Fakes and Forgers of Classical Literature: Ergo Decipiatur! 9004266410, 9789004266414

Citation preview

Fakes and Forgers of Classical Literature

Metaforms Studies in the Reception of Classical Antiquity

Editors-in-Chief

Almut-Barbara Renger, Freie Universität Berlin Jon Solomon, University of Illinois at Urbana-Champaign John T. Hamilton, Harvard University Editorial Board

Kyriakos Demetriou, University of Cyprus Constanze Güthenke, Princeton University Miriam Leonard, University College London Mira Seo, University of Michigan

VOLUME 2

The titles published in this series are listed at brill.com/srca

Fakes and Forgers of Classical Literature Ergo decipiatur! Edited by

Javier Martínez

Leiden • boston 2014

Library of Congress Cataloging-in-Publication Data Fakes and Forgers of Classical Literature : Ergo decipiatur! / edited by Javier Martinez. pages cm. — (Studies in the Reception of Classical Antiquity ; 2) ISBN 978-90-04-26641-4 (hardback : acid-free paper) — ISBN 978-90-04-26642-1 (e-book)  1. Literary forgeries and mystifications—History—To 1800. 2. Classical literature—History and criticism. I. Martínez García, Francisco Javier, 196—editor of compilation. PA3014.F6F35 2014 880.9’001—dc23

2013046700

This publication has been typeset in the multilingual “Brill” typeface. With over 5,100 characters covering Latin, IPA, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities. For more information, please see www.brill.com/brill-typeface. ISSN 2212-9405 ISBN 978-90-04-26641-4 (hardback) ISBN 978-90-04-26642-1 (e-book) Copyright 2014 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Brill Nijhoff, Global Oriental and Hotei Publishing. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. Brill has made all reasonable efforts to trace all rights holders to any copyrighted material used in this work. In cases where these efforts have not been successful the publisher welcomes communications from copyright holders, so that the appropriate acknowledgements can be made in future editions, and to settle other permission matters. This book is printed on acid-free paper.

CONTENTS Foreword ............................................................................................................ Javier Martínez Prologue / Volume Retrospect: Ergo decipiatur ..................................... John Henderson

vii xv

Libertine Erudition: José Marchena’s Fragmentum Petronii and the Power of the False ........................................................................................ Joaquín Álvarez Barrientos

1

Lucretius auctus? The Question of Interpolation in De rerum natura ........................................................................................... David Butterfield

15

Authorless Authority in Plato’s Theaetetus .............................................. Zina Giannopoulou

43

The Poet and the Forger: On Nonnus’ False Biography by Constantine Simonides ............................................................................. David Hernández de la Fuente

59

“Genuine” and “Bastard” Dialogues in the Platonic Corpus: An Inquiry into the Origins and Meaning of a Concept ................ Mark Joyal

73

Female Voice, Authorship, and Authority in Eudocia’s Homeric Centos ........................................................................................... Andromache Karanika

95

The Surgical Treatises of the Corpus Hippocraticum: Statistical Linguistics and Authorship .................................................. 109 Mikel Labiano True Plautus, False Plautus. Pellio restitutus – uxor excisa. Annotations to Plautus’ Bacchides ........................................................ 125 Klaus Lennartz

vi

contents

Athena and Pallas, Image, Copies, Fakes, and Doubles ....................... 143 Françoise Létoublon Hippias of Elis: Lessons from One Master Forger ................................. 163 Javier Martínez Reading the Fraudulent Text: Thessalus of Tralles and the Book of Nechepso ....................................................................................................... 179 Karen Ní Mheallaigh Hapax Legomena in the “Speeches of Apollodoros” and their Relation to the Corpus Demosthenicum ............................................. 187 Felipe G. Hernández Muñoz Language and (in-)Authenticity: The Case of the (Ps.‑)Lucianic Onos ..................................................................................... 195 Heinz-Günther Nesselrath SH 906 and the Apollo of Simias of Rhodes: Some Issues of (mis‑)Attribution ......................................................................................... 207 Marco Perale Order, Ambiguity, and Authority in Venantius Fortunatus, Carm. 3.26 ..................................................................................................... 219 Joseph Pucci Authors Pseudonyms in the Seventeenth Century: The Case of Gaspar Scioppio ........................................................................................... 231 Eustaquio Sánchez Salor Pseudepigraphy and Magic ........................................................................... 243 Emilio Suárez de la Torre The Sophists’ Place in the Greek Wisdom Tradition ............................ 263 Håkan Tell Forging Ancient Greek Words in Modern Times .................................. 283 Onofrio Vox Index: Names and Subjects ........................................................................... 291

Foreword In the end, forgery is a sort of crime. Let us then examine the three circumstances that need clarification when any crime, human or literary, is to be investigated: motive, means, and opportunity. A. Grafton 1990: 37

The studies conducted over the last few decades by authors such as Speyer, Grafton (his Forgers and Critics published in 1990 in particular), and more recently, Ruthven,1 have created an extremely favourable academic setting for the study of textual and literary – as opposed to traditional – forgery. The new era of postmodernism has also encouraged a difference of feeling towards the work of the forger. Until relatively recently, this “creative activity” was condemned, criticised, and even criminalised when the deceit was uncovered. However, for some time now many of these ignored and discarded works have captured the curiosity of academics interested in recovering their hidden values. It is not only their literary value that intrigues researchers; these texts can also provide information on the particular political, social, and cultural circumstances that led to their creation. Forged texts can, and in many cases should, become fullfledged members of a literary tradition studied with sensitivity in keeping with modern times and the new approaches of philology and literary criticism. In light of the quotation cited at the beginning of this introduction, we might consider forgery as “a type” of crime in itself with a view to promoting detailed research into the circumstances that motivated its creation in the first place. Our aim as scholars, far from making any moral judgement, is not to stigmatise this genre, demand punitive damages, or dismiss it entirely from literature as a whole for extra-literary reasons. Treating 1 W. Speyer, Die literarische Fälschung im heidnischen und christlichen Altertum: ein Versuch ihrer Deutung, München 1971; A. Grafton, Forgers and Critics: Creativity and Duplicity in Western Scholarship, Princeton/New Jersey 1990; “Fälschungen” in Der Neue Pauly 4 (1998): 394/97, “Correctores corruptores? Notes on the Social History of Editing” in Editing Texts – Texte edieren, ed. G.W. Most, Göttingen 1998 (= Aporemata 2): 54–76; “Forgery” in The Classical Tradition, ed. A. Grafton / G.W. Most / S. Settis, Harvard 2010: 361–364; K.K. Ruthven, Faking Literature, Cambridge 2001. See also A. Guzmán Guerra, “Pseudoliteratura, falsificación y canon: una perspectiva programática” in Actas del XI Congreso Español de Estudios Clásicos, ed. A. Alvar, Madrid 2005: 1:177–216.

viii

foreword

f­ orgery as a crime in our analysis of this genre does not mean that we actually consider it to be a crime in any way.2 Our aim should be to focus our attention on the different aspects that influence forgery in order to glean information about the circumstances surrounding these works themselves and understand the motives for the “crime.” This kind of anticanonical or secondary text that tradition has deemed worthy of transmission is not a kind of second-rate literature. These works have managed to survive the tension that exists between what tradition grants authority, that is, the canon, and the freedom that comes with all literary creation. Attempts were made from early on to categorise the different reasons that lead an author to present his own literary invention as ­another’s (which is ultimately what forgery is), to renounce his own creation (sc. authorship) and to ascribe his work to another writer. A attempt particularly interesting for its conciseness and simplicity is one of the earliest: the analysis of David, a little-known commentator from the sixth century, who in his remarks on Porphyry’s Isagoge3 divides the forged works into four types γίνεται δὲ νόθον σύγγραμμα κατὰ τέσσαρας τρόπους. The first kind arises through homonymy δι’ ὁμωνυμίαν, and could be due to a coincidence with the author’s name δι’ ὁμωνυμίαν τῶν συγγραψαμένων or the titles of the works δι’ ὁμωνυμίαν τῶν συγγραμμάτων. The second kind is the result of ambition or vanity διὰ φιλοτιμίαν ἤτοι κενοδοξίαν, whilst the third is down to a sordid love of gain δι’ αἰσχροκέρδειαν. Finally, the fourth kind of work is down to the affection of a writer’s disciple δι’ εὔνοιαν τοῦ οἰκείου διδασκάλου. David explains at this point that many ascribe their own work to their master, giving as an example Pythagoras’ followers καὶ γὰρ πολλοὶ ποιοῦσι συγγράμματα καὶ διὰ τὴν εὔνοιαν τὴν πρὸς τὸν διδάσκαλον τὸ ὄνομα τοῦ οἰκείου διδασκάλου ἐπιγράφουσιν, ὅπερ καὶ οἱ Πυθαγόρειοι ἐποίησαν καὶ γὰρ οὗτοι ἐποίησαν τὰ Χρυσᾶ ἔπη καὶ πρὸς τιμὴν τοῦ οἰκείου διδασκάλου ἐπέγραψαν τὸ ὄνομα αὐτοῦ. His final words could not be more appropriate and highlight the fact that it is common knowledge that a large amount of forged literature is in circulation throughout the world ἐπειδὴ οὖν, ὡς εἴρηται, πολλά εἰσι νόθα συγγράμματα . . . 

2 The quasi-inquisitorial stance that Robert S. Lopez adopts when he recommends that his colleagues in the area of medieval studies follow a similar strategy is interesting here. As opposed to in dubio pro reo, Lopez suggests that such an unprovenanced document should always be regarded “as guilty until proved innocent”; see “The Case Is Not Settled,” in The Proceedings of the Vinland Map Conference (PVMC), ed. W.E. Washburn, Chicago, 1971: 31. 3 Dav., in Isag. 81ff.



foreword

ix

Of these four categories of forged works, one of the most difficult to identify is the fourth kind due to the nature of the inside knowledge that it requires. The first kind is also relatively difficult to pinpoint and can often be impossible to detect, as in the case of the various Philostratuses.4 The other two kinds of text are the most predominant in literature. In the third category, the motive and efforts that have influenced the forger are purely economic, while behind the fourth kind there is always a desire for honour and respect. From a strictly literary perspective, the motivation is the same as that of the cuckoo in the natural world: the survival of one’s own creations under another’s protective wing. From a forger’s perspective, an author’s literary status and respected position are a means of endorsing his work since the higher status represents a protective blanket for the survival of the forger’s “progeny” for generations to come. With this in mind, we should regard the relationship between genuine and spurious texts as a necessary one, as almost a dialectical pairing, since it is most probably the quest to stand out from the rest that attracts a writer to take advantage of the privileged situation of another’s work that has already achieved this goal. In this sense, we agree with Mülke’s dictum that “Verfälschung negiert nicht die Autorschaft, sondern setzt sie voraus und stärkt sie”5 and extend it to the notion of forgery, “Fälschung,” not restricting it to adulteration, “Verfälschung.” Anecdotes and examples of textual forgery are numerous throughout classical literature and appear as early as the archaic period, coinciding (not by chance) with the introduction of writing and the subsequent generalisation of literacy, at a time when authors began to acquire consciousness of their identity and were openly proud of their creations. One of the first examples of this new sentiment is to be found in the Hymn to Apollo 3.170 ss., where the author uses some of his verses to highlight the “incomparable” quality of his work and offers an unmistakeable hint at his identity: ὦ κοῦραι, τίς δ᾽ ὔμμιν ἀνὴρ ἥδιστος ἀοιδῶν 170 ἐνθάδε πωλεῖται, καὶ τέῳ τέρπεσθε μάλιστα; ὑμεῖς δ᾽ εὖ μάλα πᾶσαι ὑποκρίνασθαι ἀφήμως: τυφλὸς ἀνήρ, οἰκεῖ δὲ Χίῳ ἔνι παιπαλοέσσῃ τοῦ μᾶσαι μετόπισθεν ἀριστεύσουσιν ἀοιδαί. 4 Or v.gr. similarity of the names, cf. below Nesselrath’s conclusion (regarding the name confusion of Lucius and Lucian) and his n. 74. 5 M. Mülke, Der Autor und sein Text. Die Verfälschung des Originals im Urteil antiker Autoren, Berlin/New York 2008, 263.

x

foreword “Whom think ye, girls, is the sweetest singer that comes here, and in whom do you most delight?” [170] Then answer, each and all, with one voice: “He is a blind man, and dwells in rocky Chios: his lays are evermore supreme” (trans. Hugh G. Evelyn-White).

As soon as the concept of authorship appears with clarity and the notion of the distinctive author’s identity is established, any textual creation is open to being emulated within the channels of creative mimesis typical of Greek literary creation. In the same way, however, any author of prestige can fall victim to forgery when the forger, using this mimetic process, makes significant changes to his model or ascribes a completely new work to another author. The gradual expansion of the use of writing as a channel for the dissemination of literature did not put a stop to the work of forgers, just as technology has not stopped them nowadays. Paradoxically, any innovations in the long process of transmission (e.g., metagraphé or metacharacterismos), including the introduction of new materials and of new means of transmission, seem to encourage this activity, and forgers seem to find their niche with the application of more modern techniques and means. It is admirable and praiseworthy indeed that a forger distanced historically from his object might produce something that is equal at a literary level and that manages to remain undetected by its critics. It is also commendable that he may be able to demonstrate such skill that he can pass off modern material as old. Overall, in antiquity the question over the manipulation of material was not as clear-cut as it is today, although this does not mean that it was not latent. With regard to the text itself, however, as is the case today, the author felt strongly that his work was his own and that he should actively fight against all attempts to manipulate his creations. A good example of this tension can be found in the words of Rufinus, who includes in his De adulteratione librorum Origenis a letter from Origen to his readers, concerned over suggestions that his writings were being adulterated:6 talia ergo quaedam uideo eliam nobis accidere. nam quidam auctor haereseos, cum sub praesentia multorum habita inter nos fuisset disputatio et descripta, accipiens ab his qui descripserant codicem, quae uoluit addidit et quae uoluit abstulit, et quae ei uisum est permutauit, circumferens tanquam ex nomine nostro, insultans et ostendens ea quae ipse conscripsit. pro 6 CCL 20, 11ff. See this and similar examples in M. Mülke, Der Autor und sein Text, 39ff.



foreword

xi

quibus indignantes fratres qui in Palaestina sunt, miserunt ad me Athenas hominem qui acciperet a me ipsa authentica exemplaria. quod ne relectum quidem uel recensitum a me antea fuerat, sed ita neglectum iacebat ut uix inueniri potuerit. misi tamen, et sub Deo teste loquor quoniam, cum conuenissem illum ipsum qui adulterauerat librum, arguens quare hoc fecisset, uelut satisfaciens mihi respondit: quoniam magis ornare uolui disputationem illam atque purgare. uidete quali purgatione disputationem nostram purgauit. tali nempe, quali purgatione Marcion Euangelia purgauit uel Apostolum; uel quali successor eius post ipsum Apelles. It is something of the same kind, I perceive, which is happening to us also. A certain promoter of heresy, after a discussion which had been held between us in the presence of many persons, and notes of it had been taken, procured the document from those who had written out the notes, and added or struck out whatever he chose, and changed things as he thought right, and published it abroad as if it were my work, but pointing in triumphant scorn at the expressions which he had himself inserted. The brethren in Palestine, indignant at this, sent a man to me at Athens to obtain from me an authentic copy of the work. Up to that time I had never even read it over again or revised it: it had been so completely neglected and thrown aside that it could hardly be found. Nevertheless, I sent it: and, – God is witness that I am speaking the truth, – when I met the man himself who had adulterated the work, and took him to task for having done so, he answered, as if he were giving me satisfaction: “I did it because I wished to improve that treatise and to purge away its faults.” What kind of a purging was this that he applied to my dissertation? Such a purging as Marcion or his successor Apelles after him gave to the Gospels and to the writings of the Apostle (trans. Philip Schaff).

For Origen, it is absurd that his dispute with the “promoter of heresy” which had taken place publicly and was also registered in writing sub praesentia multorum habita inter nos fuisset disputatio et descripta, could be used by his opponent, who alters and modifies it at will quae uoluit addidit et quae uoluit abstulit, et quae ei uisum est permutauit, to then circulate it under the name of Origen circumferens tanquam ex nomine nostro. It appears that some brothers in Palestine became suspicious of the (altered) writings that had come into their hands and contacted Origen, who then sent them the original copies of the text acciperet a me ipsa authentica exemplaria. Finally, Origen takes matters into his own hands and addresses the forger cum conuenissem illum ipsum qui adulterauerat librum directly. Responding to Origen’s demands for an explanation, ­Rufinus replies that his changes had in fact improved and corrected errors in the original. This passage also highlights a third critical element in dispute, which, generally speaking, often falls into oblivion: the public. It is precisely by demand of the addressees of the work, the public, that the forgery is

xii

foreword

uncovered. In this example, upon reading the text, the outraged brothers indignantes fratres must have realised that its content was not correct, probably because it did not agree with the lessons received. They sent a messenger to Origen to inform him of the matter. It is clear that these brothers were not happy about being tricked, despite the supposed predisposition of the public (i.e., mundus) to being fooled.7 In this case, the public actively participated in uncovering the forgery. With this in mind, it is important to note Ehrman’s words on pseudepigraphic literature: “People in the ancient world did not appreciate forgeries any more than people do today. There are numerous discussions of forgery in ancient Greek and Latin sources. In virtually every case practice is denounced as deceitful and ill-spirited, sometimes even in documents that are themselves forged.”8 It is thus clear that forgery affected both the author and the forger, but also the public in its capacity as the final addressee of the text in question, since it is the public’s acceptance of the work that allows the forgery to survive and even take root in literary tradition. From this perspective, the relationships between spurious and genuine literature and between the forger and the author acquire a new and essential dimension: that of the final addressee of the work, since it is the public who can be tricked velis nolis, but who may also be capable of exposing the forgery. All of these considerations pose new questions over the nature and motives behind forgery. New topics of unquestionable interest also arise, however, from a more modern approach without previous ideological hindrances, such as the study of the cultural and ideological contexts in which forged texts originated or the epistemological prejudices that led to texts that had been admired and even venerated for generations to be dismissed from the canon. Is it the simple intention that invalidates them? Does forgery not deserve a worthy position in literature, as it would have had if the deception had not been uncovered? Should we banish to oblivion a valuable text with an unquestionable literary (and also social, cultural, etc.) value just because its author may have used “non-canonical” methods to remain in the canon? 7 On the proverbial sentence mundus uult decipi, cf. J. Martínez, Mundus vult decipi, Madrid, 2012, 9ff. The phrase is most probably found in Sebastian Brant, Narrenschif, Basel 1494, 65, l. 68, but it does not appear in Jakob Locher’s Latin version, which appeared in Strasbourg in 1497 with the title Stultifera navis. The first Latin quotation was probably written by Luther in 1529, cf. WA 39.40.5, cf. Thesaurus proverbiorum medii aevi, Berlin 2002: 13 s.u. Welt, 58. 8 B.D. Ehrman, Lost Christianities: The Battles for Scripture and the Faiths We Never Knew, Oxford 22005: 10.



foreword

xiii

Bearing in mind these new ideas and concepts of forgery, during our research project’s series of ongoing conferences we invited several authors to contribute their own or general opinions on the phenomenon without imposing any restrictions on them. The result is this book. Our aim in this work is to display some of the many possibilities offered today when studying the phenomenon of forgery. I would like to take this opportunity to thank all of the authors who participated for their commitment and wonderful contributions. I am also grateful to John Henderson in particular for graciously giving us his time and for providing us with such an original and inspiring prologue. Last but not least, I would like to thank the Universidad de Oviedo and the Government of the Principality of Asturias for providing the resources that made this book possible.9 J.M.

9 This publication is based on research funded by project FFI2009-09465 of the Spanish Ministry of Science and Innovation and support granted by Universidad de Oviedo (unov12-edo09) and the Gobierno del Principado de Asturias (Plan de Ciencia, Tecnología e Innovación de Asturias / FICYT CNG13-20).

Prologue/Volume Retrospect: Ergo decipiatur John Henderson A desirable aptitude for being ‘illuded’ by fiction becomes indistinguishable from that susceptibility to the overtures of confidence tricksters which is memorialized in the proverb, mundus vult decepi [sic] (“the world wants to be deceived”). K.K. Ruthven, Faking Literature (Cambridge, 2001) 148

I This new volume of papers takes its place in an ongoing series dedicated to the topic of fakedom in textuality. In particular, the allusive title is a twin for editor Martínez’ 2012 collection, Mundus vult decipi. Completing the apocryphal and complicatedly fake- and mock-ascribed Petronian saying (see Chapter 1) will prove to be a reliably authentic guide to thoroughgoing congruence between the opening salvo and the present ­follow-up for 2013. The changes in focus signalled by the divergent sub-titles “Studies on Fakes and Forgeries of Classical Literature” after “Estudios interdisciplinares sobre falsificación textual y literaria” are more apparent than substantive. The element of interdisciplinarity is indeed less pronounced in the second gathering, but the same thrust is there still in that the praxis of classical studies as a deliberately totalizing, comprehensive conglomeration of inter-related investigations must always negotiate the porosity of borderlines with whatever specialisms involved in the evolution of First World into global culture. “Authors’ pseudonyms in the seventeenth century” (see Chapter 16) challenges criteria for belonging (but writing the grammar of classical languages is first step to Parnassus); so also, from another angle, with the lexico-graphical wordpower, “Forging Ancient Greek Words in Modern Times” (see Chapter 19), and, no less fortissimo, so too “Athena and Pallas, image, copies, fakes and doubles,” busy with statuary and sculpture fudgelore – and its mytho-graphic texts (see Chapter 9). Indeed Classics’ investment in the nesting/siege of the echt “literary” within the expansive field of the “textual” is as inevitably salient in these

xvi

prologue/volume retrospect

new essays as it was when spelled out in the previous rubric: whether supplemented or not, Lucretius even names within orthodox Latin studies the parasite cuckoo “philosophy” in the literary nest; expel his didactic epic poem from “Literature,” and the chances of comprehending textuality at Rome would be good as nullified – and besides, the brawl that would arise over the mound of deserted corpora would be unending (see Chapter 2)! Plato’s dialogues, “authorless,” “bastard,” or whatever, consist of the (not so self-evident, but certainly self-self-defining) claim to share in social discourse through textualization (see Chapters 3 and 5); their place within Literature as enclave of intellectualism is staked out by hedging round with venal rhetoric othered as “sophistry,” so that the tarring of annihilated “wisdom” texts uncontroversially belongs to the business of “Classical Literature” because it is so entwined with the range of text wars that have decided what we can, let alone have to, read (and have to read; see Chapter 18). Similarly, instinctive preclusion of texts such as The Surgical Treatises of the Corpus Hippocraticum from “literary” studies would, all would agree, constitute frankly unclassical misprision, as well as butchery, and the same goes with other “scientific/technical” as with philosophical or historical or X . . . -ical writings from antiquity (see Chapter 7); historiography as sibling within the family Literature is least disputably at home, even in the case of barefaced chronography (see Chapter 10). Which classicist today would dis-recognize the necessity of syncrisis with sub-literary writing for the negotiation of Literature in its guise of valorized canon and precisely with respect to its pretensions to stand (supremely?) for the intrinsic value of Classics? “Magic-making Pseudepigraphy” is not cleanly “non-Literature” since the textual protocols and affectations at work therein ape/assimilate those governing the Greats (see Chapter 17); likewise, the bedraggled tatters of an asinine trash novel implicate the prolific oeuvre of the slickest writer of sophisticated talkshow essays on the ancient circuit precisely because it pressurizes all claims to know what makes Lucian a star of Greek literature (see Chapter 13). With lost, abjected, and marginal textuality come the texts of literary culture – more than that, they can come ready bundled together into a single form of fusion, as when the cento cuts up Homer into biblical re-make (see Chapter 6). Writings about Literature, too, are not merely to be precariously sidelined as only in attendance at the literary show since (purist tendencies within literary theory notwithstanding, and granted enthusiastic theorophobia amongst “literary” critics) the authority of authors as decisive governing distribution point of meanings and values packaged



john henderson

xvii

in textuality near-ubiquitously rules in castle Literature: “author, author” biography ranks accordingly alongside other forms of interpretative reception/critique of Literature, and (the logic has governed this unstylishly elephantine opening paragraph) shares significant textual properties with its object (see Chapter 4, etc.). This classical cosmos (mundus) means, as a main plank in its portfolio of arguments, to represent, even realise, the omni-pertinence of humanist culture, with special emphasis on its literary-textual intrication with grammato-logy. Homer, Plato, Demosthenes, Lucian, Nonnus; Plautus, Lucretius, Petronius, Venantius . . . – this is some multi-volume on “Classical Literature”; but Simias of Rhodes, Thessalus of Tralles and Pseudo-Simias and Constantine Simonides, Gaspar Scioppio, José Marchena . . . – these texts frame and bound, define and inform the collective enterprise, beggaring the reach of any plausible polymathy. The editor’s master-pun (see Chapter 10), positioning Hippias of Elis as “The Master Forger” of all time, in that a few crafted sentences from him rustled up the paradigmatic data-founded account of the emergence of Classical history as an interlinked whole, simultaneously conjures up both the constitutively interdisciplinary panoramic claims of Classics and the mix of forging (creating) and forging (faking) necessarily in play in its objects of attention and its praxis alike. II The editor has powered these twinned volumes, as we noted, within a larger and longer project (and plenty more to come). The contents roughly match up in terms of topic and, as we just observed, spread. The subtitles at once point towards a major change of strategy: Mundus vult decipi was written mainly in Spanish, with a fraction in the other major Euro-­languages; this time, Anglophony reigns. On the other hand, plenty in both volumes appeared in translation, and a more telling underlying contrast that merits flagging up is that the Oviedo conference-based and faculty-centred attraction of mostly Iberian contributions now yields to a palpably cosmopolitan collection of volunteered scholarship in the wake, and train, of the earlier effort. A hard core of overlap and reinforcement backs the symbolism of common editorship and “set-up + punchline” titulature, in the form of a scatter of returning authors and subjects: most obviously, besides the editor, Hernández reappears, pursuing further imagined lives for Nonnus (see Chapter 4), and Álvarez reprises his principal role as volume-opener and liminary paratext for the diptych by continuing his

xviii

prologue/volume retrospect

adventures in the play of eros/power in the underworld of Marchena and the “Fragmentum Petronii” (see Chapter 1). Further echoes and matches obtain, so there is a vestigial “sophomore” tinge to this “Second Album”; but this is a freestanding collection with its own momentum and intertextual activity. The repeat formatting of ordering by abecedarian initial of the contributors’ surnames (which serendipitously permitted the precedence of A-for-Álvarez) unapologetically offers each essay to take its place as a stand-alone; but at the same time opens the volume’s readers to digesting them within the overall frame set for these “Studies.” In this regard, they are encountered as instances; instances, that is to say, of “Studies of Classical Literature” (of course), but more pointedly, as recommended interpretations of “faking and forging” in their pertinence for “Classical Literature.” They offer samples of, as well as sampling, ways to imagine, conceptualize, analyse, theorize, capitalize on, commandeer as critical tools, but also to embrace, enjoy, celebrate, lionize, heroize, (even/especially) recommend. Since the 2001 classic study by K.K. Ruthven (see my epigraph), re-­ evaluation, and in particular, revaluation, of textual “falsificación” has been on the table; and gathering atttention, too, within Classical and adjacent studies, in what must be some kind of culturally symptomatic ambience, as a po-mo/post-crash manifestation of distrust in author-ity as realised through rapidly morphing, so toxically unaccountable, forms of textualization (beyond Euro-Oviedo, the two most recent Cambridge books in my own specialist discipline of Latin Literature sitting here on my desk just happen to tally with “Fakes and Forgeries”: I. Peirano, The Rhetoric of the Roman Fake [2012] beside S. McGill, Plagiarism in Latin Literature [2012]). III One further evident contrast for me to draw between Mundus and Ergo concerns my own role! Whereas I produced a first-time review of the former (online at Bryn Mawr Classical Review 2010.10.02), this time I get to antefix this brisk preview. There is no point in the move “inside” if it just means cloning or mut.-mut.-ing my remarks, so I shall summarize the methodologies, the idioms, adopted by the contributors here, closing on them as instantiations, as themselves performances within the orbit of “falsificación.” I shall say at once that the more downright ­self-­reflexive



john henderson

xix

essays serve to throw into clear profile the commitment of the rest to various and sundry blends of authenticity rhetoric. For sure you will decide whereabouts this prologue retrospect itself belongs between these two poles; the aim, however, is to tilt reading towards inter-imbrication between verism and hustle. A preview, then, of the range of approaches on display: (1) At one self-erasing extreme, the modest “statistical linguistics” persona graphs significance levels to make the “Student’s t-test” count, out to demonstrate what the method and its methodicality can achieve (Chapter 7). No messing, either, just objectivist reticence with listing, whether in the cataloguer’s elaboration of categories for exhaustive computation (Chapter 12) or in the extraction of divergence from lexis authorized by reference to oeuvre (Chapter 13). The selection of prime cases of “semantic” alongside wholesale coinages is similarly a product of recovery by exhaustive search, this time lexicographical (Chapter 19); but sending the volume off with exposure of faked-up or expropriated pseudo-Greek terms, this patch of brevity turns out to skewer a particular idolatry and the attendant myths which buy an interested collusion from classicists (keeping Greek alive, indeed!) in a wry, understated jab that hurries past – in aside, as if scarcely worth the mention – the items “anthropology” and “biology” before rubbing our noses in the crummy stories of “epistemology” and “ethnography,” hilariously set on a par with the (henceforth unforgettable) beauties “prosopagnosy” and “zootechny.” On behalf of the collective ruse enabled, not sponsored, by the classical lexica, a few “favoured” vagaries of word-mongery take a final salute. (2) No such gesture toward tinkering with the Text from the severely programmatic classical editor (Chapter 2). A wondrously committed hunt to track down “active interpolation” can leave no stone unturned, as disagreement elicits one spanking after another, for “unnecessarily sophistical special pleading. . . . a rather sophistical response . . . convoluted argumentation and special pleading . . . sheer improbability.” There shall be no chink in the armature; the authenticity pledge registers loud in the pained “I regret to say that I am not convinced.” The price paid is to suspend “literary” appreciation for the duration, as wrong ideas line up for lopping: so the one instance of “faking/forging” that survives interrogation, poor DRN 5.1006, can’t be enjoyed because it is for felling, but slipping in the flagrantly self-contained (= easily snuck out) improba nauigii ratio tum caeca iacebat is blatantly flashy naughtiness fit to make anyone despair (shall we pretend?) of the possibility of human civilisation! Think

xx

prologue/volume retrospect

of Lucretius gurgling away as he wrote the multiply “stupid” fig. etym. into his tabl-eau of the sea frustrated of its attempt to charm humanity into trouble: . . . nec poterat quemquam placidi pellacia ponti | subdola pellicere in fraudem ridentibus undis. | ! What empathetic reader wouldn’t want to join in with this – the provocative giggling and the fraud? The poet in DRN is, even, asking for it, and deserves hyper-animated mimicry in his wake. (3) On now towards the middle ground. First, counterpose the texts that swarm around powerful images of power, as their status of copies without originals sponsors and accommodates competing tutelary statues, magicking up live-evil disseminated wherever and whenever the tales are told (Chapter 9). Next, watch out for stupefying stereotypefiction! Dare to nail forcing labelling whenever an institution establishes its own ascendancy by reifying assorted rivals into its demonized double (Chapter 18). Equably, however, keep calm and re-think the great library’s starring of particular texts as “bastards” as no crude branding as bogus; rather, appreciate their acceptance as congenially compatible supplements to the authorized corpus, intratextually espousing ethos and style and untroubling the dogma (Chapter 5). (4) The staple classicist’s business of close reading of texts takes the forms here of a hard-sell scrutiny of authority, risked (apparently) to epistolary and Christian “ambiguity,” but (we guessed) ultimately destined for restitution (Chapter 15); the genial vindication of an excised moment of proscenium-busting “Selbstironie” offset by consequential need for alternative culling as the immersed critic palms his critical stance off as attuned ad hoccery while emphatically (implicitly/explicitly) campaigning against a once-potent scholarship mightily wielding the knife (Chapter 8); and – no fakery or forgery in sight – an ascription of a fragment to a virtually non-extant writer is firmly dismantled, as the details are stacked up (Chapter 14). (5) From this point fakery forges ahead, spun variously positive. Classicists can understand that a practical schooltext could merit a figleaf nom de plume whereas a pet gospel setting out radically reformed principles for a revisionary pedagogy is coloured melancholic (Chapter 16). An army of non-existent authors for non-existent, inaccessible, or generally unaccessed texts adds up to powerful listing, this time as the generically orthodox ploy of exotic names in the “cultural koinê” of magical papyri. On they march, listen!, nothing can stand in their way (Chapter 17). (6) That zero degree writer, the centonist, can beggar any Borgesian Pierre Menard’s achievement in exact replication of a classic text to significant revisionary effect: when cut-up Homer pledges “cultural innovation,”



john henderson

xxi

this, and let’s revel in the vision, arms “social revolution,” as female voicing re-tunes patriarchy to the self-empowering creed of powerlessness. Re-think, that is to say, conventional authorship; it need not be that way (Chapter 6). At length the forgers queue for hagiography. First, welcome two rogue antiheroes who conspired in their own long-documented outing: one MSS market supplier conscientiously crafts “skilled pastiche” with “a genial touch,” in amongst the howlers and the play with mischievously homonymous authorities, to deliver a needed author-bio to evince, no less, classical-Christian symbiosis, just the way novels and films imagine up for us what’s wanted (Chapter 4 – we’re in the picture at last). From here on, it’s all prose, and all cons: another forger headed, figure-headed, and ghosted for an officer’s mess full of inventive spoof-scholarship that combined expertly informed, creative, and even (perhaps!) seriously motivated political, educational, and (pseudo-)feminist annotation to their pretend supplement to the fragmentary classic of “liberated” Latin; the “erudite joke” played its subtle part in the knowing fabrication of a picaresque identity that contrived to incorporate real adventures in the library into the always already exploded career of a rogue celebration of embarrassment to Classics (Chapter 1). Our trusty editor warms to the task, empathizing with his “master forger’s subtlety and insight into the psychology of the reader.” He runs by us here and now the ancient chronographer’s “crime of intent” (in the true “Ringolevio” spirit of the classic founding text of fakery critique, S. Stewart, Crimes if Writing: Problems in the Containment of Representation [1991]), before teaching us the unprinciples of the con-man, knowing full well that this is what the world (and his dog) wants. Wicked! The essay wears – sports – its intrication with its subject and topic proudly (Chapter 10: I too had “means-opportunitymotive” when I extracted the same pretend confessional style of (XII) “commandments” for savvy commentary making, in “Oxford Reds”: Classic Commentaries on Latin Classics [2006] Chapter 1, pp. 24–33). (7) Philosophy pulls its room-full-of mirrors-stunt whenever Plato writes Socrates discussing speaking in dialogical problematizing of the presence of text that absents the writer, and when the topic of critique is Protagoras, the thematic is guaranteed to oblige us (the mundus) to “measure up” and get with it – feel the “fictionalization” of – everything fabricated by the text. Until, perhaps, the very last, saving, pietistic non sequitur, where (somehow but how?) “Socrates’ authority is indisputable because it mingles with god’s absolute authority”; the trick here is to fold reading as such into the grain of the text, to paradox-monger purposively with literariness (Chapter 3). The ultimate trick in the box of textual insider

xxii

prologue/volume retrospect

trading sucks us into the scam: “th[e] narrative about a text within-thetext is also a story about how we all read such narratives of authenticity and fraudulence – about how we are reading the text right now.” When pseudo-documentary fiction self-styles as Ruthven’s parade of factitious “illusion,” transferential mirroring of literary montage infolds “our own anxieties about how to interpret the fictions of authenticity and authorship from antiquity.” Nem. con. IV The question, then, is whether subscription to the full (pseudo-)“proverbial” title engrosses each performing contributor within its mundus – . . . or on what grounds any (self-illuded) player can think to subtract themselves from its application. Under pressure, no fiat from authors can put themselves over as hors de combat, for attentive – “literary” – readers always appreciate style as “illusory.” As any critical essay, like other species of paratext, rhymes, shadows, clashes, contraverts, meshes with the object of criticism it posits/constructs, so the topic of “falsificación” invades, or bleeds into, the authenticity stakes that govern authorship/authoriality. Writing about sincere scamps, methodical jokers, logical hoaxers, honestto-goodness crooks, and authentic wizards? Honestly! Let’s all, in any case, let go and be captivated (we all are, anyhow).

Libertine Erudition: José Marchena’s Fragmentum Petronii and the Power of False 1 Joaquín Álvarez Barrientos In his book Questions de littérature légale, Charles Nodier sets out a categorisation or typology of forgeries, one of which he labels “la supposition des passages,” or supposititious passages. These supposed passages, if contexts and paratexts prove suitably convincing, have many features which make them credible. For the most part this can be explained by the fact that forgers play on the desire of the scientific community to possess originals as well as new works by well-known writers, and also specifically because, in addition to the supposed discovery, there is a known author whose reputation and oeuvre have to be completed. The need to fill in gaps led to the invention of supposed letters by the authors of the texts edited, which were published as appendices to their complete works in order to round out their image and complete the list of their writings. It also led to a filling-in of gaps in already known but incomplete texts. José Marchena (1768–1821) and the Practice of Falsification This is precisely what was carried out by José Marchena, a Spanish literary figure who was born in Utrera in the province of Seville on 18 November 1768 and who died in Madrid on 31 January 1821. He studied Latin and Humanities in Madrid and Salamanca; in the latter university he met other important figures of the period such as Juan Meléndez Valdés, a Francophile like himself, and Ramón Salas, a cultural radical with liberal ideas. In Salas’s home they debated and translated the latest literary and philosophical texts, such as those by Adam Smith and others, authors whose writings were not part of the university curriculum. Marchena published the periodical El Observador (The Observer), which soon closed

1 This chapter forms part of the Proyecto de Investigación del Ministerio de Educación y Ciencia: El otro Parnaso: falsificaciones literarias españolas, HUM2007-60859/FILO. It was translated by Begoña Lasa. The traslation of this paper would not have been possible without the disinterested help of my good friend Professor Philip Deacon, University of Sheffield.

2

joaquín álvarez barrientos

after clashes with the censorship apparatus; he was also persecuted for other works for which he was responsible until he finally fled to France to support the Revolution. There he worked as a propagandist for newspapers and pamphlets which were introduced clandestinely into Spain, an activity that led to a his recruitment for the French Foreign Ministry. He remained in France, suffering the vagaries of fortune, until he returned to Spain as part of Napoleon’s army in the role of secretary to General Murat. An untiring propagandist, Marchena edited the Correo Político y Militar of Córdoba and prompted Joseph Bonaparte’s expedition to Andalusia. After the defeat of the French forces, he abandoned Spain, remaining in France until 1820, when a liberal government came to power in Spain; he died in 1821. He translated Molière, Rousseau, Voltaire, Ossian, and Volney, among others, and wrote important literary and philosophical works, such as an Essai de Théologie, defending atheism, and Lecciones de filosofía moral y elocuencia, which includes an outstanding political history of Spanish literature. More importantly, he produced forgeries, such as a fragment of the Satyricon of Petronius, which I will treat later. As far as we know, this example of learned libertine literature was not repeated in any of his other works, but his interest in forgeries remained. For example, he prepared the Ossian poems by Macpherson for publication in the journal Variedades de Ciencias, Literatura y Artes in 1804, although they had already been translated and published, and it would appear that his version had notes. To date, the text has not resurfaced, but it provides proof of his interest in the activities of the Republic of Letters; it was also perhaps an attempt to impress Napoleon, who proclaimed himself Emperor that same year and who enjoyed the poetry of the supposed Gaelic bard. Rather different is his pseudo-Catullus, which appeared in 1806 under the title Fragmentum Catulli, ex Parcarum carmine fatidico, in poemate de Thetidis ac Pelei nuptiis; the work lacks publication details but was first published in Paris; in 1807 and 1808 two further editions appeared in Jena and Vitoria. According to the 1806 Journal Générale de la Littérature Française, the Catullus text was printed by Dabin and Firmin Didot. The 1807 edition was the work of Heinrich Karl Abraham Eichstaedt, a professor from Jena, who added 20 new lines of his own in addition to pointing out errors in the fragment. The Vitoria edition exhibits certain variants with respect to the 1806 text while asserting its authenticity in the following note: It would be pointless to explain where or how this fragment came to light: the style makes clear that it is not false. The first line of verse of this fragment in the ancient manuscript containing it was placed after the one in



libertine erudition

3

which the Fates predict the sacrifice of Polyxena: Alta Polixenia medefient caede sepulcra” (Spanish text in Manchón Gómez 91–92).2

As in the case of the Satyricon, this fragment from Catullus reconstructed a lacuna in the “Epithalamium of Peleus and Thetis,” but Marchena’s aim here is not to deceive, but, more probably, to eulogize Bonaparte, since he introduces into the lines various allusions to the Emperor and his victories in Prussia, Egypt, and Italy. It was in October 1806 that Napoleon had made his triumphal entry into Jena, the city where the deception of the pseudo-Petronius was discovered. Marchena’s forgery thus served both political and propagandistic aims. The work of the pseudo-Petronius survived in fragments. Catullus had been the object of numerous readings and amendments in attempting to establish the text, something which made it an easy target for forgers and plagiarists. Marchena himself had played this role in his Fragmentum Petronii since he had included in it his own version of “Ad Lesbiam,” as François Noël demonstrated in 1803 in his Traduction complète des Poésies de Catulle, suivie des Poésies de Gallus et de la Veillée des fêtes de Vénus, avec des notes . . ., les parodies des poètes latins modernes et les meilleures imitations des poètes françois. Marchena joined this world when the version of “Ad Lesbiam” was incorporated by Noël into his edition. In order to produce the new work, it is possible that Marchena made use of Noël’s own text or the edition of Catullus published in 1738 by Corradino Dall’Aglio. It is interesting that Corradino Dall’Aglio’s edition may also have been apocryphal, since he claimed to have found an ancient manuscript in Rome with very good texts of Catullus’s verse (Manchón Gómez 102). In this edition, following practice which was almost the norm for falsifiers, he fills in gaps and carries out his annotation so that text and notes support one other, just as Marchena had done in the Fragmentum Petronii. In fact, Eichstaedt underlined with some irony the debts and connections Marchena had with Corradino dall’Aglio. The Erotic and Erudition: The Meaning of the Notes in the Fragmentum Petronii The Fragmentum Petronii, issued without publication details in 1800, theoretically fits the pattern or type of justification which critics and ­historians

2 All translations are my own.

4

joaquín álvarez barrientos

conventionally claim to be the aim and objective of their professional endeavours. However, in this particular case the focus is faulty, or at least not in the right place. The book is comprised of three parts: the first is a dedication to the French armies; the second is the Latin fragment and its French translation; and the third consists of six scholarly notes relating to the fragment. Johann Decker (the printer of the text, as we shall see below), Frédéric Schoell (239–240), and Joseph-Marie Quérard (I: col. II) claim that the erudite notes were composed first and that, in order to publish them as a complete whole, Marchena made up the fragment which linked two episodes of the novel by Petronius. In fact, when writing about the Fragmentum, only the data intended to explain and clarify Petronius’s Satyricon were taken into account, while Marchena’s text was ignored. The Latin section of the book was the only part investigated, and the relevant evidence indicating that this text was written after the accompanying six explanatory notes was not taken into account. The fact, pointed out by people close to the events such as Decker, the editor of the fragment, is crucial because it provides relevant clues as to how the forgery should be understood and about the forger’s intentions. Just as one can speak of the intentions of the author – admittedly a controversial issue for some – one can also speak, perhaps with greater confidence, of the intentions of the forger, since generally the forger acts in opposition or responds to something objectively known. In both cases, and due to the passing of time, it is in reality more a question of the intentions of the texts as much as the readings of them. Thus, as far as Marchena is concerned, his purpose was not to compose a fragment to fill a gap in Petronius’s narrative – it was not to contribute to making the text known, as was the case for Nodot and others – but rather to offer to a select, erudite and enlightened readership a treatise on the history of sexual behaviour in antiquity. The information on how the forgery was effected gives an idea of its goals, which were to offer a new reading of the historical past, recover Rousseau’s concept of desire as something natural, and promote a type of individual who does not recognize the ideas of sinfulness or punishment in the realm of desire but fully accepts its diverse manifestations. Moreover, such ways of understanding love and the various figures who contributed to such an objective are seen as the expression of the development of civilization, a fact that the author emphasizes in particular when referring to courtesans and go-betweens and narrating the evolution of methods of seduction. It is true, however, that one of the features that does not change over time in such relationships is that they are conceived



libertine erudition

5

of in terms of power. The historical novelty highlighted by Marchena is the way this power is expressed, from its initial manifestations in the form of violence and rape to more civilized ones in which women are allowed to choose. In presenting this trajectory he also reveals the slow, gradually increasing presence of women in society, to the extent that one could see the explanatory notes as fragments of a social history of behaviour and customs. There are various reasons why women constitute one of the underlying focal points of the work. Apart from other considerations and matters specific to Marchena, one of the features which regularly characterizes the work of forgers is that they project onto their creations the concerns and methodology of their present circumstances. In the second half of the eighteenth century there was much questioning of the ways in which men and women related to one another, and new alternative forms came into being. Similarly, in the most modern literature of the period love and desire acted as scalpels for dissecting reality. Women were the ones who used these instruments in fiction, allowing authors to perform autopsies on society. Marchena is unable to escape this tendency, and makes use of desire and human relationships to construct his own history of that society, which takes on significance because of the sexual encounters as well as his exploration of sexual practices which were persecuted and condemned. This purpose has remained hidden because of the nature of the explanatory notes and because of the conservatism of critics and their focus of interest on literary issues. Consequently, Marchena’s insertion continued to be reprinted with other Latin texts, in full knowledge that it was a fake. What is more, the author’s image contributed to the fact that the scholars who studied the text misinterpreted its origins and read it in a probably incorrect way. Given that Marchena was a fine Latinist and a radical thinker, the fragment was generally interpreted as an erudite joke and, at best, as in the case of Menéndez Pelayo, as evidence of the author’s talent. In spite of the amount of critical work devoted to the text, there are issues still unresolved. Certain points, such as the place and date of the edition and the printer’s name, have been clarified. The Fragmentum Petronii was published in Basel by the printer Decker in 1800. The brief text appeared clandestinely without imprint, doing justice to a style of erudition which meshed perfectly with the festive, libertine, heterodox, essayistic literature that over many years emanated from Cosmopolis, a land which apparently gave rise to numerous editions with false ­printing

6

joaquín álvarez barrientos

details (Eisenstein 1997). Even though these aspects have now been solved, authorship has still not been clarified. Cultural history assumes that the author is José Marchena, yet there is evidence that, in the notes at least, various hands played a part. This raises the problem of the sources used by these military men who spent their leisure in salons. Schoell and Quérard claim that the notes, initially five, though the last is shorter, were written by a group of friends, and that Marchena is the author of the Latin text. This is not surprising given the fact that at the time it was common for one writer to take the credit although various others had contributed. Equally, the contribution of various hands to the annotation of a text was by no means exceptional. On the contrary, editions of classical texts cum notis variorum, that is, with notes by various scholars, were commonplace and, in Petronius’s case, quite standard (Grafton 1990). Marchena and his friends did not create a parody in their learned annotation, like Pope, Swift, Arbuthnot, Parnell, and Gay did as members of the Scriblerus Club, ridiculing pretentious erudition and academic jargon, but they did use the notes in a different way.3 There is no signature in this case, but from Decker onwards all bibliographers have made Marchena responsible for the apocryphal text. Given the thematic coherence and style of the scholarly notes, he may have been the one who wrote the work and gave form to the facts and ideas suggested by the rest of the contributors. There are actually various issues which go beyond the specific subject matter of each of the notes and which extend across, interlinking them. Such an effect would have been difficult to achieve without one thinking mind and a single hand to write and connect them. The end result is a scholarly artefact against the grain of the customs and the erotic norms and religious morality of the period. It is the product of a joyful and uninhibited use of erudition which makes facts not only valuable in themselves but also through the ideological discourse created; it makes links, uniting a passage of the Bible with a quote from Homer, Ovid, or Suetonius. It is a discourse similar to narrative not only because of the diachronic treatment given to the different practices under discussion but also because of the use made of anecdotes. The the Catholic Church is attacked by criticizing its representatives and denouncing their hypocritical behaviour. The praise of courtesans is a way of exalting the intelligent woman – as is also the case of the

3 On the history of footnotes see Grafton (1997). On the pseudonymous Martin Scriblerus, see Levine (1991) and Álvarez Barrientos (2006).



libertine erudition

7

g­ o-between – who mobilizes and improves the society in which she lives by the exercise of public and private virtues. It also serves as a critique of celibacy and concubinage directed at clerical figures who do the opposite of what they preach. It is evident, on the other hand, that the reader addressed in the text is a select reader, capable of identifying the allusions and characters present in its pages. Among the main sources employed by this literary circle while the army was quartered for the winter in Basel, we encounter such authors as Homer, Horace, Ovid, Plutarch, Plato, Suetonius, Tibullus, Vale­rius Maximus, and Virgil, but also works such as Pierre Bayle’s Dictionnaire historique et critique, Diderot and D’Alembert’s Encyclopédie, and the Bible. The use of classical sources is wholly justified by the examples they provide of all kinds of anecdotes and practices of a sexual nature. The Encyclopédie provides support for at least one specific case, where the author documents a moral as opposed to physical interpretation of virginity. Bayle’s dictionary and the Bible are used in a more belligerent way, not only to supply examples but also to support the negative critical view of the Church and the Catholic religion. When quoting the Bible, the author uses Christianity’s basic text as a weapon against Catholicism, demonstrating how much of what Catholicism rejects is to be found in its pages. He uses the same technique to compare characters from the Old and New Testament (especially the Old) with representatives of the Catholic Church, always to the detriment of the latter. The repeated use of this particular source, while evoking the kind of reader he is addressing, is justified as follows: On nous reprochera peut-être de citer trop souvent la sainte Bible; mais ce livre contient la science du salut, et ceux qui veulent le faire ne sauraient l’étudier avec assez d’assiduité. Nous avouons que cette étude a occupé une bonne partie de notre vie, et que nous avons toujours trouvé du profit à y faire. Cet aveu peut paraître ridicule aux esprits forts; nous n’écrivons que pour les ames pieuses, et elles applaudiront volontiers à cette profession courageuse de notre piété (Marchena 60).4 We may perhaps be criticized for quoting the Holy Bible too much, but this book contains the science of salvation, and those who wish to achieve it have only to study it assiduously. We must confess that such study has

4 Quotations from Marchena’s Fragmentum Petronii are taken from the original French edition published in 1800. This text was translated into English by W.C. Firebaugh (Petro­ nius 1912), who included it in his edition of Petronius’s Satyricon, into Dutch (Smarius 1996) and into Spanish (Álvarez Barrientos 2007).

8

joaquín álvarez barrientos o­ ccupied a great part of our life and that we have always found it profitable. This confession might sound ridiculous to independent minds, but we write only for the pious who will strongly applaud this bold profession of our piety.

The juxtaposition of erudition serves to add authority to the opinions expressed, appearing to give both types of text equal validity, but the new knowledge thus established creates its own ideology. By adding one thing to another the effect is to assert the moral, pagan line of thought and to undermine the heritage of Christian morality. The biblical narratives are placed on the same level as the myths and legends of classical antiquity; the author points out that the Homeric texts and those of the Bible are the most similar “parce que les peoples qu’ils décrivent et les hommes dont ils parlent, sont parvenus presqu’au meme degré de civilisation” (“because the peoples described and the individuals of whom they speak have reached almost the same level of civilization”; Marchena 27). The notes, which in reality constitute the body of the text, use the rhetoric of scholarship as well as certain fictional devices to create a treatise on scientia sexualis rooted in materialism, differentiated from works in the ars amandi tradition. This treatise is in turn legitimated by other authorities: Diderot, D’Holbach, and Spinoza, all theorists espousing varying degrees of philosophical materialism. What their works highlight is the defence of freedom – identifying sexual desire and freedom to exercise it as a general desire of humanity – limited by the control of that desire. This is presented as a documented historical reality in such a way that two versions are lined up against each other, one as a desideratum and the other as reality, representing the struggle between the two tendencies of humanity – freedom and control. For this reason the presence of the notes is quite essential. I have already suggested that they constitute the kernel of the story. If Marchena had gone somewhat further he might have only published the notes, as Gottlieb Rabener did in 1743 with his satire “Hinkmars von Repkow Noten ohne Text” (Notes Without a Text), in which he condemns the ignorant who add notes to the work of others in order to achieve fame (III: 6). If that had been the case here, we would merely be discussing a parody of the misuse of scholarship, but in Marchena’s case they are used in the traditional way and in part to validate a text, although the text is neither scholarly nor historical, but fictitious and false. The result is that the notes apparently validate an apocryphal text, which in turn provides a platform to display the message lying at the heart of the notes. In this way he establishes a link with the fictional pornographic world, which customarily validates its



libertine erudition

9

scenes by the use of annotation. On the other hand, notes generally provide help when determining meaning. This is the case here although the apocryphal writer orientates the reader in the appropriate direction, the one established by the note. It may be the case that the scholarship which Marchena exhibits in the notes and the notes’ length are intended as a parody of the working methods of scholarly academics, but it is equally possible and even probable that we are being presented with a treatise on erotic and sexual behaviour. In fact, although the notes are officially linked to the fragment, they can be read totally independently. However, if the model of Bayle’s Dictionary was indeed present in Marchena’s memory, with its amusing notes, sexual allusions, and pornographic interpretations of biblical extracts, it is also possible that the example of Edward Gibbon’s Decline and Fall of the Roman Empire (1776– 1788) provided a supporting role. The use Marchena makes of footnotes is quite close to that of the British historian. In both there are notes which do not only or in any sense at all constitute bibliographical references but contain, in essayistic form, ideas, criticism, or comments. In fact, although both authors use mockery and irony, their reference point is the tradition of ecclesiastical scholarship with its tendency to distinguish what is true from what is false. And it is unquestionably the case that in this text Marchena would have been able to gain the approval of the academic world, although his inverse usage of knowledge – legitimizing the false – carried with it a mockery of the standard control mechanisms of the human sciences. We can reasonably conclude that, not only because of the literary quality of the fragment and the high level of scholarship, the false Petronius is of interest for the history of the footnote. Context and Paratext If the notes are interlinked by lines of argument which go from one to the next to give a sense of unity, something similar occurs with the fragment, but in relation to the novel of Petronius. Marchena used words and phrases from other known fragments as well as alluding to previous and succeeding episodes and characters in order to give coherence and produce a sense of authenticity. This technique, which fits a strategy of creating suitable contexts and paratexts, was what convinced people for quite a long time that it should be taken as genuine, but was also, given the nature of the notes, the trigger which set in train the process by which it was recognized as false. The reviewer of this literary novelty in the

10

joaquín álvarez barrientos

­ llgemeine Literatur-Zeitung for 1801, though thinking it genuine at first, A highlighted the fact that, in a way typical of a literary dilettante, the notes in no way contributed to literary scholarship. It is likely that the expressive tone and character of the notes made him nervous, but what certainly rang alarm bells, as he made clear in his second review, was the fact that the fragment did not continue with the text by Petronius but with that of Nodot. In the end, he declared in the third review that the find was false with the nature of the fraud was now unveiled. In November of the same year, in a newspaper of a city near Jena, where the Allgemeine Literatur-Zeitung was published, the publisher Decker revealed that the author of the fraud was a Spaniard called Marchena, a friend of Brissot steeped in Latin cultural history, and someone who had played an important role in the French Revolution (Álvarez Barrientos, 2007: 25). Marchena produced a cultural object which linked falsehood and authenticity. If the fragment is his own invention, the scholarship is genuine and helps to make it believable; it is subservient to the falsity, and both features support an already expressed idea. As others were to do, both before and after, he created a web from what was true which erased the borders between fiction and reality, between the false and the genuine, all to serve the intentions he pursued. The effect of authenticity was created internally, supported by the paratext which explained the finding of the fragment and the dedication to the Army of the Rhine. In it, however, among other ironies, he inserted the claim that one could not doubt its authenticity and that “it is impossible to believe that this fragment is false” (Álvarez Barrientos 2007: 7). And Marchena could certainly have thought that his subterfuge concerning the find was convincing since it was based on, or rather copied from, something that had happened to him a year or two before, in 1798. According to the dedication, what Marchena found in the library of the monastery of Saint Gall – about 160 km from Basel – was a manuscript on vellum with texts from the eleventh century by Gennadius of Marseille in which he discussed the duties of priests. Under this writing he found, half-erased, the remains of another text which, when deciphered, proved to be by Petronius. Manuel Díaz y Díaz comments that the idea of simulating a palimpsest “came to him, no doubt, as a result of his research plan, put forward unsuccessfully in 1798, concerning a re-written codex from the Bibliothèque Nationale in Paris which contains written beneath it the Code of Euric” (1968: cvi). Thus, when inventing the necessary circumstances to give credibility to his finding, Marchena went back to



libertine erudition

11

the details of a genuine find which had occurred two years earlier in the French library. Andrés Muriel, on referring to Marchena in his Historia de Carlos IV, recounts the 1798 episode, in which a research grant was requested from the Spanish monarch in order to investigate matters relating to Spain based on materials in the Bibliothèque Nationale in Paris. In his submission Marchena referred to some of the manuscripts related to his research, among them ones treating Visigothic laws: inéditas y absolutamente desconocidas hasta ahora, que se leen en un códice del siglo VII, donde están las obras de San Jerónimo y Genadio, De viris illustribus. Estas leyes se hayan esparcidas en quince o veinte páginas, desde la 71 hasta la 144, y aunque se han raspado y sobre el mismo pergamino se han escrito los dos tratados citados, sin embargo, muchas de esas leyes son aún legibles y preciosísimas por su antigüedad (Muriel 197). unpublished and totally unknown until now, which can be found in a seventh-century codex containing De viris illustribus by Saint Jerome and Gennadius. These laws are scattered over some 15–20 pages, between pages 71 and 144, and although they are scraped and both treatises have been written on the same vellum surface, many of the laws are still legible and most precious due to their antiquity.

Marchena then explained the interest of these texts for Spanish history, but his request was rejected owing to the negative report sent by José Nicolás de Azara to the Minister Saavedra, in which he pointed out, among other things, that Marchena had written a book defending atheism, a reference to his Essai de Théologie of 1797.5 As can be seen, the scholar describes the feature of the manuscript in this genuine case; two years later he would use this type of description to make credible his fabrication: “in those ignorant ages it was customary to write ecclesiastical books on codices which contained the works of the best authors of classical Roman times” (Álvarez Barrientos 2007: 74). Marchena introduced buffers between himself and his find, as many other forgers of supposed texts do, and in a similar fashion the whole fabrication blew up in his face when a police investigation started and, among other things, people asked to see the original. 5 Menéndez Pelayo (163–164) argues that if Marchena had been able to work on these manuscripts, he would have anticipated Knust, who copied the isigothic fragments in 1828, and Bluhme, who published them in 1846. He also says that “paying attention to the palimpsest from Paris and having grasped its importance in 1798 are factors which undoubtedly demonstrate Marchena’s undoubted intelligence, as long as his encyclopedic monomania did not affect his judgement.” See too Fuentes (179–180) and Manchón Gómez (82–84).

12

joaquín álvarez barrientos

The Force of Non-Material Powers and the Power of Falsehood Because of their influence on literary culture, the works of Marchena, especially the Fragmentum Petronii, highlight the importance of certain non-material powers in our culture. One of these powers is literary tradition, which survives the passage of time and the various readings to which scholars expose texts. Readings which update texts and liberate them from the author’s intentions emphasize the potentialities or intentions of the works, which become evident with each new reading or different interpretation applied to them (Eco). I believe that this is what Borges perceived when he spoke of the versions of Don Quixote by ­Cervantes and Pierre Menard as the same, but different, in order to demonstrate the identity between reading and rewriting. An interpretation can be an essay, but also a rewriting, a sequel, or a the filling of a gap. But do these interpretations alter the value or meaning of the work? Did Marchena alter the meaning of the Satyricon or produce a different work with distinct aims, but still relying on the prior work by Petronius? Literature is truth, and, in addition, creates reality from the moment when characters, situations, or texts escape their frameworks to name realities external to them. Literature offers reflections of reality, but also models of behaviour and models of truth, although these models are imaginary. To attempt to legitimate them by way of notes, by using a body of knowledge whose prestige is generally accepted, is a strategy which indicates how people want to reinforce this literary truth, or truth in literature, as a respectable cultural reality. The desire to rewrite literature or history, as happens nowadays with war games in which the player produces sequels or endings which are different from those which really happened, is an attitude parallel to the rewriting of history or literature in texts having different formats. But although there may be new media – just think of the case of El hacedor by Borges and by Fernández Mallo – the practice is not new. Rewriting a work, continuing it or completing it, is common in literary history, partly due to the need to see the narrative or human cycle closed, and partly, perhaps, in the case of rewriting, because to do so may be an indication of a possible disagreement with the known text. Literary forgery can sometimes be disguised as rewriting, sometimes as a continuation, in the myriad manifestations which a genre assumes in the history of literature. From this point of view, literary forgery, in confronting the prestige and closed, unalterable condition of the text which serves to protect it, would



libertine erudition

13

be a way of vindicating creative liberty, since it can alter the fortunes of already closed works, as well as creating other apocryphal ones. ­Marchena required freedom to express his desires and accordingly constructed a discourse which sought protection in the prestige of knowledge; he used a text valued by the scientific community, but he rewrote and invented it. The strength of his forgery lays in the power of the fragment, which was frequently reprinted in spite of its falsity, not in the scholarly notes, which fell into oblivion. At all events, if literature is a non-material power, so is false or forged literature, which like falsehood in general has the power to inspire both actions and beliefs. References Álvarez Barrientos, Joaquín (2006), Los hombres de letras en la España del siglo XVIII. Apóstoles y arribistas, Madrid. —— (2007), Fragmentum Petronii. Edición, traducción y notas de Joaquín Álvarez Barrientos, Sevilla. Díaz y Díaz, Manuel (1968) (ed.), Petronius. Satiricón, I, Barcelona. Eco, Umberto (2005), Sobre literatura, Barcelona. Eisenstein, Elisabeth L. (1997), The Printing Press as an Agent of Change, 2 vols., Cambridge. Fuentes, Juan Francisco (1989),  José Marchena. Biografía política e intelectual, Barcelona. Grafton, Anthony (1990), “Petronius and Neo- Latin Satire: The Reception of the Cena Trimalchionis,” Journal of the Warburg and Courtauld Institutes, 53: 117–129. —— (1997), The Footnote: A Curious History, London. Laes, Christian (1998), “Forging Petronius: François Nodot and the Fake Petronian Fragments,” Humanistica Lovaniensia 47: 358–402. Levine, Joseph M. (1991), The Battle of the Books: History and Literature in the Augustan Age, Ithaca and London. Manchón Gómez, Raúl (2006), “El abate Marchena, traductor y escritor latino (pseudoPetronius, pseudo-Catullus),” in José Luis de Miguel Jover and Alejandro Jiménez Serrano (eds.), Maestro y amigo. Homenaje al profesor Juan Jiménez Fernández, Jaén: 65–106. Marchena, José (1800), Fragmentum Petronii ex Bibliothecae Sti. Galli Antiquissimo mss. Excerptum, nunc primum in lucem editum. Gallice vertit ac notis perpetuis illustravit Lallemandus. S.Theologiae Doctor, n.p.n.p. Menéndez Pelayo, Marcelino (1942), “El abate Marchena,” Estudios y discursos de crítica histórica y literaria IV, Madrid: 107–221. Muriel, Andrés (1959), Historia de Carlos IV, Carlos Seco Serrano (ed.), Madrid (BAE, 114 y 115). Nodier, Charles (1828), Questions de littérature légale, Paris. Noël, François (1776) (ed.), Traduction complète des Poésies de Catulle, suivie des Poésies de Gallus et de la Veillée des fêtes de Vénus, avec des notes . . ., les parodies des poètes latins modernes et les meilleures imitations des poètes françois, par . . ., 2 vols., Paris: Léger et Rémont, 1803, 2 vols., Rabener, G.W. Satiren, III, Bern. Petronius (1922), Satyricon: Complete and Unexpurgated Translation by W. C. Firebaugh, in Which Are Incorporated the Forgeries of Nodot and Marchena, and the Readings Introduced into the Text by De Salas, New York. Quérard, Joseph-Marie (1869), Les supercheries littéraires dévoilées, 3 vols., Paris.

14

joaquín álvarez barrientos

Schoell, François (1808), Répertoire de littérature ancienne ou Choix d’auteurs classiques grecs et latins, d’ouvrages de critique, d’archéologie, d’antiquités, de mythologie, d’histoire et de géographie anciennes, imprimés en France et en Allemagne: nomenclature de livres latins, françois et allemands, sur diverses parties de la littérature: Notice sur la stéréotypie, 2 vols., Paris. Smarius, A. (1996) (ed.), Pseudo-Petronius het Fragmentvm Petronii van José Marchena, Amsterdam.

Lucretius auctus? The Question of Interpolation in De rerum natura David Butterfield Two of the three greatest epics extant in Latin literature were apparently not completed by their authors: Lucretius’ De rerum natura and Virgil’s Aeneis: although they fulfilled their intended six- and twelve-book scopes respectively, they evidently did not receive the ultima lima from their authors, owing to Virgil’s certainly and Lucretius’ probably premature death. By contrast, Ovid’s Metamorphoses, the third great epic poem of Classical Rome, shows no sure sign in its fifteen-book form of authorial incompletion, despite its author’s later complaint that this was indeed the case.1 In the textual and literary criticism of the magna opera of Lucretius and Virgil, the scholar is therefore required continually to take care, insofar as is possible, to distinguish between the text left by the author (the editor’s goal) and the text ultimately intended by the author (an unreachable but not uninteresting goal) when analysing the text inevitably corrupted in transmission (the paradosis). When faced, however, with apparent infelicities in phraseology, consistency, and the progression of narrative or argument, this task of distinction is often extremely difficult and ultimately uncertain. The rich biographical tradition available for Virgil has allowed scholars for centuries to work confidently with the clear understanding that his Aeneis required further revision, perhaps over multiple years.2 In the case of Lucretius’ De rerum natura (henceforth DRN), the state of incompletion, declared without qualification in the first sentence of this paper, has been occasionally disputed. Some scholars have claimed that the text of the poem as we have it, although undoubtedly lacerated and corrupted by almost one thousand years of transmission from its composition to the

1  In Tristia 1.7 Ovid claims that his exile prevented him from giving the work its final polish (28 summam . . . manum, 30 ultima lima) and that he wished to destroy the poem in the fire (15–20), therefore announcing quicquid in his igitur uitii rude carmen habebit, / emendaturus, si licuisset, eram (39–40). This assertion could, however, be an attempt to align the fate of the Metamorphoses with the august tradition of Virgil and his Aeneis. 2 Cf. the comment of Julius Hyginus on Virgil’s supposedly intended but unimplemented corrections (apud Gell. 10.16.11) correcturum fuisse Vergilium putat, nisi mors occupasset.

16

david butterfield

survival of our earliest manuscripts (of the ninth century),3 nevertheless stems from a once-complete work: any evidence apparently opposing that conclusion is the result of either chance misfortune in scribal transcription or the purposeful intervention of later readers.4 The case for the latter of these factors – active interpolation in Lucretius’ poem – will be the central subject of this survey: do such fake or forged verses lurk amidst the true? The notion of spurious verses in Lucretius’ great Epicurean poem has been known to its critics since the Renaissance5 but did not merit anything approaching a systematic theory until the turn of the nineteenth century, when Heinrich Eichstädt suggested that the whole of Lucretius’ work had been improved and polished by a later author.6 Albert Forbiger subsequently took up this theory, although he claimed instead that this

3 Lucretius’ text survives in two complete codices preserved at Leiden (O: Cod. Voss. Lat. F. 30, s. ix1/4; Q: Cod. Voss. Lat. Q 94, s. ix3/4); some 45% of the text exists in Copenhagen and Vienna (Copenhagen Gl. Kgl. S 211 2o, s. ix3/4 [= 1–2.456]; Vienna ÖNB Lat. 107 ff. 9–18, s. ix3/4 [= 2.642–3.661; 6.743–1286]), which schedae (S) show shared descent with Q. OQS allow the reconstruction of an archetype (Ω, c. s. viii), which is the farthest back that the stemmatic method can take the Lucretian scholar. Over fifty Italian manuscripts of the fifteenth (and sixteenth) centuries survive, but their ultimate source, the lost codex Poggianus (π, copied on the orders of Poggio Bracciolini in 1417), was transcribed from a lost apograph of O (which I term χ); they therefore bear no textual independence from O, which of course survives complete. (For summary of this once-contentious point, see my review of Enrico Flores’ third volume of Lucretius [Naples, 2009]: Gnomon 83 [2011] 597–608.) The new Oxford Classical Text of Lucretius that I am preparing will therefore draw upon the Itali solely as a source for conjectures. 4 The case that the work was complete was made most forcefully by J. van der Valk in his 1902 Kampen thesis De Lucretiano carmine a poeta perfecto atque absoluto, which maintained that nearly all apparent signs of imperfection were the result only of the poem’s manuscript transmission. A less extreme, and more subtle, thesis along similar lines was proposed by J. Mussehl, De Lucretiani libri primi condicione ac retractatione (Diss., Greifswald, 1912). The third Teubner editor, Josef Martin (Leipzig, 1934. 19695), argued in a lengthy article (“Lukrez und Cicero,” WJA 4 [1949–50] 1–52, 309–29) that the poem’s total completion is proved by the elaborate architectural structure evident in several books, often of supposed correspondences in the number of verses for large passages involved. However, these claims are illusory, being extremely difficult for a reader of the poem to detect and subjective in their posited division; Martin was even reduced to claiming that Book 3 has no such numerical balance because it was composed at an early stage when Lucretius had no interest in such correspondences! In reality, Martin’s entire numerical construct is too elaborate to be credible. 5 Cf. n. 43 below. 6 H.C.A. Eichstädt (ed.) T. Lucreti Cari De Rerum Natura Libri Sex. Vol. 1 (Leipzig, 1801) LXXVII ff. (subsequent volumes after the first never appeared).



lucretius auctus?

17

reviser was incompetent and largely spoilt Lucretius’ poem;7 the presence of such interpolated verses was duly accepted by Karl Lachmann in his magisterial commentary.8 The high-water mark in the century following Lachmann was reached in two German dissertations of the 1870s, those of Friedrich Neumann9 and Carl Gneisse,10 which maintained that many scores of verses were either spurious repetitions or original compositions inserted by a later, non-Lucretian hand. In their wake the pendulum swung in the reverse direction, and scarcely any editor in the subsequent hundred years accepted large-scale interpolation in DRN. Direct treatment of the question of interpolation was indeed left almost entirely alone by critics until its sudden reawakening at the hands of three scholars in the latter half of the twentieth century. Gerhard Müller, in a two-part article surveying the history of this problem, proposed that some 270 transmitted verses were not from the hand of Lucretius.11 Konrad Müller, in his provocative edition, athetised a total of 223 verses from the poem.12 Finally, the most detailed treatment of the topic to date was provided by Marcus Deufert in his detailed monograph of 1996, which rejected as spurious 368 verses from 92 passages in DRN.13 These contributions show how much is at stake: the level of interpolation alleged – nearing 5% of the whole poem – exceeds that posited for almost any other work of classical ­literature.14 It will be seen, however, from what follows that I believe these allegations of wide-scale interpolation to be largely incorrect. I shall therefore wish to side with many eminent scholars of the last century, including Hermann Diels, Richard Heinze, and Cyril Bailey, who were loath to acknowledge active interpolation throughout the work.  7 A. Forbiger, De T. Lucretii Cari carmine a scriptore serioris aetatis denuo pertractato dissertatio philologica et critica (Leipzig, 1824).  8 K. Lachmann, In T. Lucretii Cari De rerum natura libros commentarius (Berlin, 1850).  9 F. Neumann, De interpolationibus Lucretianis (Halle, 1875). 10 C. Gneisse, De versibus in Lucretii carmine repetitis (Strassburg, 1878). See also, in a similar vein, A. Kannengiesser, De Lucretii versibus transponendis (Göttingen, 1878) and W. Lohmann, Quaestionum Lucretianarum capita duo (Braunschweig, 1882) Cap. 1: De repetitionibus. 11 G. Müller, “Die Problematik des Lucreztextes seit Lachmann,” Phil. 102 (1958) 247–83 and Phil. 103 (1959) 53–86; cf. also id., Die Darstellung der Kinetik bei Lucrez (Berlin, 1959) along with Anghang IV. 12  K. Müller (ed.) T. Lucretii Cari De Rerum Natura (Zurich, 1975). 13 M. Deufert, Pseudo-Lukrezisches im Lukrez: die unechte Verse in Lukrezens ‘De Rerum Natura’ (Berlin, 1996). Of this large figure it should be noted that Deufert, himself greatly influenced by Otto Zwierlein, only suggested the deletion of twenty verses at his own instigation. 14 For an overview of the history of interpolation-hunting in DRN, see Deufert (as n. 13) 15–26.

18

david butterfield

Three key issues are thrown up by the question of spurious verses: whether the poem was completed by Lucretius, whether he was capable of possible ambiguity or imprecision in his phraseology, and whether the text could have contained genuine inconsistencies on its first ­circulation.15 The hardened practitioner of Interpolationsforschung argues that Lucretius worked his poem up to a level of finish and perfection and as a result produced a wholly consistent and complete unit. It will therefore be essential to investigate the poem’s state of completion in further detail before turning to the yet thornier question of interpolation. We may thus begin by laying out the primary pieces of evidence that DRN was indeed not completed by its author. We should first run quickly over the few possible data that exist about the work’s composition. No biographical information survives about the poem and its creation (beyond the most tentative inferences possible from the text itself) from contemporary sources: indeed, it is not until Jerome’s revision of Eusebius’ Chronicon, carried out around ad 380, that any explicit evidence emerges on this matter. Jerome’s brief note (inserted at his own prompting) makes the claim not only that Lucretius composed the poem during sane intervals of his madness (per interualla insaniae) and committed suicide (propria se manu interfecit) but also that Cicero “corrected some books” of the poem (aliquot libros emendauit). These claims, even if certain aspects can be traced back to a Suetonian or similar source no longer extant, have been rightly dismissed as fictitious, originating most probably as malicious anti-Epicurean rumours which a Christian writer such as Jerome would have been all too willing to perpetuate.16 The statement about (surely Marcus Tullius) Cicero’s editorial involvement in the poem’s early history receives no positive support from either the direct transmission of DRN or any external evidence, in Cicero or other authors. Most significantly, no mention of Lucretius is made in Cicero’s published writings (largely written in the decade following the composition of DRN) and no clear engagement with the contents of the DRN occurs in his corpus of philosophical writings, despite their in-depth treatment of Epicureanism. The basis for Jerome’s assertion may be ­nothing more than the fact that, at least to judge from extant evidence, Cicero 15 The completion of the poem is much the most important of these, since it impinges directly on the following two questions. A short and subtle, but in my opinion too credulous, account of apparent inconsistency deliberately employed by Lucretius has been given by J.J. O’Hara, Inconsistency in Roman Epic (Cambridge, 2007), Chapter 3, 55–76. 16 This view was forcefully advanced by K. Ziegler, “Der Tod des Lucretius,” Hermes 71 (1936) 421–40, and has been almost universally held since.



lucretius auctus?

19

was the only ­contemporary of Lucretius who mentioned the poet’s name, albeit in a terse response to a (lost) epistle from his brother Quintus.17 Three further pieces of evidence about possible scholarly and/or editorial engagement with DRN remain, although all are at some remove from the work’s composition. First, an intriguing fragment that survives in an eighth-century manuscript (Paris BN lat. 7530, ff. 28r–29r), known as the Anecdoton Parisinum, claims that the celebrated grammarian M. Valerius Probus (late first cent. ad) added critical notes to several poetic texts, including Lucretius.18 Disappointingly, this assertion, which perhaps drew its origin from a lost work of Suetonius, is not borne out by any other evidence.19 Second, Jerome elsewhere made cursory mention of the fact that, in the late fourth century, at least one commentary upon the poem was available to him;20 although we have had cause to reject his testimony above, if it is correct here, these commentarii could perhaps be linked with Probus’ work; as before, no evidence survives to support this intriguing possibility. Third and finally, we have the so-called capitula, some 170 running titles that regularly divide DRN in its manuscript transmission. It has been supposed that these capitula could reflect editorial involvement at some stage in the poem’s history;21 however, on the basis of a detailed study of these capitula and their transmission I have made elsewhere,22 it will suffice here to record in brief my five most significant conclusions:

17  poemata Lucreti ut scribis ita sunt, multis luminibus ingeni, multae tamen artis (Epp. Q. fr. 2.10; Feb. 54 B.C.): the poemata probably refer to extracts of DRN already in circulation by that point: see further F.H. Sandbach, “Lucreti poemata and the Poet’s Death,” CR 54 (1940) 72–7; the arguments of U. Pizzani (Il Problemo del testo e delle composizione del De Rerum Natura di Lucrezio [Rome, 1959] 38–40) that poemata refers to the complete poem are not persuasive. 18  Probus qui illas [sc. notas] in Vergilio et Horatio et Lucretio apposuit, ut in Homero Aristarchus. 19  For further context see S.F. Bonner, “The ‘Anecdotum Parisinum,’ ” Hermes 88 (1960) 354–60, J.E.G. Zetzel, Latin Textual Criticism in Antiquity (New York, 1981) 41–54, and H.D. Jocelyn, “The Annotations of M. Valerius Probus,” CQ 34 (1984) 464–72, and CQ 35 (1985) 149–61. 20  Adu. Rufinum 1.16 [Migne PL XXIII 410]: puto quod puer legeris . . . commentarios . . . aliorum in alios, Plautum uidelicet, Lucretium, Flaccum, Persium atque Lucanum. 21 For the first study of these capitula see H. Fischer, De capitulis Lucretianis (Diss., Giessen, 1924); a much more unsatisfactory treatment is that of D. Pasetto, ‘I capitula Lucreziani’, AFLN 10 (1962–3) 33–50. The bizarre theory was proposed even this century – though it can be dismissed in an instant – that Lucretius himself was responsible for these capitula: see S. Sconocchia, ‘Su capitula e indices del De rerum natura di Lucrezio’ in A. Isola, E. Menestò & A. di Pilla (eds.), Curiositas: studi di cultura classica e medievale in onore di Ubaldo Pizzani (Naples, 2002) 51–90. 22  The Early Textual History of Lucretius’ De rerum natura (Cambridge, 2013) Chapter 3, 136–202.

20

david butterfield

(i) these capitula were not composed by Lucretius (or a contemporary); (ii) they do not reflect editorial involvement, but rather readers’ annotations in antiquity, probably from two distinct individuals; (iii) their purpose is typically to record the rough subject of a passage, but their sporadic nature, and often insignificant or incorrect details, do not suggest methodical work; (iv) the jottings were originally marginal but were subsequently incorporated into the main text as if authoritative headings; (v) the capitula are very probably limited to the line of transmission by which DRN survives and are insignificant in the textual survival of the poem. It may thus be regretted that our conclusions from the evidence available are so negative, but careful circumspection and critical suspicion in these matters will prove most advantageous in untangling the difficult and dark history of Lucretius’ Überlieferungsgeschichte. Since our evidence about the poem’s possible editorship is now exhausted, we may return to the crucial question of the poem’s completion. There is no doubt that the poem was only intended to be six books in length, for Lucretius tells us as much when he speaks of his being on the final leg of the poetic journey at 6.92–5.23 Nor is any book obviously deficient in length; on the contrary, all books exceed a thousand verses, the last three being notably longer than the first trio; indeed Book 5 stands in its present form of 1,457 verses as the longest (extant) book of Latin poetry.24 Nevertheless, readers from the Renaissance onwards have found the close of the poem difficult to stomach: DRN breaks off suddenly during the account of the Athenian Plague (6.1138 ff.), without any clear signal that the work has ended or how such an abrupt close should be handled by the reader (or, one supposes, the confused listener). Although the last lines are grammatically complete (1285–6: subdebantque faces, multo cum sanguine saepe / rixantes potius quam corpora desererentur) the breakneck stop is at once unexpected and perplexing. Several scholars of the twentieth century sought to defend the ending as posing a sophisticated test for the poem’s readership: faced with this bleak picture of death, destruction, and despair in Athens, can the reader, now at the end of their journey through Lucretius’ vast programme of Epicureanism, understand at this

23 Following the convention of earlier (and subsequent) poets, it would at any rate have been highly irregular for Lucretius to have written a work of anywhere between seven and eleven books, and it is impossible to believe that the DRN could have been envisaged on a twelve-book scale. 24 The length of the six books as transmitted is: 1: 1117; 2: 1174; 3: 1094; 4: 1287; 5: 1457; 6: 1286. The first half thus totals 3,385, the second 4,030.



lucretius auctus?

21

final point that death and its processes, however painful, are in reality nothing to fear?25 However, this subtle reading has not found favour with all modern critics, for it must be acknowledged that Lucretius was not the sort of poet to leave so powerful and so crucial an assertion to mere inference on the part of his addressee and readers: to suppose that he would thus allow the threads of his rigorous didactic web to dangle unconnectedly is, to my mind, incredible. Instead, if the rest of DRN may serve as a guide to a more probable conclusion, the work’s ultimate denouement would have allowed Lucretius to deliver a moral and apotropaic tour de force, one perhaps amounting to the work’s true climax and thus a worthy companion piece to the intricately crafted Hymn to Venus (1.1–43) from where it all began.26 Two possible manoeuvres could be made before concluding that Lucretius did not close the poem as transmitted: first, one may suppose that the last leaf/leaves of a codex were lost in transmission, thereby depriving the work of its last lines; second, that accidental transposition of verses has caused the wrong lines to stand at the poem’s end. No one has made a persuasive case for the first of these options, there being no positive evidence to encourage this uncertain conclusion. With regard to the second strategy, however, Friedrich Bockemüller, one of the most unconventional Lucretian commentators, tried in 1874 to mitigate the harshness of the poem’s end by transposing five lines from earlier in the book (1247–51) to after 1286, thus causing the work to end with an all-encompassing assertion of the death and grief that struck fifth-century Athens.27 This still leaves the main objection made above about the strangely tacit tone of the poem’s close. Yet a further problem arises: there is no evidence elsewhere in the 7,400 verses of DRN for an accidental transposition of 25 For this view see H.S. Commager Jr, “Lucretius’ Interpretation of the Plague,” HSCP 62 (1957) 105–18, D.F. Bright, “The Plague and the Structure of De rerum natura,” Latomus 30 (1971) 607–32, D. Clay, Lucretius and Epicurus (Ithaca, NY, 1983) 262–6, M. Gale, Myth and Poetry in Lucretius (Cambridge, 1994) Chapter 6, 208–28, P. Fowler, “Lucretian Conclusions” in D.H. Roberts, F.M. Dunn and D.P. Fowler (eds.), Classical Closure: Reading the End in Greek and Latin Literature (Princeton, 1997) 112–38. 26 As David Sedley (Lucretius and the Transformation of Greek Wisdom [Cambridge, 1998] 160–5) has suggested, perhaps the final element of the Epicurean tetrapharmakos, that what is bad/evil is easily avoidable, would have appeared after the account of the plague, so as to provide guidance as to how such terrible bodily sufferings witnessed in an age before Epicurus can now be borne steadfastly by the enlightened Epicurean. 27 This argument was supported by Bright (as n. 25) and Fowler (as n. 25), both constructing an elegant argument with several literary parallels, but I remain unconvinced for several reasons: in a forthcoming article I discuss the case for rejecting this transposition in detail.

22

david butterfield

this scale and nature. Almost all other agreed instances of transposition involve the movement of lines to a position earlier in the text than where they should stand, only one case involves more than one or two transposed lines, and no transposition covers more than twenty lines (which could occur on a single page of text): the movement of (a) five lines to (b) 40 lines (c) later in the text is therefore highly improbable.28 Needless to say, no convincing explanation of how verses 1247–51 could have made such a transition has been offered. The most natural conclusion, which avoids unnecessarily sophisticated special pleading on both interpretative and textual terms, is instead to revert to the old and simple idea that the poem’s abrupt close is a strong indicator of its incompletion. There exist several other strong indications that DRN failed to receive its final phase of composition and revision (although all have been points of contention). Most famously, at 5.155 Lucretius promises to discuss the (atomic) nature of the gods and their dwelling-place fully (largo ­sermone) – to the extent of proof (probabo) – at a future point within the poem.29 Despite this assertion, no such discussion occurs. Four possible solutions present themselves: (i) Lucretius did fulfil his promise but the relevant passage was accidentally lost during the poem’s transmission; (ii) Lucretius’ text as preserved does fulfil this promise but in a less-thanobvious way; (iii) the line has wrongly been transposed from some other place in the poem to after 5.154: if moved back to its original location, the verse’s relative pronoun (quae) may have as its referent something that Lucretius did indeed prove largo sermone; (iv) Lucretius did not fulfil his promise. No one has cogently put forward the case for (i), which would be a singularly unfortunate loss from the poem, and of a size presumably far greater than any other lacuna that the poem seems to have suffered. Solution (ii) is a rather sophistical response to the problem, and does not give due attention to the forceful phrase largo sermone probabo: scholars have pointed in desperation to brief passages that occur earlier in the work, such as 2.646–51 (= [1.44–9]), 3.18–22 and 5.82, yet these verses do nothing beyond highlighting the ataraxic existence of the gods. Even if we disregard the fact that they are positioned before the claim

28 There are two instances in which a verse is transposed to a later point in the text: 6.990 to after 6.997 (so Lambinus) and 1.155 to after 1.158 (as found in a manuscript of Serenus: see below); it is probable that the six verses 5.440–5 should be transposed to after 436 (so Reisacker). 29 Incidentally, Lucretius’ authorship of the line has never been doubted.



lucretius auctus?

23

of 5.155, they still do not actively prove the claim! A different tack was taken by Ubaldo Pizzani,30 who chose to refer quae not to 153–4 (nor 148–152) but rather back to 146–7, which assert that the gods do not live in the world: it is supposed that proof for this assertion is given by the content of Books 5–6, which discuss in several respects how the world is so abundantly suited to men and how celestial phenomena are independent of the gods, a view also promoted by Richard Minadeo.31 Yet this is an unnatural way to read quae (in the plural and with multiple possible referents intervening between 146 and 155), and the ample proof explicitly promised by largo sermone probabo does not suggest such a negative inference, one that Lucretius never presses home.32 Option (iii) was proposed by ­Gerhard Müller33 and supported by Deufert in a “kurzes Wort” of his Pseudo-­Lukrezisches im Lukrez (302–4): they proposed that 5.155 should follow 5.125, verses 5.126–45 being rejected as interpolated on Müller’s view, or 5.155 being so transposed on Deufert’s. With the line thus moved, quae would instead refer to the new preceding sentence (5.121–5), which states that the earth, sky, and sun are not divine entities but things devoid of living qualities. Yet, this undeniably violent solution still leaves a promise not fulfilled properly in the transmitted text: Lucretius nowhere proves that such bodies lack a living force. Nor is there any reason to suspect the authenticity of 5.126–44.34 By contrast, in its transmitted location 5.155 closes a carefully structured passage (5.146–54): 146–7 is the probandum, 148–52 the premises of the argument, 153–4 the conclusion. After such a unified paragraph, it is most natural that quae should refer to all of these lines, that is, to the dwelling place and nature of the gods. As a result, option (iv) remains much the most probable conclusion: Lucretius did not compose the promised treatment. Where such a passage (of, one presumes, at least 50 verses) 30 Pizzani (as n. 17) 174–80. 31 R. Minadeo, “Three Textual Problems in Lucretius,” CJ 63 (1967–8) 241–6, at 244–6. 32  Most improbable are the explanations of G. Müller (“Die fehlende Theologie im Luk­ reztext” in E. Lefèvre (ed.) Monumentum Chiloniense: Festschrift für E. Burck [Amsterdam, 1975] 277–95) that 5.1198–1203 (on the ineffectiveness of piety in religious ceremony) is a fragment of this largus sermo, or of B. Manuwald (Der Aufbau der lukrezischen Kulturentstehungslehre [Mainz, 1980] 51 n. 191) that 5.155 refers to a lost work of Lucretius De dis (a suggestion that had been tentatively suggested by J.P. Elder, “Lucretius 1.1–49,” TAPA 85 [1954] 88–120, at 89 n. 3)! 33 “Die Problematik” (as n. 11, 1959) 65. 34 This passage consists of two verses, followed by the repetition of 3.784–97 (with minor alterations), and then three further verses.

24

david butterfield

would have stood in DRN is open to debate, although towards the poem’s close may have been an attractive opportunity: it may be objected that Book 6 already comprises nearly 1,300 verses, but it and its predecessor (1,457 vv.) would presumably have been honed into a shorter form when completed. Another problematic instance is the passage at 4.45–53, which was most plausibly identified by Johannes Mewaldt as a “doublet” of the programmatic statement recorded more fully immediately beforehand, at 4.26–44.35 When carefully analysed, verses 45–53 have all the appearance of being an earlier draft of the programme provided in the passage transmitted as 4.26–44: intriguingly, verses 45–53 only show a connection with Books 1 and 2, whereas 26–44 relate to the contents of Book 3. Mewaldt therefore concluded that 45–53 originated from an earlier time in DRN’s composition when Book 4 was to follow Book 2; at a subsequent stage Lucretius changed his mind about the work’s structure and inserted Book 3 between 2 and 4, which prompted him to write a new and fuller programme for Book 4 (albeit reappropriating verses 49–51 for those transmitted now as 29–31).36 If this theory is correct – and it is much the neatest explanation offered to date – it is impossible to believe that Lucretius wished to retain

35 J. Mewaldt, “Eine Dublette in Buch IV des Lukrez,” Hermes 43 (1908) 186–95. The conclusions of Mewaldt served as a prompt to the thesis of Mussehl (as n. 4), viz that Lucretius’ rough order of book-composition was 1, 2, 5, 4, 3, 6, in many ways an attractive contention (note, inter alia, the absence of Memmius from Books 3, 4 and 6) which cannot be proved but is still by no means disproven. 36 For subsequent treatments of the opening 53 verses of Book 4, which as a whole fail to advance the question beyond Mewaldt’s persuasive hypothesis, see (alongside the commentators) H. Drexler, “Aporien in Prooemium IV des Lucrez,” Athenaeum 13 (1935) 73–100, W. Schmid, “Altes und Neues zu einer Lukrezfrage,’ Phil. 93 (1938) 338–51, H. Diller, “Die Prooemien des Lukrez und die Entstehung des lukrezischen Gedichts,” St. Ital. 25 (1951) 5–30, K. Büchner, “Die Proömien des Lukrez,” C&M 13 (1952) 159–235, G. Barra, “Qua ratione et mente IV poematis librum composuerit,” AUN 6 (1956) 5–17, L. Gompf, Die Frage der Entstehung von Lukrezens Lehrgedicht (Diss., Cologne, 1960), K. Gaiser, “Das vierte Prooemium des Lukrez und die ‘Lukrezische Frage’ ” in J. Kroymann and E. Zinn (eds.), Eranion: Festschrift für Hildebrecht Hommel (Tübingen, 1961) 19–41, T.V. Vasiljeva, “Lucretiani carminis quarti prooemium duplex an non?” in Sapientiae procerum amore: Mélanges médiévistes offerts à Jean-Pierre Mueller à l’occasion de son 70e anniversaire [Rome, 1974] 307–10, L. Canfora, “I proemi del De rerum natura,” RFIC 110 (1982) 63–77, M. Gale, “The Proems of the De rerum natura,” PCPS 40 (1994) 1–17, S. Kyriakidis, “Lucretius DRN 1.926–50 and the Proem to Book 4,” CQ 56 (2006) 606–10.



lucretius auctus?

25

this draft paragraph:37 one must therefore conclude that the book did not enjoy the poet’s summa manus.38 Several passages in the poem have been highlighted as seeming to want Lucretius’ final revision. For instance, although nine verses of the subsequent, more polished proem to Book 4 (33–41) suggest that ghosts will be an important part of the book, only eleven verses (757–67) subsequently appear on this topic. In Book 5, verses 110–234 are an extended digression that leaves 5.235 (principio . . .) as an abrupt and awkward development of the argument of 91–109; perhaps the transition would have been smoothened in due course. Later in the book, the astronomical account at 509–770 interrupts two phases in the narrative of the world’s history: whether Lucretius would have refined this arrangement is open to debate. The same applies to 5.318–23, which would sit more naturally after 305, and to 6.608–38, which would develop more naturally if the passage followed 6.711 (or had further transitional material inserted beforehand).39 In Book 4, the two similar passages 768–76 and 788–801 would probably not both have stood in their present form. Alongside these apparently under-worked passages there exist several lacunae in the poem’s fabric; however, the most obvious of these are generally agreed to be the result of textual corruption, not authorial incompletion.40 37 Attempts to avert the problem by large-scale transposition of verses (e.g., ­Marullus: 45–8, 26–43, 51–3, 49–50; Martin: 45–8, 26–30, 52–3, 31–44 [with 49–51 bracketed]; Gaiser: 45–8, 26–8, 49–53, 33–44 [with 29–32 bracketed]) are far too convoluted to be convincing. Yet more improbable is the argument of Schmid (as n. 36) and Deufert (as n. 13, 160–4) that verses 4.45–53 were an interpolation composed by a later reader: why would someone feel motivated to fashion these ill-fitting verses, let alone retain what they seem to replace? 38 Intriguingly, the fact that the later version (26–44) precedes the earlier in transmission makes it difficult to conceive how these two same versions could have been written in the same text, if the later version was not a marginal addition. It is useful to note in connection with this problem that Books 1–3 (especially 1–2) are largely complete, whereas 4–6 are evidently not fully revised (and exhibit less crafted closes). 39 Most improbable is the view held by the second Teubner editor of Lucretius, Adolf Brieger (Leipzig, 1894), who believed that almost all of these passages were written as Lucretius intended but they were wrongly ordered by a later figure; as a result he, and to a lesser extent Carlo Giussani (Turin, 1896–8), bracketed and transposed verses throughout the poem to a remarkable and incredible degree. Nevertheless, Brieger’s arguments for Lucretius’ original completion of the poem remain valuable: “Die Unfertigkeit des Lucrezischen Gedichtes,” Phil. 67 (1908) 279–303. 40 Most famous among these is that at 1.1094–1101, a key piece of evidence in ­Lachmann’s magisterial reconstruction of the 26-page archetype from which OQS descend. Other plausible lacunae have been detected after 1.43 (in lieu of [1.44–9]), 1.860, 1.1013, 2.164, 2.681, 2.748, 2.750, 2.903, 3.97, 3.823, 3.1011, 4.126, 4.144, 4.289, 5.704, 5.1012, 6.47, 6.697,

26

david butterfield

A much more controversial issue, and one which is of prime importance in tackling the question of interpolation in DRN, is presented by the several repetitions of large passages in the poem, four of which exceed ten lines in length. For many scholars of the last two centuries, it has proved hard to believe that Lucretius intended to retain such repeated passages in both of their transmitted places; if he did, one struggles to find any analogous practice on this scale in earlier or later authors in Latin literature. Even when the web is cast wider to Greek epic and didactic, the techniques of Homer and Empedocles do not provide persuasive support.41 Accordingly, repeated verses have been treated as a primary hunting ground for potential interpolations in the poem, inserted or incorporated by later readers or scribes during DRN’s transmission. The two most contested repetitions are 1.44–9 = 2.646–51 and 1.926– 50 = 4.1–25 (with minor variations).42 The six verses at 2.646–51, outlining the ataraxic state in which the gods live, fit most naturally into their context, thus dismissing the misconceptions of the cult of Magna Mater (2.600–43). These same verses also occur, however, immediately after the poem’s opening proem (1.1–43), in which Lucretius addresses Venus and highlights his allegiance to Memmius. Yet this location for them is most difficult to stomach: for directly after an impassioned prayer to the goddess for the bestowal of peace and favour there follows the assertion that the gods are in no way concerned with the life of humans. Despite the claims of several twentieth-century scholars,43 I must join the many 6.1246; perhaps also after 2.465, 4.823 and 5.25. No lacuna should be posited after 2.600: see my “Three Lucretian Emendations,” AAntHung 48 (2008) 351–64, at 354–7. 41 The novel thesis of Joseph Hellegouarc’h (“Fabricator poeta: existe-t-il une poésie formulaire en latin?, REL 62 [1984] 166–91, and “Style didactique et expression poétique dans le De rerum natura de Lucrèce,” BSL 19 [1989] 3–16), namely that Lucretius composed more Homerico by drawing upon several stock mechanical phrases according to the demands of argument and metre, is unpersuasive on close inspection. 42  1.936 sed: 4.11 nam; 1.949 perspicis: 4.24 percipis; 1.950 qua constet compta figura: 4.25 ac persentis utilitatem. For Lucretius’ habit of making minor alterations when repeating verses, see C. Bailey, Titi Lucreti Cari De Rerum Natura Libri Sex (Oxford, 1947) I 164–5; nevertheless, I suspect the text of 1.936 and 4.24. 43 See, for instance, E. Bignone, “Nuove ricerche sul proemio del poema di Lucrezio,” RFIC 47 (1919) 423–33, O. Regenbogen, Lukrez: seine Gestalt in seinem Gedicht (Leipzig, 1932), P. Friedländer, “Retractationes II,” Hermes 67 (1932) 43–6, at 43–4, N.H. Romanes, Further Notes on Lucretius (Oxford, 1935) 7, P. Friedländer, “The Epicurean Theology in Lucretius’ First Prooemium (Lucr. I.44–49),” TAPA 70 (1939) 368–79. Bignone supposed that Lucretius wished to insert further material after 1.43 to bridge the transition and explain the allegory of Venus; Regenbogen rather supposed that Lucretius added these verses marginally but failed to accommodate them to their new surroundings and duly alter the prayer to Venus; Romanes posited a lacuna of one line prior to 43; ­Friedländer



lucretius auctus?

27

scholars since the fifteenth century who cannot believe that these verses should stand here and instead reject them as spurious.44 Not only is the alarming volte face difficult to reconcile with the beginning of a didactic poem but there are three other telling indications that these verses were not placed here by Lucretius himself. First, the explanatory particle enim in 1.44, which made perfect sense after the claim of 2.644–5, makes nonsense in this place: Friedländer (as n. 43 bis, followed by Martin [as n. 4]) was forced to relate it back to pacem of 40, but this ignores the quite different assertion of 41–3 (itself introduced by nam) that Lucretius and the gens Memmia are in a state of turbulence. Second, 1.50 (whose close is defective but probably restorable from the scholia Veronensia ad Verg. Geo. 3.3) and subsequent verses seem to be directed towards the poem’s dedicatee Memmius, in which case such a direct vocatival address must have been lost in a lacuna – perhaps only of a small number of verses – in lieu of 1.44–9. Third, if we turn to the capitula in the poem’s manuscripts, we may note that a reader (probably the first author of these capitula, active in the first two books of DRN) made a few annotations in Greek in Books 1 and 2, often drawing phrases from Epicurean writings. Crucially, however, the Greek heading provided for 2.646–51 appears in its original Greek script: τὸ μακάριον καὶ ἄφθαρτον; the heading for 1.44–9, however, the first capitulum to appear in the poem, has been transliterated – incorrectly – into Roman letters: toma carion cae apitraton. The most natural conclusion is that the original capitulum was transferred along with the subsequent lines 2.646–51 but by a reader (or scribe) who was not fully cognisant of Greek script and duly made errors in transliteration into Latin script; the three other Greek capitula in the manuscript (preceding 1.951, 1.1052 and 2.14) managed to avoid such unwelcome corruption.

boldly maintained that nothing in the transmitted text requires alteration, a view in which Bailey (as n. 42, ad loc.) and more recent scholars have tentatively followed him: see G. Barra, Struttura e Composizione del “De rerum natura” di Lucrezio (Naples, 1952) 18 ff., Minadeo (as n. 31) 241–4, and F. Giancotti, Il preludio di Lucrezio e altri scritti lucreziani ed epicurei (Florence, 1978) 217–22. For calm among the storm, see E. Courtney, “The proem of Lucretius,” MH 58 (2001) 201–11. 44 The verses were athetised by Pontanus and Marullus, and removed from the 1512 Giuntine edition. The very incongruity of the passage set against Lucretius’ proem prompted his most verbose opponent, Cardinal Melchior de Polignac (1661–1742), to complain in his nine-book Anti-Lucretius (1745) vocet demens, quos tendat perdere, divos, / immemor ipse sui (5.35–6). The annotator (late s. xv) of Vat. Ottob. Lat. 2834 had already harangued Lucretius for his inconsistency: for a transcription of the note see M.D. Reeve, “The Italian Tradition of Lucretius Revisited,” Aevum 79 (2005) 115–64, at 148 n. 95.

28

david butterfield

Why and how would such a transfer of the passage (and its capitulum) take place? The most natural answer is that a keen reader of DRN chose to respond to the remarkably unepicurean nature of the proem by jotting marginally Lucretius’ own most explicit statement of the gods’ disinteredness. These verses were in due course incorporated into the text of DRN by a scribe keen to preserve as much of the poem as seemed to be transmitted, under the misconception that these were Lucretian verses intended for insertion in the location they flanked. Quite probably, their incorporation brought about the loss of lines in which Memmius was first addressed directly.45 Proof that such confusions occurred is attested in a different respect by the capitula once more: in several cases these headings have been wrongly incorporated as if verses of Lucretius, and contrariwise verses of Lucretius have been wrongly taken to be capitula.46 Is the marginal repetition of such a passage a true act of interpolation? This is more a question of semantics then textual criticism: if interpolation denotes the purposeful insertion of verses (whether composed by the author or not) into a text where they were not intended to be, such accidental incorporation of former marginalia is not interpolation in its true and full sense; if, however, verses may be called interpolated if they crept into the text contrary to the author’s design, clearly we are faced with a genuine case of interpolation. The negative stigma regularly attached to the term “interpolation,” however, encourages me to make a distinction for what follows in my discussion between accidental incorporations and active interpolations created ex nihilo (i.e., not ex Lucretio). As I shall argue, of all spurious verses in DRN, the very great majority are of the first of these two classes – accidental and/or unfortunate incorporations. A yet more complex and notorious case is provided by verses 1.926–50, repeated with only minor changes at 4.1–25. Such a 25-line repetition is almost double in size compared with the next largest repetitions in DRN 45 Excessively complicated attempts have been made to mitigate the awkward ordering of verses in the opening phase of the poem: Bockemüller read 1–43, 62–101, 136–45, 155, 50–61, 104–35; Brieger 1–43, [lacuna], 62–79, [lacuna], 136–45, 80–135; Martin 1–49, 62–135, 50–61, 136–45; P. Gimpel (“De rerum natura: prooemium restitutum,” RCCM 31 [1981] 3–41) 1–43, 80–101, 62–79, 44–9, 102–26, 50–61, 127–45. All such tampering is as idle as it is violent. 46 By the time of the archetype, two capitula were written as if Lucretian verses (1.419 and 1.705; note also that S wrote the capitulum before 2.112 as a Lucretian verse); contrariwise, seven Lucretian verses were written as capitula (2.42–3, 3.672a, 3.759, 3.805, 3.905, 3.949). O seemingly extended this error, by mistaking thirteen further Lucretian verses capitula (1.11, 1.411, 2.95, 2.502, 2.508, 2.608, 2.809, 2.887, 2.909, 2.962, 2.1012, 2.1023, 2.1112), a mistake Q made only once (1.565).



lucretius auctus?

29

and unlike anything else in extant Greco-Roman poetry. Three possible explanations exist: (i) that Lucretius wrote the verses in both places (leaving aside the question of whether he wished to make any further changes); (ii) that they were incorrectly repeated from Book 1 at the beginning of Book 4; and (iii) that they were incorrectly repeated from Book 4 after 1.925. Each of these three views have been maintained by scholars.47 It is not possible to engage in a detailed study of the problem here but it will suffice to say that, since no one has doubted that at least one of the passages was composed by the poet himself, and since it is both improbable and unfeasible that a later reader or scribe48 both repeated so large a passage in another place and made the minor verbal alterations characteristic of the poet’s repetitions (see n. 42 above), the most likely solution is that Lucretius was responsible for writing the verses in both of their transmitted locations. Two secondary questions then arise: which passage was written first by Lucretius, and did he wish for both to stay in their transmitted form? Mercifully, to answer these questions lies beyond the editor’s remit, since it verges upon emending not the poem put into circulation but rather Lucretius’ autograph on his behalf.49 From the foregoing discussion it should be clear that one can only avoid the conclusion that Lucretius’ poem is indeed incomplete either by employing convoluted argumentation and special pleading or by maintaining that many apparent problems with the work are the result of an interpolator’s keen activity. Given that the former solution rules itself out by its sheer improbability, we may now turn to discuss directly the case for non-Lucretian verses in Lucretius.50 If it can be accepted that Lucretius did not finish the work, and therefore could let himself nod in 47 For some of the more important contributions in the vast bibliography see most of the items listed above in n. 43. 48 It is of course possible that one individual performed both of these roles, or that a reader commissioned his scribe to act in this way. 49 My own belief is that Lucretius first wrote 1.921–50 and then transferred the passage as the proem to Book 4, perhaps expecting to modify the version in Book 1 so as to mitigate the anticlimactic repetition. The reverse process, as proposed by Büchner (as n. 36), seems much less probable. Deufert’s belief (as n. 13, 81–96) that the proem to Book 4 was created from Book 1 (and duly tweaked) does not persuade me: if Book 4 lacked a proem (as Deufert [ibid., 159] most improbably claims Lucretius intended), then the work was indeed incomplete (which affects much of the broader argument); if Book 4 did have a proem, what happened to it, and what would really motivate such an interpolation to oust it? Finally, I am not persuaded by the arguments of Gale and Kyriakidis (both as n. 36) that these double passages represent Lucretius’ final intentions. 50 By “non-Lucretian verses” I mean verses that appear in places where the poem as first circulated did not contain them.

30

david butterfield

c­ omposition, can it still be demonstrated that the poem was augmented by an over-active reader or an over-zealous editor? It will be useful at this point to consider briefly how DRN would have first entered circulation, if its author could not distribute it as a finished entity himself, either owing to his premature death (probable) or his choosing to abandon the poem (most improbable). How, we must ask, did it swiftly reach a wider audience, as its almost immediate effect on the poetry of Catullus, and subsequently Virgil, Horace, Ovid, and Manilius, attests?51 Since Lucretius composed squarely in the age of papyrus and centuries before the revolutionary infiltration of the codex, we must envisage that his incomplete poem either stood in multiple rolls (presumably one per book), perhaps showing signs of revision rather than a fair copy, or alternatively significant portions of it were preserved on individual papyrus schedae. In either case, the involvement of a figure other than Lucretius – perhaps someone by no means expert in Epicureanism – was necessary in order to put the poem into circulation by preparing its initial transcription(s). An eternally unknowable question is to what extent this figure had to determine in what order parts of Lucretius’ poem should stand or decide which verses should be included and excluded. In fact, Occam’s razor and the complete absence of evidence should lead us to suppose that this figure made no change to what was before him: thus to talk of “the first editor” in textual terms can be dismissed, for practical purposes, as a mere straw man.52 We are faced, instead, with analysing the text as we have it: unfortunately, of course, we can do no more than scrutinise our earliest manuscripts (of the ninth century: see n. 3), drawing evidence from the indirect tradition wherever possible. It has been well said that “there is only one feature common to all interpolations: the fact they are all extraneous additions to a genuine text.”53 Yet, in order to sift through the evidence thrown up by Lucretius’ poem, we may conveniently posit three broad classes of verses that could be spurious accretions during the transmission of DRN: (i) transposed verses, (ii) repeated verses, and (iii) interpolated verses.

51 Unfortunately it is unclear what portion of the complete work the poemata read by Marcus and Quintus Cicero (see n. 17 above) represented; it is also probable, but not provable, that their exchange of February 54 followed Lucretius’ death (I believe a strong case can be made for 94–55 bc for Lucretius’ lifespan). 52  Significantly, the existence of the apparent doublet in Book 4 could suggest that this figure was not given to excising any Lucretian verses he found. 53 R. Renehan, Greek Textual Criticism (Cambridge MA, 1969) 29.



lucretius auctus?

31

(i) Transposed Verses We may begin with cases of the transposition of verses from their incorrect place. Although such transpositions occurred primarily via mechanical errors, it is also possible that, in cases where it is agreed that a verse or passage is wrongly transposed or repeated, the error arose from a later reader/scribe recopying the verses and/or actively changing their ­position.54 Errors of transposition can be divided into two genera: scribal accident or the active involvement of a reader/scribe. Accidental error occurred primarily when a scribe’s eyes skipped down the exemplar and duly omitted one or more verses, which were then inserted marginally by the scribe (or diorthotes) but, at a later stage of transcription, were copied in the incorrect place.55 I believe that mechanical transpositions of this nature occur in 31 cases in the poem.56 Almost all are of single lines: three cases with pairs of lines are attested in Books 4–5, 54 A third possibility, namely Lucretius’ not sufficiently having revised his manuscript(s), may be set aside for present purposes. 55 Of course, if this error was not noticed, lacunae are instead introduced into the text. Less commonly, a scribe accidentally lets his eye skip back up the page to reread, and thus rewrite, text already copied, thereby producing in his apograph two versions of the same piece of text; if he or his corrector noticed this error, the second passage would have been marked for deletion. If these signs were not understood at the next stage of transcription, the accidental repetition remains; if, by contrast, a later scribe or reader misunderstood the scribe’s sigla and instead deleted the former version of the passage, thus allowing only the second to stand, the transposition of that passage forwards in a text would result. An extreme version of transposition because of the physical form of the codex has affected Lucretius’ mss: in Book 4 verses 299–322 occur wrongly before verses 323–47. In the archetype Ω (cf. n. 3) the leaf which contained verses 4.299–347 (= leaf 76) was disbound and inserted the wrong way round, so that in the next stage of transmission the original verso was copied before the recto. We know that the archetype was in a poor condition since, by the time ψ (the exemplar of QS) was copied, four further leaves of the manuscript (containing 1.734–85, 2.253–504, 2.757–806, 5.928–79) had become disbound and inserted, in an incorrect order, at the very rear of the book, where they were wrongly copied by a later scribe. 56 (The first corrector is added in parentheses): 1.14–15 (J), 1.155 / 158 (L; Sereni cod. Vat. Reg. Lat. 598 f. 33r), 1.434–5 (Flor. cod. Laur. 35,32), 1.1085–6 (Marullus), 2.5–6 (Avancius), 2.622–3 (transposui), 2.680 / 659 (Lachmann), 2.923 / 914 (Bernays), 3.46 / 43 (Bentley), 3.865 / 857 (Lachmann), 3.955 / 951 (Lachmann), 4.135 / 132 (Lambinus), 4.250–1 (Marullus), 4.260–1 (Muretus), 4.299–322 / 323–47 (Q2), 4.551–2 / 541 (Lambinus), 4.826 / 821 (Q2), 4.999 / 990 (Avancius), 4.1123–4 (Lambinus), 4.1210 / 1203 (Naugerius), 5.30 / 29 (Munro), 5.440–5 / 436 (Reisacker), 5.573 / 569 (Marullus), 5.594–5 / 585 (Cartault), 5.975 / 967 (Naugerius), 5.1131–2 / 1126 (Munro), 6.934–5 (Avancius), 6.990 / 997 (Lambinus), 6.1033 / 1025 (Avancius), 6.1178 / 1173 (Naugerius), 6.1245 / 1236 (Bentley). There is no generally accepted instance of any line or lines requiring transposition over a greater distance than 20 lines (2.680 to after 2.659), a distance that would often cover more than a given page of a codex, or column of a roll.

32

david butterfield

and, largest of all, a passage of six lines in Book 5 (440–5);57 as expected, in almost all cases the verse needs to be transposed to an earlier place in the work, having been overlooked and duly inserted later in the poem’s text. No interpolator’s hand is at work here. The second type of transposition, which arises from the involvement of a reader/scribe, occurs most commonly via two stages; first, a reader of the manuscript actively engages with the poem’s contents and makes marginal jottings; second, a scribe accidentally incorporates these annotations as part of the poetic text proper. In the most obvious case, we can imagine a reader jotting in the margin a Lucretian passage from elsewhere in the work which relates somehow to the text in question, perhaps supporting the argument, perhaps seemingly contradicting it, perhaps employing similar language, or perhaps worthy of comparison in other respects. Since these casual marginalia would naturally stand on the same leaf as the text, they could readily be interpreted by a future scribe as mistakenly omitted verses that required inclusion once more. However, it should be noted that this event can only be classed as simple transposition if the original passage (now duplicated) is subsequently deleted. I do not believe any such case of active incorporation and deletion has afflicted the transmission of DRN since its first circulation. (ii) Repeated Verses If, by contrast, the original passage and the incorporated transposition of it just described are left to stand, we are faced instead with a case of spurious repetition. This particular form of corruption is particularly difficult for the editor of Lucretius, as the poet is famous for his fondness for repetitions, which range from the smallest elements (letters, syllables, words) through to whole verses and complete passages.58 This device

57 The exceptional instance of the reversal of the archetypal leaf in containing 4.299– 347 (see n. 55 above) must be disregarded. 58 For Lucretius’ penchant for repetition see, along with Gneisse (as n. 10), C. Lenz, Die wiederholten Verse bei Lukrez (Diss., Dresden, 1937), A.E. Raubitschek, “Zu einigen Wiederholungen bei Lukrez,” AJP 59 (1938) 218–23, Bailey (as n. 42) I 161–5, W.B. Ingalls, “Repetition in Lucretius,” Phoenix 25 (1971) 227–36, J.D. Minyard, Mode and Value in Lucretius’ De Rerum Natura (Berlin, 1978), Clay (as n. 25) 180–5, and A. Schiesaro, “Problemi di formularità lucreziana,” MD 24 (1990) 47–70. For the pattern of repetitions of letters and sounds see also R.E. Deutsch, The Pattern of Sound in Lucretius (Diss., Bryn Mawr, 1939). Lenz calculated that there are 213 repeated verses in the poem, which make up almost 3% of its content as transmitted. The present discussion is limited to full verses, but as



lucretius auctus?

33

need not surprise: with no system of footnoting or quick and precise referencing available, repeating verses was an almost essential didactic practice in a work as complex as DRN. Necessarily, therefore, several repetitions in the work must be retained as genuine Lucretian practice. The question must thus always be asked by the critic: is this repetition the work of Lucretius or did it more probably result from a reader’s marginal jottings?59 Scholars, beginning with Gneisse (as n. 10), but developed by Christoph Lenz (as n. 58), rejected many such repetitions as interpolated. However, several methodical flaws of Lenz’s study were soon exposed by Antony Raubitschek, who demonstrated how many repeated verses translate extant Epicurean doctrine closely, and deserved repetition by dint of their philosophical importance.60 Although a truly thorough and sophisticated treatment of Lucretius’ employment of repetition is still required, for our present purposes we can set aside those passages that, on inspection, seem perfectly credible as Lucretius’ own intended repetitions, even if in some cases he would perhaps have altered one passage further in revision.61 Of the problematic cases, we can categorise them as either

a particularly striking example of the repetition of phrases, cf. the occurrence of tempore with certo no fewer than six times in seventeen lines at 5.656–73. 59 The intermediary answer, that the repetition is correctly transmitted but reflects the posthumous editor’s mistaken inclusion of one passage in two passages, is impossible for the modern editor to prove: to posit a distinction between Lucretius’ final arrangement of his poem and an editor’s confused arrangement thereof can be achieved in theory but scarcely in practice. 60 Note for instance the case of 1.670–1 = 1.792–3 = 2.753–4 = 3.519–20, a pair of lines which Clay (as n. 25, 191 ff.) has termed the “Epicurean axiom of change,” drawing upon Epicurus’ Epistula ad Herodotum 54. 61 For a complete list of repetitions see Lenz (as n. 58) 8–12 and Minyard (as n. 58) Appendix B, 151–72. The following repetitions rejected by Deufert (typically on the suggestion of earlier scholars) I regard as genuine (if not in all cases fully polished): 1.146–8 and 2.55–61 (= 3.87–93= 6.35–41), 1.531 and 4.672 (~ 1.429), 1.548–50 (548 = 1.609 ~ 1.574), 1.670–1 (= 1.792–3 = 2.753–4 = 3.519–20), 1.672–3 (= 1.797 = 2.756 = 2.864), 1.674 (~ 1.757), 1.883–5 (~ 1.907–10), 1.919–20 (~ 2.976–7), 1.1079–80 (~ 2.236–7), 2.32–33 (= 5.1395–6), 2.177–81 (= 5.195–9), 2.334 (~ 3.32 = 4.46), 3.701–2 (701 ~ 3.756), 3.806–18 (~ 5.351–63), 4.1–25 (= 1.926–50), 4.216–29 (= 6.922–35), 4.774 (= 4.779), 4.800–1 (~ 4.771–2), 5.111– 12 (= 1.738–9), 5.128–41 (= 3.784–97), 5.189–91 (= 5.424–6), 5.210–12 (~ 1.211–12), 5.1315 (~ 2.632), 5.1388–9 (= 5.1454–5), 5.1932–4 (~ 2.29–31), 6.56–7 (= 1.153–4 = [6.90–1]), 6.58–66 (= 5.82–90), 6.228–9 (~ 1.489–90), 6.251–4 (= 4.170–3), 6.299 (~ 6.1017), 6.383–5 (= 6.87–9). Many other Lucretian repetitions are not suspected by Deufert, for instance: 1.76–7 = 1.595–6 = 5.89–90, 1.131 ~ 4.121, 1.132 ~ 4.33, 1.135 ~ 4.734, 1.188 = 2.707, 1.428 ~ 1.505, 1.510 ~ 1.538, 1.635–6 ~ 1.705–6, 1.637 = 2.176, 1.869 ~ 1.874, 1.1080 = 2.237, 2.12–13 ~ 3.62–3, 2.82 ~ 2.229, 2.149 ~ 2.246, 2.163–4 ~ 4.207–8, 3.371 = 5.622, 3.624 ~ 3.670, 4.43 ~ 4.64, 4.44 = 5.882, 4.107 ~ 4.290, 4.165 = 4.725, 4.180–2 = 4.909–11, 4.189 ~ 5.283, 4.246–7 ~ 4.280–1, 4.585 = 5.1385, 4.590 ~ 5.845, 5.183 ~ 5.1049, 5.238 ~ 5.321, 5.266–7 ~ 5.388–9,

34

david butterfield

(a) mechanical transpositions that require deletion, or (b) repetitions originating from marginal comments that require bracketing/deletion. (a) Mechanical repetitions are easily identifiable for deletion. In seven cases a single verse is accidentally repeated,62 once two verses,63 once four.64 In three instances an accidental repetition has been subsequently altered, presumably consciously, in an ill-conceived attempt to adjust the verses crudely to their new context: at 2.105 (paucuia quae porro magnum per inane uagantur) we probably witness an accidental anticipation of 2.109 (multaque praeterea magnum per inane uagantur), abbreviations such as q; and p̄ terea perhaps encouraging the alteration to quae porro, and paucula, the correction of paucuia in several Italian manuscripts, is not attested in Lucretius; 2.636 (armat et in numerum pernice chorea) is an accidental fusion of the text of 2.635 ( . . . –um pernice chorea) and 637 (armati in numerum p- . . . ), rightly deleted by certain Itali (AB); at 4.1047, the incorrectly repeated 1034 (qui ciet inritans loca turgida semine multo) was roughly incorporated into its new location by the substitution of incitat for qui ciet, correctly athetised by Naugerius.65 In three further instances, the deletion of repeated verses is directly linked with the transposition of one or more verses.66 (b) Several repetitions more probably result not from mechanical error but rather from the incorporation of marginal verses added by a reader of the poem, perhaps for the sake of comparison with Lucretius’ text, but were misunderstood and incorporated by a later scribe: 1.44–9 = 2.646–51 (Marullus), 2.1020 = 2.726 and 5.1441 (Lachmann), 4.102–3 = 4.65–6 (Itali), 6.56–7 = 1.153–4 (Bentley), and 6.90–1 = 1.153–4 (Munro). That such a phenomenon can occur in a didactic text such as DRN need not surprise when the same has occurred in a work with as august a tradition as Vergil’s Aeneis: the case of 3.230 (repeated from 1.311 because of 5.269–72 = 6.635–8, 5.406 ~ 6.767, 5.520 = 6.200, 5.619 ~ 5.692, 5.801 ~ 5.1078, 5.866 ~ 6.1245. 62  1.769 = 1.762 (O2), 3.672a = 3.678 (Itali), 3.763 = 3.746 (Lachmann), 4.808 = 4.804 (Itali), 5.574 = 5.570 (Itali), 5.596 = 5.584 (Naugerius), 5.771 = 5.764 (Lambinus). 63 6.988–9 = 6.996–7 (Itali). 64 4.1000–3 = 4.992–5 (Itali). 65 A fourth case of this kind may exist: 5.555 (ex ineunte aeuo coniuncta atque uniter apta [Pontanus: aucta OQS]) could be an accidental or marginal repetition of 5.537 (ex ineunte aeuo coniunctam atque uniter aptam) incorrectly incorporated into the text; thereafter, the accusative was altered to the nominative to fit the sense (and apta corrupted to aucta). 66 6.990 (6.988–9 = 6.996–7), 4.999 (4.1000–3 = 4.992–5), 5.573 (5.574 = 5.571 (570)).



lucretius auctus?

35

the similar clausulae of 1.310 and 3.229) will repay inspection. More complicated are several cases where marginal (or sublinear / supralinear?) annotations appear to be doctored by a later hand, perhaps after their initial (incorrect) incorporation into the text proper, so as to scan and/or provide basic sense. At 3.474–5 one finds the anticipation of 3.510 (et quoniam mentem sanari, corpus ut aegrum) followed by a slightly doctored verse (et pariter mentem sanari corpus inani); it is possible that the opening verse(s) of 3.510 ff. were here cited marginally to flag up the discussion of medicine that would come. All subsequent changes must be purposeful alterations to preserve metre, if not sense: pariter occurs four times in the preceding 29 lines (445, 446, 457 bis), and corpus inani closes Lucretian verses at 1.524 and 6.941 (cf. also corpus inane at 1.745 and 1.1009). The two verses must be deleted, as Naugerius first saw. 5.1328 (in se fracta suo tinguentes sanguine tela) appears to be the hexametric expansion of an original gloss on infracta in the verse immediately preceding (tela infracta suo tinguentes sanguine saeui) and was properly athetised by Lambinus. Finally, at 6.1106–9, where Lucretius surveys the different effects of the world’s various climates upon people, we see him mapping out four compass points from Rome, Britain in the north, Egypt the south, the Black Sea the east, and lastly Gades the west; yet the inclusion with the last of 6.1109 (usque ad nigra uirum percocto saecla colore), a slight alteration of 6.722 (inter nigra uirum percocto saecla colore), which concerns the Ethiopians of north/central Africa, is a marked oddity. Given the verse’s inappropriateness and the very great irregularity of atque at the hexameter’s close – a metrical expedient not attested elsewhere in DRN, and very rare in Latin poetry in general67 – I have proposed that this later verse should be rejected as a non-Lucretian annotation to the passage, duly incorporated and made grammatical by the solecistic usage of atque and alteration of inter to the un-Lucretian usque ad.68 One case remains unclear to me: at 5.1388–9 we find the couplet sic unumquicquid paulatim protrahit aetas / in medium ratioque in luminis erigit oras, which also occurs at 5.1454–5, where it is much more suited to its context. It is possible that Lucretius only wrote the verses at 1454–5 and a reader jotted them as also relevant to the gradual development of human song and music (1379–87), from where they were duly brought into the

67 See D.J. Butterfield, “The Poetic Treatment of atque From Catullus to Juvenal,” Mnem. 61 (2008), 386–413. 68 See my “Lucretiana quaedam,” Phil. 152 (2008), 111–27, at 123–5.

36

david butterfield

text; more likely, however, is that Lucretius himself wrote them in both places, and neglected to delete the former passage: few have defended the existence of the passage in both locations as significant. (iii) Original Interpolations We come at last to the most controversial class: where someone other than Lucretius composed one or more verses ex nihilo that duly found their way into the poet’s text. I regret to say that I am convinced by only one of thirty-six cases supported or proposed by Deufert:69 all other cases in which spurious verses appear in our ninth-century manuscripts I regard as cases of classes (i) and (ii) above. Several problems are thrown up by the nature of verses that Deufert and his predecessors have rejected as interpolated, which cannot be dealt with here. It will suffice to remark, however, how improbable it is that such suspected verses include cases of sigmatic ecthlipsis (4.1018, 5.53),70 the rare Lucretian employment of cluere (3.207, 4.53) and quoad scanned as a monosyllable with synizesis (5.1213; cf. 5.1433). Such linguistic and prosodic features could only be inserted by an interpolator who actively strove to pass himself off as Lucretian (and had the learning to do so); as is clear from my foregoing remarks, I find no evidence to support the existence of a figure interested in such an act of literary aemulatio and emendatio, who subsequently enjoyed such success in his endeavours. The sole case of a line that seems very probably to have been composed by a non-Lucretian hand is 5.1006 (improba nauigii ratio tum caeca iacebat). This verse, which summarises the foregoing argument that primitive man was not subjected to the danger of seafaring (5.999–1005), presents too many unsurmountable problems to be overcome.71 Not only do we find 69 I am thus convinced that the following lines are Lucretian: 1.58–61, 1.192–8, 1.334, 1.527, 1.577, 1.907–14, 1.1085–6, 2.183, 2.859–64, 2.923, 3.206–7, 3.412, 3.430–3, 3.743, 3.764, 3.1031, 4.788–93, 4.885, 4.1013–14, 4.1018–19, 5.52–4, 5.123, 5.372, 5.545, 5.816–20, 5.832–6, 5.1131–2, 5.1148–50, 5.1164, 5.1190, 5.1311–17, 5.1341–9, 6.33–4, 6.565–7, 6.948–50. I likewise believe to be Lucretian other supposed interpolations not discussed by Deufert: e.g., 1.518–19, 1.841, 1.853, 2.189, 2.566, 3.97, 3.362, 3.498, 4.490, 4.1076–7, 5.175–6, 5.880–1, 6.529, 6.778–9. 70 For a survey of the constraints with which this characteristically Lucretian metrical licence was employed, see my “Sigmatic Ecthlipsis in Lucretius,” Hermes 136 (2008) 188–205. 71 For more detailed discussion see (alongside the commentaries) K. Müller, “De versibus interpolatis nonnullis” (in his 1975 Zurich edition) 364–85, and 368–9, Deufert (n. 13) 261–2; cf. also Butterfield (as n. 68) 120–1 n. 41.



lucretius auctus?

37

the anachronistic usage of the uncontracted genitive plural of an i-stem second-declension noun, first attested in non-Greek words in the poetry of Propertius, and the usage of nauigium as ‘navigation’, not attested until the second-century ad, but the line is weak in force and interrupts the progression of the argument. Single-line gnomic assertions such as this often attracted the hand of an interpolator in Greco-Roman poetry, and the same may be the case here. We need not necessarily suppose the activity of an individual who desired to foist his own compositions onto the work; it may simply be a casual jotting that came to be included in the main text of the poem; the line could even be a quotation from a late-Latin poem that has been completely lost to us. At any rate, Gustav Pinzger was surely correct to reject the verse as not Lucretius’ own. Two further difficult cases seem rather to reflect Lucretius’ lack of completion than genuine interpolation. First, 1.873 (praeterea tellus quae corpora cumque alit auget), which is not followed by text to complete the sense in our manuscripts, should probably be excluded as a variant of 1.867 (praeterea quaequomque e terra corpora crescunt); could this have been a verse existing in Lucretius’ original and ill-advisedly included by the posthumous “editor”? If so, 1.874 would round off the sense of 1.871–2, deliberately using very similar words (cf. 1.870) to 1.869. Second, 5.1359 (atque ipsi pariter durum sufferre laborem) seems to serve little purpose before 1360 (atque opere in duro durarent membra manusque), despite its making grammatical sense: it could be careless writing or reflect an alternative close to the paragraph (perhaps suggested by 5.1272: nec poterant pariter durum sufferre laborem), which Lucretius would perhaps have decided to omit. Less likely is that a reader jotted 5.1272 as similar in its content, and that this verse was subsequently adapted and incorporated into its transmitted location. Most improbable is that it represents an interpolation composed ex nihilo.72 72 A particularly complicated case is presented by the notorious verses 5.1341–9, which follow Lucretius’ infamous account of wild animals used in primitive warfare (5.1308 ff.). These nine lines, which assert Lucretius’ doubt that such animals were ever used (having just recounted it as apparently historical fact), followed by the presumption that such events did happen, have prompted more than one scholar to doubt Lucretius’ sanity and others to regard them as certainly not Lucretian. Suspicion began with Lachmann, who rejected 1344–6 and attributed them to the lector “a quo poetam, ubi incredibilia tradere videtur, versibus interpositis derisum intelleximus in libro III 412 743” (as n. 8, ad loc.). Munro went further, deleting 1341–6. Neumann, and subsequently Giussani and K. Müller, rejected all nine verses. Even more improbably, A.E. Housman, in a most surprising misjudgement (“The First Editor of Lucretius” CR 42 [1928] 122–3 = J. Diggle and F.R.D. Goodyear (eds.), The Classical Papers of A.E. Housman [Cambridge, 1972] III 1153–5),

38

david butterfield

We may at his point return to Deufert’s study (as n. 13), which divides all of his suspected passages into four categories:73 those where substantial lines are repeated (“Wiederholungen längerer Textabschnitte”), so-called “cento” passages constructed from truly Lucretian parts of the work (“Cento-Abschnitte”), shorter repeated passages of one to three lines (“Wiederholungen kürzeren Umfangs”), and freely composed passages (“Frei komponierte Interpolationen”). The repetitions have been dealt with above in (ii), and the alleged “free” interpolations dismissed as unconvincing. The final class I have found to be no more persuasive, namely, those passages declared to be “centos” constructed from several verses and phrases employed elsewhere in the work, with some further additions and alterations. The creation of such passages, which are often important in the context of the argument,74 seems hard to correlate with any possible motivation for a reader or scribe. I therefore share the deep scepticism of Francesco Nicodemo,75 and find myself unconvinced by any of the ten cases proposed.76 attributed verses 1341–3 and 1347–9 to Cicero! Yet of all possibilities the most probable is that these verses are Lucretius’ own: they may document the poet’s lack of complete certainty about the veracity of these accounts (which he presumably drew from a source – perhaps Epicurean – that no longer survives) and could have been enjoyed more neat integration in future. (Lachmann’s transposition of 1342–3, thus avoiding the isolation of futurum, may well be correct.) For a defence of the veracity of these verses, see R.B. Onians, “Lucretius V.1341–9,” CR 42 (1928) 215–17 (cf. also CR 44 [1930] 169–70), K.L. McKay, “Animals in War and Isonomia,” AJP 85 (1964) 124–35, E.J. Kenney, “The Historical Imagination of Lucretius,” G&R 19 (1972) 12–24, L. Nicastri, “Aspetti della razionalità Lucreziana,” Vichiana 16 (1987) 16–26, and C. Salemme “Animali in guerra (Lucr. 5, 1308–1349),” Inv. Luc. 31 (2009) 157–75. Kenney (ibid., 24) quotes the fascinating observation of L.P. Wilkinson that Pompey celebrated the dedication of the Temple of Venus Victrix in September 55 bc, perhaps only a month before Lucretius’ death, with games in which elephants shocked the crowds (Plin. NH 8.7): does the passage represent an (unfinished) allusion to what the poet witnessed at this very spectacle? 73 A useful three-fold categorisation of interpolation had already been proposed by R.J. Tarrant (“The Reader As Author: Collaborative Interpolation in Latin Poetry” in J.N. Grant [ed.], Editing Greek and Latin Texts [New York, 1989] 121–62): emendation, annotation, and collaborative interpolation (or emulation). 74 Cf. esp. the repetition of 1.818–19 at 1.909–10, 2.761–2 and 2.1007–9. 75 F. Nicodemo, “Le presunte interpolazioni nel De rerum natura di Lucrezio,” Vichiana 4 (2002) 288–312. This is the only concerted argument published against Deufert’s monograph, beyond the sceptical remarks of reviewers, among which see especially E.J. Kenney (CR 48 [1998] 25–7) and H.B. Gottschalk (Mnem. 52 [1999] 748–55). 76 1.907–14 (1 verse + ~ 1.817–19 + 4 verses), 2.688–99 (= 1.823–5 + 1 verse + ~ 2.336 [= 2.723] + 1 verse + = 2.336 [= 2.724] + 5 verses), 2.718–29 (5 verses + 2.336–7 + 4.649 + 5.438 (441) + 3 verses), 2.748–56 (2 verses + 1.789–93 + 1 verse + 1.673), 2.760–2 (1 verse + 1.818–19), 2.1013–22 (~ 1.823 + ~ 2.884 + ~ 1.820–1 + 3 verses [removing 2.1020 = 2.726] + 1.685 + 1 verse), 2.1058–63 (3 verses + ~ 5.429–31), 5.419–31 (= 1.1021–3 + 1.1024 = 5.187 + 5.188–91 + 1 verse + ~ 1.1026 + ~ 2.1061–3), 5.554–63 (= 3.325 + [5.537] + 8 verses), 5.1211–17



lucretius auctus?

39

Deufert’s categorisation of suspected verses was redivided into five separate classes at the close of his monograph: (a) editorial supplements (“Herausgeberergänzung”), (b) readers’ comments (“Leserkommentar”),77 (c) literary supplements and improvements (“literarisch motivierte Ergänzung, Erweiterung”), (d) a philosophical reader’s supplements, improvements, and statements (“Ergänzung, Erweiterung, Erklärung eines philosophischen Lesers”), and (e) mechanical errors of transcription (“mechanische Kopistenfehler”). Of those alleged “interpolations” to which I would give any credit, most are rightly classed as (e), which should not be regarded as any form of interpolation (for these see nn.62–4 and 66). As regards the remainder that are non-Lucretian: 1.44–9 is legitimately classed as (d); 2.105 and 5.1328 are classed as (a), and 4.1047 as (c), although they should be classed as (e). 3.474–5, 5.1388–9, 6.56–7 and 6.90–1 are classed as (d), and 5.1006 as (a)/(d). I would not attribute them to the hand of a philosophical reader desiring their insertion (d) but rather an owner who annotated his copy marginally for his own use; it is not impossible that this figure was one of the authors of the capitula throughout the work.78 The Indirect Tradition Before this survey draws to an end, we may briefly consider what evidence relevant to the question of interpolation exists beyond Lucretius’ own manuscripts. Most importantly, in several cases numerous ancient authors show in their citation of Lucretius either that supposedly interpolated verses are genuine, or that the alleged interpolation must have taken place at some remove before these authors. The anonymous author of the pseudo-Virgilian Culex, perhaps writing under Tiberius or Claudius, shows knowledge of DRN 2.24–33 in verses 62–71, although 2.32–3 were rejected as a non-Lucretian interpolation by Deufert. Such a fact means that an interpolation must have pervaded (most of ) the Lucretian tradition in the

(5 verses + 1.1004 + 5.379). It is odd that all but one of these supposed patchworks stand in Books 2 and 5. 77 Only three cases of (b) are offered – 3.206–7, 5.1341–9 and 6.565–7 – all of which I regard as Lucretian. 78 The possible exception is 5.1388–9, which could be Lucretius’ own unpolished positioning of these lines.

40

david butterfield

poem’s first seventy years of circulation.79 If so, one must suppose either that the interpolation was made at a very early stage in transmission (perhaps almost instantly after composition) or that regular comparison and collation of Lucretian manuscript rolls occurred in the late Republic or under Augustus. Neither of these suspicions seems to me at all probable. Regarding the dual appearance of the programmatic passage at 1.926–50 and 4.1–25 treated above, these verses are first cited from Book 4 by Quintilian in the late first century (Inst. 3.1.4 and 8.6.45), yet the imitation of Virgil at Geo. 3.289–93 shows imitation of 1.921–5 as well as the subsequent lines.80 Thus, if 4.1–25 were indeed copied by a non-Lucretian hand from 1.926–50, such an interpolation must have happened within the first 140 years or so of transmission, and somehow infiltrated Lucretius’ manuscripts across the board (as Quintilian shows no hesitation in providing the book reference). Several other alleged interpolations are attested among the Lucretian citations of a wide range of ancient authors, in both temporal and geographical terms.81 This evidence only militates further against the supposition of active interpolation in the poem. Finally, there are sixteen instances in which the indirect tradition presents verses not found in the Lucretian tradition: such a fact seems prima facie to encourage the notion that Lucretius’ poem can be augmented by

79 Deufert supposes that much of the interpolation could have occurred in Epicurean circles (“klar denkender Anhänger der epikureischen Philosophie” [as n. 13, 274]), perhaps at Herculaneum, which would provide a terminus ante quem of ad 79. This supposition removes the interpolations farther from Rome, and it becomes more surprising that the poet of the Culex and Quintilian show knowledge of supposedly interpolated verses. 80 Nonius also shows knowledge of the passage in Book 4, citing 4.2–3 (319,12–14; cf. also 252,9–10), 4.11–12 (190,25–7; 413,17–19) and 4.15–16 (241,32–4). 81 Seneca cited 2.55–56 at Epp. 110.6 (ad 65) in a context that suggests that he was thinking of these verses in their location in Book 2 (rather than Books 3 or 6): therefore such an interpolation must have occurred before the mid-first century ad. Aulus Gellius quoted 4.223–4 (from 4.216–29, suspected by Forbiger) at 1.21.6. Nonius cited a number of verses from passages that are alleged to be interpolated: 4.51 (from 4.45–53, suspected by Gneisse): 199,35–200,1; 4.219–20 (from 4.216–29, suspected by Forbiger): 247,12–14, explicitly cited from Book 4 (and perhaps also 311,1–2); 6.990 (from 6.988–90, suspected by G. Müller): 227,23–4. Priscian cited 5.211 (from 5.210–12, suspected by Bockemüller): GLK II 476,24–477,1 (where the error of aurarum for terraique solum must be one of memory, prompted by auras at 5.212: no similar word occurs in the environs of 1.211–12). The Bernese scholia Vergiliana cited 5.818–20 (from 5.816–20, suspected by Zwierlein) ad Geo. II.336 (cf. the citation of 5.818–19 by the Brevis Expositio of Vergil ad Geo. 2.338). Finally, Isidore appears to show knowledge of 5.1192 (suspected by Zwierlein) as well as 5.675 at Etym. 1.36.13 (cf., in turn, Julian of Toledo, De uitiis et figuris 5.32 and Ars 18.29).



lucretius auctus?

41

these or, if not, that pseudo-Lucretian verses were in circulation.82 For the most part, neither conclusion is correct. As I have argued in detail elsewhere,83 almost all of these fragments can be dismissed as incorrect ascriptions of verses to Lucretius, either through the error of the citing author (or their source) or through corruption of the original poet’s name (especially Lucilius and Lucanus) into Lucretius. There is no instance which seems to be a purposefully pseudo-Lucretian fiction. Only three fragments have a realistic possibility of being Lucretian: tantus conduxerat omnia torpor, cited by Nonius (229,1), cameraque caminis / †ecterritibus†, cited by Charisius (73.10B; cf. Beda Orth. GLK VII 266,12–14), and mensibus frigus, cited by Marius Plotius Sacerdos (GLK VI 448,7–9). However, a good case can be made to reject even this trio as non-Lucretian. Lucretius auctu non eget. To conclude this survey, I may make one final observation about the text of Lucretius. We have seen from the capitula and other marginalia that the poem, at least in its present line of transmission, was carefully read, provoking several annotations that sought to highlight difficulties and/or aid the interpretation of the text. Despite the probable existence of a Probian recension and one or more scholarly commentaries on the poem, such an active engagement on our readers’ behalf does not seem to have been carried out with any editorial purpose or active desire to change or supplement the text in circulation. Although there is undeniable evidence that verses have infiltrated parts of the poem where Lucretius did not leave them, further precision is needed henceforth in their diagnosis: it should be constantly borne in mind that not all cases of “interpolation” need be sinister and suspicious. The processes we have witnessed regularly in the transmission of DRN should not be seen as interpolation in its purest form – with the exception of the remarkable case of 5.1006. Instead, notwithstanding the contentions of the Müllers and Deufert, I assert that there has been no large-scale reworking of the text by anyone since the very first release of the poem. The significant textual damage 82  For a list of thirteen of these fragments, see the rear of Hermann Diels’ addition (Berlin, 1923–4; I 386–90), to which three further additions are made in my own survey (see following note). 83 See my monograph (as n. 22), Chapter 2, 103–35. The one fragment (XVI) that has regularly found its way into the actual printed text of the poem, viz feruere cum uideas classem lateque uagari (ex Non. 503,24–5), often inserted in the environs of 2.43, I have maintained (ibid., 118–20) to be a mistaken and contaminated reminiscence of 2.41 ( feruere cum uideas belli simulacra cientis).

42

david butterfield

inflicted by the incompetence, ignorance, and insouciance of the series of scribes through whose hands DRN passed of course nevertheless remains to be defeated in battle. Yet the modern-day editor must rest thankful that his or her job is merely to reconstruct the state of Lucretius’ poem as it finally left his house; precise answers about the poet’s ultimate designs for DRN, were it completed, must be left firmly beyond mortal ken.

Authorless Authority in Plato’s Theaetetus Zina Giannopoulou Writing lacks independence. It depends on the maid who tends the fire, on the cat warming itself by the stove; it is even dependent on the poor old human being warming himself by the stove. All these are independent activities subject to their own laws; only writing is helpless and cannot live in itself. It is comedy and despair. Franz Kafka, Diaries

1. Introduction Few authorial absences from philosophical texts have been more frequently noted and lamented than that of Plato from his dialogues. The fashionable, if arguably unanswerable, question “who speaks for Plato?” betrays, among other things, a kind of anxiety over our inability to ascribe resolutely the author’s views to any of his characters.1 To some extent, this question concerns authority: which character is authorized to speak for Plato and why? What might it mean for a character to bear this kind of authority and how might he display it? In this paper, I undertake to answer these questions by looking at the way in which authority and authorial absence affect each other in Theaetetus, Plato’s most comprehensive treatment of knowledge. Over the course of Socrates’ discussion with two mathematicians, Theodorus and his student Theaetetus, the youth offers four definitions of knowledge, none of which survives scrutiny. The dialogue proper is prefaced by an exchange between Euclides and Terpsion. In what follows, I explore the issues of authorship and authority as they arise in the preface and in the investigation of Theaetetus’ definition of knowledge as perception.

1 See, for example, Press 2000.

44

zina giannopoulou 2. Euclides and Socrates

The prefatory exchange between Euclides and Terpsion is shadowed by death.2 The two men meet in Megara and talk about Theaetetus who is being brought back home from the Athenian camp at Corinth dying of wounds and dysentery.3 They commend his conduct: he was a “fine man” (142b7) who distinguished himself on the battlefield. The incident reminds Euclides of a conversation that Socrates once had with ­Theaetetus – a reference to the dialogue proper – when the former was a few weeks away from his death, and the latter “little more than a boy” (142c6). Socrates then predicted that Theaetetus would distinguish himself, if he lived to maturity. In the fictional setting of the opening scene, Theaetetus is on the verge of death, Socrates has been dead for thirty years, and the dialogue celebrates Theaetetus’ brilliance and confirms Socrates’ prediction. The exchange is a nostalgic revisiting of the past made possible by Euclides’ painstaking work as its recorder and editor. Euclides heard Socrates narrate the discussion and jotted down memoranda. He then went home, recalled the conversation at his leisure, and transcribed it. On subsequent occasions, he corrected his account by questioning Socrates “about the points he could not remember” with the result that he “has got nearly the entire discussion in writing” (143a4–5).4 At present, Euclides and Terpsion have gathered at the former’s house where a slave reads out the written record to them. Of all Platonic works, only Theaetetus calls attention to its materiality, its status as a written text. With a striking self-awareness Euclides presents himself as the diligent transcriber of Socrates’ words, thereby usurping Plato’s role as writer of the main dialogue. William Johnson remarks that “The text itself, if the reader allows the author’s voice to intrude, can be regarded either as Plato’s copy of Euclides’ text, or Euclides’ text as remembered by Plato, and retranscribed from memory.”5 Perhaps it is more

2 These Megarians are also present at Phaedo 59c. Euclides and Terpsion are identified as natives of Megara where Plato was said to have visited them after Socrates’ death. See Diogenes Laertius 2.106, 2.108, 3.6. See further Guthrie 1969: 499–507 and Kahn 1996: 12–5. 3 The dramatic date of the prologue, and thus of the reading of the dialogue, is probably 369 bc. As alternative dates scholars have proposed 394 bc and 391 bc. For a discussion see Nails 2002: 275–7, 320–1. 4 Unless otherwise stated, translations of Theaetetus are Levett’s, revised by Burnyeat (1990); the Greek text used is the Oxford Classical Text revised in 1995 by Duke et al. 5 Johnson 1998: 586.



authorless authority in plato’s theaetetus

45

accurate to say, however, that the text is the work of the historical author Plato, which in the prologue is cast as the joint product of the oral report of the fictional author Socrates and its transcription by the fictional scribe Euclides. Socrates and Euclides are presented as the text’s efficient cause: they perform mimetically Plato’s role as author and writer of ­Theaetetus.6 In so doing, they illustrate the dialogue’s genesis: qua immaterial entity, it is a mental construct, the product of Socrates’ recollection cast in speech; qua material entity, it is a written text, Euclides’ record. Both aspects of its composition raise questions about authority. Let us first look at Socrates. As a postmodernist figure avant la lettre, he is both a character in the inner dialogue and the fictional creator of that dialogue; Socrates qua author of the narrative postdates Socrates qua participant in the actual event. Barring various factors, which I shall mention in a moment, his recollection is a faithful report of the conversation. Yet even if Socrates has an extraordinarily good memory, it is possible that he put together his account both from memory and from imagination. Since neither Theaetetus nor Theodorus is present during his report to Euclides or at the slave’s reading to Euclides and Terpsion, we cannot tell whether the dialogue is pure reportage or part reportage and part fiction. Socrates, the dead author, wields paramount authority. In his role as amanuensis, Euclides hears Socrates’ exchange, records it upon returning home at a later time, makes inquiries of points he cannot recall on subsequent visits to Socrates, and corrects his version still later at home. The temporal distancing – a past exchange is first recollected in toto and then selectively revisited – and the three stages of ­recollection – notes, later draft, even later corrections – foreground memory and its role in the transcription. Moreover, the accuracy and completeness of Euclides’ record are affected by his hearing, the correct identification of his ­memory-gaps, and the quality of his memoranda. In light of these factors, his statement that “nearly” (143a5: schedon) the entire argument has been transcribed is hardly surprising. Euclides also serves as editor: he dispenses with the “bothersome” bits of narrative between the speeches – phrases such as “I maintained” or “I said,” as spoken by Socrates, and “he agreed” or “he would not admit this,” in reference to the person ­answering – and casts the finished product in direct discourse (143c2–6). The imitative or dramatic mode eclipses the text’s oral transmission and preserves the ­liveliness of the original exchange. 6 For this notion of mimesis see Kosman 1992: 83.

46

zina giannopoulou

At the same time, however, the omission of all narrative markers renders Euclides invisible. His absence as editor is compounded by his silence throughout the inner dialogue and his failure to resume the exchange with Terpsion at the end.7 Euclides’ attitude to his task echoes Socrates’ views about writing in Phaedrus. Having written down the conversation, Euclides has lost the ability to reproduce it from memory (142d6). His inability recalls the belief that written words are reminders of what the writing is about; writing is an aid to reminding, not remembering.8 In addition, his reliance on Socrates’ words and subsequent dependence on the written text suggest that he lacks understanding of it. He thus exemplifies the claim that every discourse, once written down, “rolls about everywhere, reaching indiscriminately those with understanding no less than those who have no business with it, and it doesn’t know to whom it should speak and to whom it should not” (275e1–3). Contrast this with Socrates’ active relationship with his recollected account: not only does he relay the entire exchange from start to finish (142c8–d1), but he also takes pieces of the narrative out of sequence in order to fill the gaps of his scribe’s ­memory.9 Euclides’ estrangement from his record is complemented by the fact that the inner dialogue is read aloud by a slave, “a completely uncharacterized functionary, who serves solely as a mouthpiece or passive conduit of the discourse to his audience.”10 The slave’s reading to Euclides and Terpsion both mirrors Socrates’ narration to Euclides, insofar as both acts are oral performances conducted in front of an audience, and differs from it, insofar as it relays another’s words whose meaning the speaker cannot be presumed to understand.11 And although Euclides is present at the delivery, he listens without being able to correct a potential misreading, for he has neither memorized nor understood the text; he is a “mere site of textual inscription.”12 A parallel emerges between the two levels  7 See

Blondell 2002: 313. word “reminders” (143a1: hupomnêmata) recalls Socrates’ word “remind” (275d1: hupomnêsai). 9 Euclides is also unlike Phaedo who narrates the whole of Phaedo from memory. 10  Blondell 2002: 305. 11 The slave’s relationship with the script seems to illustrate a point Theaetetus makes later in the dialogue. When Socrates asks him whether seeing things we do not know, such as letters, signifies knowing them, the youth answers that “we know just that in them which we see; . . . we both see and know the shape and the color of the letters . . . But what schoolmasters and interpreters tell us about them, we don’t perceive by seeing . . . and we don’t know, either” (163b1–7). 12 Halperin 1992: 115.  8 Euclides’



authorless authority in plato’s theaetetus

47

of Theaetetus’ composition: on the extra-dialogic level, Plato invents and writes the dialogue from which he is absent, and on the intra-dialogic level, Socrates is said to have narrated the inner dialogue to Euclides, who removes all narrative intrusions and makes it available in Socrates’ absence, while the slave reads the script to a physically present but otherwise inert Euclides. What does this elaborate strategy of mediation and authorial unavailability tell us about authorship and authority? By fashioning a passive transcriber of Socrates’ words, Plato creates the impression that Theaetetus bears his master’s “approval and imprimatur.”13 Euclides’ and the slave’s subservience suggests that both characters are expendable intermediaries between Socrates and the dialogue’s readers.14 Even though this intimation of historicity is undercut by the fact that it occurs in the fictional world of the prologue, Socrates is still treated as the source of the inner dialogue. The authority of Euclides’ transcript is thus derivative from and parasitic upon Socrates’ authority. As David ­Halperin remarks vis-à-vis Symposium, for the narrators of the frame “philosophy seems largely to consist in a personal, not to say idolatrous, cult of Socrates. Instead of engaging in Socratic enquiry, they tell stories about Socrates . . . Plato never represents . . . his narrators doing philosophy: we never see . . . them advancing or examining philosophical claims; we only see them recapitulating uncritically the philosophical claims made by others, most of all by Socrates.”15 A similarly “cultic” atmosphere permeates the prologue of Theaetetus, this time enhanced by the two deaths against which it is set. Although Euclides intended to write down the entire conversation, his transcript may still be an incomplete record of it. Why would Plato make his transcriber self-consciously aware of this possibility? Euclides’ reservations, I suggest, are meant to signal the difficulty of transmitting in writing a conversation, which was not witnessed first-hand. To put the matter in terms familiar from Phaedrus, his (and Plato’s) misgivings concern the defective nature of the written logos, which is “an image” of the “living and animate” word (276a8–9), the only legitimate or authentic kind of speech. That Euclides’ transcript sans narrative marks is a copy not of Socrates’ report, but of the original dramatic exchange underscores the fact that it 13 Sedley 2004: 16. For Theaetetus as a vivid portrait of Socrates see Long 1998: 115–6. 14 Cf. Howland 1998: 45: “. . . the slave’s experience of reading as simply uttering the written word is well-matched with Eucleides’ superficial understanding of the act of ­writing.” 15 Halperin 1992: 118–9. Cf. Klein 1965: 11–12.

48

zina giannopoulou

is not that exchange. In his helpless reliance on another’s words, Euclides is in a cognitive position comparable to the jurors’ in the so-called Jury passage in the inner dialogue (200c7–201c6). The jurors are persuaded but not taught about matters that only an eyewitness can know. They form the opinions that the orators inculcate in them, and Euclides receives Socrates’ report of the conversation; both jury and Euclides lack first-hand knowledge. With the slave Plato inserts the reader into the cast of author/transmitter and scribe/receiver, and thus creates a second layer of intra-dialogic reception and transmission. The slave is a stranger to Socrates and thus serves as a stand-in for Plato’s readers across time and space, the incidental and unintended reader of this and any dialogue. Viewing him as a symbol is facilitated by his featurelessness and lack of individuality; the only thing we know about him is his social status, exemplified by his servile dependence on another’s script. His passive reading of the text treats it as a dead entity and fails to generate a discussion that might turn it into “live” discourse of the kind extolled in Phaedrus. When he gets to the end of the script, he “simply shuts off, like a gramophone record that has finished playing.”16 The absence of a dramatic exchange at the end of the inner dialogue suggests the intellectual lethargy into which the slave’s reading has plunged Euclides and Terpsion. The slave reads the text but cannot illuminate it. In Phaedrus Socrates compares written words to dumb paintings, which respond to questions by repeating the same words over and over again (275d4–9), and in Protagoras he compares the spoken words of the rhetoricians to dumb books, which can neither ask nor answer questions but which, like the reverberation of a gong, keep booming out the same sound until silenced (329a2–6).17 In these passages, Plato criticizes both written and spoken language, Euclides’ and the slave’s medium, respectively. The orators are like books because they cannot answer questions and keep saying the same thing forever, and the slave reads passively the text placed in his hands until it is over. His attitude to reading serves as a negative exemplar that Plato’s readers are implicitly urged to reject. For if, as Paul ­Friedländer says, “the dialogue is the only form of book that seems to suspend the book form itself,” readers must become silently active participants; lacking such participation, what is before them is nothing but a book.18 16 Halperin 1992: 112. 17 Derrida 1972: 156 brings the two passages together. 18 Friedländer 1958: 166.



authorless authority in plato’s theaetetus

49

3. Socrates and Protagoras Socrates and Protagoras cross paths during the examination of ­Theaetetus’ second definition of knowledge as perception (151d–186e). After the first definition of epistêmê as sciences and crafts has been examined and rejected, Socrates describes his art of mental midwifery. As midwife of the intellect, Socrates is barren of wisdom (150c4) and does not advance any of his own ideas. Instead, guided by the god, he draws ideas out of Theaetetus and assesses their validity. The youth offers various definitions of knowledge, none of which works, and is disabused of his epistemic pretensions. Socrates’ ad hominem reasoning foregrounds the question of authority: if the mental midwife is barren, what legitimates the results of his practice? Elsewhere, I have argued that the divine underpinnings of spiritual obstetrics validate Socrates’ assessments of his interlocutors’ beliefs.19 In this section, I focus on the richest example of spiritual obstetrics in the dialogue, the examination of Theaetetus’ definition of knowledge as perception, in an attempt to explore how Socrates receives others’ opinions, and what, if anything, constrains the interlocutor to accept his evaluation of them. Protagoras’ man-as-measure doctrine (MD) comes up for discussion as soon as Theaetetus has formulated his definition (T). Since the discussion is long and complicated, I shall restrict myself to its main points. Socrates claims that (T) is the same as the thesis advanced by Protagoras, although the latter expressed his logos “in a different way,” when he said that “man is the measure of all things, of the things which are, that they are, and of the things which are not, that they are not.” Theaetetus admits that he has often read (MD), and Socrates sets about interpreting it. He begins by saying that Protagoras “seems to mean” (152a6–8) that things are for each person as they appear to him. Finding this a “riddle” communicated to the mob, Socrates seeks illumination in the flux doctrine (H), which is usually associated only with Heraclitus but is here attributed to Protagoras, Heraclitus, Epicharmus, and Homer. According to (H), “there is nothing which in itself is just one thing, nothing which you could rightly call anything or any kind of thing . . . the things of which we naturally say that they ‘are,’ are in process of coming to be, as the result of movement and change and blending with one another” (152d2–8). He then proceeds to refine and radicalize (H) until it comes to mean that people and the things with which they interact “are tied neither to any other thing in the 19 Giannopoulou 2007.

50

zina giannopoulou

world nor to [their] respective selves. It remains then that [they] are tied to each other” (160b7–8) and that they are or become “for somebody,” or “relatively to something” (160b7–10). Perceivers and perceived objects obtain only through and for the duration of their interaction with one another. Over the course of the discussion, the range of beliefs expands to include non-perceptual beliefs. After this interlinking of theories, Socrates concludes that (T), (MD), and (H) coincide: “that of Homer and Heraclitus and all their tribe, that all things flow like streams; of Protagoras, wisest of men, that man is the measure of all things; and of Theaetetus that, these things being so, knowledge proves to be perception” (160d6–e2). This sketchy exposition foregrounds two issues that are relevant to this paper. First, Socrates invokes a dead author’s theory and is cautious about attributing beliefs to him: he is tentative about Protagoras’ meaning, uses hedges, such as “supposing Protagoras or anyone else were to ask you” (154c7), offers explanations “in accordance with the theory which we are attributing to Protagoras” (155d6), and lodges putative objections in behalf of “those people” who subscribe to (MD) (158e5–6). These expressions underscore Protagoras’ absence from the proceedings and Socrates’ role as interpreter of an intellectual “orphan” (164e3). (MD) is a piece of writing that Theaetetus has read but does not seem to comprehend. Reviving it requires that we understand its meaning through interpretation. Since it is a fragment devoid of context, Socrates associates it with (H) and its radical version in the belief that this association explains (MD)’s relativistic epistemology. For Socrates, then, theories are not explicable in isolation but must be embedded in a nexus of relevant theories and understood in relation to them. Secondly, Socrates presents the theories in support of (MD) as “truths” that must be unearthed from hidden places, since “the ancients . . . used poetical forms which concealed from the majority of men their real meaning” (180c8–d1). This suggests that meaning cannot be extracted from straightforward language but must often derive from modes of indirect discourse, such as imagery, metaphor, and various kinds of symbolic expression. Furthermore, meaning is multilayered, with each successive layer unfolding the meaning of the previous one. For Socrates (MD) has the same meaning as (T), (H) is “the hidden meaning” of (MD) that Protagoras revealed “in secret” to an inner circle of disciples (152c10), and the theory of perception underpinning the radical version of (H) is “the veiled truth in the thought of a great man, or perhaps of great men” (155d9–10).20 20 For a discussion of these connections see Lee 2005: 77–86.



authorless authority in plato’s theaetetus

51

Scholars believe that the radical version of (H) is a dialectical exercise by Plato himself, rather than something extracted from Protagoras’ own writings. As Myles Burnyeat puts it, “This is Plato’s attempt to bring out the kind of thinking which leads to a relativistic account of sensible qualities, so it already contains an element of diagnosis.”21 But although the “diagnostician” is Plato or, rather, Socrates in whose mouth Plato places the “diagnosis,” the “great man” whose thought is said to have generated the radical version of (H) is Protagoras.22 An important aspect of Socrates’ argumentative procedure emerges, according to which Protagoras’ known “truth” is linked to two other “truths,” which are also attributed to Protagoras by Plato/Socrates and are thought to explain the known “truth.” The point of the association is to bring together theories that Socrates thinks are logically interrelated and to attribute them to the same author with a view to ascertaining whether his belief set contains inconsistencies. After Socrates has interpreted (MD), he raises a few preliminary objections against it and answers them by imagining what Protagoras might say, if he were present. The dialogue form is ideally suited to the expression of these two different voices, but the answers that Socrates puts in Protagoras’ mouth are at odds with the sophist’s commitment to (MD). For example, speaking in his own voice, Socrates complains that (MD) levels all epistemological distinctions between animals, humans, and gods by equating expertise or wisdom with sense perception (161c5–d1). He imagines that Protagoras responds by berating his critics for “dragging in gods, whose existence or nonexistence [he] excludes from all discussion, written or spoken” (162d6–e2) and by insisting that they appeal not to “plausibility” and “likelihood,” but to “proof and necessity” (162e5–9). This putative answer reflects a genuine saying of the historical Protagoras: “Concerning the gods I cannot know either that they are or that they are not; for many are the obstacles impeding knowledge, both the obscurity of the question and the shortness of human life” (DK B4). The fragment emphasizes the difficulty of theorizing about the divine, and its allusive use in Theaetetus points to Protagoras’ dismissal of “the divine perspective, any universal or objective viewpoint.”23 But by correcting his critics’ attitude to his theory Protagoras appeals to authorial expertise, which is at 21 Burnyeat 1979: 76. 22 Socrates’ attribution of (H) to “all the wise men of the past” (152e2), one of whom is Protagoras (152e3), is picked up by his reference to “the thought of a great man – or perhaps I should say, of great men” (155d9–10) responsible for the formulation of the radical version of (H). 23 Polansky 1992: 113.

52

zina giannopoulou

odds with (MD)’s epistemological egalitarianism. Moreover, his request for strict argumentation is reminiscent of Socratic dialectic, and his demand for scientific reasoning subverts (MD)’s emphasis on sense perception.24 The most sustained example of this fusion of Socrates’ and Protagoras’ voices is the long speech known as Protagoras’ Apology, in which Socrates impersonates the author of (MD) and defends his maligned “orphan” against its critics’ abuse (165e8–168c1). The speech is intended to be one that Protagoras would deliver in support of his offspring, if he were alive (165e5–6), but Socrates’ spirit is reflected in it.25 As far as content is concerned, the sophist is made to speak in ways that either contradict or sit ill with (H), as when, for example, he seems to presuppose the existence of ontological stability by referring to the “same man’s knowing and not knowing the same thing” (166b5). From a methodological point of view, he is thought to reprimand Socrates for proceeding polemically rather than dialectically (167d5–168b4). By “straining every nerve to defend” the deceased sophist (168e8), Socrates becomes a hybrid, inhabiting as he does his own body and expressing the thoughts of his opponent as he, Socrates, conceives them. He is not only the speech’s author/speaker but also its addressee, for Protagoras is made to fulminate exclusively against Socrates, whom he accuses of unfair treatment (167e3), or to broach issues of interest to him, while Theaetetus is seen as “a small boy” bullied into submission (166a3–4; cf. 168d1–4).26 The Apology then is a pseudodialectical exercise: “pseudo-” because it is overtly a speech on behalf of a man who preferred long speeches to the method of question and answer; and “dialectical” because it is covertly a dialogue in which the speaker addresses a silent recipient. The spirit of dialectic also informs Socrates’ refutation of (MD). ­Protagoras’ virtual presence is here felt in a number of ways. Before his first attack, the so-called peritropê or self-refutation argument (169d10– 171c7), Socrates states that he intends to obtain Protagoras’ concession through “an appeal to his doctrine” (169e8; cf. 171b8) since the author is not present to defend himself. (MD) thus serves as a stand-in for Protagoras, its fixity representing the sophist’s deadness. During the refutation, Socrates addresses his opponent in the vocative, as if he were an active participant (170a6, c2), asks Theodorus what the consequences of people’s 24 See especially Lee 1973: 228–9. For the dichotomy between real proof and persuasion see, for example, Theaetetus 201a–c and Timaeus 51d–e. 25 I have defended this view in Giannopoulou 2009. 26 See, for example, 166a2, a6, c2, 167b5, d5, 168a3–4, a8.



authorless authority in plato’s theaetetus

53

thinking that (MD) is “true for no one” are “for Protagoras” (170e7), and at the end claims that, given his admissions to his critics, “even Protagoras himself” is bound to find his theory false (171c1; cf. 171b10). At the conclusion of the argument, he evokes Protagoras’ ghost in a memorable image: the long-dead sophist pops up as far as the neck, reproaches Socrates and Theodorus for having mistreated him, and sinks down to rush off (171c10–d3). In the second and final assault against (MD) (177c6–179b5), Socrates addresses Protagoras in the vocative once again (178b4, c1, e4) and challenges his assumption that man as the measure of all things has the “criterion of the [things that are going to happen in the future] within himself” (178b6). In his attacks against Protagoras, then, Socrates engages his opponent in an imaginary dialogue from which he extracts the sophist’s agreement with the critics’ verdict. Is his procedure a legitimate way of dealing with MD? Before I answer this question, let me briefly consider two points. First, although Socrates claims to be speaking in behalf of Protagoras throughout this part of the dialogue, he often worries that his opponent is not present to make his own case or reply to his detractors. For example, when he and Theaetetus appear to have scored a premature victory against the sophist, Socrates says, “I don’t think that this would have happened, my friend, if the father of the other tale were alive. He would find plenty of means of defending it. As things are, it is an orphan we are trampling in the mud” (164e2–4). His misgivings about the legitimacy of speaking for another echo his claim in Phaedrus that when the written logos is “faulted and unfairly attacked, it always needs its father’s support; alone it can neither defend itself nor come to its own support” (275e3–5).27 A little later, he criticizes himself and his interlocutors for “making alarming admissions” by paying little attention to their verbal expressions (165a5– 7). The peritropê is framed by references to the possibility of Protagoras’ successfully countering the charges leveled at him. Before the argument, Socrates wonders whether Protagoras would accept the view attributed to him in the Apology that “some men are superior to others in questions of better and worse, these being ‘the wise’ ” (169d6–8) and decides to turn to (MD) for answers because it might be claimed that “[the critics] had no authority on his behalf ” (169e3). His last critical gesture occurs at the end of the peritropê, where he says that if Protagoras were to show up he might be able “to convict his critics of talking nonsense” (171d2) before 27 Cf. Seventh Letter 344c.

54

zina giannopoulou

he ducked down to rush off again. Common in all these self-reproaches is the suspicion that Socrates and his interlocutors win cheap victories at ­Protagoras’ expense by abusing an absent author’s mental child. Is the suspicion meant seriously or ironically? Secondly, in his effort to find a worthy substitute for his absent opponent Socrates turns to Theodorus, Protagoras’ friend (161b8, 162a4, 168c3, e7, 171c8, 183b7). Friendship connotes like-mindedness and goodwill, and thus Theodorus is the best living spokesman of the sophist. Yet the mathematician is unwilling to join the discussion, even though he apparently disagrees with Protagoras (162a), because he agrees with Socrates that one should not answer contrary to one’s own opinions (162a6–7). He thus prefers to remain a spectator (162b5–9) and refers Socrates back to Theaetetus. A second time Socrates urges Theodorus to defend (MD), as one of the “guardians” of Protagoras’ “orphan” (164e3), and again Theodorus refuses on the grounds that he has turned from “bare words” (165a2) to mathematics and is not qualified to assist Protagoras. When Socrates goes on to assault the sophist, Theodorus protests that they are running his “friend” too hard, to which Socrates replies that they are not running past the truth (171c10–11). Despite his protestations, the reluctant “comrade” finally opts to defend the truth, a fundamentally philosophical choice, rather than his wronged friend.28 Respect for one’s own truth and for the truth, as established in the peritropê, undermines (MD). Why should the relativist Protagoras care? This question goes hand in hand with the two questions I raised earlier but did not answer: the legitimacy of Socrates’ attempt to involve Protagoras in an imaginary dialogue whose outcome implicates him in the rejection of (MD) and the seriousness of his suspicion that he wins cheap victories at the expense of an absent author who might be able to defend his mental child if he were to show up from the dead. By its very nature (MD) licenses subjective interpretations. If man is the measure of all things and if one of these things is (MD), then man can measure or judge (MD). The word “man” may be used in the collective sense of “humanity,” and this is probably how the thesis is naturally read, but Socrates gives it the individualistic sense of “each human being” by taking (MD) to mean that “as each thing appears to me, so it is for me, 28 For the difference between Theodorus’ notion of friendship and philosophical philia see Blondell 2002: 286–9. In Nicomachean Ethics 1096a16 Aristotle says about friendship and philosophy that “although both are dear, piety demands that we give greater respect to the truth.”



authorless authority in plato’s theaetetus

55

and as it appears to you, so it is for you – you and I each being a man” (152a6–8). He then illustrates this meaning with the example of the wind, which posits mutually incompatible perceptual beliefs: when the same wind blows, one of us feels cold and the other not. By breaking up humanity into its individual members, Socrates transforms univocity into disagreement and makes room for dialogue. This enables him to distinguish between Protagoras’ and his own conception of disagreement: whereas Protagoras solves disagreement by relativizing the truth of the beliefs to those holding them, Socrates views disagreement as emblematic of a genuine difference in opinion, one that calls for a dialectical arbitration of the truth. For Protagoras, a person’s beliefs are true by virtue of being his beliefs, and for Socrates a person’s beliefs are true if they survive dialectical scrutiny. We may now see why Socrates labels (H) and its radical version “truths,” as well as why (MD) is referred to as Protagoras’ “truth”: all these theories are “true” by virtue of being Protagoras’ theories and are true for him, but what makes them really or objectively true is their passing the obstetric test. The peritropê dramatizes the difference between the subjective and the objective notion of truth by showing that (MD) is self-refuting. In judging (MD) to be false, Socrates claims to be speaking on behalf of all human beings, both present and absent, Protagoras’ followers included, who “believe in the existence of both wisdom and ignorance among themselves” (170b6–7), “do not always judge what is true” (170c3), and think that “human judgments are both true and false” (170c3–4).29 Theodorus agrees with these admissions and thus turns his earlier piecemeal disagreement with various implications of his friend’s thesis into a full-fledged betrayal. The only exception from this universal consensus against MD is Protagoras, who authored it and must believe that it is true. At the dialogue’s dramatic date, however, Protagoras is not alive but dead; he is an ex-man. Therefore, his single belief in the truth of (MD) not only counts for naught but also contradicts his thesis, which states that man is the measure of all things.30 So even if Protagoras could stick his head up from the dead, he could not disprove his critics because he would speak as a ghost to a gathering of humans. In discussing Plato’s use of absent theorists in the dialogues, Mary Margaret McCabe remarks that “if these people cannot 29 References to human beings or humanity occur explicitly at 170a6, 7, b2, 6, c3, and e2, and implicitly at 170a8, b7, d6, 8. 30  This is a summary of my interpretation of the peritropê. For the entire argument see Giannopoulou 2011.

56

zina giannopoulou

really talk, or turn up, or appear live in a conversation, they may be mere figments of Plato’s imagination. To their historical status, then, is added a fictional dimension; and the use of these historical fictions has a distancing effect.”31 In Theaetetus, Protagoras’ “fictionalization,” his absence from the human circle of Socrates’ interlocutors, deprives him of his ultimate authority as author of (MD) by making him contradict both the letter and the spirit of his theory. For the Socrates of Theaetetus, mental midwifery determines the truth or falsity of a belief. Protagoras and Socrates differ in their attitude toward truth: whereas the former deems all beliefs veridical for those holding them, the latter makes mental midwifery the arbiter of doxastic truth. This difference underscores the contrast between Socrates’ reception of (MD) and Euclides’ reception of Socrates’ report in the prologue. Whereas Euclides receives Socrates’ words reverentially with a view to preserving them, Socrates receives Protagoras’ words critically and examines their associations and ramifications. Euclides is like a historian reporting his findings, and Socrates is like an archaeologist excavating the foundations of theories. In the latter’s case, authority is bound up not with nominal authorship, but with the procreation of ideas and the critical testing of them by the mental midwife, whose craft enjoys divine support. Assessing the truth of beliefs by associating them with other relevant beliefs rescues them from the death to which de-contextualization condemns them. Devoid of context, a belief may be intelligible qua utterance but cannot be understood in its implications and ramifications. (MD) is precisely a belief of this sort, an isolated, riddle-like fragment in need of explanation. By itself, it is as defunct as its author, an ontological symmetry that the peritropê exploits with spectacular results. In addition, (H) conceives of beliefs as piecemeal and episodic events: how things appear to a person at a time is how they are for him at that time. Appearances are separate cognitive events, and beliefs are logically unrelated to other beliefs.32 A person can convey another’s beliefs but cannot scrutinize or assess them in light of the evidence offered in their support. Euclides’ attitude toward Socrates’ exchange in the prologue reflects this intellectual stance, albeit without the excesses of Heraclitean flux.

31 McCabe 2000: 13. 32 McCabe 2000: 35.



authorless authority in plato’s theaetetus

57

4. Conclusion Theaetetus presents us with two ways of understanding authority in the absence of the author. In the prologue, Euclides claims to have produced a faithful transcript of Socrates’ report. His authority is that of a conscientious recorder of another’s words. If we set aside his reservations about the completeness of the written record and assume that the transcript replicates the oral report, we see that Socrates wields authority by proxy; Euclides effaces himself in order to preserve Socrates’ authority. In the main dialogue, Socrates receives Protagoras’ written doctrine and evaluates it. Throughout his examination of (MD), Socrates, a living human being, speaks on behalf of and against his absent opponent, whether as an individual in his own voice, or as Protagoras’ impersonator, or as representative of all humanity. Here authority is implicitly given to him by (MD) itself, which makes man the measure of all things. Protagoras’ thesis strips the dead author of his authority and transfers it to his human opponents. Socrates executes this authority in the context of mental midwifery, which enjoys divine support and yields infallible results. Socrates’ authority is thus indisputable and it mingles with god’s absolute authority. References Blondell, R. (2002), The Play of Character in Plato’s Dialogues, Cambridge. Burnyeat M.F. (1979), “Conflicting Appearances,” Proceedings of the British Academy 65: 69–111. —— (1990), Introduction to M.J. Levett (trans.), The Theaetetus of Plato, rev. M. Burnyeat, Indianapolis. Derrida, J. (1972), “La pharmacie de Platon,” in La Dissémination, 69–197. Paris. Diels, H. and W. Schubart (1905), Berliner Klassikertexte, ii, Berlin. Friedländer, P. (1958), Plato I, New York. Giannopoulou, Z. (2009), “Objectivizing Protagorean Relativism: The Socratic Underpinnings of Protagoras Apology in Plato’s Theaetetus,” Ancient Philosophy 29: 1–22. —— (2011), “In and Out of Worlds: Socrates’ Refutation of Protagorean Relativism in Theaetetus 170a-171c,” Ancient Philosophy 31.2: 275–94. Guthrie, W.K.C. (1969), A History of Greek Philosophy, vol. 3, Cambridge. Halperin, D.M. (1992), “Plato and the Erotics of Narrativity,” in J.C. Klagge and N.D. Smith (eds.), Methods of Interpreting Plato and His Dialogues, Oxford Studies in Ancient Philosophy, suppl., Oxford: 93–130. Howland, J. (1998), The Paradox of Political Philosophy: Socrates’ Philosophic Trial. Lanham, MD. Johnson, W.A. (1998), “Dramatic Frame and Philosophic Idea in Plato,” American Journal of Philology 119.4: 577–98. Kahn, C.H. (1996), Plato and the Socratic Dialogue: The Philosophical Use of a Literary Form, Cambridge. Klein, J. (1965), A Commentary on Plato’s Meno, Chapel Hill.

58

zina giannopoulou

Kosman, L.A. (1992), “Silence and Imitation in the Platonic Dialogues,” Oxford Studies in Ancient Philosophy (suppl.): 73–92. Lee, M.-K. (2005), Epistemology After Protagoras: Responses to Relativism in Plato, Aristotle, and Democritus, Oxford. Long, A.A. (1998), “Plato’s Apologies and Socrates in the Theaetetus,” in J. Gentzler (ed.), Method in Ancient Philosophy, Oxford: 113–36. McCabe, M.M. (2000), Plato and his Predecessors: The Dramatisation of Reason, Cambridge. Nails, D. (2002), The People of Plato: A Prosopography of Plato and Other Socratics, Indianapolis. Polansky, R.M. (1992), Philosophy and Knowledge: A Commentary on Plato’s Theaetetus, London and Toronto. Press, G.A. (2000), Who Speaks for Plato? Studies in Platonic Anonymity, Lanham, MD. Sedley, D. (2004), The Midwife of Platonism: Text and Subtext in Plato’s Theaetetus, Oxford.

The Poet and the Forger: On Nonnus’ False Biography by Constantine Simonides David Hernández de la Fuente Two characters converge in the present article:* a poet and a forger. Nonnus of Panopolis, a Late Antiquity epic poet to whom the Dionysiaca and the Paraphrase of the Gospel of John are attributed, is hardly anything but a name for most of us.1 His works are extensive and controversial: the great epic poem about Dionysus – the largest extant poem in Ancient Greek – and a complex epic version of the life of Christ, suffused with theological subtleties. Nevertheless, and despite recent attempts at shedding light on it, his life is still shrouded in mystery. Only a genial forger of the nineteenth century, Constantine Simonides, palaeographer and trader of icons and manuscripts, was able to come up with apocryphal data on the life of this poet, certainly in his most peculiar manner. This forger, about whom detailed studies have recently been published,2 tricked erudite Germany (even the well-known philologist K.W. Dindorf fell into his clutches) and was on the brink of enticing Nonnus’ nineteenth-century editor Count Marcellus, aristocratic philologist and retriever of Greek antiquities, with a false biography of Nonnus. In view of the exasperating and obstinate silence of the sources, the life of Nonnus could have only been recreated by a forger, and, shortly afterwards, by some historical fictions that will be briefly reviewed hereafter, namely, a short story by Richard Garnett and a novel by Margarete Riemschneider. Constantine Simonides left a deep imprint in Greek studies at the beginning of the nineteenth century, at the time of European philhellenism, enthusiastic manuscript-purchasing campaigns, and heated competition for the editions of Greek Classics in Germany and France between publishing houses such as Tauchnitz-Teubner and Firmin-Didot and between academics from both countries. The long shadow of this

* This paper is based on research funded by project FFI2009–09465 of the Spanish Min­istry of Science and Innovation. 1  Cf., in general, Hernández de la Fuente 2008. 2 E.g., Schaper 2011.

60

david hernández de la fuente

rivalry still reaches us today, and is deeply felt, as the case of the Artemidorus papyrus attests.3 Simonides was born in the Aegean island of Syme, and from a very young age he was on good terms with members of the Greek Orthodox Church. He spent several years of his youth, from his ninteeth to his twentysecond year, in monasteries of Mount Athos, where he did his best to acquire knowledge and expertise in the Greek manuscripts held in the sacred Chalcidice peninsula.4 Maybe his first forgeries, possibly practical exercises of palaeography at the very beginning, are also to be dated to this time, and also his first purchases of Byzantine manuscripts, with which he would start trading upon leaving the Hagion Oros. His privileged access to ancient parchments, which he could reuse imitating Byzantine palaeography, took him to Athens around 1846–49, where he would try to sell his false manuscripts as if they were authentic: among other treasures, he would offer fragments of the Gospels and Homeric poems, some of which he even tried to sell the new king of independent Greece. Putting forward as evidence his repeated stays in Mount Athos, Simonides claimed that he had found a large number of manuscripts in a secret store, containing, for example, works by Anacreon, Hesiod, and Homer, which he tried pass off as authentic in Athens. There was, however, no agreement about their authenticity, and he left the country. He arrived first to Istanbul, around 1850, where he continued with his counterfeiting activity, trying to sell manuscripts to local antiquarians. Simonides then traveled to England in 1853, where he sold some manuscripts to Sir Thomas Phillips (as reported by the journal of the time, Athenaeum, on 9 February 1857) and edited facsimiles of the Gospels whose originals he purported to date to a few years after the death of Christ. He then moved into France and Germany, offering to European dealers, libraries, and scholars manuscripts in a purportedly archaic handwriting; works by Sanchouniaton – an almost legendary Phoenician author of three lost works – and Homer, Aristotle’s poetry and a purported history by Uranius, among others. Simonides was regarded as an expert in Greek literature and palaeography and had his say in all the scholarly polemics of the time such as, for example, the dating of Irenaeus of Lyon. Simonides was denounced several times. The first time, in Istanbul, he was denounced by the German scholar A.D. Mordtmann, an expert

3 Canfora 2010 and 2011. Canfora and Bossina 2008. 4 On his life and deeds, Farrer 1907, 39 ff.



the poet and the forger

61

in manuscripts, in an article published in the Allgemeine Zeitung in Augsburg (1853) and echoed in the same year by Ludovico Lalanne in the journal Athenaeum Français (10 December, 1853). Apparently, and despite his bad name, some of his findings had been sensational. Simonides purported to have found The Shepherd of Hermas on Mount Athos and to have taken six copies to Europe; this forgery was included in an edition in Leipzig in 1856 by professor K.W. Dindorf.5 It was also there that he sold his most notable forgery, Uranius’ Historia Aegyptiae. Uranius was an author that Simonides dated to the fourth century of our era and whose manuscript he passed off as authentic with such success – he even convinced a commission of the Prussian Academy, endorsed by professors Boeckh and Lepsius – that the well-known philologist K.W. Dindorf, on behalf of the Prussian king, paid a substantial sum for it and began to prepare a critical edition of it in Leipzig. It was on the verge of being published, but the book was retrieved just in time from the printers after Professor Tischendorf and other scholars meticulously examined the purported codex rescriptus sold to Dindorf upon royal commissioning in view of the record of forgeries of Simonides in England. The matter of the false palimpsest gave rise to a famous scholarly polemic between Tischendorf and Lepsius about the authenticity of the manuscript, which led to an exchange of letters in the press and was later collected in a documented work on Simonides’ Uranius.6 A second article by Lalanne in Athenaeum Français (9 February, 1857), reporting on this very incident, accused him also of trying to sell several works, such as forty-seven comedies by Menander, an author only extant in fragments, Sophocles’ complete plays, the full catalogue of the library of Alexandria, etc.; those interested were warned against Simonides. The whole issue ended, as Lalanne reported, with the arrest of Simonides in Leipzig and impounding of the sum of money he had received for the Uranius’ palimpsest. After having tricked cultivated Europe for years, in 1862 Simonides published an article in The Guardian newspaper in which he revealed himself as a forger and claimed to be the real author of the Codex Sinaiticus, the most ancient manuscript of the Bible, discovered by Tischendorf; he claimed to have written it during his youth. This lie and many others placed him under scrutiny by manuscript experts. At the end of the day, Simonides is still relevant today because the famous Artemidorus ­papyrus,

5 Sandys 1910, v. 3. 6 Fritzsche 1856.

62

david hernández de la fuente

which contains what seems to be part of Book II of his Geography, is considered a forgery probably produced by Simonides himself by scholars such as Luciano Canfora.7 The deeds of this master forger continued to the point of falsifying his own date of death in order to carry on with his activities, but an in-depth examination of this character’s background and career should be the object of another article. For the time being, we will focus on one of his most peculiar forgeries, in which he was trying to offer biographic information about specific authors, which roused great academic interest at the time. Simonides announced that he had discovered Demetrius’ lost treaty On Namesake Poets and Writers. Demetrius of Magnesia is mentioned by Diogenes Laertius several times, as an author from the first century bc.8 However, somewhat anachronistically, Simonides would show fragments of this purported sensational finding containing biographical data on Late Antiquity authors, such as Nonnus of Panopolis, from the fifth century ad, or the above-mentioned false Uranius. Drawing on contemporary sources, Masson has dug into the particulars of Simonides’ visit to France in 1854.9 This French scholar mentions with a certain degree of admiration how, following the bad press that accompanied Simonides and despite the warnings made by Mordtmann and Lalanne,10 the forger could still have access to the most relevant men of letters in France upon his arrival to Paris, and even captivate them with his mystifications. Among others, the writer Sainte-Beuve, who gave him recommendation letters to the Imperial Library in Paris, the Hellenist Ernest Beulé, the Secretary of the Academie Française, Abel Villemain, or even the Home Secretary, Hippolyte Fortoul. All these hosted and smothered him with their attentions in Paris. Among the scholars that Simonides was introduced to was Count Marcellus, a well-known character at the time, as specified by the biographer and supporter of Simonides, Charles Stewart,11 for being one of the finders of the Venus de Milo. This character, MarieLouis-Jean-André-Charles Demartin du Tyrac (1795–1865), full name of Count Marcellus, had a stellar political career from his admittance into the personal guard of the Duke of Angoulême until his diplomatic missions in Constantinople, Palestine, London and Madrid, either as Secretary of Embassy or even as Minister Plenipotentiary. Marcellus took part in 7 Parsons 2008, Canfora 2010 and 2011. 8 Diogenes Laertius I 112, V 3. 9 Masson, 1994, 370 ff. 10 Masson, 1994, 377. 11 Stewart, 1859, apud Masson, 1994, 372.



the poet and the forger

63

1820 in the unearthing of the Venus de Milo together with the adventurer Dumont D’Urville, subsequently purchasing her on behalf of the Marquis de la Rivière to present to Louis XVIII. In the history of classical studies, this multifaceted Hellenist and aristocratic diplomat from the time of the Bourbon Restoration is well-known for having published in 1856 the first translation of the Dionysiaca by Nonnus of Panopolis, alongside a critical edition of the text. It would be followed by the edition and translation of the Paraphrase of the Gospel of John by Nonnus, which meant that the complete works of this author were translated into a modern language for the first time. Nor should the literary dimension of Count Marcellus, however, be forgotten, since it led him to write two books of travels, Souvenirs de l’Orient (1839) and Épisodes littéraires en Orient (1851), full of anecdotes and tales, which were so much to the taste of the time. The text and translation of the supposed biography of Nonnus of Panopolis run as follows: ᾽Απάνθισμα περὶ ὀμωνύμων ποιητῶν καὶ συγγράφεων τοῦ Πανιερωτάτου καὶ Θεοπροβλήτου Μητροπολίτου Λιβύης, κυρίου Διονυσίου, τοῦ κατά κόσμον Δημητρίου Μάγνητος ἀπανθισθὲν ἔκ τε τῶν συγγραμμάτων Δημητρίου Μαιάνδρου, Δημητρίου Θαρσυκλέους, Δημητρίου Κλέωνος, τῶν ἐκ τῆς κατὰ Καρίαν Μαγνησίας, καὶ Καλλιμάχου ῾Ηρωδιανοῦ τοῦ Κυρηναίου, τοῦ τοὺς Πίνακας συνταξαμένου τῶν ἐν παιδείᾳ λαμψάντων.  Νόννος Πλεονίππου, ὀ τῶν Διονυσιακῶν ποιητής, θέσει μὲν ἧν Πανοπολίτης, φύσει δὲ Ἡλιουπολίτης γεγονὼς κατὰ Κωνσταντῖνον τὸν Μέγαν. Διήκουσεν Ἀριστοκλέους Διδύμου τοῦ Ἀλεξανδρέως, ὅν καὶ διεδέξατο· ἔγραψε δὲ σὺν τοῖς Διονυσιακοῖς καὶ περὶ Αἰγυπτίων θεῶν ἐν βίβλοις δυωκαίδεκα, καί ὕμνους εἰς Ἀφροδίτην τεσσαράκοντα μέτρῳ Σαπφικῷ· μετὰ δὲ ταῦτα ὑπὸ Σεραπίωνος ἐπισκόπου τῆς κατ᾽ Αἴγυπτον Θμωίδος κατηχηθεὶς, καὶ πρὸς Ἀθανάσιον γενόμενος τὸν Ἀλεξανδρείας, καὶ τούτῳ μαθητεύσας μικρὸν, τὸν παλαιὸν ἀπεκδύεται ἄνθρωπον τῷ βαπτίσματι, καὶ Πανοπολιτῶν ἱερατεύει ἐπὶ ἔτη τριάκοντα πρὸς τοῖς ἑπτὰ, καὶ τὸ κατὰ Ἰωάννην εὐαγγέλιον ἐπικῷ παραφράζει μέτρῳ καὶ Ἠσαίαν καὶ Ἱερεμίαν τοῦς προφήτας. ἀσθενείαι δὲ καμών ἐτελεύτησεν ἐν τῷ τεσσαρακαιδεκάτῳ ἔτει τῆς Βασιλείας Θεοδοσίου τοῦ [νεωτέρου] βιοὺς ὅλα ἔτη ὀγδοήκοντα πρὸς τοῖς δύω.  Ἐγένοντο δὲ καὶ ἄλλοι Νόννοι δύω. Πρῶτος ὀ Θάχιδος Αἰγύπτιος τὸ γένος, καὶ οὖτος ἐξ Ἀρύπης πόλεως, μικρόν τοῖς χρόνοις πρεσβύτερος τοῦ Πανοπολίτου ὅς καὶ τὰ Αἰγυπτίων ἐν ἑπτὰ ἐκωμῴδησε, βίβλοις ἐμμέτρῳ λόγῳ γελοῖα ἐπιγράψας αὐτὰ. Δεύτερος ὀ Ἀρώχεως, Ἀνυσίτης τὸ γένος, τοὐπίκλην φιλόσοφος, καὶ Ὠριγένει γνώριμος. Ἔγραψε φιλόσοφα καὶ Αἰγυπτίων σοφῶν διατριβὰς. ἐτελεύτα δὲ μικρὸν πρὸ τῆς τοῦ Ὠριγένους τελευτῆς ὑπέργηρως πάνυ. Anthology about namesake poets and writers edited by Dionysus, bishop of Libya, most beloved and righteous to the Lord, whose worldly name was Demetrius of Magnesia. Compiled from the writings of Demetrius, son of Menander, Demetrius, son of Tharsicles, Demetrius, son of Cleon, all from the city of Magnesia, in Caria, and from the writings of Callimachus of

64

david hernández de la fuente Cyrene, son of Herodianus, who composed the Pinakes, and of those who have shone in erudition.  Nonnus, son of Pleonippus, the poet of the Dionysiaca, lived in Panopolis but was originally from Heliopolis, and was born in the times of Constantine the Great. He was a disciple of Aristocles Didymus of Alexandria, whom he also succeeded in his school. He wrote, as well as the Dionysiaca, twelve books about the gods of Egypt and forty hymns to Aphrodite, in Sapphic verse. After this, he was catechized by Serapion, bishop of Thmois in Egypt and afterwards by Athanasius of Alexandria. After a short time as disciple of the latter, he left his old self behind through baptism. He exercised his ministry in Panopolis for thirty-seven years. He paraphrased in epic verse the Gospel of John and the prophets Isaiah and Jeremiah. After falling ill, he died in the fourteenth year of the reign of Theodosius the Young, having lived a total of eighty-two years.  There were two other Nonnus. The first one, of Egyptian race, was the son of Thachis and originally from the city of Arype, a few years older than the one from Panopolis. This one ridiculed Egyptian uses and wrote his humorous works in seven books in verse. The second one, son of Arocheus, from the city of Anysis, was called “the philosopher” and was acquainted with Origen. He wrote philosophy and the Disputations of the Egyptian sages. He died shortly before the death of Origen, at a very advanced age.

This falsification in grandiose mode truthfully imitates the style of short Late Antiquity biographies, in the fashion of grammarians and biographers like Eunapius of Sardis or later Byzantine lexicons. It is made as only a expert forger could do, interweaving the right doses of fantasy and real data. The fact that Marcellus had been immersed for a long time in the elaboration of a great work on Nonnus was known in various circles, both in England and in France. In a couple of articles in the journal Athenaeum, in its French version, there are reports of Marcellus’ progress, and fragments of his translation of the Dionysiaca had also been published.12 It is not far-fetched to imagine that Simonides knew of Marcellus’ interest in Nonnus when he made public the news that he had a manuscript with the biography of the poet. Marcellus relates in a letter dated to 17 February, 1856 how one day, during a stay of Simonides in Paris, they were talking about the difficulties the French philologist was encountering in establishing a chronology for Nonnus and his works when the Greek forger told him that he thought he remembered that in one of his manuscripts he had read a biography of the author of the Dionysiaca. Marcellus, who, at that time, according to his own account, had no doubts about the authenticity 12 Athenaeum Français 1855, 164 and 1857, 80.



the poet and the forger

65

of Simonides’ material, asked him enthusiastically to send it to him: fifteen days later he received the copy and ascertained that there were serious chronological inaccuracies in the text. He told Simonides that he would not be using the biography for his edition since he suspected it to be a forgery. Simonides, according to Marcellus, “received my statement without any apparent embarrassment and, making use of that particular phlegm that never abandoned him, did not press the matter any further.” The quotation, reflecting the literal words of the Hellenist showing his surprise at the forger’s boldness, is taken from the letter included in the above-mentioned study by Masson. The issue of Nonnus’ biography was mentioned, finally, in an article entitled “Literary Forgery” published in the journal Athenaeum, in its English version, on 16 February, 1856, in which references were also made to the dubious business of Simonides in Leipzig. The false biography, as we mentioned, is remarkable, despite its inaccuracies, already pointed out by Marcellus and Masson. All the Demetriuses quoted at the beginning are made up, with the exception of Demetrius of Magnesia quoted by Diogenes Laertius. There is a certain degree of ambiguity regarding the authorship of the fragment, which we shall briefly examine later on. But the attested name of the poet and scholar Callimachus of Cyrene, author of the tables or catalogue of authors of the library of Alexandria, is then inserted. However, the name of the father of Callimachus, Herodianus, is also an invention, since his name was Battus, according to tradition. The same could be said about Nonnus’ supposed father, Pleonippus. The forger Simonides is extremely cautious with the characters mentioned, and carries on amalgamating reality and fiction in a skilled pastiche. The person supposedly responsible for the conversion of pagan Nonnus to Christianity is none other than Serapion, a historical character born at the end of the third century and bishop of Thmois around 340. Several treatises on psalms and against Manicheans are attributed to this Christian scholar, labelled “scholastic,” aside from gospel paraphrases that would draw him closer to the Nonnus of the Paraphrase of John. There is in the work another well-known character, Athanasius of Alexandria (296–373), patriarch and famous author of the life of Saint Anthony the Great. Simonides bases his sham on real data about the time, since Photios mentions the relationship between Serapion and Athanasius, telling how the latter used to present his writings to the wise bishop of Thmois for their correction.

66

david hernández de la fuente

The forger researched mostly real cities for the geography of Nonnus’ life. Panopolis and Heliopolis are well known. Heliopolis is said to be his place of origin, while he had become famous for his long stay, more than thirty years, in Panopolis. Alexandria, where the poet says he is writing – “across from the island of Pharos” – seems to be the place where he studied. There he followed the pagan teachings of the sage Aristocles Didymus, perhaps following the model of Hierocles, Neoplatonist master in Alexandria, and his successor, Hypatia. It is worth noting the works of the pagan period attributed to Nonnus, apart from the Dionysiaca: a work in twelve books about the gods of Egypt and forty hymns to Aphrodite. Maybe the forger took into account not only the Panopolis poet’s predilection for religious hymns, with which he interspersed his work dedicated to Dionysus, including some dedicated to Aphrodite,13 but also the substantial length of his pagan work, the Dionysiaca, in forty-eight books. His conversion to Christianity is nicely embellished: he places it in Thmuis, one of the cities of Egypt noted by Herodotus, and site of the ancient bishopric where Serapion held office. The false biography mentions, furthermore, several new Christian works by Nonnus that would have astonished his French editor: the paraphrases of prophets Isaiah and Jeremiah, which corresponded to the paraphrases in fashion at the time such as Apollinaris of Laodicea’s paraphrase of the psalms. Finally, the list of poets and writers also called Nonnus in the apocryphal fragment deserves special mention: there is a Nonnus of Aripe, son of Thachis, a Christian author of writings ridiculing Egyptian paganism. Of course, he is pure fabrication. His city, Arype, is mentioned by Stephanus of Byzantium, but his location in Egypt is erroneous and should have been in continental Greece. The third Nonnus, son of Arochus, is from the city of Anysis, also mentioned by Herodotus in his description of Egypt, like Thmuis. He is described as a pagan or converted philosopher from the time of Origen. Maybe these two false Nonnuses were simply a splitting of the poet into a homonymous pair polarized as pagan and Christian. Finally, the chronological mistakes in the false biography’s chronology of Nonnus are obvious. The main problem arises from placing his birth “around [κατά] the time of Constantine the Great.” Indeed, it is stated that he dies in the fourteenth year of the reign of Theodosius II (408–450),

13 Cf. Braun 1915. The Hymn to Aphrodite is in Dion. XLI 315 f., although interrupted by a lacunae.



the poet and the forger

67

i.e., in 422. But it is also said that he was 82 at the time of his death, which would mean that he had been born in 340, this is, three years after the death of Constantine I (337). Unless this κατά can indicate an approximate value, the arithmetic is wrong. In any case, Nonnus’ chronology is usually set between the Council of Ephesus in 431 and the earthquake that destroyed Beirut in 551 due to the reference made to the poet by Agathias of Myrina (530–582).14 However, the most monumental mistake of all, and particularly eye-catching, is, of course, pretending that the scholar who wrote On Namesake Poets and Writers, dated by Diogenes Laertius to the first century bc, made any comments about such a Late Antiquity author as Nonnus. In spite of that, one must acknowledge a genial touch in Nonnus’ apocryphal biography by Simonides, for it includes a shrewd ambiguity regarding precisely the chronology of the namesake Demetriuses. He speaks of a work On Namesake Poets and Writers (Περὶ ὁμωνύμων ποιητῶν τε καὶ συγγραφέων), which is attributed by Diogenes Laertius (fl. 3rd century ad) to a Demetrius of Magnesia on Maeander, a grammarian and biographer of the first century bc: concretely in Vitae Philosophorum I 112 (ἀλλὰ Δημήτριος ὁ Μάγνης ἐν τοῖς Περὶ ὁμωνύμων ποιητῶν τε καὶ συγγραφέων . . .) and V 3 (. . . ὥς φησι Δημήτριος ὁ Μάγνης ἐν τοῖς Περὶ ὁμωνύμων ποιητῶν τε καὶ συγγραφέων).15 Demetrius of Magnesia is also referred to16 by Cicero, who was his contemporary: according to Cicero (Ad Atticum VIII 11, IV 11), Demetrius was author of a work On Concord (Περὶ ὁμονοίας), which he sent to Atticus and which Cicero was eager to read. But Simonides here shrewdly presents this work not as Demetrius’ Περὶ ὁμωνύμων ποιητῶν τε καὶ συγγραφέων itself, but as an anthology edited by a seemingly Byzantine or Late Antique bishop of Libya, whose worldly name is Demetrius of Magnesia and who is compiling the anthology from the writings of two other Demetriuses of Magnesia and the writings of Callimachus. This manoeuvre would allow a margin of doubt; we can imagine the hesitation of Le Comte de Marcellus. Undoubtedly Simonides, seeking to satisfy both the scholarly and literary interests of Marcellus in Nonnus, had composed a short and plausible biography for the poet from Panopolis that at long last unravelled the mystery of his double work, pagan and Christian, and provided a reasonable

14 Cf. in general, Vian’s introduction, Vian 1976. 15 Cf. also FHG IV 382. 16 Cf. v. gr. Plutarch, Vit. Demosth. 15, 27, 28, 30.

68

david hernández de la fuente

chronology. In spite of the obvious mistakes and forgeries, the strategy of multiplying the personality of Demetrius of Magnesia was simply brilliant. In elegant style the scholarly author of a work On Namesake Poets and Writers is divided into three namesake grammarians as if reflected in a “game of mirrors,” that would have amused Jorge Luis Borges. Simonides’ forgery is without a doubt a work of genius. Nonetheless, he could not fool the meticulous French Hellenist as he had tricked his critics at the Academy of Berlin. On the other hand, it should be remembered that, in the midnineteenth century, the interest in historical novels set in the ancient world, particularly in Late Antiquity, was at its height. Historical fictions written during the last decades of the eighteenth century to the end of the nineteenth often described controversial or little known characters from the last centuries of antiquity. Such was the case, for example, with Hypatia of Alexandria; she is little more than a name to us, but was enormously popular due to the story of her death at the hands of Christian monks popularized by Gibbon in Decline and Fall. This work was heavily biased against the Catholic Church, and for this story Gibbons relies on late authors such as Socrates Scholasticus. The English Anglican priest Charles Kingsley (1819–1875), author of the well-known historical novel Westward Ho! (1855) made her famous in his fiction work Hypatia, published in 1853, exactly at the time Simonides was in England. A possible influence of this newly-developed interest in the final years of conflict between paganism and Christianity should not be completely overlooked in the creation of these apocryphal works on Uranius and Nonnus. Aside from the fictitious biography of Nonnus made up by Simonides, there were, from this time onwards, several attempts to recreate, in literary form, the life of such a curious character, who exemplified as few others could the symbiosis between paganism and Christianity in Late Antiquity and who could well compete with Hypatia as a protagonist of a novel about that time. The first author who treated Nonnus from a literary perspective was Richard Garnett (1835–1906), librarian of the British Library. Garnett was an extremely cultivated man and a voracious reader who worked from 1851 until his retirement in 1899 in the huge room that constituted the British Library, at the time still part of the British Museum. Most likely Garnett included Nonnus among his preferred readings. We do not know if he had access to Simonides’ forged biography that so fascinated Marcellus, but he surely knew the Marcellus’ translations. Chapter XVII “The Poet from Panopolis” in Garnett’s first literary work, the



the poet and the forger

69

book The Twilight of the Gods (1888), is devoted to the controversial character of Nonnus. This recreation of Nonnus’ life is quite amusing: Nonnus is presented in his native Thebaid competing for the bishopric of Panopolis with an angry and unwashed anchorite and having to decide between Dionysus and Saint John. Needless to say, this is a conversion narrative, and, once again, the Dionysiaca are made typical of his youthful paganism while the more serious Paraphrase come from his later role as a respectable member of the Christian clergy. It is, ultimately, an ingenious story with Nonnus about to be appointed bishop (anticipating, perhaps, Enrico Livrea’s modern theory on the identification of Nonnus with a namesake bishop of Edessa),17 and hesitating between burning his pagan works and accepting his ecclesiastical dignity, in a sort of ordeal. Garnett presents an interesting fictional reenactment to deal with the difficult relation between the Dionysiaca and the Paraphrase, works by the same author: in the end, according to this fiction, Nonnus will be fated to publish both works, sowing doubts in the minds of scholars of centuries to come. The twentieth century produced another literary work based on the fictitious biography of Nonnus of Panopolis. This is a work of fiction by the German Historian Margarete Riemschneider, born in Königsberg in 1899. Riemschneider was a prestigious scholar of antiquity in Eastern Germany, author of several monographs on Homeric questions, ancient religions, Orientalism, etc., and also of several reference articles on Nonnus of Panopolis, published in journals of the Eastern Bloc. Her other role as literary creator is attested in a series of historical novels that enjoyed a certain degree of popularity in the former German Democratic Republic.18 Her historical fictions deal with topics from the ancient East to the life of Erasmus of Rotterdam, and include titles as stimulating as Der Schwur des Espaini. Ein Roman aus dem alten Urartu (1966), Das Wunder von Jerusalem. Historischer Roman aus der Zeit Sanheribs (1967). In 1970 she published the novel Im Garten Claudias, dedicated to the life of Nonnus of Panopolis, which bears as subtitle Roman über den letzten Großen Dichter der Antike. In line with Simonides and Garnett, Riemschneider also makes Nonnus a literary character, a character in a historical fiction, reflecting the fascination arising in those who enter into contact with the mysteries surrounding the life of the Panopolitan. The eventful

17 Livrea 1987 and 2003. 18 Schregel 1991.

70

david hernández de la fuente

life of Ammonius-Nonnus, and his ultimate conversion from paganism to Christianity, is narrated in this novel set in the last years of the Western Empire, in which a prominent role is played by the imperial envoy Stilicho, the famous Stilicho of Vandal father and Roman mother, as well as by the terrible Barbaric invasions of the fourth and fifth centuries. We see how both the forger and the novelists coincide in turning the conversion of the poet Nonnus into a key element of his biography in order to explain the apparently contradictory slant of his works. Nowadays we know that the Paraphrase was prior to the Dionysiaca and it is usually considered the result of an eclectic period of his life.19 The false biography of Nonnus raised Marcellus’ hopes for finding some truth about the last great poet of antiquity; but fiction seems, still today, to be the only means to investigate the life of this author. References Braun, F. (1915), Hymnen bei Nonnos von Panopolis, Universität Königsberg. Canfora, L. (2010), Il viaggio di Artemidoro. Vita e avventure di un grande esploratore dell’antichità, Milano. —— (2011), La meravigliosa storia del falso Artemidoro, Palermo. Canfora, L. and L. Bossina (2008) (eds.), Wie kann das ein Artemidor-Papyrus sein? Ma come fa a essere un papiro di Artemidoro?, Bari. Elliott, J.K. (1982), Codex Sinaiticus and the Simonides Affair, Tessaloniki. Farrer, J.A. (1907), Literary Forgeries, London. Fritzsche, C.L. (1856), Enthüllungen über den Simonides-dindorfschen Uranios, Leipzig. Garnett, R. (1940), The Twilight of the Gods, (1st ed. 1888, 2nd ed. 1903), London. Hernández de la Fuente, D. (2008), Bakkhos Anax. Un estudio sobre Nono de Panópolis, Madrid. Kingsley, C. (1853), Hypatia: New Foes with an Old Face, Oxford. Livrea, E. (1987), “Il Poeta ed il Vescovo,” Prometheus, 13: 97–123. —— (1989), Nonno di Panapoli, Parafrasi del Vangelo di S. Giovanni, Canto XVIII, Napoli. —— (2003), “The Nonnus Question Revisited,” in D. Accorinti and P. Chuvin (eds.), Des Géants à Dionysos. Mélanges de mythologie et de poésie grecques offerts à Francis Vian, Alessandria: 447–455. Masson, O. (1994), “Le faussaire grec C. Simonides à Paris en 1854, avec deux lettres inconnues de Sainte-Beuve et un récit du comte de Marcellus,” Journal des savants 2: 367– 379. Parsons, P. (2008), “Forging Ahead: Has Simonides Struck Again?” Times Literary Supplement 22 Feb.: 14. Riemschneider, M. (1970), Im Garten Claudias. Kulturgeschichtlicher Roman über den letzten grossen Dichter der Antike, Leipzig.

19 On the former, cf. Livrea 1989, introd., 20 ff. on the latter, again Vian 1976 and Hernán­dez de la Fuente, 2008. For a recent overall survey of Nonnus, cf. Shorrock 2011.



the poet and the forger

71

Sandys, J.E. (1910), The Eighteenth Century in Germany and the Nineteenth Century in Europe and the United States of America, History of Classical Scholarship vol. 3, Cambridge. Schaper, R. (2011), Die Odyssee des Fälschers. Die abenteuerliche Geschichte des Konstantin Simonides, der Europa zum Narren hielt und nebenbei die Antike erfand, München. Schregel, F.H. (1991), Die Romanliteratur der DDR. Erzähltechniken, Leserlenkung, Kulturpolitik, Opladen. Shorrock, R. (2011), The Myth of Paganism: Nonnus, Dionysus and the World of Late Antiquity, London. Stewart, C. (1859), A Biographical Memoir of Constantine Simonides, London. Vian, F. (1976), Nonnos de Panopolis, Les Dionysiaques, vol. I: Chants I–II, Paris.

“Genuine” and “Bastard” Dialogues in the Platonic Corpus: An Inquiry into the Origins and Meaning of a Concept* Mark Joyal The Platonic Corpus is not the largest extant collection of a single ancient author’s works, but it is highly unusual insofar as it is a substantial (ca. 2,500 pages of OCT text), complete set of one writer’s production. The special circumstances behind the compilation and survival of Plato’s opera omnia have a role to play in a discussion about the nature of that collection’s contents,1 but the details of the story – many of which are disputed, of course – are only incidental to the main focus of this paper. It is enough to say at the outset that the Corpus contains works whose authenticity was either suspected or rejected by ancient scholars and commentators themselves and that the preponderance of informed modern opinion is that the number of its non-Platonic works exceeds even that which ancient writers were prepared to identify. My purpose is to shed some light on the contents and makeup of the Corpus by drawing upon evidence which has not received the treatment that it deserves. In order to keep the discussion to a manageable scope, I want to concentrate on two themes: first and foremost, the language and images which ancient commentators used in dealing with the topic of Platonic imitations and the distinction between genuine and spurious Platonica; and secondly, the nature of those works which modern scholarship has identified as certainly or probably imitative. I Some time in the late third century or early second century bc, Aristophanes of Byzantium (ca. 257–ca. 180) organized fifteen Platonic works into five trilogies. Our source for the event, as in so much that has to do

* An early version of this paper was presented in Winnipeg in December, 2007, at the conference “Complicating Value: The Uses of Imitation in ancient Greece and Rome.” I am grateful to the organizers Mark Lawall and Peter van Alfen whose invitation provided the inspiration to formulate thoughts that had been long gestating. 1  See, e.g., Grote 1865: 132–165; Heidel 1896: 10–12; Alline 1915: 1–64; Jachmann 1941; Wilamowitz-Moellendorf 1962: 324–331; Carlini 1972: 3–76.

74

mark joyal

with the early transmission of the Platonic writings, is Diogenes Laertius (3.61–2 [Vita Platonis] = Ar.Byz. fr. 403 Slater): ἔνιοι δέ, ὧν ἐστι καὶ Ἀριστοφάνης ὁ γραμματικός, εἰς τριλογίας ἕλκουσι τοὺς διαλόγους. Some – one of whom is Aristophanes the grammarian – pull the dialogues into trilogies.2

As these words demonstrate, Aristophanes was not alone in this organizational activity; an anonymous “some” (ἔνιοι) also arranged the dialogues into trilogies, whether before or after Aristophanes we do not know.3 Diogenes, following his source, lists the contents of Aristophanes’ five trilogies: 1. Republic, Timaeus, Critias 2. Sophist, Statesman, Cratylus 3. Laws, Minos, Epinomis 4. Theaetetus, Euthyphro, Apology 5. Crito, Phaedo, Letters; and he adds: “the others [i.e., the dialogues accepted as completing the Platonic Corpus?] are added individually and in no particular order” (τὰ δ᾿ ἄλλα καθ᾿ ἓν καὶ ἀτάκτως). Hence we do not know precisely what compositions, in Aristophanes’ eyes, constituted “the Platonic Corpus.” The inclusion in his trilogies of the Minos, the Epinomis, and the Letters is, however, an important datum for us, since the first of these is today widely (but not universally) condemned as inauthentic,4 the authorship of the second was frequently assigned to a pupil of Plato even in antiquity,5 and most of the Letters (often all) are today held to be un-Platonic.6 That story is instructive since it provides evidence that works which are today considered to be of dubious authenticity were admitted into the Corpus at a relatively early date. The grounds for their admission seem to have been as much literary and dramatic as philosophical (n.b. especially trilogies 4 and 5 above), and the overarching organizational principle appears to have been the events in Socrates’ life as they could be made sense of from Plato’s works.7 We do not, however, have to dwell very long 2 All translations are my own unless otherwise noted. 3 The possibility of pre-Aristophanic arrangements in general is discussed in Tarrant 1993: 14–17, with reference to earlier scholarship. 4 Modern judgements on the Minos are conveniently assembled and the case for inauthenticity presented persuasively in Manuwald 2005; see also Dalfen 2009: 27–67. 5 I.e., Philip of Opus; for the evidence, see Tarán 1975: 7–13; Brisson 2005: 21–23. 6 Thrasyllus (see below) included thirteen Letters in his arrangement of the Platonic Corpus (D.L. 3.61), and these are presumably the thirteen that have been transmitted to us in the Platonic ms. tradition, but we do not know which of them Aristophanes himself included.  7 This continued to be a principle of organization much later, among Neoplatonic commentators; cf. the anonymous Prolegomena to Platonic Philosophy 24.20–25.36 ­Westerink-Trouillard-Segonds.



“genuine” and “bastard” dialogues in the platonic corpus

75

on the activities of Aristophanes since a much more influential ordering of the Corpus occurred probably, but not certainly, at a later time.8 I am referring, of course, to the organization of the Platonic dialogues into nine tetralogies – the “Thrasyllan” organization, so called after the activity of Tiberius’ court astrologer Thrasyllus (ob. ad 36), who is credited with carrying it out. Once made, this organization overshadowed all others and came to be reflected in most of the early substantial medieval witnesses to the Platonic text: Parisinus graecus 1807 (A, ca. 875), MS E.D. Clarke 39 (B, 895), Venetus App. Cl. 4.1 (T, ca. 950), Vaticanus graecus 1 (O, ca. 900), Vindobonensis suppl. gr. 39 (F, late thirteenth/early fourteenth century), and (for the first thirteen dialogues that appear in it) Vindobonensis suppl. gr. 7 (W, second half of the eleventh century). It is the principle of presentation followed in John Burnet’s five-volume Oxford Classical Text (1900–1907), the most widely consulted Greek editions. And it is the sequence adopted in the Hackett Complete Plato, now surely the most popular English translation of its kind.9 The tetralogical ordering has been the subject of much discussion and debate for many years, especially in regard to its date, its originator, and its rationale,10 but these controversial issues can be set aside as not really central to the main purpose of this paper. What is of basic importance is the canonical status of this arrangement, which fixed the contents of the Platonic Corpus and hence shaped widespread belief about the authenticity of individual Platonic works, with only rare expressions of skepticism, until the first years of the nineteenth century.11 The introduction of the tetralogical organization and its tenacity to the present day are crucial historical facts for our discussion, since the tetralogies contain numerous works whose inauthenticity ranges from possible to probable to certain. These works were accepted into the Platonic ­Corpus not least because they could pass for Platonic, for reasons which we will consider a little later. To all these tetralogical works Diogenes Laertius applies the epithet γνήσιοι (3.57), which for now I translate as “genuine” (see section II below). Yet not all of the so-called “genuine dialogues”  8 See Wilamowitz-Moellendorf 1962: 324–325; Pfeiffer 1968: 196–197; Philip 1970: 299– 301; Tarrant 1993: 14–15; Mansfeld 1994: 62–63.  9 Cooper and Hutchinson 1997. 10 E.g., Alline 1915: 112–124; Tarrant 1993: 1–30; Mansfeld 1994; Göransson 1995. The earliest evidence for the Thrasyllan tetralogical organization is now the recently published (2009) P.Oxy. 4941 (saec. ii). 11  Questions about the authenticity of canonical Platonic works began to be asked in earnest in correspondence between Friedrich Schleiermacher, Friedrich Schlegel, and Ludwig Heindorf in 1800; see Döring 2004: 76–78; also Joyal 2000: 122 n. 6.

76

mark joyal

(γνήσιοι διάλογοι) escaped the critical eye of some ancient scholars. We have already seen that this was so for the Epinomis. Thrasyllus himself apparently suspected the Rival Lovers (Anterastai) of being un-Platonic (D.L. 9.37); it seems certain that the Hipparchus too was questioned (Ael. VH 8.2)12; and some believed that Xenophon was the author of the Alcibiades II (Athen. 11.506c).13 What is more, the Stoic Panaetius (ca. 185–109 bc) is said to have considered the Phaedo to be νόθος, a word which for now I translate as “inauthentic” but which in other, non-literary contexts is usually given the meaning “bastard” (fr. 127–129 van Straaten – the significance and interpretation of Panaetius’ judgement are examined in section II below). The criteria which the ancient doubters applied in forming their opinions – whether external evidence (i.e., direct attestation for alternative authorship) or internal (i.e., an observed stylistic, literary, or historical feature of the work) – are usually uncertain. But the raw material for our study does not end with these suppositious tetralogical children. For it is probably some time after the tetralogical arrangement was introduced that another set of works, presented in no special or fixed order, came to be attached to the now-standard catalogue of Platonic writings. This set is sometimes referred to as the Appendix Platonica. The works in the Appendix (apart from the Definitiones) are all in one way or another “Socratic”; that is, Socrates (though unnamed in Demodocus and Sisyphus) is a principal interlocutor in all of them. It might be assumed, however, that because these works failed to be included in the tetralogical arrangement – or in any other organization, for that matter – they are more clearly imitative, and therefore un-Platonic, than the spurious and dubious works within the tetralogies. Such an assumption, in my opinion, would not be correct.14 Both sets are imitative of those works that scholarly consensus holds to be genuinely Platonic (see below), and the spuria that comprise the Appendix Platonica 12 The language of our evidence in each of these two cases is similar: εἴπερ οἱ Ἀντερασταὶ Πλάτωνός εἰσι and εἰ δὴ ὁ Ἵππαρχος Πλάτωνός ἐστι τῷ ὄντι. Neither εἴπερ nor εἰ δή must be taken as implying doubt or skepticism; “since” is a possible translation in both cases. The phraseology does at least suggest, however, that the authors are responding to doubts about authorship. For recent discussions, see Mansfeld 1994: 100; Tarrant 1993: 24–25, 70–71, 228–229 and n. 7, 1995: 150–151; Göransson 1995: 88 n. 2; Carlini 2005: 27. 13 ὁ γὰρ δεύτερος [i.e., of the two Platonic Alcibiades dialogues] ὑπό τινων Ξενοφῶντος εἶναι λέγεται. 14 The Eryxias, Axiochus, Sisyphus and Demodocus compare favourably on philosophical, literary and linguistic levels with some, if not all, of the spurious and doubtful works within the tetralogies. De Virtute and de Iusto are out of place in this whole group, but that has as much to do with generic differences as qualitative ones.



“genuine” and “bastard” dialogues in the platonic corpus

77

possess their own intrinsic interest for the attentive reader; they are not manifestly “inferior” to all tetralogical works. Yet there can be no doubt that it is because of their avowedly non-Platonic status that they have suffered greater neglect than they deserve.15 This is a circumstance which can, of course, be paralleled easily from other areas of artistic expression in the ancient Greek and Roman worlds. The works of the Appendix Platonica are, however, a special case, since ancient critics acknowledged that they, unlike the tetralogical dubia and spuria, were not “genuine articles.” Our main source, once again, is Diogenes Laertius (3.62 – extant works are in bold lettering): νοθεύονται δὲ τῶν διαλόγων ὁμολογουμένως Μίδων ἢ Ἱπποτρόφος, Ἐρυξίας ἢ Ἐρασίστρατος, Ἀλκυών, Ἀκέφαλοι, Σίσυφος, Ἀξίοχος, Φαίακες, Δημόδοκος, Χελιδών, Ἑβδόμη, Ἐπιμενίδης· ὧν ἡ Ἀλκυὼν Λέοντός τινος εἶναι δοκεῖ, καθά φησι Φαβωρῖνος ἐν τῷ πέμπτῳ τῶν Ἀπομνημονευμάτων. The dialogues that are considered inauthentic by general agreement are Midon or Horsebreeder, Eryxias or Erasistratus, Alcyon, headless dialogues,16 Sisyphus, Axiochus, Phaiakes, Demodocus, Chelidon, Hebdome, Epimenides. The Alcyon is thought to be the work of a certain Leon, according to Favorinus in the fifth book of his Memorabilia.

But Diogenes is not our only source. Perhaps 350 years later, the anonymous Neoplatonic author of the Prolegomena to Platonic Philosophy discussed the contents and organization of the Platonic Corpus (26.3–6 Westerink-Trouillard-Segonds): πάντες τοίνυν κοινῶς ὁμολογοῦσι νόθους εἶναι τὸν Σίσυφον καὶ τὸν Δημόδοκον καὶ τὴν Ἀλκυόνα καὶ τὸν Ἐρυξίαν καὶ τοὺς Ὅρους, οὓς εἰς Σπεύσιππον ἀναφέρουσιν. There is universal agreement17 that Sisyphus, Demodocus, Alcyon and Eryxias are inauthentic, as are the Definitions, which are attributed to Speusippus. 15 Lesky’s comments (1966: 511 = 1971: 575) are representative of a large body of opinion: “These writings [sc. the Appendix Platonica] are productions of little importance in the tradition of the Platonic school.” On the contrary, the argument may be made that these works are more important, not less, for the possibility that they are products of the Academy that do not originate from the hand of Plato. 16 Probably signifying “without title,” and referring therefore to On Virtue and On Justice, designations which reflect the standard form of philosophical subtitles, i.e., περὶ ἀρετῆς and περὶ δικαίου (cf., e.g., Μένων ἢ περὶ ἀρετῆς and Πολιτεία ἢ περὶ δικαίου); see Müller 1975: 38–40; Oswiecinski 1979; Carlini 2005: 29–30. 17 In spite of πάντες . . . κοινῶς ὁμολογοῦσι here and ὁμολογουμένως in the passage from Diogenes above, a few ancient writers did attribute members of the Appendix Platonica to Plato himself: so Clement of Alexandria and Eusebius on the Axiochus (see Tarán 1975: 7 and n. 23); note also the colophon in P.Oxy. 3683 (saec. ii) Πλάτων[ος] Ἀλκυών. The language of Diogenes and the anonymous author invites the hypothesis that some works

78

mark joyal

The verbal resemblances between these passages are suggestive, especially the application of the terms νόθοι and νοθεύω to spurious works. But are the resemblances significant? One thing we can say with some confidence is that they do not indicate the anonymous author’s reliance on Diogenes. Apart from a difference in their lists of inauthentic works – a difference which is not decisive in any case – there is no evidence that late-antique authors, Neoplatonists in particular, ever drew upon Diogenes. Either he and the anonymous author are relying upon a common source, or they are using language in reference to the Platonic Corpus which is current and familiar to both of them. Yet these two possibilities are not mutually exclusive. My suspicion is that our two authors are drawing on a source which called the works of the Appendix Platonica νόθοι, but that this source was not being at all innovative in doing so. My reason for believing this is not so much the fact that Aristotelian and Neoplatonic commentators contemporary with and a little earlier than the anonymous author habitually use νόθος for what they consider to be inauthentic works, both Platonic and otherwise. It is rather that there is evidence for its wider use in discussions about inauthenticity well before the time of Diogenes (third century ad). I begin with a passage from Quintilian’s book on rhetorical education which, so far as I can tell, has gone unnoticed in discussions on this subject (1.4.3). His work dates to the ad 90s: quo [sc. iudicio] quidem ita severe sunt usi veteres grammatici ut non versus modo censoria quadam virgula notare et libros qui falso viderentur inscripti tamquam subditos summovere familia permiserint sibi, sed auctores alios in ordinem redegerint alios omnino exemerint numero. The grammarians of old were so strict in their judgements that they not only allowed themselves to mark lines of verse with a sign of disapproval, and disinherit, as it were, as bastards any books which seemed to be wrongly attributed, but also drew up a prescribed list of authors, to which some were admitted while others were expelled altogether (trans. Russell, with ­modifications).

There is no hint here (or in the surrounding context) that Quintilian has Plato’s dialogues specifically in mind. On the contrary, his reference to “a prescribed list of authors” (auctores alios . . . in ordinem redegerint) – a

other than those which they cite were considered inauthentic by some critics but not by all; those dialogues mentioned on p. 76 above are examples of such writings. See further Carlini 2005: 26–28.



“genuine” and “bastard” dialogues in the platonic corpus

79

canon – suggests that the literary metaphor of the rejection (summovere: OLD s.v. submoveo 5b, c) of bastard children (subditos: OLD s.v. subdo 8a) enjoyed wider application than simply to the Platonic Corpus.18 The “grammarians of old” can be none other than the Alexandrians who toiled in that city’s library in the third and second centuries bc. The reference in this passage to the text-critical use of the “sign of disapproval” (censoria virgula) – i.e., the obelos – confirms this inference.19 The question of real interest is whether the language that Quintilian uses in describing works of rejected authorship – more precisely, the simile (tamquam; cf. also quadam: OLD s.v. quidam 3b) which he adopts – is his own innovation or reflects a manner of speaking that can be traced to the Alexandrians themselves. Or to put it another way: did the description of spurious works as “illegitimate,” that is to say, as νόθοι, begin with the Alexandrians? The main items of evidence for this question are the following. 1. Certain ancient declarations about the authenticity of Aeschylean, Sophoclean, and Euripidean plays which employ the terms νόθος and νοθεύω may trace their origins back to Callimachus’ catalogue of Greek literature, the Pinakes, or to the supplementary work on the Pinakes by Aristophanes of Byzantium.20 a. Catalogue of the titles of Aeschylus’ plays appended to the Vita Aeschyli (T 78 Radt): 1d. Αἰτναῖαι γνήσιοι / 2a. Αἰτναῖαι νόθοι. b. Vita Sophoclis T 1.76–7 Radt (= Ar.Byz. fr. 385 Slater): “According to Aristophanes, Sophocles composed 130 plays; of these, 17 [or 7] have been designated as νόθα” (Ἔχει δὲ δράματα, ὥς φησιν Ἀριστοφάνης, ρλ΄, τούτων δὲ νενόθευται ιζ΄ [ζ΄ coni. Bergk]). c. Hypothesis b to [Euripides], Rhesus, transmitted in three medieval primary witnesses: “Some have suspected this play [sc. Rhesus] to be νόθον and not Euripides’ on the grounds that it instead shows signs 18  Plutarch, a contemporary of Quintilian, uses the same metaphor in similar language, but applies it to (non-Platonic) doctrine (λόγος) rather than to literary compositions (Mor. 1070b–c): γέγονε δ᾿ ἕτερος λόγος ὑφ᾿ οὗ μᾶλλον ἡ συνήθεια παρανενόμηται, τὰς μὲν γνησίας ὑφαιροῦντος αὐτοῦ καὶ ἀποσπῶντος ἐννοίας ὥσπερ τέκνα νόθας δὲ ὑποβάλλοντος [Kronenberg, προσβάλλοντος BE; cf. in n. 25 below] ἑτέρας θηριώδεις καὶ ἀλλοκότους καὶ ταύτας ἀντ᾿ ἐκείνων ἐκτιθηνεῖσθαι καὶ στέργειν ἀναγκάζοντος. 19  For the obelos, see Pfeiffer 1968: 15, 178; Reynolds and Wilson 1991: 10–11; and on this passage in general, see Colson 1924: 37–38, where, however, the metaphor is not identified or discussed. I. Peirano, The Rhetoric of the Roman Fake: Latin Pseudepigrapha in Context (Cambridge 2012), appeared too late for me to take it into account (see esp. 37–42). 20 See Pfeiffer 1968: 128, with addendum at 287.

80

mark joyal of Sophocles’ imprint. In the lists of plays, however, it is recorded as γνήσιον” (τοῦτο τὸ δρᾶμα ἔνιοι νόθον ὑπενόησαν, Εὐριπίδου δὲ μὴ εἶναι· τὸν γὰρ Σοφόκλειον μᾶλλον ὑποφαίνειν χαρακτῆρα. ἐν μέντοι ταῖς διδασκαλίαις ὡς γνήσιον ἀναγέγραπται).21

These passages, if they really do stem essentially and in detail from Callimachus or Aristophanes, would settle the question posed above, but it would be philologically naïve simply to accept surface impressions. What we need to know is the extent to which these extracts reproduce their source (Alexandrian or otherwise);22 only then can we be certain that they demonstrate that the special metaphorical use of νόθος and νοθεύω has its origins in the Alexandrian Library. It is possible – and many parallels could be cited – that each of these passages, which postdate Callimachus by many years, reflect the application of current jargon to the raw Callimachean material that their authors were using.23 Moreover, to believe that these remarks about authenticity go back to Callimachus requires that we suppose also that his Pinakes were a “critical” catalogue and not merely a series of lists and biographies. That is not, however, a serious difficulty to surmount, since Aristophanes is surely a promising candidate for the kind of critical activity that we would need to assume here.24 2. Words in the νοθ- root appear in some scholia in which the activity of the Alexandrian critics is mentioned: Aristonicus apud Σ Hom. Od. 11.547 (bis), 13.320–3, 16.281–98 (cf. Σ Od. 16.281), Σ Hes. Th. 114–15. But where a critic is cited by name as grammatical subject in these passages, it is the colourless ἀθετεῖ – the Alexandrian verb par excellence in contexts

21  For discussion of this hypothesis, see Ritchie 1964: 8–29, esp. 10–11; Liapis 2012: 59–67, esp. 59–61. The Vita Euripidis records that τὰ πάντα δ᾿ ἦν αὐτῷ δράματα Ϙβ΄, σῴζεται δὲ οη΄. τούτων νοθεύεται τρία· Τέννης, Ῥαδάμανθυς, Πειρίθοος (TrGF 5.47: 1.IA.9). For the authenticity and attribution of these three works, see Collard and Cropp 2008: 629–635. The Vita is certainly post-Alexandrian, but the denial of Euripidean authorship in this passage is doubtless Alexandrian in origin, if not in language. 22 The word-order in the extract from the Vita Sophoclis (1.b) leaves open the possibility that Aristophanes’ assessment applies only to the number of Sophoclean plays, not to the number of νόθα δράματα. 23 The hypothesis to the Rhesus (above) provides an example. We are told that in the (probably Aristotelian) chronological records of the dramatic festivals (ἐν ταῖς διδασκαλίαις), the Rhesus is recorded as γνήσιον. It is unlikely, however, that the διδασκαλίαι stated explicitly that the Rhesus was “genuine”; it probably recorded simply that Euripides was the author of a play by this name (so Ritchie 1964: 17–18), so that the characterization expressed by the term γνήσιον reflects an inference which the author or compiler of the hypothesis has drawn. 24 See Slater 1976.



“genuine” and “bastard” dialogues in the platonic corpus

81

that involve declarations of spuriousness – that is predicated of his activity, not the more evocative νοθεύει: ἀθετεῖ Ἀρίσταρχος / Ζηνόδοτος (bis) / Ἀριστοφάνης / ὁ Σέλευκος. The sole exception seems to be the rather late Herodian (second century ad) on Hom. Il. 15.71 (Vers. 358.12–13): τὸ γὰρ “Ἴλιον αἰπύ” νοθεύει Ἀρίσταρχος. Where a form of νοθεύει is used in the scholia, it appears in the passive voice without a personal agent. It may be argued that ἀθετεῖ is vox propria for marking passages (especially Homeric passages) as spurious interpolations, νοθεύει the verb for declaring the inauthenticity of whole works, but (beyond the passage from Herodian cited just now) the appearance of νοθεύονται as a gloss or variant for ἀθετοῦνται in Aristonicus’ athetesis of Odyssey 13.320–3, as well as a few other passages, throw that general principle into some doubt.25 The question, then, is not whether νόθος and νοθεύω were used only with reference to imitative works in the Platonic Corpus – it is clear that this usage was in fact widespread – but how early it is that words in the νοθ- root were applied as a label of literary inauthenticity. If this special application goes back to Alexandrian philological scholarship, then it belongs to the vocabulary that was applied in the earliest phase of scholarly research on matters of authorship, authenticity, and, by implication, imitation.26 II What does it mean to call an ancient work, Platonic or otherwise, a νόθος? A great deal has been written about this word, especially about its appearance in the context of Athenian law. In its earliest uses, in Homer’s Iliad (but not the Odyssey, and not in Hesiod), νόθος is used of an offspring 25 [νοθεύονται] ἀθετοῦνται δ΄ στίχοι. ὁ μὲν πρῶτος ὅτι ἀντὶ τοῦ ἐμῇσιν ἔχει τὸ ᾗσιν, ὅπερ ἐστὶ τρίτου προσώπου κτλ. Cf. Σ S. Ant. 45 Δίδυμος δέ φησιν ὑπὸ τῶν ὑπομνηματιστῶν τὸν ἑξῆς στίχον νενοθεῦσθαι; Σ S. Aj. 841a ταῦτα [sc. “τὼς αὐτοσφαγεῖς”] νενοθεῦσθαί φασιν, ὑποβληθέντα πρὸς σαφήνειαν τῶν λεγομένων (for ὑποβάλλω in this context, cf. n. 18 above, and LSJ s.v. II.1–2; also subditos p. 78 above); Σ Th. 1.3.3 (on Hom. Il. 2.530) τὸ γὰρ “ἐγχείῃ δ᾿ ἐκέκαστο Πανέλληνας καὶ Ἀχαιοὺς” νενόθευται κτλ. Quintilian, however, does seem to restrict application of the concept of νοθεία to whole works rather than to lines of verse (p. 78 above). 26 Schironi 2005 presents new evidence for Alexandrian work on the Platonic Corpus while also discussing the little that was previously known about this activity. Her interesting arguments are weakened, however, by a tendency to press the evidence farther than it can tolerate, in particular by assuming the existence of an Alexandrian “edition” of Plato’s works, as opposed simply to the Alexandrian arrangements of the corpus and the philological exegeses of some texts. Unlike the former, there is demonstrable evidence for the latter.

82

mark joyal

born out of wedlock, especially one whose mother is a slave or concubine (e.g., Il. 2.727). In the sixth century, Solon created a law – or so Athenians believed by the end of the fifth century bc if not much earlier (Ar. Av. 1660–63) – that denied to νόθοι the privileges carried by rights of kinship (ἀγχιστεία).27 In 451/0 Pericles introduced “the law concerning bastards” (ὁ περὶ τῶν νόθων νόμος, Plu. Per. 37.2–6), which required that Athenian citizens be born from two parents who were themselves citizens (ἄστοι; cf. [Arist.] Ath. 26.4). The regular antonym of νόθος is γνήσιος, literally “of/ belonging to the γένος.”28 The two adjectives are coupled constantly; apart from the passages in Diogenes Laertius and in the Aeschylean catalogue quoted above, e.g., Hom. Il. 11.102, S. fr. 87.1 Radt, E. Andr. 638, Hdt. 3.2.6, Ar. Av. 1650, Pl. R. 536a4, Men. fr. 248.2–3 Körte-Thierfelder. These passages and many others leave no doubt that in their primary meanings both νόθος and γνήσιος refer to blood relationships or their absence and, as a consequence, to the question of legitimacy. They are also – and I think more meaningfully – terms that designate inclusion and exclusion: in or from the family and its privileges (especially inheritance) and in or from the city and the rights afforded by it.29 If we look at the two terms in this way, it seems a simple matter to explain the application of νόθος and γνήσιος to the works transmitted in the Platonic manuscript tradition. Those that were included in the Thrasyllan canon belonged to the legitimate Platonic family and were γνήσιοι; those that fell outside the canon were of impure pedigree and were therefore νόθοι. It is, however, not so simple. For one thing, the definition and the rights of a νόθος in Athens were highly fluid through the fifth and fourth centuries, as they must have been elsewhere during this and other ­periods30; assessment of the term’s metaphorical use should take this fluidity into account.31 We must consider as well the possibility that, despite its 27 See Ruschenbusch 1966: 86–88. 28 See Ogden 1996: 17–18. 29 See Patterson 2005: 280–282. 30 See Rhodes 1993: 331–335, 496–497, with modern works cited there and in addenda to 333 (at 775) and 497 (at 777–778); Dunbar 1995: 732–734. 31  The fluid nature of νοθεία may be reflected also in the range of its metaphorical uses. For instance, νόθος was occasionally used for the main purpose of designating inauthenticity, e.g., Σ Ar. Pax 835 = Ion DK 36A2: Ion wrote in prose the work called Ambassador, which νόθον ἀξιοῦσιν εἶναί τινες καὶ οὐχὶ αὐτοῦ, “some judge to be νόθος and not Ion’s” (taking καί here as explanatory, though καὶ οὐχὶ αὐτοῦ may denote something additional to νόθον). Sometimes doctrine, not the work itself, is identified as inauthentic, e.g., Plu. Mor. 1070b–c, quoted above (see n. 18); Procl. Theol.Plat. 1.5.17–23 Saffrey-Westerink: let



“genuine” and “bastard” dialogues in the platonic corpus

83

i­ nclusion in the tetralogical organization, Thrasyllus himself doubted the authenticity of the Anterastai (see p. 76 above). That passage in Diogenes Laertius is difficult to assess, but harder still is the judgement about the Phaedo attributed to Panaetius, also mentioned above (p. 76): Elias in Arist. Cat. 133.18–23 Busse (= fr. 129 van Straaten): Συριανὸς μὲν γὰρ ὁ φιλόσοφος ἐπέγραψε τῷ Φαίδωνι νοθευομένῳ ὑπό τινος Παναιτίου· εἴ με Πλάτων οὐ γράψε, δύω ἐγένοντο Πλάτωνες· Σωκρατικῶν ὀάρων ἄνθεα πάντα φέρω. ἀλλὰ νόθον με τέλεσσε Παναίτιος· ὅς ῥ᾿ ἐτάλασσε32 καὶ ψυχὴν θνητὴν κἀμὲ νόθον τελέσαι. (= AP 9.358 = fr. 128 van Straaten). The philosopher Syrianus wrote an epigram on the Phaedo, [a dialogue] which was considered νόθος by a certain Panaetius: “If Plato did not write me, there were two Platos; for I carry all the flowers of the Socratic songs. But Panaetius designated me as νόθος – a man who had the nerve to designate both the soul as mortal and me as νόθος.” Asclepius in Arist. Metaph. 90.23–6 Hayduck (= fr. 127 van Straaten): Παναίτιος γάρ τις ἐτόλμησε νοθεῦσαι τὸν διάλογον. ἐπειδὴ γὰρ ἔλεγεν εἶναι θνητὴν τὴν ψυχήν, ἐβούλετο συγκατασπάσαι καὶ τὸν Πλάτωνα· ἐπεὶ οὖν ἐν τῷ Φαίδωνι σαφῶς ἀπαθανατίζει τὴν λογικὴν ψυχήν, τούτου χάριν ἐνόθευσε τὸν διάλογον. A certain Panaetius had the affrontery to declare the dialogue [sc. Phaedo] νόθος. Since he held that the soul is mortal, he wanted to drag Plato down with him. Plato in the Phaedo clearly presents the rational soul as immortal; that’s why Panaetius declared the dialogue to be νόθος.

us examine what characteristics in Plato’s doctrines about the gods (τὰ περὶ θεῶν δόγματα) we should focus on, so that τά τε γνήσια καὶ τὰ νόθα τῶν εἰς αὐτὸν ἀναφερομένων κρίνειν δυνησόμεθα, “of those that are attributed to him, we will be able to distinguish the genuine and the inauthentic” (sc. “dialogues” according to Saffrey and Westerink, but the adjectives agree with δόγματα). This application of the word to doctrine may help to explain those passages where a certain degree of νοθεία is implied, e.g., Porph. Plot. 16: Porphyry wrote treatises in which he demonstrated that a book by “Zoroaster” was ὅλως νόθον τε καὶ νέον, “entirely inauthentic and recent,” as well as πελπασμένον, “fabricated” (frequently used of forgeries: LSJ s.v. πλάσσω V); Procl. in Remp. 134.5–6 Kroll: according to Proclus, the Epinomis is νοθείας . . . μεστή, “full of inauthenticity”; Athen. 2.57c: in Hippocrates’ Περὶ πτισάνης (On Barley Gruel), ὃ ἐκ τοῦ ἡμίσους μὲν νοθεύεται, ὑπ᾿ ἐνίων δὲ καὶ ὅλον, “which is considered half inauthentic, though some consider it entirely inauthentic.” Rather different is Procl. in Prm. 4.872.32–36: this passage (i.e., Prm. 131d7–e2) ἔδοξέ τισιν οὕτω δυσδιάθετον εἶναι κατὰ τὴν λέξιν, ὡς καὶ ἐν τοῖς νόθοις αὐτοῖς καταλέξαι τινὰς καὶ περιγράψαι τῶν τοῦ Πλάτωνος ῥημάτων, “has seemed to some to be so difficult to construe that they count it even as inauthentic and mark it off from Plato’s words.” 32 ἐτέλεσσε is the reading of the direct ms. tradition of this epigram, but I suspect that ἐτάλασσε is the verb which Elias (and Asclepius) believed that Syrianus presented; see further n. 34 below.

84

mark joyal

Heinrich Dörrie proposed that AP 9.358 = fr. 128 van Straaten was intended as a reductio ad absurdum: the poet demonstrates the conclusion to which Panaetius’ belief in the mortality of the soul leads, namely, to the declaration that the Phaedo is νόθος.33 Dörrie’s interpretation is a plausible one, possibly compelling, and if he is right, Panaetius cannot be said to have declared the Phaedo to be un-Platonic. Under his proposal, however, the opening of the epigram is curious: εἴ με Πλάτων οὐ γράψε is a response to an original claim or suggestion that “Plato did not write the Phaedo,” not to the doctrinal position that “the soul is mortal”; Dörrie’s position would be more convincing if the Phaedo were actually mentioned in the poem. Furthermore, Dörrie strengthens his argument by asserting that the emphasis in the epigram is on Panaetius’ denial of the soul’s immortality, whereas his denial of the Phaedo’s authenticity is merely obiter dictum. The structure of the epigram, however, tells a different story: first, a reference to the dialogue’s authorship, followed by evidence which is presented in explanatory asyndeton; then the designation of the Phaedo as νόθος twice through the imposition of a-b-a structure; and reference to the soul’s mortality is restricted to the first half of line 4. Far from being “fast eine Kleiningkeit,” authenticity is the focus of the epigram; it is certainly not obvious that the epigram’s author intended the poem to be a reductio. Yet even if we accept the epigram for what it appears to be, we cannot be sure that Panaetius used the term νόθος to express his opinion about the Phaedo; once again we may be dealing with a case in which the reporters are applying the terminology that they knew best to the information transmitted to them (Syrianus belongs to the early fifth century ad, Asclepius and Elias both to the sixth). The designation Παναίτιος τις (“a certain Panaetius,” “someone named Panaetius,” possibly conveying a hint of contempt [LSJ s.v. τις A.II.6.a]) which Asclepius and Elias both use suggests strongly that their knowledge about Panaetius is not first-hand; no doubt they relied upon Syrianus himself for their information about him. The emphasis which this epigram focuses on the term νόθος, however, does suggest, at least, that this was a word that Panaetius himself applied to the Phaedo.34 33 Dörrie-Baltes 1990: 322. Dörrie has found a supporter in Barnes 1991: 116. 34 Matters are further complicated by the strong likelihood that Syrianus was not, in fact, the original author of this epigram (it is transmitted in the ms. tradition of the Palatine Anthology, unconnected with Syrianus), though it seems probable to me that he adapted it to suit his own purposes. In particular, ἐτάλασσε pro ἐτέλεσσε in l. 3 (see n. 32



“genuine” and “bastard” dialogues in the platonic corpus 85

However that may be, it is very unlikely that Panaetius believed the Phaedo not to have been written by Plato. If he called the Phaedo νόθος, it was not because he thought – too eccentric an opinion for us to credit – that Plato was not its author. Rather, Panaetius’ statement would have been an attempt to score a rhetorical point in regard to his own teachings about the ψυχή.35 In other words, if Panaetius used the term νόθος, it is unlikely to have connoted “spurious” or “pseudo-Platonic.” This inference is strengthened by another, rather more famous, statement from him (D.L. 2.64 = fr. 126 van Straaten): πάντων μέντοι τῶν Σωκρατικῶν διαλόγων Παναίτιος ἀληθεῖς εἶναι δοκεῖ τοὺς Πλάτωνος, Ξενοφῶντος, Ἀντισθένους, Αἰσχίνου· διστάζει δὲ περὶ τῶν Φαίδωνος καὶ Εὐκλείδου, τοὺς δὲ ἄλλους ἀναιρεῖ πάντας. Of all the Socratic dialogues, Panaetius thinks that those of Plato, Xenophon, Antisthenes, and Aeschines are true; he has doubts about those of Phaedo and Eucleides, but he rejects all the others.

The category which Panaetius establishes here – “true” or “genuine” or “authentic” (ἀληθεῖς) Socratic dialogues – clearly does not involve a judgement about authorship. It has to do, instead, with fidelity to an accepted standard (i.e., Panaetius’) of Socratic representation and imitation. As far as we can tell from the fragmentary remains of his work that survive, Panaetius was not interested in questions of authorship. What can we learn from the evidence presented in this section, both in general and about the Platonic dubia and spuria? The usual view that the terms γνήσιος and νόθος emphasize primarily or solely the authenticity and inauthenticity of an ancient work reflects a one-sided perspective which is called into question not only by the social, political, and legal applications of these terms but by other, philological evidence as well. In ancient Greek eyes the νόθος lacked the purity and privilege that a legitimate offspring, the γνήσιος, possessed, but for all that the νόθος did enjoy a blood-relationship with the progenitor. We need have no hesitation above) may well be his innovation, the kind of editorial liberty which was common among Middle Platonists and Neoplatonists (see Whittaker 1989). The probability that Asclepius is dependent on Syrianus for his information – already strong a priori – is then enhanced by the correspondence between Syrianus’ ἐτάλασσε . . . με νόθον τελέσαι and Asclepius’ ἐτόλμησε νοθεῦσαι τὸν διάλογον (note also Asclepius’ preservation of the a-b-a sequence of thought in his paraphrase of lines 3–4 of the epigram). Dörrie reasonably suggests (in Dörrie-Baltes 1990: 322–323) that the epigram’s date of composition is likely to be the first or second century ad. 35 See also Müller 1975: 20 n. 4 (begins at 18); Carlini 1972: 31–33. For Panaetius’ psychology, cf. Cic. Tusc. 1.32.79–33.80 = fr. 83 van Straaten, with fr. 84–5.

86

mark joyal

in pressing the metaphorical meanings and implications of these terms. Ancient critics, after all, had a variety of alternative nouns and verbs at their disposal which they could use to convey the meaning of imitation or even forgery,36 but those critics who discussed the Platonic works seem to have chosen a different path. III That the literary νόθος can be viewed as a reflection (however imperfect) of an original, authentic standard is supported by the nature of the unPlatonic works themselves. Many or most of these works establish their Platonic pedigree in a rather predictable way, in particular by casting their readers’ minds to the Platonic Socrates and to the actual words of certain Platonic dialogues, especially those works in which Socrates’ character is drawn most vividly. These are, of course, the early “Socratic” dialogues, but also other works in which Socrates plays a leading role, above all, the Apology of Socrates and the Phaedo, the two most biographical of all works in the Corpus, as well as the Theaetetus, which, too, is tied dramatically to the circumstances surrounding Socrates’ last days.37 This fact is familiar to anyone who has sampled the modern scholarly works that deal with the dubia and spuria.38 Careful readers of these dialogues know that 36 E.g., “authentic”: ἀληθής (cf. ψευδῶς later in this n.); “forge”: πλάσσω (see n. 31 above); “forgery”: πλάσμα (LSJ s.v. II); “imitate”: μιμέομαι, ἀπομάσσω (LSJ s.v. III), ἐκμάσσω (LSJ s.v. II), ἀποτυπόω (LSJ s.v. II.1–2); “a model”: παράδειγμα, ἀρχέτυπον (LSJ s.v. ἀρχέτυπος II). The range of possible phrases and circumlocutions to express inauthenticity is theoretically limitless; two familiar examples in Galen (in Hippocratis Nat. Hom. 5.91.1.57 and 55 CMG = 15.109, 105 Kühn): ἐν γὰρ τῷ κατὰ τοὺς Ἀτταλικούς τε καὶ Πτολεμαικοὺς βασιλέας χρόνῳ πρὸς ἀλλήλους ἀντιφιλοτιμουμένους περὶ κτήσεως βιβλίων ἡ περὶ τὰς ἐπιγραφάς τε καὶ διασκευὰς αὐτῶν ἤρξατο γίνεσθαι ῥᾳδιουργία κτλ., “When the Attalid and Ptolemaic kings were competing with each other for the acquisition of books, deception over their ascription and composition began to take place”; πρὶν γὰρ τοὺς ἐν Ἀλεξανδρείᾳ τε καὶ Περγάμῳ γενέσθαι βασιλεῖς ἐπὶ κτήσει παλαιῶν βιβλίων φιλοτιμηθέντας, οὐδέπω ψευδῶς ἐπέγγραπτο σύγγραμμα, “Before the kings in Alexandria and Pergamon had become keen rivals for the acquisition of old books, no composition had yet been falsely attributed.” 37 It is, without doubt, Socrates’ appearance as a central character in the Alcyon, rather than any doctrinal considerations, that encouraged its inclusion in the Appendix Platonica (most notably, in Parisinus gr. 1807, the oldest of all medieval Platonic mss.), despite the fact that this little piece is also transmitted in the body of Lucian’s works and was attributed in antiquity to “Leon the Academic” (Athen. 11.506c, D.L. 3.62); see Müller 1975: 17–20, 303–304, 317–319. 38 E.g., Heidel 1896; Shorey 1933; Taylor 1949; Müller 1975: passim; Guthrie 1978: 383– 398. The nineteenth-century editions and commentaries by Schleiermacher, Stallbaum, Ast and Fritzsche are veritable storehouses of information on Platonic parallels and



“genuine” and “bastard” dialogues in the platonic corpus 87

not all the alleged reminiscences and “borrowings” that are catalogued in these books and articles can really be proved to be just that, but a large number of them do withstand close examination. So, for instance, it is apparent that the Axiochus not only draws verbally upon the Apology and Phaedo but depends for its meaning upon the reader’s knowledge of these works.39 The Theages reproduces whole sentences from the Apology and the Theaetetus nearly verbatim and takes those writings as points of departure for its extraordinary presentation of Socrates’ famous “divine sign” (τὸ δαιμόνιον σημεῖον).40 The Clitophon paraphrases the Apology in order to characterize and criticise the shortcomings in Socrates’ pedagogical ­method.41 On Virtue is nearly a précis of the Meno, to such an extent that its text has been used by modern editors in establishing the text of the Meno itself.42 The Alcyon assumes its dramatic setting from Socrates’ extraordinary sojourn outside the walls of Athens in the Phaedrus.43 The Eryxias takes its point of departure from the Charmides, but its argument seems to have grown out of Socrates’s confrontation with the sophists Euthydemus and Dionysodorus in the Euthydemus.44 These and other works are innovative but also imitative, frequently telegraphing their relationship to Plato through the use of key phrases or sentences, dramatic setting, allusion to Socratic doctrine, and reference to Socratic traits that are familiar from Plato’s works. Above all, they are literary in their imitation; their authors, to all appearances, have no interest in the “unwritten (metaphysical) doctrines” that seem to have formed the basis of much of Plato’s teaching in the Academy and of study in the school thereafter.45 To some of them the terms “cento,” “pastiche” and “farrago” have been variously applied, sometimes, I think, with good reason. One term that is not appropriate is “forgery”: that word implies a motive – an attempt to deceive – which simply cannot be inferred from the textual

­possible borrowings to which all subsequent scholars who work on the dubia and spuria have been indebted. 39 Full details in Joyal 2005: 101–104, 109–110. 40 See Joyal 2000: 37 n. 63, 73–77, 82–89, 256–260 passim; Döring 2004: 44–47, 52–53, 57–66, 78–79. Bailly (2004: 265–279) is less confident about these dependences; I intend to deal (inter alia) with his main arguments in a forthcoming article. 41  See Slings 1999: 103–112 (with extensive reference also to Plato’s Euthydemus and Meno); Erler 2008. 42 See Bluck 1961: 147; Müller 1975: 197–220; Vancamp 2010: 100–102. 43 See Heidel 1896: 19; Müller 1975: 295–296, 317–318. 44 See Heidel 1896: 59; Schroll 1901: 10–12, 17–18, 36–38; Souilhé 1962: 84–85. 45 See, e.g., Gaiser 1963; Guthrie 1978: 418–442 for an overview of the unwritten ­doctrines.

88

mark joyal

or historical evidence available to us. But more than this: ancient use of the terms νόθος and γνήσιος suggests a literary status and an activity that are more constructive and creative than a forgery can ever be. Consistent with the connotations of this terminology, too, is the desultory way in which alternative authorship is only occasionally ascribed (as the Alcibiades II was to Xenophon and the Epinomis to Philip of Opus): the father of the νόθος may not be known, or may only be suspected. That ancient critics could apply the term νόθος with a more positive sense than standard English renderings usually convey may be suggested by the ease with which Platonic dubia and spuria were accommodated within the Corpus without, apparently, a great deal of concern over authorial identity.46 The works served their own purposes in advancing arguments and agendas – pedagogical or philosophical – within the activities of the Academy itself; that, rather than authenticity, is what mattered most to the earliest readership of these works.47 We may reasonably suppose that some dialogues were labelled γνήσιοι because they were included, not included because they were γνήσιοι (likewise for the exclusion of some νόθοι). So far as we can reconstruct the modus operandi of the Platonic imitators, then, the term νόθος, understood in its full sense, seems to be an especially appropriate one. Whether or not it is correct that the terms νόθος and γνήσιος belong to the earliest phase of philological discussion about literary imitation (see section IV below), we ought to consider the ­possibility

46 Here we should recall (p. 76 above) the report of Thrasyllus’ own doubt over the authenticity of the Anterastai, the fourth dialogue in the fourth Thrasyllan tetralogy. According to the author of the Prolegomena to Platonic Philosophy (25.36.1–4 WesterinkTrouillard-Segonds), the Epinomis was treated as γνήσιος simply to fill out the requisite number of tetralogies: φασὶν οὖν ἐννέα εἶναι τετραλογίας, ὡς λϛ΄ εἶναι τοὺς πάντας διαλόγους αὐτοῦ. οὗτοι δὲ διὰ τὸ συστῆναι τὸν τῶν τετραλογιῶν ἀριθμὸν τὸ Ἐπινόμιον, νοθευόμενον, γνήσιον ἀποφαίνουσιν, “They say that there are nine tetralogies, so that his [Plato’s] dialogues are 36 in all. In order for the number of the tetralogies to be completed, these people declare the Epinomis γνήσιον, even though it is considered νόθον.” 47 This point is borne out by the fact that some ancient Platonists recommended that students begin their study of Plato with dialogues that are today considered to be of doubtful authorship (at least), i.e., Alcibiades I (D.L. 3.62; cf. Procl. in Alc. 11.3–17 Westerink, Olymp. in Alc. 10.18–11.6 Westerink), Theages (D.L. 3.62, Alb. Eisag. 149.5 Nüsser; also Theon of Smyrna – see Tarrant 1993: 60–68) and Clitophon (D.L. 3.62). No evidence exists, however, to suggest that ancient readers thought any of these three dialogues had been written by anyone other than Plato (for its part, Theages was being copied and read in Egypt at least as early as the first half of the second century B.C.; see Joyal 2002: 43). Lea Stirling’s observations (forthcoming) on the secondary importance of artistic originals and of questions of authenticity in late-antique sculpture provide further support to the inference that the ancient audience placed a lower premium on these matters than may be customary for the modern viewer (or reader); see also Ridgway 1984, esp. 81–95.



“genuine” and “bastard” dialogues in the platonic corpus 89

that this usage traces its origins to discussions about the contents of the Platonic Corpus itself. This is, I believe, an attractive proposal because it is coherent with Plato’s own celebrated descriptions of the ­creative process in philosophical thought in at least three certainly genuine works. One of them is the Symposium. Here Diotima informs Socrates that the purpose of love is “physical and mental procreation in a beautiful medium” (τόκος ἐν καλῷ καὶ κατὰ τὸ σῶμα καὶ κατὰ τὴν ψυχήν, 206b7–8); she adds that “all people are pregnant both physically and mentally” (κυοῦσιν . . . πάντες ἄνθρωποι καὶ κατὰ τὸ σῶμα καὶ κατὰ τὴν ψυχήν, 206c1–3; in general 206b7–212a7). Another of these works is the Theaetetus. At an early point in this dialogue Socrates famously introduces the metaphor of intellectual midwifery (148e1–151d6): he is himself barren but has the uncanny ability to assist those of his associates who are “pregnant” to give birth to their philosophical offspring. Plato develops this metaphor in an elaborate and extended way and then returns to it at strategic points in the remainder of the dialogue (157c7–d3, 160e2–161a4, 184a9–b1, 210b4– d2). In the Phaedrus, Socrates explains that a man’s discourses (λόγοι) on morality are like his genuine (γνήσιοι) sons (277e5–278a6; cf. 276a1–3).48 A fourth text, Republic 603a9–b5, though it focuses on artistic rather than philosophical creativity, is nevertheless relevant in this context. Socrates and Glaucon agree that painting, and mimetic art in general, produce work that is far from the truth. What is more, this art associates with (or has intercourse with, προσομιλεῖ) the part in us which is far from intelligence; it is a companion and friend (ἑταίρα καὶ φίλη) which has no healthy or true (ἀληθεῖ) purpose. The mimetic art, an inferior mother having intercourse with (συγγιγνομένη) an inferior father, gives birth to inferior offspring. Although this proposal to find the seeds of the metaphor of “genuine” and “bastard” compositions within the writings of Plato himself must remain in the realm of informed speculation, I find support in Harold Tarrant’s attempt to trace the rationale behind the contents and order of the Thrasyllan tetralogies to the educational program which Plato presents in Republic 7.49 That hypothesis is an interpretive and methodological parallel for the theory which I am putting forward here: in both cases, Platonic doctrine as expressed in Plato’s works is held to have influenced ancient scholarly treatment and discussion of the Platonic Corpus itself.

48 The most illuminating discussion of the relevant material in Tht., Smp. and Phdr. remains Burnyeat 1977; see also Halperin 1990. 49 Tarrant 1993: 98–103.

90

mark joyal IV

For good reason have the activities of Alexandrian scholars figured heavily in the foregoing discussion, since it is surely in the context of the Alexandrian Library that the metaphorical application of literary bastardy and legitimacy first arose. Much of the evidence presented above – in particular Quintilian, didascalic records, scholia – has pointed towards this conclusion. But even apart from this evidence, the metaphors presuppose the concept of a “family” of literary works in relation to which inclusion or exclusion can be practised, and it is the Alexandrian scholars who first brought together and organized the full corpora of important authors and had to make decisions about the “membership” of individual works. The metaphors are incompatible with a context in which individual works are being assessed in isolation or against only a limited set of comparanda. Moreover, if we think of the literary “family” in physical terms, of the many papyrus rolls that comprised an author’s collected works (say, Sophocles’ or Plato’s), a νόθος could be most readily “slipped in” or “snuck out” (ὑποβάλλω / subdo, ὑφαιρέω / submoveo; see p. 79 and n. 18, and n. 25) if the roll in which it was contained was merely one among many others – a condition which suggests a large library. The possibility that Alexandrian scholars thought of papyrus rolls in this metaphorical way is not difficult to believe, since the literary personification of the roll was initiated and developed elaborately by the Alexandrians themselves.50 The attribution of the metaphor’s origin to the Alexandrians makes sense at another level. These scholars did not inherit a tradition which was preoccupied with matters of authenticity; nor would authenticity become their primary concern. Collection, organization and preservation were of the first importance, and the subordinate questions of authenticity and inauthenticity had to be incorporated into these central activities. The relationships designated by the terms νόθος and γνήσιος were ­especially 50 See Bing 2008: 10–48, esp. 18–21, 29–35. For the adoption in Roman poetry of the metaphor of the personified book, cf. Hor. Ep. 1.20, Ov. Tr. 1.1, Mart. 1.3; see Mordine 2010, esp. 540–543 for the bookroll as the author’s “child.” For a striking medieval example, cf. Richard de Bury (1287–1345), Philobiblon IV.68–71: in an extended metaphor, the book complains about the false attribution of its parentage (i.e., of its authorship) and poor reproduction (i.e., transcription) of itself. The conceit lives on: in 2013 I read the following in a contract issued by a major academic publisher (under “Moral Rights”): “The Author hereby asserts to the Publisher and its licensees the Author’s right of paternity in the Work . . .” For philologists it survives most prominently today in the familial and genetic language of Lachmannian stemmatic theory and its successors (on which see Timpanaro 2005).



“genuine” and “bastard” dialogues in the platonic corpus

91

suitable for this purpose: not only did the words themselves evoke the relatively straightforward matter of genetic source, but they did so at the same time that they reminded the reader of a given work’s traditional association with a much larger collection (its “family”). It is hard to see how this effect could have been achieved through the use of a different pair of complementary words. Finally, if this investigation provides some general insight into the Platonic Corpus itself, it is that we have seen reason not to place excessive trust in Thrasyllus’ authority as a judge or transmitter of what was genuinely Platonic.51 It is possible (at the very least) that his motives for including or excluding a work were based on considerations that have little or nothing to do with questions of authenticity. More important, the analyses in this paper suggest that the terms used in antiquity to designate inclusion in and exclusion from the Thrasyllan organization were not, or were not only, intended to assert authorial status. References Alline, H. (1915), Histoire du texte de Platon, Paris. Bailly, J. (2004), The Socratic Theages: Introduction, English Translation, Greek Text and Commentary, Hildesheim/Zürich/New York. Barnes, J. (1991), “Hellenistic Platos,” Apeiron, 24: 115–128. Bing, P. (2008), The Well-Read Muse: Present and Past in Callimachus and the Hellenistic Poets, Ann Arbor. Bluck, R.S. (1961), Plato’s Meno, Cambridge. Brisson, L. (2005), “Epinomis: Authenticity and Authorship,” in K. Döring, M. Erler, S. Schorn (eds.), Pseudoplatonica, Akten des Kongresses zu den Pseudoplatonica vom 6.–9. Juli 2003 in Bamberg, Philosophie der Antike Bd. 22, Stuttgart, 9–24. Burnyeat, M. (1977), “Socratic Midwifery, Platonic Inspiration,” BICS, 24: 7–16. Carlini, A. (1972), Studi sulla tradizione antica e medievale del Fedone, Roma. —— (2005), “Alcune considerazioni sulla tradizione testuale degli scritti pseudoplatonici,” in K. Döring, M. Erler, S. Schorn (eds.), Pseudoplatonica, Akten des Kongresses zu den Pseudoplatonica vom 6.–9. Juli 2003 in Bamberg, Philosophie der Antike Bd. 22, Stuttgart, 25–35. Collard, C. and M. Cropp (2008), Euripides: Fragments, 8, Cambridge, MA/London. Cooper, J.M. and D.S. Hutchinson (1997) (ed.), Plato: Complete Works, Indianapolis/ Cambridge, MA. Colson, F.H. (1924), M. Fabii Quintiliani Institutionis Oratoriae Liber I, Cambridge. Dalfen, J. (2009), Platon. Werke, IX.1: Plato, Minos, Göttingen.

51 Examples of this trust can be found among those who in recent years have begun their exegeses from the assumption that all Platonic works that have been preserved in the tetralogical “canon” deserve on that basis to be considered authentic, e.g., Pangle 1987: 1–16; Zuckert 2009: 10 n. 21; Altman 2012: 1–2, 31–35. This approach to the Thrasyllan canon traces its modern origins back to Grote 1865: 132–211.

92

mark joyal

Döring, K. (2004), Platon. Werke, V.1: [Plato], Theages, Göttingen. Dörrie, H. and M. Baltes (1990), Der Platonismus in der Antike, Bd. 2, Stuttgart-Bad Canstatt. Dunbar, N. (1995), Aristophanes: Birds, Oxford. Erler, M. (2008), “Dire il nuovo in modo vecchio e il vecchio in modo nuovo: Gli spuria del Corpus Platonicum fra poetica e retorica ellenistica,” in Filologia, papirologia, storia dei testi. Giornate di studio in onore di Antonio Carlini, Udine, 9–10 dicembre 2005, Pisa, Roma, 225–241. Gaiser, K. (1963), Platons ungeschriebene Lehre: Studien zur systematischen und geschichtlichen Begründung der Wissenschaften in der Platonischen Schule, Stuttgart. Göransson, T. (1995), Albinus, Alcinous, Arius Didymus, Göteborg. Guthrie, W.K.C. (1978), A History of Greek Philosophy, vol. 5, Cambridge. Halperin, D.M. (1990), “Why is Diotima a Woman? Platonic eros and the Figuration of Gender,” in Halperin, J.J. Winkler and F. Zeitlin (eds.), Before Sexuality: The Construction of Erotic Experience in the Ancient Greek World, Princeton. Heidel, W.A. (1896), Pseudo-Platonica, Diss. Chicago. Jachmann, G. (1941), “Der Platontext,” NAWG, 11: 225–389. Joyal, M. (2000), The Platonic Theages: An Introduction, Commentary and Critical Edition, Philosophie der Antike, Bd. 10, Stuttgart. —— (2002), “The Theages Papyri,” ZPE, 139: 43–45. —— (2005), “Socrates as σοφὸς ἀνήρ in the Axiochus,” in K. Döring, M. Erler, S. Schorn (eds.), Pseudoplatonica, Akten des Kongresses zu den Pseudoplatonica vom 6.–9. Juli 2003 in Bamberg, Philosophie der Antike Bd. 22, Stuttgart, 97–117. Lesky, A. (1971), Geschichte der griechischen Literatur, third edition, Bern/München (= id. [1966] A History of Greek Literature, trans. C. de Heer and J. Willis). Liapis, V. (2009), “Rhesus Revisited: The Case for a Fourth-Century Macedonian Context,” JHS, 129: 71–88. —— (2012), A Commentary on the Rhesus Attributed to Euripides, Oxford. Mansfeld, J. (1994), Prolegomena: Questions to be Settled before the Study of an Author, or a Text, Philosophia Antiqua 61, Leiden/New York/Köln. Manuwald, B. (2005), “Zum pseudoplatonischen Character des Minos: Beobachtungen zur Dialog- und Argumentationsstruktur,” in K. Döring, M. Erler, S. Schorn (eds.), Pseudoplatonica, Akten des Kongresses zu den Pseudoplatonica vom 6.–9. Juli 2003 in Bamberg, Philosophie der Antike Bd. 22, Stuttgart, 135–153. Müller, C.W. (1975), Die Kurzdialoge der Appendix Platonica, München. Ogden, D. (1996), Greek Bastardy in the Classical and Hellenistic Periods, Oxford. Oswiecinski, S. (1979), “The Acephalous Dialogues,” Eos, 67: 55–67. Patterson, C. (2005), “Athenian Citizenship Law,” in M. Gagarin and D. Cohen (eds.), The Cambridge Companion to Ancient Greek Law, Cambridge, 267–289. Pfeiffer, R. (1968), History of Classical Scholarship: From the Beginnings to the End of the Hellenistic Age, Oxford. Philip, J.A. (1970), “The Platonic Corpus,” Phoenix, 24: 296–308. Reynolds, L.D. and N.G. Wilson (1991), Scribes and Scholars, third edition, Oxford. Rhodes, P.J. (1993), A Commentary on the Aristotelian Athenaion Politeia, second edition, Oxford. Ridgway, S.B. (1984), Roman Copies of Greek Sculpture: The Problem of the Originals, Ann Arbor. Ritchie, W. (1964), The Authenticity of the Rhesus of Euripides, Cambridge. Rowe, C. (2005), “What Might We Learn from the Clitophon about the Nature of the Academy?” in K. Döring, M. Erler, S. Schorn (eds.), Pseudoplatonica, Akten des Kongresses zu den Pseudoplatonica vom 6.-9. Juli 2003 in Bamberg, Philosophie der Antike Bd. 22, Stuttgart, 213–224. Ruschenbusch, E. (1966), Solons Nomoi: Die Fragmente des solonischen Gesetzeswerkes mit einer Text- und Überlieferungsgeschichte, Historia Einzelschrift 9, Wiesbaden.



“genuine” and “bastard” dialogues in the platonic corpus 93

Schironi, F. (2005), “Plato at Alexandria: Aristophanes, Aristarchus, and the ‘Philological Tradition’ of a Philosopher,” CQ, 55: 423–434. Schroll, O. (1901), De Eryxia qui fertur Platonis, diss., Göttingen. Shorey, P. (1933), What Plato Said, Chicago. Slater, W.J. (1976), “Aristophanes of Byzantium on the Pinakes of Callimachus,” Phoenix, 30: 234–241. Slings, S.R. (1999), Plato: Clitophon, Cambridge Classical Texts and Commentaries 37, Cambridge. Souilhé, J. (1962), Platon. Œuvres complètes, vol. 13, second edition, Paris. Stirling, L. (forthcoming), “Collections, Canons, and Context: The Afterlife of Greek Masterpieces in Late Antiquity,” in S. Birk, T.M. Kristensen and B. Poulsen (eds.), Using Images in Late Antiquity, Oxford. Tarán, L. (1975), Academica: Plato, Philip of Opus, and the Pseudo-Platonic Epinomis, Philadelphia. Tarrant, H. (1993), Thrasyllan Platonism, Ithaca/London. —— (1995), “Introducing Philosophers and Philosophies,” Apeiron, 28: 141–158. Taylor, A.E. (1949), Plato: The Man and His Work, sixth edition, London. Vancamp, B. (2010), Untersuchungen zur handschriftlichen Überlieferung von Platons Menon, Palingenesia 97, Stuttgart. Whittaker, J. (1989), “The Value of Indirect Tradition in the Establishment of Greek Philosophical Texts or the Art of Misquotation,” in J.N. Grant (ed.), Editing Greek and Latin Texts, New York, 63–95. von Wilamowitz-Moellendorf, U. (1962), Platon, vol. 2, third edition, Berlin.

Female Voice, Authorship, and Authority in Eudocia’s Homeric Centos Andromache Karanika A cento, often referred to as a literary patchwork, is a technical term that delineates the kind of work that derives from lines or segments such as couplets of lines taken from other, usually famous, works. The making of cento is based on a process of fragmentation and reconstruction in a new narrative. One text is dismembered and reassembled in order to create new meanings in a different context. In the particular case of Eudocia’s Homerocentones, a combination of approximately 2349 lines (following Usher’s edition) from Homer’s Iliad and Odyssey render stories from the Bible, from the Old and New Testaments, with emphasis on the latter. In this paper, I offer a reading of Eudocia’s Homerocentones that presents it as an independent work in its own terms without delving into the connections and alluded meanings created through the communication with the “original” text of Homer, as Eudocia would know it.1 From such a perspective, Eudocia participates not simply in an intertextual relationship with versions of the Homeric text that she would know, but poses her own, distinctly female voice, as a new authorial voice. Eudocia’s fluency in Homer has become a mask that she uses in a performative game, one meant to blur and impose the voice of authority, authorship, and feminine stance. After briefly considering evidence associated with her life and her other work on the life of St. Cyprian, I will be arguing that at a meta-literary level Eudocia employs her literary sophistication to reflect her own intellectual and religious transfiguration in Homeric terms through the process of dismemberment and rearrangement of the Homeric membra. By examining the stylization of lament and female emotion and the representation of space and movement in the way Eudocia constructs the feminine ­characters in

1 Current scholarship has rediscovered and reevaluated cento poetry, and Eudocia’s centos in particular. It has mostly focused on individual scenes in her Homerocentones with a view that examines the transportation from the Homeric context to Eudocia’s late antique reading and interpretation of the biblical stories. See notes below on individual points.

96

andromache karanika

her cento, we shall see that she constructs a pattern of movement and behavior for the female speaker, including her own “authorial” voice. Perhaps with few authors is the designation Late Antiquity as accurate as with Eudocia. Much of her work is lost; it is her paraphrastic poetic compositions that survive: the centos that treat biblical scenes, from the story of Adam and Eve to the Resurrection and Ascension, to a life of the martyr St. Cyprian.2 She is positioned as heir to a long literary tradition and is, in this light, the product of a time that, even if it did not call itself “Late Antiquity” as we now do, still conceptualized its work as such. For this reason it has met not only bitter contempt in philological appraisals but, even worse, oblivion. Current scholarship is actively reevaluating and contextualizing late antique literary works and treats cento writings seriously, abandoning earlier condescension. Centos have been regarded as an ostensive act of erudition and as such they have usually been studied from the viewpoint of a twofold relationship: that between author/cento producer and author of the “original” or imitated, used, and ultimately recomposed parts. In most studies pivoting on the axis of writer, “A,” and re-writer, “B,” at least two misconceptions hold sway. First is the relation between the two texts; sometimes scholars presume an “original” author and a “pseudo-author,” who is a predator. The second misconception is that both stem from a tradition of erudition and are the products of “writing,” but are denied a culture of deictic performance, oratorical or even educational, ritual, and liturgical uses. Let us first focus on the author, her text, and the reception of her “intertext” in her times. Eudocia, daughter of Leontius, an Athenian sophist, became an empress in early Byzantium after marrying Theodosius II in 421. She spent the last part of her life, the two decades from 440 to 460, in Jerusalem, having fallen out of favor of the imperial court for reasons that are not entirely clear. In her youth, she underwent a strict rhetorical and philosophical education.3 Her voice as the author of Homerocentones comes forth in epigrammatic evidence. Both the earliest manuscript of the cento (tenth c. Paris suppl. gr. 388), and the later one (fourteenth c. Iviron 4464), link the centos to Eudocia through the addition of epigrams that function as prologues to the poem.4 The authorship issues 2 For a summary on her works with a review on the scholarship on it see Sowers 2010. 3 For Eudocia’s rhetorical education as well as the possibility that she studied not only with her father but also with famous teachers like Orius and Hyperechius of Alexandria see Cameron 1982. 4 See Whitby 2007.



female voice, authorship, and authority

97

become more complicated as Eudocia appears in those epigrams not as the sole, unquestionable author, but rather as a co-author/editor of an earlier work by Patricius, which she perfected. In fact, the earlier Paris manuscript lists three authors for this work: Bishop Patricius, the philosopher, Eudocia Augusta, and a third person identified as Cosmas of Jerusalem, a later eighth-century poet and editor of the Homerocentones.5 Rey, following Ludwich’s earlier assessment, identifies Cosmas as an eighth-century bishop, thus making him more of an editor of the earlier existing text.6 While this title might reflect notions on the very idea of authorship through different phases of Byzantium, the epigrams themselves supply two joint names: that of Patricius and Eudocia, with Eudocia being not only slightly later chronologically but appearing as the driving force and the one who gave the definitive touch to any earlier reception of this Homeric retelling or paraphrase of biblical stories. Eudocia, raised as a pagan, converted to Christianity. She is a case of an author whose work reflects the turns of her own ideological journey. As someone who had received a classical education from her father, she was exposed to Homer from her early childhood and must have memorized a good chunk of Homeric selections.7 Schoolmasters dissected a text and offered it for pedagogical use in fragmented ways, not as a whole, often lingering on details of grammar. As with any text that is taught in anything resembling a classroom setting, any ancient text would be tailored to pedagogical needs and would be appropriated accordingly so that the idea of authorship would not apply as a relevant concept.8 A reformed sense of orality would arise as a result of pedagogical processes that encouraged experimentation with ancient “authorities” such as Homer. Memorization and restitching for the purposes of telling a story, even if the story had nothing to do with earlier narratives, would shift the literary effort on to forging a new link between form and meaning. The composer of centos uses Homeric authority to become the author of a new narrative. Although the “new” text draws from an earlier spring that any learned reader or audience member can more or less easily discern, it transforms earlier forms, blending them in a new context. In a way, the concept of “old” and “new” are blended, as the “new” is made from the “old” and 5 See Whitby 2007. Rey’s edition uses the Paris manuscript. 6 Ludwich 1897: 87; Rey 1998: 56–59. See also on this Smolak 1979: 31. 7 See Usher 1998, an excellent study which brings forth the role of memorization and the ultimate re-stitching of the Homeric text, Eudocia becoming the closest analogue we have to a late antique “rhapsode.” 8 See Lamberton 1992: 137.

98

andromache karanika

the “old” becomes “new” in a harmonious circle when the two seeming opposites blend. What kind of audience should we presuppose? Usher in his pioneering work emphasized the orality and performance aspect in the crafting of the Homerocentones. He convincingly argued that Eudocia’s poem has an “oral/ aural” aspect to it, and it is indeed possible that it comes from a tradition of memorization and retelling in a showcase of wit and erudition. One can safely assume further that the poem actively involved a communicative process with an audience that could be versatile in Homeric Greek as well as in the Greek of its time, at the very least.9 Homer occupied an important role in educational settings. Moreover, the rhetorical tradition is also reflected in the text which makes a good attempt to build a character’s declamation in biblical terms and thereby offer multiple interpretations not only on biblical narratives and even christological matters but also on the symbiosis of ancient and Christian literature. In examining the authorship issue, we see that in the epigrams preceding the Homerocentones, Eudocia states that it is the result of “common labor” between Patricius and herself, “even though she is a woman.” The female authorial voice is placed in a frame when an earlier male voice is employed in conjunction with the female. ἀλλ’ ἔμπης ξυνὸς μὲν ἔφυ πόνος ἀμφοτέροισι, Πατρικίῳ κἀμοί, καὶ θηλυτέρῃ περ ἐούσῃ. (Usher, Praefatio 32–33). But nevertheless this was a common labor to both of us, meaning Patricius and myself, even though I am a woman.10

Eudocia has made it clear that Patricius was the first who conceived the work but did not say everything “truthfully” and certainly not without the necessary “harmony of the words.”11 οὕνεκα δὴ πάμπρωτος, ἐμήσατο κύδιμον ἔργον. ἀλλ’ ἔμπης οὐ πάγχυ ἐτήτυμα πάντ’ ἀγόρευεν· οὐδὲ μὲν ἁρμονίην ἐπέων ἐφύλαξεν ἅπασαν, οὐδὲ μόνων ἐπέων ἐμνήσατο κεῖνος ἀείδων, ὁππόσα χάλκεον ἦτορ ἀμεμφέος εἶπεν Ὁμήρου. (Ludwich 1897, 4–8). . . . since he was the very first to contrive glorious work. but nevertheless, he did not speak entirely truthfully

9 For Eudocia’s role in the “Graecizing” of the empire see Connor 2004: 60. 10 All translations are my own, unless otherwise indicated. The text from Eudocia’s Homerocentones is from Usher’s 1999 edition. 11 She says in line 4 that “he was the first to conceive of the ‘glorious work.’ ”



female voice, authorship, and authority

99

nor did he guard all the harmony of the epic lines, and he didn’t recite from memory, in his singing, only epic lines that the brazen heart of immaculate Homer told him in epic.

In these lines the final verb eipen relates etymologically to the repetition of epeōn and links the act of making epic lines with the very act of speaking. Etymological play and alliteration also brings together the words emēsato and emnēsato bridging the act of inventing poetry and remembering epic lines in recitation. The verb used for her criticism of Patricius’ work, who did not exclusively work from Homeric material, is ephulaxen meaning he did not “guard” the total harmony of the words. Eudocia sets for herself the role of a “guardian” of the received text and puts forth her concept of “authenticity” of the new work, when she criticizes Patricius for not speaking “truthfully.” In fact, she creates a refreshed communication circle: Homer is the Muse, and the poet’s role is to keep everything “truthful.” While most studies allude to the process of recontextualization and the allusions evident in comparison with the Homeric text itself, the poem brings an epic armor and façade to biblical paraphrase. As such, it can be put in the aesthetic milieu of its time. Canons, and with them meanings and tastes, shift, making Late Antiquity a time of flux and transformation.12 Ziolkowski compares the reshuffling and the transformations of earlier literature with the reutilization of classical material in architecture and art. In this reading, centos can be viewed as the “most extreme literary parallel.”13 Just as a church could have material components from the pagan temple, such as columns, stones, etc., or reliquaries or chests could use small ancient things like gems or coins, literary production played perhaps more freely than ever before with segments of earlier literature, presenting them in new contexts. And just like new architectural or art formations made from old material, the old material needed no introduction but stood out. In the most extreme cases, such as alterations in the Parthenon or the temple of Apollo at Didyma, any Christian alterations did not fully convert the identity of the building.14 In similar terms, the Homeric material stands out. The accenting of new meanings, though, comes not by the microscopic but rather the bird’s eye view that shows a new, harmonic landscape. Homeric authority is used to create new meanings, but also becomes the vehicle that establishes female erudition. Centos are not new to women writers. The Roman aristocrat Faltonia Betitia 12 Ziolkowski (2007: 22) described Late Antiquity as the time of “shifting canons.” 13 Ziolkowski 2007: 22. See also Karanika 2011: 266. 14 Doig 2008: 83.

100

andromache karanika

Proba composed a cento made from Virgil’s lines in a Christianized plot.15 Proba refers to herself as a vatis, a term used not simply interchangeably for poeta but denoted a difference in the kind of instructive poet of importance. Such a self-reference is “hardly an act of humility.”16 Indeed, as it has been argued, it is possible that Eudocia knew Proba’s composition. What is remarkable with Proba’s cento is that it very quickly found use in the classroom as a textbook. After all, students use the microscopic method, line by line, and the context becomes irrelevant when the lines are read and discussed for the purposes of learning better Latin. Proba’s centos survive in more manuscripts and editions than the writings of any other pre-modern woman.17 In fact, in certain Christian settings, Proba’s work was far more influential than Virgil’s own “original” work.18 As Stevenson has noted, several women in particular were attracted to the composition of centos. Although certainly not exclusively female, the path that Proba and Eudocia seem to pave makes us wonder why women took to this form. Just as Proba is hardly a humble person, Eudocia claims her own superiority over centos that she, supposedly, received. Homeric authority lends a hand to female erudition and ambition, but, in the process, not only is the idea of authorship deconstructed, but the literary edifice is reconstructed to stand for those who want a Christianized plot. In the case of Eudocia, the authorial female voice is firmly proclaimed; and, in particular, female speech is carefully crafted and echoed with considerable emphasis throughout her poem. The first female speech is that of Virgin Mary in the Annunciation scene (lines 202–268).19 Her speech is prepared with a director’s care to focus on the emotional world. Eudocia employs a full combination of Homeric clusters and formulae to show the young woman’s speechlessness and painting of her emotional world as her limbs loosen, delight and pain seize her, and her eyes are filled with tears. Even the hair of her body gets attention. Eudocia writes like a painter who would do a close-up painting, combining a number of verses from both the Iliad and the Odyssey that refer to different female characters in the respective epics to create the full effect before female speech folds itself into her own creation, with a sapphic touch. 15 On Proba see McGill 2005, Stevenson 2005. On dates see Shanzer 1994 who places Proba in 390s whereas Matthews 1992 places her earlier in the mid fourth century. 16 Stevenson 2005. 17 Stevenson 2005: 69. 18 Stevenson 2005: 70. 19 On the Annunciation theme see Smolak 1979.



female voice, authorship, and authority

101

ὣς φάτο, τῆς δ’ αὐτοῦ λύτο γούνατα καὶ φίλον ἦτορ, ἡ δ’ οὔτ’ ἀθρῆσαι δύνατ’ ἀντίη οὔτε νοῆσαι, καί ῥ’ ἀκέουσα καθῆστο, ἐπιγνάμψασα φίλον κῆρ. τὴν δ’ ἅμα χάρμα καὶ ἄλγος ἕλε φρένα, τὼ δέ οἱ ὄσσε δακρυόφιν πλῆσθεν, θαλερὴ δέ οἱ ἔσχετο φωνή, ὀρθαὶ δὲ τρίχες ἔσταν ἐνὶ γναμπτοῖσι μέλεσσιν. (Homerocentones 240–245) Thus he spoke, and her knees and heart loosened, and she couldn’t utter anything in response or think, but instead sat down silently, curbing her heart, and both delight and pain seized her mind, while her eyes were filled with tears, and her voice grew and her hair stood straight on her pliant limbs.

In this example, the role of mimesis is refracted. It is not imitating Homer or the Bible per se, but rather retelling biblical stories in an adorned fashion. Eudocia is expansive in her description in ways that the biblical stories are not. We could discern an external mimetic force, namely the Homeric verses, as the forma, and the internal narrative needs as deriving from the Bible. The Homerocentones paraphrase biblical stories in ways that force the author and reader (or even possible audiences at some point) to pause and reflect via the Homeric form on details of staging the scene or emotional responses of the characters presented. The scene of the Cana Wedding also focuses on women. In a different movement, this time from the one to the many, Eudocia directs her attention to crafting women spectators (548–547, Usher). The eight lines for Jesus’ mother (590–598) are followed by Jesus’ response. The next prominent episodes involving women are those of the hemorrhaging woman (lines 993–1045) and then of the Samaritan woman, in which Jesus appears far more reproachful than the biblical story presents it. In the first, the woman is described like a worm on the earth (line 998) and Eudocia presents the emotional outburst as she cries (δάκρυα ἔκβαλε θερμά, 1003). Her speech contains an amalgam of supplication and self-reproach, with reference to Helen’s self-reproach in Iliad 6. Language of pain and lament are intertwined throughout her plea to Jesus: ἄλγος, (1004), στενάχω (1006), κήδεα μυρία πέσσω (1006), ἕλκος μὲν γὰρ ἔχω τόδε καρτερόν (1008), ἕλκος δ’ ἰητὴρ ἐπιμάσσεται (1016), μελαινάων ὀδυνάων (1017), γόῳ φρένα τέρπομαι (1019), κόρος κρυεροῖο γόοιο (1020), ἕλκος ἄκεσσαι (1021). In the centos, the author mixes the language of intense physical pain such as the pain caused by a wound (presented with several iliadic references) with typical lament-like supplication to Jesus. This linguistic strategy highlights the woman’s condition further and brings her to the focal point

102

andromache karanika

of the cento’s narrative.20 Eudocia, though, with her general interest in depicting the space stages the entire episode as a gradual movement from a worm to the upper part of the palace (δ’ εἰς ὑψόροφον θάλαμον κίε δῖα γυναικῶν, 1041). The hemorrhaging woman’s gradual acquisition of mobility becomes the axis that transforms her position. As with the Samaritan woman, the space and time of the encounter are portrayed in detail. In fact, the time of the day, Ἦμος δ’ ἠέλιος μέσον οὐρανὸν ἀμφιβεβήκει (line 1046, and 1872) is the same line from Iliad 8.68 that marks the meeting point of Jesus and the Samaritan woman and the time around the Crucifixion. The Samaritan woman’s response to Jesus is the most extensive female speech (line 1073–1115) and uses overwhelmingly Odyssean material, addressing Jesus repeatedly as ξεῖνε (1073, 1086, 1092, 1094, 1108, 1112, 1114). This way, the Samaritan woman becomes the interrogative agent of the encounter and it is through her eyes that such a perspective is imposed. While Christian writers remark on the one hand that the Samaritan woman is a “stranger” to Jesus and refer to her as ἀλλόφυλος, yet on the other hand point to her as the one who received the “Lord’s long words,” as John Chrysostom put it, Eudocia’s presentation of the scene is unique: She makes the encounter with the Samaritan woman a central focus of her narrative and gives a lengthy section to presenting the woman’s words. ἐπεὶ καὶ ἡ Σαμαρεῖτις γυνὴ καὶ πτωχὴ ἦν καὶ ἐναγὴς καὶ ἀλλόφυλος, καὶ ὅμως ἀξία τῶν μακρῶν ἐκείνων ἐνομίσθη λόγων εἶναι τῷ τῆς οἰκουμένης Δεσπότῃ (John Chrysostom, Homilia dicta in templo sanctae Anastasiae 63.493). The Samaritan woman was both poor, guilty, and of a different tribe, yet, she was thought worthy of the long speech by the Lord of the world.

It is not the Samaritan woman who is the “stranger”; rather, Jesus is contrasted as the stranger and emphatically so. Eudocia subverts the episode.21 In this central episode Eudocia constructs in her restitching a distinctly female point of view. It is the lens of the Samaritan woman that is the central one. She is the one receiving her interlocutor, who is repeatedly referred to as “xenos.” It is her voice that proclaims the strangeness of Jesus rather than the biblical construction of the Samaritan woman as the “stranger.” Moreover, the same movement from down upwards is implied 20 For the allusions that to Glaucus’ trauma in the scene of the hemorrhaging woman see Sandnes 2011. 21 For a reading of Eudocia’s centos, with a penchant to the role of gender and the epi­ sode of the Samaritan woman with emphasis on multiple interpretation in the late antique moral sphere see Sowers, Breaking Boundaries.



female voice, authorship, and authority

103

here, as was the case earlier with the scene of the hemorrhaging woman with a movement to the city around which there was a citadel, again paraphrasing lines from Odyssey 6.262: αὐτὰρ ἐπὴν πόλιος ἐπεβήσατο ἣν πέρι πύργος (line 1118). Even when the Samaritan woman talks to the others after her encounter with Jesus, she still refers to him as xeinos (1129). The address with xeine is opposed both to the usual biblical kyrie, but also to the Homeric equivalent that Nonnus used in his paraphrase of St. John’s gospel: koirane, as the closer Homeric parallel to the biblical and liturgical kyrie. οὐτιδανὴ Σαμαρεῖτις ἀμείβετο θήλεϊ φωνῇ· κοίρανε, γινώσκω σε θεηγόρον ἄνδρα προφήτην· (Nonnus, 4.87–88) This Samaritan woman of no importance was responding in her female voice: Lord, I acknowledge you as a prophet, a man who speaks divine words.

In paraphrasing, Eudocia comes up with her own interpretation, and places, as I argue, her own authorial voice in the repetitive usage of xeine in the Samaritan woman’s address.22 She is guiding the lens of the poem from the female point of view. Female language takes the front center in the episode of the resurrection of Lazarus, where Lazarus’ sister announces in lamentation the death of Lazarus (lines 1236–1246). She uses formulaic expressions like καθίκετο πένθος ἄλαστον (1239), and ends her announcement with ululation (1247, ὣς εἰποῦσ’ ὀλόλυξε).23 Lament is the language of the woman who anointed Jesus with myrrh (lines 1300–1326). Although we do not have her speech, only her movement of supplication and tears, Jesus’ praises her by telling her that her kleos will reach the broad sky (1316). The episode of Judas’ betrayal also receives a female touch. His hanging is given in words taken from the Nekyia on Jocasta’s fate (line 1712, Odyssey 11.278), ἁψάμενος βρόχον αἰπὺν ἀφ’ ὑψορόφοιο μελάθρου. This is an episode that has received significant attention; in fact, I would like to suggest further that there are ways in which one could say that these lines are also reminiscent of the Byzantine drama of Christus Patiens, which uses Euripidean threads. In the latter, Jesus’ reproach to Judas is carefully given (compare Homerocentones lines 1700–1701 to Ch. Pa. 290, or Homerocentones 1716–7, to Ch. Pa. 335–336).24

22 As Johnson (2006: 106) argues, there is no paraphrasis without interpretation. 23 On ancient ululation see Karanika 2009. 24 Also compare lines 1706–7 to Ch. P. 290.

104

andromache karanika

Eudocia’s expansive and embellishing approach to the biblical characters is manifested once again in the scene of the lamentations for Jesus’ death. The mother’s lamentations, her gestures, and her words (2040– 2073) become the focal point with a level of detail that draws from a depth of both the Iliadic and Odyssean tradition. The last line (from Odyssey 21.58) after the description of tears, the lament, and the address to the son (2049–2050) brings the image of a departing Virgin Mary in a Penelopean mask as she moves to go to her palace βῆ ῥ’ ἴμεναι μέγαρόνδε μετὰ μνηστῆρας ἀγαυούς (line 2073, Od. 21.58).

Movement towards the palace consistently becomes part of the female framing at the end of a scene. The same pattern is repeated in the immediately following scene, that of the resurrection at the end of the Homerocentones. After the lamentation (lines 2177–78), the mother’s pain is presented like the melting of the snow: ὡς δὲ χιὼν κατατήκετ’ ἐπ’ ἀκροπόλοισιν ὄρεσσιν (line 2178)

The end of the representation of the lament coincides with a simile (taken from Od. 19.205). In a paraphrase when Jesus entrusts the care of his mother to his disciple John, following the last female speech of the Homerocentones, the pattern of the mother who leaves to go back to her house, again in allusion to Penelope’s movement, becomes the last female act. This representation of movement is not only a detail in the structure that draws from the movement of Homeric heroes around their constructed space. For Eudocia there is self-conscious metapoetic resonance. Eudocia is a figure whose life presents an extra-ordinary journey from Athens to Constantinople and ultimately to Jerusalem where she spent the last part of her life. This coincides with an intellectual journey of transformation. She experienced a Kulturkampf, which she bridged through her work, work that marries the first part of her life with the second. As a Christian centoist, she relates to her hypotext aiming to a better understanding of the Homeric text as well as the biblical story through this.25 The constant address of the Samaritan woman to Jesus as xeine with the woman being the focal point of the encounter and the emphasis on the female movement in the space that she constructs out of Homeric material is consistent with her own self-perception. While most of Greek late antique 25 See also Sandness 2011 for an assessment of Christian cento retellings of the Bible.



female voice, authorship, and authority

105

literature does not present any “real” female voice, but rather constructs feminine voice, Eudocia’s work appears as a vox in deserto with a Homeric dress, yet unequivocally self-conscious, and feminine as such.26 I submit that Eudocia uses female characters in her work to construct not only the character’s female authority but also her own female authorial voice. Eudocia also wrote the life of St. Cyprian.27 This was again a paraphrase in hexametric poetry of an earlier prose text with a contested date of composition.28 It is not difficult to see why Eudocia would take such an interest in a conversion story of a figure who used to be a magician. The underlying theme of transformation is a persistent interest both as substance and as form, through the use of centos in one case and the poeticization of available prose on the other. The authorial voice that she constructs is consistent throughout her work. Although there is some controversy regarding her Athenian birth and upbringing, and it is certainly a possibility that later generations played up the conversion story, the prologue of the Life of St. Cyprian projects a consistent self-constructed view not only of St. Cyprian, but of Eudocia herself as an Athenian.29 The sense of space that is represented there is in accordance with the movement and space of her female heroines in her centos. ᾤκεον εὐγενέων δέ τ’ Ἀθηναίων πόλιν αἰπήν. καὶ ναέτης γενόμην δέ, ἐπεὶ τόδε ἥνδαν’ ἐμεῖο πατράσιν, οἵ μ’ ἔτεκον. δεκάτῳ δ’ ἐπιβὰς λυκάβαντι Δηοῖ δᾷδας ἀνῆψα, Κόρης δ’ ἀργεννὸν ὑπέστην πένθος. Ἀθηναίης δ’, ἥτις πόλιν ἐστὶν ἐς ἄκρην, ῥέξα δρακοντείους τελετάς. ζάκορος δὲ τελεσθεὶς ἦλθον ἐς Οὐλύμποιο νάπος, τό γε δὴ καλέουσιν ἀγνῶτες μακάρων ἀμενηνῶν εἶναι ἔδεθλον. (Eudocia, 2.16–23) I dwelled in the towering city of the well-born Athenians since the time my parents lived there. At the age of ten I bound fast a torch for Demeter and bore the white mourning of Kore. I performed the snake rites of Athene who is on the citadel. Having been a temple attendant I went to the grove of Olympus which those ignorant ones say is the precinct of their feeble deities.30

26 For the construction of gender in Late Antiquity see Martin’s Introduction in Martin and Cox Miller 2005: 10–12. 27 See Holum 1982: 117–120. 28 According to Livrea 1998 the poem on St. Cyprian was composed around 439 whereas Cameron (1982: 282–3) places it at the later stage of her life in Jerusalem. 29 For the controversy see Cameron 1982. 30 Translation by G. Ronald Kastner in Wilson Kastner et al. A 1981: 159.

106

andromache karanika

This episode is the epicenter of St. Cyprian’s conversion and condemnation of his prior religion. Yet the reference to Athens and its rituals are only a reflection of Eudocia’s earlier life. As the world becomes “increasingly intolerant of religious plurality,” as Sowers put it, Eudocia does something unique: she separates form and content and through the interweaving of ancient form and Christian content builds a new cultural bridge without losing her own identity.31 Her cento, in this regard, is not just an exercise or a failed attempt to produce poetry; on the contrary, it becomes in her hands a vehicle for cultural innovation and social revolution, as her female authorial voice is at the fore. Eudocia is, after all, a newcomer to Christianity, and when talking about it she decides eloquently to return to her home, Homer. References Cameron, Alan (1982), “The Empress and the Poet: Paganism and Politics at the Court of Theodosius II,” Yale Classical Studies 27: 217–282. Cameron, Averil (1993), The Mediterranean World in Late Antiquity, London: Routledge. Connor, Carolyn L. (2004), Women of Byzantium, New Haven. Doig, Allan (2008), Liturgy and Architecture from the Early Church to the Middle Ages, London. Elm, Susanna (2005), “Hellenism and Historiography: Gregory of Nazianzus and Julian in Dialogue” in Journal of Medieval and Early Modern Studies Fall 2003 33(3): 493–515. Grig, Lucy and Gavin Kelly (2012), Two Romes: Rome and Constantinople in Late Antiquity, Oxford. Holum, Kenneth G. (1982), Theodosian Empresses, Berkeley. Johnson, Scott F. (2006), The Life and Miracles of Thekla: A Literary Study, Washington. Karanika, Andromache (2011), “Homer the Prophet: Homeric Verses and Divination in the Homeromanteion,” in A.P.M.H. Lardinois, J.H. Blok and M.G.M. van der Poel (eds.), Sacred Words: Orality, Literacy and Religion: Orality and Literacy in the Ancient World, vol. 8., Leiden: 255–277. —— (2009), “The Female Ritual Cry in the Homeric Hymn to Apollo: From Poetics to Politics,” in L. Athanassaki, R.P. Martin and J.F. Miller (eds.), Apolline Politics and Poetics, European Cultural Center of Delphi Publications, 67–77. Lamberton, Robert (1992), “The Neoplatonists and the Spiritualization of Homer,” in Lamberton and Keaney 1992: 115–33. Lamberton, Robert and John J. Keaney (1992) (eds.), Homer’s Ancient Readers: The Hermeneutics of Greek Epic’s Earliest Exegetes, Princeton. Livrea, Enrico (1998), “L’imperatrice Eudocia e Roma: per una datazione del De S. Cypri,” Byzantinische Zeitschrift 91: 70–91. Ludwich, A. (1897), Eudociae Augustae, Procli Lycii, Claudianii Carminum Graecorum Reliquiae, Leipzig. Martin, Dale B. and Patricia Cox Miller (2005), The Cultural Turn in Late Ancient Studies: Gender, Asceticism, and Historiography, Durham, NC.

31 Sowers 2010: 33.



female voice, authorship, and authority

107

Matthews, J. (1992), “The Poetess Proba and Fourth-Century Rome: Questions of Interpretation,” in M. Christol, S. Demougin, Y. Duval, C. Lepelley and L. Pietri (eds.), Institutions, société et vie politique dans l’ empire romain au IVe s. ap. J.-C. Collection de l’ École française de Rome 159: Roma: 291–304. McGill, Scott (2005), Vergil Re-Composed, Oxford. Rey, A.L. (1998) (ed.), Patricius, Eudocie, Optimus, Come de Jerusalem: Centons Homériques (Homerocentra), Sources chrétiennes 437, Paris. Sandnes, Karl Olav (2011), The Gospel ‘According to Homer and Virgil’: Cento and Canon, Supplements to Novum Testamentum 138, Leiden. Shanzer, Danuta (1994), “The Date and Identity of the Centonist Proba,” Recherches Augustiniennes 27: 75–96. Smolak, K. (1979), “Beobachtungen zur Darstellulngsweise in der Homerzentonen,” Jahrbuch der Oesterreichinischen Byzantinistik 28: 29–49. Sowers, Brian (2010), “Retelling and Misreading Jesus: Eudocia’s Homeric Cento,” in Weir and Koyzis 14–33. —— (2008), “Eudocia: The Making of a Homeric Christian,” diss., University of Cincinnati. Stevenson, Jane (2005), Women Latin Poets: Language, Gender, and Authority, from Antiquity to the Eighteenth Century, Oxford. Usher, Mark David (1999), Homerocentones Eudociae Augustae, Stuttgart and Leipzig. —— (1998), Homeric Stitchings, New York. —— (1997), “Prolegomenon to the Homeric Centos,” Journal of Philology, 118, no. 2: 305– 321. Weir, Heather and Nancy Calvert Koyzis (2010), Breaking Boundaries: Female Biblical Interpreters Who Challenged the Status Quo, New York and London. Whitby, Mary (2007), “Nonnus’ Paraphrase of St. John’s Gospel and ‘Eudocia’s’ Homeric Centos,” in J.H.G. Scourfield (ed.), Texts and Culture in Late Antiquity: Inheritance, Authority, and Change: 195–231. Wilson-Kastner, Patricia et al. (1981), A Lost Tradition: Women Writers in the Early Church, Washington. Winkler, John and Gordon Williams (2010), Later Greek Literature, Cambridge. Ziolkowski, Jan M. (2007), “Middle Ages,” in C.W. Kallendorf (ed.), A Companion to the Classical Tradition, Oxford: 17–29.

The Surgical Treatises of the Corpus Hippocraticum: Statistical Linguistics and Authorship Mikel Labiano 1. Introduction The surgical treatises1 of the Corpus Hippocraticum (CH), Joints (Art.), Fractures (Fract.), Surgery (Off.), Instruments of reduction (Mochl.), Ulcers (Vlc.), Wounds in the head (VC), Fistulas (Fist.) and Haemorrhoids (Haem.) are the most attractive among the whole corpus of medical writings gen­ erally ascribed to Hippocrates in antiquity, mainly Art., Fract. and VC, due to their high level of technical composition and scientific and method­ ological care. However, we honestly know little or nothing about their author(s), date of composition, and place within the CH. Though it is true that throughout the centuries to present times there have been many different hypotheses attempting to solve these problems, they have met with little success and no certainty. See for example Laín (1970), Bourgey (1953: 15–105), Withington (1928: XXII–XXV, 2–5, 84–93), Jouanna (1992: 527–563), Duminil (1998: 25), and Rodríguez Alfageme (2000: 174). So, from linguistic, stylistic, historical, conceptual, medical, etc. perspec­ tives, many hypotheses have been proposed, including a common author for Art., Fract., and VC, connected to the author of Prog., Acut., Aph., and Aër. Other possibilities include the following: Off. has connections with Fract.; Mochl. has connections with Art.; despite these connections Mochl. and Off. are from different authors than Fract. and Art., which share one common author as previously mentioned; Fist. and Haem. have one com­ mon author, probably the same as VC; Fist. and Haem. are from different authors than VC; Vlc. is probably from the author of VC; Fist. and Haem. share one common author with Vlc., etc. The list is endless, and, in short, chaotic. Some of these theories and hypotheses are evidently contradic­ tory and experts can reach no agreement. So we are left with an important

1 This paper, based on research funded by project FFI2012–36050 of the Spanish Minis­ try of Economy, is a revision and update of a previous paper (Labiano 2004).

110

mikel labiano

and crucial set of medical writings in the history of ancient medicine, but few clear ideas about their formation, composition, relation to each other, author(s), chronology, and adhesion to one medical school or another, among other concerns. The purpose of this paper is therefore to attempt to clarify some of this ambiguity and contradiction, and, in this respect, a statistical study applied to linguistics – that is, statistical linguistics – is proposed as a use­ ful method and a starting point for future studies. This procedure has been successfully applied to other medical writings in CH in previous studies by the research team of the Universidad Complutense de Madrid (UCM). See Ángel (2003) for Ep. V and VII, Sierra de Grado (2003) for VM, Aër., Carn., and Prorrh. II, and especially Rodríguez Alfageme for Gland. (1992), Vid. Ac. (1993), and Loc. Hom. (2010). In short, this method involves a count of syntactic links (particles of coordination, links of coordination and subordination, absolute genitive, asyndeton), in groups of 25 sentences over a text, in order to compare the average and statistical deviations, using the Student’s t-test within two texts. This will determine whether the two compared texts may originate from the same writer, in terms of statistical probability. The first stage, therefore, is to “translate” the text into numbers, in order to make statisti­ cal analysis possible (Muller 1973: 15). Through comparisons of sets of two texts and their statistical deviations, several types of relations can be estab­ lished between them based on statistical probabilities. These are relations of common/divergent authorship, proximity/distance in time, in style, etc. Analysis of syntactic distribution of sentences and links throughout one text allows us to obtain a syntactic pattern – a syntactic style, unique to each writer, with a constant rate in every era. The use and distribution of these links, present or absent (asyndeton), in relation to the content of the text, reflects the logical and psychological methods each writer applies to his text, his attention to style, etc. In short, this syntactic pattern is a unique and personal way to represent what a writer wants to communi­ cate. It is a personal brand; his personal brand. According to these statistical probabilities – and it is important to emphasize that they are statistical – it can be shown statistically that two texts are from the same writer or, at least, that they belong to the same period. It is also possible, on the other hand, to deny all these connections. This is the main hypothesis, or null hypothesis, that two texts come from the same writer/author, whereas the alternative hypothesis is the opposite



the surgical treatises of the corpus hippocraticum

111

(the two texts do not come from the same writer/author). After making comparisons between pairs of texts, this paper will explore whether it is possible to accept or reject the null hypothesis in statistical terms. At this point, it is important to explain that the act of not rejecting the null hypothesis does not necessarily imply that it has been accepted as fact. However, the act of rejecting the null hypothesis does require acceptance of the opposite. In this way, the act of rejecting the null hypothesis pro­ vides more information than the simple act of not rejecting it, because there are no statistical probabilities against it (Muller 1973: 124). The statistics give us objective evidence for further studies (linguistic, philologic, historical, medical, etc.). Given the problems listed above, this will hopefully prove to be a good starting point. This paper only deals with this statistical study, in order to establish relations of proximity between the Surgical Treatises in the CH and their (unknown) authors. For general details of the method, see Dixon & Massey (1966) and Muller (1973). For more specific details, application to CH and counting of links, see Rodríguez Alfageme (1992, 1993 and 2010) and Sierra de Grado (2003, mainly 405–409). 2. Distribution of Sentence Links: Graphs Before discussing these statistical operations, some graphs will help us to better explain this process visually, that is, the syntactic pattern of each text obtained through the syntactic distribution of links and sentences, by groups of 25, 50 or 100 sentences. The X-axis represents the groups 1, 2, 3 and so on of 25, 50, 100 or 200 sentences, as indicated, with the Y-axis representing the number of coordinated and subordinated sentences by group as well as the number of asyndeton (if remarkable).

112

mikel labiano

3. Student’s t-Test: Application In the next pages, all the surgical treatises will be compared in pairs with the Student’s t-test and the results will be carefully examined and explained. Comparison between Fist. / Haem. Fist. / Haem. (25 sentences by group) t student α

Coord. –0,0687 0,9463

Subord. 1,1815 0,2623

Asynd. –0,3246 0,7515



the surgical treatises of the corpus hippocraticum

113

In hypothesis testing, the significance level, usually denoted by the Greek symbol α, is the criterion used for rejecting the null hypothesis. Tradi­ tionally, experimenters have used either the 0.05 level (sometimes called the 5% level) or the 0.01 level (1% level), although the choice of levels is largely subjective. The lower the significance level, the more the data must diverge from the null hypothesis to be significant (Muller 1973: 127). For further safety the 0.05 level or 5% level will be used in this study as the acceptable significance level. Assuming at this first stage the null hypothesis is true (that is, the two texts come from the same writer), if the probability (α value) is less than or equal to the significance level (0.05), then the null hypothesis is rejected and the outcome is said to be statisti­ cally significant (the two texts do not come from the same writer). As we can see, the three α values for coordination, subordination, and asyndeton are much higher than 0.05, so we can safely accept the null hypothesis and assert that these two treatises come from the same writer. The α value of 0.9463 for coordination is a very high one and these results encourage us to consider whether the two treatises are in fact part of the same work, as it has been traditionally postulated, due to the importance of coordination. The α value of 0.7515 for asyndeton is also very high. Coordinated sentences appear more widely at the beginning of the trea­ tise, while asyndeton appears later. This trend changes throughout the text and moves towards a confluence of both syntactic procedures at the end of the treatise, culminating in a decreased coordination and increased asyndeton. The graph shows this trend very well. All this reflects a very strong syntactic pattern, surely due to a deliberate style decision on the part of the writer rather than by chance. We will have the occasion later to observe this marked syntactic pattern again. Comparison between Off. / Mochl. Mochl. / Off. (50 sentences by group) t student α

Coord. –0,0590 0,9538

Subord. 1,2231 0,2447

Asynd. –0,9913 0,3410

As in the previous case, we have a wide range of probabilities for not err­ ing in accepting the null hypothesis of unity of authorship. This probabil­ ity is especially high, as it was earlier, in the case of coordination, with a value of α 0.9538. With respect to the pattern shown in the graphs, we can

114

mikel labiano

also say the same for the couple Fist.-Haem. This should be interpreted as a feature of style and deliberate composition. Such behavior in both texts invites us to compare them. Comparison between Fist.-Haem. / Off.-Mochl. Haem. / Mochl. (25 sentences by group) t student α

Coord. –0,2443 0,8092

Subord. 1,1680 0,2558

Asynd. –0,4704 0,6428

Haem. / Off. (25 sentences by group) Coord. t student –0,2305 α 0,8212

Fist. / Mochl. (25 sentences by group) t student α

Coord. –0,2193 0,8282

Subord. 0,3065 0,7618

Asynd. 0,0134 0,9894

Subord. 0,0866 0,9322

Asynd. 0,3053 0,7649

Fist. / Off. (25 sentences by group) Coord. t student –0,1776 α 0,8612

Subord. –1,1637 0,2615

Asynd. 0,6795 0,5065

The results speak for themselves. The values of α again persuade us not to reject the null hypothesis – that is to say that this group of four treatises come from the same writer. Comparison between Fract. / Art. Art. / Fract. (100 sentences by group) t student α

Coord. 0,9159 0,3665

Subord. 1,5588 0,1288

Asynd. 0,0245 0,9805

The high value of α 0.9805 for asyndeton does not help us much on this occasion, though at first it may seem otherwise, because the presence of asyndeton in this pair of treatises is almost negligible, meaning that this information is completely inconsistent and proves nothing. However, the values of α for coordinated and subordinated sentences allow us to not reject the null hypothesis, and so accept the traditional interpreta­ tion (Regenbogen 1914, Withington 1928: 5, Bourgey 1953: 60–61) that Fract. and Art. are treatises by the same author. The graphs also show this clearly. The following is a comparative graph with the same number of sentences per group:



the surgical treatises of the corpus hippocraticum

115

It can be observed that the frequency levels between coordinated and subordinated sentences are fairly constant; in the case of the coordinates, they usually remain between 50 and 60 sentences per group, while in the case of the subordinates this range is between 30 and 40 sentences per group. Comparison between Vlc. / VC Duminil recently proposed (1998: 25) that these two texts were the work of one author. Let us compare them, before proceeding to search for other possible relations between these surgical treatises. Vlc. y VC (50 sentences by group) t-student α

Coord. –1,2660 0,2248

Subord. –1,1737 0,2587

Asynd. 3,9785 0,0012

The value of α 0.0012 for asyndeton invites us to reject the null hypothesis, i.e., these two treatises are not from the same author. The value of asyn­ deton is irrelevant in the case of VC (close to zero) but it has to be taken into account in the case of Vlc., because it follows the same pattern of behavior we have seen before in the case of treatises Fist., Haem., Off. and Mochl., as shown in the graphs. The values for the coordinated and subor­ dinated sentences, albeit within the margins of allowing us to accept the null hypothesis, are not as clear as those found in previous comparisons. This, therefore, leads us to reject the null hypothesis and the hypothesis formulated by Duminil. At the moment the situation with regards to acceptance or rejection of the null hypothesis is as follows: ‒ Fist. and Haem.: null hypothesis is accepted. They are by the same author.

116

mikel labiano ‒ Off. and Mochl.: null hypothesis is accepted. They are the work of a unique author, probably the same as for Fist. and Haem., as in the comparison of both treatises we are inclined to accept, with high probability, the null hypothesis. ‒ Art. and Fract.: null hypothesis is accepted. They are by the same author. ‒ Vlc. and VC: null hypothesis is rejected. It is unlikely that the two texts are by the same author.

Now we will continue our inquiries in search of more possible relations between these treatises. Comparison between Art. and Fract. / the remaining treatises2 Fist. / Fract. (25 sentences by group) t student α

Coord. 0,8111 0,4207

Subord. 3,1553 0,0025

Asynd. –5,8451  *2

Haem. / Fract. (25 sentences by group) t student α

t student α

Subord. 2,2530 0,0265

Asynd. –6,1560  

t student α

Art. / Vlc. (50 sentences by group) t student α

Coord. 6,0998 0,0004

Subord. 1,9082 0,0151

Asynd. –9,7226  

t student α

Subord. 2,5766 0,0146

2 In blank when negligible.

Asynd. –7,7802  

Asynd. –7,1425  

Coord. 0,9908 0,3243

Subord. 2,9406 0,0041

Asynd. –6,6659  

Art. / VC (50 sentences by group) t student α

Fract. / Vlc. (50 sentences by group) Coord. 3,9380 0,0004

Subord. 3,6106 0,0006

Haem. / Art. (25 sentences by group)

Fist. / Art. (25 sentences by group) Coord. 1,3502 0,1801

Coord. 0,5642 0,5749

Coord. 3,0654 0,0034

Subord. –0,6844 0,4967

Asynd. –1,2363 0,2218

Fract. / VC (50 sentences by group) t student α

Coord. 0,8759 0,3874

Subord. 0,1679 0,8676

Asynd. –1,0844 0,2860



the surgical treatises of the corpus hippocraticum Art. / Mochl. (50 sentences by group)

t student α

Coord. –2,4606 0,0172

Subord. –2,5368 0,0142

Asynd. 9,3674  

Art. / Off. (50 sentences by group) t student α

Fract. / Mochl. (50 sentences by group) t student α

Coord. –1,2194 0,2318

Subord. –3,0541 0,0046

Asynd. 7,7131  

117

Coord. –2,0337 0,0476

Subord. –4,0725 0,0001

Asynd. 8,7155  

Fract. / Off. (50 sentences by group) Coord. t student –0,9949 α 0,3282

Subord. –4,2688 0,0002

Asynd. 7,1643  

We can safely reject the null hypothesis in the comparison of Art. and Fract. with Fist. and Haem., Off. and Mochl. and Vlc. This conclusion should not prevent us from seeing the very interesting results in the val­ ues of coordination, such as α 0.4207 between Fist. and Fract., α 0.5749 between Haem. and Fract., or α 0.2318 between Fract. and Mochl. and α 0.3282 between Fract. and Off. When considering VC, it would seem that Fract. and VC present values of α for coordinated and subordinated sentences (again asyndeton is not relevant to these treaties) that invite us to accept the hypothesis that both treatises have a common author, especially in the case of α 0,8676 for the subordinates. This value of α for the subordinates decreases to 0.4967 in comparison with Art. and lies far from any threshold of reliability in the case of coordinates, with α 0.0034. VC may not be the work of the author of Art. and Fract. but it can be placed chronologically in the same or close period of composition, nearer to Fract. than Art., assuming that between Fract. and Art. – although both are from the same author – there is a noticeable time lag. Comparison between VC and the remaining treatises (except for Art., Fract. and Vlc.) Fist. / VC (25 sentences by group) t student α

Coord. 0,4820 0,6338

Subord. –1,4156 0,1687

Asynd. 2,7989 0,0095

Haem. / VC (25 sentences by group) t student α

Coord. 0,4457 0,6599

Subord. –1,6845 0,1056

Asynd. 3,2781 0,0032

118

mikel labiano VC / Mochl. (50 sentences by group)

t student α

Coord. 0,3346 0,7420

Subord. –1,4452 0,16657

Asynd. 3,5093 0,0026

VC / Off. (50 sentences by group) t student α

Coord. 0,3413 0,7382

Subord. –1,9916 0,0678

Asynd. 3,3689 0,0050

First, we must discard the data offered by the asyndeton, because of its low presence in VC compared to its strong role in Fist., Haem. Off., and Mochl. The coordinates, in turn, offer tempting α values that could persuade us to accept the null hypothesis, especially for VC and the pair Off. and Mochl. With regard to the unity of authorship, this fact provides some problems in the light of conclusions that we have previously drawn with a high level of security, namely that the group formed by Fist., Haem., Off., and Mochl. is separated from the group formed by Art. and Fract., with which, on the other hand, VC seemed to show reasonable affinities. For these reasons, and taking into account the relevance of the variable of asyndeton in the treatises Fist., Haem., Off., and Mochl., we prefer to reject the null hypothesis. Comparison between Vlc. and the remaining treatises (except for Art., Fract. and VC) Fist. / Vlc. (25 sentences by group) Coord. t student –2,0369 α 0,0544

Subord. 0,7121 0,4842

Asynd. 0,8168 0,4231

Haem. / Vlc. (25 sentences by group) t student α

Vlc. / Mochl. (50 sentences by group) t student α

Coord. 1,9401 0,0714

Subord. –0,3367 0,7409

Asynd. –1,4327 0,1724

Coord. –1,7676 0,0940

Subord. 1,3951 0,1799

Asynd. 0,3845 0,7050

Vlc. / Off. (50 sentences by group) t student α

Coord. 2,1208 0,0574

Subord. –1,8205 0,0959

Asynd. –0,2367 0,8171

For coordinated sentences α values are in all cases above 0.05, the limit for the level of significance and acceptance of the null hypothesis already discussed. They are, however, too low, too close to the limit. On the other hand, the asyndeton, a variable that has to be taken into account in these five treatises since they follow a very similar pattern, gives a value of α 0.8171 for Vlc. and Off. Because of the adjusted values of the variable for coordinates, acceptable but with a very slim margin, it is more prudent to



the surgical treatises of the corpus hippocraticum

119

reject the null hypothesis for Vlc. in relation to the group of Fist., Haem., Off., and Mochl. However, it is possible to suggest that they have been written during the same or a close chronological period, as their language shows important affinities, a situation similar to that observed for VC with respect to Fract. and Art. Therefore, the situation regarding the Student’s t-test and acceptance or rejection of the null hypothesis is as follows: Null hypothesis is accepted: ‒ Fist. and Haem.: They are by the same author. ‒ Off. and Mochl.: They are the work of a unique author, probably the same as for Fist. and Haem., as in the comparison of both treatises we are inclined to accept with high probability the null hypothesis. ‒ Art. and Fract.: They are by the same author.

Null hypothesis is rejected: ‒ Vlc. and VC: It is unlikely that the two texts are by the same author. ‒ Vlc. and treatises Fist., Haem., Off., and Mochl. ‒ Vlc. and the pair Art.-Fract. ‒ The group Fist.-Haem.-Off.-Mochl. and the pair Art.-Fract. ‒ VC and the pair Art.-Fract. (although for Fract. the null hypothesis could be accepted).

4. (Pearson) Correlation Coefficient In addition to the Student’s t-test and the results and conclusions obtained, it will be interesting to perform an additional statistical test. This allows us to compare and contrast the results obtained, in order to interpret them more accurately. In short, first we select the ten most representative syntactic links. These are then arranged by number of appearances in each text and assigned a numeric value. The most frequent link is assigned a rank of 1, the next one a rank of 2, and so on until 10 (Dixon & Massey 1974: 290). Ranks/ ranges of numbers of each text so obtained are compared with each other in pairs, and, using the Pearson correlation coefficient (Muller 1973: 210), a value ranging from –1 to +1 are obtained with the following meaning: +1 signifies that the two classifications of links are identical, –1 signifies that they are exactly the opposite, and 0 means that there is no relation between them, or weaker dependence (Muller 1973: 203). The correla­ tion coefficient provides basic information about a given state of language

120

mikel labiano

in a specific chronological period, in which some specific links are pre­ ferred over others. The probability of reaching or exceeding, just by chance, a coefficient of 0.76 is 0.01 in the present case (with N = 10). This probability is 0.02 with a coefficient of 0.71, 0.05 with 0.63, and 0.10 with 0.54. In these cases it is better to accept the null hypothesis of perfect independence of the two sets of ranks. Therefore a coefficient of 0.76 will be the reference that we will consider. Fist. / Vlc. 0,8545

Fist. / VC 0,5636

Fist. / Fract. 0,3333

Fist. / Art. 0,3818

Fist. / Off. 0,4060

Haem. / Vlc. 0,8787

Haem. / VC Haem. / Fract. Haem. / Art. Haem. / Off. 0,6 0,4303 0,4181 0,5636

Fist. / Mochl. 0,6121 Haem. / Mochl. 0,7333

Fist. / Haem. 0,939393939

Art. / Fract. 0,9030

Art. / Vlc. 0,3454

Art. / VC 0,7212

Vlc. / VC 0,6848

Off. / Mochl. 0,6121

Fract. / Vlc. 0,4303

Fract. / VC 0,8545

Off. / Fist. 0,4060

Off. / Haem. 0,5636

Off. / Vlc. 0,6969

Off. / VC 0,3818

Off. / Fract. 0,2848

Off. / Art. 0,2242

Mochl. / Vlc. 0,7818

Mochl. / VC Mochl. / Fract. Mochl. / Art. Mochl. / Fist. Mochl. / Haem. 0,6848 0,7212 0,6121 0,6121 0,7333

The coefficient for the pair Fist.-Haem. is 0.9393 and for Art.-Fract. is 0.9030, a value very close to 1 (which would mean the two ranks of links are perfectly identical), much higher than 0.76. This fully confirms the conclusions obtained by the Student’s t-test. This test also confirms the clear separation between Fist.-Haem. and Fract.-Art. – Fist. / Fract. 0.3333; Fist. / Art. 0.3818; Haem. / Fract. 0.4303; Haem. / Art. 0.4181. The coefficient for the pair Off.-Mochl. is 0.6848. Despite the high level of confidence obtained with the Student’s t-test, this coefficient is not as convincing as the coefficient of the pairs Fist.-Haem. and Fract.-Art., but even so we think the hypothesis of a common author can be maintained. The cause may be the way in which both texts have been formed, and their special and brachylogic syntax. The correlation coefficient again confirms, as with the Student’s t-test, the clear separation between Off.Mochl. and Fract.-Art., especially with the latter – Mochl. / Fract. 0.721; Mochl. / Art. 0.6121; Off. / Fract. 0.2848; Off. / Art. 0.2242. The correlation coefficient can now help us to refine some issues previ­ ously analyzed with the Student’s t-test, especially Vlc. relations with the



the surgical treatises of the corpus hippocraticum

121

group Fist.-Haem., Off.-Mochl, on the one hand, and VC relations with this same group, and the group of Fract.-Art., on the other. First, we analyze the relations of Vlc. with the group Fist.-Haem., Off.Mochl. These are the coefficients: Vlc. / Fist. 0.8545; Vlc. / Haem. 0.8787; Vlc. / Mochl. 0. 7818; Vlc. / Off. 0.6969. In general, t values are very high, except for the latter. However, as the values were weak for coordinates, we still think it is more reasonable to postulate different authors. The strong dependency between them could be explained, as mentioned earlier, by proximity in time and proximity of their authors. The higher affinity is visible between Vlc., Haem. and Mochl., just as Vlc. showed more rela­ tive affinity with Fract than with Art. Shortly we come back to this issue in order to make an attempt at a possible relative chronology of these treatises. With respect to VC, these are the coefficients: VC / Art. 0.7212; VC / Fract. 0.8545; VC / Fist. 0.5636; VC / Haem. 0.6; VC / Mochl. 0.6848; VC / Off. 0.3818. With these coefficients for the group of Fist.-Haem. and Off.Mochl, as we concluded with the Student’s t-test, it is more reasonable again to postulate different authors for VC and this set of treatises. With regard to Art. and particularly Fract., the coefficients invite us to agree with the traditional assumption of a single author for these three treatises. α values for VC and Fract. gave a very convincing argument towards this thesis, but α values for VC and Art. in coordinates (0.0034) are clearly contrary to these expectations. Moreover, the author of Fract.-Art. uses a much larger set of coordination procedures than any other treatise, as shown by the use of the particles ἀτάρ, τοίνυν, καίτοι, μέντοι, μήν, compared to its limited use in other surgical treatises. This is consistent with the idea that VC has abundant rhetorical features but fewer artistic features than those of Fract.-Art., which show in many ways a greater degree of formal elaboration. These facts would impact the behavior of coordinated sentences, which is exactly what happens. The null hypothesis cannot be accepted on the basis of these statistical observations. We hold, there­ fore, the conclusions previously developed. We reject the null hypothesis between VC and the pair formed by Art. and Fract., although their lan­ guage is not very distant chronologically. That is, VC must belong to the same or close period as Fract. The correlation coefficients have led in practice to the same conclu­ sions as the Student’s t-test results, and have helped to suitably refine some details.

122

mikel labiano 5. Comments on a Possible Relative Chronology of the Surgical Treatises of CH

The conclusions drawn from the Student’s t-test, and, especially, from the correlation coefficients, have shown a series of marked relations (direct relations of authorship, or, at least, of chronological proximity or remote­ ness) between treatises. It has also already been noted that the correlation coefficient provides basic information about a given state of language in a specific chronological period, in which some specific links are preferred over others. A simple but eloquent example: the preference of δέ for καί, or vice versa, as a link of coordination, is sufficiently expressive. From this data, we propose the following suggestions regarding a possible relative chronology of these treatises. We believe that Art. is the oldest treatise in the group of Art., Fract., and VC. Although Art.-Fract. and VC do not share the same author, the correlation coefficients, which tend to be constant in each period accord­ ing to our working hypothesis, systematically separate Art. from the other treatises. Just the opposite is the case with Fract. and VC. Moreover, the latter showed some relations with Fist. and Haem., treatises of the fourth century bc (Labiano 2002 and 2003). The correlation coefficients for the group of Art., Fract., VC and for Aër. and Prorrh. II (data from Sierra de Grado 2003), which allows us to check what information can be obtained from this test and to compare the degree of proximity or remoteness of the state of language in these treatises, are as follows: Aër. / Art. 0,8079

Aër. / Fract. 0,6440

Aër. / VC 0,6908

Prorh. II / Art. 0,7333

Prorrh. II / Fract. 0,8060

Prorrh. II / VC 0,9151

These coefficients prove that Art. is closer to Aër., probably one of the oldest treatises in CH, than to Fract. and VC. A reasonable cause for this proximity is precisely proximity in time, a state of language of the same or close chronological period. On the other hand, according to conclusions from another study (Labi­ ano 2002 and 2003), Haem. is moderately more recent than Fist. It is also likely that Mochl. is more recent than Off. Both treatises, Mochl. and Off., are for their part closer to Haem. than Fist. And from the correlation



the surgical treatises of the corpus hippocraticum

123

coefficients, Vlc. is also closer to Haem. than Fist. So the chronological sequence might be as follows: Fist., Off., Mochl., Vlc., and Haem. This is merely a conjecture based on the statistical analysis. From this starting point, a deep philological and linguistic study would be required in order to go further. 6. Conclusions In some cases we have accepted the null hypothesis and in other cases we have rejected it. Thus we have obtained some clear ideas about some of the relations of the so-called surgical treatises of CH, at least in the light of statistical linguistics. Vlc. is clearly separated from VC and from the pair Art.-Fract., both from the same author. For its part, Vlc. is also separated from the group Fist., Haem., Off., and Mochl., which are surely from the same author, but is placed around the same chronological period. We reach the same conclu­ sion about VC and the pair Art.-Fract.; they are not from the same author but from the same or close period. Although not from the same author, VC and the group Fist., Haem. Off. and Mochl. exhibit some similarities in coordination, as does the group Fist.-Haem. and Fract. On the other hand, the syntactic pattern of asyndeton in Fist., Haem. Off., Mochl., and Vlc. is very striking and characteristic, and reflects undoubtedly a purposeful care and a deliberate rhetorical style. Considering these conclusions, despite the differences, there is not much serious disparity in the language of these treatises. Therefore, it seems plausible to suggest, without too much risk, that there are no big chronological leaps – at least important or relevant leaps – between one treatise and the others. Art., the treatise with the appearance of being the oldest of the group of surgical treatises, exhibits a close affinity, despite clear differences in style, with Aër., a work of the last third of the fifth century bc, probably among the older treatises of the CH. On the other hand, Haem. is placed in the second half of the fourth century bc (Labiano 2002 and 2003). Therefore it is highly probable that, except for Art., the remaining surgical treatises of the CH, possibly including VC and Fract., are placed in the first half of the fourth century bc. But that is another story. Statistical linguistics does not go beyond this at the moment. For our part we hope to have fulfilled the aims we set at the beginning.

124

mikel labiano References

Ángel, Jesús (2003), Comentario sintáctico-estilístico de Epidemias V y VII. Madrid: E-Prints Complutense. Bourgey, Louis (1953), Observation et expérience ches les médecins de la Collection Hippocratique, Paris: Vrin. Dixon, Wilfrid Joseph and Frank J. Massey (1966), Introducción al análisis estadístico, sec­ ond edition, Madrid (trad. española). Duminil, Marie-Paule (1988), Hippocrate. Plaies, Nature des os, Coeur, Anatomie, Paris. Jouanna, Jacques (1992), Hippocrate, Paris. Labiano, Mikel (2002), “Aspectos fonéticos y morfológicos de dos tratados del Corpus Hippocraticum,” CFC (G) 12: 9–51. —— (2003), “Aproximación a la lengua de los tratados hipocráticos De fistulis y De haemorrhoidibus,” CFC (G) 13: 5–40. —— (2004), “Los tratados quirúrgicos del Corpus Hippocratic. Aplicación de análisis de estadística lingüística,” CFC (G) 14: 91–109. Laín, Pedro (1970), La medicina hipocrática, Madrid. Muller, Charles (1973), Estadística lingüística, Madrid. Regenbogen, Otto (1914), Symbola Hippocratea, Berlin. Rodríguez Alfageme, Ignacio (1992), “Sobre la fecha de Hipp. De glandulis,” Epos 8: 549– 566. —— (1993), “La atribución de Hipp. De visu,” CFC (G) 3: 57–65. —— (2000), “La medicina hipocrática: formación del Corpus Hippocraticum y su proyec­ ción,” La lengua científica griega: orígenes, desarrollo e influencia en las lenguas modernas europeas, Juan Antonio López Férez (ed.), Madrid: 173–182. —— (2010), “De locis in homine: intento de datación,” CFC (G) 20: 25–44. Sierra de Grado, Cristina (2003), Rasgos de estilo y recursos de composición en algunos tratados del Corpus Hippocraticum. Madrid: E-Prints Complutense. Withington, Edward Theodore (1928), Hippocrates, vol. 3, Cambridge (MA).

True Plautus, False Plautus Pellio restitutus – uxor excisa. Annotations to Plautus’ Bacchides Klaus Lennartz The Bacchides of T. Maccius Plautus has become a prime focus of Plautine studies since Handley’s inaugural lecture Menander and Plautus: A Study in Comparison (London 1968), as is widely known.1 Here, if anywhere, we are given the precious opportunity of direct comparison of a provable original (Menander’s Dis Exapatôn, the papyrus that enables a wealth of possible interesting observations, finally published completely as P. Oxy. 4407 in 1997) with its provable translation / remake.2 A milestone of twentieth-century Plautus philology – and so we come to the subject matter of our anthology – is of course Zwierlein’s Höhere Kritik (Zwierlein 1990ff.), whose scrupulous examinations of single verses and passages from our author, in my opinion, cannot be ignored by any serious discussion of the Sarsinate poet.3 In what follows, first I will attempt to reclaim from Zwierlein’s retractator a passage from Bacchides (Bacch. 208–217) that has some general significance concerning the pragmatics of Plautine theatre (I: “Pellio restitutus”), and at the same time explain two verses somewhat more closely than has hitherto been done (Bacch. 751f.); secondly I intend to go beyond even Zwierlein, proposing the excision of Bacch. 1009. In so doing, I hope to establish a more powerful argument in favour of Zwierlein’s negative verdict of Bacch. 1091–1100 (II: “uxor excisa”). First, however, the synopsis of the Bacchides, which cannot be easily related in brief but nevertheless must be provided as an essential

1  A fresh bibliography in Suerbaum (ed.) 2002: 186ff., esp. 192ff.; compare Questa 2008 (with lit.) with Leo’s and Lindsay’s classic editions (1895 / 1904). 2 See Lennartz 1994: passim, esp. 77–83. 3 Instead of scrutinizing the details, too often, in my opinion, one reads methodologi­ cal commonplaces concerning “il Plauto di Zwierlein,” etc. (see, e.g., Guastella 2002: 167ff.; also, e.g., Blänsdorf 2000: passim). Questa 2008 unfortunately does not make note of Zwi­ erlein’s decisions.

126

klaus lennartz

background to the following observations, is given and abridged where necessary.4 Pistoclerus has, according to a letter from his friend Mnesilochus, found in Athens the meretrix Bacchis of Samos, who has this very day sought refuge in Athens with her sister. This sister permanently lives in the city, works as meretrix, and is also called Bacchis. The friend of Pistoclerus, Mnesilochus, had met the Samian Bacchis during a two-year trading residence on Samos in Asia Minor: the Samian Bacchis returned his love, but was – and still is – under annual contract with a soldier, Cleomachus. Getting tired of him, the Samian Bacchis had left Samos, not unnoticed by Cleomachus, who followed her and is willing to take her immediately to Elis. Alternatively, he is ready to dissolve the annual contract for an advance sum of 200 Philippi (Macedonian gold coins). Unsurprisingly, Pistoclerus – quite to the dismay of his moralizing pedagogue Lydus – falls in turn to the power of the Athenian Bacchis. As Mnesilochus and his servus callidus Chrysalus reach Athens, the question is: where to get the 200 Philippi? Chrysalus is quick to come up with a plan: he leads the father of Mnesilochus, Nicobulus, to believe that the Ephesian host, to whom Nicobulus had left 1200 Philippi to make further investments, has deceived him and seized the money. At the last minute they had deposited the principal sum at Ephesus under the title of a local man, Theotimus, a priest of the Artemisium;5 Mnesilochus brought only a small portion back with him to Athens. Now, just before Nicobulus is expected to travel to Ephesus, a serious misunderstanding ensues: Mnesilochus overhears the words with which the desperate Lydus informs Pistoclerus’ father, Philoxenus, of his son’s newly evil lifestyle with a Bacchis. Mnesilochus, thinking he means the Samian Bacchis and not realising that there are actually two meretrices of the same name, is so angry at the faithless conduct of his friend that he transfers the entire Ephesian sum to his father Nicobulus, and prays, not without difficulty, for impunity for Chrysalus.6 Thus Mnesilochus is left where he began.

4 A “book of Plautine plots” might be welcome as a very useful philological tool. For instance, see Lowe 2000. 5 For v. 306ff. see Lennartz 1994: 130ff. 6 The P. Oxy. contains the exact scene of the misunderstanding of Mnesilochus, the returning of the money, the begging for Chrysalus (off stage) and the reconciliation with Pistoclerus.



true plautus, false plautus

127

A second plot is necessary, and Chrysalus is not slow to devise a new illusion: it succeeds, with the use of a Bellerophon-letter, filled with awful warnings against this ever-cunning slave, which he dictated to Mnesilochus with the aim of winning important ground against the senex. When Nicobulus reads the alleged words of his son, he is extremely angry, and wants to tie up Chrysalus according to the letter’s instruction. Chrysalus, however, pushes the action on stage so far that he is able to lead Nicobulus to the inner room of the house of Bacchis. The father has the opportunity to see, with his own eyes, the disorderly behaviour of his supposedly worthy son in the disreputable house of Bacchis. When the relationship of trust with his son is shaken, Chrysalus, in turn, becomes trustworthy again. At the same time Chrysalus skilfully insinuates to Nicobulus that the relationship of his son with the Samian Bacchis is not just a minor sexual relationship with a meretrix: Cleomachus the soldier, with staged precision, appears at the same moment, not just demanding the Samian Bacchis for his journey to Elis, but uttering brutal threats against Mnesilochus. Now, finally, Nicobulus must believe the Samian Bacchis to be the uxor of the soldier, and his son to be in danger of being killed by the soldier for adultery. Chrysalus brings about a swift agreement between the senex and Cleomachus; the soldier is pleased to receive his remaining 200 Philippi, and Nicobulus imagines he has saved his son from death. Because he will not enter into the house of Bacchis to meet his son, he wants Chrysalus to wash his head vigorously. As Chrysalus comes out of the house again, he not only delivers a song of triumph – it is the moment of the famous Troy canticum 925ff. – but presents Nicobulus with a second (dictated) letter from Mnesilochus, which will help to extract a further 200 gold coins as “entertainment money” from the senex. Mnesilochus promised – and so he wrote to his father, allegedly not daring to appear before his eyes – to give the “woman” of the soldier 200 gold coins for an unspecified purpose. Nicobulus, who cannot tolerate the idea of his son being shamed and maybe even prosecuted for permitting perjury, agrees once again. Later, when Nicobulus has completed the transaction with the soldier and the money has been delivered, he learns that he has been deceived in the civil and private law status of the alleged uxor. However, before he can break into the house of the Bacchides, angry at his own stupidity, and with the intent of punishing Mnesilochus and his cunning slave, he and Philoxenus, the more compromising father of Pistoclerus, are seduced by the Athenian Bacchis and the Samian Bacchis into attending a general reconciliation banquet loaded with aphrodisiacs. Even the

128

klaus lennartz

strict Nicobulus decides to accept this invitation, especially when Bacchis promises the return of the second 200 gold coins. I First let us inspect Plaut. Bacch. 205ff.: 211. Chrysalus stops the “message” from Pistoclerus concerning the love of the Samian Bacchis for Mnesilochus with immo hercle abiero, and – after a reassuring question from Pistoclerus – slips into an extra-fictional aside (212ff. non res, sed actor mihi cor odio sauciat. / etiam Epidicum, quam ego fabulam aeque ac me ipsum amo, / nullam aeque invitus specto, si agit Pellio); then, via a mythological joke (216), Chrysalus renews the dialogue, returning to the theme of the loving Samian Bacchis (218). The whole passage 208–217 has been excised as non-Plautine by Zwierlein 1992: 199ff. Zwierlein claims to point to a number of non-Plautine linguistic phenomena and criticizes the inhibition of plot development by these verses; the famous “Pellio-sentence,” in particular, is regarded as a break in dramatic illusion not consistent with the original author, combined moreover with a sort of onomasti kômôidein not to be expected at all on the Roman stage. Well, let us look at the details of this magisterial indictment: a) Zwierlein complained that the retractator, interpolating 208 PI. immo ut eam credis? misera amans desiderat to 217, has overlooked that 207 immo unice unum plurumi pendit “nicht mehr zu steigern war” (p. 202). It seems to me, however, that Zwierlein has missed that in desiderat 208 there is in fact a new direction of thought, and with 209 non tantulum / usquam intermittit tempus, quin eum nominet a valuable reference is made to the beginning of the comedy (transmitted only in fragments but present, of course, in the minds of the actual audience), where the character of the Samian Bacchis had been depicted precisely along these lines: for, obviously and much to the chagrin of her sister, she cannot confess her desire for her lover enough, cf. Bacch. fr. xii cor meum, spes mea, mel meum, suavitudo, cibus, gaudium and the admonition of the sister Bacchis 35 quid si hoc potis est ut tu taceas, ego loquar? SO. lepide, licet. So, it seems to me, Zwierlein unjustly censures that “was er (i.e., the retractator) hinzuträgt . . ., fällt gegenüber der lapidaren Formulierung des echten Plautus stark ab” (ibid.). b) The thought expressed in these verses and the “capping” with immo followed by this type of remark is stylistically paralleled in Curcul. 57ff. and can be recognized as an original Plautine stylistic element.



true plautus, false plautus

129

c) The announcement of abire by one of the interlocutors in Plautine dialogue is surely – as Zwierlein readily admits – not an oddity (Zwierlein 1992: 202 A. 453).7 d) What Zwierlein 1992: 203 names as non-Plautine verbal usages, in my opinion, do not amount to veritable offences: “tantulus ist dem echten Plautus fremd, der durchweg . . . das Deminutivum tantillus verwendet.” Negative tantillus can be counted three times in Plautus, but these instances do not happen to be in conjunction with non, as tantulus is here. I surmise that with non tantillus a heavy molossus would be heard that is not desirable in this semantic context. However, the possibility for comparison is altogether too small to allow höhere Kritik. “tantulus scheint frühestens rhet. Her. 4, 13 und dann häufiger bei Cicero belegt”; yes, but it is equally true (even though Zwierlein does not acknowledge it) that tantulus is found frequently in Apuleius and Gellius, who might have culled it from their beloved authors. And last but not least, when Plautus wants to use a quantifying expression such as this at the end of the senarius, he should use a cretic equivalent to tantillus! Finally, against intermittere, we read the somewhat typical statement (Zwierlein 1992: 203): “Auch intermittere kennt der echte Plautus nicht” – with note 454: “Most. 959 . . . ist unecht”: quod erit demonstrandum. Now, one could count this, like so many classic philological zetemata, among the questions described by the Kallikles of Plato Gorg. 485d6ff., were it not that Zwierlein, together with Bacch. 208–217, also deletes the famous “Pellio-locus,” and thus the veritable lifetime achievements of an ancient man of the theatre8 have fallen victim to him, so that this man runs the risk of being reduced from impresario of a genius to one of the people qui utuntur vino vetere. Should we not help him eagerly? Zwierlein, as we have seen, holds that the break of dramatic illusion displayed in our locus is impossible, even given that Barsby 1986: 115 has indicated a number of verses of a somewhat similar character; Men. 404, especially, deserves attention, in which an apparent reference is made to the name of Pellio via an amphibolic expression supellex pellionis (“tanner”): Erotium: perii misera! quam tu mihi nunc navem narras? Men. ligneam, saepe tritam, saepe fixam, saepe excussam malleo; QVASISVPELLEXPELLIONIS, palus palo proxumust.

7 Poen. 922 we find even the motive of odium under this circumstance. 8 For T. Publilius Pellio see Hanslik 1959.

130

klaus lennartz

Here Menaechmus II is speaking, who holds a somewhat fuller role than Menaechmus I. The offence against Pellio from the mouth of the main character returns in our passage; we shall see that there is a certain connection between this man and props (supellex), also, in the Bacchides.9 Our passage, Zwierlein argues, must be dated later than Plautus because of onomasti kômôidein, not thinkable ca. 200 bc on the stage of the original author; rather these come from a time terminus post quem around 160, when literary feuds – albeit in prologue and here still without onomasti kômôidein! – began to flourish on the stage. Zwierlein notes the wellknown onomasti of the poet Accius by a mime (rhet. Her. 1, 24 mimus quidam nominatim Accium poetam compellavit in scaena) and the subsequent trial (Accius obtaining his case).10 Precisely in the time of Accius, Zwierlein sees his retractator at work: the prologues of Terence showed “daß eine abschätzige Bemerkung über einen Konkurrenten, wie sie in Bacch. 214 gemacht wird, unvorstellbar innerhalb des Stückes selbst ist, sondern daß sie ein mit Stilempfinden ausgestatteter Dichter allenfalls vor die eigentliche Komödienhandlung . . . plazieren konnte. Ein späterer impresario setzte natürlich andere Schwerpunkte; wir ertappen ihn auf Schritt und Tritt bei Zutaten, die Zielstrebigkeit der Handlung und Einheitlichkeit der Charakterzeichnung zerstören, wenn er glaubt, den Bedürfnissen seines (zumindest eine Generation später lebenden) Publikums entsprechend modische Späße und Anspielungen einflechten zu können.” Now, T. Publilius Pellio was always viewed as the main impresario of Plautus (so to say, Plautus’ Ambivius Turpio). In our Bacchides-locus he is placed in context with Epidicus-performances; he should be adumbrated in the above Menaechmi-locus, and the full form of his name proudly figures in the single, almost completely preserved didascalic notice to Plautus, on the Stichus. Does it? Zwierlein holds (Zwierlein 1992: 207ff.) – the scholar is well aware of the implications and he fears that only a few will follow him in this! – that Pellio in the didascalia on Stichus11 is mentioned only as a second name, thus to be regarded as director of revivals, just like Atilius Praenestinus in the didascalic notices to Terence.12 Zwierlein refers to the didascaliae of

9 Further supellex-references in Plautus: Poen. 597 the fake money of Collybiscus, 1145 parce muliebri supellectili. 10 Note Zwierlein himself 1990: 180: “Der Mimus aber gehört einem anderen Genos an.” 11  Cod. Ambrosianus fol. 681 verso: See Studemund 1889, Leo 1896: 2, 367. 12 Cp. still Jachmann 1934: esp. 601ff. (“bequemer Überblick”: Marti 1963: 16); important groundwork by Dziatzko 1865 on the shoulders of Ritschl.



true plautus, false plautus

131

Terence, whose rows he arranges “nach dem Muster der Stichusdidaskalie” (Zwierlein 1992: 208 n. 470); accordingly in l.5 of the Stichus-didascalia due to lack of rubrication he misses something that looks like , but to me it seems too convenient to assume that two names have been recorded just as in the case of Terence. It seems to me that Zwierlein pays too little attention to the centering mise-au-page of the didascalia in the Stichus Ambrosianus page: the copyist of A structures the page – taking over from the original or self-employed – semantically, so that what belongs together stays together. This can be seen in l. 8 St. = 4 Leo and l. 15 St. = 11 Leo, where he set two names in a category into a single row, whereas, as in l. 12 St. = 8 Leo, where he had one proper name in front of him, he at least placed it solitary and centred; the predicate of the sentence (suppl. Ritschl) he put accordingly into a separate row 11 St. = 7 Leo, not due to lack of space, but for semantic reasons (as can be shown in l. 6/8/15 St. = 2/4/11 Leo, he would have, were he not working on semantic grouping, put MARCIPOR OPPII easily in one line).13 It therefore seems more obvious to me that in l. 9 St. = 5 Leo we must not suppose that one verbum finitum + nomen proprium has appeared, followed in l. 10 St. = 6 Leo by another nomen proprium PVBLILIVS PELLIO. Rather, the presentation of A offered only a single proper name, which was grouped by A semantically corresponding to the division of 11/12 St. = 7/8 Leo / MARCIPOR OPPII: Ritschl, therefore, in l. 9 St. = 5 Leo probably correctly added . Zwierlein is even willing to admit this, but “selbst wenn in Zeile 5 der Stichusdidaskalie nur – wie Ritschl annahm – EGIT stand, müßten wir in Analogie zur Überlieferung der Terenzdidaskalien immer noch damit rechnen, daß in die Stichusdidaskalie nur der ,Impresario‘ der späteren Wiederaufführung Eingang gefunden hat, nicht jener dominus gregis, der die Uraufführung leitete” (Zwierlein 1992: 211). Now the Bembinus A in the Heauton-didascalia still has the later, second impresario “suppressed”; however, he moves correctly (save the numerus in EGIT, probably from incorrect resolution of EG.) in the didascalia of Eunvchus: EGIT L. AMBIVIVS TVRPIO L. HATILIVS PRAENESTINVS. The relevant part of the Bembinus Phormio-didascalia is not preserved, so is also eliminated as evidence; in the Adelphoi the Bembinus shows its better quality in that it has retained the name of L.

13 The two lines of the Pseudolus-didascalia are not helpful; at best, they show the same tendency, not the principle named by Zwierlein (Zwierlein 1992: 211 A. 476 dubitanter), “durch Abkürzungen die Zeilen mit möglichst viel Informationen zu füllen.”

132

klaus lennartz

Ambivius Turpio against the recensio Calliopea, even if he swapped – only here! – the original order of the first and secondary impresario, that is still identified with Donat. But with this play, and only this one, there must have been some confusion in proper names, such as the recensio Calliopea shows by embezzling Ambivius Turpio, who, as Donat shows, belonged to the original didascalia, introducing in turn, next to Atilius Praenestinus, the name of another impresario Minutius Prothymus, who according to Wilmann’s conjecture was present also in the tradition of Donat. So in Adelphoi we find a “pool” of these three impresarios, obviously a special case. Overall, we can say that the didascalic tradition of Terence, except in the recensio Calliopea concerning the Adelphoi, which represents a special case, retains the original name of the actor / impresario Ambivius Turpio, and the principle “wir sind also frei, in Publilius Pellio nicht den ‘Impresario’ des Plautus, sondern einen der Regisseure zu sehen, die bei den späteren Wiederaufführungen tätig wurden” (Zwierlein 1992: 211), emerges as less convincing. For if we accept the latter fact, we must either accept in l. 9 St. = 5 Leo and waive the conscious design of the Stichus-didascalia, or accept, in positing with Ritschl and Leo, that in the case of Stichus there occurred something that in the Bembinus of Terence is never the case, and in the recensio Calliopea happens to take place only within the confusion concerning the Adelphoi: that the name of the original impresario would no longer exist, and the name of the secondary impresario would be introduced, on the other hand, as if he were the only acting director of this piece, displacing the original first director. The fact that the didascalia would take on merely a reenactment reference is indeed hardly possible due to the correct date for consuls in the year 200 – which is the year of the premiere! In addition, in my opinion the testimonium for Pellio in Symm. epist. 10, 2 (written to the Emperor Gratian [367–383]), is still valuable, a testimony that, I think, is no autoschediasma, but reflects an opinio recepta: magnis enim negotiis itidem ut magnis comoediis edecumati adponuntur actores: non idem honor in pronuntiandis fabulis Publilio Pellioni qui Ambivio (sc. Turpioni) fuit, neque par Aesopo ut Roscio fama processit.

Here, Publilius Pellio is compared unfavourably with Ambivius Turpio, the impresario and actor of Caecilius and Terence, and the assessment goes back, I think, indirectly, to a comment on our Bacchides-locus, where Pellio indeed comes off badly (concerning Aesopus there is a note by Cicero epist. 7, 1, 2, that he failed to convince at the opening of the Pompeian



true plautus, false plautus

133

theatre 55: this should be, in his case, the indirect source). But it is vitally important that Pellio, in this tradition of rank with Ambivius Turpio, seems quite on a par. Such a ranking could hardly have developed from the one Stichus-didascalia (as we can see with Donat, there circulated ample anecdotal material on Ambivius). The fact that a “secondary impresario,” Pellio, in the ancient perception of Plautus, would have appeared regularly and mistakenly as the only or first impresario, is, in view of the findings of the didascalic material on Terence unlikely in my opinion. This seems to me Publilius Pellio put back in the picture. The relevant passage viewed in this way should be considered “original” Plautus, which means that the linguistically-objective offences Zwierlein names emerge as valuable observations of style. But could Pellio have been mentioned in this way on the Bacchidesstage with an onomasti kômôidein? Must we truly draw the same conclusions that Leo 1913: 94f. draws from our passage? Leo says: . . . daß der Schauspieler Pellio den Epidicus für die Aufführung erworben hatte, wie den Stichus, für den wir es durch die Didaskalie erfahren, . . . daß das gute Geschäftsverhältnis, das früher zwischen Plautus und Pellio bestand, gestört worden und zur Feindschaft geworden ist . . .

Concerning the point of dispute and alienation, I cannot agree because if we study the testimonies of ridicule with Zwierlein 1992: 207 n. 468 – and add the locus classicus Cic. rep. 4, 12 Ziegler veteribus displicuisse Romanis laudari quemquam vel vituperari in scaena vivum hominem –, we may safely assume that if the remark was to be directed against Publilius ­Pellio, then on the Roman stage it would have been pronounced only by one person: Pellio himself. Zwierlein also acknowledges this, p. 211, in passing (“falls es sich . . . um Selbstironie handelt”), which of course ignores that onomasti kômôidein was precisely the point of his scepticism before. Selbstironie (Fraenkel 1960: 431) today is the mostly accepted interpretation on our passage: the actor played Chrysalus, a star role, which was designed just for that reason by “his” strong playwright Plautus. He could not miss this opportunity to provoke laughter para prosdokian with his own name. Fraenkel 1922: 250 calls it a “probability” that Pellio played Chrysalus, and so do others. I think that the fact can even be proved by referring to another locus of Bacchides that only fully unfolds its meaning if this is the case. At the beginning of the second plot, the double attack via the dictated letter, Chrysalus commands Pistoclerus to bring some props to the stage (748ff.):

134

klaus lennartz cedo tu ceram ac linum actutum, age obliga, opsigna cito. MN. opsecro, quid istis ad istunc usust conscriptis modum, ut tibi ne quid credat atque ut vinctum te adservet domi? CH. quia mi ita lubet. potin ut cures te atque ut ne parcas mihi? mea fiducia opus conduxi et meo periclo rem gero. MN. aeque dicis. CH. cedo tabellas. MN. accipe. CH. animum advortite.

An analogous translation of 752 (e.g., “I was relying on myself when I took on the job and I’m conducting my business at my own risk”; De Melo 2011 and most other translators; I do not find the locus to be commented on in detail), in my view, obscures the punctum saliens: here, from the mouth of the Chrysalus mask, the impresario himself says that, as is common with the contemporary stage conditions, he rented the supellex from a choragus.14 The fact that this loan relationship works here on the trivialities of cera, linum and tabellae, makes the joke about the liability for the items quite funny. II As I want to give back, in a manner, Plaut. Bacch. 208-(at least) 215 to the Sarsinate poet, I’m willing to damn on my part another verse, which Zwierlein “spared” and (unfortunately) no one has yet incriminated. We had reached the attack, via the double letter of Chrysalus and Mnesilochus, against Nicobulus. I fully agree with Zwierlein 1992, arguing against Lefèvre 1978 and Primmer 1984, in believing that both parts of the letter together constitute the second attack, that is, the second deception of Menander’s Dis Exapatôn. Adolf Primmer, whose theories regarding the broadest Plautus adaptation of the play will be ignored, in his Vienna Academy paper about the plot outline of Dis Exapatôn and Bacchides (Primmer 1984), directed attention, with complete justice, to the scene 4, 4, the very hinge between the first and second Apatê. This scene is obviously – as we have known since the release of the entire Papyrus Oxy. 4407, especially col. iii l. 22 = v. 84 – written on the back of the Greek model. As Mnesilochus stresses against Chrysalus how much he is now a liar in his father’s eyes, so much so that he would even believe the daylight to be night if Chrysalus should show him the sun and call it

14 See loci and lit. in Woytek 1982: 215, commenting on Plaut. Pers. 159; especially see Trin. 858 ipse ornamenta a chorago haec sumpsit suo periculo: the impresario played the role of Megaronides, see, e.g., Duckworth 1952: 74.



true plautus, false plautus

135

day, the slave decides to take the fight to the senex a second time (701): emungam hercle hominem probe hodie, ne id nequiquam dixerit. And so he tells the astonished Mnesilochus and Pistoclerus to do exactly as they wish, and to dedicate themselves to their beloved Bacchides in sight of the front door during a vivid convivium. He needs – we have seen this above – only table, wax, and tape for a letter. After the first letter is written on the stage, Mnesilochus and Pistoclerus disappear from the rest of the scene. In using this first letter, which accused him of inciting a second Apatê and faking his own distress, making his allusions and comments interesting (the scene is Menandrian, as is proved by the famous words 816f. quem di diligunt, / adulescens moritur, dum valet, sentit, sapit = Men. dis exap. fr. 4 Sandbach), Chrysalus alone will succeed in finding a way to move the senex Nicobulus to inspect the dissolute life of his son. And so, in the scene (835–841) Chrysalus clearly insinuates that the liaison of Mnesilochus with the Samian Bacchis is not that of an ordinary adulescens with his meretrix, as the senex first – as is only right and proper – must accept. To be sure, Chrysalus does not express himself decisively (841 ex me quidem hodie numquam fies certior). The latter is central to the strategy behind the deception: the senex should be fooled by seeing nothing but the facts. For in fact, the Samian Bacchis is currently no meretrix in the true sense; she is bound to a strict “temporary marriage contract,” as the audience, through the citation of the relevant contract in the very first scene of the Bacchides, fr. x, heard in more detail. The contents of such a contract are known from a similar annual contract between the iuvenis Diabolus and the alleged meretrix Philaenium, which the parasite of Diabolus in Asinaria, scene 4, 1 v. 756ff. maintains. I would almost suggest that the meaning of this initial scene in the Bacchides, in which the puer Cleomachi appears, lies mainly in the fact that he quotes the contract to which the Samian Bacchis is bound, as long as she does not pay off the soldier for the remainder of the year. The Greek poet has done well here, just as the poet of alazôn (Miles Gloriosus) has done, by selecting an appropriate subject matter for the entrance scene(s), setting up the intrigue to come with relevant foreshadowing at the beginning of the piece.15 Such a woman is under a contract of loyalty, and – in a sense – not meretrix, certainly no uxor, but just mulier of her partner, her vir. It is possible that in the contract between the Samian Bacchis and Cleomachus it was also mentioned that the Samian Bacchis, for a period of one year, 15 Cp. Zwierlein 1991: 212 ff. about the corresponding preparations in Miles.

136

klaus lennartz

was no longer allowed to lead the life of a meretrix (see also fr. xi limaces viri). This particular relationship is what allows Chrysalus 839ff. to speak as he does:16 CH. quid illam, meretricemne esse censes? NI. quippini? CH. frustra es. NI. quis igitur opsecrost? CH. inveneris. ex me quidem hodie numquam fies certior.

Chrysalus, who was expecting the arrival of the soldiers, had to persuade the senex to pay the 200 Philippi that would allow Mnesilochus to continue his relationship with the Samian Bacchis. If the true conditions were known to Nicobulus, of course, he would have urged the “contract” and not freed her, so to speak, for a future liaison with his son. He must therefore be tricked into a situation in which he is bound to pay. This situation arises when Cleomachus appears in scene iv 8: the two opening verses of the scene are important for understanding this, in which Plautus certainly increases the fury of miles gloriosus Cleomachus (Zwierlein has incriminated some of the particularly dramatic elements) (842f.): CL. meamne hic Mnesilochus, Nicobuli filius, per vim ut retineat mulierem? quae haec factiost?

The desire of the soldier, of course, is still to disappear with the Samian Bacchis to Elis on the same day, and in fact he insists on the retinere by Mnesilochus. This would be in the best interest of Nicobulus. But at the same time, the scene is written in a way that shows how Nicobulus must get the impression that the liaison of his son constitutes adultery with the foreign soldier’s wife. This impression is supported by Chrysalus with the use of ambiguous information (850 ff.): NI. Chrysale, quis ille est qui minitatur filio? CH. vir hic est illius mulieris quacum accubat. NI. quid vir? CH vir, inquam. NI. nuptan est illa, obsecro? CH. scies hau multo post. NI oppido interii miser.

This is the same strategy again: the senex himself sees and hears, and it is not Chrysalus – in front of whom the senex feels invulnerable – who is the author of this comic hamartia, but rather the senex’s own misunderstanding, which punishes his comic hubris. The fact that the storyline is designed in this way, or could be perceived so, is proven by a verse from Chrysalus’ famous Trojan triumph song 16 Schönbeck 1981: 131–159 has drawn attention to the suggestive remarks of Chrysalus.



true plautus, false plautus

137

Atridae duo fratres cluent etc., 925ff., which, after the first part of the second intrigue, was developed by Plautus out of a Menandrian kernel, as Gaiser 1970: 73 correctly remarked.17 961 illustrates the comparison of the actions of Chrysalus with the second fatum Troiae, the murder of Troilus: this stage of Iliupersis had occurred, quom censuit Mnesilochum cum uxore esse dudum militis. Here censuit is of crucial importance in the sense outlined above; the important meaning is that Agnoia governs, not Pseudos! Given this development, it is extremely strange that Chrysalus, who did not want to speak the truth outright, in second part of his letterintrigue in the second letter, which Mnesilochus writes from the Bacchides’ den of vice, should drop his precious deception. The letter, for which Chrysalus assumes full responsibility according to tradition, is in turn (a) necessary for the complete avoidance of any remaining doubts (see 922 numquam edepol quicquam temere credam Chrysalo: Nicobulus longs for a direct confirmation of his (false) impression given by Mnesilochus [920 [Philippi], quos non dabo / temere etiam priusquam filium convenero]), and (b) necessary to gain the still-missing entertainment money promised by Chrysalus! The second letter – interrupted by bomolochian objections – is as follows (997ff.): NI. (loudly) “pater, ducentos Philippos quaeso Chrysalo da, si esse salvom vis me aut vitalem tibi. (. . .) pudet prodire me ad te in conspectum, pater: tantum flagitium te scire audivi meum, quod cum peregrini cubui uxore militis.” pol hau derides; nam ducentis aureis Philippis redemi vitam ex flagitio tuam. CH. nihil est illorum quin ego illi dixerim. NI. “stulte fecisse fateor. sed, quaeso, pater, ne me, in stultitia si deliqui, deseras. ego animo cupido atque oculis indomitis fui; persuasumst facere quoius me nunc facti pudet.” prius te cavisse ergo quam pudere aequom fuit. CH. eadem ista verba dudum illi dixi omnia. NI. “quaeso ut sat habeas id, pater, quod Chrysalus me obiurigavit plurumis verbis malis et me meliorem fecit praeceptis suis, ut te ei habere gratiam aequom sit bonam.” 17 That we have to deal here with a number of interpolated verses is generally accepted, if not by Lefèvre and Barsby.

138

klaus lennartz (. . .) “nunc si me fas est opsecrare aps te, pater, da mihi ducentos nummos Philippos, te opsecro. (. . .) ego ius iurandum verbis conceptis dedi, daturum id me hodie mulieri ante vesperum, priusquam a me abiret. nunc, pater, ne peiierem cura atque abduce me hinc ab hac quantum potest, quam propter tantum damni feci et flagiti.”

The letter ends with a request not to take the loss of the 200 Philippi too badly; Mnesilochus will repay it (1034 tibi si vivo). This letter, and the entire scene, is a masterpiece of comic irony and ethopoiia. The impression of pudor from the son to the father occurs at every turn. This pudor (displayed of course only at the direction of Chrysalus!) – the root word appears as a leitmotif in 1007 – allows the son to secretly refer to the actual offense, v. 1008 tantum flagitium, 1013 stulte fecisse, 1014 deliqui, 1016 facere quoius me pudet nunc facti. The same intimation allows for the comic irony in which Nicobulus each time understands something different than Mnesilochus / Chrysalus and the audience. Nicobulus now finally understands the act of adultery from which he has “saved” his son with his promise of the imagined indemnity; the misconception via the correspondence of 1008 is a particularly clear example of the exploitation: tantum flagitium te scire audivi meum and 1010f. nam ducentis aureis / Philippis redemi vitam ex flagitio tuam. After he has heard, almost from the lips of his son, the flagitium and damnum – a damnum that the father thinks is adultery, but which the audience of course understands is actually the damnum of the first 200 Philippi! – for him, all doubts are dispelled. On top of the first comic irony is built a second layer, namely that Nicobulus, from the statements of Chrysalus (1012) nihil est illorum quin ego illi dixerim and (1018) eadem istaec verba dudum illi dixi omnia, understands that Chrysalus had just given such warnings to Mnesilochus on behalf of his father. The audience knows, however, that the words are really the actual dictation of the letter. A scene full of such strange complexity is fully Menandrian. Now, it would be completely incomprehensible if, within this structure, the son in 1009 suddenly handles most outspoken vocabulary and speaks boldly in his father’s face his cum peregrini cubui uxore militis. These words destroy in one blow the thoroughly constructed ethopoiia and the entire system of intrigue and comic irony. On the other hand, they are shown, not least in their completely painless excision, to be an explanatory supplement to



true plautus, false plautus

139

v. 1008 tantum flagitium18 (to my opinion, they could demonstrate their falsity by the fact that they basically construe the quod, not with flagitium, but with scire; but see Hofmann-Szantyr 576). There are two further important points. The Bacchides (Dis Exapatôn) are designed so that the denouement takes place with the handover of the first 200 Philippi to the soldier in the forum. Outside the stage action – without the controlling presence of the slave! – there happens to be the explanation of the soldier to the father, that the Samian Bacchis is in no way the uxor, but an annual contract meretrix. Nicobulus, of course, gives the money that he had promised, and can now go, in a comic autangelos scene 5, 1087ff., to display his displeasure. In this penultimate scene of the play, Zwierlein can find broad enhancements in the Auftrittsanapäste of Nicobulus (1088. 1091–1100. 1102f. del. Zw.). In fact there is in addition to a wealth of linguistic and metrical details a fact that should pose, in my opinion, a major argument for obvious Zudichtung that has not hitherto been recognized. The first verses are spoken by Nicobulus, as previously mentioned, the comic autangelos, who now sees through the plot: NI. quiquomque ubi sunt, qui fuerunt quique futuri sunt posthac [stulti, stolidi, fatui, fungi, bardi, blenni, buccones del. Zw.] solus ego omnis longe antideo stultitia et moribus indoctis: perii, pudet: hocine me aetatis ludos bis factum esse indigne?

He is dis apatômenos and knows now, after the money transfer, that under the auspices of his own stupidity he has adopted the wrong idea. That he should, after that, still strike out furiously against Chrysalus (1094ff.), might be accepted, but at least the following verse (from which the next verses cannot be separated easily), in my opinion, is not possible: [ita miles memorat meretricem esse eam quam ille uxorem esse aiebat]

First, the allegation is factually incorrect, as Chrysalus in the original play must never actually state that the Samian Bacchis was the soldier’s uxor. Nicobulus, who had sworn in his comic hubris to be no longer deceived by Chrysalus, has directed his katastrophê completely by himself: herein lies his stultitia. Had he been straight-out lied to, it would be a malicious mendacium of the son rather than a clever ploy that Nicobulus may 18 Similar interpolation of a single verse, this time against the location of the action, Poen. 372 (see Zwierlein 1990: 148: deletion by Guyet).

140

klaus lennartz

suspect Chrysalus to be behind. And further, if Nicobulus in 1009 had heard by a letter from his son the blatant lie of the uxor, any amicable solution to the conflict would be much harder, if not impossible, to achieve. For the outright lie, pronounced by the son – whether by litterae – to his father, would not be easily forgotten. The senex could hardly – before a Roman audience! – under such conditions, join the convivium of his son and the Samian Bacchis, as takes place at the end of the Bacchides in the famous quartet of the girls with the two senes. Zwierlein must therefore be correct in positing a wider interpolation of scene 5, 1, and it is, in my opinion, argue that at least 1096–8 are not original; not original because with 1096 they reference a false verse 1009, in which they depend on the characteristic diffuse elements of the un-Plautine additions.19 References Barsby, J. (1986), Plautus, Bacchides, Warminster. Blänsdorf, J. (2000), “Plautus, Bacchides oder: Die Methoden der Plautuskritik und der Geist der Komödie,” in S. Gödde u. a. (ed.), Skenika. Beiträge zum antiken Theater u. seiner Rezeption. Festschrift zum 65. Geb. v. H.-D. Blume, Darmstadt: 153–163. De Melo, W. (2011), Plautus, Amphitryon, The Comedy of Asses, The Pot of Gold, The Two Bacchides, The Captives, Cambridge, Mass. / London. Duckworth, G.E. (1952), The Nature of Roman Comedy: A Study in Popular Entertainment, Princeton. Dziatzko, K. (1865), “Über die Terenzianischen Didaskalien,” RhM N.F. 20: 570–598. Fraenkel, E. (1922), Plautinisches im Plautus, Berlin. —— (1960), Elementi Plautini in Plauto, Florenz. Gaiser, K. (1970), “Die plautinischen Bacchides und Menanders Dis Exapaton,” Philologus 114: 51–87. Guastella, G. (2002), “I monologhi di ingresso dei parasiti. Plauto e i modelli,” in Due seminari Plautini: La tradizione del testo; i modelli, C. Questa and E. Raffaelli (eds.), Urbino: 155–198. Hanslik, R. (1959), “Publilius [14],” in RE 23, 2 [1917]: 23–38. Jachmann, G. (1934), “Terentius [36]” in RE A 5, 1 [1934]: 598–650. Lefèvre, E. (1978), “Plautus-Studien II: Die Brief-Intrige in Menanders Dis Exapatôn und ihre Verdoppelung in den Bacchides,” Hermes 106: 518–538. Lennartz, K. (1994), Non verba sed vim. Kritisch-exegetische Untersuchungen zu den Fragmenten archaischer römischer Tragiker, Stuttgart und Leipzig. Leo, Fr. (1895–6) (ed.), T. Macci Plauti comoediae, rec. et emend., vol. 1–2, Berlin. —— (1913), Geschichte der römischen Literatur, vol. 1: Die archaische Literatur, Berlin. Lindsay, W.M. (1904–5) (ed.), T. Macci Plauti comoediae, vol. 1–2, Oxford. Lowe, N.J. (2000), The Classical Plot and the Invention of Western Narrative, Cambridge. Marti, H. (1963), “Terenz 1909–1959 (Fortsetzung),” Lustrum 8: 5–101. Primmer, A. (1984), Handlungsgliederung in Nea und Palliata. Dis Exapaton und Bacchides, Wien (Sitzungsberichte der Österr. Akad. d. Wiss., philos.-hist. Kl. 441).

19 Thanks to Victoria Hooper, Nottingham, for “tuning” my English.



true plautus, false plautus

141

Questa, C. (2008) (ed.), Titus Maccius Plautus, Bacchides, Sarsina und Urbino. Schönbeck, H.-P. (1981), Beiträge zur Interpretation der plautinischen Bacchides, Düsseldorf. Studemund, W. (1889) (ed.), T. Macci Plauti fabularum reliquiae Ambrosianae. Codicis rescripti Ambrosiani apographum, Berlin. Suerbaum, W. (2002) (ed.), Handbuch der lateinischen Literatur der Antike: Die archaische Literatur von den Anfängen bis Sullas Tod. Die vorliterarische Periode und die Zeit von 240 bis 78 v. Chr., München (HdA 8, 1). Woytek, E. (1982), T. Maccius Plautus, Persa. Einleitung, Text und Kommentar, Wien (Sitzungsberichte der Österr. Ak. der Wiss., philol.-hist. Klasse 385). Zwierlein, O. (1990), Zur Kritik und Exegese des Plautus I: Poenulus und Curculio, Mainz. —— (1991), Zur Kritik und Exegese des Plautus II: Miles gloriosus, Mainz. —— (1992), Zur Kritik und Exegese des Plautus IV: Bacchides, Mainz.

Athena and Pallas, Image, Copies, Fakes, and Doubles Françoise Létoublon The question of the fake in the visual arts may be still more compli­ cated than in literature. It raises three aspects of the fake: the original author, the time of creation, and the resemblance between the origi­ nal and the copy. Greek literature offers many allusions to and anecdotes about a statue named “the palladion” after the goddess Athena’s surname Pallas1 that played an important part in the legend of Troy and in several foundation myths of cities around the Greek and Roman ancient world. We shall try to analyze some aspects of the question of the fake and the genuine palladion in myth. The palladion is not mentioned in Homer, but the important part it played in the legend of Troy appears in numerous other texts, which seems to imply that it was a key of first importance to victory or defeat in the war in several parts of the Epic Cycle. Thus the hypothesis that it is alluded to in the Iliad without being explicitly mentioned appears sustainable. The paper aims to detect the ancient allusions or refer­ ences to a very complicated story and to look at the iconographical testimonies as well.2 The task is not easy, since texts, though numerous, are generally recent, but some images undoubtly ancient prove that the legend circulated early. The complications of this story might imply that the statue had some secret powers and therefore was required to stay hidden in the adyton; maybe even uttering its name was forbidden. The scarcity of ancient literary documents could then be explained by a wellpreserved secrecy; but of course this cannot be demonstrated. A first detailed narrative occurs in Apollodorus, Bibl. 3.142–146 and in two passages of Dionysius Halicarnassensis. Let us begin with 1  The reverse direction might also be pertinent. 2 The iconographical part of the research contributed by my colleague Maria Paola Castiglioni was orally presented with my work in Greece at a conference on the Epic Cycle. Unhappily, she did not have time to revise her text, so I remain solely responsible for this paper including the errors that may occur. My warm thanks also go to Javier Martínez who asked me to take part to this volume and to Stephen Rojcewicz who revised my English with a kind akribeia.

144

françoise létoublon

Ps.-Apollodorus. The passage is clearly articulated in two parts, the first one concerning the presence of the statue in Troy, the second one telling its divine origin. Though this succession may appear out of chronological order, we will follow it since it will allow us to begin the research with the statue moving in the human world and to close it with its creation by a goddess as a copy and substitute of a beloved person. This idea might imply that fabricating a statue is generally meant as the creation of a fake imitation of a figure, be it human or divine. In between, with the help of other authors, historians, or poets, we will investigate other aspects of the statue. Paragraphs 3.142–3 of the Bibliotheca first recall Ilos’ itinerary from Arcadia in Greece to Phrygia through Samothace. He was chosen as king and received an oracle telling him to follow a cow and to found a city in the place where the cow would stop, which shows a strong parallel to the better known Theban myth. He followed the instructions and founded a city named Ilios, named for himself. He asked Zeus to give him a sign, and a statue fell from the skies before Ilos’ tent: the Greek text reads τὸ διιπετὲς παλλάδιον πρὸ τῆς σκηνῆς κείμενον ἐθεάσατο: “he admired the διιπετὲς pal­ ladion.” Which meaning does the adjective διιπετὲς convey? Is it seman­ tically appropriate? We will see further that it is probably a formulaic epithet,3 which may thus appear in more or less pertinent contexts. Let us start from formal facts: it is a compound adjective, the first element of which Διι- (a variant form Διο-πετὲς will be met later) can only be a dative with an instrumental value (or in the case of Διο- a genitive form of origin) of the name of Zeus. The meaning is then either “fallen from Zeus” or “by Zeus” (instrumental value of the dative). The epithet was probably no lon­ ger understood already at the time where the Bibliotheca was composed, and it was glossed in Greek texts and translated in Latin as if it meant “fallen from the sky.” In any case, the epithet seems here quite appropriate since the object was sent by Zeus in response to Ilos’ prayer. In a more detailed narrative, Dionysius of Halicarnassus tells in great detail the foundation of Rome through ancient Pelasgians, Aborigenians, and Trojans who escaped Ilios with Aeneas during the city’s fall. In 1.67, he ventures to explain how the Roman images of the gods were taken from Troy to Italy (first to Lavinium, later to Alba) by Aeneas and kept there in a temple with an adyton (ναοῦ χωρίον ἔχοντος ἄβατον). The following

3 See Milman Parry’s definition of the formula and recall that Achilles may be called “swift-footed” (Gr. Πόδας ὠκύς. ποδωκύς) even when seated (Graziosi and Haubold 2005).



athena and pallas, image, copies, fakes, and doubles

145

night, the statues moved by themselves and were found standing on their former bases (διαμείψαντα τὰ βρέτη τὴν στάσιν ἐπὶ τῶν ἀρχαίων εὑρεθῆναι κείμενα βάθρων); the same phenomenon occurred a second time. Then they decided to leave the statues in their places and to arrange for six hundred people to care for them. The historian gives their names, and in particular the Roman one (Penates). He ends paragraph 67 with a remark on the absolute necessity of secrecy around them, and in general:4 it was forbidden to see them and even to speak or write about them. We have to remember this obligation, which may have have applied to sacred objects from the beginning. In the following paragraph, he mentions a small tem­ ple on the forum in Rome, in a place called in Velia, where Trojan gods’ images, two sitting youths with spears and military equipment, could be looked at by everybody. Among the sources for them he explicitly men­ tions Arctinos, who is known elsewhere as one of the authors of the Epic Cycle, namely of the Aithiopis and the Iliou Persis.5 The palladion occurs in this context of the sacred objects brought with them by Greek migrants first to the island of Samothrace, then to several places in Phrygia, then to Italy, and eventually to Rome: Dionysius actually uses the plural palladia, which raises the issue of a genuine palladion versus a fake palladion. The detailed narrative of 1.68.3 reads thus: Χρύσην τὴν Πάλλαντος θυγατέρα γημαμένην Δαρδάνῳ φερνὰς ἐπενέγκασθαι δωρεὰς Ἀθηνᾶς τά τε Παλλάδια καὶ τὰ ἱερὰ τῶν μεγάλων θεῶν διδαχθεῖσαν αὐτῶν τὰς τελετὰς. Chrysê, the daughter of Pallas, when she was married to Dardanus, brought for her dowry the gifts of Athena, that is, the palladia and the sacred sym­ bols of the Great Gods, in whose mysteries she had been instructed (tras. E. Cary).6

4 End of 1.67: ἐγὼ δὲ ὅσα μὲν ὁρᾶν ἅπασιν οὐ θέμις οὔτε παρὰ τῶν ὁρώντων ἀκούειν οὔτε ἀναγράϕειν οἴομαι δεῖν, νεμεσῶ δὲ καὶ τοῖς ἄλλοις, ὅσοι πλείω τῶν συγχωρουμένων ὑπὸ νόμου ζητεῖν ἢ γινώσκειν ἀξιοῦσιν “For my part, I believe that in the case of those things which it is not lawful for all to see I ought neither to hear about them from those do do see them nor to describe them; and I am indignant with every one else, too, who presumes to inquire into or to know more thant what is permitted by law” (tras. E. Cary). Beginning of 1.67: ῝Α δὲ αὐτός τε ἰδὼν ἐπίσταμαι καὶ δέος οὐδὲν ἀποκωλύει με περὶ αὐτῶν γράϕειν τοιάδε ἐστί “But the things which I myself know by having seen them and concerning which no scruple forbids me to write are as follows” (tras. E. Cary). 5 According to the summary by Photios of the Chrestomathia of Proclos. 6 Those Mysteries of the Great Gods are thus supposed to come from the Peloponnese. In this case, the word palladion seems to be related to the name of Chryse’s father, while in (Apollodorus’) version it is related to the young girl, friend of Athena.

146

françoise létoublon

Afterwards, having escaped the flood (τὴν ἐπομβρίαν φεύγοντες),7 Darda­ nos and his Arcadians fled from the Peloponnese to Samothrace, where they founded the Mysteries.8 Dardanos seems then to be the main chief in the initiation process: ’Επειδὴ δὲ τὴν ἐπομβρίαν ϕεύγοντες ’Αρκάδες Πελοπόννησον μὲν ἐξέλιπον, ἐν δὲ τῇ Θρᾳκίᾳ νήσῳ τοὺς βίους ἱδρύσαντο, κατασκευάσαι τὸν Δάρδανον ἐνταῦθα τῶν θεῶν τούτων ἱερὸν ἀρρήτους τοῖς ἄλλοις ποιοῦντα τὰς ἰδίους αὐτῶν ὀνομασίας καὶ τὰς τελετὰς αὐτοῖς τὰς καὶ εἰς τόδε χρόνου γινομένας ὑπὸ Σαμοθρᾴκων ἐπιτελεῖν. When the Arcadians, fleeing from the deluge, left the Peloponnesus and established their abode in the Thracian island, Dardanus, built there a tem­ ple to these gods, whose particular names he kept secret from all any others, and performed the mysteries in their honour which are observed to this day by the Samothracians (trans. E. Carey).

He did not stay in the island, but left for Asia Minor with the palladion and other sacred objects.9 Thereafter, Dionysius quotes the text of the oracle that mentions the aporthetos polis “the indestructible city” (then Dardania, later Ilios nearer to the coast): the statue was meant to guaran­ tee the safety of the city forever (note the formulaic expressive redupli­ cation in the text of τὸν ἀεὶ χρόνον and ἤματα πάντα).10 The well-attested links between oracles and formulaic diction in Greek tradition allow us to conclude that oral tradition helped preserve the words of a very old 7 The story is thus linked to Deukalion’s myth. 8 The secrets of the Mysteries of Samothrace have been so well kept that we do not know much about them, see Burkert 1987, Stochos 2001. 9 Dionysius’ text may contain contradictions since it seems to imply that the same sacred objects followed Dardanos in his migrations from Peloponnese to Phrygia through Samothrace, whereas the Mysteries of Samothrace lasted long after Dardanos and others left for Asia. Dionysius actually makes a distinction between the sacred objects proper to Samothrace and other ones: ὡς δὲ μετῆγε τοῦ λεὼ τὴν πλείω μοῖραν εἰς τὴν Ἀσίαν τὰ μὲν ἱερὰ τῶν θεῶν καὶ τὰς τελετὰς τοῖς ὑπομείνασιν ἐν τῇ νήσῳ καταλιπεῖν· τὰ δὲ Παλλάδια καὶ τὰς θεῶν εἰκόνας κατασκευασάμενον ἀγαγέσθαι μετ' αὐτοῦ. 10 The formula ἤματα πάντα is obviously Homeric: Il. 8.539, 12.133, 13.826, 14.235, 276, 19.226, 23.594, 24.491, Od. 2.55, 205, 4.209, 592, 5.136, 210, 219, 6.46, 7.94, 257, 8.431, 468, 9.123, 10.467, 15.54, 17.534, 21.156, 23.6, 336, 24.24; it occurs very often in the verse end, but not constantly. The phrase τὸν ἀεὶ χρόνον appears as a gloss or comment, as if the late author was aware that Homeric Greek could confuse. For more clarity, we quote here the whole text of the oracle as it stands in Roman Antiquities 1.68.4:  Εἰς πόλιν ἣν κτίζῃσθα θεοῖς σέβας ἄϕθιτον αἰεὶ / θεῖναι, καὶ ϕυλακαῖς τε σέβειν θυσίαις τε χοροῖς τε. / ἔστ' ἂν γὰρ τάδε σεμνὰ καθ' ὑμετέρην χθόνα μίμνῃ / δῶρα Διὸς κούρης ἀλόχῳ σέθεν, ἡ δὲ πόλις σοι / ἔσται ἀπόρθητος τὸν ἀεὶ χρόνον ἤματα πάντα.  In the town thou buildest worship undying found / To gods ancestral; guard them, sacrifice, / Adore with choirs. For whilst these holy things / In thy land remain, Zeus' daughter's gifts of old / Bestowed upon thy souse, secure from harm / Thy city shall abide forevermore (trans. E. Cary).



athena and pallas, image, copies, fakes, and doubles

147

text.11 We are then in Troy or Troad in the times of its occupation and foundation. Let us remark that those migrants are supposed to be Greek: thus, for Dionysius, himself of Ionian origin, the war of Troy consists not at all in a clash of Greeks against Asiatic enemies; rather it is a matter of Greek tribes fighting one another because of raped women, booty, and wealth.12 In 1.69, Dionysius explains how Dardanos set the objects in the eponymous city founded in Troad,13 and how they were later transferred to Ilios where the temple was built for them on the acro(polis), with a secret room (adyton); he adds that the remaining palladion was taken from the acropolis by Aeneas after the fall of Troy, which leads him to mention that the other one was stolen by Diomedes and Odysseus: Δάρδανον μὲν ἐν τῇ κτισθείσῃ τε ὑϕ’ ἑαυτοῦ καὶ ὀνομασίας ὁμοίας τυχούσῃ πόλει τὰ ἕδη καταλιπεῖν, ’Ιλίου δ’ ἐν ὑστέρῳ χρόνῳ συνοικισθέντος ἐκεῖσε μετενεχθῆναι πρὸς τῶν ἐγγόνων αὐτοῦ τὰ ἱερά. ποιήσασθαι δὲ τοὺς ’Ιλιεῖς νεών τε καὶ ἄδυτον αὐτοῖς ἐπὶ τῆς ἄκρας καὶ ϕυλάττειν δι’ ἐπιμελείας ὅσης ἐδύναντο πλείστης θεόπεμπτά τε ἡγουμένους εἶναι καὶ [some words are missing there in Greek] πόλει. ἁλισκομένης δὲ τῆς κάτω πόλεως τὸν Αἰνείαν καρτερὸν τῆς ἄκρας γενόμενον, ἄραντα ἐκ τῶν ἀδύτων τά τε ἱερὰ τῶν μεγάλων θεῶν καὶ ὅπερ ἔτι περιῆν Παλλάδιον (θάτερον γὰρ ’Οδυσσέα καὶ Διομήδην νυκτός ϕασιν εἰς ῎Ιλιον ἀϕικομένους κλοπῇ λαβεῖν) οἴχεσθαί τε κομίσαντα [τὸν Αἰνείαν] ἐκ τῆς πόλεως καὶ ἐλθεῖν ἄγοντα εἰς ’Ιταλίαν. Dardanus, accordingly, left the statues in the city which he founded and named after himself, but when Ilium was settled later, they were removed thither by his descendants; and the people of Ilium built a temple and a sanctuary for them upon the citadel and preserved them with all possible care, looking upon them as sent from Heaven and as pledges of the city’s safety. And while the lower town was being captured, Aeneas, possessing himself of the citadel, took out of the sanctuary the images of the Great Gods and the Palladium which still remained (for Odysseus and Diomed, they say, when they came into Ilium by night, had stolen the other away), and carrying them with him out of the city, brought them into Italy (trans. E. Cary).

After mentioning, in prose, the importance of the objects sent by the gods (θεόπεμπτά) for keeping the city safe (καὶ σωτηρίας κύρια τῇ πόλει),14 he very quickly resumes the taking of the downtown part of Troy with a 11  See the entries “Formulaic Language” and “Oracular Language” in the EAGLL (Létou­ blon, to appear). Among the formulae, let us also note σέβας ἄϕθιτον αἰεὶ and δῶρα Διὸς κούρης. 12 As Herodotus put it previously in his first book. 13 Dardanos was the city’s name for Dionysios, whereas the Iliad mentions Dardania, higher on Mount Ida’s slopes than Troy. 14 It is actually a prosaic transposition of the oracular text mentioned above.

148

françoise létoublon

genitive absolute (ἁλισκομένης δὲ τῆς κάτω πόλεως), which explains that Aeneas could take the sacred objects from Acropolis and flee with them to Italy. There are, then, two palladia: Dionysius invokes the authority of Arcti­ nos (is it a quotation or rather a translation in prose?) for a commentary on the plural: Zeus gave Dardanos one palladion, and this is the object that was kept in Ilios until its fall: Ἀρκτῖνος δέ ϕησιν ὑπὸ Διὸς δοθῆναι Δαρδάνῳ Παλλάδιον ἓν καὶ εἶναι τοῦτο ἐν Ἰλίῳ τέως ἡ πόλις ἡλίσκετο κεκρυμμένον ἐν ἀβάτῳ· But a copy of the original palladion had been made, so close to the genuine image that people who saw it could be deluded. This image was set up for viewing, and the (Achaean) thieves took the copy, the fake palladion, instead of the genuine one: εἰκόνα δ’ ἐκείνου κατεσκευασμένην ὡς μηδὲν τῆς ἀρχετύπου διαϕέρειν ἀπάτης τῶν ἐπιβουλευόντων ἕνεκεν ἐν ϕανερῷ τεθῆναι καὶ αὐτὴν Ἀχαιοὺς ἐπιβουλεύσαντας λαβεῖν. The end of paragraph 69 specifies that the genuine palladion taken from Troy by Aeneas was still kept in the temple of Hestia (Latin Vesta) with the city’s sacred fire in Dionysius’ time. Dionysius says he will come back to this issue later; he does so in 2.66–7, in the wake of the Roman heritage and tradition: he explains how the temple of Vesta/Hestia preserves not only the well known sacred public fire, but also the objects brought by Dardanos from Samothace and later by Aeneas from Troy to Italy: τοῦτο δὴ λαβόντες ὁμολογούμενον ἐπισυνάπτουσιν αὐτοὶ στοχασμούς τινας ἰδίους, οἱ μὲν ἐκ τῶν ἐν Σαμοθρᾴκῃ λέγοντες ἱερῶν μοῖραν εἶναί τινα ϕυλαττομένην τὴν ἐνθάδε, Δαρδάνου μὲν εἰς τὴν ὑϕ’ ἑαυτοῦ κτισθεῖσαν πόλιν ἐκ τῆς νήσου τὰ ἱερὰ μετενεγκαμένου, Αἰνείου δὲ, ὅτ’ ἔϕυγεν ἐκ τῆς Τρωάδος ἅμα τοῖς ἄλλοις καὶ ταῦτα κομίσαντος εἰς Ἰταλίαν, οἱ δὲ τὸ διοπετὲς Παλλάδιον ἀποϕαίνοντες εἶναι τὸ παρ’ Ἰλιεῦσι γενόμενον, ὡς Αἰνείου κομίσαντος αὐτὸ δι’ ἐμπειρίαν, Ἀχαιῶν δὲ τὸ μίμημα αὐτοῦ λαβόντων κλοπῇ· (2.66.5) Taking this incident, then, as an admitted fact, they add some conjectures of their own. Thus, some affirm that the objects preserved here are a part of those holy things which were once in Samothrace; that Dardanus removed them out of that island into the city which he himself had built, and that Aeneas, when he fled from the Troad, brought them along with the other holy things into Italy. But others declare that it is the Palladium that fell from Heaven, the same that was in the possession of the people of Ilium ; for they hold that Aeneas, being well acquainted with it, brought it into Italy, whereas the Achaeans stole away the copy (trans. E. Cary).

Thus there are for him two divergent traditions, one admitting that Dar­ danos took all the sacred objects from Samothrace, whereas in the other the palladion, qualified as διοπετὲς, was sent by Zeus. The formal coinci­ dence with Apollodorus’ τὸ διιπετὲς Παλλάδιον is striking, maybe all the



athena and pallas, image, copies, fakes, and doubles

149

more with a very slight difference in the text, implying either a dative or a genitive of Zeus’ name in the first term of the compound.15 It seems then that, according the first tradition, the palladion was part of the wed­ ding gifts given to Dardanos’s wife by Zeus’ daughter, as seen above. In the same part of the narrative, Dionysius also alludes to the theft of the palladion, and to the fact that the Achaean thieves were fooled and took only the copy (τὸ μίμημα αὐτοῦ). The occurrence of the formula τὸ διοπετὲς παλλάδιον occurs in the context of the statue’s presence in Rome. The narrative does not seem entirely consistent with Book 1 and the mention of Dardanos’ wedding gifts. But Dionysius is perhaps more consistent with the historical method here in that he recognizes that several mythological traditions may jus­ tify the state of things. In 1.69.4, he called the palladion the legendary, τὸ μυθευόμενον παλλάδιον, and some anecdotes that he tells, in agreement with other authors, seem to imply that the statue had magical powers. In 2.66.4, Lucius Caecilius Metellus, seeing Vesta’s temple burning as the priestesses fled without the sacred objects, rushed into fire without car­ ing for his own safety and saved them. Dionysius does not mention any misfortune for this important Roman citizen, who received a reward for this courageous deed. But other saviors of the objects, particularly the pal­ ladion, were less fortunate as some cases of blindness particularly testify. Metellus himself may have suffered this fate according to other sources. Plutarch, in his Minor Parallels, describes Ilos’ blindness in two ver­ sions. In one case, Ilos took (ἥρπασε may imply violence)16 the statue out of the adyton to the daylight and was blinded (ἐτυφλώθη).17 He prob­ ably had good intentions, since in both versions the temple was burn­ ing (καιόμενον). Plutarch gives the reason for his blinding: no man was allowed to gaze at the statue (οὐ γὰρ ἐξῆν ὑπ’ ἀνδρὸς βλέπεσθαι). According to Plutarch, this was the version by Derkullos, whereas Aristeides of Mile­ tos suggested that Ilos was trying to save the statue from burning. In any case, he was blinded but recovered his sight later. Plutarch includes this tale in his Minor Parallels to show that the Trojan Ilos had a fate similar to the Roman Metellus (mentioned by Dionysius) and Antylus. From those allusions, it can be deduced that the statue had the power, when brought

15 Thus the variant in the tradition could be a strong clue to the authenticity of the formula coming from an oral tradition relative to the Epic Cycle. 16 The verb ἁρπάζω is currently used in Greek for the rape of women, for instance. 17 See Létoublon 2010. ἐτυφλώθη in the aorist may imply that it happened suddenly, as if the statue were punishing his raper.

150

françoise létoublon

into daylight, to throw dangerous lightning flashes that could blind human beings. The description of the statue in Virgil’s Aeneid strikingly tallies with this hypothesis. The Greek traitor Sinon tells the Trojans that after they have stolen the palladion, Diomedes and Ulysses encountered the goddess Athena’s wrath (Virg. Aen. 2.171–179): Nec dubiis ea signa dedit Tritonia monstris. Vix positum castris simulacrum, arsere coruscae luminibus flammae arrectis, salsusque per artus sudor iit, terque ipsa solo—mirabile dictu— emicuit, parmamque ferens hastamque trementem. Extemplo temptanda fuga canit aequora Calchas, nec posse Argolicis exscindi Pergama telis, omina ni repetant Argis, numenque reducant, quod pelago et curvis secum avexere carinis. Tritonia gave signs of this by no uncertain portents. The statue had just been set down in camp. Dancing flames blazed from her upwards-turned eyes and salt sweat passed over her limbs. Three times she actually leaped from the ground (a wonder to relate), carrying her round shield and quivering spear. At once Calchas pronounced that the Greeks were to venture on the waves in flight; Pergama was not to be taken by Argive arms, if they did not seek out fresh auspices at Argos and bring back the goddess whom they had carried off with them by sea in their curved ships (trans. Horsfall).18

The fact that the context of Virgil’s interpretation of the Trojan war is Sinon’s lies does not negate the general knowledge of the brightness of the statue and its magical powers. Quintus of Smyrna 10.359 f. has an allusion to the palladion in a cryptic prediction:19 ἀλλά μιν αὐτὸς ἀπ᾽ Οὐλύμποιο Κρονίων κάββαλεν ἐς Πριάμοιο πολυχρύσοιο πόληα. 10.350–60 How on his advice the son of mighty Tydeus / Accompanied by Odysseus, would scale the lofty wall, / Inflict a miserable death upon Alcathoos, / And seize the wise goddess Tritogeneia with her consent, / The image that had protected the city and people of Troy, / Because not even a god, however great his anger, / Had strength enough to sack the prosperous

18 N. Horsfall, Virgil, Aeneid 2. A Commentary, Leiden, 2008: 11. See Horsfall’s commen­ tary on these verses, ibid.: 165–73. The statue (called sacrem effigiem in v. 167) turned her eyes away from the thieves and sent flames (arsere coruscae / luminibus flammae arrectis, v. 172–3 is particularly expressive). 19 Of course, since the word “palladion” is not there, a search of TLG missed this occurrence.



athena and pallas, image, copies, fakes, and doubles

151

city of Priam / As long as that goddess stood inviolate in iths defense; / Her immortal image had not been wrought in iron / By humans, but the son of Kronos himself had thrown it / Down from Olympos into the city of wealthy Priam (tras. A. James)

We can notice here the parallel with Dionysius Halicarnassensis’ refer­ ence to the oracle, and the mention of Zeus (called here by the peri­ phrastic “son of Cronos”) throwing the statue from Olympus: the aorist with apocope and assimilation in the beginning of the verse is typically Homeric.20 Quintus does not name the palladion as such, but uses ἄμβροτον εἶδος, a periphrastic expression which means “immortal appearence,” and strengthens the Homeric tone of the passage.21 So it seems that Quintus imitated Homer’s style, perhaps in a particular instance when the content is not actually Homeric. It is also possible that the palladion was not explicitly designated here, precisely because of the magical power of the statue: in Dionysius, the description of the foundation of the Samothrace Mysteries by Dardanos strongly implies the presence of some secret objects and words,22 so that the periphrastic expression used by Quintus might respect a language pro­ hibition. To those classical sources, the New Jacoby online now allows us to add Papyrus Rylands 22a, which alludes to the theft of the palladion by Odys­ seus and Diomedes, following a prediction given by the seer Helenus.23 Thus literary sources available for the palladion are relatively late, but some of them include pieces of formulaic style that may refer to quota­ tions from the Epic Cycle or from oracular traditions transmitted from a period close to the Homeric epics. The formal variations appear strong testimonies to archaic remains kept in recent texts as linguistic fossils. Pictorial arts might also show some interesting testimonies. To sum­ marize Castiglioni’s still unpublished study, the statue considered to be the palladion may have been linked to several episodes in the Epic Cycle, namely Cassandra’s rape by Ajax, the recovery and attempted murder of Helen by Menelaus in the Iliou Persis, the theft of the palladion in the

20 The TLG provides us with 8 occurrences of κάββαλ-: Il. 5.343, 8.249, 9.206, 12.206, 23.127, 683, Od. 6.172, 17.302. 21 ἄμβροτον is actually used in Homer with αἷμα, θεόν, κρήδεμνον, ἱστόν; εἶδος appears in several Homeric formulas, none with an epithet. Thus both words are Homeric, but their association is not. 22 Dion. Hal. 1.67 above. 23 P.Ryl. 22, see below.

152

françoise létoublon

Ilias Parva according to Proclus’ summary,24 and Aeneas carrying both the palladion and his father Anchises. The images showing two similar palladia are very interesting in light of Dionysius’s mention of a copy of the palladion made intentionally to deceive enemies. We might perhaps call it an “official fake.” If the legend admits the existence of two palladia, one genuine one and one fake, Diomedes and Odysseus taking one and Aeneas the other, it raises a serious issue: the oracles thought that the city would remain safe as long as the palladion stayed in its hiding place. If the Achaeans took the fake, the city should not have fallen unless the oracles were mistaken. If they took the genuine, how could Aeneas still have the genuine palladion to bring to Italy? The argument of a second palladion opens the way to sophistic speculations on fate and history, and maybe explains that from two, the number could be expanded ad infinitum: it may have been the main reason why several cities in ancient Greece and Italy claimed that they owned a palladion kept among their most precious heritage from the war at Troy.25 Before going back to the fabrication of the statue, it seems necessary to insist on the status of the formulae and their relation to poetry and metrics, which could reveal quotations from the Epic Cycle, particularly Διο/Διιπετὲς παλλάδιον.26 We know two probable Greek “quotations” (Apollodorus and Dionysius Halicarnassensis), and a Latin “translation” in Dictys Cretensis (a caelo). The quotation is not explicit and is not officially attributed to the Epic Cycle by either one or the other author. The fragments of the Epic Cycle are very scarce, but they seem to have used the same hexametric form as the epics and archaic poetry in general. Let us then try a metrical analysis. In the dactylic hexame­ ter παλλάδιον begins with a long vowel followed by three short vowels: 24 The first evidence is an Attic cup from the Hermitage Museum (LIMC 23), called the Makron cup (around 480 bc), with two palladia. A Stockholm Amphora (LIMC 24), an Attic production of the same period (480), represents the same scene, but Athena replaces the Greek heroes between Diomedes and Odysseus who face each other and are both fully armed. Two palladia also occur in an Apulian oenochoe from the Louvre (360–350). An Attic cup (470–460) and a Panathenaic amphora (about 420) depict only one palla­ dion in accordance with the Little Iliadʼs version. 25 Namely at least Argos, Sparta, Thebes, Athens, Amphissa, Rome, Siris, Luceria, and Lavinium. 26 This adjective, not known elsewhere if we trust the LSJ, seems to have been mod­ elled after the Homeric frequent form διοτρεφής (56 cases in Homer, among them 37 of the neuter διοτρεφές).



athena and pallas, image, copies, fakes, and doubles

153

if the final o stands in front of a lengthening consonant, the word can be scanned ‒�� / ‒. Thus the nominative form cannot be located at the end of the verse. In other cases we can imagine παλλαδίῳ or παλλαδίου: ‒�� / ‒. This form cannot stand at the end of the verse either, but it could stand at the beginning of a second hemistich, after a penthemimeral caesura. On the other hand, διοπετὲς scanned ‒�� / ‒, implies a metrical length­ ening of the first iota. The inversion of both terms does not provide a satisfactory solution. Thus the whole formula διοπετὲς (variant διιπετὲς in Apollodorus) παλλάδιον implies ‒�� / ‒ and ‒�� / ‒, which cannot occur as a metrical sequence in the hexameter, but could reflect the dislocation of the terms with other words, something like διοπετὲς �� παλλάδιον, for instance, with the epithet at the beginning of the verse, followed by two short syllables, as, for instance, διοπετὲς λάβε / παλλάδιον . . . or διοπετὲς φέρε / παλλάδιον . . . with the caesura after the verbal form. Parry’s definition of the formula helps us understand διοπετὲς (διιπετὲς) as a fixed or particularising epithet (that is, which is only used to qualify this particular object in Homeric Greek): it gives the object the status of an individual, a person, providing it with almost a proper name. We actu­ ally know that in Greek the onoma is first a proper noun, then only sec­ ondarily a “common” name.27 This settles the problem of the meaning of the epithet which becomes in Latin (and other existing translations of the Greek) “fallen from the sky,” while the first element of the compound Διοπετὲς (Διιπετὲς) can only be a genitive form of origin (or in the case of Διι- a dative with an instrumental value) of the name of Zeus. The meaning is then: “fallen from Zeus” or “by Zeus” (instrumental value of the dative). The equivalence between Zeus and the sky is well known in the IndoEuropean reconstruction of the name (the correspondents skr dyauh, Lat. dies “daytime” imply a close semantic reconstruction with the meaning “bright day”). It is also valid to speak from a “functional” point of view: the Iliad refers to the sharing of timai between the sons of Cronos, ending with Zeus reigning over the sky and the earth, Poseidon over the sea and Hades over the underworld. Nethertheless it remains linguistically impos­ sible in Greek to understand Διοπετὲς (Διιπετὲς) as meaning properly ἀπ᾽ οὐρανοῦ or κατ᾽ οὐρανοῦ. On the contrary, the excerpt from Quintus quoted above indicates that Zeus threw the palladion (Κρόνιων / κάββαλεν) from the top of Olympus (ἀπ᾽ Οὐλύμποιο). 27 We studied a similar case for Πηλίαδα μελίην in Homer.

154

françoise létoublon

We assume that both details were mentioned in the earliest texts, and that the actual meaning of διιπετές, as proved by the occurrence of such syntagms as Δία ῥῖψαι (Bibl. 3.145) and Κρόνιων / κάββαλεν (Quintus of Smyrna 10.359–60), must be “thrown by Zeus.” But as some late lexicog­ raphers attest, this phrase could no longer be properly understood during their times and required a gloss saying “fallen from the sky” or glosses such as ἀπ᾽ Οὐλύμποιο to replace it. P. Rylands 22 τὸ τῆς ᾽Αθηνᾶς οὐράνι[ον ἄγαλμα, “Athena’s celestial statue” seems consistent with this view. It is Zeus who threw the palladion from Olympus, and Dardanos or Ilos who found it at his door (we will not dwell upon whether it was a wed­ ding gift for Chryse and Dardanos).28 Zeus never uses epiphanies to show himself to mortals on earth as other gods do (Athena, Hermes, Apollon, Poseidon, Ares in the Epics), but he usually uses messengers for his com­ munication.29 In any case, in Homer’s text, Zeus always sits on the heights, whether on Mount Olympus or Mount Ida, and only leaves them for the sacred Tables of the Ethiopians; he sometimes stands up to brandish the lightning or the thunder, and to throw objects from his throne such as the rain of blood he sends out in the episode of Sarpedon’s death.30 Cf. The formula αἷμα διοπετές “blood fallen from Zeus” in Homer (Il. 5.339, 5.870). We will try now to show that the palladion was both a magical object and a talisman.31 Given by the divinity, the statue must be kept inside (in the adyton of the temple, inside the city), far from the light. In daylight, the statue emits flashes of light (Virgil, Aen. 2.) and can blind the person who sees it (cf. the blinding of Ilos as told by Derkyllos in Plutarch’s Minor Parallels, and the parallels he mentions in Roman tradition: Metellus etc.). This could be the reason why Diomedes and Ulysses steal the palladion at night.

28 Three mythological versions of origin at least may be distinguished: 1. The statue is wedding gift to Dardanos’ wife by Athena (who then may have fabricated it previously), 2. The statue is found fallen before Dardanos’ door, 3. The statue if found fallen before Ilos’ door. Dionysius Halicarnensis mentions both versions relating to Dardanos, [Apollodorus] the version relating to Ilos and Athena’s fabrication. 29 His frequent courtships imply, however, that he sometimes descends to earth and takes a human form – Amphitryon’s, for example, – to seduce women such as Alcmene. 30 The meteorological verbs are impersonal in Greek, but the imperative used in prayers asking for rain and addressed to Zeus prove that rain and snow were thought as the result of his will and action. 31  This point was to our opinion largely underestimated in Faraoneʼs book on the question; see Faraone 1994.



athena and pallas, image, copies, fakes, and doubles

155

The formula we analyzed probably comes from a fragment of the Little Iliad or the Iliupersis, as the summary by Proclus, and perhaps some icon­ ographical representations, seem to indicate. But we must be careful: the theft of the palladion could belong to the episodes of the fall of Troy, but the fall of the palladion on Ilos’ or Dardanos’s front door might have been told or alluded to in a previous episode. Naturally, if Διοπετὲς Παλλάδιον is a formula, it might be repeated in several episodes without a deep seman­ tic value or direct relevance in every one of them. In any case, Apollodoros and Dionysius probably still had knowledge of the traditions deriving from the Epic Cycle, in which the small statue of Athena (see the size of three cubits accurately indicated by Apollodorus) had a formulaic epithet, but its meaning was forgotten over time. Several texts and images show two palladia or use the plural. The most explicit version is found in Dionysius, who recalls Dardanos’ travel from Arcadia to Samothrace. He seems to have received the statue there, either as a wedding gift or rather “fallen from Zeus.” Did he then have a copy of the statue made before leaving the island for Asia Minor with sacred objects, including the palladia? Later on, several episodes provide refer­ ences, which are more or less explicit, to at least two palladia: it may be inferred that the true or original statue came from Zeus who threw it on earth. The copy might have been designed as a trap for thieves who wanted to steal the statue for its magical power. Several of our texts refer to predictions linked to the palladion (see esp. Dionysius and Quintus). The theft of the statue by Odysseus and Diomedes is a consequence of such a prediction: they knew that Troy would not fall if the palladion remained in its adyton, far from human eyes. The allusions to their expedition are numerous, but Dictys Creten­ sis alone provides an explanation for the method used by the Achaean heroes: Antenor and his wife Theano secretly agreed to steal the statue from the temple and exchange it for a great amount of money.32 We find this version very convincing because it would account for several details found in the Iliad and still unexplained: in Book 3, Antenor describes Odysseus’ sublime eloquence in the famous Teichoscopia, comparing his words to a snow storm.33 Was Antenor referring to his own experience of being swept away by this verbal storm? In Book 6, Hector asks his mother 32 Dionysios Halicarnassensis 1.46 could contain an allusion to this version: Ἰλίου κρατηθέντος ὑπʼ Ἀχαιῶν, εἴτε τοῦ δουρείου ἵππου τῇ ἀπάτῃ, ὡς ‘Ομήρῳ πεποίηται, εἴτε τῇ προδοσίᾳ τῶν Ἀντηνοριδῶν εἴτε ἄλλως πως, 33 Il.3.222 ἔπεα νιφάδεσσιν ἐοικότα, see Létoublon 1994.

156

françoise létoublon

to offer a sacrifice to Athena. She brings her most splendid veil as a gift to be offered to the goddess by her priestess Theano. The sacrifice seems in per­ fect accordance to the ritual, but the goddess rejects it (Il. 6.311 . . . ἀνένευε δὲ Παλλὰς Ἀθήνη): this can be considered a very bad omen for the city of Troy. One could assume that, even if she is a woman, Hecuba had no right to enter the adyton; therefore she asked the priestess to take care of the prayer and the offering on her behalf. But on the other hand, Ath­ ena cannot accept a veil coming from Paris’s treasure chest. Is the statue mentioned in this occurrence the palladion or another cult object? The question remains open. Later on, some authors thought that Diomedes and Odysseus stole two statues (the palladion from Dardanos’ time and a later one); others sug­ gested that they stole only one statue, but that is was the fake. It then could be inferred that Antenor and Theano were more clever than the polymetis Odysseus himself since they sold a fake statue for a high price.34 Anyway, the legend of the Trojan origins of Rome is based, among other elements, on the idea that Aeneas carried away with him, apart from his father on his back, his wife, his son, the Penates, and the original palla­ dion thrown by Zeus: this statue was then kept by the Romans in Vesta’s temple, if we follow Dionysius. The story takes on a sophistic turn that, as far as we know, received no comment on from ancient authors: if the statue was a guarantee of safety for the city, and if Diomedes and Odysseus took only a fake palladion, a copy, while the original remained in the temple, why did Troy fall after all? From the Achaean point of view, too, the palladion is the key to their victory, and once it is removed from its adyton they will somehow replace it with the wooden horse, another trap, but one which reverses the move from outside to inside. The main feature is, of course, that with perfect fakes, the genuine cannot be distinguished from the copy. Let us go back to the text of the Bibliotheca and its narrative of the stat­ ue’s creation by Athena. According to the mythographer, the goddess was raised as a child by Triton,35 with his daughter Pallas. Both of the girls used to play war games (ἀσκούσας τὰ κατὰ πόλεμον) and once came to a quarrel. 34 The sophistic issue mentioned above seems still more acute in this case. Did Theano, as the priestess of Athena, betray her city and her goddess, or did she act skillfully, in collusion with her husband, deceiving the skillful Odysseus himself? In either case, the city’s fall shows their choice was mistaken. 35 ϕασὶ γεννηθεῖσαν τὴν Ἀθηνᾶν παρὰ Τρίτωνι τρέϕεσθαι. This specific mention could explain the frequency of Tritogeneia as appellation of Athena in our texts, be they Greek or Latin.



athena and pallas, image, copies, fakes, and doubles

157

As Pallas was going to strike Athena, the young goddess was frightened. Zeus also felt afraid for her and he stretched the aigis before her as a pro­ tection. Pallas, startled, looked up, was striken by Athena, and fell down dead. Deeply sorry for her death (περίλυπον ἐπ’ αὐτῇ γενομένην), Athena made an image resembling her (ξόανον ἐκείνηc ὅμοιον καταcκευάcαι), put the aigis she was afraid of around her shoulders, and placed it besides Zeus to be honoured there (καὶ περιθεῖναι τοῖc cτέρνοιc ἣν ἔδειcεν αἰγίδα, καὶ τιμᾶν ἱδρυcαμένην παρὰ τῷ Διί). But later, Electra, the daughter of Atlas, after her rape, sought refuge next to the image, and Zeus, with Ate, threw it into the Ilian countryside (ὕcτερον δὲ Ἠλέκτραc κατὰ τὴν φθορὰν τούτῳ προcφυγούcηc, Δία ῥῖψαι μετ᾿ αὐτῆc/ ᾽Ατης καὶ τὸ παλλάδιον εἰc τὴν Ἰλιάδα χώραν), where Ilos built a temple for it and honoured it (Ἶλον δὲ τούτῳ ναὸν καταcκευάcαντα τιμᾶν). Almost every word counts in this short narrative: the statue is a xoanon, which means a wooden archaic statue; it is rather small in size (three cubits: tripekhus) and could easily fit in human arms. The statue was endowed with the attributes of the goddess (δόρυ, ἠλακάτη, ἄτρακτος); the first one, the spear, is not surprising for her, since she was competing with Pallas in war matters, but the distaff and spindle are interesting. Could it mean that grieving Pallas leads this warrior in the direction of an Athena Ergane? It may also be a specifically Ionian image as some archaeological findings suggest.36 The statue is clearly conceived with the idea of replacing the dead friend and allowing grieving and solace. This corresponds to ideas cur­ rent in Greek literature about the birth of the statues and paintings in general as shown by the myths of Protesilaos and his wife Laodamia and the myth of Boutades’ daughter: the deceased’s wife or the fiancée of a soldier leaving for war tries to keep at least an image of the beloved one.37 In addition to this essential feature, we may note a childish tone that could show that making an imitation (ἐκείνηc ὅμοιον) of her friend may also be a way of playing with a doll for young Athena.38 We particu­ larly note the detail of the aigis as a garment. Therefore the palladion is first an image of the beloved Pallas and a solace for her death, and as an image, it is a substitute for the person. But still earlier in the tale, when 36 See Villing 1998. 37 See esp. Bettini 1992 and Steiner 2001. 38 Bettini devoted a whole chapter to dolls in antiquity. They are considered a symbol of virginity, which fits particularly well for the virgin goddess playing in her childhood with another girl close to her.

158

françoise létoublon

playing with her living friend, Athena may be considered as playing with her own double: she is a girl, more or less of the same age, and together they play war games with a strong aspect of competition and danger. It is well-known that children often invent imaginary companions to play with. In this story, young Athena had an actual friend who was a kind of double for her, who provided the education of companionship, and she replaces her with an image, another kind of double. We meet with the subtle notion of the double in mythology, as studied by Maurizio Bettini in a paper published in 2004 and by Ilaria Sforza in a book published in 2007. Bettini distinguishes two kinds of creation of doubles in Greek myths, as proceeding either from defence or from attack. Here, Pallas is from the beginning a double for Athena, as her foster-father Triton is a double and substitute for Zeus. Zeus and the aigis defend Athena against the potential violence of her double, and Athena makes an image of her, which can be seen as a defence of virginity.39 Fighting against her own double is the kind of challenge that Athena likes to face.40 The question we wish to pose is the analogy between the notions of fake, double, and image: after the death of her double, Athena makes an image of her which has very strange powers, hence a strange story, full of mysteries and rid­ dles, and this image provokes many fakes and delusions. Homer’s silence on the topic may be part of the problem. Pallas as Athena’s double is incorporated into Athena’s persona when her name becomes one of her favorite epikleseis, and the palladion pre­ serves the memory of their history. But Pallas’ name also occurs frequently as a masculine name for other characters in Dardanos’ journeys; mytho­ logical matters can be very complex and entangled. Athena’s and Pallas’ history of friendship and grieving in the Bibliotheca provides an astonishing reduplication of doubles and fakes. It allows a reflection on the image as replacing a beloved absent person and on the birth of both art and religion as the sanctification of this image. This paper started with the idea that fake in art is a more complicated problem than in literature. However, since works in the visual arts largely disappeared, we may sometimes admire ancient fakes, calling them for instance “Roman copy of a Greek original.” But the texts sometimes reflect

39 However, the examples of Electra, in this version, and of Cassandra raped by Ajax although she sought refuge near the palladion, show that its protection is not sufficient. 40 For instance in the myth of Arachne as told by Ovid.



athena and pallas, image, copies, fakes, and doubles

159

the question, evoking some cases of an original and its copies.41 As in a mirror gallery, we know too many doubles of a single image of Pallas to believe in one and the same item. If the copy perfectly imitates the original, nobody can distinguish one from another after time. Moreover, the sacral character of gods’ images made the problem much more than a mere artistic quarrel: religion and politics gave the palladion such a price that they multiplied the so-called originals to every city in Magna Grecia and Italy that claimed parenthood with the once aporthetos polis. References Allen, N. (2001), “Athena and Durga: Warrior Goddesses in Greek and in Sanscrit Epic,” in Deacy-Villing, 367–382. Anderson, M.J. (1997), The Fall of Troy in Early Greek Poetry and Art, Oxford. Augé, M. (1982, 20082), Génie du paganisme, Paris. Bernabé, A. (1996), Poetae epici Graeci. Testimonia et fragmenta, I, Stuttgart-Leipzig. Bettini, M. (1992, 2008), Il ritratto dell’amante, Torino. —— (2004), “Construire l’invisible. Un dossier sur le double dans la littérature classique,” Mètis n.s. 2: 217–230. Borgeaud, Ph. (2004), Aux origines de l’histoire des religions. Paris. Burgess, J.S. (2001), The Tradition of the Trojan War in Homer & the Epic Cycle, Baltimore. Burkert, W. (1987), Ancient Mystery Cults, Cambridge Mass. Buxton, R.G.A. (1980), “Blindness and Limits: Sophokles and the Logic of Myth,” JHS 100, 22–37. Cameron, A. (2004), Greek Mythography in the Roman World, Oxford. Carrière, J.-C. and B. Massonie (1991), La Bibliothèque d’Apollodore, traduite, annotée et commentée, Paris. Cassimatis, H. and P. Demargne (1984), “Athéna,” LIMC 2 Zürich: 67–111 and 955–1044. Chantraine, P. (1968–1980 [20092]), Dictionnaire étymologique de la langue grecque. Histoire des mots, achevé par Jean Taillardat, Olivier Masson et Jean-Louis Perpillou, avec en supplément les Chroniques d’étymologie grecque rassemblées par Alain Blanc, Charles de Lamberterie et Jean-Louis Perpillou, Paris. Conio, G. (2001), Figures du double dans les littératures européennes, études réunies et dirigées par G. Conio, Paris. Cosmopoulos, M.B. (2004) (ed), The Parthenon and its Sculptures, Cambridge. Davies, M. (1988), Epicorum Graecorum Fragmenta, Göttingen. Davreux, J. (1942), La légende de la prophétesse Cassandre d’après les textes et les monuments, Liège. Deacy, S. and A. Villing (2001) (eds.), Athena in the Classical World, Leiden. Demargne, P. (1984), “Athéna,” LIMC II, Zürich-München: 965–969. Dodds, E.R. (1951), The Greeks and the Irrational, Berkeley.

41 For instance, I suggest some ancients passages and texts for reflection on the pal­ ladion: Apollodorus, Bibliotheca, I: 142–146, III: 144–145; Dionysius of Halicarnassus, I 66, 68–9, II 66; Plutarch Parallela Minora 309f (Derkyllos and Aristides of Miletos); Proclus Chrest. 206 (Iliadis Parvae enarratio and Iliupersidos enarratio); Quintus of Smyrna, 10: 350–360; Virgil, Æn. 2: 171–179; Dictys Cretensis 5.5–7 (focus on Antenor and Theano as traitors).

160

françoise létoublon

Dowden, K. (1992), The Uses of Greek Mythology, London. Dumézil, G. (1995), Le roman des jumeaux et autres essais: vingt-cinq esquisses de mythologie, Paris. Faraone, Ch. (1992), Talismans and Trojan Horses: Guardian Statues in Ancient Greek Myth and Ritual, Oxford. Fromentin, V. (1998), Denys d’Halicarnasse, Antiquités romaines, Introduction générale, livre I, Paris. Gantz, T. (1993), Early Greek Myth: A Guide to Literary and Artistic Sources, Baltimore. Grafton, A. (1990), Forgers and Critics: Creativity and Duplicity in Western Scholarship, Princeton. Graziosi, B. and J. Haubold (2005), Homer: The Resonance of Epic, London. Horsfall, N. (2008), Virgil, Aeneid 2: A Commentary, Leiden-Boston. Hurwit, J.M. (1999), The Athenian Acropolis: History, Mythology, and Archeology from the Neolithic Era to the Present, Cambridge. Kowalzig, B. (2007), Singing for the Gods: Performances of Myth and Ritual in Archaic and Classical Greece, Oxford. Létoublon, F. (1985), “Les dieux et les hommes. Le langage et sa référence dans l’Antiquité grecque archaïque,” in K. Boudouris (ed.), Language and Reality in Greek Philosophy, Athènes: 92–99. —— (1994), “Le bon orateur et le génie selon Anténor dans l’Iliade: Ménélas et Ulysse,” in La rhétorique grecque. Actes du colloque Octave Navarre, Nice, 29–40. —— (2009), “Athéna et son double,” Pallas 81, Kaina pragmata. Mélanges offerts à JeanClaude Carrière, ed. M. Bastin-Hammou and Ch. Orfanos, 179–190. —— (2010), “To See or Not to See: Blind People and Blindness in Ancient Greek Myths,” in M. Christopoulos and E. Karakantza et al. (eds.), Light and Darkness in Greek Myth and Religion, Lanham: 167–180. Liberman, G. (1991), Alcée. Fragments, Paris. Loraux, N. (1990, 2002), “Qu’est-ce qu’une déesse?,” in Histoire des femmes en Occident. I. L’Antiquité, ed. P. Schmitt Pantel, Paris: 39–79. Lowenstam, S. (2008), As Witnessed by Images: The Trojan War Tradition in Greek and Etruscan Art, Baltimore. Malamoud, Ch. (2002), Le jumeau solaire, Paris. Martinez, J. (2011), “Epic Fake? Forgery, Fraud, and the Birth of Philology,” in J. Martinez (ed.), Fakes and Forgers of Classical Literature, Madrid: 15–23. Neils, J. (1996) (ed.), Worshipping Athena: Panathenaia & Parthenon, Madison. Osborne, R. (1996), Greece in the Making: 1200–470 BC, London. Page, D. (1974), Supplementum Lyricis Graecis, Oxford. Ridgway, B.S. (1992), “Images of Athena on the Akropolis,” in J. Neils (ed.), Goddess and Polis: The Panathenaic Festival in Ancient Athens, Princeton: 119–142. Robertson, N. (1992), Festivals and Legends: The Formation of Greek Cities in the Light of Public Ritual, Toronto. —— (1992), “Athena’s Shrines and Festivals,” in J. Neils (ed.), Worshipping Athena: 27–77. —— (1996), “Athena and Early Greek Society: Palladium Shrines and Promuntory Shrines,” in M. Dillon, Religion in the Ancient World: New Themes and Approaches, 383–475. —— (2001), “Athena as Weather Goddess: the Aigis in Myth and Ritual,” in Deacy/Villing (eds.), Athena in the Classical World, 29–55. —— (2004), “Pandora and the Panathenaic Peplos,” in Cosmopoulos (ed.), The Parthenon and Its Sculptures, 4–113. Scully, S. (1990), Homer and the Sacred City, Ithaca. Severyns, A. (1928), Le Cycle épique dans l’école d’Aristarque, Paris. Sforza, I. (2007), L’eroe e il suo doppio: uno studio linguistico e iconologico, Pisa. Stanford, W.B. (1968), The Ulysses Theme: A Study in the Adaptability of a Traditional Hero, Oxford.



athena and pallas, image, copies, fakes, and doubles

161

Steiner, D. (2001), Images in Mind: Statues in Archaic and Classical Greek Literature and Thought, Princeton. Stochos, Ch. (2001), “Samothrake II, Religion,” in Der Neue Pauly: Enzyklopädie der Antike, 11: 27–29. Teffeteller, A. (2001), “Greek Athena and the Hittite Sungoddess of Arinna,” Athena in the Classical World, in Deacy-Villing, 349–365. Vernant, J.-P. (1990), Figures, idoles, masques, Paris. Vian, F. (1969), Quintus de Smyrne, La Suite d’Homère, texte et traduction, Paris. Villing, A. (1998), “Athena as Ergane and Promachos. The Iconography of Athena in archaic east Greece,” in N. Fisher and H. van Wees, Archaic Greece: New Approaches and New Evidence, London-Swansea: 147–168.

Hippias of Elis: Lessons from One Master Forger Javier Martínez Introduction* Since the inception of writing, unscrupulous individuals have sought to use the medium to deceive others for their own gain. As a matter of fact, in ancient Mesopotamia, shortly after the invention of writing, cylinder seals became exceedingly popular, not just for kings, but for local businessmen and farmers as well. The seal was an attempt to prevent others from forging receipts or other documents to claim property as their own. In Classical Greece, forgers flocked to the literary masters recognizing a demand for their work and eyeing a profit by means of supplying that demand themselves. As technology has improved, more complex means of committing and preventing forgery have been invented. But more than competence in the latest technological innovations, successful forgery relies on the forger’s intuitive knowledge of human psychology. Like the con artist, the forger must know what instills and engenders trust in his audience or readers and how to maintain that trust throughout the deception. Unlike the continual development of the technology associated with forgery, the nuances wrapped up in the psychology behind a good forgery have remained the same throughout the centuries. This leads to an interesting paradox. “The only understanding we have of deception is from failed deceptions or those revealed by their perpetrators” (Mitchell, 851). But, arguably, the most successful forgers are those who have never been caught. Studying successful historical forgers thus requires that there not be a scholarly consensus as to their status as forgers. One historical figure who has generated a great deal of scholarly discussion with regard to the legitimacy of his work is Hippias of Elis. Around the turn of the fourth century bc, Hippias wrote a work entitled Olympionikon Anagraphe the heart of which contained a list of Olympic victors purported to extend back to the first Olympics.

* This paper is based on research funded by project FFI2009-09465 of the Spanish Ministry of Science and Innovation.

164

javier martínez On the whole one can distinguish between optimists who believe that Hippias found an official list in the temple archive at Olympia which he then published (Asheri, Begston, Finley and Pleket), and pessimists or skeptics who declare Hippias of Ellis to be a liar and forger who invented the names at least before the 6th century bce (Mahaffy, Körte, Beloch, Peiser, Bilik) (Möller, 2004, 173).

The use of the term “forgery” may appear confusing in this context because the scholarship does not question the authorship of this work, but rather the integrity of its content. The reason for the use of the term “forgery” in relation to the Olympionikon Anagraphe is in the implied source material. Hippias provided names of victors somewhere around 300 years before his time. There are many types of sources that may have been available to Hippias. What places Hippias in the forger category for several scholars is the idea that he invented his information and referred to nonexistent sources. The definition provided by Mitchell in his study of human deception is as follows: . . . a proposition P is a deceptive pragmatic presupposition of the person in a given context when the person does not assume or believe that P, but does assume or believe that the addressee assumes or believes that P, and does assume or believe that the addressee assumes or believes that the person is making these assumptions or has these beliefs (Mitchell, 1996, 824).

Applying this understanding of deception (forgery) to the work of Hippias yields the following premises. 1) Hippias did not assume or believe that his Olympian victor list was accurate and he was aware that he had invented the information contained therein. 2) Hippias did assume that his readers would believe that his Olympian victor list was accurate. 3) Hippias also assumed that his readers believed that Hippias himself believed in the accuracy of his Olympian victor list. When viewed in this way, Hippias of Elis was perhaps among the greatest forgers of all time. The eventual effect his work had upon Greek history was monumental. A few centuries after he wrote, Greek scholars would continue to use his list as the basis for reckoning time throughout the Greek world. His forgery was so subtle and so well-constructed that modern scholars to this day continue to debate the authenticity of his work. This study will examine Hippias of Elis as a forger and highlight the ways in which he managed to manipulate his audience into accepting his work as genuine. While most studies of forgery focus on the technical and technological aspects of a forged work, this study will focus on the



hippias of elis: lessons from one master forger

165

psychological nuances upon which Hippias relied to deceive his audience. A close look at his work reveals that much of its success was due to his not becoming entangled in the traps that led to the discovery of many forgers since his time. Since forgery is a crime “with intent,” this paper will explore various principles that guided Hippias as they relate to the aspects of means, opportunity, and motive that underlie such crimes. The Question of Means Principle #1: Rest on Your Own Laurels Hippias was an Elean aristocrat and could easily have played up his contacts with the Elean officials who ruled over Olympia. For later forgers this became a prominent feature in their writing and often led to their discovery and downfall. Grafton describes the typical way this feature appears in the works of forgers, as follows, “the motif of the object found in an inaccessible place, then copied, and now lost, as the authority for what would have lacked credibility as the work of an individual” (1990, 9). Hippias was in the ideal situation to make use of this typical forger technique. As an Elean he could have claimed access to the temple of Zeus. It would have been easy for him to describe an ancient text, inscription, or stele upon which the names of the Olympic victors were written and claim it as his source along with an accompanying story about how it was lost. Granted, he would have needed to convince the priests to “play along,” but such a sacred fable for the greater good may not have been a tough sell for the Elean priests. An almost contemporary example of this type of legitimation story took place just across the Mediterranean Sea when King Josiah of Judah found a scroll in the Jerusalem temple. The priests of the temple were not only aware of the ruse; they may have used King Josiah as a fall guy for their forgery. But Plutarch, the only extant source mentioning Hippias, makes it clear that the credibility for his work rested with Hippias alone, ὧν τὴν ἀναγραφὴν ὀψέ φασιν Ἱππίαν ἐκδοῦναι τὸν Ἠλεῖον, ἀπ᾽ οὐδενὸς ὁρμώμενον ἀναγκαίου πρὸς πίστιν . . . (Numa 1.4) “the list of which is said to have been published at a late period by Hippias of Elis, who had no fully authoritative basis for his work” (trans. B. Perrin). The reason Plutarch can critique Hippias for having “no fully authoritative basis for his work” stems from the fact that Hippias did not cite his sources for his list, either oral or written. He bore the burden of trust solely on his own shoulders.

166

javier martínez

It would seem that Hippias was confident enough to believe that his own authority was sufficient for his work. His qualifications for the job do sound rather impressive. He had a close familiarity with the Olympic Games, since he had attended on numerous occasions (Hippias Minor 363c–d). He had served as a trusted ambassador to Sparta for Elis. In addition to this, he was known for compiling catalogues of names and places. Principle #2: Don’t Imitate; Innovate Most forgers have a particular individual after whom they pattern their work. This was not the case with Hippias. There is no argument as to whether Hippias’ victor list came from anyone other than Hippias himself. Not only that, but his style was rather unique. Wacker argues that he “probably didn’t collect the names of victors for chronological interests, a scientific field not really developed in his times . . .” (1998, 43). One of the biggest problems with this objection is that the influential historian Thucydides writing on events of direct concern to Hippias was already using Olympic victors as a way to anchor the dates of historical events. Although the event that Thucydides is reporting directly involves an Olympic festival, the import and the possibilities for such a list as a historical resource would have been evident. Wȩcowski, on the other hand, argues for a very specific literary tradition of which Hippias was a part. In the fourth century there were a number of Greek writers producing various lists of yearly events that could serve to mark time. These lists involved various institutions that were independent from the greater political spectrum that concerned most historians. Many such lists were mainly of local importance and did not connect in any significant way with other lists. These lists surely had a chronological aim, even though their scope was quite limited geographically. Wȩcowski argues that Hippias produced his Olympionikon Anagraphe in direct response to the publication of Priestesses of Hera in Argos by Hellanicus of Lesbos (2012). By being himself embedded within a larger literary tradition, Hippias is able to present himself as an author in his own right. Plato’s comments about Hippias indicate that he had a reputation for rattling off long lists of names. The fact that another literary work of his, the Ethnon Onomasiai, with geographic names, was entirely reliable had further established his credibility. This left readers with a set of implicit assumptions about Hippias and his credibility as an author upon which he could then



hippias of elis: lessons from one master forger

167

prey. Mitchell observes that “The more the deceiver [forger] relies on the victim’s [reader’s] implicit assumptions and the less the victim [reader] infers motivation on the deceiver’s [forger’s] part in setting up the signs of regularity and preying on the victim’s [reader’s] implicit assumptions, the less likely is the victim to become suspicious” (1996, 821). Principle #3: Build a Counter-Narrative Many forgers make the mistake of relying too heavily on the information accepted unquestionably by their contemporaries. Contemporary or later critics are then able to demonstrate this literary dependence once that information is later discovered to have been inaccurate. Skilful con-artists have the ability to weave a counter-narrative so compelling that their victims are unable to be dissuaded from it. One infamous con-artist “. . . knew that the more improbable his inventions sounded, the more likely people were to believe that he could not possibly be lying” (Mitchell, 1996, 845). Such appears to have been the case with Hippias’ description of the gradual introduction of the events in the Olympic Games. His counternarrative soon displaced the standard narrative current in his day. Although the Iliad does not reference the Olympic Games directly, Körte (1904) notes that the funerary games mentioned therein bear a remarkable resemblance not only to the Olympic Games, but also to the Isthmian, Nemean, and Pythian games. Neither Mahaffy (1881) nor Körte (1904) can understand how foot-races could have continued to be the only events for sixty-eight years at a festival that attracted contestants from across the Peloponnesus. But since the Iliad is silent concerning the Olympic Games directly, this evidence is indirect at best. More direct evidence comes from Pindar, who wrote Olympic odes for the victors about a generation before Hippias. In his praise of current victors, Pindar harkens back to some of the victors of the earliest Olympic festivals. Pindar knows nothing of a gradual introduction of Olympic events. Gardiner went so far as to say, The first part of this account is obviously absurd in view of the evidence given above for funeral games. There can be no doubt that in the first Olympiad the programme included at least all the events described by Pindar, the foot-race, the diskos, the spear, boxing, wrestling, and the chariot-race (1910, 52).

But whereas the evidence from Homer is indirect, Brinkmann (1915) questions whether the evidence from Pindar, which is mytho-poetic in nature, should be read as historical.

168

javier martínez

This brings us to the evidence from the historian Thucydides, another forerunner of Hippias. Thucydides uses Olympic victors to secure dates for historical events in a similar way to what would become the norm among Greek writers. In describing a trip of envoys from Mytilene to Olympia, Thucydides says, ἦν δὲ Ὀλυμπιὰς ᾗ Δωριεὺς Ῥόδιος τὸ δεύτερον ἐνίκα (3.8) “It was the Olympiad in which the Rhodian Dorieus gained his second victory” (trans. R. Crawley). But Dorieus was the victor in the pancratium, not in the stade, which is in contrast to the habitual practice of giving the name of the stadion winner. For Busolt (1893), this demonstrates that the stade victor had not yet been established as the standard reference point in regard to the Olympics. It is only after Hippias creates his list that Philistus of Syracuse uses the stadion victors as reference points. As a matter of fact, Philistus is the first known source for assigning the stadion winer as a reference for a particular Olympiad (FHG II:186 F6, 11 f.). But the power of such a counter-narrative is immediately evident in the scholarship defending the authenticity of the importance of the stadion in the origin of the Olympic Games. Both Gardiner and Lee demonstrate how effective a counter-narrative can be psychologically. For Lee, “no obvious reason comes to mind why anyone should invent so elaborate a series of dates for the evolution of the program” (2002, 111). The only reason Gardiner can come up with for such a fabrication is that “. . . the compiler dated the introduction of each new event from the first occasion on which he found mention of it” (1910, 52). But viewing Hippias as a master forger brings the motive behind Hippias’ creative story into clear focus. It is another means to present his present his “new” information in a compelling way. It was so compelling, in fact, that generations of Greek historians after him continued to use his list as definitive and no rival list ever appeared to challenge his authority. But the power behind Hippias’ counter-narrative lay not only in the fact that other scholars could find no compelling reason for him to create such a tale in the first place. Hippias knew from years of studying human behavior that “to generate fictional truths, the observer [reader] is to infer other propositions which are congruent with the representation, much as when a listener attends to the propositions in a story to make inferences” (Mitchell, 844). Scholars throughout the past century, including Weniger (1904), Burkert (1985), Siewert (1992), Brulotte (1994), Scanlon (2002), and Valvanis (2006), have clearly made such inferences, finding further support and legitimacy for his counter-narrative from ancient sources. They have each argued with slightly different nuances that the origins of the Olympic festival were primarily religious in nature with the sacrifice



hippias of elis: lessons from one master forger

169

to Zeus as the primary component. In his Gymnasticus, the third century Greek sophist Philostratus describes the invention of the stadion race (5). This description places the stadion squarely in the service of the cult of Zeus and removes it from the sphere of other Greek athletic contests that are often brought to bear on the question of the origin of the Olympic Games. The emphasis on Philostratus’ description provides an alternative and seemingly plausible explanation for why the stadion race would have been the only contest for many years and why the other non-cultic events would have only been added gradually later (Valvanis, 2006, 141). The stade was a means for Zeus to select the most athletic male to light the sacrificial fire with the torch. Other scholars have latched onto slightly different types of evidence to support the primacy of the stade in the Olympic Games. Weniger notes the comparison here between the early Olympic Games and the Heraean games for girls, which were religious in nature, where the stade was the only race (1905, 191, n. 2). Brinkmann (1915) also argued that an Attic inscription supports Hippias’ depiction of the stade as the only original race. Finally, Ziehen (1915) sees corroborating evidence in the fact that in the Panathenaic games, it is the stade victors who receive the highest prize. As for Pindar, the events he describes in his Odes were all part of the program during his day at the beginning of the fifth century bc. The fact that Pindar retrojects this group of events back to the beginning of the Olympic Games would not be surprising if the gradual introduction of events at Olympia was neither general knowledge in the Greek public nor the common received tradition. Thucydides’ citation of the victor of the pancration rather than the stadion can be seen in a similar way. Wilamowitz-Moellendorf (1922) explains away Thucydides’ use of the pancration victor as a natural tendency to select the most famous victor for reference before a general convention had been established. But the idea that a common received tradition is not reliable should not surprise ancient historians. The Question of Opportunity Principle #4: Include a Few Verifiable Details One of the earliest modern scholars to scrutinize Hippias’ Olympic victory list was Mahaffy. He questioned the authenticity and reliability of the list and concluded that he could find no fault with all of the material after

170

javier martínez

about 600 bc, or with the information after the forty-eighth or fiftieth Olympiad (Mahaffy, 1881, 175–76). In other words, Hippias’ forgery was limited to the first half of his list. But Hippias was not content to simply invent the entire first portion of his list. He made sure that he included names of verifiable victors in the early portion of his list. Centuries later Pausanias verified at least four of the victors in the early portion of Hippias’ list pointing out commemorative inscriptions of various kinds for interested tourists. The winner of the first Olympiad, Coroebus [Koroibos] of Elis, was buried on the outskirts of Elis (Pausanias, Description of Greece, 5.8.6); the winner of the sixth Olympiad, Oebotas of Dyme, had his gravesite in Dyme (Pausanias, Description of Greece 7.17.13); and the winner of the fifteenth Olympiad, Orsippus of Megara, was buried in Megara (Pausanias, Description of Greece, 1.44.1). For the latter, one need not rely on Pausanias’ account alone, for in 1769 archaeologists found the very epitaph to which Pausanias refers. Simonides (ca. 556–468 bc) is held to be the author of this epitaph, which would make it available to Hippias as a possible source (Hicks and Hill, 3–4). In addition to these gravesites, monuments at Olympia itself commemorate the victories of some of the Olympic victors Hippias listed prior to the fiftieth Olympiad. The inhabitants of Achaea erected a statue commemorating the Olympic accomplishments of Oebotas of Dyme during the eightieth Olympiad, when they learned from the Delphic Oracle that he had pronounced a curse on their region. The inscription at the foot of the statue read Οἰνία Οἰβώτας στάδιον νικῶν ὅδ᾽ Ἁχαιὸς / πατρίδα Πάλειαν θῆκ᾽ ὀνομαστοτέραν (7.17.7) “This Oebotas, an Achaean, the son of Oenias, by winning the foot-race, / Added to the renown of his fatherland Paleia” (trans. W.H.S. Jones and H.A. Ormerod). Pausanias also notes it: ἀνάκειται δὲ τῇ ἐν Ὀλυμπίᾳ καὶ στήλη λέγουσα τοῦ Λακεδαιμονίου Χιόνιδος τὰς νίκας. (6.13.2) “There is also set up in Olympia a slab recording the victories of Chionis the Lacedaemonian” (trans. W.H.S. Jones and H.A. Ormerod). This Spartan athlete was the victor in the stadion-race in three consecutive Olympiads. The Spartans altered his inscription at Olympia in 480 bc when Astylus presumably beat Chionis’ record by not only winning the stade and diaulos in three consecutive Olympics, but also emerging victorious in the hoplitodromos. The revised inscription emphasized the fact that the race in armor had not yet been added when Chionis competed, as Pausanias states (6.13.2).



hippias of elis: lessons from one master forger

171

Given the other short notices that accompany victor entries, if Hippias were looking to authenticate his writing, a short note next to any of these entries highlighting the fact that interested observers could visit the relevant monument for these victors “to this day” would surely have lent him credibility among his ancient readers. This is the approach taken in the Deuteronomistic History on multiple occasions and could have easily been employed by Hippias of Elis.1 This is part of the brilliance of Hippias of Elis. Whereas other forgers would go out of their way to legitimate their own work, Hippias knew that a better strategy would be to have others legitimate his work for him. But it was only Hippias’ deep understanding of human psychology that allowed him to pull off his deception. Most forgers and con-artists will not trust a third party that is not privy to their deception. In order for Hippias’ ploy to work, he had to rely on the regularity principle. This principle describes the human capacity to generalize from specifics to the general. Mitchell explains the way in which deceivers use the regularity principle as follows, “Any time a regularity is known or expected, the sign of the regularity can be used to deceive” (1996, 826). Providing a small number of verifiable names within his list creates the illusion for his audience that the entire list is trustworthy. The Question of Motive Principle #5: Seek No Compensation for Your Work In any crime, one of the factors prosecutors look for is motive. The profit motive seems to be one constant that drives forgers both ancient and modern. Admittedly, many legitimate writers make a living from their writing and so generating income from a written work is not by itself an indication of “foul play,” but it certainly is a common motive for many forgers. In the case of the Olympic victory list of Hippias, there is no indication that he was paid by anyone for this work. Our sources do not indicate that such payments were customary for literary works at that time. This

1 The most recent and exhaustive discussion of the use of this phrase in the Deuterono­ mistic History appears in Geoghegan, 2006.

172

javier martínez

may be one factor that was not much in Hippias’ control, but the lack of any type of official commissioning for his literary work certainly helps to avert suspicion. Principle #6: Do Not Draw Attention to Your Agenda One psychological phenomenon that deceitful individuals can rely on is that a victim will hold even more strongly to a belief that they themselves have arrived at through deductive reasoning than they will to a belief they are asked to accept after it is all laid out for them. If the victim has put together pieces of a puzzle in his/her own mind, that victim will be much more invested in the resulting belief. One can observe the way in which Hippias uses this tool of deception on multiple occasions. One example is with his choice of when to start the series of Olympic victors. Sinn has brought up the argument that Hippias chose 776 bc as the starting date for the Olympics because there was so much significance to the Olympic festival celebrated in 476 bc (1991, 31–54) According to Sinn, there were three significant events that took place during the seventy-sixth Olympiad held in 476 bc. After the famous Battle of Thermopylae in 480 bc and several other battles, the Greeks had managed to fend off the attempted Persian invasion and keep them out of the Peloponnese, and this Greek victory over the Persians would surely have been celebrated at the next pan-Hellenic Olympic Games. In addition to celebrating their military success, they had constructed a new Temple of Zeus at Olympia and had widened the Olympic stadium. Hippias himself was not born until 460 bc and would not have had a firsthand knowledge of this monumental Olympiad, but would likely have heard the oral testimony from an older generation. According to Sinn, this made Hippias’ choice of 776 bc as the first Olympiad quite natural as it fell 300 years and 75 Olympiads prior to this momentous occasion. This is where the genius of Hippias and his innate understanding of human psychology become clear. Most forgers with an agenda call the reader’s attention to it. They not only make the connection for the reader, but they emphasize their main point in various ways to ensure that the reader does not miss their intended message—Hippias knew better. The most likely reconstruction of Hippias’ entry for the Seventy-Sixth Olympiad reads as follows, Εβδομηκοστὴ ς’. Σκάμανδρος Μιτυληναῖος στάδιον. “76th Scamander of Mytilene stadion race”

(Moretti, 1957, 209).



hippias of elis: lessons from one master forger

173

Even the more extensive lists like POxy II 222 merely add more names and events (Christesen, 2007, 28). Hippias mentioned neither the monumental defeat of the Persians during the years of the previous Olympiad nor the great celebration of that victory during this Olympic festival. Hippias relied instead on his trust in the intellectual prowess of his readers and the connections that they would inevitably draw without his help. First, Hippias was confident that the name “Scamander of Mytilene” itself would be sufficient to call to mind the momentous Olympic festival that his readers had heard about on their fathers’ and grandfathers’ knees. Additionally, his readers would share Hippias’ penchant for numerology. Once they connected the name of Scamander with the famous Olympic festival, they would then be able to make the calculation of four years multiplied by seventy-five previous Olympiads to produce the significant number 300. With that number in mind, they could then deduce that the Olympics had been held for 300 years until the celebration of the 300 Spartans who faced the Persians at Thermopylae. Skeptics may argue that Hippias could not have known that his readers would make this connection, but this is precisely the point on which he relied. Skeptics may also point out that despite the massive amount of literature written about Hippias and the victor lists that he engendered, it has taken over two thousand years for any scholar to notice this “obvious” connection between 476 bc and 776 bc. Principle #7: Appear Impartial Although most all scholars agree that as an Elean, Hippias had a vested interest in his enterprise and wanted to promote the Olympic Games and Elis with the production of his list of Olympic victors, critics have found it difficult to demonstrate the partiality that they often expect to find there. Wacker finds it “quite astonishing that the victors of the disciplines newly introduced in the first century were Eleians” (1998, 46). But both statistically and from an intuitive perspective, the number of significant Elean victories is quite minimal. It has already been pointed out that the victory of Koroibus in the stade as the first “recorded” victor was corroborated in antiquity by his grave (presumably with a grave marker of some kind). Hypenus of Elis, the victor in the diaulos (double race), and Polynices of Elis are the only uncorroborated significant Elean victors in the entire first fifty Olympiads mentioned by Hippias. According to the received list, a Laconian won the first long-course race, the first wrestling match for men and for boys, and the first pentathlon for men

174

javier martínez

and for boys. An athlete from Smyrna won the first boxing match, an athlete from Thebes won the first four-horse chariot race, an athlete from Syracuse won the first pancration, an athlete from Thessaly won the first horse race, and a young athlete from Sybaris won the first boxing match for boys. Considering that the Eleans were the organizers of the games, it is probable that their athletes may have had more advanced notice prior to the introduction of a new event at an Olympic festival. This prior notice combined with the hometown advantage evident in modern sports contests and the modern Olympics would suggest that there is little that is remarkable about the presence of three first-time Elean victors among a dozen or so first-time victors overall. On the other hand, Beloch (1929) uses the fact that there are so few Elean victors in the early part of the list as a critique of the list’s authenticity. The number of Elean victors in the first fifty Olympiads is consistent with the number of Elean victors found in the later “reliable” portions of the list. In this way Hippias is easily able to pass the scrutiny of questions about local promotion. Those readers and critics looking to find a bolstering of the number of local Elean heroes in Hippias’ list are bound to come up empty-handed. He neither erases Elis completely from the early record nor does he insert Elis so frequently as to raise suspicion. Contributing Factors Principle #8: Bet on the Right Horse The rural area of Elis and Pisatis were notorious as grazing lands for horses. Thousands of bronze and terracotta votive figurines depicting cattle and horses dating from the tenth to the eighth century bc were found in the Temple of Zeus at Olympia (Barringer, 2010, 158). Indeed, chariot racing as a contest or sport was a part of Greek culture as early as the thirteenth century bc (Lee, 2002. fn. 6). Busolt (1893) cannot fathom why an Elean cavalier nobility would have waited twenty-five Olympiads before introducing the first chariot race into their Olympic Games. Körte (1904) finds this fact even more disconcerting given the archaeological evidence of halters and yokes for miniature, votive, two-horse chariots dating from the eighth century bc whereas, according to the received list, the twohorse chariot race was not even added to the games until the ninety-third Olympic festival.



hippias of elis: lessons from one master forger

175

What Körte seems to forget in leveling this critique is the generally accepted validity of the list after the fiftieth Olympiad. This suggests that the introduction of the two-horse chariot race as a new contest among the Olympic Games belonged to the recorded and verifiable period of the Olympic victor list. Weniger provides a counter-narrative of his own regarding the votive horses and chariots, arguing that they may have been dedicatory offerings to Zeus without any relationship to the Olympic festivals. Principle #7: Provide Minimal Details In the Olympic victor list that likely goes back to Hippias of Elis, the entry for the victor of the fifteenth Olympiad has an annotation: προσετέθη δόλιχος· καί γυμνοί ἔδραμον· “A long-course race was added, and they ran naked.” The victor for this fifteenth Olympiad was Orsippus of Megara. The entry goes on to describe the addition of the long race that includes a victor from Laconia, a Spartan city. It is within this additional information that the phrase “and they ran naked” appears. By Hippias’ day, athletic competitors in the Olympic Games competed naked, but Homeric tradition that spoke of athletes girding themselves before competition indicated that this tradition was not always so: ζῶμα δέ οἱ πρῶτον παρακάββαλεν, αὐτὰρ ἔπειτα δῶκεν ἱμάντας ἐϋτμήτους βοὸς ἀγραύλοιο. (Il. 23.683 f.) A girdle first he cast about him, and thereafter gave him well-cut thongs of the hide of an ox of the field (trans. A.T. Murray). ζωσαμένω δ᾽ ἄρα τώ γε βάτην ἐς μέσσον ἀγῶνα (Il. 23.710). Then the twain, when they had girded themselves (trans. A.T. Murray).

The way in which later writers interpreted this entry is telling. Dionysius of Halicarnassus interprets this tradition as indicating that running naked at the Olympic games was a Spartan innovation, connecting the practice with Acanthus the Lacedaemonian (a Spartan): τοῦτο καὶ εἰς ἐμὲ τὸ ἔθος ἐν Ῥώμῃ διέμενεν, ὡς ἐξ ἀρχῆς ἐγίνετο παρ᾽ Ἕλλησιν: ἐν δὲ τῇ Ἑλλάδι καταλέλυται Λακεδαιμονίων αὐτὸ καταλυσάντων. ὁ δὲ πρῶτος ἐπιχειρήσας ἀποδυθῆναι τὸ σῶμα καὶ γυμνὸς Ὀλυμπίασι δραμὼν ἐπὶ τῆς πεντεκαιδεκάτης ὀλυμπιάδος Ἄκανθος ὁ Λακεδαιμόνιος ἦν. τὰ δὲ πρὸ τούτων δι᾽ αἰσχύνης εἶχον ἅπαντες Ἕλληνες ὅλα γυμνὰ φαίνειν ἐν ταῖς ἀγωνίαις τὰ σώματα, ὡς Ὅμηρος τεκμηριοῖ, μαρτύρων ἀξιοπιστότατός τε καὶ ἀρχαιότατος ὢν ζωννυμένους τοὺς ἥρωας ποιῶν. (Roman Antiquities 7.72.2–3).

176

javier martínez This custom continued even to my time at Rome, as it was originally practised by the Greeks; but it is now abolished in Greece, the Lacedaemonians having put an end to it. The first man who undertook to strip and ran naked at Olympia, at the fifteenth Olympiad, was Acanthus the Lacedaemonian. Before that time, it seems, all the Greeks had been ashamed to appear entirely naked in the games, as Homer, the most credible and the most ancient of all witnesses, shows when he represents the heroes as girding up their loins. (trans. E. Cary)

Thucydides also claimed that it was the Spartans that introduced the practice of running naked to the Olympic Games (The Peloponnesian War, I.6). The idea that an innovation like nudity was introduced by one particular group when there is independent evidence for the same practice within that group in other settings makes eminent sense to the historian, both ancient and modern. Moreover, Scanlon (2005) argues for the Spartan innovation of athletic nudity based on multiple sources quite separate from Dionysius. Most forgers after Hippias would make the mistake of calling attention to such a connection in an attempt to legitimate their work. Hippias’ expertise was in the subtlety with which he carried out his craft. Some historians may object to this observation by countering that Hippias likely wrote a longer work in which his list was contained and that his larger work may indeed have called attention to this connection directly. While such a hypothesis is possible n the abstract, the literary evidence argues against it. According to Pausanias, it was not Acanthus who first ran naked, but rather Orsippus (Description of Greece, 1.44.1). He indicates that the gravesite of Orsippus is visible to visitors at Megara. The tradition in Megara can be seen in the grave epitaph itself, which reads,

5

Ὀρρίππω Μεγαρῆς με δαίφρονι τῆδε ἀρίδηλον μνᾶμα θέσαν, φάμα Δελφίδι πειθόμενοι· ὃς δὴ μακίστους μὲν ὅρους ἀπελύσατο πάτρα πολλὰν δυσμενέων γᾶν ἀποτεμνομένων, πρᾶτος δ’ Ἐλλάνων ἐν Ὀλυμπία ἐστεφανώθη γνμνός, ζωννυμένων τῶν πρὶν ἐνὶ σταδίω

For militant Orripos, the Megarians, in obedience to a Delphic oracle, set me up, a memorial seen clearly from afar. He freed the greatest boundaries for the fatherland, when the enemy were appropriating much land, and was the first of Greeks in Olympia to be crowned naked, when those of yore wore loincloths in the stadion. (trans. W.B. Tyrrell)



hippias of elis: lessons from one master forger

177

Pausanias makes it clear that the tradition of his day was not that Orsippus intentionally ran naked, but rather that his perizomata slipped off of him accidentally during the race. Pausanias then provides his take on the tradition, inferring that the slip was not accidental, but on purpose, in order to give him an advantage in the race. What can be inferred from both the epitaph and the later commentary by Pausanias is that the only thing passed down by the earliest tradition of the Olympic victor list is very close to what appears in Julius Africanus’ entry for the fifteenth Olympiad. Dionysius of Halicarnassus was familiar with a tradition that interpreted the phrase as referring to the long course race, whereas in Megara they interpreted the tradition as a reference to the stadion race. Such variant traditions would not have been possible if the original source and tradition had elaborated on this short side note to any extent. On the other hand, such a terse comment is prime fodder for later commentators. Conclusion Hippias of Elis stands out from other ancient forgers in various ways. Many scholars to this day do not recognize Hippias as a forger, but seek to defend the veracity of his work. His work demonstrates subtlety and insight into the psychology of the reader. His deception is skillfully hidden in virtually every facet of his work. He not only succeeded in deceiving his fellow Greeks for generations, but continues to this day to deceive modern scholars. References Barringer, Judith M. (2010), “Zeus at Olympia,” in Jan N Bremmer and Andrew Erskine (eds.), The Gods of Ancient Greece: Identities and Transformations, Edinburgh Leventis Studies 5, Edinburgh: 155–77. Beloch, Karl Julius (1929), “Die Siegerliste von Olympia,” Hermes 64: 192–98. Brinkmann, August (1915), “Die Olympische Chronik,” RhM 70: 622–37. Brulotte, Eric (1994), “The ‘Pillar of Oinamaos’ and the Location of Stadium I at Olympia,” American Journal of Archaeology 98: 53–64. Burkert, Walter (1985), Greek Religion, Cambridge, MA. Busolt, Georg (1850–1920), Griechische Geschichte bis zur schlacht bei Chaeroneia, Gotha. Christesen, Paul (2007), Olympic Victor Lists and Ancient Greek History, Cambridge. Collon, Dominique (1987), First Impressions: Cylinder Seals in the Ancient Near East, Chicago. Gardiner, E. Norman (1910), Greek Athletic Sports and Festivals, London. Geoghegan, Jeffrey (2006), The Time, Place, and Purpose of the Deuteronomistic History: The Evidence of “Until this Day,” Providence.

178

javier martínez

Grafton, Anthony (1990), Forgers and Critics: Creativity and Duplicity in Western Scholarship, London. Hicks, E.L. and G.F. Hill (1901), A Manual of Greek Historical Inscriptions, Oxford. Körte, Alfred (1904), “Die Entstehung der Olympionikenliste,” Hermes 39: 224–43. Lee, Hugh M. (2002), “The ‘First’ Olympic Games of 776 BC,” in Wendy J. Raschke (ed.), The Archaeology of the Olympics: the Olympics and Other Festivals in Antiquity, second edition, Madison: 110–18. Mahaffy, J.P. (1881), “On the Authenticity of the Olympian Register,” The Journal of Hellenic Studies 2: 164–78. Mitchell, Robert W. (1996), “The Psychology of Human Deception,” Social Research 63: 819–61. Möller, Astrid (2004), “Greek Chronographic Traditions about the First Olympic Games,” in Ralph Mark Rosen (ed.), Time and Temporality in the Ancient World, Philadelphia: 169–84. Moretti, Luigi (1957), Olympionikai, i vincitori negli antichi agoni olimpici, Roma. Siewert, Peter (1992), “Zum Ursprung der Olympischen Spiele,” Nikephoros 5: 7–8. Scanlon, Thomas F. (2002), Eros and Greek Athletics, Oxford. —— (2005), “The Dispersion of Pederasty and the Athletic Revolution in Sixth-Century BC Greece,” in Beert C. Verstraete and Vernon Provencal (eds.), Same-Sex Desire and Love in Greco-Roman Antiquity and in the Classical Tradition of the West, New York: 63–85. Sexton, D.J. (1986), “The Theory and Psychology of Military Deception,” in R.W. Mitchell and N.S. Thompson (eds.), Deception: Perspectives on Human and Nonhuman Deceit, Albany: 349–56. Sinn, Ulrich (1991), “Olympia: Die Stellung der Wettkämpfe im Kult des Zeus Olympios,” Nikephoros 4: 31–54. Valvanis, Panos (2006), “Thoughts on the Historical Origins of the Olympic Games and the Cult of Pelops in Olympia,” Nikephoros 19: 137–52. Wacker, Christian (1998), “The Record of the Olympic Victory List,” Nikephoros 11: 39–50. Wȩcowski, Marek (2012), “Hippias of Elis (6),” in Ian Worthington (ed.), Brill’s New Jacoby, Brill Online, accessed 4 December, 2012, . Weniger, Ludwig (1905), “Das Hochfest des Zeus in Olympia,” Klio 5: 184–218. Wilamowitz-Moellendorff, Ulrich von (1922), “Pisa und die Olympische Sieger,” in Pindaros, Berlin: 481–90. Ziehen, Ludwig (1915), “Bericht über die Griechischen Sakralaltertümer,” Jahresberichte über die Fortschritte der Klassischen Altertumswissenschaft: 66–71.

Reading the Fraudulent Text: Thessalus of Tralles and the Book of Nechepso Karen Ní Mheallaigh Pseudo-documentary fictions are narratives which claim, disingenuously, to be based on ancient documents.1 Usually these texts pretend a greater antiquity than they actually possess; in some ancient cases (for example, the famous “Trojan War diaries” of Dictys and Dares) they pretend an antiquity greater even than the Homeric poems, the oldest works of the Greek literary tradition. But pseudo-documentary fictions always hint at their own fictionality as well, generating, in addition to the basic first level of “semantic” reading (reading to find out “what happens”), a second level of “semiotic” reading (reading to find out how “what happens” is narrated) which hints at the naiveté of the first-level reader and so ironizes the very reader which the text itself constructs.2 Pseudo-documentary fictions therefore tend to embroil the reader in a paradoxical dialectics of reading: to enjoy them fully, the reader must enter into the belief that the text is authentic while knowing that these fictions of authenticity are also, at the same time, the ultimate fakes.3 For the reader who accepts the invitation to read the text at more than one level the narrative becomes a fabula de se, a story about the reader him or herself. These teasing fantasies of ancient texts dismantle and explore the mental processes which enable the reader imaginatively to enter the fiction of authenticity in the first place. As a result, the reader is tugged by conflicting waves of belief and scepticism, leading ultimately towards a savouring of the intertwining double-helix between the two that constitutes the most sophisticated and paradoxical pleasure of fiction. At a fundamental level, pseudo-documentary fictions are about books and book-culture; they are playful narratives about the author and reader (especially the antiquarian scholar or academic) and the relationship of (dis)trust between the two.

1  On pseudo-documentarism in ancient literature, see Hansen 2003 and Ní Mheallaigh 2008; on literary forgery in the ancient world, see Speyer 1971 and Martínez 2011. 2 On such ironic interplay of these levels of reading, see Eco 2006. 3 Eco 1990 examines the fake as a test-site for authenticity.

180

karen ní mheallaigh

The story of the putative ancient text’s discovery is usually narrated in the prologue, often by a voice which presents itself as that of the text’s editor or translator. By staging fictional encounters with pseudo-texts in this way, these fictions absorb the reader into their fictional worlds: the reader of the text is assimilated to the reader in the text, and both readers’ experiences seem to become entwined. Often too, by a strange inversion (or twist of fate?), the fiction of such texts can appear to preempt their real history, when these fictions of long-lost texts turn out to be partially or obliquely extant texts, which blurs the boundaries between fact and fiction still further. This is what makes them so compelling, for pseudo­documentary fictions are not just stories about the ancient reader’s encounter with the text, but also about the modern reader’s encounter with the Classical past through its fragmentary remains. When these works dramatize the excitement and the anxieties that are concomitant with the fictional “r­ediscovery” of ancient lost texts, they are also dramatizing the thrill – and the interpretive dangers – which are a part of the classicist’s encounter with the texts of the ancient Greek and Roman past.4 Their narratives about ancient pseudo-authentic texts speak also of our own anxieties about how to interpret the fictions of authenticity and authorship from antiquity. The best known pseudo-documentary fictions of the imperial period, such as the Dictys and Dares-texts, or Antonius Diogenes’ novel The Incredible Things beyond Thule, belong to a rich hinterland of scholarly treatises and wisdom-texts of the post-classical period that were prefaced by fictional narratives of origin, stories about the text itself. The primary purpose of these fictions was to guarantee the treatises’ authenticity, enhance their authority, and, by generating an aura of mystery and adventure about the text, to pique the curiosity of potential readers. Typically it is claimed that the text is derived from an origin of extreme antiquity, usually non-Greek, which the author “discovered” during his extensive travels. Dionysius Skytobrachion’s pseudo-history Phrygia was reputedly based on the “Phrygian composition” written by the pre-Homeric Trojan poet Thymoetes;5 the magical wisdom contained in Bolus of Mendes’ treatise Compounds and in the first book of the Kyranides was reputedly discovered in Phoenicia, Egypt, and Syria. In each case the arcane nature of the putative source is emphasized in order to enhance the esotericism

4 For a fascinating account of this encounter, see Obbink 2011, esp. 24–5. 5 Diodorus Siculus 3.67.1 and 4–5 (= FGrHist 32 F8).



reading the fraudulent text

181

of the knowledge gleaned from it. The first book of the Kyranides is supposedly derived from an obscure Syriac text deciphered with the help of a translator, for example:6 Bolus of Mendes claimed to have derived his work from texts by Democritus which Democritus, in turn, claimed to have copied from the ancient texts of Apollobex of Coptos and Dardanus of Phoenicia.7 These complex fictions of origin imbue their treatises with the perfume of antiquity and convert the encounter with the text into an adventure which takes the reader, imaginatively, on remote journeys, into “direct” contact with ancient books and archives, with the lost world of the past and the company of gods and holy men. This article will focus on one such text, a medical treatise whose prologue takes the reader on a thrilling journey to Alexandria and deep into the Egyptian desert to a magical encounter with the god of medicine himself. It is a narrative designed to confirm the treatise’s authenticity – which contains, paradoxically, a story of the author’s encounter with a fraud. Pseudo-Thessalus of Tralles’ On the Virtues of Plants On the Virtues of Plants is a collection of herbal remedies, complete with instructions about the appropriate times for harvesting the ingredients. The treatise is ascribed to Thessalus, a wonder-healer of the first century ad who, in an elaborate peritextual frame (prologue and epilogue), presents the text as his bootlegged dictation of his interview with the god Asklepios. The textual tradition of On the Virtues of Plants is complex, but it is possible to reconstruct an original version of the Greek “Thessalus”-text which was framed by a prologue where “Thessalus” set forth his credentials as a scholar and explained the provenance of the work and an epilogue where he thanked the god Asklepios for his revelations and received instructions to keep the precious text a secret.8

6 Kyranides prol.; text in Kaimakis 1976. 7 Pliny NH 30, 9. 8 The text is in Friedrich 1968 (see esp. 13–36 on the textual transmission). There is a helpful translation of the prologue, with some commentary, in Ogden 2009, 52–4; however, Ogden’s translation is based primarily on the BH manuscripts of the Greek Hermes-text, which differs in some details from the Thessalus-text which I refer to here. For discussion of the prologue’s narrative format, see Winkler 1985, 258–260 (with further bibliography) and Hansen 2003, 310–311.

182

karen ní mheallaigh

The preface takes the form of a letter written in the person of the historical Thessalus of Tralles to the Emperor Claudius or Nero.9 In this letter Thessalus narrates his meteoric ascent to fame and fortune as the foremost wonder-healer in Rome, beginning with his success as a Wunderkind of “grammatical science” in Asia and followed by his equally triumphant conquest of the field of medicine in Alexandria. Once his studies were complete, before returning home to Rome, he went searching through libraries for texts which he could use in his future career, and discovered a book by the pharaoh Nechepso, which contained twenty-four marvellous treatments.10 Excited by the wonders promised by the book, Thessalus wrote to his parents confidently announcing the authenticity of his discovery and his imminent return home. Disaster struck, however, when all experiments to test the treatments’ efficacy failed. Thessalus, repenting of his greed-driven gullibility and feeling Nechepso’s deception “harsher than death,” was forced to leave Alexandria in disgrace. In colourful tones of melodrama, he relates how he was too ashamed to return to Rome and wandered around Egypt as a tragic figure, “driven by a gadfly of the soul,” in a desperate quest to locate some genuine arcane medical knowledge with which to recuperate his ruined reputation and make good his premature boasting about Nechepso’s book. Stretching his hands to heaven, he begged the gods to commune with him so that he could return to Alexandria and thence to Rome with his pride restored. Failing that, he says, he had decided to “bid adieu to life.” Fortunately, however, in Diospolis (Thebes), the most ancient city of Egypt, Thessalus encountered learned high-priests who appear to possess the magic he desires, although their wary refusal to share their knowledge with him caused Thessalus to suspect they may be impostors: his bitter experience with the fraudulent Nechepso-text taught him a thing or two about the promise of Egyptian wisdom. However, one old priest was willing to help and agreed to arrange a lecanomantic encounter with a divinity of his choosing. The would-be physician selected Asklepios, the god of medicine and healing, and after three days of purification, 9 The Greek text T actually reads: “Harpokration to Caesar Augustus, greetings” Ἁρποκρατίων Καίσαρι Αὐγούστῳ χαίρειν, but later in the preface Asklepios addresses Thessalus by name (prooem. 25), thus confirming Thessalus as the letter-writer. The name “Harpokration” may have been imported from the prefatorial letter to the first books of the Kyranides, where “Harpokration” plays a similar role to Thessalus as the discoverer of a wisdom-text. 10 On the importance of novelty (novel strategies, cures, techniques etc.) for establishing one’s reputation as a doctor, see Pliny NH 29.11.



reading the fraudulent text

183

Thessalus entered a room in a house which had been prepared for the magical epiphany. Without the old priest’s knowledge, he smuggled in papyrus and ink and requested to speak to Asklepios in private.11 Reluctantly, the priest agreed. When Asklepios appeared, he assured Thessalus that he will win great success and even be worshipped as a god, and bid him to ask what he will. Thessalus first asked why he failed to make Nechepso’s treatments work. King Nechepso, the god replied, was a man of sound mind and virtue, but he did not have the good fortune to commune with the god in order to acquire the information that would soon be imparted to Thessalus; consequently, although he understood the affinities of plants and stones, he did not realise the importance of harvesting plants at the proper time and place, which is why the treatments were ineffectual. The god then began his learned exposition which Thessalus recorded in two books. After the divine discourse, in the epilogue,12 he records the god’s instructions to keep the text a closely-guarded secret because the information it contains is so powerful that, if it became common knowledge, it would eradicate the need for any other form of medicine and consequently all other branches of the discipline would decline into obsolescence. Thessalus’ book, then, is the medical book to end all medical books. The primary purpose of this narrative frame is to guarantee the authenticity of the treatise and therefore enhance the authority of the medical knowledge it contains, which is presented as a dictation direct from the highest medical authority, the healer-god himself. The entire Fundbericht is also obviously contrived to swell Thessalus’ own prestige, which matches what we know about the historical Thessalus, who was famous for aggressive self-promotion of this nature: his statue on the Appian Way, for example, sported the inscription Iatronikēs, “Doctor-Conqueror.”13 Here his success is sanctioned by the god who also predicts his future divine status, which in turn enhances the value of the treatise as well. But the presence of the problematic book of Nechepso in the treatise’s fabula 11 As Speyer (1971, 148) notes, the manner in which Thessalus equips himself with writing-materials in advance of the god’s revelation demonstrates his deceptive intent. 12  As explained above, the Greek Thessalus-text (T) breaks off in Book 1, and so does not include the epilogue; I therefore follow the version in the Latin manuscript P, which is an extension of T. In the Latin P and the Greek manuscripts of the Hermes-text the priest is absent during the divine encounter; it is clear from the introductory frame preserved in T that the priest was absent in the Greek Thessalus-text, too. In the other Latin manuscripts of the Thessalus-text (M and V), the priest appears to be present. 13 Pliny NH 29.5.9.

184

karen ní mheallaigh

de se also provokes questions, bringing the question of how to interpret these very claims to authenticity to the front of the reader’s mind. The Nechepso Book: Reading the Fraudulent Text In the midst of this heady self-advertisement, there is the account of Thessalus’ excruciating professional embarrassment when he declared his faith in a book of medicine, the Book of Nechepso, which turned out to be a fraud. It was an audacious move to include a confession of professional failure in a narrative whose purpose is precisely the opposite – to advertise and confirm the author’s credentials. One effect of such an ostentatiously “honest” disclosure (which is closely akin to Lucian’s disingenuous admissions of his “shortcomings” as a historian in True Stories) is to provide further evidence of Thessalus’ trustworthiness, as someone who has been in the reader’s position and knows, from painful personal experience, what it is like to be duped by a fraudulent text. For the compliant reader who is eager to believe the promise of Thessalus’ book, the book of Nechepso is the anti-text of Thessalus’ treatise: Nechepso’s apparent charlatanry and ignorance contrast sharply with the integrity and expertise of Thessalus, the reliable author who has gone to such lengths to seek out authentic information which he then imparts to the reader, and who even shows that he is willing to die in the Egyptian desert rather than degrade his art. For the more sceptical reader who is inclined, rather, to dismiss On the Virtues of Plants as just another book of medical mumbo-jumbo like so many books of Nechepso (and we know there were many such readers in antiquity),14 Thessalus’ experience acts as a pressure-release: it gives expression, within the narrative, to the reader’s worst suspicions about the treatise in hand. In this case, the story of the Egyptian fraud (withinthe-potential-Egyptian-fraud) becomes a concession about the nature of texts like the one which the reader is reading right now – which disarms the reader’s fear of being “taken in” and deceived.

14 For ancient scepticism regarding the work of charlatan physicians, see Columella 7.5.17 (on the falsehood of Bolus of Mendes); see also Plutarch Quaest. Conv. 641b, with Winkler 1985, 264–5; Pliny NH 29.5.9. Lucian’s Philopseudes dramatizes the interplay between extreme scepticism regarding magico-medical cures (Tychiades) and enchantment with these (and other supernatural phenomena) among the intellectual “lovers of lies” gathered at Eucrates’ bedside; in the dialogue these conflicting responses to magic and medicine become a way to theorize fiction itself.



reading the fraudulent text

185

But this gives no real guarantee that Thessalus’ text is any more trustworthy than Nechepso’s, and a more wilfully subversive reader might conclude from Thessalus’ prologal letter that the author had enough raw ambition to fuel such an audacious, career-enhancing deceit. Certainly, by his own admission, Thessalus had both the medical knowledge and – more suspiciously – the philological expertise to do it. Possibly some readers suspected that the supposed “discovery” of Nechepso’s book in the library was a ruse, which was designed to deflect from Thessalus the suspicion of forging the text himself, especially in light of the fact that forgers frequently made claims about serendipitous “discoveries” of texts or antique sources of wisdom.15 In this pseudo-documentary fiction, the book of Nechepso hovers tantalizingly between honest reassurance and wily connivance as if on a knife-edge: the reader’s feelings about the putative pharaonic text will determine his or her attitude of belief towards Thessalus’ treatise. Conclusion: Thessalus’ Anti-Fraud Device? With its embedded pseudo-documentary ploy about the book of Nechepso, the Thessalus-text executes a preemptive strike against the reader’s suspicions about the authenticity of Thessalus’ own treatise by using the story of a fraud – paradoxically – to ward off suspicions of fraudulence. But at the same time, Nechepso’s book is a trompe l’oeil which risks drawing attention to the very deceit it is calculated to conceal, as this text-withinthe-text provokes the reader to think actively about the authenticity of On the Virtues of Plants itself. The book of Nechepso, therefore, functions both as the anti-text to Thessalus’ treatise and also, potentially, its mirrortext; it is either a decoy or a collusion, depending on how you (want to) read. And the reader’s own knowledge and frame of mind seems to be at the heart of the issue, for to complicate matters even further, the god ultimately reveals that the pharaoh’s book was not, after all, the malicious fraud which Thessalus, in his acute disillusionment, had held it to be: the inefficacy of its remedies was due, not to its author’s calculated dishonesty after all, but to his incomplete understanding. Fraudulence, the text suggests, is also a matter of interpretation, which may alter depending on how much the reader knows. 15 See Speyer 1971, 62. Speyer (1971, 68–70) cites further examples of the use of libraries and archives to authenticate forged texts.

186

karen ní mheallaigh

By inscribing a story about a serendipitously discovered text into the pseudo-documentary narrative about his own treatise, “Thessalus” dramatizes the excitement and dangers which confront his own reader as well. With its cunning manipulation of the conflicting responses of credence, doubt, and intrigue which the ploy of pseudo-documentarism triggers in the reader’s mind, this narrative about a text within-the-text is also a story about how we read all such narratives of authenticity and fraudulence – about how we are reading the text right now. References Eco, U. (2006), “Intertextual Irony and Levels of Reading,” in On Literature, Orlando: 212– 235. —— (1990), “Fakes and Forgeries,” in The Limits of Interpretation, Bloomington, IN: 174– 202. Friedrich, H.-V. (1968), De virtutibus herbarum von Thessalos von Tralles, Hain. Hansen, W. (2003), “Strategies of Authentication in Ancient Popular Literature,” in S. Panayotakis, M. Zimmerman and W. Keulen (eds.), The Ancient Novel and Beyond, Leiden and Boston: 301–13. Kaimakis, D. (1976), Die Kyraniden, Meisenheim am Glan. Martínez, J. (2011) (ed.), Fakes and Forgers of Classical Literature / Falsificaciones y falsarios de la literatura Clásica, Madrid. Ní Mheallaigh, K. (2008), “Pseudo-documentarism and the Limits of Ancient Fiction,” AJP 129.3, 403–31. Obbink, D. (2011), “Vanishing conjecture: lost books and their recovery from Aristotle to Eco,” in D. Obbink and R. Rutherford (eds.), Culture in pieces. Essays on ancient texts in honour of Peter Parsons, Oxford, 20–49. Ogden, D. (2009), Magic, Witchcraft, and Ghosts in the Greek and Roman World, second edition, Oxford. Speyer, W. (1971), Die literarische Fälschung im heidnischen und christlichen Altertum: ein Versuch ihrer Deutung, München. Winkler, J.J. (1985), Auctor & Actor: A Narratological Reading of Apuleius’s Golden Ass, Berkeley and London.

Hapax Legomena in the “Speeches of Apollodoros” and their Relation to the Corpus Demosthenicum Felipe G. Hernández Muñoz To identify the hapax in a classical author is a hard task, but very productive in that it can reveal the contribution of that author to the construction of the Greek lexicon and indicate the author’s preferred morphological and semantic procedures. It can also offer valuable data bearing on the authenticity of works attributed to the author. However, it is a task with a degree of insecurity since our knowledge of the Greek language grows continuously with the appearance of new texts and the available technical tools of the trade are far from complete.1 We have studied Demosthenes, an author fundamental to Greek literature in many respects, elsewhere; the task of researching his hapax legomena was developed in a series of phases. In separate works published in 1988 and 1993, the first phase, we broached the study of the Demosthenic “relative hapax,” i.e., not “absolutely” within all preserved Greek literature, but within the preserved works regardless of whether the term has been documented outside the author. This study of the “relative hapax” allowed us to reach interesting results concerning the authenticity and chronology of the speeches collected within the so-called “Corpus Demosthenicum.”2 Later, we began the hard task of locating the Demosthenic hapax in the full, “absolute,” sense, i.e., within all extant Greek literature. Some years ago we introduced a first instalment in the framework of the IXth Spanish Congress of Classical Studies,3 restricting attention to the hapax composed by the addition of a preposition, and focussing on the possible echoes to be found in the rival orator Aeschines. Finally, in the previous volume edited by Javier Martínez,4 we returned to the hapax to consider 1 Such as the Index Demosthenicus composed by S. Preuss, Leipzig 1892. 2 “Contribución lexicográfica al estudio de la autenticidad de los discursos políticos del Corpus Demosthenicum,” Myrtia 3 (1988), 61–91 and “La noción de Hápax relativo y su aplicación a los problemas de autenticidad en literatura clásica: un caso práctico,” Epos 9 (1993), 41–49. 3 “Hápax demosténicos en compuestos preposicionales,” Actas del IX Congreso Español de Estudios Clásicos. Lingüística griega (Madrid 1997), 123–126. 4 Fakes and Forgers of Classical Literature. Falsificaciones y falsarios de la Literatura Clásica, Madrid, 2011, 137–144. Cf. also our “La autenticidad de Prometeo Encadenado a la

188

felipe g. hernández muñoz

its function within general studies of stylometry and authenticity, discussing specific examples. Responding to his kind invitation, we now offer a new instalment of our work,5 inescapably provisional, including the list of all absolute hapax currently detected in the Corpus Demosthenicum, in order to offer a wider view concerning the Demosthenic hapax, a task which remains to be completed. Furthermore, we will attempt to identify some differences with regard to a group of speeches known as the “Speeches of Apollodoros,” which, though traditionally integrated into that “corpus” (speeches 46, 47, 49, 50, 52, 53, 59), appear not in Demosthenes but in the contemporary orator Apollodoros,6 the “eleventh orator” of the Attic canon.7 Accordingly, the number of each speech collected in the Corpus Demosthenicum will be placed inside parentheses ( ) if it is a speech of doubtful authenticity and inside square brackets [ ] if it is considered definitely inauthentic. The speeches will also be divided into four main groups: 1. Political speeches before the assembly; 2. Political-judicial speeches; 3. Private speeches; and 4. Epideictic speeches, prologues and letters.8 If the number of a speech is not mentioned, this is because no “absolute hapax” in it has yet been located. At the end of the list, the data for the “Speeches of Apollodoros” are separated out and tabulated. The following is a list of these hapax, as yet provisional and still awaiting final checking and searching; they follow an initial “P” if composed by the addition of a preposition, “NP” if not. After each hapax appear the following abbreviations: A: probable absolute hapax in all extant Greek authors. G: hapax later used by some ancient commentator (Pollux, Harpocration, etc.).

luz de las frecuencias lingüísticas,” in Lógos Hellenikós. Homenaje al Prof. Gaspar Morocho, Universidad de León 2003: 149–157. 5 In the framework of the official research project entitled “Greek manuscripts in Spain and Its European Context” (ref. FFI2011–25805). I thank Sonia de Juana Calvo and Victoria Hooper for their valuable assistance in preparing the English version of this work, and Juan Muñoz Florez, who has kindly read a preliminary version of this article and provided some helpful comments concerning the documentation of some hapax. 6 Cf. the introduction, with its references to the question, included in the edition of K. Kapparis, Apollodoros Against Neaira [D. 59], Berlin-New York 1999. We understand that Kapparis is now preparing an edition of the whole oeuvre of Apollodoros. 7 Cf. L. Pearson, “Apollodorus, the Eleventh Attic Orator,” in The Classical Tradition: Literary and Historical Studies in Honor of H. Caplan, Wallach, L. (ed.) Cornell Univ. 1966: 347–359. 8 For bibliography on all these speeches, see our “Demóstenes 1965–1997: Repertorio bibliográfico,” Tempus 21 (1999), 37–74, updated on the website http://www.liceus.com.



hapax legomena in the “speeches of apollodoros”

189

D: probable first documentation9 in Greek literature (before the Middle Ages). R: hapax repeated in several passages or Demosthenic speeches.

An asterisk, *, indicates that there is an epigraphical or papyrical record for the term. Hapax in the Corpus Demosthenicum 1. Political speeches before the Assembly Speech number:    8 P: προσπαρασκευάζω (D) NP:    9 P: ἐκπιπράσκω (A) NP: (10) P: συνεπανορθοῶ (D) NP:      14 P: προαπεχθάνομαι (A) NP: ἑξηκονταταλαντία (A), πεντεκαιδεκαναία (G) [17] P: ἐναποχρῶμαι (A*) NP: βδελυρεύομαι (D) 2. Political-judicial speeches

18 P: συσκευωροῦμαι (D), συγκατακτῶμαι (D) NP: αὐτοτραγικός (A), ἰαμβειοφάγος (G), νεβρίζω (G), νεήλατος (G), τριταγωνιστής (G), ὁποστοσοῦν (D), σεμνολόγος (D), σιτώνης (D*), τριηραρχικός (D), φιλιππισμός (D) 19 P: συμπαραπόλλυμι (D), συμπαιωνίζω (D), συνδιαλλάττω (D), συστεφανῶ (D), ὑπεργέλοιος (D) NP: πυροπωλῶ (A), ἀειλογία (GR), ξυληγῶ (G) 21 P: συνιππεύς (G), προαποστερῶ (D) NP: ἀρχιθεωρῶ (G*), ἀσελγαίνω (G), βεβουλευμένως (D), βοώνης (G*), τελωνία (D) 22 P: ἀποψηφιστέον (A), προσενεχυράζω (A), ἐπεξαμαρτητέον (D), καταχωνεύω (D) NP: ἀμφορίσκος (D), τριηροποιικός (D)

9 This can sometimes be uncertain because of problems of relative chronology between authors, as with several terms documented only in Demosthenes and Aristotle.

190

felipe g. hernández muñoz

23 P: συνεξαπατῶ (D) NP: ὁσαχοῦ (D), ἀνδροληψία (D), δεκατηλόγος (D), τετυφωμένως (D) 24  P: ἐξεγγύησις (G), προσκατάβλημα (G), καταχωνεύω (DR) NP: δημοτικῶς (D) (25) P: κατεγγύη (A), προευλαβοῦμαι (A), ἀντεγγράφω (D) NP: σκυλόδεψος (G*) 3. Private Speeches  27 P: προσεπίκειμαι (A), προσεπιτροπεύομαι (A), προσπεριποιῶ (A), συνεπίτροπος (A*), ὑπέρχρεως (D) NP: κλίνειος (A), μαχαιροποιεῖον (A)  28 P: NP: πεντεκαιδεκατάλαντος (A)  29 P: καταχηρεύω (A), συγκαταμαρτυρῶ (A), συνεπίτροπος (AR*), συγχορηγός (D) NP: ἀνεπιείκεια (G)  30 P: προπηλαστικῶς (A) NP: ἀμίσθωτος (D)     31 P: προσεγγυῶμαι (A) NP: [33] P: συνδιαιτητής (D) NP: (34) P: συνεπανορθῶ (DR) NP: ἑτερόπλοος (G) [35] P: NP: πεντηκοστεύομαι (D), εἰκόσορος (D)  36 P: ἀντιμοιρεί (A) NP: σκευώρημα (D), τραπεζιτεύω (D)  37 P: ἐπικατατέμνω (G), προσφευκτέον (D) NP:  39 P: ἀντίληξις (A), προσομολογία (A) NP: [40] P: NP: φυγοδικῶ (D) 41 P: NP: σκευώρημα (GR) [42] P: NP: ὀκτωκαιδεκάδραχμος (A), δωδεκάδραχμος (A*), ὑλαγωγῶ (G)



hapax legomena in the “speeches of apollodoros”

191

[43] P: ὑπεραναίσχυντος (A) NP: [44] P: ἐμπαιδοτροφῶ (A), ἐνισχυρίζομαι (D) NP: φρατορικός (D) 45 P: NP: τραπεζιτεύω (DR) [46] P: συμμιμνήσκομαι (A) NP: [47] P: ἀντιπροσκαλοῦμαι (D) NP: ἀκάκως (D) [48] P: ἀνθυπόμνυμαι (G) NP: οἰκίσκη (G) [49] P: διαπέρχομαι (A), ἐνεπισκήπτομαι (G*) NP: [50] P: ἐπιτριηραρχῶ (A), συνεπιπλῶ (A*), ἐντελόμισθος (D), ἐπιτριηράρχημα (G), διαπρεπέστατα (D), ἐπιθάνατος (D) NP: πεντηκονταρχῶ (A), τριηράρχημα (D) 51 P: NP: ἀνδροληψία (GR) [53] P: περίστοιχος (D) NP: τραπεζιτεύω (DR) 54 P: ἐπιπόλιος (A) NP: τυπτητέος (A), ὑβριστέος (D) [56] P: μετεντίθεμαι (A), προσεξαμαρτάνω (D) NP: παλιγκαπηλεύω (A*), ἑτερόπλοος (GR), ἀκαριαῖος (D), σιτηγία (D) 57 P: NP: ἀειλογία (GR), δημοτικῶς (DR), τιτθεύω (D), τρυγήτρια (D), φρατρίαρχος (D) [58] P: συνδεκαδίζω (A), ἀνθυπόμνυμαι (GR), ἀποχειροτονία (G) NP: [59] P: συσυκοφαντῶ (A), ἀποπεφασμένως (D), παραβοῶ (D) NP: τοκοφορῶ (G) 4. Epideictic speeches, Prologues and Letters (60) – [61] P: προσσυνίστημι (A), ὑπεράκρως (A), ἀποθρασύνομαι (D) NP: εὐδιαβόλως (A)

192

felipe g. hernández muñoz

Exordia: – Epistulae: – “Speeches of Apollodoros” [46] P: συμμιμνήσκομαι (A) NP: [47] P: ἀντιπροσκαλοῦμαι (D) NP: ἀκάκως (D) [49] P: διαπέρχομαι (A), ἐνεπισκήπτομαι (G*) NP: [50] P: ἐπιτριηραρχῶ (A), συνεπιπλῶ (A*), ἐντελόμισθος (D), ἐπιτριηράρχημα (G), διαπρεπέστατα (D), ἐπιθάνατος (D) NP: πεντηκονταρχῶ (A), τριηράρχημα (D) [52] – [53] P: περίστοιχος (D) NP: τραπεζιτεύω (DR) [59] P: συσυκοφαντῶ (A), ἀποπεφασμένως (D), παραβοῶ (D) NP: τοκοφορῶ (G) We can now establish some provisional statistical conclusions with regard to the “absolute hapax” (A) of the Corpus Demosthenicum: of the total of 43 hapax, the majority (31) are compounds formed by the addition of a preposition; 23 are verbs and 20 other forms (substantives, adjectives and adverbs); more appear in the group of private speeches (30) than in the rest (13). So far as the hapax may provide a criterion of authenticity, note that they are more frequent in the group considered wholly authentic (24) than in the doubtful-apocryphal speeches (19). Although in both groups prepositional hapax outnumber the non-prepositional, in the doubtfulapocryphal group the proportion is bigger: 15 with preposition vs. 4 nonprepositional. Consequently, within the entire Corpus Demosthenicum, the absolute hapax are concentrated in the private authentic speeches, sometimes as a preposition (especially συν- o προσ-) + verb, and sometimes as a double compound (2 prepositions + verb). Within the group of doubtful speeches or those that are clearly inauthentic, the so-called “speeches of Apollodoros” (speeches 46, 47, 49, 50, 52, 53, 59) are traditionally integrated into the Demosthenic corpus, as noted above. They present 6 “absolute hapax” (A), almost all of them compound verbs with preposition, with some tendency for the preposition to be συν-, and almost no examples of προσ-, even including the first lexi-



hapax legomena in the “speeches of apollodoros”

193

cal documentations (D) and terms quoted by commentators (G), which raises the total of the terms considered in these speeches to 19. The preference in the authentic speeches of Demosthenes for terms composed with preposition προσ-, normally of double compound type (2 prepositions + verb) – procedures that are almost non-existent in the so-called “speeches of Apollodoros” –, marks a new difference in the configuration of the hapax of this group of speeches. In this case, the difference is of a lexical character10 and points again towards non-Demosthenic authorship of these work that are traditionally included within the Corpus Demosthenicum.

10 Differences in other respects are also notable, for example in the proportion of the use of some particles, cf., our “μέν > δέ < δ’ : une loi d’authenticité démosthénique?,” Les Études Clasiques 75 (2007), 249–250.

Language and (in-)Authenticity: The Case of the (Ps.-)Lucianic Onos Heinz-Günther Nesselrath 1. Introduction Can language be an indicator of inauthenticity? Yes, it surely can. When the style of student papers suddenly seems to undergo a remarkable change in syntax and words, we reasonably suspect that the author has availed himself of illicit “help” by plagiarising what seem to be suitable passages (or even whole texts) found on the internet, and, more often than not, a simple Google search confirms our suspicions. Thus language may even be called a prime indicator for determining whether a text really is the genuine property of a writer or not. 2. A War of Several Centuries: The Question of the (Ps.-)Lucianic Onos The criterion of language may also help to solve a question vexing scholars for at least 350 years: In 1664, Tanaquil Faber (Tanneguy Le Fèvre / Lefebvre) seems to have been the first to express doubts1 about the authenticity of the story “Lucius, or The Ass” (Λούκιος ἢ Ὄνος; hereafter simply called Onos). This story was transmitted among the numerous writings of Lucian of Samosata and labelled there (in a subscription found in the best manuscript, Γ = Vat. Gr. 90) as Λουκιανοῦ ἐπιτομὴ τῶν Λουκίου Μεταμορφώσεων, i.e., as an abbreviated version (epitome) of a text either entitled “The Transformations (Metamorphoses) of Lucius” or “The Transformations, written

1 In a casual remark in his commented edition of Phaedrus (Phaedri Fabulae et P. Syri Mimi. Cum Notis et Emendationibus Tanaquilli Fabri hac secunda editione auctioribus, Salmurii 1664), 193: “Alcyon, qui hodie inter Luciani opera legitur, licet Luciani non magis sit quam fabulosa illa de Asino narratio (quod ego πᾶσιν, οἷσπερ βουλομένοις ἔσται, probare possum) . . .” A year later, we find a similar remark in the second volume of his collected Letters (Epistolae. Pars Altera, Salmurii 1665), 294: “Emendo Lucianum in Asino, si tamen Luciani est hic libellus; mihi certe alia omnino sententia est . . .). In both cases, unfortunately, Faber does not give any reasons why he thinks that the Onos could not have been written by Lucian.

196

heinz-günther nesselrath

by Lucius.” Along with this ascription (which, as the alternative possibilities of its translation show, is not altogether clear in itself), scholars have always been tantalized by the remarks the Byzantine patriarch Photius devoted to the relationship between Lucian and Λουκίου Μεταμορφώσεις, when he recorded his reading of the latter text (here given as Λουκίου Πατρέως Μεταμορφώσεων λόγοι διάφοροι) in his Bibliotheca (cod. 129). On the one hand, Photius detected a very close relationship in wording and syntax between the two texts, calling the author of these Μεταμορφώσεις an ἄλλος Λουκιανός and declaring, a few lines further on, that Lucian, in his view, seemed to have reproduced a shorter version of the original text αὐταῖς τε λέξεσι καὶ συντάξεσιν. On the other hand, the patriarch claimed to have discovered two diametrically opposed orientations of the two texts: while the author of the Μεταμορφώσεις appeared to him to be a credulous person and a true believer in the fantastic transformation stories he wrote about (including his own into an ass and back again), Lucian’s shorter text seemed to exhibit the same mockery of Hellenic superstitions that Photius could detect in other Lucianic writings. But how is that possible if both texts were so similar in their linguistic and stylistic form?2 It is impossible to enumerate the manifold (and often very contorted) attempts of explanations by which scholars (especially in the late nineteenth and the early twentieth centuries) tried to make sense of Photius’ remarks.3 Possibly the first prominent defender of the “Lucianicity” of the Onos was Christoph Martin Wieland, who at the end of his translation of the story4 added a short essay (“Ueber den wahren Verfasser des vorstehenden Mährchens,” pp. 296–304) in which he declared it impossible that Lucian should have filched or epitomized another author’s work and that therefore Lucian had to regarded as the creator of the original Onos story. A similar opinion was voiced twenty-nine years later by Paul-Louis Courier, who in 1818 published a bilingual (Greek-French) edition of the text: in Courier’s eyes, the Onos had to be the Lucianic original and the Μεταμορφώσεις mentioned by Photius a later expansion of this text. Yet, all nineteenth-century editors of Lucian’s works (Carl Jacobitz, Wilhelm Dindorf, Immanuel Bekker, Franz Volkmar Fritzsche)5 regarded 2 On this inconsistency, see already K. Bürger, De Lucio Patrensi sive de ratione inter Asinum q. f. Lucianeum Apuleique Metamorphoses intercedente, Diss. Berlin 1887, 8. 3 A very thorough and systematic survey of all the different opinions (up to present times) can be found in H. Mason “Greek and Latin Versions of the Ass-Story,” in Aufstieg und Niedergang der Römischen Welt ÍI 34.2 (1994), 1665–1707. 4 In Wieland’s Lucians Sämtliche Werke, Vierter Theil, Leipzig 1789. 5 In Julius Sommerbrodt’s edition the Onos was not included.



language and (in-)authenticity

197

the Onos as spurious. Their main reason for this may already have been that the language of this text seemed to show remarkable and disturbing aberrations from the style so typical of Lucian’s many other writings. In 1860, Carel Gabriel Cobet declared the Greek in the Onos to be too late for Lucian’s time,6 and soon others voiced similar concerns: in 1869, Erwin Rohde drew attention to “eine beträchtliche Anzahl nicht nur von Lukians sonstiger Schreibweise, sondern von jedem correcten Ausdruck abweichender Wörter und Constructionen.”7 In the late nineteenth and the early twentieth centuries, two studies of Lucian’s language confirmed this impression.8 In the latter part of the century, the most eminent critics who based their verdict of non-authenticity on primarily linguistic foundations were Helmut van Thiel9 and Jennifer Hall.10 Nevertheless, Lucianic authorship of the Onos has still found a number of dedicated defenders.11 Confronted with the linguistic difficulties, they resort to three strategies: either (1), they claim that Lucian, when he sat down to write the Onos, simply had to conform to the stylistic rules of this genre of tale12 and thus deviate from his “normal” practice of writing; or (2), they go on the offensive and show that the Onos exhibits so many really “Lucianic” features of style that the aberrations are negligible.13 6 C.G. Cobet, Variae Lectiones, Leiden 1860, 260: “Quicumque scripsit Lucium sive Asinum aliquanto serius quam Lucianus vixit et Graecitate utitur aliquanto deteriore, multa negligenter et plebeiis erroribus scriptitans” (he then gives a few examples). 7 E. Rohde, Ueber Lucians Schrift ΛΟΥΚΙΟΣ Η ΟΝΟΣ und ihr Verhaeltniss zu Lucius von Patrae und den Metamorphosen des Apuleius, Leipzig 1869, 40. Still, as the title of this book shows, Rohde nevertheless regarded the Onos as Lucianic, but later he changed his view (see RhM 40, 1889, 91). 8 S. Chabert, L’atticisme de Lucien, Paris 1897; R.J. Deferrari, Lucian’s Atticism: The Morphology of the Verb, Princeton 1916. 9 H. van Thiel, Der Eselsroman, vol. 1: Untersuchungen, München 1971. 10 J. Hall, Lucian’s Satire, New York 1981 (still one of the best books ever written on the subject). 11 Just last year, a new commented edition has appeared bearing already in its title the claim that the piece was written by Lucian: Evan Hayes, Stephen Nimis (ed.) Lucian’s The Ass: An Intermediate Greek Reader. Greek Text with Running Vocabulary and Commentary. Oxford, OH 2012 (rev.: J.C. Relihan, BMCR 2012.09.55). 12 One of the greatest promoters of this thesis was Wilhelm Schmid, the author of the massive study Atticismus (5 volumes 1887–1897) and later the main author of the important Geschichte der griechischen Literatur. Schmid even created a special name for this genre, calling it “mimische Erzählung” (Neue Jahrbücher für das klassische Altertum 13, 1904, 485; see also already Philologus 50, 1891, 315–6 and still Berliner Philologische Wochenschrift 39, 1919, 168). 13 This approach was chosen by C.F.E. Knaut, De Luciano libelli qui inscribitur Lucius sive Asinus auctore, Diss. Leipzig 1868 and (more comprehensively and thoroughly) by V. Neukamm, De Luciano asini auctore, Diss. Tübingen, Leipzig 1914. Neukamm was a pupil of Wilhelm Schmid and adopted his explanation that the “un-Lucianic” stylistic features

198

heinz-günther nesselrath

(3) If they feel that these aberrations are not so negligible after all, they have tried to blame the epitomator.14 All three approaches, however, have flaws. As for (1), it proceeds from the assumption that Lucian must have written the Onos and thus runs the risks of being circular: if he had not written the Onos, he would not have had to conform with the laws of “mimische Erzählung.” And was there really such a thing with strictly defined laws? In the course of the twentieth century, papyrus finds have greatly enriched our knowledge of popular Greek narrative texts; we now know that they could be written in a whole range of linguistic levels, from Lollianus’ quite vulgar Phoenicica to the almost high-brow Aethiopica of Heliodorus. Thus Lucian might not have had to conform to any firm stylistic laws to write something like the Onos.15 As for (2), it has very rightly been pointed out16 that the overwhelming majority of stylistic features that the Onos shares with “genuine” Lucianic writings can also be found in other authors and texts and thus cannot really help us establish firm links between the Onos and Lucian. Even a refinement of this theory – namely that in the Onos Lucian deliberately mixed “lower” stylistic elements into his style “to produce the correct ‘low’ tone and mock-heroic ethos required by the subject”17 – has been thoroughly refuted by pointing out that the inconsistencies thus produced18 are still too few and too widely spaced to be noted by listeners or readers. As for (3), this presumes that there would have to be quite remarkable stylistic differences between original and epitome – but where could they come from, if this epitome is really only an epitome, i.e., an abridged version of a fuller text? The scholars who postulate such differences seem to fall into the same sort of inconsistency as Photius before them. Others

in the Onos (of which Neukamm found quite a few) were to be explained by the laws of “mimische Erzählung.” Knaut’s and Neukamm’s findings were readily accepted by Ben Edwin Perry in his dissertation of 1920 (The Metamorphoses Ascribed to Lucius of Patrae) and even by Matthew Macleod when he edited the Onos in the last (eighth) volume of the Loeb Lucian (of 1967) stating (on p. 49) that the two German scholars had “amassed a formidable list of peculiarly Lucianic usages on the Ass.” 14 Thus Perry (above n. 13), pp. 12, 73. 15 We can also see from his Dialogues of the Courtesans that Lucian was fully able to treat sexual themes like those of the Onos without having to sacrifice his style. 16 E.g., by Jennifer Hall (see above n. 10), pp. 361–2. 17 Thus G. Anderson, Studies in Lucian’s Comic Fiction, Leiden 1976, who argues on pp. 34–67 that Lucian wrote both the lost original Ass Story and the abridged Onos. 18 See Hall (above n. 10), p. 359.



language and (in-)authenticity

199

have shown that the Onos at times seems to be “rather hasty and badly done”;19 we should therefore expect little linguistic variation from it. Even after Hall’s lucid remarks on the Onos, some scholars have still clung to regarding the work as genuinely Lucianic. A case in point is Niklas Holzberg’s study of 1984,20 in which the author declares himself satisfied with Anderson’s explanations of the linguistic discrepancies in the Onos and then seeks to establish as the main argument for authenticity the fundamental identity of what both the Onos and Lucian’s prominent other writings want to demonstrate to their audience. Their central theme is (according to Holzberg) the exposure of the difference between appearance and reality in the various spheres of human society (“Aufdeckung des Unterschieds zwischen Schein und Sein in den verschiedenen Bereichen der menschlichen Gesellschaft,” p. 172). This theme, however, is of such a general nature that by using it one could surely propose other candidates for the authorship of the Onos at least as promising as Lucian (e.g., Epictetus, who would also have the advantage that his style – as recorded by Arrian, of course – in general is much less indebted to Atticism than that of Lucian). The preceding remarks (although not exhaustive in covering the already long debate about the Onos) must suffice to show that the most reliable criterion for coming to grips with the question of whether the Onos is Lucianic or not always was and still is the question of language. In the remaining parts of this paper I will give the results of my own investigation into the language of the Onos, hoping to show that it clearly points to non-Lucianic authorship. 3. The Linguistic Case for Inauthenticity First of all, here is a list (compiled independently from previous ones, such as Neukamm’s) of more than eighty words or expressions (combinations of words) in the Onos not found in other, “authentic” Lucianic writings: πατρικόν τι συμβόλαιον (ch. 1) ἁλῶν κοινωνεῖν (ch. 1) ἀργαλέα ὁδὸς (ch. 1) οἰκίδιον (ch. 1,2)21 19  Hall (above n. 10), p. 359. 20 N. Holzberg, “Apuleius und der Verfasser des griechischen Eselsromans,” Würzburger Jahrbücher für das Altertum N.F. 10, 1984, 161–177. 21  The phrase οἰκίδιον ἀνεκτόν is found nowhere else in Greek literature.

200

heinz-günther nesselrath [5] κλινίδιον (ch. 2)22 ἐξοχώτατος (ch. 2)23 οἰκίδιον . . . εὔγνωμον (ch. 2)24 ἀνεξικάκως οἰκήσας (ch. 2)25 τὸν ἑταῖρον κοιτῶνα (ch. 2)26 [10] οἰκημάτιον (ch. 3) σκιμπόδιον (ch. 3) κριθίδια (ch. 3) συνανακλίνεσθαι (ch. 3) τὴν ἑσπέραν . . . πότῳ δόντες (ch. 3) [15] πλησιαίτερον (ch. 4) ἀπωτέρω (ch. 4) ἐπιβάλλειν τὸν ὀφθαλμόν τινι (ch. 4)27 τῆς . . . γυναικὸς . . . πόρρω ἵστασο (ch. 5)28 κἀπὶ ταύτης κυλιόμενος (“rolling on this woman,” ch. 5)29 [20] μακάριος ὅστις ἐνταῦθα ἐνεβάψατο (ch. 6) κοράσιον (ch. 6, also twice in ch. 36)30 τὴν γλυκεῖαν ὀδύνην (ch. 6)31 ἀνθρωπομάγειρον (ch. 6)32 ταῖς πικραῖς καὶ ἡδείαις θεραπείαις (ch. 6)33 [25] τοῦ ὕπνου καταψευσάμενος (ch. 7)34

22 The phrase κλινίδιον στενόν is found nowhere else in Greek literature. 23 Lucian uses the word ἔξοχος (or its adverb ἔξοχα) only in poetical quotes (Gall. 7, Char. 8, Pisc. 41) and its superlative ἐξοχώτατος nowhere else. 24 The phrase οἰκίδιον . . . εὔγνωμον seems to be singular in all Greek literature. 25 Lucian uses the adjective ἀνεξίκακος (not the adverb) only in Iud. Voc. 9; the combination of the adverb with οἰκεῖν is found nowhere else in Greek literature. 26 The word κοιτών is used nowhere in Lucian’s authentic writings, and the combination with ἑταῖρος (if genuine: Harmon reads τὸν ἕτερον κοιτῶνα, a variant of the recentiores) seems to be unique in all Greek literature. 27 The expression is found in Alexis (fr. 240 K.-A.), in the Greek Novel (Ach. Tat. 4.2.1, Heliod. 1.14.4), in Libanius (decl. 32.27), Aristaenetus (1.12), even in Christian authors (Basil. De virginitate PG 30 p. 697.34–8) – but not in Lucian. 28 I have found no other instance of this expression (with the sense of “keep away from”) either in Lucian or in other Greek literature. 29 Somewhat similar, but far from identical is Sext. Adv. Math. 1.291 (συγκυλίνδεται τῇ γυναικί). 30 This word, branded as un-Attic by Aelius Dionysius and Phrynichus, is frequent in the Septuagint, the New Testament, and is also found in Epictetus, the Vita Aesopi, the Passio Perpetuae et Felicitatis and in Christian writers. 31 According to the electronic TLG, this oxymoron is only found again in Greg. Nyss. In Cant. cantic. p. 380,5 Langerbeck and in Anth. Pal. 12.99.6. 32 A hapax in Greek literature. 33 For ἡδεῖα θεραπεία, cf. Arist. EN 1157a10–11 (τῷ μὲν γὰρ οὐκ ἔστιν ἡδεῖα ἡ ὄψις, τῷ δ’ οὐ γίνεται ἡ θεραπεία); for πικρὰ θεραπεία, see Eus. PE 13.3.40 (τὰς ἀλγεινὰς καὶ πικρὰς θεραπείας); but I have found no other instance in which all three words are combined. 34 There seems to be no other instance of καταψεύδομαι with this kind of object (LSJ s.v.: “make a pretence of ”).



language and (in-)authenticity

201

πρώσας (ch. 9)35 εὐηκόως (ch. 10) γονάτιον (ch. 10) ἐπίπρωσον (ch. 10)36 [30] παρεμβολή (ch. 10) ἐπὶ νοῦν μοι ἦλθε (ch. 11) ὀπή (ch. 12, 16, 57, 52) καταλαλέω (ch. 12)37 δίχα (ch. 15)38 [35] δακὼν τὸ χεῖλος (ch. 15)39 ὕπνος ὁ γλυκύς (ch. 16)40 ἀτρίπτῳ ὁδῷ (ch. 16)41 περιβοσκόμενοι (ch. 17)42 ῥοδοδάφνη (ch. 17)43 [40] ἀχθοφορία (ch. 19) εἰς ἐμαυτὸν ἀνασωθήσεσθαι (ch. 20)44 συνάριστος (ch. 21)45 καθαρισμός (ch. 22)46 ἀναπεσόντες (ch. 23)47 35 The “real” Lucian uses προωθέω without contraction (VH 2.42, Merc. Cond. 10, Pro lapsu 16). 36 Apparently an absolute hapax in Greek literature. 37 The word seems to have a rather unusual meaning here (LSJ s.v.: “talk before”). 38 Apart from Onos, Amores, and Ocypus, δίχα is found only once in the Podagra (v. 64), the authenticity of which is not without doubt (and it is in verse), so unlike the Onos author the “authentic” Lucian uses it nowhere in his prose. 39 The “authentic” Lucian has a similar but not identical expression in Calumn. 24 (ἐνδακόντα τὸ χεῖλος, cf. Alciphr. ep. 3.22.3: ἐνδακὼν τὸ χεῖλος). The Onos author’s phrase is probably derived from poetic models (cf. Tyrt. fr. 10,2 and 11,2 W.: χεῖλος ὀδοῦσι δακών; Eubulus fr. 52.6 K.-A.: δάκνων τὰ χείλη). 40 The expression “sweet sleep” is frequently found in the Iliad (e.g., 1.610: γλυκὺς ὕπνος) and the Odyssey (e.g., 15.6: ὕπνος . . . γλυκύς) as well as in other poets, but Lucian only once comes close by quoting Il. 10.3–4 (ἀλλ’ οὐκ Ἀτρεΐδην Ἀγαμέμνονα . . . / ὕπνος ἔχε γλυκερὸς πολλὰ φρεσὶν ὁρμαίνοντα) in Gall. 25. 41 Neither the word ἄτριπτος itself nor the expression ἀτρίπτῳ ὁδῷ (for which cf. e.g., Callim. fr. 1.28 Pf. κελεύθους / ἀτρίπτο]υ̣ς, Ael. VH 12.64 κρυπταῖς ὁδοῖς καὶ ἀτρίπτοις) is found in the “genuine” Lucian. 42 The earliest prose author to use περιβόσκω (apart from the Onos author) is Gregory of Nyssa. 43 In the “genuine” Lucian, this is found only as a denigratory nickname (Pseudol. 19). 44 The “genuine” Lucian has just two instances of this verb (Vit. auct. 22, Hermot. 28), none of them with εἰς / ἐς. I have also found only one passage, where ἀνασῴζω is used in a similar way (Plut. Bruta an. 985E (εἰ χάριτι σῇ λαβὼν τούτους αὖθις ἀνθρώπους [ἑταίρους] ἀνασώσαιμι). 45 The word seems to be a hapax in Greek literature. 46 The word is often found in the LXX and the NT, then in Christian authors, but nowhere in Lucian. 47 In the meaning here required (“recline at meals”), this word is not found in the “genuine” Lucian.

202

heinz-günther nesselrath [45] γυπῶν τέκνα (ch. 23)48 ἀωρίᾳ (ch. 24) ἀρχιτεκτόνημα (ch. 25) ἐγκατοικίσωμεν (ch. 25) ἀνέστενον (ch. 26) [50] εὐαγγέλιον . . . ἐμοῦ προογκησαμένου (ch. 26)49 συναιχμαλώτου (ch. 27)50 παιδάριον ἀκάθαρτον (ch. 29) κακὸν . . . τυρεύων (ch. 31)51 πῦρ ἄπλετον (ch. 31)52 [55] ἐσκόρπισεν (ch. 32)53 ἀποσιτῆσαι (ch. 33) ἀκέραιος (ch. 33) νεόνυμφον (ch. 34) δινουμένη (ch. 35) [60] θεοφόρητος (ch. 37, 38) ὅμιλος [= μέλος] ἔνθεον (ch. 37) μαλακοῦ αἵματος (ch. 37)54 κακὸν ἐξαίσιον (ch. 39)55 ἀγέρωχος (ch. 40, 45) [65] στρόμβος (ch. 42) ἄσβεστον ἐγέλων (ch. 45)56 Ἐξ ὄνου παρακύψεως (ch. 45)57

48 A really unique expression, humorously modelled after instances like Plut. Animi an corporis aff. p. 501A παῖδες ἰατρῶν, Philo Plant. 173 παῖδας ἰατρῶν καὶ φιλοσόφων, Philo Somn. 1.51 ἰατρῶν παῖδες, Philo De Ios. 160, Luc. Hist. Conscr. 7, Luc. Dips. 5; Plat. Leg. 769b οἱ ζωγράφων παῖδες; Luc. Imag. 9 πλαστῶν καὶ γραφέων καὶ ποιητῶν παῖδες; Luc. Tox. 44 βασιλεῖς καὶ βασιλέων παῖδες; Max. Dial. 33.8b ὦ ποιηταὶ καὶ ποιητῶν παῖδες, Synes. Calv. 11 φαλακρῶν παῖδες. 49 Neither εὐαγγέλιον nor προογκάομαι is found in genuine Lucianic texts. 50 συναιχμάλωτος is found only in the NT, once in Heliodorus and in later Christian writers. 51  τυρεύω, condemned by Pollux (Onom. 6.130; though already attested in Demosthenes) is never used by the “genuine” Lucian. The expression κακὸν τυρεύω is elsewhere only found in Adamantius’ Physiognomonica (1.17) and Synesius (Aeg. 1.15). 52 For this poetic-sounding expression there is only one parallel that comes somewhat close: Empedocl. fr. 17.27 (πῦρ καὶ ὕδωρ καὶ γαῖα καὶ ἠέρος ἄπλετον ὕψος). 53 σκορπίζω, noted by Phryn. Ecl. 189 as un-Attic, is nowhere found in the “genuine” Lucian, nor in classical Attic authors, but in NT and LXX. 54 This word-combination is attested in Galen (De elem. ex Hippocr. I p. 495.1 Kühn: τὸ μὲν αἷμα τῆς σαρκὸς μαλακώτερον, De temp. III 1 p. 564.10 Kühn, De usu part. IV p. 243.7 Kühn), but apparently not found in non-medical writers (including Lucian). 55 ἐξαίσιος is nowhere found in the “genuine” Lucian, and the combination of κακόν and ἐξαίσιον seems to be otherwise attested not before Synesius (Epist. 121; Aeg. 2.3). 56 Curiously, there is no parallel for this expression (which is clearly based on the Homeric ἄσβεστος γέλως in Il. 1.599, Od. 8.326, 20.346) in the “genuine” Lucian, who nowhere uses ἄσβεστος. 57 A proverbial expression, apparently first attested in Menander (Hiereia, fr. 246 K.-A.), but not found in the “genuine” Lucian.



language and (in-)authenticity

203

σύνδουλος (ch. 46)58 σύσκηνος (ch. 46) [70] ἅρπαξ (ch. 46) τὸ . . . συμπόσιον ἐκλονεῖτο τῷ γέλωτι (ch. 47)59 συναριστῶν (ch. 50) γυνὴ . . . τὴν ὄψιν ἱκανή (ch. 50)60 μισθὸν . . . ἁδρὸν (ch. 50)61 [75] χαμεύνιον (ch. 51, twice) ἀκόρεστος (ch. 51) συνευνάζομαι (ch. 52) τὰς φιλοτιμίας ἦγεν ὁ ἐμὸς δεσπότης (ch. 53)62 ἀποσκευή (ch. 55) [80] μεταμορφόω (ch. 56) πρωκτός (ch. 56)63

Second, it is to be noted that the Onos author employs a number of favourite words or expressions that either do not appear at all in the “authentic” Lucian or only very rarely: five times he uses ἀργαλέος,64 especially in combination with ὁδός (ch. 1, 34 and twice in ch. 42);65 δωμάτιον (again five times);66 ὀπή (four times); ἁδρός (four times as well: ch. 12, 32, 36, 50);67 ἀνασῴζω (six times; ch. 20, 23 [bis], 47, 56 [bis], and see above n. 44). ἀκάθαρτος in the abusive sense of “foul/beastly” is a favourite

58 The “genuine” Lucian only uses ὁμόδουλος ( JConf. 7, Tim. 23, Philops. 11) as prescribed by Atticist rule (Moeris: ὁμόδουλος Ἀττικοί, σύνδουλος Ἕλληνες) and which the Onos author also uses (ch. 42, 47). 59 κλονέω is nowhere found in the “genuine” Lucian, and κλονέομαι γέλωτι seems unique in Greek literature. 60 This phrase is found in Plutarch (cf. Mul. virt. 256E, Alex. fort. 339E), but not in Lucian. 61  This combination is found in Aelian (NA 11.33, VH 8.8, 11.63, fr. 71, 193, 194 Hercher) and in a few later writers (Zosimus, Theodoretus), but not in Lucian. 62 The special meaning of the plural of φιλοτιμία (LSJ s.v.: “occasions for munificence”) has no parallels in authentic Lucianic writings. 63 Despite the Onos author’s assurance (ἐκ κυνὸς πρωκτοῦ, τὸ δὴ τοῦ λόγου), this “saying” seems to be attested only in the Onos. 64 The word has a distinctly poetic pedigree (it is found thirty-four times in the Iliad, twenty-six times in the Odyssey, fifteen times in Hesiod – without the fragments –, seven times in Mimnermus, four times in Solon, nine times in Theognis, eight times in Apollonius Rhodius, ten times in Oppian’s Halieutica, ten times in the Cynegetica). The only prose author (apart from the author of the Onos) who uses it frequently is Philo of Alexandria. The authentic Lucian has ἀργαλέος only once (Pseudol. 19: φεύγεις ἑκὼν πατρίδα οὔτε χεῖμα κακὴν οὔτε θέρει ἀργαλέαν), and there as a reminiscence from Hesiod (Op. 640: Ἄσκρῃ, χεῖμα κακῇ, θέρει ἀργαλέῃ). 65 In ch. 31, we have δεσμῷ ἀργαλέῳ. 66 The word is found also three times in the Pseudo-Lucianic Amores (ch. 10, 34, 53), but only once in the (probably) authentic De Parasito (55). 67 The word appears only once in the “genuine” Lucian (VH 2.40: ἁδρότερον).

204

heinz-günther nesselrath

term of the Onos author (see ch. 29, 32, 33),68 while the genuine Lucian employs it only once (Lexiph. 19) denoting a woman “quae menstrua non habet” (thus LSJ s.v.). Third, there are distinct stylistic peculiarities of the Onos author that are not characteristic of the “authentic” Lucian: a penchant for diminutives69 (οἰκίδιον [ch. 1, 2], κλινίδιον [ch. 2], οἰκημάτιον [ch. 3], σκιμπόδιον [ch. 3], κριθίδια [ch. 3, 17, 46, 47], κοράσιον [ch. 6, 36], γονάτιον [ch. 10], δωμάτιον [ch. 12, 13, 16, 51, 56]; a penchant for expressions that are given a new and obscene sense (κυλίομαι ἐπί τινος [ch. 5], ἐμβάπτομαι [ch. 6], γλυκεῖα ὀδύνη [ch. 6], ταῖς πικραῖς καὶ ἡδείαις θεραπείαις [ch. 6], παρεμβολή [ch. 10]); and a penchant for using certain words or expressions rather differently from the way the “real” Lucian uses them: ἐργαστήριον (ch. 25) in the meaning of “band of robbers” is not found in the “genuine” Lucian; by ὀρθὴν (= ὀρθίαν) ὁδόν (ch. 29) the Onos author means a “dangerously steep / precipitous way,” while the “genuine” Lucian uses the expression only to signify a “right / straight way” (Rhet. Praec. 9; D. deor. 24.2);70 θεοφόρητος is used twice by the Onos author in a sense (“carrier of a god” = θεοφόρος) apparently found nowhere else in Greek literature; likewise, the use of ὅμιλος = μέλος (ch. 37) seems to have no parallels elsewhere.71 Fourth, in a number of cases, the “authentic” Lucian uses different words from the Onos author to express the same things: instead of κριθίδια, we find κριθαί (Tim. 14, Adv. Ind. 30, Lexiph. 5, Hermot. 61); instead of ἐκ τοῦ μύρου ἀλείφεσθαι, Lucian uses χρίεσθαι μύρῳ (Char. 22, Merc. cond. 28, Luct. 11, Pseudol. 31); instead of ἐπὶ νοῦν, we find εἰς νοῦν (Somn. 14); instead of δίχα in the sense of “without,” Lucian is perfectly happy with using ἄνευ (which is found nowhere in the Onos); instead of δακὼν τὸ χεῖλος, Lucian says ἐνδακόντα τὸ χεῖλος (see above n. 39); instead of σαρδάνιον γελῶντες (ch. 24), Lucian has τὸ σαρδάνιον ἐπιγελῶν (JTrag. 16); instead of παιδάριον, Lucian uses παῖς and παιδίον; instead of δινέω, Lucian has περιδινέω (VH 1.9, Dea Syr. 36, Lexiph. 2). Finally, there are not a few grammatical constructions that are not compatible with Lucian’s classicising Greek.72 (1) Confusions of ἐν and ἐς / εἰς (or “where” and “whither”) are rather frequent: ἄνδρα οἰκοῦντα ἐς τὰ Ὕπατα (ch. 1); ὅποι τῆς πόλεως οἰκεῖ (ch. 1, 2); πέμπει παρ’ ἐμοὶ τοὺς ἑταίρους 68 Cf. Dio or. 7.134 and the vocative ἀκάθαρτε in Bato fr. 5.2 K.-A. 69 Already noticed by van Thiel (above n. 9) p. 167 n. 10. 70 Cf. for this sense already Dem. or.18.15, 322, 25.36, Plat. ep. 7.330d. 71  The “authentic” Lucian, of course, uses μέλος. 72 On these, see, e.g., Hall (above n. 10), p. 359.



language and (in-)authenticity

205

(ch. 2); Ποῖ . . . κατάγῃ; (ch. 4); καταλύομεν εἴς τινα ἔπαυλιν (ch. 17); ἐμὲ κατακλεισθέντα ποι φυλάττεσθαι (ch. 40). (2) Uses of a preposition where “Attic” Greek would not need one: ἀλειψάμενος . . . ἐκ τοῦ μύρου (ch. 9, also ch. 56);73 γραῦν . . . ἐκ τῆς πέτρας κρεμαμένην ἐν καλῳδίῳ (ch. 24); τεθνηκὼς ἐν ταῖς πληγαῖς (ch. 44); καί με ἐκ τῆς φορβειᾶς λαβομένη . . . εἷλκεν (ch. 51). 4. Conclusion Taken together, the above described features show us an author who differs from Lucian’s stylistics in some remarkable ways. One might dismiss some (but surely not all) expressions within the above catalogue as items the non-appearance of which in Lucianic works seems rather natural. Items, however, in the categories two, three, and four (and probably also five) are harder to explain if we want to go on regarding the Onos as written by Lucian: why, for instance, would Lucian use ἀργαλέος, δωμάτιον, ὀπή, ἁδρός, and ἀνασῴζω so often only in the Onos? Why should it be only in the Onos that he uses ὀρθὴν ὁδόν, θεοφόρητος or ὅμιλος in such a peculiar and uncommon way? Why are there eight instances of words which appear as “doublets” of “Lucianic” expressions with identical meanings? The most straightforward explanation for these peculiarities is that the writer of the Onos is not a Lucian who seems mysteriously bent on expressing things differently from all of his other writings, but that he isn’t Lucian at all. As for the question of how the Onos came to be included among Lucian’s writings and unquestionably regarded (apparently up to Tanaquil Faber) as Lucianic, two plausible answers have already been given: “The similarity of the names Lucius and Lucian,74 and the fact that this is an amusing story about magic and adventure – and Lucian was known to tell amusing stories involving magic and fantastic adventure (Philopseudes, Vera Historia) – caused the anonymous epitome to be incorporated among Lucian’s works.”75

73 No exact equivalent can be found elsewhere (but cf. NT Joann. 12.3: ἡ δὲ οἰκία ἐπληρώθη ἐκ τῆς ὀσμῆς τοῦ μύρου). 74 For the easy confusion between these two name forms, also Mason (above n. 3): 1680 provides some very telling examples. 75 Hall (above n. 10) 366–7.

SH 906 and the Apollo of Simias of Rhodes: Some Issues of (mis‑)Attribution Marco Perale The provenance of SH 906 = P.Mich. III 139 (MP3 1458, LDAB 3912) is unknown. The papyrus was purchased in Egypt by B.P. Grenfell and F.W. Kelsey in March or April 1920.1 It was first published by J.G. Winter in 1922,2 commented on by A. Körte,3 and republished with corrections and amendments for the Michigan Papyri series by Winter himself, who tentatively assigned the fragment to the “Hesiodic cycle and the Boeotian school.”4 This attribution was subsequently rejected by R. Merkelbach, who thought to have identified the papyrus text as an excerpt from Simias of Rhodes’ Apollo.5 Merkelbach’s analysis did not win the favour of P. Parsons and H. Lloyd-Jones, the masters in fragmentary poetry, who included the fragment in the adespota papyracea section of their Supplementum Hellenisticum.6 Finally, in one of his most recent contributions, L. Di Gregorio, following in Merkelbach’s footsteps, also ascribed the fragment to Simias.7 The present article purposes to show the unlikeliness of Merkelbach’s theory, highlighting the contradictions undermining such an ascription. I print below the text of the Supplementum, with a few alterations. The apparatus summarizes a variety of different views and approaches to text, and includes new suggestions by Peter Parsons privately communicated to the author (*). ] ϲ̣ τα φύει χθὼν τηλεθάοντα ] π̣ ων · ἱερὴ δ’ ἀποκίδναται ὀδμή ]οιο πολυπτύχου ἠγαθέοιο ]νει ἑκατηβόλου Ἀπόλλωνοϲ 05  ]’ Ἀχελωίωι εὐνηθεῖϲα. Βο ]ρέηϲ τε νότοϲ τ’ ἐν πείραϲι γαίηϲ

1   P. Heilporn, APIS entry no. 1317. 2 Winter 1922 (=Winter1). 3 Körte 1927. 4 Winter 1936, 19–22 (=Winter2). 5 Merkelbach 1951, 257–260. 6 Lloyd-Jones, Parsons 1983, 411–412. 7 Di Gregorio 2008, 87–98.

208

marco perale

10 15

]οιϲι μινυνθάνει· ἀγλαὸν ἥβην ]πηϲι νότοϲ καὶ ἀπὸ κρύοϲ ἔλθηι ] κεκαλυμμέναι· ἂψ δ’ ἐπὶ Γαῖαν ] ἐπὶ μητέρα· αἶψα δ’ ἔπειτα ] καλὰϲ ϲείουϲιν ἐθείραϲ ] τ̣ α περὶ χροῒ νύμφαι ἔχο[υ]ϲαι· ] ν̣ το παρ’ Ἱπποθόην ἰάνοφρυν· ] α̣ καὶ Ἥρην χρυϲοπέδιλον ]ωνοι ἐν οὐρανῶι ἀϲτερόεντι ] τ̣ εϲ ἀκηδέα θυμὸν ἔχοντεϲ· ] ϲειϲίχθονα ποντομέδοντα· ] ̣ ἤπιον Ἀμφιτρίτην

1 φύλλα δὲ πλε]ῖ̣ϲτα Merkelbach || 2 τηλόϲ’ ἀπ’ ἀνθρ]ώ�̣ πων Merkelbach : κ]α̣ ρ̣πῶν dub. Winter2, sed ρ non legitur || 3 κρατὸϲ ἀπ’ Οὐλύμπ]οιο Körte : τηλόϲ’ ἀπ’ Οὐλύμπ]οιο Winter1 : ὡϲεὶ ἀπ’ Οὐλύμπ]οιο Merkelbach : Ῥιπαί]οιο Livrea || 4 ἐν τεμέ]νει Winter1 : ἔνθα παραὶ vel ἔνθα δὲ πὰρ τεμέ]νει etiam possis : Θάλεια μέ]νει Della Corte ap. Traversa || 5 ναίουϲ’ ἅϲ τέκε Γαῖ]’ Merkelbach : Μελπομένηϲ] (sic) Della Corte : Τερψιχόρη χαρίεϲϲ]’ e.g., Parsons* || 6 βο]ρέηϲ omnes || 7 Μοῖρα δὲ τ]οῖϲι Winter1 : δένδρεϲι καὶ φύλλ]οιϲι Merkelbach : ἤμαϲι χειμερί]οιϲι Perale | μινυνθανει· ὰ⟦υ⟧γλαον pap. | μινυνθάνει Crönert, Lobel ap. Winter2, recc. Merkelbach et Lloyd-Jones – Parsons: μίνυνθ’ ἄνει Winter1, Körte | μινυνθάνει ἀγλαὸν ἥβην· Winter2, Körte : ἀγλαὸν ἥβην / ἀλλ’ ὁπόταν θάλ]πηϲι Merkelbach : ἀγλαὸν ἥβην / πρηὺϲ ἐπὴν τέρ]πηϲι dub. Parsons* || 8 ´̣]πηϲι vel ´̣ ]̣ πηϲι pap. | θάλ]πηϲι Körte : τέρ]πηϲι Perale | ἔλθηι̠ pap. || 9 οὐκέτι δ’ ἔγκεινται vel similia possis | κεκαλυμμέναι· ὰψ pap. || 10 επ⟦ε⟧ιμητερα⟦αγν⟧ αιψα⟦ι⟧δ’ pap. | Φοῖβοϲ Ἀπόλλων εἶϲ’] ἐπὶ μητέρα Merkelbach || 11 αἱ μὲν ἀπεϲϲύμεναι] καλὰϲ Merkelbach || 12 εἵματα ϲιγαλόεν]τα Winter, recc. Körte et Merkelbach | εχο[υ]ϲ⟦ιν⟧αι pap. || 13 παρ’ ϊπποθοηνϊανο⟦δ⟧φρυν pap., ϊάν- Winter2, vestigiis congruens : ἵάν- malint LloydJones et Parsons | ἰάνοφρυν Taillardat : ἰανόφρυν Körte : κυάνοφρυν Campbell Bonner ap. Winter1 || 14 Κρονίω]ν̣α Parsons*, recte, ut videtur : Δήμητ]ρ̣α Körte : Ζῆνά τε μητιόεν]τ̣α contra vestigia Merkelbach : Δ]ία Della Corte | χρυϲοπέδειλον pap. : corr. Lloyd-Jones et Parsons || 15 υἱ]ωνοὶ Körte : ἔνθ’ αἱ μὲν γλυκύφ]ωνοι Merkelbach || 16 ὥϲ τε θεοὶ ζώον]τεϲ Körte : ναίουϲιν παίζο]ν̣τεϲ Merkelbach || 17 αἱ δ’ αὖ πρὸϲ Κρονίδην] Merkelbach || 18 Ἰϲτίην, Δήμητρα καὶ] Winter1

. . . [abundant?] the earth nourishes in bloom / . . . a sacred smell8 spreads from the peaks(?) of rocky . . . / [temple?] of far-shooting Apollo / [Gaia?] having slept with Acheloos [begot the nymphs?] / . . . Boreas and Notos at the edges of the earth / [during the winter?] are minimal, / and when Notos [delights the nymphs’] splendid youth, and the cold goes away, / [they no longer stay? they leave?] hidden [by?] / . . . once more into the presence of Gaia / . . . before the Mother. Immediately / . . . they shake their beautiful tresses / . . . the nymphs, wearing [glittering robes?] / . . . towards violetbrowed Ippothoe / [the son of Cronos?] and gold-sandalled Hera / . . . in the starry sky / . . . untroubled in their hearts / . . . earth-shaker, lord of the sea / . . . gentle Amphitrite. 8 Di Gregorio’s “di frutti (?) uno straordinario odore si diffonde” is, if not wrong, misleading, as the connective particle δέ in v. 2 introduces a new sentence.



sh 906 and the apollo of simias of rhodes

209

It has been noted that the passage draws significantly from the archaic formulaic patrimony, recalling Hesiod in particular “in tone and epithet.”9 It can hardly be denied that the vocabulary is reminiscent of archaic models. The formulaic οὐρανὸϲ ἀϲτερόειϲ (15), occuring twenty-five times in archaic hexameter poetry, is never used by any of the Alexandrian poets.10 The same can be said about the clausola πείραϲι γαίηϲ, occurring in the same sedes in Hom. Od. 9.284, Hes. Th. 622 and h.Hom. 5.227, and accurately disassembled by A.R. 3.679–680 ἐπὶ γαίηϲ / πείραϲι and 4.1566 γαίηϲ ἐνὶ πείραϲι;11 the spondaic ending ἑκατηβόλου Ἀπόλλωνοϲ, not recorded outside the Homeric corpus; and the Hesiodic hemistich ἀκηδέα θυμόν + ­participle of ἐχω (Th 61; Op. 112, 170), otherwise attested only in [Lucian.] AP 7.308.1 = GVI 957 (II c. ad?). The consonance in terms of content, vocabulary and diction between the papyrus text and two passages of the Homeric Hymn to Aphrodite on the myth of Tithonus and the Dryads has also gone unnoticed. The two texts, paralleled in the table below, suggests a dependence of the anonymous author on the last section of Aphrodite. SH 906 shows a rather harsh treatment of hiatus, such as the one in conjunction with short element in 10] Μητέρα· αἶψα, and the one in coincidence with the main cesura in 4]νει ἑκα-.12 Unlike other Hellenistic poets, Simias’ hexameters13 are not immune to striking hiatuses (cf. fr. 6.2 Powell ῾Εκτοροϲ, ἥντε οἱ αὐτῷ ἀριϲτῆεϲ Παναχαιῶν) and “epic corrections”; limitations granting him the epithet of “pear barely edible” (βρωτὴ ἀ�́ χρας) in Meleager’s introduction to the Garland (AP IV 1.30 = HE 3955). However, the incidence of the correptio epica in Simias (9 cases in 34 verses, only 2 in conjunction with bucolic diairesis) is still much lower than in the papyrus text, which displays eleven cases of correptio in sixteen verses.

9 Winter 1936, 20. 10 The clausula is systematically avoided, as one would expect, by Nonnus, who playfully evokes it in D. 2.166 = 18.161 and 40.408 οὐρανὸν ἀcτερόεντι/-εϲ . . . χιτῶνι/-εϲ; “programmatically” varies it in 31.242 οὐρανὸν ἀμπελόεντα μετ’ ἀcτερόεντα καλέccω; and intentionally modifies it in D. 2.335 and 44.173, where he substitutes ἀϲτερόειϲ with ἀϲτερόνωτοϲ and ἀϲτερόφοιτοϲ respectively. The formula resurfaces in the form of a Homeric borrowing in Batr. 168, in the archaicizing SH 990.7 (partially supplied), in the Oracula Sibyllina (x6), and in two late Homeric Hymns (30.17, 31.13). 11 Campbell 1981, 52. 12 Note how in 4 the formulaic ἑκατηβόλου Ἀπόλλωνοϲ is strikingly used despite the prosodical difficulty implied. 13 I refer to the corpus of fragments in Powell 1925, including epigrams and the doubtful ep. 22.

210

marco perale

h.Hom. 5 225 227 228–229 258 265–266

SH 906 τὸν (Tithonus) δ’ ἦ τοι εἵωϲ μὲν ἔχεν πολυήρατοϲ ἥβη (Eos) ναῖε παρ’ Ὠκεανοῖο ῥοῆιϲ ἐπὶ πείραϲι γαίηϲ· αὐτὰρ ἐπεὶ πρῶται πολιαὶ κατέχυντο ἔθειραι / καλῆϲ ἐκ κεφαλῆϲ ὄροϲ μέγα τε ζάθεόν τε ἔφυϲαν ἐπὶ χθονὶ βωτιανείρηι / καλαὶ τηλεθάουϲαι

7

ἀγλαὸν ἥβην

6

ἐν πείραϲι γαίηϲ

10–11

αἶψα δ’ ἔπειτα / ] καλὰϲ ϲείουϲιν ἐθείραϲ πολυπτύχου ἠγαθέοιο ] ϲ̣ τα φύει χθὼν τηλεθάοντα

3 1

Such a high percentage would hardly be admissible in the work of an erudite grammatikos. The perfunctory repetition of nine ἂψ δ’ and ten αἶψα δ’ ἔπειτα also suggests, as far as we can tell from the fragmentary state of the poem, a certain syntactical clumsiness. All these factors result in a somewhat impersonal piece of poetry, attributable to the author’s own limitations rather than to a deliberate archaicizing intent. Simias’ epic vocabulary has already been thoroughly analyzed by Di Gregorio, who stressed not only a “desiderio di differenziarsi dal modello” but also a (co-existing?) adherence to Homeric formulaic language.14 It should be mentioned, however, that as many as five of the ten examples provided by Di Gregorio to illustrate Simias’ alleged close dependence on the models are actually taken from the papyrus poem; of the other five, three are from Simias’ Gorgo, a poem (at least partially) centred on Troy and characterized by a much more evident Homeric patina;15 none is from the Apollo, the work where Simias’ style, according to Fränkel, is most evidently recognizable: nullum substantivum nudum poneretur, sed omnia aut adiectivis aut attributis quae dicuntur additis condecorarentur; . . . omnino oratio mollis est et lenis, adiectivis et attributis ubique additis lenta dulcisque fit, distichorum

14 Di Gregorio 2008, 63: “Il nostro poeta . . . si limita a far sue certe espressioni formulari inserendole nei propri versi nella stessa posizione dell’ Iliade, dell’ Odissea e dei poemi che dalla loro lingua prendono fin le singole locuzioni, come quelli del ciclo e dello stesso Esiodo e gli inni omerici.” 15 Perale 2010. Of these five, three cannot be technically defined as “formulaic,” differing from the Homeric parallels in either their metrical placement or the use of preposition: see fr. 6.5 P. ἐν νηυϲὶν ἐβήϲατο ποντοπόροιϲιν ~ Hom. Il. 7.72 παρὰ νηυϲὶ δαμείετε ποντοπόροιϲιν, 13.381 ἐπὶ νηυϲὶ ϲυνώμεθα ποντοπόροιϲιν, 13.628 ἐν νηυϲὶν μενεαίνετε ποντοπόροιϲι; fr. 6.6 P. ἐκ πάντων Δαναῶν ~ Hom. Il. 16.85 πρὸϲ πάντων Δαναῶν; ep. 20.3 P. ἐν νομῷ ὕληϲ ~ Od. 10.159 ἐκ νομοῦ ὕληϲ.



sh 906 and the apollo of simias of rhodes

211

divisione perspicua et gracilis, nitet figuris et glossis modice immixtis, variata oppositionibus, imitationibus culta.16

Surprisingly enough, Merkelbach applied Fränkel’s statement to the Michigan fragment,17 where the only markedly innovative elements seem to be two previously unattested words, at 7 μινυνθάνει and at 13 ἰάνοφρυν, and the use of ἀποκίδναμαι, not recorded before A.R. 4.133. However, with regard to the first, a possible occurrence of an aorist form of μινυνθάνω in B. 5.151 μίνυθεν δέ μοι ψυχὰ γλυκεῖα was recently noticed.18 As for the second, the learned horizon of the ambitious anonymous poet, not necessarily in conflict with his inadequate writing skills, would also explain the adoption of what may well be an Apollonian coinage (cf. Livrea ad A.R. 4.133). Although the geographic setting of the fragment remains obscure, it is highly improbable that the fruitful χθών mentioned in vv. 1–2 is the land of Hyperboreans. According to Merkelbach’s reconstruction, vv. 6–7 οὐ . . . βο]ρέηϲ τε νότοϲ τ’ ἐν πείραϲι γαίηϲ / δένδρεϲι καὶ φύλλ]οιϲι μινυνθάνει could be paraphrased as follows: “Bei den Hyperboreern, am Ende der Welt, wird die Blüte der Natur nicht durch kalte Winde gestört” oder “in dieser Zeit unterbrechen am Ende der Welt, im Mittelmeergebiet (von den Hyperboreern aus gesehen) kalte Winde die Blüte der Natur.”19 However, Merkelbach failed to recognize that a seasonal change like the one described in vv. 5–6 could not have affected the territories located beyond the north wind,20 as poets refer to the peoples located in both the most southern and northern lands as not weatherbeaten.21 If the Boreas never affected the Hyperboreans, living in perennial summer, the coming of warm spring could hardly have induced a change in the nymph’s behaviour.

16 Fränkel 1915, 23. 17 Merkelbach 1951, 260. 18 Perale 2009; or, alternatively, an imperfect from *μινυνθάω, as pointed out by Valerio 2010. The Bacchylidean passage was certainly known by the anonymous author, 5.151–154 μίνυθεν . . . ἀγλαὰν ἥβαν clearly being echoed in 7 μινυνθάνει . . . ἀγλαὸν ἥβην: see Perale 2009, 4. 19 Merkelbach 1951, 258. 20 On the etymological value of the ethnonym, vd. Call. Del. 281–282 οἳ καθύπερθε βορείηϲ οἰκία θινὸϲ ἔχουϲι with Mineur ad loc.; Paus. 5.7.7 ἀνθρώπουϲ οἳ ὑπὲρ τὸν ἄνεμον οἰκοῦϲι τὸν Βορέαν. 21  Str. 1.3.22 εἰ δ’ ἄρα, τοῦ Ἡροδότου (4.6) τοῦτ’ ἐχρῆν αἰτιᾶϲθαι, ὅτι τοὺϲ ὑπερβορείουϲ τούτουϲ ὑπέλαβε λέγεϲθαι, παρ’ οἷϲ ὁ βορέαϲ οὐ πνεῖ. καὶ γὰρ εἰ οἱ ποιηταὶ μυθικώτερον οὕτω φαϲίν, οἵ γ’ ἐξηγούμενοι τὸ ὑγιὲϲ ἂν ἀκούϲαιεν, ὑπερβορείουϲ τοὺϲ βορειοτάτουϲ λέγεϲθαι. ὅροϲ δὲ τῶν μὲν βορείων ὁ πόλοϲ, τῶν δὲ νοτίων ὁ ἰϲημερινόϲ· καὶ τῶν ἀνέμων δ’ ὁ αὐτὸϲ ὅροϲ.

212

marco perale

In this land ἐν πείραϲι γαίηϲ (6),22 Boreas and Notos, normally blustering during the winter,23 are reduced (μινυνθάνει, 7). The transition from the winter (ἤμαϲι χειμερί]οιϲι, 7) to the spring, when Notus turns into a gentle, warm breeze (ἀγλαὸν ἥβην / πρηὺϲ ἐπὴν τέρ]πηϲι νότοϲ, 7–8),24 is almost imperceptible. Still, the coming of the spring is celebrated, year after year, by the choruses of nymphs in honour of their mother (μητέρα, 10), identifiable with Γαῖαν at 9 and with the spouse of the Acheloos at 5: βο]ρέηϲ τε νότοϲ τ’ ἐν πείραϲι γαίηϲ / ἤμαϲι χειμερί]οιϲι μινυνθάνει·· ἀγλαὸν ἥβην / πρηὺϲ ἐπὴν τέρ]πηϲι νότοϲ καὶ ἀπὸ κρύοϲ ἔλθηι / οὐκέτι δ’ ἔγκεινται] κεκαλυμμέναι· ἂψ δ’ ἐπὶ Γαῖαν . . .25 A different explanation would be given by capitalizing Μητέρα in 10, making her the mother not only of the nymphs (including Hippothoe at 13),26 but also of the Olympian divine couples mentioned in the fragment: (Zeus) & Hera at 14, Poseidon & Amphitrite at 17–18.27 The two pairs would be separated, the first living in the sky (ἐν οὐρανῶι ἀϲτερόεντι, 15) and the second in the sea (ποντομέδοντα, 17).28 If Gaia coincided with the

22 Perhaps on the Isles of the Blessed, as suggested by Powell 1929, 196–197. 23 Cf. Pi. fr. 94b.17–19 Maehler χειμῶνοϲ ϲθένει / φρίϲϲων Βορέαϲ ἐπι-/ϲ̣πέρ̣χ̣ηϲ’; S. Ant. 335 χειμερίῳ νότῳ. 24 Perhaps before ceasing to blow, cf. Thphr. Vent. 10 τοῖϲ μὲν βορείοιϲ χειμῶνόϲ τε καὶ θέρουϲ καὶ μετοπώρου μέχρι τοῦ λήγειν, τοῖϲ δὲ νοτίοιϲ κατὰ χειμῶνά τε καὶ ἀρχομένου (ἔαροϲ) καὶ μετοπώρου λήγοντοϲ. 25 Merkelbach’s aporetic interpretation was promptly noticed by Di Gregorio 2008, 92: “il v. 8, in cui si parla della fine della stagione fredda, non riguarda la località descritta negli esametri precedenti, lontana e misteriosa ed immaginata quasi al di fuori del mondo (v. 6), perchè in essa la natura è in pieno rigoglio (vv. 1–2) e, almeno al momento, non c’è l’inverno. Vi si dovrà vedere piuttosto un accenno all’arrivo della primavera sulla terra, menzionata al v. 9: se . . . leggiamo con il Körte ἀλλ’ ὁπόταν θάλ]πῃϲι, nei vv. 8–10a si potrebbe dire che, allorché sul nostro pianeta noto (scil. Noto), un vento caldo del sud, riesce ad intiepidire l’atmosfera e il freddo se ne va, le ninfe, lasciando il τέμενοϲ di Apollo, dove sono state quando c’era la cattiva stagione, vi fanno ritorno (ἂψ δ’ ἐπὶ γαῖαν) e si recano dalla loro madre.” Di Gregorio suggests to divide the fragment into three sections: vv. 1–7a (punctuate after μινυνθάνει) would refer to the Hyperboreans; 7b–10, providing a hint (“un accenno”) to the spring coming back to the earth (9 ἐπὶ γαῖαν); vv. 10–18 portraying the nymphs on the Echinades islands, where Hippothoe (13) was taken in captivity by Poseidon (17): cf. [Apollod.] 2.4.5. The mother mentioned in 10 would be Perimele, loved by Acheloos (Hes. fr. 10a.34–36 M.-W. e [Apollod.] 1.7.3) and turned into one of the Echinades (Ov. Met. 8.591, Perimele). However, as admitted by Di Gregorio himself, Perimede’s name in 4/5 would be impossible to scan. On the other hand, a geographically tripartite division seems to be hardly reflected in the Greek text. 26 In Hesiod, Th. 251 Hippothoe is a Nereid. 27 Amphitrite is mentioned as Poseidon’s spouse in Hes. Th. 930. 28 Parsons*, suggesting that Hades and Persephone may well have followed to complete the tripartite division.



sh 906 and the apollo of simias of rhodes

213

Mother of the Gods,29 one could even think of καλὰϲ ϲείουϲιν ἐθείραϲ at 11 as alluding to an orgiastic dance celebrating the rejuvenation of nature.30 But this is, of course, highly speculative. The most economical interpretation is to assume an attempt by a second-century31 learned amateur to recreate the traditional motif of the locus amoenus.32 The compatibility of the plot of Simias’ Apollo with the episode narrated in the papyrus text is also questionable. The longest fragment of the Apollo, fr. 1 Powell, is centred on the peregrinations of an unnamed character from the land of the Hyperboreans to the mountains of India, crossing Scythia and the river Oxus.

Τηλυγέτων δ’ ἀφνειὸν ‘Υπερβορέων ἀνὰ δῆμον, τοῖϲ δὴ καί ποτ’ ἄναξ ἥρωϲ παρεδαίϲατο Περϲεύϲ, ἔνθα δὲ Μαϲϲαγέται θούρων ἐπιβήτορεϲ ἵππων ναίουϲιν τόξοιϲι πεποιθότεϲ ὠκυβόλοιϲιν··  5 θεϲπέϲιόν τε περὶ ῥόον ἤλυθον ἀενάοιο Καcπάϲου, εἰϲ ἅλα δῖαν ὃϲ ἀθάνατον φέρει ὕδωρ. ’Εκ δ’ ἱκόμην ἐλάταιϲι περὶ χλωρῇϲιν ἐρυμνὰϲ νήϲουϲ ὑψικόμοιϲιν ἐπηρεφέαϲ δονάκεϲϲιν. ‘Ημικύνων τ’ ἐνόηϲα γένοϲ περιώϲιον ἀνδρῶν, 10 τοῖϲ ὤμων ἐφύπερθεν ἐϋϲτρεφέων κύνεοϲ κρὰϲ τέτραφε γαμϕηλῇϲι περικρατέεϲιν ἐρυμνόϲ. Τῶν μέν θ’ ὥϲτε κυνῶν ὑλακὴ πέλει, οὐδέ τι τοίγε ἄλλων ἀγνώϲϲουϲι βροτῶν ὀνομάκλυτον αὐδήν.33

29 As, for instance, in h.Hom. 30 and in P.Derv. XXII 7–12 Tsantsanoglou-Parássoglou. 30 Active forms of ϲείω alongside with objects such as πλόκαμον, κόμην etc. are often employed in bacchic contexts: see E. Ba. 184–5, ποῖ δεῖ χορεύειν, ποῖ καθιϲτάναι πόδα / καὶ κρᾶτα ϲεῖϲαι πολιόν; 240–1 παύϲω κτυποῦντα θύρϲον ἀναϲείοντά τε / κόμαϲ; 930–1 ἔνδον προϲείων αὐτὸν (scil. πλόκαμον) ἀναϲείων τ’ ἐγὼ / καὶ βακχιάζων ἐξ ἕδραϲ μεθώρμιϲα; cf. also Luc. Alex. 13 ϲείων ἅμα τὴν κόμην ἄνετον ὥϲπερ οἱ τῆι μητρὶ ἀγείροντέϲ τε καὶ ἐνθεαζόμενοι. In Pi. fr. 70b.8–14 Maehler, the voices of the Naiads mix with the clashes of the typana in honour of the Great Mother: see Bowra 1964, 62–3; cf. also P. 3.77–9, with Lehnus 1979, 10. In the Orphic Hymn to the Horai (43.7–9), the Moirai and the Graces accompany the return of Persephones from the Underworld with circular dances, pleasing the Fruit-giver (μητέρι καρποδοτείρηι). 31  That is, not much earlier than the material used to write the poem: cf. Livrea 1985, 599. 32 Associated with the nymphs’ enthusiasmos since Plato, Phdr. 230 b–c. For late antique descriptions of the coming of the spring, see De Stefani 2008. For a general survey of loci amoeni, Haß 1998. 33 Quoted from Perale 2011a, 368–369. On the geography of the fragment, see pp. 378–382.

214

marco perale

The main character of this fragmentary section was correctly identified by H. Fränkel as Cleinis,34 the protagonist of one of the transformation tales of Antoninus Liberalis (Synagoge XX = Simias fr. 2 Powell). Fränkel’s assumption was base on a marginal annotation in the only manuscript of Antoninus extant, the Palatinus Heidelbergensis graecus 398 (middle of ninth century): Κλεῖνιϲ. Ἱϲτορεῖ Βοῖοϲ β´ (= of his Ornithogony) καὶ Ϲιμμίαϲ ὁ Ῥόδιοϲ. In Antoninus’ Metamorphosis, Cleinis is represented as a wealthy man from Babylon, where he lives with his wife and four children, devoted to the cult of Apollo and Artemis. Cleinis habitually travels in the company of the two gods to the land of the Hyperboreans, where he has seen performed sacrifices of donkeys. Cleinis is once tempted to repeat the sacrifice in Babylon, but is warned by Apollo not to do so, the donkey sacrifice being prohibited outside the Hyperborean territory. Cleinis reports Apollo’s apparition to his family, but two of his children, against the god and their father’s will, proceed with the sacrifice. Apollo, in response, strikes the animals with madness, and they end up devouring Cleinis in their frenzy. Cleinis is eventually resurrected by Apollo and turned into a bird along with the members of his family.35 While it is impossible to say how extensively Simias treated the subject, it seems reasonable to think that the last section of Antoninus’ tale, entirely devoted to bird transformations, was developed exclusively 34 Fränkel 1915, 17. As pointed out by Fränkel, the use of ἀνά + accusative in 1, and the transmitted περί as opposed to Wilamowitz’s παρά in 5 may suggest that Cleinis flew over those lands, transported per aera by Apollo. 35 Τῆϲ λεγομένηϲ Μεϲοποταμίαϲ περὶ Βαβυλῶνα πόλιν ὤικηϲεν ἀνὴρ θεοφιλὴϲ καὶ πλούϲιοϲ ὄνομα Κλεῖνιϲ, ἔχων πολλοὺϲ βοῦϲ καὶ ὄνουϲ καὶ πρόβατα. τοῦτον ἐκτόπωϲ ἐφίληϲεν ’Απόλλων καὶ ῎Αρτεμιϲ καὶ πλειϲτάκιϲ ὁμοῦ τοῖϲ θεοῖϲ τούτοιϲ ἀφίκετο πρὸϲ τὸν ναὸν τοῦ ‘Απόλλωνοϲ τὸν ἐν ‘Υπερβορέοιϲ καὶ εἶδεν ἱερουργουμέναϲ αὐτῶι τὰc θυϲίαϲ τῶν ὄνων. παραγενόμενοϲ δὲ εἰϲ Βαβυλῶνα καὶ αὐτὸϲ ἐβούλετο καθάπερ ἐν ‘Υπερβορέοιϲ ἱερεύειν τῶι θεῶι καὶ τὴν ἑκατόμβην τῶν ὄνων ἔϲτηϲε παρὰ τὸν βωμόν· ’Απόλλων δὲ παραγενόμενοϲ ἠπείληϲεν ἀποκτενεῖν αὐτόν, εἰ μὴ παύϲαιτο τῆϲ θυϲίαϲ ταύτηc καὶ κατὰ τὸ ϲύνηθεϲ αἶγαϲ αὐτῶι καὶ πρόβατα καὶ βοῦϲ ἱερεύϲει· τὴν γὰρ τῶν ὄνων θυϲίαν ἐν ‘Υπερβορέοιϲ ἀγομένην αὐτῶι καθ’ ἡδονὴν εἶναι. καὶ ὁ Κλεῖνιϲ δείϲαϲ τὴν ἀπειλὴν ἀπῆγεν ἀπὸ τοῦ βωμοῦ τοὺϲ ὄνουϲ καὶ τὸν λόγον, ὃν ἤκουϲεν, ἐξέφερε πρὸϲ τοὺϲ παῖδαϲ. ἦϲαν δὲ αὐτῶι παῖδεϲ Λύκιοϲ καὶ ῞Αρπαϲοϲ καὶ θυγάτηρ ’Αρτεμίχη, παῖδεϲ ἐκ μητρὸϲ ῞Αρπηϲ. Λύκιοϲ μὲν οὖν καὶ ῞Αρπαϲοϲ ἀκούϲαντεϲ ἐκέλευον ἱερεύειν τοὺϲ ὄνουϲ καὶ τέρπεϲθαι τῆι ἑορτῆι, ‘Ορτύγιοϲ δὲ καὶ ’Αρτεμίχη πείθεϲθαι τῶι ’Απόλλωνι προϲέταϲϲον κἀπεὶ τούτοιϲ ὁ Κλεῖνιϲ ἐπείθετο μᾶλλον, ῞Αρπαϲόϲ τε καὶ Λύκιοϲ κατὰ βίαν ἐκλύϲαντεϲ τῶν δεϲμῶν τοὺϲ ὄνουϲ ἀπήλαυνον παρὰ τὸν βωμόν. καὶ ὁ θεὸϲ ἐνέβαλε τοῖϲ ὄνοιϲ λύϲϲαν· οἱ δὲ τούϲ τε παῖδαϲ καὶ τοὺϲ θέραπαϲ αὐτῶν καὶ τὸν Κλεῖνιν κατήϲθιον. οἱ δὲ ἀπολλύμενοι τοὺϲ θεοὺϲ ἐπεβοῶντο. καὶ ῞Αρπην μὲν καὶ ῞Αρπαϲον ὤικτειρε Ποϲειδῶν καὶ ἐποίηϲεν αὐτοὺϲ ὄρνιθαϲ τῶι αὐτῶι λεγομένουϲ ὀνόματι, Λητὼ δὲ καὶ ῎Αρτεμιϲ ἔγνωϲαν ἀναϲῶιϲαι τὸν Κλεῖνιν καὶ τὴν ’Αρτεμίχην καὶ τὸν ’Ορτύγιον, ὅτι οὐκ αἴτιοι τῶν ἀϲεβημάτων ἦϲαν· ’Απόλλων δὲ Λητοῖ καὶ ’Αρτέμιδι δίδωϲι τὴν χάριν καὶ πρόϲθεν ἢ ἀποθανεῖν μεταβαλὼν ἐποίηϲε πάνταϲ ὄρνιθαϲ. A detailed account of the bird species the family members are transformed into follows.



sh 906 and the apollo of simias of rhodes

215

in Boeus’ Ornithogonia. However, the circumstances of Cleinis’ trip in Simias fr. 2 seem not easily reconcilable with the content of fr. 1, where the reference to the Hyperboreans is limited to the first couple of verses. If, as believed by Fränkel, fr. 1 represents an excerpt from Cleinis’ report of the mirabilia experienced during his return trip to Babylon, it is not clear why in fr. 1 the episode of the onosphagia, central in Antoninus’ tale, was omitted. Di Gregorio’s explanation of fr. 1 as an introductory hymn serving as a framework for a multi-faceted epyllion is certainly appealing, but not grounded in suggestive pieces of evidence.36 Di Gregorio posits a short composition based on multiple myths related to Apollo and the god’s punishment of devious and impious behaviours. One section of the epyllion would concern Cleinis, the second Marsias (most probably mentioned in fr. 3 Powell), the third Niobe (fr. 5 Powell). In Powell’s edition of Simias, embedded in his Collectanea Alexandrina, the fragments ascribed to the Apollo are indeed five (fr. 1–5). Fr. 3 Powell Χρυϲῶι δ’ αἰγλήεντι προϲήρμοϲεν ἀμφιδαϲείαϲ / κόρϲαϲ καὶ ϲτόμα λάβρον ὀπιϲθοδέτοιϲιν ἱμᾶϲιν is cited by two later sources on Marsyas’ musical rivalry with Apollo: Plu. 456c (with no further attribution); and Tz. H. 1.375–377 Leone, assigning it to Simonides. Although the confusion Simias/Simonides is not unparalleled in scholiastic tradition,37 and the style of the fragment is indeed strongly reminiscent of Simias, there appears to be no strong evidence suggesting any link with the above mentioned mythical events.38 Fr. 4 Χρυϲῶι τοι φαέθοντι πολύλλιϲτοϲ φλέγεται κράϲ “in shiny gold his head blazes, much prayed for” is assigned to an unspecified work of Simias by some manuscripts of Choeroboscus, and, again, to Simonides by others. It contains an address to an unknown god, who may well also be Helios wearing a golden helmet.39 Fr. 5, a passage on Assaon’s incestuous love for Niobe from the prose epitome of Parthenius’ narrative poem on Assaon’s incestuous love for Niobe (XXXIII), is also ascribed to Simias by the annotator of the codex preserving Antoninus: Ἱϲτορεῖ Ξάνθοϲ Λυδιακοῖϲ καὶ Νεάνθηϲ β΄ καὶ Ϲιμίαϲ ὁ Ῥόδιοϲ. Here, caution is suggested not only by the parallel witness of the scholia, highlighting several discrepancies in the actual treatment of the myth by Xanthos, but also by ‘fresh evidence’ on parallel fragmentary poets. The publication of PSI XIV 1319 (MP3 371, LDAB 877), for instance, 36 Di Gregorio 2008, 78; cf. Perale 2011a, 371–372 n. 20. 37 Perale 2011b, 195. 38 Only Fränkel included it among the incertorum carminum fragmenta (fr. 5). 39 Fr. 7 Fränkel; Perale 2011b.

216

marco perale

has yielded a text of Euphorion’s Thrax that appears to be very different from the one expected from Parthenius’ prose version.40 A more prudent approach would be to assign the first two fragments to the Apollo and the other three to other works from the four-book poetic corpus assigned to Simias by Suidas (ϲ 431, p. 360.7 Adler Ϲιμμίαϲ, Ῥόδιοϲ, γραμματικόϲ. ἔγραψε Γλῶϲϲα βιβλία γ´· ποιήματα δίαφορα βιβλία δ΄). As for the Apollo, we may think of a hymn or an epyllion, in which fr. 1 P. served as a prologue. This introductory catalogue may have listed all the marvellous places accessed by Cleinis with the aid of Apollo and the adventures experienced in each place, later developed in autonomous sections. The episode of the donkey sacrifice may just have been one of the fabulous events that Apollo and the Hyperboreans had in store for Cleinis. References Bowra, Cecil Maurice (1964), Pindar, Oxford. Campbell, Malcom (1981), Echoes and Imitations of Early Epic in Apollonius Rhodius, Leiden. De Stefani, Claudio (2008), “L’epigramma longum tardogreco e bizantino e il topos dell’arrivo della primavera,” in Alfredo Maria Morelli (ed.), Epigramma longum. Da Marziale alla tarda antichità. From Martial to Late Antiquity, Cassino II, 571–602. Di Gregorio, Lamberto (2008), “Sui frammenti di Simia di Rodi, poeta alessandrino,” Aevum 82 (2008): 51–117 Fränkel, Hermann (1915), De Simia Rhodio, Gottingae. Haß, Petra (1998), Der locus amoenus in der antiken Literatur: Zu Theorie und Geschichte eines literarischen Motivs, Bamberg. Körte, Alfred (1927), “Literarische Texte mit Ausschluß der Christlichen,” Archiv für Papyrusforschung 8: 251–254. Lehnus, Luigi (1979), L’inno a Pan di Pindaro, Milano. Lightfoot, Jane (1999), Parthenius of Nicaea, Oxford. Livrea, Enrico (1985), review of Lloyd-Jones/ Parsons (1983), Gnomon 57: 592–601. Lloyd-Jones, Hugh and Peter John Parsons (1983), Supplementum Hellenisticum, Berolini – Novi Eboraci. Merkelbach, Reinhold (1951), “Über zwei epische Papyri,” Aegyptus 31: 254–260. Perale, Marco (2009), “Una nota a Bacchyl. 5.151–4,” Zeitschrift für Papyrologie und Epigraphik 171, 3–4. —— (2010), “Il. Parv. fr. 21 Bernabé e la Gorgo di Simia di Rodi,” in Ettore Cingano (ed.), Tra panellenismo e tradizioni locali: generi poetici e storiografia in Grecia, Alessandria: 497–518. —— (2011a), “Il catalogo ‘geografico’ di Esiodo: due diversi casi di ricezione nella prima età ellenistica” in Antonio Aloni, Massimiliano Ornaghi (eds.), Tra panellenismo e tradizioni locali. Nuovi contributi, Messina: 365–89. —— (2011b), “Simia e la testa del Sole: fr. 4 Powell,” Eikasmos 22, 195–200. Powell, John Undershell (1925), Collectanea Alexandrina, Oxonii.

40 Lightfoot 1999, 542–545 (on Simias and Xanthus) and 516–519 (on Euphorion).



sh 906 and the apollo of simias of rhodes

217

—— (1929), New Chapters in the History of Greek Literature, Second Series, Oxford. Valerio, Francesco (2010), “Ancora su Bacch. 5.151,” Zeitschrift für Papyrologie und Epigraphik 175: 3–4. Winter, John Garrett (1922), “Some Literary Papyri in the University of Michigan Collection,” Transactions and Proceedings of the American Philological Association 53: 128–41. —— (1936), Papyri in the University of Michigan Collection: Miscellaneous Papyri, University of Michigan Studies, Humanistic Studies vol. 40, Michigan Papyri vol. 3, Ann Arbor.

Order, Ambiguity, and Authority in Venantius Fortunatus, Carm. 3.26 Joseph Pucci Given that it increases interpretability, it would seem that ambiguity diminishes the authority of the writer who deploys it. But I argue in this paper that by allowing his readers to understand the opening and closing couplets of carm. 3.26 in discrete ways, Venantius Fortunatus exploits ambiguity in order to control the ways in which this poem might be understood, thus affirming most powerfully his poetic authority. Further, I argue that the poet understands ambiguity to be a species of order, akin to antithesis, rhyme, parallelism, and the like. Modes of poetic order found throughout the poem thus go to the larger interpretive project highlighted at the poem’s beginning and end, where ambiguity, as the most important kind of ordering, controls the poet’s words. In exploiting ambiguity in this way in carm. 3.26, I hope to suggest the ways in which Fortunatus reaches simultaneously a public and a private audience, in order to validate a notion of friendship as we understand it, that is, as “a personal relationship predicated on affection and generosity.”1 For, as I hope to suggest, there would be no need to write under ambiguity’s authority if carm. 3.26 exploited friendship simply as a rhetorical move, that is, as a way to pay a requisite compliment to a social superior. On the contrary, precisely because ambiguity connects counterpoised public and private spaces, this poem reveals the precise power of friendship’s bonds by framing private attachments against the larger fabric of public acclaim, enabling the affiliations of friendship to be seen as equivalent to the grandest of things – God, Christ, the soul, nature, and ultimately ensuring that the poet’s own voice is unambiguously, authoritatively, understood to say as much. Let me begin with a note about the term “ambiguity,” which has received much attention over the past century: more positively by the New Critics, who considered it the acme of “good literature”;2 and recently, but less 1 The phrasing is from D. Konstan, Friendship in the Classical World (Cambridge, 1997), 5. 2 Most famously by W. Empson, Seven Types of Ambiguity (London, 1931, rep. and rev. ed., 1949), 23: ambiguity is “any verbal nuance, however slight, which gives room for

220

joseph pucci

optimistically, by the Poststructuralists, who have seen in it the harbinger of the indeterminacy of the word.3 While there is much to be said for these views, and for others articulated in the theoretical space between them, I adhere in what follows to Quintilian’s definition: [in ambiguity] two things are signified and, so far as the written or spoken word is concerned, the two sides are equally balanced [Duas enim res significari manifestum est et quod ad scriptum vocemve pertinet in utramque partem par est; Inst. Orat. 7.9.14].4

On this elegantly simple view, two discrete features made emphatic by word placement stand out: first, an ambiguous word harbors two meanings (the duas of the definition’s opening) that are, second, thematically equal (the par of the definition’s ending), qualities pointed up in the term by which Quintilian designates “ambiguity” –, not the Latin ambiguitas, but the Greek amphibolía, a word whose literal sense, “a state of being attacked on both sides,”5 elicits the image of a single word controlling two discrete meanings.6 As will become clear in what follows, Fortunatus would seem to adhere to this notion of ambiguity as part of a larger project in which he says two things at once in order to clarify his thematic aims. a­ lternative reactions to the same piece of language.” On this view, however, as A. Kaplan and E. Kris, “Aesthetic Ambiguity,” Philosophical and Phenomenological Research (1948): 415–435, later observed, just about anything could be considered ambiguous. On the other hand, their articulation of categories of ambiguity –, disjunctive, additive, integrative, conjunctive, projective – relied more on subjective responses to the multiplicity of meanings inherent in words, less on anything stable in language itself that might control, or authorize, ambiguity. Their tradition of schematization, taken up subsequently by others, such S. Rimmon-Kenan, The Concept of Ambiguity: The Example of James (Chicago, 1977), led in turn to a rejection of this approach, in the words of J. Hillis Miller, “The Figure in the Carpet,” Poetics Today 1 (1980), 107–118, as “overly rigid.” 3 For an overview of the development of ambiguity from literary category to cultural force, see G. Graff, “Determinacy/Indeterminacy,” in F. Lentricchia and T. McLaughlin, eds., Critical Terms for Literary Study (Chicago, 1990, sec. ed., 1995): 163–176. 4 The Latin and English are from D.A. Russell, trans., Quintilian, The Orator’s Education, Books 6–8 (London and Cambridge, 2001): 286–87. 5 LSJ9, p. 90, s.v. amphiboleús, –ía. 6 The etymology of the Latin ambiguitas, ambo + gerere, literally, to bear something in two ways, is nearly identical, but is too close to the English “ambiguity” to stand clear of current synonym – vague, ambivalent, unclear, muddled, and so forth – that confuse, rather than clarify, the issue. It may perhaps be that Quintilian chose a Greek word for just this reason, since ambiguitas bears some of these current associations in antiquity also. Quintilian also commends doubleness and equality in his own definition, in which, in the phrase, in utramque partem par est, one sees and hears “par” twice – and in the very phrase, and concerning the very words, that articulate the qualities of doubleness (in partem) and equality (par est) that Quintilian otherwise highlights.



order, ambiguity, and authority in venantius fortunatus 221

Of the many poems written by Fortunatus that feature ambiguity, carm. 3.267 is among the more complex, and readers can see at once the ways in which rhetorical modes of ordering dominate the poem’s articulation: Altaris Domini pollens, bone Rucco, minister,   hinc tibi festinus mando salutis opus.  Nos maris Oceani tumidum circumfluit aequor,   te quoque Parisius, care sodalis, habet.  Sequana te retinet, nos unda britannica cingit: divisos terris alligat unus amor. Non furor hic pelagi vultum mihi subtrahit illum nec boreas aufert nomen, amice, tuum. Pectore sub nostro tam saepe recurris amator, tempore sub hiemis quam solet unda maris. Ut quatitur pelagus quotiens proflaverit eurus stat neque sic animus te sine, care, meus. Blanda serenato tempestas pectore fervet atque ad te varia mobilitate trahit. Sed memor esto mei votumque repende petenti ut pariter paribus det sua dona Deus, humanum mentem Christi quo gratia ditet ac Domino nostro sensus et ora vacent. Good Rucco, deacon, rock of Christ’s altar, I’m dashing off a hello from here. The ocean swells around me, friend: while Paris holds you, huddling the Seine, Breton waves harass me, dear, yet one love binds our separation. Friend, no sea-borne fury banishes your face, no northern wind carries off your name; in my heart you’re a lover who appears as often as the ocean cuffs the wintry shore like the sea shaken when the east wind blew: so stands my soul without you, dear, while a pleasant storm burns a docile heart, bearing down on you in all its complication. Requite my wish, Rucco: think of me, share the gift that God gave to us both, to let Christ’s grace make us more fully human, to speak, to think only of who commands us.8

5

10

15

5

10

15

7 If its title and the contents of book 3 are to be believed, Rucco had not yet assumed the episcopal seat at Paris when Fortunatus wrote carm. 3.26 to him, placing its composition at some point between the poet’s arrival in Gaul, in 566, and Rucco’s installation as bishop, in 576. On these particulars see J. Pucci, Venantius Fortunatus: Poems to Friends (Indianapolis, 2010), 18–19. 8 The translation is from Pucci, Venantius Fortunatus, 19.

222

joseph pucci

Globally, the poem is ordered in successive blocks, comprising three groups of six lines, each of which contains three couplets. We might see in these larger divisions a simple tendency owed to the exigencies of composing in elegiacs or consider them the result of coincidence, otherwise lacking in meaning, were similar moves on a local level not so consistently plied throughout the poem. In fact, these larger groupings would seem in part to have been designed to announce these local moves, and those moves, in turn, can be understood to ballast the ordering that the poem, on this global view, encourages. In turn, both supply part of the poem’s thematic energy, which is, after all, bound up with twos and threes – that is, with the doublets provided by addressee and author, Rucco and Fortunatus; their respective habitats, Paris and Brittany; the bodies of water with which they are linked, the Seine and the North Sea; and so forth. To this pair of friends can be added the figure of Christ and God, whose presence at the beginning and end of the poem creates a group of three, but who are depicted in the poem as a duplex figure – Father and Son – who nonetheless is also triune. But what of these more local moves that privilege twos and threes? V. 1, for example, mimics the poem’s global groupings by allowing its six words to be gathered into three sets of two, either like this: Altaris Domini pollens, bone Rucco, minister,

which respects the line’s syntax; or like this: Altaris Domini pollens bone Rucco minister

which represents the way the poet orders parts of speech in the line. The figures mentioned in the line go to the privileging of twos and threes also, for two are named, Rucco and Christ, but the latter, as I just noted, is triune. V. 2 might be ordered in an identical way, that is, three groups of two words each: hinc tibi festinus mando salutis opus

but the somewhat odd pairing of the first two words, hinc and tibi, perhaps draws attention to them by dint of their difference, for it is not clear at first under what logic they might be connected. They are not akin syntactically, as festinus mando and salutis opus are, designating respectively subject/verb and direct object. Yet they do suggest the ways in which order can augment theme, for their proximity bridges the divide that the poem



order, ambiguity, and authority in venantius fortunatus 223

otherwise narrates, insofar as hinc betokens the place from which Fortunatus writes, while tibi, designating Rucco, suggests the recipient of what is written and therefore points up the theme of separation that energizes the poem throughout. These words thus visualize in their connection the reunion of friends, attainment of which is the poem’s goal. Vv. 3–4 exploit the horizontal ordering practiced by Fortunatus in his first couplet, but its lines are vertically aligned, too, through the parallelism of their initial words, nos and te. Since they operate on a different axis, so to speak, I don’t include them on the horizontal plane in which the poet would otherwise seem to ply orderly strategies. On this plane, in any case, v. 3 appears nearly identical to v. 1 in respect of the ordering that syntax provides: both lines have two nouns in the genitive, followed by an adjective separated from the noun it modifies; here, by the verb circumfluit, and in v. 1, by the vocative bone Rucco: NOS maris Oceani tumidum circumfluit aequor,

Syntax provides this horizontal grouping of words for v. 4 – all of these gatherings going, I suggest, to the poet’s continued interest in twos and threes: TE quoque Parisius, care sodalis, habet,

Let me return for a moment to the parallelism enabled by the first words of this couplet, for the vertical energy there found provides a further way for the poet to emphasize the double business of his theme – not for nothing in verses that contain more than a few doublets, both within and across verses, including maris Oceani, tumidum aequor, care sodalis, circumfluit, and habet, among others. Moreover, in contrast to the depiction of potential unity suggested in the beginning of v. 3 in that line’s initial words, hinc tibi, the first words of these verses have an opposite function, with the separation of nos and te standing for the separation of the poet and friend in real time. This might allow us to see the opening of v. 2 as antithetical to that of vv. 3 and 4, again, with a clear bow toward playing with twos and threes – two words comprise each part of the anthesis, but within three lines, etc.: Altaris Domini pollens, bone Rucco, minister, hinc tibi festinus mando salutis opus. Nos maris Oceani tumidum circumfluit aequor, te quoque Parisius, care sodalis, habet.

224

joseph pucci

A strategy of ordering that emphasizes twos and threes remains in place in vv. 5–6. V. 5, for example, tugs forward the parallelism exploited in the previous couplet, but is now organized around parallel clauses within the line: Sequana te retinet, nos unda britannica cingit: 5

The light rhyme of retinet and cingit helps to situate the balance of the clauses more firmly, but that balance is further exploited through word placement within the clauses: the verbs of both go last, but in the initial clause the subject is first, while in the second, the object is first, making for a pleasing symmetry that also provides another instance of doubling – that is to say, parallel construction is both interior and exterior to the clauses. More than this, too, is the fact that both clauses dramatize the idea of the line, that is, that either figure is surrounded in his particular circumstance and in some way held back by water, for we “see” the te that is Rucco surrounded in his clause, while the nos that is Fortunatus is similarly surrounded, with nos standing as both the first word of the second clause and the mathematical center of the line as a whole. The urge to compose a line that dramatizes through order the idea it communicates also carries over into v. 6. Picking up on the thought of the previous line, Fortunatus declares that, despite separation, love binds him to his friend, an idea that order ramifies in this line, whose words can be gathered into three groups with the verb in the middle forming the pivot to which the other words are bound, as follows: divisos terris alligat unus amor.

Alligat thus can be understood to do in this line what it signifies, that is, bind two things that are separated – precisely the point the poet wishes to make here respecting the single love that he shares with Rucco that supercedes physical separation. Vv. 7–8, which, on a global view, recall, mark the beginning of the second of the poem’s three parts, seem to acknowledge this division by offering orderly strategies that more often involve vertical pairings rather than groupings within individual lines that seem to predominate in vv. 1–6. For example, it is difficult to discern obvious patterns of order among the individual words of vv. 7 or 8, but between these lines parallel words draw attention to the beginnings and endings of both. Thus, the initial words, non and nec, are parallel, as are the words that follow each, furor and boreas, both subjects in their respective lines. Similar is the parallelism of illum and tuum, both adjectives in final position in their respective lines,



order, ambiguity, and authority in venantius fortunatus 225

whose presence, in turn, tugs readers back to the nouns with which they agree. Both stand at the beginning of the second half of their lines, illum going with vultum, the fifth word of an eight word line, and tuum with nomen, the fourth word in a six word line. There is thus a symmetry to the words so flagged – and also, I hasten to add, a bow toward twos and threes yet again. Non furor hic pelagi vultum mihi subtrahit illum   nec boreas aufert nomen, amice, tuum.

The tendency for orderly strategies to reach vertically between the lines of couplets, rather than horizontally within lines, continues in vv. 9–10, where parallelism of a kind just examined is augmented by rhyme. One can easily see, or hear, for example, in the opening of v. 9, tempore sub nostro, the aural similarities that affiliate it to v. 10, tempore sub hiemis, including the rhyme of pectore and tempore and the repetition of sub. More than this, though, is the rhyme provided by tam and quam in their respective lines, an aural connection that confirms both words’ coordinating function in the couplet. Thinking about the lines in this vertical way, however, reveals at the same time a horizontal order of sorts, allowing for the first four words of both lines to be joined with three subsequent words following in both lines and the initial word of both beginning with “s.” Pectore sub nostro tam saepe recurris amator,   tempore sub hiemis quam solet unda maris. 10

In turn, vv. 11–12 offer horizontal and vertical orderings alike. Both lines of the couplet perhaps break down into two equal groups of words, the hexameter comprised of two groups of three words each, the pentameter of two groups of four words, these corresponding to metrical, syntactical and/or mathematical orderly schemes. But the rhyming final words of each line, eurus and meus, link the couplet vertically also, while the repetition of initial “q” and “p” in the interior words of v. 11 is perhaps meant to mimic the beating of the stormy sea or the blowing of the eastern wind. Ut quatitur pelagus quotiens proflaverit eurus   stat neque sic animus te sine, care, meus.

Vv. 13–14 mark, on the global view, the beginning of the third of the poem’s three parts. The orderly energies in these lines continue to inhabit vertical, rather than horizontal, spaces but again draw attention to the first

226

joseph pucci

and last words of both lines. For the phrase atque ad that commences v. 14, when elided, lightly rhymes with the initial word of v. 13, blanda, while the final syllables of the second last words of both lines, “re” and “te” fully rhyme. Too, there is more than a little resemblance in sound in the final words of each line, fervet and trahit: Blanda serenato tempestas pectore fervet   atque ad te varia mobilitate trahit.

A larger ordering principle would seem to be at work here, too – one that returns readers to the first third of the poem. Recall that the poem would seem to be grouped into three blocks of six lines. V. 6, that is, the last line of the first part of the poem, would seem to be controlled by the same orderly principles as v. 13, the first line of the third part of the poem. The five words of v. 6, recall, were gathered into three groups, two comprised of two words each, with the third containing one: divisos terris alligat unus amor.

The six words of v. 13, by distinction, are gathered also into three groups and can be ordered by parts of speech, that is, adjective, noun, verb: Blanda serenato tempestas pectore fervet.

or, respecting the syntax of the line, on a chiasmic model: Blanda serenato tempestas pectore fervet

Much as the separation of phrases in v. 6 underscored the power of love expressed in that line, v. 13 visualizes unity in the tempestas that is at its center, whose manifestation is in the heart that covets friendship, and whose binding power is betokened in the way the nouns and adjectives of the line fold in on each other. By distinction, in vv. 15–16 horizontal ordering prevails, relying on connections of sound and sense. In v. 15, for example, the final two words, repende petenti offer a soft rhyme, while, in v. 16, the symmetry and partial rhyme of pariter paribus help the poet confirm the unity that friendship sponsors and that God gives as a gift. At the same time, the duplex phrase pariter paribus trumpets the unity offered by the triune God, a play on twos and threes that is also ratified in the twice repeated “p” of pariter paribus and the thrice-repeated “d” of det sua dona Deus.



order, ambiguity, and authority in venantius fortunatus 227 Sed memor esto mei votumque repende petenti  ut pariter paribus det sua dona Deus,

15

But, as I said at the start, in this last third of the poem, I wish to focus on ambiguity as perhaps a more important orderly strategy, that is to say, as a way of suggesting interpretive bipolarity, rather than a unified meaning, for the poem’s situation of discourse – a binarism portended and confirmed by the strategies of order, both global and local, that otherwise infiltrate the poem. We can spy ambiguity as early as v. 13, where the “storm” there mentioned picks up on the storms described in vv. 7–12, but has now become a storm of the heart, and a pleasant one. The ambiguity of pectore ramifies this sentiment, for the “heart” in question in which this storm rages may be Rucco’s, owed to his reading of this poem; or it might be Fortunatus’, in which case the poet thereby expresses his love for Rucco, with pectore the locus of that love, that proffers the “pleasant storm” of the poet’s devotion to his good friend. Perhaps most importantly for my purposes, the final couplet of the poem provides a conclusion controlled by the word dominus, whose ambiguity, if my reading of these lines is correct, fashions a statement of friendship’s power. It is hard to ignore this word thematically, to be sure, but, as if to insist on the interpretive pivot it provides, Fortunatus seems determined in multiple ways to gain his readers’ attention by shrouding it, literally, in doubles and triples – mirroring his moves throughout the poem. Altaris Domini pollens, bone Rucco, minister,  1  hinc tibi festinus mando salutis opus. humanam mentem Christi quo gratia ditet  ac Domino nostro sensus et ora vacent. 18

Thus, the word appears in two lines in the poem, the first and last, where it is the second word in both lines; two meters are used in the two lines in question; the first words of the first and last lines both begin with the letter “a,” altaris in v. 1; ac in v. 18, thus forming two phrases of two words each in which dominus figures in two lines, i.e., Altaris Domini, v. 1 and ac Domino, v. 18; the pentameter coupled with v. 1 and the hexameter coupled with v. 18 both begin with words starting with the letter “h,” hinc, in v. 2; humanam, in v. 17; while the number of words in that figure in signalling ambiguity, both at the start and the end of the poem, is three. More than this, though, the interpretive doubling provided by the ambiguity of dominus resonates here. Recall that in v. 15 Fortunatus orders

228

joseph pucci

Rucco to remember to requite his prayer by sending some good word to the poet as a response to this poem. That response, Fortunatus goes on to say in v. 16, will be, like the poem itself, a God-given gift that the two equal friends deserve to have in equal measure. But then, seemingly on the cusp of subsuming human friendship under the rubric of something God bestows as a gift, Fortunatus trumps this view, in v. 17, by shifting his focus away from God to Christ, and then from Christ to the humanity that Rucco and Fortunatus share with him. For v. 17 is an injunction to allow Christ’s grace to enrich the human mind, that is, the humanity that affiliates one mind for the next. Emphatic in v. 17, humanam points out the way in which Fortunatus wishes to stress not merely friendship, but the human space in which friendship is found. This emphasis on the humanness of friendship in turn presses on the ways in which we might read the poem’s final line. Its words seemingly conclude the thought of v. 17, whereby the grace of Christ that enriches the human mind of the poet and of Rucco allows for both to think and to speak only, in Fortunatus’ words, of our lord –, domino nostro. To be sure, dominus can be Christ, mentioned in the previous line and also at the start of the poem. But the emphatic phrase humanam mentem leans on how we are to understand dominus, too. If we read the final four lines of the poem as creating an increasingly human and private space where friendship abides – moving, as these lines do, from God, to Christ, to the human mind, and to the specific acts of thinking and speaking (sensus et ora) that are controlled by the human mind, then the dominus of v. 18 can also be Fortunatus himself, Rucco’s “lord” in his role as a gift-giver demanding his due in return. Of course, the statement can also apply to friendship writ large, in which case Rucco is as much a dominus as Fortunatus – he has, after all, occasioned the grand gestures of this very poem and proven in the process that he is equally Fortunatus’ lord where friendship is concerned, just as, in demanding a response to this poem in the first place, Fortunatus is his. And in that move, we see the poet playing with twos and threes yet again: for ambiguity here supports not two, but three readings. As a final bow to this complicated poem I wish to notice the way in which its ambiguous conclusion steps on the way we might also understand its opening couplet. Recall that the poet establishes cues that make readers think of the opening and closing couplets in tandem – the repetition and placement of dominus, the repetition and placement of “a” and “h” in the couplets’ initial words, not to mention all of the horizontal and vertical groupings throughout the piece that trumpet duplex and triplex



order, ambiguity, and authority in venantius fortunatus 229

schemes – all of this attaches the ambiguity of the dominus of v. 18 to the dominus of v. 1. On this view, then, the dominus of both verses is read in a like way, that is, as Christ or as “one who commands us”; doing so, in turn, points out the ambiguity of the vocatives minister and pollens of v. 1. When read as the nominative subject of mando, these designate Fortunatus himself, while the figure who does the commanding on this reading can only be Rucco. Fortunatus is thus heard to call himself “a powerful minister of the altar of one who commands [me],” or better to say, he expresses his devotion to Rucco by declaring that he worships at the altar of their friendship. The pun imbedded in salutis opus then takes on further resonance, for as a statement of greeting it stands for the words the poem delivers, but it is phrased in such a way as to suggest also the salvific work involved in worship, even when that worship goes to friendship. But why compose in such a way? One motive perhaps goes to a desire on Fortunatus’ part to articulate actual devotion to Rucco, that is, to use David Konstan’s phrase, to suggest “a personal relationship predicated on affection and generosity.” For what better way to affirm his own participation in the worlds in which he finds himself than to compose pieces that meet readerly expectations in a public posture – an expected posture – but then to complicate those expectations with an ambiguity that goes to more private concerns and proves the staying power of friendship’s emotion? At the least, it seems to me, there would be no need to write under ambiguity’s authority if these poems did not seek to communicate something about a personal relationship predicated on affection and generosity – precisely because the need to construct these pieces in this way would otherwise be unnecessary. The public posture would be enough. But if I am correct respecting the moves Fortunatus would seem to make in this poem, it was not enough. In the final gestures the poet offers a grand affiliation of Rucco and of friendship to larger things – God, Christ – but he insists in connecting these larger things to the more private concerns of the poet’s resolve. For there is no getting around the ways in which the conception of Christ as dominus, so unambiguously articulated at the poem’s start, renders that much more powerful the figure of “one who commands us,” as both Christ and one’s good friend at the poem’s conclusion. On this view, then, the enrichments of the human mind that Christ’s grace bestows are, precisely, human allurements, those things that bind us in our imperfections, not least the power of friendship to affirm human connection and the power of poetry to celebrate it.

Authors Pseudonyms in the Seventeenth Century: The Case of Gaspar Scioppio Eustaquio Sánchez Salor The use of pseudonyms has a long history, but the seventeenth century is its quintessential century. This article will not go into the history of the literary pseudonym through this century, but I must, however, point out that among the reasons that have been given to explain the general use of the pseudonym, the fundamental one during the seventeenth century is to hide the real name, which may be considered praiseworthy, unlike the work headed by the pseudonym. That is, the pseudonym is used when the author writes something that either the author or his readers consider unworthy of his abilities. The author writes what his conscience encourages him to write, but that in his opinion does not enhance the fame of the real name. Such is the case with many clergymen and other good people of the seventeenth century who write reviews and satires that are unworthy of their social position and name. To avoid the clash between the work and the name of the author, they resort to the use of a pseudonym. A significant case is that of Jean-Baptiste Poquelin; as the son of a Parisian furniture merchant and the king’s upholsterer (“valet de chambre”), he was part of the Parisian bourgeoisie. In 1642 he graduated from the University of Orleans as a lawyer, a profession that he never practised but which was useful for him to draw upon the plots of some of his works. In 1643, when he was twenty-one years old, he abandoned his middle-class past and decided to devote his life to the theater. This was the moment when he took the pseudonym “Molière,” so that his surname would not be dishonoured. In the world of satire and theater the name and work of an author was often rejected, or the author even confronted, for social reasons. This article will not delve into this world, but instead will explore the liberal arts and, to be precise, grammar. In the world of grammar in the seventeenth century, writing a handbook conveyed social significance: if he wrote an introductory handbook, the author showed himself to be a lowprofile school teacher; if he wrote a rational and philological grammar, the author showed himself to be a philological connoisseur and a follower of the most important grammatical schools of the seventeenth century.

232

eustaquio sánchez salor

But what of the person who wrote grammars of both types, the author who wrote an intoductory grammar and a grammar of deep analysis? In such cases, resorting to a pseudonym was very useful. The proper name, the real one, was used for the profound grammar with a philological and rational nature, and the pseudonym was left for the introductory grammar. This way, the author eased his conscience or fulfilled his teaching obligations by writing an introductory grammar, while also satisfying and pleasing his linguistic ego by writing a grammar of deep significance. In the first case, when he achieves his compromise with society, but not his aspirations of being a grammar author, he used the pseudonym; in the second case, when he was able to fulfil his aims as a high-flying grammarian, he turned back to his real name. This is the case with Gaspar Scioppio. 1. Gaspar Scioppio: Several Pseudonyms and One Name for Two Types of Grammars Gaspar Scioppio was born in Germany in 1576; he first studied in Amberg and later in Heidelberg. In 1599 he converted, in Rome, from Protestantism to Catholicism. It has been said that he started writing books at a very early age. Indeed, many carry either his name or his pseudonym. The number of works he authored has been put at 104; many of them appear under the pseudonyms “Nicodemus Macer,” “Oporinus Grabinius,” “Pascasius Grosippus,” and “Mariangelus a Fano.” The focus of the present article is on grammatical pseudonyms. The relevance of Scioppio as a grammarian is evident if we take into account his most important contributions in this field. These are twofold. In the first place, he introduces himself as the grammarian who will bring order to the chaos afflicting the Netherlands and Germany; namely, the contemporary teaching of grammar. It is a patriotic duty which he undertakes at the behest of the elite of his homeland. In order to fulfil his commission, he writes an introductory grammar that clarifies and dispels the chaos: we refer to his Rudimenta. Second, we will examine his rational perception of grammar, which places him at the level of Sanctius and the reasoned grammar of Port Royal. This perception is the fundamental axis of his Grammatica philosophica.



the case of gaspar scioppio

233

2. The Introductory Grammar of Gaspar Scioppio Scioppio’s grammatical works are, then, two: the Rudimenta and the Grammatica philosophica. The Rudimenta was the result of the first grammatical task of Scioppio: to turn the chaos present in the Netherlands and Germany into order. For that task there is no need of a grammar with a profound scientific significance. More useful would be an introductory grammar that provides rules and clarifies pedagogic ideas. This was the Rudimenta, which was published not under his real name, but under pseudonyms; the Grammatica philosophica, on the contrary, does appear under Scioppio’s name. It would seem that he only wishes to associate his name with what, from his point of view, is the most important work, while adopting pseudonyms for the grammar that he considers less important. We are indeed looking at two very different grammatical works. The Rudimenta is a introductory work, written under pressure; the Grammatica philosophica is a work for display; it offers rational linguistics, with Scioppio in the main following the sixteenth-century Spanish grammarian considered in the seventeenth century to be the father of rational grammar, Brocense. In fact, besides his Grammatica philosophica, Scioppio also published a Minerva edition of Brocense with commentary. The pseudonyms used for the Rudimenta are “Pascasius Grosippus” and “Mariangelus a Fano.” The use of these different names is justified by insisting that all the opinions shown by each have been shared with the master, Scioppio; Mariangelo a Fano repeats that he has sought advice from the master Grosippus – that is in fact Scioppio – concerning the doctrine exposed; in the dedication of the Rudimenta, signed under the name of Mariangelo a Fano, he says: De hoc meo consilio cum deinceps ad Auctorem ipsum retulissem (“when I later discussed with the master my own decision”). Furthermore, in the dedication of the Auctarium, which is added to the Rudimenta and signed with the same pseudonym, we read: nec in eo (book) quicquam inuenies quin id vel ex libris Grosippi, qua scriptis, qua editis mutuatus sim . . . arripuerim (“in this book you will not find anything that was not inspired by Grosippus books from which I have gained inspiration, from those written, from those edited . . .”) The introductory grammar appears in 1629 with this cover: Pascasii Grosippi Rudimenta Grammaticae Philosophicae et eiusdem Mercurius bilinguis in usum Tironum paucis mensibus linguam Latinam perdiscere auientium / accessit auctarium Mariangeli a Fano benedicti, cum privilegio Pontificio, Caesario et Regio. Mediolani 1629.

234

eustaquio sánchez salor

This work has several parts, some of them signed with the pseudonym Grosippus and others with the Mariangelo a Fano moniker. There is in fact a curious play on pseudonyms. The dedication1 of the Rudimenta is neither signed, as we have already seen, with Scioppio’s name, the real author’s name, nor with Paschasius Grosippus, the name accompanying the title as the author of the work, but instead with “Mariangelo a Fano.” Having said that, this curious play can be explained. The purpose is to start with the Rudimenta in order to introduce the Grammatica philosophica to the scientific world. Right at the beginning of the dedication of the Rudimenta, signed “Mariangelo a Fano,” there is a reference to a philosophical grammar that appeared the year before authored by Paschasius Grosippus, the same author who appears on the cover of this Rudimenta. It would not have been wise for the dedication in which the philosophical grammar is praised, recently published some months before under the name of Grosippus – in fact the author was Scioppio, and it appeared with his name2 – to be signed by Grosippus; it is not a good idea for anyone to praise his own work, and that is the reason why the dedication is signed with the second pseudonym, “Mariangelo a Fano.” So the pseudonyms in the dedication of the Rudimenta had as their final purpose the introduction of the author’s advanced work to the scientific world. Indeed, Mariangelo a Fano – Scioppio himself –,  says in this dedication that several months before the philosophical grammar from Paschasius Grosippus – also Scioppio – was edited, he had the grammar in his hands, and has studied it with interest; he thought it worthy to study the work of such a well-known man, respected by Christian kings and princes, so that he would not later regret not having studied it.3

1 It is addressed to Illustrissimo et Excellentissimo D. Hyacintho de Gonzaga serenissimi D. Ferdinandi Mantuae ac Montiferrati Ducis filio. And it is dated in Milan in 1628: Hunc libellum praeclarissimo tuo nomine ornatum . . . edere decreui. . . . Mediolano Kl. Decem. anno MDCXXVIII. 2 In fact, this is the work that later will appear under the authentic name of Scioppio. In the Index Librorum Gasparis Scioppii quos latinae linguae et literarum praesertim humaniorum studio conscripsit that would appear in further editions of the philosophical grammar, it is said that the first edition is from 1628: Grammatica philosophica pro latina lingua magistris et tironibus, ut scilicet tirones magistri, magistri autem tirones fiant, Mediolani, 1628. Et Amstelodami, 1659. Nova ejus editio, beneficio V. Ampl. Petri Scavenii plurimis in locis insigniter aucta et recognita ex schedis ipsius Auctoris, et quam hic damus, Annotationibus nunc primum suis locis repositis, Franekere, 1704. 3 (A2r) Postquam superioribus mensibus Pascasii Grosippi Grammatica Philosophica in lucem prodiit, minime satis habui saepius illam domi attenteque manibus versare, quod minime poenitendum operae pretium me facturum arbitrarer congnoscendis iis,



the case of gaspar scioppio

235

He also states that, in order to corroborate the acceptance of the new ratio docendi of the philosophical grammar among knowledgeable people, he has gone to the city’s bookstores pretty well every day to listen incognito to the opinion of the learned people that frequent such places.4 Here he also put to the test Theognis’ ancient saying, and found it true, “Rain never pleases everyone” [“Jupiter never pleases everyone, whether he rains or not.”]5 Some people, the ones who thought of themselves as the wisest, despised his philosophical grammar for being arrogant: it was not accepted, he claims, by those who had set up, so to speak, a literatary eclave, who thought it important that everyone knew that there could be no aspiration to scale any heights of erudition without their input, and that “literature was born with them and will die with them,” as Palaemon boasted of himself in Suetonius.6 Some others, he continues, the honorable, though grammar ignoramuses, would think that there was a need to give reasons for linguistic usage. Truly those who speak without passion and interest, and those who think that this has nothing to do with them, confess, with ingenuity and without remorse, that they have spent all their lives in murky ignorance of the written arts. Some of this type, those who can write something in Latin, say no one – neither Cicero, nor Varro, nor Quintilianus7 – has given them a real and rational explanation of anything that has been written correctly in Latin; so they say that all grammarians refer to language use,

quae ab eius celebritatis viro profecta essent, quem omnes, quorum vel praecipuum in literis nomen est, atque adeo maximi quique Christiani orbis Reges ac Principes pendunt plurimi. 4 . . . Sed hoc amplius curae mihi fuit, ut quibus eorum iudiciis qui se literatos perhibere postulant, noua ista docendi ratio exciperetur, comperirem. Quae me quoque causa impulit, ut quasi quotidie ad tabernam librariam urbis huius, ubi fere homines eruditi ani/(A2v) mi causas confidere solent accederem. 5 Ibi experiundo tandem verum esse sensi, quod tanto ante a Theognide dictum habemus, nec Iuppiter ipse / siue pluat, seu non, omnibus ille placet. 6 Illi nempe, qui quandam quasi literarum officinam instituêre, quique sua plurimum omnem in partem interesse censent, in ea omnes esse sententia vix ad ullam eruditionis summam, nisi ipsis operam addentibus, adspirari posse, secumque natas et morituras literas (quod apud Suetonium de se Palaemon ille Grammaticus iactabat). 7 Scioppio quotes several times these three Latin grammarians. They are the Latin paradigm. That is the reason for the following title given to the Paradoxa in the Index that appears at the beginning of the Grammatica Philosophica: Paradoxa literaria, in quibus multa de literis novè contra Ciceronis, Varronis, Quintiliani aliorumqe hominum, tan veterum, quam recentiorum sententiam disputantur. And it is added that the work was published with Grosippo pseudonym: sub nomine Pasc. Grosippi, Mediolani 1628, et Amstelodami, 1659.

236

eustaquio sánchez salor

not to ratio. They behave the same way that empiricist doctors do, who only trust their experience and do not provide any rationale for to why a disease is better treated with this or that medicine.8 Now, the person who has written the dedication, – Mariangelo á Fano, i.e., Scioppio – following the opinion of those who speak with no interest or passion, declares that our author – Grosippus, i.e., Scioppio – has decided to write a philosophical grammar, since when learning Latin it is not necessary to go by the rules, but simply to know the Latin language. It is outrageous, he says, to teach the students the rules of Latin constructions before they are able to understand Latin. This opinion is so sensible, he claims, that thinking otherwise is typical of someone who is out of his mind and needs treating with hellebore.9 To learn Latin, he suggests the following program: Some months for learning the declensions and conjugations.  Learn by heart the 1141 sentences of the Ianua,10 that he reproduces here in the Rudimenta, with the name of Mercurius bilinguis.  After that he will be able to learn, in two or three months, to translate Cicero and the rules and figures of rational grammar.

The author of the dedication follows up with the news, gotten through letters, that this same method is being used successfully in several towns in Italy, and that it even encourages young people to study Latin with pleasure.11

8 Alii rursum, qui sine ullo studio aut cupiditate loquuntur, nullamque in eo rem suam agi intelligunt, ingenue ac sine retractatione, adhuc in literarum caligine actam omnibus fuisse aetatem, confitentur; eorumque qui aliquo usque latine scribendo processis/(A3r) sent, extitisse prorsus neminem (ne ipsum quidem Ciceronem, Varronem aut Quintilianum) qui eorum, quae vel ipse vel alius recte scripsisset, veram aut idoneam sciret reddere rationem, cum non nisi quotidiana loquendi consuetudine omnes niterentur, haud dissimiliter medicis empiricis sola fretis experientia, cum rationem expedire non possent, cur hoc aut illo potius remedii genere morbi persanentur. 9 Quin immo iidem toto caelo aberrauisse praeceptores sentiunt, qui latina prius Grammaticae praecepta discipulis ac tironibus tradiderint, quam illi latinam linguam intelligerent, quod contra noster hic auctor latinam linguam necessario prius intelligi debere, post id locorum Grammaticae praecepta vel ludibundos percipere posse demonstrat. Quae sententia tantum eorumdem iudicio pondus habet, ut aliter sentire, non nisi eius censeant “cui parum sanum sit sinciput, quique helleboro indigeat”. 10 Handbook from the sixteenth century where the Latin language was taught by using sentences. 11 Cui eorum opinioni plerique alii in diuersis Italiae ciuitatibus (quemadmodum ex eorum mihi constat Epistolis) accedunt, ex quibus sunt qui affirment nouum hoc docendi modum iis locis, quibus eius experimentum capi coeptum sit, satis ex sententiae procedere, atque etiam puerorum / (A3v) animos ad discendum nunc infllammari, quippe



the case of gaspar scioppio

237

But in this program two different phases can be clearly identified: on one side is learning Latin, which is the fundamental objective of the students, achieved through the first two stages; and on the other side is analytical and rational learning of the Latin language, which is in the main an aim for the teachers.12 That is Scioppio’s motivation for writing two treatises. The Rudimenta supports the first stage. The Grammatica philosophica, the second. Thus, the Rudimenta is an introductory grammar, which serves the first objective – to teach Latin to students – in which some additions on rational or philosophical grammar have been included. To justify this addition, and to explain the two grammars’ teaching programmes – the introductory and the philosophical – Scioppio has leaned on the fiction of the two pseudonyms: he publishes the introductory grammar under the name of Paschasius Grosippus – we have already seen in the dedication that this work is not considered worthy of such an important master – but at the same time he adds pieces belonging to the philosophical grammar; and those pieces are added under a different pseudonym: Mariangelus a Fano, i.e., the one signing the Auctarium in which the rational doctrine is depicted. In short, Scioppio’s Rudimenta, which was first published in 1629, is an introductory grammar. Even the author admits several times that it is not the same to write a grammar for students as to write one for the teachers. This one is for students; the Grammatica philosophica will be for teachers. The first one is published under pseudonyms – Paschasius Grosippus and Mariangelus a Fano – and the second one appears under his real name, Gaspar Scioppio. Scioppio does not want the Rudimenta to be known as his work. 3. Scioppio’s Rational Grammar This is the Grammatica philosophica. As written in the Index Librorum Gasparis Scioppii quos latinae linguae et literarum praesertim humaniorum studio conscripsit which can be found at its beginning, the first edition is qui non minorem linguae latinae intelligentiam se octo mensibus, quam alios fortasse vix ipso quinquenio aassecuturos intelligant. 12 Verum nihilominus est, cum in scribenda sua Grammatica Magistris potius quam discipulis auctor consuluerit, hinc usu uenire ut illi non minimum sudent atque elaborent in iis excerpendis quae tironibus ediscenda sunt. Atque eam ipsam ob caussam optarent illi, ut peculiari libello ea solum comprehenderentur, quorum tironibus usus sit, magistro subinde ad ea quae in Grammatica fusius docentur, reiecto.

238

eustaquio sánchez salor

from 1628,13 published again in 1659 by Herzog,14 and in 1664 by Scavenius, in Amsterdam. Scavenius’ edition will show up later: 1704, 1712, 1728. The author understands that this is a work with profound scientific significance, which is why he signs with his own name. The proof that this grammar is understood to have profound scientific significance is evident in the defence that Scioppio mounts in this work, of Sanctius’ rational grammar against the regulation or introductory grammar of, for example, the Jesuits. In the preface, authored by Marquardus Gudius,15 it is made clear that the starting point for this grammar is the Minerva from the Brocense. In the Rudimenta the author used the artifice of pseudonyms to praise philosophical grammar; in this work, he uses, with the same purpose, the option of a friend writing the preface. Indeed, Gudius announces that this work of Scioppio has discovered the real roots of the Latin language, eradicating all the filth that former grammarians have put on it.16 This does not imply that Gudius is taking merit away from Sanctius, a divine man, whose own immortal services are fittingly assessed. Truly Sanctius was the first author who, with sincere and weighty judgment, mended the work of former grammarians; in this way, Gudius hails Sanctius as a Spanish Hercules who cleaned from Augeas’17 stables the vast and oppressive work of earlier authors of grammars.18

13 This is the first work referred to in that Index; these editions are mentioned: Grammatica philosophica pro latina lingua magistris et tironibus, ut scilicet tirones magistri, magistri autem tirones fiant, Mediolani, 1628. Et Amstelodami, 1659. Nova ejus editio, beneficio V. Ampl. Petri Scavenii plurimis in locis insigniter aucta et recognita ex schedis ipsius Auctoris, et quam hic damus, Annotationibus nunc primum suis locis repositis, Franekere, 1704. 14 Gasparis Scioppii Grammatica philosophica: quam cum annotationibus ipsius Scioppii cuilibet paginæ subnexis et quadam observationum accessione instruxit m. Jo. Christianus Herzog. 15 Marquardus Gudius (1635–1689) is a humanist from the seventeenth century, the author of a collection of fables in verse and of a Bibliotheca exquisitissimis libris in theologia, jure, medicina, historia literaria omnique omnium alio studiorum genere instructissima. 16 . . . jam vero dictum est eam partem eruditionis reconditae quae latini sermonis aut caussam exigit aut munditiam observat, perfectissime omnium explicasse Scioppium. 17 In myth, the owner of a herd of bulls which, by the will of the gods, suffered from no disease, and was therefore the largest herd in the country. His stables were famous for never being cleaned. Eurystheus charged Hercules, as one of the labours imposed on him, to clean the stables in a single day. There was enough dung in the stables to make this impossible, but Hercules succeeded by re-routing the rivers Alpheus and Peneus through the stables to wash out the filth. 18 Nihil tamen nostra opinione delibabimus de laude Sanctii, divini hominis, proprio atque immortali quodam beneficio suo satis commendati. Primus enim ille sincero grauissimoque judicio exercuit grammaticos, quorum in doctrinis purgandis, tanquam in sordi-



the case of gaspar scioppio

239

So, Scioppio – says Gudius – follows in the line of Sanctius; he enlightens the doctrine received from Sanctius; he destroys even more monsters than the Spaniard;19 and he adds things that not even Sanctius could suspect.20 Gudius ends by addressing the bad grammarians, forcing them to stop spouting nonsense (nugari); offering them the option of taking advantage of what it is presented in Scioppio’s Grammar; and asking them to adhere, together with Priscian, who is in fact followed by the bad grammarians,21 to Sanctius’ doctrine, properly learning the Ars grammatica, which they have been teaching so badly.22 Therefore, the philosophical grammar is presented as an advance on the Brocense Minerva, superseding the rulegenerated Grammars. After Gudius’ preface, it is Scioppio himself, in the core of his grammar, who defends the need for a rational grammar with philological and scientific significance and disdains the grammar of rules and norms, despite himself being the author of one of them. He is of the conviction, like many in the seventeenth century, that to learn Latin one need not learn a grammar consisting of rules; rather, it could be learned through familiar treatises such as Ianua linguarum, which were no more than a compilation of frequently used elementary Latin sentences. Once Latin was learned this way, there was no need to learn rules. Consequently, the grammar should not be composed of rules, but rational and linguistic, à la Brocense grammar. In this way, Scioppio brags of his knowledge of linguistic historiography. While on the topic of reviewing several grammars, it turns out that he already knew many of them, the old ones – thanks to Putschius’23 ­compilation dissi/mo Augiae stabulo alter Hispaniae suae Hercules immensam operam molestiamque devoravit. 19 Sanctius, in his Minerva, often defines as “monsters” many of the explanations of the preceding grammarians; as an example, in chapter IV of the book Minerva dedicated to the Antiphrasis, he says: Ad nauseam usque repetunt indocti grammatici antiphrasin figuram; Parcas enim affirmant dici quia nemini parcunt, et lucum, quia minime lucet; et huiusmodi sexcenta monstra. 20 Tanti viri vestigiis inhaerens Scioppio passim illuminat quae acceperat, multa adhuc monstra conficit, multa somnia explodit, inveteratam plurimorum verborum peregrinitatem exterminante, plures alias veterum frustationes, varia commenta ostendit, quae ne Sanctius quidem potuit olfacere. 21 Gudius did not know that one of the old teachers who Sanctius followed was precisely Priscian, esp. in books XVII and XVIII of his Grammar. 22   Ita tandem desitum est in Grammaticis nugari! . . . Corrigant (grammatici) istum morbum, cedant occassioni et tandem sapient. Sapient autem si confestim sese cum Prisciano suo tradant in disciplinam Sanctianam et patientissimo animo discant artem, quam frustra docent. . . . 23 Helias Putschius (1580–1606) is the author of a collection of Grammaticae latinae auctores antiqui, 1605.

240

eustaquio sánchez salor

of ancient grammarians – and the contemporary ones from different European countries.24 Among the latter, father Álvarez, a Portuguese Jesuit who published an Art of Grammar in 1572, stood out above all other grammarians as the author of a treatise in which the rules of the grammatical art were clear. Because of this, it was not necessary to redo what he had already done. Given that Álvarez departed from the well-established rules of the old grammarians, he could refrain from making any further analysis on the subject.25 He follows with an analysis of the origins of grammar, and here makes his famous reference to “the cloacina grammar.” Next he harshly criticizes the grammars that list one rule after another;26 there are grammars, he says, that never move away from the construction of names, verbs, and participles, and already accumulate “more than five hundred rules” to remember, which not even five years of full dedication would be enough to learn.27 And it is not only an issue of all these rules, but of the multifarious exceptions that make the rules lose value, all the appendices, annotations, caveats, and sundry other addenda to the rules. They ensure, Scioppio sarcastically assures us, that when the student has learned all the rules he will always be afraid of missing one, or missing an exception, or an appendix.28 Furthermore, those rules are often not even followed

24 Sed factum est, ut veteres omnes quotquot ante mille annos aliquid in ea arte literis consignatum reliquerunt (qui quidem duobus voluminibus Francofurti, anni MDCV editis continentur) tum recentiorum complures, ut quisque in Hispania, Gallia, Germania et Italia praeter caeteros in ea praestitisse visi sunt, cognitos haberem. 25 Inter quos, ut verum fatear, Emmanueli Alvaro primas deberi animadverti, nam et ipse longe cultius dicendi genus, quam non dico veterum quisquam (nam illi pessime omnes latine scrisperunt), sed quam recentiores plerosque in arte tradenda praestitit, et ea ex optimo quoque veterum auctorum exempla seligere habuit, quibus regulae artis plurimum stabiliri et sine negoti a tironibus intelligi possent. Verum enim quoniam ille a veterum Grammaticorum praeceptis, omnium praesertim consensu traditis, discedere religioni habuit, effugere non potuit, quin eosdem fere omnes errores, quos in aliis Sanctius agitavit, in ipsius quoque rite inveniamus. 26 He has written, as we have already seen, an introductory grammar, not a grammar full of sets of rules. In this grammar he proposed a fast learning of declensions and conjugations, and then to move immediately to the learning of short sentences by means of the Mercurius bilinguis. 27 . . . Sed enim in illis qui artem meliore quam caeteri ordine et elegantia scripsisse judicantur, numerum inire velint, de sola Nominum, Verborum et participiorum Syntaxi plus quingentas regulas traditas invenient, quibus ediscendis haud scio an quinquennium solidum satis sit. 28 Sed etiam cum qui eas memoriae mandarit, tamen vel propter earum turbam, vel propter tot exceptiones, quibus potestas earum restringitur, semet haesitet animique incertus sit, necesse est, cum merito vereatur, ne aliqua ei sive regula, sive exceptio, sive appendix effluxerit.



the case of gaspar scioppio

241

by the most famous authors, such as Cicero; even authors of Cicero’s day had to put up with the fact that their rules were not obeyed by the most famous author in the Latin language.29 This idea is presented again at the end of the section on syntaxes. He begins and ends the Grammatica philosophica with the same point: the futility of rule-bound syntax. To show this, at the end of the syntaxes he adds some Annotationes in Syntaxin (p. 183), in which he warns that it would help if grammar teachers possessed a clear understanding of his syntaxes.30 The first thing they should know when reference is made to rules for syntax as transmitted through the rule-generated grammars is that there are too many rules, and that most of them are confusing given that they do not differentiate between concord and regime; and, even worse, they are so full of exceptions, observations, and appendixes, that if they do not drive those learning them crazy, they will bring them to despair.31 Certainly, he declares, the damage that learning these rules causes adds up to: “waste of time, waste of estate, and waste of esteem.” Damna porro quae id genus regulae discentibus afferunt, jactura cùm temporis, tùm rei familiaris, tum denique existimationis spectantur.

That is why Scioppio, in his introductory grammar, was not a slave of the rules, but instead emphasized usage of the language; and that is why, following Sanctius, he wrote his philosophical grammar. 4. Conclusion Scioppio was obliged by the pressure of the noblemen of his country to write an introductory grammar, so that he could shed light on the chaotic grammatical panorama that confused the teaching of Latin. However,

29 . . . Quo accedit, quod ipsi quoque qui Ciceronis aequales aut suppares fuerunt Grammatici, aperte tulerunt regulas suas plerasque ab optimo quoque linguae auctore parum observari. 30 Annotationes in Syntaxin. Primum omnium ad rem pertinere et ex usu discentium fore arbitror, ut magister de Genere Syntaxeos nostrae, qualis scilicet illa sit et quid ab aliorum praeceptionibus ea intersit, tum de usu eius breviter admoneatur. 31 Quod ad aliorum de Syntaxi praeceptiones attinet, res in propatulo posita vociferatur, regulas et nimis multas esse, et confusas, nec certis generibus distinctas, nullo scilicet Concordiae et rectionis discrimine; quodque multo majoris est incommode, tot exceptionibus, appendicibus et observationibus oneratas, ut discentem si minus ad insaniam, ad desperationem quidem certe possint redigere.

242

eustaquio sánchez salor

his certainty as a grammarian was that Latin could not be learned from rule-generated grammars, but through a treatise that would provide a compilation of sentences (Ianuae), while the real grammar was the rational one with linguistic significance; that is why he wrote a Grammatica philosophica. The grammar written for the use of teachers was published, not with his name, but using pseudonyms. But the grammar written to satisfy his philological and linguistic ambitions was published under his real name.

Pseudepigraphy and Magic1 Emilio Suárez de la Torre 1. The “Greek and Demotic Magical Papyri” as Representative of Contemporary Trends, Including Literary Forms In the following pages I will focus on the presence of pseudepigraphy in the magical papyri of Roman Imperial Egypt. However, I want to underline that this is but one among many features which connect these texts to cultural trends of diverse origin, belonging not only to this period, but also to the time of the original composition of the recipes, many of which come from Hellenistic Alexandria. The archives of the Egyptian House of Live accumulated not only the old Egyptian religious, medical and – generally speaking – literary works inherited since ancestral times, but also other texts of disparate origin and content, used by the scribes for different aims, according to their interests. Several of these volumes were of “pseudepigraphical” origin, that is, books falsely attributed to a wide range of authors, some of them very famous, some other less known, but likewise reinforcing the authority of the recipe, spell, or treatise.2 This trend is not exclusive of the magical papyri; on the contrary, it is a wellknown trend shared by ancient literary corpora, particularly well-attested in Greece since the fourth century bc onwards, with a large variety of possibilities, as the chapters of the present volume show. In this aspect, the magical papyri follow a common tendency that must be added to other features aiming to endow the magical texts with instruments of appeal for a wide, Greek-speaking audience. Thus, to the use of the Greek language should be added the recourse to literary forms usual in religious contexts – for example, the Hymns – or in the transmission of wisdom. In this case, it is important to note that several magical texts are conceived as didactic, including also characteristic features of the initiation of the magician.3

1 This article has been made under the frame of the Research Project FFI2011–27438 of the Spanish Ministry of Science and Innovation. 2 Betz (1990) has brilliantly described this feature. 3 See Moyer (2003a).

244

emilio suárez de la torre

Thus, we find epistolary forms4 filtering even into small treatises in which the author addresses the recipient in terms that correspond to a masterdisciple relationship.5 Some of the names attested in the papyri were part of the catalogue of experimented magicians and authors of specialized works circulating since an early date. This list underwent additions and modifications through the centuries. Thus, for instance, Pliny6 mentions an impressive series of medical and magical authors who coincide in good measure with the names used as authorities in the magical papyri: certe Pythagoras, Empedocles, Democritus, Plato ad hanc discendam navigavere exiliis verius quam peregrinationibus susceptis, hanc reversi praedicavere, hanc in arcanis habuere. Democritus Apollobechen Coptiten et Dardanum e phoenice inlustravit voluminibus Dardani in sepulchrum eius petitis, suis vero ex disciplina eorum editis. The witness of Apuleius,7 as Abt observed, is of great importance, as much for the coincidence in many aspects of magical practice as for his mentions of authors and treatises. The most cited passage is found in chapter 90 of his Apologia: si una causa uel minima fuerit inuenta, cur ego debuerim Pudentillae nubtias ob aliquod meum commodum appetere, si quamlibet modicum emolumentum probaueritis, ego ille sim Carmendas uel Damigeron uel † his Moses uel Iohannes uel Apollobex uel ipse Dardanus uel quicumque alius post Zoroastren et Hostanen inter magos celebratus est. These two Roman authors give witness to the normal circulation of books on magic falsely attributed to a series of authors partially coincident with the names appearing in the papyri. It must be stressed that in some cases these names are only that, i.e., mere names – or “empty names” –, with no real work circulating as such, but there were actually magical texts, with a traceable tradition, very often attributed to them.

4 PGM I 42, IV 152, 2006. For the peculiarities of “magical” Greek letters and their deviations from the epistolary normative of antiquity, as well as for their differences with the “malediction” letters, see López Jimeno (1990). 5 PGM XIII 342. 6 NH 30, 2. 7 Apologia 90. Abt (1908): 318–325.



pseudepigraphy and magic

245

2. Pseudepigraphy: List of Names8 Egyptian9 Zminis of Tentyra10 The name is quite usual in private demotic documents.11 It is adapted from Ns-Mn, “He-who-belongs-to-(the god)-Min.” Tentyra (Dendera) is not far (60 km.) from Thebes.12 He is the author of an oneiropompos, in which the name of Ostanes13 also appears. Paysakh from Cusae The aim of the spell is to experiment a ph ntr or σύστασις. The name Paysakh (or Pasash)14 is otherwise unattested and it has even been argued that it is doubtful that Paysakh is a personal name.15 But a name appearing only once in the corpus is not unique. It seems that names of famous magicians are combined with those of more modest priests because they can guarantee an effective and real use of the recipes with more proximity to other practitioners. Moreover, it may not be casual that the name appears in a demotic papyrus, destined to more “local” customers.   8 A good analysis of pseudepigraphy in the magical papyri has been made by Dieleman (2005). This work, as well as the article by Betz (1990; see above, n. 2) have been mostly helpful for this chapter. 9 In this catalogue I will exclude two references to anonymous physicians of the Oxyrhynchite nome, which appear in PDM XIV (1 and 528 respectively: both are spells for vessel divination, see Dieleman 2005: 264–265), because they do not belong properly to pseudepigraphy. Of course, they may be considered a case of resource to prestige authority, if we take into account the fact that physicians had a high reputation in Egyptian society and in their practice there was not a rigid boundary between science and magic. I will also exclude the mention of the hierogrammateusThphe (PGM XIII 958), because the allusion is not to a written work, but to a phrase pronounced by him in presence of King Ochos (Artaxerxes III, 358–338 bc). 10 PGM XII 121. 11 See Henderson-Thompson-Milne (1913), 29, 30, 34, 38, 40, 41, 63. 12 See Dieleman (2005), 264, with references. 13 For this important “mage hellénisé,” see Bidez-Cumont (1938), 167–225 (doctrines) and 267–358 (texts). For his presence in Apuleius and magical texts see Abt (1908), 325– 326. See infra, under the heading “King Pitys and Pitys the Thessalian.” 14 This form was the transcription of Griffith and Thompson (1904–1909), 65. 15 PDM XIV 232–238 (a priest). “The name Paysakh is otherwise unattested. The name, which is written without the seated man determinative, consists of the demonstrative or possessive article pзy and the substantive sh “gall”. Cusae (Kis in Egyptian) was the capital of the fourteenth Upper Egyptian nome in Middle Egypt, about 200 kilometers to the north of Thebes. Nothing is known about its economic, political or religious significance in the Greco-Roman period” (Dieleman [2005], 265, n. 203).

246

emilio suárez de la torre

Pnouthis the scribe16 He appears as the author of a letter17 with instructions to a disciple to acquire an assistant. The name seems Pз-ntr or Pa-ntr “the god” or “he of the god.”18 The recourse to a letter deserves special attention. First, as I have noted above, it is a proof that the authors of recipes are in accordance with contemporary trends in the use of this “genre” for the diffusion of their wisdom. Properly speaking, it is not (or may not be) a case of pseudepigraphy. Or, more exactly, we may imagine these different situations: (a) the sender and the recipient are real people – then there would be no “fake” – who actually maintain a correspondence; (b) the sender is in fact a priest, but he invents a fictional character as recipient (this is again no pseudepigraphical letter); (c) both sender and recipient are real, but the interchange of letters did not happen: this would be a case of pseudepigraphy sensu stricto. Manetho19 His name is well justified in this context: he is a good example of the hellenization of priests. The fragments of his historical work have deserved much attention in the controversy about Moses (see infra).20 According to the spell, he receives from the hands of Osiris a copy of a hierà bíblos, defined as πρόγνω[σ]ις καὶ μνημον[ική]. This is a very interesting version of the Homeric description of the “panchronic” oracular powers of the mantis, as reflected in the following logos: ποίησόν με προγνῶναι τὰ μέλλοντα καὶ τὰ προμέλλοντα καὶ πεπραγμένα καὶ πάντα τὰ ἔργα τῇ σήμερον καθάπαξ (428–30). In the text (in a very fragmentary part) the name of Moses appears in a corrupt passage, but there is no particular objection to the ascription. Nephotes21 This is probably another priest.22 The demotic name23 is Nfr-hētp (epithet of Osiris). This recipe has some features that characterize it as a “highlevel” text if compared to other, more simple spells. To begin with, the 16 PGM I 42–195. 17 Dieleman (2005), 257–258. 18 Dieleman (2005), 257, n. 171. See Thissen (1991), 295. 19 PGM III 424–266. 20   On his work, see Dillery (1999) and Gozzoli (2006), 191–225. 21   PGM IV 154–285. 22 Dieleman (1995), 258. 23 Thissen (1991), 295.



pseudepigraphy and magic

247

author chooses the form of the letter, but not one sent to a pupil, as is the case with Pnouthis. He opts for an effective resource to reinforce the efficacy of the procedure, which we will find again in Pachrates: the recipient is an ancient Pharaoh, Psammetichus.24 Pa(n)crates/Pachrates25 The priest mentioned in this text, where the name of the Emperor Hadrianus appears (a terminus post quem), seems to coincide with the (quasi) homonymous magician of the Philopseudês of Lucianus, in the story related by Eucrates with the theme of the “apprentice of the sorcerer.” He is described there as a man who does not speak Greek very well,26 but is endowed with surprising powers, including the capacity of driving a crocodile and understanding the language of animals.27 Since it happens in the story of Thessalos, it shows that Egyptian priests were renowned and their wisdom sought out by many people.28 First it is called an ἀγωγή, though in fact it has multiple effects; the name of Pachrates is preceded by the title ἐπίθυμα, a “burnt offering,” but it seems applied to the whole series of spells of this part of the papyri, which are of a very varied nature. A remarkable feature is the inclusion of two versified prayers, the first in hexameters and the other in iambic catalectic trimeter. As noted in Betz,29 this group of spells is followed by a similar one,30 including a poetic iambic text. King Pitys (and Pitys the Thessalian)31 The identification with the high priest (prophetes) Bitys is generally accepted: cf. Iamblichus, VIII.5, X.7 (he translated hermetic texts into Greek). The name means “He-who-belongs-to-the-country (Egypt)”

24 Which one? Psammetichus I is perhaps a better candidate than Psammetichus II, because the former unified Egypt and had perhaps more prestige. Psammetichus III seems to be discarded, because he was defeated by the Persians, who would dominate Egypt for many years (525–404 bc). 25 PGM IV 2441–2621 (see 2446–2455). On his identity see now Ogden (2007), 248– 259. 26 Οὐ καθαρῶς ἑλληνίζοντα, 33.21. 27 ἄλλα τε πολλὰ τεράστια ἐργαζόμενον, καὶ δή καὶ ἐπὶ κροκοδείλων ὀχούμενον και συννέοντα τοῖς θηρίοις, 33, 25–26. 28 Moyer (2003b). 29 By O’Neil, ad loc., 82. 30 PGM IV 2622–2784. 31 Described as king in PGM IV 1928 and 2006; and identified with a Thessalian in 2140.

248

emilio suárez de la torre

(derived from Pa-tз).32 The three practices attributed to Pitys share their necromantic nature but differ considerably in many features. The first one (1928–2005) is described as an ἀγωγή, contains a long hymnic prayer to Helios, and seems to have lost the relation with the title, because the described effect is far from being a simple attraction. The second one again takes the form of a letter, this time allegedly sent to King Ostanes33 in response to a consultation of the king about the enquiry of scull cups. Properly said this is a long spell of attraction. Finally, there is a short spell in which Pitys is named “the Thessalian.” As Dieleman34 points out, it reflects only “secondary buildings,” undoubtedly prompted out by the celebrity of Thessalian magic in matters of necromancy. Pibechis35 (+ Apollobex?) This is the Greek Form of Egyptian Pз-bik “The Falcon” or Pa-bik, “He of the Falcon” (Thissen 295). It is an exorcism. Dieleman comments on the possibility that it could be identified with the Apollobex of XII 121.36 This supposition is based on the assumption that the Egyptian name Horbik “Horus-the-Falcon” is translated by Greeks through identification of Apollo and Horus.37 This text has a strong Jewish colour.38 Ourbikos39 This name arises a difficult problem due to the context where it appears. At first sight, one is tempted to explain it as a transliteration of the Roman name Urbicus. At least two Roman citizens are recorded with this name in the history of the Roman Empire: a Governor of Britain in the reign of Hadrian (Quintus Lollius Urbicus), and an agrimensor and jurist, endowed with very good rhetorical qualities, who probably lived in the fourth Century ad (Aggenus Urbicus).40 However, the main difficulty with this possi-

32 See Thissen (1991), 295. 33 See infra. 34 Dieleman (2005), 267, n. 212. 35 PGM IV 3007. 36 For Apollobex see Apuleius, Apologia 90; Plinius, NH 30,9, who says: si una causa uel minima fuerit inuenta, cur ego debuerim Pudentillae nubtias ob aliquod meum commodum appetere, si quamlibet modicum emolumentum probaueritis, ego ille sim Carmendas uel Damigeron uel † his Moses uel Iohannes uel Apollobex uel ipse Dardanus uel quicumque alius post Zoroastren et Hostanen inter magos celebratus est. 37 See Hopfner I (1921), §210, 102. 38 Knox (1938), Lidonnici (2007). 39 PGM XII 316–350. 40 See Castillo (1998).



pseudepigraphy and magic

249

bility is that the name appears in a recipe in which the important Egyptian ceremony of “the opening of the mouth” is mentioned, as Dieleman and Moyer,41 following a proposal of Vergote,42 have convincingly defended. Ieu43 This is the ζωγράφος. Another exorcism (and see also Jewish elements). The reference is to a stele –,  included or transcribed in a letter. Greek/Roman (a) Three “Orphic” authors The Leiden Papyrus J 395 has been usually studied due to an important text it contains – in fact, two versions of the same text, but with additions in each of them – attributed to Moses (see infra). In the second version there is a list of names and references used to reinforce the efficacy of an invocation to have a σύστασις with a god. The first three names are Orpheus,44 Erotylos,45 and Hieros.46 They represent the clearest appearance of an orphic tradition in the magical papyri. The presence of orphic elements in ancient magic –,  and therefore in the PGM – has been exhaustively studied in detail by Raquel Martín,47 and I refer to her study for detailed analysis. (b) Philosophers Pythagoras48 His name is found exclusively in the title of the recipe attributed also to Democritus, with the specification that his work was an ὀνεραιτητόν whereas that of Democritus is described as ὀνειρόμαντις μαθηματικός. The inclusion of Pythagoras and Democritus in the list of authors of this kind of works is perfectly coherent not only with the development of their biographies, which accumulate legendary features, but also with a particularity shared by them which allows an ideal link between Egypt and Greece: I 41  Dieleman-Moyer (2003) and Dieleman (2005), 173–175 and 177–183. For the ceremony, see Otto (1960) and Assmann (2003), 408–431. 42 Vergote (1961). 43 PGM V 96. 44 PGM XIII 934 ss. 45 Author of Orphica, PGM XIII 947. 46 PGM XIII 954. 47 Martín (2006). See 346–358. 48 PGM VII 795–845 (795), together with Democritus.

250

emilio suárez de la torre

point to the tradition concerning the travels to Egypt and contacts with the priests attributed to both philosophers.49 The spell combines astrological elements (zodiac) with other usual indications, which function as a new feature contributing to the fusion with Greek traditions because they include the use of laurel for divination. Democritus50 The name of Democritus was very soon linked to magic and astrology. In PGM VII 167–186 he appears as the author of παίγνια, usually translated as “table tricks” or “table gimmicks.” Kotansky51 suggests that this collection of recipes is designed to be performed or used at a symposiastic dinner and mentions the parallels found in Athenaeus’ Deipnosophistae. As for VII 795–845, the spell bears, as said above, the double attribution ὀνεραιτητὸν Πυθαγόρου and Δημοκρίτου ὀνειρόμαντις μαθηματικός, in this case with a particular use of the term ὀνειρόμαντις. In fact it displays first the mathematical part and then the instructions to receive a revelation by dreams. Attention must be paid to the words: διὸ ἐν στέρνοις ἔχων τὰ ὑπ’ἐμοῦ δηλοθέντα καὶ μετὰ τὴν ἀπόλυσιν τοῦ βίου τοῦ ἐμοῦ ὁσίως πράξας ἐπιτεύξῃ: this and other details point to a master-pupil relationship. As for the transformation of Democritus into a magician,52 the name of Bolus of Mendes is often cited – the main names in this line go from Nietzsche to Wellmann –, but recently Leszl53 has observed that there are not sufficient reasons to maintain this opinion. He argues that Collumella (VII 5, 17) does not say that Bolus has published works under the name of Democritus, but that “they are published under the false name of Democritus.”54 He plays with the possibility of Thrasillus’s influence in this process, and he supports his argument by adducing that he was one of the most cited sources referred to by Pliny the Elder – an author who contributed enormously to Democritus’s fame as a magician. He also analyzes the list of Democritean works cited by Diogenes Laertius, taken from Thrasillus.

49 This detail can be found in Diogenes Laertius, Porphyrius, and Jamblichus. This trait is shared with the biographies of other “wise men” (Solon, Plato, etc.), but they do not appear in the magical tradition. 50 PGM VII 167–186; 795–845; XII 351. Important: Leszl (2007). 51  In Betz (19922), 120. 52 See Abt (1908), 326–327. 53 Leszl (2007). 54 Leszl (2007), 60.



pseudepigraphy and magic

251

In PGM XII 351 (351–364) the work attributed to Democritus is the Sphere (Σφαῖρα), in fact a rectangle that allows some calculations of life and death.55 (c) A mythical hero Dardanus56 According to Diodorus Siculus (5. 48–49) he was the founder of the mystery rites of the Kabeiroi on Samothrace.57 As a magician, he is mentioned by Pliny, NH, 30,9, and since then he will reappear in late lists of Ancient magicians, as for instance H. Cornelius Agrippa’s De occulta philosophia.58 Dardanus’s lineage is also important for the “mythical” construction of Roman legendary ancestors. The recipe with his name is The Sword of Dardanus. The term “sword”59 is sometimes applied to magical formularies. The other best known example is The Sword of Moses. The spell included in PGM is an erotic incantation, and it can be illustrated with some magical gems which have incised the same – or similar – images described here. It is related to the theme of Eros and Psyche, as Reizenstein observed,60 and is therefore again an interesting witness of the cultural environment shared by magical recipes. (d) Some possible “historical” figures Euenos61 This is the only mention of Evenus’s Ἀπομνημονεύματα. It is included in the series of authors cited to illustrate the different forms of the “holy name” (in this case, φθεθων, as pronounced by Egyptians and Syrians). A qualified candidate could be the Elegiac poet (who can be paralleled to Phocylides),62 but this is a mere guess (is it a Hellenization of an Egyptian or Semitic name?).

55 In Betz (19922), 166, Hershbell remembers that a similar ‘sphere’ is attributed to Petosiris, the mythical Egyptian astrologer. It can be a circle or a rectangle. See BouchéLeclercq (1899), r. 1963, 537–542. 56 PGM IV 1716. 57 Hermann (1957). 58 Cap. II. 59 A short sword or ξίφος (fr. ‘glaive’). See Mouterde (1930), 51–56. 60 Reitzenstein (1912), 19–21. 61   PGM XIII 965. 62 Both have a pseudepigraphical tradition of their own.

252

emilio suárez de la torre

Pyrrhus63 The reference is minimal (again in the list of the “Holy Names”). I do not dare to suggest a candidate.64 Himerius65 It is difficult to find an alternative to Preisendanz’s proposal of identification with the homonymous physician and veterinarian of the fourth century.66 The papyrus apparently includes three spells: one is a drawing made with Typhonian ink, and two are spells for good fortune at a workshop; both include the use of an egg. In the first recipe the egg is a part of the magical praxis, whereas in the second one there is no specification about the egg, and several Egyptian gods are named: it is a prayer to Ammon, though the egg is related to Egyptian cosmogonies. Agathokles67 A possible candidate is a homonymous author of pharmacological recipes, who lived in the Augustan period and was cited by Andromachus, according to Galenus (13.82), and by Plinius.68 In this papyrus it goes on a “dream sender” (ὀνειροπομπός) formula. At the end of the spell, to show its efficacy, it is said that “it was also used by Apollobex.” (e) A theios anêr Apollonius of Tyana69 Philostratus’s biography of Apollonius shows that he was a popular figure with a profile exceeding that of a simple philosopher. Philostratus shaped the later development of that profile by narrating that Apollonius has become a reputed magician with effective amulets, spells, and practices.70

63 PGM XIII 970. 64 I have speculated with Pyrrhus, the younger member of the Epirotean dynasty, who was taken in for a while by king Ptolemy, and also with a wrong transcription of the philosopher Pyrrhon’s name, given the usual assimilation of philosophers to magicians (as Empedocles, Democritus and so on). 65 PGM XII 96. The reading of the papyrus is Ἡμέριος. It may be a case of itacism, but both forms are normal in texts and inscriptions. An haruspex Hemerius is mentioned in CIL 13.06394 (Augusta Treverorum) 66 Preisendanz refers to RE 8, 2, 1635. 67 PGM XII 107. 68 NH 22.90. 69 PGM XIa 1. Included as T43 by Jones (ed. 2006). 70 See Bowie (1978), Dzielska (1986).



pseudepigraphy and magic

253

This recipe belongs to this magical profile of the philosopher. It is not a usual recipe. It has some elements of folk-tale and in some aspects it recalls the tales recorded by Apuleius in his Metamorphosis: the skull of an ass71 will cause the appearance of a beautiful woman on an ass, who will be later transformed into an old woman, who will immediately appear as a goddess of extraordinary beauty. The narrative shows then a kind of incoherence because it seems that in fact the old woman has a different identity from that of the goddess, who will finally find the means to warrant the assistance of the old woman permanently. This old woman, according to the title of the spell, should be a servant of Apollonius. We have here again a case of shared cultural milieu, at least in the phase of the first redaction of the recipe. From Thessaly to Egypt Philinna the Thessalian72 This recipe and another attributed to a Syrian woman from Gadara,73 both with a medico-magical content, have frequently been the subject of scholarly attention because of the early chronology of the papyrus (first century bc) and not least for their important particularities relevant to the history of magical collections of spells in Greek language. Dickie74 – who emphasizes the rare use of women as magical authorities in the papyri – has convincingly shown that at least Philinna may be a historical name. His research leads him to assert that “Philinna was, accordingly, one of those notorious women in Hellenistic historiography who had succeeded in capturing a ruler’s affections” (something that he illustrates with several examples). Additionally, the tradition of Thessalian sorcery adds a strong aura of authority to the spell. Moreover, these two spells, both in hexameters and belonging to a type known as historiolae – for obvious reasons –, give witness to the circulation of collections of Greek magical spells in Hellenistic Egypt, are in relation to the medico-magical treatises of the Hellenistic period, and can help to illustrate the circulation of anthologies.75 71   The ass is related to Seth-Typhon. 72 PGM XX 13. 73 PGM XX 4. 74 Dickie (1994). I refer to his complete list of references. 75 Faraone (2000). To this list might be added a Pseudo-ptolemaic treatise entitled Ἓν καὶ τὸ Πᾶν (PGM XIII 980). It is described as belonging to a group of “Ptolemaic” works and is referred to in the group including the “sacred name.”

254

emilio suárez de la torre

(f ) Romans Claudianus76 He is named as the author of a lunar spell. In his apparatus criticus, Preisendanz mentions a homonymous alchemist cited by Berthelot-Ruelle as one of the ancient “faiseurs d’or.”77 Semitic Since the only non-Jewish figure is an anonymous Syrian woman from Gadara to whom only one recipe is assigned,78 this group should be properly called “Jewish.” We find here two of the most famous and recurrent authorities in ancient pseudepigraphical magic, Solomon and Moses who would both have a great influence in their own Jewish tradition, in all the surrounding cultures of antiquity, and in the magical tradition of later times. However, we must take into account that now we are not dealing with the problem of the actual ‘Jewishness’ of the magical elements of these papyri.79 As mentioned above, the magical papyri are basically Egyptian in spirit and details, although they incorporate other elements in such a way that it is impossible to determine a precise precedence. Solomon80 Although his name appears only once in the magical papyri, Solomon, whose wisdom is emphasized in the Bible and became proverbial, was a perfect candidate to be included in the list of the best magicians, beyond Jewish limits and with an important late development of this profile. The work usually attributed to him was the Testament of Solomon,81 in which the sage receives from god a ring with which he can defeat all kinds of demons and “demonesses.” Hence the high frequency of the name of Solomon in magical gems and amulets, with the inscription σφραγὶς θεοῦ,82 where he is represented as a horseman armed with a lance and defeating a demon. 76 PGM VII 862. 77 Berthelot-Ruelle (eds. 1887), vol. I, 26, 1. 78 See above, under Philinna the Thessalian. 79 Cf. Smith (1996), Bohak (2003) and (2008). 80 PGM IV 850–929. 81   See Mc Cown (1922), Johnston (2002), Torijano (2002). See also the introductions and translations of Conybeare (English 1898), McCown (English 1922) and Piñero (Spanish, 1987), among many others. 82 Cf. Perea (2000).



pseudepigraphy and magic

255

Not surprisingly, in PGM IV a spell for divination through medium bears his name, or more exactly, The “Trance” of Solomon, Σολομόνος κατάπτωσις, literally a “collapse,” a medical term used also for epileptic attacks,83 which shows the care of the authors in their choice of vocabulary. Moses84 The agon between the magicians of Pharaoh and those of Yahweh, Aaron and Moses, described in Exodus (7. 11–13), would have an enormous impact in later times, as much in Jewish as in Christian sources, and traces of it can be also found in “pagan”85 authors. The consequences of this impact will be noticed in two important themes in the evolution of Moses’s “biography,” such as was elaborated – a posteriori – by different authors. On the one hand, it contributes to reinforce the image of Moses as a fundamental piece in the “victorious fight” of monotheism vs. polytheism or, more exactly, in the consolidation of the image of true religion vs. false superstition.86 On the other hand – and this is what interests us in this study – it will raise up the figure of Moses to the top-ranking magician of all time. However, outside Jewish or Christian circles the anonymous Egyptian magicians of Exodus will not only be baptised in later times – as Jannes and Jambres – but will also have a first rank position in the world of magic and will be mentioned side by side with Moses. This is what we find, for instance, in Pliny the Elder,87 who is probably following Apion. 83 LSJM: “ἡ συγκοπή ἐστι κ. δυνάμεως” Gal. 10.837; of epileptic seizures, Alex.Aphr.Pr.2.64, cf. Vett.Val.38.13 (pl.). 84 PGM VII 619–27, XIII 1–343 (the Unique or Eighth book of Moses), 343–646 (the Eighth or Holy Book of Moses), 971–973 (the book of Moses entitled Archangelic). I will not deal here with the alchemical writings attributed to Moses nor with other magical implements (amulets, phylacteria) attributed to him, but not recorded in the magical papyri. See Gager (1972), 152–159; Bohak (2003) and (2008), 201–215. 85 I acknowledge that this term charges the weight of the “alterity” upon the rest of authors who were neither Jewish nor Christian, but I use it for the sake of brevity. 86 Assmann (1998). He speaks of Gegenreligion. This aspect of the traditions around Moses will not be treated here. I refer to Smith (1996) for detailed analysis and different points of view. As a matter of fact the discussion of these aspects has its fundament in the information given – with different versions – of Moses’s role by ancient historians, substantially Manetho, from the Egyptian side, and Flavius Josephus from the Jewish one. The latter made a biased manipulation of the data offered by Manetho – for instance, the fusion of two chronologically distant episodes, the invasion of the Iksos and the foundation of a monotheistic religion by Amenophis IV (who gave himself the name of Echnaton) – in order to defend the superioriy of Jewish religion thanks to Moses. But in this process some other authors of diverse origin have contributed to the final result: Hekataeus of Abdera, Bochoris, Artapanus, etc. 87 NH 30.1 est et alia magices factio a mose et ianne et lotape ac iudaeis pendens, sed multis milibus annorum post zoroastren. The name Lotapes is otherwise unattested, and may

256

emilio suárez de la torre

For its relationship with the magical texts of the papyri, it is important to underline that, outside those texts, the profile of Moses as a magician is attested as early as the first century, as Gager pointed out.88 The influence of Pliny may perhaps be detected in Apuleius (see above), though he seems to have had personal access to many texts. Gager observed that, whereas Pliny – and apparently Apuleius – probably depend on Apion, the mention of Moses found in Numenius,89 who calls him Mousaios,90 depends on a different tradition.91 But let’s go now back to the magical papyri. The works attributed to Moses are as follows. In PGM VII (619–627) a short text, containing two recipes, is presented as part of the Διάδημα of Moses.92 The two formulae are recommended, respectively, to become invisible and to seduce a woman (plus other possible applications to get “whatever you wish”).93 The only Jewish hints of the spells are in the series of voces magiace, at the end of which the names of Iao, Sabaoth, and Adonai appear. However, the greatest number of pseudepigraphical works by Moses come from a sole grimmorium, the Leiden magical papyrus J 395 (PGM XIII), an interesting book containing an impressive document probably original from the Alexandrian milieu, but denoting an important number of manipulations and additions. Its most important text is precisely the Holy Book Called Unique or Eighth (Book) of Moses concerning the Holy Name.94 This treatise is a composite one. We detect the presence of two main versions, the first of which includes an alternative version of one of its parts.95 Their content is the so

represent an alternative to the more usual couple Jannes-Jambres or it may be the product of a misreading. See Gager (1972), n. 11. 88 Gager (1972), 140 89 Apud Euseb. PE 9.8.1–2 (411d); cf. Orig., c. Cels. 4.51. 90 See Gager (1972), 139 for this identification. It appears also in Artapanus (in Eusebius, PE 9.27.3 432a). 91  In addition to the Greek texts mentioned here, I must cite other Jewish apocryphal works, such as The Sword of Moses (Garter [1896]). 92 The title of the book may be related to a biblical passage where a diadem is mentioned or may evoke a magical instrument (rings, amulets, etc.), which has been adopted as title. 93 But only in love-affairs, and not for any other aim whatsoever, as sometimes is said (the adequate spell is ‘ποίησον τὴν δεῖνα πρὸς τὸν δεῖνα’!!). 94 It was studied in detail, among others, by Dieterich (1890) and Merkelbach (1992), but from very different points of view. 95 Merkelbach (1991) gives a very good description of the different parts and of the differences of the two main versions. He also prints the text of these versions in parallel in order to make the differences visible.



pseudepigraphy and magic

257

called Leiden Kosmopoiia, a complex text composed by an elaboration of Egyptian cosmogonic and theogonic elements. Inside this text we find an allusion to another pseudepigraphical work, The Key of Moses, and even a reference to an alternative version of the Eighth book, called The Hidden Book of Moses on the Great Name. Later on, included in the same list in which the Orphic and other treatises are mentioned, an Archangelic Book (Ἀρχαγγελική) enlarges the number of Mosaic books. Some lines later is the turn of another “secret” text, perhaps not a complete book, but a simple prayer to the moon, called Σεληνιακή. Quite surprisingly, at the end of this part we find a new title: The Tenth Hidden Book. Finally, a possible allusion to Moses (but not to a written work) is the formula included in a prayer of the other Leiden papyrus (J384), with the motif of the identification of the practitioner with a divinity (in this case not a God, but a prestigious religious authority): “I am he whom you met at the foot of the holy mount and to whom you gave the knowledge of your most great [name].”96 To confirm that identification is the mention of Moses in PDM XIV 130–131 (col V 14): “reveal yourself to me / here today in the manner of the form of revealing yourself to Moses which you made upon the mountain, before which you have created darkness and light” (a formula included in a spell for a “god’s arrival”).97 Jacob98 This Berlin papyrus (P.gr. 13895) has a long prayer attributed to Jacob asking for important benefits such as wisdom or strength. It denotes a strong Jewish influence. As Preisendanz (ad loc.) observes, it seems designed for amulets. Persian The magical spells included in PGM did not escape the widespread celebrity of the Mages hellénisés.99 In the magical spells Persian wisdom is represented by two of them, Zoroaster100 and Ostanes,101 but there is no trace

96 ἐγώ εἰμι, ὧ συνήντησας ὑπὸ το ἱερὸν ὄρος καὶ ἐδωρήσω // τὴν τοῦ μεγίστου ὀνόματός σου γνῶσιν κτλ. Trans. of Hubert Martin Jr., in Betz (19922), 156. 97 Translation of Janet Johnson in Betz (19922), 202, following Griffith-Thompson (1904), 34. 98 PGM XXIIb 1. 99 Bidez-Cumont (1938). 100 PGM XIII 967. 101   PGM IV 2006, XII 123.

258

emilio suárez de la torre

of Hystaspes. However, we find a love-charm attributed to an Ἀστραψοῦχος, a variant of the more usual name Astrampsouchos or Astrampsychos,102 the purported “author” of the collection of lot oracles known as sortes Astrampsychi103 (second–third century ad). Zoroaster appears in the list of “authority” names added to the second version included in the above cited Leiden Papyrus 395. He is called Ζωροάστρης ὁ Πέρσης, but the name of the assigned book is lost. Ostanes appears twice: first as the addressee of the letter sent by Pitys with an ἀγωγή,104 and then as the dream sender of Zminis, the one who recommends writing the spell with myrrh ink. As for the so called φιλτροκατάδεσμος of Astrapsouchos, it is much more than a simple love spell: it is in fact a long prayer to Hermes asking for “favor, nourishment, victory, prosperity, sex appeal, beauty of face, strength of all men and women” (5–6), as well as “business to the workshop” (62). The papyrus105 is to be dated to the fourth or fifth centuries ad, perfectly in accordance with the late spread of Astrampsychos’ fame. 3. Conclusions Given the disparate origins of these magical texts in terms of chronology, culture, and types of recipes, the resort to pseudepigraphy is far from being homogeneous. However, it is possible to summarize the main characteristics of this procedure. First of all, the pseudepigraphical practice reflected in these papyri agrees with the more general process of cultural “cross-fertilization” detectable in them, which is a result of a kind of cultural – and, in this case, magical – koiné. The spells of the Greek magical papyri share in the same trends of the contemporary religious, philosophical, and technical literature of the surrounding cultures, mainly that of Greece. Since the authors and compilers of these recipes choose the Greek language as a prestigious vehicle of transmission, it is not surprising that are familiar with contemporary tendencies in the literary genres cited above. They also incorporate other elements from the Jewish, Persian, and Mesopotamic traditions, creating a specific blend of the pseudepigraphical

102 PGM VIII 1–63. 103 See now Naether (2010). 104 See above, on Pitys. 105 Gr. P. CXXII Brit. Mus. London.



pseudepigraphy and magic

259

heritage of those cultures. Moreover, the relevant pseudepigraphical tradition in Egypt’s own religious literature must also be taken into account. The form of the pseudepigraphical texts oscillates from the simple recipe with a false attribution in its incipit to the great grimmorium with a large text, either monographic or with more than one spell, including the use of a well-known genre of didactic transmission: the letter. This is a well-attested genre in Egyptian history, used for very different purposes, and matches the same trend attested in Greek literature. A relevant characteristic is that there is a mixture of “consecrated” magicians and other less known names. It may be reasonable to speak of an alternation of the “universal” and the “local.” The distribution of these two different groups does not follow a fixed model, but it is in accordance with the different types of texts which compose the magical papyri as a whole. Some of them cite only one name as an authority, but for different reasons in each case: the old anonymous assistant of Apollonius in PGM XIa, Philinna in PGM XX, and Jacob in XXIIb. But these are short papyri with a very concrete content. Pnouthis is the only authority of PGM I, Manetho of PGM III,106 Ieu of PGM V, and Astrampsychos of PGM VIII. The rest of the papyri display the mixture cited above: Nephotes, Panchrates, Pitys, Pibechis, Dardanus, Solomon, and Ostanes appear in PGM IV; Pitys again, Democritus, Claudianus, and Moses in PGM VII; Zminis, Ourbikos, Democritus, Himerius, Agathocles, and Zoroaster in PGM XII; finally, the “orphic trio,” Evenus, Pyrrhus, and, again, Moses are in PGM XIII. Thus, the only names repeated in more than one papyrus are Moses, Pitys, Democritus, and Ostanes. The demotic PGM XIV has the name of Moses, Paysakh, and other anonymous magicians. This coexistence of “consecrated” authors and others cited only once – and even anonymous authors – is in accordance with the different ways of conveying magic and with two different models of practice: a popular one and a restricted one. Furthermore, we must differentiate between the magical operations leading to immediate and practical results and the mixture of magic with different mystic trends. For all these reasons, the magical papyri deserve to be included in a study of pseudepigraphy in the ancient world.

106 In other aspects PGM I-III share some peculiar traits: for instance, they are more ‘Apollonian’ than others.

260

emilio suárez de la torre References

Abt, A. (1908), Die Apologie des Apuleius von Madaura und die antike Zauberei. Beiträge zur Erläuterung der Schrift De magia, Giessen. Assmann, J. (1998), Moses der Ägypter. Entziferung einer Gedächtnisspur, (trans. from the original English version: Moses the Egyptian: The Memory of Egypt in Western Monotheism, Cambridge, Mass., 1987), München, Wien. —— (2003), Tod und Jenseits im alten Ägypten, München. Bakir, A. el-M. (1970), Egyptian Epistolography from the Eighteenth to the Twenty-First Dynasty, Le Caire, Institut Français d’Archéologie Orientale. Berthelot, M. and Ch. E. Ruelle (1887) (eds.), Catalogue des anciens alchimistes Grecs, 3 vols., Paris. Betz, H.D. (1990), “The Formation of Authoritative Tradition in the Greek Magical Papyri,” in H.D. Betz (ed.), Hellenismus und Urchristentum (Gesammelte Aufsätze I), Tübingen: 173–183. —— (19922) (ed.), The Greek Magical Papyri in Translation: Including the Demotic Spells (vol. I: Texts), Chicago/London. —— (1997), “Jewish Magic in the Greek Magical Papyri (PGM VII.260–271), in P. Schäfer and H. G. Kippenberg (eds.), Envisioning Magic: A Princeton Seminar and Symposium, Leiden: 45–63. Bidez, J. and F. Cumont (1938), Les mages hellénisés. Zoroastre, Ostanès et Hystaspe d’après la tradition grecque (vols. I Introduction, vol. II Les Textes), Paris, Les Belles Lettres, (reimpr. New York, 1975; Paris, 2007). Bohak, G. (2003), “Hebrew, Hebrew Everywhere? Notes on the Interpretation of voces magicae,” in S. Noegel, J. Walker, B. Wheeler (eds.), Prayer, Magic, and the Stars in the Ancient and Late Antique World, The Pensilvania State University. —— (2008), Ancient Jewish Magic: A History, Cambridge. Bouché-Leclercq, A. (1899), L’astrologie grecque, Paris. Bowie, E.L. (1978), “Apollonius of Tyana: Tradition and Reality,” ANRW II, 16, 2, Berlin: 1652–1699. Conybeare, D.C. (1898), “The Testament of Solomon,” JQR 11: 1–45. Cosentino, A. (2002), “La tradizione del re Salomone come mago ed esorcista,” in A. Mastrocinque (ed.), Gemme gnostiche e cultura ellenistica (Atti del incontro di studio, Verona, 22–23 ott. 1999), Bologna. Dickie, M. (1994), “The Identity of Philinna in the Philinna Papyrus (PGM2 XX 15, SM 200.15),” ZPE 100: 119–122. Dieterich, A. (1890), Abraxas. Studien zur Religionsgeschichte des Spätens Altertums, Leipzig, Teubner. Dieleman, J. (2005), Priests, Tongues, and Rites: The London-Leiden Magical Manuscripts and Translation in Egyptian Ritual (100–300 CE), Leiden/Boston. Dieleman, J. and I. Moyer (2003), “Miniaturization and the Opening of the Mouth in a Greek Magical Text (PGM XII.270–350),” JANER 3: 47–72. Dillery, J. (1999), “The First Egyptian Narrative History: Manetho and Greek Historiography,” ZPE 127: 93–116. Dzielska, M. (1986), Apollonius of Tyana in Legend and History, Roma. Faraone, Ch. A.D. Obbink, D. (1991) (eds.), Magika Hiera: Ancient Greek Magic and Religion, New York / Oxford. Faraone, Ch. A. (2000), “Handbooks and Anthologies: The Collection of Greek and Egyptian Incantations in Late Hellenistic Egypt,” AfR 2: 195–214. Gager, J.G. (1972), Moses in Greco-Roman Paganism, SBL Monograph Series 16, Nashville Gardiner, A.H., H. Thompson, and J.G. Milne (1913), Theban Ostraca, London/Oxford. Garter, M. (1896) (ed.), The Sword of Moses, London.



pseudepigraphy and magic

261

Gordon, R. (1997), “Reporting the Marvelous: Private Divination in the Greek Magical Papyri,” in P. Schäfer and H.G. Kippenberg (eds.), Envisioning Magic: A Princeton Seminar and Symposium, Leiden: 65–92. Gozzoli, R. B. (2006), The Writing of History in Ancient Egypt during the First Millenium BC (ca. 1070–180 BC): Trends and Perspectives, Egyptology 5, London. Griffith, F.L. and H. Thompson (1904–1909), The Demotic Magical Papyrus of London and Leiden, 3 vols., London. Hermann, A. (1957), “Dardanus,” RAC 3: 593–594. Hopfner, Th. (1921), Griechisch-Ägyptischer Offenbarungszauber, I Leipzig, reimpr. Amsterdam, A.M. Hakkert, 1974; II1 Leipzig 1924, reimpr. Amsterdam, 1983; II2 Leipzig 1924, reimpr. Amsterdam, 1990. Johnston, S.I. (2002), “The Testament of Solomon, from Late Antiquity to the Renaissance,” in J.N. Bremmer, J.N. Venstra (eds.), The Metamorphosis of Magic: From Late Antiquity to the Early Modern Period, Leuven: 35–50. Jones, Ch. P. (2006) (ed.), Philostratus, Apollonius of Tyana: Letters of Apollonius. Ancient Testimonia. Eusebius Reply to Hierocles, LCL, Harvard. Knox, W.L. (1938), “Jewish Liturgical Exorcism,” HTR 31: 191–203. Kotansky, R. (1991), “Incantations and Prayers for Salvation on Inscribed Amulets,” in Faraone-Obbink: 107–137. Leszl, W. (2007), “Democritus’ Works: From their Titles to their Contents,” in A. Brancacci and P.M. Morel (eds.), Democritus: Science, The Arts of the Care of the Soul, Proceedings of the International Colloquium on Democritus (Paris, 18–20 September, 2003), Leiden/ Boston: 11–76. Lidonnici, L. (2007), “ ‘According to the Jews’: Identified (and Identifying) ‘Jewish’ Elements in the Greek Magical Papyri,” in Lynn Lidonnici and Andrea Lieber (eds.), Heavenly Tablets: Interpretation, Identity, and Tradition in Ancient Judaism, Leiden: 87–108. López Jimeno, A. (1990), “Las cartas de maldición,” Minerva 4: 121–144. Martín, R. (2006), El orfismo y la magia, diss. (unedited), Universidad Complutense, Madrid. McCown, Ch. Ch. (1922), The Testament of Solomon, Leipzig. Mastrocinque, A. (2005), From Jewish Magic to Gnosticism, Tübingen, Mohr Siebeck. Merkelbach, R. (1992) (ed.), Abrasax. Ausgewählte Papyri religiösen und magischen Inhalts, Teil 3: Zwei griechisch-ägyptische Weihezeremonien (Die Leidener Weltschöpfung. Die Pschai-Aion-Liturgie), Opladen, (Papyrologica Coloniensia XVII, 3). Mouterde, R. (1930), Le glaive de Dardanus: Objects et inscriptions magiques de Syrie, Beirut. Moyer, I.S. (2003a), “The Initiation of the Magician,” in D.B. Dodd and Ch. A Faraone (eds.), Initiation in Ancient Greek Rituals and Narratives, London: 219–238. —— (2003b), “Thessalos of Tralles and Cultural Exchange,” in S. Noegel, J. Walker, and B. Wheeler (eds.), Prayer, Magic, and the Stars in the Ancient and Late Antique World, Pennsylvania State University: 39–56. Naether, F. (2010), Die Sortes Astrampsychi: Problemlosungsstrategien durch Orakel im römischen Ägypten (Orientalische Religionen in der Antike 3), Tübingen. Noegel, S.B. (1996), “Moses and Magic: Notes on the Book of Exodus,” JANES 24: 45–59. Ogden, D. (2007), In Search of the Sorcerer‘s Apprentice, Swansea. Otto, E. (1960), Das ägyptische Mundöffnungsritual, Wiesbaden. Perdrizet, P. (1903), Σφραγὶς Σολομόνος, REG 16: 42–61. Piñero, A. (1987), Testamento de Salomón, en A. Díez Macho (ed.), Apócrifos del Antiguo Testamento, vol. V, Madrid: 325–390. Preisendanz, K. and A. Henrichs, A. (19732 and 19742) (eds.), Papyri Graecae Magicae, Die griechischen Zauberpapyri, I Suttgartt, 19732, II Stuttgart, 19742. Reizenstein, R. (1912), Das Märchen von Amor und Psyche bei Apuleius, Leipzig/Berlin. Smith, M. (1984), “The Eighth Book of Moses and How It Grew (PLeid. J 395),” Atti XVII Congresso internazionale di Papirologia (Napoli 1984), Napoli: 683–93.

262

emilio suárez de la torre

—— (1986a), “The Jewish Elements in the Magical Papyri,” in id., Studies in the Cult of Yahweh, Leiden, 1996, vol. 2, 242–256 (improved version of the same article in SBL Seminar Papers 25: 455–462). —— (1986), “P Leid J 395 (PGM XIII) and its Creation Legend,” in A. Caquot, M. HadasLebel, J. Riaud (eds.), Hellenica et Judaica. Hommage à Valentin Nikiprowetzky, LeuvenParis: 491–498. Thissen, H.-J. (1991), “Ägyptologische Beiträge zu den griechischen magischen Papyri,” in U. Verhoeven and E. Grafe (eds.), Religion und Philosophie in alten Ägypten. Fs. Derchain (OLA 39), Leuven: 293–302, 295. Torijano, J.P. (2002), Solomon the Esoteric King: From King to Magus, Development of a Tradition, Leiden-Boston-Köln. Vergote, J. (1961), “Sur les mots composés en Égyptien et en Copte,” BIOR 18: 208–214.

The Sophists’ Place in the Greek Wisdom Tradition Håkan Tell This paper is based on a book that I recently completed and is thus part of a larger project. Plato’s Counterfeit Sophists is a study of the intellectual activities spearheaded by the older sophists in fifth- and fourth-century bc Greece.1 The main argument is that the sophists as we know them are a Platonic design; that there never existed in antiquity a group of people consistently labeled sophists. Plato successfully managed to create an intellectual genealogy for the sophists that cut them off from their precursors and contemporaries and disrupted meaningful intellectual continuities with the earlier Greek wisdom tradition. In contrast to this Platonic characterization, I suggest that the sophists should be understood as individual thinkers and as integral and in many ways crucial to the development of increasingly specialized practices within the field of sophia in the fifth and fourth centuries bc. We seem so trapped in Platonic categories that it becomes almost impossible even to discuss this group of practitioners of wisdom without simultaneously reinforcing their unique status as championed by Plato. But since this is an association that I am trying to unravel, what is the appropriate terminology to adopt to distinguish between the Platonically designed sophists, on the one hand, and the historical individuals, on the other? Given that the label sophist is so ingrained, it seems impractical and unnecessarily counterintuitive to avoid it. I will thus use the term sophist to refer to the group of people that Plato designated as such. Who are they?2 The list is short: Protagoras,3 Gorgias,4 Prodicus,5 Hippias,6 Euthydemus and Dionysodorus,7 and Miccus.8 Commonly included in this 1   Plato’s Counterfeit Sophists. 2011. Cambridge, MA: Center for Hellenic Studies, Harvard University Press. 2 This is not an attempt to treat exhaustively all the various contexts in which a figure occurs in conjunction with the labels “sophistry,” “sophistic,” etc. The present focus is simply to explore whom Plato explicitly labels a sophist. 3 Cra. 391c; Prt. 311e, 313c, 317b, 317c, 349a, and 357e. 4 Meno 95c; Hp. mai. 282b. 5 Prt. 357e; Symp. 177b; Lach. 197d; Erx. 399c; Euthd. 277e. 6 Prt. 357e; Hp. maai. 282e. 7 Euthd. 271c. 8 Ly. 204a.

264

håkan tell

group, although never so labeled by Plato, are Antiphon, Thrasymachus, Xeniades, Lycophron, Critias, and Callicles.9 More advisable, however, and to the degree possible, I suggest that we avoid group labels altogether and refer to each individual by his own name. To highlight the fluidity of categories in the Greek wisdom tradition and to emphasize the lack of intellectual specialization, I will occasionally use the term sophos as an unmarked designation to refer to individuals with a claim to sophia. This term is common in Greek and embraces a wide variety of groups that we would hesitate to juxtapose, such as sages, religious experts, poets, and philosophers. Here I will focus on one part of the argument in my book, the different criteria that Plato used to characterize the sophists as qualitatively distinctive – their philosophical otherness, if you will – and argue that he is in fact fashioning the category “sophist” by projecting a set of undesirable characteristics onto a relatively heterogeneous group of sophoi. By reviewing the primary material, I hope to illustrate this Platonic intellectual construct and offer a historically reasonable alternate interpretation. At the end of the paper, I will hint at some of my broader conclusions and point to areas of intellectual continuity in the activities of the sophists. But since the sophists’ assumed otherness is such a stumbling block in trying to articulate such areas of sophistic intellectual continuity, I will focus here on reassessing their unique status. The sophists have had an almost exclusively hostile reception in antiquity as well as in modern times. Though there are no easy answers to account for this treatment, it seems reasonable to suggest that at least part of it originates with Plato’s powerful articulation of sophistry as the opposite of philosophy – philosophy’s “daemonic double,” to quote Michael MacDonald.10 This classification has later been expanded on and upheld by the Platonic tradition all the way up to contemporary scholarship. By treating Plato, Xenophon, and Aristotle as a group in that tradition, I do not intend to imply that there are not fundamental differences in their intellectual outlooks, only that they all fundamentally agree in their attitudes regarding the sophists and in the importance of establishing their status as intellectual outsiders. By Platonic tradition I also mean 9 Euenus is not labeled a sophist in the Apology. He is referred to as a σοφός (20a3), though he is featured in a discussion in which the word “sophist” occurs. 10 “The refutation of sophistry constitutes one of the founding acts of philosophy. Philosophy, it seems, creates itself by purging the sophists as its other, its daemonic double, even its ‘counteressence’ (Gegenwesen),” M. MacDonald 2006: 39.



the sophists’ place in the greek wisdom tradition

265

later commentators and doxographers who rely on – without necessarily being sympathetic to – Plato and Aristotle, such as Themistius and Olympiodorus. Their portrayal as outsiders is a large contributing factor to the persistence in the negative attitudes towards the sophists – and this is also the reason why it is so important to expose this assumption to renewed critical attention. There have of course been attempts to rehabilitate the sophists philosophically and to give voice to their muted contributions. But even in these attempts at rehabilitation, they are understood as different from and outside of the mainstream philosophical tradition. Havelock, one of the sophists’ most vocal supporters, even goes so far as to speculate that modern political thought would have been radically different – indeed that the destructive military rivalry between Germany and France could have been avoided – if the liberal temper of the sophists had prevailed over the totalitarian philosophical systems of Plato and Aristotle.11 I am not interested in rehabilitating the sophists philosophically or in arguing for a reappraisal of their misunderstood or suppressed intellectual achievements. Instead, I am concerned with the formation in ancient Greece of a set of practices that have come down to us as philosophical. In conjunction with that, I am also concerned with understanding and documenting the cultural authority attached to the position of the sophos – the practitioner of wisdom – in antiquity. It is widely accepted today that philosophy as a specialized discipline was not developed before Plato, but that he was instrumental in creating and defining this new field of intellectual practices.12 Indeed, his articulation

11 “Had their [sc. the sophists’] doctrine been allowed to prevail and influence the mind of Europe at a crucial stage in its development, who is to say what happier and sunnier societies would not have in time arisen on the plains of Gaul and Germany?” E. Havelock 1957: 308. 12 A. Nightingale 1995: 10: “Indeed, as a careful analysis of the terminology will attest, φιλοσοφεῖν does not take on a specialized and technical meaning until Plato appropriates the term for his own enterprise. When Plato set forth a specific and quite narrow definition of this term, I will suggest, he created a new and specialized discipline.” But see A. Laks 2006, esp. pp. 55–82, for a critique of this position. According to Laks, the development of philosophy as an autonomous field of activities had already emerged before Plato, as evidenced by the discussions in Gorgias Helen (13), Hippocrates VM (20), and Plato Euthydemus 305c. If we take into consideration the doxographical nature of Hippias’ Synagoge, it seems very likely that Laks is right in his critique of the view of Plato as philosophy’s inventor. What matters to my argument, however, is not that Plato was the first to establish philosophy as an autonomous field, only that his articulation was among the first ones and that it deliberately portrayed philosophy as an old discipline without acknowledging either its newfangled or contested status.

266

håkan tell

of philosophy was so powerful that it remained largely uncontested into modern times. But the establishment of the discipline of philosophy in antiquity was anything but a straightforward and uncontested enterprise. My underlying assumption – and here I draw heavily on the cultural sociology of Pierre Bourdieu – is that the terms “philosophy” and “sophist” were contested in antiquity and that it is problematic for moderns to naturalize them by adopting them as neutral classifications – thereby removing them from their original, contentious context. The use of the word “sophist,” for example, is intertwined with the development of increasingly specialized practices within the field of sophia and the corollary struggles over the appropriation of the terms “philosophy” and “philosopher.” If the term “sophist” occurred in a struggle over orthodoxy and heterodoxy – where it was employed (mainly by Plato) to denigrate certain sophoi as less orthodox and thus as less legitimate – then the uncritical adoption of Platonic terminology effectively runs the risk of taking sides in an ideologically driven battle over legitimacy. Plato, Xenophon, and Aristotle ultimately won the definitional struggle, and the pejorative designation of certain sophoi as sophists has become the “historical” truth. But this truth still does not support our embrace of the term as inevitable or natural. Our task is not directly to perpetuate the Platonic victory of classification, but rather to examine the conditions under which the term “sophist” came to be attached to certain individuals and take on derogatory connotations. If I am right in postulating that the sophists are a Platonic invention, how was Plato able to pull this off without anyone noticing it or calling his bluff? Since almost all of the early philosophical texts have been lost, we have to rely on the doxographical tradition for our information. By doxography I mean the quasi-scholarly activity in antiquity that Jaap Mansfeld has defined as “the systematical collection and sometimes analysis of views held by others, for the purposes of polemics, research, edification, or delectation.”13 When it comes to the sophists, however, there is close to no doxographical tradition. Again, the reason for this – as George Kerferd speculates – is probably Plato and Aristotle’s rejection of them as counterfeit thinkers, since this “meant they were virtually excluded from the series of surveys commissioned in Aristotle’s school, which were a major source of subsequent information.”14 In other words, to reconstruct the intellectual scope and interests of the sophists, we have to rely almost

13 J. Mansfeld 1990: 24. 14 G.B. Kerferd 1981: 36.



the sophists’ place in the greek wisdom tradition

267

exclusively on the accounts of Plato and Aristotle – accounts that are both hostile and polemical. This reconstruction is especially precarious since the hostile stance on the part of the main witnesses occasionally undermines their credibility. But by juxtaposing Plato and Aristotle to accounts that are either earlier – especially Old Comedy – or relatively independent of their influence – Isocrates, for example – I hope to cast doubt on the Platonic genealogy of the sophists. This, in turn, will be the first step in recovering a different intellectual genealogy, one where the sophists are understood as members integral to the developing intellectual field in fifth- and fourth-centuries bc Greece. Sophistic Otherness To see some ways in which the sophists have been portrayed as qualitatively different by the Platonic tradition, I will examine seven different criteria of otherness. In some cases, there exists a large scholarly body of literature critical of the Platonic characterization of the sophists. In these instances, I will for the most part limit myself to summarizing the major findings of this literature. In other instances, however, the Platonic verdict has largely gone unchallenged, and I shall have more to say about those points. My first two criteria of sophistic distinctiveness fall into this category. 1) To begin with, they were labeled “sophists.” This classificatory designation clearly marks them off as a distinct group of intellectuals, separated from their “philosophical” precursors and contemporaries.

The standard account of the word “sophist,” σοφιστής, states that in early Greek society σοφιστής was indistinguishable from σοφός and that only later, with the emergence of a novel group of practitioners of wisdom (Gorgias, Protagoras et al.), it took on a new, specialized meaning. This “limited” meaning was from then on consistently applied to these figures, although the old, “indeterminate” sense was occasionally still in use; and it was the association with the new group of thinkers that gave it its pejorative ring, similar to the modern connotations of “sophistry.”15 According to this explanatory model, there is a precise shift from a general to a specific meaning, a process that seems to mirror the concomitant process of intellectual specialization in Greek society. 15 The terms “limited” and “indeterminate” are A. Grant’s, 1885, vol. 1: 110.

268

håkan tell

It is my contention that this is an artificial demarcation, and that the time limit imposed is arbitrary. Instead, the use of σοφιστής continued well into the second half of the fifth century and beyond, and it was frequently used with respect to what we would call “philosophers,” that is, to those practitioners of wisdom traditionally referred to as Presocratics, Socratics, Platonists, or Aristotelians. We have numerous examples of Presocratics being called sophists. Diogenes of Apollonia (a contemporary of Herodotus) labels the (Ionian) natural philosophers (φυσιόλογοι) sophists.16 Xenophon (Mem. 1.1.11) calls those who engage in natural investigations and discussions about the cosmos by the same name, and elsewhere (Mem. 4.2.1) he mentions that Euthydemus had collected many works of the renowned poets and σοφισταί, presumably referring to the Presocratics.17 Isocrates attests that Pericles studied with two “sophists,” Anaxagoras and Damon. Later, he warns students not to get too caught up in the subtleties of the early “sophists,” and he mentions Empedocles, Ion, Alcmaeon, Parmenides, Melissus, and Gorgias as examples of this group.18 Diodorus Siculus (12.39.2) also refers to Anaxagoras as Pericles’ teacher and a σοφιστής, as does Athenaeus (220B). Finally, Plato (Meno 85b), in an apparent reference to the mathematicians, labels them σοφισταί. Socrates and the Socratics were also repeatedly called sophists: in Aristophanes’ Clouds (360–1), the chorus professes that Prodicus and Socrates are the foremost of the astrological sophists (μετεωροσοφισταί). But it is not only Aristophanes who refers to Socrates as a sophist; Androtion and the orator Aeschines (1.173) also call him by that name.19 A number of Socrates’ disciples (the so-called Socratics) were categorized as sophists: Lysias thus designates Aeschines (not to be confused with the orator), Aristotle uses the term in reference to Aristippus, and Xenophon in reference to Antisthenes.20 Isocrates is called a sophist by Plutarch (Quast. conv. 1.1); and to judge from Isocrates’ defensive attitude against his detractors in the Antidosis, it seems reasonable to assume that Plutarch had many predecessors in that practice.21 Isocrates, in turn, delivers an attack against what seems to 16 Diogenes of Apollonia in Simplicius Phys. 151.20 = DK 64A4. 17 See G.B. Kerferd 1950: 8, who also understands this in reference to the Presocratics. 18 Isocrates Antid. 235 and 268. 19 Jacoby, FGrH 324 F 69 = Aristides XLVI: 407 Dindorf = III, 677 Behr. 20 Aristides XLVI: 407 Dindorf = III, 677 Behr; cf. Athenaeus 611D–612F (Lysias); Aristotle Metaph. 3.2 (Aristippus); Xenophon Symp. 4.4 (Antisthenes). 21  Isocrates Antid. 166, 196, 213, and 231.



the sophists’ place in the greek wisdom tradition

269

be Plato’s Laws and Republic in Philippus (12), where he dismisses them as sophistic works. Lysias also calls Plato a sophist.22 Timon is less discriminating in his use of the term: he labels all philosophers – Plato and Aristotle included – sophists.23 Finally, Timaeus in his abuse of Aristotle refers to him as a pedantic sophist (σοφιστὴς ὀψιμαθής).24 As is clear from this brief survey, the word σοφιστής was in use in reference to a wide category of sophoi both during and after Plato’s lifetime. Its broad continuous application is an indication that Plato’s restricted use of the term was neither in general use nor universally accepted. It thus does not seem that the word went through a substantial change or crystallization in its definition after the emergence of the sophists. Indeed, in light of these other examples, we might question Plato’s generally privileged position as a witness for the meaning of σοφιστής.25 Isocrates is a perfect example of how unsettled and contentious the use of “sophist” was in antiquity. When explicitly defining who he means by that label, he names Solon, Empedocles, Ion, Alcmaeon, Parmenides, Melissus, Zeno, Protagoras and Gorgias. As we have seen, he appears to refer to Plato’s Republic and Laws as sophistical works.26 Except for Gorgias and Protagoras, there is no compatibility among the individuals that occur under the label sophist in Isocrates and in Plato. When defining philosophy, in turn, Isocrates singles out the study of how to speak well (λέγειν εὖ) and of how to persuade one’s hearers (πείθειν . . . τοὺς ἀκούοντας).27 Isocrates thus offers a complete reinterpretation of Platonic philosophy, where the focus is shifted from Plato’s apolitical sage – who is now labeled a sophist! – to the politically active and rhetorically astute figure of his own standing. In the words of H. Sidgwick: 22 Fr. 281 Baiter-Sauppe = Aristides XLVI: 407 Dindorf = III, 677 Behr. 23 Timon in D. L. 9. 65 and 112. Cf. Grote 1872: 33. 24 Timaeus in Polybius 12. 8. 4 = FGrH 566 F156. 25 Although the word σοφιστής was still in wide application during and after Plato’s lifetime and did not crystallize in its meaning at any given point in time, it nevertheless appears to have acquired a more pejorative tinge in its later use. But this is far from absolute: Xenophon, for example, in Cyropaedia (3.1.14 and 38) relates the tragic fate of a sophist (σοφιστής) who was unjustly put to death by the Armenian king on charges of corrupting (διαφθείρειν) his son Tigranes. When he was about to die, the sophist sent after Tigranes and asked him not to feel any anger towards his father, since he was acting out of ignorance (ἄγνοια) and not malice (κακόνοια), and, since he was acting out of ignorance, he was acting against his own will. This story is surely meant to allude to the fate of Socrates, and it would be particularly odd if Xenophon chose to use a strictly pejorative epithet in such a context. 26 Antid. 268–71, 235, 313; Helen 2–3. 27 Antid. 275.

270

håkan tell The testimony of Isocrates then comes to this: he attacks the Sophists in the same style as Plato: only Isocrates calls Sophists just those whom Plato and posterity call Philosophers, while the more honourable title of ‘Philosophy’ he reserves for his own special industry, the Art of Public Speaking.28

Plato’s consistent application of the term sophist to a set group of individuals is at great variance with how the word was otherwise used in antiquity and is thus of questionable quality as evidence for the distinctiveness of the sophists. 2) The sophists were said to trade in wisdom. Kerferd succinctly articulates modern and ancient criticism: What is wrong is that the sophists sell wisdom to all comers without discrimination – by charging fees they have deprived themselves of the right to pick and choose among their pupils.29

As Harrison has noted, Plato is almost incapable of referring to the sophists without simultaneously emphasizing their pecuniary ambitions.30 Indeed, establishing this association is absolutely integral to Plato’s characterization of the sophists and their practices. Why? Teaching for pay was not a neutral labeling but a highly inflammatory charge to which all sophoi, to one degree or another, were susceptible. We need to allow for a split between historical reality – about which we often know very little – and the way that that reality was expressed. A fee can be described as a gift or a bribe, but the practice of charging money can also go without comment. The language surrounding monetary transactions in antiquity is notoriously difficult to assess. Of particular sensitivity is the language surrounding sophia and the commodification of wisdom. Indeed, a favorite way to undermine the authority of a sophos (or public figures in general) was to suggest that they had monetary motivations and were driven by greed. I suggest that we need to read the accusations (διαβολή) – 28 H. Sidgwick 1872: 293. J.S. Morrison (1958: 218) expresses similar sentiments: “It is fascinating to observe how Isocrates has exactly reversed Plato’s terminology, so that while the former’s practical, purely professional training becomes philosophy, those who are concerned with physis are left with the name of sophist. ‘Philosopher’ and ‘sophist’ are labels now attached respectively to those teachers who do and those who do not subscribe to the speaker’s view of the proper subjects for higher education.” 29 G.B. Kerferd 1981: 25. 30 “Indeed, it is no exaggeration to say that he is almost incapable of using the term sophist without at the same time making some explicit reference to this professionalism. And it comes as no surprise when this professionalism looms larger than any other element in each of the definitions of the sophist which appear in the dialogue of that name.” E.L. Harrison 1964: 191 and fn. 44.



the sophists’ place in the greek wisdom tradition

271

which is really what they are – against the sophists of exacting fees in light of these considerations. To get a sense of just how redolent of disparagement and bias is the close association of the sophists and their sophia with money and fees, let us review Xenophon’s remarks in the Memorabilia (1.6.13), where he comments on the cultural implications associated with offering one’s personal qualities for sale: Among us (παρ’ ἡμῖν) it is considered that there is a good and a shameful way to dispose of one’s beauty and wisdom. If a man sells his beauty to any one who wants it, he is called a prostitute (πόρνον), but if he befriends someone he knows to be a noble and good lover, he is thought of as prudent. And in the same way we call those who sell their wisdom to anyone who wants it sophists, just as if they were prostitutes (ὥσπερ πόρνους).

Here Xenophon establishes a thematic sequence consisting of wisdom, money, and prostitution, in which the interference of the intermediary phase – money – runs the danger of corrupting and even conflating the things of the mind with the sphere of the body. This is, of course, the exact opposite trajectory of what we are wont to see in Plato.31 In the Symposium and Alcibiades I, for example, ἔρως (sexual desire) is limited to using the physical as an initial stepping stone to climb the philosophical ladder and ultimately to reject the body in favor of the mind, thus gradually transforming itself from a physical, sexual desire directed at a specific individual to a generic non-physical love of the beautiful. By reversing this trajectory and by introducing the concept of intellectual promiscuity, Xenophon invites us to appreciate the contentiousness of his portrayal of the sophists; and in this antagonistic depiction, the mention of teaching for money is crucial in allowing the association of sophistic sophia with the body and, ultimately, with prostitution. Read in this way, then, teaching for pay takes on a more sinister facet than has previously been recognized and paves the way for the successive ubiquitous complaints of the speciousness of sophistic wisdom. As a whole, the portrayal of the sophists by the Platonic tradition exhibits a thematic emphasis on money over wisdom, on body over mind – in stark opposition to this tradition’s own valorization of the intellect. Most importantly, though, is that the definition of sophist becomes based on

31 This observation is indebted to L. Kurke’s discussion (unpublished manuscript) of the philosophical trajectory from the body to the soul in Alcibiades I.

272

håkan tell

a formal characteristic – teaching for pay – rather than on intellectual content. Almost all later testimonia to the effect that the sophists taught for pay are derived from the Platonic tradition. Of particular importance, however, is that the same association is not maintained in Old Comedy or in Isocrates – traditions relatively independent of Plato. In fact, in Old Comedy it is Socrates who is labeled sophist and accused of teaching for pay. Again, why did Plato settle on this formal characteristic as a distinguishing feature? To understand this better, we have to examine other ancient charges against sophoi of venality. G.E.R. Lloyd has called attention to how the author of the medical text On the Sacred Disease charges his rivals with fraud and greed.32 In this treatise, the accusations of charlatanism motivated by greed seem to be part of a stock repertoire of invectives employed against opponents in an attempt to undermine their claims to sophia. Lloyd argues that we need to separate between the “empirical content” – where there are striking similarities in the proposed treatments between the author of On the Sacred Disease and the practitioners he accuses of fraud – and the “argumentative weaponry” – where intellectual differences are emphatically stressed.33 In On the Sacred Disease, then, accusations of venality operate mainly on the rhetorical level and are aimed at deflating the opponent’s authority. This is a theme familiar from comedy and tragedy, where attacks against sophoi for greed are frequent. In tragedy, for example, the Theban seer Teiresias is an illustrative expression of this development. Frequently referred to as σοφός, he is welcomed as the savior of the state by Oedipus (O.T. l. 314), but when he delivers undesirable information – that Oedipus is the slayer of his own father Laius – he is quickly accused of having eyes only for profit and of being an impostor (ἐν τοῖς κέρδεσιν / μόνον δέδορκε, τὴν τέχνην δ’ ἔφυ τυφλός, lines 388–9). Similar patterns are visible in Antigone (lines 1035–6 and 1055: τὸ μαντικὸν γὰρ πᾶν φιλάργυρον γένος) and Bacchae (lines 255–7). The charge against Teiresias is not informed by a historical reality, but by the rhetorical strategy of seeking to invalidate his divinatory authority. And the regularity with which the invective of greed 32 Fraud: 2.1–10 and greed: 4.17 (βίου δεόμενοι). G.E.R. Lloyd 1979: 16–7. For practitioners of medicine as sophoi and members of the Greek wisdom tradition, see, e.g., G.E.R. Lloyd 1979 and R. Thomas 2000. There is a strong link between wisdom and healing in figures such as Epimenides, Empedocles, and Alcmaeon. For examples of doctors as sophoi in Plato, see, for instance, Ly. 210a, Euthd. 280a, Thg. 123d–e, and Epin. 976a. All references to the Sacred Disease are from the edition of W.H.S. Jones 1923. 33 The formulations are those of Lloyd 1979: 125.



the sophists’ place in the greek wisdom tradition

273

is directed at seers in comedy and tragedy suggests that this was a topos familiar to the audience. I propose that we use the findings from On the Sacred Disease and tragedy as an analogy to the charges against the sophists of teaching for pay in the Platonic tradition. There are important thematic overlaps in the way the accusations of venality are treated in these authors and genres: they often occur as part of a more general invective discourse; their force is mainly destructive, aimed at undercutting the authoritative position of the adversary; finally, they are not motivated by an ambition to establish actual historical differences in social practices but are rather an expression of what Lloyd calls “argumentative weaponry.” With this analogy in mind, the Platonic characterization of the sophists as greedy peddlers of specious sophia takes on a new light. We are invited to rethink the deeprooted identification of them with the formal characteristics of teaching for pay. 3) The sophists were predominantly focusing on rhetoric, not philosophy.

Both Edward Schiappa and Thomas Cole have argued persuasively that the close association of rhetoric with the sophists is another Platonic and Aristotelian definitional imposition attempting to control and circumscribe the intellectual practices of the sophists so as to subordinate them to their own philosophical enterprises. Schiappa argues that “the word rhêtorikê may have been coined by Plato in the process of composing the Gorgias around 385 bce.”34 As Andrea Nightingale has pointed out, it was in the Gorgias that Plato first offered an exhaustive definition of philosophy and applied it to his own activities. She goes on to note that, If Schiappa is right about rhêtorikê, then the first explicit and systematic definition of the art of rhetoric would go hand in hand with the first attempt to define philosophy. In attempting to define his own territory, as is seems, Plato had to define and delimit his rival’s terrain.35

There are thus good reasons to question the assertion that the sophists exhibited a predominantly rhetorical interest and expertise, especially 34 E. Schiappa 1990: 457. But see G. Pendrick’s 1998 criticism of Schiappa’s argument: “the significance that Schiappa attributes to Plato’s supposed invention of the name and notion of rhetoric appears illusory, despite the formidable array of modern theorizing he summons in support of his contentions. The term ῥητορική itself certainly antedated its appearance in the Gorgias, and there is no reason to think that Plato either invented or redefined it in the way, and with the motives, Schiappa suggests”: 22. 35 A. Nightingale 1995: 72.

274

håkan tell

given the negative assessment of rhetoric in comparison with philosophy in both Plato and Aristotle. It seems more likely that it is another attempt to differentiate and separate them as un-philosophic, that is, as illegitimate. As for the rhetorical features that we do find in the writings of the sophists, Mark Griffith has argued that, far from being novel in nature, they exhibit strong traditional characteristics: Such writers as Gorgias, Protagoras, and Euripides, though they certainly struck their contemporaries as doing something strange and (to many) shocking, were not for the most part introducing radically new techniques or attitudes, but rather exploiting, systematizing, and exaggerating possibilities that they found already well developed by their poetic predecessors.36

But it was not only in the poetic predecessors that models for their rhetorical techniques were found. Already in antiquity, Isocrates (Helen 2–3) compared the rhetorical writings of Gorgias and Protagoras to those of Zeno and Melissus – traditionally included among the Presocratics – and Aristotle, in his lost work, the Sophist, wrote that Empedocles invented rhetoric and Zeno of Elea dialectic (DK 29A10 = D. L. 8. 57). There is thus no doubt that public oratory was a theme of immense interest and rapid development in the fifth and fourth centuries bc. But it was not exclusive to the sophists nor can it be used as the defining characteristic of their intellectual activities. 4) The sophists teach what appears to be F, but actually is not, thus actively deceiving their pupils.

This is a position upheld by both Plato and Aristotle – and many modern scholars, too. In the words of Aristotle: The art of the Sophist is the semblance of wisdom without the reality, and the sophist is one who makes money from an apparent but unreal wisdom.37

In the Metaphysics, Aristotle adds that, “dialectic is tentative regarding the things where philosophy claims knowledge, and sophistic is what appears to be philosophy but is not.”38 The seventh definition of sophist in the Platonic dialogue of the same name unequivocally underscores the Platonic position:

36 M. Griffith 1990: 187. 37 SE 165a21; cf. 183b36ff, and EN 1164a30. 38 1004b25–26: ἔστι δὲ ἡ διαλεκτικὴ πειραστικὴ περὶ ὧν ἡ φιλοσοφία γνωριστική, ἡ δὲ σοφιστικὴ φαινομένη, οὖσα δ’ οὔ.



the sophists’ place in the greek wisdom tradition

275

VISITOR: Imitation of the contrary-speech-producing, insincere and unknowing sort, of the appearance-making kind of copy-making, the word-juggling part of production that’s marked off as human and not divine. Anyone who says that the sophist is of this “blood and family” will be saying, it seems, the complete truth. THEAETETUS: Absolutely (trans. N.P. White).39

Xenophon takes a less convoluted stance when he asserts that, “the sophists talk to deceive, write for their own profit, and are of use to nobody.”40 The main part of my book consists of an exploration of the traditional elements in the sophists’ intellectual practices and highlights how deeply embedded they were in the Greek wisdom tradition. This strong continuity – historical as well as intellectual – serves as a counter-argument to their alleged tendencies towards charlatanism and philosophical counterfeiting. It also appears odd – not to say anachronistic – that they should be accused of falling short of the Platonic and Aristotelian definitions of philosophy, definitions that were articulated long after the heyday of the sophists themselves. 5) The sophists subscribed to extreme relativism and denied the existence of objective judgments

The basis for this view is mainly to be found in Plato’s treatment of Protagoras’ homo mensura doctrine in the Theaetetus. Modern scholars, treating Protagoras as the intellectual spearhead of the sophistic movement, seem to have generalized and applied similar views to the sophists as a group.41 But extreme relativism, strictly speaking, can only be attributed to Protagoras, and even in his case with great uncertainty. It is of course possible to identify and ascribe relativistic ideas to the sophists – for example, that what qualifies as a virtuous action depends on the context and circumstances – but such reasoning is in no way exclusive to the sophists; if this is the criterion for relativism, we would have to include both Aristotle and Socrates in this group.42 Richard Bett has compellingly argued that if by relativism is meant the rejection of objectivity, then it is misguided to attribute this position to any of the sophists, with the possible

39 Sph. 268c8–d5. 40 Cyn. 13.8: οἱ σοφισταὶ δ’ ἐπὶ τῷ ἐξαπατᾶν λέγουσι καὶ γράφουσιν ἐπὶ τῷ ἑαυτῶν κέρδει, καὶ οὐδένα οὐδὲν ὠφελοῦσιν. 41  Cf. R. Bett 1989: 139, who also provides a list of modern scholars who assume that the sophists subscribed to relativism, p. 139, esp. fn. 1. 42 For examples, see R. Bett 1989: 149.

276

håkan tell

exception of Protagoras. Relativism is thus not an intellectual position of any value in establishing the sophists as qualitatively different. As a way effectively to question their intellectual credibility, however, it has served an important function. 6) The sophists led an itinerant lifestyle and traveled all over Greece in search of employment. As opposed to their precursors and contemporaries, their travels were not motivated by intellectual curiosity, only by the prospects of attracting students and of increasing profits.

In Plato’s Hippias Major (282d–e), for example, we hear how Hippias visited Sicily and made good money while Protagoras was simultaneously touring the island and offering his teachings. In the Apology (19e) Socrates draws a clear distinction between himself and Gorgias, Prodicus, and Hippias, who, according to him, could travel effortlessly into any city they wanted and persuade the young men to associate with them for pay. Socrates expresses his misgivings in the Timaeus (19e) about the intellectual reliability of the sophists, since they are a traveling group (γένος . . . πλανητόν) who do not have a permanent residence. Silvia Montiglio has picked up on this Platonic portrayal of the itinerant (and mercenary) disposition of the sophists. She argues that they – in sharp contrast to the positive paradigm of travel personified by Solon and Democritus – had merely professional and monetary motives for their travels: But unlike Solon or Democritus, the Sophists did not travel in order to acquire knowledge. Hardly any ‘theory’, any abstract curiosity or ethnographic interest motivated their travels. Rather, the Sophists traveled to sell their skill . . . Travelling was part of their professional activity.43

Again, this distinction between the theoretical travels of the philosophers and the professional travels of the sophists is difficult to establish. Travel and wisdom were intrinsically linked in ancient Greece from at least the archaic period onward, and we have ample ancient evidence of this connection in our sources. Most of the so-called Presocratics traveled far and wide. Democritus, for example, attests to his many travels: Among the men of my day, I traveled over most land researching the greatest things and I saw most climates and lands and I listened to most learned men.44

43 S. Montiglio 2000: 92. Cf. S. Montiglio 2005. 44 DK 68B299.



the sophists’ place in the greek wisdom tradition

277

What is more, there are significant overlaps and regularities in terms of where the Presocratics and the sophists went and in what activities they partook. Many of them were involved in colonizing expeditions. The most notorious example of this is the Panhellenic (though mainly Athenian) establishment of Thurii in 444/3 in southern Italy. In connection with this expedition we hear of the active involvement of Herodotus and Lysias as colonists, Protagoras as lawgiver, and the Milesian Hippodamus as town planner – all of whom are perfect examples of practitioners of wisdom – for which the Greeks used the term sophoi. We also hear that Empedocles traveled to and spent time in Thurii, presumably in his capacity as a sophos. Significant numbers of sophoi also gravitated towards the Panhellenic sanctuaries. These centers served as meeting places: this was where they attracted listeners and students, where they read their works, and where they, in turn, could familiarize themselves with the work of others. From this it becomes clear that travel was a fundamental aspect of the archaic Greek institutions of wisdom and of crucial importance to the circulation and dissemination of wisdom. There are no clearly identifiable differences between the travels of the sophists and those of other practitioners of wisdom. To be sure, Socrates is the odd man out with his insistence on remaining fixed in Athens and with his concomitant rejection of (intellectual) travel. In addition to these predominantly Platonic and Aristotelian characterizations of the sophists’ otherness, modern scholars have often employed another criterion: 7) The sophistic movement developed in response to specific Athenian social forces, and the sophists spent most of their time in Athens.

De Romilly articulates this modern position well: But the fact remains that it is in Athens that we find them all . . . Were it not for Athens, we should probably not even know the name ‘Sophists’. And even if we did, it would have no meaning or interest. Without doubt, the vogue for the Sophists only came about thanks to a catalyst which Periclean Athens alone could provide.45

Against this Athenocentric view of the sophistic movement, Robert Wallace has recently argued that Athens – and its democratic form of government – has been given too much importance in understanding the sophists: 45 J. de Romilly 1992: 18.

278

håkan tell Virtually all the various characteristics of their teaching and careers are attested earlier than 450, and not in narrow association with Athens. These characteristics developed not simply as a result of or always in conjunction with democracy – though they certainly flourished under open democratic systems – but also for internal philosophical reasons. Even during the second half of the fifth century, most of those whom we regularly call sophists spent most of their time outside Athens.46

Rosalind Thomas, in her work on Herodotus, is also highly critical of the way scholars tend to over-emphasize the significance of Athens, both as the sole impetus for the intellectual movement in the second half of the fifth century (at the expense of, say, East Greece), but also as the undisputed center of all intellectual activity in general. There is an abundance of evidence of the sophists’ activities outside of Athens. In fact, to judge from Plato’s characterization in the dialogue Protagoras, for example, many years had elapsed since Protagoras’ last visit – years during which he was presumably active elsewhere. There is thus a very weak case to be made that the sophists emerged in response to Athens’ need for a higher education and that this polis was the sole impetus for their activities – or their intellectual distinctiveness. * * * We may speculate about Plato’s reasons for branding certain sophoi as sophists and forcefully rejecting their intellectual positions. To begin with, most of the major sophists were Socrates’ contemporaries and intellectual rivals. As is well known, Socrates himself was accused of being a sophist in Aristophanes’ Clouds, and in the Apology of Socrates Plato has Socrates state that this and similar characterizations of him had contributed to bringing the trial against him. We might assume, then, that Plato would want to purge from Socrates’ legacy the charge of being a sophist, that is, of being an intellectual charlatan who taught the young Athenian elite to disrespect the laws and morals of the land. Indeed, many of the Platonic dialogues are named after and treat extensively the prominent sophists, and throughout, Socrates is carefully distinguished from the sophists and portrayed almost as their intellectual opposite. Another reason might have to do with the competing notions of philosophy promoted by some of the sophists well before Plato. We have substantial evidence, for example, that Hippias wrote a comprehensive history of philosophy labeled Synagoge – a work that predated both Plato 46 R. Wallace 1998: 205.



the sophists’ place in the greek wisdom tradition

279

and Aristotle’s articulations of philosophy.47 In it, Hippias stated his ambition of creating a new and varied work by compiling statements from Orpheus and Musaeus, Hesiod and Homer, poets and prose writers, and Greeks and non-Greeks. In Hippias’ view, philosophy was not unique to the Greeks, nor was it sufficiently distinct to require a separate treatment from the earliest poetic expressions. Such views are of course anathema to both Plato and Aristotle. Aristotle, in particular, traces the beginning of philosophy back to Thales and strongly rejects the notion of an earlier beginning or more inclusive approach to what might qualify as philosophy.48 Scholars have showed that both Plato and Aristotle drew on Hippias’ work in their own writings – without acknowledging their source, obviously.49 Both Protagoras and Gorgias are also known to have written compilations and commentaries on their precursors.50 We might speculate that the Platonic rejection of the sophists is also a rejection of the different articulations of philosophy that some of them championed. On this point, perhaps Schiappa is right when he argues that the founding act of philosophy requires a negative double. It is well known that Plato takes great liberties when fashioning a historic lineage for his own intellectual practices. Similar strategies might be at work when dealing with his intellectual rivals. An indispensable part of the Platonic articulation of philosophy was accomplished by repressing or misrepresenting rival claims. This is a phenomenon that Schiappa, following earlier scholars, has labeled “dissociation,” that is, the rhetorical splitting of a coherent concept into a positive and negative part. 47 We owe a fragment from Hippias’ Synagoge to Clement of Alexandria, who in the Stromateis (VI.15 = DK 86B6) quotes what is supposedly the proem: τούτων ἴσως εἴρηται τὰ μὲν Ὀρφεῖ, τὰ δὲ Μουσαίῳ κατὰ βραχὺ ἄλλῳ ἀλλαχοῦ, τὰ δὲ Ἡσιόδῳ τὰ δὲ Ὁμήρῳ, τὰ δὲ τοῖς ἄλλοις τῶν ποιητῶν, τὰ δὲ ἐν συγγραφαῖς τὰ μὲν Ἕλλησι τὰ δὲ βαρβάροις· ἐγὼ δὲ ἐκ πάντων τούτων τὰ μέγιστα καὶ ὁμόφυλα συνθεὶς τοῦτον καινὸν καὶ πολυειδῆ τὸν λόγον ποιήσομαι. Some of these things have probably been said by Orpheus, others by Musaeus briefly in different places, yet others by Hesiod and Homer, others by the other poets, others in the prose writings of Greeks and non-Greeks alike. But I will make this account new and varied by putting together the most important and related sayings from all of them. 48 See M. Frede’s 2004 exploration of Aristotle’s reasons for settling on Thales as the first philosopher and how it relates to Aristotle’s conception of philosophy. 49 See, for example, B. Snell 1944; C.J. Classen 1965; A. Patzer 1986; J. Mansfeld 1990. 50 “The assumption that the rudimentary beginnings of the historiography of Greek philosophy may be dated to the period of the Sophists is a very safe one,” J. Mansfeld 1990: 27. “Their [sc. Plato and Aristotle’s] famous discussions of the problems of being and becoming, of unity and plurality, and of genesis and change or motion, are ultimately rooted in the preliminary doxographies of the Sophists,” J. Mansfeld 1990: 69.

280

håkan tell In this instance Plato was attempting to dissociate the general and traditional meaning of sophistês as a wise person or teacher into two concepts, one of which (the Sophist as a possessor of counterfeit knowledge) would be negatively valued, the other (the philosopher as the seeker of true wisdom) would be positively valued.51

By reassessing the sophists we are allowed a glimpse of an alternate development and course of philosophy. Many of the binaries that some take for granted in the development of philosophy are challenged, such as the opposition between rhetoric and philosophy or political involvement and activism versus the apolitical and disinterested stance of the true philosopher. Already the Greeks themselves were struggling with the nature and origin of thinking. Hippias can perhaps be said to offer a more inclusive articulation of philosophy by considering non-Greeks, barbaroi, as well as representatives of the poetic tradition. The uniformity and coherence that emerges from the Platonic tradition is counterfeit. In origin, philosophy was more heterogeneous and tentative. References Bett, R. (1989), “The Sophists and Relativism,” Phronesis 34: 139–69. Classen, C.J. (1965), “Bemerkungen zu zwei griechischen ‘Philosophiehistorikern’,” Philologus 109: 175–81. Cole, T. (1991), The Origins of Rhetoric in Ancient Greece, Baltimore. Frede, M. (2004), “Aristotle’s Account of the Origins of Philosophy,” Rhizai. 1: 9–44. Grant, A. (1885), The Ethics of Aristotle, fourth edition, 2 vols, London. Griffith, M. (1990), “Contest and Contradiction in Early Greek Poetry,” in M. Griffith and D.J. Mastronarde (eds.), Cabinet of the Muses: Essays on Classical and Comparative Literature in Honor of Thomas G. Rosenmeyer, Atlanta: 185–207. Grote, G. (1872), A History of Greece, fourth edition, vol. 7, London. Harrison, E.L. (1964), “Was Gorgias a Sophist?” Phoenix 18: 183–192. Havelock, E.A. (1957), The Liberal Temper in Greek Politics, New Haven. Jones, W.H.S. (1923), Hippocrates, vol. 2, Cambridge, Mass. Kerferd, G.B. (1950), “The First Greek Sophists,” CR 64:8–10. —— (1981), The Sophistic Movement, Cambridge. Laks, A. (2006), Introduction à la philosophie présocratique, Paris. Lloyd, G.E.R. (1979), Magic, Reason and Experience, Cambridge. MacDonald, M. (2006), “Encomium of Hegel,” Philosophy and Rhetoric 39: 22–44. Mansfeld, J. (1990), Studies in the Historiography of Greek Philosophy, Assen. Montiglio, S. (2000), “Wandering Philosophers in Classical Greece,” JHS 120: 86–105. —— (2005), Wandering in Ancient Greek Culture, Chicago / London. Morrison, J.S. (1958), “The Origins of Plato’s Philosopher-Statesman,” CQ 8: 198–218. Nightingale, A. (1995), Genres in Dialogue: Plato and the Construct of Philosophy, Cambridge/New York.

51 E. Schiappa 1991: 6.



the sophists’ place in the greek wisdom tradition

281

Patzer, A. (1986), Der Sophist Hippias als Philosophiehistoriker, Freiburg. Pendrick, G. (1998), “Plato and ῥητορική,” RhM 141: 10–23. Romilly, J. de (1992), The Great Sophists in Periclean Athens, Oxford. Schiappa, E. (1990), “Did Plato Coin Rhêtorikê?” AJP 111: 460–73. —— (1991), Protagoras and logos: A Study in Greek Philosophy and Rhetoric, Columbia. Sidgwick, H. (1872), “The Sophists,” JP 4: 288–307. Snell, B. (1944), “Die Nachrichten über die Lehren des Thales und die Anfänge der griechischen Philosophie- und Literaturgeschichte,” Philologus 96: 170–82. Thomas, R. (2000), Herodotus in Context: Ethnography, Science and the Art of Persuasion, Cambridge/New York. Wallace, R.W. (1998), “The Sophists in Athens,” in D. Boedeker and K.A. Raaflaub (eds.), Democracy, Empire, and the Arts in Fifth-Century Athens, Cambridge, Mass: 203–22.

Forging Ancient Greek Words in Modern Times Onofrio Vox The ancient Greek language really isn’t dead: its plurimillenary existence continues not only in medieval and modern Greek languages, but also outside the Hellenophone geographical area, in the various modern European languages – French, English, German, Spanish, Italian – often acting within them or in combination with the Latin language or through its filter.1 The vitality of ancient Greek within modern languages occurs in “onomaturgy,” i.e., the making of names,2 especially, as well known, in science, or rather in the sciences, both natural and human: with an obvious or hidden influence, either through direct or indirect loans (semantic calques) or through coinages, whether of words or of whole semantic families, made on the basis of old linguistic material. If we consider the phenomenon from the ancient language’s point of view, the loans should be evaluated as “true graecisms,” and vice versa the coinages as “false graecisms.” Where words that are old but have been recycled in modernity with an entirely or partly new meaning (semantic coinages), these should be considered “false graecisms.” Two striking examples are the semantic families of anthropology3 and of biology.4 Here 1 This was recalled with precision and breadth of documentation by Rodríguez Adrados (2005): 270 ff. For the influence of Greek on modern European languages, it is also important to take into account the collection by Munske–Kirkness (1996), and in particular for the Italian and German languages respectively the studies by Janni (1987) and Eisenberg (2011). A list of graecisms in modern languages can be found on the internet at http://en.wikipedia.org/wiki/Category:Greek_loanwords (consulted 01.27.2012). Cf. Hogben (1970), which serves as handbook for the creation of modern scientific terminology, providing a kind of grammar as well as a bilingual dictionary, English and Greek-Latin; Kytzler–Redemund–Eberl (20073), a voluminous dictionary of Greek words used in modern German, usually in composition, followed by a list of modern terms, annotated and analysed for components. 2 See Migliorini (1975): 2 ff. 3 In ancient times we only meet the nomen agentis ἀνθρωπολόγος, “speaking of man, i.e., fond of personal conversation, Arist. EN 1125a5” (LSJ: 141), and the verb ἀνθρωπολογεῖν, “describe or represent in the form of man, Ph. 1.282: Pass., ib. 181” (LSJ: ibid.). For modern history of the semantic family see Janni (1987): 13–21. 4 The abstract noun is non-existent in ancient times, while the verb βιολογεῖσθαι, “Pass., to be sketched from life, esp. common life, τὰ βιολογούμενα Longin. 9.15” (LSJ: 316), the widespread nomen agentis βιολόγος, “kind of mimic actor or mime” (LSJ, Rev. Suppl.: 69), and the adjective βιολογικός, only with reference to κωμῳδίαι, “= μῖμοι, Suid. s.v. Φιλιστίων” (LSJ:

284

onofrio vox

I point out a handful of modern coinages, of various importance, but all instructive in terms of genesis and diffusion mechanisms. 1. The term epistemology was coined in 1854 by the Scottish philosopher James Frederick Ferrier (1808–64)5 in the Institutes of Metaphysic: The Theory of Knowing and Being, p. 46: “the doctrine or theory of knowing . . . (λόγος τῆς ἐπιστήμης – the science of true knowing)”; and he also introduced, on p. 51, agnoiology, “the theory of ignorance (λόγος τῆς ἀγνοίας, the theory of true ignorance).” Ferrier, presenting epistemology as a polar opposite term to ontology,6 knew exactly what he meant, as a specialist who oversaw a well-loved course Lectures on Greek Philosophy, published posthumously along with Other Philosophical Remains (Edinburgh and London, 1866, 18882). For both coinages, epistemology and agnoiology, he noted, as we have seen, the ancient Greek component words which he assumed to be the basis for the term he composed. But in fact the second component word’s group is never attested in ancient times, whereas the first (λόγος τῆς ἐπιστήμης), however rare, can be found in imperial age prose, especially in the later commentators on Aristotle and Plato, with whom Ferrier had particular familiarity.7 It comes as no surprise, then, that the new term epistemology achieved immediate success, unlike agnoiology: both because he presented its meaning clearly and also because he gave an expected name to an insistent cultural phenomenon of the nineteenth-century scientific world, the reflexive study of mechanisms of knowledge associated with the search for new knowledge itself.

316) are attested. Thus, so far does it diverge from usage in ancient Greek, the modern semantic family of biology (a term that appears in 1686 with the meaning, however rare, of biography, and since 1766 as a branch of science) in effect constitutes a series of “false friends.” Unless indicated otherwise, I draw information concerning modern words from the OED. 5 See the short note by Fanfani (1976): 90. 6 This is really another “false graecism” because it is only attested since 1613, even if written in Greek characters in Rodolphus Goclenius, Lexicon philosophicum, clave quo tanquam philosophiae aperiuntur fores, Francofurti 1613: 16, in the entry abstractio, annotated “philosophia de ente.” The term spread first in the “international” Latin form, e.g., in J. Clauberg, Metaphysica, 1646 (see Migliorini [1975]: 77), then in the national languages: English ontology 1663, French ontologie 1692, German Ontologie about 1764, in Italian 1739. 7 E.g., Procl. in Plat. Alcib. i 154, 6, and in Plat. rem publicam I, 22.15, and I, 264.1; Simplic. in Arist. categ. VIII 86.20 = Anon. in Arist. cat. 12.17 cf. earlier Phil. Quod deterius potiori insidiari soleat 25.



forging ancient greek words in modern times

285

2. The term prosopagnosy indicates today the inability of a person to properly recognize other people’s faces. The coinage was presented with less erudite credentials than epistemology, and only in 1947, by the German neurologist Joachim Bodamer.8 When explaining the term he used just these dry words: “Griechisch prosopon = das Gesicht”; the cited Greek term is economically glossed. In his “onomaturgic” intervention, Bodamer took and improved the description of this neurological disorder from one of the most famous nineteenth-century researchers: Sigmund Freud. It is to Freud that we owe the introduction of the term agnosy to define this disorder, which up to that time had been given various identifications;9 Freud introduced the coinage in this way: “Störungen im Erkennen von Gegenständen, welche Finkelnburg als Asymbolie zusammenfasst, möchte ich vorschlagen ‘Agnosie’ zu nennen.”10 Nevertheless, the graecism is not completely successful, in that no compound of ἀγνωσία is attested from ancient times, the noun being unknown in medical language. The same objection can be made concerning the other modern compounds formed with agnosy (ancient ἀγνωσία). To list those I know: anosognosy, formed by a-noso(s)-gnōsia, indicates an inability to perceive that one is ill, and it was coined by a former student of Jean-Martin Charcot, the French-Polish Joseph Babinski.11 Phonagnosy, formed by phōn(ē)agnōsy, indicates an inability to recognise sounds and voices.12 For an inability to recognise colours we have acromatognosy or cromatognosy (< chromat(a)-agnōsia), but the most used, with its various results, seems to be the German coinage Achromatopsie (about 1821, about 1842 in French, in English 1844), formed by a-chromat(a)-opsia. 3. Ethnology appears for the first time in the Latin form, ethnologia, in 1767, used by Johann Friedrich Schöpperlin, opposed to geographia; ethnography instead in the German Ethnographie, by August Ludwig Schlözer in 1771, followed by Johann Christoph Gatterer in 1778 (1775).13 The two terms mark an essential novelty in the historical-cultural field, but for their Greek lexicon they draw on Byzantine scholarship. Actually attested  8 Bodamer (1947).     9 E.g., with asymbolia by Ferdinand Carl Finkelnburg (1870), imperception by Hulings Jackson (1876), Seelingblindheit by Herbert Munk (1881). 10 Freud (1891): 80. 11  Babinski (1914). 12 Van Lancker–Canter (1982). 13 See Vermeulen (2006): 127–8. But the first proof of the term ethnology is dated 1842 in OED, s. v. (with reference to J.C. Prichard, Nat. Hist. Man 132).

286

onofrio vox

are at least the two nomina agentis ἐθν(ικ)ολόγος and ἐθν(ικ)ογράφος, interpretable respectively as “Erklärer der Völker(namen)” and “Beschreiber der Völkernamen,” always referred, for near antonomasia, to the grammarian Stephanus of Byzantium, an ethnic and geographical onomastics specialist.14 4. The physicist and mathematician André-Marie Ampère (1775–1836) is an eminent example of the untiring onomaturgic activity carried out by scientists in the nineteenth century, mainly using the lexical tools of ancient languages. In this he can be compared with his contemporary William Whewell (1794–1866), responsible not only for the professional qualifications scientist and physicist, but also for the suggestion to Michael Faraday (1791–1867) of the terms anode, cathode, electrode, and ion for the process of electrolysis, “from noncommittal Greek roots.”15 Ampère carried out an intense nomenclatorial activity particularly in the powerfully rigorous exposition of the classification of sciences in the two volumes that make up the Essai sur la Philosophie des Sciences ou exposition analytique d’une classification naturelle de toutes les connaissances humaines:16 a classification based first on the distinction between “monde cosmologique” and “monde noologique,” and performed with philological

14 See LBG: 445. 15 Heilbron (2002), who says about the nomenclatorial activity of Whewell: “His use of ancient languages and his conviction that the right words would establish correct and durable concepts fit his age perfectly. Most of the people whom he wanted to call scientists knew some Latin and Greek; and they and he built as if their constructions would be in use for ever.” Bailey (2008): 235 draws attention on the meditated coinage of terms for electrolysis as it results from the correspondence between Faraday and Whewell. 16 Paris, Bachelier, tome 1, 1834, tome 2, 1843 (the second volume was published posthumously). A previous more concise attempt at classification was published in Révue encyclopédique 1832: 223–29: it included a table of only 32 “sciences du premier ordre,” followed by explanatory Latin hexameters, entitled Carmina mnemonica (and also directed, after the Argumentum, in verse, Optimo Filio et carissimo). Even outside these systematic classifications, however, Ampère loved to forge new terms “in the Greek style” boldly, as the published correspondence makes clear. Thus, for example, in the letter to his son JeanJacques dated July 16th 1830, he proposed litolbologie for “la science des richesses et de la félicité publique,” revealing: “J’ai trouvé heureusement dans un dictionnaire grec, que όλβος voulait dire à la fois richesse et félicité. Quant à l’épithète publique, il y a trois mots λειτος, δημίος, δημησιος . . . Comme λειτουργέω veut dire remplir une fonction, [fonction] se nomme en grec λειτουργημα ou λειτοργια d’où vient notre mot liturgie, j’ai pensé que les Grecs auraient dit λειτολβος pour félicité publique et j’ai, sauf ton avis que je te prie de me donner là-dessus préféré, quoiqu’avec doute, le mot litolbologie.” (I reproduce the text, including the rough writing of the Greek terms, from the printed edition of the letter: De Launay (1936): 718).



forging ancient greek words in modern times

287

care for the modern rendering of the old terms.17 I here pick out three of his most successful lexical novelties. Cinématique is the term for the science dealing with movement, coined, as Ampère specifies, from “κίνημα, mouvement” (Essay I: 50 ff., on p. 52). This innovation, soon followed in other European languages,18 represents a “false graecism” because in ancient times the adjective connected with κίνημα is actually κινητικός,19 from which will later emerge the English kinetic, the Italian cinetico.20 Cybernétique, indicating the science of state and society government (Essay II: 140), is a term taken from the “ mot κυβερνητική qui, pris d’abord, dans une acception restreinte, pour l’art de gouverner un vaisseau, reçut de l’usage, chez les Grecs même, la signification, tout autrement étendue, de l’art de gouverner en général.” Ampère, as we see, was well aware that he was taking a particular value of the ancient κυβερνητική (cf. τέχνη), the metaphor in which the action of piloting was about men and not a ship, even going as far as synonymy with politics.21 The Greek loanword was certainly unsuccessful in the meaning “correctly” proposed by Ampère, unlike the much more recent proposal to refer the term by extension to control and communication science, whether in the world of machines or of living creatures.22 Zootechnie (english zootechny) is the consciously chosen name for the science connected “à l’éducation des animaux et aux avantages qu’ils nous procurent” (Essay I: 120).23 The coinage does not completely follow the ancient language, because the compounds with -τεχνία (to indicate activity versus the simple τέχνη) provide an adjective or an adverb qualifying the way to practise the art (as εὐτεχνία, κακο-, καλλι- etc.), and the only compound with a noun, to indicate the object, is the Platonic χειροτεχνία 17 See the wide explanation of the coinage methods in the Essai I: XL ff., also with discussion of the spelling problems on p. XLV ff. 18 See OED, s. v. “kinematics”; Migliorini (1975), 30. 19 Expressed with “of or for putting in motion” in LSJ: 953.  20   The occurrences in English from 1855 and 1864, in Italian since 1905. 21  So in Plat. Clit. 408b . . . τὴν τῶν ἀνθρώπων κυβερνητικήν, ἣν δὴ σὺ πολιτικήν, ὦ Σώκρατες, ἐπονομάζεις πολλάκις, τὴν αὐτὴν δὴ ταύτην δικαστικήν τε καὶ δικαιοσύνην ὡς ἔστιν λέγων. 22 Wiener (1948), 19: “We have decided to call the entire field of control and communication theory, whether in the machine or in the animal, by the name Cybernetics.” And, since 1959, another value emerges, by extension of the previous: “the field of study concerned with the integration of living organisms and electronic or other technological devices; robotics” (OED, s. v. “Cybernetics”). 23 For awareness of this choice see Essai I: XXXIV. The coinage is marked by Migliorini (1975): 108. The first attestation of zootechny in OED, s. v., is instead ascribed to 1879 (tr. A. de Quatrefages de Bréau, Human Species 61).

288

onofrio vox

“handicraft, βαναυσία καὶ χ. Pl. R. 590c: pl., γεωργιῶν ἀπέχεσθαι . . . καὶ χ. ib. 547d; αἱ περὶ χειροτεχνίας ἐπιστῆμαι Id. Plt. 304b” (LSJ: 1986). Ampère was led to coin this word by the need for a word distinct from zoology,24 another, partial, “false graecism,” familiar in the international scientific lexicon because of previous use, starting from Latin works of the seventeenth century.25 References Babinski, Joseph (1914), “Contribution à l’étude des troubles mentaux dans l’hémiplégie cérébrale,” Revue Neurologique 27: 845–848. Bailey, Richard W. (2008), “British English Since 1830,” in H. Momma and M. Matto (eds.), A Companion to the History of the English Language, London: 235–242. Bodamer, Joachim (1947), “Die Prosop-Agnosie,” Archiv f. Psychiatrie und Nervenkrankheiten 179: 6–53. De Launay, Louis (1936), Correspondance du Grand Ampère, tome II, Paris. Eisenberg, Peter (2011), Das Fremdwort im Deutschen, Berlin-New York. Fanfani, Massimo (1976), “Ancora qualche ‘parola d’autore’ nel vocabolario filosofico,” Lingua Nostra 36: 90. Freud, Sigmund (1891), Zur Auffassung der Aphasien. Eine kritische Studie, Leipzig und Wien. Heilbron, John L. (2002), “Coming to Terms,” Nature 415: 585. Hogben, Lancelot (1970), The Vocabulary of Science, New York. Janni, Pietro (1987), Il nostro greco quotidiano: i grecismi dei mass-media, Roma-Bari. Kytzler, Bernhard, Lutz Redemund and Nikolaus Eberl (20073), Unser tägliches Griechisch: Lexikon des griechischen Spracherbes, Mainz. LBG = Lexikon zur byzantinischen Grazität, 1, Bd. A-K, erstellt von E. Trapp, Wien 2001. LSJ = A Greek-English Lexicon, compiled by H.G. Liddell-R. Scott, rev. and augm. by H.S. Jones-R. McKenzie, Oxford 19409; Revised Supplement, ed. by P.G.W. Glare-A.A. Thompson, Oxford 1996. Migliorini, Bruno (1975), Parole d’autore (Onomaturgia), Firenze. Munske, Horst Haider–Kirkness, Alan (Hrsg.) (1996), Eurolatein: das griechische und lateinische Erbe in den europäischen Sprachen, Tübingen.

24 The difference between zoology and zootechny is so explained: “La zootechnie est relativement à la zoologie ce que l’agriculture est par rapport à la botanique” (Essay I: 118). 25 The semantic family seems unknown in ancient times, however strange that may seem to us moderns, accustomed to the classification as zoological of certain Aristotelian works, according to a partition of Aristotelian themes that appears, totally isolated, in the late Sch. Luc. Vit. auct. 26 (perhaps contributed by the learned Arethas in the early Tenth Century). In modern times the noun zoology comes to light in the Latin of Pharmacopoeia Medico-Chymica, sive Thesaurus Pharmacologicus . . . authore Johanne Schrödero, Ulmae 1644, as the title of book V (Pharmacopoeiae medico-chymicae, Liber V. De Zoologia), but it is presented in Greek characters in the title of the book, when separately translated into English by T. Bateson (Ζωολογία: or, The history of animals as they are useful in physick and chirurgery, London 1659), and in English it immediately begins to be used, since 1669, always referring to the work of Johann Karl von Schröder. It appears later in other European languages: in French since 1776, in German since 1778, in Italian since 1792.



forging ancient greek words in modern times

289

OED = Oxford English Dictionary, online version December 2011, http://www.oed.com/. Rodriguez Adrados, Francisco (2005), A History of the Greek Language: From Its Origins to the Present (1999), Engl. trans., Leiden. Van Lancker, D.R. and G.J. Canter (1982), “Impairment of Voice and Face Recognition in Patients with Hemispheric Damage,” Brain and Cognition 1: 185–195. Vermeulen, Han F. (2006), “The German Invention of Völkerkunde: Ethnological Discourse in Europe and Asia, 1740–1798,” in S. Eigen-M. Larrimore (ed.), The German Invention of Race, Albany: 123–45. Wiener, Norbert (1948), Cybernetics, New York.

Index: names and subjects Acheloos 209 Aeneas 144 aigis 157 Agenda 88, 172 Agrippa 251 Alexandrians 79–81, 90–91, 209, 256 see also Aristophanes of Byzantium; Callimachus; Aristonicus Ambivius Turpio 132–133 ambiguitas 220 Ambiguity xx, 18, 65, 67, 110, 227 and New Critics 219–220 and Poststructuralists 220 Quintilian’s definition of 220 Ambroton eidos (ἄμβροτον εἶδος) 151 amphibolía 129, 220 Anderson, Graham 198n17, 199 Annotation 3, 6, 9, 38n72, 41, 175, 240 Annotationes in Syntaxin 241 Incorporated 32 marginal 35, 214 (pseudo-) feminist xxi Readers’ 20, 27 Antenor 155f, 159n Antonius Diogenes The incredible things beyond Thule 180 Antylus 149 Apocryphal texts xv, 1–14, 59, 66–68, 192 Apollobex of Coptos 181 Apuleius 129, 199n20, 244, 245n13, 248n36, 253, 256 aporthetos polis 146, 159 Aristonicus 81 Aristophanes of Byzantium 73–75, 79–80 Artemidorus’ papyrus 60f Arctinos 145, 148 Arts liberal 231 visual 143, 151, 158 written 235 Asclepius (Asklepios) 83–85, 181–3 Assaon 215 Athena 143–161 Attributes doru, elakate, atraktos 157 Athenaeus 250, 268 Athens xi, 20f, 60, 82, 87, 104, 126, 152n25 As center of sophistic movement 277

Athetein (ἀθετεῖν) 80–81 Attribution 50 False 86n36, 90n, 244, 259 To Alexandrians 90 To Ἀστραψοῦχος 258 To Demetrius of Magnesia 67 To Democritus 249f, 251 To Epic Cycle 152 To Euripides 80n21 To Jacob 257 To Leon the Academic 86n37 To Moses 249, 255f To Nonnus 59, 65f To Panaetius 83 To Petosiris 251n55 To Plato 77n17, 78 To Protagoras 50f, 53 To Pythagoras 249f To Simmias 207ff To Speusippus 77 Double 250 Authority viii, xvi, xxf, 8, 43–57, 91, 95f, 99f, 148, 165–168, 180, 183, 219, 229, 243, 245n9, 253, 257f, 272f. Cultural 265 Magical 243–259; see also a list under magical authorities Religious 243f, 257–259, 264, 272 sophos’ 270 Authorship viii, x, xxi, xxii, 6, 43, 47, 56, 65, 74, 76, 79, 81, 84f, 88, 90, 95–100, 110, 113, 118, 122, 164, 180,193 Alternative 76, 86, 88, 110 Epictetus’ 199 Euripides’ 80n21 Lucian’s 197, 199 Lucretius’ 22n28 Azara, José Nicolás de 11 Bayle, Pierre 7, 9 Bacchides Comparison w. Greek model 125, 134 Dramatic illusion in, break of 129 hamartia, comic 136 hubris, comic 136, 139 Intrigue 135–140 Irony, comic 138 Line 1009 125, 134–140

292

index: names and subjects

Lines 208–217 128–134 Lines 1096–1098 140 Marriage contract, temporary 135–136 Non-Plautine passages, signals of  128–129, 140 onomasti kômôidein 130, 133 Props 130, 133 Synopsis 125–8 Barbaros (βάρβαρος) 280 Bekker, Immanuel 196 Bolus of Mendes 180–1 Borges, Jorge Luis 12 Bourdieu, Pierre Cultural sociology of 266 Callimachus 63, 65, 67, 79–80 Canfora, Luciano 24n35, 60n3, 62 Catalogues Aeschylean 82 Alexandrian 61, 65, 79 Callimachus’ 79f Geographical 166, 216 Hippias’ 166 Magicians 244 Platonic writings 76 Cervantes Saavedra, Miguel de 12 Choragus 134 Cloacina grammar 240 Cobet, Carel Gabriel 197 Code of Euric 10 Commodification of wisdom 270 Corpus Hippocraticum, see Hippocratic Corpus correptio epica 210 Courier, Paul-Louis 196 Critics vii, x, xvi, 5, 16f, 21, 51–55, 68, 78n, 86, 173, 197 Alexandrian 80 Ancient 77, 86, 88 Contemporary 167 New 219 Cyprian, Saint 95, 96, 105,106 Dall’Aglio, Corradino 3 Dares 179, 180 Dardanus of Phoenicia 181, 244, 251, 259 Decker, Johann 4, 6, 10 Demetrius On Namesake Poets and Writers 67–68 Democritus 181, 244, 249–252n, 259, 276 [Ps.]-Demosthenes 188–193 Derkyllos 154, 159n Diabole (διαβολή) 270–271

Dialectical ix, 51, 55, 179, 274 pseudo-dialectical 52 Dialectics of reading 179 Didascaliae Design 131–132 didascalic records 90 Plautus’ 130, 131n, 132 Terence’s 131–132 Dindorf, Wilhelm 60–61, 196 Diogenes Laertius 62, 65, 67, 74, 75, 77–78, 83, 250 Diodorus Siculus, 180n, 251, 268 Diomedes see Odysseus Diopetes, diipetes (Διι-/Διο-πετές) 144, 148–9, 152 doll 157 dominus 227–229 gregis 131 double 158–9 daemonic 264 letter 134 negative 279 passages 29n48 work 67 See also attribution doublets 24, 30n51, 205, 222f Dictys 152, 155, 159n, 179f Dionysius Skytobrachion 180 Doxography Definition of 266 Egyptian priests 182–3 Eichstaedt, Heinrich Karl Abraham 2 Electra 157–8 enthusiasmos 213n32 epiphany 154, 183 Epithet Archaic 209 fixed 153 formulaic 144, 155 γνήσιοι 75 hesiodic? 209 Osiris’ 246 Pejorative 269n25 Simias’ 210 Euclides 43–48, 56f, 57 Eudocia 95–107 Eusebius 256n89, n90 Faber, Tanaquil 195, 205 Falsification techniques x, 3–12, 165–177, 179 Fernández Mallo, Agustín 12



index: names and subjects

Fictional pornographic world 8 Firebaugh, W.C. 7n Forgery vii–ix, xi–xiii, xx, 163–165 Literary forgery vii–xiii, 4–14, 61–68, 86–88, 163–165, 179n1 Forgeries xii, xviii, 1f, 60f, 68, 83, 87f Hippias’ 165–177 Marchena’s 3–13 C. Simonides’ 60–63 πελπασμένον “fabricated” 83n31 successful 163 Typology of 1 Fragment xx, 2–13, 19, 23n31, 41, 50, 56, 62, 65f, 85, 128, 155, 207, 211–215 fraudulent text 182–6 friendship 54, 158, 219, 226, 229 Fritzsche, Franz Volkmar 196 Garnett (Richard) The Twilight of the Gods 68–69 Genadius 11 Gibbon, Edward 9 Graecisms False 283–289 Grammar 97, 231–242 Cloacina 240 of classical languages xv Introductory grammar 237 Latin 236–238, 241 Philosophical grammar 231–234, 236–238, 241 Rational grammar 231f, 236, 238f Grammarian 64, 68, 78f, 232–242 Aristophanes the 74 Demetrius of Magnesia 67 M. Valerius Probus 7, 19 Stephanus of Byzantium 66, 286 Greed Accusation of 270, 272 -driven gullibility 182 Grenfell, Bernard 207 Hapax [legomenon] 188–193, 200n32, 201n36, n45 Hall, Jennifer 197, 198n16 Harpocration / Harpokration 182n8, 188 Hellanicus List of Priestesses of Hera 166 Hiatus 209–210 Hippias of Elis xvii, 163–177, 263, 265n12, 276, 278f, 280 Compilation 279

293

Ethnon Onomasiai 166 Synagoge 265n12, 278 Hippocratic Corpus relations of proximity between the Surgical Treatises of 111, 123 relative chronology of the Surgical Treatises of 122–123 Holzberg, Niklas 199 Homer Homeric language 97–100, 146n10, 151, 202n56, 209–211, 246 See also Iliad; Odyssey Homerocentones 95–107 Hypatia 66 Charles Kingsley’s 68 Hyperboreans 211–215 Iamblichus 247 Iliad 81, 95, 100–102, 143, 153, 155, 167 Little Iliad 137, 152n24, 155 sweet sleep in 201n40 Impartiality 173 Interpolation xix, 81, 128, 137n, 139n, 140 in Lucretius 30–41, esp. 36–9 Jacobitz, Carl 196 Jerome, Saint 11, 18f Jesus 101–104, 107 kabbale (κάββαλεν) 150f Kelsey, Francis Willey 207 Kritik Höhere 125, 129 Kyranides 180–1, 182n8 Lament 95, 101, 103f Le Fèvre / Lefebvre, Tanneguy see Faber Legitimation 8, 12, 49, 53, 82, 85, 90, 163, 165, 168, 171, 176, 266, 274 Libraries xx, 10, 60, 79, 90, 182, 185 Lucian xvi Corpus 86n37, 195, 196 Name confusion ixn4, 205n74 Style 197 True Stories 184 Philopseudes 184n [Ps.]-Lucian Alcyon 195n1 Onos 195–205 Lucius author of Metamorphoses (?) 196 Lucretius 15–41 See also interpolation; transposition; repetition

294

index: names and subjects

Macleod, Matthew 198n12 Magical authorities Aaron 255 Agathocles 252 Apollonius of Tyana 252, 253, 259 Astrampsychos, 257–258–259 Bolus of Mendes 250 Claudianus 254, 259 Dardanus 244, 251, 259 Democritus 244, 249, 250, 251, 252n64, 259 Erotylos 249 Euenos (Evenus) 251 Hieros 249 Himerius 252, 259 Ieu 249, 259 Jacob 257, 259 Manetho 246, 255n86, 259 Moses 244, 246, 248n36, 249, 251, 254, 255, 256, 257, 259 Nephotes 246, 259 Orpheus 249 Ostanes 257, 258, 259 Ourbikos (Urbicus) 248, 259 Pachrates (Pa[n]krates) 247 Paysakh (Pasash) 245, 259 Philinna the Thessalian 253, 254n78, 259 Pibechis 248, 259 Pythagoras 244, 249 Pitys (Bitys) 245, 247, 248, 258, 259 Pnouthis 246, 247, 259 Pyrrhus 252, 259 Solomon 254, 255, 259 Thphe 245n9 Zminis of Tentyra 245, 258, 259 Zoroaster 257, 258, 259 Marchena, José 1–14 Ars amandi 8 As a propagandist 2 As a translator 2 Clandestin editions 5 Ancient Sources 6–8 Cosmopolis 5 Courtesans 4, 6 Cum notis variorum 6 “Epithalamium of Peleus and Thetis” 3 Essai de Théologie 2, 11 Fragmentum Catulli 2, 3 Fragmentum Petronii 1–13 The history of the footnote 9 see also scholarly notes Interpretation of virginity 7 Literary salons 6 8

Multiple Authorship 6 Philosophical materialism 8 Satyricon of Petronius 1–12 Scholarly notes 3–9 See also Forgery Marsias 215 Means vii–ix, xxi, 87, 91, 165, 168f, 171 Meléndez Valdés, Juan 1 Menéndez Pelayo, Marcelino 5, 11n Monastery of Saint Gall 10 Motive vii, ixf, xxi, 165, 168f Murat, Joachim 2 Muriel, Andrés 11 Mysteries 145f Aura of 180 Nonnus’ 67, 69f. Samothracian 146, 151, 251 Napoleon 2 Nechepso book of 182–5 Niobe 215 Nodier, Charles 1 Questions de littérature légale 1 Noël, François 3 Traduction complète des Poésies de Catulle 3 Nonnus Dionysiaca 59, 63, 66, 69, 70 False Biography of 63–64 Paraphrase 59, 63, 65, 69, 70, 103, 107 Numenius 256 Nymphs choruses of 211–213 Obelos 78–79 Odysseus and Diomedes 147, 150–152, 155–6 Odyssey 81, 95, 100, 103, 104 Athetesis of 81 sweet sleep in 201n40 Olympic Olympionikon Anagraphe 163f, 166 Victors 163, 165–177 Oracle 144, 146f, 150–152 Delphic 170, 176 Oracula Sibyllina 209n10 Oracular Language 147n11 Oracular tradition 151 Oracular powers 246 Sortes Astrampsychi 258 Origen (Origenes) x–xii, 64, 66, 256n89



index: names and subjects

Palimpsest 10–12, 61 Palladion 143–161 Pallas A man 145 A young girl 145, 156–8 Panaetius 76, 83–85 Paratext xvii, xxii, 9–12 Patricius 97–99 Pearson Correlation Coefficient 119–121, 122, 188 Pelasgians 144 Perry, Ben Edwin 198n12 Penates 145, 156 Petronius 1–12 Phaedrus 46–48, 53, 87, 89, 195n Philostratus ix Biography of Apollonius 252 Gymnasticus 169 Phocylides 251 Photius 196 Plato Anonymous Prolegomena to Platonic Philosophy 77–78, 88n46 Apology of Socrates 86–87 Appendix Platonica 76–78 Authenticity 74–77 Dubia and spuria 86–88 γνήσιοι “genuine” 75–76, 80n23, 82, 90–91 νόθοι “bastards” 76–91 Paternity and legitimacy 81–82, 85–86, 89–91 Phaedo 76, 83–85, 86–87 Platonic Corpus 44, 73, 75, 88–89 Trilogies 74–75 Tetralogies 75–77, 89 Text and manuscripts 75, 77n17, 82, 86n37, 88n47 Theaetetus 43–45, 51, 56f, 74, 86–87, 89, 275 Platonic tradition 77n15, 264–265, 272 Plautus See Bacchides; Didascaliae; Publilius Pellio Pliny 38n71, 181n6, 182n9, 184n13, 244, 250, 251 Power Non-material 12–13 Of a counter-narrative 168 Of friendship’s bonds 219, 226f, 229 Of love 226 Oracular 246f Secret/magical 144, 149–151

Psychology forger’s knowledge of 163, 171f, 177 Panaetius’ 85 reader’s xxi Prediction Cryptic 150 Helenus’ 151 Socrates’ 44 See also oracle Proba 100, 107 Proclus 83n31, 145n5, 152, 155 Protagoras xxi, 48–57, 263, 267, 269, 274–279 Pseudo-documentary fiction xxii, 179–86 In ancient scholarly texts 180–1 Pseudonyms xv, 6n, 231–242 Grammatical 232–242 Publilius Pellio 128–134 Quérard, Joseph-Marie 4, 6 Quintilian 40, 78–79, 220, 235 And ambiguitas 220 And amphibolía 220 Rabener, Gottlieb 8 “Hinkmars von Repkow Noten ohne Text” 8 Refutation 52, 264n10 Self-refutation 52 Regularity principle 167, 171 Relativism 57, 275f Repetition 99, 103, 210, 225, 228 in Lucretius 32–36 Rewriting of history and literature 12 Rhetoric xvi, xix, 85, 121, 123, 221, 272–273 Of scholarship 8 Rhetorical education 78, 96 Rhetorical move 219 Rhetorical qualities 248 Riemschneider, Margarete Im Garten Claudias 69–70 Rites Bacchic 213n30 Kabeiroi 251 Sacrifice of donkeys 214–215 Sacrifice of Polyxena 3 Sacrifice to Athena 156 Sacrifice to Zeus 168f Snake 105 Rohde, Erwin 197 Rome xvi, 144–145, 148 Rucco, Deacon of Paris 221–229

295

296

index: names and subjects

Samothrace 145–148, 155, 161 Salas, Ramón 1 Samaritan woman 101–104 Schmid, Wilhelm 197n12, n13 Scientia sexualis 8 See also Sexual behaviour Schoell, Frédéric 4, 6 Scriblerus Club 6 Sexual behaviour in antiquity 4, 9 Simias of Rhodes 207–217 Simonides 170, 215 Simonides, Constantinos 60–70 See also Artemidorus’ papyrus Forgeries 60–63 Life of 59–60 Uranius’ Historia Aegyptiae 60–61 Smith, Adam 1 Socrates xxi, 43–57, 74, 76, 86–89, 268, 272, 275–278 Scholasticus 68 and Socratic dialogues 74, 76, 85, 86–87 Solon 82, 203n64, 250n49, 269, 276 Sommerbrodt, Julius 196n5 Statue xx, 143–157 Olympic 157, 170 Thessalus’ 183 Sophia (σοφία) 264, 271, 273 Field of 263, 266 Medicine as 272, 272n32 Sophists As counterfeiters of wisdom 274–275, 280 As Platonic design 263 As qualitatively distinctive 264, 276 Lack of doxographical tradition 266 Philosophical rehabilitation of 265 Platonic definition of 263–264 Sophistes (σοφιστής) 267 Ancient use of 268, 269 Used in reference to Aristotle 269 Used in reference to Isocrates 268 Used in reference to Plato 268–269 Used in reference to Presocratics 268 Used in reference to Socrates and Socratics 268 Used in reference to Solon 269 Sophistry xvi, 264 Sophos (σοφός) 264, 265, 267, 269, 270, 272, 277, 278 Accusation of greed against 272 Student’s t-test xix, 110, 112–119, 122

syntactic Chiasmic model 226 Distribution of sentences 110, 111 Pattern 110, 113, 195, 196, 222f. Relationship 196 Syntax 222–226 Brachylogic 120 Rule-bound 241 Teaching xxi about the ψυχή 85 for pay 270–273 pagan 66 Plato’s 87 Scioppio’s 232, 237, 239, 241 Teiresias 272 Terence Prologues 130 Didascaliae 131–132 See also Ambivius Turpio; Didascaliae Thales 279 [ps.]-Thessalus of Tralles 181–186 On the virtues of plants 181–186 Thrasyllus 75, 76, 83, 89, 91 Thurii 277 Thymoetes 180 Transposition 147n14 in Lucretius 31–2 Travel 60, 63, 126, 155, 180, 214, 250, 276–277 Triton 156 Tritogeneia 156 Troy 144 Typology of forgeries Invention of supposed letters 1 Supposititious passages 1 van Thiel, Helmut 197 Venantius Fortunatus And authority 219 And order 219 And public and private audiences 219 And Rucco 221–229 And uses of ambiguity 219–220 Vesta 148f, 156 Wieland, Christoph Martin 196 Women in 18th century 5 Xenophon 76, 85, 88, 264, 266, 268f, 271, 275 Xoanon see statue; palladion