ Glossary of the Language of Tragedy. *38 has the same problem. 53 See Braswell 2011 for examples. 54 S. OC 389–90: σὲ τοῖς ἐκεῖ ζητητὸν ἀνθρώποις ποτὲ / θανόντ᾿ ἔσεσθαι ζῶντά τ᾿ εὐσοίας χάριν (‘That you shall one day be sought by the people there in death and in life for their preservation’s sake’). Alc. fr. 286a.6 Liberman (ε]ὐσοΐα{ι}ς) and Crinagoras 36.4 Ypsilanti (εὐσο) with Scattolin 2013: 30–31 with nn. 14–15. 55 Timachidas of Rhodian Lindos (F27 Matijašić) had access to a few different copies of the text of Aristophanes. Aristarchus consulted the older copies (ἐν τοῖς ἀρχαίοις [ἀντιγράφοις?], Σ Hom. Il. 6.4a Erbse). 56 182.

42



Coward

Sophocles’ Amphityron (122 TrGF4) to support this reading.57 This interpretation of εὔσοια is also backed up in the lexicographical corpus.58 The variation in spelling and meaning is likely to be due not to a spelling error, but to the similar meaning between the words and εὐνοία being more common than εὐσοία.59 The final passage of possible Didymean commentary shows the use of antiquarian and literary sources (ΣLrT S. OC 683–4 Xenis). In the first stasimon, the scholia comment on the narcissus flower and discuss the identification of the goddesses who used this flower as a crown (681–85).60 The anonymous critic quotes οἱ ὑπομνηματισάμενοι who said that the narcissus was not the crown of Demeter and Kore, because they used ears of corn. Euphorion is cited to back up the argument that the Eumenides are meant here.61 Elsewhere Hesychius (δ 185 Cunningham) refers to the δαμάτριον (named after Demeter) as a name for the narcissus flower.62 They also say that the narcissus grew around tombs and that νάρκισσος suits goddesses who produce narcotic effects.63 The anonymous critic also adds that it was forbidden to carry this flower during the Thesmophoria, and also asks if the dual form μεγάλαιν θεαῖν should be understood as a reference by syllepsis to Kore, who was described as picking up narcissus flowers when she was seized by Pluto.64 The last piece of evidence is from Istros the Callimachean, who said that the crown of Demeter was made with myrtle and smilax in order to reinforce the reference to Kore.65 The passage is a mixture of opinions, but Didymean authorship can be proposed, as the fragments of Didymus’ commentary on Pindar demonstrate, because he liked to quote prose, usually historical and antiquarian sources, and passages of poetry.66 These quotations sometimes could be extensive, and they were used for comparison, or to establish historical context or allusions, or to express disagreement with earlier commentators and scholars. Here there are quotations from Euphorion and Istros the Callimachean, who predate Didymus. Didymus uses or cites Euphorion and Istros elsewhere, e. g. 119, °281, 293ab on the former and 112a, 306 on the latter. Therefore, it is possible that this is from Didymus, based on his use of verse and prose sources, but this passage contains a jumble of opinions indicating a few degrees of separation between its current form and the commentary passage it was based upon. There is another passage from the scholia to Sophocles, without reference to commentators, that could be considered fortasse Didymi: διχ(ῶς)· ‘καὶ δῆλός ἐστιν, ὥς τι σημανῶν νέον.’ In two senses: ‘and it is clear that he is signalling something untoward.’ 57 Tosi (2003: 363) also proposed Didymean authorship of this passage, although he argues that Didymus is one of οἱ ὑπομνηματισάμενοι, whereas I think Didymus, if the unnamed critic here, is reporting their views. 58 Hsch. ε 7192 Cunningham; Phot. ε 2337 Theodoridis. Scattolin 2013: 30–34 proposes, on the basis of these parallels in the lexicographical corpus, the loss of ἢ (vel καὶ) σωτηρίαν (found in Hesychius) after τὴν εὐθένειαν in ΣLrT S. OC 390a1 Xenis, which is not mentioned in Xenis 2018: 118–19. Cf. also Suda ε 3757 Adler, which is derived from the scholia to Sophocles. Theocritus (Id. 24.8) also has εὔσοα τέκνα, and cf. also Bato fr. 5.10 PCG4 with app. crit. and Hsch. ε 7195 Cunningham. 59 Didymus shows that the change of alphabet had generated many variants in Σ Hom. Il. 7.238c2: αἱ Ἀριστάρχου ‘βῶν’. ἡ Ἀριστοφάνους ‘βοῦν’. ἡ Ῥιανοῦ ‘βῶ’ […] ἐν τοῖς παλαιοῖς ἐγέγραπτο ΒΟΝ, ὅπερ οὐκ ἐνόησαν οἱ διορθωταί (‘the editions of Aristarchus [have] βῶν; that of Aristophanes [has] βοῦν; that of Rhianus [has] βῶ […] in the ancient [manuscripts] it was written ΒΟΝ, which the editors did not understand’). 60 Cf. also Plu. Quaest. Con. III 674b and Clem. Al. Paed. 2.72.3. Cf. also Verg. G. 4.122 on the narcissus flower as a crown and Thphr. HP 6.6.9 on the flower. 61 See Villari 2013 on the grove of the Eumenides in Sophocles’ OC. 62 Cf. also Dsc. 4.60.2. Thphr. HP 6.8.1–3 remarks that the polyanthus narcissus was one of the earliest-used flowers for making garlands. 63 Berti 2009: 181 and commentary on BNJ 334 F29. On the etymological connections between νάρκισσος, ναρκάω, νάρκη, and νάρκησις, see Chirassi 1968: 143–44; cf. also Baumann 1986: 69–72. 64 The critic alludes to H. Hom. 2.8; cf. also Paus. 8.31.1, 9.31.9. 65 The original context of the quotation is lost, but possibly it was drawn from Istros’ Atakta (see BNJ 334 F17). On the use of myrtle during the Eleusinian Mysteries, see Ar. Ran. 156, 300 with Σ ad loc. The lawsuit referred to by Istros can be possibly identified with a speech falsely attributed to Deinarchos; see De Marco 1961. 66 See Braswell 2011 for examples.

Didymus on Attic tragedy



43

(Σ S. Aj. 1225a Christodoulou/Papageorgiou = Comm. in S. F9, p. 242 Schmidt) διχ in L introduces this portmanteau. The proposed variant is a blend of Aj. 326 (καὶ δῆλός ἐστιν ὥς τι δρασείων κακόν) and Ant. 242 (δηλοῖς δ᾽ὥς τι σημανῶν νέον). The actual line spoken by Teucer is δῆλος δέ μοὐστὶ σκαιὸν ἐκλύσων στόμα (‘and it is clear that he will open his mouth in jarring speech’). Elmsley, followed by Schmidt, Jebb, and Colonna, took διχ to mean Δίδυμος with the chi used to draw attention to the name.67 But it is easier for διχ to mean διχῶς, which is commonly used in scholia to introduce variants, e.g. Σ S. El. 28 Xenis; Σ [A.] Pr. 3a Herington. However, Didymus in his Iliadic commentaries also uses distinct phrasing: a reading is often introduced with οὕτως, ‘so’, while concurrent readings (both accepted by Aristarchus) are introduced by the adverb διχῶς, ‘according to two readings’, ‘in two ways’.68 So, this could be a piece of Didymus, and suggest a variant that was noted by Aristarchus and reported by Didymus.69

4. DIDYMUS ON ION OF CHIOS There are two known testimonies found in Athenaeus of Didymus’ works on Ion of Chios, and there is another fragment in Stephanus of Byzantium.70 These references indicate a commentary on Ion’s Agamemnon, and the other reference indicates a general work of refutations against other scholars who had commented on Ion. Athenaeus (11.468c) quotes a passage of Ion’s Agamemnon to define the daktulōtos (a type of drinking vessel):

οἴσῃ δὲ δῶρον ἄξιον δραμήματος ἔκπωμα δακτυλωτόν, ἄχραντον πυρί, Πελίου μέγ’ ἆθλον, Κάστορος δ’ ἔργον ποδῶν You will carry off a gift worthy of the race you ran, a daktulōtos drinking vessel, untouched by fire, a prize cherished by Pelias, to commemorate what Castor’s feet accomplished. (Ion F1 Leurini = TrGF1 19 F1) The meaning of daktulōtos is discussed, although the passage (Ath. 11.468c–f) is a garbled mix of several sources, e.g. 468c and 468f are very similar other than that the definition of daktulōtos is assigned to Epigenes (F4 SGG, fourth/third century BC) in the former and in the latter Philemon records his sources without cited authorities.71 The first pair of definitions are that the daktulōtos is a drinking vessel of two handles, in which one can insert one’s fingers on either side; unnamed authorities say that it is a cup with impressions resembling fingers all around it. The third explanation, attributed to anonymous sources, was that the daktulōtos was a form of drinking-horn, with a reference to Theopompus about horns being used to make drinking vessels.72 Then Athenaeus offers Didymus’ explanation: 67 Elmsley 1825 ad loc. followed by Jebb 1900: 183 and Colonna 1983: 139. See Turner 1968: 116–17 and Pontani 2018 on the many ways in which χ is used. 68 Schironi 2018: 13. 69 On Aristarchus’ writings on Sophocles, see Σ S. El. 6–9 Xenis (with Hsch. λ 1390 Cunningham for authorship), frr. 728 TrGF4 (Chryses), 449 TrGF4 (°4: Niobe), and 624 TrGF4 (Troilus). 70 Hsch. ν 714 Cunningham cites Ion 19 F52 TrGF1 = *62 Leurini, but Meineke’s emendation of Didymus’ name into the text of Hesychius is not secure. 71 See Massara 2001 on this vase. 72 Theopomp. BNJ 115 F284.

44



Coward

‘βέλτιον δὲ λέγειν,’ φησὶν ὁ Δίδυμος ἐν τῷ τοῦ δράματος (sc. τοῦ Ἀγαμέμνονος) ἐξηγητικῷ, ‘ὅτι παρήκουσεν Ὁμήρου λέγοντος (Il. 23.270)· πέμπτῳ δ’ ἀμφίθετον φιάλην ἀπύρωτον ἔθηκεν. 'ἔδοξε γὰρ ἔκπωμα εἶναι· ἐστὶ δὲ χαλκίον ἐκπέταλον λεβητῶδες, ἐπιτηδείως ἔχον πρὸς ὑδάτων ψυχρῶν ὑποδοχάς. δακτυλωτὸν δ᾽ οἷον κύκλῳ τὴν φιάλην κοιλότητας ἔχουσαν ἔνδοθεν οἷον δακτύλων, ἢ ἐπεὶ περιείληπται τοῖς τῶν πινόντων δακτύλοις. τινὲς δὲ ἀπύρωτον φιάλην τὸ κέρας· οὐ γὰρ γίνεται διὰ πυρός. λέγοι δ᾿ ἂν ἴσως κατὰ μεταφορὰν ἔκπωμα τὴν φιάλην.᾽ Didymus in his (commentary) on the play says ‘A better interpretation is that Ion misunderstood Homer (Il. 23.270) when he said: “and as fifth prize he set a two-handled phialē untouched by fire.” ‘Because he (sc. Ion) took this (sc. phialē) to be a drinking vessel, whereas it is in fact an open bronze vessel that resembles a basin and is suited to having cold water poured into it. The phialē is daktulōtos in that it has depressions all around its interior, like those produced by fingers (daktuloi), or else because people grasp it with their fingers when they drink from it. Some authorities claim that a “phialē untouched by fire” is a drinking-horn, because fire is not used to produce drinking-horns. But perhaps he is referring metaphorically to the phialē as a drinking vessel.’ (Ion Chius T31a Leurini ap. Ath. 11.468e (43 = 204a)) Athenaeus elucidates Ion’s Homeric allusion, where Ion’s ἄχραντον πυρί (F1.2) glosses Homer’s ἀπύρωτος (Il. 23.270). He uses Didymus to correct Ion, who has misunderstood (παρήκουσεν Ὁμήρου) a phialē as an ekpōma, a fifth-century BC word.73 A phialē is not a drinking vessel as Ion thought, but a bronze basin used for cold water. Athenaeus also indicates that there were several opinions on the meaning of ἀμφίθετος, which qualifies phialē: firstly it signified a vessel without a base; or a vessel which could not be placed and supported on its base, but on its top; or a vessel able to be carried by two handles; or a vessel that could be positioned on either side.74 The title of the commentary is inferred from Ἴωνος Ἀγαμέμνονος ἐξηγητικόν (cf. Ion F1 Leurini) and ἐξηγητικόν suggests a work of one book.75 βέλτιον δὲ λέγειν indicates that the summaries of interpretations by Epigenes and others likely came from Didymus’ commentary, though the precise connection between the analysis of Epigenes and Didymus is unclear.76 This passage, however, shows overlaps between the exegesis of the Homeric poems and Didymus’ lexicographical works. Schmidt naturally attributed this passage to the Glossary of the Language of Tragedy, due to Hesychius’ stated sources (Ep. ad Eulogium line 4 Cunningham) 73 Stevens 2007: 259. Cf. Hdt. 9.80.1 (φιάλας τε καὶ ἄλλα ἐκπώματα) on the booty taken from the Persians after Plataea, where the former are shallow handleless bowls and the latter are generic drinking vessels. 74 The last opinion is by Aristarchus (see n. 78), who clearly thinks it cannot be a drinking vessel, because of the adjectives that accompany it, ἀμφίθετος and ἀπύρωτος, which he does not think can refer to a drinking vessel. Antimachus of Colophon (fr. 20.1 Matthews) uses the Homeric epithet ἀμφίθετος to describe κελέβειον, a container for honey, to indicate that it had two handles and that it was seated on one side, anticipating Aristarchus. Cf. Ath. 11.783b–c for a similar discussion on ἀμφικύπελλος, another vase. 75 Cf. Lycus Neapol. fr. 316 and Heraclid. fr. 348 Deichgräber. Paolo Scattolin per litteras notes that the Ἴωνος Ἀγαμέμνονος ἐξηγητικόν could be a section of the Counter-Explanations on Ion. 76 Cohn 1907: 65.25–30. See e.g. the difference between ἐπεὶ περιείληπται τοῖς τῶν πινόντων δακτύλοις in Didymus and the equivalent τὸ ἄμφωτον ποτήριον, εἰς ὃ οἷόν τε τοὺς δακτύλους διείρειν ἑκατέρωθεν in Epigenes, and how τῶν πινόντων is written after it and it is denied that δακτυλωτόν is an ἔκπωμα; Didymus presupposes the alternative τὸ ἐπιτήδειον εἰς ψυχρῶν ὑδάτων ὑποδοχὴν ἢ τὸ πρὸς ψυχροποσίαν εὔθετον.

Didymus on Attic tragedy



45

and Hesychius glosses δακτυλωτόν with ἔκπωμα, though Didymus is not named.77 The citation of the Iliad in the Athenaeus passage above refers to the phialē. Later on, Athenaeus (500f–502a), using the same section of the Iliad, analyses the word phialē. Aristarchus thought that the Homeric phialē was a shallow bowl-shaped cauldron, and ἀμφίθετος means that it can stand either upright or upside down on its rim, an interpretation which is also found in the scholia to the Iliad.78 Immediately following the summary of Aristarchus is a comment by Dionysius Thrax, who maintains that a phialē runs around (amphitheousa) in a circular shape.79 Athenaeus then quotes a more detailed discussion by Asclepiades of Myrlea, who glosses phialē as derived from πιάλη, and he thinks an ἀμφίθετος φιάλη is a large-sized cup. ‘ἡ μὲν φιάλη,’ φησί, ‘κατ’ ἀντιστοιχίαν ἐστὶ πιάλη, ἡ τὸ πιεῖν ἅλις παρέχουσα· μείζων γὰρ τοῦ ποτηρίου. ἡ δὲ [ἀμφίθετος καὶ] ἀπύρωτος ἢ ψυχρήλατος ἢ ἐπὶ πῦρ οὐκ ἐπιτιθεμένη, καθότι καὶ λέβητα καλεῖ ὁ ποιητὴς τὸν μὲν (Il. 23.702) ‘ἐμπυριβήτην’, τὸν δὲ ‘ἄπυρον’ (Il. 23.885)· κὰδ δὲ λέβητ’ ἄπυρον βοὸς ἄξιον ἀνθεμόεντα, τὸν δεχόμενον ἴσως ὕδωρ ψυχρόν, ὥστε καὶ τὴν φιάλην εἶναι χαλκίῳ προσεοικυῖαν ἐκπετάλῳ, δεχομένην ψυχρὸν ὕδωρ. τὴν δ’ ἀμφίθετον πότερα δύο βάσεις ἔχειν δεῖ νομίζειν ἐξ ἑκατέρου μέρους, ἢ τὸ μὲν ἀμφὶ σημαίνει τὸ περί, τοῦτο δ’ αὖ τὸ περιττόν; ὥστε λέγεσθαι τὴν περιττῶς πεποιημένην ἀμφίθετον, ἐπεὶ τὸ ποιῆσαι θεῖναι πρὸς τῶν ἀρχαίων ἐλέγετο. δύναται δὲ καὶ ἡ ἐπὶ πυθμένα καὶ στόμα τιθεμένη· ἡ δὲ τοιαύτη θέσις τῶν φιαλῶν Ἰωνική ἐστι καὶ ἀρχαία. καὶ Μασσαλιῶται ἔτι τιθέασι τὰς φιάλας ἐπὶ πρόσωπον.’ ‘The phialē is, via substitution of a letter, a pialē, that is, a cup that provides one with enough to drink (piein halis), because it is larger than an ordinary cup. As for a phialē that is amphithetos and apurōtos, the latter means that it has either been cold-forged or has never been placed over a fire, in the same way that Homer refers to a cauldron as either empuribētēs (“having gone over a fire”) or apuros (Il. 23.885): “And he set down an apuros cauldron, which was worth an ox and had a floral design worked into it,” ‘Perhaps meaning one used to hold cold water, so that the phialē in question resembles a shallow bronze vessel and is used to hold cold water. As for a phialē that is amphithetos, should we imagine that it has two bases, one on either side? Or is amphi used here in the sense of peri, which is in turn to be understood as peritton (“extraordinary”)? In that case, an exquisitely made phialē could be referred to as amphithetos, since the ancients used theinai (“to put, set, place”) to mean poiēsai (“to make”). But the reference might be instead to a phialē that is set on either its base or its lip, which is an ancient Ionian way of storing phialai. Even today, in fact, the Massaliotes stores their phialai upside down.’ (Asclep. Myrl. 8 Pagani ap. Ath. 11.501b–d) A phialē that is amphithetos and apurōtos: the latter means that it has either been cold-forged (ψυχρήλατος) or has never been placed over a fire, citing Homeric parallels of similar descriptions for other pots (Il. 23.702, 885). Asclepiades agrees with the Homeric echo of Ion’s verse, but quotes Il. 23.267 (ἄπυρον […] λέβητα), a different type of vessel. He points out that the rewriting of Ion is incorrect: the Homeric φιάλη was not an ἔκπωμα, so, as Didymus testifies, it is better to say that Ion misunderstood Homer (παρήκουσεν Ὁμήρου). 77 Hsch. δ 145 Cunningham. Schmidt 1854: 89, his Lex. Trag. fr. 8. Cf. also IDidym. 433.7–13 (270/269 BC). 78 Σ Hom. Il. 23.270a Erbse and Ath. 11.501b (for authorship). 79 Διονύσιος δ᾿ ὁ Θρᾲξ (fr. 28 Linke) τὴν στρογγύλην, τὴν ἀμφιθέουσαν κυκλοτερεῖ τῷ σχήματι. The meaning of ἀμφιθέουσα is unclear; see Linke 1977: 52–53 and Pagani 2007: 216–18.

46



Coward

There is an overlap between Didymus’ ἐπιτηδείως ἔχον πρὸς ὑδάτων ψυχρῶν ὑποδοχάς (‘and is suited to having cold water poured into it’) and Asclepiades’ comments here, especially on the purpose of the vessel in holding water (τὸν δεχόμενον ἴσως ὕδωρ ψυχρόν, ὥστε καὶ τὴν φιάλην εἶναι χαλκίῳ προσεοικυῖαν ἐκπετάλῳ, δεχομένην ψυχρὸν ὕδωρ).80 Didymus’ comments on Ion here seem to be informed by Asclepiades’ work on Homer, indicating that Didymus may have also discussed this matter in his works on the Iliad, and this discussion may go back to Epigenes. The discourse on the phialē continues on to comedy with another form of phialē (βαλανειόμφαλοι φιάλαι), which contains a reference to Didymus’ entry on this phrase in his Glossary of the Language of Comedy and states that he followed Asclepiades’ interpretation, indicating that Didymus made further use of Asclepiades’ works.81 The other work attested in Athenaeus is called The Counter-Explanations on Ion (Εἰς Ἴωνα ἀντεξηγήσεις).82 The discussion here is about whether the name for a musical instrument refers to a pipe or a lyre: πολλάκις καὶ αὐτὸς ἐν ἐννοίᾳ γίνομαι, μουσικῆς ὢν ἐραστής, περὶ τῆς μαγάδιδος καλουμένης, πότερον αὐλῶν εἶδος ἢ κιθάρας ἐστίν. ὁ μὲν γὰρ ἥδιστος Ἀνακρέων (374 PMG) λέγει που· ψάλλω δ’ εἴκοσι †χορδαῖσι μάγαδιν† ἔχων, ὦ Λεύκασπι, σὺ δ’ ἡβᾷς. Ἴων δ’ ὁ Χῖος ἐν Ὀμφάλῃ (TrGF1 19 F23 = F26b Leurini) ὡς περὶ αὐλῶν λέγει διὰ τούτων· Λυδός τε μάγαδις αὐλὸς ἡγείσθω βοῆς. ὅπερ ἐξηγούμενος ἰαμβεῖον Ἀρίσταρχος ὁ γραμματικός, ὃν μάντιν ἐκάλει Παναίτιος ὁ Ῥόδιος φιλόσοφος (T154 Alesse = F93 van Straaten) διὰ τὸ ῥᾳδίως καταμαντεύεσθαι τῆς τῶν ποιημάτων διανοίας, γένος αὐλοῦ φησιν εἶναι τὸν μάγαδιν, οὔτ’ Ἀριστοξένου (100 Wehrli) τοῦτ’ εἰπόντος ἐν τοῖς περὶ Αὐλητῶν ἢ ἐν τοῖς περὶ Αὐλῶν καὶ Ὀργάνων,83 ἀλλὰ μὴν οὐδὲ Ἀρχεστράτου· πεποίηται γὰρ καὶ τούτῳ δύο βυβλία περὶ Αὐλητῶν. οὐκ εἶπεν δὲ τοῦτο οὐδὲ Πύρρανδρος ἐν τῷ περὶ Αὐλητῶν (FHG iv.486), οὐδὲ Φίλλις ὁ Δήλιος (F6 FHG iv.476)· ξυνέγραψε γὰρ καὶ οὗτος περὶ Αὐλητῶν καὶ Εὐφράνωρ. Τρύφων (110 Velsen) δ’ ἐν δευτέρῳ περὶ Ὀνομασιῶν λέγει οὕτως. ‘ὁ δὲ μάγαδις καλούμενος αὐλός.’ καὶ πάλιν· ‘ μάγαδις ἐν ταὐτῷ ὀξὺν καὶ βαρὺν φθόγγον ἐπιδείκνυται, ὡς Ἀναξανδρίδης (36 PCG) ἐν Ὁπλομάχῳ φησίν· μαγάδι λαλήσω μικρὸν ἅμα σοι καὶ μέγα.84 τὴν ἀπορίαν οὖν μοι ταύτην οὐδεὶς ἄλλος δυνήσεται ἀπολύσασθαι, καλὲ Μασούριε, ἢ σύ.’ καὶ ὃς ἔφη· ‘Δίδυμος ὁ γραμματικὸς ἐν ταῖς Εἰς Ἴωνα Ἀντεξηγήσεσιν, ἑταῖρε Αἰμιλιανέ, μάγαδιν αὐλὸν ἀκούει τὸν κιθαριστήριον· οὗ μνημονεύειν Ἀριστόξενον ἐν πρώτῳ περὶ Αὐλῶν Τρήσεως (101 Wehrli) λέγοντα πέντε 80 Pagani 2007: 213. 81 8 and Asclepiades’ interpretation (fr. 9 Pagani = Crat. T*40 PCG); see on Didymus in Athenaeus see DiGiulio, this volume, 86–90. 82 The transmitted title is Πρὸς Ἴωνα ἀντεξηγήσεις (Counter-Explanations against Ion), which was amended by Wilamowitz ap. Kaibel 1887– 90 to εἰς. Schmidt 1854: 305 proposed ἐν ταῖς πρὸς Ἰώβαν Ἀντεξηγήσεσιν, based on Suda ι 399 Adler, which stated that Didymus wrote works against King Juba of Mauretania, but clearly the passage is about Ion’s interpretation of the word magadis, so Schmidt’s emendation should be discarded. 83 Olson 2019: 232 emends to ἢ περὶ Αὐλῶν καὶ Ὀργάνων. 84 Olson 2019: 233 has μάγαδιν λαλήσω μικρὸν ἅμα σοι καὶ μέγαν, following the transmitted text in A. See Olson 2019: vii–viii on how and why his edition presents quotations of ancient authors as rendered in the manuscript tradition and not in the critical editions of these fragmentary authors.

Didymus on Attic tragedy



γένη εἶναι αὐλῶν, παρθενίους, παιδικούς, κιθαριστηρίους, τελείους, ὑπερτελείους. ἢ ἐλλείπειν οὖν δεῖ παρὰ τῷ Ἴωνι τόν τε σύνδεσμον, ἵν’ ᾖ μάγαδις αὐλός ὁ προσαυλούμενος τῇ μαγάδιδι. ἡ γὰρ μάγαδις ὄργανόν ἐστι ψαλτικόν, ὡς Ἀνακρέων φησί (374 PMG), Λυδῶν τε εὕρημα. διὸ καὶ τὰς Λυδὰς ψαλτρίας φησὶν εἶναι ὁ Ἴων ἐν τῇ Ὀμφάλῃ διὰ τούτων (TrGF1 19 F22 = F26a Leurini)· ἀλλ’ εἶα, Λυδαὶ ψάλτριαι παλαιθέτων ὕμνων ἀοιδοί, τὸν ξένον κοσμήσατε.’ ‘I often wonder whether what is referred to as a magadis is a type of pipe or a lyre. For the delightful Anacreon says somewhere: “I play the harp, holding a twenty-stringed magadis, Leucaspis, but you are young and beautiful.” Ion of Chios in his Omphale refers to them as if they were pipes, in the following passage: “And let the Lydian magadis-pipe initiate the noise!” In his explication of this iambic line, the grammarian Aristarchus—the philosopher Panaetius of Rhodes used to refer to him as a mantis, because he could easily divine the point of a poem—claims that the magadis is a type of pipe, even though Aristoxenus does not say this in either his On Pipe-Players or his On Pipes and Instruments, and neither does Archestratus; the latter also produced an On Pipe-Players in two books. Pyrrhandrus also omits any mention of this instrument in his On Pipe-Players, as does Phillis of Delos; for he too composed an On Pipe-Players, as did Euphranor. But Tryphon in Book 2 of On Terminology says the following: “the pipe referred to as a magadis.” And again: “The magadis produces high and low tones simultaneously, as Anaxandrides says in the The Hoplite-Trainer: ‘I’ll speak along with you soft and loud, like a magadis.’” No one other than you, therefore, my good Masurius, will be able to resolve this puzzle for me.’ And Musurius said: ‘The grammarian Didymus in his The Counter-Explanations on Ion, my friend Aemilianus, takes a magadis to be a pipe played to accompany a lyre; (he also claims that) Aristoxenus mentions the instrument in Book One of his On the Boring of Pipes, where he asserts that there are five types of pipes: girls’ pipes, boys’ pipes, pipes played to accompany a lyre, adult-pipes, and more-than-adult pipes. Alternatively, the conjunction must be missing from (the text of) Ion, meaning that it ought to read: “the magadis and the pipe”, which is played along with the magadis. For a magadis is an instrument that resembles a harp, as Anacreon says, and was invented by the Lydians. This is why Ion in his Omphale refers to the Lydian women as harp-players, in the following passage: “But come, Lydian women who play the harp, singers of ancient hymns—tend to the stranger!”’

47

48



Coward

(Ion Chius T31b Leurini ap. Ath. 14.634e–f (°206 = 47)) Athenaeus’ first speaker, Aemilianus, in this passage presents the overall problem: whether the magadis is a pipe or stringed instrument, and if it was Lydian in origin.85 This topic remains an insoluble question.86 Athenaeus cites examples from Anacreon and Ion to show where the word is used to mean one thing and then another.87 He then turns to scholarly authorities, starting with Aristarchus, who was in favour of the pipe and seems to have interpreted this line of Omphale as a separate work on Ion or as part of his exegesis on Anacreon.88 An anonymous source notes that this definition was not found in several authors of works on pipe instruments (Aristoxenus, Archestratus, Pyrrhandrus, Phillis of Delos, and Euphranor). Didymus’ contemporary, Tryphon (c.60–10 BC) reaffirms Aristarchus’ definition and cites Anaxandrides for proof of its vocal range rather than referring to another learned authority. The second speaker, Masurius, puts forward Didymus’ proposal, which was that the magadis was a pipe instrument, but one that accompanies a lyre (μάγαδιν αὐλὸν ἀκούει τὸν κιθαριστήριον).89 Didymus then claims that such a definition is found in another work of Aristoxenus (On the Boring of Pipes), which had not been cited earlier, and in referring to this previously uncited work, Didymus makes a tacit defence of Aristarchus. All these interpretations are based on literary references alone. Εἰς Ἴωνα Ἀντεξηγήσεσιν indicates that Didymus wrote a dedicated treatise responding to the interpretations of other scholars (including Aristarchus) on the works of Ion or on his Omphale in particular.90 The discussion then continues to consider, if it is a harp, whether it is the same as the pēktis. The confusion over magadis is that the word applies both to the lyre and aulos, which could be understood as a way of producing music in octave concords rather than as a type of instrument.91 Didymus’ interpretation suggests that his Counter-Explanations on Ion contained a summary of the doxography as outlined by Aemilianus before Masurius supplies Didymus’ interpretation, as Didymus refers to an Aristoxenan work not cited in the discussion so far. ἢ ἐλλείπειν οὖν δεῖ παρὰ τῷ Ἴωνι τόν τε σύνδεσμον is likely a continuation of Didymus’ explanation, as ὁ προσαυλούμενος τῇ μαγάδιδι would correspond to Aristoxenus’ αὐλὸς κιθαριστήριος. A further suggestion reported by Aemilianus proposed an ellipsis of τε (μάγαδις αὐλός ) rather than emending the particle into the text, which makes the magadis a stringed instrument, using literary sources to support this, and reaffirms its Lydian origins with another quote from Ion’s Omphale.92 Another likely fragment of Didymus’ exegesis on Ion and/or Glossary of the Language of Tragedy is from Stephanus of Byzantium: Πνύξ· τὸ παρ’ Ἀθηναίοις δικαστήριον, περὶ οὗ φησιν Ἡρωδιανὸς ἐν τοῖς Περὶ παθῶν οὕτως “ἔστιν οὖν Πύκν κύριον. οὐ δύναται δὲ καταλῆξαι, καὶ ὑπερετέθη τὸ ν, καὶ ἐτράπη τὸ κ εἰς ξ, Πνύξ. ὑπέρθεσιν ἄρα ἔχει διὰ τὴν κατάληξιν. ἐν δὲ τῇ γενικῇ ἀπέλαβε τὴν τάξιν μετὰ τοῦ κ Πνύξ Πυκνός, ‘τῆς 85 Ath. 14.634c–637a, foreshadowed at 4.182d–e. 86 See Barker 1984: 294 and West 1992: 79 on this passage. 87 The transmitted text of Ion has been questioned. West 1983: 79 proposed that [τε]? μάγαδις Λυδὸς ἡγείσθω βοῆς was the correct reading and that Athenaeus’ quotation was meaningless. A further suggestion is also offered through textual emendation of the citation of Ion’s Omphale suggested by Aemilianus: μάγαδις αὐλὸς ἡγείσθω > μάγαδις αὐλός [ἡγείσθω]. add. Kaibel. The same line of Ion is also cited by Hesychius (μ 3 Cunningham), who follows the interpretation of Didymus. 88 Ion T26 Leurini. On Aristarchan exegesis on Anacreon, cf. e.g. Ath. 15.671f–672a: Ἀρίσταρχος ὁ γραμματικώτατος, ἑταῖρε, ἐξηγούμενος τὸ χωρίον ἔφη ὅτι … (‘Aristarchus, the ultimate grammarian, explaining the passage, my friend, said that …’); NB the use of ἐξηγούμενος here and the passage under discussion. On whether Aristarchus wrote an edition and/or commentary on Anacreon, see Hadjimichael 2019: 235 n. 62 for references. 89 Elsewhere, Didymus (270ab) claims that, rather than playing a lyre, some people make it a practice to strike shells or potsherds against one another to produce a rhythmic sound to accompany dancers, per Ar. Ran. 1304–06. 90 ἀντεξήγησις itself is a hapax, a term coined by Didymus or the only surviving instance? Cf. Didymus’ Περὶ τῆς Ἀρισταρχείου διορθώσεως (50–68) for a similarly orientated treatise, with Pagani, this volume, 10–20. 91 See Barker 1988 and West 1992: 73 n. 110, who independently came to the same observation. S. 238 TrGF4 ap. Ath. 14.637a (after our passage of Didymus) further suggests this notion. Cf. also Hsch. μ 5 Cunningham and Phot. μ 2 Theodoridis. 92 See 176.

Didymus on Attic tragedy



49

Πυκνὸς τὰς ἡνίας’‚ ‘τοῦ λίθου τοῦ ’ν τῇ Πυκνί’, καὶ ‘Πύκνα’. κέκληται δὲ παρὰ τὸ πυκνὸν τῶν πάλαι συνῳκισμένων οἰκιῶν. ὁ ἐν τούτῳ οἰκῶν Πυκνίτης. λέγεται καὶ Πυκναία. Δίδυμος δέ φησιν, ὅτι ‘τὴν Πύκνα ἂν εἴη λέγων. τὸ δὲ Πυκναία ἀπὸ τοῦ πυκνή παρῆκται, ὅπερ οὐκ εἴρηται· οὐδεὶς γὰρ εἶπε προσηγορικῶς ἢ κυρίως τὸ πυκνή· τὰ γὰρ τοιαῦτα ὡς ἐπίπαν ἀπὸ μακροκαταλήκτων παράγεται, εὐνή εὐναία, Γυρή Γυραία, ἀνάγκη ἀναγκαία. τὸ ὡς ἐπίπαν εἴρηται διὰ τὸ ἅμαξα ἁμαξαία καὶ κέρας κεραία.’ Pnyx, the court of the Athenians, about which Herodian in the Περὶ παθῶν says the following: ‘So there is Πύκν as a proper . It cannot end with ν, and so the ν has been transposed, and the κ has changed into a ξ, so you say Πνύξ. thus has a transposition due to the ending. In the genitive it kept the order with κ, Πνύξ Πυκνός, “the reins of the Pnyx” (Ar. Eq. 1109), “the stone in the Pnyx” (Ar. Pax 680) and “ Pnyx” (Ar. Ec. 283 or Ar. Th. 658). The name comes from the dense crowds (πυκνόν) of the houses that were once crowded together there. Anyone who lives in this place is called Πυκνίτης. It is also called Πυκναία, Didymus says: ‘ should have said Πνύξ . The Πυκναία comes from πυκνή, which does not exist; because nobody says πυκνή as an appellative or as a proper name. Such are usually derived from words with a long-syllable ending, εὐνή (“bed”) εὐναία (“couch”), Γυρή Γυραία, ἀνάγκη (“necessity”) ἀναγκαία (“necessity”). The restriction “usually” is due to ἅμαξα (“wagon”) ἁμαξαία (“truck”) and κέρας (“horn”) κεραία (“horn/bow”).’ (St. Byz. π 188 Billerbeck–Neuman-Hartmann (Πνύξ) (205 = 48)) The etymology of Πνύξ was debated in antiquity, though all the definitions share the notion of a concentration of persons or buildings.93 After Herodian’s analysis of the word and its etymologies, Stephanus offers an alternative name for the Pnyx as Πυκναία. Then Didymus is quoted, who says that Ion’s Πυκναία meant Πνύξ. Hesychius supplies the information that Ion used Πυκναία, although the word cannot be assigned to any particular work or play.94 This suggests, given the word is glossed in Hesychius and Didymus’ discussion of the proper usage of the noun with cited parallels, that it was used in a Commentary on Ion or in the Counter-Explanations, and/or Glossary of the Language of Tragedy. If it is from a commentary, then this suggests a commentary on a tragedy of Ion other than Agamemnon. Papyri may yield some further Didymean work on Ion, but these claims are not secure. P.Oxy. 13.1611 fr. 16 probably refers to a discussion about the Omphale (F37* Leurini = TrGF1 19 F33a), and P.Oxy. 13.1611 fr. 2 col.i.121–27 quotes lines of the play attested elsewhere (F22 Leurini = TrGF1 19 F17a). P.Oxy. 13.1611 preserves an unnamed scholar’s notes on literary problems including references to the Omphale. Parts of the papyrus may contain material that goes back to Didymus, per Arrighetti, but doubts remain due to the miscellaneous nature of the papyrus.95 Another papyrus, a commentary on Aristophanes (possibly on the Anagyros) very likely contains Didymean material.96 P.Oxy. 35.2737 col.i.19–27 offers an interpretation of a section of a play and points to possible intertexts between the Aristophanic passage and other poets. In essence, Aristarchus thought the Aristophanic passage under discussion was imitating Terpander (S6 SLG), Euphronius thought Ion (S316 SLG = fr. 85 Leurini/7 Valerio), and an unnamed author of a Παραπλοκή 93 See McInerney 1994: 31 for a summary and references. I add Σ Ar. Eq. 42b, 165b Mervyn Jones–Wilson; Σ Pl. Criti. 112a bis Greene; Σ Luc. 21.11 Rabe; Phot. π 976, 1527–28 Theodoridis. In Photius (π 1526–27 Theodoridis) there is Πυκν†ή† for the Pnyx, but this is likely a corruption of the dative Πυκνί; see Harp. π 125 Keaney  Phot. π 1526–27 Theodoridis with app. crit. 94 This fragment could be from Ion’s Epidemiai (Visits), his prose memoir; cf. Ion BNJ 392 F14, where the Spartans appeal to Athens for support during the Helots’ revolt (463/2 BC) and Cimon persuades the Athenians (presumably on the Pnyx) to help them. Hsch. π 4331–32 Hansen preserves two glosses of Πυκναία; the first is for πτερά and the second for Pnyx as used by Ion. On the former, see Hom. Od. 5.53, Il. 11.454, 23.879; Sapph. fr. 1.11 Voigt; on the latter Ion is the only example. 95 On the papyrus, see Grenfell and Hunt 1919; Arrighetti 1968 = 1987: 204–28 (reprinted with minor changes); Trojahn 2000: 208; and the cautionary notes by Montana in CLGP II.4: 115 and Perrone in CLGP I.1.4 (ed. alt. 2012): 238–39. 96 P.Oxy. 35.2737 = CLGP Ar. 27. See Benuzzi, this volume, 56; Montana in CLGP ad loc.; Valerio 2013: 112–15.

50



Coward

thought Alcman (S2 SLG). The author of the commentary thought the quote derived from a Homeric Hymn (21.1). Clearly the phrase was a topos. The compilation of opinions suits Didymus, but authorship is not secure, as it could be another scholar or a compilation of scholarly opinions. 5. CONCLUSIONS: DIDYMUS ON ATTIC TRAGEDY The extant fragments of Didymus’ works on tragedy discussed above concern several problems of both a textual and non-textual (interpretative) nature. They cover a range of approaches found in ancient literary scholarship and demonstrate the resources available to Didymus and how he used them. They show him continuing the methodologies of his predecessors, but also making his own contributions. He is not just compiling the work of his predecessors but engaging and building upon these activities to offer his own interpretation (e.g. 43 = 204, 47 = °206). Sometimes these contributions are wrong, but further investigation demonstrates the thinking behind these ideas and how and why he erred (e.g. 176, 183). Some examples show him underlining the limits of interpretation based on the available evidence, some problems which remain insoluble to this day (e.g. 179). The passages discussed above also show overlaps between different sorts of scholarly works where lines or passages are analysed in different formats, such as the glossary (a reference work), the commentary (an interpretative work), and the treatise (a discursive work). Not only do these indicate that Didymus recycled material but also the working methods of the ancient scholar in compiling and analysing literary works and earmarking materials that are better suited for one type of scholarly work than another. Difficulties remain on securely identifying Didymean material where he is not named; some examples can be proposed on methodological grounds or stylistic convention but must be classified as probable fragments of Didymus rather than as certain fragments. Didymus’ contribution to the study of Greek tragedy concerned matters of literary and textual criticism (e.g. 176, 179, 182, 205 = 48), interpretations based on grammar and syntax (e.g. 176, 183, 205 = 48), explicatory comments on the meaning of words and passages (e.g. °4, 44 = 184), and antiquarian researches using other verse and prose works to support these interpretations (e.g. °206 = 47). The surviving fragments of Didymus’ works on Sophocles and Ion of Chios demonstrate him utilizing a wide range of approaches and resources, and the enduring interests of these literary works into the Imperial era.

Bulletin of the Institute of Classical Studies, 2020, 63, 51–61 doi: 10.1093/bics/qbaa017 Original article

Chapter Four Didymus and comedy* Federica Benuzzi Cà Foscari University, Venice, Italy

ABSTRACT Didymus’ commentaries on the comic playwrights and his Comic Vocabulary responded to the interests of the readership of Attic comedy primarily in two ways: by summarizing the opinions of previous scholars and by offering a wide range of explanations, useful also to less specialized readers. Although his exegesis of comedy is now preserved only through quotations (mainly in the scholia to Aristophanes), it is still possible to identify the main features and interests of Didymus’ interpretative work and highlight its relevance for both ancient and contemporary readers of Greek comedy.

The exegesis of Attic comedy played an undeniably important role within Didymus’ tireless interpretative work. Not only did he write running commentaries to Aristophanes and other playwrights, but he also compiled a Comic Vocabulary that had a strong influence on subsequent lexicographers. Living in the first century BCE, he was able to make full use of a long and illustrious tradition of studia comica that had its roots in Aristotle and the Peripatetic school1 and had found renowned representatives in Alexandria (from Callimachus, Lycophron, and Eratosthenes to Aristophanes of Byzantium, Aristarchus, and their pupils),2 as well as in other centres such as Pergamum and Rhodes3 throughout the Hellenistic era. Didymus’ position at the end of such a remarkable succession of studies was one of the main reasons that caused his exegetic endeavour to be seen, at best, as uncritical derivative work, of worth only as a conduit of more ancient information.4 The transmission of Didymean scholarship on comedy through quotations in other works and in papyri will be the focus of the first part of this chapter, while the second part will consist of an overview of the * I am very grateful to the editors of this volume and to the reviewers for the precious observations that greatly improved this chapter. Unless otherwise stated, the translations are my own. Numbers in bold refer to testimonia and fragments in the Checklist (pp. 95–120). 1 Examples are Aristotle’s catalogue of theatrical productions, the Διδασκαλίαι (frr. 415-462 Gigon [618–30 Rose]; see Jachmann 1909; Pfeiffer 1968: 81; Bagordo 1998: 19–20), Theophrastus’ On Comedy (frr. 1–2 Bagordo; see Nesselrath 1990: 149–61; Bagordo 1998: 22; Novokhatko 2015: 55), Dicaearchus’ On the Dionysiac Agones (frr. 2–7 Bagordo; see Martini 1903: 555–56; Nesselrath 1990: 162–63; Bagordo 1998: 24; Novokhatko 2015: 57), Chamaeleon’s On Comedy (frr. 10–11 Bagordo; see Podlecki 1969: 120–24; Arrighetti 1987: 141–59; Nesselrath 1990: 163–64; Bagordo 1998: 26–28; Martano 2012; Schorn 2012; Novokhatko 2015: 57) and Eumelus’ On Old Comedy (fr. 1 Bagordo; see Nesselrath 1990: 165; Bagordo 1998: 32). 2 Overviews e.g. in Pfeiffer 1968: 123–233; Nesselrath 1990: 172–87; Montana 2015b: 107–43. 3 See e.g. Pfeiffer 1968: 234–51; Broggiato 2014; Montana 2015b: 73–74, 143–53; Matijasic 2020, 21–31; Coward forthcoming. 4 See e.g. White 1914: xxv: ‘he was a collector, not a scholar of original and creative genius’. Even more negative assessments are found in Rutherford 1905: 432 and Roemer 1908. C The Author(s) 2021. Published by Oxford University Press on behalf of the Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected].



51

52



Benuzzi

content of the fragments, according to their kind, in order to show that Didymus was a commentator in his own right and that his interpretations were not only grounded in the great tradition of the past, but also responded to the interests of the readership of Attic comedy at his time. Indeed, between the late Republican and the early Augustan era, the focus of Alexandrian scholars was ‘not so much on the library and its contents any more, but on the needs of the community ‘outside’, whether in Egypt, Rome or elsewhere’:5 Didymus’ exegesis was not aimed at the small intellectual circle of the Mouseion, but at a much broader and less specialized group of readers.6 In all likelihood, their interest in Old Comedy was closely linked to the paradigmatic value of the genre within the rhetorical training, both in terms of exempla of political rivalry and, more generally, in terms of a polemical, agonistic mode of discourse.7 This perspective allows a better understanding of the main features of Didymus’ work on comedy: on the one hand its focus on lexical elucidation, explanation of references to historical figures and events, and generic textual clarification (see below); on the other hand its inherently compilatory nature, stemming from the need to make his predecessors’ scholarship accessible for this new, broader readership. Since he relied heavily on the previous philological tradition, he is responsible for the preservation of numerous ancient interpretations: he quoted explicitly, for instance, from Lycophron’s On Comedy8 and from Ammonius’ Κωμῳδούμενοι.9 Besides these, citations from Euphronius’ and Callistratus’ Aristophanic commentaries were in all likelihood embedded in his work, as suggested (but not proved beyond doubt) by a number of fragments.10 Moreover, some rich doxographies— including citations from Eratosthenes, Herodicus, Lycophron, Polemon, and Chaeris—are probably owed to him.11 However, the sources of Didymus’ exegesis of comedy were not exclusively philological. On the contrary, he made extensive use of historians and antiquarians, such as Androtion and Autoclides,12 and he probably quoted Aristotle’s Διδασκαλίαι, as well as Antigonus’ paradoxographical collection.13 Another typical trait of Didymus’ exegesis of comedy is the accumulation of parallels from other (now lost) comic plays.14 Overall, the grammarian played a central role as vehicle of transmission of a wide range of ancient information that would not have been preserved otherwise. In fact, his contribution in this sense must have been much more extensive than what the evidence suggests, since in most cases Didymus’ name was probably obliterated due to epitomation. Didymus wrote running commentaries (hypomnēmata) to Aristophanes’ plays, basing his work on the text established by Aristophanes of Byzantium.15 Since the scholia to Aristophanes are the most important means of transmission of Didymean exegesis of comedy, the eleven canonical plays preserved by the manuscripts are inevitably 5 Hatzimichali 2013: 177. For an overview on the reception of Alexandrian philology in Late Republican and Augustan Rome, see Montana 2016: 105–13. 6 On the basis of the provenance of papyrological finds of Old Comedy from Egypt, this readership can be identified with the urban elites (see Del Corso 2017: 237). Moreover, the chronological distribution of the papyri of Old Comedy places the beginning of the popularity of the genre approximately in Didymus’ time (first century BCE – first century CE), with a sharp increase in the Imperial age (see Perrone 2011; Del Corso 2017). 7 The paradigmatic value of Greek Old Comedy for Roman literature of the Republican age (satire and oratory in particular) has been investigated by Ruffell 2014. 8 See Pellettieri 2020. 9 See Ath. 11.501d, Lyc. fr. 25 Strecker (8); Phot. τ 219 Theodoridis, Lyc. fr. 134 Strecker (36a); Σ Av. 1297–99 Holwerda (247; see below). 10 See e.g. Σ Av. 530b–a–c–f (230), 1379a–b Holwerda (249). 11 See Σ Ar. Ran. 1028aα–bα–c–eα–f Chantry (269), discussed by Montana 2017: 215–21; Ath. 4.139c–140b (15). 12 See Σ Ar. Av. 13a–b Holwerda (222); Harp. ο 25 Keaney (309a). On Didymus’ scholarship on and frequent use of historians, see Montana in this volume. 13 See Σ Ar. Av. 299a–aα–aβ–b (227), 1379a–b Holwerda (249). 14 See e.g. Eupolis’ Demes (fr. 136 K.–A., 39 Telò) and Flatterers (fr. 177 K.–A.) in Σ Ar. Av. 876f Holwerda (237); Eupolis’ Taxiarchoi (fr. 282 K.–A.) and Atalantai (Call. fr. *4 K.–A.) in Σ Ar. Av. 1294 Holwerda (245); and Platon’s Victories and Man in Terrible Pain (frr. 85 and 116 K.–A.), Phrynichus’ Ephialtes and Grass-cutters (frr. 4 and 43 K.–A.), and Metagenes’ Homer (fr. 12 K.–A.) in Σ Ar. Av. 1297–99 Holwerda (247). 15 See Boudreaux 1919: 94.

Didymus and comedy



53

the most represented among his fragments. These are predominantly found in the scholia to Birds (thiry-three fragments) and Frogs (fourteen fragments). However, Didymus’ commentary covered in all likelihood other plays of the Alexandrian edition of Aristophanes (such as Old Age, °275), if not all of them.16 Moreover, both a hypomnēma to at least two plays by Menander and one to Phrynichus’ Cronus are attested (see below).17 Therefore, it can be supposed that Didymus’ commentaries spanned the main poets of Old Comedy, as well as Menander, although the scanty evidence outside the Aristophanic scholia does not allow further speculation. Overall, with sixty-six quotations, Didymus is by far the most frequently cited ancient commentator in the Aristophanic scholia. Nevertheless, he is not mentioned in any of the subscriptiones that name the sources of the scholia to Birds, Peace, and Clouds in MS V (ad Av. παραγέγραπται ἐκ τῶν Συμμάχου καὶ ἄλλων σχολίων, ad Pac. παραγέγραπται ἐκ Φαείνου καὶ Συμμάχου, ad Nub. παραγέγραπται ἐκ τῶν Φαείνου καὶ Συμμάχου καὶ ἄλλων τινῶν). This is probably due to the fact that his commentary to Aristophanes did not survive into Late Antiquity as a self-standing hypomnēma, but was soon replaced by the work of later Aristophanic commentators, who—by excerpting and reusing Didymus’ interpretations—prevented the survival of his commentary in its original redaction, and at the same time constituted the only means of its fragmentary transmission.18 A primary role in this process, as suggested by the subscriptiones, must have been played by Symmachus, whose scholarly activity can be placed in the first to second century CE.19 Given the importance of Didymus’ commentary within the Aristophanic scholia, modern critics have tried to detect further anonymous quotations from that work: comments ascribed exclusively to Symmachus have been often suspected of preserving Didymean material, especially those containing quotations from other comic playwrights and Alexandrian commentators.20 The search for unascribed fragments has also been carried out by comparing the Aristophanic scholia with other works that relied on Didymean material, such as Zenobius’ paroemiographical collection21 and Hesychius’ lexicon (see below)22 or—with more far-fetched attempts—by looking for allegedly distinctive linguistic features, above all the adverb μήποτε.23 The scholia to Aristophanes are not the only means of transmission of Didymus’ exegesis of comedy. An important role is played also by Athenaeus’ Learned Banqueters, which preserves material drawn from the hypomnēma to Aristophanes’ Wealth (see Ath. 2.67d = 26): Ἀριστοφάνης δὲ ἐν Πλούτῳ φησίν· ‘ὄξει διέμενος Σφηττίῳ’ (720). Δίδυμος δ’ ἐξηγούμενος τὸ ἰαμβεῖόν φησιν· ‘ἴσως διότι οἱ Σφήττιοι ὀξεῖς’. Aristophanes says in Wealth: ‘Drenched in Sphettian vinegar’. Didymus explains the verse by saying: ‘perhaps because the Sphettians are sharp-tempered’.24 16 As hypothesized by Boudreaux 1919. On the number of Aristophanes’ plays (i.e. approximately fifty), see Proll. de com. III 41, XXXa 7–20 Koster. See also e.g. Novati 1879; Boudreaux 1919: 14–17; Kassel and Austin 1984: 4–5. 17 See Et. Gud. p. 338.20–25 Sturz (°278) and Ath. 9.371f (°279; see below) respectively. On the number Didymus’ hypomnēmata to Menander, see Theodoridis 1973. 18 For an overview of the debate on the genesis of the Aristophanic scholia (and of scholiography in general) and a detailed analysis of the available evidence, see Montana 2006 and 2011. 19 He was active before the redaction of Herodian’s On Lexical Singularity, where he is quoted (Περὶ μονήρους λέξεως, II p. 245 Lentz). His name occurs along with Didymus’ thirteen times in the scholia to Birds and once in the scholia to Wealth. On Symmachus’ dependence on Didymus in general, see Schneider 1838: 97–99; Schmidt 1854: 289; Schnee 1879: 35–46; Schauenburg 1881: 5–33; White 1914: l–liii; Boudreaux 1919: 153–58; Dunbar 1995: 40–41; Montana 2003. 20 See e.g. Σ Av. 1296a Holwerda and Schauenburg 1881: 11–13. 21 See e.g. Meiners 1890: 16–17; Boudreaux 1919: 104–05. 22 See e.g. Schmidt 1854: 298–99; Schnee 1879: 46; Schauenburg 1881: 1820; Novati 1882; Meiners 1890: 14–16; Boudreaux 1919: 103– 04; Montana 1996: 30 n. 64. 23 Wilson 1984: 93 (see also Montana 1996: 30 n. 64) has cautioned against this approach, generally accepted by most scholars before him (see e.g. Schmidt 1854: 212; Schnee 1879: 37–38; Schauenburg 1881: 11–12; Meiners 1890: 10–14; and Boudreaux 1919: 110–18). 24 The same interpretation (without Didymus’ name) is found in Σ Ar. Pl. 720iα–β Chantry. The translation of all of Athenaeus’ passages is by Olson (2006–12: I 381, IV 214–15, V 355 and 369–70).

54



Benuzzi

and from the one to Phrynichus’ Cronus (see Ath. 9.371f = 36a):25 οἳ δὲ τὸ γήθυον καλούμενον τοῦτό (sc. τὴν γηθυλλίδα) φασιν εἶναι, οὗ μνημονεύει Φρύνιχος ἐν Κρόνῳ (fr. 12 K.–A.)· ὅπερ ἐξηγούμενος δρᾶμα Δίδυμος ὅμοιά φησιν εἶναι τὰ γήθυα τοῖς λεγομένοις ἀμπελοπράσοις, τὰ δ’ αὐτὰ καὶ γηθυλλίδας λέγεσθαι. But other authorities claim that this (i.e. the gēthullis) is what is referred to as gēthuon, which Phrynichus mentions in Cronus. In his explication of the play, Didymus says that gēthua resemble what are called ampeloprasoi (literally ‘grapevine-leeks’), and that the same vegetables are also referred to as gēthullides. Other pieces of Didymean exegesis preserved by Athenaeus are not closely linked to specific comic passages, but rather deal with terms frequently occurring in comedy, for instance names of different drinking vessels, as in Ath. 11.484f (18, on the λαβρώνιος) and 487c–f (21): μάνης. ποτηρίου εἶδος. Νίκων Κιθαρῳδῷ (fr. 1 K.–A.)· ‘καὶ πάνυ τις εὐκαίρως “προπίνω, φησί, , / πατριῶτα.” μάνην δ’ εἶχε κεραμεοῦν ἁδρόν, / χωροῦντα κοτύλας πέντ’ ἴσως. ἐδεξάμην.’ παρέθετο τὰ ἰαμβεῖα καὶ Δίδυμος καὶ Πάμφιλος. καλεῖται δὲ μάνης καὶ τὸ ἐπὶ τοῦ κοττάβου ἐφεστηκός, ἐφ’ οὗ τὰς λάταγας ἐν παιδιᾷ ἔπεμπον. Manēs. A type of cup. Nicon in The Citharode: ‘And at just the right moment someone says: “I drink to you / my fellow-countryman!”. He had a large ceramic manēs / that held maybe five kotylai. I accepted it.’ Didymus and Pamphilus both cited these lines. The term manēs is also used for the object that rests on top of the kottabos-stand, at which they used to throw their wine-lees in the course of the game. In all likelihood, these quotations derive ultimately from Didymus’ lexicographic work, the Comic Vocabulary (Λέξις κωμική), a lemmatized collection of explanations of specific terms found in comedy. This lexicon must have shared much of the lexicographical material present in the hypomnēmata on the comic playwrights: whether it developed as a by-product of the running commentaries or alongside them, the redaction of the Comic Vocabulary must have involved a massive compilatory effort, possibly carried out by a wider scholarly entourage, rather than single-handedly by Didymus.26 The work was a primary source for the lexicographers of the Imperial age: not only did Pamphilus (see Ath. 11.487c–f above)27 and his epitomator Diogenianus (see Hsch. ep. ad Eul. 1–8) have it as their source, but Galen also made his own epitome of the work ‘in six thousand lines’.28 Moreover, some suspect that the Λέξις κωμική was among the sources of the lost rhetorical lexicon of the grammarian Iulianus.29 Similarly, Harpocration might have made use of the Comic Vocabulary alongside Didymus’ commentaries to the orators,30 as seems to be suggested by the presence of the same quotation (again regarding a drinking vessel, the

25 See Stama 2014: 14; Montana 2018: 518–19. 26 On its relationship with the commentaries, see Schmidt 1854: 27–29; Boudreaux 1919: 102; see also Wilamowitz 1889–95: I 165. That Didymus availed himself of ‘assistants’ is a suggestion of West 1970: 288 n. 2. A similar hypothesis for Galen’s Hippocratic lexicon is found in Perilli 2017: 124–25. 27 See Schmidt 1854: 74: ‘alterum καὶ innuit fontem Pamphili fuisse Didymum’. See also Rohde 1870: 7; Cohn 1903: 463; Wendel 1949: 339; Nesselrath 1990: 77. 28 The book then was lost in the fire that deprived the physician of a considerable part of his library (see Gal. De indol. 23b–24b). 29 See Cohn 1913; Alpers 1981: 117–18; Ucciardello 2006, 2013: 11 n. 3. 30 Schmidt 1854: 39–40 assigned five Didymean quotations in Harpocration’s lexicon to the Comic Vocabulary, whereas Gibson 2002: 138 considered only four of them to be derived ‘from a work of Didymus having nothing to do with Demosthenes’.

Didymus and comedy



55

κυμβίον) in Ath. 9.481f and in Harp. κ 92 Κeaney (17).31 The Comic Vocabulary played an important role for Late Antique and Byzantine lexicographers as well, whether as a primary source or, more likely, through the mediation of lexicographers of the Imperial age. Actually, the title Λέξις κωμική is clearly linked to Didymus’ name only in the etymological tradition, in an entry that goes back to the fifth-century CE grammarian Orus; see Et. Gen. s.v. καρύκκη (13):32 καρύκκη: ἕψημά τι δι’ αἵματος καὶ ἀρτυμάτων παντοίων. ὁ δὲ Δίδυμος Λύδιον βρῶμα φησὶν ἐκ πολυτελοῦς σκευασίας συγκείμενον· ἐγὼ δὲ οἶμαι ‘καρύκην’ αὐτὸ λέγεσθαι, ἐπειδὴ μέλαν ἐστίν· φησὶ γὰρ ἐν τῇ κωμικῇ λέξει οὕτω, ‘Καρυκεῖον, καρυοβαφές’· καὶ ‘κάρυος ῥόος οἷον μέλανος ῥόος’. οὐκοῦν ἐκ τούτου οἶμαι λέγεσθαι καρύκη. Ωρ. karykkē: something boiled with blood and all sorts of seasonings. Didymus says it is a Lydian food made with a refined preparation. I believe that the spelling is karykē, because it is black. Indeed, in the Comic Vocabulary he says the following: ‘karykeion (means) “dipped in walnut-juice”’ and ‘“walnut flow”, that is to say “black flow”’. Therefore, on this basis, I believe the spelling is karykē. Orus.33 In addition, Hesychius explicitly mentions Didymus among his sources (ep. ad Eul. 3–4 Cunningham) and quotes him in several entries concerning words found in Cratinus and Aristophanes (β 1152, κ 3661, μ 1705, ο 409 Cunningham, σ 147, 742 Hansen = 9, 16, 22, 25, 30, 33).34 The fifth-century CE lexicographer Orion seems to have used the Comic Vocabulary as well, seemingly through a later Aristophanic hypomnēma.35 As far as later lexicography is concerned, Photius preserves several entries ultimately deriving from the Comic Vocabulary (e.g. η 152, ο 127, 377, σ 148, τ 219 Theodoridis = 12a, 24, 309b, 31, 36a).36 Traces of Didymean exegesis of comedy can be found in non-Aristophanic scholia as well, especially in the scholiastic corpus to Apollonius of Rhodes: the grammarian is mentioned twice, with regard to words occurring respectively in Eupolis (Σ 1.1134–39b Wendel = 29) and Menander (Σ 4.1058 Wendel = 35). Moreover, the scholia to Apollonius make two references to an anonymous Comic Vocabulary (Σ 4.973 Wendel, ἐν τῇ Κωμικῇ λέξει τῇ συμμίκτῳ and 1613–16c Wendel, ἐν τῇ Κωμικῇ λέξει). These, however, cannot be linked to Didymus with certainty.37 Finally, Didymus’ presence in comic papyri (where the names of grammatikoi are almost always omitted)38 is limited to two notes added to the main text of P.Flor. II 112 (second century CE), a commentary to an unidentified Aristophanic comedy (CLGP Ar. 28):39 the first additamentum has to do with the

31 It cannot be excluded, however, that Didymus himself offered the same interpretation both in the Comic Vocabulary and in the commentary to Demosthenes. 32 The entry is at ff. 195r–v in MS Vat. gr. 1818 (A) and at f. 149v in MS Laur. S. Marco 304 (B). These are the discrepancies between the two manuscripts: καρύκκη Α, καρύκκη B; καρυκεῖον A, καρύκι(ον) B; Ωρ A, om. B. See also EM p. 492.51–57 Gaisford (where Orus’ name is omitted) and Et. Gud. p. 301.37–43 Sturz. On Orus’ use of Didymus, see Alpers 1981: 118–20. 33 On the ambiguity of the abbreviation Ωρ in the tradition of the etymologica, see Reitzenstein 1897: 100 n. 2; Alpers 2015: 304–05. 34 See Schmidt 1854: 29–36. 35 If he did not directly quote Didymus’ hypomnēma. Indeed, when discussing Demeter’s epithet Ἀχαιά (Ar. Ach. 709), Orion (coll. 18.21–19.5 Sturz) gives the etymology ἀπὸ τοῦ ἄχους τοῦ ἐπὶ τὴν Περσεφόνην (‘from the pain for Persephone’), declaring an Aristophanic hypomnēma as his source (see ibid., οὔτως εὗρον ἐν Ὑπομνήματι εἰς Ἀριστοφάνην). The same etymology is ascribed to Didymus in the Etymologicum Gudianum (d2 p. 248.13–22 De Stefani Ἀχαιά· ἡ Δημήτηρ παρὰ Ἀττικοῖς, εἴρηται παρὰ τὸ ἄχος τῆς Κόρης. οὕτω Δίδυμος). Therefore, the commentary mentioned by Orion either relied on Didymus’ Comic Vocabulary or was actually Didymus’ Aristophanic commentary itself. 36 See Schmidt 1854: 37–39, 46–48. 37 Works titled Λέξις κωμική are attested also for Theon of Alexandria (see Hsch. ep. ad Eul. 1–4; Meliadò 2015) and Palamedes of Elea (see Suda π 43 Adler; Et. Gen. α 1203 Lasserre–Livadaras; Σ Ar. V. 710a, 1108b, 1122b Koster, Pac. 916 Holwerda). 38 See Montana 2004: 372. 39 Frr. C+D+E col. I. On the relationship between the main text of the commentary and Didymus, see Montana 2012b: 188–90.

56



Benuzzi

identification of a lyric nomos as hypotext for one of the lines commented upon, while the second is too fragmentary for any attempt at a reconstruction.40 Moreover, a scrap of parchment codex from the Louvre (CLGP Ar. 4, seemingly from the sixth century CE,41 now lost) preserved remnants of Aristophanes’ Birds with Didymus’ explanation of the term πρηγωρεών at v. 1113 (Didymus is not mentioned, but the interpretation is clearly ascribed to him in Σ Ar. Av. 1113a Holwerda = 241).42 As far as potential attributions are concerned, P.Oxy. XXXV 2737 (CLGP Ar. 27), a hypomnēma to another unidentified Aristophanic comedy, may be an epitomized version of Didymus’ commentary.43 Much more speculative are, instead, the identification of P.Oxy. XIII 1611 as excerpts from a Didymean hypomnēma to a comedy by Aristophanes and that of P. Oxy. XV 1801 (CLGP Ar. 3, 7, 12, 24, 26, 30, 31) as deriving from the Comic Vocabulary.44 Despite representing only a fraction of what Didymus’ exegesis of comedy must have been, the fragments preserved by the scholia to Aristophanes and by the other sources allow us to highlight some of the grammarian’s interests and some of the features of his interpretation of comic texts. 1. EXPLANATION OF RARE WORDS As already mentioned, the readership of Old Comedy in Didymus’ time was a much broader and less specialized group than the small intellectual circle of the Alexandrian Mouseion. Therefore, the explanation of obsolete vocabulary—a field of enquiry already firmly established by Didymus’ predecessors45—was an essential part of the running commentaries to the plays, and constituted at the same time the core of the Comic Vocabulary (see above). As a consequence, the presence of such material in the Aristophanic scholia can be owed to either (or both) of these traditions. These explanations may consist merely of an equivalent word or expression, sometimes introduced by ἀντὶ τοῦ, as in Σ Ar. Av. 530b Holwerda (230):

Δίδυμος δέ· βλιμάζοντες: ἀντὶ τοῦ ‘κακοῦντες’ (fort. κρατοῦντες legendum).46 RVΓ3M ἀποτίλλουσι γὰρ καὶ κατεσθίουσιν. RVΓ3 Didymus: βλιμάζοντες in the sense of ‘to maltreat’ (‘to hold in the hand’?). Indeed, they pluck them and eat them up. Lexicological analysis, however, is often closely linked with (folk) etymology, as shown for instance by Σ Ar. Av. 149a Holwerda (225): Δίδυμος δέ φησι Λέπρεον ὠνομάσθαι ἢ διὰ τὸ τὴν χώραν αὐτῶν †λέπειν† (fort. λέπ εἶν legendum).47 διαφαίνονται γὰρ ἐκ τῆς ὀρεινῆς RVEΓMLh […] ἢ διὰ τὸ τοὺς πρώτως οἰκήσαντας τὴν πόλιν ταύτῃ τῇ νόσῳ (sc. τῇ λέπρᾳ) κατεσχῆσθαι. VEΓMLh 40 41 42 43 44 45 46

See Montana 2009c and 2012b: 203–04, 206–07. See Montana 2012b: 45–46. See Montana 2012b: 45–47. See Montana 2012b: 145–82. See Montana 2012b: 238–39; Esposito 2012 and 2017: 39. For an overview on the glossographic tradition in Alexandria, see Tosi 1994: 143–78. The alternative reading κρατοῦντες comes from Tzetzes’ commentary to Birds (Σ Tz. Ar. Av. 530 Koster). The secondary meaning of κρατέω, ‘to hold in the hand’, would be a much more adequate equivalent for βλιμάζειν, given the sense required by the passage in Birds. In particular, it cannot be excluded that Didymus originally glossed the verb βλιμάζειν with κρατεῖν, referring to this secondary meaning, and that his interpretation was misunderstood by later readers who took ἀντὶ τοῦ κρατοῦντες as ‘in the sense of “prevailing”’ and felt the need to clarify the interpretation, by adding ἀποτίλλουσι γὰρ καὶ κατεσθίουσιν (thus facilitating the corruption κρατοῦντες > κακοῦντες). 47 The neuter form τὸ λέπας occurs with the meaning ‘peak’ (exclusively as nominative and accusative) several times in fifth-century Attic prose and drama (e.g. A. Ag. 283, πρὸς Ἑρμαῖον λέπας; Th. 8.78.5, ἐκαλεῖτο δὲ Ἀκραῖον λέπας) and in later poetry (Lyc. 420; Nic. Th. 147, 634). The word must have interested ancient and Late Antique lexicographers because of the potential confusion with ἡ λεπάς (‘limpet’; see Ammon. Diff. 295 Nickau, Hsch. λ 662 Cunningham, Suda λ 284 Αdler) and was apparently perceived to be linked to the adjective λεπρός, as the exegesis of the rare nominative feminine λεπράς in Theoc. 1.40 demonstrates (see Σ Theoc. 1.40a Wendel λεπράς: ἡ τραχεῖα πέτρα, ἡ ὑπὸ τῶν κυμάτων λεπιζομένη. ἢ ἡ ὑψηλή, ἐπεὶ καὶ λέπας τὸ ἄκρον τοῦ ὄρους φασίν. ἢ λευκὴ καὶ λέπρᾳ ἐοικυῖα). This proposed emendation would be consistent not

Didymus and comedy



57

Didymus says that (the city) is called Lepreon either because their territory †crumbles† (is a peak?)— indeed they stand out against the mountainous region—[…] or because the first inhabitants of the city were affected by this condition (i.e. leprosy, lepra). A specific folk-etymological concept that seems to have been favoured by Didymus was syncope, i.e. the derivation of a word through ‘contraction’ from an originally longer form.48 This method is applied to Pan’s epithet κεροβάτης (from κερατοβάτης; see Σ Ar. Ran. 230aα Chantry = 263) and to the obscure βρίκελοι in Cratinus’ Seriphioi (from βροτῷ εἴκελοι; see Hsch. β 1152 Cunningham = 9). This procedure is explicitly mentioned in another Didymean fragment, probably deriving from the Comic Vocabulary and preserved in Harpocration’s Lexicon to the Ten Attic Orators (π 76 Κeaney = 249a):49 ποδοκάκκη· Δημοσθένης Κατὰ Τιμοκράτους (24.105). τὸ ξύλον τὸ ἐν τῷ δεσμωτηρίῳ οὕτως ἐκαλεῖτο, ἤτοι παρεμβεβλημένου τοῦ ἑτέρου κ, ποδῶν τις κάκωσις οὖσα, ἢ κατὰ συγκοπήν, ὥς φησι Δίδυμος, οἷον ποδοκατοχή. Podokakkē: Demosthenes in Against Timocrates. The wooden stock in the prison was called by this name, either because the second kappa is superfluous, because it is an evil (kakōsis) for the feet (podōn), or as Didymus says, it is a syncopated form of podokatochē (‘foot hindrance’).50 2. KŌMŌDOUMENOI A field that seems to have been of particular interest to Didymus and to readers of Aristophanes in his time is that of the identification and description of the targets of direct personal attacks, with the consequent discussion of possible namesakes and of the meaning and origin of the mockery.51 In this respect, Didymus could avail himself of a tradition of kōmōdoumenoi literature of which now no more than a few names and titles remain.52 Among these, the Κωμῳδούμενοι of Aristarchus’ pupil Ammonius were a prime source for Didymus.53 Not only did he quote directly from his predecessor, but he also engaged critically with his interpretations, as shown for instance by the scholion to Av. 1297–99, on the kōmōdoumenos Meidias (247):

Μειδίας δ’ ἐκεῖ REΓ ὄρτυξ E: ὁ μὲν Δίδυμος οὕτως· ὁ [δὲ] Ἀμμώνιος (BNJ 350 F4 = fr. 4 Bagordo) ᾠήθη ἐξ ἐπιθέτου Μειδίαν ὄρτυγα καλεῖσθαι. γελοίως διὰ τὸ κυβευτὴν εἶναι καὶ ἐν γύρῳ τοὺς ὄρτυγας κόπτειν RVEΓM οὕτως αὐτὸν νῦν Ἀριστοφάνης προσεῖπεν. VEΓM δηλοῖ δὲ τοῦτο Πλάτων ἐν Περιαλγεῖ

48 49

50 51

52

53

only with the location of Lepreon ‘on a steep hill north of modern Lepreo (formerly Strovitzi), c.7 km from the coast’ (Heine Nielsen 2004: 544; see also Baladié 1978: 280), but also with Photius’ entry concerning the city (λ 196 Theodoridis), where the first proposed etymology is indeed ἀπὸ τοῦ παρακειμένου τραχέος ὄρους. For different emendation proposals, see White 1914: 42; Kakridis 1974: 50; Rutherford 1896: 441; and Holwerda 1991: 28. Etymologies with syncope are attested already for Aristarchus; see Schironi 2018: 343. The similarity between the gloss preserved by Harpocration and Hesychius’ entry ποδοκάκη· […] Πλάτων (fr. 281 K.–A.). καὶ ἤτοι ποδοκάκη παρεμβεβλημένου τοῦ ἑτέρου , ἢ κατὰ συγκοπὴν ποδοκατοχή (π 2679 Hansen) points towards a common derivation from the Comic Vocabulary (a Didymean commentary to Plato is not attested and Harpocration’s entry would be the only evidence for one on Demosthenes’ Against Timocrates). Translation by Gibson 2002: 142 (with alterations). Other Didymean fragments that deal with kōmōdoumenoi besides the ones discussed below are in Σ Ar. Av. 876f (237, on Cleocritus), 1294 (245, on Opuntius), 1295bα (246, on Philocles and Teagenes), and 1379a–b Holwerda (249, on Cinesias); Σ Lys. 313a Hangard (254, on Phrynichus the general); and Σ Ra. 13a–b (257, on the comic playwright Phrynichus), 965a (266, on Phormisius), and 990c Chantry (268, on Mammacythus and Melitides). The most comprehensive study on the topic is still Steinhausen 1910. On Herodicus’ Κωμῳδούμενοι, see also Broggiato 2014: 46–48. Nesselrath 1990: 76–77 supposes that Didymus made use of Antiochus’ Περὶ τῶν ἐν τῇ μέσῃ κωμῳδίᾳ κωμῳδουμένων ποιητῶν (On Poets Ridiculed in Middle Comedy). The Hypsicrates mentioned as the author of a collection of kōmōdoumenoi in P.Oxy. XVIII 2192 (see Roberts 1941: 152) is probably to be identified with the grammarian and historian that lived approximately in Didymus’ time (see Perrone 2018). On Ammonius’ life and works, see D’Alessandro 2018.

58



Benuzzi

(fr. 116 K.–A.) κτλ VEΓ ‘and Meidias over there, (they call him) quail’: Didymus (says) the following: Ammonius believed that Meidias was called ‘quail’ by epithet. But Aristophanes here calls him this way mockingly, because he was a gambler and he would tap the quails in the ring. This is proved by Plato in the Man in Terrible Pain… Ammonius inferred from the lines of Birds that ‘quail’ was Meidias’ actual nickname, while Didymus argued that this was a specifically Aristophanic invention aimed at mocking Meidias’ enthusiasm for gambling on quail- and cock-fights, a conclusion based on other comic passages apparently overlooked or misinterpreted by the author of the Κωμῳδούμενοι. The treatment of Molon—a famous actor indirectly mocked in Ran. 55 for his large body—shows at the same time Didymus’ failure to understand the irony of the passage and his attempt to link the kōmōdoumenos to a namesake mentioned both by Eupolis and Teleclides. See Σ Ar. Ran. 55c–a–b Chantry (259): μικρός, (Ald) ἡλίκος EBarb(Ald) Μόλων: MEBarb(Ald) παίζει. ἔστι γὰρ μεγαλόσωμος ὁ Μόλων. VMEΘBarb(Ald) ἄλλως. V Δίδυμος δέ φησιν ὅτι δύο Μόλωνές εἰσιν, ὁ ὑποκριτὴς καὶ ὁ λωποδύτης· καὶ μᾶλλον τὸν λωποδύτην λέγει, ὅς ἐστι μικρὸς τὸ σῶμα. Τιμαχίδας δὲ (fr. 20 Matjašić) τὸν ὑποκριτὴν νῦν λέγεσθαί φησι Μόλωνα. RVMEΘBarb(Ald) ‘Small … the size of Molon!’: he is joking, because Molon is large-bodied. Otherwise. Didymus says that there are two Molons, the actor and the thief. And (Aristophanes) rather refers to the thief, who is of small build. Timachidas, instead, says that here he means the actor Molon. Indeed, although unanimously deemed by modern scholars to be an autoschediastic invention,54 a thief by the name of Molon is in all likelihood attested by a lemma of the papyrus commentary to Eupolis’ Taxiarchoi (P.Oxy. XXXV 2740 = Eup. fr. 268 K.–A.),55 where the relevant gloss signalled the presence of the same figure in a play by Teleclides (fr. 73 K.–A.). The idea of Didymus making up a kōmōdoumenos just to fit his interpretation of a passage is even less plausible given that the grammarian plainly declared when he lacked information on the characters mentioned in the text, as shown by Σ Ar. V. 1178a Koster (215): Δίδυμος· ὁ Καρδοπίων ζητητέος. οὐδαμοῦ κωμῳδεῖται. Didymus: Cardopion has to be investigated. He is not mocked anywhere (else). At the same time, when he has plenty of information on a kōmōdoumenos, the grammarian seems to favour the most overcomplicated or simply abstruse explanation. Such is the case of the Alcaeus mentioned (along with Ibycus and Anacreon) in Th. 162: this was identified by Didymus with an obscure citharode attacked by Eupolis in the Golden Race (and listed as object of mockery in Eratosthenes’ On Ancient Comedy)56 rather than with the lyric poet clearly implied by the passage (see Σ Ar. Th. 162a Regtuit = 256). In fact, Didymus’ treatment of the Alcaeus of Th. 162 and of the Molon of Ran. 55 might indicate that the grammarian tended

54 See e.g. van Leeuwen 1896: 16; Dover 1993: 197; Sommerstein 1999: 161. 55 The integration Μόλ]ωνι at line 41 is proposed in Benuzzi 2019. 56 Hsch. α 3086 Cunningham ἀλκαῖον· *ὅπλον P ἢ ἀλεξιφάρμακον φυτόν. Ἐρατοσθένης (fr. 6 Strecker) δὲ ὄνομα εἶναί τινος ἀνθρώπου κωμῳδουμένου. See Olson 2016: 478.

Didymus and comedy



59

to overlook the sense of a single passage, if this allowed him to highlight a link (in this case, the presence of the same kōmōdoumenos) between different plays and authors.

3. ALLUSIONS TO OTHER TEXTS Another facet of Didymus’ tendency to draw connections between Aristophanic plays and other works and authors is the detection of hypotexts, sometimes rather far-fetched ones. For instance, in Σ Ar. Av. 1121a Holwerda (242) the solemn description of the messenger rushing on scene as ‘breathing Alpheus’ (see Ar. Av. 1121, ἀλλ’ οὑτοσὶ τρέχει τις Ἀλφειὸν πνέων) is considered—quite implausibly—a reference to the beginning of Pindar’s Nemean 1:

ἀλλ’ οὑτοσὶ VEΓ τρέχει E: Σύμμαχος· οὗτος οὕτω συντόνως τρέχει ὡσεὶ ὀλυμπιακὸς σταδιοδρόμος. VEΓLh ὁ δὲ Δίδυμος· παρὰ τὸ Πινδάρου ‘ἄμπνευμα σεμνὸν Ἀλφειοῦ’ (N. 1.1). διχῶς δέ τινες ‘Ἀλφειὸν πνέων’. VEΓMLh ‘but that one comes running’. Symmachus: he runs as rapidly as if he were an Olympic runner. Didymus: (with an allusion) to Pindar’s line ‘Alpheus’ noble sigh of rest’.57 Some interpret ‘breathing Alpheus’ in two ways. Similarly, the grammarian saw in Ran. 704 (τὴν πόλιν καὶ ταῦτ’ ἔχοντες κυμάτων ἐν ἀγκάλαις, ‘having the city and these affairs in the arms of the waves’—a verbatim quotation from Archilochus, fr. 213 West) an echo of an Aeschylean line, now unidentifiable due to loss of text in the scholion (see 264).58 In other cases, it is impossible to verify how plausible the hypotext detected by Didymus is. One example comes from the already mentioned hypomnēma P.Flor. II 112 (CLGP Ar. 28), where an interlinear addition attests that Didymus identified a nomos as hypotext for the Aristophanic lines commented upon; another is offered by Σ Ar. V. 1064a Koster, where the hypotext presupposed by Didymus is a lost song by Timocreon of Rhodes (PMG 733; 214).59 The search for hypotexts often led to the identification of proverbs: the grammarian’s paroemiographical interests are well attested not only within his exegesis of Aristophanes (267)60 and of the orators,61 but also as a self-standing field of research. Indeed, Didymus compiled his own collection of proverbs (later epitomized by Zenobius),62 probably drawing on the previous paroemiographical tradition initiated by Aristotle.63 Hypotexts and proverbs were not the only types of allusions investigated by Didymus. In a clear effort to display his vast erudition and his access to ancient sources, the grammarian dealt extensively with historical references as well. A telling example is Σ Ar. Av. 13a Holwerda (222), where the grammarian overinterprets 57 The comparison between Aristophanes’ and Pindar’s passages proves that Didymus understood the word ἄμπνευμα in the etymological sense of ‘breath exhaled […], sigh of rest’ (Carey 1981: 104–05) instead of ‘resting place’. It was this interpretation (and not simply the similarity in wording) that made him perceive such a similarity between the beginning of the first Nemean—with the imagery of Alpheus passing under the sea from Elis to emerge in Ortygia—and the scene of the messenger rushing on scene out of breath from the long run (the emphatic repetition of ποῦ in the messenger’s first line, i.e. Av. 1122–23, has in fact been interpreted as an expression of his breathlessness; see Thompson 1940: 188; Dunbar 1995: 594). 58 Δίδυμός φησι· παρὰ τὸ Αἰσχύλου (fr. dub. 462 Radt) . VMEΘBarb(Ald) ἔστι δὲ ὄντως παρὰ Ἀρχιλόχῳ (fr. 213 West) VMEΘBarb ‘ψυχὰς ἔχοντες κυμάτων ἐν ἀγκάλαις.’ VMEΘBarb[Ald]. I follow the text printed by Süss 1959: 49 as it reflects the opposition between Didymus’ detection of a literary allusion (παρὰ τὸ Αἰσχύλου) and the anonymous remark on the Aristophanic expression being a verbatim quotation from Archilochus (παρὰ Ἀρχιλόχῳ). 59 See Montana 2009c. 60 Δίδυμος δέ φησιν ὅτι δύναται (sc. ὁ Ἀριστοφάνης) καὶ τῆς παροιμίας μεμνῆσθαι, οὐ Χῖος, ἀλλὰ Κεῖος (fort. Κῷος leg.) VMEΘBarb(Ald). See also Σ Ar. Ran. 186b, 990c Chantry. 61 See e.g. Harp. π 54, τ 19 Keaney (283b = 355b; 318a). 62 On the relationship between Didymus’ and Zenobius’ works, see Bühler 1987: 36–37 n. 16. 63 For an overview, see Tosi 1994: 179–92.

60



Benuzzi

the expression οὑκ τῶν ὀρνέων, ‘the one from the bird-market’, as a nod to Orneai and to its inhabitants, who had fought with the Athenians at Mantinea in 418 BCE: οὑκ τῶν ὀρνέων: ἀντὶ τοῦ ‘ὀρνεοπωλίων’. Δίδυμος δὲ δεινὰ φάσκειν αὐτοὺς ἐκ τῶν ὀρνέων πεπονθέναι, ἐπεὶ Ὀρνεαὶ τῆς Λακωνικῆς εἰσι, πρὸ δὲ ἐτῶν δʹ κακῶς περὶ Μαντίνειαν ἀπήλλαξαν, ὡς καὶ τοὺς στρατηγοὺς ἀποβαλεῖν Λάχητα καὶ Νικόστρατον, καθὰ καὶ Ἀνδροτίων φησίν (BNJ 324 F41). VEΓM ‘the man from the birds’: in the sense of ‘the bird-market’. Didymus (says that) they claim to suffer terribly because of the birds (ornea), since there is a place named Orneai in Laconia, and four years earlier (i.e. 418 BCE) they (sc. the Athenians) failed badly around Mantinea, so that they even lost the generals, Laches and Nicostratus, according to what Androtion also says.64 4. ORTHOGRAPHY AND TEXTUAL CRITICISM In accordance with the interest of the Aristophanic readership of his time, matters of orthography seem to have been dealt with by Didymus only when the meaning of the text was at stake. For instance, Didymus’ discussion of Av. 299–300 focuses on the orthography of the bird’s name, κηρύλος, in order to clarify the origin of Aristophanes’ comic coinage κειρύλος (from κείρειν, ‘to cut’); see Σ Ar. Av. 299a Holwerda (227):

Δίδυμος τὸ κατὰ φύσιν ὄνομα κηρύλος65 λέγεσθαι. VΓ2Lh […] μήποτε παρὰ τὸ κείρειν ἐσχημάτικεν (sc. ὁ Ἀριστοφάνης). VM9Γ2Lh Didymus says that the name said according to nature is kērylos. […] Maybe he (sc. Aristophanes) modelled the word on the verb ‘to cut’ (keirein). As far as textual criticism is concerned,66 the material confusedly transmitted by the scholia to Ran. 775— where the slave describes how Euripides has gained the support of the dead with ‘his argumentative speeches and his twistings and weavings’67—seems to preserve a (correct) textual choice by Didymus: λυγισμῶν (‘twistings’) against the corrupt λογισμῶν (‘reasonings’), which is in fact attested by part of the manuscript tradition of Frogs. Σ Ar. Ran. 775d Chantry (265): Δίδυμος δὲ ‘λυγισμῶν’ VMEΘ(Ald) γράφει. M Didymus writes lygismōn. Unfortunately, due to its long-acknowledged ambiguity in the scholiastic jargon,68 the verb γράφειν does not allow us to ascertain whether λυγισμῶν was Didymus’ own conjectural emendation (γράφειν in the sense of 64 Didymus’ explanation is not entirely clear, possibly due to epitomation in the central section of the scholion, as can be argued from the lack of a logical link connecting the statement about the existence of a place named Orneai and the account of the Athenian defeat at Mantinea in 418 BCE. What might have to be supplied is that Orneai, as an ally of Athens and Argos, had had significant responsibility for the defeat of its own front. On this quotation of Androtion by Didymus, see also Montana, this volume, 72. 65 I follow White 1914: 72 in printing κηρύλος. All other editors (see Bekker 1829: 237; Dindorf 1838: 174; Dübner 1842: 217; Holwerda 1991: 54) choose κειρύλος, despite the unanimity of the manuscripts on the wrongly accented form κήρυλος (except the insertion of the diphthong ει above the line in Γ2). 66 In general, when dealing with textual criticism, Didymus shows a conservative approach in some cases (see e.g. Σ Pi. O. 10.55c = 123 and Σ P. P. 7.6a = 135), but offers emendation attempts in other instances (see e.g. Σ P. N. 10.114a = 165). 67 Sommerstein 1999: 99. 68 See Slater 1989: 46–53; Neri 1996: 34–37.

Didymus and comedy



61

μεταγράφειν) or if it was a varia lectio that the grammarian found in some Aristophanic copies (γράφειν in the sense of τῇ γραφῇ κεχρῆσθαι). 5. TEXTUAL CLARIFICATION Aside from the specific exegetic interests discussed above, the bulk of Didymus’ fragments on Aristophanes actually consists of basic textual clarification. Indeed, a passage could be made unclear by obscure references, such as the expression ἐπὶ πετρῶν discussed in Σ Ar. Av. 836a–b Holwerda (236):

ὡς δ’ ὁ θεὸς ἐπιτήδειος EΓ3: Δίδυμός φησι τὸ πελαργικὸν τεῖχος ἐπὶ πετρῶν κεῖσθαι. RVEΓ3MLh καὶ ὁ ἀλεκτρυὼν οὖν ὡς θεὸς οἰκήσει ἐπὶ πετρῶν. Lh ‘How well suited the god (is to living on rocks)’: Didymus says that the Pelargic wall stood on rocks. Therefore, the cock will live as a god on the rocks. or by complex jokes, such as the pun between the city-name Sparta and the word sparton (‘rope’) in Σ Ar. Av. 816a Holwerda (233): οὐδ’ ἂν χαμεύνη REΓ: Δίδυμός φησιν· οὐδ’ ἂν σπάρτῳ χρησαίμην, VEΓM οὕτω μισῶ τὴν Σπάρτην. χαμεύνη δὲ ταπεινὴ κλίνη. ὅθεν καὶ τὸ ὄνομα εἴληφεν. RVEΓM ‘(I would not attach this name) even to the chameunē’: Didymus says: I would never use a rope (sparton), so deep is my hate for Sparta. The chameunē is a low bedstead. From this (i.e. from chamai, ‘on the ground’) it received its name. The high number of fragments of this kind further confirms that the readership of Old Comedy in Didymus’ time was mostly composed of non-specialists that needed explanations not only of obsolete vocabulary, literary references and long-forgotten historical figures and events (all topics already investigated by Didymus’ predecessors), but also of the puns and convoluted expressions typical of comic language. To sum up, the grammarian’s work on comedy responded to the needs of his audience primarily in two ways: firstly, it summarized the opinions of previous scholars on debated passages, making them easily accessible; secondly, it offered a wide typological range of elucidations (on vocabulary, Realien, proverbs, doubles entendres, historical references, etc.) that were useful also to less specialized readers. Although Didymus’ commentaries also included valuable insights potentially coming from his own research (especially on rare words, intertextual references, historical allusions, and kōmōdoumenoi), these two main components of his exegesis are the ones that earned him detractors both in antiquity and in modern times. However, the relevance of his commentaries and Comic Vocabulary for subsequent scholarly literature up to this day cannot be denied: Didymean exegesis was—and still is—an invaluable source that, due to its main features, was destined to be widely used (but seldom praised) by readers of Greek comedy of all times.

Bulletin of the Institute of Classical Studies, 2020, 63, 62–78 doi: 10.1093/bics/qbaa019 Original article

Chapter Five Didymus and the Greek historians* Fausto Montana University of Pavia, Pavia, Italy

ABSTRACT We lack positive evidence that Didymus composed scholarly works specifically devoted to Greek historians. Even more, the very origins and characteristics of the Alexandrian interest in the historiographic genre, not to say in literary prose, represent open issues for the historian of Hellenistic scholarship. In this chapter, the rare and sparse pieces of information are gathered, in order to obtain a possibly systematic and organic overview on the very defective puzzle of the Didymean approach in this field. In the first part, clues from testimonies and fragments directly concerning historians (Herodotus, Thucydides) and antiquarian subjects (Solon’s axones) are taken into special consideration. The second part deals with the use and abuse of history and historical sources detectable in fragments from Didymean works which were devoted to fields and genres other than historiography. As a result, the testimony of the Pindar scholia and of the writing On Demosthenes attested to by P.Berol. inv. 9780 recto (second century AD) proves to be especially decisive for a reassessment of Didymus’ approach to historiography and history.

In grateful memory of Luigi Piccirilli (1939–2002)

1. ANCIENT GREEK SCHOLARSHIP AND HISTORIOGRAPHY The origins and early features of Hellenistic grammarians’ interest in Greek literary prose are open issues for the historian of ancient scholarship.1 Prose writers may appear to have been excluded from the immediate horizon of Hellenistic philology until the first century BC, when Alexandrian scholars such as Asclepiades of Myrlea, Didymus, Artemidorus of Tarsus, and his son Theon devoted much attention to these fields. As for the historians, we lack clear evidence that Didymus composed works that specifically focused on any one of them. But these impressions are almost certainly misleading. Dionysius Thrax (late second century BC), the pupil of Aristarchus, defined grammatikē as ‘empirical knowledge of what is mostly said by poets and prose writers (συγγραφεῦσιν)’.2 P.Amh. II 12 (CLGP I.2.6 Herodotus 4, third century AD), containing the final lines of a commentary on Hdt. 1 which the preserved subscription attributes to Aristarchus, directly attests a high level of scholarly interest in this sphere.3 These traces show how fragmentary and insufficient our * Numbers in bold refer to testimonia and fragments in the Checklist (pp. 95–120). This chapter was translated into English by Orla Mulholland. 1 Montana 2009b: 166–75; Montanari 2013; Montana 2015b: 95–97, 2020: 168–70; Matijašić 2018: 147–60. 2 D.T. Tekhnē grammatikē 1; cf. S.E. M. 1.57. As an instance, the commentaries on the Attic orators composed by Didymus cannot be considered a beginning, but appear to be the final culmination of previous lines of research carried out in this field at Alexandria. 3 Most recently: Matijašić 2018: 150–51, 156–57; Montana 2019: 39–61 (new commented edition of the Amherst papyrus). C The Author(s) 2021. Published by Oxford University Press on behalf of the Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected].



62

Didymus and the Greek historians



63

documentation is, compelling us to proceed on the basis of inferences and to assume that the prose writers actually did make their entrance among the interests of the Hellenistic grammarians, although at a second stage, when philology on works of poetry had already been developed extensively.4 This chapter gathers the pieces of positive information about Didymus’ interest in historians and history. I will first deal with the meagre clues to scholarly works that could have been specifically devoted to historians (below, §2), and then continue with the evidence of the use of history and historical sources that can be found in fragments from Didymus’ works concerning fields and literary genres other than historiography (§3). 2. DIDYMUS’ WORKS FOCUSED ON HISTORIANS AND FACTUAL SUBJECTS

2.1. Herodotus It would seem that Didymus, following in the footsteps of Aristarchus, composed his own commentary on Herodotus’ Histories. This inference is supported by an anonymous hypomnēma, P.Oxy. LXV 4455 (CLGP I.2.6 Herodotus 7, third century AD).5 The extant text concerns Hdt. 5.52–55. What stands out are traces of a discussion of the information given by Herodotus on the length and staging points of the Persian Royal Road. The commentator seems to be comparing the data provided by the historian to those reported in Xenophon’s Anabasis (1.2.5–23), and he uses terms and concepts proper to historical-geographical investigations: comparison of the sources for land measurement; evaluation of their discrepancy (διαφωνία); judgement on their reliability (ἀξιοπιστία). By this the commentator is revealed to have been a learned person applying a rationalistic approach to his reading of Herodotus. In col. i.12–14 we read a literal (φησίν) but unfortunately very lacunose quotation of an observation by Didymus (322) on 5.52.2 (readings and restorations by the editor princeps): ± 3 ὁ Ἅλ]υ**ς π* [ο]τ* α* μός* , *ἐ π᾽ ὧι πύ* [λαι] *τ ε εἰσί ± 6 γε]φ * ύ* ρ* α* ς φησὶν* ὁ Δ * [ί]δυ* μο**ς π* υ* ± 8 διαβά]ν**τι κτλ. [the Hal]ys river, on which there are gates | [… bri]dges(?) Didymus says ga|[tes(?)] etc. (new lemma). In line 13, Haslam plausibly restored γε]φ * ύ* ρ* α* ς, comparing Hdt. 1.75.3, ὡς δὲ ἀπίκετο ἐπὶ τὸν Ἅλυν ποταμὸν ὁ Κροῖσος, τὸ ἐντεῦθεν, ὡς μὲν ἐγὼ λέγω, κατὰ τὰς ἐούσας γεφύρας διεβίβασε τὸν στρατόν; therefore, he suggested completing the entire sentence thus: ἀντὶ τοῦ γε]φ * ύ* ρα* ς φησὶν* ὁ Δ * [ί]δυ* μο**ς π* ύ* |[λας λέγεσθαι. It is hardly probable, however, that Didymus and the anonymous commentator would have been discussing a usage, which is not attested, of πύλαι, ‘gates’, as a metonymy for γέφυρα, ‘bridge’. If there is a difficulty in the Herodotean passage, it is the brachylogy of locating the gates on the river (ἐπ᾽ ᾧ). We could, hence, restore Didymus’ explanation as e.g. ἐπὶ τῆς γε]φ * ύ* ρα* ς φησὶν* ὁ Δ * [ί]δυ* μο**ς π* ύ* |[λας εἶναι, ‘Didymus says that there are gates just on the bridge’ (or π* ύ* |[λαι εἰσίν, with φησὶν* ὁ Δ * [ί]δυ* μο**ς as a parenthetical clause). Didymus’ explanation would then concern not vocabulary (πύλαι is certainly not a glōssa), but the factual information on the presence of gates on the bridge that crosses the river Halys along the course of the Royal Road. If so, the Didymean remark was drawn not from a lexicon, but from a hypomnēma to the Histories.6

4 Was this delay influenced by Aristotle’s view on the different natures of poetry and historiography (Po. 9, 1451a36–1451b11)? Peripatetic writings such as Praxiphanes’ Περὶ ἱστορίας were probably devoted to literary history: Matelli 2012: 189 and 277–81. 5 Ed. Montana 2019: 74–89. Photo at Oxyrhynchus Online . 6 Montana 2009a: 253–54 and 2019: 82–83; Braswell 2017: 40 n. 48. For further evidence of Didymus’ use of Herodotus in grammatical writings and commentaries, see below, §§3.1.1, 3.1.2, 3.2.3.

64



Montana

2.2. Thucydides The existence of Alexandrian scholarship on Thucydides is conjectural. Dionysius of Halicarnassus knew a whole tradition of grammatikē exēgēsis to Thucydides’ work (Th. 51, 55).7 It would be hard to read that notice as excluding Didymus, Dionysius’ contemporary and the ‘administrator of the estate’ of Alexandrian scholarship.8 An observation by Didymus on Thucydidean vocabulary (see below, §3.1.1) is not enough to demonstrate that he composed a hypomnēma.9 However, he did deal with the historian in depth, if he is the same Didymus quoted by name three times in Marcellinus’ Life of Thucydides:10 i. 2–3 (323). In the initial part of the Life, dedicated to the historian’s descent from Miltiades (IV) the Younger and Cimon and to his membership of the genos of the Philaids, the biographer cites a Didymus as auctoritas, who in turn explicitly asserts that he has taken the genealogy up to Miltiades (III) the Elder from the first book of Pherecydes’ (of Athens) Histories (fr. 2 Fowler): καὶ τούτοις Δίδυμος μαρτυρεῖ, Φερεκύδην ἐν τῇ πρώτῃ τῶν Ἱστοριῶν φάσκων οὕτως λέγειν· κτλ., ‘and Didymus attests to these matters, saying that Pherecydes, in the first book of his Histories, says as follows: etc.’. Didymus’ use of Pherecydes is documented twice more in the Iliad scholia (below, §3.2.1): this fact lends at least some support to the identification of Marcellinus’ Didymus as the Chalcenterus. The Vita proceeds (4) by observing that the same information is attested in Hellanicus’ Asopis (BNJ 4 F22); and then it goes into more depth on the question of Thucydides’ relationship with the Philaids through the story of the events that led Miltiades the Elder to leave Attica for Thrace and then Miltiades the Younger to return there and to assume the position of stratēgos in the first Persian War (4–13). The digression closes with a recapitulating phrase that contained the name of the source, which, however, has fallen out (14). A credible hypothesis is that the name to be supplied is that of Didymus (fr. 4 Schmidt) and that the entire section 3–14 goes back to him and was opened and closed by his name.11 The restoration extends back Didymus’ contribution to the reconstruction of the history of the Philaids to the late archaic period, completing Pherecydes’ genealogy. The Didymean authorship of this section seems to be supported by the scholion to Pi. N. 2.19 (145 = 326): seeking a solution to the zētēma on the connection between the dedicatee of the ode, Timodemus, and the island of Salamis, mentioned in lines 12–13, ‘Didymus says that perhaps it is better to assume that (Timodemus) traced his family to Aias, as also (did) Miltiades and Cimon and Alcibiades and the historian Thucydides the son of Olorus’. ii. 16–17 (324).12 In the discussion of the correct form of the name of Thucydides’ father, the biographer declares in favour of Ὄλορος and against Ὄρολος, following Didymus’ opinion. Next there is documentary proof suggesting that this is the very argument adopted by Didymus (16): αὕτη γὰρ ἡ γραφή, ὡς καὶ Διδύμῳ δοκεῖ, ἡμάρτηται. ὅτι γὰρ Ὄλορός ἐστιν, ἡ στήλη δηλοῖ ἡ ἐπὶ τοῦ τάφου αὐτοῦ κειμένη, ἔνθα κεχάρακται· Θουκυδίδης Ὀλόρου Ἁλιμούσιος. This spelling (sc. Ὄρολος) is incorrect, as held also by Didymus. In fact, that (the correct form) is Ὄλορος is demonstrated by the inscription placed on his tomb, where it is carved: Thucydides, son of Olorus, from (the deme) Halimous. 7 Usener 1889: 71 ff.; Pfeiffer 1968: 225. A pair of instances of Aristarchus’ use of Thucydides in his scholarship on Homer (Σ Ariston. Il. 2.867a (A) Erbse and Od. 3.71a (DHMa) Pontani) are pointed out by Matijašić 2018: 154–55, 159; cf. Schironi 2018: 660. 8 Luschnat 1954: 23. 9 Pace Schmidt 1854: 334. 10 Schmidt 1854: 321–34, no. 27; Braswell 2017: 86–88, no. 47. Marcellinus’ Didymus is identified instead with Didymus Claudius by Mazzarino 1966: II 2.466; cf. Piccirilli 1985: 67–68, 89–90. Contra Arrighetti 1968: 97 n. 94 and 1987: 226–27 n. 202; Porciani 2001: 45 n. 106; Montana 2015a. 11 Ritter 1845: 335; cf. Piccirilli 1985: 16, 86–87, who is inclined to restore in 14 ἀπὸ τούτου οὖν κατάγεσθαί φησι τὸ Θουκυδίδου γένος, ‘ affirms, therefore, that the lineage of Thucydides descends from that man’. 12 I follow the text established by Piccirilli 1985.

Didymus and the Greek historians



65

Didymus probably drew the text of the inscription from a work by Polemon of Ilium. The latter, in fact, is cited a little later in Marcellinus’ Life (17) as a witness to the same notices in his writing On the Acropolis (of Athens) (fr. 5 Preller = fr. 5 Capel Badino: καὶ Πολέμων δὲ ἐν τῷ περὶ ἀκροπόλεως τούτοις μαρτυρεῖ), a work also quoted by Didymus elsewhere (see below, §3.2.4; cf. §3.2.2). iii. 32 (325). Didymus is cited also for the circumstances of the historian’s death and burial. According to the grammarian, Thucydides returned from exile and in Athens he died a violent death, receiving burial among the tombs of Cimon’s family. It seems that Didymus discussed the thesis of Zopyrus (FHG IV 533, fr. 6), according to which the historian died in Thrace.13 If the words that follow still have Didymus as their subject,14 we learn that the latter connected the return of Thucydides to Athens with the amnesty granted by the polis to the exiles after the disaster in Sicily and he was prompted to ‘reprove the ingenuity of those who thought he died outside Attica’, since this contrasts with the presence of the tomb at Athens. The point about the amnesty is picked up again at the end of the argument: ‘But it is evident that the exiles were allowed to return, as said by Philochorus (BNJ 328 F137) and Demetrius (of Phalerum) in The Archons (BNJ 228 F3 = fr. 92 SOD)’.15 The uncertainty over the extent of the Didymean fragments complicates the task of determining which work they were drawn from.16 Probably the compiler of the Life did not cite Didymus’ passages at first hand but found them already reduced to excerpts (σχόλια) in the exegetical tradition.17

2.3. Didymus the Antiquarian Plutarch, at the beginning of his Life of Solon (1.1), quotes Didymus’ work Reply to Asclepiades on Solon’s Axones (Περὶ τῶν ἀξόνων τῶν Σόλωνος ἀντιγραφὴ πρὸς Ἀσκληπιάδην: °361; BNJ 340 F1 = test. 3 Ruschenbusch) concerning the name of the Athenian statesman’s father (Euphorion, according to an unknown Philocles mentioned by Didymus; Execestides, according to others).18 This monograph must have had an antiquarian character and commented on the shape and content of the axones or slabs on which the text of Solon’s laws was fixed and put on public display at Athens. The question lay at the origin of a rich ancient discussion, which started, it seems, from Aristotle (Περὶ τῶν Σόλωνος ἀξόνων ε´: test. 1 Ruschenbusch) and perhaps took off again only after the rediscovery of his exoteric writings in the first century BC.19 To this period are dated the syngramma Τῶν ἀξόνων ἐξηγητικά composed by Asclepiades (BNJ 339 F1 = test. 2 Ruschenbusch), supposedly the scholar from Myrlea,20 then Didymus’ Reply to Asclepiades, and finally Seleucus Homericus’ Ὑπόμνημα τῶν Σόλωνος ἀξόνων (BNJ 341 F1 = test. 4 Ruschenbusch). Some erudite details present in Plutarch’s Life of Solon may derive from Didymus’ work.21 Didymus’ familiarity with Solon’s

13 Piccirilli 1985: 120–22. 14 Cf. Jacoby’s division of the fragment in FGrHist and Jacoby 1954: I 508 with n. 5 = II 410; Piccirilli 1985: 122; Jones 2016. 15 In 33 there is a critique in the first person (ἐγὼ δὲ […] νομίζω) of the opinions of, again, Zopyrus and, further, of Cratippus (BNJ 64 F2), Timaeus (BNJ 566 F136) and ‘others’. It is improbable that the subject is still Didymus, who is cited in the previous paragraph in the third person (φησιν, φησι, ἔφη). 16 Thus Braswell 2017: 88, against Schmidt 1854: 334 (a Life of Thucydides) and Cohn 1903: 461 (Συμποσιακά). 17 Piccirilli 1985: xxiv, xxvi. The MS Pal. gr. 252 (tenth/eleventh century) transmits the title Μαρκελλίνου ἐκ τῶν εἰς Θουκυδίδην σχολίων περὶ τοῦ βίου αὐτοῦ Θουκυδίδου καὶ τῆς τοῦ λόγου ἰδέας. 18 The second view possibly reached Plutarch from Hermippus of Smyrna: Piccirilli 1998: 108–09. 19 The Hellenistic debate on ἄξονες and κύρβεις (Eratosthenes, Polemon of Ilium, Aristophanes of Byzantium, Apollodorus of Athens) was probably lexicographical in nature, a viewpoint already documented by the comic play on Solonian glōssai in Ar. Daitaleis fr. 233 K.–A. In the first century BC, the revival of the antiquarian approach, begun by Aristotle, would have been inserted into the previous lexicographical discussion: Ruschenbusch 1966: 50; cf. Piccirilli 1998: 108. On the material, shape, and content of ἄξονες and κύρβεις: Rhodes 1993: 131–35. 20 Ruschenbusch 1966: 50 n. 135. Jacoby 1955: I 91 thought of Asclepiades of Nicaea or Alexandria (BNJ 339 F1); cf. Piccirilli 1998: 107–08; Braswell 2017: 99. On the two Asclepiades: Montanari 1997; Pagani 2007: 121–29. 21 e.g. Sol. 19.3–4 and 25.2, according to Jacoby 1955: I 92; cf. Ruschenbusch 1966: 46–47; Piccirilli 1998: xix, 107.

66



Montana

axones is confirmed by the remnants of his commentary on Demosthenes’ Against Aristocrates (see below, §3.2.4).22 3. DIDYMUS’ USE OF HISTORIOGRAPHICAL AND ANTIQUARIAN SOURCES In contrast to the scanty evidence just considered, we are comparatively fortunate with the traces of history and historiographical sources used by Didymus in works devoted to fields other than historiography. Here the direct tradition is represented by a very remarkable fragment of the Περὶ Δημοσθένους (°281; see below, §3.2.4). However, the greater part of the evidence comes via indirect channels, and identifying the Didymean material securely is an impossible task. All this makes it advisable to concentrate on fragments that are certain and to adopt as the guiding criterion for the following survey not the historians who interested Didymus, but his own works and the corresponding fields of interest.

3.1. Lexicographical Writings 3.1.1. Σ Ar. Pl. 388b Chantry, on ἀπαρτί, reports that Didymus in his Περὶ διαφορᾶς λέξεων (°3, Schmidt 1854: 20 and 412–13; Braswell 2017: 41–42)23 presented examples of the use of ἄρτι by Callimachus (fr. 609 Pfeiffer) and Plato (Ly. 215c): συνωνυμεῖ ἡ λέξις ( add. Schneider, prob. Chantry)· ἔσθ᾽ ὅτε γὰρ καὶ χρονικὸν ἐπίρρημα δηλοῖ VMEΘBarbAld ὡς καὶ Καλλίμαχος MEΘBarbAld (fr. 609 Pfeiffer) ‘ἄρτι θέναρ βάλλει’,24 καὶ Πλάτων (Ly. 215c) ἐπὶ τοῦ ‘νῦν’· ‘ἤδη ποτέ του ἤκουσα λέγοντος, καὶ ἄρτι ἀναμιμνήσκομαι’. VM ταῦτα ἐκ τῶν Διδύμου περὶ διαφορᾶς λέξεων. V The word (sc. ἀπαρτί?) has a synonymous sense: at times in fact it also has the sense of a temporal adverb, for example in Callimachus, ‘now (ἄρτι) the palm of the hand throws’, and in Plato, in the sense of ‘now’: ‘I have already once heard someone say, and now (ἄρτι) it has come to my mind.’ This (is) from Didymus’ work On the Different Meaning of Words. Apparently, the topic of Didymus’ discussion is the adverb ἄρτι: therefore, Schneider supplements at the start of the scholion, and is followed by Chantry.25 But, in that case, one cannot understand what other sense of ἄρτι could have been illustrated by Didymus, aside from the temporal one (ἔσθ᾽ ὅτε γὰρ καὶ χρονικὸν ἐπίρρημα δηλοῖ). Alternatively, the understood subject could be ἀπαρτί, the Aristophanic word (ἡ λέξις) that is the topic of the commentary. One could suppose, then, that Didymus’ observation was reported in the scholion in abbreviated form26 and was originally centred on the lexical and semantic relation between ἀπαρτί/ἀπάρτι with temporal sense and ἄρτι.27 This implies that Didymus also recorded one or both of the other senses of ἀπαρτί/ἀπάρτι documented in the learned tradition: ἀκριβῶς or ἀπηρτισμένως, ‘perfectly/ precisely’, and the derived antiphrastic acceptation ἐκ τοῦ ἐναντίου, ‘to the contrary’.28 Now, the 22 Gibson 1997: 380 n. 15; Sickinger 2007 and 2016. 23 The title is uniquely attested in this very scholion to Aristophanes (Ven. Marc. gr. Z 474), where it was corrected to Περὶ διεφθορυίας λέξεως by Dindorf, followed by Schmidt 1854: 20, 412–13 and Cohn 1903: 464. The transmitted title is convincingly defended by Braswell 2017: 41–42. 24 As conjectured by Schneider; ἄρτιθεν ἀβραλλεῖ M, ἄρτιθεν ἃ βράλλει V. 25 Cf. Chantry’s translation (2009: 378): ‘le terme […] est synonyme (de ἀπαρτί)’. 26 Pfeiffer 1949–53: I 416. 27 Braswell 2017: 41 n. 53. Antiatt. α 68 Valente (ἀπάρτι cod.) attests the temporal usage of the adverb in Plato’s Sophists (fr. 155 K.–A.). 28 The most complete treatment is Synag. B α 1637 Cunningham.

Didymus and the Greek historians



67

well-established locus classicus in the erudite tradition for the meaning ‘precisely’ is Hdt. 2.158.4.29 The possibility thus exists that the whole discussion on ἀπαρτί and the mention of Herodotus go back to Didymus’ Περὶ διαφορᾶς λέξεων. The same monograph seems to be indicated by a testimony in Ammonius’ lexicon of synonyms. In Diff. 451 Nickau (327b), the quotation of Didymus’ view on the different meanings of συμμαχεῖν (‘to be allied’ in either a defensive or offensive sense) and ἐπιμαχεῖν (‘to be allied’ only in a defensive sense) is closely connected to Thucydides’ vocabulary in 1.44.1: συμμαχεῖν καὶ ἐπιμαχεῖν διαφέρει. συμμαχεῖν μὲν γὰρ λέγουσι τὸ σὺν ἑαυτοῖς, φησὶ Δίδυμος, εἴτ᾽ αὐτοῖς ἐπίοιεν πολέμιοι, εἴτε αὐτοὶ ἑτέροις ἐπιστρατεύοιεν. ἐπιμαχεῖν δὲ ὅταν τοὺς ἐπιόντας ἀμύνωνται μόνον. διέσταλκε Θουκυδίδης ἐν τῇ πρώτῃ (1.44.1) λέγων ‘Κερκυραίοις Ἀθηναίους συμμαχίαν μὲν οὐ ποιήσασθαι, ἐπιμαχίαν δέ’. συμμαχεῖν and ἐπιμαχεῖν differ. For they call συμμαχεῖν the (action of fighting) ‘together with themselves’, says Didymus, whether enemies attack them or whether it is they who attack others. (They say,) on the other hand, ἐπιμαχεῖν only in the case where they need to defend themselves against attackers. It is Thucydides who fixed the distinction, when in Book 1 he says: ‘that the Athenians made with the Corcyraeans not a symmachia, but an epimachia’.30 It is probable that it was Didymus who produced the Thucydidean passage as the basis of his own lexical distinction. Schmidt assigns the explanation to a commentary on Thucydides,31 but the Περὶ διαφορᾶς λέξεων seems more appropriate, as the ancestor of the work attributed to Ammonius.

3.1.2. There are hints that in his Λέξις κωμική Didymus drew on evidence supplied by Greek historians.32 Hsch. μ 1824 Latte = 23 preserves a clue to Didymus’ use of Herodotus: Μυκερίνα· ἡ Μέμφις, Δίδυμος τὴν Σάϊν. ταύτης γὰρ λέγει Ἡρόδοτος (2.130.1) βασιλεῦσαι Μυκερῖνον, ‘Mycerina: Memphis, Didymus (says that it is the city also called) Sais, for Herodotus says that Mycerinus reigned over it’. The testimony of Herodotus was produced by Didymus himself, since it forms the argument that proves (γάρ) the identification of Μυκερίνα with Σάϊς. The pertinence of this fragment to the Λέξις κωμική is plausible because this work is among Hesychius’ stated sources.33 Harpocration (o 25 Keaney = 309a) attests that Didymus took an interest in the term ὀξυθύμια on two different occasions.34 The explanation begins by giving the attestation of the term in Hyperides’ Against Demades (fr. 79 Jensen), where the orator states that it would be more correct to put the inscribed stele (containing the decree of proxeny for Euthycrates, proposed by Demades) ἐν τοῖς ὀξυθυμίοις […] ἢ ἐν τοῖς ἡμετέροις ἱεροῖς. There follows a collection of opinions on the meaning of ὀξυθύμια, among them, based on the testimony of Didymus (Λέξις κωμική, 308a), that of the exēgētēs Autoclides (BNJ 353 F2): 29 Quoted in the form ἀπὸ τούτου εἰσὶ στάδιοι χίλιοι ἀπαρτί in Synag. B α 1637 Cunningham and in Σ Ar. Pl. 388a Chantry. In the direct tradition of the Histories the adverb is absent; however, after Bekker, Hude, and Rosén, N. G. Wilson also prints it in his critical edition (2015a; cf. 2015b: 44–45). The lexicographical explanation of ἀπαρτί as ἀπηρτισμένον could also occur in the anonymous hypomnēma P.Oxy. LXV 4455, col. ii.6–7 (see above, §2.1), concerning Hdt. 5.53. 30 Cf. 325a; Et. Gud. p. 515.7–9 Sturz; Σ Thuc. 1.44.1d Kleinlogel–Alpers. 31 Schmidt 1854: 334. 32 On Didymus’ Λέξις κωμική, see Benuzzi, this volume, 54–55. 33 Schmidt 1854: 27–29. 34 On Harpocration’s use of Didymus, see DiGiulio, this volume, 84–86.

68



Montana

Δίδυμος δὲ Αὐτοκλείδου (BNJ 353 F2) λέξιν παραγράψας ἐκ τῶν Ἐξηγητικῶν φησίν· ὀξυθύμια τὰ καθάρματα λέγεται καὶ ἀπολύματα. ταῦτα γὰρ ἀποφέρεσθαι εἰς τὰς τριόδους, ὅταν τὰς οἰκίας καθαίρωσιν. Didymus, quoting35 a passage from Autoclides’ Explanations, says: ὀξυθύμια means the refuse and dirt. For these are brought out to the crossroads when they clean the houses. Harpocration records at this point that Didymus, in his commentary on Hyperides’ speech (309a),36 also explained ὀξυθύμια as designating the offerings for Hecate deposited at crossroads: an interpretation corroborated by Eupolis’ Demoi fr. 132 K.–A. (which was presumably also part of Didymus’ argument).37 The first of Didymus’ explanations probably comes from the Λέξις κωμική, as is suggested by the correspondence with Hsch. ο 948 Latte38; but one cannot exclude that it would have found its place (also) in Didymus’ hypomnēma to Eupolis,39 if he ever composed one.40

3.1.3 We possess a testimony for the use of historical sources in Didymus’ Λέξις τραγική (39ab). In Saturnalia 5.18.2–5, Macrobius takes an interest in the antonomasia ‘Achelous = fresh water’ present in Virgil (G. 1.7– 9) and cites Aristophanes as a parallel (fr. 365 K.–A.).41 He continues (5.18.6–8) by reporting the rationalistic explanation supplied by Ephorus (BNJ 70 F20a). According to the latter, the habit of calling any watercourse ‘Achelous’ derived from the generalization, or mistaken interpretation, of the instruction, which was formulaic in the oracular responses of the sanctuary of Dodona, to offer sacrifices to this river. At this point, Macrobius cites the opinion of Didymus (5.18.9–10 = °39a). The latter reported Ephorus’ explanation, but preferred the mythographic interpretation proposed by Acusilaus (Didymus […] posita causa quam superius Ephorus dixit, alteram quoque adiecit his uerbis): ἄμεινον δὲ ἐκεῖνο λέγειν, ὅτι διὰ τὸ πάντων τῶν ποταμῶν πρεσβύτατον εἶναι Ἀχελῷον τιμὴν ἀπονέμοντας αὐτῷ τοὺς ἀνθρώπους πάντα ἁπλῶς τὰ νάματα τῷ ἐκείνου ὀνόματι προσαγορεύειν. ὁ γοῦν Ἀκουσίλαος (BNJ 2 F1 = fr. 1 Fowler; Ἀγησίλαος vel Ἀγεσ- codd., corr. Gronovius)42 διὰ τῆς πρώτης ἱστορίας δεδήλωκεν ὅτι Ἀχελῷος πάντων τῶν ποταμῶν πρεσβύτατος. ἔφη γάρ· Ὠκεανὸς δὲ γαμεῖ Τηθὺν ἑαυτοῦ ἀδελφήν, τῶν δὲ γίνονται τρισχίλιοι ποταμοί, Ἀχελῷος δὲ αὐτῶν πρεσβύτατος καὶ τετίμηται μάλιστα. It is preferable to say as follows, that men call all watercourses by the name Achelous as an act of homage, since it is the most ancient of all rivers. Or, at least, Acusilaus, in the first book of his History, has made clear that Achelous is the most ancient of all rivers. For he says: ‘Ocean wed his sister Tethys and from them were born three thousand rivers, and Achelous is the eldest of them and is held in greatest honour.’ Macrobius continues by adding a further observation of Didymus, taken from his lexicon of tragedy (5.18.11–12: etiam Euripidis nobilissimi tragoediarum scriptoris addetur auctoritas, quam idem Didymus grammaticus in his libris quos τραγῳδουμένης λέξεως scripsit posuit his verbis): 35 λέξιν παραγράψας ἐκ does not mean ‘commenting upon an expression from’ (Jones 2010): see Montana 2014a: 35–38; Braswell 2017: 82 with n. 222. 36 Cf. Braswell 2017: 82. 37 Olson 2017: 454. 38 Cf. Phot. Lex. ο 379 Theodoridis (from the epitome of Harpocration’s lexicon); Moer. ο 26 Hansen. 39 This is suggested by the sentence ταῦτα γὰρ ἀποφέρεσθαι εἰς τὰς τριόδους, ὅταν τὰς οἰκίας καθαίρωσιν, which picks up and expands the hendiadys present at the start of Eupolis’ fragment cited in the second explanation (ὃν χρῆν ἔν ταῖς τριόδοις κἀν τοῖς ὀξυθυμίοις κτλ.). 40 The sources collected under Eup. T48 K.–A. document Didymus’ interest in Eupolis (cf. Olson 2017: 88), but not that he composed a hypomnēma on his comedies: Braswell 2017: 60–63. 41 Further Greek parallels in Parker 2011. On Macrobius’ use of Didymus, see DiGiulio, this volume, 90–93. 42 Kaster 2011a: 328 keeps Ἀγησίλαος in the text as an author’s error (cf. Kaster 2011b: II 427 n. 74).

Didymus and the Greek historians



69

Ἀχελῷον πᾶν ὕδωρ Εὐριπίδης φησὶν ἐν Ὑψιπύλῃ (fr. 753 Kannicht). λέγων γὰρ περὶ ὕδατος ὄντος σφόδρα πόρρω τῆς Ἀκαρνανίας, ἐν ᾗ ἐστιν ὁ ποταμὸς Ἀχελῷος, φησίν· ‘δείξω μὲν Ἀργείοισιν Ἀχελῴου ῥόον’. Euripides in Hypsipyle calls every kind of water ‘Achelous’: speaking of a watercourse that is rather far from Acarnania, which is where the river Achelous is, he says: ‘I will show to the Argives the course of the Achelous’. Macrobius says nothing about the origin of the first Didymean quotation.43 The discussion of the polysemy of the name Ἀχελῷος would have had an apt place in Didymus’ Περὶ διαφορᾶς λέξεων. Here it is important to stress that, in his investigation into the origin of this antonomasia, Didymus makes use of two historical-mythographic sources (Ephorus, Acusilaus) and one poetic authority (Euripides), and that he is not content merely to report them, but takes up a position. It is not superfluous to highlight that the point of the discussion is not, naively, the truthfulness of the mythical tale, but the origin of a linguistic usage.44

3.2. Commentaries and Critical Treatises 3.2.1. As we know, in the field of Homeric scholarship Didymus produced a monograph on the Aristarchean diorthōsis of the poems (Περὶ τῆς Ἀρισταρχείου διορθώσεως), which was then excerpted by the compiler of the so-called VMK scholia to the Iliad, and in addition commentaries of his own on the poems.45 Among the countless annotations of alleged Didymean origin preserved in the VMK class there is a remarkable lack of historiographical sources. In Σ Did. Il. 24.110b1 (A) Apollodorus (of Athens; BNJ 244 F268) and Aretades (the historiographer from Cnidos?)46 are mentioned, but only as witnesses for a variant reading.47 That, it seems, is all. This absence is likely to be explained by the purely text-critical focus of Didymus’ work on the Aristarchean diorthōsis, and of the latter itself as well. We get a little more from the class of the exegetical scholia to the Iliad. Two of them explicitly quote Didymus, who in turn is relying on Pherecydes (of Athens).48 In Σ ex. Il. 7.135c (AT) = 56a, concerning the Homeric line Φειᾶς πὰρ τείχεσσιν, Ἰαρδάνου ἀμφὶ ῥέεθρα, we read: καὶ ἡ Φειὰ παραθαλάσσιός ἐστι, καὶ Ἰάρδανος οὐχ ὁρᾶται ποταμὸς αὐτόθι. ἄμεινον οὖν ὡς Δίδυμος ‘Φηρᾶς’ γράφειν καὶ ‘Δαρδάνου ἀμφὶ ῥέεθρα’. οὕτω γὰρ καὶ Φερεκύδης ἱστορεῖ (BNJ 3 F159 = fr. 159 Fowler). Pheia is near the sea and, in addition, a river (called) Iardanus is not visible there. Therefore, it is better to write instead, with Didymus, Φηρᾶς and Δαρδάνου ἀμφὶ ῥέεθρα. For Pherecydes too tells it in this way.

43 Schmidt places the whole passage among the fragments of the Λέξις τραγική. 44 Among the fragments of Didymus’ Λέξις τραγική, Schmidt (1854: 90–91; cf. Cappelletto 2003: 296–98) lists Hsch. β 237 Cunningham, where the words Βαρκαίοις ὄχοις (S. El. 727) are explained by adducing Μνασέας ἐν τοῖς περὶ Λιβύης (FHG III 156 fr. 40 = fr. 43 Mehler = fr. 43 Cappelletto). However, this attribution rests on speculative arguments. 45 See Pagani in this volume. 46 In favour of the identification: Müller in FHG IV 316 (cf. Ippolito 2006). Jacoby holds that Aretades of Cnidos is an invented person: introduction to FGrHist 285; cf. FGrHist 1759 (Chrubasik 2015). 47 προτιάπτω, against προιάπτω, preferred by Zenodotus and Aristarchus. Apollodorus is probably also the source of the varia lectio κέαται instead of νέαται in Il. 9.153 (BNJ 244 F306), although the name Ἀπολλόδωρος adduced in Σ Did.(?) Il. 9.153d1 (T) competes with Ἀπολλώνιος in Σ Did.(?) Il. 9.153c (A). 48 On Pherecydes quoted by Didymus, see also above, §2.2.

70



Montana

The excerptor of this remark is apparently attributing the correction in the Homeric line to Didymus himself (that is, not to Aristarchus via Didymus)49 and referring to Didymus’ original argument, as pointed out by οὖν. It is beyond doubt that the quotation of Pherecydes’ testimony goes back to Didymus, because of the determining weight of this source for the argument and the presence of γάρ as a clue in this direction. The second instance is Σ Did. vel ex. Il. 19.116a1 Erbse (A) = 83: ἄλοχον Σθενέλου: Δίδυμος (p. 162 Schmidt) παρατίθεται Φερεκύδην (BNJ 3 F68 = test. 7Aa, fr. 68 Fowler) μὲν λέγοντα αὐτὴν τὴν Πέλοπος Ἀμφιβίαν, Ἡσίοδος (fr. 191 Merkelbach–West) δὲ δὲ Ἀντιβίαν τὴν Ἀμφιδάμαντος ἀποφαίνεται. This scholion attests that Didymus dealt with the problem of identifying Sthenelus’ wife, whose name is not mentioned in the Homeric passage, and that he drew on Pherecydes, who argued for Amphibia, against the alternatives offered by Hesiod (Nicippe) and another author whose name is not recorded (Antibia).50 Both of these quotations of Pherecydes go beyond the Aristarchean diorthōsis and, hence, in all likelihood, come from Didymus’ own hypomnēma to the Iliad. Moreover, they are perfectly in line with what we find in the twenty or so explicit mentions of Pherecydes in the Iliad scholia, which are usually adduced in the identification of characters and settings of mythological episodes recalled in the poem. Therefore, it is not overly bold to suppose that (some of) these quotations also go back to Didymus.51 The Odyssey scholia are apparently even less generous than those to the Iliad, because we can trace only one possible instance of Didymus’ (indirect) use of historical sources. According to Σ Did. Od. 1.259e (HMa) Pontani, in the Homeric line τινὲς ‘Ἴρου’ γράφουσιν, ἐπεὶ καὶ Πρόξενος ἐν Ἠπειρωτικοῖς (BNJ 703 F3) Ἶρόν φησι Μερμέρου παῖδα, ‘some write Ἴρου (sc. instead of Ἴλου), because Proxenus in the History of Epirus speaks of Irus the son of Mermerus’.52 The notice can be traced back to Didymus on the basis of the class of scholia that transmit it (MSS HMa). We cannot say whether behind τινές stands the Chalcenterus himself or if he simply reported a critical intervention formulated by others.

3.2.2. This last case introduces Didymus’ use of local historiography, something which is frequently attested in the Pindar scholia. ‘A prominent feature of Didymus’ exegesis (sc. on Pindar) is his use of historians to explain references in the text and settle points of dispute.’53 The quantitative divide from the testimony of the Homeric scholia is striking, to the point that it may appear not only to be related to the intrinsic character of Didymus’ writings, but also to depend on the choices of the excerptors, readers, and copyists in the different fields. We owe to Braswell a catalogue of Didymus’ quotations of historical sources extant in the Pindar scholia.54 Local history is represented by Ister the Callimachean (BNJ 334 F41; 112ab), Theotimus of Cyrene (BNJ 470 F1; 132), Pythaenetus of Aegina (BNJ 299 F2a; 157), and three further sources now anonymous (115, perhaps Ergias of Rhodes, BNJ 513 F1; 136, 137). Chronographic and antiquarian sources are quoted in no less than five Didymean comments: Apollodorus of Athens twice (BNJ 244 F69 and 127; 104 and 109); once each Polemon of Ilium (or so it seems: fr. 26 Preller = fr. 11 Capel Badino; 116; cf. Schmidt 1854: 220),55 Theophrastus (Περὶ εὑρημάτων fr. 734 FHS&G; 126), and a list of Nemean victories (ἐν τοῖς 49 Lehrs 1882: 238 n. 151; Ludwich 1884–85: I 276. Aristarchus’ reading should be Φειᾶς, commented on in Σ Ariston. Il. 7.135a (A) and Σ Hrd. Il. 7.135b (A). 50 Despite Erbse’s hesitation concerning the scholiastic class of this annotation (VMK or exegetica), its Didymean content is beyond doubt. 51 Only three of the other Iliad scholia that quote Pherecydes are not labelled exegetica by Erbse: Σ Hrd. 2.585 (A), where Pherecydes (fr. 168 Fowler) is adduced to confirm the right form of the name Οἴτυλος; Σ Hrd. 2.592 (A), where Pherecydes’ reading Ἐΰκτιτον αἰπύ (fr. 169 Fowler) is rejected in favour of ἐΰκτιτον Αἰπύ; and Σ Ariston. 16.175b (A), a ὅτι-scholion, where [Aristarchus] points out a mistake by Pherecydes (fr. 61a Fowler) in identifying the Peleus mentioned in the Homeric passage with Achilles’ father (cf. Σ ex. 16.175c1 (T)). 52 On Proxenus, court historian to King Pyrrhus, see Rzepka 2011b. 53 Braswell 2017: 113. 54 Braswell 2017: 113–16. See also Prodi, this volume, 26–28. 55 Cf. Capel Badino 2018: 183–84, who is also inclined to include his fr. 20 = fr. 23 Preller among Didymus’ quotations of Polemon within the Pindar scholia (pp. 253–54).

Didymus and the Greek historians



71

Νεμεονίκαις, 162). The citations from works of general history, however, form the most strongly represented category, amounting to eight or nine instances: one or possibly two from Ephorus (BNJ 70 F18b and 21; 125 and °172a),56 the others from Philistus of Syracuse (BNJ 556 F49; 114) and Timaeus of Tauromenium, who is the most frequently quoted overall (six mentions: BNJ 566 F93b, 96, 39b, 145, 142a, 18; 105ab, 114, 117, 127, 140, 163). All these sources are mentioned in order to illustrate Realien: customs and practices, athletic games, a divine epiclesis, toponyms, anthroponyms, genealogies, city settlement, chronology. Philistus and Timaeus are usually adduced to address issues pertaining to Greek Sicily. The ratio of the historical quotations (20) to the total number of certain fragments of Didymus’ exegesis on Pindar (68 in Braswell’s edition) is close to 1 in 3.57 This short review allows us to appreciate how and how much Didymus could draw on historians when performing his own scholarship.

3.2.3. Ancient scholarship on Attic drama is much less generous ground in yielding evidence for our investigation.58 Unfortunately, we do not have explicit information on the reuse of Didymus’ writings in the preserved scholiastic corpora to dramatic works, with the exception of the final subscription to Euripides’ Medea in MS Par. gr. 2713 (fol. 129r: διάφορα ἀντίγραφα Διονυσίου ὁλοσχερὲς καί τινα Διδύμου). In addition, only a few scholia, out of the very many that quote historical sources, are explicitly derived from Didymus: some in the published scholia to Aristophanes and just one in those to Euripides; none in those to Aeschylus (where the name of Didymus never appears) or to Sophocles. Of the several ‘fragments’ collected by Schmidt (1854: 240–61), the majority are the product of his heavily attributionist approach (1854: 261–99). Some criteria of attribution are based on aspects that are documented in the erudite tradition and do ensure some reliability;59 others appeal to linguistic features and analogies in expression that are solely formal and external, with an indicative weight that is hardly reliable, and easily cross over into begging the question.60 Notwithstanding the scepticism of some, Schmidt’s method won a following.61 Wilhelm Meiners applied it to the scholia to Aristophanes’ comedies when searching for historiographical subjects commented upon by Didymus. Beginning from the four annotations that explicitly cite Didymus for issues related to history and historians, he reached the amazing total of forty-nine additional fragments that he attributed to him by conjecture.62 The progressive increase in our knowledge of (and questions about) ancient scholarship invites us to look at the documentation with greater respect and prudence. The few secure cases allow us to describe Didymus’ approach to historical matters in Aristophanes’ comedies in the following terms: i. The explanation of a reference to a political-military circumstance by adducing a piece of epigraphic evidence reported in an antiquarian source (254). According to ‘Didymus and Craterus’ (i.e. Didymus quoting Craterus: BNJ 342 F17), in Lys. 313 the Chorus of Old Athenians mentions ‘the stratēgoi stationed at Samos’ in order to attack allusively (αἰνίττεσθαι) Phrynichus the son of Stratonides: the Athenians inflicted severe penalties on him for his conduct during his generalship at Samos (411 BC; Th. 8.54.3).63 The idea that the 56 The latter is the only certain remnant of Didymus’ commentary on the Paeans and one of only two fragments collected by Braswell that are not transmitted among the Pindar scholia. 57 The calculation does not consider the citation of Mnaseas, Περὶ χρησμῶν fr. 56 Cappelletto = fr. 372 Parke-Wormell (fr. dub. 70 Braswell), which Cappelletto 2003: 337 believes to be genuinely Didymean. 58 On Didymus’ interest in Eupolis, see above, §3.1.2. See also Coward and Benuzzi in this volume, on tragedy and comedy respectively. 59 An acceptable criterion is convergence between scholia and Hesychius’ Lexicon, since the latter lists Didymus’ Λέξις κωμική and Λέξις τραγική among its sources (via Diogenianus); e.g. Montana 1996: 69–83. 60 e.g. Schmidt 1854: 296–97. 61 As well as Meiners 1890 (on whom see next in the text), this method attracted Boudreaux 1919: 110–19. Caution is urged by e.g. White 1914: lxix–lxx; Gudeman 1921: 678; Wilson 1983: 93; Gibson 2002: 28. 62 Meiners 1890: 239: ‘ita paene certum est Didymum, […] omnia quae sibi non satis nota uiderentur, commentatum esse, ut scholia historica in uniuersum dicere liceat ex eodem fonte, Didymi commentario, fluxisse’. 63 The link established by Didymus appears wrong: Jacoby 1955 ad loc.; Henderson 1987: 106; cf. Carawan 2007 (BNJ) ad loc.

72



Montana

comic poet is alluding specifically to Phrynichus ought to belong to Didymus, whereas Craterus’ Ψηφισμάτων συναγωγή would have provided only the text of the decree.64 ii. The recognition of a comic double entendre with a political background by consulting an Atthidographic source (257). The expression οὑκ τῶν ὀρνέων in Av. 13, explained by others as a periphrasis for the birdseller, is taken by Didymus to allude to the locality of Ὀρνεαί and is meant to recall the defeat suffered by the Athenians at Mantinea four years previously (418 BC), καθὰ καὶ Ἀνδροτίων φησίν, ‘as also Androtion says’ (BNJ 324 F41). The explanation is overingenious, but it is important to note that Didymus’ specific interest here is in the historical background and chronology of Aristophanic dramas.65 iii. The enquiry into the characterization of a real individual mentioned in a play (267). Ran. 970, οὐ Χῖος ἀλλὰ Κεῖος, connotes ironically the political opportunism of Theramenes, which permits him to escape unscathed from every difficult situation. Aristarchus (Σ 970b Chantry) saw in the expression a reference to the name of two opposite scores in dice-throwing (Χῖος = 1, Κῷος = 6). Demetrius (Ixion) (Σ 970c) held that Aristarchus was mistaken, since ‘(Theramenes) was from Κεῖος’, and that therefore Κεῖος (not Κῷος) is the genuine reading in the passage. An anonymous commentator defended Aristarchus, observing that his explanation remains valid if we read Κεῖος and understand it as an aprosdokēton with respect to the expected Κῷος (Σ 970d).66 Finally, Didymus (Σ 970e), who seems to take for granted the entire previous discussion, also maintained the reading Κεῖος and asserted a possible (δύναται) proverbial sense of the phrase67 to indicate the mutability (ποικίλος τις ὢν καὶ ἀγχίστροφος) that permits an individual to adapt to circumstances thanks to small and subtle changes (like that from Χῖος to Κεῖος).68 iv. The identification of real persons mentioned in a play (271). Line 550 of Plutus, where a Thrasybulus and a Dionysius are recalled as examples of individuals who are opposed and cannot be assimilated (ὑμεῖς γ᾽ οἵπερ καὶ Θρασυβούλῳ Διονύσιόν φατ᾽ εἶναι ὅμοιον), is accompanied by a composite array of annotations in the medieval manuscripts.69 Three of these notes (Σ 550a, b, e Chantry), which can be attributed to the same source, quite obviously recognize in the two individuals the Athenian democratic leader and the tyrant of Syracuse contemporary to Aristophanes. A different commentator (Σ 550c), though accepting this identification, is inclined to hypothesize (μήποτε) that the opposition is playing on the polarity not between love of democracy and of tyranny, but rather between arrogance (Thrasybulus) and madness (Dionysius): negative qualities of the two that are attested in comedy, respectively, by Strattis (fr. 20 K.–A.) and Polyzelus (fr. 12 K.–A.). Someone (Σ 550d) did not exclude, again by conjecture (μήποτε), that the counterpoint focuses ironically on a physical (inasmuch as visible: περὶ τὴν ὄψιν) difference between the two individuals. Finally, someone attempted an entirely different path (again Σ 550e) by hypothesizing (μᾶλλον ἄν τις ὑπονοήσειε) that the Dionysius mentioned in the Plutus is the brother of Thrasybulus of Collytus, namesake and friend of the democratic leader. This last annotation is preserved only in the Estensis manuscript (E),70 where it is followed by an unintelligible mention of Didymus (text according to Chantry): †καὶ ἐν τῇ Ἰλιάδι† σαφέστερον 64 Meiners 1890: 225–26 and 373; cf. Erdas 2002: 209–10; Carawan 2007 (BNJ) ad loc. 65 The temporal clarification belongs to Didymus, it being improbable that Androtion recorded the chronology of Aristophanes’ Birds: Jacoby 1954 and Jones 2015 ad loc. 66 Muzzolon 2005: 97–98. 67 Proverbs with the pair Χῖος/Κῷος: Zen. 4.74 and Diogenian. 5.70, Χῖος πρὸς Κῷον; Stratt. fr. 24 K.–A. ap. Arsen. in Apostol. 18.26a, Χῖος παραστὰς Κῷον οὐκ ἐᾷ λέγειν, cf. App. Prov. 5.28, Eust. in Od. 1397.39 Stallbaum. Perhaps Didymus collected them in his Περὶ παροιμιῶν and was inspired by them in his conjecture of a proverb that plays instead on Χῖος/Κεῖος. The Aristophanic expression, labelled παροιμία by Didymus, is in Arsen. in Apostol. 14.16b. 68 Cf. Rogers 1902: 150. Meiners 1890: 306 indicates Xenophon (HG 2.3.3–33) as Didymus’ hidden source for the characterization of Theramenes. An interest in this politician is documented in P.Flor. II 112 (CLGP I.1.4 Aristophanes 28, second century AD), fragments of an anonymous (Didymus’? Crönert 1908a: 1199 and 1908b: 1391) commentary on a lost comedy attributable to Aristophanes (frr. C+D+E, col. i 23–25): Montana 2012b: 183–211. 69 Schmidt 1854: 290–91; Meiners 1890: 227–30; Allen 1929: 29. 70 Modena, Est. α.U.5.10, known for its conservative character (e.g. D. M. Jones in Jones and Wilson 1969: xi–xii). The scholion is repeated in the Aldina, which depends on the Estensis.

Didymus and the Greek historians



73

οὐδὲν (con. Holwerda, οἶδεν cod.) ἢ κατὰ Δίδυμον.71 This annotation is at the origin of a lively debate, not least due to the difficulty of recognizing the actual contribution of Didymus in these scholia.72 One may note only that the explanation given in Σ 550a/b/e points to the political character of the counterpoint (democracy vs tyranny), whereas Σ 550c and 550d bring it back within the limits of satirical caricature and so to a question that concerns an artistic aspect and is literary in nature. These few examples positively attest some of Didymus’ hermeneutical approaches to real individuals and facts cited in Aristophanes’ comedies. Explicit references to people as well as hidden and potentially misleading allusions (i. and ii.; αἰνίττεσθαι in i.) prompt Didymus to consult specialized sources (Craterus, Androtion) and sometimes demand intuitive insight and interpretative effort, including conjecture (i.–iv.; δύναται in iii. and, possibly, μήποτε twice in iv.). Furthermore, one perceives the interest in chronological evidence through establishing synchronisms between comedies and historical events (i. and ii.) – a critical issue that was typical of Alexandrian scholarship on drama from its earliest times. Here our concern is not to judge the correctness of Didymus’ explanations, which modern scholars deny often and on good grounds, but to acknowledge his problematizing attitude and his use of specialized tools. This dossier is enriched by a comment of Didymus reported in a scholion to Euripides’ Hecuba. In line 887, the Trojan queen reminds Agamemnon that women are capable of overcoming men and gives the examples of the Danaids and the women of Lemnos. As regards the second case, the scholia (ed. Schwartz) adduce a few narrative variants of two distinct episodes that are in some way linked: i. The women of Lemnos had snubbed Aphrodite and the goddess punished them by making them repugnant to their husbands; the latter had intercourse with some Thracian slaves and therefore the Lemnian women killed their own husbands and fathers (with only Hypsipyle sparing her father Thoas).73 ii. According to Didymus (198), (a) the Pelasgians abducted the Athenian women and carried them off to Lemnos, had intercourse with them and had children. The mothers raised them like true Athenians and they grew up as a tightly knit community foreign to that of the Pelasgians. But the latter, frightened by their behaviour, killed them together with their mothers; ‘after that’, the scholion concludes, (b) ‘the Lemnian women also killed all the men who were together with Thoas (cf. i.); and due to both these events, the paroimia arose of Λήμνια κακά, Lemnian misdeeds’.74 In Didymus’ account (ii.), the link between (a) and (b) under the common denominator of the proverbial saying goes back at least to Herodotus (6.138). After a diffuse narration of episode (a), the historian alludes succinctly also to episode (b), indicating that it was prior (προτέρου) to the other; and he concludes that it is from these events that the Greek usage derives of ‘calling all wicked actions Λήμνια’. The shared narrative and train of thought demonstrate that Didymus’ explanation derives directly, though condensed, from the Herodotean passage. The only essential difference concerns the relative chronology of the two episodes: episode (b) preceded episode (a) according to Herodotus, while in Didymus’ account the order is the reverse. But this may depend on an error by Didymus or on an accident that occurred in the textual tradition, due to the order in which the two facts are actually narrated in the Histories, namely (a) before (b).75 The most important point is that Didymus is compiling from Herodotus, apparently keeping as the main focus of his own exegesis the origin and meaning of the proverb Λήμνια κακά rather than Hecuba’s words in the Euripidean passage.

71 In the view of Allen 1929: 29 (following Dindorf) the phrase should be referred rather to Ra. 458 or 545. 72 Meiners 1890: 226 with n. 10 assigned Σ 550c and 550d to Didymus, since they begin with the adverb μήποτε, which was held to be a distinctive mark of Didymean authorship. 73 Cf. Caucal. BNJ 38 F2; Myrs. Metymn. BNJ 477 F1; Apollod. Bibl. 1.9.17, 3.6.4. 74 Cf. Zen. 4.91; Diogenian. 6.10; Apostol. 10.65; Phot. λ 271 Theodoridis; Suda λ 451 Adler. 75 Photius and the Suda (see previous n.) seem to imply Didymus’ temporal sequence.

74



Montana

3.2.4. In the field of prose writers, the evidence clusters around Demosthenes. The indirect tradition restores three or four Didymean fragments that adduce historiographical and antiquarian authorities. The most generous witness is Harpocration’s Lexicon of the Ten Orators.76 i. A comment by Didymus on the expression ὁ κάτωθεν νόμος occurring in Demosthenes’ Against Aristocrates (23.28) is preserved twice, in Harp. ο 14 Keaney (292b) and in the fragmentary lexicon P.Berol. inv. 5008 verso 7–24 (292a).77 The two testimonia go back to a lost shared source, which combined three different explanations of ὁ κάτωθεν νόμος (292b: see Did. fr. e Schmidt 1854; Did. ap. Harp. fr. 10, Gibson 2002: 141–42; Braswell 2017: 72–73, no. 30). From Harpocration it emerges that all three explanations belonged to Didymus.78 According to the latter, the word κάτωθεν in Demosthenes’ passage may (a) refer to the Heliaia, understood as the ‘lower’ court (in contrast to the ‘upper’ court of the Areopagus?), or (b) mean ‘from below’ in the sense of ‘from the lower left (of the stone)’, because ‘Euphorion in Apollodorus (fr. 6 Powell = 8 Lightfoot) made it clear that axones and kyrbeis (sc. the Solonian laws thereon) were written boustrophēdon’. The third interpretation (c), the Didymean authorship of which is disputed, hypothetically connects Demosthenes’ phrase to the fact that Ephialtes had the axones and kyrbeis transferred from the Acropolis down to the bouleutērion and the agora, as Anaximenes says in his Philippika – thus in Harpocration’s words: τοὺς ἄξονας καὶ τοὺς κύρβεις ἄνωθεν ἐκ τῆς ἀκροπόλεως εἰς τὸ βουλευτήριον καὶ τὴν ἀγορὰν μετέστησεν Ἐφιάλτης, ὥς φησιν Ἀναξιμένης ἐν Φιλιππικοῖς (BNJ 72 F13). Two interpretations (b–c) out of three, therefore, appeal to the Solonian axones and kyrbeis. Plausibly Didymus took advantage of the same information in his monograph on this subject (see above, §2.3), as well as in a lexicographical work or his exegesis of Demosthenes’ speech.79 ii. Ath. 11.486c–d and Harp. λ 31 Keaney (299ab) attest that the word Λυκιουργεῖς, an attribute of φιάλας in Demosthenes’ Against Timotheus (49.31), was explained by Didymus as meaning ‘manufactured by Lycius’ (Did. fr. g Schmidt 1854; Did. ap. Harp. fr. 17, Gibson 2002: 144–45; Braswell 2017: 74–75, no. 32). We learn from Athenaeus that Didymus identified the artisan based on (epigraphic evidence produced by) Polemon of Ilium: ἦν δὲ οὗτος τὸ γένος Βοιώτιος ἐξ Ἐλευθερῶν, υἱὸς Μύρωνος τοῦ ἀνδριαντοποιοῦ, ὡς Πολέμων φησὶν ἐν α´ περὶ ἀκροπόλεως, ‘he (sc. Lycius) was by birth a Boeotian from Eleutherae, son of Myron the sculptor, as Polemon says in the first book of On the Acropolis (of Athens: fr. 2 Preller = fr. 2 Capel Badino)’. But this interpretation was rejected by a commentator, now anonymous, because words shaped like Λυκιουργεῖς usually derive from names of towns or peoples.80 iii. Harp. π 63 Keaney asserts that, according to Didymus (300a: see Did. fr. h Schmidt 1854; Did. ap. Harp. fr. 18, Gibson 2002: 145; Braswell 2017: 76, No. 33), the word περίστοιχοι occurring in Demosthenes’ Against Nicostratus (53.15) designates a variety of olive tree and corresponds to what Philochorus called στοιχάδες (BNJ 328 F180): Δίδυμος δέ τι γένος ἐλαιῶν περιστοίχους καλεῖ, ἃς Φιλόχορος στοιχάδας προσηγόρευσε. The wording raises the impression that Philochorus was cited by Didymus himself.81 iv. Similar to these cases is a probable attestation of Didymean misuse of a historiographical source. Harp. γ 2 Keaney s.v. γαμηλία, ‘wedding feast’, reports that the grammarian explained this word in his commentaries 76 See also DiGiulio, this volume, 84–86. 77 The Berlin papyrus, from a codex of the fourth or fifth century AD, lists in alphabetical order and explains six lemmata drawn from Demosthenes’ Against Aristocrates by quoting Philochorus, Theopompus, Anaximenes, Aristotle, Xenophon, Ephorus, and Didymus. Gibson 2002: 157–71; Harding 2006: 32–34; cf. Stroppa 2009: 323. Photo at BerlPap, . 78 Gibson 1997 and 2002: 169 maintains that only the first two concern Didymus. But the lacunae in the papyrus impede any secure conclusion about the third explanation: Ucciardello 2012: 44 n. 37. 79 Braswell 2017: 73 argues for a lexicographical work, perhaps Περὶ ἀπορουμένης λέξεως. 80 Examples in Preller 1838: 37–38. Similarly, Harpocration, who, however, omits mention of Didymus’ source Polemon. For further quotations of the latter by Didymus, see above, §§2.2.ii, 3.2.2. 81 Braswell 2017: 76 n. 199.

Didymus and the Greek historians



75

on Isaeus (cf. Is. 3.76, 3.79, 8.18, 8.20; Schmidt 1854: 320; Braswell 2017: 83, no. 42) and on Demosthenes’ Against Eubulides as well (57.43, 69; °302  °313; see Did. fr. i Schmidt 1854; Did. ap. Harp. fr. 20, Gibson 2002: 145; Braswell 2017: 77, No. 34). In the former, Didymus understood the word as ‘(the feast) given at weddings to one’s fellow phratry members’ and corroborated the explanation by ‘adducing a passage of Phanodemus (BNJ 325 F17), in which nothing of the kind is written’, παρατιθέμενος λέξιν Φανοδήμου, ἐν ᾗ οὐδὲν τοιοῦτον γέγραπται. However—Harpocration continues—in his writings on Demosthenes (ἐν δὲ τοῖς εἰς Δημοσθένην) Didymus explained the word as ‘the introduction of wives to the phratry members’, this time ‘without giving any argument in support of his interpretation’, οὐδεμίαν ἀπόδειξιν τῆς ἐξηγήσεως παρατιθέμενος. The lexicographer (or his source) seems to imply that Didymus noticed the error and changed his mind—unless he is insinuating Didymus’ unreliability or his proverbial forgetfulness due to the countless number of his own writings (βιβλιολάθας, T5).82 The variety in the spheres represented in such a small sample of fragments is impressive: lexicography, antiquarianism, local history. Equally important is the inventory of antiquarian and Atthidographic sources quoted (Polemon, Philochorus, Phanodemus) or presupposed (the long-standing debate on the Solonian axones and kyrbeis). Beyond the actual value of Didymus’ explanations, these scant remains make it regrettable that the indirect tradition has not preserved more explicit quotations of his scholarship on the Attic orators.83 However, we cannot really lament the ill fortune of Didymean scholarship on Demosthenes, since we have a genuine ‘pearl’ of direct tradition: P.Berol. inv. 9780 (°281), a large portion of a roll found at Hermoupolis and datable to the second century AD, containing on the recto substantial portions of a work which the preserved subscription identifies as Didymus’ Περὶ Δημοσθένους.84 In what remains of fifteen columns of text we read lemmata drawn from the Philippics (speeches 9–11 and 13) followed by comments. Here we find concentrated a truly significant number of sources, literary, erudite, and above all historiographical, which are frequently quoted word for word, and generously so. Philochorus (thirteen mentions) and Theopompus (seven, of which five are securely from his Philippika) are favourite sources to illustrate aspects of political and military life in the age of Philip II. The Atthidographic tradition is also represented by Androtion and Demon; general historiography (Makedonika and histories of Alexander) also by Anaximenes of Lampsacus, Marsyas of Pella, Duris of Samos, and Callisthenes of Olynthus. Fourth- and third-century historians of Athens and Macedonia appear to be the sources preferred by Didymus for Realien, chronology, and historical facts.85 Didymus’ manner of proceeding is exemplified well in the section dedicated to the capture and death of Hermias, the tyrant of Atarneus (coll. iv.59–vi.62). Didymus begins by observing that there are opposing judgements regarding this historical figure, and for that reason he finds it appropriate to examine the different points of view to satisfy his readers’ ‘curiosity’ (iv.59–63: ἐπεὶ δ᾽ εἰς | [μεγίστην διαφ]ο* ρὰν ἥκ* ο* υ* σιν ο[ἱ] τὰ περὶ τὸν | [Ἑρμίαν π(αρα)δεδω]κότες, τῆς φιληκοίας ἕ|[νεκα τῶν κ(αὶ) νῦν π]ολυπραγμονούντ(ων) τὰ τοιαῦ|

82 Symmetrically, Harp. π 48 (Schmidt 1854: 310–11) attests, concerning the Polystratus mentioned in Demosthenes’ Philippics (4.24) as the person in charge of a mercenary force at Corinth, that ‘Didymus says he has not found (this information) anywhere’. Gibson 2002: 146. 83 It is worth remarking that P.Rain. 7, remnants of a lexicon of Demosthenes’ Against Midias, datable to the fourth or fifth century AD as the Berlin papyrus, quotes Arist. Ath. 53.2–6: Gibson 2002: 190–99. 84 Diels and Schubart 1904; Pearson and Stephens 1983; Gibson 2002; Harding 2006; Braswell 2017: 80 (no. 38). Photo at BerlPap, . 85 Gibson 2002: 26–42, esp. 36–37; Harding 2006: 20–31. The preponderance of historical sources could depend on a selection from Didymus’ original work: Gibson 2002: 68–69.

76



Montana

[τα ἐπὶ πλέον] δ* [ο]κῶ μοι περὶ τούτων [εἰ]πεῖν). Among the judgements hostile to Hermias he quotes two passages of Theopompus, Περὶ [Φίλι]ππον (BNJ 115 F291: twenty-four lines of word-for-word quotation in the papyrus) and [ἐν τῇ πρ]ὸς Φί[λιππον ἐπισ]τολῇ (BNJ 115 F250: quotation of indeterminable size). He then passes to the opinions favourable to the tyrant: the first one, anonymous due to the lacuna,86 is followed by a quotation of Callisthenes (BNJ 124 F2: twenty-three lines) and by the ‘paean’ and funerary epigram composed by Aristotle in memory of the tyrant (PMG 842 and EG 622–26). Aristotle’s ‘paean’ is reported in its entirety (vi.20–22): κοὐκ ἂν [ἔ]χ[ο]ι φαύ|λως αὐτὸν ἀναγρά[ψαι δι]ὰ τὸ μὴ πολλοῖς | πρὸ χειρὸς (εἶναι), ‘it will not be superfluous to report it, since it is not easily accessible to the many’. At this point Didymus returns to adducing an item unfavourable to the tyrant, namely an epigram composed by Theocritus (of Chios) against Aristotle’s encomiastic portrait of Hermias and preserved in Bryon’s On Theocritus (SH 738). Finally, Didymus underlines the variety of historiographical versions on the crucial segment of the biography of the tyrant, namely his capture and death. The sources cited are Hermippus of Smyrna (Life of Aristotle: FGrHistCont 1026 F31), Theopompus (fragments cited above), and Callisthenes (BNJ 124 F3). It is not worth the effort, Didymus concludes, of citing the testimony of Anaximenes’ Περὶ Φ[ίλιππ]ον (BNJ 72 F9), ‘[because it is use]less’ (ο[ὐ γ(ὰρ) ὄφε]λος). The current consensus is that for his Περὶ Δημοσθένους Didymus conducted original investigations and cited the historical sources at first hand.87 In the example just cited, he expressly assumes responsibility (δ* [ο]κῶ μοι) for citing and arranging the sources. In this light, to remark on the disagreement of the testimonies relative to Hermias is equivalent to affirming the necessity of critical scrutiny and presenting a justification of the choice made. The same criterion corresponds to the careful preservation of the hard-to-find Aristotelian paean and, conversely, the condemnation of the testimony of Anaximenes. The unwillingness to accept contradiction among the sources and the acidic judgement on Anaximenes give a very personal colour to the conclusion of the digression, which seems to have cost Didymus no little time and effort. This instance definitively documents that the commentator, though presumably also benefiting from previous exegetical works, did apply his personal discernment and took into consideration interests and goals of his own and of his addressees.

3.3. Further Fragments St. Byz. θ 45 Billerbeck–Zubler reports the testimony of Δίδυμος ἐν β´ Συμποσιακῶν (°334) on the inhabitants of the Pontic city of Θιβαΐς: ‘their breath kills whomever it comes to, and their bodies, if they are thrown into the sea, do not sink to the bottom’. Didymus could have acquired the notice from Phylarchus (BNJ 81 F79a): ‘they are lethal not only for children but also for adults: for, if they come into contact with their gaze or breath or speech, they weaken and fall ill’. St. Byz. ο 28 Billerbeck reads: Οἰνοῦς· πολίχνιον Λακωνικῆς, ὡς Ἀνδροτίων (BNJ 324 F49) καὶ Δίδυμος (389). τὸ ἐθνικὸν Οἰνούντιος καὶ Οἰνουντιάς, ‘Oinus: a little town in Laconia, as Androtion and Didymus say. The name of the population is Oinountios and Oinuntias’. Its source could be a commentary of Didymus on a poetic work, in the light of Alcm. PMGF 92 = fr. 134 Calame, οἶνον δ᾽ Οἰνυντιάδα. Phot. κ 1280 Theodoridis and Suda κ 2804 Adler = 402ab reproduce diverse opinions concerning the golden colossus of the Cypselids at Olympia (Pl. Phdr. 236b). Agaclytus in his work On Olympia (BNJ 411 F1) asserted that the statue was located in a temple dedicated to Hera by the inhabitants of Scillus and that Cypselus had donated it after seizing power. Didymus attributed the initiative to Periander, who with this massive expense would have been intentionally weakening the Corinthians, his subjects; and he cited the opinion of Theophrastus (Περὶ καιρῶν, Book 2; fr. 609 FHS&G),88 who recognized the same strategy of 86 Hermippus, Theophrastus, Theopompus? See Bollansée 1999: 92 n. 89; Gibson 2002: 108–09; Harding 2006: 139–40. 87 Gibson 2002: 32–34, 69; Harding 2006: 36–39. For the opposite view: Diels and Schubart 1904; Lossau 1964. 88 This work could coincide with Πολιτικῶν πρὸς τοὺς καιρούς (fr. 589.4a FHS&G): Podlecki 1985: 238 and 243. Contra Rzepka 2011a.

Didymus and the Greek historians



77

social subjection in the wars constantly waged by the tyrant Dionysius and in the construction of the Egyptian pyramids.89 The comment closes by providing two slightly different versions of the votive epigram for the colossus, the second of which is documented by Apellas Ponticus (BNJ 266 F6). For Jacoby the entire excerpt goes back to Didymus.90 There is no way to confirm this assumption, but it is certain that Chalcenterus addressed the zētēma without avoiding its historical and political implications. 4. CONCLUSIONS: TOWARDS A REASSESSMENT Our scrutiny of the evidence prompts some conclusions. Only a slight hint supports the hypothesis of a hypomnēma to Herodotus by Didymus, though it would not be unexpected, given the Aristarchean precedent. If the Didymus cited in Marcellinus’ Life of Thucydides is the Chalcenterus, then we have proof of a study of this historian that dealt at least with biographical and antiquarian issues and drew on specialized sources (Pherecydes of Athens and Zopyrus, possibly also Hellanicus, Demetrius Phalereus, Philochorus, Polemon). In any case, the tradition of grammatikē exēgēsis to Thucydides that was known to Dionysius of Halicarnassus could hardly have bypassed an omnivorous scholar like Didymus. His antiquarian interest in the Solonian axones is documented by his syngramma on the topic and in a fragment of his exegesis to Demosthenes. All this is too little to gain a clear idea of Didymean scholarship on Greek historians and factual subjects, but it offers enough episodes to confirm the character of his approach as both philological and antiquarian, his propensity to take a position in relation to the sources and to insert himself into traditional debates, and the use of the whole gamut of tools or ‘genres’ of erudite learning (lexicography, commentaries, monographic writings, biography) with the transfer and refocusing of ideas and materials from one to the other. Quite paradoxically, the relation of Didymus to history and the historiographical tradition emerges in a richer way in other fields of his work. Several fragments that derive from his lexicographical and exegetical or critical writings attest the use of works of genealogy and mythography (Acusilaus, Hellanicus?, Pherecydes), general and universal history (Herodotus, Thucydides?, Philistus, Ephorus, Theopompus, Timaeus), histories of Macedonia and of Alexander (Anaximenes, Theopompus again, Marsyas, Duris, Callisthenes), local history (Proxenus, Theotimus of Cyrene, Pythaenetus of Aegina) and especially Atthidography (Androtion, Phanodemus, Philochorus, Ister, the exēgētēs Autoclides), biography (Hermippus), and chronography and antiquarian enquiries, including collections of inscriptional records (Theophrastus, Craterus, Polemon, Apollodorus, a list of Nemean victories). It is necessary to underline the very ragged and disparate nature of this documentation as well as its extreme imbalance. The surviving exegesis to Pindar and that to Demosthenes display very different characteristics from each other, but such as to permit us to appreciate a large number of Didymean applications of historiographic and antiquarian sources. The scholia to Homer or Aristophanes do not make this possible to the same degree. For the VMK scholia to Homer, this is reasonably explained by the text-critical character of Didymus’ work that was used as a source. The commentators on Aristophanes’ plays prove to have been strongly focused on Attic-centered historians, but Didymus’ contribution is scarcely mentioned and likely remains widely hidden below the surface. Similarly, the scholia to Aeschylus and Sophocles, which allegedly owe a great deal to Didymus, in their present state do not preserve any explicit evidence of his use of historians. We may reasonably ascribe this scarcity, at least partly, to the choices made by compilers and copyists of the scholiastic corpora, our main channels of tradition in this respect91—and so on, each field demanding an individual evaluation, without allowing us hasty or generalizing judgements.92 Therefore, in order to try to capture the characteristics of Didymus’ approach to history at its best we must look above all to the direct tradition (On Demosthenes) and the more generous sectors of the indirect tradition (Pindar scholia). There we 89 90 91 92

Cf. Arist. Pol. 1313b18–25. See his comments in FGrHist; Anderson 2008; Rzepka 2011a. See e.g. Montanari 2016. Cf. Matijašić 2018: 153–54.

78



Montana

recognize a rich library of historiographical sources and the personal insight (whatever its ultimate value) which the scholar applied to identifying and illustrating facts, usages, and persons mentioned in the literary works; to establishing connections in chronology and context; and to comparing discordant reconstructions and taking a position on them. This perspective can function as an important factor in our evaluation of Didymus’ remaining scholarship on and through historians, and can help us look at this intellectual experience in a more cautious and respectful way, within the broader framework of ancient philology.93

93 Overviews of opposing modern opinions on Didymus: Gibson 2002: 51–69; Harding 2006: 31–39; Braswell 2017: 123–26; and the Introduction to this volume.

Bulletin of the Institute of Classical Studies, 2020, 63, 79–94 doi: 10.1093/bics/qbaa018 Original article

Chapter Six The compiler compiled: Didymus in Imperial scholarly and miscellanistic literature* Scott J. DiGiulio Mississippi State University, Starkville, Mississippi

ABSTRACT Few ancient scholars were as prolific as Didymus of Alexandria, who was hailed by some in antiquity as the greatest of the grammarians. Yet, despite his polymathic output and seemingly positive ancient reputation, Didymus was much maligned for his carelessness and the compilatory nature of his work, attitudes which have continued in modern scholarship. This chapter aims to reassess the earliest period of Didymus’ reception by looking closely at the scholarly and miscellaneous texts of the Roman Empire that cite and discuss the Alexandrian. By examining four particular points of reception—Roman educationalists, Harpocration, Athenaeus, and Macrobius—this chapter illustrates that while Roman authors were particularly interested in using a caricature of Didymus as a straw man for their own arguments, Greek figures more readily engaged with Didymus and his work per se. These two traditions evolve in parallel and use Didymus to represent both the positive and negative facets of polymathy; each freely cites, and transforms, him to suit their literary purposes. As a result, Didymus’ reception throughout antiquity is considerably more complex than has been previously acknowledged: he is a figure of both authority and consternation, even among the authors most similar to himself.

1. INTRODUCTION For all his learning and voluminous output, Didymus’ modern reputation has suffered considerably, with any charity towards an author of scholarly demeanour set aside to focus upon his failings. Consistently attacked for carelessness, or the derivative nature of his compiled work, or his failure to focus on questions of interest to modern scholars, Didymus has largely served as an object of scorn.1 At least in part, these modern reactions are conditioned by the ancient testimonia; yet, a critical examination of Didymus’ reception at Rome, and particularly among the authors of ancient miscellanies and other scholarly writers that cite him directly, * Numbers in bold refer to testimonia and fragments in the Checklist (pp. 95–120). 1 Thus Pfeiffer 1968: 274 suggests that ‘his reputation has never been very high’. Accusations of negligence are not uncommon, suggesting that Didymus was careless in compiling his various sources into a coherent commentary (West 1970), that he provided incorrect cross references within his own works (Bliquez 1972), or that he fundamentally misinterpreted the historical details discussed in his commentaries (Harris 1989). However, many of these critiques have their origins in Didymus’ ancient reputation, which was polemical and based less on his own works and more on his status as one of the last great Alexandrian grammarians. For a survey, see Gibson 2002: 51–69. C The Author(s) 2021. Published by Oxford University Press on behalf of the Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected].



79

80



DiGiulio

suggests a more complicated attitude than his modern reputation might indicate. To that end, in this chapter I survey four distinct responses to Didymus in the Roman Empire, examining his reception among Roman writers on education, in the lexical work of Harpocration, Athenaeus’ Deipnosophistae, and Macrobius’ Saturnalia.2 Each offers a distinct view of Didymus and his work, revealing a more complex understanding of the grammarian and his work in the ancient world. What emerges out of this survey is the existence of two images of Didymus that roughly correlate to his treatment in each linguistic tradition. Among the Greek authors of the Imperial period, Didymus is freely quoted, cited, and discussed as much for the information he provides as he is for the authority which he carries. In these contexts, he represents a legitimate source to be studied and scrutinized: far from the consternation with which he is often treated by modern scholars, Didymus often served as a primary, if not absolute, authority against whose learning others might be measured. In contrast, the Latin sources have a fundamentally different understanding of Didymus, more in line with modern critical opinion, that does not seem to be rooted primarily in approaching his work. Rather, these Romans construct an essential idea of Didymus—a grammaticus par excellence—through which character they might assail the perceived faults of Greek learning. In examining his Imperial reception, it is evident that these two traditions represent competing visions of Didymus, both in his own right and as a representative example of the broader tradition of Hellenistic polymathic scholarship. 2. ROMAN IMPERIAL EDUCATION: DIDYMUS AMONG THE GRAMMATICI What little information for which there is any certainty about Didymus’ reputation at Rome derives from the Suda, and precious little is known of his reception in the city prior to the Neronian period.3 To be sure, in the entry for Didymus (T1) almost nothing is said of the scholar’s reputation at Rome, or whether he had ever seen the city itself.4 Glimpses of his reputation survive in entries dedicated to others. Heraclides Ponticus the Younger, who had been a student of Didymus and later served as head of a school at Rome under Claudius and Nero, undertook a defence of his master upon hearing of Didymus’ poor reception and moved to the city (T4 = Suda η 463 Adler):

οὗτος ἐπειδὴ ἤκουσεν Ἄπερως τοῦ Ἀριστάρχου μαθητοῦ εὐδοκιμοῦντος κατὰ τὴν Ῥώμην, πολλά τε τοὺς Δίδυμον διασύροντας, ἔγραψε μέτρῳ Σαπφικῷ ἤτοι Φαλαικίῳ βιβλία γ, δυσερμήνευτα καὶ πολλὴν τὴν ἀπορίαν ἔχοντα προβαλλομένων ζητημάτων, ἅτινα Λέσχας ἐκάλεσεν. εἰς Ῥώμην δὲ κομίσας καὶ τοῦ Ἄπερως καταφανεὶς κατέμεινε σχολαρχῶν ἐν αὐτῇ ἐπὶ Κλαυδίου καὶ Νέρωνος. When he [sc. Heraclides Ponticus] heard that Aper, a student of Aristarchus, was thought of highly in Rome, and that they [i.e. those in Rome] greatly ridiculed Didymus, he wrote three books in Sapphic or Phalaecian metre, difficult to understand and presenting much difficulty in the questions posed, which he titled Leschai. After he brought them to Rome and denounced Aper, he stayed there as the head of a school in the reigns of Claudius and Nero.5 2 These four examples provide the most extensive references to Didymus in a primarily literary milieu throughout the Imperial period. He is additionally cited briefly at Plu. Sol. 1 (°361), Hermog. Id. 2.11 (398), Clem. Al. Protr. 2.24 (399), and Amm. Marc. 22.16 (406 = T7); in each case, he is invoked as an authority on a minor point of interest, and not extensively utilized by the authors that cite him. 3 For a fuller survey of the material from the Suda and the testimony provided for Didymus’ life, see Braswell 2017: 27–36. 4 Based on the extant sources, there is little indication that Didymus ever taught at Rome, though the claim was common in earlier scholarship. The origin of this claim appears to have been Schmidt’s (1854: 335–36) adventurous emendation of the Suda to conflate the entries for Didymus the Younger, who did in fact work at Rome, and Didymus of Alexandria. 5 The grammarian Aper mentioned here is otherwise unknown, and the reading Ἄπερος may be an error, either for Ἀπίωνος (which would align with Apion’s known period of activity and acclaim at Rome), or else Ἀντέρωτος (which would indicate Apollonius Anteros, mentioned as a contemporary of Heraclides elsewhere in the Suda); on the identification of this figure, and the attendant textual issues, see Braswell 2017: 34–35; Matthaios 2015: 223–24.

The compiler compiled



81

The anecdote is suggestive both of Didymus’ place in discussions of scholarly literature at Rome, and of the influence of his work. On the one hand, his prominence placed him in the crosshairs of rival scholars (much like his contemporary, Aristonicus) and made him the target of polemic.6 On the other hand, the Suda’s brief biographical entries note the persistent influence of Didymus’ own pupils—while far from the only grammatici based in Rome, his students flocked to the capital, promulgating his teachings, defending his work, and establishing schools of their own. Despite this prominence, when Didymus begins to be discussed directly by the Romans, he is primarily considered an industrious, though unoriginal, compiler first, and a philological thinker second. This attitude is particularly clear within the context of Roman education. For all his care in digesting the Alexandrian philological tradition into his own commentaries and other exegetical works, and his importance for the broader tradition of Alexandrian scholarship, it is perhaps a surprise that Didymus received such a relatively cool reception in Rome. Authors within the Latin tradition call particular attention to the volume of his output, casting him as negligent and otherwise thoughtless in assembling his works. Seneca the Younger, in the earliest extant discussion of Didymus and his work, attacks him for the sheer volume of his output, which Seneca determines to lack utility for leading a good life (Sen. Ep. 88.37 = T14 = F394–97): Quod ista liberalium artium consectatio molestos, verbosos, intempestivos, sibi placentes facit et ideo non discentes necessaria quia supervacua didicerunt. Quattuor milia librorum Didymus grammaticus scripsit; misererer si tam multa supervacua legisset. In his libris de patria Homeri quaeritur, in his de Aeneae matre vera, in his libidinosior Anacreon an ebriosior vixerit, in his an Sappho publica fuerit, et alia quae erant dediscenda si scires. Because the excessive pursuit of the liberal arts makes men boorish, verbose, rash, and self-satisfied, not learning the essentials since they have learned superfluous points. Didymus the grammarian wrote four thousand books; I would pity him, if he had only read so many unnecessary things. In these books he investigates Homer’s birthplace, Aeneas’ true mother, whether Anacreon was more lustful or prone to drinking, whether Sappho was a woman of ill repute, and other things that you ought to forget, if you were to know them. Notably, Seneca does not appear to be criticizing Didymus on the basis of any specific text, nor does his response appear to be based on any first-hand knowledge of his works. Rather, Seneca’s objection is predicated upon a conventional caricature of Didymus: he attacks the scholar for his (admittedly) extensive output, criticizing the superfluous knowledge that he had gathered and accumulated into his vast oeuvre. Within the broader context of Seneca’s critique of education and the studia liberalia, Didymus stands as an exemplum of the sorts of inane, prototypically Greek learning that abounded in the miscellanies of the Imperial era.7 Seneca singles out Didymus as a paragon of indulging in trivia, and a representative of the vice of intemperantia (Ep. 88.36), a lack of restraint that borders on πολυπραγμοσύνη.8 In this respect, he serves as a stand-in for grammatici generally, and his representation here is comparable to other prominent grammarians like Apion (Ep. 88.40) and Aristarchus (Ep. 88.39), who come under scrutiny in Seneca for their focus on minutiae.9 Like the others mentioned here, Didymus’ learning is deemed superficial, too concerned with volume and breadth that detracts from true wisdom. Didymus, along with Apion and Aristarchus, is representative of 6 Indeed, attacks on Didymus and his grammatical work may have persisted until at least the second century, as it is noted in the Suda (τ 895 Adler) that Suetonius wrote a critical response to a work on Cicero’s De re publica, assigned to Didymus by Ammianus Marcellinus (22.16.16 = 406). The authenticity of this work is, however, disputed; see Braswell 2017: 100. 7 Compare the attacks of Gellius on the Greek miscellanies from which he attempts to distinguish himself (NA praef. 11), or the criticism of his anonymous friend’s miscellany for his approach in collecting (NA 14.6). 8 On this latter quality in Seneca, see Leigh 2013: 175–77. 9 Lucian similarly holds up Aristarchus as a pedant par excellence (VH 2.20); cf. Montana 2015b: 64–66.

82



DiGiulio

their broad fields of enquiry for Seneca; as such, they serve as useful straw men for him to tear down as he distances systematic study of the texts of Homer qua texts from his ideal liberal curriculum.10 Like Seneca, Quintilian finds the volume of Didymus’ output and his injudiciousness in assembling his works remarkable; in articulating ideal reading practices for young men, in which he argues that they should avoid tracking down every possible text that they can find, he expressly chastises Didymus for this approach (Inst. 1.8.18–20 = T6): Persequi quidem quid quis umquam vel contemptissimorum hominum dixerit aut nimiae miseriae aut inanis iactantiae est, et detinet atque obruit ingenia melius aliis vacatura. Nam qui omnis etiam indignas lectione scidas excutit, anilibus quoque fabulis accommodare operam potest: atqui pleni sunt eius modi impedimentis grammaticorum commentarii, vix ipsis qui composuerunt satis noti. Nam Didymo, quo nemo plura scripsit, accidisse compertum est ut, cum historiae cuidam tamquam vanae repugnaret, ipsius proferretur liber qui eam continebat. To hunt down what even any of the most despised writers said is either excessively pitiful or empty boasting, and it detracts from and ruins talents better left open for other things. Anyone who goes through every scrap of paper, even the things unworthy of reading, can also spend his energy on old wives’ tales. And the commentaries of the grammatici are full of this sort of baggage, barely enough known by those that compiled them. For it is said that it happened to Didymus, than whom no one wrote more, that when he refuted a certain historical account as being false, one of his own books was brought forward which contained that account. Quintilian recognizes Didymus’ polymathy, but he couches it in particularly negative terms; Didymus is only invoked as an example of pedantry (nimiae miseriae aut inanis iactantiae) and a futile desire to assemble a vast array of knowledge.11 Indeed, such thoughtless compilation leads to the sorts of errors, and anecdotes, mentioned by Quintilian, as Didymus assails a narrative while forgetting that he himself wrote it. There is little benefit to the breadth of Didymus’ learning, Quintilian suggests, if he forgets his own works; furthermore, his credulousness and lack of discernment in compiling lead to self-contradiction, as the anecdote illustrates.12 It is important to note here that Quintilian’s anecdote, and to a lesser extent Seneca’s, is not concerned with the substantive content of Didymus’ scholarly commentaries, and offers little insight into the Imperial reception of his works broadly speaking.13 Rather, both of these caricatures of the scholar and his work situate Didymus firmly within a tradition of Imperial authors writing about education and particularly the utility of grammatical learning.14 Quintilian adduces Didymus as an example in the context of structuring the 10 Braswell 2017: 38. 11 In contrast, the Stoic polymath Posidonius appears to receive little to no criticism for his extensive learning and participation in a range of fields; the extant Roman sources consider him a philosopher of the first rank (e.g. Cic. Tusc. 2.61, Fin. 1.6; Strabo 16.2.10; Sen. Ep. 90.20, 108.36–38). The difference between Didymus’ and Posidonius’ reputations may derive from their statuses as grammarian and philosopher respectively, with the latter more respected among the Romans, to say nothing of the fact that one of the chief sources for Posidonius’ reception at Rome, Cicero, had studied under Posidonius and would thus have incentive to paint a flattering portrait. 12 To be sure, in singling out the sheer volume of work that Didymus produced, and his tendency to forget what he had written, Seneca and Quintilian may not necessarily turn their audiences against the grammarian. However, the framing of their criticism, which centres on needless polymathy and its detrimental effects for the education of the young Roman, suggests that such forgetfulness is to be avoided. Rather, one should focus deeper attention on a narrower range of topics and texts. By contrast, in the Greek tradition the volume of Didymus’ output and his accompanying slips of memory are mentioned but are less a cause for criticism than his practice as a polymath; see further below, on Didymus in Athenaeus. 13 Gibson 2002: 55–59, who goes on to note that these particular testimonia are frequently adduced in discussions of Didymus’ commentaries, albeit probably wrongly, as neither passage suggests any familiarity with Didymus’ works beyond the most general knowledge. 14 Gibson 2002: 56–58.

The compiler compiled



83

reading of young men as they progress to the upper phases of their education, emphasizing in particular the futility of pursuing narrow questions of the sort that abound in the works of the grammatici: his concern is not with helping a potential student interpret or appreciate Didymus per se, but cautioning them from indulging in his stereotyped habits of mind. Indeed, Quintilian does not appear to expect his budding orator to encounter Didymus, except as a conventional exemplum, as he deploys him in the Institutio Oratoria. Likewise, for Seneca Didymus is a cautionary example, representative of the acquisitiveness and excess that characterizes the undisciplined habits of mind possessed by the grammatically inclined. He assails grammatici for not offering a path to virtue (Ep. 88.3), and bemoans that even the philosophers, in adopting the grammarians’ fastidiousness for language, have likewise lost sight of the goal of bettering individuals (Ep. 88.42–43). These attitudes are common throughout Seneca’s corpus, setting up the grammatici and their teachings as impediments to true knowledge. While he calls them guardians of the Latin language (Ep. 95.65), this title is not without significant irony, as Seneca complains frequently that grammatici do little other than waste their time on pedantic questions of linguistic curiosity.15 Explicit within these critiques is a strong degree of cultural chauvinism, with the methods of those like Didymus held up as fundamentally Greek, and thus un-Roman. For Seneca, the intellectual habits of the grammatici, and their fixation on useless detail, are a disease that then comes to infect the Romans (Graecorum iste morbus, Dial. 10.13.2). To be sure, he attacks those Romans that now pursue similar enquiries, but it is because they have limited utility; while those things that touch on elements of the Roman state appear interesting, they nevertheless fail to provide any substantive benefit (10.13.3), either to the individual or the state. This perspective is echoed later by Aulus Gellius, who shares Seneca’s hostility to grammatici; in his case, the polemic centres on his construction of a distinctly Roman elite identity, which has little use for the impracticalities of trite Greek learning.16 While the ars grammatica certainly has utility, Gellius shows great disdain for slight detail (he condemns the miscellany of a certain friend for the inclusion of such detail, for instance, at 14.6), and even goes so far to attack eminent thinkers for focusing too narrowly on semantics (e.g. 2.9, where Plutarch is attacked for ‘word-hunting’, λεξιθηρεῖν). He assigns the pedantic features of the grammatici to Greek influence, and he excludes these elements from his own self-presentation as an idealized Roman. At least in part, the hostility to Didymus in the Roman sources emerges from this set of biases. Didymus typifies a particular strand of Greek scholarship deemed trivial by Seneca and Quintilian, who argue that grammatical study of the sort typified by Didymus distracts from the true purpose of education, and it is these opinions that in turn have conditioned Didymus’ modern critical reception. He represents a tradition of compilation of other sources, collecting the opinions of his predecessors and valuable solely for that reason. To be sure, Seneca, Quintilian, and the like say very little about the actual work of Didymus per se: he is instead synonymous with grammatical enquiry, and both deploy him as a stand-in for the grammarian’s enterprise.17 While the Roman criticism of Didymus offers a portrait of a prolific author, who was nevertheless negligent with his own words on occasion, the explicit critiques speak little to the accuracy of his work, his perceived intelligence, or his reputation as a scholar. In fact, there is little in these criticisms to suggest that Didymus’ work would not be valuable to the well-trained student; what require care, especially for the student, are the grammaticus and his methods, and Didymus serves as a straw grammaticus for both Seneca and Quintilian, in part owing to the absence of any specific textual references. While Roman educationalists frequently talk about Didymus, we can say little about the role they envision for Didymus beyond acting as cipher for a particular kind of scholarship; their own engagement with him is 15 Kaster 1988: 53. 16 At least in part, the attack on grammatici in Gellius is predicated upon their professional status, which aligns them with Greek sophists rather than the Roman upper classes, who in the NA pursue such questions and topics only in their negotium; see Keulen 2009: 28–32. 17 Compare Gibson 2002: 58: ‘Didymus is the perfect foil for their criticism because he was the best-known (and perhaps the best) practitioner of the discipline.’

84



DiGiulio

limited to exemplary anecdotes, and there seems to be no expectation that the average Roman student would have reason to engage with Didymus directly. Indeed, even in compilatory works such as Gellius’ Noctes Atticae, in which one might reasonably expect the works of Didymus to find a ready home, Latin authors of the early Empire are conspicuously quiet. However, several Greek authors provide a window onto how Didymus’ actual works could be integrated into their intellectual projects. Such examples offer a more fruitful path for investigating the reception of Didymus’ work beyond the grammatical tradition. In the remainder of this chapter, I examine citations of Didymus in the compilatory literature of the later Imperial period, focusing on the manner in which his works have been redigested. Didymus provided natural fodder for miscellanists and other scholarly authors in the second to fourth centuries, with his treatment of an extensive array of texts and genres. For all their notional kinship, the miscellanists’ use of Didymus wrestles with two images of the scholar, as an authoritative Hellenistic critic and a grammarian who fails to see the forest for the trees. 3. DIDYMUS AND/IN HARPOCRATION Before turning to Athenaeus, I wish to look briefly at the fragments of Didymus preserved in Harpocration’s Lexeis of the Ten Orators, which may serve as a control example for the ways in which Didymus’ works were cited in the Imperial period.18 The lexicon, consisting of some 1300 largely alphabetized glosses on words or phrases from the canonical Attic orators in its full form, likely dates to the end of the second century, and was epitomized at some point prior to 850 CE. In the lexicon Harpocration cites Didymus by name thirty-five times; across these entries, he names at least five distinct works, with many of the fragments so closely tied to passages from Demosthenes that it is unlikely that they derive from anything other than Didymus’ commentary on the orator.19 Across these references, Harpocration draws on Didymus’ corpus, particularly his earlier commentaries on the Attic orators, as a source for his explanations of difficult vocabulary and expressions in those texts. In the sample of fragments directly related to Didymus and his works in the Lexeis, it is apparent that Didymus’ word carried a good deal of authority, and Harpocration is judicious in his use of these citations. He cites Didymus at length on a range of different issues. Unlike the Roman educationalists, who were interested in Didymus qua grammaticus primarily as a convenient shorthand, Harpocration uses Didymus directly and engages with his works seriously. Harpocration largely works within the same scholarly tradition as Didymus, and thus he is more concerned with the content of Didymus’ thought rather than his conventional character. Broadly speaking, Harpocration refers to his Didymean source material for both linguistic and historical questions; while he does on occasion quote extensively from Didymus himself, he is largely content to reproduce Didymus’ arguments in paraphrase. Often, the citations are quite neutral and relatively brief, as the gloss on κυμβίον (κ 92 Keaney = 291) illustrates: Δημοσθένης ἐν τῷ Κατὰ Μειδίου. εἶδός τι ἐκπώματος τὸ κυμβίον. φησὶ δὲ Δίδυμος ἐπίμηκες αὐτὸ εἶναι καὶ στενὸν καὶ τῷ σχήματι παρόμοιον τῷ πλοίῳ ὃ καλεῖται κυμβίον (‘Demosthenes, in his speech Against Meidias. A κυμβίον is a kind of drinking cup; but Didymus says that it is oblong and narrow, and in shape it closely resembles the ship that is called κυμβίον’). Quotations at length are an exception; the most notable instance is the entry for ὁ κάτωθεν νόμος (ο 14 Keaney = 292b), in which Harpocration devotes almost the entire entry to Didymus’ own words.20 In the discussion which is relayed, Didymus provides a historical explanation of Demosthenes’ expression, drawing on several other sources including Euphorion and Anaximenes. The extensive reference to other sources which he had compiled in his research lends authority to Didymus’ words; their inclusion vouches for both Didymus’ explanation and the 18 The most recent, albeit problematic, edition is Keaney 1991. On the Lexeis generally, see Keaney 1973; Whitehead 1997; on the dating, see Hemmerdinger 1959. 19 On the Didymus fragments in Harpocration, including edition and commentary, see Gibson 2002: 137–56. 20 For the entry and its parallels with P.Berol. inv. 5008 = 292a, see Gibson 1997. Didymus is clearly quoted, rather than paraphrased, at Harp. ε 94 (298a), ε 178 (285), ο 38 (296a), and π 99 Keaney (316a); in each of these cases, the passages cited offer Didymus’ own explanation of a peculiar term, or a relatively concise treatment of a historical question as it relates to a term.

The compiler compiled



85

authority of Harpocration’s glosses. Similarly, the gloss for πρόπεμπτα (π 99 Keaney = 316a), in which Harpocration notes the works of Lysias and Demosthenes in question before quoting Didymus’ discussion that provides further textual parallels from both corpora. The composition of the entry, in which Harpocration foregrounds Didymus’ analysis, leans heavily on the parallels that the latter found in his research. It is the depth of this research here that leads Harpocration to reproduce the discussion without additional commentary. While Didymus’ ultimate conclusion is aporetic (he offers solutions based on both Lysias and Demosthenes without asserting one or the other as being correct), the discussion is extensive enough to warrant no further comment. For, as authoritative as he appears to be, Didymus is not immune to criticism by Harpocration, who assails his predecessor’s opinions several times in the Lexeis; in fact, some of these challenges share some parallels with the traditional deprecations of Didymus among modern scholars. While his critiques can be based on difference of interpretation (for instance, ε 66 Keaney = 301a: μήποτε ἀντὶ τοῦ ξένος ἦν, καὶ οὐχ ὡς Δίδυμός φησιν, ἀντὶ τοῦ οὐκ Ἀττικῶς διελέγετο ἀλλὰ ξενικῶς; ‘perhaps this was said in place of “he was a foreigner”, and not, as Didymus claims, instead of “he did not speak Attic, but foreignly”’), in several cases the objections are more substantive. A clear example of the criteria applied to Didymus’ work appears under the lemma φαρμακός (φ 5 Keaney = 297a), in which Harpocration reports that Didymus accented the word with a circumflex on the penult: Δίδυμος προπερισπᾶν ἀξιοῖ τοὔνομα. ἀλλ’ ἡμεῖς οὐχ εὕρομεν οὕτω που τὴν χρῆσιν (‘Didymus thinks that the word should have a circumflex on the penult; but we have not found this usage anywhere’). Harpocration’s objection rests on the lack of parallel for Didymus’ claim, which seemingly provides a key criterion for judging Didymus’ contributions.21 Similarly damning is Harpocration’s claim in his discussion of ἔνθρυπτα (ε 55 Keaney = 287), in which he argues that Didymus’ explanation lacks parallel: Δίδυμος ὁ γραμματικὸς ἐν τῷ ὑπομνήματι τοῦ λόγου εἰπὼν ὡς τὰ ἔνθρυπτα ἐκκείμενον καὶ γνώριμον ἀπ’ αὐτῆς ἔχει τῆς φωνῆς τὸ σημαινόμενον, περιεργότερόν τινα ἐκτίθεται ἐξήγησιν ἀμάρτυρον (‘When Didymus the grammarian, in his commentary on the speech, says that “crumb cakes” possess a meaning that is both recognizable and evident from the very sound of the word, he sets out an explanation that is overwrought and unparalleled’). Even more, Harpocration sees as overwrought Didymus’ explanation that the phonetic sound of ἔνθρυπτα is suggestive of its meaning,22 and his description of this gloss as περιεργότερoν suggests the needless labour involved in producing this definition.23 In trying to articulate an explanation, Harpocration suggests, he allows himself to indulge in one of the great flaws of the grammaticus, producing an answer with relatively little grounding. All of these faults come together in the lemma for γαμηλία, in which Harpocration cites Didymus twice, only to highlight the grave failings of each reference. To explain the precise nature of a wedding feast, Harpocration first adduces Didymus in Isaeum (°313a), only to note that a source that Didymus himself cited did not contain anything related to the question at hand (παρατιθέμενος λέξιν Φανοδήμου, ἐν ᾗ οὐδὲν τοιοῦτον γέγραπται, ‘he offers a saying of Phanodemus, in which nothing of the sort has been written’). Further, he provides a second instance in which Didymus offers the same explanation (°302), only to note that no parallel or support is provided (οὐδεμίαν ἀπόδειξιν τῆς ἐξηγήσεως παραθέμενος, ‘but he provides no evidence for his interpretation’). In each of these cases, Didymus is reprehended for producing information without adequate parallels, or else being mistaken that a particular source supports his claims. Didymus is far 21 Photius φ 64 Theodoridis, cited by Schmidt 1854: 314 in his discussion of the fragment and attributed by Erbse to the Hadrianic grammarian Aelius Dionysius, suggests that the accentuation is Ionic, with Naber suggesting five parallels in Hipponax; see also Gibson 2002: 151– 52. The immediately preceding gloss (φ 63 Theodoridis) is a nearly verbatim quotation of Harp. φ 5 Keaney. 22 Cf. Gibson 2002: 148, who notes that while ancient grammarians on occasion demonstrated interest in the connections between meaning and sound, Harpocration’s citation of Didymus here represents the only instance of such an argument being put forward in any of the ancient commentaries on Demosthenes. 23 Περιεργία, which is closely linked with πολυπραγμοσύνη in the critical tradition, indicates a broad range of stylistic or intellectual vices, including superfluous labour and a desire for useless learning; see Leigh 2013: 161–94. By alluding to Didymus’ περιεργία here, Harpocration touches on one of the principal vices assigned to Didymus, namely his boundless appetite for compiling his sources.

86



DiGiulio

from unimpeachable in Harpocration’s view, and this criticism is based on the merits of the argument: Harpocration identifies two places where Didymus had faltered in his own analysis and thus rebukes his predecessors for his methodological shortcomings here. His challenge is typical of scholarly polemical modes of contesting authority: by highlighting the errors of Didymus, in particular by returning to Didymus’ sources, Harpocration presents his superior knowledge not only of the orators but of the Greek literary tradition more broadly. For all of the authority that it possessed, Harpocration’s Lexeis demonstrates that a degree of caution was to be taken when working with Didymus’ commentaries and other scholarly works. While authoritative, especially when the research underlying his works was thorough, Didymus nevertheless could be prone to misinterpretation and careless treatment of his sources.24 Indeed, in several cases the arguments put forward by Didymus, and in turn critiqued by Harpocration, lack supplementary evidence from his own compiled authorities, particularly in instances where it appears that Didymus is in fact incorrect.25 While Didymus’ commentaries and lexicographical work were valuable for his endeavours, nevertheless Harpocration maintained a healthy dose of scepticism, understanding that Didymus was far from infallible, and at times prone to the sorts of slips frequently assigned to grammatici. Even if he attests to the breadth and quality of Didymus’ ideas, he is not neutral in his citations. Importantly, he takes Didymus and his corpus seriously even as he pushes back on its content and approach. In challenging the great Alexandrian scholar, Harpocration positions himself as a linguistic authority in his own right, versed in the language of the Attic orators and better able to explain it than one of the most learned of the Alexandrian scholars. 4. DINING WITH DIDYMUS: ATHENAEUS’ DEIPNOSOPHISTAE Throughout the Lexeis of the Ten Orators, Harpocration engages with Didymus as a source, rather than as a clichéd exemplum; in some respects, this is more typical of Didymus’ appearances in the Imperial Greek tradition. However, Athenaeus and his treatment of Didymus’ intellectual reputation provides an intriguing test case for the limits of this reception. In the Deipnosophistae, Athenaeus produced a work which, like Didymus’ scholarly output, is predominantly fashioned out of compilation from, and quotation of, other works, albeit with a fundamentally different purpose. As with many authors of miscellanistic and compilatory texts from the Imperial period, the principal scholarly interest in Athenaeus has been in his utility as a mine of these fragments, rather than focusing on the text in its own right.26 However, as Athenaeus interweaves these fragments throughout his dinner-party frame, the banquet of the Deipnosophistae participates in a tradition of sympotic literature, which frequently challenges the audience to participate in, and meditate upon, the reading habits of his characters and their treatment of the texts which they read.27 As traditional approaches to the Deipnosophistae were keenly interested in the ample quotations throughout the text, this last point has

24 West 1970: 295–96 offers several examples in the papyrus of the commentary on Demosthenes (P.Berol. inv. 9780 = °281) where Didymus played fast and loose with his sources, including his consultation of just enough of Philochorus to provide an answer without fully grasping the historical context (col. vii.19 ff.), and his lack of concern for the authenticity of Demosthenes 11 (col. xi.7 ff.); on the latter, see also Gibson 2002: 125–26. 25 Cf. West 1970: 293–94, on Didymus’ omission of evidence in at least one instance where he misinterpreted the historical facts underlying his own views. 26 Recent years have seen a rehabilitation of the Deipnosophistae in particular, with serious attempts to understand the sophistication of the work’s construction, and its relationship to other symposiastic literature, and to the Greek culture of the Roman Empire more broadly; see, among others, Braund and Wilkins 2000; Paulas 2012; König 2012: 90–120; Jacob 2013. Continued further study of Athenaeus will be greatly benefited by the new critical edition of the text by Olson (the first volumes of which began to appear in 2019), his re-edition of the Loeb Classical Library text (Olson 2006–12), and the tools provided by the Digital Athenaeus Project. 27 Proper critical reading practice is a central concern of numerous Imperial miscellanies, whether those texts employ the sympotic frame or not. On the importance of the critical reading tradition for the Imperial miscellanists, and how these practices provide unity to such texts, see König 2007: 50–62; Howley 2018: 66–111. On reading Athenaeus in such a way, see Paulas 2012: 406–10.

The compiler compiled



87

been centrally important as scholars ask to what extent the quotations and paraphrases are true to the texts from which they originate.28 Because of the fragmentary nature of his work, it is exceedingly difficult to determine the accuracy of the quotations from Didymus with real certainty; nevertheless, Athenaeus’ compilation of material from the grammarian offers a window onto his use of the latter as a source. In approach he stands between Harpocration, for whom Didymus is a source deserving serious scholarly attention, and the Romans that were concerned with an imagined Didymus that they could target as emblematic of the grammatici. Within the Deipnosophistae, Athenaeus looks to him instead as representative of the tradition of Greek learning, and deploys citations of Didymus throughout the work to develop an implicit criticism of his contemporary intellectual culture. Perhaps the best-known testimony for Didymus in the Deipnosophistae appears to fit well within the critical tradition discussed above, as Athenaeus, citing Demetrius of Troezen, labels Didymus βιβλιολάθας, the ‘book-forgetter’ (4.139c = T5), because of the sheer mass of works he produced. At first glance, this epithet and its stated origin conjure up the criticisms of Seneca and (especially) Quintilian, with their disdain for the volume of Didymus’ output: Athenaeus is clearly aware of the Roman caricature of the Greek polymath as mindless compiler. However, Athenaeus’ diverse use of Didymus reveals less criticism in this claim than the similarity to the Roman authors’ attacks might initially suggest. Throughout the Deipnosophistae, Didymus is cited by name eighteen times;29 of those eighteen citations, seven cluster in the discussion of different kinds of drinking cups in Book 11. The majority of these references draw on Didymus in paraphrase to support the various interlocutors’ claims about diction and other points of usage, invoking his authority on the proper terminology for tree-ripened olives (2.56d = 11) or the etymology of Pramnian wine (1.30e = 27a). These brief citations treat Didymus as authoritative, and Athenaeus’ banqueters rarely disagree with him openly when they introduce him into their discussions. Even in the cases in which Athenaeus has his characters disagree, their critiques are relatively mild, suggesting that other sources present a different opinion (as in the case of Pramnian wine, 1.30d–e). What serves as the strongest rebuke of Didymus in the Deipnosophistae is still a gentle correction, as Athenaeus’ character Plutarch notes that Didymus confuses a patronymic with an ethnic in explaining a passage of Herodotus (ἀγνοεῖ δ᾿ ὁ γραμματικὸς ὅτι τὸν τοιοῦτον σχηματισμὸν ἀπὸ κυρίων ὀνομάτων οὐκ ἄν τις εὕροι γινόμενον, ἀλλ᾿ ἀπὸ πόλεων ἢ ἐθνῶν, ‘the grammarian is ignorant of the fact that one would not find such a formation made from personal names, but from the names of cities or peoples’, 11.486d = 299a). Plutarch mentions the mistake, but with the requisite correction made he does not belabour the point, with his own authoritative position confirmed and his knowledge, which surpassed Didymus’ at least on this point, demonstrated. For the reader to recognize Plutarch’s expertise as he corrects the grammarian, Didymus must remain authoritative in the eyes of the audience, and Athenaeus depicts him as such. Athenaeus’ characters do not openly challenge Didymus except to provide a mild correction; however, the citations themselves, and the manner in which Athenaeus integrates them into his work, are more revelatory of their role as part of his broader literary project. This raises the related problem of the extent to which Athenaeus manipulates Didymus’ words, which is exceptionally difficult to determine.30 However, two brief 28 The question of accuracy is frequently raised, including by Pelling 2000, the contributions in Lenfant 2007a, and most recently by Olson 2018, who builds upon the typology proposed by Lenfant 2007b to suggest that, while Athenaeus appears to handle all of his prose sources in a comparable fashion, there are frequent small changes that can fundamentally alter the meaning of the quoted passages, sometimes positioning them at odds with their original contexts. While Olson suggests that these ‘artfully distorted versions of the original’ exist (p. 448), he stops short of suggesting that such alterations participate in any literary or intellectual programme in the work. The problem of accuracy is particularly acute for texts for which Athenaeus serves as the only source, as in the case of certain of the fragments of Didymus; while any rewriting of the text necessarily remains invisible to the modern reader, nevertheless we should remain aware that Athenaeus may manipulate Didymus, much as he does his other prose sources. 29 Ath. 30e (27a), 56d (10), 67d (26 = 355), 70c (44 = 184), 139c (15 = 341), 368b (°1), 371f (°279), 392f (9), 468e (43 = °204a), 477e (14), 481f (17 = 291), 484e (18), 486d (299a), 487c (21), 501e (8), 634e (46 = 206), 636d (270b), 689b (207). 30 Athenaeus’ quotation habits, and his manipulation of the passages that he chooses to excerpt, may demonstrate deliberate aesthetic choices as part of a broader literary project; the discomfort with these changes in modern scholarship is based largely on a desire to

88



DiGiulio

examples may offer a glimpse into his general treatment. In Book 11, deep in the midst of a discussion of different kinds of drinking vessels, Athenaeus cites Didymus’ claim that the kumbion is shaped like the boat of the same name: φησὶ δὲ Δίδυμος ὁ γραμματικὸς ἐπίμηκες εἶναι τὸ ποτήριον καὶ στενὸν τῷ σχήματι, παρόμοιον πλοίῳ (‘Didymus the grammarian says that the cup is oblong and narrow in shape, comparable to the ship’, Ath. 11.481f = 17). Harpocration reports this detail with almost the exact same words (κ 92 = 291, φησὶ δὲ Δίδυμος ἐπίμήκες αὐτὸ εἶναι καὶ στενὸν καὶ τῷ σχήματι παρόμοιον τῷ πλοίῳ ὃ καλεῖται κυμβίον), as do the later entries for the word in the Suda, Hesychius, Eustathius, and the scholia to Lucian.31 Macrobius too discusses the kumbion, suggesting that the word is not a syncopation of kissibion, but rather a diminutive of kumba (Macr. Sat. 5.21.11); both the kumbion and the kissibion are discussed in close proximity by Athenaeus.32 Similarly, there are close parallels between Athenaeus’ use of Didymus to explain λυκιουργίδες (Ath. 11.486d–e = 299a) and Harpocration’s own entry for λυκιουργεῖς (λ 31 = 299b): both correct Didymus’ error on the basis of Herodotus and Critias, though Athenaeus quotes the latter more fully and supplements his explanation with a quotation of Aristophanes.33 The parallels between these passages suggest that in these instances Athenaeus accurately represented what Didymus had to say, or at least cited his own intermediate source’s report of Didymus’ words more or less precisely.34 In this regard, Athenaeus appears to respect the authority of the grammarian, even if he looks to correct his claims on occasion. Ultimately, in the majority of citations, Didymus and his works bolster the claims made by various participants, with his name carrying significant weight with respect to a range of texts. However, even in these relatively offhand references, a knowledge of Didymus and his work signals the learning of the banqueters on multiple levels, not simply demonstrating their acquaintance with the texts that abound throughout the Deipnosophistae, but also illustrating the resources at these figures’ disposal for locating different texts, as when a passage of Nicon (fr. 1 K.–A.) is cited through Didymus’ own quote (παρέθετο τὰ ἰαμβεῖα καὶ Δίδυμος καὶ Πάμφιλος, ‘both Didymus and Pamphilus present the iambs’, 11.487c = 21).35 Such quotation (and the indirect quotation of the passage of Nicon) draws from the memories of each of the Deipnosophists themselves, a reflection of the depth of their learning. At the same time, it is a manifestation of the compiler’s mental practice, extracting material from distinct sources and reintegrating it into a new whole; in the case of Athenaeus, the process of deconstruction and reconstitution reinforces the cultured nature of his characters. To return to the critical label ‘book-forgetter’ briefly, even though Didymus is challenged for potentially forgetting his own works, the comment is otherwise passed over without incident. Athenaeus then proceeds to cite Didymus at length, employing his work to access Polycrates’ History of Sparta (FGrH 588 F1) for information on the Hyacinthia festival (4.139c–e = 342). Ironically, while Didymus is criticized by Demetrius

31 32 33 34

35

understand his work as an anthology, rather than a literary work per se. The object of the audience’s derision in these instances, though, is likely to be Athenaeus’ characters, rather than the authors whom he excerpts, as in the case of Cynulcus and his attempts to make a basic math problem in Pl. Lg. 819b into a significant challenge (noted by Olson 2018: 443). Compare the ideas of creative intertext and its uses suggested by Paulas 2012: 408–10. Gibson 2002: 149–50. Macrobius seems to know Athenaeus first-hand, though he never cites the Deipnosophistae by name; see Cameron 2010: 583–84. Zecchini 2000: 154, who sees the parallels as suggesting a shared, post-Didymean source, rather than Athenaeus directly working with Harpocration. Lenfant 2007b and Maisonneuve 2007 suggest different paradigms for understanding Athenaeus’ quotation practice for Herodotus and Xenophon respectively; the former in particular distinguishes between quotations and paraphrases, arguing that quotations are more or less accurate, but Athenaeus’ paraphrases take more liberty with the text. Building on these approaches, Olson 2018 notes that the patterns identified by Lenfant largely translate to other authors as well, including the manner in which Athenaeus signals a quotation: as he suggests (428–31), when quoting Herodotus Athenaeus writes Ἡρόδοτός φησι; when paraphrasing, he writes ὥς φησιν Ἡρόδοτος, or else a range of indirect constructions. Translating this typology to the case of Didymus, only four of his citations should be treated as direct quotations: 2.67d (26 = 356), 2.70c–d (44 = 184), 9.392e–f (32), and 501e–f (8), if we understand the Lycophron passage reported to be quoted directly from Didymus. If the noted pattern were to hold in the case of Didymus, the vast majority of citations should be treated with a healthy dose of caution; however, at least two of the paraphrases and their parallels in Harpocration appear to indicate that those texts are reported fairly closely, even if through an intermediary. Jacob 2013: 99.

The compiler compiled



89

of Troezen, in the citation Athenaeus expressly cites Didymus, in order to access at a remove the books to which Didymus himself referred.36 While he may have forgotten certain of his own works, for Athenaeus and his characters Didymus’ citations of other texts are secure enough to include, and to consider accurate representations of the source texts. Likewise, Didymus’ quotation of Lycophron’s explanation of the etymology of βαλανειόμφαλοι is reported unproblematically—in fact, Didymus serves as a bridge linking the two parts of the passage, confirming the initial answer and using his quotation of Lycophron to support the next claim (11.501e–f = 8). Didymus thus provides a wealth of detail to be redeployed, woven through the Deipnosophistae to heighten the appearance of learning for Athenaeus’ characters; rather than condemning his absentmindedness, they borrow freely from Didymus as an authority on the issues which they discuss throughout. The enthusiastic compilation that produced Didymus’ works recombined these fragments into a new library, out of which his later readers like Athenaeus could excerpt material without issue. Like each of the Deipnosophists (and Athenaeus), Didymus is himself a librarian cataloguing and preserving a literary tradition.37 However, even if Didymus is useful both as scholar and as second-hand compiler of other texts, Athenaeus’ inclusion of his forgetfulness casts a shadow of doubt over his authority in the work.38 Although the various indirect quotations are included without question, the criticism of Didymus as a ‘book-forgetter’ alerts the reader to take care with the various quotations of Didymus’ own works in the Deipnosophistae, to say nothing of the works that are cited through his commentaries and other works. Indeed, for all that we might expect Didymus to be welcome among the banqueters for their intellectual similarities as compilers and excerpters, calling attention to his failing memory brings his reception into question. Memory, and the ability to recall texts, is central to the pepaideumenos’ identity as a cultured individual; when one of the characters’ memories lapses, this failing sets him up as an object of teasing and scorn.39 The same appears to apply to Athenaeus’ excerpted authors like Didymus, who is singled out for his faulty memory and who is most sternly criticized when he appears to have overlooked a fact that the Deipnosophists believe he should have known (11.486d = 299a). To embody the library, and to have an encyclopaedic memory of all the books that one has read, is one of the most prized qualities among Athenaeus’ érudits. That Didymus is given the epithet βιβλιολάθας includes him among this group, despite his generally secure position through the Deipnosophistae. Though he is not the pedantic grammaticus of Seneca and Quintilian, Didymus’ exhaustive learning is not without fault.40 While he is a valuable source for the banqueters, both as an authoritative 36 To be sure, the criticism levelled by Demetrius (and, implicitly, Athenaeus in repeating it) may be seen as sophistic if, like Quintilian, he is imagining a discrepancy between two different works separated by some time. The latter work may act as a correction to the original, hence explaining the divergence, as Braswell 2017: 39 suggests. 37 Athenaeus’ depiction demonstrates a tendency through the second and third centuries to understand individual sophists as representative of the books from which their learning derives. In this sense they become ‘breathing libraries’, as Eunapius termed the sophist Longinus (βιβλιοθήκη τις ἦν ἔμψυχος, VS 456); see Too 2000: 122–23. Virtually all of the characters in the Deipnosophistae demonstrate some degree of bibliophilia, with deep interest in and engagement with books and libraries, and their own practice throughout the work suggest at least some acquaintance with habits of librarians in the ancient world. Athenaeus’ banquet thus allows each of his characters to act as a kind of librarian in the Hellenistic tradition (Jacob 2000: 87–90; Jacob 2013: 55–69); Didymus’ work (and working habits) are of a piece with Athenaeus’ characters and his project, and would thus be seen as unproblematic. 38 The citations of Didymus, and Demetrius’ criticism, are given in the midst of a lengthy sermon on sympotic customs by the character Plutarch of Alexandria, noted to be a clever scholar and grammarian at the outset (παρῆν δὲ καὶ Πλούταρχος καὶ Λεωνίδης ὁ λεῖος καὶ Αἰμιλιανὸς ὁ Μαυρούσιος καὶ Ζωίλος, γραμματικῶν οἱ χαριέστατοι, Ath. 1.1c). Plutarch himself falls victim to a similar failure of memory at 8.359d–e (καὶ ὁ Πλούταρχος ἔφη Ῥοδιακὴν εἶναι λεγομένην ἱστορίαν, ἣν ἐπὶ τοῦ παρόντος ἀποστοματίζειν οὐ δύνασθαι τῷ πάνυ πρὸ πολλοῦ ἐντετυχηκέναι τῷ ταῦτα περιέχοντι βιβλίῳ), but is nevertheless able to introduce other intertexts to further the conversation (on this use of creative intertext as a model for reading in the work, see Paulas 2012: 429–34). Between his place of origin, profession, and forgetfulness, Athenaeus appears to invite comparison between Didymus and Plutarch, illustrating an avenue for the proper use of broad reading, even when one’s memory fails. 39 Jacob 2013: 74–75. 40 While the supposed pedantry of Didymus is never openly mentioned or alluded to by Athenaeus, his critiques of reading more generally may suggest a similar hostility to that demonstrated in the Latin sources. Athenaeus often juxtaposes intensive reading, focused on interrogating texts and producing attestations, with a more aesthetic reading, which avoids getting lost in the weeds of narrow philology; the two

90



DiGiulio

grammarian and as an excerpter of other texts, his readers nevertheless cannot overlook his failing memory for his own works, with its implications for the other texts indirectly quoted through him. 5. MACROBIUS, DIDYMUS, AND THE PRAISE OF COLLECTION Throughout much of the Imperial period there is an evident bifurcation between the Greek and Roman attitudes towards Didymus and his polymathic compilation: the Latin sources look down on Hellenistic scholarly volume, while Greek authors approach Didymus more critically, directly engaging with his work and incorporating it into their intellectual projects. In the fourth century CE these latter Greek attitudes emerge even in Latin, as evidenced by Macrobius, a voracious compiler in his own right who looks more kindly on Didymus’ voluminous corpus. Standing near the end of the classical literary tradition, Macrobius considers the Alexandrian to be ‘easily the most learned of all the grammarians’, grammaticorum omnium facile eruditissimus (Sat. 5.18.9 = °39), and similarly names him to be the best equipped of all grammarians that there was or had been (5.22.10 = 336). In Macrobius, then, Didymus encountered a kindred spirit that admired his work; perhaps the later writer’s own compositional strategies led him to be sympathetic to his predecessors’ work, alleviating any of the concerns raised by Quintilian or Seneca. Although criticism was levelled against Didymus for being a mere compiler, the same charge could be levelled against Macrobius, who not infrequently copies passages from other sources in their entirety without any indication that he is doing so.41 Beyond the affinities between his own compilatory methodology in the Saturnalia and the traditional interpretation of Didymus’ working habits, Macrobius embraces Didymus as a grammatical and linguistic authority, set apart from the plebeia grammaticorum cohors and respected as an eminent thinker, even as late as the fourth century 42 CE in the Latin West. In discussing Didymus, Macrobius includes him among his intellectual heroes, like Servius, and reaffirms Didymus’ place in the top tier of grammarians. Despite this lavish praise, it is perhaps surprising that Macrobius (and his characters) only cites Didymus three times, with two of the citations in quick succession and all three coming within the context of an extended discussion of Virgilian emulation and imitation of other poets. In each case, the citations all support Macrobius’ broader discussion, elucidating points of grammar and diction. However, while Macrobius preserves these substantial extracts from of the Λέξις Τραγική and the Ξένη ἱστορία respectively, he appears to be drawing on the previous commentators of Virgil in adducing these passages.43 To take the first example, on Virgil’s use of Achelous as a periphrasis for ‘water’ at Georgics 1.9 (°39), he cites Aristophanes and Ephorus as support and quotes extensively from each, before adding Didymus as an additional authority and quoting two further passages at length. Of these, the first two examples are also cited by Servius, who does not provide the Greek; however, he does include mention of Orpheus while omitting any reference to Didymus. On the one hand, this discrepancy between the two can be explained simply as the product of a common source, with Servius and Macrobius each choosing the portions of that source to suit their needs.44 However, this raises a related question: what, for Macrobius, is the utility of deploying Didymus as a source, rather than omitting it as Servius does? The context of the discussion may offer some insight. The majority of the fifth and sixth books of the Saturnalia focus on questions of Virgil’s use of Greek and Latin sources respectively, with the ultimate aim of

41

42 43 44

models are best represented by Ulpian and Cynulcus, respectively. The former is singled out throughout the work for his narrow focus in picking apart the texts that he cites, pedantically extracting material from his reading while leaving aside its aesthetic virtues; see Paulas 2012: 414–28; Jacob 2013: 73–74, 80–82. Gellius is among the more common targets in this regard, though material in Macrobius frequently has its roots in Plutarch, Athenaeus, and others. On the use of (and possible plagiarism from) Gellius, see Gunderson 2009: 259–64. On discussions of plagiarism in the Saturnalia more generally, especially with respect to Virgil, see McGill 2012: 178–209, and below. Cf. Macr. Sat. 1.24.8, where it is Servius that Macrobius praises as greater than the plebeia cohors; see Kaster 1988: 171. Cameron 2004: 198–200. For Cameron 2004: 199 the omission of Didymus by Servius, and of Orpheus by Macrobius, are ‘classic pointers to a common source’, with the DS scholia relaying from an earlier treatise what was exegetically necessary and Macrobius, in typical fashion, copying much of the passage in extenso.

The compiler compiled



91

defending the poet from charges of plagiarism.45 Book 5 in particular addresses the question of Virgil’s debt to Homer and other Greek sources. Eustathius defends the Roman poet against the critique levelled by the routinely bothersome Evangelus; both are Greeks celebrating the Saturnalia among Macrobius’ other characters, and are thus well qualified to drive the debate.46 Within this broader discourse, the cluster of Didymean quotations all focus on explicating the most obscure elements of the text, which would be clear only to those that have a full understanding of Greek literature (ad illa venio quae de Graecarum litterarum penetralibus eruta nullis cognita sunt, nisi qui Graecam doctrinam diligenter hauserunt, ‘I come to those things that have been dug up from the depths of Greek literature, known to no [Romans], except those that have attentively drunk Greek learning’, 5.18.1). Given the opacity of the reference that Eustathius here attempts to resolve, Didymus’ extensive learning and research makes him an ideal choice to illuminate the passage, which the grammaticorum cohors has otherwise failed to interpret correctly. The Alexandrian scholar’s reputation as a figure of deep learning is well suited to such a question; further, the citation of passages in which Didymus himself cites other ancient texts for their authorizing value (addetur auctoritas, 5.18.11) signals his discernment in collecting material for his commentaries, perhaps also serving as a reminder of his reputation as a compiler. However, the third and final reference to Didymus lacks the same degree of development as those in the discussion of Achelous and reflects a more complicated usage of the citation. In commenting on the appearance of Pan at Georgics 3.391–3, Macrobius’ character remarks that Valerius Probus, the eminent grammarian and commentator of the first century CE, could not identify Virgil’s source (5.22.9).47 He expresses his dismay that Probus would not know this, observing that it originates from Nicander before sharing Didymus’ evaluation of the poet: nam Nicander huius est auctor historiae, poeta quem Didymus, grammaticorum omnium quique sint quique fuerint instructissimus, fabulosum vocat (‘for Nicander is the author of this story, a poet whom Didymus, the best trained of all the grammarians who are or have been, calls “full of myths”’, 336). As an authoritative commentator on Greek poetry, Eustathius posits Didymus as a logical choice to consult when facing a challenging author such as Nicander—a point which Valerius Probus seems to have neglected, and his opinion suffers as a result.48 Once again it would seem that Didymus serves to provide information on obscure Hellenistic poets that his reputation guarantees, and in this regard Eustathius cites Didymus in much the same way that many of Athenaeus’ Deipnosophists did: to demonstrate his own learning at the expense of those, like Probus, whose own knowledge falls short. Yet, the framing of the reference implies Macrobius is using the citation of Didymus here for a purpose other than simple clarification of the source. As the discussion continues, Macrobius suggests that Virgil 45 As McGill 2012: 178–79 observes, such charges emerged quite early, and the Suetonian–Donatan Vita notes (44–46) that Perellius Faustus and Quintus Octavius Avitus produced works aiming to illustrate Virgil’s plagiarism; these questions, based on the Vita, engendered significant debate, with works published by Virgil’s supporters and detractors. Macrobius and the Saturnalia fall into the former camp. 46 Evangelus receives a consistently negative portrayal throughout the Saturnalia, serving as a nuisance for the other guests with his constant complaints and interjections, as typified by his attack on Servius during the banquet (2.2.12); see Kaster 1980: 226–29, 241–43; König 2012: 224–26. Eustathius’ depiction is more measured, yet he is nevertheless reproached in the work, like Evangelus, for his Greek cleverness. His authority is implicitly challenged, despite his status as a Greek philosopher, and his intellectual proclivities in explaining Virgil frequently border on the methods he seeks to condemn. 47 Valerius Probus is generally well regarded in antiquity, including among Macrobius’ sources, such as Gellius, who does not explicitly disagree with the grammarian; indeed, he depicts him as so great an authority in the latter part of the second century that his name could be invoked to lend support to arguments, whether accurately or speciously (e.g. NA 15.30; on Gellius’ portrait of Probus, see HolfordStrevens 2003: 163–65). While he includes him among the grammatici and professores, Suetonius notes that he never took real students (hic non tam discipulos quam sectatores aliquot habuit, DGR 24.3), which may suggest a note of disapproval; however, his discussion skews towards the positive (though perhaps not to the extreme posited by Grisart 1962: 389–90, who sees Probus as an idealized grammaticus). On the depiction of Probus in the DGR, see Kaster 1995: 248–50. 48 The citation of Didymus may be less logical than Eustathius thinks, though, as his critical work does not seem to have covered Hellenistic literature; a member of Macrobius’ audience well versed in the grammatical tradition would surely have recognized this fact, which implicitly challenges the authority of Macrobius’ character.

92



DiGiulio

provides confirmation of his use of Nicander through his own hedging: he notes that the story might be beyond belief, but the inference drawn by Macrobius hardly seems definitive. Unlike before, Didymus here in fact lacks real authorizing force—he is cited only to note Nicander’s reputation rather than provide more detailed commentary—and his opinion suggests the need to take care in incorporating sources that may not be entirely credible, or have a reputation for embellishment. The passing reference to Didymus and his opinion on Nicander challenges the authority of Valerius Probus, with the citation serving to confirm the claims that Probus should have known the ultimate source for Virgil’s own description of Pan. Perhaps there is a hint of irony here, as Macrobius muddies the waters by drawing on the traditional Latin ideas about Didymus. This citation, attacking Probus’ forgetfulness as it does, bears resemblance to Quintilian’s anecdote about Didymus, who questioned a narrative as vana, only to discover that in fact it stemmed from his own pen. Moreover, in that same passage Quintilian goes on to argue that grammatici are particularly prone to questioning the source of every mythological allusion in texts in an effort to demonstrate the breadth of their learning, taking these enquiries to absurd extremes and even fabricating potential sources (Inst. 1.8.21).49 This sets up a clear dichotomy between the two claims that are laid out in the Saturnalia. Valerius Probus, as Macrobius notes, confessed that he was not able to locate a source underlying Virgil; on the terms laid out by Quintilian, this acknowledgement that his own learning had limits demonstrates Probus’ virtue as a grammaticus. In contrast, Eustathius’ insistence on uncovering what Valerius Probus could not leads him to Didymus for his indirect confirmation, and for the assertion that he considered Nicander a poeta fabulosus. Eustathius suggests that Virgil was aware of this reputation, and hence included the lines si credere dignum est as a kind of Alexandrian footnote. He does not, however, provide the actual Didymean source or text, as he did earlier, but invokes only the scholar’s name.50 The absence of a specific source ostensibly places the emphasis on Eustathius’ own knowledge and authority, but Macrobius’ framing of the reference, with its reminiscences of Quintilian’s discussion of Didymus, provides a degree of implicit challenge to Eustathius and his claims. Unlike Probus, who admitted his ignorance, Eustathius produces Nicander as the source while giving voice to Didymus’ doubts about the poet’s accuracy. In the light of Quintilian, one may question the ultimate source and the veracity of the claim —the grammaticus names Nicander, a challenging and recondite poet, as the source for Virgil’s borrowing in an effort to demonstrate his own intellectual superiority, and uses an offhand reference to Didymus to buttress these claims. However, unlike his previous citations, in which he can point to specific passages, here he relies only on Didymus’ name. Further, both the Quintilianic reminiscence and the kind of argument that is being made problematize Didymus as a source and the intellectual habits being demonstrated. Didymus himself, who for Quintilian served as a negative exemplar of this particular tactic of challenging obscure passages by hunting down their ultimate sources, hardly seems the surest support here. While Macrobius is sympathetic to the arguments defending Virgil against charges of plagiarism, the scene by this point has become a satire of the grammaticus’ endeavour; indeed, Macrobius has had his character perform a point-by-point defence at length.51 He even goes so far as to performatively conclude his discussion, noting that he could go on at length and fill huge book-rolls on the topic, but will stop to avoid offending his audience’s sensibilities (nam si fastidium facere non timerem, ingentia poteram volumina de his quae a penitissima Graecorum doctrina transtulisset implere, ‘if I weren’t afraid of repelling you, I could fill huge books of these 49 Quod evenit praecipue in fabulosis usque ad deridicula quaedam, quaedam etiam pudenda, unde improbissimo cuique pleraque fingendi licentia est, adeo ut de libris totis et auctoribus, ut succurrit, mentiantur tuto, quia inveniri qui numquam fuere non possunt: nam in notioribus frequentissime deprenduntur a curiosis. Ex quo mihi inter virtutes grammatici habebitur aliqua nescire. 50 Schmidt 1854 suggests that the passage indicates the narrative of Pan and Luna was likely included in the Ξένη ἱστορία (336), and that this was possibly Virgil’s ultimate source, or at the very least among the sources for the Virgil scholiasts (DServ. ad G. 3.391–3 suggests Nicander as the source but omits Didymus). However, to the best of my knowledge none of the Virgil scholia identify the particular work, and while Schmidt’s assignment of the fragment seems likely, the ambiguity left here by Macrobius appears deliberate. 51 The apologia runs for almost the entirety of Book 5; by this point, almost the end of the book, the comparison of Virgil and his Greek sources has gone on for nearly 100 pages in Kaster’s Oxford edition.

The compiler compiled



93

sorts of things that he translated from the most obscure knowledge of the Greeks’, 5.22.15). Given the extensive length of the discussion thus far, it is ironic that Macrobius stops the passage here, before then turning to Virgil’s Roman sources in Book 6: he has already delved deeply into the Greek tradition, drawing from quintessential grammatical figures like Didymus, but stops himself for fear of producing the sort of voluminous output that typified Didymus himself. Eustathius claims to check himself before falling too far down the grammaticus’ rabbit hole, but his final citation of Didymus suggests that, at least based on Quintilian’s definition, he has already reached that point. Macrobius offers a curious final glimpse of Didymus and his reception in the miscellanistic tradition. For all of Macrobius’ claims about Didymus’ excellence as a grammarian and critic, he makes little use of him throughout the Saturnalia as a whole, instead relying on his discussions of minor linguistic points. Further, he appears to take some licence with his actual use of Didymus, harmonizing citations with both his context and a broader critical tradition in Latin literature. Quintilian in particular lurks in the background of Macrobius’ citations, and the praise of Didymus as a grammaticus still leaves him open to earlier critiques. Macrobius reveals Didymus to be in the first instance a learned author with great exegetical value, and in the second instance a cautionary example of the grammaticus’ method, especially when read alongside Quintilian. By using Didymus as he does in the latter instance, Macrobius offers a wink to his audience at the expense of his character, who has the industry and breadth of knowledge which typified Didymus, and claims to stop himself before fully offering the same degree of output. Macrobius redeems Didymus among Latin literary authors, and unlike Seneca or Quintilian appears to be familiar with the actual text of his works; in this regard he marks a break from the earlier Latin tradition, and as a grammarian he may have expected his like-minded audience to have first-hand familiarity with Didymus. However, he still sees an opportunity to employ the scholar as a literary tool at the expense of one of his characters. If Macrobius’ treatment of him is any indication, Didymus’ reputation among the Latin grammatici of the fourth century was greatly improved relative to Quintilian and Seneca’s appreciation of him; such a reception likely goes hand in hand with the increased esteem in which grammarians were held in the fourth century. Yet, close attention to these brief citations suggests that elements of the critical assessments of the early Empire lingered, with Didymus emblematic of some of the grammarian’s greatest vices. Macrobius recombines these earlier caricatures with the critical eye of our Imperial Greek sources in order to fashion a Didymus that can serve his literary enterprise. 6. CONCLUSION As I hope to have shown in this survey, Didymus’ reception throughout the tradition of compiled literature is complex, as he was recognized as a prodigious polymath and kindred spirit to the later compilers, yet he could not escape their reproach for his intellectual habits. His status as one of the last great grammarians made him an easy straw man for those seeking to attack the grammatici and their enterprise, particularly in the first and second centuries CE. Yet, this line of attack exposes a degree of cultural chauvinism, as it is the Latin authors (Seneca, Quintilian, and to a lesser extent Macrobius as he reflects on the tradition) that call Didymus to task for his working habits and proclivities. In Didymus, there was a ready target through whom one might rebuke the worst elements of the grammatici and Greek scholarship, such as their pedantry and insistence on the pursuit of trivial facts and details. To be sure, the Roman grammarians and their ilk engaged in similar forms of enquiry, but, as Seneca suggests, only through the influence of Greek thinkers like Didymus. For the Romans, Didymus’ pursuit of grammatical knowledge lacked any practical use, and that he produced so great a volume of writing only reinforced such an enterprise’s lack of utility. In some regards, this image of Didymus resembles the critical portrait of Pliny the Elder constructed by Aulus Gellius: a compiler indiscriminately heaping together masses of material to little avail.52 While the two likely shared work habits, Pliny still places a premium on utility (e.g. HN. praef. 17), which situates his work within the 52 On the Gellian portrait of Pliny and the challenges it presents, see Howley 2018: 112–56.

94



DiGiulio

mainstream of Roman elite culture and renders himself safe from consternation. Didymus, in contrast, serves as a cipher for Greek learning and its potential pitfalls. Indeed, the Romans are concerned far more with the idea of Didymus, rather than his work: while they are happy to lambast his reputation, there is virtually no direct engagement with his work itself. The Greek compilers and their use of Didymus appear kinder and more willing to work with Didymus on his own terms; in fact, both Harpocration and Athenaeus see in Didymus an authoritative grammarian, even if his erudition falters on occasion. In the Lexeis and the Deipnosophistae, Didymus is regularly cited both for his own opinions as well as for the texts of other authors and commentators that he recorded. Whether or not his authority is unimpeachable, Didymus is nevertheless engaged with directly—his works are read and directly cited, unlike in Seneca and Quintilian—and this in part may be owed to the similarity in genre. Both Harpocration and Athenaeus work adroitly within traditions of learned Greek literature, freely incorporating other grammarians and linguistic authorities into their own works, as opposed to the polemical discussions within the Latin tradition. Yet, even as Didymus is invoked, his reputation as a ‘book-forgetter’ and his predilection for minute detail loom in the background. While he and his works carry the weight of the Alexandrian tradition, nevertheless he must be treated with care. In Macrobius (a grammaticus himself), these two traditions blend together, as he appears to hold Didymus in the highest esteem. As he does with many of his sources, however, Macrobius integrates his citations of Didymus into his work in such a way as to challenge his authority and unites the positive and negative traditions together. As one of the Alexandrian grammarians par excellence, Didymus is a model for Macrobius and his characters; yet the eager invocation of Didymus, and thus the authority of the character citing him, is problematized by the Roman frame of the discussion. Even among the miscellanists in Late Antiquity most kindred to him in methodology, Didymus’ reputation is complex, a figure of both authority and consternation who was fundamentally difficult to digest and reintegrate into the miscellanistic tradition. These glimpses of Didymus complicate the picture of polymathic scholarship in antiquity, and of Didymus’ own intellectual reputation. Whether treating him as a source or as a stereotype, by the Imperial period Didymus had become a shorthand for thinking about polymathy. While he was doubtless important for his monumental corpus and contributions to ancient learning, our surviving fragments and testimonia demonstrate the sophistication with which these intermediaries cited and transformed him to suit their own purposes.

Bulletin of the Institute of Classical Studies, 2020, 63, 95–120 doi: 10.1093/bics/qbaa022 Original article

A checklist of the testimonia and fragments of Didymus Thomas R. P. Coward1 and Enrico Emanuele Prodi2 1

Ca’ Foscari University, Venice, Italy 2 University of Oxford, Oxford, UK

This Checklist aims to collect the testimonia to Didymus’ life and references to all extant fragments of his works. It is not, and could not have been, a proper edition to supersede the one published by Moritz Schmidt in 1854; the latter endeavour we leave to guts more brazen than ours. But we believe that a Checklist of fragments, organised by works, will be a useful instrument for those working on Didymus in the interim. Scholars will be able to have as complete an overview as possible of the evidence for Didymus’ output and to cite any fragment of it in a simple and unambiguous way. We have only included those fragments which are explicitly attributed to Didymus by an ancient source, though we have sought to remark when other sources preserve the same information without attribution. While this meant leaving out much that is probably or certainly Didymean (one need only think of Didymus’ quarter of the Iliadic VMK, or the doxographies in the Pindar scholia), we believe that securely attributed material is a safer starting point for the interpretative and comparative work that will turn up more. We have, however, included two fragments where the source does not cite Didymus by name, but does cite a title which is attested among his works. In these cases, the number is prefaced by an asterisk as a mark of uncertainty (*38, *49). When the source cites both Didymus and the title of the work, the number is prefaced by a circlet (°1, °2, °3, etc.). Given our policy of only including named fragments, we have excluded several papyri which have been connected to Didymus at one point or another, with varying degrees of plausibility but without conclusive proof. We offer a list here, with no pretence of exhaustiveness. The miscellany P.Oxy. XIII 1611, with its wealth of quotations and its overlaps with the Aristophanes scholia and Marcellinus’ Life of Thucydides, seems a promising candidate for the Symmikta. P.Oxy. XV 1801 recto (CLGP I.1.4 Aristophanes 3) preserves a lexicon of comic words with a great deal of parallels in Hesychius; it could be Didymus’ Λέξις κωμική, or Theon’s, or another’s. Several hypomnēmata to Aristophanes have been linked to Didymus or regarded as excerpts from his work: P.Oxy. VI 856, XXXV 2737, P.Flor. II 112 (CLGP I.1.4 Aristophanes 1, 27, 28; two marginal annotations on this last papyrus mention Didymus explicitly; see 276, 277). The same goes for two Pindaric commentaries, P.Berol. inv. 13875 and PSI XIV 1391 recto, and the hypomnēma to Bacchylides’ Dithyrambs, P.Oxy. XXIII 2368 (CLGP I.1.4 Bacchylides 4). Didymean authorship has also been suggested for P.Oxy. XXXV 2744, which appears to be a commentary to an unknown text. Two mutilated titles mention someone who can be supplemented as Didymus: P.Berol. inv. 11903 = BKT X 20 (the book was entitled ]τεχνος, unattested among our Didymus’ works) and P.Oxy. inv. 51 B 44/G(b) (a commentary to

© The Author(s) 2021. Published by Oxford University Press on behalf of Institute of Classical Studies. All rights reserved. For permissions, please email: [email protected]



95

96



Coward and Prodi

Aristophanes—but other supplements are possible for the author’s name).1 A Didymus and his πρόγραφα or πρόσγραφα are mentioned in P.Lond. inv. 2110 recto (SB XX 14599) in connection with the copying of books;2 there is no way of telling whether this was our man or one of the many other writers who shared his name. In many cases our attribution of a fragment to a specific work is conjectural. We believe that an educated guess based on content, when possible, is preferable to lumping large numbers of fragments together among those incerti operis. Too little is known about the Περὶ παθῶν, however, for the many etymological fragments transmitted by the Byzantine Etymologica to be attributed to that work by editorial fiat: attractive though that course was, they have been relegated to the incerti operis instead. Often a degree of ambiguity remains, especially when a given interpretation could have occurred either in a commentary to a text or in a lexicographical work. The ambiguity is compounded by the fact that Didymus must have reused material across different works on the same topic, and therefore some things must have been found both in a lexicon and in the commentary to the relevant text, or in works of other kinds (see for instance the triplet formed by 221, °281 col. xi.63 ff., and 355). Altogether we make no claim to inerrancy in this matter, even more than in others. Most of the titles in our checklist are attested by the ancient sources, but we have accepted a few which can reasonably be hypothesised even without direct attestation (e.g. X, XI, XV). We have discarded some of Schmidt’s less felicitous inventions, such as the Metamorphoses and On Geography. Likewise, the Polemical Explanations against Juba have been returned to their transmitted, and more satisfactory, title (XVIII). We have retained On the Different Meanings of Words (Περὶ διαφορᾶς λέξεων) as a separate work from On the Wrong Use of Words (Περὶ διεφθορυίας λέξεως). We share Schmidt’s—and others’—suspicion that the titles Συμποτικά and Σύμμικτα may refer to one and the same work, but in the absence of positive evidence to contradict the transmitted titles, we keep them separate (XXXVII, XXXVIII). We accept the Suda’s testimony on the existence of ‘Didymus the Younger’ (δ 873 Adler) and his authorship of Πιθανά and Περὶ ὀρθογραφίας; we have consequently excluded those two works from our count, and all the Didymean fragments which Schmidt assigned to the Περὶ ὀρθογραφίας are now incerti operis. We have also attributed to the younger Didymus, and therefore kept off our list, the Hippocratic lexicon mentioned by Erotian in his preface (p. 5 Nachmanson), since the younger Didymus—unlike his elder namesake—is independently credited with a fragment on a medical topic (Et. Gud. s.v. ἀρτηρία, I p. 208 add. De Stefani).3 Likewise we accept the Suda’s attribution of On Analogy among the Romans to Didymus Claudius (δ 874 Adler), and the resulting absence of Roman interests among Didymus’ genuine fragments speaks against his authorship of the polemic against Cicero’s De re publica (Suda τ 895 Adler), pace Amm. Marc. 22.16.16. The Checklist broadly follows the organising principles of Schmidt’s table of contents and Braswell’s critical catalogue of Didymus’ works.4 Lexicographical works come first (I–VI, °1–*49), then commentaries to individual authors and related works (VII–XXXIII, 50–327), grammatical treatises (now reduced to XXXIV, °328), and works of antiquarian interest (XXXV–XLI, 329–°361); fragmenta incerti operis (362–402) and the list of dubia and spuria (403–05, 406–14) conclude the Checklist. Titles of works are given Roman numerals, while we use Arabic numerals for fragments. Unlike Schmidt with his idiosyncratic reference system, we number all the fragments continuously throughout the Checklist. When a fragment is preserved by two or more sources naming Didymus, we number them as 11a, 11b, etc.; further sources transmitting the same material without attribution are listed after the relevant entry, introduced by ‘cf.’. For reference purposes the Checklist can be cited with the respective testimonium or fragment number followed by C.–Pr., e.g. Didym. Lex. Com. F36 C.–Pr. or Didym. V 36 C.–Pr., or Didym. T11 C.–Pr. 1 2 3 4

Montana 2012b: 6–7 and n. 23. Re-edition and commentary in Caroli 2019. Schmidt 1854: 27. Schmidt 1854: 11–13; Braswell 2017: 40–103. Note that the actual order of the fragments in Schmidt’s edition differs from the one presented in the table of contents. Our Concordance includes page nos. as well as fragment nos.

A checklist of the testimonia and fragments of Didymus



97

Testimonia T1 (1 Br.) T2 (2 Br.) T3 (3 Br.) T4 (5 Br.) T5 T6 T7 T8 T9 T10 T11 T12 T13 T14

Suda δ 872 Adler Suda ι 399 Adler Suda α 3215 Adler Suda η 463 Adler Demetrius of Troezen, SH 376 ap. Ath. 4.139c Quint. Inst. 1.8.20 Amm. Marc. 22.16.16 Hier. Epist. 33.1 Isid. Orig. 6.7.1 Suda χ 29 Adler  T1 Tz. Prol. in Ar. 1 Const. Manasses, Chron. 6534–36 Manuel Holobolus, Or. 2, 92.2 Sen. Epist. 88.37 = 394–97

Fragmenta I. On the wrong use of words (Περὶ παρεφθορυίας/διεφθορυίας λέξεως) °1 °2

Ath. 9.368b Σ Ar. Av. 768e Holwerda cf. Hsch. ε 1604 Cunningham

II. On the different meaning of words (Περὶ διαφορᾶς λέξεων) °3

Σ Ar. Pl. 388b Chantry

III. On puzzling words (Περὶ ἀπορουμένης λέξεως) °4

Harp. δ 23 Keaney (δερμηστής) cf. Hsch. δ 684 Cunningham; Phot. δ 201 Theodoridis; Suda δ 261 Adler; Et. Gen. s.v. δερμηστής

IV. Glossary of figurative expressions (Λέξις τροπική) °5

a Syn. B α 233 Cunningham (ἀγαθοεργοί) b Phot. α 79 Theodoridis

V. Glossary of the language of comedy (Λέξις κωμική) Cf. Hsch. Ep. ad Eulogium line 4 Cunningham; Gal. De ind. 24a 6

Gal. De ind. 27 (ἀβυδοκόμη) = 350 cf. Hsch. α 225 Cunningham; Phot. α 63 Theodoridis; Syn. Β α 46 Cunningham; Eust. in Il. 2.836, I pp. 559–60 van der Valk (Paus. Lex. α 3 Erbse); Zen. 1.1

98



7

8 9

10 11

12

°13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

28 29 30

Coward and Prodi

Gal. De ind. 27 (ἀβυρτάκη) cf. Syn. B α 61 Cunningham; Phot. α 66 Theodoridis; Suda α 103 Adler; Et. Gen. α 2 Lasserre–Livadaras; EM α 30 Lasserre–Livadaras; Et. Sym. α 16 = MG α 19 Lasserre– Livadaras; [Zonar.] I p. 8 Tittmann; Σ Luc. 46.6 Rabe; Eust. in Od. 19.28, II p. 188 Stallbaum (Ael. Dion. α 8 Erbse; Paus. Lex. α 4 Erbse) Ath. 11.501e (βαλανειόμφαλοι φιάλαι) Hsch. β 1152 Cunningham (βρίκελοι) cf. Eust. in Od. 1.101, I p. 26 Stallbaum (Paus. β 20 Erbse); Et. Gen. β 259 Lasserre– Livadaras; EM β 324 Lasserre–Livadaras; Et. Sym. β 210 = MG β 224 Lasserre–Livadaras Ath. 2.56d (δρυπεπεῖς ἐλᾶαι) a Et. Gen. s.vv. ἐπίδαι, ἐμπίδαι (Baldi) b EM. s.v. ἐπίβδαι, pp. 357–58 Gaisford c Et. Sym. ε 601 Baldi cf. Tim. Soph. ε 46 Valente; Hsch. ε 4622 Cunningham; Σ Pi. P. 4.247a Drachmann; [Zonar.] s.v. ἐπίβδαι, I p. 800 Tittmann a Phot. η 152 Theodoridis (ἡμεροκαλλές) b EM s.v. ἡμεροκαλλές, p. 429 Gaisford cf. Ath. 15.681e; Hsch. η 470 Cunningham; Et. Gen. p. 149 Miller a Et. Gud. s.v. καρύκη, p. 301 Sturz b EM s.v. καρύκκη, p. 492 Gaisford Ath. 11.477e (κιβώριον) Ath. 4.139c–e (κοπίς) cf. P.Oxy. XV 1801 col. ii.31–32 = CLGP I.1.4 Aristophanes 31 Hsch. κ 3661 Cunningham (κόρσακις) Ath. 11.481f (κυμβία) = 291 cf. Macr. Sat. 5.21.9; Hsch. κ 4542 Cunningham Ath. 11.484e–f (λαβρώνια) Hsch. λ 699 Cunningham (λέσπιν) Hsch. λ 1041 Cunningham (Λιμοδωριεῖς) = 354 cf. Phot. λ 314 Theodoridis; Suda λ 555 Adler; [Plu.] Par. Alex. 1.34 (CPG I p. 326) Ath. 11.487d (μάνης) Hsch. μ 1705 Cunningham (Μοσσυνικὰ μαζονομεῖα) Hsch. μ 1824 Cunningham (Μυκερίνα) Phot. ο 127 Theodoridis (οἰνόπται) Hsch. ο 409 Cunningham (οἰσπώτη) cf. Phot. ο 154 Theodoridis; Poll. 5.91 Bethe; Σ Ar. Lys. 575 Hangard Ath. 2.67d (ὄξος) = 357 cf. Hsch. ο 939 Cunningham; App. Provv. 4.29 (CPG I p. 440) a Ath. 1.30d–e (Πράμνιον/Πράμνια) b Eust. in Il. 11.638–39, III p. 284 van der Valk cf. Hsch. π 3196 Hansen; Suda π 2207 Adler; Apostol. 14.74; Σ Ar. Eq. 107a–b Jones– Wilson; Σ Nic. Alex. 163a Geymonat Phot. ρ 53 Theodoridis (ῥάχετρον) cf. Hsch. ρ 151 Hansen Σ A.R. 1.1134–39b Wendel (ῥόμβος) Hsch. σ 147 Hansen (Σαμιακὸς τρόπος) cf. Phot. σ 59 Theodoridis

A checklist of the testimonia and fragments of Didymus

31 32 33 34

35 36

37 *38



99

Phot. σ 148 Theodoridis (σέρφοι) cf. Hsch. σ 433 Hansen Ath. 9.392f (σιαλίς) Hsch. σ 742 Hansen (σισύμβριον) Hsch. σ 1014 Hansen (σκινθαρίζειν) cf. Suet. Περὶ παιδιῶν A.20 Taillardat; Poll. 9.126 Bethe; Eust. in Il. 11.536, III p. 248 van der Valk Σ A.R. 4.1058a Wendel (στρευγομένοις) a Phot. τ 219 Theodoridis (τευτάζειν) b Suda τ 431 Adler c EM s.v. τευτάζειν, p. 755 Gaisford cf. Phot. Amphil. 21.54–58 Westerink; Eust. in Il. 14.80s., III p. 579 van der Valk (Ael. Dion. τ 11 Erbse); Et. Gen. p. 280 Miller Choerob. GG IV.1, p. 104 Hilgard (ψάλυξ) cf. Hsch. ψ 57 Hansen–Cunningham Σ Ar. Pl. 313e Chantry (μίνθον)

VI. Glossary of the language of tragedy (Λέξις τραγική vel τραγῳδουμένη) °39

40

41 42 43 44 45 46 °47

48 *49

a Macr. Sat. 5.18.9 (Ἀχελῷος) b Eust. in Il. 4.475, I p. 793 van der Valk cf. DServ. in Georg. 1.8 Thilo–Hagen; Hsch. α 8841 Cunningham; Σ Il. 21.194a1 Erbse; Σ Il. 24.616b Erbse; Σ E. Andr. 167 Schwartz; Eust. in Il. 26.616, IV p. 962 van der Valk a Syn. B α 426 Cunningham (ἀηδόνα) b Phot. α 441 Theodoridis cf. Hsch. α 1498, 1500 Cunningham Σ E. Tr. 1079 Schwartz (αἰθέρα) = 200 cf. Hsch. α 1855 Cunningham; Phot. α 570 Theodoridis; EM s.v. αἴθρα, p. 33 Gaisford Eust. in Od. 2.10, I p. 78 Stallbaum (ἄργεμα) cf. Hsch. α 7026 Cunningham Ath. 11.468c–f (δακτυλωτὸν ἔκπωμα) = 204 cf. Hsch. δ 145 Cunningham Ath. 2.70c (κύναρος) = 184 cf. Hsch. κ 4562 Cunningham Hsch. λ 311 Cunningham (Λαπέρσαι) cf. Eust. in Il. 2.319, I p. 350 van der Valk; Σ Lyc. Alex. 1369b Leone; St. Byz. λ 1 Billerbeck Ath. 14.634c–f (μάγαδις) = 206 cf. Hsch. μ 3 Cunningham; Phot. μ 2 Theodoridis Harp. ξ 4 Keaney (ξηραλοιφεῖν) cf. Hsch. ξ 65 Cunningham; Phot. ξ 29 Theodoridis; Suda ξ 61 Adler; Lex. Rhet. p. 284 Bekker; EM p. 612 Gaisford; Et. Sym. cod. V fol. 129r (Gaisford); [Zonar.] II p. 1417 Tittmann; Σ Aeschin. 1.138 Dilts; Σ Luc. 46.2–4 Rabe St. Byz. π 188 Billerbeck–Neuman-Hartmann (Πνύξ) = 205 cf. Hsch. π 4331–32 Hansen (πυκναῖα) Σ S. Tr. 1159 Xenis (πρόσφαντον)

100



Coward and Prodi

VII. On the Aristarchan recension (Περὶ τῆς Ἀριστάρχου διορθώσεως) Cf. Il. cod. A subscr. libr. 1–16, 18–23 Iliad 50 51 52 53 54 55 56 57 58 59 60 61 62 °63 64 °65 °66 67

Σ Il. 2.111b Erbse Σ Il. 2.258a1 Erbse Σ Il. 2.397b Erbse Σ Il. 2.435a1 Erbse Σ Il. 3.10b Erbse Σ Il. 6.71a1 Erbse a Σ Il. 7.135c Erbse b Et. Gen. s.v. Φειά, p. 299 Miller Σ Il. 8.535–37 Erbse Σ Il. 10.53a1 Erbse Σ Il. 10.397–99a Erbse Σ Il. 11.441a1 Erbse Σ Il. 13.450a1 Erbse Σ Il. 15.86c Erbse Σ Il. 17.607c1 Erbse Σ Il. 19.76–77 Erbse a Σ Il. 21.110a Erbse b Σ Il. 21.110c Erbse Σ Il. 24.557a Erbse Σ Il. 24.663b2 Erbse

Odyssey 68

Σ Od. 2.260d1 Pontani

VIII. Commentaries to the Iliad 69 70

°71 °72 73 74

a Σ Il. 1.399 Erbse b Σ D Il. 1.399 van Thiel Et. Gen. α 1391 Lasserre–Livadaras (αὖθι) cf. Σ D Il. 1.492 van Thiel; EM α 2076 Lasserre–Livadaras; Et. Sym. α 1555 = MG α 1561 Lasserre–Livadaras Ammon. Diff. 513 Nickau cf. Σ Il. 2.183b Erbse Ammon. Diff. 300 Nickau cf. Σ Il. 2.494 Erbse St. Byz. α 361 Billerbeck (Ἀπολλωνία) cf. Σ Il. 2.519a–b Erbse a St. Byz. σ 122 Billerbeck–Neumann-Hartmann (Σήσαμον) b Eust. in Il. 2.853, I p. 569 van der Valk cf. Σ Il. 2.853 Erbse; Σ A.R. 2.941–42 Wendel

A checklist of the testimonia and fragments of Didymus

75 76 77 78 79 80

81 82 83 84 85 86 87

88

89

Σ D Il. 4.475 van Thiel Σ Il. 10.428b Erbse St. Byz. α 8 Billerbeck (Ἄβιοι) cf. Σ Il. 13.6 c/e Erbse St. Byz. α 24 Billerbeck (Ἀγάθυρσοι) cf. Σ Il. 13.363b Erbse; St. Byz. κ 2 Billerbeck Hdn. Orth. ap. Reitzenstein 1897: 307 a Syn. B α 2550 Cunningham (ἄφλαστα) b Phot. α 3369 Theodoridis c Eust. in Il. 15.717, III p. 790 van der Valk (Paus. Lex. α 177 Erbse) cf. Hsch. α 8701 Cunningham; Σ. Il. 15.717a Erbse Σ Il. 19.81–82 Erbse Σ Il. 19.90c, d Erbse Σ Il. 19.116a1 Erbse Σ Il. 21.203 in P.Oxy. II 221 col. x.12 cf. Σ Il. 21.203b Erbse Σ Il. 21.363 in P.Oxy. II 221 col. xvii.27–28 cf. Σ Il. 21.363c Erbse Σ D Il. 22.126 van Thiel a Cyrill. auct. s.v. ἐνδίνων, An. Par. IV p. 182 Cramer b Et. Gud. ΙΙ p. 469 add. De Stefani cf. Σ Il. 23.806d Erbse; Orion p. 52 Sturz; EM s.v. ἔνδινα, p. 339 Gaisford a Et. Gen. s.v. πείρινθος ap. Neitzel 1977: 269 b EM s.v. πείρινθος, p. 668 Gaisford c [Zonar.] II p. 1531 Tittmann d EM s.v. κάνναθρον, p. 489 Gaisford cf. Σ Il. 24.190c Erbse Σ Il. 24.640 Erbse

IX. Commentaries to the Odyssey 90

°91

°92 °93

°94 °95

a Σ Od. 4.797c Pontani b Eust. in Od. 1.344, I p. 65 Stallbaum cf. Σ Pi. O. 9.79d Drachmann; Tz. in Lyc. 792 Scheer a Et. Gen. α 1133 Lasserre–Livadaras (Ἀργινοῦσαι) b Et. Gen. s.v. Σκύρος ap. Reitzenstein 1897: 48 c EM s.v. Σκύρος, p. 720 Gaisford cf. EM α 1744 Lasserre–Livadaras St. Byz. α 410 Billerbeck (Ἀρέθουσα) a Et. Gen. s.v. περισκέπτῳ ap. Alpers 1981: 112 b EM p. 664 Gaisford cf. Σ Od. 14.6 Dindorf Et. Gud. s.v. ἀνέγναμψαν, I p. 138 De Stefani cf. Et. Gen. α 845; EM α 1345; Et. Sym. α 1008; MG α 1007 Lasserre–Livadaras a Orion, s.v. ἀρετή, col. 1 Sturz b Et. Gud. I p. 190 add. De Stefani cf. Et. Gen. α 1142; EM α 1761; Et. Sym. α 1341 = MG α 1345 Lasserre–Livadaras



101

102



Coward and Prodi

96 a Syn. B α 931 Cunningham (αλύειν) b Phot. α 1030 Theodoridis (Ael. Dion. α 81 Erbse) c Suda α 1428 Adler 97 Phot. α 752 Theodoridis (ἀκιδνότερον) cf. Hsch. α 2399 Cunningham X. Commentary to Hesiod 98 Σ Hes. Th. 126 Di Gregorio 99 Σ Procl. Hes. Op. 304b Pertusi XI. Commentary to Alcaeus 100 Σ Alc. fr. 117b.20 Voigt in P.Oxy. XV 1788 (CLGP I.1.1 Alcaeus 4) fr. 4 101 Σ Alc. fr. 117b.40 Voigt in P.Oxy. XV 1788 (CLGP I.1.1 Alcaeus 4) fr. 4 102 Σ Lesb. inc. fr. 35.7 Voigt in P.Oxy. XXI 2299 (CLGP I.1.1 Alcaeus 18) fr. 10b XII. Commentary to Anacreon 103 P.Oxy. LIV 3722 (CLGP I.1.2.2. Anacreon 2), fr. 2.4–9  396 XIII. Commentary to Pindar Olympians 104 Σ Pi. O. 1.35b–c Drachmann 105 a Σ Pi. O. 2.29b Drachmann b Σ Pi. O. 2.29d Drachmann 106 Σ Pi. O. 2.82a Drachmann 107 Σ Pi. O. 2.140a Drachmann cf. Σ Pi. O. 2.138b Drachmann 108 a Σ Pi. O. 3.1a Drachmann b Σ Pi. O. 3.1d Drachmann 109 Σ Pi. O. 3.54a Drachmann °110 Inscr. a Pi. O. 5 Drachmann 111 a Σ Pi. O. 5.20e Drachmann b Σ Pi. O. 5.27b Drachmann 112 a Σ Pi. O. 6.55a Drachmann b Σ Pi. O. 6.55b Drachmann 113 Σ Pi. O. 6.115a Drachmann cf. Σ Pi. O. 6.115c Drachmann 114 Σ Pi. O. 6.158b–c Drachmann cf. Σ Pi. O. 6.158a Drachmann; Σ Pi. O. 6.160d Drachmann 115 Σ Pi. O. 7.34a Drachmann 116 Σ Pi. O. 7.154a Drachmann cf. Σ Pi. O. 7.153e Drachmann; Σ Pi. O. 6.154c Drachmann 117 Σ Pi. O. 7.160c Drachmann cf. Σ Pi. O. 7.159g Drachmann

A checklist of the testimonia and fragments of Didymus

118 Inscr. a Pi. O. 8 Drachmann 119 Σ Pi. O. 8.41a Drachmann cf. Σ Pi. O. 8.44c Drachmann 120 Σ Pi. O. 9.34c Drachmann 121 Σ Pi. O. 9.44a Drachmann cf. Σ Pi. O. 9.48 Drachmann; Σ Pi. O. 9.43 Drachmann 122 Σ Pi. O. 10.17c Drachmann cf. Σ Pi. O. 10.17f–i Drachmann 123 Σ Pi. O. 10.55c Drachmann 124 Σ Pi. O. 10.83a Drachmann 125 Σ Pi. O. 13.17c Drachmann 126 Σ Pi. O. 13.27a–d Drachmann 127 Σ Pi. O. 13.29b Drachmann cf. Σ Pi. O. 13.29a Drachmann; Σ Ar. Av. 1110a–b Holwerda; Suda α 578 Adler Pythians 128 Σ Pi. P. 4.44b Drachmann 129 Lact. Div. Inst. 1.22.19–27 cf. Apollod. Bibl. 1.1.6; Zen. 2.48 130 Σ Pi. P. 4.446 Drachmann 131 Σ Pi. P. 4.455d–e Drachmann cf. 133 132 Σ Pi. P. 5.34 Drachmann 133 a Σ Pi. P. 5.78a Drachmann b Σ Pi. P. 5.80a–b Drachmann cf. Σ Pi. P. 5.76b Drachmann 134 Σ Pi. P. 6.35a–b Drachmann 135 Σ Pi. P. 7.6a Drachmann 136 Σ Pi. P. 8.113c Drachmann 137 a Σ Pi. P. 9.177 Drachmann b Σ Pi. P. 9.178 Drachmann 138 Σ Pi. P. 9.207b Drachmann 139 Σ Pi. P. 10.51b Drachmann cf. Σ Pi. P. 10.46b Drachmann; Σ Pi. P. 10.56a Drachmann Nemeans 140 141 142 143 144 145 146 147

Inscr. a Pi. N. 1 Drachmann Σ Pi. N. 1.7b Drachmann Σ Pi. N. 1.36 Drachmann Σ Pi. N. 1.38 Drachmann Σ Pi. N. 1.49c Drachmann Σ Pi. N. 2.19 Drachmann = 326 Σ Pi. N. 3.1c Drachmann Σ Pi. N. 3.16b Drachmann



103

104



Coward and Prodi

148 Σ Pi. N. 3.72a–b Drachmann 149 Σ Pi. N. 4.5 Drachmann cf. Σ Pi. N. 4.3 Drachmann; Σ Pi. N. 4.6a–b Drachmann 150 Σ Pi. N. 4.14a Drachmann 151 Σ Pi. N. 4.32 Drachmann 152 Σ Pi. N. 4.95a–c Drachmann 153 Σ Pi. N. 4.151a Drachmann cf. Σ Pi. N. 4.151b; 154 154 Σ Pi. N. 4.153 Drachmann cf. 153; Σ Pi. N. 4.155a Drachmann 155 Σ Pi. N. 5.10a Drachmann cf. Σ Pi. N. 5.10b–c Drachmann 156 Σ Pi. N. 6.30 Drachmann 157 Σ Pi. N. 6.53a Drachmann 158 Σ Pi. N. 7.1a Drachmann cf. 160 159 Σ Pi. N. 7.47 Drachmann cf. Σ Pi. N. 7.46b Drachmann 160 Σ Pi. N. 7.56a Drachmann cf. 158 161 Σ Pi. N. 7.89b Drachmann cf. Σ Pi. N. 7.89c Drachmann 162 Inscr. Pi. N. 8 Drachmann cf. Σ Pi. N. 8.26 Drachmann 163 Σ Pi. N. 9.95a Drachmann cf. Σ Pi. N. 9.93b Drachmann; Σ Pi. N. 9.95b Drachmann 164 Σ Pi. N. 10.49b Drachmann 165 Σ Pi. N. 10.114a Drachmann cf. Σ Pi. N. 10.114c Drachmann 166 Inscr. a Pi. N. 11 Drachmann cf. inscr. b Pi. N. 11 Drachmann Isthmians 167 Inscr. a Pi. I. 1 Drachmann cf. Σ Pi. I. 1.52a Drachmann; 168 168 Σ Pi. I. 1.52b Drachmann cf. 167 169 Σ Pi. I. 1.60–61a Drachmann 170 Σ Pi. I. 1.67 Drachmann 171 Σ Pi. I. 2.19a Drachmann Paeans °172 a Her. Phil. 91 Palmieri b Ammon. Diff. 231 Nickau cf. Σ Pi. P. 11.5 Drachmann

A checklist of the testimonia and fragments of Didymus

Incertae sedis 173 Σ Pi. Pae. 22b.5 in P.Oxy. XXVI 2442 fr. 39 174 Σ Pi. fr. 140b.36–37 Snell–Maehler in P.Oxy. XXVI 2442 fr. 97 XIV. Commentary to Bacchylides’ Victory Odes °175 a Her. Phil. 123 Palmieri b Ammon. Diff. 333 Nickau cf. Eust. in Od. 24.47, 58, II p. 314 Stallbaum XV. Commentary to Sophocles Ajax 176 Σ S. Aj. 83 Christodoulos 177 Σ S. Aj. 408 Papageorgius Antigone 178 Σ S. Ant. 4 Papageorgius 179 Σ S. Ant. 45 Papageorgius 180 Σ S. Ant. 722 Papageorgius Oedipus at Colonus 181 Σ S. OC 155–56 Xenis 182 Σ S. OC 237 Xenis 183 Σ S. OC 763 Xenis Incertae sedis 184 Ath. 2.70c–d = 44 XVI. Commentary to Euripides Medea Cf. cod. B subscr. 185 186 187 188

Σ E. Med. 148 Schwartz Σ E. Med. 169 Schwartz Σ E. Med. 264 Schwartz Σ E. Med. 356 Schwartz



105

106



Coward and Prodi

189 Σ E. Med. 380 Schwartz 190 Σ E. Med. 737 Schwartz Andromache 191 192 193 194

Σ E. Andr. 330 Schwartz Σ E. Andr. 362 Schwartz Σ E. Andr. 885 Schwartz Σ E. Andr. 1077 Schwartz

Hecuba 195 196 197 198 199

Σ E. Hec. 13 Schwartz Σ E. Hec. 736 Schwartz Σ E. Hec. 847 Schwartz Σ E. Hec. 887 Schwartz Σ E. Hec. 1029 Schwartz

Trojan Women 200 Σ E. Tr. 1079 Schwartz = 41 cf. Hsch. α 1855 Cunningham; Phot. α 570 Theodoridis; EM s.v. αἴθρα, pp. 32–33 Gaisford Phoenician Women 201 Σ E. Ph. 751 Schwartz 202 Σ E. Ph. 1747 Schwartz Orestes 203 a Σ E. Οr. 1384 Schwartz b Phot. α 2835 Theodoridis (ἁρμάτειον μέλος) c Et. Gen. α 1203 Lasserre–Livadaras d EM α 1828 Lasserre–Livadaras cf. Hsch. α 7302 Cunningham; Et. Sym. α 1392 = MG α 1395 Lasserre–Livadaras XVII. Commentary to Ion °204 a Ath. 11.468e = 43 b Eust. in Il. 23.270, IV pp. 726–27 van der Valk cf. Hsch. δ 145 Cunningham 205 St. Byz. π 188 Billerbeck–Neuman-Hartmann (Πνύξ) = 48 cf. Hsch. π 4331, 4332 Hansen XVIII. Polemical explanations of Ion (Πρὸς Ἴωνα ἀντεξηγήσεις) °206 Ath. 14.634e–f = 46 cf. Hsch. μ 3 Cunningham

A checklist of the testimonia and fragments of Didymus

XIX. Commentary to Achaeus 207 Ath. 15.689b–c XX. Commentary to Aristophanes Acharnians 208 Et. Gud. s.v. Ἀχαιά, I p. 248 add. De Stefani cf. Σ Ar. Ach. 708c Wilson; Orion coll. 18–19 Sturz 209 Σ Ar. Ach. 1076a(II) Wilson cf. Suda χ 22 Adler 210 a Σ Ar. Ach. 1101a Wilson b Suda θ 489 Adler cf. Σ Ar. Eq. 954b(I) Jones–Wilson Clouds 211 Σ Ar. Nub. 133 Holwerda = °281 col. xi.63 ff. = 359 Wasps 212 Σ Ar. Vesp. 772b Koster 213 Σ Ar. Vesp. 1038a Koster 214 Σ Ar. Vesp. 1064a Koster cf. P.Flor. II 112 (CLGP I.1.4 Aristophanes 28) fr. C+D+E ad col. i.8 s.l. 215 Σ Ar. Vesp. 1178a Koster 216 Σ Ar. Vesp. 1309a Koster 217 Σ Ar. Vesp. 1388a Koster Peace 218 Σ Ar. Pax 758c Holwerda cf. Suda λ 85 Adler 219 Σ Ar. Pax 831d Holwerda 220 Σ Ar. Pax 932b Holwerda cf. Suda ο 88 Adler 221 Σ Ar. Pax 1254α Holwerda Birds 222 Σ Ar. Av. 13a Holwerda 223 a Σ Ar. Av. 43aα Holwerda b Σ Ar. Av. 43aβ(2) Holwerda cf. Suda κ 318 Adler



107

108



Coward and Prodi

224 Σ Ar. Av. 58 Holwerda cf. Suda ε 2807 Adler 225 Σ Ar. Av. 149a Holwerda cf. Suda λ 288 Adler 226 Σ Ar. Av. 217 Holwerda = 345 cf. Suda ε 774 Adler 227 Σ Ar. Av. 299a Holwerda 228 Σ Ar. Av. 304b Holwerda 229 Σ Ar. Av. 440 Holwerda 230 Σ Ar. Av. 530b Holwerda cf. Suda β 341 Adler 231 Σ Ar. Av. 704α Holwerda 232 Σ Ar. Av. 737fα Holwerda cf. Suda λ 714 Adler 233 Σ Ar. Av. 816a Holwerda cf. Suda χ 75 Adler 234 Σ Ar. Av. 824a Holwerda 235 Σ Ar. Av. 835a Holwerda 236 Σ Ar. Av. 836a Holwerda 237 Σ Ar. Av. 876f Holwerda cf. Suda κ 1721, 2587 Adler 238 Σ Ar. Av. 994b Holwerda 239 Σ Ar. Av. 1001b Holwerda 240 Σ Ar. Av. 1112 Holwerda 241 Σ Ar. Av. 1113a Holwerda cf. Σ Ar. Av. 1113 in P.Louvre s.n. (CLGP I.1.4 Aristophanes 4); Hsch. π 3259, 3566 Hansen; Suda π 2412 Adler 242 Σ Ar. Av. 1121a Holwerda 243 Σ Ar. Av. 1273a Holwerda cf. Suda κ 573 Adler 244 Σ Ar. Av. 1283a Holwerda 245 Σ Ar. Av. 1294 Holwerda 246 Σ Ar. Av. 1295bα Holwerda 247 Σ Ar. Av. 1297–99 Holwerda 248 Σ Ar. Av. 1363aα Holwerda 249 Σ Ar. Av. 1379b Holwerda 250 Σ Ar. Av. 1521b Holwerda cf. Suda ι 327 Adler 251 Σ Ar. Av. 1678 Holwerda 252 Σ Ar. Av. 1681b Holwerda 253 Σ Ar. Av. 1705a Holwerda Lysistrata 254 Σ Ar. Lys. 313a Hangard

A checklist of the testimonia and fragments of Didymus



109

Thesmophoriazusae 255 Σ Ar. Th. 31 Regtuit 256 Σ Ar. Th. 162a Regtuit Frogs 257 Σ Ar. Ra. 13a Chantry cf. Suda λ 808 Adler 258 Σ Ar. Ra. 41a Chantry 259 Σ Ar. Ra. 55a Chantry cf. Suda μ 1053 Adler 260 Σ Ar. Ra. 104a Chantry 261 Σ Ar. Ra. 186b Chantry cf. Suda α 1998, τ 655 Adler 262 Σ Ar. Ra. 222c Chantry 263 Σ Ar. Ra. 230aα Chantry cf. Suda κ 1415 Adler 264 Σ Ar. Ra. 704a Chantry 265 Σ Ar. Ra. 775d Chantry 266 Σ Ar. Ra. 965a Chantry cf. D.H. Lys. 32 267 Σ Ar. Ra. 970e Chantry 268 Σ Ar. Ra. 990c Chantry 269 Σ Ar. Ra. 1028aβ Chantry 270 a Σ Ar. Ra. 1305a Chantry b Ath. 14.636e Wealth 271 Σ Ar. Pl. 550e Chantry 272 Ath. 2.67d ad Ar. Pl. 720 273 Σ Ar. Pl. 1011a Chantry cf. Suda ν 430 Adler 274 a Σ Ar. Pl. 1129dα Chantry b Harp. add. 24 (ἀσκώλια) ap. Keaney 1967: 210 cf. Hsch. α 7722; Suda α 4177 Adler; Eust. in Od. 14.463, II p. 85 Stallbaum (Paus. Lex. α 161 Erbse); Σ Ar. Pl. 1129a Chantry; Σ Hes. Op. 366bis Gaisford Geras (?) °275 Her. Phil. 114 Palmieri Incertae sedis 276 P.Flor II 112 (CLGP I.1.4 Aristophanes 28) fr. C + D+E ad col. i.10 s.l. 277 P.Flor II 112 (CLGP I.1.4 Aristophanes 28) fr. C + D+E marg.

110



Coward and Prodi

XXI. Commentary to Menander °278 Et. Gud. s.v. Κορύβαντες, p. 338 Sturz XXII. Commentary to Phrynichus’ Kronos °279 Ath. 9.371f XXIII. Commentaries to the Orators °280 a Her. Phil. 122 Palmieri b Ammon. Diff. 334 Nickau c Eust. in Il. 7.341, II p. 474 van der Valk d Eust. in Od. 1.183, I p. 46 Stallbaum e Eust. in Od. 12.387, II p. 31 Stallbaum cf. Et. Gud. p. 407 Sturz XXIV. On Demosthenes/Commentary to Demosthenes (Περὶ Δημοσθένους) °281 P.Berol. inv. 9780 recto  211 = 359 Philippics 282 a Harp. π 84 Keaney (Πολύστρατος) b Phot. π 1059 Theodoridis c Suda π 2006 Adler 283 a P.Berol. 21188 fr. 1 = 355 b Harp. π 54 Keaney (Περὶ τῆς ἐν Δελφοῖς σκιᾶς) c Phot. π 787 Theodoridis d Suda π 1216 Adler cf. Zen. 6.28; App. Provv. 4.26 (CPG I p. 439) In Defence of Ctesiphon/On the Crown 284 P.Berol. inv. 9780 col. xii.35–37 285 Harp. ε 178 Keaney (ἑωλοκρασία) cf. Phot. ε 2524 Theodoridis; Suda ε 1885 Adler; Ps.-Greg. in Hermog. Μεθ., Rhet. Gr. VII/ 2 p. 1159 Walz; Tz. H. 6.877–83 (85) (Suet. Περὶ παιδιῶν A.4 Taillardat); Eust. in Od. 2.395, I pp. 105–06 Stallbaum (ibid. C.IV.101); EM s.v. ἑωλοκρασία, p. 352 Gaisford 286 Harp. ο 7 Keaney (οἰκίσκῳ) cf. Ammon. Diff. 350 Nickau; Phot. ο 93 Theodoridis; Suda οι 63 Adler; Σ D. 18.97 Dilts 287 Harp. ε 55 Keaney (ἔνθρυπτα) cf. Suda ε 1381 Adler Against Aeschines/On the False Embassy 288 Harp. ε 143 Keaney (ἐσπαθᾶτο) cf. Phot. ε 2015 Theodoridis; Suda ε 3177 Adler; Et. Gen. s.v. ἐσπαθᾶτο; Σ D. 19.43 Dilts

A checklist of the testimonia and fragments of Didymus



111

Against Meidias 289 P.Lond. I 131 = P. Lond. Lit. 179 col. i.25–27 290 a Harp. π 29 Keaney (παρασκήνια) b Phot. π 300 Theodoridis c Suda π 972 Adler cf. Σ D. 21.76 Dilts; Lex. Rhet. p. 292 Bekker 291 Harp. κ 92 Keaney (κυμβίον) = 17 cf. Phot. κ 1199 Theodoridis; Suda κ 2683 Adler; Σ D. 21.13 Dilts; Lex. Rhet. p. 274 Bekker Against Aristocrates 292 a P.Berol. inv. 5008 verso 7–24 (ὁ κάτωθεν νόμος) b Harp. ο 14 Keaney cf. Phot. ο 173 Theodoridis; Suda ο 104 Adler 293 St. Byz. κ 145 Billerbeck (Κεβρηνία) cf. Harp. κ 32 Keaney 294 Hyp. Ael. Aristid. Cimon, III p. 515 Dindorf Against Timocrates 295 a Harp. π 76 Keaney (ποδοκάκκη) b Phot. π 986 Theodoridis c Suda π 1848 Adler cf. Hsch. π 2679 Hansen 296 a Harp. ο 38 Keaney (ὅσιον) b Phot. ο 554 Theodoridis c Suda ο 687 Adler cf. Hsch. ο 1406 Cunningham; Ammon. Diff. 358 Nickau  Ptol. Diff. 394, 26; Σ Aeschin. 1.23 (55) Dilts; Lex. Rhet. p. 288 Bekker Against Aristogeiton 297 a Harp. φ 5 Keaney (φαρμακός) b Phot. φ 63 Theodoridis c Suda φ 105 Adler d EM s.v. φάρμακος, p. 788 Gaisford cf. Et. Gen. p. 299 Miller Against Stephanos 298 a Harp. ε 94 Keaney (ἐπιδιετὲς ἡβῆσαι)

112



Coward and Prodi

b Σ Aeschin. 3.122 Schultz cf. Phot. ε 1526–26b Theodoridis; Suda ε 2297 Adler; Lex. Rhet. p. 255 Bekker; Et. Gen. s.v. ἐπιδίετες ἡβῆσαι (Baldi); EM p. 359 Gaisford; Et. Sym. ε 613 Baldi; [Zonar.] I p. 809 Tittmann Against Timotheos 299 a Ath. 11.486c–d b Harp. λ 31 Keaney (Λυκιουργεῖς) c Phot. λ 449 Theodoridis d Suda λ 807 Adler cf. Harp. ι 5 Keaney (ἰερὰ τριήρης); Eust. in Il. 12.311, III p. 400 van der Valk; Lex. Rhet. p. 277 Bekker Against Nikostratos 300 a Harp. π 62/3 Keaney (περίστοιχοι) b Phot. π 776 Theodoridis c Suda π 1313 Adler Against Euboulides 301 a Harp. ε 66 Keaney (ἐξένιζε) b Phot. ε 1161 Theodoridis (ἐξένιζεν) c Suda ε 1628 Adler °302 Harp. γ 2 Keaney (γαμηλία)  °313 cf. Phot. γ 26 Theodoridis; Et. Gen. p. 74 Miller Against Neaira 303 a Harp. π 131 Keaney (πωλῶσι) b Phot. π 1592 Theodoridis c Suda π 2174 Adler Against Medon 304 Harp. δ 16 Keaney (δεκατεύειν) cf. Hsch. δ 563 Cunningham; Phot. δ 151 Theodoridis; Suda δ 182 Adler XXV. Commentary to Aeschines Against Ctesiphon 305 a Harp. π 106 Keaney (προστασία) b Phot. π 1362 Theodoridis c Suda π 2803 Adler 306 Harp. π 2 Keaney (Παιανιεῖς καὶ Παιονίδαι) cf. Hsch. π 45 Hansen; Phot. π 13 Theodoridis; Suda π 839 Adler

A checklist of the testimonia and fragments of Didymus



113

307 Harp. κ 80 Keaney (Κραυαλλίδαι) cf. Hsch. κ 3933 Cunningham; Phot. κ 1064 Theodoridis; Suda κ 2349 Adler 308 Harp. θ 34 Keaney (Θύστιον) cf. Phot. θ 286 Theodoridis; Suda θ 622 Adler XXVI. Commentary to Hyperides Against Demades 309 a Harp. ο 25 Keaney (ὀξυθύμια) b Phot. ο 377 Theodoridis c Suda ο 425 Adler cf. Hsch. ο 948 Cunningham; Phot. ο 378, 379 Theodoridis 310 a Harp. ε 35 Keaney (Ἐλευθέριος Ζεύς) b Phot. ε 597 Theodoridis c Suda ε 804 Adler d Et. Gen. s.v. ἐλευθέριος (Baldi) e EM s.v. ἐλευθέριος, p. 329 Gaisford cf. Hsch. ε 2020 Cunningham; Σ Pl. Erx. 392a Greene; Et. Sym. ε 310 Baldi Against Apellaios 311 Harp. π 124 Keaney (Πυθαέα) XXVII. Commentary to Isaeus Against Medon 312 Harp. π 9 (πανδαισία) cf. Hsch. π 319 Hansen; Phot. π 132 Theodoridis; Suda π 167 Adler; Σ Il. 24.5 Erbse; Σ Od. 9.373 Dindorf Incertae sedis °313 a Harp. γ 2 Keaney (γαμηλία)  °302 b Phot. γ 25 Theodoridis c Suda γ 48 Adler cf. Et. Gen. p. 74 Miller XXVIII. Commentary to Isocrates Areopagiticum 314 Harp. α 196 Keaney (ἀπὸ μισθωμάτων)

114



Coward and Prodi

XXIX. Commentary to Dinarchus Against Proxenos 315 Harp. μ 11 Keaney (ματρυλεῖον) cf. Phot. μ 138 Theodoridis; Suda μ 290 Adler; Et. Gen. s.v. ματρυλεῖον (Theodoridis) XXX. Commentary to Lysias Against Mixidemos 316 a Harp. π 99 Keaney (πρόπεμπτα) b Phot. π 1282 Theodoridis c Suda π 2557 Adler XXXI. Commentary on Lycurgus Against Demades 317 a Harp. τ 19 Keaney (τοὺς ἑτέρους τραγῳδοὺς ἀγωνιεῖται) b Phot. τ 404 Theodoridis c Suda τ 833 Adler On the Priestess 318 a Harp. π 44 Keaney (πέλανος) b Phot. π 538 Theodoridis c Suda π 912 Adler cf. Σ A.R. 1.1075–77a.b Wendel; Et. Gen. s.v. πελανός; EM p. 659 Gaisford; Et. Sym. cod. V fol. 146v (Gaisford); Σ E. Or. 220 Schwartz; Σ [E.] Rh. 430 Merro Against Ischyrios 319 Harp. σ 51 Keaney (στρωτήρ) Against Menesaichmos 320 a Harp. π 96 Keaney (προκώνια) b Phot. π 1254 Theodoridis c Suda π 2482 Adler cf. Erot. α 142 Nachmanson, Gal. Gloss. Hipp. XIX p. 133 Kuhn; Poll. 6.77 Bethe; Phot. π 1255, 1256 Theodoridis; EM p. 700 Gaisford

A checklist of the testimonia and fragments of Didymus

Rhetorica incertae sedis 321 a Harp. θ 15 Keaney (θέτης) b Suda θ 293 Adler cf. Phot. θ 140 Theodoridis; Lex. Rhet. p. 264 Bekker XXXII. Commentary to Herodotus 322 Σ Hdt. 5.52.2 in P.Oxy. LXV 4455 (CLGP I.2.6 Herodotus 7) col. i.12–14 XXXIII. Life of Thucydides 323 324 325 326 327

Marcellin. Vit. Thuc. 2–3 Stuart-Jones Marcellin. Vit. Thuc. 16–17 Stuart-Jones Marcellin. Vit. Thuc. 32–33 Stuart-Jones Σ Pi. N. 2.19 Drachmann = 145 a Her. Phil. 163 Palmieri b Ammon. Diff. 451 Nickau cf. Ammon. De impr. 31 Nickau; Et. Gud. pp. 201, 515 Sturz

XXXIV. On modifications of word forms (Περὶ παθῶν) Cf. Σ Il. 3.272a1, 11.160, 17.201c Erbse °328 Et. Gen. α 1312 Lasserre–Livadaras (ἄστρις) cf. EM α 1975 Lasserre–Livadaras XXXV. List of demes 329 St. Byz. τ 189 Billerbeck–Neumann-Hartmann (Τρικόρυνθον) 330 St. Byz. τ 191 Billerbeck–Neumann-Hartmann (Τρινεμεῖς) 331 St. Byz. χ 48 Billerbeck–Neumann-Hartmann (Χολαργός) XXXVI. Strange stories (Ἱστορία ξένη) Cf. Ioann. Malal. prol. p. 3 Thurn °332 a St. Byz. μ 229 Billerbeck b Georg. Sync. Chron. pp. 189–90 Mosshammer cf. Σ Il. 2.869a Erbse; Orion s.v. Μυκάλη, col. 97 Sturz; EM p. 594 Gaisford 333 a Ioann. Malal. 4.10, p. 57 Thurn b Georg. Sync. Chron. p. 189 Mosshammer 334 Ioann. Malal. 4.17, p. 62 Thurn 335 Ioann. Malal. 6.22, p. 128 Thurn 336 Macr. Sat. 5.22.10



115

116



Coward and Prodi

XXXVII. Conversations at table (Συμποσιακά) St. Byz. θ 45 Billerbeck–Zubler (Θιβαΐς) EM s.v. σκολιά, p. 718 Gaisford St. Byz. η 27 Billerbeck–Zubler (Ἠτία) Clem. Al. Strom. 1.14.61.1–2 D.L. 5.75–76 cf. Ath. 13.593e–f °342 St. Byz. μ 184 Billerbeck (Μίλητος) °343 Clem. Al. Strom. 4.19.122.4–123.1 Stählin–Früchtel

°337 °338 °339 340 °341

XXXVIII. Miscellanea (Σύμμικτα) °344 a Her. Phil. 42 Palmieri b Ammon. 117 Nickau c Et. Gud. s.v. γέρων καὶ πρεσβύτης καὶ προβεβηκώς διαφέρει, II p. 307 De Stefani XXXIX. On poets/On lyric poets (Περὶ ποιητῶν/Περὶ λυρικῶν ποιητῶν) °345 a Orion s.v. ἔλεγος, p. 58 Sturz = 226 b Et. Gen. s.v. ἐλεγεῖα (Baldi) c Et. Gud. s.v. ἐλεγεῖα, II p. 451 add. De Stefani d EM s.v. ἐλεγεῖα, p. 327 Gaisford e Et. Sym. ε 287 Baldi cf. Procl. Chr. 24–26 Severyns; Et. Gud. s.v. ἔλεγος, II p. 452 add. De Stefani 346 Σ Pl. Lg. 630a Greene cf. Harp. θ 6 Keaney °347 a Orion s.v. ὕμνος, pp. 155–56 Sturz b Et. Gen. s.v. ὕμνος ap. Grandolini 1999: 9 c EM s.v. ὕμνος, p. 777 Gaisford d EM s.v. προσῳδίαι, p. 690 Gaisford e [Zonar.] II p. 1767 Tittmann f Σ Theoc. 1.61b Wendel cf. Procl. Chr. 38–40 Severyns; Io. Sardianus in Aphth. Prog. 21.8 Rabe 348 a Cyrill. auct. s.v. παιᾶνες ap. Bühler 1967: 102 b Lex. cod. Crypt. Za. III ap. Theodoridis 1976: 56 c Et. Gud. p. 446 Sturz cf. Procl. Chr. 41 Severyns; Orion p. 133 Sturz; Et. Gen. s.v. παιᾶνες, p. 234 Miller; EM p. 657 Gaisford 349 Σ A.R. 1.972 Wendel (ἴουλος) cf. Poll. 1.38 Bethe; Σ A.R. 2.43 Wendel XL. On proverbs (Περὶ παροιμιῶν) Cf. Zen. tit. 350 Gal. De ind. 27 (ἀβυδοκόμη) = 6

A checklist of the testimonia and fragments of Didymus

351 352 353 354 355

356 357 358 359 360



117

cf. Hsch. α 225 Cunningham; Phot. α 63 Theodoridis; Syn. Β α 46 Cunningham; Eust. in Il. 2.836, I pp. 559–60 van der Valk (Paus. Lex. α 3 Erbse); Zen. 1.1 Helladios ap. Phot. Bibl. 279, p. 533b Bekker Zen. 2.31 Zen. 4.20 Hsch. λ 1041 Cunningham (Λιμοδωριεῖς) = 20 cf. Phot. λ 314 Theodoridis; Suda λ 555 Adler; [Plu.] Par. Alex. 1.34 (CPG I p. 326) a P.Berol. 21188 fr. 1 = 283 b Harp. π 54 (Περὶ τῆς ἐν Δελφοῖς σκιᾶς) c Phot. π 787 Theodoridis d Suda π 1216 Adler cf. Zen. 6.28; App. Provv. 4.26 (CPG I p. 439) Ath. 2.67d (ὄξος) = 26 cf. Hsch. ο 939 Cunningham; App. Provv. 4.29 (CPG I p. 440) Σ Pl. Cra. 3 Cufalo Apostol. 19.41 Σ Ar. Nub. 133 Holwerda = 211 = 281 col. xi.63 ff. Tz. H. 8.10–14 (159) cf. Zen. 1.83; Diogenian. 1.72; Suda α 1504, 1574 Adler; Apostol. 2.63; Σ Ar. Pax 299d Holwerda

XLI. Reply to Asklepiades on Solon’s tables of law (Περὶ τῶν ἀξόνων τῶν Σόλωνος ἀντιγραφὴ πρὸς Ἀσκληπιάδην) °361 Plu. Sol. 1.1 = BNJ 340 F1 Fragmenta incerti operis 362 St. Byz. α 80 Billerbeck (Ἀθῆναι) 363 a Syn. B α 834 Cunningham (ἀλάστωρ) b Phot. α 896 Theodoridis cf. Suda α 1082 Adler; Paus. Lex. α 61 Erbse; Harp. α 68 Keaney 364 St. Byz. α 200 Billerbeck (Ἀλεξάνδρειαι) 365 Phot. α 1179 Theodoridis (ἀμείβεται) 366 Et. Gud. s.v. Ἀμφιτρίτη, I p. 125 De Stefani cf. Hsch. α 4130 Cunningham; Et. Gen. α 739, EM α 1223, Et. Sym. α 832, MG α 831 Lasserre–Livadaras 367 Suda α 2146 Adler (ἄνδηρα) cf. Hsch. α 4704 Cunningham (ἀνδειράδες); Syn. B α 1232 Cunningham 368 a Cyrill. auct. s.v. Ἀργειοφόνφης (sic), An. Par. IV p. 179 Cramer b Et. Gen. α 1124 = α 1587 Lasserre–Livadaras (Ἀργειφόντης) c Et. Gud. I pp. 185–86 De Stefani d EM α 1741 Lasserre–Livadaras cf. Et. Sym. α 1326 = MG α 1330 Lasserre–Livadaras; Eust. in Od.5.51, I p. 198 Stallbaum 369 a Et. Gen. α 1220 Lasserre–Livadaras (ἁρπεδόεσσα) b EM α 1850 Lasserre–Livadaras cf. Et. Sym. α 1807 = MG α 1810 Lasserre–Livadaras; [Zonar.] I p. 300 Tittmann

118



Coward and Prodi

370 Et. Gud. s.v. ἄφενος, I p. 240 De Stefani cf. Ap. Soph. p. 48 Bekker; Et. Gen. α 1480, EM α 2175 Lasserre–Livadaras 371 a Epim. Hom. α 215 Dyck (Ἀφροδιτη) b Et. Gud. s.v. Ἀφροδίτη, I p. 246 De Stefani c EM α 2189 Lasserre–Livadaras cf. Et. Gen. α 1490 Lasserre–Livadaras 372 Choerob. in An. Ox. II p. 290 Cramer (βείδιοι) 373 Et. Gud. s.v. βελόνη, I p. 267 add. De Stefani 374 a Et. Gen. β 100 Lasserre–Livadaras (βεύδεα) b EM β 124 Lasserre–Livadaras cf. Et. Sym. β 86 = MG β 87 Lasserre–Livadaras 375 a Et. Gen. β 210 Lasserre–Livadaras (βουκόλος) b EM β 268 Lasserre–Livadaras c Et. Sym. β 173 = MG β 176 Lasserre–Livadaras d [Zonar.] I p. 397 Tittmann cf. Σ Il. 13.571 Erbse; Et. Gud. I p. 281 De Stefani 376 a Tz. Exeg. in Il. 1.245, 270 Papathomopoulos  379 b Eust. in Od. 1.68, I p. 21 Stallbaum cf. Tz. Exeg. in Il. 1.62 Papathomopoulos, Σ Opp. Hal. 2.32 Bussemaker 377 a Et. Gen. γ 42 Casadio (γεγωνεῖν) b Et. Gud. s.v. γεγωνῶ, II p. 301 add. De Stefani c EM p. 224 Gaisford d Et. Sym. γ 36 Baldi (γεγωνῶ) e [Zonar.] I p. 432 Tittmann (γεγωνῶ) 378 Et. Gud. s.v. γείτων, II p. 303 add. De Stefani (p. 402) 379 a Et. Gen. γ 102 Casadio (γῄδιον)  376 b EM s.v. γήδιον, p. 230 Gaisford c Et. Sym. γ 86 Baldi d Zonar. I p. 435 Tittmann cf. Et. Gen. AB γ 49 Casadio, EM p. 229 Gaisford 380 a Et. Gen. s.v. δάνειον (Baldi) b Et. Gud. s.v. δάνειον, II p. 334 add. De Stefani c EM p. 247 Gaisford d Et. Sym. δ 37 Baldi e [Zonar.] I p. 470 Tittmann 381 EM s.v. ἐνωπίδες cf. Hsch. ε 5592 Cunningham; Σ Lyc. Alex. 1176c–d Leone; EM s.v. ἐπωπίδα, p. 368 Gaisford; [Zonar.] s.v. ἐνωπίδες, I p. 732 Tittmann 382 Hsch. ε 5578 Cunningham (ἐπωβελία) cf. Harp. ε 125 Keaney; Phot. ε 1868 Theodoridis 383 a Et. Gen. s.v. ἐρείπω (Baldi) b Et. Gud. s.v. ἐρίπω, II p. 525 add. De Stefani c EM s.v. ἐρείπω, p. 372 Gaisford cf. Σ Il. 17.201c Erbse 384 a Epim. Hom. θ 2 Dyck (θνῄσκω) b EM s.v. θνήσκω, p. 452 Gaisford cf. Et. Gen. p. 159 Miller; Σ Il. 11.799a Erbse

A checklist of the testimonia and fragments of Didymus

385 a Epim. Hom. θ 33 Dyck (θρῴσκω) b EM s.v. θρῴσκων, p. 456 Gaisford c [Zonar.] II p. 1056 Tittmann 386 Eust. in Il. 18.352, IV p. 190 van der Valk cf. [Arc.] epit. Hdn. Pros. Cath. p. 265 Roussou 387 a Et. Gen. λ 183 Adler–Alpers (λιτός) b Et. Sym. cod. V (Gaisford) cf. EM p. 567 Gaisford 388 Theodos. Περὶ κλίσεως τῶν εἰς ων βαρυτόνων p. 18 Hilgard cf. St. Byz. ν 29 Billerbeck (Νέδων); Suda ν 131 Adler 389 St. Byz. ο 28 Billerbeck (Οἰνοῦς) 390 EM s.v. ὄχεσφιν, p. 645 Gaisford cf. Et. Gen. s.v. ὄχεσφι, p. 232 Miller; [Zonar.] II p. 1491 Tittmann 391 Cyrill. auct. s.v. Ποσειδῶν, An. Par. IV p. 188 Cramer cf. Σ Il. 15.192–93 Erbse 392 a EM s.v. σῴζω, p. 741 Gaisford b EM s.v. σωτήρια, p. 741 Gaisford c [Zonar.] s.v. σῶος, II p. 1703 Tittmann cf. Et. Gen. p. 277 Miller 393 a A.D. Adv. 159 Schneider b Σ Pl. Apol. 35 Cufalo c Σ Pl. Epist. 319e Greene d Suda ω 260 Adler (ὦ τάν) cf. Hsch. ω 70 Hansen–Cunningham 394 Sen. Epist. 88.37 (de patria Homeri) 395 Sen. Epist. 88.37 (de Aeneae matre vera) 396 Sen. Epist. 88.37 (libidinosior an ebriosior Anacreon vixerit)  103 397 Sen. Epist. 88.37 (an Sappho publica fuerit) 398 Hermog. Id. 2.11 Rabe 399 Clem. Al. Protr. 2.28.3 400 Clem. Al. Strom. 5.8.46.1–2 Stählin/Früchtel 401 Serv. in Aen. 4.261 Guillaumin 402 a Phot. κ 1280 Theodoridis (Κυψηλιδῶν ἀνάθημα ἐν Ὀλυμπίαι) b Suda κ 2804 Adler Fragmenta dubia 403 On Pythagorean philosophy: Clem. Al. Strom. 1.16.80.4 cf. Suda δ 875 Adler as amended by Schmidt 404 EM α 1834 Lasserre–Livadaras (ἀρνεῖα) Δυ codd: Didymus (Sylburg) or Diogenianus (Lasserre–Livadaras)? cf. Hsch. α 7339 Cunningham 405 Serv. in Aen. 3.64



119

120



Coward and Prodi

Fragmenta falsa Against the six books of Cicero’s Republic Cf. Suda τ 895 Adler 406 Amm. Marc. 22.16.16 Herodian, Commentary to Didymus’ On modifications of word forms (Περὶ παθῶν) 407 Σ Il. 3.272a1 Erbse 408 Σ Il. 11.160 Erbse 409 Σ Il. 17.201c Erbse On analogy among the Romans/On the Latin language (Περὶ τῆς παρὰ Ῥωμαίοις ἀναλογίας/ De Latinitate) Cf. Suda δ 874 Adler 410 411 412 413 414

Priscian. Inst. 1.20, II p. 15 Hertz–Keil Priscian. Inst. 8.96, II p. 445 Hertz–Keil Priscian. Inst. 11.1, II p. 548 Hertz–Keil Priscian. De fig. num. 9, III p. 408 Hertz–Keil Priscian. De fig. num. 17, III pp. 411–12 Hertz–Keil

Bulletin of the Institute of Classical Studies, 2020, 63, 121 doi: 10.1093/bics/qbaa020 Abbreviations

Abbreviations An.Ox.

J. A. Cramer, Anecdota Graeca e codd. manuscriptis bibliothecarum Oxoniensium, 4 vols, Oxford 1835–37. An.Par. J. A. Cramer, Anecdota Graeca e codd. manuscriptis bibliothecae regiae Parisiensis, 4 vols, Oxford 1839–41. BDAG F. Montanari, The Brill Dictionary of Ancient Greek, Leiden–Boston 2015. BNJ I. Worthington (ed.), Brill’s New Jacoby (). CGCG E. van Emde Boas, A. Rijksbaron, L. Huitink, and M. de Bakker, The Cambridge Grammar of Classical Greek, Cambridge 2019. CLGP G. Bastianini, M. Haslam, H. Maehler, F. Montanari, and C. Ro¨mer (eds.), Commentaria et lexica Graeca in papyris reperta, Munich–Berlin–Boston 2004–. CPG E. L. van Leutsch and F. G. Schneidewin, Corpus paroemiographorum Graecorum, 2 vols, Go¨ttingen 1839–51. FGrHist F. Jacoby, Die Fragmente der griechischen Historiker, 15 vols, Berlin–Leiden 1923–58. FGrHistCont G. Schepens et al. (eds.), Die Fragmente der grieschischen Historiker Continued, 5 vols, Leiden 1998–. FHG C. Mu¨ller, Fragmenta historicorum Graecorum, 5 vols, Paris 1841–83. FHS&G W. W. Fortenbaugh, P. M. Huby, R. W. Sharples, and D. Gutas (eds.), Theophrastus of Eresus: Sources for His Life, Writings, Thoughts, and Influence, 2 vols, Leiden–New York 1992. K. G. R. Ku¨hner, Ausfu¨hrliche Grammatik der griechischen Sprache. Zweiter Teil: Satzlehre (2nd ed.), 2 vols, rev. B. Gerth, Hannover–Leipzig 1898–1904. LSJ H. G. Liddell, R. Scott, and H. S. Jones, A Greek–English Lexicon (9th ed.), Oxford 1941, with P. G. W. Glare, Revised Supplement, Oxford 1996. PA R. Goulet (ed.), Dictionnaire des philosophes antiques, 7 vols, Paris, 1989–2018. PCG R. Kassel and C. F. L. Austin, Poetae comici Graeci, 8 vols, Berlin–New York 1983–98. PMG D. Page, Poetae melici Graeci, Oxford 1962. Rhet.Gr. C. Walz, Rhetores Graeci, 9 vols, Stuttgart and Tu¨bingen 1832–36. SH H. Lloyd-Jones and P. J. Parsons, Supplementum Hellenisticum, Berlin–New York 1983. SLG D. Page, Supplementum lyricis Graecis, Oxford 1974. TrGF Tragicorum Graecorum fragmenta, I: Didascaliae tragicae, catalogi tragicorum et tragoediarum, testimonia et fragmenta tragicorum minorum (ed. B. Snell, Go¨ttingen 19711, 19862); II: Fragmenta adespota (eds. R. Kannicht and B. Snell, 1981); III: Aeschylus (ed. S. L. Radt, 1977); IV: Sophocles (ed. S. L. Radt, 19851, 19992); V: Euripides (ed. R. Kannicht, 2 parts, 2004). Papyri and inscriptions are cited according to the Checklist of Editions of Greek, Latin, Demotic, and Coptic Papyri, Ostraca, and Tablets () and the List of Abbreviations of Editions and Works of Reference for Alphabetic Greek Epigraphy (), respectively.

C The Author(s) 2021. Published by Oxford University Press on behalf of Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected]



121

Bulletin of the Institute of Classical Studies, 2020, 63, 122–132 doi: 10.1093/bics/qbaa021 Bibliography

Bibliography Agosto, M. 2018: ‘Ἄτης ἄτερ: A note on Sophocles, Antigone 2–6’, Mnemosyne 71.6, 1–18. Alesse, F. 1997: Panezio di Rodi: testimonianze, Naples. Allen, T. W. 1929: ‘Miscellanea III’, CQ 23, 28–30. Alpers, K. 1981: Das attizistische Lexikon des Oros, Berlin–New York. ———. 2015: ‘Difficult problems in the transmission and interrelation of the Greek etymologica’, in G. A. Xenis (ed.), Literature, Scholarship, Philosophy, and History: Classical Studies in Memory of Ioannis Taifacos, Stuttgart, 293–314. Anderson, G. 2008: ‘Agaklytos (411)’, in Brill’s New Jacoby. Available from . Andolfi, I. 2019: Acusilaus of Argos’ Rhapsody in Prose: Introduction, Text, and Commentary, Berlin–Boston. Arrighetti, G. 1968: ‘Il POx XIII 1611: alcuni problemi di erudizione antica’, SCO 17, 76–98. ———. 1987: Poeti, eruditi, biografi: momenti della riflessione dei Greci sulla letteratura, Pisa. Atzert, C. 1907: De Cicerone interprete Graecorum, Göttingen. Baege, M. 1882: De Ptolemaeo Ascalonita: dissertatio inauguralis philologica, Halle. Bagordo, A. 1998: Die antiken Traktate über das Drama: mit einer Sammlung der Fragmente, Stuttgart–Leipzig. Baladié, R. 1978: Strabon: Géographie, V: Livre VIII, Paris. Barker, A. 1984: Greek Musical Writings, I: The Musician and His Art, Cambridge. ———. 1988: ‘Che cos’era la mágadis?’, in B. Gentili and R. Pretagostini (eds), La musica in Grecia, Rome–Bari, 96–107. Baumann, H. 1986: Die griechische Pflanzenwelt in Mythos, Kunst und Literatur 2, Munich. Bekker, I. 1829: Ἀριστοφάνους κωμωιδίαι. Aristophanis Comœdiæ cum scholiis et varietate lectionis, II: Versio latina. Deperditarum fabularum fragmenta. Scholia, London. Benaissa, A. 2015: ‘5202. Copy of an inscription for Apion’, in The Oxyrhynchus Papyri 79, London, 125–37. Benelli, L. 2011: ‘Beobachtungen zu einige Stellen des P.Oxy. LIV 3722: Kommentar zu Anakreon’, Aegyptus 91: 51–74. Benuzzi, F. 2019: ‘Una proposta di integrazione nel commentario ai Tassiarchi di Eupoli’, ZPE 209, 40–42. Bernsdorff, H. 2020: Anacreon of Teos: Testimonia and Fragments, 2 vols, Oxford. Berti, M. 2009: Istro il Callimacheo, I: Testimonianze e frammenti su Atene e sull’Attica, Tivoli. Bishop, C. 2019: Cicero, Greek Learning and the Making of a Roman Classic, Oxford. Bitto, G. 2012: Lyrik als Philologie: zur Rezeption hellenistischer Pindarkommentierung in den Oden des Horaz, mit einer rhetorisch-literarkritischen Analyse der Pindarscholien, Rahden. Bliquez, L. J. 1972: ‘A note on the Didymus papyrus XII.35’, CJ 67, 356–57. Boeckh, A. 1819: Pindari opera quae supersunt, II, Leipzig. Boissonade, J. F. 1819: Herodiani partitiones, London. ———. 1826: Aristophanes II, Paris. Bollansée, J. 1999: Hermippos of Smyrna. FGrHistCont IVA.3, Leiden–Boston–Cologne. Bömer, F. 1953: ‘Der commentarius: zur Vorgeschichte und literarischen Form des Schriften Caesars’, Hermes 81, 210–50. Bothe, F. 1829: Poetae scenici Graecorum, VI, Leipzig. Boudreaux, P. 1919: Le texte d’Aristophane et ses commentateurs, Paris. Braswell, B. K. 2011: ‘Didymus on Pindar’, in S. Matthaios, F. Montanari, and A. Rengakos (eds), Ancient Scholarship and Grammar: Archetypes, Concepts and Contexts. TiC Supplement 8, Berlin–New York, 181–97. ———. 2012: ‘Reading Pindar in antiquity’, MHelv 69, 12–28. ———. 2015: ‘The grammarian Chrysippos’, in M. Tziatzi, et al. (eds), Lemmata: Beiträge zum Gedenken an Christos Theodoridis, Berlin–Boston, 130–52. ———. 2017: Didymos of Alexandria, Commentary on Pindar (2nd edn), Basel. © The Author(s) 2021. Published by Oxford University Press on behalf of Institute of Classical Studies. All rights reserved. For permissions, please email: [email protected]



122

Bibliography



123

Braund, D. and , and J. Wilkins (eds) 2000: Athenaeus and His World: Reading Greek Culture in the Roman Empire, Exeter. Broggiato, M. 2001: Cratete di Mallo: i frammenti, La Spezia. ———. 2014: Filologia e interpretazione a Pergamo: la scuola di Cratete, Rome. Brunel, J. 1971: ‘La Ve Olympique et la reconstruction de Camarine en 461/460’, REA 73, 326–42. Bühler, W. 1967: ‘Beiträge zu den griechischen Lexikographen’, Quaderni dell’Istituto di Filologia Greca 2, 93–107. ———. Zenobii Athoi proverbia, I: Prolegomena, Göttingen. Calvani Mariotti, G. 1987: ‘Ricerche sulla tecnica esegetica degli scholia vetera a Pindaro’, Ricerche di filologia classica III, Pisa, 83–167. ———. 1999: ‘Οἱ προυπομνηματισάμενοι negli scoli a Pindaro’, Athenaeum 87, 51–56. ———. 2006: ‘Ἱστορία negli scholia vetera a Pindaro’, in G. Arrighetti and M. Tulli (eds), Esegesi letteraria e riflessione sulla lingua nella cultura greca, Pisa, 159–72. Calvani Mariotti, G. 2012: ‘Cheride interprete di Pindaro’, A&R n.s. 6, 150–56. Cameron, A. 2004: Greek Mythography in the Roman World, Oxford. ———. 2010: The Last Pagans of Rome, Oxford. Cannatà Fera, M. 2018: ‘Criticism of Pindar’s poetry in the scholia vetera’, in M. Ercoles et al. (eds), 2018: 233–59. Cannatà Fera, M. 2020: Pindaro, Le Nemee, Milan. Capel Badino, R. 2018: Polemone di Ilio e la Grecia: testimonianze e frammenti di periegesi antiquaria, Milan. Cappelletto, P. 2003: I frammenti di Mnasea: introduzione, testo e commento, Milan. Carawan, E. 2007: ‘Krateros the Macedonian (342)’, in Brill’s New Jacoby. Available from . Carey, C. 1981: A Commentary on Five Odes of Pindar: Pythian 2, Pythian 9, Nemean 1, Nemean 7, Isthmian 8, Salem. Carnevali, P. 1980: Didimo e Pindaro, Rome. Carnuth, O. 1869: Aristonici Peri semeion Odysseias reliquiae emendatiores, Leipzig. ———. 1875: Nicanoris Peri Odysseiakes stigmes reliquiae emendatiores, Berlin. Caroli, M. 2019: ‘P.Lond. inv. 2110 recto (= SB XX 14599): riedizione e commento’, APapyrol 31, 75–94. Cavallo, G. 1984: ‘I rotoli di Ercolano come prodotti scritti: quattro riflessioni’, S&C 8, 5–30. Chantraine, P. 1958–63: Grammaire Homérique, 2 vols, Paris. Chantry, M. 1999: Scholia in Aristophanem, III/1A: scholia vetera in Aristophanis Ranas, Groningen. ———. 2009: Scholies anciennes aux Grenouilles et au Ploutos d’Aristophanes, Paris. Chirassi, I. 1968: Elementi di culture precereali nei miti e riti greci, Rome. Christodoulou, G. A. 1977: Τὰ ἀρχαῖα σχόλια εἰς Αἴαντα τοῦ Σοφοκλέους, Athens. Chrubasik, B. 2015: ‘Aretades of Knidos (1759)’, in Die Fragmente der Griechischen Historiker Part IV. Available from . Cohn, L. 1903: ‘Didymos (8)’, in RE 5.1, 445–73. ———. 1907: ‘Epigenes (16)’, in RE 6.1, 64–65. ———. 1913: ‘Griechische Lexikographie’, in K. Brugmann (ed.), Griechische Grammatik: Lautlehre, Stammbildungs- und Flexions-lehre, Syntax, Munich, 679–730. Coker, A. 2019: ‘Galen and the language of Old Comedy: Glimpses of a lost treatise at Ind. 23b–28’, in C. Petit (ed.), 2019, 63–90. Colonna, A. 1983: Sophoclis Fabulae, III: Philoctetes, Oedipus Coloneus, Indices, Turin. Corradi, M. 2015: ‘Aristocles’, in Lexicon of Greek Grammarians of Antiquity. Available from . Coward, T. R. P. Forthcoming: Intellectual Life and Learning on Rhodes (168 BC–AD 44). Crönert, W. 1908a: review of D. Comparetti (ed.), Papiri Greco-Egizii. Papiri Fiorentini 2, Milan 1908, LZ 59, 1199–1201. ———. 1908b: ‘Zu den neuen Aristophanesscholien (P. Fior. 112)’, BPhW 28, 1390–92. Dain, A. 1954: Le ‘Philétaeros’, attribué à Hérodien, Paris. D’Alessandro, G. 2018: ‘Ammonio di Alessandria: testimonia’, SCO 64, 107–70. D’Alessio, G. B. 1997: ‘Pindar’s Prosodia and the classification of Pindaric papyrus fragments’, ZPE 118, 23–60. ———. 2000: review of Schröder 1999, BMCR 2000.01.24. David, S. et al. (eds) 2009: Traduire les scholies de Pindare, I: De la traduction au commentaire. Problèmes de méthode. DHA Supplement 2, Besançon. ———. (eds) 2015: Traduire les scholies de Pindare, II: Interprétation, histoire, spectacle. DHA Supplement 13, Besançon. Deas, H. T. 1931: ‘The scholia vetera to Pindar’, HSCPh 42, 1–78. Deichgräber, K. 1965: Die griechische Empirikerschule (2nd edn), Berlin–Zurich.

124



Bibliography

De Kreij, M. 2016: ‘From σύνδεσμοι to particulae’, in A. Bonifazi, A. Drummen, and M. de Kreij, Particles in Ancient Greek Discourse: Exploring Particle Use across Genres. Hellenic Studies 79. Volume I.2, Washington, DC. Available from . Delattre, D. 2007: Philodème de Gadara: Sur la Musique, Livre IV, 2 vols, Paris. Del Mastro, G. 2012: ‘Altri frammenti dal PHerc 1691: Filodemo, Historia Academicorum e Di III’, CErc 42, 277–92. De Marco, V. 1952: Scholia in Sophoclis Oedipum Coloneum, Rome. ———. 1961: ‘Un’orazione attribuita a Dinarco e un frammento di Istro’, PP 77, 81–92. Dickey, E. 2007: Ancient Greek Scholarship: A Guide to Finding, Reading, and Understanding Scholia, Commentaries, Lexica, and Grammatical Treatises, from Their Beginnings to the Byzantine Period, Oxford–New York. Diels, H. and W. Schubart 1904: ‘P.Berol. 9780’, in BKT I, Berlin. Dimitrijević, M. R. 1899: Studia Hesiodea, Leipzig. Dindorf, W. 1838: Aristophanis comoediae. Accedunt perditarum fabularum fragmenta, IV/3: Scholia Graeca ex codicibus aucta et emendata, Oxford. ———. 1851: Aeschyli tragoediae superstites, III: Scholia, Oxford. ———. 1852: Scholia in Sophoclis tragoedias, Oxford. ———. 1855: Scholia Graeca in Homeri Odysseam ex codicibus aucta et emendata, 2 vols, Oxford. ———. 1869: Poetarum scenicorum Graecorum Aeschyli Sophoclis Euripidis et Aristophanis fabulae superstites et perditarum fragmenta, IV (5th edn), Leipzig. Dix, T. K. 2013: “Beware of promising your library to anyone”: Assembling a private library at Rome’, in J. K. König, K. Oikonomopoulou, and G. Woolf (eds), Ancient Libraries, Cambridge, 209–34. Doody, A. 2010: Pliny’s Encyclopaedia: The Reception of Natural History, Cambridge. Dorandi, T. 1991: Filodemo, Storia dei Filosofi: Platone e l’Academia (PHerc. 1021 e 164). La Scuola di Epicuro 12 Naples. ———. 2007: Nell’officina dei classici: come lavoravano gli autori antichi, Rome. Drachmann, A. B. 1913: Scholia vetera in Pindari carmina, I: Scholia in Olympionicas, Leipzig. Dübner, F. 1842: Scholia Graeca in Aristophanem, Paris. Dué, C. (ed.) 2009: Recapturing a Homeric Legacy: Images and Insights from the Venetus A Manuscript of the Iliad, Washington, DC–Cambridge, MA. Dueck, D. 2000: Strabo of Amasia: A Greek Man of Letters in Augustan Rome, London. Dunbar, N. 1995: Aristophanes Birds, Oxford. Easterling, P. E. 2007: ‘Looking for Omphale’ in V. Jennings and A. Katsaros (eds), The World of Ion of Chios. Mnemosyne Supplement 288, Leiden–Boston, 282–92. Edelstein, L. and I. G., Kidd 1989–99: Posidonius (2nd edn), 3 vols, Cambridge. Elmsley, P. 1823: Sophoclis Oedipus Coloneus, Oxford. Elsperger, W. 1907: Reste und Spuren antiker Kritik gegen Euripides gesammelt aus den Euripidesscholien. Philologus Supplement 11.1, Leipzig. Erbse, H. 1955: review of De Marco 1952, Gnomon 27, 84–87. ———. 1959: ‘Über Aristarchs Iliasausgaben’, Hermes 87, 275–303. ———. 1969–88: Scholia Graeca in Homeri Iliadem (Scholia vetera), 7 vols, Berlin. Erdas, D. 2002: Cratero il Macedone: testimonianze e frammenti, Tivoli. Ernst, N. (ed.) 2006: Die D-Scholien zur Odyssee. Kritische Ausgabe, Köln (digital edition: http://kups.ub.uni-koeln.de/volltexte/ 2006/1831/). ———. 2006: Die D-Scholien zur Odyssee, Diss. Cologne. Available from . Esposito, E. 2012: ‘Aristophanes 3, 7, 12, 24, 26, 30, 31’, in Commentaria et lexica Graeca in papyris reperta I.1.4 (2nd edn), 41– 44, 67, 81–83, 146–48, 155–56, 219–23. ———. 2017: Tra filologia e grammatica: ricerche di papirologia e lessicografia greca, Bologna. Essler, H. 2019: ‘Copy-Paste in der Antike’, ZPE 212, 1–24. Farrell, J. 2016: ‘Ancient commentaries on Theocritus’ Idylls and Virgil’s Eclogues’, in C. S. Kraus and C. Stray (eds), Classical Commentaries: Explorations in a Scholarly Genre, Oxford, 397–418. Feine, P. 1883: ‘De Aristarcho Pindari interprete’, CPhI 2, 253–327. Ferrari, F. 1992: ‘La sigla ζη/ζ nei papiri pindarici’, SCO 42, 273–76. ———. 2006: ‘Molto rumore per nulla? L’inscriptio a all’Olimpica 5 e l’autenticità dell’ode’, Eikasmos 17, 75–78. Finglass, P. J. 2011: Sophocles Ajax, Cambridge. Flach, H. 1876: Glossen und Scholien zur hesiodischen Theogonie mit Prolegomena, Leipzig.

Bibliography



125

Fontenrose, J. 1978: The Delphic Oracle: Its Responses and Operations with a Catalogue of Responses, Berkeley. Fränkel, H. 1961: ‘Schrullen in den Scholien zu Pindars Nemeen 7 und Olympien 3’, Hermes 89, 385–97. Fraser, P. M. 1972: Ptolemaic Alexandria, 3 vols, Oxford. Gentili, B. et al. 2013: Pindaro: le Olimpiche, Milan. Gibson, C. A. 1997: ‘P.Berol inv. 5008, Didymus, and Harpocration reconsidered’, CPh 92, 375–81 ———. 2002, Interpreting a Classic: Demosthenes and His Ancient Commentators, Berkeley–Los Angeles–London. Goodwin, W. W. 1889: Syntax of the Moods and Tenses of the Greek Verb (2nd edn), London. Grandolini, S. 1999: ‘Didimo e la classificazione della poesia lirica’, GIF 51, 1–22. Grenfell, B. P. and A. S. Hunt 1919: ‘1611. Extracts from a work on literary criticism’, in The Oxyrhynchus Papyri 13, London, 127–48. Griffith, M. 1999: Sophocles Antigone, Cambridge. Griffith, R. D. 1987: ‘The hoopoe’s name (a note on Birds 48)’, QUCC 26.2, 59–63. Grisart, A. 1962: ‘Valerius Probus de Beyrouth’, Helikon 2, 379–414. Grisolia, R. 2001: Oikonomia: struttura e tecnica drammatica negli scoli antichi ai testi drammatici, Naples. Gudeman, A. 1921: ‘Scholien’, in RE IIA.1, 625–705. Guhl, C. 1969: Die Fragmente des Alexandrinischen Grammatikers Theon, Diss. Hamburg. Gunderson, E. 2009: Nox Philologiae: Aulus Gellius and the Fantasy of the Roman Library, Madison, WI. Hadjimichael, T. A. 2019: The Emergence of the Lyric Canon, Oxford. Harding, P. 2006: Didymos: On Demosthenes, Oxford. Harris, E. M. 1989: ‘More chalcenteric negligence’, CPh 84, 36–44. Haslam, M. 1998: ‘P.Oxy. 4455’, in The Oxyrhynchus Papyri 65, London, 55–59. Hatzimichali, M. 2006: Greek Scholarship of the Early Imperial Period: The Work of Pamphilus of Alexandria, Diss. Oxford. Havekoss, J. 1960: Untersuchungen zu den Sophokles-Scholien, Diss. Hamburg. Heine Nielsen, T. 2004: ‘Triphylia’, in M. H. Hansen and T. Heine Nielsen (eds), An Inventory of Archaic and Classical Poleis, Oxford. Hemmerdinger, B. 1959: ‘Deux notes papyrologiques’, REG 72, 106–09. Henry, W. B. 2005: Pindar’s Nemeans: A Selection, Munich–Leipzig. Henderson, J. (ed.) 1987: Aristophanes, Lysistrata, Oxford. Herington, C. J. 1972: The Older Scholia on the Prometheus Bound, Leiden. Hiller, E. 1869: ‘Der grammatiker Pius und die ἀπολογίαι πρὸς τὰς ἀθετήσεις Ἀριστάρχου’, Philologus 28, 86–115. Holford-Strevens, L. 2003: Aulus Gellius: An Antonine Scholar and His Achievement (2nd edn), Oxford. Holwerda, D. 1991: Scholia in Aristophanem, II/3: Scholia vetera et recentiora in Aristophanis Aves, Groningen. Horn, E. 1883: De Aristarchi studiis Pindaricis, Greifswald. Howley, J. A. 2018: Aulus Gellius and Roman Reading Culture: Text, Presence, and Imperial Knowledge in the Noctes Atticae, Cambridge. Ippolito, A. 2006: ‘Aretades’, in Lexicon of Greek Grammarians of Antiquity. Available from . Irigoin, J. 1952: Histoire du texte de Pindare, Paris. Jacob, C. 2000: ‘Athenaeus the librarian’, in Braund and Wilkins (eds) 2000: 85–110. ———. 2013: The Web of Athenaeus. Hellenic Studies 61, Washington, DC. Jacoby, F. 1949: Atthis: The Local Chronicles of Ancient Athens, Oxford. ———. 1954: Die Fragmente der griechischen Historiker IIIb (Supplement). A Commentary on the Ancient Historians of Athens, 2 vols, Leiden. ———. 1955: Die Fragmente der griechischen Historiker IIIb. Kommentar zu Nr. 297–607, 2 vols, Leiden. Janko, R. 2011: Philodemus, On Poems Books 3 and 4: With the Fragments of Aristotle, On Poets, Oxford. Jebb, R. C. 1900: Sophocles: The Oedipus Coloneus, Cambridge. Jones, D. M., and N. G. Wilson (eds) 1969: Scholia in Aristophanem, Pars I: Prolegomena de comoedia, Scholia in Acharnenses, Equites, Nubes, Fasc. II: Scholia vetera in Aristophanis Equites, Scholia Tricliniana in Aristophanis Equites, Groningen. Jones, N. F. 2010: ‘Autokleides (353)’, in Brill’s New Jacoby. Available from . ———. 2015: ‘Androtion of Athens (324)’, in Brill’s New Jacoby. Available from . ———. 2016: ‘Philochoros of Athens (328)’, in Brill’s New Jacoby. Available from . Kaibel, G. 1887–90: Athenaei Naucratitae Dipnosophistarum libri XV, 3 vols, Leipzig.

126



Bibliography

Kakridis, F. I. 1974: Αριστοφάνους Όρνιθες: ερμενευτική έκδοση, Athens. Käppel, L. 1992: ‘Pindar, Pae. fr. 67, 68, 69, 70 + 249b Sn.-M.’, RhM n.s. 135, 44–48. Kassel, R. and C. Austin 1984: Poetae comici Graeci, III/2, Berlin–New York. Kaster, R. A. 1980: ‘Macrobius and Servius: Verecundia and the grammarian’s function’, HSCPh 84: 219–62. ———. 1988: Guardians of Language: The Grammarian and Society in Late Antiquity, Berkeley. ———. 1995: C. Suetonius Tranquillus De grammaticis et rhetoribus, Oxford. ———. 2011a: Macrobii Ambrosii Theodosii Saturnalia, Oxford. ———. 2011b: Macrobius: Saturnalia, 3 vols, Cambridge, MA–London. Keaney, J. J. 1967: ‘New fragments of Greek authors in codex Marc. Gr. 444’, TAPhA 98, 205–19. ———. 1973: ‘Alphabetization in Harpocration’s Lexicon’, GRBS 14, 415–23. ———. 1991: Harpocration: Lexeis of the Ten Orators, Amsterdam. Keulen, W. 2009: Gellius the Satirist: Roman Cultural Authority in Attic Nights. Mnemosyne Supplement 297, Leiden–Boston. Kock, T. 1894: Ausgewählte Komödien des Aristophanes, IV: Die Vögel, Berlin. König, J. 2007: ‘Fragmentation and coherence in Plutarch’s Sympotic Questions’, in J. König and T. Whitmarsh (eds), Ordering Knowledge in the Roman Empire, Cambridge, 43–68. ———. 2012: Saints and Symposiasts: The Literature of Food and the Symposium in Greco-Roman and Early Christian Culture, Cambridge. Koster, W. J. W. 1975: Scholia in Aristophanem, IA: Prolegomena de comoedia, Groningen. ———. 1978: Scholia in Aristophanem, II/1: Scholia vetera et recentiora in Aristophanis Vespas, Groningen. La Bua, G. 2019: Cicero and Roman Education: The Reception of the Speeches and Ancient Scholarship, Cambridge. Langslow, D. R. 2009: Lectures on Syntax with Special Reference to Greek, Latin, and Germanic, Oxford [transl. of Wackernagel 1926–28]. Lascaris, J. 1518: Commentarii in septem tragedias (sic) Sophoclis quae ex aliis eius compluribus iniuria temporum amissis solae superfuerunt. Σχόλια τῶν πάνυ δοκίμων εἰς τὰς σωζομένας τῶν Σοφοκλέους τραγῳδιῶν, Rome. Lasserre, F. 1989: Sappho: une autre lecture, Padua. Lefkowitz, M. R. 1985: ‘The Pindar scholia’, AJPh 106, 269–82. Lehrs, K. 1882: De Aristarchi studiis Homericis (3rd edn), Leipzig. Leigh, M. 2013: From Polypragmon to Curiosus: Ancient Concepts of Curious and Meddlesome Behaviour, Oxford. Lenfant, D. ‘Les "fragments" d’Hérodote dans les Deipnosophistes’, in D. Lenfant (ed.) 2007: 43–72. Lenfant, D. (ed.) 2007: Athénée et les fragments d’historiens: actes du colloque de Strasbourg (16–18 juin 2005), Paris. Lentini, G. 2007: ‘P.Oxy. XV 1788: Alceo o Saffo?’, in B. Palme (ed.), Akten des 23. internationalen Papyrologenkongresses (Wien 22.–28. Juli 2001), Vienna, 387–91. Lentz, A. 1867–70: Grammatici Graeci, III: Herodiani technici reliquiae, 3 vols, Leipzig. Leurini, A. 2000: Testimonia et fragmenta Ionis Chii (2nd edn), Amsterdam. Liberman, G. 1999: Alcée: fragments, 2 vols, Paris. Lightfoot, J. L. 2009: Hellenistic Collection: Philitas, Alexander of Aetolia, Hermesianax, Euphorion, Parthenius, Cambridge, MA. Linke, K. 1977: ‘Die Fragmente des Grammatikers Dionysios Thrax’, in Die Fragmente der Grammatikers Dionysios Thrax, hrsg. K. Linke; Die Fragmente der Grammatiker Tyrannion und Diokles, hrsg. W. Haas; Apions Γλῶσσαι Ὁμηρικαί, hrsg. S. Neitzel. SGLG 3, Berlin–New York, 1–77. Lloyd-Jones, H. J. and N. Wilson 1990: Sophoclea: Studies on the Text of Sophocles, Oxford. Lobel, E. 1951a: ‘2299. Sappho or Alcaeus’, in The Oxyrhynchus Papyri 21, London, 61–71. ———. 1951b: ‘Addenda’, in The Oxyrhynchus Papyri 21, London, 122–47. ———. 1961: ‘2442. Pindar, Ὕμνοι, Παιᾶνες,? other pieces’, in The Oxyrhynchus Papyri 26, London, 31–78. Lomiento, L. 2006: ‘Pindaro Olimpica V e Camarina: una nuova proposta di datazione’, in P. Pelagatti, G. Di Stefano, and L. de Lachenal (eds), Camarina 2600 anni dopo la fondazione: nuovi studi sulla città e sul territorio. Atti del convegno internazionale, Ragusa, 7 dicembre 2002/7–9 aprile 2003, Rome, 285–93. Lorimer, E. H. L. 1936–37: ‘Δίπαλτος’, ABSA 37, 172–86. Lossau, M. J. 1964: Untersuchungen zur antiken Demosthenesexegese, Bad Homburg–Berlin–Zurich. Ludwich, A. 1884–85: Aristarchs Homerische Textkritik nach den Fragmenten des Didymos, 2 vols, Leipzig. Luschnat, O. 1954: ‘Die Thukydidesscholien: zu ihrer handschriftlichen Grundlage, Herkunft und Geschichte’, Philologus 98, 14–58. Luzzatto, M. T. 2011: ‘Commentare Demostene (le strategie dell’hypomnema nel Didimo di Berlino)’, BollClass 3rd ser. 32: 3–72. MacDowell, D. M. 1971: Aristophanes Wasps, Oxford. McGill, S. 2012: Plagiarism in Latin Literature. Cambridge. McInerney, J. 1994: ‘Politicizing the past: The Atthis of Kleidemos’, ClAnt 13.1, 17–37.

Bibliography



127

McNamee, K. 2007: Annotations in Greek and Latin Texts from Egypt. ASP 45, [Atlanta]. MacPhail, J. A. 2011: Porphyry’s Homeric Questions on the Iliad: Text, Translation, Commentary, Berlin. Mader, W. 1990: Die Psaumis-Oden Pindars (O. 4 und O. 5): ein Kommentar, Innsbruck. Maisonneuve, C. 2007: ‘Les "fragments" de Xénophon dans les Deipnosophistes’, in Lenfant (ed.) 2007: 73–106. Martini, E. 1903: ‘Dikaiarchos (3)’, in RE V.1, 546–63. Massara, M. F. 2001: ‘δακτυλωτὸν ἔκπωμα’, in P. Radici Colace and M. F. Massara, Lexicon Vasorum Graecorum, IV, Pisa, 27–33. Mastronarde, D. 2017: Preliminary Studies on the Scholia to Euripides. California Classical Studies 6, Berkeley. Matelli, E. 2012: Prassifane: testimonianze e frammenti. Filosofia e grammatica in età ellenistica, Milan. Matijašić, I. 2014: ‘Timachidas di Rodi: introduzione, edizione dei frammenti, traduzione e commento’, in C. Ampolo, D. Erdas, and A. Magnetto (eds), La Gloria di Athana Lindia, ASNP s. 5, 6.1, 113–86. ———. 2018: Shaping the Canons of Ancient Greek Historiography: Imitation, Classicism, and Literary Criticism, Berlin–Boston. Matthaios, S. 2015: ‘Greek scholarship in the Imperial era and Late Antiquity’, in Matthaios, Montanari, and Rengakos (eds) 2015: I 184–296. Matthews, V. J. 1996: Antimachus of Colophon: Text and Commentary. Mnemosyne Supplement 155, Leiden. Mazzarino, S. 1966: Il pensiero storico classico, 3 vols, Bari. Meineke, A. 1860: Aristophanis comoediae, II, Leipzig. ———. 1868: Sophoclis Oedipus Coloneus cum scholiis Graecis, Berlin. Meiners, W. 1890: ‘Quaestiones ad scholia Aristophanea historica pertinentes’, Diss. Philologicae Halenses 11, 217–403. Meliadò, C. 2015: ‘Theon [1]’, in Lexicon of the Greek Grammarians of Antiquity. Available from . Merro, G. 2018: ‘Theon’s Pindaric exegesis: new materials from marginalia on papyri’, in M. Ercoles et al. (eds), 2018: 213–31. Montana, F. 1996: L’Athenaion Politeia di Aristotele negli scholia vetera ad Aristofane, Pisa–Rome. ———. 2003: ‘Symmachus [1]’, in Lexicon of the Greek Grammarians of Antiquity. Available from . ———. 2004: ‘I grammatici alessandrini nei papiri di Aristofane’, in R. Pretagostini and E. Dettori (eds), La cultura ellenistica: l’opera letteraria e l’esegesi antica, Rome, 371–84. ———. 2006: ‘L’anello mancante: l’esegesi ad Aristofane tra l’antichità e Bisanzio’, in G. Avezzù and P. Scattolin (eds), I classici greci e i loro commentatori: dai papiri ai marginalia rinascimentali, Rovereto, 17–34. ———. 2009a: ‘A hypomnema on Herodotus’ Bk. 5: notes to P.Oxy. LXV 4455, Col. I’, in F. Montanari and S. Perrone (eds), Fragments of the Past: Ancient Scholarship and Greek Papyri. TiC 1.2, 241–54. ———. 2009b: ‘Storici, filologi, storici-filologi: intersezioni nella cultura ellenistica’, in F. Gazzano, G. Ottone, and L. Santi Amantini (eds), Ingenia Asiatica: fortuna e tradizione di storici d’Asia Minore. Atti della prima giornata di studio sulla storiografia greca frammentaria, Genova, 31 maggio 2007, Tivoli, 157–81. ———. 2009c: ‘Terpandrean hypotexts in Aristophanes’, TiC 1, 36–54. ———. 2010: ‘Due note di lessico scoliastico: παρακείμενον σχόλιον e παραγράφεσθαι’, in Montana (ed.) 2010: 185–95. ———. (ed.) 2010: Aner polytropos: ricerche di filologia greca antica dedicate dagli allievi a Franco Montanari, Rome. ———. 2011: ‘The making of the Greek scholiastic corpora’, in Montanari and Pagani (eds) 2011: 105–61. ———. 2012a: La filologia ellenistica: lineamenti di una storia culturale, Pavia. ———. 2012b: ‘Aristophanes’, in Commentaria et lexica Graeca in papyris reperta I.1.4 (2nd edn), Berlin–Boston, 3–241. ———. 2014a: ‘Anything but a marginal question: On the meaning of παρακείμενον σχόλιον and παραγράφεσθαι’, in Montana and Porro (eds) 2014, 24–38. ———. 2014b: ‘Gli Omeri di Aristarco in fumo: Gal. Indol. 13’, Prometheus 3, 262–70. ———. 2015a: ‘Didymus [2] Claudius’, in Lexicon of Greek Grammarians of Antiquity. Available from . ———. 2015b: ‘Hellenistic scholarship’, in Matthaios, Montanari, and Rengakos (eds) 2015: II 60–183. ———. 2016: ‘Leggere i Greci nella Roma di Ovidio’, AevAnt n.s. 16, 105–29. ———. 2017: ‘Zetemata alessandrini negli scoli alle Rane di Aristofane: riflessioni ecdotiche’, in G. Mastromarco, P. Totaro, and B. Zimmermann (eds), La commedia attica antica: forme e contenuti, Lecce–Rovato, 195–229. ———. 2018: ‘Commentare i Poetae Comici Graeci: il progetto Kommentierung der Fragmente der griechischen Komödie di Bernhard Zimmermann’, RFIC 146, 506–26. ———. 2019: ‘Herodotus’, in Commentaria et Lexica Graeca in papyris reperta I.2.6, Berlin–Boston, 17–92. ———. 2020: ‘Hellenistic Scholarship’, in F. Montanari (ed.), History of Ancient Greek Scholarship from the Beginnings to the End of the Byzantine Age, Leiden-Boston, 132–259. Montana, F. and A. Porro (eds) 2014: The Birth of Scholiography: From Types to Texts. TiC 6.1, Berlin–New York.

128



Bibliography

Montanari, F. 1997: ‘Asklepiades (7)’, in H. Cancik and H. Schneider (eds), Der neue Pauly: Enzyklopädie der Antike, Stuttgart– Weimar, II, 92. ———. 1998: ‘Zenodotus, Aristarchus and the ekdosis of Homer’, in G. W. Most (ed.), Editing Texts/Texte edieren, Göttingen, 1–21. ———. 2000: ‘Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis’, in M. Cannatà Fera and S. Grandolini (eds), Poesia e religione in Grecia: studi in onore di G. A. Privitera, Naples, 479–86. ———. 2002: ‘Alexandrian Homeric philology: the form of the ekdosis and the variae lectiones’, in M. Reichel and A. Rengakos (eds), Epea pteroenta: Beiträge zur Homerforschung. Festschrift für Wolfgang Kullmann zum 75. Geburtstag, Stuttgart, 119–40. ———. 2003: ‘L’ekdosis di Omero e i ripensamenti di Aristarco’, in Per Paola Venini: atti della giornata di studio (Pavia, 14 maggio 1999), Pisa, 29–43. ———. 2004: ‘La filologia omerica antica e la storia del testo omerico’, in A. Bierl, A. Schmitt, and A. Willi (eds), Antike Literatur in neuer Deutung: Festschrift für Joachim Latacz, Munich–Leipzig, 127–43. ———. 2009a: ‘Ancient Scholarship on Hesiod’, in F. Montanari, A. Rengakos, and C. Tsagalis (eds), Brill’s Companion to Hesiod, Leiden–Boston, 313–42. ———. 2009b: ‘Ekdosis alessandrina: il libro e il testo’, in M. Sanz Morales and M. Librán Moreno (eds), Verae lectiones: estudios de crítica textual y edición de textos griegos, Huelva, 143–67. ———. 2009c: ‘L’esegesi antica di Eschilo da Aristotele a Didimo’, in J. Jouanna and F. Montanari (eds), Eschyle à l’aube du théâtre occidental. Entr. Hardt 55, Vandœuvres–Geneva, 379–433. ———. 2011: ‘Correcting a copy, editing a text: Alexandrian ekdosis and papyri’, in Montanari and Pagani (eds) 2011: 1–15. ———. 2013: ‘Gli storici greci e la filologia di età ellenistico-romana’, in F. Gazzano and G. Ottone (eds), Le età della trasmissione: Alessandria, Roma, Bisanzio. Atti delle Giornate di Studio sulla storiografia greca frammentaria (Genova, 29–30 maggio 2012), Tivoli, 1–32. ———. 2015a: ‘Aristarchus’ conjectures (once) again’, in M. Tziatzi (eds), Lemmata: Beiträge zum Gedenken an Christos Theodoridis, Berlin–Boston, 119–29. ———. 2015b: ‘Ekdosis: A product of the ancient scholarship’, in Matthaios, Montanari, and Rengakos (eds) 2015: II 641–72. ———. 2015c: ‘From book to edition: philology in ancient Greece’, in S. Pollock, B. A. Elman, and K. K. Chang (eds), World Philology, Cambridge, MA–London, 25–44. ———. 2016: ‘Remarks on the citations of authors and works in ancient scholarship’, TiC 8, 73–82. ———. 2018: ‘Ancient scholarship today’, in Ercoles et al. (eds) 2018: 345–54. ———. 2020: ‘Ekdosis alessandrina e collazione di copie: un problema superato’, in M. Tulli (ed.), Graziano Arrighetti e la produzione letteraria dei Greci: Atti del convegno (Pisa 9–10 gennaio 2018), Pisa–Rome, 179–99. Montanari, F., S. Matthaios, and A. Rengakos (eds) 2015: Brill’s Companion to Ancient Greek Scholarship, 2 vols, Leiden–Boston. Montanari, F. and L. Pagani (eds) 2011: From Scholars to Scholia: Chapters in the History of Ancient Greek Scholarship. TiC Supplement 9, Berlin–New York. Moretti, J.-C. 1998: ‘L’implantation du théâtre d’Argos dans un lieu plein de sanctuaires’, in A. Pariente and G. Touchais (eds), Άργος και Αργολίδα: τοπογραφία και πολεοδομία, Paris, 233–84. Most, G. W. 1986: ‘Pindar, Nem. 7, 31–36’, Hermes 114, 262–71. Murphy, T. 2004: Pliny the Elder’s Natural History: The Empire in the Encyclopaedia, Oxford. Muzzolon, M. L. 2005: ‘Aristarco negli scolii ad Aristofane’, in F. Montana (ed.), Interpretazioni antiche di Aristofane, La Spezia, 55–109. Nagy, G. 2004: Homer’s Text and Language, Urbana–Chicago. ———. 2009: Homer the Classic, Washington, DC. Negri, M. 2004: Pindaro ad Alessandria: le edizioni e gli editori, Brescia. Neitzel, S. 1977: ‘Apions Γλῶσσαι Ὁμηρικαί’, in Die Fragmente der Grammatikers Dionysios Thrax, hrsg. K. Linke; Die Fragmente der Grammatiker Tyrannion und Diokles, hrsg. W. Haas; Apions Γλῶσσαι Ὁμηρικαί, hrsg. S. Neitzel. SGLG 3, Berlin–New York, 185–328. Neri, C. 1996: ‘Poeti, filologi e patelle’, Eikasmos 7, 25–55. Nesselrath, H. G. 1990: Die attische mittlere Komödie, Berlin–New York. Novati, F. 1879: ‘Index Fabularum Aristophanis, ex Codice Ambrosiano L 39 Sup.’, Hermes 14, 461–64. ———. 1882: ‘Saggio sulle glosse aristofanesche nel lessico d’Esichio’, in Studi di filologia greca pubblicati da E. Piccolomini, I, 59–105. Nünlist, R. 2009: The Ancient Critic at Work: Terms and Concepts of Literary Criticism in Greek Scholia, Cambridge. ———. 2019: ‘Two cornerstones of Nicanor’s syntactic explanations’, RFIC 147, 395–416.

Bibliography



129

———. Forthcoming: ‘Nicanor – more than a punctuator’, in B. Beck, A. Kelly, and T. Phillips (eds), The Ancient Scholia to Homer’s Iliad: Exegesis and Interpretation. BICS 64/1 (2021). Olson, S. D. 2006–12: Athenaeus: The Learned Banqueters, 8 vols, Cambridge, MA. ———. 2016: Eupolis: Heilotes–Chrysoun genos (frr. 147–325). Translation and Commentary. Fragmenta comica 8.2, Heidelberg. ———. 2017: Eupolis: Testimonia and Aiges–Demoi (frr. 1–146). Introduction, Translation, Commentary. Fragmenta comica 8.1, Heidelberg. ———. 2018: ‘Athenaeus’ “fragments” of non-fragmentary prose authors and their implications’, AJPh 138, 423–50. ———. 2019: Athenaeus Naucratites: Deipnosophistae, IV.A: Libri XII–XV, Berlin–Boston. Pagani, L. 2007: Asclepiade di Mirlea, I frammenti degli scritti omerici, Rome. ———. 2014: ‘Through the warping glass: A reconsideration on Venetus A subscriptions and the birth of scholiography’, in Montana and Porro (eds) 2014: 39–53. ———. 2015: ‘L’eredità di Aristarco: uno scholikon agnoema nella tradizione esegetica antica (Sch. Il. 2, 111). SIFC s. 4, 13.1, 73–94. Pagani, L., and S. Perrone 2012: ‘Le ekdoseis antiche di Omero’, in G. Bastianini and A. Casanova (eds), I papiri omerici: atti del Convegno Internazionale di Studi. Firenze, 9–10 giugno 2011, Florence, 97–124. Papageorgiou, P. N. 1888: Scholia in Sophoclis tragoedias vetera, Leipzig. Parker, V. 2011: ‘Ephorus (70)’, in Brill’s New Jacoby. Available from . Paulas, J. 2012: ‘How to read Athenaeus’ Deipnosophists’, AJPh 133: 403–39. Pearson, L., and S. Stephens 1983: Didymi in Demosthenem Commenta, Stuttgart. Pellettieri, A. Forthcoming: ‘Lycophron’, in Supplementum grammaticum Graecum, III. Pelling, C. 2000: ‘Fun with fragments: Athenaeus and the historians’, in Braund and Wilkins (eds) 2000: 171–90. Perilli, L. 2017: Galeni vocum Hippocratis glossarium: interpretazione delle parole difficili di Ippocrate, Berlin–Boston. Perrone, S. 2011: ‘La tradizione papiracea della commedia attica antica’, in A. M. Andrisano (ed.), Ritmo, parola, immagine: il teatro classico e la sua tradizione, Palermo, 201–20. ———. 2018: ‘Hypsicrates’, in Lexicon of Greek Grammarians of Antiquity. Available from . Petit, C. (ed.) 2019: Galen’s Treatise Περὶ Ἀλυπίας (De indolentia) in Context: A Tale of Resilience. Studies in Ancient Medicine 52, Leiden–Boston. Pfeiffer, R. 1949–53: Callimachus, 2 vols, Oxford. ———. 1968: History of Classical Scholarship from the Beginnings to the End of the Hellenistic Age, Oxford. Phillips, T. 2013: ‘Callimachus in the Pindar scholia’, CCJ 59, 152–77. ———. 2020: ‘Historiography and ancient Pindaric scholarship’, in B. Currie and I. Rutherford (eds), The Reception of Greek Lyric Poetry in the Ancient World: Transmission, Canonization and Paratext. Mnemosyne Supplement 430, Leiden–Boston, 441–60. Piccirilli, L. 1985: Storie dello storico Tucidide: edizione, traduzione e commento delle Vite, Genoa. ———. 1998: Plutarco, La vita di Solone, a cura di M. Manfredini e L. Piccirilli (5th edn), Milan. Podlecki, A. 1985: ‘Theophrastus on history and politics’, in W. W. Fortenbaugh et al. (eds), Theophrastus of Eresus: Sources for His Life, Writings, Thought and Influence, VIII: Sources on Rhetoric and Poetics, New Brunswick, NJ, 231–50. Pontani, F. 2005: Sguardi su Ulisse: la tradizione esegetica greca all’Odissea, Rome. ———. 2007: Scholia Graeca in Odysseam, I–(IV), Rome. ———. 2018: ‘Chi’, in M. Ercoles et al. (eds), 2018, 51–60. Porciani, L. 2001: Prime forme della storiografia greca: prospettiva locale e generale nella narrazione storica, Stuttgart. Preller, L. 1838: Polemonis periegetae fragmenta, Leipzig. Prodi, E. E. 2014: review of Braswell 2013 (1st edn of Braswell 2017), BMCR 2014.04.16. ———. 2020: ‘Poem-titles in Simonides, Pindar, and Bacchylides’, in B. Currie and I. Rutherford (eds), The Reception of Greek Lyric Poetry in the Ancient World: Transmission, Canonization, and Paratext. Mnemosyne Supplement 430, Leiden– Boston, 462–515. ———. forthcoming: ‘Da hypomnemata a scholia: l’esegesi pindarica di età ellenistica e il ruolo di Didimo nella sua trasmissione’, in E. Cingano (ed.), Commentaries on Greek Texts: Problems, Methods, and Trends of Ancient and Byzantine Scholarship, Rome. Rabe, H. 1906: Scholia in Lucianum, Leipzig. Razzetti, F. 2000: ‘Aristonico e Pindaro’, in G. Arrighetti and M. Tulli (eds), Letteratura e riflessione sulla letteratura nella cultura classica: Atti del convegno di Pisa, 7–9 giugno 1999, Pisa, 253–64. ———. 2010: ‘Aristonico fra Aristarco e Didimo: alcune questioni di filologia omerica antica’, in Montana (ed.) 2010: 41–61. Reitzenstein, R. 1897: Geschichte der griechischen Etymologika, Leipzig. ———. 1907: ‘Etymologika’, in RE VI.1, 807–17.

130



Bibliography

Rengakos, A. 2012: ‘Bemerkungen zum antiken Homertext’, in M. Meier-Brügger (ed.), Homer, gedeutet durch ein großes Lexicon: Akten des Hamburger Kolloquiums vom 6.–8. Oktober 2010 zum Abschluss des Lexikons des frühgriechischen Epos, Berlin– Boston, 239–52. Rhodes, P. J. 1993 A Commentary on the Aristotelian Athenaion Politeia, Oxford (first ed. 1981). Ritter, F. 1845: ‘Das Leben des Thucydides, aus Scholien zur thucydideischen Geschichte geschöpft von Marcellinus: Quellen und geschichtlicher Werth dieser Lebensbeschreibung’, RhM 3, 321–59. Roberts, C. H. 1941: ‘2192. Letter about books’, in The Oxyrhynchus Papyri 18, London, 150–52. Roemer, A. 1908: ‘Philologie und Afterphilologie im griechischen Altertum, II: Didymus als Erklärer des Aristophanes’, Philologus 67, 366–410. Rogers, B. B. 1902: The Frogs of Aristophanes, London. ———. 1906: The Birds of Aristophanes, London. Rohde, E. 1870: De Julii Pollucis in apparatu scaenico enarrando fontibus, Leipzig. Römer, C. 2013: ‘Alcman’, in Commentaria et Lexica Graeca in papyris reperta I.1.2.1, Berlin–Boston. Ruffa, M. 2001: ‘La questione dell’autenticità dell’Olimpica 5 di Pindaro’, in M. Cannatà Fera and G. B. D’Alessio (eds), I lirici greci: forme della comunicazione e storia del testo. Atti dell’Incontro di Studi, Messina 5–6 novembre 1999, Messina, 27–45. Ruffell, I. 2014: ‘Old Comedy at Rome: Rhetorical model and satirical problem’, in S. D. Olson (ed.), Ancient Comedy and Reception, Berlin–Boston, 275–308. Ruschenbusch, E. 1966: Σόλωνος νόμοι: die Fragmente des solonischen Gesetzeswerkes mit einer Text- und Überlieferungsgeschichte, Wiesbaden. Rutherford, W. G. 1896: Scholia Aristophanica: Being Such Comments Adscript to the Text of Aristophanes as Have Been Preserved in the Codex Ravennas, I, London–New York. ———. 1905: Scholia Aristophanica, III: A Chapter in the History of Annotation, London–New York. Rzepka, J. 2011a: ‘Apellas of Pontos (266)’, in Brill’s New Jacoby. Available from . ———. 2011b: ‘Proxenos (703)’, in Brill’s New Jacoby. Available from . Sandys, J. E. 1903: A History of Classical Scholarship from the Sixth Century B.C. to the End of the Middle Ages, Cambridge. Scattolin, P. 2013: ‘Tra Didimo ed Esichio: tre casi di tradizione indiretta dell’ Edipo a Colono (vv. 312, 390, 900)’, Prometheus 39, 25–43. Schauenburg, A. 1881: De Symmachi in Aristophanis interpretatione subsidiis, Halle. Schironi, F. 2015: ‘Aristarchus’ work in progress: What did Aristonicus and Didymus read of Aristarchus?’, CQ n.s. 65, 609–27. ———. 2018: The Best of the Grammarians: Aristarchus of Samothrace on the Iliad, Ann Arbor. ———. Forthcoming: ‘Aristarchus in his own words? What his ‘most secure’ fragments can tell us about Aristarchus’ commentaries and their transmission’, in B. Beck, A. Kelly, and T. Phillips (eds), The Ancient Scholia to Homer’s Iliad: Exegesis and Interpretation. BICS 62/1 (2021). Schlunk, R. R. 1974: The Homeric Scholia and the Aeneid: A Study of the Influence of Ancient Homeric Literary Criticism on Virgil, Ann Arbor. Schmidt, Martin 1997: ‘Variae lectiones oder Parallelstellen: was notierten Zenodot und Aristarch zu Homer?’, ZPE 115, 1–12. Schmidt, Moritz 1854: Didymi Chalcenteri grammatici Alexandrini fragmenta quae supersunt omnia, Stuttgart. Schnee, R. 1879: Ein Beitrag zur Kritik der Aristophanesscholien, Berlin. Schneider, O. 1838: De veterum in Aristophanem scholiorum fontibus commentatio, Stralsund. Schneidewin, F. W. 1853: review of Dindorf 1852, Neue Jahrbücher für Philologie und Pädagogik 67, 497–510. Schrader, H. 1880–82: Porphyrii Quaestionum Homericarum ad Iliadem pertinentium reliquias, 2 vols, Leipzig. Schröder, S. 1999: Geschichte und Theorie der Gattung Paian, Stuttgart–Leipzig. Severyns, A. 1938: Recherches sur la Chrestomathie de Proclus, II: Le codex 239 de Proclus, Liège. Sickinger, J. P. 2007: ‘Didymos von Alexandria (340)’, in Brill’s New Jacoby. Available from . ———. 2016: ‘Didymos von Alexandria (340)’ (2nd edn), in Brill’s New Jacoby. Available from . Slater, W. J. 1986: Aristophanis Byzantii fragmenta, Berlin–New York. ———. 1989: ‘Problems in interpreting scholia on Greek texts’, in J. N. Grant (ed.), Editing Greek and Latin Texts, New York, 37–61. Smith, O. L. 1982: Scholia Graeca in Aeschylus quae exstant omnia, II.2: Scholia in Septem adversus Thebas, Leipzig. Snell, B. 1961: Bacchylidis carmina cum fragmentis (8th edn), Leipzig. Sommerstein, A. H. 1987: The Comedies of Aristophanes, VI: Birds, Warminster. ———. 1999: The Comedies of Aristophanes, IX: Frogs (2nd edn), Warminster.

Bibliography



131

———. 2008: Aeschylus, I: Persians, Seven against Thebes, Suppliants, Prometheus Bound, Cambridge, MA–London. Soubiran, J. 1972: Cicéron: Aratea, fragments poétiques, Paris. Stama, F. 2014: Frinico: introduzione, traduzione e commento, Heidelberg. Steinhausen, J. 1910: Κωμωιδούμενοι: de grammaticorum veterum studiis ad homines in comoedia Attica irrisos pertinentibus, Bonn. Stevens, A. 2007: ‘Ion of Chios: Tragedy as commodity at the Athenian exchange’, in V. Jennings and A. Katsaros (eds), The World of Ion of Chios. Mnemosyne Supplement 288, Leiden–Boston, 241–65. Stroppa, M. 2009: ‘Some remarks regarding commentaries on codex from Late Antiquity’, in F. Montanari and S. Perrone (eds), Fragments of the Past: Ancient Scholarship and Greek Papyri. TiC 1.2, 298–327. Susanetti, D. 2012: Sofocle: Antigone, Rome. Susemihl, F. 1891–92: Geschichte der griechischen Litteratur in der Alexandrinerzeit, 2 vols, Leipzig. Süss, W. 1959: Die Frösche des Aristophanes mit ausgewählten antiken Scholien (2nd edn), Berlin. Swiggers, P. and A. (eds) Wouters 2002: Grammatical Theory and Philosophy of Language in Antiquity, Leuven–Paris–Sterling, VA. Taub, L. 2017: Science Writing in Greco-Roman Antiquity, Cambridge. Theodoridis, C. 1976: Die Fragmente der Grammatikers Philoxenos, Berlin–New York. ———. 2013: Photii Patriarchae Lexicon, III: Ν–Φ, Berlin. Thompson, D. W. 1940: ‘Aristophanes: Birds 1122’, CR 54, 188. Too, Y. L. 2000: ‘The walking library: The performance of cultural memories’, in Braund and Wilkins (eds) 2000: 111–23. Tosi, R. 1994: ‘La lessicografia e la paremiografia in età alessandrina ed il loro sviluppo successivo’, in F. Montanari (ed.), La philologie grecque à l’époque hellénistique et romaine. Entr. Hardt 40, Vandœuvres–Geneva, 143–97. ———. 2003: ‘Osservazioni sulla tradizione indiretta dell’Edipo a Colono’ in G. Avezzù (ed.), Il dramma sofocleo: testo, lingua, interpretazione, Stuttgart–Weimar, 357–69. Trendelenburg, A. 1867: Grammaticorum Graecorum de arte tragica iudiciorum reliquiae, Bonn. Trojahn, S. 2000: Die auf Papyri erhaltenen Kommentare zur alten Komödie, Munich–Leipzig. Turner, E. G. 1968: Greek Papyri: An Introduction, Oxford. Ucciardello, G. 2006: ‘Iulianus [1]’, in Lexicon of Greek Grammarians of Antiquity. Available from . ———. 2012: Hypomnemata papiracei e lessicografia: tra Alessandria e Bisanzio, Messina. ———. 2013: ‘Il “lessico retorico” di Eudemo di Argo e alcuni frammenti inediti di Lisia e Iperide nei marginalia del Vat. gr. 7’, MD 71, 9–44. Usener, H. 1889: Dionysii Halicarnassensis Librorum de imitatione reliquiae epistulaeque criticae duae, Bonn. Valerio, F. 2013: Ione di Chio: frammenti elegiaci e melici. Eikasmos studi 21, Bologna. van der Valk, M. 1963–64: Researches on the Text and Scholia of the Iliad, 2 vols, Leiden. van Leeuwen, J. 1896: Aristophanis Ranae cum prolegomenis et commentariis, Leiden. ———. 1902: Aristophanis Aves cum prolegomenis et commentariis, Leiden. van Straaten, M. 1962: Panaetii Rhodii fragmenta. Philosophia antiqua 5, Leiden. van Thiel, H. 1992: ‘Zenodot, Aristarch und andere’, ZPE 90, 1–32. ———. 1997: ‘Der Homertext in Alexandria’, ZPE 115, 13–36. ———. 2000: ‘Die D-Scholien der Ilias in den Handschriften’, ZPE 132, 1–62. ———. 2014a: Aristarch, Aristophanes Byzantios, Demetrios Ixion, Zenodot: Fragmente zur Ilias gesammelt, neu herausgegeben und kommentiert. 4 vols, Berlin–Boston. ———. 2014b: Scholia D in Iliadem: proecdosis aucta et correctior, Cologne. Available from . Vassilaki, E. 2009: ‘Aristarque interprète des odes siciliennes de Pindare: explication interne et explication externe’, in David et al. (eds) 2009: 121–45. ———. 2015: ‘Entre historie et légende: recherche sur les emplois des mots ἱστορία, ἱστορεῖν, ἱστοριοφράφος et ἱστορικός dans les scholies aux Olympiques de Pindare’, in David et al. (eds) 2015: 93–117. Vecchiato, S. 2018a: ‘Chrysippus “Pindaricus” fr. 19 Braswell’, MHelv 75, 138–39. ———. 2018b: ‘Note interpretative ai frammenti del commentatore pindarico Crisippo’, SemRom n.s. 7, 185–93. von Velsen, A. 1853: Tryphonis Grammatici Alexandrini fragmenta, Berlin. ———. 1881: Aristophanis Ranae, Leipzig. Villari, E. 2013: ‘Dal kepos al temenos. Note sul paesaggio dell’Attica tra locus amoenus e “giardino funerario” nell’Edipo a Colono: il bosco delle Eumenidi e l’Ur-Athen pre-sinecistica’, in E. Villari (ed.), Il paesaggio e il sacro. L’evoluzione dello spazio di culto in Grecia: interpretazioni e rappresentazioni, Genoa, 117–53. de Villoison, J. B. G. D’Ansse 1788: Homeri Ilias ad veteris codicis Veneti fidem recensita, Venice. Voigt, E.-M. 1971: Sappho et Alcaeus: fragmenta, Amsterdam.

132



Bibliography

Wackernagel, J. 1926–28: Vorlesungen über Syntax mit besonderer Berücksichtigung von Griechisch, Lateinisch und Deutsch, 2 vols, Basel. Wartelle, A. 1971: Histoire du texte d’Eschyle dans l’antiquité, Paris. Wehrli, F. 1967–69: Die Schule des Aristoteles: Texte und Kommentar (2nd edn), 10 vols, Basel. Wendel, C. 1949: ‘Pamphilos (25)’, in RE XVIII.3, 336–49. West, M. L. 1978: Hesiod: Works and Days, Oxford. ———. 1983: ‘Tragica VI’, BICS 30, 63–82. ———. 1992: Ancient Greek Music, Oxford. ———. 2001: Studies in the Text and Transmission of the Iliad, Munich–Leipzig. ———. 2007: ‘Forward into the past’, in P. J. Finglass, C. Collard, and N. J. Richardson (eds), Hesperos: Studies in Ancient Greek Poetry Presented to M. L. West on His Seventieth Birthday, Oxford, xx–xxviii. West, S. 1970: ‘Chalcenteric negligence’, CQ n.s. 20, 288–96. White, J. W. 1914: The Scholia on the Aves of Aristophanes, Boston–London. White, P. 2009: ‘Bookshops in the literary culture of Rome’, in W. A. Johnson and H. N. Parker (eds), Ancient Literacies: The Culture of Reading in Greece and Rome, Oxford, 268–87. Whitehead, D. 1997: ‘Harpocrationiana’, Eikasmos 8, 157–64. von Wilamowitz, U. 1889–95: Euripides: Herakles, 3 vols, Berlin. Willcock, M. M. 1995: Pindar: Victory Odes. Olympians 2, 7, 11; Nemean 4; Isthmians 3, 4, 7, Cambridge. Wilson, N. G. 1983: ‘Scoliasti e commentatori’, SCO 33, 83–112 [updated and repr. as ‘Scholiasts and commentators’, GRBS 47, 2007, 39–70]. ———. 2015a: Herodoti Historiae, 2 vols, Oxford. ———. 2015b: Herodotea: Studies on the Text of Herodotus, Oxford. Wolf, F. A. [1795] 1985: Prolegomena to Homer: 1795, transl. with introd. and notes by A. Grafton, G. W. Most, and J. E. G. Zetzel, Princeton. Xenis, G. A. 2010a: Scholia vetera in Sophoclis Electram. SGLG 12, Berlin–New York. ———. 2010b: Scholia vetera in Sophoclis Trachinias. SGLG 13, Berlin–New York. ———. 2018: Scholia vetera in Sophoclis Oedipum Coloneum. SGLG 18, Berlin–New York. Young, D. C. 1970: ‘Pindar Nemean 7: some preliminary remarks (vv. 1–20)’, TAPhA 101, 633–43. Ypsilanti, M. 2018: The Epigrams of Crinagoras of Mytilene: Introduction, Text, Commentary, Oxford. Zecchini, G. 2000: ‘Harpocration and Athenaeus: Historiographical relationships’, in Braund and Wilkins (eds) 2000: 153–60. Zuntz, G. 1965: An Inquiry into the Transmission of the Plays of Euripides, Cambridge.

Bulletin of the Institute of Classical Studies, 2020, 63, 133–158 doi: 10.1093/bics/qbaa023 Concordances

Concordances WORKS C.–Pr. I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI XVII XVIII XIX XX XXI XXII XXIII XXIV XXV XXVI XXVII XXVIII XIX XXX XXXI

Schmidt no. I.1

Schmidt p. 15–20

I.2 I.3 I.5 I.6 II.1 II.2 II.3 II.4

20–23 23 27–82 82–111 112–16 179–82 183–86 299–300

II.5 II.6 II.8 II.9 II.10 II.11 II.17 II.14 II.7 II.16 II.18 II.19 II.20 II.21 II.22 II.23 II.24 II.19.l

214–40 300–01 241–42 242–46 301–05 302–05 305–06 246–61 307 306–07 321 310–17 317–19 319 320 320 320 316 320

Braswell no. 1 2 3 4 6 7 8 9 10 11

12 13 15 16 17 18 24 21 22 23 25 26–37 39 40–41 42 43 44 36

C The Author(s) 2021. Published by Oxford University Press on behalf of Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected].



133

134



Concordances

C.–Pr. XXXII XXXIII XXXIV XXXV XXXVI XXXVII XXXVIII XXXIX XL XLI

Schmidt no.

Schmidt p.

Braswell no.

II.27 III.2

321–34 343–45 352–55 356–63 368–79 368–79 386–96 396–98 399

47 49 54 55 57 58 63–64 65 66

IV.1

IV.9 IV.10 IV.11

FRAGMENTS C.–Pr. 1 2 3 4 5a 5b 6 7 8 9 10 11a 11b 11c 12a 12b 13 14 15 16 17 18 19 20 21 22 23 24

Schmidt no. I.1.1 I.1.2 I.1.3 I.2.1 I.3.1

Schmidt p. 19 19–20 20 20–23 23

I.5.24 I.5.2 I.5.42

42–44 29–30 75

I.5.23

41

I.5.12 “ I.5.1 I.5.39 I.5.25 I.5.3 I.5.40 I.5.41 I.5.4 I.5.5 I.5.38 I.5.6 I.5.7 I.5.14

37 “ 28 75 44 30–31 75 75 31–34 34 73–75 34 35 38–39

Braswell

Pearson–Stephens

Concordances

C.–Pr. 25 26 27a 27b 28 29 30 31 32 33 34 35 36a 36b 36c 37 *38 39a 39b 40a 40b 41 42 43 44 45 46 47 48 *49 50 51 52 53 54 55 56a 56b 57 58 59 60 61 62  63 64

Schmidt no. I.5.8 I.5.43 I.5.47 “ I.5.20 I.5.33 I.5.9 I.5.13 I.5.46 I.5.10 I.5.11 I.5.34 I.5.31

Schmidt p. 35 76 77–79 “ 41 69 35–36 37–38 76–77 36 36–37 69–70 46–50

“ I.5.50 I.5.36 I.6.2 “ I.6.3

“ 81–82 72–73 84–86 “ 86

I.6.7 I.6.6 I.6.8 I.6.5 I.6.4 I.6.9 I.6.1 I.6.10 I.6.12

89 88 89 88 87–88 89 84 89 90–91 112–13 113 113 113–14 114 114 114–15 115 115 115 115 115 115–16 112 116

Braswell



135

Pearson–Stephens

136



C.–Pr.  65a 65b  66 67 68 69a 69b 70  71  72 73 74a 74b 75 76 77 78 79 80a 80b 80c 81 82 83 84 85 86 87a 87b 88a 88b 88c 88d 89 90a 90b 91a 91b  91c  92  93  94  95a  95b 96a 96b

Concordances

Schmidt no.

Schmidt p. 112 112 112 116 183

II.2.1

179

II.2.5 II.2.4 II.2.2 II.2.3

180 180 179 179

II.2.7 II.2.8 II.2.9 II.2.10

180–81 181 181 181

I.5.48, II.2.11

79–80, 181

I.5.48, II.2.11 II.2.12 II.2.13 II.2.14

79–80, 181 181–82 182 182

II.2.15

182

I.5.45 II.2.16 II.3.2 “

76 182 183 “

II.3.3 II.3.4 II.3.5 II.3.6 II.3.7 “ II.3.9

183–84 184 184–85 185 185 “ 185–86

Braswell

Pearson–Stephens

Concordances

C.–Pr. 96c 97 98 99 100 101 102 103 104 105a 105b 106 107 108a 108b 109  110 111a 111b 112a 112b 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133a 133b 134 135 136

Schmidt no.

Schmidt p.

II.4.1 II.4.2

300 300

II.5.2

214

II.5.3 II.5.4 II.5.5 II.5.6 II.5.6 II.5.7 II.5.1 II.5.8 “ II.5.9 “ II.5.10 II.5.10b II.5.11 II.5.12 II.5.13 II.5.15

215–16 216 216–17 217–18 217 218 214 218 “ 219 “ 219 219 219 219–20 220 221

II.5.16 II.5.17 II.5.18 II.5.19 II.5.20

221 221–22 222 222–23 223–24

II.5.21 II.5.22 II.5.23 II.5.14 II.5.24 II.5.25

224–25 225 225–26 220–21 226 226

II.5.26

226–27

II.5.27 II.5.28 II.5.29

227 227 228

Braswell

1 2a 2b 3 4 5a 5b 6 7 8a 8b 9a 9b 10 11 12 13 14a 15 16 17 18 19 20 21a 22 23 24 25 14b 26 27 28 29b 29c 30 31 32



137

Pearson–Stephens

138



C.–Pr. 137a 137b 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171  172a  172b 173 174  175a  175b 176 177 178 179

Concordances

Schmidt no. II.5.30

Schmidt p. 228

II.5.31 II.5.32 II.5.33 II.5.34 II.5.35

228 228 229 229 229

II.5.36 II.5.37 II.5.38 II.5.39 II.5.40 II.5.41

229 230 230 230 230–31 231–32

II.5.42 II.5.43 II.5.44 II.5.45 II.5.46 II.5.47

232 232 232–33 233 233 233–34

II.5.48 II.5.49 II.5.50 II.5.51 II.5.52 II.5.52 [bis] II.5.53 II.5.54

234 234 234–35 235 235–36 236 236–37 237

II.5.55 II.5.56 II.5.57 II.5.58

237 237 237 237

II.5.59

237–38

Braswell 33a 33b 34 35a 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 “

II.6 II.8.7 II.8.8 II.8.4 II.8.5

300–01 242 242 241–42 242

7 8 4 5

Pearson–Stephens

Concordances

C.–Pr. 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203  204a 204b 205  206 207 208 209 210a 210b 211 212 213 214 215 216 217 218 219 220 221 222 223a

Schmidt no. II.8.6 II.8.1 II.8.2 II.8.3 II.8.10 II.9.7 II.9.8 II.9.10 II.9.11 II.9.12 II.9.13 II.9.1 II.9.2 II.9.3 II.9.4

Schmidt p. 242 241 241 241 242 243–44 244 244 244–45 245 245 242–43 243 243 243

II.9.17 II.9.18 II.9.19 II.9.6 [bis] II.9.14 II.9.15 II.9.16

246 246 246 246 243 245 245 245–46 301–02

I.5.49 II.14.57 II.14.58

302 302–03 305–06 80–81 258 258

II.14.(5) II.14.59 II.14.60 II.14.61 II.14.62 II.14.63 II.14.64 II.14.52 II.14.53 II.14.54 II.14.55 II.14.19 II.14.20

248 259 259 259 259 259 259 257 258 258 258 251 251

Braswell 6 1 2 3 10



139

Pearson–Stephens

140



C.–Pr. 223b 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268

Concordances

Schmidt no.

Schmidt p.

II.14.21 II.14.22 II.14.23 II.14.24 II.14.25 II.14.26 II.14.27 II.14.28 II.14.29 II.14.30a II.14.31 II.14.32 II.14.33 II.14.34 II.14.35 II.14.36 II.14.37 II.14.38 II.14.39 II.14.40 II.14.41 II.14.42 II.14.43 II.14.44 II.14.45 II.14.46 II.14.47 II.14.48 II.14.49 II.14.50 II.14.56 II.14.65 II.14.66 II.14.5bis II.14.6 II.14.7 II.14.8 II.14.9 II.14.10 II.14.11 II.14.12 II.14.13 II.14.14 II.14.15 II.14.16

251 251–52 252 252 252 252–53 253 253 253 253 253 253–54 254 254 254 254 254 254 255 255 255 255 255 255 256 256 256 256 256 256 258 259–60 260, 309 248 248 248 248 248–49 249 249 249 249 249 249–50 250

Braswell

Pearson–Stephens

Concordances

C.–Pr. 269 270a 270b 271 272 273 274a 274b  275 276 277  278  279  280a  280b 280c 280d 280e  281 282a 282b 282c 283a 283b 283c 283d 284 285 286 287 288 289 290a 290b 290c 291 292a 292b 293 294 295a 295b 295c 296a 296b 296c

Schmidt no. II.14.17 II.14.18 “ II.14.1 II.14.2 II.14.3 II.14.4

Schmidt p. 250 250–51 “ 246 247 247 247–48 320

Braswell



141

Pearson–Stephens

307 306–07 321 321

II.19.a “

310–11 “

1

II.19.a “ “

311 “ “

2

II.19.b II.19.b II.19.b II.19.c

311 311–12 312 312

3 4 5 6

II.19.d “ “ “

312–13 “ “ 313

7

II.19.e

313–14

II.27.5 I.5.18 “ “ II.25 “ “

324 40 “ “ 316 “ “

8 10 “

11

12

142



C.–Pr. 297a 297b 297c 297d 298a 298b 299a 299b 299c 299d 300a 300b 300c 301a 301b 301c  302 303a 303b 303c 304 305a 305b 305c 306 307 308 309a 309b 309c 310a 310b 310c 310d 310e 311 312  313a 313b 313c 314 315 316a 316b 316c 317a

Concordances

Schmidt no. II.19.f “ “

Schmidt p. 314 “ “

Braswell

Pearson–Stephens 13

16 II.19.g “ “ “ II.19.h “ “ II.19.i

314–15 “ “ “ 315 “ “ 315

17 “

II.19.i, II.22 II.19.k “ “ II.25 II.19.m “ “

20 21

I.5.15, II.21a “ “ II.21a

315, 320 315 “ “ 315–16 316–17 “ “ 317 317–18 318–19 39, 319 “ “ 319

II.21a

319

II.21a II.21b

319 319 320 315, 320

II.19.i, II.22

II.19.l “ “

320 320 316 “ “ 321

18

19

9 14

20

15

Concordances

C.–Pr. 317b 317c 318a 318b 318c 319 320a 320b 320c 321a 321b 322 323 324 325 326 327a 327b  328 329 330 331 332a  332b 333a  333b 334 335 336  337  338  339 340  341  342  343  344a  344b  344c  345a 345b 345c 345d 345e 346  347a

Schmidt no.

I.5.17 “ “ I.5.19 “ “

Schmidt p. “ “ 40 “ “ 320–21 40 “ “ 320 “

II.27.1 II.27.2 II.27.3 II.27.6

321–22 322–23 323–24 324

II.27.7

333–34

IV.1.1 “ IV.1.2a IV.1.2 IV.1.3 IV.1.4 IV.1.5 IV.3/4.1 IV.3/4.2 IV.3/4.3 IV.3/4.4 IV.3/4.5 IV.3/4.6 IV.3/4.7

352 352–53 353 358–59 “ 360–61 359–60 361 361–62 362 370–71 371 371–72 372–74 374 374–75 375–78

IV.3/4.*8 “ IV.9.1

378–79 “ 387–88

“ “

“ “

IV.9.2 IV.9.3

388–89 389–90

Braswell

41



143

Pearson–Stephens

144



C.–Pr. 347b  347c  347d 347e 347f 348a 348b 348c 349 350 351 352 353 354 355a 355b 355c 355d 356 357 358 359 360  361 362 363a 363b 364 365 366 367 368a 368b 368c 368d 369a 369b 370 371a 371b 371c 372 373 374a 374b 375a

Concordances

Schmidt no.

Schmidt p.

“ IV.9.4

“ 390

IV.9.5 I.5.32

390 66–69

IV.10.1 IV.10.2 IV.10.3 IV.10.4

397 “ “ 397–98

IV.10.5 IV.10.5

398 398

IV.10.6 IV.10.7 IV.10.8 IV.10.9 IV.10.10 III.5.b V.1

398 “ “ “ “ 399 351 401

III.5.c

351–52

III.1.5 III.1.2

338–39 321 337–38

“ “

“ “

III.2.3 V.2

344 401

V.3 “ III.1.1

401–02 “ 337 x

V.4

402

Braswell

2

Pearson–Stephens

Concordances

C.–Pr. 375b 375c 375d 376a 376b 377a 377b 377c 377d 377e 378 379a 379b 379c 379d 380a 380b 380c 380d 380e 381 382 383a 383b 383c 384a 384b 385a 385b 385c 386 387a 387b 388 389 390 391 392a 392b 392c 393a 393b 393c 393d 394 395

Schmidt no. V.5

Schmidt p. 402

V.6 “

402 “

V.7

402

III.1.10

341

V.8 “

402 “

I.5.22 V.8a

41–42 402

III.1.3 “ III.1.7 “ III.1.8

338 “ 339–40 “ 340

III.1.11

341–42

III.1.6 V.11 III.5.a V.9 III.1.4 III.1.9 “ “ V.10 “

339 403 351 403 338 340–41 “ “ 403 “

V.10 IV.5 IV.6

403 384–86 “

Braswell



145

Pearson–Stephens

146



C.–Pr. 396 397 398 399 400 401 402a 402b 403 404 405 406 407 408 409 410 411 412 413 414

Concordances

Schmidt no. IV.7 IV.8 II.26 IV.1.6 IV.3/4.*9, V.13 379, 404–05 V.12 V.11a “ IV.[14] I.5.21 IV.2.3 IV.12 III.2.1 III.2.2 III.2.3 III.3.1 III.3.2 III.3.3 III.3.4 III.3.5

Schmidt p. “ “ 310 363

404 404 “ 380–82 41 364–66 399–400 343 344 “ 346 347 “ 347–48 348–49

Braswell

Pearson–Stephens

Concordances

Schmidt no. I.1.1 I.1.2 I.1.3 I.2.1 I.3.1 I.4 I.5.1 I.5.2 I.5.3 I.5.4 I.5.5 I.5.6 I.5.7 I.5.8 I.5.9 I.5.10 I.5.11 I.5.12 I.5.13 I.5.14 I.5.15 I.5.16 I.5.17 I.5.18 I.5.19 I.5.20 I.5.21 I.5.22 I.5.23 I.5.24 I.5.25 I.5.26 I.5.27 I.5.28 I.5.29 I.5.30 I.5.31 I.5.32 I.5.33 I.5.34 I.5.34a I.5.35 I.5.36 I.5.37

Schmidt p. 19 19–20 20 20–23 23 24–27 28 29–30 30–31 31–34 34 34 35 35 35–36 36 36–37 37 37–38 38–39 39 40 40 40 40 41 41 “ “ 42–44 44 44–45 45 45 45–46 46 46–50 66–67 69 69–70 70–71 71–72 72–73 73

C.–Pr. 1 2 3 4 5 

13a, b 9 16 19 20 22 23 25 30 33 34 12a, b 31 24 309a, b 296a, b 318a, b, c 295a, b, c 320a, b, c 28 404 381 11b 8 15

36a, c 349 29 35

*38



147

148



Concordances

Schmidt no. I.5.38 I.5.39 I.5.40 I.5.41 I.5.42 I.5.43 I.5.44 I.5.45 I.5.46 I.5.47 I.5.48 I.5.49 I.5.50 I.6.1 I.6.2 I.6.3 I.6.4 I.6.5 I.6.6 I.6.7 I.6.8 I.6.9 I.6.10 I.6.11 I.6.12 II.1 “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ II.2.1 II.2.2

Schmidt p. 73–75 75 75 75 75 76 76 76 76–77 77–79 79–80 80–81 81–82 84 84–86 86 87–88 88 88 89 89 89 89 89–90 90–91 112 112 112 112–13 113 113 113–14 114 114 114–15 115 115 115 115 115–16 116 116 116–79 179 179

C.–Pr. 21 14 17 18 10 26 88d 32 27a, b 80a, c 208 37  47 39a, b 40a 45 44 42 41 43 46 ¼  206 48 *49 63  65a, b  66 50 51 52 53 54 55 56a 57 58 59 61 62 64 67 

69b 73

Concordances

Schmidt no. II.2.3 II.2.4 II.2.5 II.2.6 II.2.7 II.2.8 II.2.9 II.2.10 II.2.11 II.2.12 II.2.13 II.2.14 II.2.15 II.2.16 II.3.1 II.3.2 II.3.3 II.3.4 II.3.5 II.3.6 II.3.7 II.3.8 II.3.9 II.3 II.4.1 II.4.2 II.5.1 II.5.2 II.5.3 II.5.4 II.5.5 II.5.6 II.5.7 II.5.8 II.5.9 II.5.10 II.5.10b II.5.11 II.5.12 II.5.13 II.5.14 II.5.15 II.5.16 II.5.17 II.5.18

Schmidt p. 179 180 180 180 180–81 181 181 181 181 181–82 182 182 182 182 183 183 183–84 184 184–85 185 185 185 185–86 186–211 300 300 214 214 215–16 216 216–17 217–18 218 218 219 219 219 219 219–20 220 220–21 221 221 221–22 222

C.–Pr. 74a  72  71 75 76 77 78 80a, c 81 82 83 86 89 68 90a, b  91c  92  93  94  95a, b 96 98 99  110 104 105 106 107 108a, b 109 111a, b 112a, b 113 114 115 116 117 129 118 120 121 122



149

150



Concordances

Schmidt no. II.5.19 II.5.20 II.5.21 II.5.22 II.5.23 II.5.24 II.5.25 II.5.26 II.5.27 II.5.28 II.5.29 II.5.30 II.5.31 II.5.32 II.5.33 II.5.34 II.5.35 II.5.36 II.5.37 II.5.38 II.5.39 II.5.40 II.5.41 II.5.42 II.5.43 II.5.44 II.5.45 II.5.46 II.5.47 II.5.48 II.5.49 II.5.50 II.5.51 II.5.52 II.5.52 [bis] II.5.53 II.5.54 II.5.55 II.5.56 II.5.57 II.5.58 II.5.59 II.6 II.7 II.7 [l. 15]

Schmidt p. 222–23 223–24 224–25 225 225–26 226 226 226–27 227 227 228 228 228 228 229 229 229 229 230 230 230 230–31 231–32 232 232 232–33 233 233 233–34 234 234 234–35 235 235–36 236 23637 237 237 237 237 237 237–38 300–01 240–41 307

C.–Pr. 123 124 126 127 128 130 131 133a 134 135 136 137a 138 139 140 141 142 145 146 147 148 149 150 152 153 154 155 156 157 159 160 161 162 163 164 165 166 168 169 170 171  172b  175b 

278

Concordances

Schmidt no. II.8.1 II.8.2 II.8.3 II.8.4 II.8.5 II.8.6 II.8.7 II.8.8 II.8.9 II.8.10 II.9.1 II.9.2 II.9.3 II.9.4 II.9.5 II.9.6 II.9.6 [bis] II.9.7 II.9.8 II.9.9 II.9.10 II.9.11 II.9.12 II.9.13 II.9.14 II.9.15 II.9.16 II.9.17 II.9.18 II.9.19 II.10 “ II.11 II.12 II.13 II.14.1 II.14.2 II.14.3 II.14.4 II.14.(5) II.14.5 II.14.6 II.14.7 II.14.8 II.14.9

Schmidt p. 241 241 241 241–42 242 242 242 242 242 242 242–43 243 243 243 243 243 243 243–44 244 244 244 244–45 245 245 245 245 245 246 246 246 301–02 302 302–05 307–08 308–10 246 247 247 247–48 248 248 248 248 248 248–49

C.–Pr. 181 182 183 178 179 180 176 177 184 191 192 193 194

200 185 186 187 188 189 190 201 202 203 197 198 199  204 205  206

271 272 273 274a 211 257 258 259 260 261



151

152



Concordances

Schmidt no. II.14.10 II.14.11 II.14.12 II.14.13 II.14.14 II.14.15 II.14.16 II.14.17 II.14.18 II.14.19 II.14.20 II.14.21 II.14.22 II.14.23 II.14.24 II.14.25 II.14.26 II.14.27 II.14.28 II.14.29 II.14.30 II.14.30a II.14.31 II.14.32 II.14.33 II.14.34 II.14.35 II.14.36 II.14.37 II.14.38 II.14.39 II.14.40 II.14.41 II.14.42 II.14.43 II.14.44 II.14.45 II.14.46 II.14.47 II.14.48 II.14.49 II.14.50 II.14.51 II.14.52 II.14.53

Schmidt p. 249 249 249 249 249 249–50 250 250 250–51 251 251 251 251–52 252 252 252 252–53 253 253 253 253 253 253 253–54 254 254 254 254 254 254 255 255 255 255 255 255 256 256 256 256 256 256 257 257 258

C.–Pr. 262 263 264 265 266 267 268 269 270a, b 222 223a 224 225 226 227 228 229 230 231 232  2 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 210 218 219

Concordances

Schmidt no. II.14.54 II.14.55 II.14.56 II.14.57 II.14.58 II.14.59 II.14.60 II.14.61 II.14.62 II.14.63 II.14.64 II.14.65 II.14.66 II.14.(67) II.14.(68) II.15 II.16 II.17 II.18 “ II.19.a “ II.19.b “ “ II.19.c II.19.d “ II.19.e II.19.f II.19.g II.19.h II.19.i “ II.19.k II.19.l II.19.m II.20 “ “ II.21.a “ II.21.b II.22 II.23

Schmidt p. 258 258 258 258 258 259 259 259 259 259 259 259–60 260 260 260 307 306–07 305–06 321 321 310–11 311 311 311 312 312 312–13 313 313–14 314 314–15 315 315 315 315 316 316–17 317 317 318–19 319 319 319 320 320

C.–Pr. 220 221 254 209 210a 212 213 214 215 216 217 255 256



278 279 207  280b, c 368 282a, b 283b, c, d, e 285 286 287 288 290a, b, c 291 292b 297a, b, c 299a, b, c, d 300a, b, c 301a, c  302 303a, b, c 316a, b, c 305a, b, c 306 307 308 309a, b, c 310a, c, e 311  313a 314 



153

154



Concordances

Schmidt no. II.24 “ “ “ “ “ II.25 “ II.26 II.27.1 II.27.2 II.27.3 II.27.4 II.27.5 II.27.6 II.27.7 III.1.1 III.1.2 III.1.3 III.1.4 III.1.5 III.1.6 III.1.7 III.1.8 III.1.9 III.1.10 III.1.11 III.2.1 III.2.2 III.2.3 “ III.3.1 III.3.2 III.3.3 III.3.4 III.3.5 III.4 III.5.a III.5.a.a III.5.a.b III.5.a.c III.5.b “ “ IV.1.1

Schmidt p. 320 320 320 320 320–21 321 315–16 316 310 321–22 322–23 323–24 324 324 324–33 333–34 337 337–38 338 338 338–39 339 339 340 340 341 341–42 343 344 344 “ 346 347 347 347–48 348–49 349 350 351 351 351–52 352 352–53 353 358–59

C.–Pr. 315 274b 321 312 319 317a, b, c 304 296a, b, c 398 323 324 325 294 326 327b 372 368b, c 383b, c 391 366 387b 386a, b 385a 392a, b, c 379b 386 407 408 409 369b 410 412 411 413 414 78 389 362 364 330 329 331  332a, b

Concordances

Schmidt no. IV.1.2 IV.1.2a IV.1.3 IV.1.4 IV.1.5 IV.1.6 IV.2.1 IV.2.2 IV.2.3 IV.3/4.1 IV.3/4.2 IV.3/4.3 IV.3/4.4 IV.3/4.5 IV.3/4.6 IV.3/4.7 IV.3/4.*8 IV.3/4.*9 IV.5 IV.6 IV.7 IV.8 IV.9.1 IV.9.2 IV.9.3 IV.9.4 IV.9.5 IV.9.(6) IV.10.1 IV.10.2 IV.10.3 IV.10.4 IV.10.5 IV.10.6 IV.10.7 IV.10.8 IV.10.(9) IV.10.(10) IV.11 IV.12 IV.[13] IV.[14] V.1 V.2 V.3

Schmidt p. 359–60 360–01 361 361 362–63 363 363 364 364–66 370–71 371 371–72 372–73 374 374 375–77 378 379 384–86 “ “ “ 387–88 388 389–90 390 390 395–96 397 397 397 397–98 398 “ 398 398 398 “ 399 399–400 400 380–82 401 401 401–02

C.–Pr. 333b 333a 334 335 336 399 90 405 337  338  339 340  341  342  343  344b, c 400 394 395 396 397  345a, c, d 346 347a, c 347d 348 

351 352 353 354 355 356 357 358 359 360  361 406 403 363a 370 371a, b, c



155

156



Concordances

Schmidt no. V.4 V.5 V.6 V.7 V.8 V.8a V.9 V.10 V.11 V.11a V.12 V.13 V.14 V.15

Pearson–Stephens 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Schmidt p. 402 402 402 402 402 402 403 403 403 404 404 404–05 405 405

C.–Pr. 282a 283b ¼ 355b 285 286 287 288 290a 291 304 292a, b 295a 296a 297a 305a 316a 298a 299a, b 300a 301a  302,  313a 303a

C.–Pr. 374b 375b 377b, c 378 380b, c 382 390 393a, b, c, d 388 402a, b 401 400

Concordances

Braswell (Pindar) 1 2a 2b 3 4 5a 5b 6 7 8a 8b 9a 9b 10 11 12 13 14a 14b 15 16 17 18 19 20 21a 21b 22 23 24 25 26 27 28 29a 29b 29c 30 31 32 33a 33b 34 35a 35b 36

C.–Pr. 104 105a 105b 106 107 108a 108b 109  110 111a 111b 112a 112b 113 114 115 116 117 129 118 119 120 121 122 123 124 125 126 127 128 130 131 132 133a 133b 134 135 136 137a 137b 138 139 140



157

158



Concordances

Braswell (Pindar) 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68

Braswell (Sophocles) 1 2 3 4 5 6 7 8 9 10

C.–Pr. 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171  172a, b

C.–Pr. 181 182 183 178 179 180 176 177 184

Bulletin of the Institute of Classical Studies, 2020, 63, 159–172 doi: 10.1093/bics/qbab012 Index

INDEX

Index of passages cited Acusilaus BNJ 2 F1 ¼ 1 Fowler Aeschylus Agamemnon 283 643 Seven against Thebes 985

Anaximenes BNJ 72 F9 BNJ 72 F13

76 74

56 n.47 37 n.21

Androtion BNJ 324 F41 BNJ 324 F49

72 76

37

Antiatticist a 68 Valente

66 n.27

Antimachus of Colophon F20.1 Matthews

44 n.74

68

Suppliants 146

37 n.21

Fragments F151 TrGF3 F462 (dub.) TrGF3

34 n.2 59 n.58

Agaclytus BNJ 411 F1 Alcaeus F286a.6 Liberman Alcman F92 PMGF ¼ F134 Calame S2 SLG Ammonius Diff. 295 Nickau Anacreon 374 PMG 376.2 PMG 450 PMG Anaxandrides F36 PCG

76 41 n.54 76 50 56 n.47 46 22 n.4 22 n.4 46-7

Apellas Ponticus BNJ 266 F6

77

Apollodorus of Athens BNJ 244 F69 BNJ 244 F127 BNJ 244 F148 BNJ 244 F268 BNJ 244 F306

26, 70 27, 70 29 69 69 n.47

[Apollodorus] Bibl. 1.9.17 Bibl. 3.6.4

73 n.73 73 n.73

Apostolius 10.65

73 n.74

App. Prov. 5.28

72 n.67

Archilochus F213 West

59

Aristarchus of Samothrace F12 Horn ¼ F11 Feine

28

C The Author(s) 2021. Published by Oxford University Press on behalf of Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected]



159

160



Index

F17 Horn ¼ F16 Feine F54–56 Feine F56 Horn ¼ F40 Feine F58 Horn ¼ F42 Feine F65 Horn ¼ F59 Feine F68 Horn ¼ F51 Feine Aristodemus BNJ 383F14 Aristophanes Acharnians 709

27 25 n.34 30 n.52 25 29 30 25-6

55 n.35

Birds 33-35 299-300 461 1121 1122-3 1297-9

37 60 36 n.14 59 59 n.57 57

Frogs 55 156 300 704 755 1304–06

58 42 n.65 42 n.65 59 60 48 n.89

Thesmophoriazusae 162

58

Wasps 394

36 n.14

Wealth 720

53-4

Fragments F233 PCG F365 PCG Aristophanes of Byzantium F381 Slater Aristotle Ath. 53.2–6 Pol. 1313b18–25 Pol. 1451a36-51b11 Didascaliae F415-62 Gigon ¼ 618-30 Rose

65 n.19 68 24 75 n.83 77 n.89 63 n.4 51n.1

On Solon’s Axones T1 Ruschenbusch Aristoxenus F100 Wehrli F101 Wehrli Asclepiades of Myrlea F8 Pagani F9 Pagani Asclepiades of Nicaea/Alexandria BNJ 339 F1 ¼ T2 Ruschenbusch Athenaeus 1.1c 1.30d-e 2.70d 2.71a 4.182d–e 8.359d–e 11.468c 11.468c-f 11.500f–502a 11.501b 11.783b–c 14.634c–637a 15.671f–672a Athenodorus of Tarsus BNJ 746 T4 ¼ F5

65 46-7 46-7 45-6 46 n.81 65, 65 n.20 89 n.38 87 40 n.41 40 n.40 48 n.85 89 n.38 43 43 45 45 n.78 44 n.74 48 n.85 48 n.88 2 n.11

Autoclides BNJ 353 F2

67-8

Bacchylides 18.49

37 n.21

Bato F5.10 PCG4

42 n.58

Bryon SH 738 Callias F*4 PCG

76 52 n.14

Callimachus F609 Pfeiffer

66

Callisthenes BNJ 124 F2 BNJ 124 F3

76 76

Index

Caucalus of Chios BNJ 38 F2 Chaeris F20 Berndt Chamaeleon On Comedy F10-11 Bagordo Choeroboscus in Theod. II 175.10 (GG IV.2) Chrysippus F2 Braswell

73 n.73 25

51n.1 38 n.27 25

Cicero Att. 2.1 [21] Att. 16.11.4 [420] Att. 16.14.4 Fam. 3.7.5 Fin. 1.6 Inv. 2.4 QF 3.4.5, 3.5.6 Tusc. 2.61

2 n.10 2 n.11 2 n.11 2 n.11 82 n.11 6 2 n.11 82 n.11

Clement of Alexandria Paed. 2.72.3

42 n.60

CLGP Alcaeus 4 (I.1.1) Alcaeus 18 Alcman 5 (I.1.2.1) Anacreon 3 (I.1.2.2) Aristophanes 1 Aristophanes 3 Aristophanes 4 Aristophanes 7 Aristophanes 12 Aristophanes 24 Aristophanes 26 Aristophanes 27 Aristophanes 28 Aristophanes 30 Aristophanes 31 Bacchylides 4 (I.1.4) Herodotus 4 (I.2.6) Herodotus 7 Craterus BNJ 342 F17

22 n.6 22 n.7 22 n.8 21-22 95 56, 95 56 56 56 56 56 49 n.96, 56, 95 55-6, 59, 72 n.68, 95 56 56 22 n.10, 95 62 63 71



161

Crates of Mallus F55 Broggiato

38 n.29

Cratinus T*40 PCG

46 n.81

Cratippus BNJ 64 F2

65 n.15

Crinagoras of Mytilene Ep. 36.4 Ypsilanti

41 n.54

Cypria F15 Bernabe´

29

Demetrius of Phalerum The Archons BNJ 228 F3 ¼ F92 SOD

65

Demetrius of Troezen SH 376 Demosthenes Or. 4.24 Or. 23.28 Or. 24.105 Or. 49.31 Or. 53.15 Or. 57.43, 69 Dicaearchus On the Dionysiac Agones F2–7 Bagordo Didymus Testimonia T1 C.-Pr. T2 C.-Pr. T3 C.-Pr. T4 C.-Pr. T5 C.-Pr. T6 C.-Pr. T7 C.-Pr. T8-10 C.-Pr. T8 C.-Pr. T9 C.-Pr. T10 C.-Pr. T12 C.-Pr. T13 C.-Pr. T14 C.-Pr.

4 75 n.82 74 57 74 74 75

51n.1

1, 4, 7, 10, 28, 80 2, 4 2, 4 2, 80 2, 4, 5, 75, 87 2, 4, 5, 82 3, 7, 80 n.2 2 4 4 4 4 4 2, 3, 4, 7, 21, 81

162



Index

Fragments  1 C.-Pr.  3 C.-Pr.  4 C.-Pr. 6–*38 C.-Pr. 8 C.-Pr. 9 C.-Pr. 10 C.-Pr. 11 C.-Pr. 12a C.-Pr. 13 C.-Pr. 14 C.-Pr. 15 ¼ 341 C.-Pr. 16 C.-Pr. 17 C.-Pr. 18 C.-Pr. 21 C.-Pr. 22 C.-Pr. 23 C.-Pr. 24 C.-Pr. 25 C.-Pr. 26 ¼ 355 C.-Pr. 27a C.-Pr. 29 C.-Pr. 30 C.-Pr. 31 C.-Pr. 32 C.-Pr. 33 C.-Pr. 35 C.-Pr. 36a C.-Pr. *38 C.-Pr.  39 C.-Pr.  39a C.-Pr.  39ab C.-Pr.  39–*49 C.-Pr. 40ab C.-Pr. 43 ¼  204a C.-Pr. 44 ¼ 184 C.-Pr. 46 ¼ 206 C.-Pr. 47 ¼  206 C.-Pr. 48 ¼ 205 C.-Pr. *49 C.-Pr. 50-68 C.-Pr. 50 C.-Pr. 56a C.-Pr. 57 C.-Pr. 59 C.-Pr.

87 n.29 66 34 n.3, 43 n.69; 50 34 52 n.9, 87 n.29, 88 n.34, 89 55, 57, 87 n.29 87 n.29 87 55 55 87 n.29 52 n.11, 87 n.29 55 55, 87 n.29, 88 54, 87 n.29 54, 87 n.29, 88 55 67 55 55 53, 87 n.29, 88 n.34 87, 87 n.29 55 55 55 88 n.34 55 55 52 n.9, 54, 55 41 n.52 4, 5, 90 68 68 34 37 n.22 44, 87 n.29 40, 50, 87 n.29, 88 n.34 87 n.29 46-8, 50 48-9, 50 41 n.52 48 n.90 12, 14, 15 n.53, 16, 16 n.66 69 12 17

62 C.-Pr. 63 C.-Pr.  65a C.-Pr. 68 C.-Pr. 69b C.-Pr.  71 C.-Pr.  72 C.-Pr. 73a C.-Pr. 75 C.-Pr. 76 C.-Pr. 77 C.-Pr. 78 C.-Pr. 80b C.-Pr. 81 C.-Pr. 82 C.-Pr. 83 C.-Pr. 84 C.-Pr. 85 C.-Pr. 86 C.-Pr. 89 C.-Pr. 90a C.-Pr.  91a C.-Pr.  91b C.-Pr.  93a C.-Pr.  94 C.-Pr. 95b C.-Pr. 96b C.-Pr. 98-99 C.-Pr. 100 C.-Pr. 101 C.-Pr. 102 C.-Pr. 103 C.-Pr. 104 C.-Pr. 105 C.-Pr. 105ab C.-Pr. 105b C.-Pr. 107 C.-Pr. 108a C.-Pr. 108b C.-Pr. 109 C.-Pr.  110 C.-Pr. 111 C.-Pr. 111a C.-Pr. 111b C.-Pr. 112a C.-Pr. 114 C.-Pr. 116 C.-Pr. 

12 11 11 11 9 9 9 9 9 9 9 9 10 10 10 9, 70 5, 10 9 9 10 10 10 10 10 10 10 10 8 22 22 22 21-2 26, 70 5 71 28 30 28 28 27, 70 30 5, 28 27 27 27, 41 n.46, 42, 70 26, 71 70

Index

117 C.-Pr. 119 C.-Pr. 120 C.-Pr. 122 C.-Pr. 123 C.-Pr. 125 C.-Pr. 126 C.-Pr. 127 C.-Pr. 132 C.-Pr. 135 C.-Pr. 136 C.-Pr. 137a-b C.-Pr. 139 C.-Pr. 140 C.-Pr. 141 C.-Pr. 142 C.-Pr. 143 C.-Pr. 144 C.-Pr. 145 ¼ 326 C.-Pr. 146 C.-Pr. 150 C.-Pr. 151 C.-Pr. 152 C.-Pr. 153 C.-Pr. 154 C.-Pr. 155 C.-Pr. 157 C.-Pr. 158 C.-Pr. 159 C.-Pr. 160 C.-Pr. 161 C.-Pr. 162 C.-Pr. 163 C.-Pr. 164 C.-Pr. 165 C.-Pr. 166 C.-Pr. 170 C.-Pr. 171 C.-Pr.  172 C.-Pr. 173-4 C.-Pr.  175 C.-Pr. 176 C.-Pr. 177 C.-Pr. 178 C.-Pr. 179 C.-Pr. 180 C.-Pr. 182 C.-Pr.

71 42 31 30, 32 n.62 30, 60 n.66 71 33, 70 71 26, 70 30, 60 n.66 27, 70 27, 70 33 5, 6, 26, 71 31 31 31 25, 32 64 41 n.46 33 27 28 30, 33 33 31 6, 28, 70 5, 24 n.27, 25 30, 32 25 n.34 31 71 71 32 29, 60 n.66 30 32 n.62 31 23, 31, 71 24 22 35, 50 36-7 36 n.9 41 nn.46-47, 50 35 n.5 39, 41 n.47

183 C.-Pr. 184 ¼ 44 C.-Pr. 185–203d C.-Pr. 190 C.-Pr. 197 C.-Pr. 198 C.-Pr. 201 C.-Pr. 202 C.-Pr.  204a ¼ 43 C.-Pr. 205 ¼ 48 C.-Pr.  206 ¼ 47 C.-Pr. 207 C.-Pr. 214 C.-Pr. 215 C.-Pr. 222 C.-Pr. 225 C.-Pr. 227 C.-Pr. 230 C.-Pr. 233 C.-Pr. 236 C.-Pr. 237 C.-Pr. 241 C.-Pr. 242 C.-Pr. 245 C.-Pr. 246 C.-Pr. 247 C.-Pr. 249 C.-Pr. 249a C.-Pr. 254 C.-Pr. 256 C.-Pr. 257 C.-Pr. 259 C.-Pr. 263 C.-Pr. 264 C.-Pr. 265 C.-Pr. 266 C.-Pr. 267 C.-Pr. 268 C.-Pr. 269 C.-Pr. 270ab C.-Pr. 270b C.-Pr. 271 C.-Pr.  275 C.-Pr.  278 C.-Pr.  279 C.-Pr.  281 C.-Pr.



163

38, 50 40, 50, 87 n.29, 88 n.34 34 n.1 38 n.28 36 n.9, 38 n.28 73 34 n.2 38 n.28 44, 50, 87 n.29 48-9, 50 46-8, 50, 87 n.29 34 n.2, 87 n.29 59 58 52 n.12, 59-60 56-7 52 n.13, 60 52 n.10, 56 61 61 52 n.14, 57 n.51 56 59 52 n.14, 57 n.51 57 n.51 52 n.9, n.14, 57-8 52 n.10, n.13, 57 n.51 57 57 n.51, 71 6, 59 72 6, 58 57 34 n.2, 59 60 34 n.2, 57 n.51 59, 72 57 n.51 34 n.2, 52 n.11 48 n.89 87 n.29 72 53 53 n.17 53 n.17, 87 n.29 3, 5, 6, 19 n.90, 24, 26, 42, 66, 75, 86 n.24

164



Index

283b ¼ 355b C.-Pr. 285 C.-Pr. 287 C.-Pr. 291 C.-Pr. 292ab C.-Pr. 292a C.-Pr. 292b C.-Pr. 293ab C.-Pr. 296a C.-Pr. 297a C.-Pr. 298a C.-Pr. 299a C.-Pr. 299b C.-Pr. 299ab C.-Pr. 300a C.-Pr. 301a C.-Pr.  302   313 C.-Pr. 305b C.-Pr. 306 C.-Pr. 308a C.-Pr. 309a C.-Pr. 309b C.-Pr.  313   302 C.-Pr.  313a C.-Pr. 316a C.-Pr. 318a C.-Pr. 322 C.-Pr. 324 C.-Pr. 325 C.-Pr. 325a C.-Pr. 326 ¼ 145 C.-Pr. 327b C.-Pr.  334 C.-Pr. 336 C.-Pr.  338 C.-Pr. 341 ¼ 15 C.-Pr. 342 C.-Pr.  345a C.-Pr. 346 C.-Pr.  347a, c, d C.-Pr. 348 C.-Pr. 349 C.-Pr. 355 ¼ 26 C.-Pr. 355b ¼ 283b C.-Pr.  361 C.-Pr. 389 C.-Pr. 394-7 C.-Pr.

59 n.61 84 n.20 85 84, 87 n.29, 88 74 84 n.20 84 42 84 n.20 85 84 n.20 87, 87 n.29, 88, 89 88 74 74 85 75, 85 34 n.2 42 67 52 n.12, 67, 68 55 75 85 84 n.20, 85 59 n.61 63 63 65 67 n.30 64 67 76 90, 91, 92 n.50 23 87 n.29 88 22, 23 23 22 23 23 87 n.29, 88 n.34 59 n.61 65, 80 n.2 76 81

394 C.-Pr. 396 C.-Pr. 397 C.-Pr. 398 C.-Pr. 399 C.-Pr. 402ab C.-Pr. 406 C.-Pr. 410-14 C.-Pr. (Schmidt) Fr 4 ex fr. 6 Comm. in S. F9 (p. 242) Diogenianus 5.70 6.10 Dionysius of Halicarnassus Th. 51, 55

5 21 21, 23 80 n.2 80 n.2 76 2 n.3, 80 n.2, 81 n.6 2 n.3 74 10 43 72 n.67 73 n.74 64

Dionysius Thrax F28 Linke Tekhne grammatike 1

45 n.79 62 n.2

Dioscorides 4.60.2

42 n.62

Ephorus BNJ 70 F20a

68

Epigenes SGG F4

43

Eratosthenes F6 Strecker Ergias of Rhodes BNJ 513 F1

58 n.56 70

Etymologica Etymologicum Genuinum a 1133 Lasserre-Livadaras b 158 Lasserre-Livadaras

10 38 n.27

Etymologicum Gudianum p. 301.37-43 Sturz p. 458, 12 Sturz p. 515.7-9 Sturz p. 208 add. De Stefani p. 248.13-22 De Stefani

55 n.32 36 n.12 67 n.30 96 55 n.35

Index

Etymologicum Magnum EM s.v. sgsa (p. 757.50–55 Gaisford) Eumelus, On Old Comedy F1 Bagordo Eunapius VS 456 Euphorion Apollodorus F6 Powell ¼ 8 Lightfoot

38 n.27

51 n.1 89 n.37

74

Galen de alim. facult. 2.51 Ind. 13 Ind. 23b–24b Libr. Propr. praef. 1–2 Gellius NA praef. 11 NA 2.9 NA 14.6 NA 15.30 Harpocration y 6 Keaney p 48 Keaney p 125 Keaney u 5 Keaney



165

40 n.40 2 n.11 54 n.28 2 n.11 81 n.7 83 81 n.7, 83 91 n.47 23 75 n.82 49 n.93 85 n.21

Eupolis Testimonia T48 PCG

68 n.38

Demes F132 PCG F136 PCG

68 52 n.14

Hecataeus of Miletus BNJ 1 F291–92a

40

52 n.14

Hellanicus BNJ 4 F22

64

Flatterers F177 PCG Taxiarchoi F268 PCG F282 PCG

58 52 n.14

Euripides Hecuba 1155-56

37 n.21

Hecules Furens 192

36 n.12

Hypsipyle F753 TrGF

69

Iphigeneia in Tauris 323

37

Medea 356 380 Phoenician Women 159-60 Trojan Women 1102 Eustathius in Il. 500.43–44 van der Valk in Od. 1397.39 Stallbaum

39 39 36 n.12 37 9 72 n.67

Heraclides F348 Deichgra¨ber Hermippus of Smyrna FGrHistCont 1026 F31

44 n.75 76

Herodotus 1.75.3 2.130.1 2.158.4 5.52.2 5.52–55 6.138 9.80.1

63 67 67 63 63 73 44 n.73

Herodian Philet. 210 Dain [Hdn.] Part. p. 105 Boissonade

40 n.40 36 n.12

Hesiodic Catalogue of Women F191 Merkelbach–West F209 Merkelbach–West Hesychius Ep. ad Eul. 4 Cunningham Ep. ad Eul. 1-8 Cunningham a 3086 Cunningham b 237 Cunningham d 145 Cunningham

70 28 44, 55 54 58 n.56 69 n.44 45 n.77

166



Index

d 185 Cunningham e 2679 Cunningham e 7192 Cunningham e 7195 Cunningham k 4562 Cunningham k 662 Cunningham k 1390 Cunningham l 3 Cunningham l 5 Cunningham m 714 Cunningham p 407 Hansen p 1290 Cunningham p 2679 Hansen p 4331–32 Hansen s 1751 Hansen–Cunningham

42 34 n.2 42 n.58 42 n.58 40 n.42 56 n.47 43 n.69 48 n.87 48 n.91 43 n.70 37 n.25 36 n.12 57 n.49 49 n.94 36 n.12

Homer Iliad 3.18-19 5.495 6.104 9.174–78 10.76 11.43 11.212 11.454, 12.298 14.6 15.717 18.482 21.162–3 21.459 23.267 23.270 23.702 23.879 23.885 24.390 24.433

37 37 37 18 37 37 37 49 n.94 37 10 10 29 37 n.17 38 n.32 45 44 45 49 n.94 45 38 n.32 38 n.32

Odyssey 1.256 5.53 11.509 13.408 14.348 14.488

37 49 n.94 10 10 10 38 n.29

Homeric Hymns 2.8 21.1 Hyperides Against Demades F79 Jensen Ion of Chios Testimonia T26 Leurini T31a Leurini T31b Leurini Fragments F1 Leurini ¼ F1 TrGF1 F1.2 Leurini ¼ F1.2 TrGF1 F22 Leurini ¼ F17a TrGF1 F26a Leurini ¼ F22 TrGF1 F26b Leurini ¼ F23 TrGF1 F37* Leurini ¼ F33a TrGF1 F*62 Leurini ¼ F52 TrGF1 F85 Leurini ¼ 7 Valerio ¼ S316 SLG BNJ 392 F14 Inscriptions IDidym. 433.7–13 Isaeus 3.76 3.79 8.18 8.20 Istros the Callimachean BNJ 334 F17 BNJ 334 F29 BNJ 334 F41 Jerome Adv. Rufin. 1.16 Lucian VH 2.20

42 n.64 50

67

48 n.88 44 46-8 43 44 49 47-8 46 49 43 n.70 49 49 n.94 45 n.77 75 75 75 75 42 n.65 42 n.63 27, 70 5 81 n.9

Lycophron Alex. 420

56 n.47

Lycus Neapolitanus F316 Deichgra¨ber

44 n.75

Index

Macrobius Sat. 2.2.12 Sat. 5.18.1 Sat. 5.18.2–5 Sat. 5.18.11 Sat. 5.18.11–12 Sat. 5.21.11 Sat. 5.22.9 Sat. 5.22.15 Manuscripts Cod. Laur. S. Marco 304 Cod. Mod. Est. a.U.5.10 Cod. Neap. II F31 Cod. Pal. gr. 252 Cod. Par. gr. 2713 fol. 129r Cod. Vat. gr. 1818 Marcellinus Life of Thucydides §17 Metagenes F12 PCG Mnaseas FHG III 156 F40 ¼ F43 Mehler ¼ F43 Cappelletto

91 n.46 91 68 91 68 88 91 92-3 55 n.32 72 n.70 37 n.18 65 n.17 34 n.1, 7 55 n.32 65 52 n.14 69 n.44

Moeris o 26 Hansen

68 n.38

Myrsilus of Metymna BNJ 477 F1

73 n.73

Nepos, Cornelius Att. 10.6

2 n.10

Nicander Ther. 147 Ther. 634

56 n.47 56 n.47

Nicon F1 PCG

54, 88

Orion col. 18.21–19.5 Sturz

55 n.35

Panaetius of Rhodes T154 Alesse ¼ F93 van Straaten

46

Papyri P.Amh. II 12

62

P.Berol. inv. 9780 recto col. iv.59–vi.63 col. vi.20–22 col. vi.62 col. viii.54 P.Berol. inv. 11903 P.Berol. inv. 13875 P.Flor. II 112 P.Herc. 1021 P.Herc. 1691 pz. 2 P.Lond. inv. 2110 recto P.Oxy. VI 856 P.Oxy. IX 1174 P.Oxy. XII 1611 fr. 2 col.i.121–27 P.Oxy. XIII 1611 fr.16 P.Oxy. XV 1801 P.Oxy. XV 1805 P.Oxy. XVII 2076 col. i P.Oxy. XVIII 2192 P.Oxy. XXIII 2368 P.Oxy. XXVI 2442 P.Oxy. 2452 P.Oxy. XXXV 2737 col.i.19–27 P.Oxy. XXXV 2740 P.Oxy. XXXV 2744 P.Oxy. LIV 3722 fr. 2 P.Oxy. LXV 4455 P.Oxy. inv. 51 B 44/G(b) P.Paris 71 col. ii.27, iv.4 P.Rain. 7 PSI XIV 1391 Pausanias 8.31.1 9.31.9 Phanodemus BNJ 325 F17 Pherecydes of Athens F2 Fowler F68 Fowler ¼ BNJ 3 F68 ¼ T7Aa F159 Fowler ¼ BNJ 3 F159 F168 Fowler F169 Fowler



167

6 75-6 25 6 n.38 7 n.39 95 95 55-6, 59, 95 6 6 n.33 96 95 35 n.6, 39 n.36 49, 56 49 56, 95 35 n.6 22 n.7 57 n.52 22, 95 24, 24 n.24 35 n.6 56, 95 49 58 95 21-2 63 95-6 22 n.8 75 n.83 95 42 n.64 42 n.64 75 64 70 69 70 n.51 70 n.51

168



Index

Philistus of Syracuse BNJ 556 F49 Phillis of Delos FHG iv.476 F6 Philochorus BNJ 328 F137 BNJ 328 F180 Philodemus On Poems On Music (Bk. 4) On Rhetoric Photius e 2337 Theodoridis k 196 Theodoridis k 271 Theodoridis l 2 Theodoridis o 379 Theodoridis p 173 Theodoridis p 542 Theodoridis p 976 Theodoridis p 1526–27 Theodoridis p 1527–28 Theodoridis r 561 Theodoridis r 897 Theodoridis u 63 Theodoridis u 64 Theodoridis Phrynichus F4 PCG F12 PCG F43 PCG Pindar Olympians 2.77 5.10-14 6.93 9.22 10.13 10.45 10.91-3 Pythians 1.33-4 4.79 5.26

26, 71 46-7 65 74 6 6 6 42 n.58 57-8 n.47 73 n.74 48 n.91 68 n.38 37 n.25 36 n.12 49 n.93 49 n.93 49 n.93 36 n.12 36 n.12 85 n.21 85 n.21 52 n.14 54 52 n.14

30 27 26 31 30 30 31 25, 33 37 n.21 26

5.124 7.6 8.79 9.101-2 10.36

32 30 27 27 33

Nemeans 1.1 1.6 1.24 1.25 1.33-72 4.11 4.59 4.93 5.6 6.31 7.14-16 7.30-31 7.30-37 7.61 7.102-4 9.9-10 9.19-21 10.25-8 10.28 10.29-33

59 31 30 n.52 31 25 33 28 30 31 28 31 32 32 31 32 n.60 30 31 32 32 32

Isthmians 1.64–67 2.8 2.12 6.7–9

32 31 31 32

Fragments F6c–f Snell-Maehler F66 Snell-Maehler F94b–c Snell-Maehler

23 23 23

Plato Lg. 819b Ly. 215c Phdr. 236b Plato Comicus F85 PCG F116 PCG F155 PCG F281 PCG

87-8 n.30 66 76 52 n.14 52 n.14, 57-8 66 n.27 57 n.49

Index

Plautus Bac. 1037

36 n.14

Pliny the Elder NH Praef. 17

6, 93-4

Pliny the Younger Ep. 3.5 Plutarch Life of Solon 1.1 19.3-4 25.2 Quaest. Con. III 674b PMG F842 F849 Polemon of Ilium F2 Capel Badino ¼ F2 Preller F5 Capel Badino ¼ F5 Preller F11 Capel Badino ¼ F26 Preller F20 Capel Badino ¼ F23 Preller F90 Preller

6

65 65 n.21 65 n.21 42 n.60 76 23 74 65 70 30 n.55 40

Polycrates FGrH 588 F1

88

Polyzelus F12 PCG

72

Posidonius T34 Edelstein–Kidd T44 Edelstein–Kidd F7 Edelstein–Kidd F41a, b Edelstein–Kidd

2 n.10 2 n.11 2 n.11 2 n.11

Proclus, Chrestomathy 24-99 Severyns 79–86 Severyns

23 23

Proxenus BNJ 703 F3

70

Ptolemy of Ascalon p. 197 Baege

38 n.29

Pyrrhandrus FHG iv.486

46-7

Pythaenetus BNJ 299 F2a

28

Quintilian Inst. 1.8.20 Inst. 1.8.21 Sappho F1.11 Voigt



169

5 92 49 n.94

Scholia R A. Pers. hyp. I Dindorf 40 n.44 R A. Th. 636m Smith 36 n.12 R A. Th. 669g Smith 36 n.12 R A. Th. 973m–n Smith 36 n.12 R [A.] Pr. 3a Herrington 43 R A.R. 4.973 Wendel 55 R A.R. 4.1613–16c Wendel 55 Proll. de com. III 41 Koster 53 n. 16 Proll. de com. XXXa 7–20 Koster 53 n. 16 R Ar. Av. 35a Holwerda 36 n.15, 37 n.16, n.2 R Ar. Av. 1296a Holwerda 53 n.20 R Ar. Eq. 42b, 165b Jones–Wilson 49 n.93 R Ar. Pl. 388a Chantry 67 n.29 R Ar. Pl. 720ia–b Chantry 53 n.24 R Ar. Ran. 186b, 990c Chantry 59 n.60 R Ar. Ran. 970b-e Chantry 72 R Dem. Or. 3.48, 49a Dilts 37 n.25 R Eur. Andr. 32 Schwartz 40 n.44, 41 R Eur. Hec. 484 Schwartz 36 n.12 R Hom. Il. 1.97-99 Erbse 13 R Did. Hom. Il. 1.106e Erbse 19, 37 n.17 R Did. Hom. Il. 1.423–24 Erbse 14 R Hom. Il. 1.512b Erbse 36 n.11 R Did. Hom. Il. 1.522a1 Erbse 17 n.77 R Hom. Il. 1.524c Erbse 13 R Hom. Il. 2.111b Erbse 13 R Hom. Il. 2.131a1 Erbse 16 R Hom. Il. 2.133a Erbse 16, 39 n.36 R Hom. Il.2.139b Erbse 36 n.11 R ex. Hom. Il. 2.183b Erbse 9 1 R Did. Hom.Il. 2.192b Erbse 17 n.77 R Did. Hom. Il. 2.221 Erbse 17 n.77 R Hom. Il. 2.335a2 Erbse 19 R Hom. Il. 2.462a2 Erbse 19 R Hom. Il. 2.463b Erbse 36 n.11 R Hom. Il. 2.517a Erbse 16 R Hom. Il. 2.798a Erbse 13 R Ariston. Hom. Il. 2.867a Erbse 64 n.7 R Hom. Il. 3.18a Erbse 19 R Hom. Il. 3.31b Erbse 36 n.11

170



Index

R Hom. Il. 3.292a1 Erbse 19 R Did. Hom. Il. 3.406a Erbse 17 n.77, 19 R Hom. Il. 4.3a Erbse 16 n.67, 17 n.77 R. Hom. Il. 6.4a Erbse 41 n.55 1 R Hom. Il. 7.130a Erbse 16 n.66 R Ariston. Hom. Il. 7.135a Erbse 70 n.49 R Hrd. Hom. Il. 7.135b Erbse 70 n.49 R Hom. Il. 7.238c2 Erbse 42 n.59 R Did. j Ap.H. Hom. Il. 7.256–57 Erbse 12 R Hom. Il. 7.428a1 Erbse 19 R Hom. Il. 7.452a Erbse 40 n.44 R Hom. Il. 8.37a Erbse 39 n.36 R Hom. Il. 8.163c Erbse 17 n.77 R Did. Hom. Il. 8.296b Erbse 13 n.39 R Did. Hom. Il. 8.441a Erbse 13 n.39 R Did.(?) Hom. Il. 9.153c Erbse 69 n.47 R Did.(?) Hom. Il. 9.153d1 Erbse 69 n.47 R Hom. Il. 9.222a Erbse 18 RA Hom. Il. 9.222b1 Erbse 18 RA Hom. Il. 9.222b2 Erbse 18 3 RT Hom. Il. 9.222b Erbse 18 R Hom. Il. 9.349-50 Erbse 13 R Hom. Il. 9.584a1-a2 Erbse 19 R Hom. Il. 12.435a1 Erbse 13 RT Hom. Il. 14.199a1 Erbse 37 n.26, 38 n.29 R Hom. Il. 14.382d1 Erbse 40 n.44 R Hom. Il. 16.21b Erbse 40 n.44 R Hom. Il. 16.467a1 Erbse 13 n.39, 40 n.44, 41 RA Hom. Il. 17.1–2a2 Erbse 37 n.24 RbT Hom. Il. 18.274b Erbse 37 n.25 R Hom. Il. 19.26a Erbse 39 n.36 R Did. Hom. Il. 19.365–681 Erbse 16, 17 R Did. Hom. Il. 21.130–35a1 Erbse 16 R Hom. Il. 21.162a1 Erbse 19, 37 n.17 R Hom. Il. 21.162a2 Erbse 19 R. Hom. Il. 21.162b2 Erbse 37 n.17 R Hom. Il. 23.270a Erbse 45 n.78 R Hom. Il. 24.110b1 Erbse 13, 69 R Did. Hom. Od. 1.259e Pontani 70 R Hom. Od. 3.71a Pontani 64 n.7 RD Hom. Od. 10.516b Ernst 36 n.12 R Hom. Od. 11.221 Dindorf 37 n.26, 38 n.29 R Luc. 21.11 Rabe 49 n.93 R Pind. O. 3.22a Drachmann 26 n.40 R Pind. O. 5.16 Drachmann 27 n.44 R Pind. O. 5.19a Drachmann 27 n.44

R Pind. O. 5.19b Drachmann R Pind. O. 5.19c Drachmann R Pind. O. 5.19d Drachmann R Pind. O. 6.154a–b Drachmann R O. 7.95a Drachmann R Pind. O. 13.27a Drachmann R Pind. P. 1.67, 69 Drachmann R Pind. P. 3.18a Drachmann R Pind. N. 6.53b-e Drachmann R Pind. N. 6.54a Drachmann R Pind. N. 7.44a Drachmann R Pind. N. 7.150a Drachmann

27 n.44 27 n.44 27 n.44 22 n.9 26 n.40 33 33 n.63 40 n.44 28 28 32 n.61 25 n.35, 32 n.60, 33 n.64 R Pl. Criti. 112a bis Greene 49 n.93 R S. El. 6–9 Xenis 43 n.69 R S. El. 28 Xenis 43 Hyp. IV.14–15 OC Xenis 39 n.34 RLrT S. OC 388 Xenis 41 RLrT S. OC 390a1 Xenis 41 RLrT S. OC 683-4 Xenis 42 R S. Aj. 1171b Christodoulou 36 n.12 R S. Aj. 1225a Papageorgiou 42-3 R Theoc. 1.40a Wendel 56 n.47 R Thuc. 1.44.1d Kleinlogel–Alpers 67 n.30 R Thuc. 5.23 Hude 37 n.25 Seleucus Homericus BNJ 341 F1 ¼ T4 Ruschenbusch Seneca Dial. 10.13.2, 3 Ep. 88.3 Ep. 88.36 Ep. 88.37 Ep. 88.39 Ep. 88.40 Ep. 88.42–43 Ep. 90.20 Ep. 95.65 Ep.108.36-38 Sextus Empiricus 1.57 7.1-262 7.263-68 Servius DServ. ad G. 3.391–3

65 83 83 81 5 81 81 83 82 n.11 83 82 n.11 62 n.2 6 n.35 6 n.35 92 n.50

Index

Skylax BNJ 709 F3–4

40

F220a.84.2 TrGF4 F238 TrGF4 F314.72 TrGF4 F314.108 TrGF4 F314.146 TrGF4 F314.156 TrGF4 F449 TrGF4 F624 TrGF4 F718 TrGF4 F728 TrGF4 F730e.16-17 TrGF4 F904 TrGF4



171

35 n.6 48 n.91 35 n.6 35 n.6 35 n.6 35 n.6 43 n.69 43 n.69 40 n.39 43 n.69 35 n.6 37 n.21, n.25

Sophocles Ajax 74-8 76 80 82-3 83-5 326 408-9

35 36 n.13 36 n.13 35 35 42 36

Antigone 4-6 45-6 242

35-6 41 n.47 43

Strabo 13.1.54 16.2.10 17.3.5

2 n.11 82 n.11 2 n.5

Electra 727 1029

69 n.44 36 n.14

Strattis F20 PCG F24 PCG

72 72 n.67

Oedipus at Colonus 207-24 226-36 237-54 258-91 292-309 389-90 681-5 761-4 761-99 774 782 794-5 981 986

38 38 38 38 38 41 n.54 42 38 38 38, 38 n.32 38 38 38 n.31 38 n.31

Oedipus Tyrannus 360 399

38 n.32 38 n.32

Philoctetes 103 611–13

36 n.14 36 n.14

Trachinian Women 744 Fragments F122 TrGF4

35 n.6 42

Suda a 1793 Adler a 3215 Adler d 871-6 Adler d 873 Adler d 874 Adler d 1256 Adler e 3757 Adler i 399 Adler k 284 Adler k 451 Adler m 450 Adler p 232 Adler s 895 Adler

37 n.23 2 3 3, 96 3, 96 37 n.20 42 n.58 2, 46 n.82 56 n.47 73 n.74 37 n.23 37 n.25 3, 81 n.6, 86

Suetonius DGR 24.3

91 n.47

Suetonian–Donatan Vita 44–46

91 n.45

Synagoge Syn.a p 82 Cunningham Syn.a p 274 Cunningham Syn. B a 1637 Cunningham Teleclides F73 PCG

37 n.25 36 n.12 66 n.28, 67 n.29 58

172



Index

Terence An. 205

36 n.14

Terpander S6 SLG

49

Theocritus Id. 1.40 Id. 24.8

56 n.47 42 n.58

Theophrastus HP 3.18.4 HP 6.6.9 HP 6.8.1-3 On Comedy F1-2 Bagordo Pokisikom pq o1 sou´ 1 kaiqou´ 1 F589.4a FHS&G Peqi kaiqom F609 FHS&G On Discoveries F734 FHS&G On Nemean Victories À

À

Theopompus BNJ 115 F250 BNJ 115 F284 BNJ 115 F291

40 n.41 42 n.60 42 n.62 51n.1 76 n.88 76 70 70 76 43 n.72 76

Theon F19-35 Guhl

39 n.36

Theotimos BNJ 470 F1

26, 70

Thucydides 1.44.1 1.86.3 5.23 8.54.3 8.78.5

67 37 37 71 56 n.47

Timachidas of Rhodian Lindos F20 Matijasic F27 Matijasic

58 41 n.55

Timaeus of Tauromenium BNJ 566 F18 BNJ 566 F39b BNJ 566 F93b BNJ 566 F96 BNJ 566 F136 BNJ 566 F142a BNJ 566 F145

71 71 71 26 65 n.15 71 71

Timocreon of Rhodes F733 PMG

59

Tryphon F61 Velsen F110 Velsen

36 n.12 46-7

Tzetzes, John R Tz. Ar. Av. 530 Koster

56 n.46

Vergil Georgics 1.7-9 3.391-3 4.122

68 91 42 n.60

Xenophon Anab. 1.2.5–23 Hell. 2.3.3-33

63 72 n.68

Zenobius, On Proverbs 4.74 4.91 Ps.-Zonaras s.v. dipakso1 p. 509 Tittmann Zopyrus FHG IV 533 F6

23, 23 n.14 72 n.67 73 n.74 37 n.20 65

Bulletin of the Institute of Classical Studies, 2020, 63, 173–174 doi: 10.1093/bics/qbab013 Index

INDEX

Achaeus 34 Achelous 68–69, 90 Aeschylus 34 Accuracy 27–28 lack of 4, 58–59, 75, 79, 87–88 Alcaeus 22, 58–59 Alcman 22 Allusion 44–46, 49–50, 59 Ammonius of Alexandria 17, 52, 57–58 Anacreon 21–22, 48 Apion 2, 7 Archilochus 22, 59 Aristarchus 1–2, 7, 8, 10–19, 24–30, 48, 62, 72 Aristonicus 2–3, 11–13, 24 Aristophanes 49–50, 55–61, 66–67, 71–73 Aristophanes of Byzantium 30 Aristotle 25, 51, 65–66 Aristoxenus 48 Asclepiades of Myrlea 45–46, 62, 65–66 Athenaeus 43–48, 53–54, 86–90 Authenticity 30 Bacchylides 22 Book circulation 2 Callimachus 7, 51, 66 Cicero 2, 6 Chaeris 25 Chrysippus (grammarian) 25 Comedy 49–61, 71–73 Commentaries 6–7, 9–10, 24–33, 34–46, 52–53, 56–61, 63, 69–76 Criticism 33, 35, 39–40, 76 of D. 3–4, 75, 79–83, 85–87 Demetrius Ixion 72 Demosthenes 26, 74–76

Didymus and earlier scholars 7, 10–20, 24–30, 40–48, 51, 64–65, 67–69, 71–72, 74–76 and his contemporaries 2–3, 7, 11–13, 48, 62 and later scholars 53, 79–94 and Rome 2, 4, 80–84, 90–94 attribution 3, 22, 26, 40–43, 49, 53, 55–56, 71, 95–96 chronology 2–3 life 1–2, 80 methods 4–7, 19–20, 24–28, 31–33, 51–52, 63, 77–78 nicknames 4, 75, 87–90 reputation 4, 7, 51, 79–94 works 3–4, 95–120 Didymus (others) 3, 95–96 Diogenianus 54 Doxography 4–7, 24–26, 52, 67–72, 74–77 Drinking vessels 43–46, 54 Edition 13–19, 30 Etymology 8–9, 23, 56–57, 96 Eupolis 55, 58, 68 Euripides 35, 68–69 Excerption 6–7 Food

53, 55, 87

Gambling 57–58, 72 Gellius, Aulus 83–84, 93 Genre 23, 30–31 Geography 27–28, 59–60, 63, 69–70, 76 Halys 63 Harpocration 23, 54–55, 67–68, 74–75, 84–86 Heliaia 74 Hellanicus 64

C The Author(s) 2021. Published by Oxford University Press on behalf of Institute of Classical Studies. All rights reserved. V

For permissions, please email: [email protected]



173

174



Index

Heraclides Ponticus (the Younger) 2, 80–81 Hermias of Atarneus 75–76 Herodotus 62–63, 67, 73, 87 Hesiod 8–9 Hesychius 54–55, 67 History 5–6, 26–28, 52, 59–60, 62–78, 84–85 Homer 9–20, 44–46, 91 hypomnema 6 hypomnematisamenoi 24, 40–42 Inscriptions 64–65, 77 Interpretation 31–33, 36–37, 61 Ion of Chios 43–50 Juba

2, 46

komodoumenoi 26, 52, 57–60 Lexicography 7, 8, 10, 40, 43–50, 54–57, 66–69 Lycophron 7, 52, 88–89 Lyric 21–33 Macrobius 68–69, 90–93 Mantinea, battle of 60, 72 Marcellinus 64–65 Menander 3, 7, 53, 55 Miltiades 64 Morphology 29–30 Musical instruments 23, 46–48 Omissions 6–7 Orion (grammarian) 55 Orthography 60, 64 Orus 55 Pamphilus 54 Papyri ascribed to D. 3, 22, 26, 49, 55–56, 75–76, 95–96 Parallels 25, 29–33, 41–42, 45, 52, 58–59 Performance 30–31 Pherecydes 64, 70 Philodemus 6 Photius 55 Phrynichus (comedian) 53–54 Pindar 24–33, 59 Pius 36–37 Plants 40, 54

Pliny the Elder 6, 93–94 Pnyx 49 Polemon of Ilium 65, 74 Praise 32–33 of D. 4, 90–91 Probus, Valerius 91 Prose 3, 62–66, 74–76 Proverbs 3, 23, 59, 72–73 Quintilian 4, 82–83, 93 Sappho 21–23 Scholia ‘Didymi’ 8 to Apollonius of Rhodes 55 to Aristophanes 52–53, 56–61, 66–67 to Euripides 34, 71, 73 to Hesiod 8–9 to Homer 9–20, 69–70, 95 to Pindar 24–33, 70–71, 95 to Sophocles 35–43 Seneca 4, 21, 81–83 Solon 65–66, 74 Sophocles 35–43 Symmachus 53 Syntax 35–36 Textual criticism 4–5, 13–19, 28–30, 37–40, 60–61, 69–70 Theognis 23 Theon 2–3, 7, 24, 62 Theramenes 72 Thucydides 64–65, 67 Tragedy 34–50 Tryphon 48 Unattested 25–27, 58 Variant readings 11, 13–14, 18–19, 30, 60–61 Virgil 90–93 VMK 10–11, 69, 95 Word formation Zenobius 23, 53 Zopyrus 65

48–49, 57