Papers of the Leeds International Latin Seminar, vol. 7: Roman poetry and prose, Greek rhetoric and poetry 9780905205878, 0905205871

Citation preview

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR SEVENTH

VOLUME

1993

ARCA Classical and Medieval Texts, Papers and Monographs

32

General Editors: Francis Cairns, Robin Seager, Frederick Williams Assistant Editors: Neil Adkin, Sandra Cairns ISSN 0309-5541

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR SEVENTH VOLUME

1993

Roman Poetry and Prose

Greek Rhetoric and Poetry

Edited by Francis Cairns & Malcolm Heath

X FRANCIS CAIRNS

Published by Francis Cairns (Publications) Ltd c/o The University, LEEDS, LS2 9JT, Great Britain

First published 1993 Copyright © Francis Cairns (Publications) 1993

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording,

or otherwise, without the prior permission of the Publisher.

British Library Cataloguing in Publication Data A catalogue record for this title is available from the British Library ISBN 0-905205-87-1

Printed in Great Britain by Redwood Books, Trowbridge, Wiltshire

CONTENTS Roman Poetry and Prose NICHOLAS

HORSFALL

Cicero and poetry: the place of prejudice in literary history MATTHEW

DICKIE (University of Illinois)

Malice, envy and inquisitiveness in Catullus 5 and 7 ALLAN

KERSHAW

(The Pennsylvania State University)

A! at Catullus 68.85 W. JEFFREY

TATUM

(The Florida State University)

Catullus 79: personal invective or political discourse? ROBERT MALTBY (University of Leeds) Varro's attitude to Latin derivations from Greek PETER

E. KNOX

(University of Colorado at Boulder)

Philetas and Roman poetry S.J. HEYWORTH (Wadham College, Oxford) Horace's Ibis: on the titles, unity, and contents of the Epodes

STRATIS KYRIAKIDIS (University of Thessaloniki) Aeneid 6.268: Ibant obscuri sola sub nocte per umbram

FRANCIS CAIRNS (University of Leeds) Imitation and originality in Ovid Amores 1.3

101

MARTIN HELZLE (Case Western Reserve University) Ovid's Cosmogony: Metamorphoses 1.5-88 and the traditions of ancient poetry

123

VINCENT HUNINK (Catholic University, Nijmegen) Lucan's praise of Nero

135

Greek Rhetoric and Poetry G.M.

PAUL

(McMaster University)

Josephus Bellum Judaicum 4.559-63: Invective as history DOUGLAS

L. CAIRNS

141

(University of Leeds)

Affronts and quarrels in the Jliad

155

MALCOLM HEATH (University of Leeds) Ancient interpretations of Pindar's Nemean 7

169

VIRGINIA KNIGHT (University of Manchester) Landscape and the gods in Callimachus! Hymns

201

C. ANNE WILSON (University of Leeds) Dionysian ritual objects in Euphorion and Nonnus

213

Preface

Papers of the Leeds International Latin Seminar, Seventh Volume, 1993 (PLLS 7) continues the series begun with the five volumes of Papers of the Liverpool Latin Seminar and followed by Papers of the Leeds International Latin Seminar, Sixth Volume, 1990. PLLS 7 includes some of the papers presented at meetings of the Leeds International Latin Seminar over the years 1990-92 in revised form (among which the Colloquium of May 1991, ‘Enmity, Envy and Insult in Ancient Poetry,’ was a particularly fertile source), together with other contributions. The reference conventions followed are those of PLLS 6. The Editors are again grateful to their referees for advice, and to the Assistant Editor of ARCA, Professor Neil Adkin, for his customarily meticulous final proof reading. Francis Cairns, Malcolm Heath

School of Classics

University of Leeds June, 1993

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 1-7 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

CICERO AND POETRY THE PLACE OF PREJUDICE IN LITERARY HISTORY NICHOLAS

HORSFALL

No prizes for reinventing the wheel, even if in an unusual way, and certainly no justification therein for rushing into print. No reason, then, to reiterate! that Cicero had strong and retrograde views about much contemporary poetry, but that this acknowledged fact conceals a whole string of details neglected by those who hurry on to study Catullus and his fragmentary friends. Exquisite Alexandrian "technique is, after all, more interesting to most than the sources and

expressions

of unilluminated

(or so they seem) and

reactionary

jeremiads. But it is the jeremiads that I have examined, and I venture to summarise briefly the results. Ways of writing the history of literature? do tend to exploit the ancient texts with polemic and partiality; it may, however, also prove rewarding to ponder one author's

outlook

on

contemporary

poetry

to try to reach some

understanding of the prejudices implied and their possible origins. The full span of Cicero's literary activity, from juvenilia on, covers the years 90-43 BC; he might have known both Accius and Virgil and we might expect to bow reverently before his expert judgements. But if one looks with care at Schanz-Hosius, at Teuffel-Schwabe-Kroll and at Bardon's unknowns, a list emerges of forty-six (near enough) poets of whom we know something (if only a name), active during that period. But by name and in respect of poetry, Cicero mentions only: (i) himself; (ii) his brother; (iii) Lucretius;? (iv) the impossible Empedoclea of Sallustius;* (v) Varro's Menippea (Academica 1.9); (vi) Licinius Calvus' Hipponactean attack on Hermogenes Tigellius (Ad Familiares 7.24.1); (vii) Furius Antias (Brutus 132); (viii) Q. Mucius

2

Scaevola (De Legibus

NICHOLAS

HORSFALL

1.1; cf. Pliny Epistulae 5.3.5.); (ix) Lutatius

Catulus (De Natura Deorum 1.79). One fifth of the total. Catullus mentions Cicero (49) and not vice versa (Malcovati (n.5) 226). One

might also contrast the long exchange of compliments between Catullus and Nepos. Cicero's silence is, moreover, even more deafening than might at

first appear: at Atticus' house there was always a reading after dinner (Nepos Atticus 14.1), though it is surely thanks merely to accidents of transmission that all our allusions in Cicero are to recitals of prose (see below): it becomes ever clearer that Seneca Controversiae 4 praef. 2: primus enim (sc. Asinius Pollio) omnium

Romanorum

advocatis

hominibus scripta sua recitavit, whatever exactly it means, does not mean exactly what it says.’ Dinners as a literary context at Rome merit further study: one read or wrote letters*; one might talk of legal problems (Ad Familiares 7.22), one might envisage a reading of the De Gloria (Ad Atticum 16.2.6, 3.1); Cicero even says that he would not mind reading a letter of his son's (4d Atticum 15.7.2). When he writes

to Atticus (4d Atticum 12.4.2) non adsequor ut scribam quod tui convivae non modo libenter sed etiam aequo animo legere possint, a public reading of the Cato does not seem to be implied: legere possint after all, not audire. Not a word, however, of poetry, at least in Latin (though this may be mere accident). But Greek is not to be excluded: Cicero writes of Philodemus' poems as et lecta et audita (In Pisonem 71) and it is clear from Pro Archia 18 that Cicero had heard a lot of

Archias' poetry performed, and not just his famous improvisations.? With Thyillus, Cicero was in contact (Ad Atticum 1.16.5 etc.; Malcovati (n.5) 85); of Alexander Lychnus (Ad Atticum 2.20.6) he

snaps neglegentis hominis et non boni poetae sed tamen non inutilis (sc. as a geographical source). Of contemporary Greek poetry no more will be said. There is no more to say. Cicero's own freedman Tullius Laurea wrote verse in both Greek and Latin (Schanz-Hosius

1.548) and will hardly have been denied

the chance to display his talents after dinner. Atticus himself wrote epigrams — those under the /magines, compiled after Cicero's death;'? 4d Atticum 1.16.15 may allude to Atticus' Greek verses under the statues in the Amaltheum of his garden in Epirus.!! It still bears emphasis that Cicero was himself a skilled and indefatigable poet: the Aratea he wrote admodum adulescens (De Natura Deorum 2.104); in time, they fall between Porcius Licinus and the Pompeian

epigrams discussed by David Ross.'? Cicero's whole political and literary biography makes it highly unlikely that the Prognostica (Ad

CICERO AND POETRY

3

Atticum 2.1.11, 60 B.C.) is a new, or different work: rather, he refers

to different parts of Aratus' poem in different ways, in keeping with Alexandrian usage. Prognostica is the title of the third part of the Phaenomena of Aratus," and Maass' hypothesis that Cicero learned the correct terminology from Boethus of Sidon's Peri Aratou is still attractive.!^ A year before his death, Cicero continues to insert highly

polished verses in his philosophica. As a young man, Cicero reached advanced levels of technical elegance and competence,

but Alexandrianism

he forswore and

‘Neoteri’ he disliked.'* However, a problem recently raised!" remains to be clarified: the suggestion that the ‘Neoteri’ belong to the years 65-55 BC and tha: Cicero's criticisms in 50-46 BC are therefore the flogging of a dead horse. Licinius Calvus probably died not long after

54 BC.'5 Only if "Lucius Iunius Calidus’ at Nepos Atticus 12.4 is emended to ‘Gaius Licinius Calvus’,'® “the most elegant poet that

our age has produced after the death of Lucretius and Catullus", do we have evidence for a slightly later date of death. If we do not emend the text of Nepos, in the place of the deceased Calvus we have the living and brilliant Calidus! Calvus apart, Cinna lives on till 44 BC,

Varro Atacinus till perhaps 36 BC (though he is anyway 'neoteric'

only in a limited sense);?° Cornificius is quaestor in 48 BC. In the 40's BC the young Gallus is likewise active.?! Poets really did start young; it was not a prerogative of the English romantics. In Cicero's correspondence with Calvus, Pollio and Cornificius, though, there is not a word about poetry. Nor is there any chronological foundation for the view that the *Neorteri were played out by 50 BC. Cicero, then, forswears his youthful pipings, imitated by Lucretius

and Catullus,” and uses traditional Alexandrian pejorative language (neoteri, novi)? against those now regarded as the leading poets of the day. Plutarch refers (Cicero 40.2) to Cicero's speed at writing verse. Poetry is hardly the word. Quintus was cursed with the same capacity (Ad Quintum Fratrem 3.5.7). One wonders whether they had not both

learned too much from Archias. At all events such tachygraphic

composition is not compatible with the technique of the Aratea and the observation must apply to the De Consulatu, the De Temporibus,?^* or the British epic. It is beneath Cicero's dignity to refer to the occasional poetry of great men, as recorded by Pliny, Ovid and Apuleius;? it might indeed be bad manners towards them to do so. Scaevola and Catullus (supra) are long dead. Poetam non audio in nugis (Paradoxa Stoicorum

3.26). The summary judgement

on Lucretius betrays no serious

4

NICHOLAS

HORSFALL

interest in contemporary literature: our histories of Latin poetry are cluttered with dozens of equally banal and unhelpful epigrams. Quos postea Cicero emendavit writes Jerome of Lucretius’ libri, after Suetonius. If he is right (and the rest of the entry does not inspire confidence),

we

have

still to evaluate

emendavit correctly;?’ the

word should not convey a picture of Cicero, after a hard day in the Senate, sitting down with the precision of a Delz or a Zwierlein. The word need imply no more than the elimination of those obvious inconcinnities, resulting from the author's death, which impeded ‘publication’. Anyway, one might suspect that Atticus’ slaves will in practice have undertaken the hard and delicate parts of the job.?® The answer to the problem, or paradox, here raised is at least not political: the poets hitherto mentioned belong to various groupings or factions, or slide easily between them. The Atticist controversy is equally irrelevant: Calvus was indeed involved, but on account of his

prose.? It is sentiment that lies at the heart of Cicero's irritable disdain; as an elderly consular, he has a sentimental objection to the

novi, and it is no accident that, in two of the key passages, such objections are linked to praise of older poetry:?? in his youth Cicero had omitted final s quod iam subrusticum videtur, olim autem politius . ita non erat offensio in versibus quam nunc fugiunt poetae novi (Orator 161); likewise, Cicero's admiration for Ennius' Andromacha

stands against the cantores Euphorionis by whom the poeta egregius nunc contemnitur (Tusculan Disputations 3.45). His informed love for

the great comic poets is amply acknowledged;?! he laments the loss of the carmina convivalia.? He knew (‘thought he knew’, perhaps) old manuscripts of Ennius: ipsius antiqui declarant libri. The grown man loved the second century, as De Amicitia, De Senectute and De

Republica amply confirm; the youth had flirted with Alexandrianism, ultra quam concessum Romano et senatori one might almost say. Part, then, of the explanation, is simply a question of ageing, of fashion, of the passing of the generations. There are a couple of hints in the Rhetorica" which could be taken as attacks on Alexandrian verbal preciousness. The scatter of archaisms in Cicero's own poetry? occur from Aratea on and are hardly significant to this argument. To call Euphorion nimis obscurus (De Divinatione 2.133) is no less than the truth, but, equally, coheres with Cicero's middle-aged sense of being out of step. I do think we may invoke old Roman prejudice against

admitting to knowledge of Greek culture,’ or indeed those traces of a certain opposition at Rome to poetry in general? But only in a public speech did Cicero need to pretend to be coy about knowing

CICERO

AND

POETRY

5

who Praxiteles or Polyclitus were; among the friends to whom he wrote letters, the ignorance and suspicion faced (e.g.) by Nepos in the Preface to the Foreign Generals were most unlikely to have prevailed to any degree at all. The technical skill entailed in rendering Aratus in Latin?? is, moreover, precisely that of which he is justly

proud in the public, and Roman, philosophica.* Plutarch says (Cicero 2.4) that, as time went on, Cicero ἔδοξεν οὐ

μόνον ῥήτωρ ἀλλὰ καὶ ποιητὴς ἄριστος εἶναι Ρωμαίων but, many good poets having since flourished, his ποιητική has become altogether ἀκλεῆ καὶ ἄτιμον. This does not necessarily mean (pace Soubiran (n.22) 3f.) that Cicero thought himself Rome's best poet. It could; ἔδοξεν could also imply ‘appeared’, ‘was held to be’. He did not demonstrably share Marsyas’ ambitions. There was, though, much irony at the expense of his verse, from the first Sallustian invective against Cicero (6) and a speech attributed to Fufius Calenus by Dio (46.21.4)*! on. Neither text is datable, but both must reflect early disapproval, based, one imagines, on the less felicitous passages of the De Consulatu and De Temporibus. In the De Ira, Seneca writes (3.37.5):

et Cicero,

si derideres carmina

eius,

inimicus esset. The

remark is, alas, quite hypothetical, though it would be notably more telling if it were not altogether fanciful: the language is, of course, very strong. Nec enim satis commoveor animo Cicero complained to Quintus in November 54 BC (Ad Quintum Fratrem 3.5.4); abest enim ἔνθουσιασμός he had lamented in the previous letter (3.4.4). Caesar's victories had clearly kindled no vital spark, and while Furius Bibaculus and Varro of Atax got on with the job, age and disappointment worsened Cicero's ‘block’, though somehow or other, invitis Musis he did finish the epic by December 54 BC (Ad Quintum Fratrem 3.7(9).6). Where, one wonders, and when did he

mira benignitate ingenia poetarum fovisse (Pliny Epistulae 3.15.1)? It is of course possible that, were they good, he could have asked Atticus to lend a hand with a suitable dinner for a reading, or with the production of a few copies. That does not make Atticus a publisher,*? nor does it require mention in a letter (clearly: there is nothing of the kind). Cicero is a skilled poet who loses interest in poetry; the artist who falls out of step with his contemporaries has been, in passing, recognised, and, often, mocked, but hardly, hitherto, understood.

NICHOLAS HORSFALL

NOTES W.V. Clausen CHCL 2, 178; A. Traglia Poetae Novi (Rome 1974); E. Paratore Poetiche e correnti letterarie (RCCM Quaderni 10, 1970) 56; cf. n.30. Cf. G.J. de Vries Mnem.

36 (1983) 241-59.

QF 2.10.3; cf. n.26.

virum te putabo si Sallusti Empedoclea legeris, hominem non putabo: QF 2.9(11).3 (54 BC). E. Malcovati Cicerone e la poesia (Pavia

1943) 227f.

Nep. Att. ed. Horsfall (1989) xv, 117.

G. Funaioli PW s.v. recitationes, E. Rawson Intellectual Life in the Late Roman Republic (London 1985) 52; T.P. Wiseman in Literary and Artistic Patronage ed. B.K. Gold (Austin 1982) 36 = T.P. Wiseman Roman Studies (Liverpool 263; id. History 66 (1981) 385 = Roman Studies 254.

1987)

Cic. Fam. 9.26; OF 3.1.19; Fam. 4.12.2, 9.7.1; Att. 9.8.1 etc. G.W. Williams in Gold (n.7) 8ff.; Wiseman ib. 32ff.; A. Hardie Statius and the

Silvae (ARCA 9, Liverpool 1983) 22.

Cf. Horsfall (n.6) 102. Cf. Horsfall (n.10) foc. cit.; F.G. Moore CPA.

1 (1906) 25ff.

YCS 21 (1969) 125ff. J. Martin Sur /e texte des Phénoménes d'Aratos (Paris 1956) 155 line 3. Aratea (Berlin

1892) 157.

N.B. Crowther LCM 5 (1980) 182f.: W.V. Clausen HSCP 90 (1986) 159-70. Clausen (n.15) 161; irony in Cat. 49 is by no means excluded (and may also be present even in Cat. 1). G. Bowersock in Le classicisme à Rome (Entr. Hardt 25, 1978) 62.

Münzer PW 13.1.433.10ff.; P.L. Schmidt KP s.v. C. Cichorius Rómische Studien (Leipzig 1922) 88f.; Horsfall (n.6) 86. Cf., most recently, N.B. Crowther AC 56 (1987) 266f. Clausen (n.15) 161. Cicero Aratea (Budé) (ed. J. Soubiran) 72ff.

R.O.A.M. Lyne CQ 28 (1978) 168; A.D.E. Cameron HSCP 84 (1980) 136. Cf. S. Harrison ‘Cicero’s De Temporibus suis: the evidence reconsidered’ Hermes 118 (1990) 455-63. Cf. Plin. Ep. 5.3.4 (with Sherwin-White ad /oc.), Ov. Trist. 2.421ff. (with Owen ad loc.).

CICERO

26.

AND

POETRY

A. Dalzell CHCL

7

2.213 is irreproachably balanced; the issue is clouded by

unending debate: cf. E. Paratore (n.1) 53 n.62; ANRW

1.4.138-43.

27.

J.E.G. Zetzel Latin Textual Criticism (Salem NH 1984) 206ff.; id. HSCP 77 (1973) 227ff.. S. Timpanaro Per la storia della filologia virgiliana (Roma 1986) 23f., 31f.

28.

Horsfall (n.6) 88f.

29.

Cf. Paratore (n.1) 56; Clausen (n.15) 160 n.3.

30.

Cf. N.B. Crowther CQ 20 (1970) 324-6; id. Mnem.

31.

J. Griffin Latin Poets and Roman Life (London 1985) 200; F.W. Wright Cicero and the Theater (Northampton, Mass. 1931); Malcovati (n.5) 89ff.

32.

Cic. TD 4.3; Brutus 75 (utinam exstarent ...).

33.

Orator 160; one seems almost to catch a whiff of Gellius here; but 1 have no wish to glean where Skutsch (Annales 26ff.), Jocelyn (The Tragedies of Ennius 26ff.)

29 (1976) 68.

and Malcovati (n.5) 95ff. have reaped. Malcovati (n.5) 224. Soubiran (n.22) 96f.

Landgraf on Cic. Rosc. Am.

46; H. Jucker

Vom

Verhältnis der Römer zur

bildenden Kunst der Griechen (Frankfurt 1950) 90f.; N. Attitudes to the Greeks (Athens 1974) 194f. Cic. Arch.

Petrochilos

Roman

12; TD 1.3f.; Petrochilos (n.36) 170f.; Cic. ap. Sen. Ep. 49.5.

Horsfall (n.6) xx. Cf. also, on the geographical project, Schanz-Hosius

14.534f.

J. Kaimio The Romans and the Greek Language (Helsinki 1979) 224f.; Petrochilos (n.36) 155ff.

al.

Cicerone: I frammenti poetici (ed. A. Traglia) 19.

42.

Horsfall (n.6) 12, 88f., 95.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 9-26 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA

32. ISBN 0-905205-87-1

MALICE, ENVY AND INQUISITIVENESS IN CATULLUS MATTHEW

5 AND 7 DICKIE

1. Introduction

Catullus 5 and 7 are variations on the theme of Catullus' desire for Lesbia's kisses.! In 5 he proposes that when they have kissed many thousands of times, they should make a muddle of their counting, to prevent themselves knowing the total and to prevent any malevolent person directing his invidia against them when he knows how often they have kissed (conturbabimus illa, ne sciamus,/ aut ne quis malus invidere possit,/ cum tantum sciat esse basiorum, 11-13). In 7 Catullus declares that he will be satisfied with a number of kisses beyond the power of busybodies to count or the evil tongue to put a spell on (quae nec pernumerare curiosi/ possint nec mala fascinare lingua, 1100. These poems have been carefully studied; in particular, their Alexandrian characteristics, realand imagined, have received minute scrutiny, and little can be added.? My main concern is with Catullus’ use of essentially Greek ideas about envy, malice, inquisitiveness and the Evil Eye, and with the intelligibility of these Greek motifs to a Roman audience. The paper’s main focus is therefore on the Roman world’s absorption of Greek patterns of thought. 2. The motif of love threatened by envy The danger that Catullus seeks to avert in 5 and 7 is theenvy that the felicity of his and Lesbia's love, symbolised in the number of kisses they have exchanged, may draw on itself.” The motif of the envy 9

10

MATTHEW

DICKIE

aroused by the shared happiness of love is less familiar to us than the envy which the successful lover engenders in his rivals and the jealousy which makes men or women want to keep their partners to themselves. Nonetheless, there is evidence that the theme of envy destroying mutual love by creating dissension between lovers or between husband and wife existed in Greek literature. The idea that a harmonious marriage is something highly desirable, which will be a source of distress to one's enemies, goes back to Homer. Odysseus prays that the gods may grant Nausicaa a husband, a household and fair ὁμοφροσύνη than which nothing better exists; its virtue lies in the distress it causes enemies, the joy it gives those who are well-disposed and in the couple's consciousness of their ὁμοφροσύνη (Odyssey 6.181-5). Precisely because dpoφροσύνη represents to the ancient mind an elevated form of felicity, it was thought to arouse the Evil Eye of envy. Already in the Odyssey Penelope attributes her own and Odysseus' suffering to the envy aroused in the gods by the thought of their being able to enjoy each other's company, of their taking delight in their youth and finally of their reaching old age together (23.210-12). In the Odyssey the enjoyment that men derive from each other's company is also vulnerable to the jealousy of the gods. Thus Menelaus sadly concludes that Zeus must have begrudged both Odysseus and himself the unceasing pleasure of each other's company by preventing Odysseus’ return and so foiling his (Menelaus’) plan to bring Odysseus from Ithaca and settle him near Sparta, so that they might be able to delight without end in each other, until the black cloud of death enveloped them (4.171-82). The motif of an envious supernatural force that cannot abide to see ὁμόνοια in a marriage can be traced from Homer to the Greek romances, and receives its most extended treatment in Chariton. The

mishap that separates Chaireas and Callirrhoe shortly after their marriage is brought about by the action of a βάσκανος δαίμων to whom the narrator introduces us immediately after describing the rejoicing that surrounded the wedding (1.1.16). The rejected suitors, guided by φθόνος, attempt to break up the marriage by inspiring jealousy (ζηλοτυπία) in Chaireas' breast. Their initial efforts fail, although Chaireas' discovery of evidence suggesting that his bride has entertained other suitors in his absence provokes a quarrel between the lovers. This quarrel is made up and both sets of parents pronounce themselves blessed in the harmony (ὁμόνοια) they witness in their childrens' relationship (1.2.1-3.7). The second

MALICE,

ENVY

AND

INQUISITIVENESS

1

attempt is successful: Chaireas in his insane jealousy kicks Callirrhoe in the solar plexus and apparently kills her (1.4.1-12). The second wedding-day in the novel is that of Callirrhoe and the Milesian Dionysius. The wedding is to take place at the sanctuary of Ὁμόνοια, where we are told bridegrooms customarily received their brides (3.2.16). Callirrhoe is acclaimed as Aphrodite herself, her path is strewn by flowers, everyone is in the streets to see her and people even climb up on to the tiles of the roofs; but again the envious demon takes out his spite on that day (ἀλλ᾽ ἐνεμέσε καὶ ταύτῃ τῇ ἡμέρᾳ πάλιν ὁ βάσκανος δαίμων ἐκεῖνος 3.1.17). In a letter of Aristaenetus, a woman, Melissa,‘ writing to her lover, says that Eros has scared away (dneoößnoev)the βασκανία that had affected them, that Aphrodite has appeared in the röle of ἀλεξίkaxos,’ and that both deities have put an end to the total warfare that had existed between the lovers. Those envious of their love (οἱ βασκαίνοντες ἡμῖν τῆς φιλίας), she declares, had rejoiced in vain and their plot had ended in failure. She goes on to describe her reconciliation the day before with her lover: as she entered the lover's house at something approaching a run she had wept for joy, had been insatiate in the kisses she bestowed on this house of love and had touched the walls and then kissed her fingers, all the while smiling sweetly and rejoicing; she had begun to mistrust her sanity, but when her lover saw her, he had stretched a finger towards her and had then

turned it round in a welcoming gesture (2.14.1-17 Mazal). Aristaenetus is here concerned not just with envy but with the Evil Eye. That is evident from the apotropaic röle he assigns Eros and Aphrodite in scaring off βασκανία. The effect of the Evil Eye’s bewitchment on the lovers is to make them quarrel with each other.® The description of the kisses Melissa bestows on the lover’s house as she returns for the reconciliation (a reversal of the usual motif of the maiden kissing her bed, the doorposts of her room and its walls before leaving it for the last time, or the wife kissing the same objects as she takes leave of her marriage-chamber)’ suggests that the story comes from elegy or a

novella.? In a novella in Plutarch's Mulierum Virtutes, Aretaphile, the wife

of Nicocrates, tyrant of Cyrene, is caught trying to poison her husband and defends her actions as an attempt to counteract the magic evil women had directed at her marriage; the women were envious of Nicocrates' good will towards her and the glory and power that were the rewards she reaped from his love (ἀλλ᾽ ὑπὲρ μεγάλων, εἶπεν, ὦ ἄνερ, ἀγωνίζομαι, τῆς σῆς εὐνοίας πρὸς ἐμὲ καὶ δόξης καὶ

12

MATTHEW DICKIE

δυνάμεως, ἣν διὰ σὲ καρποῦμαι πολλαῖς γυναιξίν 256b-c). These parallels? suggest that Catullus shared love can arouse the Evil Eye of working from Hellenistic literature." He

ἐπίφθονος οὖσα κακαῖς has taken the idea that envy or envious magicdoes not say what he

imagines the effect of the Evil Eye will be on Lesbia and himself, but the most obvious danger is a quarrel; other possibilities such as

impotence or Lesbia's falling for someone else cannot be excluded. The idea of an envious force ready to cause a schism between husband and wife or between friends can also be found in later Latin literature. At the beginning of Silvae 3.5 Statius states confidently that his wife is not sighing because she longs for another lover, and boasts that, if put to thesame test as Penelope, she would remain true to him (1-10). To his confident assertion that his wife allows no entry

to love's arrows he adds parenthetically that Rhamnusia may with

hostile look hear him say this: nullis in te datur ire sagittis (audiat infesto licet hoc Rhamnusia vultu), non datur. (4-6)

Confirmation that the infestus vultus with which Rhamnusia hears Statius boast is the malevolent and maleficent gaze of envy aroused by a husband confident in his wife's affection and loyalty is provided by Silvae 2.6. In this poem, a consolatio addressed to Flavius Ursus for the loss of a favourite boy-slave, Statius treats Rhamnusia and [nvidia as one and the same entity, and endows Rhamnusia with the baleful gaze of the Evil Eye: he says that infelix Invidia knows how to hurt, then that grim Rhamnusia stands by with baleful gaze (attendit torvo tristis Rhamnusia vultu 73) as the boy attempts to reach his eighteenth year and that she causes the boy's muscles to fill out, adding brilliance to his eyes and raising his looks to an extraordinary degree of beauty (68-75); this done, the sight of the boy is torture for Rhamnusia and in her envy she embraces him in the folds of death and tears at his face with her hooked hands (76-8). Ausonius in Epistles 24, one of two poems to Paulinus of Nola dealing with a rift in their friendship, attributes their falling out to Rhamnusia's having been provoked by his excessive confidence in the solidity of the bond between them (40-44). At the same time he

speaks of Rhamnusia's being motivated by envy in the assaults she makes on friendship: she, as a foreign deity, should not trouble the sons of Romulus but should take herself to her Medes and Arabs and seek friends to attack in places where her envy and her corrosive

MALICE. ENVY AND INQUISITIVENESS

venom

13

will create hostilities in suitable breasts (50-55): illic quaere alios oppugnatura sodales, livor ubi iste tuus ferrugineumque venenum opportuna tuis inimicat pectora fucis (53-5)

When Sidonius Apollinaris explains why he feels greater affection for his wife's estate than for his own paternal property by asserting that he lives under God’s protection in a high state of concordia with his wife and her family, he qualifies that assertion by suggesting that it may arouse fear in his correspondent of the Evil Eye (haec mihi cum meis praesule deo, nisi quid tu fascinum verere, concordia 2.2.3). Unlike Statius and Ausonius, Sidonius identifies the force which threatens with the Evil Eye (fascinus) rather than with Rhamnusia or

Nemesis; but this makes little difference in practice. Rhamnusia is Often envious and endowed with a hostile gaze that blights, and there is no clear separation in Greek and Latin between Nemesis asa force which punishes boastful talk and as a jealous or envious being who resents excessive happiness. An epitaph from Odessos in Thrace of the second or third centuries AD ascribes the death of an eighteenyear old girl who had died unmarried and without having borne children to her having looked on φθονερὰ Νέμεσις (GV 1.982). Another epitaph from Thrace, this time for a pontarch from Tomis, complains that he was struck down by φθόνος and suffered νέμεσις (ταῦτα δ᾽ ἅπαντα ἀν[ύ]σας φθόνῳ πληγεὶς νεμεσήθ[ην] GV I.

1040.7).!! The fear that a happy or harmonious relationship between two persons may draw the Evil Eye of envy on itself is no mere literary conceit. Reitzenstein published parts of a long apotropaic text intended for use as an amulet containing an exorcism of the demon Βασκοσύνη 2 in which the demon includes the separation of husband and wife among the ills she is responsible for (ἄνδρας καὶ γυναῖκας dnoywpicat 298).'? Although the manuscript (Parisinus Graecus 2316) containing the amulet belongs to the fifteenth century,

much of the material in it is of much greater antiquity. Α manuscript of the seventeenth or eighteenth century published by Delatte contains virtually the same exorcism of the demon Βασκοσύνη, and it too credits Backocóvr with causing the separation of husband and wife (ἀνδρόγυνα ἀποχωρίσαι 248).'* Amongst the Byzantine texts published by Vassiliev there is a confutation of the Devil by Jesus that lists the separation of husband from wife (χωρισμὸς ἀνδροyóvov) as one of the works of the Devil (8). Other forms of magic-working besides the Evil Eye were also

14

MATTHEW

DICKIE

believed to have the power to disrupt or restore harmonious relationships. Philo Judaeus mentions witches who boast of being able through philtres and spells to create incurable enmities between those who love each other and to make those who hate each other into the best of friends (De Specialibus Legibus 2.101), and in the great Leiden magical papyrus there are two recipes for creating strife between

lovers or between

husband

and wife (PGM

XII.365-75,

455-64). The lapidaries tell of stones that are able to create ὁμοφροσύνη or concordia between husband and wife,'$ and the collection of magical recipes known as Cyranides recounts three wa ys of creating or restoring ὁμόνοια between husband and wife." Delatte and Vassiliev both publish spells to restore concord between husband and wife. The latter spell exorcises a demon called MovonovrjA — apparently the spirit that puts man and woman asunder to exist on their own — and invokes the Seal of Solomon to that end. It is impossible to tell whether such fears had by Catullus! time become a real part of the Roman world, or whether the threat posed

by the Evil Eye to his affair with Lesbia remained for him no more than a literary conceit. 3. Number magic Catullus' aim in 5 is to ensure, by confusing the addition, that neither

he nor Lesbia nor any malicious person knows the precise number of kisses he has received. In 7 he aims to extract so many kisses from Lesbia that no busybody will be able to make an accurate accounting of their kisses. Catullus thus fears two things: that he and Lesbia should know how many kisses they have exchanged; and that some unfriendly party should know that number. The source of the idea that knowing a number is dangerous, if it is large and is the number of something we possess or have enjoyed, is unknown. Ramminger cites Javeh's punishing David for the census of the Israelites (1 Chronicles 21),'? but classical antiquity seems to provide no parallel. Nor have classical parallels been adduced for the danger Catullus fears may ensue if a malevolent or inquisitive person learns how many kisses Lesbia and he have enjoyed. The closest

parallel I can find is the belief that knowing the numerical value of a god's name gives those who wish to summon him to help them in their magic-working a better chance of success, an idea found throughout the magical papyri.? Similarly, knowing how many

MALICE, ENVY AND INQUISITIVENESS

15

kisses have passed between Catullus and Lesbia presumably gives the malevolent and inquisitive a purchase on Catullus and Lesbia that will help them to bewitch the lovers. In ancient magic in general knowing something about someone is akin to possessing something that belongs to that person, and is thought to make any magic done against him the more effective. For example, a word that someone is remembered to have uttered, or a fragment from a fringe of a gown belonging to that same person, can be used in a love spell in combination with a word of one's own or a piece of one's own clothing to help win the love of that person (scholium to Euripides Hippolytus 514)?! The belief that knowing a number associated with a person may enable a hostile party to bewitch that person probably did not originate in Italy. To judge from Pindar, classical Greece knew nothing of the fear either: Pindar is conscious of and careful to avert the dangers that praising a man's victories may create, but he is not at all concerned about announcing the precise number of victories a man, his family or his clan have won in the games. In all probability this form of number magic is Egyptian or Babylonian, and was taken up by the Greeks of the East and perhaps made known to a wider audience through the medium of poetry. 4. Invidere and mala fascinare lingua It is generally assumed that invidere (5.12) refers to fascination by the

Evil Eye and that mala fascinare lingua (7.12) is a different form of magic-working, namely bewitchment by means of spells or imprecations.? It is more likely, however, that they refer to the same procedure: drawing the Evil Eye of some hostile force onto the good fortune of another by mentioning that good fortune. The principal reason for thinking that invidere refers to fascination by the Evil Eye and not just to envy is that Catullus speaks of it as something a malus may be able to do if he possesses certain information; it would be strange to speak of someone having the ability to envy another. There are at least two ways in which a person may be said to fascinate by the Evil Eye: he may cause harm by casting his own eyes on some being or object; or he may speak in praise of someone or something, and so draw the Evil Eye of another onto what is praised. Catullus probably has the latter kind of fascination in mind, since it is

difficult to see why someone would need to know a number if he

himself has the ability to cast the Evil Eye. Catullus is, accordingly,

16

MATTHEW

DICKIE

using invidere in an extended sense. In Greek βασκαίνειν is used in this way, as are fascinare and effascinare in Latin, probably under the influence of Greek usage. Catullus! use of invidere at 5.12 is therefcre likely to be a Grecism. The practice of fascinating by deliberately making mention of another's good fortune is first attested in Aeschylus’ Agamemnon. Clytemnestra praises Agamemnon in the most fulsome and exaggerated terms and then utters an apotropaic prayer against φθόνος (896—905); Agamemnon is not reassured (916f.). An audience would

probably interpret Clytemnestra's actions as an attempt to destroy Agamemnon and would discount her apotropaic prayer as insincere. Pliny (Natura! History 7.16) cites the paradoxographers Isigonus and Nymphodorus as the source of the story that in a part of Africa there are families of fascinators whose praise kills, causes flocks and herds to perish, trees to wither and infants to die (in eadem Africa familias

quasdam effascinantium Isigonus et

Nymphodorus, quorum laudatione

intereant probata, arescant arbores, emoriantur infantes). A secondary

development from the notion that praise may destroy is the paradoxical idea that looking admiringly at one's reflection may be dangerous. Theocritus has Polyphemus spit thrice into his breast to avoid being bewitched (ὡς μὴ βασκανθῶ) after he has admired his

reflection

in the sea (6.35-9).^

The

same

topos

was

used by

Euphorion (fr. 175 CA Ξ Plutarch Quaestiones convivales 682b). He told of Eutelides' bewitching himself after admiring his own beauty reflected in the eddies of a stream and of his falling ill and losing his good looks. The danger incurred by mention of one's good fortune is that it will arouse the attention of some hostile force, which will cast the Evil Eye on the object or person mentioned. The fear aroused by excessive praise may very often have had no clear focus, but in the Agamemnon it seems to be specifically the Evil Eye of the envious gods that Clytemnestra is trying to draw down on Agamemnon by her praise and then by her having him walk on crimson cloths. It is at any rate the god's eye of envy that Agamemnon, as he has his boots taken off, prays may not strike him from afar (ur) τις πρόσωθεν ὄμματος βάλοι

φθόνος 947).?5 Those who take mala fascinare lingua in Catullus to refer to bewitching by means of spells or imprecations regularly cite Vergil Eclogue 7.28: ne vati noceat mala lingua. In Eclogue 7 Thyrsis bids the Arcadian shepherds crown him with ivy as a rising poet, so that

Codrus’ loins may burst with envy;?é but if Codrus praises him

MALICE, ENVY AND INQUISITIVENESS

17

beyond due measure, he asks them to use another plant, baccar, so that the mala lingua may not harm the poet to be (25-8): pastores, hedera crescentem ornate poetam, Arcades, invidia rumpantur ut ilia Codro; aut, si ultra placitum laudarit, baccare frontem cingite, ne vati noceat mala lingua futuro.

Coleman distinguishes Codrus’ excessive praise from the mala lingua that may harm the rising Thyrsis; the latter expression "refers not to the evil eye but to a verbal spell or curse".?' But there are no grounds for this distinction. Vergil envisages two possibilities that may arise out of Thyrsis’ being crowned: 1) Codrus will burst with envy, but do nothing else; 2) Codrus in his envy will deliberately praise Thyrsis to excess, in which event Thyrsis should be crowned with an apotropaic

plant to avert the harm Codrus' malicious tongue may create.” Hence excessive praise and the mala lingua are the same danger, which is how Servius Auctus interprets the passage (on Eclogue 7.27): whatever is praised to excess is said to have the Evil Eye put on it: (quicquid autem ultra meritum laudatur, dicitur fascinari). He glosses mala in mala lingua by fascinatoria. Vergil Eclogue 7.28, therefore, provides no support for the thesis that mala fascinare lingua in Catullus refers to bewitching by means of imprecation. It suggests a quite different interpretation: drawing the Evil Eye on someone or something by praising him or it with malicious intent.?? There is in any case no clear-cut evidence that fascinare can be used of forms of bewitchment other than the Evil Eye. It seems then that invidere at 5.12 and mala fascinare lingua both signify drawing the Evil Eye on to someone or something by speaking in such a way as to attract the Evil Eye of some other being to the person or object mentioned. If so, how will knowing the precise number of kisses that have passed between Lesbia and Catullus enable a malicious or inquisitive person to bring the Evil Eye to bear?

That is, why is it more dangerous for someone to say that Catullus and Lesbia have exchanged 3089 kisses than it is for him to say they have exchanged an uncountably large number of kisses? There are no

parallels to shed light on the problem. We have perhaps a contamination of the belief that mention of good fortune may attract the Evil Eye and the belief (discussed in the previous section) that knowing some precise fact about someone affords in magic-working a special purchase on that person. Finally, the phrase mala lingua may be the rendering in Latin of what was probably a magical terminus technicus, γλῶττα πονηρά, a

18

MATTHEW

DICKIE

notion associated with βασκανία to the extent at any rate that in a magical papyrus it appears in a prayer to Hermes immediately after

Baoxocóvn:? διάσωσόν με πάντοτε εἰς τὸν αἰῶνα ἀπὸ φαρμάκων καὶ δολίων καὶ βασκοσύνης καὶ γλωττῶν πονηρῶν, ἀπὸ πάσης συνοχῆς, ἀπὸ παντὸς μίσους θ[ε]ῶν τε καὶ ἀνθρώπων. (PGM VIII.33-5)

5. The Evil Eye in Rome Fear of the Evil Eye in its various manifestations was a feature of Roman and Italian life long before Catullus’ time.?! This fear may have existed in Rome independently of Greck influence; but there are

several reasons for thinking that it came early to Rome and Italy through contact with Greeks, perhaps being grafted onto some native Italian superstition: 1) invidia in Latin is like φθόνος and

βασκανία in Greek in meaning both ‘envy’ and ‘the Evil Eye’; 2) fascinatio and fascinare were felt to be the same words as βασκανία and βασκαίνειν and were, probably in consequence, used with the same acceptation;? 3) praefiscine is used, as Charisius correctly

points out (4rs Grammatica 306.10 Barwick), in the same way as ἀβάσκαντα; 4) in both cultures the phallus is employed to ward off the Evil Eye.?? These coincidences suggest that belief in the Evil Eye reached Italy and Rome from Greece and that it was not an

independent growth. There is from an early date evidence in Latin of men's fearing that mention of good luck would draw the Evil Eye. In Plautus men who

have spoken in an unguarded way of their own or another's good fortune utter the word praefiscine to ward off the attentions of the Evil Eye.’* As we have already noted, Servius Auctus reports that what was praised excessively was said to be exposed to fascination by the Evil Eye. According to a scholium to Juvenal 7.112 (conspuiturque sinus), people spit thrice into their breasts because of verbal

fascination (propter fascinum verborum) and are thought by so doing to ward off the Evil Eye (videntur fascinum arcere). By fascinus verborum the commentator means boastful expressions or words of

praise likely to attract the attention of a hostile force endowed with the Evil Eye; the phrase suggests that this form of fascination is being distinguished from other forms. Finally, at the end of pagan antiquity, Symmachus, after reprimanding his correspondent Marinianus for even suggesting that he will repay Symmachus for the

winter clothing he has asked Symmachus to send him, says that he

MALICE, ENVY AND INQUISITIVENESS

will pass over a slight instance of fascination arising that has been praised; his debt to Marinianus for precludes his examining the intent of what he has 3.25.2: sed praetereo laudatae epistulae levem fascinum;

19

out of a letter his kindness said (Epistles neque enim me

tantae gratiae obligatio sermonum esse interpretem sinit).”° In effect,

Symmachus tells Marinianus that it was naughty of him to have exposed Symmachus to the slight danger of fascination by praising one of his letters. It goes without saying that Symmachus is not in earnest but has contrived a clever way of returning a compliment with a further compliment. 6.

Curiositas and Invidia

Itisthe curiosi whom Catullus fears may fascinate Lesbia and himself by the maja lingua. It is not at first sight clear to us why they should be supposed to threaten the well-being and felicity of a pair of lovers, but to a Greek or Roman the connection in thought would have been obvious: because busybodies are prone to envy they are likely to do what the envious do, cast the Evil Eye or speak maliciously of the good fortune of others so as to expose them to the Evil Eye.’

The idea that busybodies are envious is found or implied in a number of Greek sources. In an anonymous comic fragment someone who is addressed as the most envious of men is asked why he is so keen-eyed in espying the ills of others but so blear-eyed in seeing his own (τί τἀλλότριον, ἄνθρωπε βασκανώτατε,, κακὸν ὀξυδορκεῖς, τὸ δ᾽ ἴδιον παραβλέπεις; CAF Adesp. 359 [Kock [11.476] = Plutarch De Curiositate 515d). Philo in the De Abrahamo

says of base men that because they are inquisitive and busybodies (ἕνεκα πολυπράγμονος περιεργίας 20) they have their ears wide open to learn about other people's affairs, whether good or bad, in the former case to envy, in the latter to rejoice (ὡς αὐτίκα τοῖς μὲν φθονεῖν, ἐφ᾽ οἷς δὲ ἥδεσθαι 21); Philo attributes this state of affairs to the base man's being envious, hating what is fair and loving what is evil (βάσκανον yàp καὶ μισόκαλον καὶ φιλοπόνηρον 21). Plutarch gives ἃ less straightforward account of the relationship between πολυπραγμοσύνη and φθόνος in the De Curiositate. He says that πολυπραγμοσύνηῃ is a desire to learn about the ills of others and as such it is a disease not free from the taint of φθόνος and ἐπιχαιρεκακία (5154). He later explains that the association of πολυπραγμοσύνη with φθόνος is indirect: the busybody, inasmuch as he is eager to learn about the ills of others, is guilty of

20

MATTHEW

DICKIE

ἐπιχαιρεκακία, which is sister to φθόνος, as both are the offspring of κακοήθεια (518c).?’ Epictetus, discussing bad qualities to which men will not confess in the slightest, speaks in one breath about φθόνος and περιεργία as though they are linked (ἀκρατῆ οὐ ῥᾳδίως ὁμολογήσει τις, ἄδικον οὐδ᾽ ὅλως, φθονερὸν οὐ πάνυ ἢ περίεργον Dissertations 2.21.3). Finally, Libanius twice pairs πολυπραγμοσύνη with φθόνος (τί μοι βασκαίνεις; τί πολυπραγμονεῖς Declamation 3.85; πανταχοῦ τὰ ἐμὰ πολυπραγμονῶν μόνα, λέγοντι πολλάκις ἐβάσκαινεν Declamation 45.5). Such is the evidence for an association in the Greek mind of πολυπραγμοσύνη or nepıepyia with φθόνος or βασκανία. It is

possible that the Romans independently made the same connection, but more probably they took the idea over from the Greeks. The

translation and adaptation of Greek New Comedy may have been one way in which the idea became naturalised in a Roman setting. It plays a considerable róle in Plautus' Stichus. The parasite Gelasimus in his long auction-speech declares that there are many evil busybodies whose principal concern is with other people's affairs and who take no interest in their own (sed curiosi sunt hic complures mali,/ alienas res qui curant studio maxumo,/ quibus ipsis nullast res quam procurent sua 198-200). Such men are always eager to know the reasons for a man's auctioning his goods (201-4). He will not make

them suffer by delaying to explain his reasons for holding an auction, even though they deserve to suffer (205f.). He will announce the reason, so that they may rejoice in his loss, since there is no busybody who is not at the same time malevolent (dicam auctionis caussam, ut

damno gaudeant;/ nam curiosus nemo est quin sit malivolus 207f.). In the next act, when the parasite learns that his patron Epignomus has returned from the East laden with riches he renounces the idea of an auction and pronounces a curse on malevolent inquirers into auctions (384f.). He has a little later to eat his words when he is told that Epignomus has brought a crowd of parasites with him. The malevolent have now got, he says, a misfortune of his to delight them (iam meo malost, quod malevolentes gaudeant 394). What Plautus says here about the curiosi coincides with what Plutarch says about busybodies in the De Curiositate: busybodies take no note of their own affairs and, in particular, ignore their own personal ills (515d-516d); the delight they take in the misfortune of others (ἐπιχαιρεκακία) is fuelled by malevolence. It seems questionable then whether Leo, followed by Fraenkel, was right in supposing that the auction-scene was a purely Plautine invention.?? The most

MALICE, ENVY AND INQUISITIVENESS

21

recent editor of Stichus, Petersmann, has undermined much of Leo's argument; he is of the opinion that the auction-scene may have been in Menander’s Adelphoe.?? Plautus' use of the notion of curiositas supports this view. He is likely to have found the theme of roAurpaynocövn in Greek New Comedy, probably in Menander, where we have evidence of its presence; we may particularly note monostichos 653: πολυπραγμονεῖν τἀλλότρια μὴ βούλου κακά." The passage in Stichus is then evidence for a Roman playwright taking over from New Comedy the idea that noAunpäynoves are malevolent (κακοήθεις) and thus rejoice to hear of the ills of others (ἐπιχαιρεκακία). Admittedly, Plautus nowhere mentions envy; but there is little room for doubt that he is drawing on the same complex of ideas about πολυπραγμοσύνη, ἐπιχαιρεκακία, φθόνος and κακοήθεια that Plutarch discusses in the De Curiositate and Philo appeals to in the De Abrahamo. 7. The identity of the mali and curiosi We cannot be sure who Catullus! audience would have thought the

mali and curiosi of 5 and 7 were. There is clearly a temptation to identify them with the senes severiores whose gossip Catullus encourages Lesbia to ignore. There is a good deal of evidence, especially from New Comedy, that the stern older man, censorious of youthful follies, was an accepted stereotype. The figure recurs in elegy in the duri senes who find fault with convivia (Propertius 2.30.13). What is surprising about Catullus" senes severiores is that they deign to gossip, an activity hardly compatible with the dignified conduct to which an older man might be expected to aspire. Although it cannot be demonstrated, one suspects that the ideal older man will have had much in common with Aristotle's μεγαλόψυχος. The latter does not talk about others (Nicomachean Ethics 1125a5) and does not speak ill of others, even his enemies (112528),

since he is above such activities. A fondness for gossip is, however, characteristic of busybodies, as Plutarch emphasises: they have a prurient appetite for scandal, they go out of their way to find out about what is hidden and they are always eager for fresh and juicy titbits (De Curiositate 516d-e, 517e-f); they pay no attention to a description of a wedding, festival or procession but are all ears when the talk is of a maiden's seduction,

a wife's adultery, impending law-suits or quarrels between brothers (518a). As for busybodies' gossiping themselves, Plutarch says that

22

MATTHEW

DICKIE

garrulousness is a concomitant of a thirst for knowledge and that scandal-mongering (kakoAoyia) goes hand-in-hand with nepıepyia (51%). No evidence comes to mind from Greek or Roman literature for the association of old age with malicious inquisitiveness. But

Aristotle says in the Rhetoric that the old are wont to be jealous of the young, because the young aspire to possess what the old already have (1388a21-3). In his general characterisation of the old Aristotle, without actually saying that they have a propensity towards envy, does endow them with traits closely connected with envy, such as κακοήθεια (1389b 19-21) and μικροψυχία (1389b25f., 35f.). Catullus may, therefore, have conflated two different images of the

older man: as stern and unyielding censor, and as gossip. That Catullus! senes severiores do gossip supports the identification of them with the mali and the curiosi. Against this identification, however, is Catullus’ indifference to their gossiping in contrast to the measures he takes to foil the mali and curiosi. I suspect therefore that Catullus’ readers would have not have identified the curiosi with the senes severiores but would have taken it for granted that maliciously inquisitive persons were to be found everywhere. They will have

occupied the same réle in the popular imagination as the φθονερὸς γείτων. 8. Conclusion

Catullus 5 and 7 both deal with the theme of the felicity of love

provoking envy. It seems likely that Catullus took that theme from Hellenistic literature, along with the idea that the precise number of

kisses he and Lesbia had exchanged was information that could be used by hostile persons for the purpose of sorcery. Catullus perhaps also owes the notion of a mala lingua to his reading of Greek literature. What I have been concerned to show in this paper is how these Hellenistic themes were accessible to a Roman audience. Catullus can take it for granted that his readers will understand that to speak of the amount of happiness he and Lesbia have enjoyed may draw the Evil Eye of Envy, and that they will have no difficulty grasping the reason for his secking to keep curiosi from knowing the number of their kisses. These essentially Greek ideas have become so much part of the fabric of Roman culture that Catullus can take themes from Hellenistic literature that arise out of these ideas and use

them in his poetry confident that his audience will be able to follow his train of thought.*!

MALICE, ENVY AND INQUISITIVENESS

23

NOTES On the practice of writing variations on a theme cf. F. Cairns ‘Catullus’ basia poems' Mnem.

26 (1973) 18.

Besides the commentaries of W. Kroll and C.J. Fordyce, see A. Ramminger Motivgeschichtliche Studien zu Catulls Basiagedichten (Diss. Tübingen, Würzburg 1937); Cairns (n.1) 15-22; C. Segal ‘More Alexandrianism in Catullus VII?” Mnem. 27 (1974) 139-43; H.-P. Syndikus Catull: Eine Interpretation 1 (Impulse

der Forschung 46, Darmstadt 1984) 92-6, 99-104.

For Fortune begrudging lovers their kisses, cf. Ach. Tat. 5.7.9: ἀλλ᾽ ἐπεί pot τῶν ἐν τῷ προσώπῳ φιλημάτων ἐφθόνησεν ἡ Τύχη, φέρε σου καταφιλήσω τὴν σφαγήν. Also called Melissarion (2.14.15),

a name that Aristaenetus uses in 1.19.

Cf. the inscription on a herm, probabiy from the hippodrome of Septimius Severus in Byzantium, published in E. Mamboury and T. Wiegand Die Kaiserpaläste von Konstantinopel zwischen Hippodrom und Marmara-Meer (Berlin and Leipzig 1934) 49 and Abb. 26: θεοῖς ἀποτροπαίοις καὶ ἀλεξικάκοις. At Aristaen. 1.28.26f. a young man who had shared with a friend a courtesan who has turned against him prays the Evil Eye of Envy may not disrupt their friendship further (φθόνος δὲ μηδεὶς τῆς ἄλλης ἡμῶν ἐπικρατήσῃ yıkiac)and that the girl's change of heart may help his friend. See W. Kroll RE Suppl. S (1931) s.v. ‘Kuss’ 517; Pease on Aen. 4.659; Vian's supplementary note to Quint. Smyrn. 7.336-43 (Quintus de Smyrne, La Suite d'Homére II [Paris 1966] 213). Melissa's touching the walls and then kissing her fingers (kai τῶν τοίχων ἐφαπτομένη τοὺς δακτύλους ἐφίλουν) has its counter-

part in Medea's kissing her bed and the door of her bedchamber and touching its walls (kai τοίχων ἐπαφήσατο), before running off with Jason (A.R. 4.26f.). For Aristaenetus' debt to Hellenistic literature, see W. Schmid RE 2 (1896), s.v. Aristaenetos (8) 851f.; A. Lesky Aristaenetos, Erotische Briefe (Zurich 1951) 32-48; for Aristaenetus' debt to Menander, cf. W.G. Arnott ‘Aristaenetus and

Menander’s Dyskolos’ Hermes 96 (1968) 384; id. ‘Some passages in Aristaenetus' BICS 15 (1968) 119-24; for his debt to Callimachus cf. Pfeiffer's notes on Aetia II, frr. 80-83; for a general appreciation cf. W.G. Arnott ‘Pastiche, pleasantry, prudish eroticism: the letters of Aristaenetus"'

YCS 27 (1982) 291-320.

Noted by Lesky (n.8) 178, but not, to the best of my knowledge, by anyone whose

primary concern is with Catullus. Rufinus AP 5.22 (ὄμμα βάλοι δέ, μήποτ᾽ ἐφ᾽ ἡμετέραις

ἐλπίσι

βασκανίη,

5f.), a lover asking that his aspirations to be

enslaved by a woman should not be confounded by the Evil Eye, is not a parallel, since it is not a question of mutual love's having aroused envy. 10.

Kroll ad 5.12 cites Plaut. Most. 307 as an instance of a lover fearing that invidia may harm his good luck. But the wish Philolaches expresses there is that those who rejoice in his love may have goods of their own to rejoice in, while those who envy it may have no goods that they may be envied for (haec qui gaudent, gaudeant perpetuo suo semper bono;/ qui invident, ne umquam eorum quisquam invideat prosus commodis). This is plain and simple envy and has nothing to do with the Evil Eye. Nor is Plaut. Curc. 178-80, adduced by Ramminger (n.2) 59f.,

an instance of love threatened by the Evil Eye.

11.

L. Robert Les gladiateurs dans l'orient grec (Paris 1940) 102 cites as a parallel another epitaph from Tomis in which Dionysodorus laments that he has suffered the attentions of νέμεσις (νεμεσσήθῃη) in seeing the children who should have

been his heirs predecease him (AEM 6 [1882] no. 59, 29).

MATTHEW

24

12.

For Backavía or Βασκοσύνη

as demon

DICKIE

in pagan magic, cf. PGM IV.1400,

1450f.; A. Vassiliev Anecdota graeco-byzantina (Moscow

1893) 333; a Christian

magical papyrus speaks of τὸν δαίμονα προβασκανίας (PGM P9.10). 13.

R. Reitzenstein Poimandres: Studien zur griechisch-ägyptischen und frühchristlichen Literatur (Leipzig 1904) 291-303.

14.

A. Delatte Anecdota Atheniensia I: Textes grecs inédits relatifs à l'histoire des religions (Liege and Paris 1927) 228-63.

15.

Vassiliev (n.12).

16.

ἐλαφοκερατίτης: Orph. Lith. 259f.; Orph. Lith. Kerygm. 4.4 (Halleux-Schamp); [Hippocr.] Ἑρμηνεία περὶ ἐνεργῶν λίθων (de Mely-Ruelle); μάγνης: Orph. Lith. Kerygm.

11.10 (Halleux-Schamp); Damigeron-Evax 30.9-11 (Halleux-Schamp);

Vassiliev (n.12) 340; berillus: Damigeron-Evax 35.3 (Halleux-Schamp).

17.

1.2.20-6; 3.7.7-9, 22.1f. Kaimakis.

18.

Delatte (n.14) 620; Vassiliev (n.12) 340.

19.

Ramminger (n.2) 58 n.11. See also H. Lewy 'Dümonenglaube'" Arch. f. Religionswiss. 28 (1930) 250.

20.

Cf. PGM 1.325, 11.125, V.435, XII.465. I am grateful to Professor H.-D. Betz for

21.

I am indebted to David Jordan for this example.

22.

this suggestion.

So Kroll ad loc.; Ramminger (n.2) 59-61; Fordyce ad loc.; C. Segal ‘Catullus 5 and 7: a study in complementaries' AJP 89 (1968) 297f.

23.

Cf. Pl. NH 7.16; Sch. in Juv. 7.112; Serv. auct. ad Verg. Ecl. 7.27.

24.

D. Gershenson 'Averting βασκανία in Theocritus: a compliment’ CSCA 2 (1969) 145-50 thinks that Polyphemus cannot be afraid of putting the Evil Eye on himself, because Polyphemus is not envious of himself. But the Evil Eye is not necessarily a manifestation of envy, although it usually is. The Colchians avoid Medea's gaze as she drives through the town (A.R. 3.885f.), even though there

can be no question of her envying them. Her eyes are dangerous in themselves and

can

be used

to bewitch

those whom

she does not envy such as Talos

(4.1665-88). Cf. M. Dickie PLLS 6 (1990) 267-96. 25.

For divine jealousy blighting with the eye, cf. Aesch. Ag. 468-74; Pi. P. 8.71f.

26.

Bursting with envy is a common figure of Greek origin; ῥηγνύμενος φθόνῳ: Phil. Iud. 7n Flacc. 29; Dio Prus. 43.2; Luc. Tim. 40; Ael. Arist. 50.69 Keil; Lib. Or.

1.207; 29.13; Decl. 29.28; Eunap. V.S. 6.2.3; invidia rumpor: Ov. Rem. Am. 389, Her. 15.223; Mart. 9.97. There is a certain coarseness to Thyrsis' ilia rumpantur; cf. Serv. ad 7.26: est autem hoc dictum per amaritudinem rusticam.

27.

R. Coleman Vergil, Eclogues (Cambridge 1977) ad loc.

28.

The identity of baccar is a mystery. Servius Auct. says it is a plant that wards off the Evil Eye (Aerba est ad depellendum fascinum). Coleman ad loc. thinks this may be an inference from the context and that Thyrsis may prefer baccar because less

glory attaches to it than ivy. If it is an inference, it is surely a correct inference. 29.

So also A. Ronconi 'Malum carmen e malus poeta', in Synteleia Arangio Ruiz (Naples 1964) 960 — Filologia e linguistica (Rome 1968) 129.

MALICE, ENVY AND INQUISITIVENESS

30.

31.

25

Vassiliev (n.12) 344f. and Delatte (n.14) 623. publish a spell that binds τὴν

κακὴν βλέψιν καὶ τὰ ἕτερα, xai τὰ πονηρὰ καὶ δόλια αὐτῶν στόματα. The spell goes on to pray: αἱ πονηραὶ καὶ δόλιαι αὐτῶν γλῶσσαι καὶ τὰ πονηρὰ αὐτῶν στόματα παυσάσθων καὶ μὴ λαλείτωσαν κατ᾽ ἐμοῦ τοῦ δούλου τοῦ Θεοῦ κακά. Cf. the use of the expression praefiscine(i) after words of praise have been uttered: Plaut. Asin. 491-93; Rud. 459-62; Afran. com. 36; Titin. com. 110; and the use of the phallus, presumably to ward off the Evil Eye: Lucil. frr. 78, 959 Marx.

32.

On the ancient etymologies of fascino, -are, R. Maltby A Lexicon of Ancient Latin Etymologies (ARCA 25, Leeds 1991) s.v. On theimpossibility of fascinus being the

loan-word βάσκανος, P. Chantraine Dictionnaire étymologique de la langue grecque (Paris 1968-1980) s.v. βάσκανος: A. Walde and J.B. Hofmann Lateinisches etymologisches Wörterbuch 1-11 (Heidelberg 1938-1954) s.v. fascinum

think

it may

be a loan-word; A. Ernout and A. Meillet Dictionnaire

étymologique de la langue latine* (Paris 1967) s.v. fascinus posit a common origin with βάσκανος in Thraco-Illyrian.

33.

Early evidence that pre-dates anything known from Italy for the phallus used to ward off the Evil Eye is: 1)a terracotta plaque of the 6th century BC found to the

west of the Potters' Quarter of Corinth showing an ithyphallic steatopygous figure standing over the fire-door of a kiln (Berlin, Staatliche Museen, Stiftung preussischer Kulturbesitz F683/757/829/822. AD 11 39 no.12; illustrated in Antike Werkstattbilder [Berlin 1982] 31 Abb. 14); 2) a boustrophedon inscription of the 6th century BC from the fortification wall of Itanos in Crete with a prayer for the town's well-being; to the left of the inscription there is inscribed a hand with the fingers outstretched and a phallus (F. Dümmler “Inschrift aus Itanos' Ath. Mitt. 16 [1891] 127-9).

34.

See n.31 above.

35.

J.-P. Callu Symmaque, Lettres Il (Paris 1982) 38 renders this: “mais je passe sur ce léger maléfice dans une lettre digne d'éloge. Et de fait, l'obligation que je vous ai d'un si grand bienfait ne me permet pas de commenter vos propos." I do not see how the Latin can be made to yield such a sense.

36.

Discussions of curiositas in Latin tend to concentrate on the notion of pointless intellectual endeavour and ignore the more mundane aspects of being a busybody;

see J. Mette

‘Curiositas’, Festschrift Bruno

Snell (Munich

1960)

227-35; A. Labhardt ‘Curiositas: Notes sur l'histoire d'un mot et d'une notion’ MH 17 (1960) 206-24; R. Joly 'Curiositas' Ant. Cl. 30 (1961) 33-44; P.G. Walsh ‘The rights and wrongs of curiosity (Plutarch to Augustine)’ GR 35 (1988) 73-85. I am indebted to Stephen Ryle for bibliography on this point.

For ἐπιχαιρεκακία and φθόνος as the offspring of κακοήθεια, cf. De Malign. Herod. 858b. F. Leo ‘Uber den Stichus des Plautus’ NGG (1902) 379-82; E. Fraenkel Elementi Plautini in Plauto (Florence 1960) 101.

H. Petersmann 7. Maccius Plautus, Stichus (Heidelberg 1973) 31f. See also Epitrep. fr. 2 Sandbach; Sam. 298-300, 573-76, 655; Perikeir. 374.

41.

I am grateful to D.R. Jordan and J.C. McKeown for their comments on an earlier version of this paper.

26

MATTHEW

DICKIE

ADDENDUM It is worth mentioning that two points made in this paper had already been made in the Renaissance by Leonardus Vairus, the Prior of the Benedictine

Abbey of St Sophia in Benevento, in his De Fascino Tres Libri: 1) invidere at Catullus 5.12 has the same meaning as fascinare at Catullus 7.12; 2) those

who know the number of something are able to fascinate it. The De Fascino was first published in Paris in 1583 (apud N. Chesneau) and was immediately

translated into French. It was re-published in 1589 by Aldus in Venice. 1 cite Vairus from the Aldine edition (314):

Nam sicut sacris scriptoribus consuetum ac familiare est, fascinare pro invidere uti; itaque contra profani invidere pro fascinare assumunt, cum fascinandi libido ab invidia oriri soleat. Unde Catullus: Conturbabimus illa, ne sciamus,

aut ne quis malus invidere possit. Quod autem invidere pro fascinare Poeta intellexerit, ex alterius epigrammatis versibus in eodem sensu compositis colligimus, ubi sic ait: Possit nec mala fascinare lingua Quin nec pernumerare curiosi [sic].

Ex enim rebus fascinatio nocere non posse credebatur, quarum numerus ignoraretur: sicut nostrae tempestatis mulierculae in superstitiosis rebus exercitate non ignorant.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 27-9 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

A! AT CATULLUS 68.85 ALLAN

KERSHAW

Like the Augustan elegists, Catullus uses the particle a/ in familiar patterns, even in non-elegiac meters: with miser (15.17; 61.132; 63.61;

64.71), and with nimis (60.5). My main purpose is to propose the removal of a/ from Catullus 68.85, but, for the sake of completeness,

I begin with a few observations on the other passages where a/ appears.!

64.135 immemor a! devota domum periuria portas?

This is an uncommon usage which probably influenced Virgil's immemor heu (Georgics 4.491), and Ovid's heu devota domus (Heroides 9.153).

64.178 Idaeosne petam montes? a, gurgite lato discernens ponti truculentum ubi dividit aequor? ah B. Guarinus at Muretus: a V

The particle here has been put to rest, along with ubi (179), by Mynors, Fordyce, and Goold in their respective editions; it has now been revived by Lee.? Such a purely parenthetical use would be, to say the least, most unusual.? It is *defended' by Kroll, ad loc., on the grounds that ‘‘a’ liebt C. sehr".^ Finally, 66.85, where the context needs to be given: nunc vos, optato quas iunxit lumine taeda, non prius unanimis corpora coniugibus

80 27

28

ALLAN

KERSHAW

tradite nudantes reiecta veste papillas, quam iucunda mihi munera libet onyx, vester onyx, casto colitis quae iura cubili. sed quae se impuro dedit adulterio,

illius a mala dona levis bibat irrita pulvis: namque ego ab indignis praemia nulla peto.

85

85 illius amala levis b.d.i.p. V: corr. edd.

The words make sense and have been accepted without question, but I doubt their Latinity and that Catullus wrote them.? What he did write, I suggest, was aemula dona.® The Lock is warning the adultress that her gifts, though they equal those of the virgin bride, will not be welcome, nor will they legitimise her participation in the ritual." True, when aemulus does not qualify a person, there is usually a genitive or dative to fill out the sense, as in Martial 8.15: sed licet haec primis nivibus sint aemula dona. In the present passage, however, the adjective more properly describes the person giving the gifts: the adultress attempting to equate herself to the bride — cf. 86, where the concern is clearly about the giver rather than the things given. Indeed I wonder whether /evis is not really genitive with i//ius rather than nominative with pulvis as all commentators suppose; this adjective perfectly describes the adultress, but when applied to dust, earth, and the like it is usually in a funereal context signifying the benevolence of the speaker. A final point of curious interest: the first words of Maximianus are aemula quid cessas finem properare senectus? (1.1). In his edition of that poet (Poetae Latini Minores) Baehrens suggested for aemula the reading a, mala!

NOTES l.

For the strict patterns observed in the usage of this particle, I refer the reader to my notes ‘Emendation and usage: two readings of Propertius’ CPh. 75 (1980)

71f., and ‘A! and the elegists: more observations’ CPA. 78 (1983) 232f.

2.

G. Lee The Poems of Catullus (Oxford

1990); C.J. Fordyce Catullus (Oxford

1961) 300: '*V's a... ubi gives no satisfactory sense; at, as often, introduces the speaker's own objection." Lee also accepts Camps' a, cette huc at 55.9.

3.

This particle should arouse suspicion when it appears without one of the words by which it is usually accompanied: cf. Culex 1.373: cogor adire lacus viduos, a, lumine Phoebi, which is equally unconvincing and which I discuss in CQ 42 (1992) 566f.

4.

W. Kroll, Catull (Stuttgart 1980) 168. Cf. H.A.J. Munro, Criticisms and Elucidations of Catullus (New York 1938) 186: "Catullus is fond of af.’ So saying, he proposed quare cur te iam a! amplius excrucies? (76.10).

A! AT CATULLUS 68.85

29

5.

Nor does it seem likely that Callimachus wrote ἀ: cf. his usage of the particle: á πάντ]ος (Aet. 1.33); & μή pe ποιήσῃς (amb. 195.30); à πάντη πάντα (fr. 736. 1); & δειλὸς βασιλέων (Hymn 3.255).

6.

On this passage R. Ellis A Commentary on Catullus (repr. New York 1979) 380 cites Prop. 2.16.46, where the wish is similar (i.e. ‘may gifts be carried off by . Storms’), and where the gifts are those of a rival.

7.

Cf. nowadays those rumores severiorum which avouch that a bride is wearing a white dress sub specie virginitatis.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 31-45 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

CATULLUS

79: PERSONAL INVECTIVE OR

POLITICAL DISCOURSE? W.

JEFFREY TATUM

Lesbius est pulcher. quid ni? quem Lesbia malit quam te cum tota gente, Catulle, tua. sed tamen hic pulcher vendat cum gente Catullum, si tria notorum suavia reppererit.

The biographical and prosopographical significance of Poem 79 is too well known to require rehearsal. The literary dimensions of the poem, or whether the brief epigram even possesses literary qualities worth discussing, are matters which have yielded a bibliography of considerably lighter heft. So far as Iam aware, only H.D. Rankin and M.B. Skinner have deemed the poem worthy of extensive literarycritical scrutiny, although with rather different results.! For Rankin,

the poem signals “a real sense of injury" inflicted upon the poet by Clodia's

incest with

her brother (a charge, incidentally, assumed

without argument to be historically accurate) and explains the poet's series of attacks on the incest of Gellius, who was also a lover of Lesbia, an indiscretion

that the poet viewed as a form of incest,

Gellius and Catullus having been on terms intimate enough to render the poet's girlfriend sufficiently forbidden fruit. The poet, in Rankin's view, could not openly attack Lesbia's incest, so, in essence,

the poet vented his spleen on another perverse noble.? Skinner's detailed and scrupulously researched study is more ambitious: in her view, Catullus’ epigram, by employing several, mostly sexual, invective tropes, represents a denunciation not merely of Clodius' (and Clodia's) values, but also “the widespread moral corruption

within contemporary

Roman

society," an argument that in large 31

32

measure

W. JEFFREY TATUM

depends

on Skinner’s unrelenting insistence that sexual

description encodes political messages.’ It is the purpose of this paper to try a tack somewhat different from that of Rankin or Skinner. Unfortunately, this involves at least some polemic, which I hope will be taken as an indication not of disregard but rather of the respect in which I hold the scholarship of my predecessors. Furthermore, since gainsaying is the easier if less

pleasant aspect of scholarship,

I hope

to make

some

positive

suggestions for the proper interpretation of Poem 79. I shall argue that the delicacy of the poet's sexual invective in this epigram is not without significance: given the rude forcefulness of Catullus' sexual satire, especially on the subjects of incest and os impurum, I believe that the comparative discretion of this poem, while it does not eliminate the element of salacious wit for those equipped to perceive it, ought to direct our attentions at least equally and perhaps

primarily to the social implications of the poem's criticism. For the poem reveals motifs other, and more obvious, than merely sexual ones: patrician exclusivity, the interplay between exclusivity and isolation, and the potential antagonism between the ‘Lesbii’ and Catullus! gens. Herein, I shall further suggest, may well lie a particular political criticism, but one of a considerably less comprehensive nature than those apprehended by Skinner. Whatever critical orientation one inclines to adopt in these days of abundant theoretical alternatives, it remains virtually impossible to ascribe to Poem 79 anything like literary autonomy. Lesbius est pulcher relies entirely on intertextuality and historical contextualisation if it is to mean anything at all: Lesbius is a masculine name wholly predicated on Lesbia, and only because we know that Lesbia is the poetical manifestation of Clodia (Apuleius Apologia 10) can we see, with greater confidence than ordinarily attends Catullan

prosopography, that Lesbius est pulcher is a punning allusion to P. Clodius Pulcher (tr. pl. 58).* Thus we come to an epigram whose interpretation the poet has firmly embedded in both his oeuvre and his milieu. It is a situation to bring out the historicist in any but the most determined to resist. However, this embedding ought not to detract from the purely artistic merits of the poem, whether or not we ultimately categorise Poem 79 as a gem or a squib. Catullan equivocation, as ever, lends

complexity to what is ostensibly a straightforward insult. From the opening word, whose strange familiarity — we recognise the stem but are surprised by the termination — promises intricateness and

CATULLUS 79: PERSONAL

INVECTIVE OR POLITICAL DISCOURSE?

33

contrariety, we are obliged to read closely. It may be necessary to remind ourselves that the opening pun was probably felt to be cleverer at the time of the poem’s appearance (we may compare Cicero’s gleeful use of the same game) than when it is one item of many in a Catullan quaestio.? Once the identification of Lesbius has been made, it is immediately vouched for by Lesbia's explicit presence in the poem, at which moment a new issue emerges in Lesbia's preference for her brother, deriving from and reinforcing his status as pulcher, over not merely Catullus, who is the addressee of

the poem, but the entire gens Valeria as well. Significantly, whether metri causa or not, Catullus' nomen gentilicium goes unmentioned. In the opening couplet Catullus is addressed familiarly, even friendlily, by his cognomen; at the same time, it is the pun on Lesbius' cognomen that accentuates the nobilitas of Lesbius and Lesbia.$ And it is in this aristocratic atmosphere of Claudian exclusivity that incest is alluded to: Catullus is estranged from Lesbia both sexually and socially. Now while the whiff of incest is undeniable, for the rumours attending Clodius were ubiquitous, the language of the poem clearly

emphasises the poet's social exclusion; for if the poem's malit must be understood exclusively or even principally in sexual terms, then the implications of the linkage of Catullus with his entire clan as the object of Lesbia's (sexual) rejection become flabbergasting to say the least. In a sense, the speaker of the poem is Catullus' only friend in this unhappy couplet, a circumstance made more pathetic yet if we take line two to be self-apostrophe.’ The strong adversative sed tamen conveys a shift. Lesbius is now hic pulcher (= hic Pulcher, the familiar cognomen rather than the formidable gentile name, Lesbius = Clodius). But Pulcher, the

affection of his sister and the eminence of his gens notwithstanding, cannot lay claim to even a trivial number of acquaintances. And just as the first couplet implied incest with discreet indirectness, so here the poem, by representing Pulcher's friendless condition through his failure to secure even three kisses, hints at Pulcher's os impurum, or so it is ordinarily and naturally assumed.* Whatever the sexual

implications of this line, a question to which we shall return, the mere failure to obtain friendly greetings is clearly contrary to any claim to be pulcher. Thus the second couplet inverts the circumstances of the first. Lesbius' pulchritude and popularity, asseverated in the opening lines, are denied in the closing ones; the reversal of his fortunes derives from the transformation of the poem's frame of reference: in the first couplet, what mattered most, even to the speaker of the

34

poem,

W. JEFFREY TATUM

was the exclusive world of the gens Lesbia; in the second,

Pulcher has been thrust into the world of public estimation, in which the tastes of noti not cognati count. The problem of line three (or, to put it another way, precisely how to construe the verbs vendat and reppererit) remains to be considered. Before doing so, however, I should like to reconsider the extent to

which Poem 79 is a poem about incest, as Rankin by and large takes it to be. Clodius was smothered in this infamy during his trial in 61; Cicero found the accusation to be no less effective a weapon at Caelius' trial in 56.? Indeed, in view of Clodius' ultimate reputation for engaging in incestuous relations with his brothers as well as with his three sisters,!° we should scarcely wonder that the reproach rears its ugly head in our poem. Nor should we forget that, in Catullus' day, Clodius' affair with his sister was a familiar subject for obscene

versification.!! One hesitates to speak of generic expectations for Clodian ridicule, but perhaps one may fairly say that a nod in the direction of incest was, like the pun on pulcher, felt by the poet to be natural to anti-Clodian rhetoric and crucial to the proper identification of Lesbius. Incest was by no means an uncommon slur in the late republic: even Cicero's reputation was sullied by the accusation.'? That incest was thought of as a distinctly aristocratic vice, as Rankin asserts," is neither likely nor necessary. If anything, the Romans will have associated the practice with eastern cultures: it pervades Greek mythology, and even in Greek rhetoric the practice is linked to the yet more eastern cultures of Egypt and Persia.'* Consequently, Ph. Moreau, who accepts Shackleton Bailey's likely conclusion that Clodius and Clodia Metelli were children of different mothers,'> has

speculated that their incestuous relationship, in which he tends to believe, was an act of philhellenic snobbery.!6 However that may be, there is ample literary evidence, even in Catullus himself, that incest was by no means uniquely regal or patrician.!’ Which is not to say that incest did not convey an impression of exclusivity. Sociologists inform us that incest taboos exist not so much to prevent brothers from marrying sisters as to require that sisters be given in marriage to others; in other words, from a sociological perspective at least, incest can represent a denial of exogamy, of openness, and therefore is inherently exclusive.'* Of course it is difficult to translate sociological theory into historically particularised social or literary perceptions obtaining in late first-century Rome. Still, in the context of Catullus' poem, the sociologists' conclusions are suggestive of the use to which

CATULLUS 79: PERSONAL

INVECTIVE OR POLITICAL DISCOURSE?

35

the poet has put the incest motif. For Rankin has rightly observed

Catullus’ disapproving fascination with incest and he is very likely correct in detecting a connection between Catullus’ hostile treatment of incestuous practices and his inclination to represent his love for Lesbia in familial terms.'? Catullus, in an unforgettable if puzzling passage, described his love for Lesbia in terms of the mutual respect

and affection felt between fathers-in-law and sons-in-law.?? The clash between this image and Lesbia's preference for Lesbius in line one of Poem 79 invites us to see in our poem not merely another dirty joke at Clodius' expense, though of course it is that, but also a prelude to the

isolation of Catullus which is so patent in line two of the poem. In brief, then, Poem

79 is not primarily about Lesbia's incest; rather,

incest is a topos employed to further the identification of Lesbius by

applying the standard insult and to reinforce the themes of Claudian exclusivity and Catullan isolation which we have already observed in our initial reading of the poem. Let us turn now to the syntax of vendat and reppererit in the second couplet. The usual construal of these lines takes vendat as a jussive subjunctive and reppererit asa future perfect indicative, thus yielding a future more vivid construction but one whose protasis is clearly understood to be impossible.?! However, K. Quinn, it appears, and Skinner, most definitely, read the Latin differently: reppererit may also be perfect subjunctive,

in which instance the construction

is

future less rather than future more vivid.?? The second couplet, read this way, has less bravado on Catullus' part and more menace on Lesbius’. If, remote possibility though it is, Lesbius should obtain tria notorum suavia, then he would proceed to put the Valerian clan up for sale. Syntactical ambiguity suits Catullan technique, and Skinner is no doubt sound in accepting the ambiguity posed by these verbs, although I think the usual construction is more natural than Quinn's and Skinner’s.?? I mention each construal because both have in common the potential merchandising of Catullus' family and the blurring of Claudian exclusivity with Lesbius' isolation — and these aspects of the final couplet will be the most relevant to our subsequent discussion of the poem. It is time to talk about the unattainable kisses. It is universally assumed that the concluding line of this poem, by insinuating that Lesbius cannot in fact obtain even a trivial number of kisses, a reference to the ordinary Roman habit of greeting one's friends with

a kiss, implies that Lesbius suffers from os impurum.?* Now in view of the frequency (and occasional obliqueness) with which Greek and

36

W. JEFFREY

TATUM

Latin epigram alike mention the need to avoid the kiss of the fellator

or cunnilingus, this assumption is hardly unreasonable.” Still, given Catullus' extensive and explicit use of obscenity, including the motif of os impurum, I rather think that the point of our poem's last line lies

elsewhere than in Lesbius' oral indulgements.” This is not to say that I would deny the obscene reference in the poem's last line, but nor do I believe that sexual invective is the end of the matter. Just as the topos of incest was earlier employed to emphasise the social dimensions of the epigram by underlining the themes of exclusion and status, so here os impurum stresses the reversal of Lesbius’ condition from the first to the second couplet: híc Pulcher has no amici. At this point, before attempting to probe further the possible

significance of Lesbius' friendless condition, it may be suitable to consider briefly Marilyn Skinner's general interpretation of this poem, if only to distinguish our two approaches. Skinner carefully searches out every political (and moral) criticism she can discern in Poem 79, and from her perspective there are quite a number to be found. The web of associations she perceives — and it is such that she interprets the poem as “the cynical pronouncement of a man who believes, rightly or wrongly, that the moral structure of his world is falling apart"? — is predicated on her assumption that topoi of sexual invective in Catullus are regularly metaphors of political corruption as well as on her acceptance of the thesis of R. Reitzenstein and D.O. Ross that Catullus" use of the language of friendship, i.e. terminology like fides, pietas, officium, gratia and amicitia, derives from Roman party politics.?* Treating these two points Aysteron proteron, let me begin by recalling the observations of R.O.A.M. Lyne, who has argued convincingly what perhaps ought not to have required argumentation, that Catullus has adapted to his love-affair the “emotive

language of aristocratic obligation.'"?? Put in different terms, Catullus has delineated his relationship with Lesbia not along the lines of merely political alliances, but along the lines of the aristocratic conception of moral commitment — the same conception that informs Roman thinking about public friendship and, consequently, political expression. This understanding not only of Catullus'

language but of amicitia itself conforms much more closely than that of Ross or Skinner to P.A. Brunt's magisterial discussion of Roman

friendship, in which he points out that amicitiae should not simply be identified with political alliances but involve a good deal more,

CATULLUS

79: PERSONAL

including

affection

INVECTIVE OR POLITICAL DISCOURSE?

and

principles held

in common.’

37

Another

objection, though one less obviously damaging, to the “political vocabulary" approach is that the thesis depends too heavily on the important but problematic work of B. Axelson. While not without merit, Axelson’s idea of ‘‘prosaic” vocabulary, at least when it is applied too mechanically, is open to exception, as readers of Tradition and Originality in Roman Poetry will know well.?! The efforts of Reitzenstein and Ross to confine terms like fides to a narrow notional sphere are, I think, overly schematic, as Lyne's comments demonstrate. And finally, and I make no apologies for the inevitable subjectivity of this assertion, Lyne's conclusions are simply more sensitive to Catullus' evident interest in the literary construction of a morally elevated, if ultimately unsuccessful, loveaffair.? All of which ought to give one pause before immediately transforming Catullus' ostensibly personal objections into a political manifesto, though it would of course be equally wrong to think that Catullus is never interested in his political environment.” Skinner's treatment of sexual invective as a political metaphor is also, in my view, overly rigid and unpersuasive. When one considers that personal abuse suffused political discourse in Rome, it hardly surprises that sexual revilement was plentiful and striking, or that it often expresses not literal truth but political rivalry and opposition.*4 But this observation is different from the reductionist conclusion that sexual invective invariably encodes political messages, that execration should be read foremost as metaphor. It is to someextent because political thinking in Rome focused on individual character — on virtus — rather than on plans and platforms that ad hominem vituperation played not a more significant but perhaps a more conspicuous role in political rhetoric than did statesmanlike reasoning, a circumstance that could, but should not, lend plausibility to

Skinner's

interpretative

method.

As ancient

rhetorical theory

makes clear, the purpose of invective was to blacken character, a

strategy that undermined an opponent's credibility even while it

humiliated him.’ Viewed in this light, aspersions of sexual misconduct and threats of sexual degradation are better apprehended by

stressing their literal (if admittedly fictive) content." A Roman politician attacked his rivals as perverts not because they actually were depraved but because he hoped to persuade someone that they were actually depraved and consequently unworthy of credence or loyalty or honour. Such rhetoric was a supplement to, but not a

substitute for, genuine political argument.** I take it that in Poem 79

38

W. JEFFREY TATUM

Catullus employs sexual invective similarly. However vendat is ultimately taken, the possibility arises of Lesbius' molesting Catullus and his family, of his robbing them of their very status. This suggestion, unless the poet chose it with utter caprice, may well give us further insight into the nature of the opposition obtaining between Catullus and Lesbius, something that goes beyond Lesbius' snobbishness or their rivalry for Lesbia's affections. The threat which Lesbius at least hypothetically poses to the Valerian gens finds something of a parallel in Cicero's De Domo, a

speech delivered before the pontiffs in 57 BC, much admired by its author and, consequently, quickly published for a wider, if nonetheless elite, audience.*? In this oration, well known for its attack on Clodius and his associates, Cicero makes the claim that Clodius' law de exsilio Ciceronis was tantamount to proscription inasmuch as the tribune was thereby able to deprive an ex-consul of his citizenship

and

his property.

In order to render

Clodius'

actions

more

frightening, the orator goes on to assert that, if a Roman of Cicero's stature, who was protected by honos, by dignitas, in short, by political splendour, could be so disgraced and so disgracefully plundered by a corrupt patrician-turned-tribune, then there could be little hope for a member of the non-political elite. Indeed, Cicero foresees whole colleges of tribunes cooperating with the greedy nobility to despoil the property of the wealthiest boni — especially when the booty could

be earmarked for a popularis cause.*! Cicero's prophecy has recognisable antecedents. The danger posed by a depraved patrician, a noble-gone-bad, especially one susceptible to being labelled as a popularis, belongs in general to the rhetoric of anti-popularis sentiment and, more specifically for Cicero, to the rhetoric attending the Catilinarian conspiracy.? Nor was the De Domo the first time when Cicero assimilated Clodius to his partisan characterisation of Catiline: the likeness had been drawn, fairly or

not, as early as the Bona Dea scandal.* With respect to the /ex de exsilio Ciceronis, however, Cicero could make a persuasive case. This notorious law was suspicious for its character asa privilegium and its retrospective aspect. The details of the law, recently discussed by Ph. Moreau, required that Cicero's property be confiscated and sold, with Clodius himself put in charge of the publicatio bonorum instead of, as was customary, entrusting the urban praetor and quaestors with the responsibility.** Moreover, Cicero's home on the Palatine and his villas (at least the villa in Tusculum) were slated by the law for destruction, provision having been made for designating the Palatine

CATULLUS 79: PERSONAL

site for

INVECTIVE OR POLITICAL DISCOURSE?

39

a monument to Libertas. Again, the law apparently placed

this public business also solely in the tribune’s hands.“ All of which allowed Cicero, in the aftermath of his recall, to claim that Clodius

was not serving the commonwealth when he inflicted punishment on the orator's estate, but rather that Clodius had for personal reasons seized his property under the pretence of popularis legislation. All boni,the orator maintained, had much to fear from such a man — in

particular the comparatively vulnerable members of the wealthy classes.^$ The threat described so effectively in Cicero's oration seems too similar to the one implied by line three of Catullus' poem to be unrelated. And one ought to recall that when Clodius' law was ultimately superseded by the vote of the centuriate assembly, that vote had followed from a senatorial division in which Clodius had stood absolutely alone against the pro-Ciceronian forces marshalled by the consul Lentulus Spinther, but especially by Pompey the Great. Of 417 senators, only Clodius voted against Cicero's recall.*’ This isolation continued even after Cicero's recall, at least with respect to the senate's disposition regarding Cicero's domus. When the full senate met on 1 October 57 to restore Cicero's property in response to the pontiffs' decision in the orator's favour, only the threat of a tribune's veto postponed for a single night the decree's passage omnibus praeter unum adsentientibus (Cicero ad Atticum 4.2.4), and

Sex. Serranus, the troublesome tribune, was immediately intimidated by the overwhelming unanimity of the senate — the delay was ultimately a face-saving measure for young Serranus.* When Catullus represents his rival as isolated in the world of noti, there exist in the background particular historical moments, tied tightly to a single thread of events, when the poet's description was true — and one finds it difficult to believe that Clodius' enemies in 57 and subsequently did not relish and recount the crushing senatorial defeat of the formidable popularis, as Cicero most certainly did.*? As a member of the non-political elite whom Cicero proclaimed to be most in danger from the likes of Clodius Pulcher, Catullus had ample reason to allude to the anti-Clodian propaganda advanced by

Cicero.? Clodius' humiliating defeat in the matter of Cicero's recall and his subsequent house (or, to put it shrine which he had force to the poem's

failure to prevent the restoration of Cicero's another way, to prevent the destruction of the dedicated and constructed)?! lend considerable assertion of Lesbius' social isolation. Poem 79

denounces and insults Lesbius, in terms both sexual and social, even

40

W. JEFFREY TATUM

while it provides some explanation for the attack: personal rivalry and political fear. If this seems overly precise or overly historicist, it must be remembered that it is the poet who has forced his readers, from the opening word of the poem, to employ just such a strategy in

reading.?? That the interplay between fiction and reality constitutes a dimension vital to the proper appreciation of Latin poetry generally and of Catullan poetry specifically — especially Catullan parody — has been demonstrated often enough and eloquently.?? Poem 79, however much or little its success at provoking laughter or derision or pain, certainly belongs in the category of Catullan parody, and we find in its miniature operation the application of complaints both personal (in so far as the literary relationship between the poet and Lesbia can be deemed personal) and public. Yet these complaints find expression in a way that bases their ostensibly personal facet in the literary world of Catullus' Lesbia epigrams and their public facet in allusion to political reality, or at least one version of it. The poet once again appears as victim: erotic victim, social victim, moral victim.’* But in striking his helpless pose, Catullus lashes out at the reputation of Publius Clodius Pulcher, the historically situated popularis politician (however refracted through the rhetoric of Cicero), even while he continues to elaborate the story of his love-

affair with Lesbia, his fictional lover (however derivative of the fleshand-blood Clodia). APPENDIX: Clodius Elsewhere in Catullus" Poetry Clodius has been spotted by commentators in two other Catullan poems and it may be germane to mention briefly those sightings, which I consider false, here.

The

manifold

difficulties

of Poem

56 are

added

to

by

the

suggestion that the pupulum of line five might be Clodius, if, that is, the puella of the poem refers to Lesbia. Few scholars have taken the idea seriously; even Ellis pronounced it “very improbable" (Comm., p. 199). But this identification was taken up nevertheless by W.K. Scott (CP 64[1969] 24-9) ina

self-contradictory article: on p.26 Scott

accepts that the identification pupulum = Clodius depends on the puella's being Lesbia, yet on p.28 this condition has been forgotten. Scott is prompted by Cicero's well-known allusion to Clodius as a pusio in Cael. 36 and by the orator's use of Pulchellus in his correspondence to conclude that “Clodius had a nickname of ‘Little

CATULLUS 79: PERSONAL

INVECTIVE OR POLITICAL DISCOURSE?

4

Boy'." Now, while Cicero's letters reveal that he most certainly used

the pejorative diminutive within his circle of friends and perhaps more widely, they do not justify Scott's postulating a common nickname. Nor does Cael. 36support Scott's proposal: the marvelous image of little Clodius snuggling in bed with his older sister is completely self-sufficient in its context, requiring no special nickname as a referent, and, in any case, because the point of the remark

is to represent Clodius' incestual conduct as inter alia an inability to put away childish things (thereby combining invective tropes), Cicero is at Cael. 36 describing Clodius! conduct when he was actually a little boy. Whether or not the puella of 56 is Lesbia, the proposition that pupulum refers to Clodius rests only on cobweb speculations;

from

which

it follows that I consider H. Dettmar's

efforts to develop Scott's interpretation, in LCM 11.8 (1986) 130, unconvincing. Recent, helpful treatments of the real problems of Poem 56 include: J.N. Adams The Latin Sexual Vocabulary (London 1982)

146 n. 1; D. Fehling, Hermes

102 (1974) 376; R.G. Tanner,

Hermes 100 (1972) 5O6ff., all with important bibliography. In Poem 106 the phrase cum puero bello has been taken to refer to Clodius (e.g. Ellis Comm. 485f.), again without substantiation. Yet one finds this far-fetched suggestion even in so sensible a book as A. Richlin The Garden of Priapus (New Haven 1983) 249 n.22. The couplet, which is perhaps only a fragment and whose significance is elusive in any event, seems to me as likely to be cautionary advice about the company one keeps as an attack on some unidentifiable puer bellus: i.e. handsome is as handsome does.

NOTES l.

H.D. Rankin ‘Catullus and incest’

Lesbius' TAPA

Eranos 74 (1976) 113-21; M.B. Skinner ‘Pretty

112 (1982) 197-208.

2.

The Gellius poems: 74, 88, 89, 91 (Gellius’ "incest" with Lesbia). Gellius" identity: T.P. Wiseman Cinna the Poet and other Roman Essays (Leicester 1974) 119-29; H. Benner Die Politik des P. Clodius Pulcher (Stuttgart 1987) 160f.

3.

A view she argues more fully in ‘Parasites and strange bedfellows: a study in Catullus' political imagery' Ramus 9 (1970) 137-52; cf. also Skinner (n.1) 201 n.10. J. Granarolo L’oeuvre de Catulle. Aspects religieux, éthiques et stylistiques (Paris 1967) 242f. contents himself with the assertion that this epigram contrasts the virtue of Catullus' family with the vices of Lesbia's.

4.

The construction of Lesbius can be paralleled in Cicero's Cytherius, a pejorative reference

to Cytheris’

lover Mark

Antony

(Att.

15.22.5); cf. H.P.

Catull: Eine Interpretation vol. | (Darmstadt 1984) 30.

Syndikus

The early, and not unreasonable, belief that Lesbius refers to Sex. Clodius

W. JEFFREY TATUM

42

(whose description in Cic. Dom. 25 is so apparently apt) can safely be recanted thanks to D.R. Shackleton Bailey's incontrovertible (though, alas, not uni-

versally appreciated) proof that P. Clodius' right-hand man was in fact named

Sex. Cloelius; cf. Shackleton Bailey CQ

10 (1960) 41-3 (numerous continental

attempts at refutation have supervened). For a speculative discussion of Cloelius’ background, cf. W.J. Tatum ZPE 83 (1990) 299-304. This is not the place to explore the implications of Clodius' (or Clodia's) choice to eschew the normal form Claudius/Claudia. A judicious survey of the many attempts to explain Clodius’ nomenclature can be found in T. W. Hillard The Claudii Pulchri 76-48 B.C. Studies in their Political Cohesion (Diss. Macquarie University 1976) 425-34. Cf. Cic. Art. 1.16.10; 2.1.4; 2.18.3; 2.22.1. H.P. Syndikus Carull: Eine Interpretation vol. 3 (Darmstadt 1987) 38 infers that puns on Pulcher were

commonplace in the fifties. The phrase Lesbius est Pulcher is marked by quid ni? — an interrogative

employed by Catullus to encourage his reader to reconsider the assertion just

preceding; cf. Syndikus ibid.

The use of cognomina in communication is cordial and intimate, cf. Cic. Dom. 22 and the commentary provided by R.G. Nisbet, M. Tulli Ciceronis De Domo Sua

ad Pontifices Oratio (Oxford 1939) ad loc. On cognomina generally see I. Kajanto The Latin Cognomina (Helsinki 1965); on their usage during the republic see R.

Syme Roman Papers | (Oxford 1979) 361-77; E. Badian B/CS Suppl. 51 (1988) 6-12; J.N. Adams CQ 28 (1978) 145-66; D.R. Shackleton Bailey Onomasticon to Cicero's Speeches (Norman and London

1988) 3-8.

Syndikus (n.5) 38 stresses the heartfelt tone of self-apostrophe in Catullus' poetry. E.A. Havelock The Lyric Genius of Catullus 2nd ed. (New York 1967) 112 is wrong to see in the opening couplet a bold assertion of family pride on Catullus' part, pace Syndikus (n.5) 39 n.7, although he is of course correct in his recognition that Catullus, provincial though he was, reveals in his poetry not even a “slight inferiority complex" when he stands up to the great men of his day. Suavia does not refer to passionate kisses here, but to the familiar Roman custom of greeting with a kiss; cf. Ph. Moreau RPh 52 (1978) 87-97. The suggestion of P.Y. Forsyth The Poems of Catullus: A Teaching Text (Lanham 1986) 510, misses any point: it is both strained and (worse) incompatible with Roman notions of sexual depravity. Ph. Moreau Clodiana Religio. Un proces politique en 61 av. J.-C. (Paris 1982) 168ff. As for the Pro Caelio, see inter alia Cael. 32 and 36. Cic. Sest. 16.

Cic. ©. Frat, 2.3.2. Dio 46.18; [Sall.] Inv. in Cic. 2. (n. 1) 120, accepted by Skinner, (n.1) 204. Rankin (n.1) 120 cites a good deal of evidence. This is of course not to say that

incest was alien to the Roman aristocracy, only that it should not be considered inherently and unambiguously aristocratic when it surfaces in Roman poetry. 15.

D.R. Shackleton Bailey AJAH 2 (1977) 148-50; a different reconstruction has

been suggested by T.W. Hillard, M. Taverne and C. Zawawi in an article forthcoming in AJAH (I am grateful to Dr Hillard for sending me an early draft of his paper).

CATULLUS

79: PERSONAL INVECTIVE OR POLITICAL DISCOURSE?

43

16.

Moreau (n.9) 174.

17.

Cf. Cat. 59 (oral sex), 78, 111 — all incidents of incest outside the bounds of nobility.

18.

Cf. G. Hanard ‘Inceste et société romaine républicaine: un essai d'interprétation ethno-juridique du fragment du livre XX de l'histoire romaine de Tite-Live' RBPh 64 (1986) 32-61. Bibliography on this point is collected on p.33, n.3. Roman

society

was

naturally

exogamous

in

practice,

whatever

the

norm

throughout the remainder of Mediterranean societies; cf. B.D. Shaw and R.P. Saller ‘Close-Kin Marriage in Roman Society? Man 19 (1984) 432-44. 19.

(n.1) 113. Cf. R.O.A.M. Lyne The Latin Love Poets (Oxford 1980) 38ff. See also the interesting remarks of P. Walcot Anc. Soc. 18 (1987) 32f.

Cat. 72.3-4. Thus e.g. A. Baehrens Catulli Veronensis Liber (Leipzig 1885) 560; R. Ellis A Commentary on Catullus 2nd ed. (Oxford 1889) 454; W. Kroll C. Valerius Catullus 5th ed. (Stuttgart 1968) 253; Syndikus (n.5) 39. Cf. Plaut. Mil. 21ff.; Sen. Apoc. 11.

22.

K. Quinn Catullus. The Poems 2nd ed. (Glasgow 1973) 414f.; Skinner (n.1) 199;J. Ferguson Catullus (Lawrence, Kansas 1985) 265f.; Forsyth (n.8) 510. Quinn is included, though his note is unclear, and his comment on p.415 (“‘Lesbius is the

sort of man to commit a major crime, just to get a few friendly greetings.") indicates a misunderstanding of the passage. 23.

Catullan ambiguity: D.N. Levin CP 54 (1959) 109-11; TAPA 100 (1969) 221-36. Cf. Skinner (n.1) 199,

24.

Learned friends have made learned suggestions as to the import of tria in line four, but [ cannot with confidence go beyond the usual reckoning, that tria simply indicates an insignificant number.

25.

H.D. Jocelyn PCPS 206 (1980) 12-66; AJP 101 (1980) 431; PLLS 3 (1981) 278f. and 283 n.7; D. Bain CQ 41 (1991) 74ff. Examples of comparably oblique references (by the shunning of a kiss) to os impurum include AP 11.219 and Mart. 2.21. Catullus' attitude toward the cunnilingus or pathicus: M.G. Morgan AJP 100 (1979) 377ff.; A. Richlin The Garden of Priapus: Sexuality and Aggression in Roman Humor (New Haven 1983) 148. It is perhaps worth reminding ourselves that a cunnilingus was not necessarily assumed to be impotent; cf. Jocelyn LCM 10.3 (1985) 41f., which means that there is no necessity of introducing that imputation to Poem 79 (as does Skinner [n.1] 198f.). D. Lateiner ‘Obscenity in Catullus" Ramus

16 (1977) 15-32.

27.

Skinner (n.1) 207.

28.

R. Reitzenstein Sitzungsber. der Heidelb. Ak. der Wiss. 12. Abhandlung (1912) 9-36; D.O. Ross Style and Tradition in Catullus (Cambridge, Mass. 1969) 80-95. Lyne (n. 19) 24ff.

P.A. Brunt The Fall of Rome and Related Essays (Oxford 1988) 351-81. This is the latest version of an essay which first appeared in PCPS 191 (1965) 1-20, a piece known both to Ross and to Skinner.

31.

B. Axelson Unpoetische Wörter (Lund 1945). Commentary on Axelson: G. Williams Tradition and Originality in Roman Poetry (Oxford 1968) 743-50; D.T. Benediktson Phoenix 31 (1977) 341-8; P. Watson CQ 35 (1985) 430-48. The

W. JEFFREY TATUM

approach adopted by R.O.A.M. Lyne Words and the Poet: Characteristic Techniques of Style in Vergil's Aeneid (Oxford 1989) 4ff., seems eminently balanced and reasonable. The dependence of Ross' method on that of Axelson's is made evident at Ross (n.27) 4ff.

32.

This is not to say that Lyne's discussion resolves everything in Catullus' Lesbia poems: his argument that Catullus conceived of his relationship to Lesbia as a marriage fails, in my opinion, to counter Ross' perceptive criticisms of this idea; cf. Ross (n.28) 81f.

33.

See the recent discussion by H.P. Syndikus Gymnasium 93 (1986) 34-47.

Cf. Rhet. ad Her. 2.48f.; Cic. Inv. 1.100-105. Discussion of Roman invective: R. Syme The Roman Revolution (Oxford 1939) 149ff.; R.G.M. Nisbet Cicero: in L. Calpurnium Pisonem Oratio (Oxford 1961) 192ff.; F. Gonfroy Dialogues d'histoire ancienne 4 (1978) 219-62; S. Koster Die Invektive in der griechischen und rómischen Literatur (Meisenheim am Glan

3-30, esp. p.10.

1980); P. Veyne Latomus 42 (1983)

Virtus in Roman political thinking: D.C. Earl The Moral and Political Tradition of Rome (Ithaca 1967) 11-43. Sources collected and discussed in Koster (n.34) 7-21.

One recent attempt is that of Richlin (n.25). To state the obvious, Cicero's political rhetoric included but was hardly limited to invective. A recent and valuable examination of Cicero's political thought (one which examines the speeches as well as the treatises) is provided by T.N. Mitchell Cicero: The Senior Statesman (New

Haven

1991) 9-62.

Cic. Att. 4.2.2. Cic. Dom. 43f. Cf. Red. Sen. 48; Sest. 65, 133; Pis. 30. Cic. Dom. 46f. Brunt (n.30) 53-65 (with abundant citations). W.J. Tatum CP 85 (1990) 205ff. The recent comparative study of Catiline and

Clodius by T. Loposzko and H. Kowlaski Klio 72 (1990) 199-210 adds little. Ph. Moreau Athenaeum 75 (1987) 465-92.

45.

Ibid. Clodius' attacks on Cicero's property: W. Nippel Aufruhr und Polizei in der

46.

Cicero's rhetoric of course reflects certain realities. Though often of comparable wealth and culture, equites could not match senators for dignity or splendour — or clout. As a result, they were genuinely more vulnerable, and not only to

römischen Republik (Stuttgart 1988) 116ff., with ample bibliography.

populares:

Sulla's first proscription

list, according to Appian

(B. Civ.

1.95),

included 40 senators and 1600 knights; for difficulties in determining precise figures for the Sullan proscriptions and for a discussion of the Roman recollection of the knights’ victimisation see F. Hinard Les proscriptions de la

Rome républicaine (Paris and Rome 1985) 116-20. On equestrians, municipal and

otherwise, see C. Nicolet L'ordre équestre à l'époque républicaine (312-43 av. J.-C.) vol. 1 (Paris 1974); T.P. Wiseman, Roman Studies (Liverpool 1987) 57-82 and 297-305; and Brunt (n.30) 144-93. Catullus' family background is explored in Wiseman ibid. 335-48 and F. Della Corte Maia 41 (1989) 229-34, whose inference from Catullus" poetic treatment of fellow provincials that the poet regarded Verona and her inhabitants as inferior to Rome — even if it is correct —

CATULLUS 79: PERSONAL

INVECTIVE OR POLITICAL DISCOURSE?

45

is no impediment to the view advanced here. 4.

Cf. Red. Sen. 25f.; Dom. 30; Sest. 129f.; Mil. 39; Pis. 34-6. On the circumstances of Cicero's recall, see C. Habicht Cicero the Politician (Baltimore and London 1990) 35-52 and Mitchell (n.38) 153ff. In a letter written shortly after these events, Cicero reported that, owing to his

excessive recourse to violence, even Clodius' followers were beginning to leave him: Clodius is described as desertus a suis (Att. 4.3.2). 49.

One may compare Cicero's ridicule of Clodius' poorly attended contio in the In Clod.

et Cur. (Schol. Bob. 88.1-2 (St.]): accesserunt ita pauci, ut eum non ad

contionem,

sed sponsum

diceres advocasse,

which

of course was a far less

momentous moment than the vote over Cicero's recall. Although Cicero, in the

In Clod. et Cur., portrays Clodius as politically isolated in the year 61, this speech

apparently saw the light of day as a published text only in 58 (inadvertently at that) and was in circulation at the time of Cicero's exile; cf. J. Crawford M.

Tullius Cicero: The Lost and Unpublished Orations (Göttingen 1989) 106ff. The

embarrassment, literary as well as political, caused by the circulation in 58 of Cicero's attack on Curio is discussed in W.J. Tatum Mnemosyne 44 (1991)

364-71.

During the late first century the local aristocracies of Italy were not a natural constituency of popularis politicians, whose efforts were directed chiefly at the urban plebs (and, often, at enhancing the careers of political grandees like Pompey or Caesar); cf. C. Meier RE Suppl. 10.580-89. 51. 52.

Again, for a recent and amply documented Mitchell (n.38) 160ff.

discussion of these events, see

It is no impediment to my reading of the poem that Catullus was not in Rome

during the time of Clodius' defeat or Cicero's triumph with his De Domo; in fact, it renders Catullus" recourse to a Ciceronian mode of criticism all the more understandable. Catullus’ attack on Clodius clearly derives from the period of his disenchantment with Lesbia, and, if Syndikus (n.33) is correct that Catullus'

hostility for the dynasts stems from the arrangements made at Luca in 56 (at which time the Claudii Pulchri changed their political stance to join forces with Pompey, cf. W.J. Tatum Kl/io 78 [1991] 122-9), it may well be that the poem isto

be dated after 56. The period leading up

to Luca, as Nippel (n.45) 123 observes,

was rife with political invective hurled at Pompey and Clodius (and Clodia),

some of it originating with the poets. For the chronology of Catullus' works, cf. T.P. Wiseman

53.

Catullan Questions (Leicester 1969).

G. Williams (n.31) 518ff.; F. Cairns Generic

Poetry (Edinburgh

Composition in Greek and Roman

1972), C.W. Macleod CQ 23 (1973) 274-303 = Collected

Essays (Oxford 1983) 171-80; J. Griffin Latin Poets and Roman Life (London 1985); O. Murray JRS 75 (1985) 39-50.

54.

M. Putnam HSCP 65 (1961) 165-206; Ramus 13 (1974) 70-86 (= Essays on Latin

Lyric Elegy, and Epic (Princeton 1982) 45-85 and 13-29 respectively); Richlin (n.25) 145.

55.

The importance of this poem in the Lesbia cycle is discussed by Syndikus (n.4) 29f. and (n.5) 37ff.

An earlier version of this paper was read to the Leeds Latin Seminar in May 1991. I very much benefited from the stimulating contributions (not always gentle) of the audience. The anonymous referee for PLLS was equally helpful.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 47-60 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

VARRO’S ATTITUDE TO LATIN DERIVATIONS FROM

GREEK

ROBERT MALTBY

According to Suetonius the serious study of grammar in Rome began in 169 BC when Crates of Mallos, on an embassy from king Attalus to the senate, fell down a sewer, broke his leg and spent his lengthy convalescence lecturing to his hosts on Greek language and literature (De Grammaticis 2). But it is not until the first century BC, perhaps as a result of a new influx of Greek grammarians captured in the wars with Mithridates, that we hear of any serious speculation concerning the relationship of the Latin language to Greek. Romans like Varro, who were well versed in both languages, must have been struck by the clear correspondences between them not only in matters of vocabulary, but also at the deeper levels of morphology and syntax. Such similarities demanded an explanation. As the ancients never developed any idea of what we would call comparative philology, the

true explanation, that Latin and Greek were derived from a common ancestor, forever eluded them. Nevertheless it was a problem of considerable philosophical as well as social and political importance.! At the philosophical level it fuelled debates concerning the ultimate origin of language. Was there, as the Stoics believed, a natural connection between the sound of a word and the thing it named (the theory of physis), or was language a system of conventionally imposed signs which would differ completely from one people to another (the theory of thesis)? The theory of physis, which goes back at least as far as Heraclitus,? is supported by Cratylus in Plato's dialogue of that name.? One obvious objection to it is that different languages use different words for the same thing. If a 47

48

ROBERT MALTBY

natural connection did exist, then, as Sextus Empiricus points out,‘ Greeks would be able to understand barbarians and vice versa. Hence Democritus,* Hermogenes in the Cratylus, perhaps taking his lead from

Parmenides, and later Aristotle,’ developed the theory

that language was a purely conventional set of signs that would differ from one language community to another. A comparative study of Latin and Greek could provide ammunition for either camp. Varro's interest in language, however, was not primarily philosophical? but antiquarian and grammatical: the

history of Latin words could illuminate the history of the Roman people and vice versa.? Furthermore, he saw a close link between etymology and semantics. His purpose in revealing the etymologies of Latin words was in part to throw light on their true meaning and correct usage (De Lingua Latina 5.2): cum unius cuiusque uerbi naturae sint duae, a qua re et in qua re uocabulum sit impositum ... priorem illam partem, ubi cur et undesint uerba scrutantur, Graeci uocant ἐτυμολογίαν, illam alteram περὶ σημαινομένων. de quibus duabus rebus in his libris promiscue dicam, sed exilius de posteriore. (Inasmuch as each and every word has two innate features, from what thing and to what thing the name is applied ... that former part, where they examine why and whence words are, the Greeks call etymology, that other part they call semantics. Of these two matters I shall speak in the following books, not keeping them apart, but giving less attention to the second. [tr. Kent])

But like the Alexandrian grammarians before him he made use of sources whose main interest was philosophical. He mentions specifically

the

Stoics

Cleanthes

(5.9) and

Chrysippus

(6.2), so it is

legitimate to look in Varro's work for traces of their theories and possibly also for some kind of philosophical synthesis of his own. A simpler and at the same time politically attractive explanation for the similarities between Latin and Greek, compatible with either the natural or the conventional theory of language, was to argue that they were not separate languages at all, but that Latin was in fact a dialect of Greek. A number of Greek grammarians working in Rome wrote on Latin's debt to Greek. Among them the most influential were probably Hypsicrates, whose works super his quae a Graecis accepta sunt are mentioned by Aulus Gellius (Attic Nights 16.12); the Alexandrian scholar Philoxenus, who wrote books on various . dialects of Greek, the Laconian, the Ionic, the Syracusan and the *Roman'; and, thirdly, the elder Tyrannio, friend of Atticus and

Cicero, who was probably the author of a work entitled περὶ τῆς

VARRO'S ATTITUDE TO LATIN DERIVATIONS FROM GREEK

49

Ῥωμαϊκῆς διαλέκτου ὅτι ἔστιν ἐκ τῆς ΕἘλληνικῆς wrongly ascribed by the Suda to the younger Tyrannio. This simpler explanation would have an obvious appeal in the Rome of the first century BC. Greek grammarians could flatter their Roman patrons’ desire for cultural respectability by persuading them that they were in fact Greek speakers and that Rome was just another Greek city. The Greeks themselves might be comforted to learn that they had not been conquered by βάρβαροι of an inferior race, but by fellow Greek speakers. The aim of this paper is to investigate how the relationship between Latin and Greek is reflected in Varro’s etymologies and to establish, if possible, his position in regard to the various theories that were current in his time. A number of earlier works have covered this subject in passing as part of wider surveys of Varro's linguistic theories; in particular I would acknowledge a debt to Jean Collart,

whose Varron, grammairien latin cleared most of the ground on which the present study is based.'? 1.

Varro on the derivation of Latin from Greek

Before attempting to analyse Varro's philosophical standpoint, it would perhaps be advisable to deal with the less contentious issue of his attitude to the question of whether Latin was a dialect of Greek. Dionysius of Halicarnassus, who like Varro combined antiquarian with linguistic interests, tells us (Roman Antiquities 1.90.1): Ῥωμαῖοι δὲ φωνὴν μὲν οὔτ᾽ ἄκρως βάρβαρον οὔτ᾽ ἀπηρτισμένως Ἑλλάδα φθέγγονται, μικτὴν δέ τινα ἐξ ἀμφοῖν, ἧς ἐστιν ἡ πλείων Αἰολίς. (The language spoken by the Romans is neither utterly barbarous nor absolutely Greek, but a mixture, as it were, of both, the greater part of which is Aeolic.)

John of Lydia attributes to Cato in his Origines and to Varro in the De Origine Linguae Latinae the view that the people of Romulus' time learnt Aeolic Greek from Evander and the Arcadians who came with him to Italy.'! Healso tells us that Varro (probably this time in the De Sermone Latino) thought that Latin consisted of a mixture of Aeolic Greek, Gallic and Etruscan.!? The idea that Latin was derived largely from Aeolic Greek lives on well after Varro's time and is mentioned,

for example, by Quintilian," as well as in later grammarians such as Diomedes and Priscian. By Aeolic, as Collart points out,'^ the ancients meant any Greek dialect that was not Attic, Ionic or Doric.

50

ROBERT MALTBY

It had a wider extension than what we would think of as Aeolic (i.e. Thessalian, Boeotian and Lesbian), since they associated it mainly with the dialect spoken by the Arcadian Evander. The main similarities with Latin that struck the ancients were its preservation of digamma, which Quintilian refers to as the Aeolic letter, and the use of long a for ἢ in words such as mater for μήτηρ. Varro, as one would expect from his antiquarian studies, was well aware of the various traditions which linked the Greeks with early Rome, and on a number of occasions in De Lingua Latina 5 he refers

to these legends while discussing individual etymologies. At 5.21, for example, he gives the Greek τέρμων asan alternative to the Latin tero as the source of the word terminus, and connects this with the arrival of Evander in Italy.'? At 5.45 the Argei, straw puppets thrown in the Tiber as part of an annual purification ceremony, are connected with the Argive princes who came with Hercules to Italy and settled there.!° At 5.53 the Pallantes or Palatini are given as one of a number of alternative origins of Palatium, the Palatine district of Rome." Finally, at 5.101, when correcting Aelius Stilo's derivation of lepus from leuipes, Varro suggests instead as its origin the Greek word λέπορις which he ascribes to the Siculi, who, he says, according to the annalists started out in Rome before moving on to Sicily.!* As the other Varronian references to this etymology show, the Siculi are variously associated with the antiqui Graeci and the Aeolians. The scanty references to them in the annalists show them as responsible for various place-names on the mainland of Italy.'® They crop up on four further occasions closely grouped together at the end of Book

Five in etymologies of what are alleged to be Greek-based words,” suggesting that Varro had them from a common source. In addition to its association with the Siculi in 5.101, Aeolic Greek is specifically mentioned as the origin of Latin words on a further four occasions in

Varro.?! Only

a very small proportion,

then, of the one hundred and

seventy or so Greek-based etymologies in Varro are specifically linked with the Aeolic dialect or with early contacts between Greeks

and Romans. Greek etymologies are suggested for about one in eight of the words examined in the De Lingua Latina, and in about a quarter of these cases the Greek is given in second place as an

alternative to a Latin-based derivation." Derivations from other foreign languages, mostly Sabine, account for a further one in twenty-five words.?? It is clear from Varro's practice, then, that he was not a strong adherent of the view that Latin was derived largely

VARRO'S ATTITUDE TO LATIN DERIVATIONS FROM GREEK

51

from Greek. The majority of Latin words are to be derived from native Latin roots. He is aware of the danger, first mentioned by Plato in his discussion of the derivation of the Greek words fire and

water πῦρ and ὕδωρ from Phrygian roots,?* of trying to derive nonnative words from native roots; witness his discussion of lepus (n.18). But he is equally aware of the opposite danger, also mentioned by Plato,” that the assumption of a foreign root can often be an evasion, covering one's ignorance of a native origin; witness Varro's etymology of faenerator in contrast with the Greek derivation proposed by

Hypsicrates.?$ As his statements in 5.3 and 5.10 show, Varro accepts that not all Latin words are of native origin; where foreign words are concerned the etymologist's task is simply to say unde sint, what language they came from.?' Elsewhere Varro often divides words into peregrina and uernacula, foreign and native words; although the

majority of the foreign words are Greek, no distinction is made within the category of peregrina between Greek and other foreign words.?® There is, therefore, no evidence for Greek having in Varroa

special position among foreign languages as the ultimate source of the Latin language. Words

derived from

Greek

roots are, however, often easier to

identify than those from other languages, and that must be the implication of a rather obscure passage at De Lingua Latina 6.40: uerborum quae tempora adsignificant, ideo locus difficillimus ropa, quod neque his fere societas cum Graeca lingua ... (Of the words which also indicate time the most difficult feature is their radicals [ἔτυμα], for the reason that these have in general no communion [societas] with the Greek language ... [tr. Kent])

By *words which also indicate time' Varro means verbs. The phrase quae tempora adsignificant has already occurred at 6.36 to denote verbs and seems to be a deliberate calque on the Aristotelian definition of a verb in De Interpretatione 16b6: ῥῆμα δέ ἐστι τὸ προσσημαῖνον χρόνον; the prefix ad- in adsignificant, like npoo-in npooonpaivov, indicates that the expression of time is an additional feature of the meaning of verbs. Verbs, then, are difficult because

they generally have no societas with the Greek language. Book Six ends nevertheless with a list of verbs which some scholars would,

according to Varro, derive from Greek roots (6.96): sed quoniam in hoc de paucis rebus uerba feci plura, de pluribus rebus uerba faciam pauca, et potissimum quae in Graeca lingua putant Latina [sc. originem habere], ut scalpere a σκαλεύειν et sternere a στρωννύειν, lingere a λιχμᾶσθαι, i ab εἶ, ite ab ite, gignitur a γίγνεται,

52

ROBERT MALTBY

ferte a φέρετε, prouidere a προϊδεῖν, errare ab Eppetv, ab eo quod dicunt στραγγαλᾶν strangulare, tinguere a τέγγειν. praeterea depsere a δέψειν; ab eo quod illi μαλάσσειν nos malaxare, ut gargarissare ab ἀναγαργαρίζεσθαι, putere a πύθεσθαι, domare a δαμάζειν, mulgere ab ἀμέλγειν, pectere ἃ πέκειν, stringere a στλεγγίζειν, id enim a στλεγγίς, ut runcinare a runcina, cuius ῥυκάνη origo Graeca. (But since in this connection I have spoken at length on a few matters,

I shall speak briefly on a number of topics, and especially on the Latin words whose origin they think to be in the Greek tongue: as scalpere ‘to engrave’ from σκαλεύειν ‘to scratch’, sternere ‘to spread out’ from στρωννύειν, lingere ‘to lick’ from λιχμᾶσθαι, i ‘go’ from εἶ, ite from ἴτε, gignitur ‘he is born’ from γίγνεται, ferte ‘bring’ from φέρετε, prouidere ‘to foresee’ from προϊδεῖν, errare ‘to stray’ from Eppetv ‘to go away’, strangulare ‘to strangle’ from otpayyaA Gv, tinguere ‘to dip, dye’ from τέγγειν. Besides, there is depsere ‘to knead’ from δέψειν; from the word which they call μαλάσσειν, we say malaxare ‘to soften’, as gargarissare ‘to gargle’ from ἀναγαργαρίζεσθαι, putere ‘to stink’ from πύθεσθαι ‘to decay’, domare ‘to subdue’ from δαμάζειν, mulgere ‘to milk’ from ἀμέλγειν, pectere ‘to comb’ from πέκειν, stringere ‘to scrape’ from στλεγγίζειν: for this is from στλεγγίς ‘scraper’, as runcinare ‘to plane’ from runcina ‘plane’, of which ῥυκάνη is the Greek source.)

Varro is not specific about who putant refers to, nor does he tell us whether he himself accepts their derivations. But one of the examples in the list, errare from Eppetv, suggests Hypsicrates as a possible source, since Gellius ascribes this derivation to Cloatius Verus, who,

he says, based his work on that of Hypsicrates (n.26). In fact all the words in this list except errare and stringere are either cognate with or derived from the Greek words mentioned. But the list well exemplifies the inability of the Romans to distinguish between these two groups: borrowed words like strangulare, depsere and malaxare are listed side by side with cognate words such as tinguere, putere and domare and the same formula with a or ab is used throughout the whole list. As far as the introductory formulae for etymologies from Greek

are concerned, Collart identified three main types:?? 1. A (Latin word) quod Graeci dicunt B (Greek word) 2. A a Graeco/a Graecis B 3. A Graecum uerbum/vocabulum/nomen B He came to the conclusion, however, that there was little distinction

between them. On reinvestigating I found that the quod type (1) is used

most

often (though

not exclusively) where

Varro

wishes to

specify a borrowing from a particular dialect, as at 5.120 with the Siculi (n.19) and 5.102 with the Aeolis (n.25). The distinction between

VARRO'S ATTITUDE TO LATIN DERIVATIONS FROM GREEK

53

types (2) and (3) is well illustrated by the discussion of Castor and Pollux at 5.73: Castoris

nomen

Graecum,

Pollucis a Graecis;

in Latinis litteris

ueteribus nomen quodest, inscribitur ut Πολυδεύκης Polluces, non ut nunc Pollux.

Type (3), the nomen graecum type, is used where the Latin word is of exactly the same form as the Greek, and in most cases the Greek is not quoted, as with Castor; type (2), a Graeco or a Graecis, is used

where the Latin and Greek forms are no longer identical and both need to be given, as in the case of Pollux. 2. Varro's philosophical standpoint What, then, does Varro's attitude to the derivation of Latin words

from Greek reveal about his philosophical position on the origins of language? In most cases very little. Normally when Varro thinks ofa societas between Latin and Greek he is thinking of a situation in which

the Latin word

was borrowed

or derived from the Greek,

either at an early stage in the language's history, or more recently, as for example with his own borrowing of technical grammatical terms such as etymologia. On occasion, however, one catches a glimpse of more subtle parallelisms which may go back to his Stoic sources or to the views of Varro's Pythagorean contemporary, Nigidius Figulus. Aulus Gellius gives an example of how Nigidius argued for the natural origin of language using as an illustration the personal pronouns whose sound was naturally fitted to their sense (Attic Nights 10.4). To put it crudely, his argument is that when you say tu or vos your lips point outwards to the person referred to, whereas with ego, me and nos they are kept closer in to oneself. This of course could be made to work for Greek as well as Latin, and at the end he comments that “the same principle that we have noted in our speech

applies also to Greek words". In fact Nigidius' discussion of the appropriateness of ego follows closely the Greek Stoic Chrysippus" discussion of that pronoun in Greek.?? In Varro a similar type of argument occurs in 5.96, where he suggests that the similarity between Latin and Greek in the names of domestic animals could be put down to the fact that onomatopoeia gives rise to the same results in both languages: ex quo fructus maior, hic est qui Graecis usus; sus, quod óc, bos, quod βοῦς, taurus, quod ταῦρος, item ouis, quod ὄις: ita enim antiqui

54

ROBERT MALTBY

dicebant, non ut nunc πρόβατον. possunt in Latio quoque ut in Graecia ab suis uocibus haec eadem ficta. (Regarding livestock from which there is larger profit, there is the same use of names here as among the Greeks: sus ‘pig’, the same as ὗς; bos ‘ox’ the same as βοῦς; taurus ‘bull’, the same as ταῦρος; likewise ouis ‘sheep’ the same as ὄις: for thus the ancients used to say, not πρόβατον as they do now. This identity of the names in Latium and in Greece may be the result of invention after the natural utterances of the animals.)

The same argument is used again shortly afterwards to explain why

the word for bear is the same in Lucanian and Latin (5.100): ursi Lucana origo vel, unde illi, nostri ab ipsius voce (the name ‘bear’ is of Lucanian origin, or our ancestors called it from its voice, as did the

Lucanians). Another very important principle in Latin etymologising was that the rationale behind the etymology of a Latin word could be parallel

to that of its Greek equivalent. Straightforward examples of this principle occur in Varro at 5.156 (senaculum vocatum, ubi senatus aut ubi seniores consisterent, dictum ut γερουσία apud Graecos), where the Latin derivation of senaculum ('senate stand") is said to parallel

that of the Greek γερουσία (‘assembly of elders’), and at 5.66 (Aelius Dium Fidium dicebat Diouis filium, ut Graeci Διόσκορον Castorem) where Varro quotes Aelius Stilo's view that Dius Fidius is parallel to the Greek Διόσκορος meaning 'son of Zeus'. A number of such parallels occur in the speech of the Stoic Balbus in Cicero's De Natura Deorum, suggesting that they originate from attempts to translate into Latin etymologies contained in Greek Stoic philosophers. At 2.64, for example, Balbus gives an etymology of Saturn from saturat annis on the analogy of the derivation of the Greek Kronos from χρόνος (‘time’).?' Quintilian criticises the ingenuity of those like Gavius Bassus, who support their Latin etymologies by reference to Greek parallels: caelibes (‘bachelors’) are like caelites (‘heaven dwellers’) just as ἠίθεοι (‘young men’) in Greek are 1j θεοί (‘like

gods’).?? For the most part in Varro where such parallels with Greek do exist this is not made explicit. Macrobius, for example, points to a parallel between the Stoic Greek etymology of Apollo and the Latin etymology of Sol his Roman mythological counterpart.?? Just as

Apollo comes from alpha privative and πολλοί (i.e. only one), so Sol in Latin is connected derivation for Sol but Apollo (Sol uel quod ita in the same paragraph

with solus. At 5.68 Varro gives the solus does not mention explicitly the parallel with Sabini, uel quod solus ita lucet), although later Apollo is mentioned as the Greek equivalent

VARRO'S ATTITUDE TO LATIN DERIVATIONS FROM GREEK

55

of Sol.?* Varro's etymologies of /una at 5.68 and memoria at 6.49 can also be shown to have Greek parallels in this way, and more examples

are to be found in Reitzenstein and Dahlmann.? The frequency of such parallels in passages dealing with the names of gods and heavenly bodies lends weight to the suggestion that they have their origin in a Latin adaptation of a Greek Stoic source. It is not the purpose of the present paper to analyse at length the general theoretical basis of Varro's linguistic teaching, particularly as there are adequate discussions of this elsewhere. However, in

order to set his derivations from Greek in this wider context, certain key passages from De Lingua Latina may be briefly considered: duo igitur omnino uerborum principia, impositio «et declinatio», alterum ut fons, alterum ut riuus. impositicia nomina esse uoluerunt

quam paucissima, quo citius ediscere possent, declinata quam plurima, quo facilius omnes quibus ad usum opus esset dicerent. (8.5) (The origins of words are therefore two in number, and no more: imposition and inflection; the one is as it were the spring, the other the brook. Men have wished that imposed nouns should be as few as possible, that they might be able to learn them more quickly; but derivative nouns they have wished to be as numerous as possible, that ali might the more easily say those nouns which they needed to use.) in hominum partibus esse analogias dicunt, quod eas natura faciat, in uerbis non esse, quod ea homines ad suam quisque uoluntatem fingat, itaque de eisdem rebus alia uerba habere Graecos, alia Syrios, alia Latinos: ego declinatus uerborum et uoluntarios et naturalis esse puto, uoluntarios quibus homines uocabula imposuerint rebus quaedam, ut ab Romulo Roma ... naturales ut ab impositis uocabulis quae

inclinantur in tempora aut in casus, ut ab Romulo Romuli Romulum et ab dico dicebam dixeram. (9.34) (In the parts of men they say there are regularities, because nature makes them, but there is none in words, because men shape them each as he wills, and therefore as names for the same things the Greeks have one set of words, the Syrians another, the Latins still another. I firmly think that there are both voluntary and natural derivations of words,

voluntary for the things on which men have imposed certain names, as Rome from Romulus ... and natural as those which are inflected for tenses or for cases from the imposed names, as genitive Romuli and accusative Romulum from Romulus, and from dico ‘I say’ the imperfect dicebam and the pluperfect dixeram.) natura ... dux fuit ad uocabula imponenda homini. (6.3) (Nature was man's guide to the imposition of names.) illi, qui primi nomina imposuerunt rebus. (8.7) (Those who first imposed names upon things). utilitatis causa uerba ideo sunt imposita rebus ut eas significent. (8.27)

56

ROBERT

MALTBY

(It was for the sake of usefulness that words were imposed on things in

order to designate them).

From these passages it is clear that Varro’s philosophical position on the question of physis versus thesis is somewhat different from that of the Stoics. He believed in the voluntary impositio by a number of original name givers of a small initial stock of vocabulary (8.7) which was then increased through the process of declinatio, by which Varro meant both the human process of derivation and the natural process

of morphological inflexion (9.34).°” The emphasis on utilitas (8.27), and the idea of language development as a continual process are features of Varro’s theory which are closer to Epicurean than to Stoic thought. But, as 6.3 shows, there was still a place for natura, for similarity between word and thing, in the process of impositio itself. It was the human involvement in this process (9.34), however, which

accounted for the differences between one language and another. 3. Conclusion

To sum up Varro’s position, we can say that he preferred to derive Latin words from native Latin roots, and did not regard Greek asthe main source ofthe language. Where he does admit to Greek influence he usually sees it in the form of Latin words being borrowed directly from

Greek,

but in only a few such cases is it specified that this

borrowing took place at an early stage of Rome's history. Other forms of societas between Greek and Latin, such as analogical formations of the type Dius Fidius/Aióokopoc, or the influence of onomatopoeia giving rise to similar sounding words in both languages, these Varro may have derived from his philosophical sources, probably via a Latin intermediary. Finally, it would seem pertinent to set Varro's teaching on the relation between Greek and Latin in its wider literary context. This was a period when Roman writers were borrowing extensively from

Greek: in the poets as a conscious stylistic device, a means of aligning themselves more closely with their Hellenistic Greek models; in prose writers as a means of compensating for the patrii sermonis egestas on technical subjects. Horace's advice that new words will win accep-

tance if drawn from a Greek source is well known.?* But in fact straight loan words from Greek, like Horace's diota in Odes 1.9.8, are relatively rare in this period, except in Varro's Menippean Satires. Far more common are calques, or words formed on the analogy of Greek, which have the advantage of not sounding foreign. We have

VARRO'S ATTITUDE TO LATIN DERIVATIONS

FROM GREEK

57

already seen this in Varro's own use of the phrase quae tempora adsignificant for verbs. So in Ars Poetica 234 Horace uses dominantia nomina, based on the Greek κύρια ὀνόματα, to designate words in their literal meaning. With Varro's use of the straight loan word etymologia (De Lingua Latina 7.109) we may compare Cicero's attempts at finding a Latin equivalent in Topica 35, where he suggests notatio, based on the Aristotelian σύμβολον, and ueriloquium, based on étupoAoyía. Neither of these had much of a future, but the equivalence of the Latin uerus with the Greek ἔτυμος in the first element of ἐτυμολογία becomes important in Augustan poetry where uerus, uere and equivalents are used as pointers to etymological

word play.?? The question of poetic etymologising, however, and its relation to Varro is a complicated topic which deserves a separate treatment of its own.

NOTES |l.

On the study of language in Rome

in Varro's time see Elizabeth Rawson

Intellectual Life in the Late Roman Republic (London

1985) 117-31.

2.

ἈΚ. Pfeiffer History of Classical Scholarship (Oxford 1968) 12, 59: F. Cavazza Studio su Varrone etimologico e grammatico (Firenze 1981) 21.

3.

Pl. Crat. 383a-b Κρατύλος φησὶν ὅδε ... ὀνόματος ὀρθότητα elvai ἑκάστῳ τῶν ὄντων φύσει πεφυκυῖαν ... καὶ Ἕλλησι xai βαρβάροις τὴν αὐτὴν ἅπασιν. (Cratylus here says ... there is a correctness in the name of all things which comes by nature ... and is the same for all men, both Greeks and barbarians.)

4.

Sextus Empiricus Adv. Math. 1.145 εἴπερ γὰρ φύσει τὰ ὀνόματα ἦν Kai μὴ τῇ καθ᾽ ἕκαστον θέσει σημαίνει, ἐχρῆν πάντας πάντων ἀκούειν, Ἕλληνας βαρβάρων καὶ βαρβάρους Ἑλλήνων καὶ βαρβάρους βαρβάρων. (If names came about by nature and did not indicate each thing by convention, then everyone would necessarily understand everyone else, Greeks would understand barbarians, barbarians Greeks and barbarians barbarians.)

5.

According to Proclus Jn Cratylum 16.7 (Pasquali), Democritus maintained that τύχῃ ἄρα καὶ οὐ φύσει τὰ ὀνόματα (names come about through chance not by nature).

6.

Pl. Crat. 384c-d καὶ μὴν ἔγωγε ... οὐ δύναμαι πεισθῆναι ὡς ἄλλη τις ὀρθότης ὀνόματος ἢ ξυνθήκη καὶ ὁμολογία (I for my part ... cannot be persuaded that there is any correctness of names other than convention and agreement).

7.

Arist. De Interp. 1624-7 ἔστι μὲν οὖν tà £v τῇ φωνῇ τῶν Ev τῇ ψυχῇ παθημάτων σύμβολα, xai τὰ γραφόμενα τῶν ἐν τῇ φωνῇ᾽ xal ὥσπερ οὐδὲ γράμματα πᾶσι cà αὐτά,

οὐδὲ

φωναὶ

αἱ αὐταί.

(Spoken

words

are the symbols

of the soul’s

experiences; written words are the symbols of spoken words; and just as allraces do not have the same writing, so they do not have the same speech.) 8.

H. Dahlmann

Varro und die hellenistische Sprachtheorie (Berlin 1932) 11.

9.

Dahlmann (n.8) 26-32; R. Schröter ‘Die varronische Etymologie’ Entretiens sur

58

ROBERT MALTBY

l'antiquité classique IX: Varron (Gentve 1963) 79-100; W. Pfaffel Quartus gradus etymologiae: Untersuchungen zur Etymologie Varros in "De Lingua Latina" (Kónigstein im Taunus 1981) 32-5. 10.

J. Collart Varron, grammairien latin (Paris 1954) 211-28; Pfaffel (n.9) 35-37; Cavazza (n.2) 88-97.

11.

Lydus De Mag. 1.5 (p. 11.16f. W) οὐδὲ yàp ἀγνοήσας ὁ Ῥωμύλος, fj of kat αὐτόν, δείκνυται κατ᾽ ἐκεῖνο καιροῦ τὴν Ἑλλάδα φωνήν, τὴν Αἰολίδα λέγω, ὥς φασιν ὅ τε Κάτων ἐν τῷ περὶ ᾿Ρωμαϊκῆς ἀρχαιότητος, Báppov τε ὁ πολυμαθέστατος ἐν προοιμίοις τῶν πρὸς Πομπήϊον αὐτῷ γεγραμμένων, Εὐάνδρου καὶ τῶν ἄλλων ᾿Αρκάδων εἰς Ἰταλίαν ἐλθόντων ποτὲ καὶ τὴν Αἰολίδα τοῖς βαρβάροις ἐνσπειράντων φωνήν. (For Romulus, or his contemporaries, is not shown at that point of time to have been ignorant of the Greek language, | mean the Aeolic, as both Cato in his work On Roman Antiquities and the most erudite

Varro in his Introduction to Pompey state, because Evander and the other Arcadians, when they had gone to Italy in olden times, had disseminated the Aeolic speech among the barbarians. [tr. Bandy]) 12.

Lydus De Mag. 2.13 (p. 69.18f. W) μάρτυς ὁ Ῥωμαῖος Báppov ἐν βιβλίῳ πέμπτῳ περὶ ρΡωμαϊκῆς διαλέκτου, ἐν ᾧ διαρθροῦται, ποία μέν τις λέξις ἐστὶν Αἰολική, ποία δὲ Γαλλική, καὶ ὅτι ἑτέρα μὲν Θούσκων, ἄλλη δὲ Ἑτρούσκων, ὧν συγχυθεισῶν ἡ νῦν κρατοῦσα τῶν Ρωμαίων ἀπετελέσθη φωνή. (The Roman Varro attests this in his work On the Roman Language, in which it is precisely

defined what sort of word is Aeolic and what sort is Gallic; and that a word deriving from the Tuscans is of one sort, while that deriving from the Etruscans is of another, from the blending of which was formed the now prevailing language of the Romans. [tr. Bandy])

13.

Quint. Inst. 1.6.31 continet autem [sc. etymologia) in se multam eruditionem, siue ex Graecis orta tractemus, quae sunt plurima, praecipueque Aeolica ratione (cui est sermo noster simillimus) declinata, sive ex

historiarum ueterum notitia nomina

hominum, locorum, gentium, urbium requiramus.

14.

Collart (n.10) 215f.

15.

Varr. LL 5.21 in Latio aliquot locis dicitur ... non terminus, sed termen. hoc Graeci quod τέρμονα, pote vel illinc. Euander enim, qui uenit in Palatium, e Graecia Arcas.

16.

Varr. LL 5.45 Argeos dictos putant a principibus, qui cum Hercule Argivo uenerunt Romam et in Saturnia subsederunt.

17.

Varr. LL 5.53 quartae regionis Palatium, quod Pallantes cum Euandro uenerunt, qui et Palatini.

18.

Varr. LL 5.101 lepus, quod Siculi, ut Aeolis quidam Graeci, dicunt λέποριν; a Roma quod orti Siculi, ut annales ueteres nostri dicunt, fortasse hinc illuc tulerunt et hic reliquerunt id nomen. For Aelius, cf. Varr. RR 3.12.6 L. Aelius putabat ab eo dictum leporem a celeritudine, quod leuipes esset. ego arbitror a Graeco uocabulo

antico, quod eum Aeolis λέποριν appellabant. Cf. Quint. Inst. 1.6.33; Gell. 1.18.2.

19.

Q. Fabius Pictor (HRF 2 Peter; GRF Funaioli 7.1) Fabius a Siculis profectos corrupto nomine Vulscos ait dictos. L. Cassius Hemina (HRF 2 Peter; GRF Funaioli 17.2) notum est ... constitutam ... Ariciam ab Archilocho Siculo, unde et

nomen, ut Heminae placet, tractum. L. Cassius Hemina (HRF 3 Peter; GRF Funaioli 17.3) Cassius Hemina tradidit, Siculum quendam nomine uxoris suae Clytemestrae ictum.

20.

condidisse

Clytemestrum,

mox

corrupto

nomine

Crustumerium

Varr. LL 5.120 a capiendo catinum nominarunt, nisi quod Siculi dicunt xátwov ubi assa ponebant; 5.173 in argento nummi, id ab Siculis; 5.175 dos, si nuptiarum causa

VARRO'S ATTITUDE TO LATIN DERIVATIONS FROM GREEK

59

data: haec Graece δωτίνη: ita enim Siculi, 5.179 mutuum, quod Siculi pottov.

21.

Varr. GRF Funaioli 357.417 (= Plutarch Roman

τοὺς παῖδας Βάρρωνα

τοῖς περιδεραίοις, ἃ βούλλας

Questions 101) διὰ τί κοσμοῦσι

καλοῦσι; ... ὃ μὲν yàp ol περὶ

λέγουσιν, οὐ πιθανόν ἐστι, τῆς βουλῆς βόλλας προσαγορευομένης,

τοῦτο σύμβολον εὐβουλίας περιτίθεσθαι τοὺς παῖδας. Varr. LL 5.25 unde sumi pote, puteus. nisi potius quod Aeolis dicebant ut πύταμον sic πύτεον a potu, non ut nunc φρέαρ; 5.102 malum, quod Graeci Aeolis dicunt μᾶλον; 5.175 ab eodem donum: nam Graece ut δόνειον.

22.

E.g. Varr. LL 6.6 quod nocet, nox, nisi quod graece νύξ nox. Cf. 5.120 (n.20 above).

23.

Pfaffel (n.9) 35.

24.

PI. Crat. 409d εἴ τις ζητοῖ ταῦτα κατὰ τὴν Ἑλληνικὴν φωνὴν εἰκότως κεῖται, ἀλλὰ μὴ κατ᾽ ἐκείνην ἐξ ἧς τὸ ὄνομα τυγχάνει ὄν, οἶσθα ὅτι ἀποροῖ ἄν. (If we should try to demonstrate the fitness of those words in accordance with the Greek language, and not in accordance with the language from which they are derived, you know we should get into trouble.)

25.

Pl. Crat. 425e-426a ἢ ἐκεῖνος, ὅτι παρὰ βαρβάρων τινῶν αὐτὰ παρειλήφαμεν: ... αὗται γὰρ ἂν πᾶσαι ἐκδύσεις εἶεν. (Or what of the idea that we got the earliest names from some foreign folk? ... All these are merely very clever evasions.)

26.

Gell.

16.12.5f.

Cloatius

Verus in libro quarto uerborum a Graecis tractorum

"faenerator"' inquit "appellatus est quasi φαινεράτωρ, ἀπὸ τοῦ φαίνεσθαι ἐπὶ τὸ χρηστότερον ... idque dixisse ait Hypsicraten: ... nihil potest dici insulsius. "faenerator" enim, sicut M. Varro in libro tertio de sermone Latino scripsit, “a

faenore est nominatus".

27.

Varr. LL 5.3 neque omnis origo est nostrae linguae e uernaculis uerbis; 5.10 igitur quoniam in haec sunt tripertita uerba, quae sunt aut nostra aut aliena aut obliuia, de nostris dicam cur sint, de alienis unde sint, de obliuiis relinquam: quorum partim quid tamen inuenerim aut opiner scribam.

28.

Varr. LL 5.77 aquatilium uocabula animalium partim sunt uernacula, partim peregrina (all Greek); 5.100 ferarum uocabula item partim peregrina (Greek, Armenian, Syrian, Lucanian); 5.103 quae in hortis nascuntur, alía peregrinis uocabulis (Greek, contrasted with uernacula in 104); 5.167 in his [cloaks and blankets] muita peregrina (Greek, Gallic). Collart (n.10) 221.

Chrysippus SVF 2.895 οὕτως δὲ xai τὸ ἐγὼ λέγομεν κατὰ τοῦτο, δεικνύντες αὐτοὺς ἐν τῷ ἀποφαίνεσθαι τὴν διάνοιαν εἶναι, τῆς δείξεως φυσικῶς καὶ οἰκείως ἐνταῦθα φερομένης. Cf. Dahlmann (n.8) 8f. 3.

Cic. DND 2.64 Κρόνος ... dicitur, qui est idem χρόνος, id est spatium temporis. Saturnus autem est appellatus quod saturaretur annis; ex se enim natos comesse fingitur solitus, quía consumit aetas temporum spatia annisque praeteritis insaturabiliter expletur.

32.

Quint. Inst. 1.6.36 ingeniose uisus est Gauius caelibes dicere ueluti caelites, quod onere gravissimo uacent, idque graeco argumento iuuit; ἠιθέους enim eadem de causa dici adfirmat.

33.

Macrob. Sat. 1.17.7 Chrysippus Apollinem, ὡς οὐχὶ τῶν πολλῶν καὶ φαύλων οὐσιῶν τοῦ πυρὸς ὄντα, primam enim nominis litteram retinere significationem negandi, fj ὅτι μόνος ἐστί καὶ οὐχὶ πολλοί, nam et Latinitas eum, quia tantam claritudinem solus obtinuit, solem uocauit.

ROBERT MALTBY

34.

I suspect that Catullus is playing on this etymology at 64.298f. inde pater diuum

sancta cum coniuge natisque/ aduenit, caelo te solum, Phoebe, relinquens. This would give added significance to the juxtaposition of solum and Phoebe in the reference to Apollo's absence from Thetis's wedding. 35.

Cf. Varr. LL 5.68 luna ... quod sola lucet noctu with Et. Mag. σελήνη παρὰ τὸ σέλας νέον ἔχειν, ἢ παρὰ τὸ σέλας ἀεὶ £v ἔχειν; 6.49 [memoria] a manendo ut

manimoria potest esse dicta with Et. Gen. μνήμη παρὰ τὸ μένω μενήμη καὶ κατὰ συγκοπὴν μνήμη. Cf. R. Reitzenstein M. Terentius Varro und Johannes Mauropus von Euchaita. Eine Studie zur Geschichte der Sprachwissenschaft (Leipzig 1901) 35f.; Dahlmann (n.8) 20f.

36.

Most recently Daniel J. Taylor Declinatio: A Study of the Linguistic Theory of Marcus Terentius Varro (Amsterdam 1974).

37.

Taylor (n.36) 117.

38.

Hor. AP 52ἴ. et noua fictaque nuper habebunt uerba fidem, si/ Graeco fonte cadent parce detorta.

39,

Tib. 1.10.2 quam ferus, et vere ferreus, ille fuit; Ov. Am. 3.9.4f. flebilis indignos, Elegia, solue capillos:/ a, nimis ex uero nunc tibi nomen erit.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 61-83 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

PHILETAS AND ROMAN

POETRY

PETER E. KNOX In memoriam John William Knox

The “poetic triumph of Callimachus" was shared with his older contemporary, Philetas of Cos, a poet and a critic of the immediately

preceding generation.'

In ancient literary canons Philetas keeps

company with Callimachus as the second leading elegist of his day; thus, for example, Photius Bibliotheca 319b, citing Proclus: λέγει δὲ kai ἀριστεῦσαι τῷ μέτρῳ Καλλῖνόν τε τὸν Ἐφέσιον xai Μίμνερμον τὸν Κολοφώνιον, ἀλλὰ καὶ τὸν Τηλέφου Φιλήταν τὸν Κῷον καὶ Καλλίμαχον τὸν Βάττου, Κυρηναῖος δ᾽ οὗτος ἦν.2 The two representatives of archaic elegy are balanced by a pair from the ‘modern’ period of Hellenistic Greece. References to the poetry of Philetas

by

his

contemporaries,

although

few

in

number,

are

uniformly flattering. Given the large number of writers in this genre during this period, the elevation of Philetas to equivalence with Callimachus is telling: his influence on the succeeding generation of poets must have been considerable. The almost complete loss of his work is generally recognized as a serious gap in our knowledge of the formative period of Hellenistic poetry; it once prompted Wilamowitz to remark on the rather meagre remains of his poetry, "Man móchte mehr von ihm lesen, ein Wunsch, den Hermesianax nicht erweckt.'?

The wish has not been granted. J.U. Powell's Collectanea Alexandrina, published in 1925, contains only 26 fragments attributed to Philetas, consisting of 48 partial or complete lines of verse. Only seven additional scraps are to be found in the Supplementum Hellenisticum, and of these fragments, four had already been included in the separate edition of Philetas by W. Kuchenmüller.* We can be certain of the titles of only two individual works, possibly three, in addition to collections of slighter verse, epigrams and paegnia, to which half of 61

62

PETER E. KNOX

the surviving fragments can be assigned.? This is not much to go on in assessing the importance of a writer who is commonly regarded as "the precursor of much that came subsequently to be regarded as fundamentally characteristic of Hellenistic poetry." His importance as a precursor of Roman elegy might also seem obvious: Philetas is invoked by Propertius more often than any other poet except Callimachus, with whom he is usually paired, and when

Ovid mentions Philetas it is also almost always in Callimachus' company.’ It is often easy to assume that the recovery of some lost work might explain a great many mysteries in our surviving texts, but it is especially difficult to resist the temptation to believe that the disappearance of Philetas' poetry has left a gaping hole in the literary record for students of Latin literature. Many are prepared to resist, however, and modern scholarship appears close to reaching a

consensus

that, in spite of the well documented importance of

Philetas for his Hellenistic contemporaries, his works vanished not long after the third century BC. Or as one recent critic puts it, “it is not easy to believe that Propertius and Ovid knew much about Philetas."? This is an idea worth challenging, not only because it is inherently improbable that two different Roman poets would cite as an authority a poet that neither they nor any of their contemporaries had ever read, but also because it is based on a view of Philetas that

was formed in the earliest stages of the criticism of Hellenistic poetry in this century before the recovery of a number of relevant texts preserved in fragmentary form on papyri. Some of the broader claims for Philetas' influence may have to be dismissed, or at least suspended, but the picture that emerges may be all the clearer for being more concise. 1. The First Epyllion Philetas was best known in antiquity as an elegist, but the biographical notice in the Suda informs us that he wrote “epigrams, elegies, and other works" (ἔγραψεν ἐπιγράμματα xai ἐλεγείας καὶ ἄλλα). Among these other works was a hexameter poem known as the Hermes (fr. 5-9 Powell, 1-4 K). It is commonly believed that we

know a great deal about the subject of this poem, information which derives for the most part from the Erotika Pathemata of Parthenius, and substantial claims for the influence of Philetas have been based on it. Parthenius, as is well known, collected a number of otherwise obscure tales and presented them in prose in summary form for the

PHILETAS AND ROMAN

POETRY

63

convenience, as he says (οἱονεὶ ... ὑπομνηματίων τρόπον, Praef. 2),

of his Roman reader. The collection as a whole is dedicated to the poet Cornelius Gallus and designed for use in hisown work (αὐτῷ τέ σοι παρέσται εἰς ἔπη καὶ ἐλεγείας ἀνάγειν τὰ μάλιστα ἐξ αὐτῶν ἁρμόδια, Praef. 2). Most of Parthenius’ stories are accompanied by marginal notes which seem to indicate the sources of the account offered by him; the notation on the second entry assigns it to the Hermes of Philetas (Erotika Pathemata 2): Ὀδυσσεὺς

ἀλώμενος

περὶ Σικελίαν

καὶ τὴν Τυρρηνῶν

xai τὴν

Σικελῶν θάλασσαν, ἀφίκετο πρὸς Αἴολον καὶ Μελιγουνίδα νῆσον,

ὃς αὐτὸν κατὰ κλέος σοφίας τεθηπὼς ἐν πολλῇ φροντίδι eye: τὰ περὶ Τροίης ἅλωσιν καὶ ὃν τρόπον αὐτοῖς ἐσκεδάσθησαν αἱ νῆες κομιζομένοις ἀπὸ τῆς Ἰλίου διεπυνθάνετο, ξενίζων τε αὐτὸν πολὺν χρόνον διῆγεν. τῷ δ᾽ ἄρα καὶ αὐτῷ ἦν ἡ μονὴ ἡδομένῳ' Πολυμήλη γὰρ τῶν Αἰολίδων τις ἐρασθεῖσα αὐτοῦ κρύφα συνῆν. ὡς δὲ τοὺς ἀνέμους ἐγκεκλεισμένους παραλαβὼν ἀπέπλευσεν, ἡ κόρη φωρᾶταί τινα τῶν Τρωϊκῶν λαφύρων ἔχουσα καὶ τούτοις μετὰ πολλῶν δακρύων ἀλινδουμένη. ἔνθα δὴ ὁ Αἴολος τὸν μὲν Ὀδυσσέα καίπερ οὐ παρόντα ἑκάκισεν, τὴν δὲ Πολυμήλην ἐν νῷ ἔσχε τίσασθαι. ἔτυχε δὲ αὐτῆς ἠρασμένος ὁ ἀδελφὸς Διώρης, ὃς αὐτὴν παραιτεῖταί τε καὶ πείθει τὸν πατέρα αὑτῷ συνοικίσαι.

Broad-ranging conclusions have been drawn from this summary about the content and manner of the Hermes, about the nature of Philetas’ poetry as a whole, and about his influence on Latin literature. Philetas' editors characterise the Hermes as an epyllion with an adventure of Odysseus as its subject,? and this view has become received opinion. Extravagant claims have been, and continue to be made, for this reconstruction: R. Heinze asserts that

the Hermes served as a formative influence on Virgil in the fourth book of the Aeneid and that Virgil modelled his account of the affair of Dido and Aeneas on Polymele and Odysseus.!! Kuchenmiiller adds that Virgil was able to make this connection because his friend Cornelius Gallus probably lent him a copy of Parthenius.'? A.W. Bulloch summarises the plot of the Hermes, compares its structure with Callimachus' Hecale, and suggests that Apollonius was influenced by it in his portrayal of Medea." All of these imaginings, however, are based upon a view of Parthenius that has been rendered untenable by the accrual of new evidence in this century. A casual reading of the Erotika Pathemata in a modern edition may easily leave the impression that the notations of source provided for the stories derive from Parthenius himself. Editors print these notations as chapter headings, but they are actually to be found in the

64

PETER E. KNOX

margins of the single manuscript that preserves the text of Parthenius and the Heteroioumena of Antoninus Liberalis, which is similarly annotated. The attribution of these notations to Parthenius was challenged long ago by R. Hercher,'* and a number of hypotheses have been advanced over the years to explain their genesis. The discovery in 1935 of a papyrus fragment of the Thrax of Euphorion has confirmed the intuition of G. Pasquali that they are the work ofa later annotator, not the author.'® In Powell's Collectanea Alexandrina seven fragments are assigned to this lost poem by Euphorion: seven lines of verse and two narratives from Parthenius' collection where Euphorion's poem is indicated as source. They relate the story of Harpalyce (Erotika Pathemata 13), who served to her husband Clymenus a meal consisting of their own son, and Apriate (Erotika Pathemata 26), whom Trambelus tried to rape. With this evidence and nothing more it once seemed reasonable to assume that Euphorion's Thrax was a collection of mythological tales of the type that we know best from Ovid's Metamorphoses, a view apparently not relinquished in recent surveys despite the evidence of the new fragment.! For it now appears that Euphorion's Thrax was a cursepoem like Ovid's Ibis, consisting of a string of brief exempla, not long

narratives.!5 The significance of this discovery for the present case lies in the fact that among the exempla offered in the new fragment are Harpalyce and Apriate, dealt with in the space of five (SH 413.12-16) and seven lines (SH 415.12-18) respectively and in a manner that bears little resemblance to Parthenius' accounts.!? The annotator's indications of Parthenius' ‘sources’ are not false, but without further corroborating evidence they cannot be taken by themselves as evidence that Parthenius is providing an epitome of a longer narrative.” Parthenius did not compose the Erotika Pathemata as an index to Hellenistic poetry, but as a collection of interesting stories, He should be taken at his word when he states the advantages of using his own collection over tracking down the myths found in earlier poetry, which could be allusive and obscure (μὴ αὐτοτελῶς λελεγμένα, Praef. 1), and his summaries should be used only with great caution in reconstructing earlier narratives.

Only two fragments of actual verse survive from the Hermes because they were cited by Stobaeus, and they have no clear connection with the love story of Odysseus and Polymele. One refers to a journey to the underworld (fr. 6 P = 2K), ἀτραπὸν εἷς ᾿Αίδαο ἥνυσα, τὴν οὔπω τις ἐναντίον ἦλθεν ὁδίτης.

PHILETAS

AND

ROMAN

POETRY

65

The other speaks of the compulsion of necessity (fr. 8 P = 3 Κ): ἰσχυρὰ γὰρ ἐπικρατεῖ ἀνδρὸς ᾿Ανάγκη, f| ῥ᾽ οὐδ᾽ ἀθανάτους ὑποδείδιεν, οἵ t' ἐν Ὀλύμπῳ ἔκτοσθεν χαλεπῶν ἀχέων οἴκους Exdpovto.

These lines, it is confidently asserted, belong to the account of his wanderings and visit to the underworld with which Odysseus entertained Aeolus, and which made such an impression upon his

daughter that she fell madly in love with him. But without the evidence for Philetas in Erotika Pathemata 2, it would be impossible to draw such conclusions. Indeed, without the benefit of Parthenius' testimony it is unlikely that any reader would connect fr. 6 P (7 2 K) with Odysseus, through a recollection of Odyssey 10.490f. ἀλλ᾽ ἄλλην χρὴ πρῶτον ὁδὸν τελέσαι καὶ ἱκέσθαι, εἰς ᾿Αίδαο δόμους. But it might have been connected with a passage in the Homeric Hymn to Hermes (572) alluding to the god's function as psychopomp: olov δ᾽ εἰς ᾿Αίδην τετελεσμένον ἄγγελον εἶναι. In short, the title of the poem, which must remain a puzzle for an epyllion about

Odysseus,?! would make sense for a narrative poem, or a hymn, about Hermes. In a later generation, Eratosthenes composed a poem of the same title, apparently containing “an account of the childhood of Hermes which included some amusing and charming episodes, the story of his ascent into the skies, and a description of the universe ... as the god saw it."?? It is of course not possible to disregard entirely the evidence provided by the scholiast to Parthenius. It may not be pure coincidence that we know of a different Polymele, the daughter of Phylas, who was loved by Hermes, as recounted by Homer (Iliad 16.180-6) in a brief genealogical digression.? Two explanations suggest themselves: first, the scholiast may have erred in citing Philetas, having recalled a mention

of this Polymele

in Philetas'

poem. That is perhaps unlikely since the source indications in Parthenius have not yet been shown to be entirely wrong in any given instance. Secondly, it is possible that Philetas did indeed refer to Odysseus' affair, even if only briefly, in a digression triggered by the name of Polymele. What will not do, however, is simply to assume the sort of epyllion that might have had a profound influence on the development of Hellenistic and Roman narrative poetry. In any case, it is probably misguided to look for evidence of a great impact by this particular poem, for whatever else it may have been, it was certainly composed in hexameters, and it was as an elegist that Philetas made his mark: that is the only context in which later poets refer to him by name. The question remains what kind of elegy Philetas wrote.

66

PETER E. KNOX

2. For the Love of Bittis

No aspect of Philetas' career has attracted as much speculation from scholars as the possibility that he composed love poetry in elegiacs that served as a model for Roman love elegy. Nothing remotely resembling such poetry survives from the hand of Philetas or any other Hellenistic poet, but the fact that both he and Callimachus are

invoked so often by the Romans continues to invite speculation about works that appears a fragment philosophers

that have been lost. Attention first focuses on a poem to be attested by his pupil Hermesianax of Colophon, in of his Leontion containing a catalogue of poets and who have suffered in love (fr. 7.75-8 P):

οἶσθα δὲ xai τὸν ἀοιδόν, ὃν Εὐρυπύλου πολιῆται Κῷοι χάλκειον στῆσαν ὑπὸ πλατάνῳ Βιττίδα μολπάζοντα θοήν, περὶ πάντα Φιλήταν ῥήματα καὶ πᾶσαν τρυόμενον λαλιήν.

These lines refer to a statue erected by the citizens of Philetas’ native Cos which represents the poet singing of the ‘volatile’ (dorjv) Bittis, the woman he loved, presumably his wife.?* That the reference is toa specific poem called Bittis is to be inferred from the testimony of Ovid, who draws a comparison with the Lyde of Antimachus (Tristia

1.6.1-3):2° nec tantum Clario Lyde dilecta poetae, nec tantum Coo Bittis amata suo est, pectoribus quantum tu nostris, uxor, inhaeres.

We know that the Lyde was composed by Antimachus to console himself for the death of his wife, after whom he titled his work.26 And we have the testimony of Athenaeus (13.597B) that the Leontion,

which mentions Antimachus' Lyde as well, was the same sort of work. Ovid's poem is addressed to his own wife, and it is reasonable to infer that he refers to the Bittis because it was addressed to Philetas’ wife. There is thus every reason to believe that Philetas wrote a poem named after the woman he loved; the pressing question

is whether it influenced Latin love elegy.?' This question must be treated with some humility, for without further evidence it can never be answered definitively. Although no surviving verses can be attributed to the Bittis, it is

clear that it must have been an elegy. The Lyde, the Leontion, and the Nanno of Mimnermus, the supposed forerunner of such poems, were all elegies and part of the point of Ovid's juxtaposition of the Bittis

and the Lyde in an elegy addressed to his wife is that he is referring to

PHILETAS AND ROMAN

POETRY

67

earlier works in that genre. Antimachus’ Lyde and Hermesianax’ Leontion seem to have been cut from the same cloth. We know little about the reception of the Leontion, but the Lyde provoked much

comment in the third century, most of it positive.?* But Callimachus" famous assessment is unequivocal: Λύδη kai παχὺ γράμμα καὶ οὐ

topóv (fr. 398 Pf.). Few readers?? would think unfair a similar verdict on the 49 couplets of the Leontion preserved by Athenaeus (13.597B, — fr. 7 P), a catalogue of poets and philosophers in love that borders on the absurd. It would be useful to know what Callimachus thought of Philetas' effort in this area. Hermesianax was not the only disciple to express admiration for Philetas. From the so-called Florentine Scholia to the Aetia, published after Kuchenmüller's

edition of Philetas, we learn that

Callimachus praised one of his poems in the Prologue (/r. 1.9f. Pf.):°° Kóiog οὐκ ἄ]ρ᾽ Env ὀλιγόστιχος᾽ ἀλλὰ καθέλκει "m πο]λὺ τὴν μακρὴν ὄμπνια OGecuogópo[c

Almost every point of interpretation in this passage is controversial: which poets, and what poems by them are being compared, and why? The simplest solution is provided by the Florentine Scholia which inform us that in this couplet and the one following it short poems by Philetas and Mimnermus are being compared to their longer works, to the disadvantage of the latter: παρατίθεται ἐν συγκρίσει τὰ ὀλίγων στίχων ὄντα ποιήματα Μιμνέρμου Tod Κολοφωνίου Kai Φιλίτα τοῦ Κῴου βελτίονα τῶν πολυστίχων αὐτῶν φάσκων elvat?! The identification of one poem as Philetas’ Demeter, designated here as ὄμπνια Θεσμοφόρος, has generally been accepted by scholars.?? But the identity of the objectionable longer poem to which it is compared has seemed to hinge on the choice of supplement to fill the short lacuna at the beginning of line 10. A number of proposals have been made, none convincing: for example, δρῦν, presumably a lost

poem by Philetas with that title? Κῶν, a foundation poem by Philetas on his home island;?* or θεῦν, an unknown poem by Philetas

about another

goddess.?

One

problem

shared by all of these

proposals is that they must postulate the existence of an otherwise unattested work. That is perhaps unnecessary. The poem in question was surely an elegy to match the elegiac Demeter: the works of Mimnermus that are being compared were certainly elegies, the only genre in which he wrote, and it is a logical inference that the same holds true for Philetas. The only other elegy by Philetas of which we have any knowledge is the Bittis, and in other respects it fits the bill here: a ‘tall lady’ to match ἡ μεγάλη ... γυνή

68

PETER E. ΚΝΟΧ

(12) of one of Mimnermus' books.” To accommodate her here, Gallavotti proposed γραῦν to fill out the line-opening, although this seems a bit inappropriate as a reference to the poet's wife. Whatever the missing word might be, the best candidate for Philetas' bad long poem remains the Bittis, which stood firmly in the tradition of Antimachus' Lyde. And the Aetia Prologue may not have been Callimachus' only reference to the poem. Tantalising by its brevity, fr. 532 Pf. seems to make polemical reference to a poem by Philetas: τῷ ἴκελον τὸ γράμμα τὸ Kóiov. Pfeiffer notes the likelihood that a “Coan ypdéppa” surely alludes to a poem by Philetas, but it is frustrating not to know what it is being compared to. Pfeiffer suggested the delicate weaving of Coan fabric,** but another poem is

perhaps more likely.?? No sure answer can be reached, but the telltale word γράμμα, used by Callimachus to characterise the noxious Lyde (fr. 398 Pf.), opens the possibility that the comparison is not flattering to Philetas, and this scrap may then belong to the same controversy over the relative worth of literary predecessors that lies

behind Aetia 1.9-12.* If the Bittis was the long poem of Philetas censured by Callimachus, the enthusiasm for making it out to be a precursor for Latin love elegy should abate. Other attempts to make Philetas into a love poet are based upon ingenious, but far-fetched reconstructions. Recently there have been attempts to reconstruct some form of erotic pastoral for Philetas, based on the hypothesis that the character named Philetas in Longus' Daphnis and Chloe represents in some way the Hellenistic poet.* The case rests on parallels between Longus, the Roman elegists and Theocritus, some of them real. A common source may well lurk behind a few, but it requires too great a leap of faith to attribute these to an otherwise unattested pastoral by

Philetas.*? The name of Philetas will always be an attractive peg on which to hang a lost work.*? Enough survives of the wreck of Hellenistic poetry to suggest that Latin love elegy was not created ina vacuum. In addition to the contribution of narrative elegy and epigram, there was monologue poetry about amatory situations. The eleventh /dyll of Theocritus is one surviving example, and doubtless there were others;*‘ but the evidence does not permit us to father one on Philetas yet. No Roman poet invokes the Bittis as a model of style or subject: it is only Ovid who follows Hermesianax in singling out this work because he happens to have been the only elegist to address

poetry to his wife.

PHILETAS AND ROMAN

POETRY

69

3. In Search of Demeter

Four fragments, amounting to only ten lines of elegiac verse, survive

from

Philetas’

Demeter.“

All

are

preserved

as quotations

by

Stobaeus, and the gnomic character of his selections leaves much

room for speculation about the context of the poem from which he or his source excerpted. Assessments of the prospects for reconstructing

the poem vary from the wildly optimistic to extreme scepticism.* One of the fragments preserved by Stobaeus represents the complaint of a person suffering pain (fr. 1 P Ξ 5 K), while two of the remaining fragments (fr. 2P Ξ frr. 7f. K) represent words of consolation addressed to someone suffering from grief. Another quotation, less certainly attributed to the Demeter (fr. 3 P = 6 K), tells of wandering over land and sea. On these points scholars are in agreement; what cannot be proved is the assertion that the grieving individual is Demeter, mourning for her lost daughter. But neither can it be ruled out of the question.*' An extravagant effort to identify the Demeter as

the source for Ovid's two accounts of the rape of Persephone in the Metamorphoses (5.341-661) and the Fasti (4.417-620) was made by C. Cessi in 1908, following up on a suggestion by Maass.* The sweeping claims made by him were convincingly refuted by Kuchenmüller,* and Philetas subsequently disappeared from investigations into sources for Ovid's narrative. In his place the names of Nicander and Callimachus have been offered alternately as authors of the lost

Hellenistic narrative posited by scholars.°® It is neither possible nor desirable to rehabilitate Cessi's detailed reconstruction of the poem, but with the addition of new evidence a new hearing is warranted,

even at the risk of appearing to indulge in what the most recent study of this problem characterises as an "obsessive concern with the

reconstruction

of lost poetry".!!

Since neither Callimachus nor

Nicander can be shown on independent evidence to have written on the subject of Demeter and Persephone, speculation about the form such a narrative might have taken could rightly be considered as indulging in fantasy. On the other hand, speculation about a poem known to have existed is a legitimate concern of the literary historian, and we know that Philetas devoted an elegy to the goddess. We also know that he was not the only poet of his time to do so. The speculations of Maass and Cessi took place within a historical vacuum, and were founded upon the unjustifiable assumption that since Ovid's account of the rape of Persephone disagrees in some details with the one found in the Homeric Hymn, it must be based exclusively upon a narrative of the myth found in some other Greek

70

PETER E. ΚΝΟΧ

poem, presumably of Hellenistic date, now lost. Because Philetas was

the only name that they could attach to a poem about Demeter in

that period, it therefore stood to reason that this was the poem that Ovid adopted as model. This but the conclusions about grounds. The publication of landscape since these issues

simplistic argument must be rejected, Philetas might be justified on other new texts from papyri has altered the were last aired, and they may now be

used to sketch in the picture. The popularity of the narrative hymn as a

literary form is well

documented for the Hellenistic period.?? For earlier scholars appraising the value of Philetas, the only evidence adduced was the hymns of Callimachus, but other poems surviving from this period fit the description, for example the twenty-second Idyll of Theocritus,

commonly known as the Hymn to the Dioskouroi. A number of other poems, known to us only as titles, reflect the popularity of the genre: an Apollo in hexameters by Simias of Rhodes,5* and a poem with the same title by Alexander of Aetolia in elegiacs.*> A Dionysus

by Euphorion,’® a Hermes by Eratosthenes, and hymns to Zeus by Cleanthes,°® to Poseidon by Moero (fr. 5 P), and to Pan by Aratus (SH 115) make up only a partial list. Two new entries to the field can be dated to this period. A third century papyrus preserves substantial portions of a narrative about Demeter written in an unusual meter, the choriambic hexameter (SH 678-80). Its author is identified as Philicus, known from Hephaestion as author of a poem on Demeter

composed in that meter. The work was relatively long — parts of at least 62 lines survive — and plainly aimed at a sophisticated audience, as the poet boasts in his proem (SH 677): καινογράφου συνθέσεως τῆς Φιλίκου, γραμματικοί, δῶρα φέρω πρὸς byac.© Apparently it went on to provide an account of Demeter's search for Persephone: the most intelligible fragments of the hymn (SH 680.51-62) introduce Iambe as an old rustic who makes Demeter laugh.*' Perhaps even more suggestive than learned works such as this one addressed to an intellectual elite is the opening of a hymn to Demeter found in another third-century papyrus (SH 990). It preserves the first eleven lines of an anonymous poem that is also distinguished by its use of an unusual meter, this time couplets of dactylic hexameter and tetrameter. After an opening invocation, the poet relates how Poseidon, Zeus and Hades cast lots for their realms (3-10), before the papyrus breaks off. “The poem may have gone on to tell how Pluto stole Persephone, and how Demeter sought her", notes Page,°? a guess, but not an improbable one considering that

PHILETAS AND ROMAN

POETRY

7

Hades has already been introduced in the extant portion of the text. The author was not a great poet, but he was not ignorant: he knew

both the Homeric Hymn to Demeter and Callimachus.® The story was popular. In this climate there is room to accommodate a poem by Philetas. The disclaimer by Callimachus in the opening of his own Hymn to Demeter seems to imply the existence of other narratives. Just as the hymn seems to be settling into an account of Demeter's loss of Persephone and her fruitless search for her daughter (8-14), Callimachus breaks off (Hymn 6.17): μὴ μὴ ταῦτα λέγωμες ἃ δάκρυον ἄγαγε Δηοῖ. The device of such a disclaimer is as old as Pindar,™ but it is at least

possible that Callimachus

uses the old formula as an oblique

reference to another contemporary work as he moves on to relate a

different, more obscure myth of the goddess. In this charged atmosphere we may now return to the only contemporary testimony to Philetas’ Demeter, in the Aetia Prologue (fr. 1.9-12 Pf.): Κώϊος οὐκ ἄ]ρ᾽ Env ὀλιγόστιχος᾽ ἀλλὰ καθέλκει ven πο]λὺ τὴν μακρὴν ὄμπνια Θεσμοφόροϊ[ς" τοῖν δὲ] δυοῖν Μίμνερμος ὅτι γλυκύς, afi κατὰ λεπτὸν ῥήσιες] ἡ μεγάλη δ᾽ οὐκ ἐδίδαξε γυνή.

There is one constant in all hypotheses about this passage: the poem indicated by ὄμπνια Θεσμοφόρος and designated as superior is the Demeter of Philetas. If later poets knew any work by Philetas, it would certainly be the one singled out as exemplary by Callimachus in the Aetia Prologue, the best known and most often imitated piece

of poetic doctrine from Alexandria. This hypothesis would receive further support if it proved possible to add to the testimony another compliment to the Demeter by a poet of Callimachus' generation. Theocritus certainly did admire Philetas, and unlike Callimachus, his praise is not qualified. In /dyl/ 7 the narrator Simichidas, who is probably to be identified with Theocritus, modestly declines to be ranked as a poet with Sicelidas and

Philetas (Jdy/l 7.39-41): οὐ γάρ πω κατ᾽ ἐμὸν νόον οὔτε τὸν ἐσθλὸν Σικελίδαν νίκημι τὸν ἐκ Σάμω οὔτε Φιλήταν ἀείδων, βάτραχος δὲ ποτ᾽ ἀκρίδας dc τις ἐρίσδω.

The literature that has grown up on these lines is prodigious. The entire poem has come to be regarded as a sophisticated and allusive

72

PETER Ε. ΚΝΟΧ

meditation on Theocritean pastoral and contemporary poetics, and a

detailed examination of the wide array of interpretations of this puzzling poem

is beyond

the scope of this paper.

But a brief

consideration of the probable rationale for invoking Philetas is in order. About the second poet paired with him, Sicelidas, there has

been little controversy: he is plausibly identified by the scholiast as the epigrammatist (ἐπιγραμματογράφον) Asclepiades of Samos.55 In spite of the fact that Asclepiades' reputation among later generations depended largely upon his epigrams, it seems unlikely that he is cited here only, or even primarily, because of his accomplishments in that genre. No trace of his longer works survives, and it is curious that his appearance here has provoked no speculation on them, perhaps because, unlike Philetas, he was never linked to Callimachus.*’ He presumably did compose lyric poems in the meter that bears his name, but they do not survive.** He may also have written hymns

about

the gods as we learn from Tzetzes.

It is probably his

contemporary fame in such fields, as well as epigram, that justifies his

appearance here with Philetas."? In contrast with Sicelidas, the invocation of Philetas in this Idyll

has given rise to a great deal of conjecture, particularly about his lost and unattested contributions to the development of bucolic poetry.”! Such speculations may one day be proved to be true by the discovery of new papyri, but at present there is simply no evidence to support the hypothesis that Philetas (or Asclepiades) wrote bucolic poetry. A

simpler interpretation of this enigmatic passage may be closer to the truth: Asclepiades and Philetas are invoked only as outstanding poets of the preceding generation. This does not mean that they were chosen casually — such choices are clearly important for poets — but it suggests that even if we were to recover all of the poetry of Philetas and Asclepiades, this 7dyl/ would still make better sense as a self-

contained work of art.” An acknowledgement by Theocritus of his debt to Philetas is likely to have been far less elaborate than has been imagined, rather more like Virgil’s allusions to Callimachus in the

Eclogues than his treatment of Cornelius Gallus. The connection between the setting of Idyll 7 at a local Coan festival to Demeter with Philetas' most famous elegy is the most plausible suggestion

made by E.L. Bowie in his ingenious study of this poem,” but it leads to a different conclusion about the work being acknowledged

by Theocritus. The cult of Demeter was well established on Cos,” and it has

been plausibly argued that Philetas’ Demeter had a Coan setting."? It

PHILETAS AND ROMAN

POETRY

73

is not difficult to conceive of how Philetas might have introduced

Coan antiquities into a poem about Demeter. In the opening of Theocritus /dyll 7 the narrator tells of how he made his way to the festival where offerings were being made to the goddess by Phrasidamas and Antigenes, two Coan luminaries whom he describes as follows (Jdyll 7.4-6): δύο τέκνα Λυκωπέος, εἴ τί περ ἐσθλὸν

χαῶν τῶν ἐπάνωθεν ἀπὸ Κλυτίας τε καὶ αὐτῷ Χάλκωνος, Βούριναν ὃς ἐκ ποδὸς ἄνυε κράναν.

The scholiast on this passage provides information about the sons of

Eurypalus and Clytia, Chalcon and Antagoras: οὗτοι δέ εἰσιν οἱ ἐπὶ τὴς Ἡρακλέους πολιορκίας τὴν Κῶ κατοικήσαντες Kal ὑποδεδεγμένοι τὴν Δημήτραν καθ᾽ ὃν καιρὸν περιήει τὴν Κόρην ζητοῦσα. The wanderings of Demeter in search of her daughter were a convenient device for legitimising local cults to the goddess; and this may have been a connection made by Philetas. This suspicion is deepened by the mention of the spring Burina in line 6, identified by the scholiast as a Coan spring and certified by citation of Philetas (/r.

24 P): νάσσατο δ᾽ ἐν npoxofjot μελαμπέτροιο Βυρίνης. ὁ Further speculation about the content of the Demeter is probably fruitless. The hypotheses advanced by earlier scholars who ‘recon-

structed' the poem from Ovid's versions of the rape of Persephone and Demeter's wanderings were based upon naive assumptions about lost Hellenistic models for Latin poetry and posited the

existence of a source rendered more or less directly into Latin by Ovid. But it is equally naive to dismiss entirely the considerable evidence for the popularity of the myth of Demeter and Persephone in Hellenistic poetry and focus exclusively on the Homeric Hymn as the only relevant text for Ovid's narratives.”’ It is true enough that this is the only text that modern readers have available for comparison, but that says little about Ovid. He certainly knew the earlier hymn, but it is Ovid's manner to demonstrate his knowledge of traditional material through reference to Hellenistic intermediaries,

and it is clear enough that the Homeric Hymn cannot have been his exclusive source. We know that Ovid transformed Callimachus' Hymn to Demeter in the eighth book of the Metamorphoses (725884), adapting Callimachus' story of Erysichthon while also alluding to Hesiod's narrative in the Catalogue."* It is not unreasonable to suggest, and only suggest, a similar engagement with Philetas' Demeter, the work that Callimachus advertised as a precedent for his

own poetry.

74

PETER E. KNOX

4. The Roman Poets

What importance did Philetas have for the Roman poets who invoke him? The evidence thus far has tended to suggest that Philetas' fame in antiquity rested primarily upon the success of his elegiac Demeter. It is time to consider the poets who refer to him by name as a representative of the genre of elegy in which he wrote. We may well suspect that he owed his position of eminence to the fact that he figured so prominently in the Aetia Prologue. It is probably not a coincidence that, although Mimnermus and Callinus are commonly

connected as exemplars of early elegy, only Mimnermus is cited as a precedent, and that only once, by Propertius (1.9.11): plus in amore

ualet Mimnermi uersus Homero. His point is polemical: this elegy is addressed to Ponticus, a poet who writes traditional mythological epic, and Propertius advances the claims of his genre by setting its founder against the archetypal epic poet. It cannot be seriously argued, nor hasit been, that heis advocating imitation of Mimnermus, and the choice of Mimnermus as a balance to Homer is surely influenced by Callimachus' advocacy of one of his books in the

Aetia.” Philetas is invoked together with Callimachus by both Propertius

and Ovid as representatives of Hellenistic elegy when they press the advantages of that genre for the poet in love. Nowhere does either poet actually assert that these authors wrote love poems, although it is an easy inference to draw, and some have been fooled.®° Only once is subject matter rather than genre demonstrably the point of such an allusion, and that is by Ovid, who could consider any poet a love poet

if he put his mind to it (as he did in Tristia 2). In a defence of amatory poetry in the Remedia Amoris, Ovid comments on the subject of generic decorum, restricting epic subjects to the hexameter and amatory topics to the elegiac couplet, using Homer and Callimachus

as representatives of each, much as Propertius had used Mimnermus and Homer in addressing Ponticus (Remedia Amoris 381f.): Callimachi numeris non est dicendus Achilles;

Cydippe non est oris, Homere, tui.

It is clear from this passage that when Ovid calls Callimachus a love poet, he has in mind the Acontius and Cydippe episode in the Aetia, which we know did not resemble the love elegies of the Roman poets,

influential thought it was.?! Philetas’ contribution, like that of Callimachus, appears on the evidence to have consisted of offering a style and a manner to the

PHILETAS AND ROMAN

POETRY

75

Roman elegist. Some references seem to point to specific passages in his work, again very much like the allusions to the opening of Callimachus’ Aetia. In the opening of Propertius’ third book he is located in a sacred grove along with Callimachus (3.1.1f.): Callimachi Manes et Coi sacra Philetae, in uestrum, quaeso, me sinite ire nemus.

And references elsewhere to a spring (/ymphisque a fonte petitis/ ora Philitea nostra rigauit aqua, 3.3.51f.) and to ivy (Philiteis ... corymbis, 4.6.3) appear so specific that we may suppose that they reflect specific passages in the poetry of Philetas. The same may be true of Propertius' only other allusion to Philetas by name. In 2.34b Propertius advises the poet in love to imitate Philetas and Callimachus by writing elegy,*? in a passage printed by most editors as follows (2.34b.31f.): tu fsatius memorem musist imitere Philetan et non inflati somnia Callimachi.

From the context it is clear that the genre of elegy is at issue, not love poetry per se: it is the Aetía of Callimachus that is proposed as a model, signalled by the reference to the dream sequence (fr. 2 Pf.) that followed the Prologue. Many critics have noted that the opening half-line of the hexameter must surely contain a similarly precise reference to the works of Philetas. The corruption of the manuscripts

makes certain recovery impossible,9 but our review of Philetas' career prompts a word in favor of a long despised candidate. One fifteenth-century manuscript reads tu satius Meropem Musis," understanding Meropem as an epithet of Philetas. If this is a conjecture, it is extraordinarily learned, drawing upon the identi-

fication of the Meropes as early inhabitants of Cos.55 Against this reading it can be argued that we require some reference to a poem by Philetas to balance the Aetia of Callimachus in the pentameter. G. Luck has proposed tu satius Musis Meropen imitere Philetae, with

a transposition and a change of case to accommodate such a title.95 A poem dealing with ancient Coan traditions may not be unexpected from

Philetas: as we have seen, there is reason to believe that the

Demeter contained such material. Indeed, the reference might well have been to the poem that was endorsed by Callimachus, as was

intuited by W. Stroh when he proposed tu satius Cererem Musis imitere Philitae.*' If the Demeter did deal prominently with Coan

antiquities, it may well be the poem alluded to in the last reference to Philetas in Roman poetry, by Statius. In the epithalamium for Stella

76

PETER E. KNOX

and Violentilla he calls upon the most famous elegiac poets of the past (Silvae 1.2.250-5): ... sed praecipui, qui nobile gressu extremo fraudatis opus, date carmina festis digna toris. hunc ipse Coo plaudente Philetas Callimachusque senex Umbroque Propertius antro ambissent laudare diem, nec tristis in ipsis Naso Tomis diuesque foco lucente Tibullus.

Each poet is accompanied by a clearly identifiable reference to his

work: Propertius 3.1 is signalled by the cave,®* Ovid's Tristia by the epithet tristis and the reference to Tomi, and Tibullus' first elegy by the hearth.*? Callimachus, too, is invoked with reference to his most famous poem, the Aetia, recalled by the epithet senex, which points to

the Prologue (fr. 1.32-6 Pf.). As A. La Penna has noted,” the rhetorical thrust of this passage requires a similar treatment of Philetas: Coo plaudente cannot simply be intended to identify his homeland, and may well refer to a poem in which the island of Cos

figured prominently?! La Penna conjectured a lost work; among the works that we know of, the Demeter is the best candidate.

References to Callimachus and Philetas by the Roman elegists were not casual or uninformed. As a final example, consider the opening of Propertius! Actium elegy (4.6.1-4): sacra facit uates: sint ora fauentia sacris et cadat ante meos icta iuuenca focos. cera Phileteis certet Romana corymbis,

et Cyrenaeas urna ministret aquas.

Long

an embarrassment

to critics with a distaste for Hellenistic

poetry and the politics of Augustus, the poem has lately benefited from an overdue re-appraisal.?? In the Actium elegy Propertius’ most immediate model is not the long aetiological collection, but Callimachus' hymns and in particular his hymn to Apollo. The poem opens with Propertius in the role of poet-priest, urging proper offerings and excluding the profane. The principal elements of the elegy, the ecphrastic narrative and the address by the god, both have formal parallels in the surviving hymns of Callimachus. The extent of this influence is substantiated by the large number of close verbal imitations of the hymn to Apollo found in this poem, most of which have long been recognised by commentators on the two poets. The reference to Callimachus in line 4 (Cyrenaeas ... aquas) should not be considered casual or uninformed; and the invocation of Philetas here may be accorded the same precision: we know of three major works

PHILETAS AND ROMAN

POETRY

77

by Philetas, two of them, Hermes and Demeter, narrative poems about the gods, and Propertius’ reference is doubtless to that aspect of his work. There is, then, no paradox when Propertius, who had dodged the

obligation to compose patriotic verse by invoking Callimachus and Philetas, calls on the same models when composing his encomium of

Augustus and his victory at Actium.” In the first three books Callimachus and Philetas were models of style as the principal exponents of elegiac verse on lighter subjects; but as Propertius turned to aetiological subjects and encomiastic hymns, Callimachus and Philetas are evoked again for the content of their poetry as well as style. For his amatory verse Propertius names no such model, nor do the other elegists. It is highly unlikely that any subsequent accretion to our store of Callimachean or Philetan fragments will

significantly alter this perception of their methods. It was not the art of translation that they learned from their acknowledged masters, but a style for manipulating language and a technique for assembling subject matter. This much at least is clear from all that we have learned of Callimachus. It is only an accident of history that we are not better able to judge about Philetas.

NOTES ]l.

The quotation is from W. Clausen ‘Callimachus and Latin poetry’ GRBS 2 (1964) 181. It isStrabo (15.657) who styles Philetas ποιητὴς ἅμα καὶ κριτικός; cf. Suda s.v. Φιλήτας.

2.

Cf. Quint. Inst. 10.1.58 tunc et elegiam uacabit in manus sumere, cuius princeps habetur Callimachus, secundas confessione plurimorum Philetas occupauit.

3.

U. von Wilamowitz-Moellendorff Hellenistische Dichtung (Berlin 1924) 1.117.

4.

W.Kuchenmüller Philetae Coi reliquiae (Diss. Berlin 1928).

5.

The Hermes and Demeter are confirmed titles; the Bittis can be inferred from

:

references to it by Hermesianax and Ovid. His Telephus (fr. 15 P = 9 K) is less securely attested, and may be a chimera. 6.

A.W. Bulloch ‘Alexandrian poetry’, in Cambridge History of Classical Literature ed. P. Easterling and B.M.W. Knox (Cambridge 1985) 1.545.

7.

Cf. Prop. 2.34.31, 3.1.1, 3.3.52, 3.9.44, 4.6.4; Ov. Ars 3.329, Rem. 760.

8.

J. Griffin Latin Poets and Roman Life (London

1985) 201 n.22, an extreme

position first formulated by Kuchenmüller (n.4) 28-35. Since then it has been repeated often: e.g. A.W. Bulloch ‘Tibullus and the Alexandrians’ PCPS 19 (1973) 84f., who restates his convictions in his more recent work on Hellenistic poetry (n.6) 545, and M. Puelma ‘Die Aitien des Kallimachos als Vorbild der

römischen Amores-Elegie' MH 39 (1982) 225. D.O. Ross Backgrounds to Augustan Poetry: Gallus, Elegy and Rome (Cambridge 1975) 120 n.2 puts it

78

PETER Ε. ΚΝΟΧ

differently: “1 leave Philetas out of account in this discussion (in spite of the fact that he shares the first line of Book III with Callimachus), because it was primarily Callimachus who influenced and formed the concept of poetics at Rome

(or at least served as a convenient precedent), and because so little is

known of Philetas that it is simply impossible to know why Propertius acknowledges him particularly and how his poetry might be reflected by Propertius — if at all." See Powell's note on fr. 5; Kuchenmüller (n.4) 55 n.2 prefers the term ‘idyllion’. Cf.

A.

von

Blumenthal

Alexandria (Oxford

‘Philetas’

RE

19 (1938) 2167;

P.M.

Fraser Prolemaic

1972) 1.556.

R. Heinze Virgils epische Technik? (Leipzig 1915) 117 n.2. Kuchenmüller (n.4) 38. Of course, even if Philetas should be kept out of account, it would be a mistake to ignore Parthenius; cf. W. Clausen ‘Virgil and Parthenius" HSCP 80 (1976) 179. Cf. Bulloch (n.6) 546. R. Hercher ‘Zur Textkritik der Verwandlungen des Antoninus Liberalis! Hermes 12 (1877) 306-19. 15.

For

a summary

with reference to earlier work,

cf. M.

Papathomopoulos

Antoninus Liberalis: Les Métamorphoses (Paris 1968) xv-xix.

16.

See G. Pasquali ‘I due Nicandri' SIFC 20 (1913) 101-4, esp. 104 on the ‘headings’ in Antoninus Liberalis, preserved in the same ms, which he compares to the subscriptions in the Homeric scholia: “Io ne concludo, per conto mio, che degli scolii ad Antonino non é lecito fare altero uso di quello che facciamo di quelle subscriptiones. Esse ci insegnano che una forma di una certa favola ricorreva anche in Nicandro; non dicono quale forma. Ogni volta si puó trattare anche solo di un accenno fuggitivo.” Cf. C. Cessi ‘Gli indici delle fonti di Partenio e di Ant. Liberale’ Atti del Istit. Veneto 81 (1921-2) 345-60.

17.

Cf. Bulloch (n.6) 608; OCD s.v. ‘Euphorion (2).

18.

For the fragment see SH 413-15; on the genre of curse-poems, cf. F. Cairns Generic Composition in Greek and Roman Poetry (Edinburgh

1972) 93-5.

As was noted by V. Bartoletti ‘Euforione e Partenio' RFIC 26 (1948) 33: ora invece, grazie al frammento fiorentino, noi possiamo constatare che Partenio non presenta nessuna particolare affinità con il racconto di Euforione.” This mistake is still quite common, and it applies equally to Antoninus Liberalis: e.g. F. Cairns Tibullus: A Hellenistic Poet at Rome (Cambridge 1979) 21; S. Hinds The Metamorphosis of Persephone (Cambridge 1987) esp. 14f., 54f. 21.

22.

Cf. Fraser (n.10) 11.793, n.28: “the reason for the title remains uncertain". Kuchenmüller (n.4) 46f. surveys earlier hypotheses: (e.g.) Hermes helped Odysseus in his adventures (Maass), or the title alludes to Odysseus' descent from the god (Kuiper). Fraser (n.10) 1.623f.; cf. 641. For the fragments of Eratosthenes' Hermes, see CA

pp. 58-63, SH 397-8.

23.

As noticed by Kuchenmüller (n.4) 47f., who draws different conclusions.

24.

The presumption is based on analogy with the Lyde of Antimachus, a model for such poems. Under the circumstances it is impossible to discover precisely what is meant by the epithet Oorjv.

PHILETAS AND ROMAN

25.

POETRY

79

Kuchenmüller (n.4) 25-8 supposed that Hermesianax was making a joke and that there was no poem about Bittis. His highly implausible hypothesis to explain these lines has won few supporters, but deserves none.

26. 27.

See the testimonia in B. Wyss Antimachi Colophonii reliquiae (Berlin 1936) nos. 6, 7, 14 and 15. See A.A. Day The Origins of Latin Love Elegy (Oxford 1938) for a judicious account of the issue to that date. More recently the case for Antimachus' Lyde,

Philetas' Bittis, and Hermesianax' Leontion as precursors of Roman love elegy

has been re-stated by Cairns (n.20) 214-30. He concedes that much of the content

of these poems was narrative, and not all of it about love, but urges that the *subjective frame' for the narrative was the crucial link. The importance of Philetas was specifically pressed by M. Lenchantin, 'Callimaco, l'acqua filetea e Properzio III 3' RFIC 13(1935) 172-6, and E. Martini ‘Ovid und seine Bedeutung für die römische Poesie’ in Epitymbion H. Swoboda (Reichenberg 1927) 190-92.

28.

See P.E. Knox ‘Wine, water and Callimachean polemics’ HSCP 89 (1985) 112-16 with reference to earlier literature.

G. Giangrande is an exception: see his attempt at rehabilitating Hermesianax in Scripta Minora Alexandrina (Amsterdam 1981) 11.387-410. I adopt the supplement at the beginning of line 9 proposed by W. Wimmel Kallimachos

in

Rom

(Hermes

Einzelschrift

identifiable reference to Philetas is needed.

31.

16,

Wiesbaden

1960)

87f.

An

In his recent commentary on fr. 1 Pf., N. Hopkinson A Hellenistic Anthology (Cambridge 1988) 93f. adopts this interpretation, citing “the advantage of simplicity". The burden of proof rests with those who assume that the Florentine Scholiast is wrong; on which the recent arguments of V.J. Matthews 'Antimachus in the Aetia Prologue’ Mnem. 32 (1979) 128-37 are unconvincing. For a lucid presentation of the problems in this passage, see A.S. Hollis ‘Callimachus,

Aetia Fr. 1.9-12' CQ 28 (1978) 402-6. 32.

It was first suggested by W.M. Edwards ‘The Callimachus Prologue and Apollonius Rhodius! CQ 24 (1930) 110, writing before the publication of the scholia.

33.

Thus G. Coppola 'Cirene e il nuovo Callimaco' RFIC 13 (1935) 15f., although it was

originally proposed

by Housman

with the meaning simply of ‘tree’ to

balance corn represented by Demeter. Later critics have often followed Edwards

(n.32) in understanding a reference to the Argo and hence the Argonautica of

Apollonius.

34.

G. Vitelli *Frammenti di Scholia agli Αἴτια di Callimaco' PS7 11 (1935) 141 n.2.

35.

Holtis (n.31). The same supplement was proposed independently by Matthews (n.31), although he takes it as a reference to the Artemis of Antimachus.

36.

This suggestion was made in typically lucid fashion by A. Rostagni 'I nuovi

37.

C. Gallavotti ‘Il prologo e l'epilogo degli **Aetia"" SIFC 10 (1933) 233f., with γρηῦν offered as a refinement by P. Maas ‘Neue Papyri von Kallimachos Attia’

frammenti di commento agli Aitia e la polemica letteraria di Callimaco' RFIC 11 (1933) 202f.

Gnomon

38.

10 (1934) 163.

This is perhaps unlikely inasmuch as there is no evidence to suggest that Philetas took any polemical position on poetic refinement. The well-known anecdote about his thinness has no relationship to his poetry: cf. A. Cameron 'How thin

was Philetas?' CQ 41 (1991) 534-8.

80

39.

PETER E. KNOX

Cf. Rostagni (n.36) 203, Pfeiffer ad loc.

The many strands of evidence are assembled by A. Cameron in his forthcoming Callimachus and his Critics (Princeton U.P.). He suggests that the antecedent of

τῷ was Antimachus and that fr. 532 Pf. belonged to an epigram comparing a poem of Philetas, presumably the Bittis, to the Lyde. I have benefited a great deal from discussions with Prof. Cameron about the Aetia Prologue, as well as from an advance look at his completed manuscript, but he should escape censure for any defects in the arguments presented here. 41.

The hypothesis is old, dating back to P.-E. Legrand Étude sur Théocrite (Paris 1898) 155-7. It has been revived by Cairns (n.20) 25-7 and argued in great detail by R.L. Hunter A Study of Daphnis and Chloe (Cambridge 1983) 76-83 and E.L.

Bowie ‘Theocritus’ Seventh Jdy//, Philetas and Longus’ CQ 35 (1985) 67-81. See too the ingenious and suggestive paper by I.M. LeM. DuQuesnay 'Vergil's First Eclogue' PLLS 3 (1981) esp. 38-51, and 60. 42.

Cf. G.O. Hutchinson Hellenistic Poetry (Oxford

43.

Consider the attractive speculation by F. Cairns about Philetas and the Milanion exemplum in Prop. 1.1 in 'The Milanion/Atalanta exemplum in Prop. 1.1: uidere

1988) 144 n.5.

feras (12) and Greek models’, in Hommages à Jozef Veremans ed. F. Decreus and

C. Deroux (Brussels 1986) esp. 33-8.

Cf. SH 964, which contains parts of two elegies which perhaps represent an address to a woman in love. POxy. 3725 contains the remains of a Greek elegy of at least 23 lines that strikingly resemble the manner of Latin love elegy. The papyrus and the poem date to the imperial period, but the poet probably had a model, perhaps a Hellenistic elegy. The possibilities are tantalising, but must for the moment at least remain only tantalising: cf. P.J. Parsons ‘Eine neugefundene griechische Liebeselegie' MH 45 (1988) 65-74.

45.

A fifth fragment (17 K), a couplet cited by Steph. Byz. s.v. Φλιοῦς, is assigned to the Demeter by E. Maass De tribus Philetae carminibus (Marburg 1895) ix, and

accepted by Powell (fr. 4), who also follows Maass in combining two other quotations from Stobaeus as a single fragment (fr. 2 = frr. 3f. K). 46.

Cf. E. Maass Orpheus (Munich 1895) 280 n.67: “In seiner ‘Demeter’ schilderte Philetas den Raub der Persephone und Demeters Irren, sicher auch die Einsetzung der eleusinischen Weihen in eigenartiger Weise, bald utopisch, bald idyllisch. Das Gedicht lässt sich inhaltlich leicht herstellen." Kuchenmüller (n.4)

53-8 is more sanguine about the prospects of restoring the poem.

47.

This neutral stance was adopted by A. von Blumenthal ‘Philetas’ RE 19 (1938)

2167f. Cf. A.W. Bulloch Callimachus: The Fifth Hymn (Cambridge 1985) 36 n. 1: “this Demeter ... was an elegiac poem and seems to have been at least partially narrative"; N.J. Richardson The Homeric Hymn to Demeter (Oxford 1974) 69.

48.

Cf. C. Cessi ‘De Philetae carminibus quaestiones’ Eranos 8 (1908) 122-37; Maass (n.45) vii-ix.

49.

Kuchenmüller (n.4) 53-8.

50.

The focus was first fixed on Nicander by E. Bethe ‘Ovid und Nikander' Hermes

39 (1904) 1-4, who argued that Ovid took substantially all of the material for Met. 5 from Nicander. Subsequent modifications of this hypothesis are found in K. Barwick 'Ovids Erzáhlung vom Raub der Proserpina und Nikanders Ἑτεροιούμενα᾽ Philologus 80 (1925) 454-66. Callimachus was introduced into the debate by L. Malten ‘Ein alexandrinisches Gedicht vom Raube der Kore'

Hermes 45 (1910) 506-53, and these two names predominate in later discussions of the issue: e.g. H. Herter ‘Ovids Persephone-Erzählung und ihre hellenistischen

PHILETAS AND ROMAN

Quellen’

Rh.M

POETRY

90 (1941)

81

236-68,

F. Montanari

‘L’episodio

Eleusino

delle

peregrinazioni di Demetra: a proposito delle fonti di Ovidio, Fast. 4.502-62 e Metam. 5.446-61' ASNP 4 (1974) 109-37. Hinds (n.20) 55.

52.

Cf. R. Wünsch *Hymnos' RE 9 (1914) 164-70; Fraser (n.10) 1.649-74.

53.

The ancient title is simply AIOZKOYPOI. Powell assigns five fragments to Simias' Apollo in CA, but only the first two are definitely ascribed to that poem. The first consists of 13 lines on the Hyperboreans, while the second is the story of Kleinis from Ant. Lib. 20 attributed by the scholiast to Simias' Apollo. It is not unreasonable to suppose the poem a hymn (thus, apparently, Fraser [n. 10] I.555), an interpretation consistent with another fragment (9 P) of a poem to Hestia. For the little that is known of Simias, see P. Maas 'Simmias (6) RE 3A (1927) 155-8; H. Fränkel De Simia Rhodio (Diss. Berlin 1915); A.S.F. Gow and D. Page The Greek Anthology:

Hellenistic Epigrams (Cambridge 1965) 11.511f. He wrote poems in lyric meters,

epigrams, and technopaegnia. Like Philetas he had scholarly interests as well,

and wrote three books of Γλῶσσαι.

55.

Parthenius preserves the only fragment (Erot. Path.

14 = fr. 3 P), 17 couplets

which report the story of Antheus in the form of a prophecy, “ganz wie bei Lycophron" (Wilamowitz [n.3] 198), with the difference that Alexander is intelligible and offers more narrative. On his life and works, see G. Knaack *Alexandros (84)' RE 1 (1894) 1447f. Although the contents do suggest a work

like the Alexandra of Lycophron, with whom he was associated as a member of the so-called ‘Pleiad’, this can only be conjecture, and it is unlikely that Apolio

was treated just like Priam's daughter.

Cf. ΒΑ. van Groningen Euphorion (Amsterdam

1977) 39-61.

See above, p.65.

CA pp. 227-9. Admittedly, this Stoic exercise is in a different category; see the commentary by Hopkinson (n.31) 131-6. He preserves the opening line: SH 676 τῇ χθονίῃ

Φερσεφόνῃ καὶ Κλυμένῳ tà δῶρα.

μυστικὰ Δήμητρί

τε καὶ

Cf. C. Gallavotti ‘Inno a Demetra di Filico' SIFC 9 (1932) 58f.: “l’inno αἱ Filico ἃ l'esposizione mitica di un artista, chedi nuovo rappresenta e fa rivivere le antiche leggende, per vederle riapparire nella loro multiforme varietà e ricrearle con nuovo spirito e con nuove forme." 61.

Cf. Fraser (n.10) 1.650-52; K. Latte ‘Der Demeterhymnos des Philikos’ MH 11 (1954)

1-19 (= Kleine

Schriften

Callimaco' SIFC 41 (1969) 13-18.

539-61),

C.

Previtali

D.L. Page Greek Literary Papyri (Cambridge, Mass.

‘Filico

di Corcira

e

1942) 408f.

See the notes on SH 990. Cf. N. Hopkinson Callimachus: Hymn to Demeter (Cambridge

1984) ad loc.

In addition to the commentaries of Gow and Dover, see the recent study by Hutchinson (n.42) 201-13, which includes references to earlier work. See Gow and Page (n.54) II.115 for the evidence for this name.

82

67.

PETER E. KNOX

Indeed, he may even have been hostile to Callimachus: his name is included in the list of Telchines provided by the Florentine Scholiast. Cf. Gow and Page (n.54) 11.115. He includes him in a list of ὑμνογράφοι with Orpheus, Homer, and Hesiod at Schol. Lyc. Alex. p.3 Scheer (= SH 218). His appearance in this company is so

surprising that it might even have some basis in fact. 70.

71.

E.L.

Bowie

(n.41)

78

infers that

reference

is to the erotic content of

E.g. Cairns (n.20) 25; M. Puelma ‘Die Dichterbegegnung in Theocrits "Thalysien"' MH

17 (1960) 150; and most recently Bowie (n.41) 67-91.

72.

Cf. Hutchinson (n.42) 201-13.

73.

Cf. Bowie (n.41) 80.

74.

See S.M. 305-12.

75.

Kuchenmüller (n.4) 53-8.

76.

the

Asclepiades’ epigrams, but this cannot be the whole story: presumably only a relatively small proportion of Asclepiades' epigrams survive and they will have dealt with many themes. It is also worth noting that there is no allusion to the extant poetry of Asclepiades in the Idyll on the scale often assumed for Philetas.

Sherwin-White

Ancient

Cos (Hypomnemata

51, Göttingen

1978)

A Coan setting has sometimes been argued for Callimachus" Hymn to Demeter: cf. Fraser (n.10) II.916f. n.290; Sherwin-White (n.74) 307-11.

77.

Doubtless many of the subtle connections between Ovid and the Homeric Hymn detected by Hinds (n.20) would seem even more subtle to a reader who had more

material available. The myth of Demeter also figured in Antimachus (frr. 32.5, 109, 67, 96 Wyss), which might explain Callimachus' vigorous endorsement of

Philetas’ Demeter. And some parallels can be noted with Philicus in Ovid's accounts: e.g. SH 680.3 and Fast. 4.485, Met. 5.438; SH 680.5 λῃστήν and Fast.

4.591 praedone marito. But cf. Gallavotti (n.60) 50: “Io credo peró che sia piuttosto da escludere una derivazione diretta della redazione ovidiana da Filico. Anche un altro poeta alessandrino, pid letto dai Romani, aveva trattato il mito di Demetra in versi elegiaci, Fileta; e non & escluso che la fonte di Ovidio sia piuttosto da ricercare nell'epica ellenistica.” 78.

See A.S. Hollis Ovid, Metamorphoses: Book VIII (Oxford

79.

I accept the argument advanced by M.L. West Studies in Greek Elegy and lambus (Berlin 1974) 74 that by the Hellenistic period Mimnermus' works consisted of two books, Nanno and Smyrneis. In that case the 'big lady' disparaged by Callimachus

recommended

would

1970) 128-30.

have been the Smyrneis, while the Nanno would

volume

of short

poems,

as first suggested

*Mimnermo e Callimaco' Athenaeum 30 (1952)

be the

by A. Colonna

191-5. Cf. E.L. Bowie ‘Early

Greek elegy, symposium and public festival' JHS 106 (1986) 13-35.

EE Griffin (n.8) 201 n.22, "could they [sc. Propertius and Ovid] have gone on referring to Callimachus as a love poet if they had really read much of his work?"

81.

For the lingering controversy over the origins of Latin love elegy in Hellenistic poetry, see n.27 above. My own view is that more isto be gained by following the evidence pointing to 'sources' for love elegy in the so-called objective narratives of Callimachus and his contemporaries than by conjuring up lost Greek love

elegies that more closely resembled the surviving Latin poems: see A. Rostagni *L'influenza greca sulle origini dell'elegia erotica latina’ Fond. Hardt (Van-

doeuvres-Gentve 1956) 59-82 and more recently Puelma (n.8) 221-46, 285-304.

PHILETAS AND ROMAN

POETRY

83

82.

So too at 3.9.37-44, and at Ovid Ars 3.329f., Rem. 759f. Cf. Hutchinson (n.42) 280.

83.

It is sobering to think what sense we would make of the reference to the somnia of Callimachus if we had no information about the opening of the Aetía. See the sensible remarks on this couplet by J.-P. Boucher Etudes sur Properce (Paris 1965) 213-15. Burman's ‘codex Lusaticus’, now Breslau, Bibl. Univ. Akc 1948 ΚΝ 1469).

85.

See Sherwin-White (n.74) 47-9. According name of an ancient king of Cos: cf. Schol.

197 (a.

to some traditions, Merops was the Theocr. 7.3. Butler and Barber in their

note ad loc. cite the relevant texts in support of this reading. 86.

G. Luck ‘Beiträge zum Text der römischen Elegiker' RA.M 105 (1962) 347. R.

Hanslik follows Luck in his edition (Leipzig 1979) but omits the necessary

change to Philitae at line end. Some years after Luck's article a fragment of a Hellenistic commentary on an epic poem known as the Meropis was published by L. Koenen and R. Merkelbach ‘Apollodorus (ΠΕΡῚ @EQN), Epicharm und die Meropis', in Collectanea Papyrologica: Texts Published in Honour of H.C. Youtie ed. A.E. Hanson (Bonn 1976) 1.3-26, now available as SH 903A. It deals with

Heracles’ invasion of the island and his war with the Meropes; it is probably

sixth-century, not Hellenistic (cf. H. Lloyd-Jones 'The Meropis (SH 903A) Atti

del XVII Congresso Internazionale di Papirologia [Naples 1984] 1.141-50), but its discovery makes the attribution of such a poem to Philetas less unlikely. W. Stroh Die rómische Liebeselegie als werbende Dichtung (Amsterdam 1971) 96 n.134.

Cf. Prop. 3.1.5 dicite, quo pariter carmen tenuastis in antro? Cf. Tib. 1.1.5f. me mea paupertas uita traducat inerti,/ dum meus adsiduo luceat igne focus.

A. La Penna ‘Ipse Coo plaudente Philetas (Stat. Silv. 1,2,252): ὕπ᾽ ipotesi su Fileta di Cos’ RFIC 116 (1988) 318-20. It is worth recalling that if the Aetia Prologue had not been recovered it is unlikely that anyone would have thought senex an allusion to his poetry. H.E. Pillinger 'Some Callimachean influences on Propertius, Book 4' HSCP 73 (1969) 189-99; but see esp. F. Cairns ‘Propertius and the Battle of Actium (4.6)’,

in Poetry and Politics in the Age of Augustus ed. A.J. Woodman and D. West

(Cambridge

9.

1984) 129-66.

Pace Pillinger (n.92) 192 n.41, reflecting general sentiment: “it might seem odd that Propertius alludes to Callimachean principle at the beginning of one of his most ‘epic’ (and theoretically non-Callimachean) efforts.”

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 85-96 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

HORACE’S IBIS: ON THE TITLES, UNITY, AND CONTENTS

OF THE EPODES

S.J. HEYWORTH

The Epodes! form a book of surprising variety: four poems addressed to Maecenas (1, 3, 9, 14); four with political accents (1, 7, 9, 16); a number full of invective (3, 4, 5, 6, 8, 10, 12) or otherwise dependent onan Archilochean model (2, 16); some on sex (8, 12); others on love

(11, 14, 15), this last group almost like odes in style, as is also 13; and finally a palinode (17), which half-heartedly recants the attack made on the witch Canidia in Epode 5 (and in Satires 1.8).? The Iambi (to

give the Epodes their other traditional title) also display metrical variety. All but the last are epodic, or, to put it in other terms, in couplets, consisting of one verse alternating with a different adjunct. But Epode 12 has dactylic hexameters followed by dactylic tetrameters, and is thus without iambs; and Epode 17 is a continuous run of iambic trimeters, and so is not epodic at all. Thus neither of the traditional titles fully describes the work's metrical nature. In Latin, moreover, the word iambi? has the meaning ‘invective’ when it is not used purely to describe metre; thus, although it may have the warrant of Archilochean practice* or of Alexandrian collections, it misleads as an announcement of the book's genre.5 Epodi has no traditional force as an indicator of genre. This essay will explore the generic complexities of the book (and its unity), and the ways in which Horace reflects upon its changing form.

1. Ibis Besides titles, opening words were also used in antiquity as a way of 85

86

S.J. HEYWORTH

referring to works of literature, and poets often took advantage of

this fact in giving their initia special resonance and point. Thus Vergil's Arma echoes the Ἄνδρα of Odyssey 1.1, whilst his virumque reproduces its sense: the epic's genre, and its combination of Iliad and Odyssey, are clearly announced. Ovid, in turn, repeatedly echoes Vergil's opening: so Arma itself misleadingly introduces the Amores; and the paired prologues of Fasti 3 and 4, one addressed to Mars, the other to Venus, begin with Bellice (for this variant on the Vergilian Arma, cf. Lucan 1.1), and A/ma respectively. Francis Cairns has

recently suggested that the initial Odi of Horace Odes 3.1 hints at the Greek φδή." However, as far as I know, no one has dwelt in print on the opening to Horace's book of invective: bis. Here too we are given a striking intimation of genre. What could announce the character of an iambic book better than this allusion to Callimachus' invective poem Ibis? Most of what we can say about that poem comes from Ovid's Ibis and the scholia thereon:? in metre unknown," of no great length (cf. Ovid Ibis 447: exiguo ... libello, but surely more than epigrammatic), it attacked and cursed a man under the pseudonym Ibis, the name of an Egyptian bird that was supposed to purge itself

by using its beak to squirt water up its rectum.!! The association between Horace's book and Callimachus may be strengthened by observing that the earlier poet's Jambi" is a book with a similar diversity of metrical forms (including three poems in couplets:

5-7).

It is, I suspect, no coincidence that the poems

in

Horace's iambic book do not number a round ten or twelve, but seventeen, the next prime number after Callimachus' thirteen." There is also a manifest allusion to Callimachus at Epode 15.7-9: dum pecori lupus et nautis infestus Orion/ turbaret hibernum mare,/ intonsosque agitaret Apollinis aura capillos. This echoes Jambi 12.69f. (fr. 202 Pf.), where Apollo is speaking: ἔστ᾽ ἐμὸν γένειον ἀγνεύῃ τριχός kai ἐρίφοις χαίρωσιν ἅρπαγίες X]óx[o]1.'^ Callimachus is thus an acknowledged model of Horace, significant to him as the writer of Jambi, as a composer of curse poetry, and as a predecessor in imitating the archaic iambists, Archilochus and Hipponax. 2. Unity The exploitation of Archilochus and Hipponax is one of the unifying forces of Horace's collection, and has been dealt with often enough,

scholars! extending the analysis which the poet himself offered at Epode 6.11-14:

HORACE'S 1815

87

caue, caue: namque in malos asperrimus parata tollo cornua, qualis Lycambae spretus infido gener, aut acer hostis Bupalo,

and Epistles 1.19.23-5: Parios ego primus iambos ostendi Latio, numeros animosque secutus Archilochi, non res et agentia uerba Lycamben.

In addition details within poems are used to give some coherence to what would otherwise be quite disparate. Thus the garlic which Horace has consumed

in Epode 3 is compared

to the poisonous

potions of Canidia and Medea (7-14; cf. Epode 5, especially 61-6); the magic of Canidia is aimed at regaining the love of Varus for the

aged witch (Epode 5.73-82), and the erotic passions of aging women are the subject also of Epodes 8 and 12; thelabours of Ulysses and his crew feature in similar lines at Epode 16.60 and Epode 17.16. There

are also repeated generic touches, such as the ater dens shared by Epode 6.15.6 and Epode 8.3, and the libera bilis of Epode 11.16 recalling the /iberrima indignatio of Epode 4.10. 3. Sex and love

Poems of invective form the largest of the categories into which I

have analysed the book: to that extent the title /bis is appropriate. The

pieces

are,

however,

largely gathered

together,

with

none

occurring after Epode 12, which is the only invective after the first change of metre. Horace begins Epode 11, the point at which the metre first changes, with an announcement, and an explanation, of the change: Petti, nihil me scribere uersiculos amore qui me mollibus in pueris

sicut antea iuuat amore percussum graui, praeter omnis expetit aut in puellis urere.

* Pettius, I gain no pleasure from writing little verses, as I did in the past, now I'm deeply smitten with love." Horace is still writing poetry — these verses prove that — but there is a difference. The word uersiculos would not be enough in isolation to differentiate the

metre of Epodes 1-10 from the metre of Epode 11, but sicut antea confirms the change: the verses no longer come out as pure iambics. As a lover, Horace must write love poetry, not invective. This

88

5... HEYWORTH

interpretation, propounded by W. Wili,"" E.A. Schmidt!* and others, was rejected by L.C. Watson;?? he objects that it places a strain on uersiculos,

but

fails

to

see

that

sicut

antea

(which

he

himself

emphasises) eases the strain; nor is it true, as he claims, that change of genre, having been raised in the opening couplet, is thereafter ignored by Horace: much of what follows in his own article is rightly

concerned with the prominence

of elegiac themes.? The poet's

current love for Lyciscus appears against a background of constant emotional

turmoil (3f.), and in particular it is compared

with his

situation over two years earlier, when he was madly in love with Inachia (5-22). His actions and words at that time gained him the notoriety of the elegiac lover and poet (7-10). Wine brought selfrevelation, not consolation (11-14); and, when friends gave advice, he was unable to follow it (20-22). Then, however, he could conceive

of letting his anger overwhelm his longings: quodsi meis inaestuet praecordiis

I5

libera bilis, ut haec ingrata uentis diuidat fomenta uulnus nil malum leuantia,

desinat?! imparibus certare summotus pudor: ubi haec seuerus te palam laudaueram ...

“But if anger were to boil up freely in my breast, so that I might distribute to the winds these unavailing remedies that fail to lighten a bad wound, I would cease to compete, shame removed, with those

who are not equal to me. When I had sensibly praised this course of action in front of you ...””. /ibera bilis is the emotion that the iambic poet reveals;? fomenta (17) implies “the poetry that tries to assuage love", as Watson himself argues. If Horace could have summoned up sufficient bile, he might have bidden farewell to his ineffectual love poetry. Epode 12 will in fact give us an example of such bile discharged against one who was a rival to Inachia; but such respite is temporary, and now, overwhelmed by the mollities of Lyciscus, he has come to a state where invective is impossible. It is hardly surprising that the metre, like the subject matter, has assumed an

elegiac tinge. We should observe that he describes how, in the past, amidst the tumult of his emotions, he proceeded to the beloved's unfriendly doorstep incerto pede (20), and that he thus once more draws attention to the inconsistency of his metre. The poem, in form and content, indicates

a move away from iambics and invective; but

it is not leading Horace to elegy, despite the intrusive hemiepes. At the end he reveals that his affair with Lyciscus will end, not with the recovery of good sense (nor indeed with death in love), but with a new

HORACE'S

2815

89

affair: he is already manifesting something that will distinguish his erotic poetry from that of the elegists — a multiplicity of lovers. This programme begins to be fulfilled in Epodes 12, 14, and 15.

Epode 12, as we have seen, returns us to the invective of the past. Ironically, this poem, which marks an escape from elegiac passion, is written in the metre closest to elegy, dactylic tetrameters replacing

pentameters. Epodes 14 and 15 introduce us to two additional figures from Horace's love life: Phryne and Neaera. Epode 14 stresses the difficulty that the lover has in bringing to completion a book of iambi: the comparison of his love to that of Anacreon (depicted as a lover, and thus as a lyric poet) makes a very specific gesture towards the future. Mollis, the poem's opening word, immediately marks its

distance from abuse: contrast Mala at Epode 10.1, and the misleading Horrida at Epode 13.1. But iambic notes are not entirely omitted: occidis (Epode 14.5) is colloquial, and a touch aggressive; and Epode 15 has its smutty pun in line 1225 and threatens both Neaera (11) and her current lover (23f.).

To sum up: the treatment of sex and love in the Epodes mimics the abandonment of concerted invective. After the change of metre Epode 12 is the only poem that treats sex with disgust or attacks an individual; and the poet has taken care to indicate that it is out of chronological order by his statement about the end of his love for Inachia in Epode 11.5f. His susceptibility to love coincides with changes in metre, tone and appropriate genre. 4. Politics and recantation

It is, I believe, a correct as well as a conventional view that Horace was attracted to the idea of writing Archilochean iambics because of

his position as an outsider in the years after Philippi. Fighting on the losing side had alienated him from the hopes of his youth, from wealth and from power. So Fraenkel wrote “11 is probable that the θυμός, the animus, of Archilochus, expressed with particular vigour

in the poet's pronouncements on the affairs of the πόλις, the res publica, was among the major incentives that caused Horace to turn to the Parian knight who called himself a servant of the god of war”. Similar points are made by Nisbet in his exploration of the political

poems

and

of Horace's

acceptance

into the circle of

Maecenas." The four political Epodes fall into two pairs, quite different

in tone. Epodes

7 and

16 are general and pessimistic,

attacking the whole people for their willingness to turn once more to

90

S.J. HEYWORTH

arms. Epodes | and 9, on the other hand, show enthusiasm for the Caesarian cause and a readiness to support Maecenas and his patron, Octavianus. Those apparently early odes that inveigh against the recurrence of civil war (Odes 1.35 and, if the traditional reading of the allegory is correct, Odes 1.14) do not return to the pessimism of Epodes 7 and 16, but they either indicate a change of attitude (Odes 1.14.17£.: nuper sollicitum quae mihi taedium,/ nunc desiderium curaque non leuis) or else they name Caesar as the potential victim of current strife (Odes 1.35.2940): serues iturum Caesarem in ultimos orbis Britannos et iuuenum recens examen Eois timendum

30

partibus Oceanoque rubro. heu heu, cicatricum et sceleris pudet fratrumque. quid nos dura refugimus aetas? quid intactum nefasti liquimus? unde manum iuuentus metu deorum continuit? quibus pepercit aris? o utinam noua incude diffingas retusum in Massagetas Arabasue ferrum.

35

40

The series of urgent questions recalls Epode 7, and so does the enthusiasm for foreign conquest; but whereas in the ode the treatment of Caesar offers a real possibility of peace at home, and

warfare on the margins of the empire, in the earlier epode the primal fratricide suggests a fate that is unalterable. Such pessimism surely belongs to the period before Horace's reception into the circle of

Maecenas, not later than 37 BC.?’ Yet there are questions raised by the publication seven or eight years later of poems that express an emotion which the poet has long lost, and E. Kraggerud has attempted to shake this, the orthodox position. He poses the problem thus: *Man kónnte freilich das skizzierte Prinzip 'retten', indem man diese Gedichte vor der Aufnahme des Horaz in den Maecenas-Kreis ansetzt; aber derselbe Horaz, der angeblich in solchen Epoden seiner Verzweiflung und seiner politischen Frustration Luft machte, hat sie ja mehrere Jahre später, unter gänzlich verschiedenen Umständen, der Welt vorge-

stellt.?* But his solution — that Epodes 7 and 16 are partisan and forward-looking Caesarian poems composed after Actium — does

not carry conviction.? What he fails to see is that the manifest turncoat provides useful propaganda for those he comes to support:

HORACE'S

1815

91

his favour carries far more emotional weight than that of the life-long partisan. So we need not be surprised that Maecenas permitted Horace to publish poems composed in his disaffected youth: the evidence of later works, where Horace’s republican past is a number of times rather unnecessarily brought to the reader’s attention,’ implies in fact that Maecenas would have favoured such revelations. Epodes | and 9 declare Horace’s attitude towards the Actium campaign,?! and so rule out the possibility that this is the background for his sentiments in Epodes 7 and 16. Kraggerud makes a strong case for distinguishing between the assumed and the actual date of Epode 9, which Horace composes after Actium

but as if the battle were

currently progressing. A similar technique is employed in the publication of Epodes 7 and 16 several years after their emotional date: the reader has to bring to each poem a broad knowledge of the poet. Horace has dated his involvement with Maecenas, and thus with Caesar's heir, in the Satires (1.6.61f.; 2.6.41f.). In Epode 9.1-10, Quando repostum Caecubum ad festas dapes uictore laetus Caesare tecum sub alta — sic? Ioui gratum — domo, beate Maecenas, bibam sonante mixtum tibiis carmen lyra,

5

hac Dorium, illis barbarum, ut nuper, actus cum freto Neptunius dux fugit ustis nauibus, minatus Vrbi uincla, quae detraxerat seruis amicus perfidis,

10

he compares Actium to the past triumph over Sextus Pompeius, mentioned at Epode 4.17-20; this perhaps serves as a hint at the date of his conversion, and of the earlier poems — he celebrated the defeat of the Neptunius dux, although the campaign began when he was still alienated. The key point, however, lies in the shape of the book asa whole: it moves from invective to love; from iamb to lyric, near elegiac; from

attack to palinode; and it thus presents the poet's metamorphosis

from outsider to insider, from anti-Caesarian to poet of the regime.?? Without

the political invectives we would

lack a major part of

Horace's motivation for producing an Archilochean book; and we

would not see the negative views of contemporary politics that Epodes | and 9 serve to correct. It is true that this involves the reader in a mental re-ordering of Epodes 1, 7, 9 and 16. The poet has achieved a balanced pattern, but with all the purely iambic epodes in the first section, and some Archilochean material in the second half:

92

S.J. HEYWORTH

compare the way that Epode 12 is placed after Epode 11 despite being marked as chronologically prior. The palinodic structure of the whole clarifies the interpretive order of the political poems; and the book's opening couplets serve as an announcement of the poet's transmogrification (Epode 1.1-4): Ibis Liburnis amice, paratus omne subire,

inter alta nauium, propugnacula, Caesari" periculum Maecenas, tuo.

The couplet that opens with the aggressive /bis quickly reveals that the word is part of eo?) and it proceeds in the epodic verse to an amice that destroys any expectations of invective.’* Horace offers to Maecenas the same aid that Maecenas gives to Caesar. Later in the poem the iambist will acknowledge that his nature is in fact imbellis ac firmus parum (16); far from being wolf, sheepdog, or bull, as in Epode 6, he appears as a timid mother-bird (17-22). Already in their first lines, then, Horace's Jambi demonstrate their dual nature. /bis itself is no more accurate a guide to what lies within than /ambi or Epodi; but Ibis ... amice is. Horace's attempt at sustained invective has been thwarted by social circumstances. Instead of enemies, he has friends; instead of disgust, he now turns out to have desire. Small

surprise that after less than four hundred verses the metre, the tone, and the genre, begin to waver; there are difficulties in finishing the proposed book. The iambist loses his touch in the rows he conducts: in Epode 12 the mulier nigris dignissima barris is given the last words; and in Epode 17 even the arch-witch Canidia has her opportunity to have the final say. It is a fitting irony that Horace, having illustrated the dual nature of his book by producing a series of epodes, uses pure iambics for the apologetic finale, a further attack masquerading as a palinode. There are, too, ontological complexities in a poet's being cursed by one of his own creations; consider the final words of Canidia — and of the

book — appropriately marked as such by exitus?! (Epode 17.76-81): an quae mouere cereas imagines, ut ipse nosti curiosus, et polo deripere lunam uocibus possim meis, possim crematos excitare mortuos desiderique temperare pocula, plorem artis in te nil agentis exitus?

“Am I to lament the end of an art that achieves nothing against you?" Her rhetorical question implies that her witchcraft will not be in

HORACE'S

1815

93

vain, her art is not about to give out. But her success Horace, and by destroying Horace, the writer ofiambics, Canidia destroys herself. Their enmity leads to mutual Horace the iambist is dead; long live Horace the lyricist, Maecenas, the lover of many.

will destroy her creator, destruction. the friend of

5. Epodic Odes Just as the Epodes look ahead to the Odes, so the Odes at various points look back to the Epodes. Certain odes?? are in epodic metres, Odes 1.7 and 1.28 in the same combination as Epode 12. I have examined the relationship between Odes 1.1 and Epode 2 inanearlier article.” Odes 1.16 (which in my opinion forms a single poem with Odes 1.17)'? has traditionally been read as a palinode,*! in which Horace asks the recipient to destroy the iambi that he has written in the past (1-4): O matre pulchra filia pulchrior, quem criminosis cumque uoles modum pones iambis, siue flamma

siue mari libet Hadriano. He goes on to describe the power and destructiveness of ira; then asks the addressee to restrain the anger his invectives have aroused in her. He now regrets submitting to anger and attributes it to his youthful past. He wishes to offer her verses of friendship not hatred — as long as she too will give up her anger (Odes 1.16.22-8): compesce mentem: me quoque pectoris

temptauit in dulci iuuenta feruor et in celeris iambos

misit furentem: nunc ego mitibus mutare quaero tristia, dum mihi fias recantatis amica opprobriis animumque reddas.

25

The following twenty-eight lines (i.e. Odes 1.17) provide the promised mitia, with an account of the /ocus amoenus which the poet inhabits and an invitation to the addressee to share its delights with him. And now she is named: Tyndaris, that is Helen, truly a more beautiful daughter of a beautiful mother, and traditionally the recipient of the first palinode, that of Stesichorus.

Horace thus sets an invitation,

material typical of his lyrics, against the background of his iambic youth; and in doing so recalls the most famous poems of one of his Greek predecessors. The tone of all this is far more friendly than that

S.J. HEYWORTH

94

of the earlier palinode, Epode 17, where the poet's recantation is scarcely skin-deep; Odes 1.16-17 thus suits the collection in which it is placed. Although friendship predominates in the Odes (Maecenas is of course the first word), hatred is not entirely absent. It serves mostly as a foil, even where it misleadingly opens poems (Persicos odi: Odes 1.38), and books (Odes 3.1.1): Odi profanum uulgus et arceo.

NOTES A shortened version of this paper was delivered at the Leeds International Latin Seminar on 3rd May W.

Fitzgerald

‘Power

1991. and

impotence

in Horace's Epodes’ Ramus

17 (1988)

176-91 starts from a similar desire to find coherence amidst the work's diversity; his search leads in different directions.

Cf. esp. Cat. 36.5: truces ... iambos, 40.2, 54.6 (all in hendecasyllables). On this, see M.L. West Studies in Greek Elegy and lambus (Berlin 1974) 22-39. Pace N.M. Horsfall 'Some problems of titulature in Roman BICS 28 (1981) 103-14.

literary history'

Cf., e.g., E.J. Kenney ‘That incomparable poem the /lle ego’ CR 20 (1970) 290. So, e.g., F. Cairns Virgil's Augustan Epic (Cambridge

1989) 19].

In 'La prima delle Odi romane di Orazio: Contenuto greco, logica latina' forthcoming in the proceedings of the Congresso internazionale: Modelli testuali

e prassi poetica: Grecia ellenistica

e Roma, held in Milan in May 1992.

Fuller discussion and bibliography can be found in L. Watson Arae: the Curse Poetry of Antiquity (Arca 26, Leeds 1991).

10.

But presumably not elegiac: given the proximity of Ov. Ibis 43f. (prima quidem coepto committam proelia uersu,/ non soleant quamuis hoc pede bella geri), modo in [bis 53f. (nunc, quo Battiades inimicum deuouet Ibin,/ hoc ego deuoueo teque tuosque modo) means ‘manner’, not ‘metre’.

Ht.

See Callim. frr. 381f. Pf.; Ov. Ibis 53-60, 447f.

12.

On the /ambi as a book, see D.L. Clayman 'Callimachus' Thirteenth Iamb: the last word' Hermes 104 (1976) 29-35.

13.

D.L. Clayman Callimachus's lambi (Mnemosyne Suppl. 59, Leiden 1980) 7, argues that four miscellaneous poems (frr. 226-9 Pf.) were appended to the

original book of 13, and that it was a 17 poem collection that Horace knew. If so, Horace will have been imitating not merely the number of poems in his model but

also its extraordinary variety. 14.

M.W. Dickie ‘The disavowal of Jnuidia in Roman iamb and satire’

PLLS 3(1981)

183-208, 198 links the horns of Epod. 6.12 with those at Callim. Jamb. 13.52 (Jr.

203 Pf.); L.C. Watson ‘Problems in Epode 11’ CQ 33 (1983) 229-38 suggests that

the theme of Epod. 11.11f. is derived from Callim. amb. 3, and the theme of Epod. 11.17 from Callim. Ep. 46 Pf.

95

HORACE'S IBIS 15.

E.g. E. Fraenkel Horace (Oxford

16.

On this, see Dickie (n.14).

17.

Horaz (Basel

18.

‘Amica uis pastoribus: der Jambiker Horaz in seinem Epodenbuch' Gymnasium 84 (1977) 401-23, esp. 413f.

19.

Op. cit. (n.14).

20.

Well analysed also by G. Luck /CS 1 (1976) 122-6.

21.

desinat Shackleton Bailey: desinet codd.

22.

Cf. Epod. 4.10.

23.

For other feet that have similarly significant epithets, cf. Cat. 63.2: cítato pede (an observation which I owe to Dr. Stephen Hinds, who has also discussed metaphorical uses of pes [pp.16-18] and ungula [pp.21-4] in The Metamorphosis of Persephone [Cambridge 1987]), Prop. 1.1.4: impositis pressit Amor pedibus (Love imposes his metre on the poet); and the passages discussed by Fedeli on

1957) 24-75.

1948) 61.

Prop. 3.1.6. The parallel couplet, Tib. 2.6.13f., has the same play: iuraui quotiens

rediturum ad limina nunquam!/ cum bene iuraui, pes tamen ipse redit.

24.

On the fun which Flaccus has with his name, see esp. Fitzgerald (n.2).

25.

Op. cit. (n.15) 47.

26.

R.G.M. Nisbet *Horace's Epodes and history’, in A.J. Woodman and D.A. West (edd.) Poetry and Politics in the Age of Augustus (Cambridge 1984) 1-17, 197-200:

*His ambition to write epodes in the manner of Archilochus had its origins in the

bitterness of defeat... Poets do not choose their personae at random, but to match something they would like to see in themselves" (2).

27.

The dates proposed by Nisbet (n.26), following D. Ableitinger-Grünberger Der junge Horaz und die Politik: Studien zur 7. und 16. Epode (Heidelberg 1971), for Epod. 7 (39 or early 38) and for Epod. 16 (the first half of 38) seem to me highly probable. Epod. 4.17-20 (quid attinet tot ora nauium graui/ rostrata duci pondere/ contra latrones atque seruilem manum/ hoc, hoc tribuno militum?) implicitly condemn the leaders of the current campaign against Sextus Pompeius for the

elevation of an upstart ex-slave to the rank of tribunus militum (as Catullus in poem 29 blames Pompey and Caesar for the ostentation of Mamurra). It is thus likely to belong to just the same period. But we must always remember that the poet can make changes to poems up to the moment of publication; and Horace, who was undoubtedly influenced by the Eclogues, may well have made alterations to increase his book's internal coherence.

28.

Horaz und Actium: Studien zu den politischen Epoden (Oslo 1984), 10. Cf. also: "dann wird allenfalis die Frage drángend, ob der Dichter dem Leser die nótigen Hinweise gibt, damit dieser das Gedicht als geschichtliches Dokument beurteilen und es so tatsächlich richtig deuten kann" (11).

29.

See the reviews by E. Burck Gnomon 58 (1986) 15-21; J. den Boeft Mnemosyne 39 (1986) 517f.; L.C. Watson JRS 78 (1988) 227f.; each attacks from a different angle.

30.

So Od. 2.7.9-12. 3.4.26, 3.14.27f.; Ep. 2.2.47f.

31.

L. Watson LCM 8 (1983) 66-8, disposes of the notion that Epod. 1 refers to Octavian's war against S. Pompeius in 36 BC.

S.J. HEYWORTH

32.

si Shackleton Bailey; fortasse sit? In the discussion

which

followed

the delivery of this paper at the Leeds

International Latin Seminar Dr. Susanna Braund pointed out that the first book of Satires can also be read as marking Horace's movement from outsider to insider: so J.E.G. Zetzel 'Horace's Liber sermonum: the structure of ambiguity’ Arethusa 13 (1980) 59-77, esp. 69.

Caesari x p.c.: Caesaris cett. 35. 36.

Cf. Tib. 1.3.1f.: /bitis Aegaeas sine me, Messalla, per undas./ o utinam memores ipse cohorsque mei.

A similar technique of misleading the reader in the opening verse is employed at Callim. Jamb. 1.1ff. (fr. 191 Pf.), where the aggressive Hipponax is introduced, only to preach peace and tolerance. Cf. also Prop. 2.13.1f. (which, I believe, was

the beginning of the original third book; see G.P. Goold's Loeb edition [1990], 16-18):

Non

tot Achaemeniis armatur tEtruscat sagittis,/ spicula quot nostro

pectore fixit Amor, and Ov. Am. 1.1.1-4 (see McKeown ad loc.). In those cases potential epic is immediately transformed into elegy.

37.

For a work beginning and ending with (addressed by Medea herself); and 1027: (addressed to Medea by Jason). Not really the boundaries of Sen. Ag.: Opaca (1) and

pointed words, cf. Sen. Med. 1: Di testare nullos esse qua ueheris deos pointed, but certainly expressive are furor (1012).

On these, see F. della Corte ‘Quattro epodi extravaganti' Maia 42 (1990) 101-20.

39.

*Horace's Second Epode’ AJP 109 (1988) 71-85, esp. 80-82. This view was held by Porphyrio and, eventually, by Heinze. Unpublished papers by Alan Griffiths have convinced me that all three pairs of poems in Odes 1 that share the same metre are to beamalgamated into single poems: thus Od. 1.26 and

27, and Od. 1.34 and 35, as well as Od. 1.16 and 17. It is an attractive principle

that Horatian odes divide only where the metre changes, and I should extend it to

books 2, number support), Professor

3, and perhaps 4, thus combining Od. 2.13-15 (14-15 still joined in a of manuscripts), Od. 2.19-20, Od. 3.1-6 (again with Porphyrio's Od. 3.24-5; and, with less confidence, Od. 4.14-15. 1 must thank Reeve for drawing my attention to Mr. Griffiths’ arguments, and for

his comments on this paper as a whole.

41.

A view rejected by Nisbet- Hubbard

ad loc. because they could not see that the

anger of Horace and the anger of his addressee are implied, and linked, in the

opening stanza (in iambis and siue flamma siue mari respectively); for further argument on this point, see F. Cairns 'The genre palinode and three Horatian examples’ Ant. Class. 47 (1978) 546-52, esp. 546.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 97-100 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

AENEID 6.268: IBANT OBSCURI

SOLA

SUB NOCTE PER UMBRAM

STRATIS KYRIAKIDIS

Solus can mean ‘alone’ in two senses: ‘solitary’, ‘lonely’, ‘deserted’;! or ‘only’, 'single'.? The question raised here is which meaning sola has

at Aeneid 6.268, i.e. does sola sub nocte mean “under the solitary (lonely) night" or "under the single (only) night”. Servius ad loc. clearly regarded both meanings as possible: aut hypallage est 'sub obscura nocte soli ibant' aut 'sub sola nocte', id est ubi nihil aliud est praeter noctem. Commentators on Virgil in the nineteenth and early twentieth century have tended for slightly differing reasons to accept Servius’ first (hypallage) interpretation, which makes sola mean "solitary", “lonely”. Heyne? regarded obscura and soli as the more normal adjective distribution, while Forbiger* was even more specific: “solus hic i.q. solitarius, vacuus; die einsame, óde Nacht per quam nemo obviam fit euntibus." Benoist? agreed. Conington noted Heyne's view and provided an example of solus “applied to things where persons are really thought of ... though in each case there is of course a certain propriety in the epithet as applied to the thing." Norden? believed that darkness and loneliness are linked here, adding “‘sola nox wie νὺξ épnpaín", with no reference. The Greek phrase may derive from Empedocles fr. 49:5} but nothing there compels us to accept Norden's interpretation.” Norden translated 268 as “Sie schritten in der Einsamkeit der Nacht." More recent commentators also accept the first Servian view, deducing from it the emotional effect of the line. Fletcher? refers to Servius but, like the rest, exploits only the first half of his comment: “the darkness of the night has 97

98

STRATIS KYRIAKIDIS

passed into the hearts of the travellers and the loneliness of their

feelings seems to be part of the night itself." For Austin!! “the interchange of epithet is very striking" and the “memorable” arrangement of words “brings out, with great impact on the reader, the dim groping figures ina terrifying loneliness of night." Paratore!? says much the same. This consensus view of sola sub nocte has been subject to little

challenge,"

despite the heavy reliance put upon the phrase for

interpretations of the entire scene. Only Quinn has diverged from it, asserting briefly that so/a sub nocte must mean "only the gloom of

night".'^ This note will argue that Servius’ second suggestion was correct, and that the phrase means ubi nihil aliud est praeter noctem. That such a meaning is linguistically possible cannot be doubted: it conforms with everyday Latin usage and is paralleled at Aeneid 11.582f. The argument about Aeneid 6.268 takes as its point of departure the context of the line. At 236ff. Aeneas sacrifices before entering the Underworld. The scene is suitably dusky: spelunca ... tuta lacu nigro nemorumque tenebris (237f.). So too is the principal sacrificial offering: quattuor ... nigrantis terga iuvencos (243), as well as one of the subsidiary victims: atri velleris agnam (249), and also the altars which Aeneas sets up for Dis: nocturnas ... aras (252). At 255 this

gloomy anticipatory imagery is interrupted by a clearand contrasting indication of the time of day (in the outer world) at which Aeneas will

enter the nether world: ecce autem primi sub limina solis et ortus (255). Once the time of day is thus fixed, the gloomy atmosphere returns in the umbra of 257, in the umbrae (in a different sense) silentes of 264, and in the invocation of the gods of the Underworld (264-7) who control (inter alia) loca nocte

tacentia late (265) and who

inspire

Virgil to pandere res alta terra et caligine mersas (267). The context, then, has already evoked a gloom-ridden ambience both in the outer and nether worlds, but has made the point that in the outer world it is sunrise. Hence at 268, after the invocation of the

nether gods has come to an end and the narrator resumes with /bant obscuri sola sub nocte per umbram, nocte must refer to the night, i.e.

the darkness,!? of the Underworld. This identification is rendered unmistakable in 270-2, where the journey of Aeneas and the Sibyl through that ‘night’ is compared to a journey through the woods in the real night of the outer world: ubi caelum condidit umbra/ Iuppiter et rebus nox abstulit atra colorem (271£.). The interpretation of sola in line 268 must, then, begin with that fact. Now no traditional

AENEID 6.268: IBANT OBSCURI

association

exists

between

99

the

‘night’

of the

Underworld

and

solitariness. But there is a strong ancient link, which begins in Homer, between the darkness of Hades and the concept of its

permanence. At the beginning of the Nekyia, Odysseus’ ship arrives in the land of the Cimmerians, which lies near tne entrance to the

Underworld (Odyssey 11.14-19): ἔνθα δὲ Κιμμερίων ἀνδρῶν δῆμός TE πόλις TE, ἠέρι καὶ νεφέλῃ κεκαλυμμένοι᾽ οὐδέ ποτ᾽ αὐτοὺς Ἠέλιος φαέθων καταδέρκεται ἀκτίνεσσιν, οὔθ᾽ ὁπότ᾽ ἂν στείχῃσι πρὸς οὐρανὸν ἀστερόεντα, οὔθ᾽ ὅτ᾽ ἂν ἂψ ἐπὶ γαῖαν an’ οὐρανόθεν προτράπηται, ἀλλ᾽ ἐπὶ νὺξ ὀλοὴ τέταται δειλοῖσι βροτοῖσι.

That the permanent darkness of this area extends to the Underworld itself is confirmed at Odyssey 11.155 and 223. The Homeric Hymn to Demeter 80 and 402f. provides further testimony, as does Aristophanes Frogs 273 and also 145ff. when taken along with 154ff.: Heracles describes the two distinct regions of the Underworld. That Virgil is working within this tradition is confirmed by Deiphobus' words to Aeneas at 6.534, asking him what fortune brings him to the tristis sine sole domos (cf. also 6.390 umbrarum hic locus est, somni noctisque soporae, and 264, already noted above).'® Virgil's emphasis

on the perpetuity of night in the Underworld is intended to draw a clear distinction between the greater part of the infernal world, where

darkness prevails, and ‘Elysium’ (6.637ff.). There, as Virgil is at pains to explain, light prevails because the region has its own sun and stars: largior hic campos aether et lumine vestit/ purpureo, solemque suum, sua sidera norunt (6.640f.)." At this later point in Book 6 Virgil is concerned to stress that, immediately Aeneas enters *Elysium', he is in a region of light. So earlier, at 268, Virgil wished to emphasise in sola sub nocte that Aeneas and the Sibyl had entered the realm of unique, i.e. perpetual, night.

NOTES l

LSsv.B;

OLD sv. 2, 3.

2.

LSs.v. A; OLD sv. S.

3.

C.G.Heyne P. Vergili Maronis Opera ed. quarta; curavit G.P.E. 1830-41) ad loc.

4.

A. Forbiger P. Vergili Maronis Opera pars II (Leipzig 1872) ad loc.

Wagner

(Leipzig

100

STRATIS

KYRIAKIDIS

E. Benoist Oeuvres de Virgile vol. 1 (Paris 1882) ad loc. J. Conington and H. Nettleship The Works of Virgil with (4th edn London

a Commentary vol. II

1884) ad loc.

E. Norden P. Vergilius Maro. Aeneis. Buch VI (4th edn Stuttgart 1957) ad loc. H. Diels and W. Kranz Die Fragmente der Vorsokratiker (Dublin/Zürich 1972) 1.331.

Cf. ibid. 318, fr. 17.33. F.F. Fletcher Virgil. Aeneid VI (Oxford 1951) ad loc. R.G. Austin P. Vergili Maronis Aeneidos liber sextus (Oxford 1977) ad loc. E. Paratore Vergilio Eneide III (Mondadori, 2nd edn 1988) ad loc. I have purposely confined myself to commentators for economy's sake, since their view has been broadly accepted by other scholars. OLD's interpretation should be added (s.v. 3 [of the darkness]) if only because the very line from the

Aeneid is given as one of the two examples in this subsection, the other being Stat. Theb. 4.439. 14.

K. Quinn

15.

J. Henry Aeneidea or Critical, Exegetical and Aesthetical Remarks on the Aeneid (1873-92, repr. New York 1972) 111.275f. also emphasised that nocte (268) is

Virgil's Aeneid: A Critical Description (London

Pollard ‘Something odd about Virgil' PVS 7 (1967-8) 43f.

1968) 165; cf. J.R.T.

darkness, and not the time of night. 16.

Later writers reiterate the perpetual darkness of the Underworld: e.g. Lucian

Cataplous sive Tyrannus 2: τὰ δ᾽ ἄλλα ζόφος καὶ ὀμίχλη xai σκότος, ἐν δὲ τῷ οὐρανῷ φαιδρὰ πάντα; ibid. 22; De Luctu 2; Menipp. 1l; Sen. Herc. Fur. 610-12, 855-7; Sil. Ital. Punic. 13.523f.; Claud. Rapt. Proserp. 1.55f.,cf. 260f. with 262f.; 2.330f. The theme survives into Dante's Inferno (c. 111.87: “nelle tenebre eterne in

caldo e 'n gelo") where the poet's guide to the lower world is Virgil himself. 17.

For

the

earlier

and

later tradition

of illumination

in Elysium,

cf. Heyne,

Forbiger, Conington-Nettleship, Benoist and Austin ad loc.; J. Knight ‘Vergil’s Elysium’, in Vergi! ed. D.R. Dudley (London

1969) 164f. and e.g. Sil. Ital. Pun.

13.550ff.; Val. Flacc. 1.842ff.; Claud. Rapt. Proserp. 2.282ff.; Lucian Ver. Hist. 2.12.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 101-22 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

IMITATION AND ORIGINALITY IN OVID AMORES

1.3

FRANCIS CAIRNS

Iusta precor: quae me nuper praedata puella est, aut amet aut faciat cur ego semper amem. a, nimium volui! tantum patiatur amari;

audierit nostras tot Cytherea preces. accipe, per longos tibi qui deserviat annos; accipe, qui pura norit amare fide. si me non veterum commendant magna parentum nomina, si nostri sanguinis auctor eques, nec meus innumeris renovatur campus aratris, temperat et sumptus parcus uterque parens, at Phoebus comitesque novem vitisque repertor hac faciunt et me qui tibi donat, Amor, et nulli cessura fides, sine crimine mores, nudaque simplicitas purpureusque pudor. non mihi mille placent, non sum desultor amoris: tu mihi, si qua fides, cura perennis eris.

5

10

15

tecum, quos dederint annos mihi fila sororum,

vivere contingat teque dolente mori. te mihi materiem felicem in carmina praebe,

provenient causa carmina digna sua. carmine nomen habent exterrita cornibus Io et quam fluminea lusit adulter ave quaeque super pontum simulato vecta iuvenco virginea tenuit cornua vara manu. nos quoque per totum pariter cantabimur orbem iunctaque semper erunt nomina nostra tuis. (Ovid Amores

20

25

1.3)

101

102

FRANCIS CAIRNS

I. Introduction

Amores 1.3 concludes the programmatic triad of elegies which prefaces the Amores, an arrangement paralleling the beginning of Propertius Book 3.! Amores 1.1 recounts Ovid's forcible conversion from epic to elegy, when Amor caused him to fall in love; 1.2 narrates Ovid's seizure as praeda by the god, and the ensuing triumph of Amor; and 1.3 first introduces the puella with whom Ovid is in love. She, and not Love, is now said to have made him praeda (1.3.1), and

Ovid finally addresses her directly. The studies of Amores 1.3 which antedate McKeown's commentary? already touch on the main concerns of the present paper, the elegy's literary indebtedness, its originality or lack of it, and its humour, all matters covered also by McKeown. As regards sources and originality, the manifest indebtedness of the central section of

Amores 1.3 to Propertius 3.2 has frequently been noted,’ although the Ovidian elegy has more often been seen as derivative in a general way.* McKeown writes that it presents “ἃ nexus of familiar themes and ideas", although he finds some Ovidian originality in its supposed dramatic framing as a suasoria.? As for humour, apart from pointers to individual jokes, it has frequently, although not universally," been claimed that Ovid undercuts his protestations of fidelity to the puella in various ways, notably in his choice of exempla in lines 21-4. A comprehensive treatment of these questions would require inter alia an examination of the triad Amores 1.1-1.3 against the background of Propertius 3.1-3.3, and an investigation of the links between Amores 1.3 and other Amores, especially 2.17.* The present paper has more modest aims: it concentrates on Amores 1.3 in isolation from this wider context, dealing in turn with its opening lines (SII), its genre ($III), and a major, although almost unnoticed, literary influence upon it ($IV), and it ends with a brief consideration of the elegy's originality ($V). Two related propositions inform this paper. The first is that, while the elegy is indeed, as McKeown wrote, *a nexus of familiar themes and ideas", it is more: not just in its central section, but throughout, Ovid has specific literary models. With the exception of Propertius 3.2, they have not survived, but their shadows can be detected. The second proposition is that, by focussing on these models, by analysing Ovid's interaction with them, and by paying closer attention to the language and genre of the elegy, we can delineate more sharply Ovid's intellectual independence from his models and his witty originality.

IMITATION

AND ORIGINALITY

IN OVID AMORES

1.3

103

II. The Opening Quatrain and its Setting

The opening quatrain of Amores

1.3 is succeeded by a striking

syntactical and logical discontinuity. Syntax first: in lines 1-4 the

puella is referred to in the third person; and the addressee, as we learn — only obliquely and only in line 4 through the (again) third person reference to her there — is the goddess Venus. But in line 5 Ovid suddenly starts to speak to his puella directly in the second person. Similarly logic: the first quatrain consists of prayers to Venus, first to cause the puella to love Ovid, then (at least) to cause the girl to permit Ovid to love her. But line 5, accipe, per longos tibi qui deserviat annos, introduces a concept new to this elegy — servitium amoris.? Now, in one sense deserviat (5) follows on conceptually from tantum patiatur amari (3), as the latter does from amet etc. (2): Ovid is moving through the erotic gamut from one extreme (the girl loving him) to

the other (his being the girl's slave); and serviat also brings out one implication of praedata ... est (1), since in Roman legal terms the *captured' Ovid would ipso facto be reduced to servile status. But in another sense the transition from ‘love’ (of one sort or another) in

1-4 to ‘slavery’ in 5 involves a logical jump, and is intended to come

as a shock to the reader. Since such logical discontinuities were cultivated by hellenistic poets, both Greek and Roman,'? this example, with its surprise enlivening effect, is not uncharacteristic. But its sharpness draws attention to the self-contained nature of the initial quatrain; and this in turn prompts the hypothesis that in lines 1-4 Ovid has translated, or adapted, four lines of a Greek epigram.'! Certain lexical and syntactical details of lines 1-4 support this notion: nuper (1) 7 ἄρτι," praedata ... est (1) =a part of ἀγρεύω, or συλάω, or ἕλκω, or aipéw,!?

aut ... aut (2) = ἥ ... 1$,'^ the grammatical variation of person, mood and voice in amet (2), amem (2), and amari (3), 5 and even the Greek

epithet Cytherea (4),'5 all these recall the Greek Anthology.'’ Various metrical and rhythmic features of the quatrain, particularly the strong sense-breaks after precor in 1 and volui in 3, further reinforce the impression of an epigrammatic model. The insertion of Latin ‘renderings’ of Greek epigrams into Roman elegies and their £&epyaoía within elegies are well-attested and wellstudied phenomena.!* Initial incorporation of a ‘translation’ of four Greek epigrammatic lines is paralleled at the start of Propertius' Monobiblos,

(Meleager)

where

1-4, and

1.1.1-4

‘translates’

elsewhere.

AP

Horace’s

12.101

‘motto’

=

G-P

technique

103

is

analogous.?° Ovid's hypothesised epigrammatic original does not

104

FRANCIS CAIRNS

survive; but Prof. W.G. Arnott suggests an epigram of Asclepiades. The sense-breaks at precor (1) and volui (3), the avoidance of a final

witty ‘point’ (compensated for by Ovid — see below), the invocation of Venus at the end of the quatrain (4), but most of all the overall

plaintive tone of 1-4 and their ending in mingled despair, hope and poignant uncertainty, all support this suggestion.?! Ovid's use of DSSS in line 1 is, to be sure, alien to Asclepiades (as to Posidippus); but would Ovid have known this? Line 3's DDSD occurs seven times

in Asclepiades.?? The first quatrain also contains wit and conceptual dexterity characteristic of Ovid; if the model was Asclepiades, Ovid has added these. Two minor jokes appear in line 2, in which Ovid asks that the puella should either love him (i.e. in return for his love), or should act

in such a way that he always loves her. First, if the puella ‘loves’ Ovid, she will grant him her favours; and if she acts in such a way as to make his love perpetual, she will also be granting him her favours.? Hence the seemingly strongly disjunctive aut ... aut of 2 in fact convey ‘alternatives’ which

amount

to the same thing.?* Secondly, line 2

makes a humorous implied threat: if Ovid is not granted the girl's favours, his love will not be perpetual, i.e. he will find another girl. A third, more covert joke extends from lines 1-4 into 5: it derives from the successive reductions in status which Ovid suffers because he experiences not one, but two rejections. The first rejection is transparent, since the initial exclamation of line 3, a, nimium volui!,

signals that the puella has rejected him between lines 2 and 3. Then Ovid immediately makes a second request, that the girl should allow herself to be loved, that is, by himself (3).

This less ambitious aspiration is also rebuffed, but that fact has been obscured by doubts over the interpretation of line 4: audierit nostras tot Cytherea preces. McKeown II ad loc. construes audierit as

a future perfect indicative, translating this and the second half of the preceding line (tantum patiatur amari) as: "let her only allow me to love her (3), Venus will have heard my many prayers (4)". This interpretation assumes that patiatur is equivalent to si patietur. McKeown then goes on to reject the alternative view of audierit as a jussive perfect subjunctive, asserting that “When Ovid recited this elegy, his intonation presumably prevented confusion". It is surely unacceptable to introduce recitation as a means of solving this problem, since Ovid was well aware that his readers would greatly outnumber the audiences at his recitations. In any case, the linguistic

context makes it overwhelmingly clear that audierit is a jussive

IMITATION AND ORIGINALITY

ΙΝ OVID AMORES

1.3

105

perfect subjunctive. This form is standard in wishes and prayers, and

it can be used in solemn invocations of deities.? McKeown himself (II ad loc.) notes

the phrase

nostras tot ... preces as "perhaps

stylistically elevated", and he remarks on some sacral aspects of line 3 (ad loc. — and it will emerge (SIII) that the entire ambience is one

of prayer formulae. The reason for dwelling on this dispute is that, if audierit were

indeed a future perfect indicative, then the girl's second rejection of Ovid would be obscured. Thus some of the benefits of recognising lines 1-4 as an ‘epigram’ followed by a logical discontinuity, including appreciation of the full extent of Ovid's wit, would be lost.

The correct interpretation of line 4 as a third-person entreaty to Venus to listen to (i.e. to assent to) Ovid's preces throws light on the fact that in line 5 Ovid suddenly stops addressing Venus and starts to plead with the girl herself. The reader assumes, prompted by recent experience of lines 2f. (where the same thing occurred), and more generally by knowledge of similar phenomena in ancient poetry,”’ that something has happened after line 4 which must be understood from the context. In this case it is that the girl has again said “No”,

this time to Ovid's aspiration of line 3, and that she has done so despite Ovid's use of Venus as his mediatrix. That is why, in line 5, Ovid abandons Venus and now pleads directly with the puella. The turning of the lover for help away from the love-gods to humans is yet another reflection in Amores 1.3 of the traditions of hellenistic epigram. For example, at AP 12.84 = G-P 114 (Meleager) line 1 ἄνθρωποι and at AP 12.85 = G-P 115 (Meleager) lines 1 and 8f. oivondtat/Esivor are the ultimate recourse of the ‘shipwrecked’ lover rather than the gods of love. In Amores 1.3 Ovid wittily turns, not to third parties, but to the beloved herself as, in effect, a more powerful goddess in matters of love than the love-goddess; and the remainder of the elegy is accordingly addressed to the puella. Amores 3.2.55-60 offer a parallel train of thought: an invocation of Venus which repeats the plea patiatur amari (referring to the girl, 57) is followed by Ovid's explicit words to the girl: pace loquar Veneris, tu dea maior eris (60). III. Genre

The sacral language of the prayers to Venus of lines 1-4, and the redirection of those prayers to the puella at line 5ff., have already

been touched on briefly. In fact the entire elegy contains prayer

106

FRANCIS CAIRNS

elements, both linguistic and non-linguistic. These will be surveyed rapidly as a preliminary to asking whether it would be helpful to classify the elegy as a ‘prayer’ or a ‘hymn’. Line 1 begins iusta precor.? precor signals that ‘prayers’ are being made, and iusta is reminiscent of Menander Rhetor's pronouncement in his section on ‘euktic and apeuktic hymns’ that “prayers must be just" (343.5 Spengel: τὰς μὲν yàp εὐχὰς δικαίας εἶναι χρή). As noted, the jussive perfect subjunctive form of audierit is connected with prayer formulae; and audierit is combined with preces (4), and

employs hymnic third person ‘Er-Stil’?® in its reference to Venus (Cytherea), while the verb audire is a technical term of prayer.?? So too is accipe (5f.), although accipere is used of prayers, not of the person praying.*! The fact that lines 5ff. are addressed to the puella, not to Venus, is unproblematic, since requests addressed to human

beings are a standard feature not just of parodic ancient prayers and hymns but of some serious ones too.?? It was commonplace in antiquity for petitioners to advance their own merits as a reason why the deity being beseeched should grant their prayers.? Sometimes these merits consisted in earlier worship of the same god, but they might, as here, be personal and moral. Of the many parallel passages collected by Appel (n.30) 151, two may be quoted: Catullus 76.19: si vitam puriter egi; and Horace Satires

2.6.4ff.: nil amplius oro,

Maia nate, nisi ut si neque maiorem nec sum facturus si veneror stultus

propria haec mihi munera faxis, feci ratione mala rem vitio culpave minorem, nihil horum ...

5

Ovid's long argument about his own merits (Amores 1.3.7-14) conforms with this practice, and lines 13f. in particular reflect, like the parallels just quoted, standard expressions. Again, at lines 16-19 Ovid employs second person ‘Du-Stil’ in polyptoton: tu (16), tecum (17), teque (ablative, 18), te (accusative, 19), cf. tuis (26). Both are

typical features of prayers and hymns.?^ Lines 16-20 and the final couplet (25f.) also incorporate a further topos of prayer, the benefits

which the deity will receive if the suppliant's prayers are heard.?? Is Amores 1.3, then, a εὐκτικὸς ὕμνος, in which Ovid initiates an example of a genre normally addressed to a deity with prayers to Venus, and then wittily substitutes as addressee for that deity (who appears powerless to help him) the mortal girl who has it entirely in

her power to help him or not? This view would make Amores 1.3

IMITATION AND ORIGINALITY IN OVID AMORES

1.3

107

Gebetsparodie, not covered, as many other ancient examples are not, by Kleinknecht (n.30), but nevertheless just as substantial, if more

sophisticated, than many he does cover. In order to secure this thesis two alternative generic possibilities need to be eliminated. The first is the komos/paraclausithyron, which portrays the activities, including the pleas, usually of the excluded or rejected lover.? The repeated accipe of Amores 1.3.5f., especially in combination with iusta precor (1) and preces (4), might seem to mark the

elegy as a komos: cf. the earliest surviving komastic fragment, Alcaeus fr. 374 LP: δέξαι ne κωμάσδοντα, δέξαι, λίσσομαί σε, λίσσομαι, with its repeated 5€Ea1 and λίσσομαι. That the conceptual

linkage 6£5av/Xíccopat was strongly komastic is shown by its reappearance in Pindar Paean 6.3-5 Sn.-Mae.;" cf. also the repeated δέξασθε of an epigrammatic komos by Meleager (AP 12.85 = G-P 115, 1 and 10), which declares itself one in κωμάζω (7). A further komastic pointer might be seen in praedata est (Amores 1.3.1) — cf. ἀγρεύσας ἕλκει in the same Meleager komos (AP 85.4), and also

ἕλκει at Meleager AP 84.3.” However the rest of Amores 1.3 is embarrassingly poor in komastic topoi. Lines 7-16 could, of course, be classified as a “Vanto delle

proprie virtü, ricchezze etc.”, for such ‘vanti’ do sometimes occur in komoi;? and there is, admittedly, nothing in them which a komast could not properly utter. But there is no door, or its like, in Amores 1.3, and there are no further indications of a komastic situation or of

one the poet considered komastic: thus — and this is perhaps the telling point — not enough of the primary elements of the genre (for which,

cf.

Pinotti

(n.36)

61)

are

present.

The

overall

thematic

congruences of the two Meleager epigrams with Amores 1.3.1-6 certainly reflect an ambience common to them and to the elegy —

and also to the lost model of Amores 1.3.1-4. But although the elegy begins (apparently) in komastic mode, it seems not to continue in it. Two ancient generic techniques (which are not mutually exclusive)

may explain what is happening. The first is *inclusion', whereby an example of one genre is ‘included’ in a subordinate capacity in a poem

belonging

to another

genre.

The

second

is ‘deception’:

Roman and Greek hellenistic poets in particular may tempt readers into premature and false judgements about the genres of poems

before disabusing them.*! If deception is involved in Amores 1.3, it takes a sophisticated form: the elegy begins with signals of a εὐκτικὸς ὕμνος; but then lines Sff., with their new addressee and their verbal

pointers to the komos, lead the reader to reinterpret lines 1-4 as

108

FRANCIS CAIRNS

komastic. However 7-26 are not unequivocally komastic, and the reader eventually reverts to the earlier reading of the elegy asa euktic hymn.

The sophistication lies in the location of these (un)deceptions: most often deceptions occur at the beginnings of poems; the reader is then shortly disabused. Here, however, initial signals turn out to have been accurate after all, and it is the subsequent strand of generic

information that eventually is perceived as misinformation.* The effect of Amores 1.3's deception and undeception is to underline Ovid's other original touch in directing a euktic hymn to a mortal. In other cases of deception the poet arguably has objectives beyond the merely dramatic and enlivening: a literary-historical or thematic link between two different genres may implicitly be suggested. Thus Ovid may here, through the preces/accipe collocation, be hinting mischievously

that

the

komos

(or

at least the lover's

komastic

speech) is a kind of euktikon, just as Tibullus in 2.6 had implied a link between komos and propemptikon by ‘including’ the latter in the former.^* Alternatively Ovid may just be touching en passant within a

euktic hymn upon the komastic aspect of love, which crops up in many elegies not generically komoi.*55 But all we know about Ovidian self-awareness and originality points rather in the first direction. The second generic possibility to be eliminated is that Amores 1.3 is

“framed dramatically as a suasoria”‚* an identification which goes back to Briick,*” who emphasised (62) the elegy's use as an argument of τὸ συμφέρον με, a standard topos of the suasoria. Iusta (Amores 1.3.1) might also be advanced in this context since to óíkatov/iustum is another symbouleutic/suasoria topic. One might even (but with little plausibility) argue for the presence in the elegy of two of the remaining three standard suasoria topoi — τὸ πρέπον (in digna, 20, cf. TLL s.vv. decere, decus, dignus) and τὸ ὅσιονΖτὸ νόμιμον (in lines 13f.). But Amores 1.3 lacks a suasoria framework in any recognisable sense; it looks back rather to a long tradition of erotic persuasion

ultimately inspired by Plato's Phaedrus, but generically it is a witty euktic hymn which dismisses divine help as being far less efficacious

than the practical assistance of Ovid's human puella who becomes after line 4 its primary addressee. IV. The presence of C. Cornelius Gallus The canon of Roman elegiac poets, as established by Ovid at Tristia 4.10.51-4 (cf. also Tristia 2.445-68), contains four names: those of

IMITATION

AND ORIGINALITY

IN OVID AMORES

1.3

109

Gallus, Tibullus, Propertius, and Ovid himself. Until recently only one line of Gallus’ poetry was known; but, even so, by the 1900s Franz Skutsch had already advanced bold speculations about

Cornelius Gallus’ importance for Virgil’s Bucolics, for the Appendix Vergiliana, and for Roman elegy.*? Skutsch's suggestions were for the

most part“ either rejected or ignored up to the 1970s, when David O. Ross began to reinstate Gallus as a major voice in the work of other Augustan poets. The publication in 19795! of the Qasr Ibrim papyrus, containing some eight new lines of Gallan elegy (= frr. 2-4 Büchner, but referred to in this paper as ‘Gallus’ with their line numbers), stimulated further interest in Gallus; and there is now a

party which argues, following Franz Skutsch, that the Gallus of most elegies of Propertius Book | is the poet C. Cornelius Gallus,?? with all the implications which that view entails. In this section it will be claimed that, in addition to the presence of a Greek epigrammatist and of Propertius in Amores 1.3, Cornelius Gallus has contributed largely to it. The starting point is Stephen Hinds'? inclusion of Amores 1.3.19f., the couplet which introduces the final section of the elegy: te mihi materiem felicem in carmina praebe: provenient causa carmina digna sua.

in his list of poetic passages of the late republic and early empire where carmen dignum, carmina digna, and the like appear. The most celebrated, of course, is Gallus 6f.: ... tandem fecerunt c[ar]mina Musae

quae possem domina deicere digna mea.

As Hinds observed, the contexts in which carmina digna etc. appears may have something further to tell us about Gallus and his influence. Hence the case will be put for the presence of more Gallan material in Amores 1.3 first by stressing what some of Hinds' contexts for carmina digna etc. have in common with other elements of Amores 1.3, and then by arguing from other known facts about Gallus and his work. The earliest parallel in Hinds' list is Lucretius 3.420: digna tua pergam disponere carmina vita.*

It has various possible implications: a) that Lucretius imitated Gallus. b) that Gallus imitated Lucretius.

110

FRANCIS CAIRNS

c) that both had a common source.

d) that the phrase carmina digna is merely a poeticcommonplace. Brevity demands dogmatic resolutions: a) Relative chronology makes it unlikely that Lucretius imitated Gallus. Gallus was probably born in 69 or 68 BC, Lucretius between

98 and 94 BC.” b) That Lucretius may have influenced Gallus is plausible, both for

chronological reasons and because the preceding line of Lucretius (conquisita diu dulcique reperta labore, 3.419) is a complex of hellenistic

programmatic

concepts:

sweetness,’®

time-consuming

poetic composition," and toil.5* These concepts, along with the fact that

Lucretius' poetry and his patron

Memmius'

vita are being

approximated? in line 420 could have added to the intrinsic appeal of Lucretius’ language for the ‘neoteric’ Gallus; and they could have led Gallus to imitate Lucretius in a programme for his own vita and

poetry which in turn provided a model for Ovid in Amores 1.3. €)

A

plausible

alternative

(or

additional)

hypothesis

Lucretius and Gallus were both drawing on a common

is that

source, an

older Roman poetic or, less likely, religious text where carmina digna or something similar (carmen dignum?) appeared. In this connection the curiously numerous verbal coincidences between the lines surrounding Lucretius 3.420, Gallus 6f., and Amores 1.3.19f. and 25f. (listed below) come into play. So doces the fact that, while in Gallus and Ovid their contexts are entirely literary, in Lucretius only lines 3.419f. discuss poetry, with the result that most of the Lucretian verbal coincidences appear in lines which have nothing to do with literature. A possible explanation runs as follows: carmina digna came to Lucretius, along with other concomitant terminology, from

a source in which they were all used about poetry. Lucretius applied digna ... carmina to his own poetry, but reemployed the concomitant terminology, perhaps unconsciously, in adjacent lines dealing with philosophy. In Ovid, however, the same concomitant terminology is found in a context entirely to do with poetry, which suggests that in Ovid's model it also appeared in a fully literary context. Ovid's model must, then, have been Gallus, not Lucretius; and Gallus must have

had independent access to the same source as Lucretius, since he too used the concomitant terminology in a poetic context. The common source of Gallus and Lucretius may be conjectured from Lucretius’ use of dignum ... carmen at 5.1: quis potis est dignum pollenti pectore carmen

IMITATION

AND ORIGINALITY

IN OVID AMORES

1.3

Hit

This, the first line of the literary prologue of De Rerum Natura 5, has an Ennian beginning. More specifically it echoes: quis potis ingentis oras evolvere belli. (164 Skutsch)

Skutsch confirms that the latter was the first line of Annales Book 6, and comments “from Virgil [i.e. Aeneid 9.528] we gain the impression that the line was followed by a prayer to the Muses".! Even if the Muses — and the easy, but uncompelling, thoughts of Musae at Gallus 6 and Phoebus comitesque novem at Amores 1.3.11 — are set aside, the literary prologue of a key book of Ennius' Annales makes an attractive starting point for the story of dignum carmen/carmina digna. d) The view of carmina digna etc. as a mere poetic commonplace is implausible. Too much contextual material accompanies it in too many of its appearances, particularly the earlier ones, for it to be just a topos. On the basis of this discussion, and of Hinds' demonstration that Gallus was the main vector for the phrase carmina digna in subsequent poetry, we can return to Amores 1.3. Lucretius provides further illumination for its final line, although here too the immediate source was Gallus (see also below on nomen etc.). Lucretius 3.420,

digna tua pergam disponere carmina vita, is immediately followed by: tu fac utrumque uno sub iungas nomine eorum. (3.421)

Although the subject matter of this line is non-literary (it refers to the anima and animus), the line nevertheless relates clearly to Amores 1.3.26 (here cited as part of the final couplet): nos quoque per totum pariter cantabimur orbem iunctaque semper erunt nomina nostra tuis. (25f.)

AWPwWnN >

At this point the remaining verbal coincidences Lucretian context, Gallus and Ovid may be listed.

between

this

unum inter se coniuncta (3.424), cf. Amores 1.3.25f. totum ... orbem (3.410), orbis (3.415), cf. Amores 1.3.25 causa (3.422, 429), cf. Amores 1.3.20 dicere (3.422, 423), cf. Gallus 7

possis (3.418), cf. Gallus 7: possem noscere (3.418), cf. Amores 1.3.8, 21, 26 — more impressive if

seen in the light of the Gallan complex involving nomen and its cognates discussed below

112

FRANCIS

CAIRNS

7. personal pronouns and adjectives: tua (3.420) and tu (3.421), cf. Gallus 1, 2 (three), 4, 5, 7, 8 (two), 9; Amores 1.3.1, 4, 5, 7, 8,9,

12 (two), 15, 16, 17 (two), 18, 19 (two), 25, 26 (two). Another passage which further confirms the Gallan antecedents of the final couplet of Amores 1.3, by providing a parallel for line 25 and linking it with carmina digna, is Virgil Eclogue 8.9f.: en erit ut liceat totum mihi ferre per orbem sola Sophocleo tua carmina digna coturno.

Its literary programmatic content and the presence in it of a patron with literary pretensions (Pollio) may be noted. Eclogue 8.7f. offers yet more to demonstrate the Gallan ancestry of this context: ... en erit umquam ille dies, mihi cum liceat tua dicere facta?

Here the d(e)icere of Gallus 7, and of Lucretius 3.422 and 423, recurs; compare also umquam (7) with tandem (Gallus 6). Noteworthy too are facta (8) (albeit in a different sense), fac (Lucretius 3.421), and fecerunt (Gallus 6).? Another dignum carmen reminiscence, Catalepton 14.2£.: Troius Aeneas Romana per oppida digno iam tandem ut tecum carmine vectus eat

again displays in Romana per oppida an equivalent of totum (per) orbem, while tecum recalls pariter (Amores 1.3.25). That Gallus is the

direct source here seems guaranteed by tandem (cf. Gallus 6); and tecum additionally suggests Gallus' predilection for personal pronouns/adjectives. semper (Amores 1.3.26) finds its analogue in saeclis accepta futuris within a further carmina digna context at Catalepton 9.15f.: carmina quae Phrygium, saeclis accepta futuris, carmina quae Pylium vincere digna senem.

Together these parallel passages appear to demonstrate that Amores 1.3.19f. and 25f. are completely Gallan in inspiration. Now these two couplets surround a triad of mythical paradeigmata of learned hellenistic type (21-4). This naturally suggests the notion that the encapsulated lines may also contain Gallan material. It should be stressed at the outset that this hypothesis cannot be proved; but it certainly deserves discussion, particularly since scholars have quite naturally been curious about what the women in

IMITATION AND ORIGINALITY

IN OVID AMORES

1.3

113

the exempla might have in common besides having been seduced and impregnated by Jupiter, and then having been celebrated in poetryas a result. One correct answer has been that they were all travellers.5* A further refinement is that two of them (but not Leda, see below) moved between continents. Europa (23f.) travelled on the back of the

‘bull’ from Phoenicia (Asia) to Crete (Europe) to become the eponym

of Europe.® Io (21) was even more adventurous. Her wanderings took her from Greece (Europe) to the Black Sea and the Caucasus

(Asia), from where she eventually arrived in Egypt (Africa) by the circuitous land route.56 By a curious coincidence the single preserved line of Gallus (fr. 1 Büchner), apart from those of the Qasr Ibrim papyrus, is: uno tellures dividit amne duas.

It refers to the river Hypanis in the Caucasus, probably the modern Kuban, as dividing "two continents with its single stream." This line

was imitated by Propertius at 1.12.3f.:9? tam multa illa meo divisa est milia lecto

quantum Hypanis Veneto dissidet Eridano.

It is tempting to think that Propertius 1.12.10, lecta Prometheis dividit herba iugis

(i.e. on the Caucasus), was also inspired by the same Gallan passage. If so, Propertius will have found in that Gallan context a reference to Prometheus, whom Io met and spoke with on her journey (cf., e.g., Aeschylus Prometheus

Vinctus 561ff.). It may be, then, that Gallus’

line about the Hypanis (which Io crossed as she crossed the Bosphorus and the Nile, other boundaries between continents) came from a description of Io's wanderings, a favourite hellenistic subject treated closer to Ovid's time also by Calvus.®* Indeed, somewhere in his own Amores (whose title, it should not be forgotten, Ovid

appropriates), Gallus could have treated a group of stories about women loved by Jupiter, including Io, Leda, and Europa. If the journeys of Io and Europa between continents, and also perhaps their crossing

of famous

rivers and/or

seas in the process, were

prominent in Gallus or his original, then the stay-at-home Leda, as noted, would not fit in. But curiously Helen, Leda's daughter, would:

she crossed from Europe to Asia (and indeed brought Europe and Asia into conflict), and later, in one version, visited Egypt. If, then, the real focus of interest for Gallus, or his original, was the children

H4

FRANCIS CAIRNS

of the three ladies (Io and Europa mothered 'founders' — cf. the standard obsession of hellenistic poets with foundations of cities), this would help explain the grouping. Alternatively, Gallus may have tucked away his least apt exemplum (Leda) in the middle of a triadic paradigm, as Propertius often does, and Ovid may have followed him in this patterning. Ovid's theme at the end of Amores 1.3 is the fame to be achieved through poetry. That too is significant for the present discussion, since elsewhere (see below) Ovid testifies that Gallus also treated the theme of his own and his mistress' fame throughout the world, fame again spread through poetry. This thematic contiguity enhances the

importance

of the last feature in lines 19-26 of Gallan origin,

particularly since it hints at Gallan influence earlier in the elegy too: nomen in line 21 and nomina in line 26. Even if nothing else in the final lines was indubitably Gallan, these words would provide a basis for argument, since it is virtually certain that Gallus made great play with a group of words linked 'etymologically', namely notus (and ignotus), noscere and nosse, nomen, nota and nobilitas, and that one context for this verbal complex was claims of poetic fame for himself and his mistress Lycoris. This has been shown by a number of

scholars," and the arguments do not need to be repeated here; but the Ovidian passages incorporating the theme are worth quoting, inter alia because they both involve the ‘East and West’ theme, which

amounts to totum (per) orbem: Gallus et Hesperiis et Gallus notus Eois,

et sua cum Gallo nota Lycoris erit. Amores

1.15.29f.

Vesper et Eoae novere Lycorida terrae. Ars Amatoria 3.537

The observation of McKeown II on nomina etc. at Amores 1.3.26 that *the girl's name is conspicuously absent from this poem" (and his conclusion from its absence) gain additional force if the Gallan model celebrated the fame of the named Lycoris. In Amores 1.3 the key words nomen and nomina appear twice in a Gallan ambience

(19-26),

and

this draws

attention to the earlier

appearance of nomina in lines 7f.: veterum ... magna parentum/ nomina. There nomina might be dismissed as non-significant but for three considerations: 1. Propertius, who is the known major influence upon Ovid in these lines, was also a frequent imitator of Gallus.

IMITATION AND ORIGINALITY IN OVID AMORES

1.3

115

2. The description of 7ff. would fit not only Ovid (and Propertius) but also, and preeminently, Gallus. All were equites who retained this rank: cf. eques 8, and below.

3. If the identification of the Gallus of Propertius 1.5 asthe poet is correct (see above and nn.48-52), then it is almost certain that

Gallus made some poetic play about his nobilitas, i.e. his inherited social status or rather his lack of it, and moreover that he did so in relation to his success in love. Further than this we cannot go with security: but if nomina at Amores 1.3.8 is Gallan, then the suspicion must arise that behind the entire

centre section of the elegy lies not only Propertius 3.2 but also a Gallan original of Propertius 3.2. Such combinations of influence have recently become a subject of intense interest among students of

Latin poetry," and further study of Propertius 3.2 might further illuminate the matter. All in all then, a strong argument can be made out for powerful Gallan influence towards the end of the elegy, and possibly in the central portion. V. Ovid’s originality in Amores 1.3 A number of points have already In particular his incorporation material in lines 1-4 have been innovations (SIII). Both involve

been made about Ovid's originality. and manipulation of epigrammatic discussed (SIT), as have his generic typical Ovidian wit. His model for

lines 1-4 will certainly not have continued as do lines 5f., and it may well have left unclear whether or not Aphrodite would help the speaker. Ovid rejects such pathetic uncertainty, and moves in lines 5f. into a long, elaborate and heartfelt prayer to the girl. The witty

implication of this move is that it is useless to invoke the gods in matters of love. The gods, so Ovid later suggests, may help themselves in such circumstances, as Jupiter does in 21-4, but they

are not interested legerdemain

has

in helping mortals. a humorous

impact.

interrupted by hints of a komos

Similarly Ovid's generic The mock euktic hymn

is

which are not sustained, but the

hymn is thereby further undercut. Ovidian originality is also evident in his handling of his Roman models. It has been suggested (8IV)that, in the middle of Amores 1.3, Ovid blends Gallus and Propertius in a novel way. Another feature apparently

original

to

Ovid

is his

excessive,

and

even

comic,

insistence on his lack of senatorial background. Ovid was technically

116

FRANCIS CAIRNS

an eques and the son of an eques; but his father was head of the leading family among the Paeligni.’? Ovid could perfectly well have entered the senate, but declined to do so (Tristia 4.10.35ff.), and he

certainly had a senatorial census, since as a youth he was granted the

latus clavus (Tristia 4.10.29). His son-in-law was a senator." Ovid seems here to beoverstressing humorously a mode of self-presentation

adopted first by Gallus, and then reiterated in 3.2 by Propertius, whose social position was virtually parallel to that of Ovid.” Gallus’

origins are debated,"5 but he was certainly of equestrian status. His Caesarian links, his significant and trusted role in the post-Actian campaign as the general 'placed in charge of Egypt' by Octavianus

and dispatched there to finish off M. Antonius and Cleopatra, indicate that a glowing senatorial career would have been open to Gallus on demand. That Gallus made great play of being a (non-Jnobilis has already been argued. Presumably claims of (relative) poverty, a common poetic topos, and one notably exploited

by Catullus, son of the great villa at Sirmio, were also part of Gallus' self-portrayal, and Ovid is guying them, partly to distinguish himself from Tibullus" and Propertius, who imitated in all seriousness these same Gallan topoi. Ovid's self-portrayal may involve further humour with a local edge. temperat et sumptus parcus uterque parens (10) naturally reflects in one dimension the traditional claims of Italic hill-folk, particularly the Sabines, to thrift. But in another dimension the combination of love and poverty was considered intrinsically ridiculous in antiquity. Catullus exploited this prejudice in his satiric verses on Furius and Aurelius;’® and when attacking Furius he described in minute detail both his parents, their thrifty ways, and their consequent physical condition. Amores 1.3.10 must, therefore, contain an element of satire upon the frugal ways of the hill-folk, and also a joke at Ovid's

own expense, since he was one of them. This element may additionally, in ways more obscure to us, play off against the self-portrayal of

the other elegists. Ovid also®° sounds a sharp, cynical, and personal note in his invitation to the girl to give herself to him, i.e. both physically, and as materies for his poetry (19f.). The cycle Amores 1.1-3 documents the induction into erotic elegy of a poet who started with no intention of becoming a love-elegist, and who throughout is more concerned with literature than with love; hence this cynical dual request confirms Ovid's innermost thoughts. Ovid's selection, to exemplify the fame

conveyed by poetry, of three of Jupiter's female conquests, none of

IMITATION AND ORIGINALITY IN OVID AMORES

1.3

117

whom had a particularly pleasant time as a result of yielding to love, and who were celebrated not by the lover who enjoyed them but by subsequent generations,*! again goes hand-in-hand with Ovid's overall tone and perceptions. Ovid presents love as a necessary fiction, one always subordinate to his literary aspirations, which themselves are not so much to greatness and profundity as to wit.

NOTES The Italian language precursor of 881, II, IV and V of this paper was read on 16 May 1990 at a conference held in the University of Salerno. That script was subsequently printed with minimal annotations as "Ovidio, Amores

1.3: Dipendenza letteraria vs

indipendenza intellettuale", in 1. Gallo and L. Nicastri (edd.) Cultura poesia ideologia nell'opera di Ovidio (Pubblicazioni dell' Università degli Studi di Salerno. Sezione Atti,

Convegni, Miscellanee 33, Naples 1991) 27-40. The present paper incorporates one additional section ($IIT); the other sections have been revised and expanded; and full annotations have been provided. The work of

revision has induced some ‘canonical’ version. I am stimulated its production, given on the basis of the |.

major changes of mind, especially in 8IV. Hence this is the grateful to Prof. Italo Gallo for the invitation which first and to Prof. S.E. Hinds and Prof. J.C. McKeown for advice Italian script.

This universally understood arrangement presumably dates from the second edition of the Amores, in three rather than five books, which appeared probably

after 2 BC: cf. J.C. McKeown Ovid Amores: Text, Prolegomena and

Commentary

in four volumes |: Text and Prolegomena (Arca 20, Liverpool 1987) [henceforth McKeown I] 76f. With the exception of semi-colon for comma at the end of 1.3, the text of Ov. Am. 1.3 printed above is that of McKeown. 2.

J.C. McKeown Ovid Amores: Text, Prolegomena and Commentary in four volumes II: A Commentary on Book One (Arca 22, Leeds 1989) [henceforth McKeown II]

60-75. 3.

Asearly, for example, as R. Neumann Qua ratione Ovidius in Amoribus scribendis Properti elegiis usus sit (Diss. Göttingen 1919) 26-8. Cf. also McKeown II.60f.

4.

For a variant on this approach, cf. W. Stroh Die rómische Liebeselegie als werbende Dichtung (Amsterdam 1971) 170.

5.

McKeown IL.61. A link between Am. 1.3 and the suasoria was first suggested by C. Brück De Ovidio scholasticarum declamationum imitatore (Diss. Giessen 1909) 62. It will be rejected in 8III below.

6.

Eg.L.C. Curran 'Desultores amoris: Ovid Amores 1.3° CPh. 61 (1966) 47-9; K. Olstein ‘Amores 1.3 and Duplicity as a Way of Love’ TAPA 105 (1975) 241-57.

7.

J.F. Davidson ‘Some thoughts on Ovid Amores I, 3’, in C. Deroux (ed.) Studies in Latin Literature and Roman History II (Coll. Latomus 168, Brussels 1980) 278-85

challenges this approach and adds some strictures (of general application) upon the untramelled use of myth for interpretative purposes. I am substantially in agreement with him about the mythical content of Am. 8.

Cf,e.g., Neumann

9.

Cf. McKeown

1.3.

(n.3) 29; Davidson (n.7) 282; McKeown

IL64f. (on deserviar).

II.61f.

FRANCIS CAIRNS

10. 11.

Cf. F. Cairns Tibullus: A Hellenistic Poet at Rome (Cambridge 1979) [henceforth

Tibullus] Chh. 5, 7.

This proposal is apparently new, although A.-F. Sabot Ovide poéte de l'amour dans ses oeuvres de jeunesse (Gap 1976) 175 wrote (about lines If. only) "Ces vers,

par leur concision, leur style nerveux ressemblent à une épigramme.”

12.

Cf. An Index to the Anthologia Graeca: Anthologia Palatina and Planudea (Amsterdam 1985-90) [henceforth AP Index] s.v. ἄρτι; McKeown I1.40n modo in Am. Epig. 1.

13.

Cf. AP Index s.vv. ἀγρεύω, αἱρέω, ἕλκω vel ἑλκύω, συλάω. The notion (for which, cf. J.A. Barsby (ed.) Ovid's Amores: Book One (Oxford 1973) 51) that

praedata est “is taken over from Propertius" [2.1.55] seems less likely than dependence on Greck models by both poets.

14.

Cf. AP Index s.v. ἥ (frequently twice in a single line).

15.

Such poikilia/variatio in grammatical forms is, of course, characteristic of all hellenistic poetry, not just of epigrams: cf., e.g., F. Lapp De Callimachi Cyrenaei tropis et figuris (Diss. Bonn

16. 17.

1965) 109-11, covering all works of Callimachus.

Cf. AP Index s.v. Κυθέρεια. The fleeting ring-composition of precor (1) and preces (4) is more difficult to

evaluate. Ring-composition is fairly universal in ancient poetry and prose: cf. Tibullus Ch.8 (with 194f. n.4 for bibliography). But the brevity of epigrams and their need of movement to achieve their 'point' perhaps makes ring-composition less frequent there than in more lengthy pieces. A similar effect is, however, sometimes found, i.e. first and last lines include forms of two cognate terms. Cf.,

e.g., Asclepiades (on whom see further below) AP 5.85 (= G-P 2); AP 5.158 (=

G-P 4), AP 5.167 (= G-P 14); AP 6.308 (= G-P 27); AP 2.209 (= G-P 36); AP 12.17 (= G-P 37); AP 9.752 (= G-P 44).

18.

E.g. A.A. Day The Origins of Latin Love-Elegy (Oxford 1938) Ch.6; E. Schulz-

Vanheyden

Properz und das griechische Epigramm (Diss. Münster

1969); G.

Giangrande ‘Motivi epigrammatici ellenistici nell'elegia romana’ in E. Flores

(ed.) Dall'epigramma ellenistico all'elegia romana. Atti del Convegno della S.LS.A.C. (Napoli 27 novembre 1981) (Naples 1984) 29-58. Cf. also below, n.19. 19.

On Propertian examples, cf. Schulz-Vanheyden (n.18) 101-32. Ovid's ἐξεργασία of epigrams in his elegies has been discussed most recently by McKeown (1.70; 11.337f., 364f.). It does not necessarily involve initial ‘quotation’. Cf. also Day (n.18) 126f. (Propertius), 113-16 (Tibullus) and 127-37 (Ovid).

20.

On the further complexities of Horatian 'motto' technique, cf. F. Cairns aus Hymn to Hermes, P. Oxy. 2734 Fr. | and Horace Odes 1,10° QUCC 42

(1983)

21.

29-35.

The first three features could also be exemplified from many other epigrammatists. The last two are rarer. On the style of Asclepiades, cf. O. Knauer Die Epigramme des Asklepiades v. Samos (Diss. Tübingen, publ. Würzburg, 1935) 51-7; H. Ouvré Quae fuerint dicendi genus ratioque metrica apud Asclepiaden, Posidippum et Hedylum (Diss. Paris 1894) Ch.6 discusses the styles of the three poets.

22.

Cf. Ouvré (n.21) 105f.

23.

Cf. McKeown

24.

For a similar Ovidian use of aur... aut, cf. M. Paschalis **Aut ego fallor aut ego

II.62f. ad loc.

IMITATION AND ORIGINALITY IN OVID AMORES

1.3

119

laedor" (Ovid Metamorphoses 1,607-608). A witty tautology?’ Eranos 84 (1986) 62f. 25. 26.

E.g. Plaut. Men. 295; Ter. Hec.

McKeown

II ad loc. also notes (specifically) iusta precor (1) as a “solemn

expression", and comments treatment as a goddess.

27.

102; Cic. Rep. 4.8; Virg. Aen. 6.62; Liv. 28.28.

(on 3f.) on tantum

For such ‘pauses’ in ancient

in prayers and on the girl's

poems, during which the response of the person

addressed is taken as made, cf., e.g., E. Fraenkel Horace (Oxford 1957) 181 (on Hor. Od. 1.27); F. Cairns ‘Theocritus Idyll 10° Hermes 98 (1970) 38-44, 39 (on

Call. Ep. 30 Pf. = AP 12.71).

28.

Cf. also above, n.26.

29.

On ‘Er-Stil’, cf. E. Norden Agnostos Theos. Untersuchungen zur Formengeschichte religióser

Rede

(Leipzig-Berlin

1913)

163-6;

L.

Deubner

'Ein

Stilprinzip

hellenistischer Dichtkunst’ NJb. 24 (1921) 361-78, 363-5; H. Meyer Hymnische

Stilelemente in der frühgriechischen Dichtung (Diss. Würzburg 1933) 4, 39, 63; K. Keyssner, Gottesvorstellung und Lebensauffassung im griechischen Hymnus, (Würzburger Studien zur Altertumswissenschaft 2, Stuttgart 1932) 2, 30f.; P.J.T. Beckmann Das Gebet bei Homer (Diss. Würzburg 1932) 57-9. On ‘Er-Stil’ combined with ‘Du-Stil’ (on which cf. below, n.34), cf. Norden (above) 163f.; Deubner (above) 363-5; Beckmann (above) 56-9; W.H. Mineur (ed.) Calli-

machus. Hymn to Delos. Introduction and 1984) 6, 74, 193.

30.

Cf. McKeown

Commentary (Mnemosyne Suppl. 83,

II on 1l.3f.; G. Appel De Romanorum precationibus (Religions-

geschichtliche Versuche und Vorarbeiten V1.2, Giessen 1909) 119. For Greek and

Roman material, cf. K. Ziegler De Precationum apud Graecos Formis Quaestiones

Selectae (Diss. Breslau 1905) 62-7, and (for Greek equivalents) F. Schwenn Gebet und Opfer. Studien zum griechischen Kultus (Religionswissenschaftliche Bibliothek 8, Heidelberg 1927) 28, 58; Keyssner (n.29) 88, 99; H. Kleinknecht Die Gebetsparodie in der Antike (Tübinger Beiträge zur Altertumswissenschaft 28,

Stuttgart-Berlin 1937) 22, 139.

31.

Cf. Appel (n.30) 138.

32.

Cf. Kleinknecht (n.30) Index s.v». Hymnus — auf Menschen; S.R.F. Price Rituals and Power (Cambridge 1984) Index s.v. choirs honouring emperors; and, e.g., the dithyramb in honour of Demetrius Poliorcetes (Athen. Deip. 253C-F) and the

paean for T. Flamininus (Coll. Alex. 173).

33. 34.

Cf. Beckmann (n.29) 46-9; Appel (n.30) 149-51. For 'Du-Stil', including polyptotic examples, cf. Norden (n.29) 143-63; Deubner (n.29) 363-5; Meyer (n.29); Keyssner (n.29) 2, 20f., 29; Beckmann (n.29) 56-8; G. Maurach Enchiridion Poeticum. Hilfsbuch zur lateinischen Dichtersprache (Darm-

stadt 1983) 17. 35.

Cf. Appel (n.30) 152; Keyssner (n.29) 133f.; Kleinknecht (n.30) Index B. Namen-

und Sachregister s.v. Gelübde. 36.

On the komos, cf. esp. F.O. Copley Exclusus amator: A Study in Latin Love Poetry (Philological Monographs published by the American Philological Association 17, 1956); F. Cairns Generic Composition in Greek and Roman Poetry (Edinburgh 1972) [henceforth GC] Index of Genres and Examples and General Index s.v. komos; P. Pinotti ‘Propert. IV 9: Alessandrinismo e arte allusiva' GIF n.s. 8 [29] (1977) 50-71. The link often made between komos and exclusion is not absolute: cf. F. Cairns ‘Two unidentified komoi of Propertius. I 3 and II 29’ Emerita 45

120

FRANCIS CAIRNS

(1977) 325-53. 37.

The links between this paean and the komastic tradition are discussed by F. Cairns ‘Propertius 4.9: “Hercules exclusus" and the dimensions of genre’, in K. Galinsky (ed.) The Interpretation of Roman Poetry: Empiricism or Hermeneutics? (Studien zur klassischen Philologie 67, Frankfurt am Main-Bern-New

Paris 1992) 65-95, 70-77.

38.

York-

Other links between AP 12.84 and 85 (Meleager) and Am. 1.3 — viz. ἄρτι (84.2;

85.3) and nuper (1); Venus as 'Cypris' (84.8) and Eros (85.9) and Ovid's prayers to *Cytherea' (1-4); and pleas of all the speakers to humans not gods — are matters

of ethos and style rather than genre. Cf. Pinotti (n.36) 64, 67f. n.69.

Cf. GC Ch.7. Cf. Tibullus Ch.7. For other such sophistications, cf. Tibullus Ch.7. Cf. GC Ch.7; Tibullus Ch.7. Cf. Tibullus 187. Cf. Copley (n.36) Ch.5. Cf. McKeown 11.61. Cf. above, n.5. F. Skutsch Aus Vergils Frühzeit (Leipzig 1901); Gallus und Vergil. Aus Vergils

Frühzeit 11 (Leipzig and Berlin 1906).

For an exception to the general disapprobation of F. Skutsch's views, cf. L. Alfonsi *L'Elegia di Gallo’ RFIC 21 (1943, publ. 1944) 46-56, 54-6. D.O. Ross Jr. Backgrounds (Cambridge 1975).

to Augustan

Poetry:

Gallus,

Elegy

and Rome

R.D. Anderson, P.J. Parsons, and R.G.M. Nisbet 'Elegiacs by Gallus from Qasr

Ibrim’ JRS 69 (1979) 125-55.

Esp. ΚΕ. Thomas ‘New Comedy, Callimachus, and Roman Poetry’ HSCPh. 83 (1979) 179-206, 203-5; J.K. King “The Two Galluses of Propertius’ Monobiblos' Philologus 124 (1980) 212-30; F. Cairns ‘Propertius 1,4 and 1,5 and the ‘Gallus’ of the Monobiblos' Papers of the Liverpool Latin Seminar Fourth Volume 1983 (Arca 11, Liverpool 1984), 61-103, 83-91. 53.

S. Hinds ‘Carmina Digna. Gallus P Qasr Ibrim 6-7 Metamorphosed' Papers of the Liverpool Latin Seminar Fourth Volume 1983 (Arca 11, Liverpool 1984) 43-54. He deals with Ov. Am. 1.3.19f. at 48f. The recent remark of E. Courtney (ed.) The Fragmentary Latin Poets (Oxford 1993) 267: “I do not think that any significance should be attached to the banal phrase dicere digna in other passages of Augustan poetry" misses the point. The key phrase is carmina/carmen digna/dignum.

54.

Hinds (n.53) 45 discusses the influence of Lucretius on the tradition, referringto R.O.A.M. Lyne (ed.) Ciris: A Poem Attributed to Vergil (CCTC 20, Cambridge

1978) on Ciris 5. Lyne gives high precedence to Lucr. 5. If.

55.

On the uncertainties over their dates of birth, cf.

J.-P. Boucher Caius Cornélius

IMITATION

AND

ORIGINALITY

IN OVID AMORES

1.3

121

Gallus (Bibliothéque de la Faculté des Lettres de Lyon 11, Paris 1966) 5f.; Kl.-P. s.v. Lucretius B.I (p.759).

56. 57.

Cf. Tibullus 5 n.23, 6 n.24.

Cat. 95.1f., referring to Cinna's Zmyrna; Hor. AP 388, with Brink ad loc. Other hellenistic programmatic concepts also imply it, esp. ‘sleeplessness’ (e.g. Callim.

Ep. 27.4 Pf. — AP 9.507.4; Cinna fr. 11.1 Büchner) and polemic against rapid autoschediastic composition (e.g. Cat. 95.3; Hor. Sat. 1.4.9f.).

Cf. Tibullus 5 n.20. For identification of the poet's interests with those of the addressee, cf. GC 222-5. Hinds (n.53) 45 notes that thesingular form is much less common than the plural. O. Skutsch The Annals of Quintus Ennius (Oxford 1985) 329. 1 observe a probable

coincidence, that Auson. Mos. 298, also imitating the Ennian formula, begins: quis potis innumeros ..., cf. Ov. Am. 1.3.9: nec meus innumeris ... .

Hinds (n.53), adding (45f.) to a predecessor's emphasis on dicere/ digna at Ecl. 9.35f., observes that neque adhuc at Ecl. 9.34 “looks forward to a tandem" (46).

Cf. Hinds (n.53) 47f. Cf. Davidson (n.7) 279f. KL-P. s.v. Europa.

KL-P. s.v. Io. On the Hypanis and for this suggestion, cf. P. Fedeli (ed.) Sesto Properzio: Il primo libro delle elegie. Introduzione, testo critico e commento (Accademia Toscana di Scienze e Lettere La Colombaria. Studi 53, Florence 1988) 290f. on

Prop. 1.12.3f.

Cf. Calvus frr. 9-14 Büchner; K/.—P. s.v. Io lists other authors who treated her story. McKeown

1.103-7 seems hesitant to accept the title ‘Amores’ for the works of

Gallus and Ovid. But the evidence he assembles is surely sufficient.

70.

Cf. King (n.52) 213f.; Cairns (n.52) 84-6, and the further bibliography there cited.

71.

Cf. e.g. 1.M.LeM. DuQuesnay ‘Vergil’s Fourth Eclogue' Papers of the Liverpool Latin Seminar 1976 25-99 (Arca 2, Liverpool 1977) 55, 99; Tibullus Ch.2; R.F. Thomas ‘Virgil's Georgics and the art of reference’ HSCPh. 90 (1986) 171-98;

McKeown I.37-45; W. Clausen Virgil's Aeneid and the Tradition of Hellenistic

Poetry (Sather Classical Lectures 51, Berkeley-Los Angeles-London 1987) 20f.,

61-9; F. Cairns Virgil’s Augustan Epic (Cambridge 1989) Chh. 8, 9.

72.

For aspects of Ovid mentioned here, cf. R. Syme The Augustan Aristocracy (Oxford 1986) Index of Persons s.v. OVID.

73.

Cornelius Fidus, cf. Sen. Dial. 2.17.1.

74.

Cf. Syme (n.72) Index of Persons s.v. PROPERTIUS.

75.

Cf. Syme (n.72) Index of Persons s.v. CORNELIUS GALLUS, C.; Cairns (n.52) 86-91.

122

FRANCIS CAIRNS

76.

Cf., for the villa, T.P. Wiseman Roman Studies: Literary and Historical (Collected

77.

For the theme in Tibullus, cf. Tibullus General Index s.v. poverty.

78.

Cf. esp. Cat.

79.

Esp. Cat. 23, with Macleod (n.78).

80.

The extent to which Am.

Classical Papers 1, Liverpool 1987) 307-70.

15, 16, 21, 23, with C.W. Macleod *Parody and Personalities in

Catullus" CQ 23 (1973) 294-303.

1.3 also contains obscene double-entendres is more

obscure. desultor (15) has been seen as having “delicately obscene implications" (McKeown II. ad /oc.); and what would Roman readers have thought about nuda

simplicitas purpureusque pudor (14) — cf. Barsby (n. 13) 53; or the strange virginea tenuit cornua vara manu (24)?

81].

A point made by Barsby (n.13) 55.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 123-34 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

OVID'S COSMOGONY METAMORPHOSES

1.5-88 AND THE TRADITIONS

OF ANCIENT POETRY MARTIN

HELZLE

“Die ovidische Kosmogonie ist eines der am meisten interpretierten Stücke antiker Dichtung." So wrote Franz Bómer in his commentary on the Metamorphoses.! Perhaps the most complete treatments are

those of Walter Spoerri and Franz Lámmli,? both of whom trace the transmission of individual motifs and phrases of the cosmogony through a number of ancient sources; and, at the end of the day, the best summing-up may well be, again, Bómer's: “4116 diese Kriterien weisen in den Bereich einer auf philosophischem, speziell stoischem

Wege ... gereinigten hellenistischen Form alter Vorstellungen, unter denen vor allem auch platonische ..., aristotelische ... und speziell auch, aber schon nicht mehr unbestritten, andere, vorsokratische

Motive fortlebten.'? But although such philosophically-based studies have contributed valuably to our understanding of Ovid's epic poem,‘ philosophy alone cannot provide the key to a work which does not purport to

expound philosophic theory, but rather uses it justifying its own central theme. The purpose of this investigate the poetic precedents and associations of to analyse some of the poetic techniques used in the

as a means of paper, then, is to Ovid's narrative, cosmogony, and

finally to offer an overall assessment of the implications of this first

metamorphosis at this initial point in the epic. The closest parallel to the Ovidian cosmogony in matter, manner

and time is found in the much discussed song of Silenus in Vergil's sixth Eclogue (31-3):5 123

124

MARTIN HELZLE

Namque canebat uti magnum per inane coacta semina terrarumque animaeque marisque fuissent et liquidi simul ignis;

The resemblance goes beyond the similarity between Vergil's coacta semina (31f.) and Ovid's congesta ... semina (8f.), or the comparable references to the four elements (Eclogue 6.32f.) and the three regions (Metamorphoses 1.5). Vergil's Silenus is also emphatically compared to Orpheus, just as Ovid had linked himself with Orpheus' song in Apollonius’ Argonautica. Vergil goes on to make Silenus describe the separation of land and sea by means of the metonymy Nereus (Eclogue 6.35), which might be analogous to Ovid's describing the as yet non-existent sea as Amphitrite (14). A possible verbal echo may also be found in Ovid's nulli sua forma manebat (17),' which could go back to Vergil's coeperit et rerum paulatim sumere formas (Eclogue 6.36). Ovid seems to use these recollections of his predecessor to evoke the entire song of Silenus in a way perceptively described by Gian Biagio Conte: "Meglio parlare di situazione poetica che non di versi singoli: basta molte volte una sola parola a condensare un'intera situazione poetica e ad evocare la Stimmung."? In his cosmogony therefore Ovid emulates Vergil's creation through a number of fleeting allusions to Eclogue 6.

Furthermore, the subject matter of both versions is surprisingly similar. Both poets include the formation of clouds (Eclogue 6.38, Metamorphoses 1.54) and trees (Eclogue 6.39, Metamorphoses 1.44). These are standard elements of cosmogonies; Ovid, however, takes

more than just τόποι from Silenus' song when fleshing it out in his creation. Vergil had reported Silenus as singing of Pyrrha (cf. Metamorphoses 1.313-415), the Golden Age of Saturn (cf. Metamorphoses 1.89-150), Prometheus (Metamorphoses 1.78-86), Hercules and Hylas, Pasiphae and the bull (Metamorphoses 8.152-8), the Proetides (Metamorphoses 15.326), Atalanta (Metamorphoses 10.560—

707), the Heliades (Metamorphoses 2.340-66), Gallus’ *Dichterweihe',

Scylla (Metamorphoses 8.145-51), Odysseus (Metamorphoses 14.223308), Tereus (Metamorphoses 6.424-674), and possibly the story of Apollo and Hyacinthus (Metamorphoses 10.162-219). Ovid, then, seems to have taken Vergil's summary of Silenus' song and, with the exception of Hercules and Gallus,? used it as the basis for his own collection of mythical tales in fifteen books. What had been a 31 verse summary in the Eclogue has become a carmen perpetuum in the hands of Ovid.'? The whole Metamorphoses thus become Ovid's version of Silenus' song.

OVID'S COSMOGONY

125

Before returning to Ovid, a closer look at Eclogue 6 is now necessary. Since Vergil's poem starts with and develops elements of

the recusatio,

philosophical, dialectical, and even existentialist

readings seem far-fetched.!? The key to the poem, in my opinion, lies in a combination of Wimmel's findings about the recusatio and Otto

Skutsch's and Zeph Stuart's interpretation of it as a catalogue of ‘Alexandrian subjects. Wimmel showed!* that most recusationes contain a rejection of grand subjects, Apollo's warning to the poet, the use of a symbol for the poetic process, an invocation of predecessors, the expression of the poet's ‘Dichtungsideal’, and the justification of the poet's poetics by his biographical background. The first four of these elements are not hard to find in Eclogue 6. Vergil refuses to sing of reges et proelia (3); he is warned against this by Apollo (3-5); he invokes Theocritus (1); and Apollo contrasts the grand and the slender poem by means of the symbol of the fat sheep as opposed to the lean, ‘spun out’ poem (4f.). There may bea

trace of

biographical justification when Vergil points out parenthetically that he does not need to sing about Varus' exploits since there are others who wish to do so (6f.). The most important feature of the Eclogue, however, is the catalogue of ‘Alexandrian subjects’ which covers the *Dichtungsideal' at length. It is this aspect of Eclogue 6 that Ovid capitalises upon when launching into his cosmogony. By alluding to Vergil's programmatic Eclogue he associates himself with his predecessor's ‘Dichtungsideal'. That is not to say that Ovid claims the pastoral aspect as his own, but rather the Metamorphoses is cast as a full-blown treatment

of the ‘Alexandrian subjects’ given by Silenus and reported by Vergil.'? Seen in this light, Ovid's cosmogony is more than an eclectic account of the first metamorphosis in the history of the universe: it is also a statement about Ovid's poetics. Unlike his Augustan predecessors, he does not use the recusatio to make his point, but he uses a

mythical tale to associate himself with earlier Roman followers of

Callimachus.'® There are only few previous surviving poetic treatments of the subject. Lucretius, for instance, had laid out the Epicurean view at De Rerum Natura 5.416ff. Much earlier Hesiod had summed up the cosmogony in his Theogony (116ff.). He offered this summary at the beginning of his narrative proper, and just after his encounter with the Muses which had led him to tell of their ancestry and to invoke them at some length. While Lucretius provides Ovid with some of his vocabulary, Hesiod's pattern of starting with the creation of the

126

MARTIN HELZLE

world

is, of course, paralleled in the structure of Ovid’s carmen

perpetuum. After a relatively brief proem'® for a hexameter poem of 11995 lines, the Augustan poet also starts at the beginning of the cosmos. This way of proceeding seems perfectly logical; but it ought

to raise more than one eye-brow in a poem which proclaims a certain amount of Callimachean poetic theory, and which is constructed along lines closely resembling the Alexandrian poet's Aetia. For Hesiod had been invoked by Callimachus too as his poetic model, not only directly in Epigram 27 (Pfeiffer), but also indirectly in the dream-scene which is placed after the Aetia prologue on poetics. Hence, by opening his narrative with his own version of the

cosmogony, Ovid not only follows the structural pattern of Hesiod's Theogony, but also that of his much closer poetic ancestor Callimachus. The mere fact that the Augustan chooses to echo Hesiod at

this point in the poem places him therefore in the tradition of the episodic Callimachus and his model Hesiod. Although the narrative

per se contains no statement about poetics, the method employed might be called ‘demonstration of polemical attitude."?? Hesiod, however, deserves special attention in this argument since

Ovid's relationship with Hesiod is particularly close. Vergil's Silenus referred to Gallus' initiation-scene (Eclogue 6.64-73) and linked Gallus with Hesiod in that scene (hos tibi dant calamos ... Musae,/ Ascraeo quos ante seni, Eclogue 6.69f.). Propertius had also considered

it necessary to refer to Hesiod on a number of occasions.?! Unlike the obviously Hesiodic Georgics, the Eclogues and Propertius' elegies are not directly indebted to the Works and Days or to the Theogony. These references therefore stress the importance attributed to Hesiod

as Callimachus' model in the Aetia.?? However, when Ovid starts his narrative with an item which recalls the beginning of Hesiod's narrative in the Theogony, he is doing more than following the practice of his Augustan forebears. He is establishing a valid link between the subject matter of Hesiodic epic and his own. It could therefore be argued that in this passage Ovid achieves three things at the same time: he aligns himself with the Roman followers of

Callimachus, with Callimachus' invocation of Hesiod at the beginning of the Aetia and finally also with Hesiod's epic, which was viewed as an alternative to the Homeric branch of the genre. At this point a broader consideration of Augustan poetry is required. It has long been recognized that any Augustan poet was

faced with enormous obstacles when considering the composition of an epic poem.?? Vergil had risen to the occasion and had written a

OVID’S COSMOGONY

127

national epic in the tradition of Homer. His instant success only intensified the problem of writing epic for anyone after him. A poet with epic aspirations now had to contend not only with Homer, but

also with Vergil. The poets of the Tiberian age responded to the challenge in two different ways. Either they tried to follow the mythical epic tradition like Rabirius, Domitius Marsus, Albinovanus Pedo, Carus, Largus, Camerinus?* and others, or they resorted to historical epic rather than compete with Vergil’s monumental achievement. One of the latter epics was the Res Romanae of Cornelius Severus.?? Most of these poets are known to us only from Ovid Epistula ex Ponto 4.16. Their success cannot have been more than moderate, since their works do not survive and their names have

come down to us only in a list of ‘nobodies’ who provide a foil for Ovid's fame. Moreover, the nature of the next extant epic speaks volumes. Lucan's unorthodox historical epic was the first subsequent work in the genre to be hailed as great by its near contemporaries — and this in spite of personal hostility between the author and the princeps.?$ To conclude, then, an epic poet in the wake of Vergil had to explore avenues other than the Homeric epic; and this is exactly what Ovid is doing in the Metamorphoses. He avoids the challenge of Homer and Vergil by writing in the Hesiodic tradition, just as Lucan would evade the issue of both the Vergilian and the Ovidian

precedents by moving into historical epic.?" The fact that Ovid's cosmogony can be read as an oblique statement of its author's poetic stand-point is matched by a number of detailed observations in the body of the narrative. The first two pieces of evidence come from lines 11 and 14. Although Ovid excludes the mythical trappings of poetic cosmogonies, he does use Phoebe and Amphitrite respectively in these two lines. As commentators have pointed out, both are used here as the metonymies of the

poeta doctus.?® The former example is not out of the ordinary,? but Amphitrite deserves closer scrutiny. Bömer ad loc. comments on the spondeiazon formed, as often,?? by means of a Greek word. While spondaic hexameter clausulae already point towards neoteric verse

technique,?! this particular verse-ending has specific associations. Forms of the name Amphitrite are found as spondeiazontes four times in Homer's Odyssey," once in the Homeric Hymns," twice in Hesiod,^ twice in Apollonius Rhodius, once in Theocritus, frequently in Oppian's notoriously Callimachising Halieutica,! and,

in the wake of Oppian, in Quintus Smyrnaeus' Posthomerica.** The statistics for Greek poetry thus suggest Callimachean associations

128

MARTIN

HELZLE

for this feature even though no examples are found in Callimachus’

extant fragments.?? This picture is confirmed by the usage of the Roman poets: only Catullus and Ovid offer examples. Catullus' instance (64.11) is especially revealing: it is the second spondeiazon of his epyllion. The use of two spondaic verse-endings within such a short space is indicative of his Callimachean poetic predilections and programme. Ovid's second example of Amphitrite (Fasti 5.729) points in the same direction: it occurs in an elaborate periphrastic description of a day's end: hanc ubi dives aquis acceperit Amphitrite,/ grata Iovi fulvae rostra videbis avis (When Amphitrite rich in water has received this [day], you will see the beak of the tawny bird which is dear to Jupiter); and it Occurs in a narrative elegy whose theme (Tempora cum causis, Fasti 1.1) consciously alludes to Callimachus' Aetia. The appearance of Amphitrite as the first spondeiazon in the Metamorphoses therefore

introduces particularly rich Callimachean associations, even though there is no extant example of it in Callimachus. One of the best-known features of Callimachus' poetry was his learning,” the characteristic which seems to have left the most visible

mark on the Roman poets influenced by him.*! A few minor learned details in Ovid's cosmogony are the metonymies Titan (10) and Phoebe (11) and the recondite reference to the Nabathaei.? In addition, it has recently been shown

that, in at least some of the

Augustan poets, including Ovid, doctrina manifests itself also inter

alia in etymological word-play and in double allusions.* The latter mechanism seems cosmogony. There, establish a link with song in Apollonius

to be employed right at the outset of Ovid's as noted, some verbal and many thematic echoes Eclogue 6. But Ovid's words also recall Orpheus’ Rhodius:

Ante mare et terras et, quod tegit omnia, caelum unus erat toto naturae vultus in orbe,

quem dixere Chaos, rudis indigestaque moles nec quicquam nisi pondus iners congestaque eodem non bene iunctarum discordia semina rerum. (Metamorphoses

1.5-9)

Ἤειδεν δ᾽ ὡς γαῖα kai οὐρανὸς ἠδὲ θάλασσα, τὸ πρὶν En’ ἀλλήλοισι μιῇ συναρηρότα μορφῇ, νείκεος ἐξ ὀλοοῖο διέκριθεν ἀμφὶς Exacta’ (Apollonius Rhodius 1.496-8)

The echo of the Apollonian model is highly appropriate, since in the Argonautica Orpheus enthralls his audience totally with his

OVID'S COSMOGONY

129

song (1.512-15). Ovid's reminiscence of Orpheus' song therefore stakes a claim to poetic perfection. However, Ovid's verbal debt to Lucretius in this section^^ also gives his version a ‘scientific’ appearance, although he departs from the Epicurean atomist Lucretius in his understanding of elementa (29) as στοιχεῖα rather than atoms.* The Lucretian tone, then, lays claims to a degree of technical expertise. Ovid is therefore alluding to the ‘Alexandrian’ song of Silenus at the same time as adapting a passage sung by Orpheus, the greatest singer of all times, in Apollonius' epic; and he

does not fail to include some Lucretian terminology to give his narrative the appearance of scientific expertise. As observed, another way of displaying doctrina was etymological

word-play. One ancient derivation relevant to the present context is the Stoic etymology of χάος from the Greek verb yéw.** This etymology seems to be implied by Ovid when he talks of the creator as one who evolvit caecoque exemit acervo (24), where acervus is the Latin equivalent of χῶμα. Similarly Amphitrite's name was some-

times derived from ἀμφί, and τρίτος“7 or tpéw* or tpitw* or ῥέω. Ὁ The last of these seems to be Ovid's choice in verse 30 where the epitheton ornans circumfluus translates ἀμφί and ῥέω. The poet may also be playing on this etymology in 13f.: nec bracchia longo/ margine

terrarum porrexerat Amphitrite, and in 37, where he describes the earth as ambitae ... terrae. In the latter passage ambitae even sounds

like Amphitrite.?! The passage most fertile in etymologies, however, is the account of the creation of the winds (61-6). Eurus ad Auroram ... recessit (61) clearly implies a connection between Eurus and "Hac, which is also made by Isidore Origines 13.11.4: Eurus eo quod ab ἠῶ

fluat id est ab oriente. Ovid repeats this etymology in 62.?? In the whole passage only Boreas escapes Ovid's verbal playfulness. Zephyrus is derived from ζόφος (darkness) in 63: Vesper, et occiduo

quae litora sole tepescunt/ proxima sunt Zephyro.?? Auster rounds off the etymologising; it is, as usual, connected with rain: Auster ab auriendo aquas (Isidore Origines 13.11.6). It is not surprising, then, that Ovid suggests rain three times in a single line: nubibus adsiduis pluvioque madescit (64).

To sum up: by beginning his narrative with a cosmogony Ovid makes a statement about the poetic tradition within which he sees himself as working. The parallel between Silenus' song in Vergil's sixth Eclogue and the Metamorphoses as a whole indicates that Ovid wants to associate himself with the Roman followers of Callimachus. The Hesiodic chord struck by the same passage only re-enforces this

130

MARTIN HELZLE

association, since Callimachus had also defined his art in terms of Hesiod, This is borne out by the technique of multiple allusion and by some etymologising, both of which have been recognised as typical ways in which the Roman poets influenced by Callimachus show their learning. Ovid’s cosmogony thus becomes an oblique secondary proem which follows after the bare bones of the first four lines. But, rather than ‘apologising’ for his poetry, or attacking literary opponents or a demanding patron, Ovid merely makes a statement of fact about the place of the Metamorphosesin the literary

tradition. David Ross? puts the influence of Callimachus on Roman poetry in a nutshell: “What the neoterics discovered was the poet's place in poetry. There are two aspects of this discovery: the individuality of the poet in his own poems, and his placein the poetry of the past." The latter aspect takes center-stage in Ovid's creation narrative.

NOTES 1. 2.

Bömer on Met. 1.5-88 (“the Ovidian cosmogony is one of the most intensively studied pieces of ancient poetry"). W. Spoerri Späthellenistische Berichte über Welt, Kultur und Götter (Schweizerische Beiträge zur Altertumswissenschaft 9, Basel 1959); F. Lammli Vom Chaos

zum Kosmos. Zur Geschichte einer Idee (Schweizerische Beiträge zur Altertumswissenschaft 10, 2 vols, Basel 1962). 3.

Bómer on Met. 1.55-88 (“all these criteria point to a philosophically and specifically Stoic based and harmonised version of older ideas, a version in which above all Platonic ..., Aristotelian ..., and in particular, although certainly no

longer undisputed pre-Socratic motifs survived").

4.

Apart from metre and length, the description of the Metamorphoses as ‘epic’ relies e.g. on R. Heinze Ovids elegische Erzählung (Berichte über die Verhandlungen der sächsichen Akademie der Wissenschaften zu Leipzig. phil.-hist. KI. 71.7,

Leipzig 1919); S.E. Hinds The Metamorphosis of Persephone. Ovid and the Self-

conscious Muse (Cambridge 1987) and P. Knox Ovid's Metamorphoses and the Traditions of Augustan Poetry (Cambr. Philol. Soc. Supplementary Vol. 11, Cambridge 1986). The whole question is dealt with at length by E. Bernbeck

Beobachtungen zur Darstellungsart München 1967) 123-38. 5.

in Ovids

Metamorphosen

(Zetemata

43,

E.deSaint-Denis ‘Le chant de Silene a la lumitre d'une découverte récente’ RPA.

37 (1963) 27-9; C. Segal ‘Vergil’s sixth Eclogue and the problem of evil’

TAPA

100 (1969) 407-35; E.W. Leach ‘The unity of Eclogue 6° Latomus 27 (1968) 13-32 (hereafter ‘Unity’); Vergil’s Eclogues. Landscapes of Experience (Ithaca and

London

1974) 241 (hereafter Landscapes), H. Bauza ‘La natura del canto di

Sileno’ Sileno 13 (1987) 21-31.

6.

nectantum Rhodope miratur et Ismarus Orphea (Ecl. 6.30), cf. Ov. Met. 1.5f.; Ap. Rhod 106-8; on the significance of Orpheus both in Vergil and Ovid, cf. Knox (n.4) 48-64.

OVID’S COSMOGONY

131

It seems tempting, if somewhat frivolous, to detect in the parechema forMA MAnebat a reference to a part of female anatomy which can also undergo visible changes; on parechema in general, cf. Servius on Verg. Aen. 2.27; Bómer on Met. PAL G. Luck ‘Notes on the language and text of Ovid's Tristia! HSCP 65(1961)

G.B. Conte Memoria dei poeti e sistema letterario. Catullo, Lucano (Torino

Virgilio, Ovidio,

1974) 10f. with n.15.

R. Coleman ‘Gallus, the Bucolics, and the ending of the fourth Georgic’ AJP 83 (1962) 57 n.7; Knox (n.4) 48.

10.

On Ovid's penchant for large-scale poetry, cf. E. Paratore *L'evoluzione della "sphragis" dalle prime alle ultime opere di Ovidio', in Asti del convegno internazionale Sulmoniano (Roma

11. 12.

1959) 1.201.

W. Wimmel Kallimachos in Rom. Die Nachfolge seines apologetischen Dichtens in

der Augusteerzeit (Hermes Einzelschriften 16, Wiesbaden 1960) 131-47.

Segal (n.5) 407-35; Leach ‘Unity’ 13-31; Landscapes 241; W. Spoerri ‘L’épicurisme et la cosmogonie de Siléne’, in Actes du VIJe Congrés G. Bude 1969 (Paris

1969) 447-56; ‘Zur Kosmogonie in Vergils 6. Ekloge' MH 27 (1970), 144-63; *Antike Vergilerklärer und die Silenuskosmologie’ MH

27 (1970) 265-72; G.

Lieberg, *L'armonia delle sfere in Virgilio? Osservazioni sull'epilogo della sesta ecloga’, in Atti del Convegno Virgiliano Bimillenario delle Georgiche (Napoli 1977) 405-20; Bauza (n.5) passim.

13.

Wimmel (n.11) passim; O. Skutsch ‘Zu Vergils Eklogen’ RAM N.F. 99 (1956) 194f.; Z. Stewart ‘The song of Silenus' HSCP 64 (1959) 179-205; cf. also J.P. Elder 'non iniussa cano: Virgil'ssixth Eclogue' HSCP 65(1961) 109-25; D.O. Ross Jr. Backgrounds to Augustan Poetry. Gallus, Elegy and Rome (Cambridge 1975) 18-38; W. Clausen 'Cynthius' AJPh. 97 (1976) 245-7. The fact that Silenus is depicted as ‘drunk’ has sometimes been used as an argument to deny him any

connection with the water-drinking followers of Callimachus, e.g. by de Saint-

Denis (n.5) 27-9; Leach *Unity’ 26 n.1. However inflatum hesterno venas, ut semper, laccho (Ecl. 6.15) means ‘hung over’ rather than ‘drunk’. Furthermore,

the follower of Bacchus is tied up, i.e. symbolically restrained, for this scene. It may also be indicative that a Dionysiac creature like Silenus gives a summary of

*sober' Alexandrian subjects and ends with the word Apollo.

14.

Op. cit. (n.11) 119.

15.

Cf. H. Hofmann 'Ovid's Metamorphoses: carmen perpetuum, carmen deductum'

16.

Unlike Propertius, for instance, Ovid never directly invokes Callimachus. The terms ‘Alexandrianism’ and *Callimacheanism' will be used loosely here (pace

PLLS 5 (1986) 227.

W. Clausen ‘Callimachus and Latin poetry’ GRBS 5 (1964) 181-96) to refer to the way in which the Roman poets used and adapted the poetics of Callimachus and the poets who shared his theoretical stance; cf. N.B. Crowther ‘oi νεώτεροι, Poetae Novi, and Cantores Euphorionis’ CQ n.s. 20 (1970) 322-4; Wimmel (n.11) passim; Ross (n.13) 3-8, 163f.; F. Cairns Tibullus. A Hellenistic Poet at Rome

(Cambridge 1979) 6-35; P. Fedeli Properzio. Il Libro Terzo delle Elegie (Bari 1985) 28, 40-2; J.C. McKeown Ovid: Amores. Text, Prolegomena and Commentary Vol. I (Arca 20, Liverpool

17.

1987) 32-62.

E.g. semina rerum (9), cf. Lucr. 1.59, 176 and often; Mer. 1.5, cf. Lucr. 5.434; discordia semina rerum (9), cf. Lucr. 5.440 de principiis, discordia quorum; mutatas ... formas (1), cf. Lucr. 5.443 dissimilis formas; iunctarum ... rerum (9), cf. Lucr.

5.444 non ... poterant coniuncta manere; cf. also Bömer on Met. 1.9. Further on in the narrative Lucretian influence seems to fade fast (cf. Bómer passim).

132

MARTIN HELZLE

18.

U. Fleischer 'Zur Zweitausendjahrfeier Ovids' A&A 6 (1957) 32; E.J. Kenney ‘Ovidius prooemians’ PCPAS n.s. 22 (1976) 46.

19.

Kenney (n.18) 46-53, against H. Herter ‘Ovids Kunstprinzip in den Metamorphosen' AJP 69 (1948) 129-48 = Ovid. Wege der Forschung ed. M. von Albrecht (Darmstadt

1968) 340-61

Hofmann (n.15) 223-41.

20.

R. Thomas

(1982) 144.

(references will be to the latter edition);

‘Catullus and the polemics of reference (Poem 64.1-18)’ AJP 103

21.

2.10.25f.; 2.13.4; 3.3.

22.

Wimmel (n.11) 56-8, 238-40.

33.

B. Otis Virgil. A Study in Civilized Poetry (Oxford 1963) 5-40; Ovid as an Epic Poet (2nd ed., Cambridge 1970) 1-3; S. Dópp Virgilischer Einfluß im Werk Ovids (München 1968) 104f.

24.

On these authors and their works see M. Helzle P. Ouidii Nasonis Epistularum ex Ponto liber IV: A Commentary on Poems 1-7 and 16 (Diss. Cambridge 1988, Hildesheim 1989) 182-97.

25.

Helzle (n.24) 60f.

26.

Statius Silv. 2.7; even the criticism of Eumolpus at Petron. 118-24 demonstrates this. For the quarrel, cf. Vacca Vita Lucani in Adnotationes super Lucanum ed. J.

Endt (Stuttgart 1909) 2. Unlike Gallus, Lucan seems not to have been affected by damnatio memoriae.

27.

He could rely on yet another branch of epic established in Hellenistic times: cf. K. Ziegler Das hellenistische Epos. Ein vergessenes Kapitel griechischer Dichtung

(Leipzig and Berlin 1934), and also on the historical epics by Ennius, Naevius and Pacuvius. These all had the advantage of being much older than Ovid and Vergil. Historical epic therefore lent itself to a new approach. Wimmel op. cit. (n.11) 239f. attributes asimilar role to Hesiod in Callimachus' own poetic development.

28.

Haupt-Ehwald-von Albrecht on Met. 1.10-14; Bómer on Met. 1.14. On metonymy in general, cf. W. Kroll on Catull. 64.11; Studien zum Verständnis der rómischen Literatur (Stuttgart 1924) 365 n.42; G. Maurach Enchiridion Poeticum

(Darmstadt 1983) 72f. 29. 30.

Verg. Georg. 1.431; Aen. 3.371; Ov. Am. 3.2.51 et al. A good collection of foci may be found in J. Soubiran ‘Hexamétres spondaiques à quadrisyllabe final’ GIF 21 (1969) 329-49.

31.

E. Norden P. Vergilius Maro: Aeneis Buch VI (7th ed., Darmstadt

1981) 441-6;

Fordyce on Catull. 64.11; Fedeli on Prop. 1.13.31; W.H. Mineur Callimachus. Hymn to Delos (Leiden

1984) 37.

32.

Od. 3.91; 5.422; 12.60, 97.

33.

Hymn. Hom.

34.

Theog. 243, 254.

35.

4.1325, 1355.

36.

23.55.

3.94.

OVID’S COSMOGONY

133

37.

1.2, 84, 343, 387, 423, 600, 619, 791; 3.459; 4.677, 5.113, 222, 333.

38.

7.374; 8.63; 14.535, 609, 644.

39,

For the positive attitude of Callimachus and his followers towards Homer, cf. Theocr. 7.45-8; H. Herter ‘Kallimachos’ RE Suppl. 5.447.48-448. 34; Williams on Call H; mn 2.105-13; P. Bing The Well-Read Muse (Hypomnemata 90, Göttingen 1988) 55f. n.11. Most recently illustrated again by A. Hollis Callimachus. Hecale (Oxford 1990) 11-13.

al.

Cf. Cairns (n.16) 36-110; McKeown (n. 16) 32-62; see also Kroll's remark (n.28) 37: "Das charakteristische Beiwort des alexandrinischen Dichters war doctus";

and further Fordyce on Catull. 35.18f.; E. Kenney 'Doctus Lucretius’ Mnemosyne 4th series 23 (1970) 366-9. 42.

Cf. Bómer on Met. 1.61; Strabo 16.760, 767, 777 and often; Diod. Sic. 2.48; 3.43; 14.31, 48, 103; Plin. Nar. Hist 5.65; 6.144, 157; 12.73; Tac. Ann. 2.57 with Koestermann ad loc.

43.

For double allusions and etymologising, cf. McKeown (n.16) 32-62; for etymologising, Cairns (n.16) 90-99; D. Porte L’étiologie religieuse dans les Fastes d'Ovide (Paris

1985)

197-264; Thomas

(n.20)

148-54; J. Farrell ‘Dialogue of

genres in Ovid's "Lovesong of Polyphemus" (Mer. 13.719—-897)' AJP 113 (1992) 235-68.

See above, n. 17.

45.

Bómer

ad loc. compares

15.237-40 (Pythagoras talking): Haec quoque non

perstant, quae nos elementa vocamus;/ quasque vices peragant, - animos adhibete! - docebo./ quattuor aeternus genitalia corpora mundus/ continet and Cic. Acad. 1.26: Itaque aer ... et ignis et aqua et terra prima sunt; ex his autem ortae animantium formae earumque rerum quae gignuntur e terra. Ergo illa initia et (ut e

Graeco vertam) elementa dicuntur. F. Robinson ‘The creation story in Ovid’ CPh.

8 (1913) 404 and P. De Lacy ‘Philosophical doctrine and poetic technique in Ovid’ CJ 43 (1947/48) 156 argued in favour of Bómer's position. However OLD s.v. elementum 2 and TLL 5.2.343.14 take the Ovidian passage as referring to

atoms in the Democritean and Epicurean sense. Two arguments speak in favour

of Bómer: first, it seems incongruous that the poet should mention ignea... vis,

aer, tellus and umor within four lines (26-30) while at the same time talking of elementa as atoms; second, Spoerri (n.2) 10-25 and 34-7 demonstrated that Ovid is drawing here on Diodorus rather than on Lucretius. Therefore Ovid's elementa are much more likely to reproduce the στοιχεῖα of his source. Zeno SVF 1.29 (von Arnim); Plut. Mor. 955e; Bómer on Mer.

1.24.

Plut. Is. Osir. 75; Schol. Od. 3.91; RE 1.1963.47ff. Didymus iunior 338 (Schmidt). W.

Roscher Ausführliches Lexicon der griechischen und rómischen Mythologie

(Leipzig 1884-1927) 1.318. 15ff. Schol. Opp. Hal. 1.2.

51.

The word-order ambitae circumdare litora terrae seems to re-enforce the sense because ambitae and terrae enclose the infinitive and its direct object. This is all

the more likely here since Ovid is ostentatiously playing with the interaction between word-order and sense two lines later in the versus aureus. On enclosing word-order in general, cf. T.E.W. Pearce ‘The enclosing word-order in the Latin

134

MARTIN HELZLE

hexameter' (Ὁ n.s. 40 (1966) 140-71, 298-320. 52.

et radiis iuga subdita matutinis (62). “Theme and variation' is an Ovidian technique repeatedly stressed by E.J. Kenney 'materie superatur opus’ CR n.s. 22 (1972) 39 n.1; ‘The style of the Metamorphoses’ , in Ovid ed. J.

and Boston 1973) 132 with n.106. 53.

Cf. Doxogr. Gr. 374.23ff. (Diels). Bómer ad loc. Op. cit. (n.13) 4.

W. Binns (London

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 135-40 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

LUCAN’S PRAISE OF NERO VINCENT HUNINK

In his study of ancient invective Severin Koster hailed Lucan as the first poet to introduce a 'subjective' form of invective into epic, thus

paving the way for the celebrated epic invectives of Claudian.! And certainly parts of Lucan's Bellum Civile (BC), which treats the mid

first century BC civil war between Pompey and Julius Caesar, direct large quantities of abuse, insult and scorn at the character of Caesar. Lucan uses every available rhetorical device to paint his Caesar in the darkest possible colours, giving him all the traits of a devil incarnate: he is bloodthirsty, tyrannical, cruel, arrogant and entirely devoid of moral scruples. He uses men, money and nature to pursue his aim, which is, not the acquisition of power so as to implement new policy, but the total destruction of all religious, social and political order.’ Since antiquity many students and readers of Lucan have suspected that there is more to these passages than meets the eye: Lucan’s invective and insults have seemed to them too strong to be aimed merely at the long-dead Julius Caesar. In their view, the real targets are the principate in general and the Emperor Nero in particular. Certain pieces of historical evidence would appear to support this view.? Initially Lucan held a privileged position in Nero's cohors amicorum. But on both sides growing jealousy and pride accompanied the growth of literary ambition. One anecdote tells of Lucan’s anger when the emperor walked out of his recitatio. As a result Lucan supposedly criticised Nero and his most powerful friends in an ‘insulting poem’ (famosum carmen). Lucan eventually took part in the Pisonian conspiracy of AD 65 against Nero. Its discovery led to the execution or enforced suicide of all participants, including Lucan. So the events of Lucan’s life seem to argue for an interpretation 135

136

VINCENT HUNINK

of BC as a manifesto of political opposition to the principate. But then a curious problem arises. The only explicit mention of Nero in BC is in the proem of Book 1, where the work is dedicated to him as patron of arts and culture* in a fervent, even extravagant,

panegyric. After picturing the devastating results of the civil war on the Italian landscape, Lucan proceeds (1.44-66): multum Roma tamen debet civilibus armis, quod tibi res acta est. te, cum statione peracta astra petes serus, praelati regia caeli excipiet gaudente polo; seu sceptra tenere, seu te flammigeros Phoebi conscendere currus, telluremque nihil mutato sole timentem igne vago lustrare iuvet, tibi numine ab omni cedetur, iurisque tui natura relinquet, quis deus esse velis, ubi regnum ponere mundi. sed neque in arctoo sedem tibi legeris orbe, nec polus aversi calidus qua vergitur austri, unde tuam videas obliquo sidere Romam. aetheris inmensi partem si presseris unam, sentiet axis onus. librati pondera caeli orbe tene medio; pars aetheris illa sereni tota vacet, nullaeque obstent a Caesare nubes. tum genus humanum positis sibi consulat armis, inque vicem gens omnis amet; pax missa per orbem ferrea belligeri conpescat limina Iani. sed mihi iam numen; nec, si te pectore vates accipio, Cirrhaea velim secreta moventem sollicitare deum Bacchumque avertere Nysa: tu satis ad vires Romana in carmina dandas.

45

50

55

60

65

Itis a strange paradox that the alleged target of BC is here addressed in the most enthusiastic terms. We know that in his early years Lucan did write a poem in praise of Nero.* But the laudatory proem of BC seems hard to reconcile with the apparent anti-Neronian nature of the rest of the work. Hitherto,

in essence, two

solutions

to this problem

have been

advanced. The first assumes that the praise of Nero in the proem is

simply insincere, and that it contains ironic hidden meanings and veiled insults. Thus Nero poised as a star in orbit and threatening to disturb the balance of the sky with his weight (53-7) might be a hint at

his corpulence; the ‘cloudless, clear sky’ (58f.) could be a glance at his baldness; his ‘looking askance’ (55) might allude to his squint; and, in

general, the dedication of BC to a man whose own poetry gave rise to nothing but laughter and scorn would have seemed patently absurd and insincere. This ironic interpretation, which goes back to

LUCAN’S PRAISE OF NERO

137

antiquity, has enjoyed considerable popularity. It allows BC to preserve its basic unity of function, since even in his apparent praise of Nero Lucan would remain fiercely anti-monarchical and antiNeronian. Today, however, a majority of scholars subscribe to a second, non-

ironic view of the proem. Their position is that, although Lucan's praise of Nero may seem extravagant and implausible to us, it is in fact firmly rooted in the traditions of Roman imperial panegyric. If we read Seneca's Apocolocyntosis or Calpurnius' Eclogae, the same flattering and adulatory tone can be perceived.’ Again, the identification of Nero with Apollo and Bacchus is in line with Neronian ideology; and even Nero's personal interest in astronomy is well attested.? Finally there are no linguistic indications in the actual text of the proem which give us reason to doubt its sincerity or to assume

any ironic intention.'? But how, then, are we toexplain the apparent contrast between the proem and the rest of BC? An ingenious solution has been proposed: the contrast must simply be accepted asa fact, and BC seen as divided into two distinct parts. The first part is the work of a Lucan still on speaking terms with the emperor, and so is not anti-Neronian. The second part, composed in a later phase, reflects Lucan's changed feelings for Nero, i.e. hatred and enmity towards Nero, disguised as ferocious invective against Julius Caesar, Nero's ancestor,!! and hence antipathy in general towards imperial rule. This solution is founded on the brief biography of Lucan by Vacca: it states that Lucan published three books of BC (often identified as Books 1-3) before a ban was imposed by Nero on its further publication; the remaining books were published after Lucan's death.'? Although I agree fully with most modern scholars that the ironic view of the proem must be rejected, the notion of dividing BC into two chronologically and ideologically distinct parts is very unsatisfactory, and fails to do justice to Lucan's work. First of all, there is no

clear-cut division between Books 1-3 and Books 4-10 as far as hostility to Caesar is concerned. Caesar 5 blamed and censured from beginning to end. The greater sharpness of tone in the later books simply reflects the development of the theme. Thus Book 7, with all its criticism of Caesar, its account of loss of freedom and its censure of the gods, is also the book where the main battle is fought. It is only natural for the poet to intensify pathos and invective as the epic's climax approaches. Furthermore, even though BC exhibits many inconsistencies, such a sharp ideological shift would, in my view,

138

VINCENT HUNINK

effectively destroy it by any standards as a work ofart. Perhaps most importantly, the bipartite explanation, like the ironic approach,

emphasises extra-literary elements rather than actual themes of the epic. If examined critically, the very concept of anti-Neronianism in Lucan is problematic. Nero is mentioned by name only in the proem, where, as I have shown, he is eulogised. The rest of the epic contains

invective and insults against Julius Caesar, but they seem to be directed at no-one except Caesar. For the Romans, the concept of ‘anti-Caesarism’, i.e. personal hatred of the successors of Caesar and Augustus, did not exist; and modern historians tend to deny rather

than assert continuity between the positions of Julius Caesar and Nero." In addition, to regard Nero as the real target behind Caesar is to assume that the ancient Vitae are trustworthy and also to subscribe to the biographical fallacy, which equates an author's life and work. Contrariwise, if BC had any political significance at all, it was not a

pamphlet about the contemporary state of affairs. If we discard the traditional view of BC, Lucan's criticism of Caesar can be better understood. What gives BC its own special ‘unity’ is its consistently rhetorical and pathetic tone. Through a variety of paradoxes, exclamations, bizarre themes and original adaptations of old topics, Lucan incessantly searches for pathos, tension and contrasts to illustrate the fundamental paradox underlying his work, that of ‘civil war'.!* This means that maximum effect is sought in each individual passage: and Lucan's praise of Nero is perfectly in accordance with contemporary genre rules, as is his pathetic censure of Caesar. Just as the prologue required praise of the emperor and Lucan supplied it, so, when elsewhere indignation and moral protest were needed, Lucan supplied that. Lucan was well trained in the schools of declamation. Praise or censure of tyrants were stock themes for pupils, as we may see in the works of Seneca the Elder. Lucan's consistently negative image of Caesar in this way matches his post-Vergilian, rhetorical mode of epic composition. The invective against Caesar is, then, of a literary and rhetorical type and is not intended for direct political consumption; and there is no personal hatred, envy or insult against Nero, or any other living person in BC. Lucan can, with Koster, be counted as Claudian's predecessor in epic invective, but the contemporary political dimension so prominent in Claudian's Jn Rufinum! is not yet present in BC. This is not to say that BC can be dismissed as empty rhetoric, as it often was before the current renaissance of Lucan studies. Rather, it

LUCAN’S PRAISE OF NERO

139

should be appreciated in the light of its own pre-Romantic standards: in these terms BC is brilliant, full of vigour, talent and wit, but politically quite innocent. Lucan seems to have found a much more effective (if somewhat vulgar) way of insulting the emperor in real life. Suetonius Vita Lucani 4 tells how one day the poet visited a public latrine and, clariore cum crepitu ventris, as the Latin decently says, recited a halfline by Nero: sub terris tonuisse putes, *you would have thought that it thundered under the earth'. This intertextual joke upon Nero's solemn line about earthquakes is likely to have annoyed the Emperor a good deal more than all the rhetoric of BC.

NOTES 1.

S.Koster Die Invektive in der griechischen und römischen Literatur (Meisenheim am Glan 1978) 168; on Claudian's invective, cf. recently E. Potz ‘Claudian’s In Rufinum' Philol. 134 (1990) 66-81.

2.

Lucan also uses ‘invective’ on other levels. Koster (n.1) deals with his condemnation of Egypt and Cleopatra in Book 10. Many other examples might be adduced, from traditional censure of riches, to full-scale condemnation of civil war.

3.

Onthe personal relationship of Lucan and Nero, cf. Suet. Vit. Lucani 4-6; Vacca Vit. Lucani 12-16. On Lucan's death, cf. Tac. Ann. 15.70; Suet. Vit. Lucani 9; Vacca Vit. Lucani 17; Vit. Lucani 3.5. On the poet's last words, cf. V.J.C. Hunink

*Lucan's last words', in Studies in Latin literature and Roman history 6 ed. C. Deroux (Collection Latomus 217, Bruxelles 1992) 390—407.

4.

Cf. in general M. Morford ‘Nero's patronage and participation in literature and

5.

Cf. Suet. Vit. Lucani 1: prima ingenii experimenta in Neronis laudibus dedit; Vacca Vit. Lucani 13: certamine pentaeterico acto in Pompei theatro laudibus recitatis in Neronem fuerat coronatus.

6.

Cf.eg. E. Griset ‘Die Eloge auf Nero’, in Lucan ed. W. Rutz (Darmstadt 1970)

the arts’ ANRW 11.32.3 (1985) 2005-31.

318-25; I. Lana, ‘Il proemio di Lucano’ Studi di storiografia antica in memoria di L. Ferrero (Torino 1971) 131-47; F.M. Ahl Lucan, an introduction (Ithaca/ London 1976) 30. 47-8 a.o.

7.

For references to Calpurnius, the Einsiedeln eclogues and Seneca, cf. Morford (n.4) 2011, 2037. For Lucan's general inspiration, cf. also Verg. Georg. 1.24-42.

8.

Cf. Morford (n.4) 2014.

9.

On astronomical aspects of the proem, cf. P. Arnaud 'L'apothéose de Néron Kosmokrator et la cosmographie de Lucain au premier livre de la Pharsale (1,45-66) REL 65 (1987) BAGB (1989) 165-71.

10.

A

non-ironic

view

167-93; A. Le Boeuffle *Le séjour céleste de Néron’

of the

proem

is generally

adopted

by

modern

Lucan

scholarship. For a convenient survey of recent literature on this question, cf. W. Rutz 'Lucans "Pharsalia" im Lichte der neuesten Forschung’, mit einem

140

VINCENT

HUNINK

bibliographischen Nachtrag 1979-1982 vom Verf. und 1980-1985 von Heinrich Tuitje, ANRW

11.32.3 (1985)

1457-537; esp.1482-5. To this may be added P.

Grimal ‘Le prologue de la Pharsale et les intentions de Lucain' VL 96 (1984) 2-9; Morford

(n.4) esp. 2014-15; A.M.

Dumont

'L'éloge de Neron’ BAGB (1986)

22-40; F. Brena *L'elogio di Nerone nella Pharsalia: moduli ufficiali e riflessione politica’ MD 20-21 (1988) 133-46.

Clear examples of this line of thought are W.D. Lebek Lucans Pharsalia. Dichtungsstruktur und Zeitbezug (Góttingen 1976) e.g. 10-27 and 279-84; Grimal

(n. 10).

12.

Vacca Vit. Lucani 13: ediderat ... tres libros quales videmus; 18: reliqui enim VII belli civilis libri ....

13.

Cf. P.H. Schrijvers Crise poétique et poésie de crise; la réception de Lucain aux XIXe et XXe siécles, suivi d'une interprétation de la scene ‘César à Troie" (La

Pharsale,

9,950-999) (Koninklijke Nederlandse Akademie van Wetenschappen:

Mededelingen van de afdeling Letterkunde. Nieuwe Reeks, deel 53 no. 1, Amster-

dam/New York etc. 1990) 13f. 14.

For this interpretation of Lucan's work, cf. V.J.C. Hunink M. Annaeus Lucanus

Bellum Civile Book III: A

Commentary (Amsterdam 1992); cf. further J. Masters

Lucan's Bellum Civile (Cambridge 15.

Cf. Potz (n.1). Potz shows Rufinus with panegyric of the two men. In the case contrasted with Caesar in

1992).

how /n Rufinum combines fierce criticism of the dead Stilicho. Thus the effect comes from the opposition of of BC such an opposition is lacking, Nero not being any way.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 141-54 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

JOSEPHUS BELLUM JUDAICUM 4.559-63: INVECTIVE AS HISTORY G.M. PAUL

In the preface to the Bellum Judaicum (1.11) Josephus writes:! Should, however, any critic censure me for my strictures upon the

tyrants or their bands of marauders (τὸ λῃστρικὸν αὐτῶν) or for my lamentations over my country's misfortunes, I ask his indulgence for a compassion which falls outside an historian’s province.

Lamentations for the fate of the city of Jerusalem duly appear at Bellum Judaicum 5.19f. (note the reference to a νόμος, asat 1.11), but it is also clear throughout the work that the rebels and their leaders arouse in Josephus strong passions which are reflected in his

language. One rebel leader, John of Gischala, is a particular focus for Josephus’ feelings, perhaps because at the outset of the war, before

Josephus

had gone over to the Romans, he and John had co-

operated more closely than Josephus felt it expedient to recall. So when he came to write his account, Josephus sought to distance

himself from John. Several recent commentators, in different ways and with different emphases, have attempted to modify the ex-. aggerations and distortions they find in Josephus’ depiction of John, who was arguably a more respectable figure than Josephus’ account allows.” Observed differences between the Bellum Judaicum and the account of the same events in the later Life contribute to the correction of Josephus’ accounts. As Rajak well observes, “ἃ historian writing of events in which he was involved will be weakest

when it comes to personalities"? The rhetoric of Josephus' depiction does not make the modern critic's task any easier. In the Bellum Judaicum John himself is 141

142

Ο.Μ. PAUL

introduced in a “defamatory set ρὶςος᾽" (2.285f.; cf. 4.85) which Thackeray thought to be "partly based" on Sallust's sketch of Catiline (Sallust Catilina 5).5 In fact, despite two or three similarities in content of a fairly general nature, there is a lack of formal similarity between Josephus' sketch and Sallust's. As Thackeray himself notes (118), Josephus “had but a slight acquaintance with Latin literature"; therefore Thackeray, in accordance with his theory, attributes the use of the Sallustian model to one of Josephus'

ubiquitous assistants. There is no need, however, to posit a Latin model for the Josephan sketch, since Greek models already existed and W. Theissen has plausibly argued that Sallust himself based his sketch of Catiline on the practice of Poseidonios (as recovered from

Diodoros).’ The hostile language of Josephus’ description of John may certainly be influenced by a particular forerunner; but it also belongs, as will become clear below, to a stock of verbal abuse which

originates no doubt in the free exchanges of daily life but finds literary expression in Greek epigram, Old Comedy and the Attic orators, and in the language of religious and philosophical controversy. This stock was drawn on by later orators, Greek and Roman, and also on occasion by historians such as Theopompos and Timaeus.* Apart from any Jewish tradition of this kind, Josephus, it seems, was sufficiently knowledgeable of Greek and Greek literature to be aware of this stock and to draw on it. It is perhaps surprising that Thackeray did not choose to comment on the similarities between Josephus' description of the Galilean supporters of John of Gischala (Bellum Judaicum 4.559-63) and Sallust's description of the friends and supporters of Catiline (especially Catilina 14). Not that there is any reason to believe that

this passage of Sallust provided a model for Josephus: Sallust in turn owes something not only to Cicero's Catilinarian speeches (especially 2.7-10, 22f.; cf. 1.13) but also, as many scholars believe, to a passage of Theopompos which claims to describe the ἑταῖροι of Philip. Parts of the Theopompan passage are known from quotations in three authors, and it may have been a passage studied for imitation in school.? Each of these three descriptions, in Theopompos, Sallust and Josephus, attributes (for example) perverted sexual activity to

the groups described, Philip's ἑταῖροι, Catiline's associates and John's Galilean followers respectively. Further points of similarity will be discussed below, but for the present it is enough to note that the Theopompan and Sallustian passages, and (it will be argued) the Josephan one, all exemplify the influence of an ancient tradition of

INVECTIVE AS HISTORY

143

verbal abuse summed up in the terms ψόγος and vituperatio. A standard element of such forms was comment on the allegediy perverted sexual proclivities of the object of abuse.!° It may be useful to recall at this point that praise and blame were accepted features of ancient historiography. When it suited him, a writer might criticise the appearance of κατηγορία and ἐγκώμιον in the writings of other historians, as Josephus does in his preface (1.2). But the historian's duty to praise and blame was generally acknowledged.!! This function of historiography naturally encourages rhetorical treatment and helps to explain why writers, including Cicero,

believed

there was a close relationship between

historio-

graphy and epideictic oratory in particular."?! Josephus' description of John's Galilean followers is notable for generalities. No specific instance of the followers' alleged misdeeds is cited, and the verbs in the passage are predominantly imperfects

denoting generalised description. In addition, while deeds of cruelty and barbarism are readily attributed to John's followers elsewhere in the narrative, no specific instances of drinking blood (6.372 is again a generalised statement), effeminate behaviour or homosexual practices on their part are reported. This is one reason why it isappropriate to label this passage as invective rather than history. Before attempting to demonstrate this in more detail, however, it should be pointed out that ‘invective’ is here used to mean the type of verbal abuse which may be found in a variety of ancient literary forms, including comedy, epigram and satire as well as oratory and history. It is not

limited to the sense of the term recently defined by S. Koster:!? “a structured literary form whose aim it is with all appropriate means to disparage publicly a named person as a personality against the background of contemporarily valid values and norms", as for example in Cicero's In Pisonem. Koster himself recognises the relationship between the ‘invectivistic’ language employed from time to time in other literary forms and the language of what he regards as

invective proper. Invective (ψόγος, vituperatio) was one of the elementary exercises or progymnasmata practised in Graeco-Roman schools of rhetoric. * Along with encomium, of which it was the converse, treating the same range of topics with dispraise rather than praise, it fell within

the province of epideictic oratory. Quintilian believed that the pupil should engage in exercises in /audatio and vituperatio early in his training (3.4.20; cf. 3.4.1) and the later rhetors devote much care and attention to progymnasmata of this type. Topics for encomia of a

144

G.M. PAUL

man (and likewise for invective) included his race, city, family, education, deeds (to which particular attention was devoted) and external resources, such as kin, friends, possessions, household, fortune and so forth.'* One may observe in surviving speeches the tendency for the actions and behaviour of an adversary to be discussed in terms of stock themes. The principal ones are noted by Wilhelm Süss and R.G.M. Nisbet:" family status and social background, foreign birth, practice of a trade, avarice leading to theft, sexual misdemeanours, peculiar clothing, demeanour and appearance, cowardice, gluttony and drunkenness, often ‘supported’

by circumstantial detail which was invented or grossly exaggerated. The stereotyped nature of these themes originates, no doubt, in the negative value accorded by Greek and Roman society to certain

practices; but the language used to describe them also became stereotyped, and was handed on not only by imitation but also in the forms and procedures of rhetorical education.

With this information to hand, and bearing in mind the descriptions of Catiline's associates and Philip's ἑταῖροι already noted as well as the invectivistic language of other literary genres, we may turn to a more detailed consideration of Josephus' description of John's

Galilean supporters. Friends were considered an external good, and therefore formed a customary topic of an encomium or invective.!* Besides, pares cum paribus facillime congregantur was a well known

proverb! and it was natural to typify

a man by his friends or

followers. The nature of Josephus' description and ofthe language in which it is couched will become clearer if we engage in a step by step analysis of the passage, seeking linguistic parallels and social and cultural connections for its various phrases; the analysis begins at 4.560: With an insatiable lust for loot, they ransacked the houses of the wealthy: according to I. Opelt, "the most frequent designation of the

political opponent as disturber of public order, as one who acts illegally, as revolutionary and anarchist ... is the legal metaphor latro", and she has collected numerous examples from invectivistic contexts, while Süss cites a number of unlikely charges of theft in the Greek orators.? Nisbet notes that “avarice was another popular charge", and with avarice was associated rapacity.?! Theopompos alleges that Philip’s ἑταῖροι live τοῖς λῃσταῖς παραπλησίως (FGrH 115 F 224; cf. F 225b ἁρπάζειν καὶ φονεύειν). Josephus' own constant use of λῃσταί to designate the insurgents owes at least

INVECTIVE AS HISTORY

145

something to the rhetorical tendency to describe one’s political opponents

as thieves and

robbers.

M.

Hengel

has argued

that

Josephus uses the term especially in contexts of moral condemnation.?? The thief is a recurrent figure in mime, comedy and epigram.?? It needs no stressing that theft was condemned in Greek and Roman law-codes as well as in the Decalogue (Exodus 20.15). Theft is, of course, likely to be condemned in any society in which the

possession of property confers social status. the murder of men and the violation of women were their sport: murder, too, was condemned in Greek, Roman and Hebrew codes.

The word parricida which originally meant the murderer of a near relative is employed metaphorically in Latin invective to designate opponents, especially those who can be represented as revolu-

tionaries or public enemies.?* In the New Testament, in the list of vices at 1 Timothy 1.9 “‘patricides and matricides" are perhaps not intended to be taken literally, but to be examples of the grossest sins. In Sallust's Catilina 52.31 Cato is represented as describing the Catilinarians as crudelissimi parricidae (cf. Cicero In Catilinam 2.22; Pseudo-Cicero In Sallustium 6.18), and Cicero employs the term to

designate Clodius, Antony and Dolabella.?* Carnifex, ‘hangman’, is similarly used, applied memorably to the young Pompey (Valerius Maximus 6.2.8). Lysias 10 deals with a case involving a slanderous charge of murdering a father. Parricidae are one of the alleged components of Catiline's supporters (Sallust Catilina 14.3), while Theopompos describes Philip's ἑταῖροι as ἀνδροφόνοι ... τὴν φύσιν (FGrH 225a-c). Mistreatment of women, especially sexual mistreatment, was another frequent charge in ancient invective along with other types of sexual misconduct (see below). Catiline's followers allegedly included adulterers (Sallust Catilina 14.2), examples of the general

decline in pudor and pudicitia claimed by Sallust (Catilina 12.2; 13.3). Related themes are to be found in satire and epigram, Greek and

Roman.”’ Libidines figure among other conventional charges which, Nisbet says, “should not be too readily believed".?* Even Cicero was so charged (Cicero De domo sua 93; cf. Dio 46.18.4, 6).

They caroused on their spoils with blood to wash them down: the Old Testament contains strong prohibitions against the shedding of blood (e.g. Genesis 9.5f.). Blood was regarded as “the life of the flesh" and in Leviticus the eating of blood with the flesh is strictly forbidden (17.10-14; cf. 3.17; 7.26f£.; Genesis 9.4; Deuteronomy 12.23). The behaviour of John's followers as described here would

146

G.M. PAUL

arouse particular revulsion in Jewish readers. But in Greek and

Roman contexts as well the drinking of blood could have negative connotations.? There were rumours that Catiline when seeking to bind his followers to the conspiracy had given them to drink human

blood mixed with wine (Sallust Catilina 22.1f.).? According to Herodotos, this was a Scythian practice (4.70), and the same writer

tells us that when the Greek and Carian mercenaries serving in Egypt

sought revenge against Phanes they killed his sons and drank their blood mingled with wine (3.11). A similar allegation is made by

Josephus about the behaviour of 'the tyrants' and their accompanying λῃστρικόν as the complete fall of Jerusalem drew near (6.372): “and if ever they found a victim with food, they snatched it

from him and devoured it, all defiled with blood". Note again the generalised description (conveyed by the optative of past general condition) and the absence of any particular case. In default of any specific instances, the allegation at 4.561 must be considered more akin to invective than to sober history. Human sacrifice and libations of blood constitute a topos of reports of conspiracies (e.g. Plutarch Publicola 4.1). One may perhaps also compare the charge of human sacrifice levelled against Pythagoreans and early Christians and

encouraged by their secret meetings.?! and from mere satiety unscrupulously indulged in effeminate practices: ἐνθηλυπαθέω is a hapax in Josephus, and this is the only passage in Greek literature to use the word revealed by a search of the Thesaurus Linguae Graecae database using an Ibycus computer. The meaning of the word, however, may be safely inferred

from a number

of related terms cited by H. Herter in his useful article s.v. effeminatus

in the Reallexikon für Antike und Christentum.? Muliebria pati (Sallust Catilina 13.3: note the context) seems to be the Latin equivalent. As Herter says, the designation 'effeminate', both in Greek and Latin, has the character of a reproach. As will become clear below, except for example in ritual contexts various ‘effemi-

nate' practices were condemned

not only in Greek and Roman

society but also in Jewish. What were seen by Greek and Roman society as sexual misdemeanours or excesses were a frequent topic of invective (whether the charges were true or not) and ‘effeminate’

behaviour was a typical imputation.?? Paul in Romans 1.26f. may be taken as representative of the Jewish condemnation of the exchange of sexual roles: Their women exchanged natural relations for unnatural, and the men

likewise gave up natural relations with women and were consumed

INVECTIVE AS HISTORY

147

with passion for one another, men committing shameless acts with men.

For the appearance of ἀρσενοκοῖται in New Testament ‘vice lists’ see also 1 Corinthians 6.9; 1 Timothy 1.10. plaiting their hair and attiring themselves in women's apparel: Herter has collected much evidence for men's attention to the hair and elaborate hair-styles as signs of effeminacy.** The compti capilli of Gabinius (Cicero /n Pisonem 25) and pexus capillus of Catiline's associates (Cicero Jn Catilinam 2.22) are taken to imply effeminate or debauched conduct. Commenting on In Pisonem 25, Nisbet aptly quotes Titinius 112: quasi hermaphroditus fimbriatum frontem gestas. In Antiquities 19.30 Josephus describes the Emperor Gaius as devising περιθέσεις πλοκαμίδων in order to counterfeit a feminine appearance; compare the alleged behaviour of Elagabalus (Dio 79.13.2. A man who wears women's apparel is described in Deuteronomy 22.5 as an “abomination to the Lord your God"; in Antiquities 4.301 Josephus adds to this prohibition “above all in battle". Whatever the reason for this addition, it is generally assumed that the prohibition against the sexes exchanging clothes was to differentiate the Israelites from the adherents of surrounding cults whose worshippers put on the clothes of the opposite sex in certain rituals. Ritual transvestism is known also in Greek and Roman society.?? In ordinary life, however, Seneca considers the exchange of clothing with a woman as contra naturam vivere (Epistulae 122.7) and Diogenes Laertius cites it as an example of a practice forbidden by unwritten law (3.86). In myth also the wearing of women's clothing by a male is considered degrading: the locus classicus here is Herakles' exchange of clothes with Omphale, described by Lucian as θέαμα αἴσχιστον and frequently portrayed in works of art.” Achilles' concealment in girl's clothing among the daughters of Lykomedes on Skyros may also be considered shameful. Works of art naturally focus on the throwing off of the disguise," but in the Achilleid Statius represents his hero as not only bewailing his absence from the field but also his ‘female’ behaviour (1.624-39, especially

634-6; cf. [Bion] 2.15-20). drenching themselves with perfumes and painting their eyelids to enhance their beauty: though the use of perfumes by men, at least in some contexts, was tolerated and even expected in Israelite society, painting of the eyes, even by women, was not well regarded.?* Though Marquardt declares that in Imperial Rome men challenged women in the use of cosmetics, in Greek and Roman society in earlier

148

G.M. PAUL

days the use of perfumes by men could earn for them the label

κίναιδος

and

is a frequent charge in invective.? Likewise the

painting of the described by the And not only women, devising

eyes was considered an effeminate practice; it is Elder Pliny as peculiar to women.* did they imitate the dress, but also the passions of in their excess of lasciviousness unlawful pleasures:

the

'effeminate

charge

of

practices’,

mentioned

above,

is here

extended and emphasised. One is reminded of the passage in Theopompos which makes similar charges about the ἑταῖροι of Philip declaring that one would justly have supposed them to be not ἑταῖροι but ἑταῖραι. The bitterness and exaggerations of this passage are condemned by Polybios.*' Sallust mentions similar charges as allegations against the associates of Catiline (Catilina 14.7); at 13.3

viri muliebria pati is cited by Sallust as one of the aspects of the decadence of his day. The charge of homosexual behaviour was a common one in invective, as can be seen for example in the Znvectiva in Sallustium (3.9; 8.21); the work was erroneously attributed to Cicero. Sexual perversion is but one of a long catalogue of vices with which the Wisdom of Solomon 14.26 charges idolaters. Similarly, Colotes is alleged to have made such charges against other philosophers (Plutarch Moralia 1086e).*? wallowing as in a brothel in the city, which they polluted from end to end with their foul deeds: the exaggeration and hostility of the language hardly require comment. In its tone it again recalls

Theopompos' description of the ἑταῖροι of Philip. Yet while they wore women's faces, their hands were murderous, and approaching with mincing steps they would suddenly become warriors and whipping out their swords from under their dyed mantles transfix whomsoever they met: the use of women's clothing as disguise by attackers is alleged in other cases also, for example, Pelopidas' attack

on the Theban tyrants.* The effectiveness of any such disguise, at least in daylight and if the men were bearded, is perhaps open to question, though the motif of disguise in the clothing of the opposite

sex appears in folk-tales.*^ Of course, as far as the Jewish War was concerned, the possibility that such actions might have been carried out does not prove that they were. Josephus certainly cites no specific

incident. The tone of his report is hostile. In the account the use of "dyed mantles” is highlighted, if the reading βεβαμμένον is correct. One can perhaps compare a passage in the invective against Cicero

ascribed to Q. Fufius Calenus where Cicero's λεπτὰ χλανίδια, along with allegedly perfumed hair, are made a matter of reproach (Dio

INVECTIVE AS HISTORY

149

46.18.3). Also Herter hascollected a quantity of evidence to illustrate that coloured garments, especially of purple, scarlet or saffron, were characteristic of effeminate dress.*5 θρυπτόμενοι is “being enervated or unmanned”,

hence “being effeminate” (**mincing" in the Loeb

translation). The combination of unmanly behaviour and dress with

violent action has something in common with Cicero's description of Catiline's closest associates (Jn Catilinam 2.23): hi pueri tam lepidi ac delicati non solum amare et amari neque saltare et cantare sed etiam sicas vibrare et spargere venena didicerunt.

The language and tone of this passage, then, have much in common

with acknowledged

passages

of invective. The life and

career of Josephus were passed at the intersection of three cultures, Jewish, Greek and Roman. Each of these had its own native tradition of verbal abuse which came to the fore, for example, in religious

controversy, as in the Wisdom of Solomon (quoted above)“ or in the kind of invective directed by Colotes against other philosophers. The

so-called ‘vice lists’ or ‘catalogues of vices’ employed by many New Testament writers (e.g. 1 Timothy 1.9f.: “the law is not laid down for the just but for the lawless and disobedient, for the ungodly and sinners, for the unholy and profane, for murderers of fathers and murderers of mothers, for manslayers, immoral persons, sodomites,

kidnappers, liars, perjurers, and whatever else is contrary to sound doctrine ...”) exhibit a similar basic stock of words of abuse. The beginnings of such lists can be traced back to Plato (e.g. Gorgias §25a) and the Stoics, and also to the lists of cardinal vices. These were

adopted by Hellenistic popular philosophy and also by contemporary Judaism (e.g. Philo), and from such sources passed to the New Testament writers.*’ The terms of abuse that occur in these lists are

characteristic also of Greek and Roman comedy. Many years ago Adolf Deissmann pointed out that there is a considerable overlap between

1 Timothy

1.9f. and the list of maledicta with which the

pander Ballio is loaded in Plautus Pseudolus 360-9: inpudice ... sceleste ... bustirape ... furcifer... sociofraude ... parricida ... sacrilege ... periure ... legirupa ... permities adulescentum ... fur ... fugitive ... fraudulente ... inpure ... leno ... caenum.* Similar invectivistic language, of course, is already to be found in Old Comedy, epigram and satire. As we have seen, Süss has collected examples of such

Bescheltungstopik

in oratory

and

pointed

to its relation

to the

language of comedy.*? When the study of oratory was formalised in schools of rhetoric, the pupils studied examples of invective along with other types of speech. A laterexample of how this might work, if

150

G.M. PAUL

Cicero's claim is accurate, is the alleged study of the Jn Pisonem by schoolboys within about a year of its publication. But, as Theo-

pompos’ description of the ἑταῖροι of Philip shows, invective had also invaded historiography at an early date. Not only so, but the way the Theopompos fragment is transmitted at least in part by three separate authors strongly suggests that the passage had been

excerpted for study and imitation in school.*! In this way theory and practice constantly interacted.

The vocabulary of abuse which so strongly colours Josephus' portrait of the supporters of John could, therefore, have been drawn from a variety of sources. As the later Contra Apionem shows, Josephus was quite familiar with the language of religious contro-

versy, and given his Jewish education he was undoubtedly acquainted with the catalogues of vices cited in contemporary Jewish paraenesis, the nature of which is best known from the similar catalogues in New Testament writers. But one should also not overlook the extent to which Josephus' account of the supporters of John of Gischala recalls Theopompos' attack on the ἑταῖροι of Philip. Both sets of

companions are described as murderers and robbers and especially as given to homosexual excesses. In Sallust's description of the associates

of Catiline,

which

is often

taken

to be modelled

on

Theopompos' attack on the ἑταῖροι of Philip, homosexual activity is mentioned

only to be denied (Catilina

14.7), so in respect of this

charge the passage in Josephus is closer to Theopompos. Admittedly,

as the evidence cited above shows, charges of murder, robbery, as well as homosexual excesses, were commonplaces of ancient invective, but the combination of these charges in a similar setting (namely an attack on a man's associates) raises the possibility of a dependence,

direct or indirect, of Josephus

on Theopompos,

es-

pecially since the Theopompos passage seems to have been widely known. The fact that Josephus is not otherwise known to have read

Theopompos, whom he does not even name in his extant writings, need not tell entirely against this view. If, as suggested above, the

attack on Philip's ἑταῖροι circulated as an excerpt for study in school, Josephus may have met with it there, though it is impossible to be certain of this since we are unfortunately not at all informed on how Josephus acquired his knowledge of Greek. An alternative possibility

would be to assume that Josephus in this passage is modelling himself on the work of some intermediate orator or rhetorical historian

whose

work

had

been

influenced by Theopompos.

A

passage in Contra Apionem, describing the attacks of critics on

INVECTIVE AS HISTORY

151

Bellum Judaicum, is relevant here; according to one translation they

treated the Bellum Judaicum

“as an exercise for the display of

perverse accusation and calumny, such as is set to boys at school”.32 To illustrate such a possibility, one could point to Cicero's description

of the character of Catiline's forces, especially the postremum genus (In Catilinam 2.22f.), which seems to owe something to Theopompos'

description of Philip's ἑταῖροι. In the present state of our

knowledge,

these

can

only

be

suggestions. Certainly, however, the relationship in ideas between

the alleged descriptions of Philip's ἑταῖροι and John of Gischala's supporters is a close one and amply justifies the contention that in the case of Josephus' description also we are dealing with invective

rather than with history in its modern sense.*? NOTES 1l. 2.

The Loeb translation of Josephus is followed here and throughout the paper. Cf.SJ.D. Cohen Josephus in Galilee and Rome (Leiden 1979) 221-3; T. Rajak Josephus, the Historian and his Society (London

1983) 131f., 160—5; M. Goodman

The Ruling Class of Judaea (Cambridge 1987) 201f. E.M. Smallwood The Jews

under Roman

merchant".

Rule

(Leiden

1976)

304

n.42,

regards

John

as

a "wealthy

3.

Rajak (n.2) 161.

4.

Rajak (n.2) 161.

5.

H.St.J. Thackeray Josephus: the Man and the Historian (New York 1929) 119f.; cf. his notes in the Loeb translation, 1.548, II.27.

6.

K. Vretska "Bemerkungen zum Bau der Charakteristik bei Sallust" SO 31 (1955)

105-18, argues persuasively that Sallust in this sketch follows the rhetorical schema for an encomium; cf. also his commentary on the Catiline (Heidelberg 1976) I.121f. Josephus' sketch of John does not follow such a rhetorical schema. 7.

DeSallustii, Livii, Taciti digressionibus (Diss. Berlin 1912) 11-13; cf. Diod. 33. 1f.;

35.5; 37.10.1. For Diodoros' dependence on Poseidonios see E. Schwartz RE 5 (1905) 690f.

8.

Cf.e.g. FGrH 115 T 20 (= Dion. Hal. Ad Pomp. 6.9). 566 T 1. See further the references to praise and blame in historiography below and in n.11.

9.

Cf. Theopompos FGrH 115 F 224, 225a-c; Vretska's commentary, [.249f.; G.M. Paul Phoenix 39 (1985) 161.

10.

Cf. W. Süss Ethos: Studien zur älteren griechischen Rhetorik (Leipzig 1910, repr. Aalen

11.

1975) 249f.

Cf.e.g. Polyb. 1.14.57; 2.61.6; 12.15.9: Diod. 11.46.1; 15.1.1; 30.15 fin.; Tac. Ann. 3.65; Lucian Hist. Conscr. 59; G. Avenarius Lukians Schrift zur Geschichtsschreibung (Meisenheim am Glan 1956) 157, 163; A.J. Woodman Rhetoric in Classical Historiography (London 1988) 40-4; M. Heath Unity in Greek Poetics

(Oxford 1989) 87f.

G.M. PAUL

152

Woodman (n.11) 95-8. S. Koster Die Invektive in der griechischen und römischen Literatur (Beiträge zur klassischen Philologie 99, Meisenheim am Glan 1980) 39; cf. 354. E.g. D.L. Clark Rhetoric in Greco-Roman Education (New York

1957) 194-8.

Cf. Hermog. Prog. 7; Aphthon. 9; Theon 8; Prisc. 7; Aphthonius includes a model progymnasma, a ψόγος of Philip. Cf. H.I. Marrou A History of Education

in Antiquity tr. G. Lamb (London 1956) 198f., listing the topics of encomium as detailed by Theon. Cf. also Rhet. ad Her. 3.6.10; Quint. 3.7, esp. 10-18 (laudatio), 19-22 (vituperatio).

See Clark (n.14) 195f. for a translation of Hermogenes' remarks on the nature and procedure of encomium. Süss (n.10) 246-54; R.G.M. (Oxford 1961) 192-7.

Nisbet Cicero: In L. Calpurnium Pisonem Oratio

Cf. e.g. Rhet. ad Her. 3.6.10; Hermog. Prog. 7; Theon 8; Cic. De Or. 2.46; cf. Arist.

Rhet. 1360620.

Otto Sprichwórter s.v. par (1); Cic. De Sen. 3.7; Quint. 5.11.41; cf. Homer Od.

17.218.

I. Opelt Die lateinischen Schimpfwórter und verwandte sprachliche Erscheinungen (Heidelberg 1965) 132f.; praedo and pirata are similarly used; Süss (n.10) 249. Nisbet (n.17) 195; cf. Süss (n.10) 249.

M. Hengel Die Zeloten? (Leiden 1976) 42-7, esp. 44. Cf. F.J. Brecht Motiv- und Typengeschichte des griechischen Spottepigramms (Philologus Suppl. 22.2, Leipzig 1930) 68-70.

Opelt (n.20) 131f. Cic. De Dom. 57; Sest. 82; Phil. 4.5; 13.42; other references in Opelt (n.20) 132.

Süss (n. 10) 249f. Cf. Brecht (n.23) 52f. Nisbet (n.17) 194f. Cf. J.H. Waszink RAC 11.459-73 s.v. ‘Blut’, esp. 465. See Vretska's commentary ad loc., citing also Dio 37.30.3; Plut. Cic. 10.4; cf. Tac. Ann. 12.47.2 and Koestermann ad loc. Cf. Cic. In Vat. 14 and Pocock ad loc. 32.

RAC IV.620-50; for the terminology see esp. 620.

33.

Süss (n.10) 249f.; Opelt (n.20) 154-7, esp. 155, for examples of effeminatus and similar terms; see also Brecht (n.23) 527 for sexual abuse in comedy, satire and epigram. For allegations of homosexuality and other departures from conventional sexual norms in Greek comedy and political oratory see K.J. Dover Greek Popular Morality (Oxford 1974) 33; attacks on Greek and Roman

politicians and rulers for effeminacy, Herter (n.32) 625f.

INVECTIVE AS HISTORY

153

Herter (n.32) 6321. Herter (n.32) 622-4.

Lucian Hist. Conscr. 10 and Homeyer adloc.; G. Herzog-Hauser RE XVIII.385ff. 5. V. ‘Omphale’, esp. 393f. K. Schefold

La peinture pompéienne (Collection Latomus

108. Brussels

187-9: LIMC 1.1.56-65, esp. nos. 105-75; Hor. Od. 1.8.13-16.

1972)

Perfume: Song of Sol. 1.3; painting of eyes: 2 Kings 9.30; Jer. 4.30; Ezek. 23.40.

J. Marquardt Das Privatleben der Römer? (Leipzig 1886) 786-8; Athen. 13.565e: invective: Cic. Red. in Sen. 12, 13, 16: Sest. 18; Pis. 25; Dio 46.18.3 possibly implied in Cic. Car. 2.22f.

(of Cicero!);

Athen. 12.528f.. 529a; Juv. 2.935; Plin. NH 11.54. Polyb. 8.9.513 = Theopompos FGrH 115 F 225a; cf. Athen. 6.260d, 261a = F 225b; Polybios' condemnation, 8.10.14.

For sexual perversion as a rhetorical topos of invective see Brecht (n.23) 9 and n.60, 54f.; on κίναιδοι, 55f.; Süss (n.10) 249f. Plut. Pel. 9.4ff.; Mor. 594e, 597c; Xen. HG 5.4.4-7.

Stith Thompson Motif-Index of Folk Literature (Bloomington, Indiana 1955-8) 1V.439, K1836; H. King PCPS n.s. 32(1986) 53-75, discusses instances of women

disguised as men. 45.

Herter (n.32) 630. Wisdom of Solomon

14.216; L.T. Johnson JBL 108 (1989) 419-41, esp. 434-41,

examines the use of slander in polemic by NT writers and their contemporaries, including Jews. 4.

These lists and their topoi have frequently been discussed; for ancient references and citations of modern literature see e.g. Martin Dibelius and Hans Conzelmann A Commentary on the Pastoral Epistles (Philadelphia 1972) 223, on 1 Τίνι. 1.9f.; Hans Conzeimann A Commentary on the First Epistle to the Corinthians (Philadelphia 1975) 100f., 106; Hans Dieter Betz A Commentary on Paul's Letter

to the Churches in Galatia (Philadelphia 1979) 281-3. It is often assumed that the NT lists originate in Hellenistic Jewish paraenesis. 48.

A. Deissmann Light from the Ancient East tr. L.R.M. Strachan (London

49.

Süss (n.10) 245-54.

50.

Ad Q. Fr. 3.1.11 (Sept. 54 BC): ... praesertim cum ... meam (sc. orationem) in illum

315-18, esp. 317.

(sc. Pisonem) pueri omnes tamquam dictata

51. 52.

1927)

perdiscant.

See n.9 above. 1.53, in Thackeray's alternative translation (I. 184, note a). I take it that Josephus

is being charged with having composed a school-exercise in invective rather than his critics; Josephus responds by maintaining his accuracy and truth. Ifthere isa reminiscence of Thuc. 1.22.4, as sometimes maintained, yópvacpa here corresponds to ἀγώνισμα in Thuc. γύμνασμα was in Josephus' day the normal term for what later was more usually called προγύμνασμα. cf. R.F. Hock and E.N. O'Neill The Chreia in Ancient Rhetoric (Atlanta 1986) L 12f.

154

53,

G.M. PAUL

Anearlier version ofthis paper was read at the Faculty-Graduate Symposium of

the Department of Classics, McMaster University. I should like to thank Dr.

Aileen Ajootian for bibliographical advice; also the editors for their comments.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 155-67 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

AFFRONTS AND QUARRELS DOUGLAS

IN THE ILIAD

L. CAIRNS

This paper! deals both with the usage and meaning of key terms, and with the wider significance of the contexts in which they are used, in

that order. In the first part, I use evidence from the Odyssey as well as the 7/iad; but the actual quarrels I consider in the second part are from the /liad. It will be obvious that the first section has as its target the theories of A. W.H. Adkins. This is not ‘flogging a dead horse’, but recognition (a) of the importance of Adkins' work in setting the agenda for his successors and (b) of the enduring persuasiveness of many of his theories, even among those who would claim to repudiate them. Adkins, moreover, has not been decisively refuted;

in fact, he often comes off best in exchanges with his opponents.” So I make no apologies for dealing first with Adkins' approach. Adkins' edifice is constructed on the basis of a positive distinction, which he traces back to the Homeric poems, between competitive

and co-operative values.? This distinction is mirrored in another, between strong and weak terms of approval and disapproval; the strong are those which operate in the competitive sphere alone. A further element is supplied by close attention to the reference of words of disapproval to agents or patients; this distinction is particularly relevant to the theme of this paper, because, in the context of the affront, it subsumes those between co-operative and competitive spheres, weak and strong terms: if a term ‘discredits’ (Adkins' word) agents alone, it will decry a co-operative failure, whereas if its reference is to the patient, the recipient of the affront, then it highlights a competitive failure to defend oneself. With these categories in mind let us now examine certain terms crucial to Adkins' thesis which are also important for our understanding of the attitude to affronts and quarrels. 155

156

DOUGLAS

L. CAIRNS

First alox oc. The basic sense of this term is ‘ugliness’ (it continues

to be used of physical ugliness long after αἰσχύνη has become regular

in the transferred sense of 'disgrace"),* and in this sense αἶσχος can refer to a state of affairs, as in Odyssey

18.220-5, where Penelope

(225) tells Telemachus that he will suffer αἶσχος and λώβη ('disfigurement’) among men for allowing a guest under his protection to be ‘disfigured’ (ἀεικισθήμεναι, 222). Here αἴσχος is a state of affairs resulting from an affront, and it reflects badly on one whose honour is associated with the patient of the affront.’ In the plural, ato yea can refer to the affront itself, both verbal and non-verbal, and to taunts

which draw attention to the ‘ugliness’ of another's situation. As a verbal affront, or as a taunt, αἴσχεα thus refers to comments on the (alleged or supposed) αἶσχος of another, remarks which are intended to inflict, cause, or compound the ‘ugliness’ in which the recipient is implicated. So αἴσχεα, ‘insults’, cannot be considered in isolation

from αἶσχος, the condition in which the affront may leave (and is intended to leave) its recipient: αἶσχος is both the aim and the frequent result of αἴσχεα, insults may in fact constitute the ugliness, and the ugliness itself may call forth further αἴσχεα. So far, we have seen no reference to the discredit of agents in the

use of these terms; an affront is clearly intended to diminish the honour of its patient, and, as Penelope warns Telemachus, dishonour, αἶσχος, may well be its result. In Odyssey 1.227-9, however, Athena comments on the excessive ὕβρις of the suitors (227, ὑβρίζοντες ὑπερφιάλως): any sensible onlooker would experience νέμεσις upon witnessing the many αἴσχεα taking place before him. According to

Adkins’

the precise reference of αἴσχεα is to the discredit of

Telemachus; αἴσχεα can only (and indeed must) be αἰσχρόν for their recipient. This, however, is clearly wrong. The reference of atoyea is

to the suitors' affronts, and so to their attempts to dishonour another; but this attempt rebounds on them. Athena is masquerading as a guest in Telemachus' house, and it is incredible that she should describe a situation in terms designed to draw attention to her host's disgrace.! Rather her designation of the suitors' acts as αἴσχεα belongs with her description of their conduct as ὕβρις (illegitimate dishonouring of another person)/? it is to this that αἴσχεα refers, the illegitimacy being highlighted by the reference to the universal indignation (νέμεσις) which such actions excite. Since this is so, it is clear that an affront designated αἴσχεα can be ugly for its agent rather than for its patient, and that the connotation of ‘ugliness’ in the word can be used to draw attention to this fact.'? It is not that

AFFRONTS AND QUARRELS

behaviour which discreditable for and νέμεσις; on αἶσχος for its

IN THE /LIAD

157

discredits Telemachus is incidentally made to seem the suitors by the application of terms such as ὕβρις the contrary, conduct intended or likely to create recipient can, in fact, constitute aloxog for its

perpetrator. Next, ‘ugly words'.!' In this phrase the adjective αἰσχρός is glossed by Hesychius as τοῖς αἰσχύνην ἐνεγκεῖν δυναμένοις, an explanation obviously intended to forestall the assumption (natural for users of Greek in later centuries) that the adjective conveys disapproval of the manner of the address. This clearly is not the purpose of the phrase, and Hesychius is entirely right to describe *ugly words' as those which are potentially αἰσχρόν for their recipient. Yet this observation gives us no warrant to assume that the adjective itself inevitably refers to the actual discredit of patients. The primary reference of the adjective, 1 suggest, is not to the dishonour of patients, but to the fact of the affront itself; αἰσχρὰ ἔπη, I suggest, means ‘words that constitute an aloyoc’ in much the same way as ὀνείδεια ἔπη means ‘words that constitute an öveidog’."? But to back this up I have to turn to the usage of the adverb αἰσχρῶς, and to the wider values exhibited in Iliadic quarrels. αἰσχρῶς occurs only twice in the poems, once in each (Iliad 23.473; Odyssey 18.321). Once again, the word occurs in the context of an affront; in both cases it is the act of addressing another that is performed *in an ugly manner'. Again, this focal reference to the affront entails a reference to the attempt to dishonour or humiliate, to place the recipient of the affront in a humiliating position or category. But even a brief look at the two passages reveals that the patient is not inevitably discredited. In Odyssey 18.321 the words delivered αἰσχρῶς are directed at the disguised Odysseus by the unfaithful maidservant, Melantho. While it is certainly humiliating for Odysseus to have to endure such abuse from an inferior, it can

hardly be that we are not also supposed to disapprove of Melantho's behaviour; ill-treatment of guests and abuse of one's master, even if he is incognito, scarcely constitute the sort of conduct commended

by the Odyssey. To be addressed αἰσχρῶς is, then, to be placed ina situation which is potentially αἰσχρόν for oneself, one which one may even oneself regard as αἰσχρόν; but still αἰσχρός does not mean *discreditable for the patient’, and the mere use of a form of ato y póg does not entail that the recipient of an affront is inevitably dishonoured. Instead, I suggest two possibilities: (a) that there is a focal reference in such locutions to the occurrence of an affront; and

158

DOUGLAS

(b) that the aesthetic aspect of the adjective and characterise the situation in a more general way.

L. CAIRNS

adverb

may

To pursue this approach further we need to look at the other passage in which the adverb occurs, which provides a convenient

bridge from consideration of the behaviour of lexical items to examination of wider contexts. In the quarrel in J/íad 23 between the

lesser Ajax and Idomeneus the former addresses the latter αἰσχρῶς (473), and Idomeneus clearly resents the affront, returning Ajax's insults in kind (482-7). Crucially, however, the quarrel is quickly brought to an end by Achilles (492-4), who comments on the

‘inappropriateness’ of the quarrel itself (οὐδὲ ἔοικε, 493). This inappropriateness, moreoever, is such that disapproval of it is presented as a universal response, one which rests on values which even the participants in the quarrel share (καὶ δ᾽ ἄλλῳ νεμεσᾶτον, ὅτις τοιαῦτά γε ῥέζοι, 494).

Thus the quarrel itself is condemned

under the same quasi-

aesthetic standards as obtain in any use of the adjective αἰσχρός:

quarrels actually αἰσχρός passage

do not look nice, and even if this distaste for quarrels is not part ofthe reference of the adverb αἰσχρῶς and the adjective in such contexts (as I think), it is perfectly clear from this that the very fact of participation in a quarrel renders both

parties open to criticism. As represented by Achilles, then, neither party stands to gain honour by pursuing the quarrel, neither is dishonoured by backing down, and, on the contrary, it is continuation of the quarrel itself which would, on the grounds of its unseemliness, meet with general disapproval. The use of νέμεσις in this passage is of the utmost importance. First, there is the implication that Achilles’ response is itself vépeotc, a response, he claims, the opposing parties would share if they could

step back from their emotional involvement in the situation and see themselves as they see others. Note, then, that Achilles’ νέμεσις is impartial; he is neither victim of the affront nor partisan of either of

the actors. This is an important fact about νέμεσις: it need not signify the response of the victim of an affront, but can express the reaction of the disinterested bystander, who represents the values of society as a whole. Moreover, these values are capable of universalisation, as

the present passage makes clear; Ajax and Idomeneus would resent such behaviour in another, and therefore should recognise that they are in breach of universal standards personally endorsed by themselves. νέμεσις is often used in locutions which make the same appeal

to universal standards,'* and one of the standards on which νέμεσις

AFFRONTS AND QUARRELS

IN THE ILIAD

159

may be based holds that quarrels, violence and excess are to be avoided. That quarrels of this sort are unseemly and an occasion for νέμεσις implies a lack of αἰδώς in their participants, for αἰδώς is an emotion which both rejects the unseemly (in which case the verb, αἰδέομαι, is typically followed, in Homer, by an infinitive) and forestalls the νέμεσις of others, both those who are the direct recipients of one's actions (‘respect’) and those who are their witnesses (‘shame’).'® One very important and fundamental area in which αἰδώς is operative is that of φιλότης. One's φίλοι enjoy a special claim to τιμή to which

one's αἰδώς responds; any failure to accord such αἰδώς is regarded as disreputable, and so liable to the νέμεσις of others." Accordingly, αἰδώς often inhibits affronts and quarrels between φίλοι. φιλότης, of course, can exist where there is no blood tie; in particular, the prominent members of the opposing armies at Troy are each the φίλοι of the others,'? and this tie of φιλότης demands that they acknowledge each other's honour on a reciprocal basis.?? The force of such αἰδώς in defusing quarrels within the group is demonstrated in a passage of Jliad4. Agamemnon abuses (νείκεσσεν, 368) Diomedes, accusing him of slacking and comparing him unfavourably with his father. Diomedes himself does not immediately respond, but his companion Sthenelus does; he clearly regards Agamemnon's remarks asan affront to his honour, and is apparently ready to commit his honour to wiping out the insult. Diomedes' reponse, however, is totally different. ‘‘Out of αἰδώς for the rebuke of

the reverend king" (αἰδεσθεὶς βασιλῆος ἐνιπὴν αἰδοίοιο, 402) he refrains from making any reply at all, and after Sthenelus has spoken he explains why he feels no νέμεσις (413): he realises (412-18) that Agamemnon is attempting to encourage the troops, and accepts his right to do so; for, as he points out, both the glory of success and the sorrow of failure rest with him. Diomedes, then, feels no νέμεσις at remarks which another has taken as an affront because he can understand why they were made; he makes allowance for Agamemnon's aims, but his aims are not separable from his status as a superior king, itself inseparable from the pressures and responsibilities which that status entails. So Diomedes refuses to commit his honour because he accepts his place as a member of a group of píAoi in which some enjoy more status than others. (If helater gets his own back when a suitable occasion arises [9.33ff.], this does not detract from the reflective nature of his response in Book 4.) If Diomedes' response is the ideal, this does not mean that a

160

DOUGLAS L. CAIRNS

response such as that of Sthenelus is not likely. Rather it is sometimes accepted as normal that νέμεσις (on the part of the recipient or his

partisans) will follow remarks which can be construed as insulting.?! Clearly, whether one takes offence (feels νέμεσις) or overrides one's νέμεσις with αἰδώς will depend on the seriousness of the affront, the status

of the

parties,

and

the

character

of the

individual;

but

everything we have seen so far suggests that the ideal in every case is

compromise rather than resentment, so that to regard oneself as affronted without good cause is deprecated. Now to the quarrel that stands at the heart of the 7/iad: in Iliad | the high emotional involvement of the parties which leads them to commit their honour and the abnormal situation in which a god inspires the return of a legitimate mark of honour combine to make the dispute difficult to resolve, but the normal values still apply, as Nestor's attempt to calm the two makes clear. He deprecates the fact that a quarrel is occurring at all (254-8), expresses the opinion that compromise is the better course (274), and urges each to consider the legitimate claim to τιμή of the other (275-84: Briseis is a γέρας, a legitimate acknowledgement of Achilles' τιμή as a warrior, mentioned by Nestor in 283f., while Agamemnon is a superior βασιλεύς). These points have been made before,?? and could not be clearer. Nestor's

disapproval of Agamemnon's part in the quarrel is conciliatory, and therefore

muted,

but his is an evaluation

of the situation which

resurfaces repeatedly until the quarrel is resolved, an evaluation which adumbrates a universal feeling of νέμεσις at Agamemnon's breach of αἰδώς.2" Nestor's remarks on Agamemnon's failure to respect the τιμή of another therefore identify the same phenomenon as do Achilles' charges of ἀναιδείη (149, 158) and ὕβρις (203, 214); Achilles, for his part, is reminded of the kind of response, the αἰδώς for a superior which might override νέμεσις at an affront, which is

manifested by Diomedes in Book 4. So the ideals are quite clear: quarrels are to be avoided, out of consideration for the τιμή of others. In default of the achievement of this ideal, however, there is more sympathy for Achilles than for Agamemnon, in so far as retaliation for an affront is regarded more indulgently than its initiation. This is why Phoenix can say (9.523) with justification that Achilles' anger against Agamemnon has, up to that point, not been

νεμεσσητόν. Yet the implication of that phrase, with its pointed πρίν, is precisely that there comes a point at which even understandable retaliation becomes illegitimate; at every stage resolution of conflict is the ideal, and prolongation of a quarrel after

AFFRONTS AND QUARRELS

IN THE /LIAD

161

honourable compromise has become possible is occasion for νέμεσις in the same way as is provocation of a quarrel in the first place or an

excessive tendency to regard oneself as affronted. The major upshot of all this is that, behind the ideal that quarrels are to be avoided, lies a notion of legitimacy in one's title to honour. If such a thing exists, then clearly not all attempts to dishonour will meet with the validation of those who matter, i.e. the generality of ‘other people’. In any affront, the patient may feel himself dis-

honoured; but honour depends not on oneself, but on popular opinion,

on the audience,

and

the audience may side with either

party, with neither, or may reserve judgement, depending on the circumstances. τιμή is awarded by the group as a whole, and the group will not award τιμή for an action which it does not regard as valuable (and clearly, society will not regard disruption of its own

rules as valuable).2* Accordingly, when the view is expressed that *anyone' would experience νέμεσις at αἴσχεα, this must exclude that recognition of the success of the agent and failure of the patient which is required if the one is to win τιμή and the other to be dishonoured. An audience which regards one's conduct as νεμεσσητόν cannot with consistency reward that conduct with τιμή; and

the general attitude is that to attempt to increase one's honour at the expense of another member of the group is occasion for νέμεσις." This idea of a legitimate title to honour entails a concept of fairness

in the recognition of others' rights or entitlements.? This means that itis by no means misplaced to talk about a concept of ‘justice’ in this context, and so Lloyd-Jones is right to suggest that such a concept is

operative in Nestor's remarks in Zliad 1." What needs to be stressed is that this concept of justice is part of the code of honour, not in any

way extraneous or added to it. In later literature, this idea of ‘justice in honour' is not separable from that of justice simpliciter, no δίκη-

word occurs in Nestor's evaluation of the quarrel in Jliad 1,?* but in Hesiod, Tyrtaeus and Theognis δίκη frequently subsumes the requirement that legitimate claims to honour be recognised.?? This application of δίκη may also be traced in classical literature (for example, Odysseus' recognition and validation of Ajax's τιμή in Sophocles' Ajax is presented in terms of the δίκη which recognises

the right to honour),’ and lies behind those elements of Athenian law which demonstrate that that society takes considerable care to

prevent illegitimate acts of dishonour among its members (witness the laws against seduction,

an offence against the honour of the

woman's κύριος, and against ὕβρις, a concept which has a regular,

162

DOUGLAS

L. CAIRNS

some would say a necessary, reference to dishonour). Even though, then, there is no mention of δίκη in the passages of the liad discussed above, still those passages do exemplify the requirement that legitimate claims to honour be recognised, which later Greeks

describe quite regularly in terms of δίκη. The relevant ‘rights’ may well be conferred on the basis of one's power or abilities, or of one's membership of some specific group or relationship; but, as legitimate expectations backed up by the social sanction of disapproval, they

are clearly the pre- or proto-legal forerunners of the kind of rights which may be established by statutory law,?! and the concept of τιμή contributes to this implicit concept of rights in much the same way as it does in Classical Athens, where τιμή quite regularly denotes the basic level of entitlement under the law that is enjoyed by all citizens

alike. These observations entail the rejection of several of the most cherished and widely held assumptions about the nature of Homeric society: (1) The model of the ‘zero-sum’ game — the idea that honour is a commodity which passes inevitably from one side to the other in affront — is no longer applicable. If I commit an act by which I intend to dishonour you, and am myself dishonoured by doing so, then there is no simple exchange of honour; to dishonour another is

not always to win honour for oneself.? The importance of the audience of ‘other people’ in validating or refusing to validate a claim to honour in itself renders the zero-sum view untenable. (2) No sharp dichotomy exists between competitive

and

co-

operative values. The code of honour is inclusive, and the protectiveness towards one's own honour which promotes self-assertion also entails a sensitivity to the vulnerability of one's honour at the point at which self-assertion, by violating the claims of others, becomes dishonourable for oneself. This inclusiveness we saw in the usage of αἶσχος and αἴσχεα (since my αἴσχεα intended to dishonour you may in fact be ugly for me), and it is also apparent in the usage of αἰδέομαι, which signifies concern both for one's own and for others'

honour.?? (3) Honour is often combined with shame, and the two concepts are indeed correlates. Homeric society, moreover, is one which is thoroughly permeated by profound attachment to standards of the honourable. But this does not make that society a 'shame-culture', for a shame-culture is one in which fear of the external sanction of disapproval takes the place of concern for the intrinsic character of

AFFRONTS AND QUARRELS

IN THE /LJAD

163

one's actions,” and we have seen that Homeric characters are capable of explicit recognition that the values under which they live

are universalisable, capable of being personally endorsed by individual members

of society.

If I can point out that any impartial

individual would feel νέμεσις at a certain course of action, if I can

argue that you too would feel νέμεσις were another to act as you do, if I can feel νέμεσις at my own conduct or reject conduct because it is of the sort at which I should feel νέμεσις, then I acknowledge that individuals can endorse, appropriate, and internalise the values of

their society, and so it is wrong to suggest that Homeric man simply conforms

to external

standards

out

of fear of punishment

or

disgrace.** Examination of the attitude to affronts and quarrels, then, and of the terms which refer to such situations, can alert us to quite major features of Homeric values.

NOTES 1.

Τῆς following paper developed my

book

Aidös:

from further reflection on material discussed in

The Psychology and Ethics of Honour and Shame

in Greek

Literature (Oxford 1992), and was presented at the Leeds International Latin Seminar colloquium on ‘Enmity, Envy, and Insult in Ancient Poetry’, 3 May

1991. Iam grateful to several of those present, and especially to the colloquium director, Professor M. Dickie, for helpful comments and discussion. 2.

See the exchange of views between Gagarin, Lloyd-Jones, and Adkins in CP 82

3.

For this distinction, and those discussed below, see A.W.H. Adkins Merit and Responsibility (Oxford 1960) c.3.

4.

αἰσχύνη occurs first at Thgn. 1272; for the classical, physical sense of aloyoc, see

(1987).

(e.g.) Pl. Symp.

201a10.

5.

Cf. IL 13.622.

6.

Eg.theatoysa about Paris which so disturb his brother (//. 6.524), and to which his wife wishes he would pay attention (6.351). Cf. 3.242; Od. 19.373.

7.

Adkins (n.3) 42.

8.

See A.A. Long ‘Morals and values in Homer’ JHS 90 (1970) 130f.

9.

On the view of Fisher and Cantarella ὕβρις has a necessary connection with dishonour, but even on the opposing view of MacDowell there is still a frequent reference to dishonour. See N.R.E. Fisher ‘Hybris and dishonour' GR 23 (1976) 177-93 and 26 (1979) 32-47; id. Hybris: a Study in the Values of Honour and Shame in Ancient Greece (Warminster 1992); E. Cantarella ‘Spunti di riflessione

critica su ὕβρις e τιμή in Omero' Symposium 1979. Actesdu iv€ colloque intern. de

droit grec et hell. (Athens 1981) 85-96 = “Studi sul lessico giuridico greco: ὕβρις in Omero' Incontri Linguistici 7 (1982) 19-30; D. M. MacDowell ‘Hybris in Athens’ GR 23 (1976) 14-31; id. (ed.) Demosthenes: Against Meidias (Oxford 1990) 18-23.

164

10.

DOUGLAS

L. CAIRNS

So Long (n.8) 131, comparing Od. 2.85f. (Antinous complains that Telemachus’ charges are calculated to αἰσχύνειν the suitors); cf. below, n.26. One might also compare Od. 11.433: by her extreme disloyalty to her husband Clytemnestra has

inflicted αἶσχος on her entire sex; she has inflicted dishonour on her husband, but she is herself dishonoured in the eyes of society. This observation, however,

does not conflict with Adkins’ thesis, for he allows that co-operative failure may

be disgraceful for women, while denying that it can be so for men ((n.3] 36f., 45);

but in fact the possibility he allows in the case of women also extends to men, and the relevant value-terms can behave identically in the context of both male and

female values. HH.

αἰσχροῖς ἐπέεσσιν, Il. 3.38; 6.325; 13.768; ἔπεσσ᾽ aloypoiow, 24.238.

12.

See I]. 1.519; 2.277; 16.628; 21.480; Od. 18.326; cf. I. 22.497 (ὀνειδείοισιν); also Il. 21.393, 471 (ὀνείδειον μῦθον).

13.

See J.-C. Riedinger ‘Les deux αἰδώς chez Homére' RPh. 54 (1980) 70-5, 76.

14.

See Od. 6.286: 15.69-71 (“I too would feel νέμεσις); with the present passage (//. 23.494, “You too would feel νέμεσις᾽") cf. 6.329f. (“You too would fight, fall out

with anyone else who did such things"). On the significance of these passages, cf.

M.W. Dickie 'Diké as a moral term in Homer and Hesiod’ CP 73 (1978) 94; I.M. Hohendahl-Zoetelief Manners in the Homeric Epic (Leiden 1980) 11-13. Od. 1.228f. (discussed above) is also relevant, since there νέμεσις is described as the likely reaction of any sensible witness. vépeoic-verbs, too, can be reflexive, and so denote one's own reaction to a (potential) breach of αἰδώς in oneself (//. 16.544; 17.254; Od. 4.158f., etc.); the relevance of this to Democritus’ ἑωυτὸν αἰδεῖσθαι is noted by J.F. Procopé ‘Democritus on Politics and the Care of the

Soul' CQ 39 (1989) 323.

15.

E.g. νέμεσις is the reaction of the other gods to a god who becomes involved in

human strife (JJ. 5.757. 872), because this involvement indicates over-eagerness

to quarrel. Similarly, the expression πρὶν δ᾽ οὔ tt νεμεσσητὸν κεχολῶσθαι (Il. 9.523; Od. 22.59) indicates that there are occasions when anger at an affront is νεμεσσητόν. Cf. Riedinger (n.13) 72 (and 72-5 passim on the particular reference of νέμεσις to violent and quarrelsome behaviour). 16.

On the two uses of the verb αἰδέομαι with a (personal) accusative in Homer, see Riedinger (n.13); also Cairns (n.1) 2-4. Riedinger's excellent article is the best account of αἰδώς in Homer known to me, but also worthy of mention are: C.E. von Erffa αἰδώς und verwandte Begriffe in ihrer Entwicklung von Homer bis Demokrit (Philologus Suppl. 30.2, 1937); W.J. Verdenius "αἰδώς bei Homer’

Mnem.

3.12 (1944-5) 47-60:

A. Beil "αἰδώς bei Homer’

Der altsprachliche

Unterricht 5.1 (1961) 51-64: A. Cheyns Senses et valeurs du mot aidos dans les contextes homériques (Trav. Fac. de Philos. et Lettres, Louvain, Sect. de Philol. Class., Rech. de Philol. et Ling. 1, 1967); J.M. Redfield Nature and Culture in the Iliad (Chicago 1975) 113-19; I cannot recommend M. Scott *Aidés and Nemesis in the works of Homer, and their relevance to social or co-operative values’ Act.

Class. 23 (1980) 1-14, but others may disagree.

17.

The connexion between αἰδώς and φιλότης is fundamental, already apparent in

the frequent combination of αἰδοῖος and φίλος. The links between τιμή, φιλότης

and αἰδώς are well brought out by Riedinger in ‘Remarques sur la τιμή chez

Homere’ REG 89 (1976) 244-64, and (η. 13); cf. E. Benveniste Indo-European

Language and Society (ET London 1973) 277f. See further Cairns (n.1) 87-119. For νέμεσις at failure to show αἰδώς for one's φίλοι see (e.g.) J. 17.91-5; Od.

2.136f. (the νέμεσις Telemachus fears here is that which his αἰδώς forestalls at 20.343).

18,

See //. 21.468 (Apollo and Poseidon); Od. 6.329f. (Athena and Poseidon).

AFFRONTS

19.

AND QUARRELS

IN THE

/LIAD

165

For copious evidence, see H.J. Kakridis La Notion de l'amitié et de l'hospitalité chez Homére (Thessaloniki 1963) 56-8 and passim; cf. H. van Wees Status Warriors: War. Violence and Society in Homer and History

(Amsterdam 1992) 48,

337 n.80. Thus the statement of Adkins, ‘‘‘Friendship” and “Self-Sufficiency” in Homer and Aristotle' CQ 13(1963) 37, that "Diomedes is far more closely bound to a Lycian who is his philos (Jl. 6.119-236] than to a Greek who is not, even during the Trojan War" requires the important qualification that, unless an active state of enmity exists between Diomedes and another of the Greek leaders, all the Greeks at Troy are his φίλοι.

20.

See the remarks of Ajax to Achilles in // 9, esp. 630f.: we honoured you with our φιλότης; 640-2: we are your φίλοι and your guests, so show αἰδώς for us and for the obligations which exist between us.

21.

See, e.g., Il. 10.114f.: Nestor foresees νέμεσις on the part of Agamemnon should he charge Menelaus with dereliction of duty; cf. 129f., where Nestor's denial that anyone will feel νέμεσις at Menelaus' orders in the present instance presupposes

that brusque commands might otherwise be taken as insults. Cf. the frequent

precaution μὴ νεμέσα (Il. 10.145; Zoetelief (n.14) 22-4.

22.

15.115; 16.22; Od. 23.213); see Hohendahl-

See H. Lloyd-Jones The Justice of Zeus (Berkeley 1971) 12f.; cf. M. Schofield ‘Euboulia in the Iliad’ CQ 36 (1986) 28; also Long (n.8) 127; Riedinger (n.17) 260f. and (n.13) 73f. To claim, as do Adkins (n.3) 37f. and (n.19) 8f., and M.M.

Mackenzie Plato on Punishment (Berkeley 1981) 72-5. that Nestor advances only

prudential arguments, is to misrepresent the situation, as Schofield points out. For additional discussion, largely centring on Adkins’ interpretation of the

phrase ἀγαθός περ ἑών in 275, see also K.J. Dover ‘The portrayal of moral evaluation

Homer' CP

23.

in Greek

poetry’ JHS

82 (1987) 285-7. 303-5.

103 (1983) 37f.; M. Gagarin

‘Morality in

See 9.104-11 (Nestor; note the similarity to Thersites' criticism in 2.239f.), 13.107-15 (Poseidon, implying criticism among the Greeks), 14.49-51 (Aga-

memnon himself) and 19.181-3 (echoing Agamemnon's recognition of his own

fault at 2.375-8) with Riedinger (n.13) 73f., and O. Taplin *Agamemnon’s role in the /liad', in C.B.R. Pelling (ed.) Characterization and Individuality in Greek Literature (Oxford 1990) 74.

24.

The importance of the community in awarding τιμή is well stressed by van Wees (n.19) 61-165 passim; for some criticisms (along the same lines as the argument of this paper) see my review in CR 43 (1993) 5-9.

25.

The failure to see that terms of disapproval such as αἰσχρόν cannot be considered without the closest reference to the concept of νέμεσις (and that αἰσχρόν etc. must therefore also be related to αἰδώς, in the sense that any action inhibited by αἰδώς must be regarded by the αἰδούμενος as αἰσχρόν), is one of the greatest weaknesses of Adkins' approach, a weakness which lies in a surprising unwillingness to consider Homeric values as a whole.

26.

The notion of legitimacy in one's title to honour also explains why the victims of an affront tend to draw attention to the fact, to appeal to the impartial judgement of witnesses and to complain at the conduct of the perpetrator (see Riedinger

[n.13] 76); the indignation of Menelaus at the disfiguring (αἰσχύνειν) of his ἀρετή by Antilochus (//. 23.570-85) and of Antinous at what he regards as unjust criticism on the part of Telemachus (criticism which he feels is capable of bringing disgrace on [αἰσχύνειν]) himself and his fellow suitors, Od. 2.85f.; cf. above, n.10) illustrates an appeal to a standard which categorises certain forms of

affront as illegitimate; cf. //. 13.622f. Note that, in 11 23, Antilochus, when he cools down, is immediately impressed by the force of Menelaus' (strikingly ‘legalistic’) arguments; when his judgement is no longer clouded by emotional involvement and the excitement of his apparent success, therefore, he accepts

166

DOUGLAS

L. CAIRNS

that Menelaus’ appeal to impartial standards of fairness is justified, that in his youthful impetuosity he had failed to observe the norms of accepted conduct; cf. A. Schmitt Selbständigkeit und Abhängigkeit menschlichen Handelns bei Homer (Abh. Mainz 1990.5) 207-11.

27.

Lloyd-Jones (n.22) 13; cf. Schofield (n.22) 28. Note,

however,

that at

19.181

Odysseus describes Agamemnon's

conduct

towards Achilles as a failure to be δίκαιος. I see no reason not to take this remark at its face value.

29.

See Hes. WD 190-4, 213-18, 237f., 327-34 (discussed in Cairns [n.1] 152-6); Tyrt.

12.37-42 W (ibid. 163); Thgn. 27-30, 292 (ibid. 172). The view of M. Gagarin,

*Diké in the Works and Days’ CP 68 (1973) 81-94, and 'Diké in archaic Greek

thought’ CP 69 (1974) 186-97, that δίκη refers, exclusively in Hesiod and usually in the archaic poets, to aspects of the legal process, and that the entitlements involved in the process of δίκῃ are exclusively economic, is frequently disproved by the evidence. Cf. D.B. Claus ‘Defining moral terms in Works and Days’ TAPA 107 (1977) 73-84; Dickie (n.14). Gagarin tacitly modifies his former account of

δίκη in Hesiod in his Early Greek Law (Berkeley 1986) 46-9, especially 49.

30.

See Aj. 1332-45, and cf. R.P. Winnington-Ingram Sophocles: an Interpretation (Cambridge 1980) 66. 1 discuss this point more fully in (n.1) 238-40.

31.

That αἰδώς, the emotion which recognises legitimate claims to τιμή in others, implies a sense of obligation which may be regarded as a precursor to a juridical mode of thought is the position of L. Gernet 'Law and prelaw in Ancient Greece", in The Anthropology of Ancient Greece (ET Baltimore 1981) 147-9.

32.

For the ‘zero-sum’ view see

A. W.H. Adkins B/CS 7 (1960) 31: A.W. Gouldner

Enter Plato (London 1965) 49-51; Redfield (n. 16) 33; Mackenzie (n.22) 75; L.B. Carter The Quiet Athenian (Oxford 1986) 5f.; J.J. Winkler The Constraints of Desire (London

1990) 47. The

existence of a term such as ὕβρις, denoting

illegitimate dishonouring of another person, already demonstrates the falsity of

the thesis; the honour of the illegitimately dishonoured does not pass to the agent of the affront (consider the case of Aegisthus, who dishonours a ‘‘much better

man", Od. 3.248-50, 258-66, but who, at least among the impartial, gains no credit as a result). Where, however, individual A commits his honour to the achievement of a specific goal, then clearly achievement of the same goal by

individual B will mean that B will win honour and A will lose. This is why

Achilles’ remarks to Patroclus at //. 16.83-90, to the effect that success for Patroclus against the Trojans would diminish Achilles' honour, seem to bear out the zero-sum view. This view, however, is definitely false: to dishonour another is

not necessarily to win honour for oneself, nor need honour be won at the expense of another member of the group — at Od. 1.95 it is Athena's purpose that

Telemachus should win κλέος on his journey to Pylos and Sparta; in the episodes which follow, this acquisition of κλέος involves his general growth in heroic stature, not just the amassing of guest-gifts, and there is no one who suffers a loss

corresponding to Telemachus' gain. The economic metaphor of the zero-sum game belongs with a view of honour as a commadity invested in scarce material goods; that such a model is inapplicable to Homer is demonstrated by van Wees

(n.19) 69-77 and 349 n.30. 33.

This last point is well brought out by Riedinger (n.13). On the inclusivity of the Homeric code of honour, see also id. (n.17) and Schofield (n.22), especially 17f.

34.

For this basic premiss of the shame-culture/guilt-culture antithesis sce: M. Mead (ed.) Co-operation and Competition among Primitive Peoples (New York 1937)

esp. 493f.; ‘Social change and cultural surrogates’, in C. Kluckhohn and H.A. Murray

511-22;

(edd.) Personality in Nature,

'Some

anthropological

Society,

and Culture

considerations

concerning

(New York

guilt',

1948)

in M.L.

AFFRONTS AND QUARRELS

IN THE /LIAD

167

Reymert (ed.) Feelings and Emotions: The Mooseheart Symposium (New York 1950) 362-73; R. Benedict The Chrysanthemum and the Sword (London

1947) esp.

221-3; and D. Leighton and C. Ktuckhohn Children of the People: The Navaho Individual and his Development (Cambridge MA 1948) esp. 104-7. I criticise these views extensively in (n.1) 27-47, drawing on earlier criticisms by G. Piers and M.B. Singer Shame and Guilt: A Psychoanalytic anda Cultural Study (Springfield IL 1953) and D.P. Ausubel ‘Relationships between shame and guilt in the socializing process’ Psych. Rev. 62 (1955) 379-90. For general criticism of the cultural determinism which lies behind the approach of Mead and the rest, see D. Freeman Margaret Mead and Samoa: The Making and Unmaking of an Anthropological Myth (Canberra 1983). 35.

Cf. Dickie (n.14) 94f.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 169-99 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

ANCIENT INTERPRETATIONS OF PINDAR’S NEMEAN 7 MALCOLM

HEATH

Pindar’s Nemean 7 was written to celebrate the victory of Sogenes son of Thearion of Aegina in the boys’ class of the pentathlon at the Nemean games, date unknown. It accords a prominent place to Neoptolemus, a hero who was (like the other Aeacids) of great importance to the Aeginetans; the myth narrates his death at Delphi, a theme also handled in Paean 6. Ancient scholars believed that there was a connection between these two poems. It was thought that the Aeginetans had been offended by an uncomplimentary portrayal of their hero in Paean 6, and that Pindar consequently incorporated into Nemean 7 a defence on the charge of having slandered Neoptolemus. My primary aim in this paper is to understand the scholia where this theory is propounded, rather than to pronounce on its correctness. Understanding is obviously a precondition of informed evaluation, and the barriers to an adequate understanding

of ancient interpretations of Pindar have not yet been fully overcome. In this paper I shall be concerned in particular with the unfamiliar literary and rhetorical assumptions which informed the ancient

interpreters! work.! The first section sets out the evidence, with commentary.? Section (2) looks at some of the general assumptions which would have informed ancient commentators' approach to Pindar's poetry. In section (3) I try to synthesise this material into a conjectural history of the interpretations of the poem in antiquity.

169

170

MALCOLM

HEATH

1. The evidence

la (116.8-117.18): ζητεῖται διατί ἀπὸ τῆς Εἰλειθυίας εἰσβέβληκε, Kai τί δήποτε τῇ Εἰλειθυίᾳ προσδιαλέγεται. ἔνιοι μὲν οὖν φασι νέον ὄντα τὸν Σωγένην ἑτέρου νικήσαντος αὐτὸν κατὰ χάριν ἀνηγγέλθαι τοῦ πατρὸς Θεαρίωνος εἰς τοῦτο φιλοτιμηθέντος, τὸν δὲ ἀγωνισάμενον Νεοπτόλεμον τοὔνομα ᾿Αχαιόν' παρὸ καὶ εἰς τοὺς ὑπὲρ Νεοπτολέμου τοῦ Ἀχιλλέως τὸν ποιητὴν παρεκβῆναι λόγους. αὐτοσχέδιον δέ φησι τοῦτο εἶναι ὁ Δίδυμος οὐδεὶς γὰρ ἀφ᾽ ἑτέρου ὀνόματος κηρύττεται, μόνον δὲ ἐν τοῖς ἱππικοῖς ἀγῶσι νενόμισται τὸ ἔθος ... ἔνιοι δέ φασι πρὸς τοὔνομα τοῦ Σωγένους παρειλκύσθαι τὴν Εἰλείθυιαν" εἶναι γὰρ αὐτὴν σωγενῆ nva διὰ τὸ τὰ γεννώμενα ἀνασῴζειν: τὸν δὲ Πίνδαρον ψυχρευσάμενον πρὸς τοὔνομα τῆς Εἰλειθυίας μεμνῆσθαι. καὶ τοῦτο δὲ οὐκ eb τότε γὰρ καταφέρεται εἰς τοῦτο ὁ Πίνδαρος, ὅταν ὑπῇ τις ὁμωνυμία

... νῦν δὲ οὐδὲν τοιοῦτόν ἐστι.

ἔνιοι δὲ τὸν πατέρα τοῦ Σωγένους Θεαρίωνα τῇ Εἰλειθυίᾳ ἔφασαν ἱερᾶσθαι’ ἀμάρτυρον δὲ καὶ τοῦτο. οἱ δὲ ὅτι ἐν γειτόνων ἦν τῷ Σωγένει ἱερὸν Εἰλειθυίας" οὐδὲ τοῦτο δὲ ἱστορεῖται. ἄλλοι δὲ ἔφασαν ὅτι τοὺς ἐκ φύσεως ἀγαθοὺς ἐπαινεῖ πάντοτε ὁ Πίνδαρος μᾶλλον τῶν ἐκ διδασκαλίας" καὶ ἡ Εἰλείθυια οὖν τὸν Σωγένην εὐθέως ὑπὸ τὴν γένεσιν ἐπιτήδειον κατεσκεύασε πρὸς ἄθλησιν. οἱ μὲν γὰρ ἄλλοι τῶν θεῶν αὐξανομένοις ἀγαθὰ δωροῦνται, ἡ δὲ Εἰλείθυια αὐτὴν τὴν πρώτην καταβολήν. ἔστι δὲ καὶ πρὸς τοῦτο λέγειν' τί δήποτε τὴν Εἰλείθυιαν ἰδίως ἐπὶ τοῦ Σωγένους παρέλαβεν, ἀλλ᾽ οὐχὶ καὶ ἐπ᾽ ἄλλων εὐφυΐᾳ διενεγκόντων; Apıotöönpog δὲ ὁ ᾿Αριστάρχου μαθητὴς βέλτιον οὕτω φησίν’ ὀψέ ποτε τῷ Θεαρίωνι

καὶ παρὰ

τὴν ἡλικίαν ἤδη προήκοντι, εὐξαμένῳ τῇ

θεῷ Σωγένην τεκνωθῆναι, καὶ τὴν τοῦ παιδὸς αὐτοῦ γέννησιν οἷον Εἰλειθυίας εἶναι χάριν. διὰ τὴν ἰδιότητα οὖν τῆς γενέσεως τοῦ ἀθλητοῦ πρὸς τὴν θεὸν ταύτην ἐπήρεισε τὸν λόγον. ἐπιστοῦτο δὲ τοῦτο ἐξ ἐπιγράμματος Σιμωνίδου. (There is a question why Pindar begins from Eileithyia, and why he addresses himself to Eileithyia:

(1) Some say that because of his youth Sogenes was proclaimed victor himself although someone else won the victory, to satisfy the ambition of his father Thearion, and that the competitor was an Achaean named Neoptolemus; this is also the reason why the poet digresses into what is said in defence of Neoptolemus, Achilles’ son. Didymus objects that this is makeshift, since no one is proclaimed victor on someone else's account; this

is customary only in equestrian events ... (2) Some say that Eileithyia is dragged in with a view to Sogenes' name, since she is herself a 'Sogenes', because she keeps the new-born safe; so, by a lapse of taste, Pindar mentions Eileithyia with a view to his name. But this

too is unsatisfactory; Pindar only descends to this when there is some underlying similarity in the names ... But there is nothing of the kind here. (3) Some say that Sogenes' father Thearion was a priest of Eileithyia; but

ANCIENT INTERPRETATIONS OF PINDAR'S NEMEAN 7

171

there is no record of this. (4) Others say that there was a temple of Eileithyia in Sogenes’ neighbourhood; but this too is unattested. (5) Others say that Pindar consistently praises people of innate excellence above those who have acquired it; and so Eileithyia ensured Sogenes’ aptitude for athletics at birth. Other gods bestow good gifts on people asthey

mature, but Eileithyia provides the initial foundation. But this too is open to objection: why did Pindar introduce Eileithyia specifically in the case of Sogenes,

and

not

also

in the case

of others

with outstanding

natural

endowments? (6) Aristarchus' pupil Aristodemus has a better suggestion: Sogenes was born late, when Thearion had already passed the normal age and after he had

prayed to the goddess; his son's birth was thus as it were a gift of Eileithyia. So because of this peculiarity of the athlete's birth Pindar addressed his discourse to this goddess. He confirmed this from an epigram of Simonides.)

This long scholion reviews six explanations for the poem's opening invocation of Eileithyia; reasons are given for rejecting the first five, and the last is approved. The compiler is probably Didymus

(116.15).? The

first theory

mentioned

is that

Sogenes

was

young;

the

implication is that Eileithyia is invoked because she has care of the young, a solution which has also appealed to some modern scholars.* Here, however, the idea is elaborated with a further speculation: because Sogenes was young (presumably, too young to compete) the victory was in fact won by someone else, who had Sogenes proclaimed in his place to gratify the father Thearion; the proxy victor was called Neoptolemus, which helps to explain the hero's

prominence in the poem.? The theory is attributed to Aristarchus in 56a, 124.13-15.5 The implausibility of Aristarchus' substitute competitor (perhaps inspired by Cimon's gesture to Peisistratus, recorded in Herodotus 6.103) was pointed out by Didymus; it was presumably already clear to Aristarchus' pupil Aristodemus, whose alternative identification of the hypothetical Neoptolemus is given in 56a, 124.16-18 (see below). One consequence of Aristodemus' modification of the theory is that Sogenes is recognised as a competitor in his own right; his youth therefore ceases to be a distinctive factor, and Aristarchus' explanation of the address to Eileithyia collapses." Aristodemus had therefore to propose another explanation for this address. His theory is the last mentioned in this review: Sogenes was born late in his father's life after prayer to the goddess, so that his birth was “as it were a gift of Eileithyia". Aristodemus believed that his account of Sogenes' birth was confirmed by an epigram of Simonides (166

172

MALCOLM

HEATH

Bergk*). The epigram is not quoted, and most scholars have supposed that no other trace of it survives; but David Young has proposed an identification.! According to Young, Aristodemus' conjecture was prompted by the name Sogenes itself, together with the fact of Eileithyia's invocation; and Aristodemus cited an extant epigram “not ... as a historical source for the biography of Sogenes, but as a literary parallel for the kind of onomastic pun which he has just attributed to Pindar". The epigram in question is Simonides 168 Bergk*: Σῶσος kai Σωσώ, Σῶτερ, σοὶ τόνδ᾽ ἀνέθηκαν, Σῶσος μὲν σωθείς, Σωσὼ δ᾽ ὅτι Σῶσος ἐσώθη.

If Young’s theory is right, Aristodemus emerges with little credit. This epigram has, as Young points out, absolutely no value as evidence in the case of Sogenes. Didymus fares even worse. The second of the six theories discussed posits without biographical elaboration a connection between Sogenes’ name and theinvocation of the goddess; Didymus dismisses the pun as frigid and contrary to

Pindar's practice.” The next two theories are rejected for postulating unevidenced facts about Thearion and Sogenes. But Didymus proceeds to accept Aristodemus' theory, which (on Young's view) turns on an onomastic pun and an unevidenced claim about the victor's biography. The inconsistency could hardly be plainer. It is a tacit premise of Young's conjecture that we may attribute arbitrary flights of fancy to ancient Pindar scholars (a presumption reflected in his sniping epithets for Aristodemus: "this inventive man ... our dauntless scholar"). But even if Young's conjecture is right (which is far from certain), this premise is wrong in principle. No understanding of the Pindar scholia is possible unless we are prepared to take them seriously, and to interpret them with care and sympathy. It is true that, because of the complex history of their transmission,'^

the scholia as preserved

may

have been selected,

conflated and summarised into incoherence, or corrupted by copyists. But they are at worst the remains of something which at least one intelligent and educated man in antiquity found plausible, however grossly implausible they may seem to us; it is therefore pointless to interpret notes in the scholia as if they were the product of arbitrary irrationality. The ancient commentators may have been wrong, and

their errors may reflect profound and systematic misconceptions. But if what they say seems arbitrary or irrational then (faults of transmission apart) it is virtually certain that we have misunderstood

ANCIENT INTERPRETATIONS OF PINDAR'S NEMEAN 7

173

it, or that we have failed to grasp the background assumptions against which a reasonable person might have thought it a reasonable thing to say.!!

46b (122.2830): ὁ δὲ voüg τούτων δὲ τιμὴ γίνεται τεθνηκότων, dv ἂν θεὸς αὔξῃ λόγον, τουτέστιν ὁ ποιητὴς ἐμπνευσθεὶς ὑπὸ τῶν θεῶν. τινὲς δὲ τὸ τεθνακότων πρὸς τὸ ἑξῆς. (The sense is: honour accrues to those dead whose repute is enhanced by a god, i.e. by a poet divinely inspired. But some connect “dead” with what follows.)

47 (123.1-18): πρὸς τὸ £v ἀκουστέον ἄχρι τοῦ ἐπεὶ πράθε, καὶ γραπτέον ἐν τῷ ε ἔμολεν, ὡς Δίδυμος, ὃς καὶ ὑφ᾽ Ev ἀπὸ τοῦ τεθνακότων ἕως τοῦ πράθεν ἀναγινώσκει. τὸ δὲ

ὅλον

τοιοῦτον

ἔμολε

παρὰ

μέγαν

ὀμφαλὸν

εὐρυκόλπου

χθονὸς

Νεοπτόλεμος, Πριάμου πόλιν ἐπεὶ npáOe: τεθνηκότων δὴ τῶν βοηθῶν ἐν Πυθίοις δαπέδοις κεῖται. ποίων δὲ βοηθῶν; τῶν περὶ τὸν Εὐρύπυλον, oc αὐτὸς ὁ Νεοπτόλεμος ἀνελὼν ἐπόρθησε τὴν Ἴλιον. οὕτω γὰρ ἴδιος ὁ πόνος ἔσται τοῦ ἥρωος. δοκεῖ δὲ ὁ Νεοπτόλεμος ἐν Δελφοῖς ἀνῃρῆσθαι καὶ τεθεῖσθαι ἐν τῷ ναῷ. ἐὰν δὲ ἐπὶ τῶν περὶ τὸν Ἕκτορα, καθὼς 'Apiotapyóc φησιν, ἀναδράμωμεν, πρῶτον μὲν μακρόθεν ἔσται τὴν πόρθησιν συνἅπτων, δεύτερον δὲ ἐπὶ τὴν κοινότητα μεταβησόμεθα, δι᾽ ἧς τὸ ἐγκώμιον

οὐκ ἔσται. εἰ δὲ γράφεται ἔμολον, τῷ πληθυντικῷ ἀντὶ τοῦ ἑνικοῦ χρᾶται, ὥστε ὅμοιον εἶναι tH ὁ δὲ ἀποπλέων Σκύρου μὲν ἅμαρτεν, ἵκοντο δὲ εἰς Ἔφυραν, ἀντὶ τοῦ ἵκετο. εἰς δὲ τὸ προκείμενον τείνων παρέθηκε τὴν τοῦ Νεοπτολέμου ἀπώλειαν, ὅτι ἀδοκήτως αὐτῷ συνέβη. (One should read as a single unit as far as ἐπεὶ πράθε, and write ἔμολεν with an £ (thus Didymus, who also takes from τεθνακότων to πράθεν as a single unit). The overall sense is: Neoptolemus came to the great navel of the broadbosomed earth when he sacked Priam's city; after the death of the allies he

lies in the Pythian ground. What allies are these? The companions of Eurypylus, whom Neoptolemus himself killed before the sack of Troy; this makes the “labour” (36) specific to Neoptolemus. If we go back as far as Hector's companions, as Aristarchus suggests, then (a) the connection with

the sack of Troy will be far-fetched, and (b) we will be switching to a generalisation, which is not a constituent of encomium. If ÉuoXov is written, Pindar is using plural for singular; cf. “sailing away he missed Scyros, and they came to Ephyra" (36f.), instead of “he came". The death of Neoptolemus is included with reference to what precedes, because it befell him unexpectedly.)

The interpretation of 31ff. was disputed in ancient as in modern times. Some ancient readers punctuated after τεθνακότων (46b), as do modern editors; we know nothing of how they understood the

174

MALCOLM

HEATH

following sentence. Others punctuated after λόγον, taking tedvaκότων βοαθόων as a genitive absolute and reading either ἔμολεν or ἔμολον, construing the latter as a third-person plural. Slater

explains the plural variant thus:!? The ancient [sic] took boathoon as a plural genitive, and therefore some according to the scholia took toi as a relative instead of a particle. This caused them problems as the scholia say and it seems obvious to me that someone ... emended the text to get rid of the anomaly of a singular verb after a plural relative. But it cannot be true (nor do the scholia say) that tot was taken as a

relative, since the dead βοηθοί were identified as Trojans or allies of Troy, who certainly did not come to Delphi. Aristarchus saw a reference to Hector; Didymus thought the reference was to Eurypylus, pointing out that this would constitute a contextually apt allusion to

one of Neoptolemus' exploits. In fact, it was not the singular form which excited comment, but the plural verb where one would expect a singular reference to Neoptolemus; to show its consistency with

Pindaric usage the scholia cite the alternation of singular and plural in 36f.? The second objection raised against Aristarchus reflects the view that excessive generality is inappropriate in encomium; see Nicolaus Sophista Progymnasmata 111.480.16f. Spengel: ἀεὶ δεῖ σπεύδειν καὶ ἐπείγεσθαι ἐπὶ τὰ ἴδια xai μόνῳ αὐτῷ ὑπάρχοντα (cf. n.40 below). Note that the poem is viewed as an encomium (cf. 44b, 122.24f.; 46a,

122.27; 112-114b, 132.22-133.3); the implications of this important assumption are discussed in section (2) below. 56a (124.12-18): παρεκβαίνει δὲ εἰς tà περὶ Νεοπτολέμου, ὡς μὲν Ἀρίσταρχος ὅτι ὁ νικήσας καὶ χαρισάμενος τὴν ἀνάρρησιν τῷ Σωγένει Νεοπτόλεμος ἦν τοὔνομα, Ἠπειρώτης τὸ γένος, ὅπερ οὐχ οὕτω φησὶν ἔχειν ὁ Δίδυμος. ὡς δὲ ᾿Αριστόδημος ὅτι ἀλείπτῃ ἐκέχρητο τῷ Νεοπτολέμφ᾽ διὸ εἰς ἔπαινον τοῦ ὀνόματος τῇ ἡρωϊκῇ κέχρηται παρεκβάσει. (He digresses to the story of Neoptolemus, according to Aristarchus because the victor, who conceded the proclamation to Sogenes asa favour, was called Neoptolemus, an Epirote; but Didymus denies that this was the case.

According to Aristodemus, it is because Sogenes’ trainer was Neoptolemus; for this reason he uses the heroic digression to reflect glory on the name.)

The Neoptolemus myth is described as a digression (napéxBaotc) also in /a, 116.14, and 76, 127.7f.; see section (2a) below for the implications of this term. Aristarchus' hypothetical Neoptolemus, identified as an Achaean

ANCIENT INTERPRETATIONS OF PINDAR'S NEMEAN 7

175

at Ja, 116.13, is here described as an Epirote (for the equivalence see 94a, 128.23-129.1). Why should Aristarchus have supposed that his

hypothetical competitor came from Epirus? The only possible reason is that he identified him with the "Achaean man living above the Ionian sea” to whom Pindar refers in line 64. If we then ask why this hypothetical competitor should have found no fault with Pindar, we can see why Aristarchus believed that his name was Neoptolemus.

The hypothesis must be that Pindar has incorporated into a poem in praise of Sogenes, the proclaimed victor, an oblique compliment to Neoptolemus, the actual competitor; and he has done this by means

of the myth of the competitor's namesake, the hero Neoptolemus. Neoptolemus the competitor therefore finds no fault with Pindar (that is to say, he is positively pleased with him) just because his role in the victory has not been overlooked. Aristodemus' alternative identification of the hypothetical Neoptolemus as Sogenes' trainer makes it possible to discard the most glaringly implausible component of this interpretation while retaining its structure: the myth of Neoptolemus obliquely acknowledges a contemporary homonym's contribution to the victory. Aristodemus could have appealed to the appearance of the trainer in other odes (e.g. Olympian 8) for parallels.'‘ 62a (125.912): φασὶ τοῦ Νεοπτολέμου θύοντος τοὺς Δελφοὺς ἁρπάζειν τὰ θύματα, ὡς ἔθος αὐτοῖς" τὸν δὲ Νεοπτόλεμον δυσανασχέτως ἔχοντα διακωλύειν᾽ αὐτοὺς δὲ διαχρήσασθαι αὐτὸν ξίφη ἔχοντας. (They say that while Neoptolemus was sacrificingthe Delphians tried to seize the victims, in accordance with their custom; Neoptolemus resented this and

tried to prevent them; they killed him, being armed with swords.)

70 (126.8-23): ᾿Αρίσταρχος οὕτως᾽ ἐπεὶ μέμψις τοῖς Αἰγινήταις πρὸς τὸν ποιητὴν ἦν χάριν τοῦ Νεοπτολέμου, εἰς τὴν ἀπολογίαν τὴν περὶ Νεοπτόλεμον δικαίως διαρκέσει τρία ἔπεα, φησίν' ὅτι μόρσιμον ἦν αὐτῷ οὕτω τελευτῆσαι, καὶ ὅτι πεπρωμένον ἦν ἕνα τῶν Αἰακιδῶν ἀποθανόντα τεθῆναι ἐν τῷ ναῷ, τρίτον ὅτι ταῖς ἡρωΐαις πομπαῖς ἕνα τῶν Αἰακιδῶν θεμίσκοπον εἶναι. τρεῖς οὖν φησι λόγους ἱκανοὺς εἶναι εἰς τὴν εὐώνυμον δίκην᾽ οἷον ἱκανὰς εἶναι τρεῖς αἰτίας, ἃς ἐξαριθμεῖται, τοῦ δικαίως αὐτῷ συμβεβηκέναι τὸν θάνατον. ἕνιοι δὲ οὕτως ἀπολογοῦνται περὶ τῆς τοῦ Νεοπτολέμου τελευτῆς, τρία φέροντες ταῦτα, ὅτι τὸ ὑπὲρ τῶν κρεῶν εἰπεῖν ἀνῃρῆσθαι, καὶ ὅτι οἱ Δελφοὶ ἐβαρύνθησαν ἐπὶ τῇ ἀπωλείᾳ αὐτοῦ, καὶ ὅτι πεπρωμένον ἦν αὐτὸν ἀναιρεθῆναι. ὁ δὲ ᾿Αριστόδημος ἔπεα τρία τὰς τριάδας ἀκούει" τὴν γὰρ ἀρχὴν ἔπος προσηγόρευον τὴν στροφὴν καὶ τὴν ἀντιστροφὴν καὶ τὴν ἐπῳδόν᾽ τρεῖς δὲ

176

MALCOLM

HEATH

λοιπὸν ὑπεῖναι τῷ ἐπινίκῳ τριάδας. (According to Aristarchus: since the Aeginetans criticised the poet because of Neoptolemus, he says that three words will be enough for his defence concerning Neoptolemus, i.e. (1) that it was fated for him to die thus; (2) that

it was destined that one of Aeacus' descendants should be buried in the temple; and (3) that it was destined that one of Aeacus’ descendants should oversee the processions for the heroes. So he says that these three words are enough for justice, whose name is beautiful, i.e. three reasons, which he enumerates, are sufficient for death having befallen him justly. ᾿ς Some make the defence concerning Neoptolemus' death by adducing these three points: (1) that hesaid he was killed for the meat; (2) that the Delphians

were grieved at his death; (3) that it was fated for him to be killed. Aristodemus understands the three words as triads, on the grounds that strophe, antistrophe and epode were originally called *word', and there are three triads remaining for the victory-song.

Aristarchus found in the statement that "for justice whose name is beautiful, three words will be enough" (48) a pointer to the core of Pindar's apologia. Turning to the immediate context for a fuller explanation, he identified the three words as three statements in 44-7

which combine to show that death befell Neoptolemus “justly”: his death was fated (ἀλλὰ τὸ μόρσιμον ἀπέδωκεν); it was fated that an Aeacid should be buried at the sanctuary at Delphi (ἐχρῆν δὲ ... δόμον); and it was fated that one of the Aeacids should be θεμίσκοπος of the processions there (ἡροΐαις δὲ ... πολυθύτοις). To say that death befell Neoptolemus justly presumably does not mean that Neoptolemus deserved to die the death described in Paean 6 (a claim which would have compounded the original offence); rather,

Aristarchus sees in these lines evidence that the death described in the Paean was not such as to dishonour the hero. This would mean that those who, like the Aeginetans, honour the hero need not regard Pindar as a liar; Aristarchus presumably construed the next line as a summary of Pindar's vindication of himself as a truthful witness to Neoptolemus' deeds: οὐ ψεῦδις ὁ μάρτυς ἔργμασιν ἐπιστατεῖ." Aristodemus rejected Aristarchus' interpretation of the three words, perhaps because he realised that δέ in 44 must be explanatory rather than connective. His own solution is that ἔπος here means

*triad', and that the reference is to the last three triads of the poem. We know.that Aristodemus was interested in the length of poems, believing that payment was made πρὸς ἀριθμὸν ἐπῶν xai τριάδων (7. 1.85b, 210.6-11: cf. n.39 below). But there appears to be no parallel for the sense given to ἔπος, by which I am baffled. The

second,

unattributed

interpretation

returns

to something

more like Aristarchus' view; but by extending the relevant context

ANCIENT INTERPRETATIONS OF PINDAR'S NEMEAN 7

177

back to 42 it is able to identify a different set of three statements: that

Pindar had said Neoptolemus died fighting over the meat (ἵνα κρεῶν ... μαχαίρᾳ); that the Delphians regretted the death (βάρυνθεν δὲ ... ξεναγέται); and that it was fated (ἀλλὰ τὸ μόρσιμον ... ἐχρῆν δὲ ... πολυθύτοις: on this interpretation the whole of 44-7 is correctly recognised as a single unit). To determine the significance of the meat in the first of these statements we must turn to Paean 6. The text of the Paean available to ancient scholars had Neoptolemus killed in a fight μυριᾶν περὶ τιμᾶν (118). This opaque phrase is unlikely to have been what Pindar wrote," but the

corruption is an old one, since Zenodotus tried to emend it (proposing Πυθιᾶν). The scholia to the Paean propose two possible explanations of the “myriad honours”:!® ἤτοι τῶν κρεῶν ἃ διαρπαζόντων συνηθῶς τῶν Δελφῶν ἐδυσχέραινε καὶ ἐκώλυε, διὸ καὶ ἀνήρηται' ἢ τῶν χρημάτων ἃ διαρπάζων εἰς ἐκδικίαν τοῦ πατρὸς ἀνῃρέθη. (Either meat, which the Delphians seized after their custom, to Neoptolemus’ annoyance, so that he was killed because he tried to prevent them; or property which he was trying to plunder as reparations for his father's death when he was killed.)

On one view the reference is to sacrificial meat: the attendants at the temple at Delphi seized the meat of a sacrifice which Neoptolemus had offered; Neoptolemus angrily tried to prevent them, and was killed in the ensuing struggle.!? Alternatively, the myriad honours are the temple's treasures: on this interpretation, Neoptolemus had come

to Delphi to claim reparations for the death of his father Achilles at Apollo's hands. Given the existence of these two interpretations of the Paean in antiquity it is easy to see the significance attached to the reference to meat in Nemean 7.42. The theory must be that the Aeginetans had interpreted the reference to myriad honours in the Paean as meaning valuables, and resented the portrayal of Neoptolemus as a temple-robber (cf. 750a, 137.4-6, ἐπὶ ἱεροσυλίᾳ); hence in Nemean 7 Pindar asserts that he did not mean valuables, but

meat, so that it was not impiety that led to the fatal quarrel but Neoptolemus' sense of honour (see on 94a below). On this theory, Pindar defends himself by clarifying the implications of the myth in the Paean: he did not mean to imply that the actions which led to the

hero's death at Delphi were impious. The theory that the reference to meat is crucial to Pindar's defence is ascribed to Aristodemus in the scholion to 102 (750a, 137.7).?? But since Aristodemus' understanding of the “three words" has nothing

178

MALCOLM

HEATH

to do with Pindar’s self-defence, the anonymous interpretation cited

here is probably a later reconciliation of Aristodemus’ emphasis on the meat with a modified version of Aristarchus’ approach to the “three words”.

76 (127.7f.): βούλεται ἤδη λῆξαι τῆς περὶ Νεοπτόλεμον παρεκβάσεως. ἀναπαύσομαι ἤδη λέγων περὶ αὐτοῦ, ἐπεὶ κόρος ἔπεστι πᾶσι ... (He wants now to bring the digression about Neoptolemus to a close. I will

cease now to speak everything ...)

about

him, since there is a potential for excess in

For the Neoptolemus myth as a parekbasis cf. 56a, 124.12-18, and (2a) below. ἀνάπαυσις and κόρος are key terms in ancient criticism's

characteristic demand for variety.?! 94a (128.23-129.12): ᾿Αχαιὸν ἄνδρα τὸν Ἠπειρώτην ἢ αὐτὸν τὸν Νεοπτόλεμον fj ἕνα τῶν Ἠπειρωτῶν ... οὐ μέμψοιτο ἂν οὖν με 'Ayaióg ἀνὴρ ἐπὶ τῷ δοκεῖν μικρολόγον παρεστακέναι τὸν Νεοπτόλεμον, ὅτι εἶπον αὐτὸν περὶ κρεῶν μεμαχῆσθαι καὶ διὰ τοῦτο ἀπολωλέναι. καθόλου γὰρ ἀπολογεῖσθαι βούλεται περὶ τοῦ Νεοπτολέμου θανάτου πρὸς τοὺς Αἰγινήτας. ἐκεῖνοι γὰρ ἠτιῶντο τὸν Πίνδαρον ὅτι γράφων Δελφοῖς τὸν Παιᾶνα ἔφη᾽ ἀμφιπόλοισι μαρνάμενον μυριᾶν περὶ τιμᾶν ἀπολωλέναι. ἀπολογούμενος γάρ τι ἀντεισήγαγε τοῦτο ὁ Πίνδαρος, ὅτι οὐκ ἔφησε περὶ χρημάτων γεγονέναι τῷ

Νεοπτολέμῳ τὴν μάχην, ἀλλὰ περὶ τῶν νομιζομένων τιμῶν τοῖς Δελφοῖς. ὃ οὖν λέγει τοιοῦτόν ἐστι᾿ παρὼν δέ τις ἐνταῦθα Δωδωναῖος οὐ μέμψεταί μοι ὡς ἐνυβρίσαντι τῷ Νεοπτολέμῳ. (By "Achaean man" he means an Epirote, either Neoptolemus himself or someone from Epirus ... So an Achaean man will not find fault with me for seeming to represent Neoptolemus as petty-minded, because I said that he fought over meat and died for that reason. For Pindar's general intention is to defend himself before the Aeginetans with regard to Neoptolemus' death. They criticised him because in a Paean written for the Delphians he said that Neoptolemus died fighting the attendants "about a myriad honours". By way of defence he makes this reply, that he did not say that Neoptolemus was involved in a fight about property, but about the honours customary at Delphi. So what he says is like this: a man from Dodona, if present here, will

find no fault with me for insulting Neoptolemus.)

Pindar's reference in 64f. to an “Achaean man living above the Ionian sea" who “finds no fault” with the poet is here linked to the

apologia theory. The Achaean man (Neoptolemus himself, or some inhabitant of Epirus with a sense of loyalty to the hero as the mythical founder of his homeland — an aspect of Neoptolemus’ career

to which

Pindar refers in 38f.) will not blame

Pindar for

ANCIENT INTERPRETATIONS OF PINDAR'S NEMEAN 7

179

representing Neoptolemus as “‘petty-minded”; a fortiori there is no

reason why the Aeginetans should take offence. The underlying assumption seems to be that Pindar faced a dilemma: to rescue the hero from the imputation of temple-robbery by invoking the meat-

motif was to expose him to the equally offensive imputation of pettymindedness. This note does not specify how the objection is to be met,? but the reference to φιλοτιμία in the note to 102 (150a, 137.7) supplies the deficiency: the Delphians' attempt to seize the meat was

an affront to Neoptolemus' sense of honour, which he had reason to resist.

Some scholars attribute this view to Aristarchus? But we have seen that he identified the Achaean man with the hypothetical substitute competitor, who finds no fault with Pindar because his contribution to the victory has been recognised; the mention of fault-

finding would therefore have no relation to Pindar's offence to the Aeginetans (see on 56a). Moreover, the meat-motif played no role in Aristarchus' analysis of Pindar's self-defence (see on 70). The issues addressed in this scholion would therefore not have arisen for Aristarchus. Aristodemus is at first sight a more promising candidate. He did make use of the meat-motif and he invoked Neoptolemus' honour, as we learn from the scholion to 102 (150a, 137.7). However, Aristodemus retained in a modified form Aristarchus' theory that the

Neoptolemus myth was obliquely complimentary to a man called Neoptolemus who was involved (in one way or another) in Sogenes' victory (see on 56a). This theory identifies the hypothetical Neo-

ptolemus with the Achaean man of 64; but the note we are now considering proposes a completely different identification of the Achaean man as either the hero Neoptolemus himself, or an inhabitant of Epirus loyal to his memory. The scholar who was

responsible for this re-identification of the Achaean man abolished the only possible grounds for believing in the existence of the hypothetical Neoptolemus.

hypothetical

Neoptolemus,

Since Aristodemus still believed in the

we

should

not (without compelling

reason) attribute the re-identification to him. The note on 64 is therefore probably derived from some other, unidentified and unidentifiable, scholar. Since 70, 126.16-20, proves the existence of an anonymous modifier of Aristarchus! views

distinct from economy.

Aristodemus,

this conclusion

involves no lack of

180

MALCOLM

HEATH

150a (136.24-137.13): οὐκ ἀτρόπως φησὶν οὐδ᾽ ἀπεοικότως εἰς μνήμην ἐληλυθέναι τοῦ Neoπτολέμου. ὁ μὲν οὖν Καλλίστρατος, ἐπεὶ ἔφησε τὸν Σωγένη γείτονα εἶναι τοῦ Ἡρακλέους, καὶ ὁ Νεοπτόλεμος δὲ γειτνιᾷ τῷ θεῷ τῷ ἐν Δελφοῖς, κατὰ τοῦτό φησιν οἰκείως μεμνῆσθαι τοῦ Νεοπτολέμου, διὰ τὴν ἀμφοτέρων γειτνίασιν. ὁ δὲ ᾿Αριστόδημος, ὅτι μεμφθεὶς ὑπὸ Αἰγινητῶν ἐπὶ τῷ δοκεῖν ἐν Παιᾶσιν εἰπεῖν τὸν Νεοπτόλεμον ἐπὶ ἱεροσυλίᾳ ἐληλυθέναι εἰς Δελφούς, νῦν ὥσπερ ὑπεραπολογεῖται εἰπὼν ὅτι οὐχ ἱεροσυλῶν ἐτελεύτησεν, ἀλλ᾽ ὑπὲρ κρεῶν φιλοτιμηθεὶς ἀνῃρέθη ... τὸ δὲ ἑλκύσαι ἀντὶ τοῦ ἐνυβρίσαι φησίν" ἡ δὲ μεταφορὰ ἀπὸ τῶν κυνῶν τῶν ἑλκόντων τὰ σώματα. (Pindar says that he did not mention Neoptolemus ἀτρόπως, i.e. inappositely. According to Callistratus the mention of Neoptolemus is apt because Pindar said that Sogenes is Heracles’ neighbour and Neoptolemus is neighbour to the god at Delphi, i.e. because of their both being neighbours. According to Aristodemus Pindar was criticised by the Aeginetans for seeming to say in the Paeans that Neoptolemus came to Delphi to rob the temple, and so he now offers a defence by saying that he did not die robbing the temple but was killed because of the meat, on a point of honour ... He says "drag" instead of "abuse"; the metaphor is from dogs worrying a corpse.) At the end of the poem Pindar returns to the myth of Neoptolemus,

and comments on his treatment of the myth. Callistratus saw this comment as, in effect, a literary-critical assertion about the appropriateness of the myth to the poem; Pindar is claiming that the myth had not been “dragged in" inappropriately.?* For a clue to what Pindar meant Callistratus looked to the preceding lines, presumably on the assumption that the claim must cohere with its immediate context. In the preceding lines, Pindar has spoken of Sogenes as living near a temple of Heracles, and Callistratus perceived a parallel between this and Neoptolemus' burial at Delphi: Sogenes is to Heracles as Neoptolemus is to Apollo. Hero and victor are both neighbour to a god, and Neoptolemus is mentioned appropriately just because of his and Sogenes' respective neighbourship to a god. This interpretation is certainly wrong; the sense it gives to ἄτροπος is impossible. Nevertheless, Callistratus’ procedure was not irrational. Pindar's abrupt recurrence to the myth at the end of the poem, and the terms in which he expresses himself there, are unique in the Pindaric corpus and constitute a genuine problem which still excites controversy; Callistratus was right to seek an explanation. For Aristodemus' version of the apologia theory see on 70 (the meat motif) and 94a (Neoptolemus' honour).

ANCIENT

INTERPRETATIONS

OF PINDAR’S

NEMEAN 7

181

2. Underlying assumptions

It is

a communis

opinio that the scholiasts never attempted to

interpret an ode as a whole; instead, their procedure was to tackle

individual problems in isolation.?° In its crudest form, the thesis that ancient Pindarists focused exclusively on isolated contexts is easily refuted; it is not hard to find examples in the Pindaric scholia of

interpretations which bring different parts of an ode into relation. Aristarchus! interpretation of Nemean 7 is a case in point. He developed a single hypothesis which solved two problems (the opening invocation and the ““Achaean man" of 64) in a way which linked them to the Neoptolemus myth. Economy is the one conspicuous merit which this interpretation possesses,?* and it is reasonable to infer that Aristarchus was attracted to his theory, and

blinded to its manifest weaknesses, precisely by this economy. Itis at

any rate clear that he was not treating individual problems in isolation from other parts of the poem.?’ But even if Aristarchus did not approach Nemean 7 as a collection of isolated problems, it does not follow that he or any other ancient Pindarist ever attempted to view the poem 'asa whole', in the sense of an artistically unified composition. However, this version of the isolationist thesis is not as straightforward as it may seem. ‘Whole’ in this sense is not a neutral descriptive term, but one laden with

evaluative implications and meaningful only in the context of a particular literary aesthetic.?* It may be true that the scholiasts did not attempt to view a poem as the kind of whole which we would regard as appropriate (whether in terms of our own aesthetic preference or of our concept of Pindar's compositional techniques);

but the unqualified claim that they did not attempt to view a poem as a whole is far more radical, and intrinsically less plausible. Structural questions were of considerable importance in ancient literary criticism, and the Homeric scholia certainly concern themselves with

large-scale compositional techniques.? Before we can attempt a responsible adjudication on this point, therefore, we must be clear what making sense of a poem as a whole might have meant to a Hellenistic commentator on Pindar. (a) Myth, digression and encomium Let us first consider the term παρέκβασις, applied to the Neoptolemus myth of Nemean 7 ina number of scholia (la, 116.13f.; 56a, 124.12f.,

18; 76, 127.7). Modern

readers of the scholia are apt to

assume that this term is equivalent in meaning and implications to

182

MALCOLM

HEATH

our 'digression', and that its use of itself implies irrelevance.?? This

assumption is mistaken, and a correct understanding of ancient thinking on digressions will prove crucial to our grasp of ancient interpretations of this poem. The scholia to Nemean 3 provide a convenient starting-point. The reference to Heracles in 22-6 is described as a parekbasis (45c, 49.24); and Pindar is seen as restraining himself in 26ff. on the grounds that he had gone further than he ought into the praise of Heracles when he should have been speaking of the victor's Aeacid ancestors (45b, 49.10-15). The Aeacid material which the poet annexes is duly recognised as germane (ἐντόπιά τινα kai οἰκεῖα 45b, 24); but it too is

described as a parekbasis in a subsequent note, which comments appreciatively on Pindar's technique (//4b, 58.5-10 πιθανῶς ... οὐκ ἀπρόσλογον

...; cf. 114a, 58.1, τεχνικώτατα). It is clear from these

appreciative comments that the term parekbasis is not itself pejorative; it designates a deviation from the immediate theme, but is

neutral with regard to that deviation's oblique relevance or contextual appropriateness. Two further points are worth noting. First, appropriateness is assessed here in relation to the victor as the laudandus of an encomium.?! We noted in section (1) that Nemean 7 was seen as an encomium (see on 47, 123.12-14); as we shall see, it is a consistent

assumption in the Pindaric scholia that the victory-song is a species of encomium, and the criteria of relevance applied are based on that premise. The idea that patrial and ancestral myth is particularly appropriate to encomium is attested elsewhere in the scholia (N. 8. 7a, 140.16 ἐπικωμιαστικῶς τῶν πατρίων ἐφάπτεται; cf. O. 9.119a-e, 295.5-7,

14-16), and is consistent with ancient rhetorical precept.??

Secondly, the phrasing of the note on 26f. (ἐπιλαμβάνεται ἑαυτοῦ ὡς περαιτέρω ... τραπέντος) is consistent with an awareness that Pindar is adopting a rhetorical pose; it need not assert a critical judgement that Pindar has in fact gone too far from his theme.? The Heracles myth in Nemean 1 was subjected to more extended scrutiny by ancient commentators, precisely because Heracles appears to have no relevance (οἰκείωσις) to the present subject (49c, 19.8-11). A number of solutions which attempt to establish a link between the hero and the victor (innate excellence; toil and reward; Heracles as founder of the Nemean games) are rejected because they fail to account for the narrative of Heracles’ infant prowess in particular (19.11-20.13); Didymus" suggestion that it is parallel to Chromius' first victory and implies subsequent victories is approved

ANCIENT

INTERPRETATIONS

OF PINDAR'S

NEMEAN 7

183

(20.13-21.6).^* Thus ancient commentators on Nemean 1 were aware that Heracles could in general be construed as relevant in a number of ways (another suggestion is recorded in 49b, 19.3-8), and were exercised primarily by the unusual choice of event from Heracles' career in this instance. This tends to confirm that the scholia to Nemean 3 need not imply that Pindar had in fact erred in treating of Heracles. It seems, then, that a parekbasis is not as such objectionable; it may well prove to be οἰκεῖος or, to use an equivalent term which appears in the scholia to Nemean 1, εὔκαιρος (49b, 19.2). Accordingly, the myth of Pythian 11 is condemned not because it is a parekbasis but because

it is ἄκαιρος

(23b,

257.5f.,

cf. 58a, 259.21; 58b, 260.9y;

similarly there is explicit criticism for the use of an “irrational” (ἄλογος) parekbasis in Pythian 10 (46b, 245.23). Thecriterion is again encomiastic. In Pythian 10 Pindar writes a good epinician up to 29, but then **missed the mark" (votóy nos 46b, 245.21f.); in Pythian 11 the myth spoils an excellently composed encomium (23b, 257.4f.). Here the break-off at 38ff. is taken literally: since the poet is under contract to write an encomium of the victor, he should be speaking of the victor himself or his father rather than of other things in a parekbasis (58a, 260.3-8). In these poems the myths lack the patrial or ancestral relevance which the scholia to Nemean 3 take as the norm, and ancient scholars were presumably unable to discern any

alternative ways of making a connection, as they were able to do in

Nemean 1.55 Encomiastic relevance to the victor is therefore one of the criteria of appropriateness which ancient commentators applied when evaluating a parekbasis, another is proportion. The ancestral relevance of the Argonautic myth in Pythian 4 was recognised (119,

115.10-13; 455b,

160.25-161.5; 4554,

161.14-18) but the lengthy

foundation-narrative was nevertheless thought to be inappropriate to an encomium both in content and in scale; thus the poem was criticised for its length as well as for having a “historical” parekbasis, and Pythian 5 — written for the same victory — was regarded as more appropriate in content and structure (οἰκειοτέρα ... κατὰ τὰ

νοήματα καὶ κατὰ τὴν olkovopíav).?? Similarly, the scholia matter (προκείμενον) 383.12; cf. 136, 383.15f. some sense a diversion

on Olympian 13 identify the poem's subjectas praise of the victor (733b, 383.2; 133d, (va ... ἐπαινέσω), from whom the myth is in (ἀφέμενος ... τοῦ νικηφόρου 136b, 383.6).

The patrial content of the myth is noted, however (136b, 383.6), and

184

MALCOLM HEATH

such a 'diversion' would, by the scholiasts' standards, cally relevant; so we should not assume that this intrinsically hostile. What is at issue here is again not much as the length of the digression (πολυλογεῖν

be encomiastiobservation is the content so ... πλείονι τῇ

παρεκβάσει 133b,383.1-7; περιττολογεῖν 133d, 383.11). In Pythian 8 the theme is likewise identified as the victor's praise (40a, 209.22f.) or the victory itself (43a, 210.7-9), but (as Pindar's text suggests) it is

only a complete, long-winded rehearsal of Aegina's glory which would be προσκορές (40a, 209.21f.; 40b, 209.25-210.1); in due

proportion, such encomiastically appropriate material would not be out of place. Similarly in Nemean 8 Pindar is observed restraining himself from developing a paradigmatic reference to Cinyras into an excessively long parekbasis (32a, 143.4-8). Nemean 4.33 (xà μακρὰ δ᾽ &&evénew ἐρύκει pe τεθμός) was glossed as a claim that the norm (νόμος) of the encomium exclude the use of long

parekbases

(53a,

73.3f.);

a parallel

note

adds

that

it was

customary to treat such material concisely (συντόμως 53c, 73.12).

However, Ammonius (another of Aristarchus' pupils) was prompted by the following references to "pressing seasons" and the new-moon festival to identify the ‘law’ in question not as a generic norm but as the poet's contract with the victor, specifying a delivery-date (53b, 73.4-9); other interpreters were apparently content to regard that as

a supplementary motive (53a, 72.20-21; cf. 56b, 74.1f.). Similarly we are

offered

two

interpretations

of Isthmian

1.60-3:

either due

proportion (συμμετρία) forbids extension, or the poem has already reached the length which had been commissioned and paid for (85b, 210.2-11).?? If we return to Nemean 7 in the light of this discussion, we see at once that the use of the term parekbasis has no pejorative implication; on the contrary, it is clear that 76, 127.7, is an appreciative comment on Pindar's graceful, and evidently self-conscious, handling of the return to his main subject-matter (as in N. 3.7 14b, 58.5-10, discussed

above). The question which would interest ancient commentators is concerned with the particular content of this parekbasis: in what way is Neoptolemus appropriate to an encomium of Sogenes? The

obvious answer is that Neoptolemus, as an Aeacid, is of patrial significance to an Aeginetan victor. Ancient interpreters must have been aware of this connection, since the alleged anger of the Aeginetans over the poet's treatment of Neoptolemus would not make sense without it. It follows that ancient interpreters cannot (if they were consistent) have found the appearance of Neoptolemus in

ANCIENT

INTERPRETATIONS ΟΕ PINDAR’S NEMEAN 7

185

Nemean 7 intrinsically problematic, and no reading of the scholia is likely to be correct which supposes that they were exercised by this problem. There is, however, an apparent anomaly here. Ifour sources are to be trusted, ancient interpreters of Nemean 7 proposed a variety of

explanations for the Neoptolemus myth without mentioning its appropriateness according to the standard criteria of encomium. The explanation of the myth which (on the account I have just given) should have been obvious is ignored in our sources, and a range of

quite different interpretations is proposed. This apparent contradiction is one of the problems which must be resolved in any reconstruction of ancient scholarly discussion of this poem. (b) Encomium and apologia The belief that the scholiasts did not see poems as wholes has been

sustained in part by the belief that they were willing to import irrelevancies into them — including such irrelevancies as a personal apologia on the part of the poet.*? But would an apologia have been an irrelevance by the scholiasts’ own criteria? We have seen that ancient interpreters regarded epinician as a species of encomium and judged the appropriateness of a poem's contents accordingly. We must ask, then, how the alleged apologia fits into this approach. The apologetic theory does not imply that the poem is to be read as an apologia instead of an encomium;

it is an encomium,

but has

apologetic elements.*! This point can be illustrated by comparison with the scholia to Pythian 2. Ancient interpreters of that poem conjectured that Pindar had been slandered to Hiero, and that the latter part of the poem consequently embodied an apologetically motivated paraenesis (132b, 54.14-18: for the conjectural status of this theory note μήποτε ... εἰκός ...); some identified the putative slanderer with Bacchylides (731b, 54.2f.: note ἔνιοι; cf. 132c-f, 54.19-55.3 etc.). This theory influenced the interpretation of 52f. (αἰνίττεται δὲ εἰς BaxxvAiönv ἀεὶ yàp αὐτὸν τῷ Ἱέρωνι διέσυρεν 97, 48.1f.); but the passage as a whole was still taken in an encomiastic sense, as a rejection of the vituperative model of Bacchylides (implicit) and Archilochus (explicit) in favour of the poet’s

role

as

Hiero's

encomiast

(see

97,

47.22-48.1;

113a-c,

50.5-14). Furthermore, the discussion of the parekbasis on Cinyras in 15ff. was cast in terms of its relevance to the praise of Hiero (275, 35.17f.; 27c, 35.24); and the encomiastic understanding of the poem

is evident also in the parallel that is recognised between Cypriot

186

MALCOLM

HEATH

praise of Cinyras and the praise of Hiero by the poet himself, as well as by the Syracusans and Locrians (276, 36.5-7; cf. 31b,

16f.; 32,

37.2). The situation with regards to Nemean 7 seems therefore to be as

follows. The poem is understood as an encomium, and is thought to be the vehicle of an apologia; the ancient commentators display no qualms about the appropriateness of apologia in an encomium; yet this is not because they were in general unconcerned about structural questions.

Either, then, they were thoroughly inconsistent in their

approach

or they had some reason for regarding apologia as in

principle at home in encomium. It is not hard to see what this reason was, given the rhetorical basis of their critical outlook. An orator has to ensure that his audience is well-disposed, and this will involve (among other things) disarming any prejudice that exists against the speaker himself.*? It might be objected that Pindar is not an orator, but this line of argument is obviously not open to those who follow

Bundy,® nor in general to those who see in epinician a species of persuasive occasional discourse; on such a view the logic of the rhetorical

theorists’

analysis

seems

inescapable.

A statement

of

praise is worthless if it fails to carry conviction; it will fail to carry conviction if the audience is inclined to discount the praise; they might be so inclined if they are prejudiced against the speaker; and they might be prejudiced against him if they think he has previously

insulted someone whose reputation they value. It follows that if the ancient commentators were right in believing that Pindar composed his song knowing that his potential audience was prejudiced against him because they thought he had slandered Neoptolemus in an earlier poem, then apologia is a proper part of the poem, understood as a species of encomium, even though apologia is not as such encomiastic. Aristarchus and Aristodemus are cited both for the apologetic

theory and for explanations based on a hypothetical contemporary Neoptolemus. This is sometimes seen asevidence either of corruption in the scholia, or of vacillation in the scholars.“ A better explanation may be that they were giving two different answers to two different questions; that is to say, the contemporary Neoptolemus and Pindar's apologia were devised to solve two distinct problems. For

reasons explained in the previous subsection, the Neoptolemus myth as such

is unlikely

to have

seemed

in need

of explanation;?

a

reconstruction of ancient discussion must therefore try to identify other problems which may have given rise to these theories.

ANCIENT INTERPRETATIONS ΟΕ PINDAR’S NEMEAN 7

187

3. Ancient interpretations: a conjectural history I shall now try to show how the identification of some of the ancient commentators' basic assumptions in section (2) helps us to organise the material surveyed in section (1) into an intelligible, if necessarily conjectural, history. (i) Callistratus: My story begins with Callistratus, a pupil of Aristophanes of Byzantium whose work on Pindar is generally thought

to have preceded that of Aristarchus.*‘ As we know from 150a, Callistratus saw in the poem's concluding lines a denial that the Neoptolemus myth had been dragged into the poem inappropriately, and he inferred from the preceding lines that the myth's appropriateness was grounded in an analogy between Sogenes (neighbour to Heracles) and Neoptolemus (neighbour to Apollo). It would clearly be flattering to Sogenes to have his place of residence assimilated in this way to a heroic paradigm. It would be particularly flattering in view of the paradigm's importance to him as an Aeginetan; Callistratus’ interpretation therefore works better if it presupposes the special significance which Neoptolemus had for an Aeginetan victor. I argued in section (2) that this would not have seemed problematic to ancient scholars, and there is no evidence that Callistratus felt that the myth demanded an explanation in itself; rather, he is responding to the fact that in these concluding lines (as he reads them) Pindar himself draws attention to the existence of an explanation. Since the myth may be over-determined, there is no inconsistency between saying (i) that the myth is appropriate because hero and victor are both neighbours to a god, and (ii) that the myth is appropriate because hero and victor are both Aeginetan. So Callistratus may have taken for granted the standard encomiastic significance of the myth, while seeing in the concluding lines a device by which the poet draws attention to an additional and less obvious point of contact between hero and victor. Callistratus saw that the poem's concluding lines were in need of explanation, but his explanation makes no reference to Pindar's apologia. This is surely significant. The connection between these lines and the apologia is easy to make. The denial that the hero has

been "dragged about” or “savaged” with remorseless words sums up the self-defence offered earlier these lines as a metaphor drawn corpse was already familiar to 137.12f.). Modern defenders of

in the poem; the interpretation of from a scavenging dog worrying a ancient apologetic theorists (/50a, the apologetic theory point to the

188

MALCOLM

HEATH

passage as the one irreducible piece of evidence in its support.*' Since Callistratus

apparently

did

not

make

this

connection,

it seems

unlikely that he accepted the apologia theory; there is in fact no evidence that he was even aware of it. (ii) Aristarchus: Aristarchus is the earliest scholar for whom use of the apology-theory is attested (70, 126.8-16); it is a plausible, if not a certain, inference that the theory originated with him. The connection between that theory and the concluding lines of the poem is not attested until the following generation, with Aristodemus (750a,

137.3-13); but I suspect that Aristarchus had

already made the connection.“ His personal relations with Callistratus were apparently not such that he would be likely to overlook the shortcomings in his rival's interpretation;? and, as already suggested, it is intrinsically probable that anyone who holds the apologetic theory will see and exploit the possibility of connecting it with these lines. More importantly, it must surely have been this

passage which prompted Aristarchus to formulate the apologetic theory in the first instance. The alternative, that he was motivated by the reference to fault-finding in 64, is ruled out by his identification of the Achaean man with his hypothetical Neoptolemus; if Aristarchus understood that passage as a reflection of thesurrogate competitor's satisfaction with the poem's oblique acknowledgement of his contribution to the victory, he is unlikely to have felt driven by it to develop the theory of Pindar's self-defence in the face of Aeginetan criticisms. The motivation for that theory is therefore more likely to

have been supplied by the concluding lines. I suspect that Aristarchus rejected Callistratus' he saw that it gave consequence to find Pindar's self-defence

interpretation of this passage (perhaps because an impossible sense to ἀτρόποισι) and had in some other explanation of them; the theory of was developed initially to meet that need.

Once the notion of Pindar's self-defence had been introduced, the

next step was comparatively easy. Aristarchus saw in Pindar's reference to justice and truth at the end of the myth the culmination of his defence, and (like Callistratus interpreting 102-4) sought elucidation in context. The defence rests on the "three words" of 44-7, i.e. the three propositions about fate which show that Neoptolemus' death was honourable (70, 126.8-16).

As for Aristarchus" hypothetical Neoptolemus (Ja, 116.10-14; 56a, 124.12-14), there is no reason to suppose that his invention was motivated by a sense that the Neoptolemus myth is intrinsically problematic. I have argued that the theory was designed to explain

ANCIENT INTERPRETATIONS OF PINDAR'S NEMEAN 7

189

the reference to an Achaean man in 64, and rests on the insight that two perplexing passages (the opening invocation and the *Achaean man") can be solved economically by a single hypothesis, i.e. by assuming that Sogenes' victory was won by a proxy competitor called

Neoptolemus.” This interpretation (like that of Callistratus) works better if we take it for granted that Neoptolemus the hero is appropriate in an encomium of an Aeginetan victor such as Sogenes.

The economy of the poem is far stronger if the oblique praise of Neoptolemus the proxy competitor is conveyed through a myth that is also appropriate to Sogenes the proclaimed victor; otherwise the compliment to Neoptolemus would be an alien and inartistic intrusion. It is worth noting that Aristarchus" theory, thus inter-

preted, is similar to that of Callistratus in this respect also: each interprets a difficult passage asa pointer to a less obvious significance of the myth over and above its patent relevance to an Aeginetan

victor.?! (iti) Aristodemus: Aristodemus modified Aristarchus' interpretation in a number of respects. First, he rejected the implausible notion of a victory won by proxy; Neoptolemus was not the real victor, but Sogenes' trainer, honoured by the reference to his heroic homonym (56a, 124.16-18). This entailed a new explanation of Eileithyia; Aristodemus believed (and was able to convince at least

one other ancient scholar) to support the conjecture Sogenes was a late-born crucially reformulated the

that he had found evidence in Simonides (perhaps based on the victor's name) that son (/a, 117.12-18). Aristodemus also apologia theory. He seems to have shifted

the focus of the defence from 44f. (the role of fate in its various

aspects in Neoptolemus' death, emphasised by Aristarchus) to 42 (the fact that Neoptolemus died in a fight over sacrificial meat). In

this interpretation, Pindar denies that in mentioning ‘myriad honours” in the Paean he had accused Neoptolemus of an impious ambition to rob Apollo’s temple; instead, the hero was killed in a fight over the meat, in which his honour was at stake (/50a, 137.3-13). Aristodemus accordingly had to find a new interpretation of the “three words”, and came up with the idiosyncratic suggestion that they were the poem’s three remaining triads (70, 126.20-3).

(iv) Subsequent modifications: An unidentified scholar?? was able to integrate Aristarchus’ view that the “three words" of 48 are three

propositions contained in the preceding lines with Aristodemus’ introduction of the meat-motif by extending the relevant context; the claim that the Paean referred to a fight over sacrificial meat rather

190

MALCOLM

HEATH

than to temple-robbery is now the first of the three propositions (70, 126.16-20).

The

same

scholar

may

have

been

responsible

for

connecting the Achaean man who finds no fault with the poet in 64 with Pindar’s apologia (94a, 128.13-129.12); in so doing he rendered the hypothesis of Neoptolemus the trainer or proxy victor super-

fluous. With this development the ancient interpretation ofthe poem reached its most satisfying and economical form. Eccentricities such as the hypothetical Neoptolemus have been discarded, and the apologia theory now offers integrated explanations ofthree problem

passages (the “three words”, the Achaean man and the concluding lines) in a manner consistent with the commentators’ rhetorical presuppositions. What account would now be given of the Neoptolemus myth? The final

sentence

Neoptolemus on those who Observation is much as an

of

47,

123.16-18,

points

out

that

the

death

of

illustrates the general statement in 30f., that death falls do not expect it as well as on those who do. The not perhaps an attempt to explain the myth as such, so observation on how the myth coheres with the

immediately preceding context. In any case, this interpretation cannot have been accepted universally since it requires that in 30f. ἀδόκητον and δοκέοντα are interpreted in terms of expectation, not renown, and we know that there was a variety of opinions on this

point among ancient interpreters (44a-b, 122.10-25).** No other explanation of the Neoptolemus myth subsequent to Aristodemus is recorded in our sources. It seems probable, therefore, that ancient scholars after Aristodemus had no special explanation to offer for the Neoptolemus myth. But that should not surprise us since, as I have argued, ancient scholars did not need a special explanation; the standard encomiastic appropriateness of the myth would have been self-evident to them. This conclusion gains support from the way in which our sources

present the various theories about the myth which we have mentioned. In the note on the opening invocation the various solutions are set out systematically as a zerema, a formal enquiry: the problem is stated; five proposals are mentioned and rejected, with reasons given in each case; a sixth proposal is mentioned and approved. The same

systematic format is adopted for the myth of Heracles in the scholia to Nemean 1 (49c, 19.8-21.6), which as we have seen constituted a genuine difficulty for ancient interpreters since it could not be explained on any of the standard encomiastic lines. But the Neoptolemus myth in our poem is not treated in this systematic

ANCIENT INTERPRETATIONS OF PINDAR'S NEMEAN 7

191

manner; opinions are recorded haphazardly, and they are not subjected to competitive evaluation. The reason may be that the myth itself was not seen as a problem in need of a solution; rather, there was a variety of separate problems, each of which prompted

supplementary theories about the myth. The obvious encomiastic explanation is not mentioned for the simple reason that it is obvious,

and the myth as such therefore requires no comment. 4. Evaluation What,

in the light of this reconstruction,

are we to make

of the

apologetic theory? Note first of all that in its ancient versions it does

not imply that Pindar was retracting an earlier treatment of the Neoptolemus story, nor that he was (in the modern sense) apologising for an acknowledged insult; it is therefore not quite correct to say “ Aristarchus and his pupil Aristodemus believed that ... Pindar tried to excuse himself for an uncomplimentary reference to Neoptolemus

made in a Paean he had written earlier." Aristodemus explicitly interprets Pindar's self-defence as a denial that his earlier reference was uncomplimentary; the poet maintains that the Aeginetans had misinterpreted the Paean. Aristarchus" position is preserved in less

detail, but may well have run along similar lines: Pindar denies that the death as recorded in the Paean reflects badly on Neoptolemus, as the offended Aeginetans supposed, and sets out explicitly the three reasons why it does not. Secondly, in its dominant variant (that initiated by Aristodemus)

the apologetic theory cannot possibly be accepted, since it presupposes a scholarly dispute about the correct interpretation of what

is in all probability a corruption in the text of Paean 6.57 Finally, there is no reason at all to suppose that Aristarchus had any evidence outside the poem to support his theory that the Aeginetans were angry. It has, in any case, always been impossible to explain how such a tradition might have been preserved over three centuries to fall into the hands of an Alexandrian scholar; but, given what we have observed of Aristarchus' methods, there is no need to

make this supposition. The hypothesis of Sogenes' victory by proxy was clearly not the work a man cautious in conjecture. The purely conjectural basis of much that purports to be historical or biographical information in the Pindar scholia is not, of course, a novel conclusion,’® and on this point the modern consensus seems to me

correct. It follows that, whether or not one accepts the apologetic

192

MALCOLM

HEATH

theory in some form, the scholia offer no independent evidence whatsoever that the Aeginetans really were, as a matter of historical fact, offended by the Paean; that has to be be inferred, if at all, from the text of Nemean 7. In other words, we may legitimately conclude that the Aeginetans were offended by Paean 6 only if we conclude, on

other grounds, that Nemean 7 cannot be satisfactorily interpreted

except apologetically. In all probability, such a conclusion will be founded

on

the

same

evidence

which

(on

my

reconstruction)

prompted Aristarchus to initiate the apologetic theory — the poem's concluding lines. But whether these lines do warrant the theory is an issue on which there is currently no consensus, and which it is not my present concern to adjudicate.?? I will, however, briefly indicate why 1 do not think that the apologetic theory, if accepted at all, should be accepted in a form which would satisfy the ancient commentators. For them, the

apologia must be seen as a rhetorical move in the context of an encomium; as we have seen, the premise that epinician is a species of

encomium was fundamental to their interpretations. But this premise is misleading. It is true that Pindar freely applies such terms as ἐγκώμιον μέλος to epinician; but by this he meant that epinician was

a species of komos-song, not of encomium in the later rhetorical sense. Ancient commentators can be shown to have been misled by this shift of meaning into a number of detailed misinterpretations; their failure to grasp clearly the komastic nature of epinician led them to gloss κῶμος inaccurately as χορός and as ὕμνος (I believe

that they may

have been wrong to suppose that epinician was

chorally performed, but my point here does not depend on that belief).! It is arguable that the consequences of these errors of detail

live on in the form of a reluctance to allow the odes to refer beyond themselves to their context of performance.‘? More fundamentally,

there seems to be no reason to grant that a song which is, in essence, a contribution to a komastic celebration need be governed by the relatively strict criteria of relevance appropriate to the rhetorical encomium; the ‘centrifugal’ tendency of ancient literary aesthetics means that more flexible principles of composition than the scholia

apply to epinician could readily be accommodated, were such flexibility generically appropriate. From this point of view, one might (for example) be able to argue both that the myth of Nemean 1 has only a limited point of departure in the celebration of Chromius' victory (as ancient commentators feared) and that there is no reason

to find fault with it on that account. But I do not propose to develop

ANCIENT

INTERPRETATIONS

OF PINDAR'S

NEMEAN

7

193

these speculations further here. 5. Conclusion

I have argued that the Pindar scholia have been widely misunderstood; but when these misunderstandings are corrected, it remains

true that the ancient commentators themselves misunderstood Pindar, which is what everyone believes anyway. Is such a result worth the effort of serious, sympathetic interpretation which (I have maintained) is required if we are to make any responsible reference to

the scholia at all? Can we not simply discard them? I offer two brief concluding observations. First, the perspective of the Pindarist is not the only legitimate one;

an understanding of the Pindar scholia is of obvious importance in the study of, for example, Hellenistic literary scholarship. To put the point more generally: for students of antiquity Aristarchus and Aristodemus have a claim to serious attention in their own right, and

not simply as adjuncts to Pindaric studies. Secondly, although the Pindar scholia are of limited direct use to the Pindarist, they may still have an indirect bearing on our approach to Pindar, and indeed to other ancient poetry. The scholia constitute one source of evidence on which we may draw when enquiring into the structure and evolution of Greek literary aesthetics. Such an enquiry is, in my view, a fundamental prerequisite for the study of ancient literature; for a genuinely critical understanding of ancient

literature will be impossible until we have grasped the distance between our own literary culture and those of the ancient world. If that is so, then we cannot afford to ignore any source of evidence —

not even ancient misinterpretations of Pindar.‘*

NOTES l.

Onthescholia to N. 7see especially Ole L. Smith *Pindar's Seventh Nemean Ode’ C&M 35 (1984) 5-17; although in what follows 1 dissent from Smith's conclusions on a number of points, his article provided an invaluable stimulus at

an early stage of my work on this material. 2.

The Pindar scholia are cited from the edition of A.B. Drachmann (3 volumes,

Leipzig 1903-27), following the convention adopted by M.R. Lefkowitz ‘The

influential fictions in the scholia to Pindar's Pythian 8' CP 70 (1975) 173-85, revised in First-Person Fictions: Pindar's Poetic ‘I’ (Oxford 1991) 72-88: i.e.,

references in Roman type are to the lineation of the poems in modern editions, while references in italics give the lineation used by Drachmann, with his subdivisions, to which references by page and line in the relevant volume are added. Note that Drachmann's subdivisions often fail to reflect the underlying

194

MALCOLM

HEATH

structure of the scholia, as he himself warned: ""magis commodo eorum qui hac editione usuri sunt quam ipsi rei et veritati consului" (I.xv).

H. Fränkel 'Schrullen in den Scholien zu Pindars Nemeen 7 und Olympien 3' Hermes 89 (1961) 391 n.1. Notably U. von Wilamowitz Pindaros (Berlin 1922) 161.

Because of its bearing on the Neoptolemus myth Frünkel (n.3) 391 n.1 and 388 n.2 refuses to count Aristarchus' contribution to the discussion of Eileithyia as such, claiming that it is “mistakenly” attached to this discussion. This is to force onto the evidence the assumption that ancient scholars concerned themselves only with isolated questions (see n.25). The attribution has been doubted, incorrectly: see n.44 below. If Sogenes is young simply in the sense that he competed in the boys' class, then

his youthfulness is not distinctive and would not explain why Pindar invokes Eileithyia in this poem and in no other; it would therefore fall to the objection which Didymus

raises against the fifth theory: why does Pindar not mention

Eileithyia in other, similar cases? This kind of objection is rejected by Frünkel

(n.3) 393f.; for a defence see G.W. Most The Measures of Praise (Hypomnemata 83, Góttingen 1985) 28 (the principle of sufficient reason is "a necessary

regulative ideal for any interpretation whatsoever”) and 35 n.35 ("Chrysippus'

suggestion [at N. 1.49c, 20.8-11) is ignored here, as it could not explain why the

myth of Herakles does not appear in every Nemean ode"). Also relevant here is

the idea that encomium should not be excessively generalised: see on 47 below. D.C. Young ‘Pindar Nemean 7: some preliminary remarks (vv. 1-20)’ TAPA 101 (1974) 636f.; accepted e.g. by C. Carey 4 Commentary on Five Odes of Pindar (New York 1981) 137 (“its illusory support in an epigram of Simonides is refuted

by Young"), and presumably by W.J. Slater ‘Problems in interpreting scholia on

Greek poetic texts' in J. Grant (ed.) Editing Greek and Latin Texts (New York 1989) 50 (“the quotation from Simonides he uses as evidence is completely irrelevant"). By contrast C.O. Pavese ‘La settima Nemea di Pindaro' in E. Livrea & G. Privitera (edd.) Studi in Onore di Anthos Ardizzoni (Rome 1978) 667 and 687 n.1, endorses Aristodemus' theory. Frankel (n.3) 392f. has a helpful discussion of

Aristodemus' proposal.

His objection is incorrect: H. Lloyd-Jones ‘Modern interpretation of Pindar’ JHS 93 (1973) 129 n.117 (reprinted in Greek Epic, Lyric, and Tragedy [Oxford

1990] 110-53).

10.

On the transmission of the scholia see H.T. Deas ‘The Scholia Vetera to Pindar’

11.

The assumption that scholia are not worth serious attention is doubtless to blame for the extraordinary carelessness with which they are often treated, on which see

HSCP 42 (1931) 1-78.

Smith (n.1) 6 (with 15 n.10, 16 n.13). 12.

W.J. Slater ‘Doubts about Pindaric interpretation’ CJ 72 (1976/7) 206.

13.

Cf. 53, 124.6-9 συνήθως δὲ ὁ Πίνδαρος τῷ τῆς λέξεως κέχρηται σχήματι; and see Drachmann's index, [11.354f., for other scholia on the interchange of singular and plural.

14.

Aristodemus' suggestion is accepted by Pavese (n.8) 681f.

15.

Note that ancient readers punctuated after ἐπιστατεῖ, taking the address to Aegina with what follows: 73, 126.24-127.5.

ANCIENT INTERPRETATIONS OF PINDAR'S NEMEAN 7

16.

195

A guess, but consistent with the modification of Aristarchus’ view discussed below. δέ for γάρ is discussed in a number of Pindaric scholia: O. 2.106a (88.21-3: the attribution to Aristarchus is convincing); O. 4.34b-c (137.12-15);

O. 6.4b-c (155.3-11); O. 13.83 (376.4-6); P. 3.22} (66.12-14).

17.

μοιριᾶν Boeckh (after J.G. Schneider), κυριᾶν Housman; see n.57 below.

18.

For the text of this scholion see S. Radt Pindars Zweiter und Sechster Paian: Text, Scholien und Kommentar (Amsterdam 1958) 165 n.1.

19.

On this myth see G. Nagy The Best of the Achaeans (Baltimore 1979) 118-41; L. Woodbury Phoenix 33 (1979) 95-133.

20.

Smith (n.1) 17 n.31 infers from ἱεροσυλία in 150a, 137.4-6, “which cannot be based on Pindar's text, but only on sch. 94a" (which he attributes to Aristarchus:

I argue below that this is a mistake), that Aristodemus “did not even look into the text of the

Paean";

in my

view

it suggests

that he was

understood the implications of the scholia to the Paean.

21. 22.

familiar with

and

See M. Heath Unity in Greek Poetics (Oxford 1989) e.g. 81f., 102, 107, 109-11. The following sentences (129.4-11) are explanatory (γάρ), but do not seem to

explain

why

the claim

that Neoptolemus

died fighting about meat is not

blameworthy; instead, they are designed to clarify the general background (note

καθόλου), and explain why that claim is in question at all: i.e., because Pindar in defending himself against Aeginetan criticisms (distinct, of course, from the potential Epirote criticism with which the note begins) asserts that in the Paean he did not say that Neoptolemus fought about valuables (which would imply

temple-robbery), but about “the customary honours of the Delphians", i.e. the meat. 23.

See Smith (n.1) 14; E. Horn De Aristarchi studiis pindaricis (Diss. Greifswald

24.

Helpful on Callistratus' interpretation of 102-5 is Fränkel (n.3) 385-7.

25.

U. Wilamowitz *Pindars siebentes Nemeisches Gedicht’ SBBerlin, ph.-hist. KI. 15 (1908) 345 = W.M. Calder ἃ J. Stern (edd.) Pindaros und Bakchylides (Wege der Forschung 134, Darmstadt 1970) 148: "Dan die antiken Grammatiker das Gedicht als Ganzes zu verstehen keinen Versuch gemacht haben. kann nicht

1883) 64 n.92.

befremden: das haben sie nie getan”; likewise (n.4) 162 n.2: “Ein Gedicht als Ganzes aufzufassen haben sie ja nie versucht." Cf. S. Fogelmark Studies in

Pindar with particular reference to Paean VI and Nemean VII (Lund 1972) 111 n.75 (“reading the scholia one gets the impression that every crucial passage is being explained in isolation and without reference to similar passages"); Lefkowitz (n.2) 177 = First-Person Fictions (n.2) 78 ("By reading one line at a time, with the possibility of easy reference within or outside the poem restricted by the format of the papyrus roll, the commentators concentrated on individual words and so lost the sense of context that gives Pindar's words their full meaning"); Most (n.7) 36f.; Slater (n.12) 203 n.60 (“Radt’s account ... is vitiated by a belief that Alexandrian scholars produced coherent interpretations of poems: in fact they thought in zetemata and luseis as our Pindar scholia amply demonstrate ..."’); Smith (n.1) 10 ("Aristarchus and Aristodemus did not look upon the text of Pindar in the same way that we do. For them the text was not an organized whole. They tried to solve the problems as they appeared"); P. Wilson ‘Pindar and his reputation in antiquity' PCPS 26 (1980) 106 (but see n.27 below). See also

Fränkel in n.5 above.

26.

Cf. Lefkowitz (n.2) 176 = First Person Fictions (n.2) 751.

27.

Wilson (n.25) 107f. cites some other pertinent examples.

196

MALCOLM

HEATH

28.

For the point of principle here see Heath (n.21) 1-3; id. ‘The origins of modern Pindaric criticism’ JHS 106 (1986) 97f.

29.

See N.J. Richardson ‘Literary criticism in the exegetical scholia to the Iliad: a sketch’ CQ 30 (1980) esp. 267-70; R. Meijering Literary and Rhetorical Theories

in Greek Scholia (Groningen

30.

1987) 134-225.

Thus Smith (n.1) 16n.20 (following A.B. Drachmann Moderne Pindarfortolkning [Copenhagen 1891] 13ff.) says that "the scholiasts regularly use the term παρέκβασις ... where they were unable to account for the choice of myth. They

regarded myths as digressions unless there was an obvious and direct connection

between subject and myth, in which case the scholiasts did not comment on the connection.” Cf. M.R. Lefkowitz ‘The Pindaric scholia" AJP 106(1985) 274f. = First-Person Fictions (n.2) 152: contra Heath (n.21) 160f., a brief discussion of the Pindar scholia which is here expanded and corrected in the light of further reflection on their overall approach to epinician poetry.

31.

Note ἐπαινούμενος in 45b (49.14, 20, 23); ἔπαινος in 45b (49.19), 57 (50.15), and

especially 54a (50.23), where ποτίφορον κόσμον (31) is glossed οἰκεῖον ἔπαινον (cf. 54b, 50.26f.); and ἐγκώμιον in 53a (50.17), 53b (50.20), 54a (50.25). Cf. Wilson (n.25) 107.

32.

The Ἰοροὶ of encomium include τὸ γένος, ὃ διαιρήσεις εἰς ἔθνος πατρίδα προγόνους καὶ πατέρας (Aphthonius Prog. 11.36.10 Sp., cf. Nic. Soph. Prog. 111.479.32-480.16 Sp.; Quint. 3.7.10); cf. also sch. O. 2.735 (61.15-17). See T.C.

Burgess Epideictic Literature (Chicago Studies in Classical Philology 3, 1902) 113-27. 33.

The noncommittal ὡς + participle of 45b appears as γάρ + indicative in 45c. This may

be a case

in which

parallel

notes allow

us to observe

misleading process of epitomisation. Cf. n.38 below. 34,

the potentially

It should not be supposed that Didymus' belief that this is Chromius’ first victory, won in his first competition, was other than speculative: cf. n.58 below.

More recent interpreters do not necessarily show greater restraint in speculation; W.J. Slater, in D.E. Gerber (ed.) Greek Poetry and Philosophy: studies in honour of Leonard Woodbury (Chico 1984) 254, conjures a reference to “this first military

distinction of Chromius" out of Heracles' first fight in N. 1.43. 35.

Wilson

(n.25)

108 remarks

that "terms

like ἄκαιρος and ἄλογος manifestly

reflect upon the limitations of the commentators' understanding rather than upon Pindar"; this is no doubt true, but modern interpreters have also found the myths in question difficult.

36.

37.

Thus the difference in scale between the treatments of Heracles in N. | and N. 3 also helps to explain why the one case was found more problematic than the other. Ón length as a genre-specific criterion of structural integrity in ancient criticism cf. Arist. Poet. 1450b34-51a15, 59b17-31, 62b5-10; Heath (n.21) 37 n.18, 43f., 95; the proem to Callimachus' Aetia is obviously relevant. P. 4.inscr.a, 92.15-18; cf. 1a, 93.24-94.3 (μεῖζόν τι ἢ κατὰ ἐπίνικον οὖσα διὰ τὸ

μεμηκύνθαι καὶ πραγμάτων ἔχειν ἀφήγησιν

ἐντοπίων), P. 5.inscr., 171.25-172.2

(διήγημα ... μᾶλλον ... ἤπερ ἐγκώμιον).

38.

As transmitted

733b, 383.7, appears to state that this

excessive in length; but comparison with N. 3.45b-c (see n.33

parekbasis is in fact above) suggests that

we should not rule out an awareness of the rhetorical nature of Pindar's break-off

device. Similarly, N. 4.60b, 74.16-75.3, might suggest that a parekbasis as such

would be found blameworthy; but I would again be cautious, since 60c, 75.4f. implies that here too proportion is at issue: ἐὰν μὴ ἐπιμένωμεν talc παρεκβάσεσι ... Also in P. 8.40c, 210.2f., οὐκ ἔστι καιρὸς τὰ ἐγκώμια τῆς Αἰγίνης ...

ANCIENT INTERPRETATIONS ΟΕ PINDAR’S NEMEAN 7

197

λέγειν is oversimplified by comparison with 40a, 209.21f., οὐκ edxaipw πᾶσαν [n.b.] αὐτῆς τὴν ἀρετὴν ἐμπεριλαβεῖν τῷ ὕμνῳ (see text below). In these instances parallel notes allow us to observe the potentially misleading consequences of epitomisation. In N. 6.94a, 113.7-10, I would likewise see an interpretation ofthe break-off device, rather than a criticism for irrelevance, for

the patrial significance of the Aeacids is clearly understood: 75, 111.6-13.

39.

For

Aristodemus'

adherence

to the theory

of contractual

length, see his

interpretation of the “three words" of N. 7.48, discussed above (70, 126.20-23).

The conjectured terms of a commission were also invoked by one of the explanations offered for the Heracles myth in N. I; the suggestion that this theme (ὑπόθεσις) was given to Pindar was rejected as implausible as early as Aristarchus (49c, 19.11-13), but it does not follow that the myth would have been

rejected as artistically unsatisfactory had the conjecture been true. Thus the parekbasis constituted by the opening catalogue of Argive glories in N. 10 is explained as a way of achieving adequate length despite the insignificance of the

competition in which the victory was won (35. 170.2-4; compare O. 8.inscr.b., 237.9-12). The idea that technical artifice, such as the need to compensate for an

otherwise unpromising subject-matter, can provide aesthetic justification for what would otherwise be a fault can be paralleled in later criticism: see Heath ‘Origins’ (n.28) 92. The specific fault in question here is presumably a use of

generality beyond what is appropriate in an encomium: compare 35, 170.23f. (ἐπὶ τὴν

κοινότητα

123. 13f., above.

τῶν

ἐπαίνων

τῆς

πατρίδος)

with the discussion

of N. 7.47,

For the assumption that apologia is "strictly personal", and therefore irrelevant, cf. (e.g.) D.C. Young Three Odes of Pindar (Mnemosyne Supplement 9. Leiden

1968) 1f.

al.

Thus the scholiasts would agree neither with Bundy's description of the poem nor with the view he attributes to their followers (Studia Pindarica [Berkeley 1962] 4): "N. 7. a straightforward enkomion, has been canonized by those who follow one guess reported in the scholia as the poet's personal apology for offensive references to Neoptolemus in the ode we now possess fragmentarily as Pa. 6". They would regard it as an encomium that is not entirely straightfoward, and would — as we shall see — be perplexed by Bundy's apparent belief that their

position can be ruled out in principle on rhetorical grounds. For the fearthat the

ode would, if read apologetically, turn into something other than an encomium

of Sogenes cf. Smith (n.1) 14.

22.

This topic is most developed in discussions of the exordium of legal oratory; see J. Martin Antike Rhetorik: Technik und Methode (Munich 1974) 60-70. Cf. Cic. Inv. 1.22 (benevolentia ... ab nostra [sc. persona), si de nostris factis et officiis sine arrogantia dicemus; si crimina [n.b.) illata et aliquas minus honestas suspiciones

iniectas diluemus ...), Quint. 4.1.5-22 (note especially 4.1. 10, on the need to avoid the appearance of being a slanderer, especially eorum qui laedi nisi adversa iudicum voluntate non possint; 4.1.12: pertinere ad causam puto quidquid ad

dicentem pertinet, cum sit naturale ut iudices iis quos libentius audiunt etiam facilius credant; and 4.1.16, which provides a parallel for the paraenetic element

detected at P. 2.132b, 54.17f., and 133b, 55.7f.). Of course, in kinds of oratory other than forensic the principles must be applied mutatis mutandis: Quint. 3.7.23-5. It is important to distinguish apologia on the speaker's behalf from apologia on behalf of the subject, which ancient theory generally does exclude from encomium: Theon Prog. 11.112.8-13 Sp. (quoting Isocr. Helen 14), Nic. Soph. Prog. 111.481.28 Sp.; Quint. 3.7.6 is flexible: cf. Burgess (n.32) 118.

43.

Cf. Bundy (n.41) 3f., 35. Horn (n.23) 65 n.94, followed by S. Radt (n.18) 85 n.4, substitutes the name

Aristonicus; contra Smith (n.1) 9, who however tries to resolve the problem by claiming that for Aristarchus and Aristodemus the text “was not an organized

MALCOLM

198

HEATH

whole”: cf. n.25 above. Frankel (n.3) 388 n.1 also denies that the text is in error, but sees the explanations as alternatives, as does Young (n.8) 636 n.20 (‘See his (second?...) explanation of Neoptolemus’ relevance").

45.

Smith

(n.1)

10, who

recognises

the

possibility

that

the two

theories were

addressed to two different problems as a rational alternative to his preferred explanation (see previous note), goes astray at this point: "we may still wonder why they could not see that if the story of the Aeginetans could explain Pindar's treatment, it could explain the introduction of the myth as well". Cf. R. Pfeiffer History of classical scholarship: from the beginnings to the end of the Hellenistic age (Oxford 1968) 221 (“probably ... the first to comment on some of

Pindar's poems"); J. Irigoin Histoire du texte de Pindare (Etudes et

Commentaires

13. Paris 1952) 49. Slater (n.12) 203 and n.60 reverses the order, citing the evidence of Callistratus' personal hostility towards his rival in Athenaeus (21c ἐν συγγράμματι κακῶς εἴρηκεν ἐπὶ τῷ μὴ εὐρύθμως ἀμπέχεσθαι); but this might

just as plausibly be seen as giving Aristarchus motive to reject Callistratus' interpretations

(and

the assumption

that only personal

motives

need to be

considered is not one that I could accept). Precedence cannot be definitively established, but I have found it easier to produce a reconstruction on the conventional view. 47.

See (e.g.) E. Tugendhat ‘Zum Rechtfertigungsproblem in Pindars 7. Nemeischen Gedicht‘ Hermes 88 (1960) 385-409 (404: "An einer einzigen Stelle jedoch — und

zwar am Schluß ... — kommt der Dichter unmittelbar auf seine Haltung im Paian zu sprechen"); Lloyd-Jones (n.9) 127-37 (esp. 136).

Smith (n.1) 14 suggests that the link with the concluding lines was Aristodemus' contribution.

49.

See n.46 above. For conjectural homonyms as a resource in ancient scholarship see D. Fehling Herodotus and his 'Sources' (Leeds 1989) 180f. (a reference I owe to Gordon Howie).

51.

A rhetorically trained reader would recognise in this a species of λόγος ἐσχηματισμένος (i.e., a text which is meant to convey a secondary sense in addition to, or instead of, its primary sense). See especially the treatises falsely attributed to Dionysius of Halicarnassus (Opuscula 11.295-385 Usener-Rademacher); cf. K. Schópsdau Rh.M 118 (1975) 83-123.

52.

Éviot at 126.16 probably refers to a single scholar, as at /a, 116.10. it designates Aristarchus.

53.

For this sense of προκείμενον cf. P. 3.25a, 67.1f., N. 6.67a, 119.21f. For myth as an illustration of a general principle cf. P. 2.40a, 38.15-21: if the Ixion parekbasis is applied specifically to Hiero, it is pointless (εἰς οὐδέν); he must therefore illustrate the general principle (καθόλου παρελήλυθεν) that benefactors are to be rewarded.

54.

As among modern interpreters: D.C. Young ‘A note on Pindar Nemean 7.30ff.' CSCA 4 (1971) 249-53, is in my view correct.

55.

As Smith observes (quoted n.30 above), the scholia tend not to comment on connections which are obvious.

56.

Lloyd-Jones (n.9) 128; cf. Bundy, quoted n.41 above; likewise A. Koehnken Die

Funktion des Mythos bei Pindar (Untersuchungen zur antiken Literatur und Geschichte 12, Berlin 1971) 37, finds in the scholia an interpretation of the poem as a "palinode".

ANCIENT INTERPRETATIONS ΟΕ PINDAR’S NEMEAN 7

57.

199

As Radt (n.18) 166f. points out (after Verrall), the Aeginetans could not have been offended by μοιριᾶν or κυριᾶν (see n.17 above) in the text of the Paean, since these epithets would clearly exclude the ‘temple-robbing’ interpretation; but this only counts in favour οὔμυριᾶν if one accepts that the Aeginetans were in fact offended (and that Aristodemus correctly identified the cause of their offence). The case for emendation is admittedly weakened by the realisation that

the scholiast's νομιζομένων is not evidence (νομιζομένων ... tolg Δελφοῖς is parallel to συνηθῶς in the Paean scholia, cf. 62a, 135. 10f. ὡς ἔθος αὐτοῖς; as

Smith (n.1) 17 n.30 observes, it is an explanation of the events rather than a

paraphrase of the text of the Paean, and so does not provide evidence for the emendations of puptav). Even so, no satisfactory interpretation of nopi&v has yet been offered. Even if the text is correct, however, Aristodemus' theory can be shown to presuppose the Paean scholia: Smith (n.1) 13. 58. 59.

See (e.g.) Frünkel (n.3); Lefkowitz (n.2); Smith (n.1).

For supporters of the theory see n.47 above; contrast Bundy (n.41)4; W.J. Slater ‘Futures in Pindar’ CQ 19 (1969) 91-4 and (n.12) 203-8; Koehnken (n.56) 37-86;

M.R. Lefkowitz ‘Autobiographical fiction in Pindar’ HSCP 84 (1980) 39-47 = First-Person Fictions (n.2) 127-46. Most (n.7) 203-9 attempts to reach a mediating conclusion. My sympathies lie with the sceptics; and if (as 1 am inclined to believe) lan Rutherford is right to argue (in his forthcoming commentary) that Paean 6 was composed for the Aeginetans themselves, then the case for the apologetic theory is weakened still further. But I am not confident

that 1 understand the concluding lines.

See A.E. Harvey ‘The classification of Greek lyric poetry’ CQ 5 (1955) 160-4; M. Heath *Receiving the κῶμος: the context and performance of epinician' AJP 109 (1988) 182f. 61.

On the misconstrual of κῶμος see Heath (n.60) 184f., with additional references

in M. Heath ἃ M.R. Lefkowitz ‘Epinician performance’ CP86(1991) 175n.4. An

awareness of the komastic context is occasionally discernible: P. 4. Ic, 94.21f.; P.

8.99b, 215.25; N. 6.55b, 131.

62. 63.

Heath & Lefkowitz (n.61) 190. See Heath (n.21). For the restrictiveness of the scholiasts' concept of encomium

cf. Wilson (n.25) 111; Lefkowitz (n.30) 272f. — First-Person Fictions (n.2) 150f. But it would be wrong to imagine that they simply impose an external concept of encomium; for their attempts to infer special criteria of Pindaric encomium from Pindar's text see Wilson (n.25) 107.

The first draft of this paper benefitted greatly from Mary Lefkowitz's comments; parts of a subsequent revision formed the basis of papers given at Harvard and

Leeds in the autumn of 1990.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 201-11 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

LANDSCAPE

AND THE GODS

IN

CALLIMACHUS’ HYMNS (with particular reference to Hymn 3) VIRGINIA KNIGHT

I. Introduction

This paper will treat aspects of Callimachus' presentation of landscape in his Hymns; the contrast between city and country in Hymn 3 (to Artemis) will be examined in detail, and other Hymns discussed as background. The interrelatedness of Callimachus' Hymns is thus presumed and will be further demonstrated. On this topic surprisingly little has been written,! although it might be

expected that extended poems in similar form by the same poet would have links, and that an author who alludes to the work of

other poets at every turn would also allude to his own. Overschematism

must

however

be avoided: a theme shared by some

Hymns will not necessarily appear in all. One obvious way in which the Hymns are comparable is that the same gods appear in different Hymns. The Olympian gods are never presented 'in session' in the Hymns, as they are, for example, in Homeric Hymn 3.309-31 (to Apollo) or in Callimachus' own Jamb 12; instead they move in and out of one another's hymns. In the hymns to Artemis (3.4-40) and to Athena (5.131-6) Zeus exercises the

authority attributed to him in Hymn 1. Artemis and Apollo, who are born in the hymn 3), and Artemis qua hymns will machus' Hymns

to Delos (4), each have a hymn of their own (2and appears in the hymn to Apollo (2).? Again, hymns have hymnic features in common. Each of Callicontains myths concerning its subject, and other 201

202

VIRGINIA KNIGHT

sacral elements appear in more than one: thus three (2, 5 and 6) are located in a purported ceremony which unfolds as the hymn is performed. Then the mythical narratives (and other passages) ofthe Hymns invite comparison ofthemes which appear throughout Greek literature, e.g. the portrayal of relationships and institutions. Finally, there are stylistic and linguistic links. This paper confines itself to landscape because it is prominent in several Hymns, and not unexpectedly, since a concern with geography and place-names is characteristic of hellenistic poetry, which thus reflects a poetic preoccupation going back to the Jliad’s catalogue of ships. Callimachus and his contemporaries were fascinated by local colour and introduced it frequently into myths and aetiologies;? he himself also wrote in prose on geographical themes.* Some hellenistic poets celebrated cities in particular: there are numerous epigrams on cities in the Anthologia Palatina,’ and Apollonius wrote a series of Κτίσεις narrating foundation myths. This hellenistic interest in places coincided with another typical feature of hymns, i.e. frequent references in narrative and divine encomium to locations. Thus Menander Rhetor stresses that geography is an important part of kletic and apopemptic hymns (Spengel 3.334-6). Hymn 4 (to Delos) best illustrates this tendency in Callimachus, since it is addressed to a

place, and its myth is enacted largely by personified geographical features. Elsewhere in the Hymns landscape appears in mythical references to places connected with the god, in accounts of the god's devotees (e.g. the settings of the rites of Hymns 2, 5 and 6), and in generalisations about

the god's relationship to cities, mountains,

forests, fields, and the sea. It will be argued that some geographical references in the Hymns do more than provide local colour, an excuse for an aetion, or an

opportunity for Callimachus to display his learning: they are also a means of characterisation, or a unifying force within a hymn, or they forge a link between different hymns. As noted, Hymn 3's contrast between 'city' and 'country' will be highlighted. Here 'city' translates πόλις and refers to a community centred on a settlement, while *country' refers not to the cultivated land within the boundaries of the ‘city’ but to the wild land outside its borders.* II. Landscape and the Attributes of Artemis in Hymn 3 It is typical of hellenistic poetry, including Callimachus, to achieve surprise by subverting the rules of a genre, by treating a familiar story

LANDSCAPE AND THE GODS IN CALLIMACHUS' HYMNS

203

or type of scene in an unexpected way (cf. the accounts of the infancy and youth of the major Olympian gods in the Hymns), by introducing obscure cults or cult-titles or myths, or by wrong-footing the reader through unusual variations in diction. This device is especially effective against the formal, stereotyped background of hymns.’Callimachus shows interest in the attributes of the gods by giving them diverse roles even in single hymns, although his emphasis on their multi-naturedness (linked with geography) appears most startlingly at the opening of Jamb 10: τὰς Appoditag — ἡ θεὸς yap οὐ μία — ἡ Καστνιῆτις τῷ φρονεῖν ὑπερφέρει πάσας

(The Aphrodites — for the goddess is not one — are all less wise than Aphrodite of Kastnion)

where the ‘different Aphrodites’ are those worshipped in different locations.’ Jamb

10 proceeded to deal with Artemis Kolaenis; since

she is mentioned as the object ofa local cult, she may, like Aphrodite, have been discriminated geographically from other *Artemises'. In Hymn 4 ‘geography’ is used to to give Artemis unusual attributes by removing her from her traditional sphere, the ‘country’, and attaching her to the ‘city’. The opening lines (1-3) give Artemis her traditional attributes and activities; the bow, hunting, dancing and playing games on the mountains. We then hear more about these from an amusing angle; the young Artemis sits on her father’s knee and asks for them (6ff.). Virginity tops her list, followed by the bearing of many names, hunting and its paraphernalia, and an escort of nymphs. After demanding “all mountains” she refuses to choose a particular city (18f.), because, she says, she will stay in the mountains and rarely come to town. Only when she comes to help women in childbirth will she visit cities (20-3), and this solely because it is a duty imposed on her by the Fates at her birth. Thus Artemis apparently repudiates further association with cities. After briefly agreeing to all she asks (31f.), Zeus offers her more: thirty cities of her very own (33-5) and others to share with other gods (36-9). Although Zeus complies with his daughter’s wishes, he shows more interest in making her a goddess of cities than in allowing her to be a goddess of the wild, which is how she thinks of herself.? Instead of giving her freedom to range the mountains, she is made protectress of cities (πόλεις) with the trappings of cult (38) in a passage similar to the end of the hymn to Athena. The novelty of having Zeus assign cities to Artemis' care becomes

204

VIRGINIA KNIGHT

clearer when the scene is compared both with Homer’s accounts of Artemis and with the main model for Hymn 3, a fragmentary Aeolic

hymn to Artemis variously ascribed to Sappho (fr. 44a Voigt) or

Alcaeus (fr. 304 L-P). Artemis is not prominent in Homer, but her attributes and behaviour at Hymn 3.1ff. recall Iliad 21.470-513, where she flees to Zeus from the battlefield, and Odyssey 6.102-8, the

simile in which Nausicaa is exploited by Callimachus situation owes much to 7/iad in Callimachus. Odyssey 6

compared to her.'° These passages are in complementary ways. The overall 21, where Zeus humours his daughter as supplies the details of the activities

Artemis asks for: dancing with her nymphs (13-15, Odyssey 6.105f.) on the mountains (18, Odyssey 6.102f.), and hunting (8-12, Odyssey 6.102-4). More generally, the many names which Artemis requests remind us of her numerous epithets in Homer, as well as suggesting

that this goddess has many aspects. In the fragmentary Aeolic hymn to Artemis she swears to remain a virgin and live on the mountains; Zeus simply assents in Homeric fashion with a nod, but apparently

does not speak. Zeus' speech to his daughter in Callimachus is therefore an innovation in two different ways: it is an addition to the scene, and it gives additional attributes to Artemis.'! It must be stressed, however, that there are precedents in earlier hymns for associating Artemis with cities as well as with the wild. The areas Callimachus assigns to Artemis in Hymn 3.1ff. expand those

given to her in Homeric Hymn 5.16-20 (to Aphrodite); hunting, music, dancing, groves, and finally the cities of just men. The context of this passage is Artemis' perpetual virginity, which in Callimachus

is her first request to Zeus. The list of Artemis' interests here becomes the plan of Callimachus' hymn to her, which deals successively with hunting (98-169), dancing (170-80) and cities (251-8). The hymn to Artemis by Anacreon (fr. 348 Page) also blends the wild and the civilised attributes of Artemis; in it the goddess cares for both the ἀγρίων ... θηρῶν (wild beasts) and the οὐ ... ἀνημέρους ... πολιήτας (civilised citizens) of Magnesia.'? Only its opening survives, leaving us uncertain whether it continued to play on the contrasting sides of her nature. In Hymn 3 Zeus’ emphasis on cities (33-6), islands (37), roads (38) and harbours (39), and his silence about mountains, do not have an immediate effect on Artemis. After leaving him, she promptly tours areas he has not referred to: a wooded mountain on Crete (41), Oceanus (42), Hephaestus' underground forges on Lipara (46-88), Pan's cave in Arcadia (87-97), and finally another mountain (99),

LANDSCAPE AND THE GODS IN CALLIMACHUS' HYMNS

205

Parrhasion, where she hunts. The description of her hunting leads

eventually to an account of how she tries out her bow against various targets (120-4); and here Artemis is again associated with cities, since

an unjust city abruptly appears as the last of her targets (122). Artemis does not usually defend justice, though she is concerned for it in Homeric Hymn 5.20.? Callimachus is emphasising here a nontraditional characteristic of the goddess. He then moves swiftly to the corollary of her punishment of injustice, her blessings on her favoured cities (129ff.). Artemis' concern for human society fulfils

the promises made to her by Zeus earlier in the hymn; she has been brought in from the wild and she blesses or blights the lives of men and the produce of their fields.

Although cities are not mentioned after 122, they remain prominent in the reader's mind because the contrast of just and unjust

communities recalls a passage of Hesiod (Works and Days 225-47).'* In both poets the difference between the two types of city affects the fertility of crops, animals and human beings. In Hesiod the activities are those of Zeus (Works and Days 229, 239, 242, 245, 247); in

Callimachus, Zeus has in effect delegated them to his daughter (cf. the case of Athena at Hymn 5.131-6). By reworking a famous passage of Hesiod and transferring to Artemis the function of rewarding justice, so characteristic of Zeus in Hesiod, Callimachus reinforces

the role given

to Artemis

in the opening

scene.

Later (183-8)

Callimachus asks Artemis a series of questions about which places and peoples she likes best. These recap the areas Zeus had assigned to her; islands (cf. 37), harbours (cf. 39), cities (cf. 33-6), as well as mountains, Artemis' own choice (cf. 18), and companions (cf. 13-17). Artemis’ favourite city is revealed as Perge in Asia Minor.'*

This passage is followed by a lengthier description of Artemis' relationships with some of her favourite hunting companions (191-224), although it is made clear that these relationships are all in

the past. Towards

the end

of the hymn,

Callimachus

recalls

in 'ring-

composition' points made in the opening scene. At 225f. some of the names Artemis asked for in 7 are listed, among them πολύπτολι (Lady of many cities), referring to Zeus' promise of the thirty cities. She is then addressed as Μιλήτῳ ἐπίδημε (dweller in Miletus), which implies not just her favour to that city but also her physical presence in it; and there follows a brief list of places with which Artemis is

associated, ending emphatically with the Argive temples built by Proitus. The myth connected with them is told at length, in contrast

206

VIRGINIA KNIGHT

with the brief allusive introduction of myths linked with the other

places. In the variant narrated by Callimachus, Artemis tamed Proitus’ daughters as they rampaged over the mountains, and brought them back home. Although, of course, the goddess of mountains is best placed to control people on mountains, Artemis behaves in a surprising way by supporting civilised values against the

call of the wild. Callimachus’ use of this variant is all the more significant because in other versions not Artemis but Melampus performed the cure.! Typically for Callimachus, the tale is given an aetiological twist and explains the origin of the two temples to Artemis. A similar implication recurs in the penultimate scene of the hymn (251-8), which narrates Artemis' protection of Ephesus against the Cimmerians.!? In this way by the end of the hymn Artemis has matured into the role chosen for her by Zeusat its beginning. She now

protects

cities and

their

inhabitants,

and

is concerned

for

communal values such as justice. Why should Artemis have been selected for such treatment? Indeed why did Callimachus write a hymn to Artemis at all, since

there was little devotion to her in Alexandria? Part of the answer may be that Artemis is more closely associated with a particular environment than most other deities, so that Callimachus could achieve a more striking poetic effect by manipulating that association. It is noticeable that Callimachus virtually ignores one of Artemis' more important roles in Greek life, her patronage of transitions in the life-cycle of women, introducing only her grudging acknowledgment that she is obliged to help women in childbirth (20-3). Instead we see Artemis herself changing and gradually taking on the duties her father has assigned to her.? It is this developing

relationship between the goddess and different kinds of landscape, and in particular the growing closeness between Artemis and cities,

which unify a poem at first sight loosely structured. III. Landscape and Deities in other Hymns Artemis’ twin Apollo, celebrated in Hymn 2, is less tied to particular scenery than she is; one of his main mythical activities is founding cities?!

but he has also served

as a shepherd.

These

aspects are

combined at the heart of Hymn 2 — Apollo as city-builder (55-64) and as Nomios (47-54). Apollo undertakes his first construction venture, the altar on Lemnos, when he is four years old. Artemis makes her only appearance in the hymn here, helping her brother

LANDSCAPE

AND THE GODS IN CALLIMACHUS' HYMNS

207

(58f.). The altar is described in terms appropriate to building a city

(δείματο ... ἐδέθλια, ὑπεβάλλετο τοίχους, θεμείλια).22 It is, then, a *trial run' for the big cities Apollo will build when he grows up, and Artemis' assistance implies that she too is concerned with cities. The

sharing of roles is reciprocal since Apollo, like Artemis, can give fertility to flocks (50-4, cf. Hymn 3.130f.). Hymn 2 continues appropriately with Apollo's city-founding; his favourite is Cyrene

(65-8), which he helps Battus to build, and he also founds Sparta and Thera (72f.). Apollo's love of cities becomes an aspect of his relationship with Callimachus, a native of his ‘favourite’ city. Hymn 2 thus resembles Hymn 3 in stressing that its deity is concerned with

cities. However, that emphasis is less striking in the hymn to Apollo since he is the city-founding god par excellence. Nevertheless, Apollo

and Artemis are presented in Hymn 2 in a way which complements the depiction of Artemis in Hymn 3. Hymn 1 (to Zeus) shows, like Hymn 3, a god making the transition

from wilderness to civilisation. Zeus is born in the wildest possible place,

a mountain where the thickets are deepest (11, 31), and he is

brought up on another (51). Even the alternative story of his birth which Callimachus rejects locates it on a mountain (6). After Zeus is born Rhea makes rivers miraculously spring up; thus the landscape reverts to a wild state, as river beds which once carried carts flood

with water (22-7). Zeus is assigned his sphere of activity, control over kings, at 73. Zeus is normally 'king of kings'; but the unusual πτολίαρχοι (city-rulers), instead of ἄνακτες or βασιλεῖς (kings), stresses Zeus' links with cities. A little later Callimachus relates the duties of kings firmly to the city: δῶκας δὲ πτολίεθρα φυλασσέμεν, Teo δ᾽ αὐτὸς, ἄκρῃσ᾽ £v πολίεσσιν (You gave them cities to guard and yourself sit on the city heights, 81f.). The progress of Zeus from his birthplace in the wilds to the city acropolis parallels Artemis’ progress in Hymn 3. The transition is less surprising here because we expect Zeus to be more concerned with cities than Artemis; indeed, it

is the responsibility for cities and for upholding justice which Zeus delegates in part to Artemis in her hymn. Hymn 6 (to Demeter), like Hymn 3, brings closely, but more discreetly, into contact with cities a goddess whose domain is elsewhere.

Demeter

does

not

control

the

urban

space,

but

the

cultivated land just outside it; nor does she belong to wild places. But Hymn 6 emphasises her concern for cities: as Demeter Θεσμοφόρος she gives them laws (18); and in 134f., adapted from Homeric Hymn

13.3 (to Demeter), she extends her protection to them. But this

208

VIRGINIA KNIGHT

concern does not detract from her usual interest in crops, dealt with

at 19-21 and

takes

place

celebration

121-3. The processional rite described in the hymn

in a city, which of Demeter

was not unusual?!

Callimachus’

as a benefactor of cities, like his parallel

treatment of Zeus and Apollo, involves no striking innovations; but it gains added significance within the overall context of the Hymns. Hymn 5 (to Athena) reverses the process seen in other Hymns. Athena, more than other deities, is guardian of cities and especially citadels,?* and Callimachus by various means displaces her. First, he both acknowledges and rejects one of the goddess’ usual titles: Hymn 5 is set during a purported procession with Athena's image at a temple — almost certainly that of Athena Polias at Argos. But Callimachus does not introduce this cult-title, thus suggesting but ignoring her patronage of a city. Then, at 35-42, the myth of the Palladion associates Athena with mountains. Eumedes escaped from Argos with this image of Athena and established it on the Kpeiov ὄρος (40), the rocky crags subsequently called *Pallatides'. Kpeiov ὄρος is repeated in 41 to emphasise the location. Next, Athena is invoked at 43 as a destroyer of cities (περσέπτολι), an attribute which follows from her role as war-goddess (43f.), but also alienates her from city life. Even the water which washes her comes from the mountains (49-51).

In the principal myth of Hymn 5 Athena, protectress of many cities and guardian of civilisation, is shown bathing in the wild. She behaves like Artemis in Hymn 3, ranging over woods and mountains (60-7) accompanied by friends, some mothers of famous mythical figures (37-9, cf. 3.209-14). She haunts mountains (72, 74), favourite places of Artemis (Hymn 3.20, 30), and although she herself does not hunt, Teiresias does (75, 91f.). The parallelism between Athena and Artemis is finally made explicit when Athena compares her own treatment of Teiresias with Artemis’ punishment of Actaeon (107-18),

which in mythological time lies in the future. Callimachus may have adapted one story to increase its similarity to the other.?$ Hymn 5 has been compared to Hymn 6 in its structure and narrative pattern; but, as shown, it can also be contrasted with Hymn 3. The goddesses

exchange attributes: Artemis defends cities while Athena punishes a hunter. Finally, Hymn 4 (to Delos); as noted, it contains a wealth of geographical and topological interest, to which only a separate study could do justice. But its main contrast, between land and sea, deserves a few words here. Callimachus narrates the wanderings of

LANDSCAPE

AND THE GODS IN CALLIMACHUS' HYMNS

209

Leto, expanding the catalogue of Homeric Hymn 3.30—48 (to Apollo)

into a series of encounters.?? Leto is finally welcomed by a place that defies geography itself, the floating island Asterie (Delos); but as

Delos accepts the goddess its independence ends and it takes its place on the map. Asterie thus ‘becomes’ a location as a consequence of contact with the gods: the birthplace of Apollo and Artemis must have a geographical position. In a sense landscape in Hymn 4 turns into narrative, and the extreme way Callimachus plays here with geography further argues for its importance in the other Hymns. IV.

Conclusions

It might be tempting to see in the Hymns’ relocations of gods in cities a reflection of the large conurbations of Callimachus' time. However πόλις need not imply size; and in any case the cities portrayed in the Hymns are those of classical Greece and Callimachus takes no

account of the changed circumstances of the hellenistic period. The same conservative tendency appears in the hellenistic epigrams on

cities.2? What the gods’ relationship with cities does show, however, is that they are closely concerned with everyday human life and with the values of the πόλις; when Callimachus places the city-goddess

Athena in the country, he piquantly counterpoints this tendency. Again it shows that the gods have many interests and are not simply personifications of aspects of human existence. Finally it reveals how

the Olympians themselves are interdependent. If the Hymns somehow comment on one another, should this affect our view of their composition? It is commonly assumed that the Hymns were written at different times. Their structural similarities and shared themes do not in themselves prove that they were written in the same period, since Callimachus could perfectly well have

developed or recalled in a later hymn ideas already hinted at or mentioned in an earlier one. But the similarities between the Hymns imply at least their literary unity, so that each should be read with an

eye to the others. Other themes too might with profit be followed through the Hymns — and indeed through the rest of Callimachus" work.’

NOTES 1.

See, however, e.g., A.W.

Bulloch ‘The Future of a Hellenistic Illusion: some

observations on Callimachus and religion’ MH 41 (1984) 209-30; some formal

210

VIRGINIA

KNIGHT

comparisons are made in E.L. Bundy ‘The Quarrel between Kallimachos and Apollonios’ CSCA 5 (1972) 39-94. Hymns 5 and 6 arecompared in A.W. Bulloch Callimachus: the Fifth Hymn. Edited with an Introduction and Commentary

(Cambridge classical texts and commentaries 26, Cambridge 1985), on Hymn 5.51f.; K.J. McKay The poet at play: Kallimachos, the Bath of Pallas (Leiden 1962) 106-24; C.W. Müller Erysichthon (Stuttgart 1987) 46-64. Hymn 3.138-41, in which the poet outlines his programme, could describe the hymns as a whole; the references to Apollo and to Leto suggest Hymns 2 and 4.

E.g. Ap. Rhod. Arg. 2.728-45 and presumably in the Aetia, passim; G. Zanker Realism in Alexandrian Poetry

(London

1987) 115-20.

Works entitled Περὶ τῶν £v τῇ οἰκουμένῃ ποταμῶν (On the rivers of the world) and Κτίσεις νήσων καὶ πόλεων καὶ petovopaciat (Foundations of islands and cities and their changes of name) and a paradoxography arranged geographically

(O. Schneider, Callimachea II (Leipzig 1873) 323-50).

K. Hartigan The Poets and the Cities: Selections from the Greek Anthology about Greek Cities (Meisenheim

1979).

M.I. Finley 'The Ancient City: from Fustel de Coulanges to Max Weber and Beyond' Comparative Studies in Society and History 19 (1979) 305-27; F. de Polignac La naissance de la cité grecque (Paris 1984). A. Ludwich Homerischer Hymnenbau (Leipzig 1908); R. Wünsch RE 9 (1916) 140-83 s.v. Hymnos; K. Keyssner Gottesvorstellung und Lebensauffassung im griechischen Hymnus (Stuttgart 1932); A.H. Meyer Hymnische Stilelemente in der

frühgriechischen Dichtung (Diss. Würzburg (Stuttgart

1933); E. Norden Agnostos Theos

1956) 143-76; W.H. Race ‘Aspects of Rhetoric and Form in Greek

Hymns' GRBS 23 (1982) 5-14.

Cf. Cic. DND 3.21-3, where it is claimed that different gods bearing the same name can be distinguished by their descent.

K.J. McKay ‘Mischief in Kallimachos' (1963) 243-56, 246 and n.2. 10.

This simile

is used

more

directly

Hymn

to Artemis’ Mnemosyne 4.16

by Apollonius

(Arg.

3.876-86).

Although

Artemis' virginity is not specifically mentioned in Homer, it is implied in her similarity to Nausicaa. On the use of Homer in this scene, H. Herter *Kallimachos und Homer; Ein Beitrag zur Interpretation des Hymnos auf Artemis' Xenia Bonnensia (Bonn 1929) 50-105 (Ξ Kleine Schriften

371-416) 62f.

(Munich 1975)

For a blending of a hymn by Alcaeus with Homer in Horace Odes 1.10, cf. F. Cairns “Alcaeus’ Hymn

to Hermes, P.Oxy.

QUCC n.s.13 (1983) 29-35.

12.

2734 fr. 1 and Horace Odes 1.10"

On devotion to Artemis in Magnesia in the third century, cf. J. Ebert 'Zur Stiftungsurkunde der AEYKO®PYHNA in Magnesia am Mäander’ Philologus 126 (1982) 198-216.

13.

Bornmann

14.

H. Reinsch-Werner Callimachus Hesiodicus (Berlin 1976) 74-86, esp. 74.

15.

On Artemis at Perge, B. Pace ‘Diana Pergaea' Anatolian Studies presented to Sir William Mitchell Ramsay (Manchester 1923) 297-314, R. Fleischer Artemis von Ephesos und verwandte Kultstatuen aus Anatolien und Syrien (Leiden 1973) 233-54.

on 122-35.

LANDSCAPE

16.

AND THE GODS

πολύπτολι

IN CALLIMACHUS HYMNS

also occurs at Hymn

4.266, where

211

the island of Delos,

Artemis’

birthplace, is describing herself. It is unattested elsewhere, except in Philo 2.541 in a different sense. ἐπίδημος is only attested in literature in this sense in Callimachus, the other occurrence (fr. 75.26 Pf., from the story of Acontius and

Cydippe in the Aetia) also being used of Artemis. 17.

Hdt.

9.34,

Apollod.

2.22. The

closest version

to Callimachus

is Bacchyl.

10.95-109, where Artemis intercedes with Hera. According to the Suda, Theocritus wrote a poem on the subject, but we know no more of it than the title (A.G. Gow Theocritus 1.xxiv).

18.

The

vocabulary

describing

her

opponents

associates

them

with

animals;

ἱππημολγῶν (mare-milkers) (252), a word normally used of Scythians (cf. 1].

13.5, Hes. fr. 150.15), and βοὸς πόρον (ox-ford) (254). The Cimmerians areshot

down in the same way as animals are targetted earlier in the hymn.

19. 20.

P.M. Fraser Prolemaic Alexandria (Oxford

1972) 1.195f.

For a summary of Artemis' activities in Greek religion and mythology, cf. W. Burkert Greek Religion (Oxford

1985) 149-52.

21.

P.B. Schmid Studien zu griechischen Ktisissagen (Freiburg 1947) 154-67.

22.

Apollo is a wall builder in //. 7.452f. and Theogn. 773.

23.

On processions in Greek religion, M.P. Nilsson ‘Die Prozessionstypen griechischen Kult' Opuscula Selecta ΠῚ (Lund 1951) 166-214.

24.

M.P. Nilsson Geschichte der griechischen Religion (Munich

25.

Bulloch (Commentary, n.1) 15f.

26.

L. Malten Kyrene (Berlin 1911) 34; Dodds on Eur. Bacch. 337-40. In earlier literature Actaeon was punished for desiring Artemis (Stes. fr. 236 Page) or for

boasting of his skill at 27.

1941) 1.405-10.

hunting (Eur. /oc. cit.).

N. Hopkinson Callimachus: Hymn to Demeter. Edited with an Introduction and Commentary (Cambridge classical texts and commentaries 27, Cambridge 13-17; McKay (n.1), 106, 113.

28.

im

1984)

P. Bing The Well-Read Muse (Góttingen 1988) 115-21. On the catalogue in the Homeric Hymn, A. Miller From Delos to Delphi: a Literary Study of the Homeric

Hymn to Apollo (Leiden 1986) 31-4. Hartigan (n.5) 106.

This paper is based on one given in Cambridge and Leeds in 1990. I would like to acknowledge the helpful comments of Francis Cairns and others on earlier versions.

PAPERS OF THE LEEDS INTERNATIONAL LATIN SEMINAR Seventh Volume (1993) 213-19 Published by Francis Cairns (Publications) Ltd (Leeds 1993). ARCA 32. ISBN 0-905205-87-1

DIONYSIAN RITUAL OBJECTS IN EUPHORION AND NONNUS C. ANNE

WILSON

This paper aims to elucidate two short literary descriptions of the use of the φιάλη in Dionysus rituals. They are: (i) Nonnus Dionysiaca 9.125f.: xai φιάλας γυμνοῖσιν ἐπὶ otépvotoi καθάψαι χαλκείας ἐνόησε καὶ ἰξύι δέρματα νεβρῶν'

In an account of the mystical practices taught to the infant Dionysus by Mystis, Nonnus says that it was she who first thought of fitting bronze φιάλαι over the naked breasts. No other author mentions this practice, and there is no iconographic evidence for it. It is unlikely to be Nonnus' own invention, so it must be ascribed to one of his several lost sources. (ii) Euphorion fr. 13 Powell: ἐν πυρὶ Βάκχον δῖον ὑπὲρ φιάλην ἐβάλοντο.

Referring to the story of the Cretan

Dionysus, son of Zeus and

Persephone, who while still a child was torn apart by the Titans and then boiled, Tzetzes told how the Titans them to his brother lost Dionysus poem

reports that both Callimachus and Euphorion threw Dionysus’ limbs into a cauldron and gave Apollo. Tzetzes then quotes from Euphorion’s this single-verse fragment.!

I shall argue that these lines are elucidated by a ritual practised in some of the newer cults which arose under the Roman Empire; this ritual depended on secret technological knowledge in which the φιάλη played a part. The first part of the paper sets out the technical 213

214

C. ANNE WILSON

and ritual background; the second part considers the interpretative

implications. 1. The ritual background Book

4 of the second-century AD Refutatio Omnium Haeresium

ascribed to Hippolytus contains an attack on the tricks and crafty practices of the magi and gnostics, and gives details of some of these tricks. Diels recognised in Chapter 31 a recipe for the distillation of wine.? The ingredients are sweet wine and salt, and they are ‘boiled’ together (the verb Éyetv denotes distilling both here and in some of

the protochemical recipes collected by M. Berthelot).? If the product of this process is heated again and is set alight with a lamp, then it will

burn safely upon the head. It catches fire more readily if manna is added to it as it burns, and it does so better still if some sulphur is applied to it.* When the Refutatio speaks of the product of distillation burning safely on the head, parallels show that it is referring to a ‘baptism’

ritual practised in a number of cults: (i) Several of the so-called tricks described in the Refutatio are thought to derive from Anaxilaus of Larissa. He was a neo-

Pythagorean who spent some years in Rome and who was banished from Italy in 28 BC on a charge of practising magic. He may have recompiled the protochemical recipe-book of Bolos-Democritus, as Diels and Wellmann believed. He was apparently also the author of

a book which contained recipes for magic lamps and wine-distilling, which Irenaeus probably had in mind when he wrote of the Paignia or

‘Playtricks’ of Anaxilaus used by Marcus the gnostic (Adversus Haereses 1.13.1). Marcus' followers baptised initiates on the head with *water and oil, by contrast with contemporary Christian

baptism which required total immersion in water (Adversus Haereses 1.21.4). The word ‘water’ in the context of gnostic initiation may have the special significance of ‘distillate’ which it has in the protochemical texts, and which was to pass via medieval Latin into modern western European languages (e.g. rosewater, eau de cologne,

Kirschwasser).® (ii) The followers of Simon Magus had a form of baptism in which **... fire appears straight away upon the water, whether this is done by some trick which, as many affirm, is a trick of Anaxilaus ..." (Pseudo-

Cyprian De Rebaptismate 16). Pseudo-Cyprian specifies that the fire

.

DIONYSIAN

RITUAL OBJECTS IN EUPHORION

AND NONNUS

215

appears directly after the initiand has entered the waterofthe river or pool where the ceremony is being conducted." (iii) Tertullian too was aware that heretical baptism was performed upon the head (De Praescriptione Haereticorum 40), and in a curious

passage he contrasts the heretics, who summon a

spirit of such

brightness (tantae claritatis) into the water by human ingenuity, with

the Christians who draw sublime spiritual music from God's organ, i.e. the water-organ (De Baptismo 8). The mention ofthe water-organ

here implies a contrast with the ‘fire-organ’ of the heretics; and it is interesting to find that ὄργανον is one of the names given to distilling apparatus in the protochemical texts.* The origins of that ὄργανον (= distilling apparatus) and of the

technique of wine-distilling could be very early. Once it was realised that the active, fiery principle of wine (the alcohol) escaped irretrievably if the wine was boiled in an open container, the question must have arisen how that principle might be captured. If the wine

was heated slowly in a vessel with an inverted bowl or cup fixed over its mouth, then the steam which condensed on the inner side of the

bowl in the early stages of the operation would have been found to be even richer in alcohol than the original wine. Collection of the condensed liquid would have been difficult, but not impossible if a succession of cups was used, each yielding a very small amount, and

each being kept as cool as possible to facilitate condensation of the steam.? Eventually an outlet tube was invented, attached near the mouth of the inverted cup and joined at its further end to a receiving vessel; this enabled a greater proportion of the spirits of wine to be collected. Simple stills of this type could well have been the prototypes from

which was developed the ‘Hellenistic still’. Its stillhead was the glass μαστάριον, or breast-shaped cup, with an inturned rim to which was fixed an outlet tube to carry off the condensed vapour into a receiver.'? Although modern historians of science have named the

still equipped with the glass μαστάριον the ‘Hellenistic still’, it is really a *late-Hellenistic' still: no metal predecessor has yet been recognised, and the glass μαστάριον is unlikely to have predated the

invention of glassblowing about the mid-first century BC.!! Descriptions and drawings of even more sophisticated distilling apparatus have come down to us in the encyclopaedia of chemical

recipes compiled by Zosimus about AD

300. Here are recorded

ingredients and methods used by groups founded in Egypt at least

216

C. ANNE WILSON

two or three hundred years earlier, especially the followers of the selfstyled ‘Democritus’, Bolos of Mendes, and of Maria the Jewess. The

text of the earliest surviving manuscript of this encyclopaedia, the tenth-century Marcianus 299, was published by Berthelot.'? It describes a still called the τρίβικος, said to have been devised by Maria the Jewess. An earthenware base-vessel was joined to a wide rising tube of bronze, the σωλήν. To this was fixed the stillhead, also of bronze and named χαλκεῖον. Three narrow tubes led from the χαλκεῖον to the βῖκοι or receiving vessels. The δίβικος was similar, but with two βῖκοι instead of three. An example appears as part of a composite drawing called ‘the gold-making of Cleopatra’ on the verso of folio 188 of the manuscript. Here the name given to the stillhead is φιάλη." Since this almost globular stillhead is clearly not of φιάλη shape (unlike the uactáptov which, as its name implies, was a breast-shaped cup), it may be deduced that φιάλῃ had become a general term for a stillhead over a long period before the appearance of the tpiBixoc, whose supposed inventor, Maria the Jewess, is

believed to have lived in the first century AD.'* From the texts and the drawings, then, we learn that a stillhead

could be called either ‘bronze’ (χαλκεῖον) or φιάλη; while a third term for it, μαστάριον, was applied especially to the stillhead of the *Hellenistic

still’, which

had

an inturned rim to help collect the

distillate. As explained above, a wine-distilling recipe and the ritual use of alcoholic fire among gnostics were linked to the name of Anaxilaus, the neo-Pythagorean and compiler of recipe collections. Dillon has suggested that Anaxilaus actually invented the technique of wine-distilling.? But Anaxilaus transmitted earlier material elsewhere, and he could have done so in the case of the wine-distilling

recipe; and if a long-established tradition already existed within the Greek world of an ‘unspeakable’ initiation rite employing the fire within distilled wine, then the most likely context for it would have

been the mysteries of Dionysus. 2. The literary descriptions Bearing this possibility in mind we can return to Nonnus' phrase about Maenads fitting bronze φιάλαι over naked breasts. The φιάλαι have puzzled modern scholars. Eitrem believed that they were bronze

cups or cymbals from which the Maenads drank milk, and A.B. Cook thought that they were cups used for making libations of milk. G. Chrétien suggested that they were large pendant ornaments, and

DIONYSIAN RITUAL OBJECTS IN EUPHORION

AND NONNUS

217

R. Turcan that they could be small bells.'® I suggest that Nonnus took his account of Mystis and the sacred

objects from an earlier source, and that he misunderstood a phrase in that source which defined the bronze φιάλη as a μαστάριον. For all that the cultured elite of Panopolis in the fifth century AD clung to their pagan literary heritage, Nonnus lived at a time when the gnostics had been banished beyond the frontiers, and when pagan cults were in terminal decline. Hence it is unlikely that he knew anything about ritual wine-distilling. He therefore interpreted μαστάριον as a covering for the breast. But his source was probably writing about a breast-shaped φιάλῃ stillhead for the bronze version of the ‘Hellenistic still’, or for its more primitive predecessor. That source could have been Euphorion himself or a contemporary of Euphorion. But since rituals are persistent, a writer much later than Euphorion could have been Nonnus' source, if he was referring to an

object which had originated far in the past. In Euphorion's verse about Bacchus we again encounter the φιάλῃ. The translation of this line has caused difficulties. Lobeck, Meinecke and Scheidweiler emended φιάλην to φιάλης, which yields the reading “On the fire they threw godly Bacchus over a φιάλη᾽". It makes little sense; a φιάλη (in the sense ‘drinking-cup’) is not put onto a fire, and the god's limbs had already been thrown into a cauldron (λέβης), a vessel which is suitable for setting on the fire." The words ὑπὲρ and φιάλην are in all the manuscripts, but there are variants in other words, e.g. Bákyav (-οις, -ac); δίαν (δῖαν, δῖον), perhaps reflecting perplexity over meaning as well as scribal error.

The verse links Bacchus with the φιάλῃ and an activity performed over a fire. If ὑπὲρ expressing motion is allowed to govern a noun in the accusative case (rarer than the genitive usage, but not impossible),

two interpretations are possible: either Bacchus is thrown over the φιάλη; or the φιάλη is ‘thrown’ (i.e. ‘put’) over Bacchus. The second interpretation (assuming poetic hyperbaton with ὑπέρ following its noun) would involve a hint at the φιάλη stillhead, and ‘on the fire’ would imply that Bacchus was boiled. As noted, ‘to boil' can also mean ‘to distil' (see above, p.214 and n.3). The verse could thus refer to Bacchus in the form of wine, enclosed in a vessel fitted with a φιάλη stillhead and placed upon a fire. But the uncertainty created as to its subject matter could go back to Euphorion himself, writing of the ‘unspeakable’ in terms which had one meaning for the initiated and another for the non-initiated. The possibilities suggested by the above account are extremely

218

C. ANNE WILSON

interesting.!* The next development could lie with the archaeologists,

whose discoveries may in due course cast further illumination on these and other texts connected with the mysteries of Dionysus.

NOTES 1.

2.

"Tzetzes In Lycophronis Alexandram 208 = Callimachus fr. 643 Pf.; Euphorion

fr.

13 Powell (= 15 Meinecke; 12 Scheidweiler; 13 de Cuenca; 14 van Groningen). I.M. Linforth The Arts of Orpheus (Berkeley 1941) 309f.

Cf.

Ἡ. Diels Die Entdeckung des Alkohols (Abhandlungen Preuss. Akad. Wiss., phil.-

hist. Kl. 3, 1913) 21-5, 35.

3.

M. Berthelot Collection des anciens alchimistes grecs (Paris 1887-8) 11.3.47.7 etc.

4.

Thesulphur probably date: sulphur burned 3566, at which point could be ignited and Alkoholrezept aus dem

refers to the sulphur test, devised at an early but unknown in distilled spirits of wine reduced its alcohol content to the sulphur itself was extinguished. Thereafter the liquid would burn safely on cloth or hair. See H. Degering Ein 8. Jahrhundert (Sitzungsber. Preuss. Akad. Wiss., phil.-hist.

KL 1917) 513; further discussion and examples in C. A. Wilson Philosophers, losis

and Water of Life (Proceedings of the Leeds Philosophical and Literary Society 19.5, 1984) 47f. 5.

Diels (n.2) 24; M. Wellmann Die Physika des Bolos Demokritos und der Magier Anaxilaos aus Larissa (Abhandlungen Preuss. Akad. Wiss., phil.-hist. Kl. 7, 1928)

56f.; but contra, R. Halleux Les Alchimistes grecs 1 (Paris 1981) 69. For a fuller

discussion of the evidence given below, see Wilson (n.4) 56-9.

6.

Berthelot (n.3) I1.2.1.29; 3.25.1, etc.

7.

The word 'water' appeared was not just stepped; and been helpful to technology.

8.

See e.g. Berthelot (n.3) 11.3.10.1; 3.47.2; 3.49.15.

9.

Zosimus (ca AD 300), in Berthelot (n.3) II.3.47.2, advocated the use of a sponge and a krater of water to cool the upper part of the more sophisticated distilling apparatus of his day.

10.

Berthelot (n.3) 11.3.8.1 line 10 explains that the outlet-tube of the naotáptov is firmly luted to the receiving vessel (ῥογίον).

ll.

E.g. A.R. Butler and J. Needham

is used twice in this passage, but the *water' on which the fire necessarily the same as the water into which the initiand had the specialised usage of *water' to mean distillate would have the gnostics who wished to preserve the secrecy of their

‘An experimental comparison of the east

Asian, Hellenistic and Indian (Ganharan) stills ...” Ambix 27 (1980) 67-76, who

recreated the μαστάριον stillhead and succeeded in producing spirits of wine with an alcoholic content of nearly 60%. 12.

Details of this and of two later MSS Berthelot (n.3) 1.173-210.

in the Bibliotheque Nationale, Paris in

13.

Berthelot (n.3) I1.3.50.1. The drawing which the words τὸ δὲ σχῆμα τοῦτο, is not present diagram has been placed on the verso of the 1.139). The diagram of ‘the gold-making

was in the original text, signalled by there in MS Marc. 299, but a stylised leaf, f. 194 (reproduced at Berthelot of Cleopatra’ which includes the

DIONYSIAN RITUAL OBJECTS IN EUPHORION

AND NONNUS

219

δίβικος with the stillhead labelled φιάλη is on the verso of f. 188, and is reproduced at 1.132.

14.

F.S. Taylor *A survey of Greek alchemy' JHS 50 (1930) 116.

15.

J. Dillon *Greek alchemy' CR 36 (1986) 36f.

16.

S. Eitrem RE 16.2 (1935) s.v. 'Mystis', following E. Maass Orpheus (Munich 1895), A.B. Cook Zeus? (Cambridge 1925) 346f.; G. Chrétien Les Dionysiaques

IV (Paris 1985) 110, note to 9.125; R. Turcan Sarcophages (Paris 1966) 548-51.

17.

A.B. van Groningen, describing the text in his edition (Amsterdam 1977) 39f. as

“trés corrompu”, recognised that the meaning produced by emending φιάλην to φιάλης did not make sense. His solution, following O. Müller, was to emend to ὑπερφίαλοι (of the Titans), and to divide the verse between two lines. 18.

Another possible manifestation is the ‘bronze’ carried by the Maenads in Eur. Ba. 755-7, in juxtaposition with the fire which burned harmlessly on their hair (757f.). Cf. also Hypothesis to Eur. Medea, on Medea ‘boiling’ the ‘nurses’, i.e. Maenads, in Aesch. Trophoi in order to make them young, which may hint at an initiation ritual for Maenads involving ‘boiling’ or distilling.