Lucretius and His Sources: A Study of Lucretius, "De rerum natura" I 635-920 311019452X, 9783110194524

Table of contents :
Introduction
Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic
1.1 Lucretius’ information is second-hand
1.2 Lucretius’ source was an Epicurean text
1.2.1 Lucretius’ use of homoeomeria
1.2.2 The choice of Heraclitus as representative monist
1.2.2.1 The Stoics as fire monists?
1.2.2.2 The Stoic denial of void in the world?
1.2.3 Lucretius’ arguments against the limited pluralists
1.2.3.1 Lines 753–781
1.2.3.2 Lines 782–802
1.2.4 The Epicurean angle
1.3 Conclusion
Chapter 2. Books XIV and XV of Epicurus’ Περὶ φύσεως
2.1 The content of books XIV and XV
2.1.1 Book XIV was not dedicated to polemic
2.1.1.1 Evidence from the format of PHerc. 1148
2.1.1.2 Columns I–XXII
2.1.1.3 Columns XXIII and XXIV
2.1.1.4 Evidence from the sezioni
2.1.2 Epicurus did not discuss Heraclitus’ theory ΠΦ XIV
2.1.3 Epicurus did not refute Empedocles’ theory in ΠΦ XIV
2.1.4 Book XV was not dedicated to criticism of Anaxagoras
2.1.4.1 Cornice 2
2.1.4.2 Cornice 3
2.1.4.3 Cornice 4
2.1.4.4 Cornice 5
2.1.4.5 Cornici 6 and 7
2.1.4.6 Cornice 8
2.2 Other considerations intrinsic to Epicurus’ work
2.3 Do ΠΦ XIV and XV depend on Theophrastus’ Φυσιϰαὶ δόξαι?
2.3.1 Was Plato the last of the limited pluralists in Theophrastus’ Φυσιϰαὶ δόξαι?
2.3.2 The detail of the arguments against Plato and air monism
2.3.3 The dating of ΠΦ XIV and of Theophrastus’ Φυσιϰαὶ δόξαι
2.4 Conclusion
Chapter 3. Lucretius’ use of sources in DRN I
3.1 The source of DRN I.156–598 and 951–1107
3.2 Did Lucretius change source after line 598 of DRN I?
3.3 The Critique does not derive from the same source as 155 ff
3.4 The connection between lines 634 and 635
3.5 Why did Lucretius have the Critique at the centre of book I?
3.6 Was Epicurus the source of the Critique?
3.7 Did Lucretius use a later Epicurean source?
3.7.1 The choice of Heraclitus
3.7.2 Lucretius’ use of homoeomeria
3.8 Conclusion
Chapter 4. Lucretius in the Critique
4.1 Heraclitus as a general
4.2 Heraclitus’ army
4.2.1 Stolidi and inanes Graii
4.2.2 Sound and truth
4.2.3 Inversis sub verbis
4.3 The theme of the path and the search for truth
4.3.1 Lines 657–59
4.3.2 Lines 690–700
4.4 Empedocles and Sicily
4.4.1 Empedocles’ language: poetry as revelation
4.4.2 Lucretius’ praise
4.4.3 Etna
4.4.4 Lucretius’ endorsement of Empedocles’ discoveries
4.4.5 The four elements: Empedocles’ disastrous fall
4.5 Lucretius’ presentation of Anaxagoras’ theory
4.5.1 Lucretius’ transliteration homoeomeria
4.5.2 Parody of Anaxagoras
4.6 The mortality of Anaxagoras’ primordia
4.7 Lucretius’ strategy in lines 859–74
4.8 The analogy of letters and atoms
4.8.1 Lines 823–29
4.8.2 Intertextuality
4.8.3 Lines 906–14
4.9 Formularity
4.10 The parallelism between lines 803–29 and 897–920
4.11 The Critique as ‘dialogue’
4.12 Conclusion
Appendix (A) Two stages of composition?
Appendix (B) The format of PHerc. 1148 and PHerc. 1151
Appendix (C) Do Epicurus’ Ad Herodotum and Ad Pythoclem reflect continuous books of ΠΦ?
Abbreviations
Bibliography

Citation preview

Francesco Montarese Lucretius and His Sources

Sozomena Studies in the Recovery of Ancient Texts Edited on behalf of the Herculaneum Society by Alessandro Barchiesi, Robert Fowler, Dirk Obbink and Nigel Wilson Vol. 12

De Gruyter

Francesco Montarese

Lucretius and His Sources A Study of Lucretius, De rerum natura I 635⫺920

De Gruyter

ISBN 978-3-11-019452-4 e-ISBN 978-3-11-021881-7 ISSN 1869-6368 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.dnb.de. © 2012 Walter de Gruyter GmbH & Co. KG, Berlin/Boston Printing: Hubert & Co. GmbH & Co. KG, Göttingen ⬁ Printed on acid-free paper Printed in Germany www.degruyter.com

V

For Linda

VI

VII

Foreword I have been helped by many people while preparing this book and the PhD thesis on which it is based. My greatest debt is to my former supervisor, Bob Sharples. Many ideas presented here are the result of conversations that I had with him. Alessandro Schiesaro also made suggestions that I have incorporated into the present volume. Richard Janko introduced me to the study of the Herculaneum papyri. Cornelia Römer helped me establish a better text of Epicurus as found in the papyri. I am indebted to David Sedley for comments regarding the restoration and interpretation of some fragments. My mother, Francesca Bugliani, Clover Peake, Dilwyn Knox, Martin Davies, Michael Wigodsky and Tom Preece-Smith commented on various parts of the work. The hours I spent in Naples’ Officina were fewer than I would have ideally liked, but the experience of studying these papyri was as inspiring as it was brief. For their help and courtesy while I was in Naples I would like to thank Agnese Travaglione, Francesca Longo Auricchio, Giuliana Leone, Gianluca del Mastro and Jeff Fish. For advice on the problems relating to the presence – or otherwise – of DRN in Herculaneum I wish to thank Mario Capasso and Dirk Obbink. For help with editorial matters I am grateful to Sabine Vogt. I should also like to thank the University College London Graduate School for the award of a three-year research degree scholarship.

VIII

IX

Table of contents

Table of contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 1 Lucretius drew the Critique from an earlier Epicurean polemic . 1.1 Lucretius’ information is second-hand . . . . . . . . . 1.2 Lucretius’ source was an Epicurean text . . . . . . . . 1.2.1 Lucretius’ use of homoeomeria . . . . . . . . . 1.2.2 The choice of Heraclitus as representative monist 1.2.2.1 The Stoics as fire monists? . . . . . . . 1.2.2.2 The Stoic denial of void in the world? . . 1.2.3 Lucretius’ arguments against the limited pluralists . . . . . . . . . . . . . . . . . . . . . 1.2.3.1 Lines 753–781 . . . . . . . . . . . . . . 1.2.3.2 Lines 782–802 . . . . . . . . . . . . . . 1.2.4 The Epicurean angle . . . . . . . . . . . . . . . 1.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . Chapter 2 Books XIV and XV of Epicurus’ Pλ φ« . . . . . . . . . 2.1 The content of books XIV and XV . . . . . . . . . . . 2.1.1 Book XIV was not dedicated to polemic . . . . . 2.1.1.1 Evidence from the format of PHerc. 1148 2.1.1.2 Columns I–XXII . . . . . . . . . . . . . 2.1.1.3 Columns XXIII and XXIV . . . . . . . . 2.1.1.4 Evidence from the sezioni . . . . . . . . 2.1.2 Epicurus did not discuss Heraclitus’ theory P XIV . . . . . . . . . . . . . . . . . . . . . 2.1.3 Epicurus did not refute Empedocles’ theory in P XIV . . . . . . . . . . . . . . . . . . . . 2.1.4 Book XV was not dedicated to criticism of Anaxagoras . . . . . . . . . . . . . . . . . . . . 2.1.4.1 Cornice 2 . . . . . . . . . . . . . . . . 2.1.4.2 Cornice 3 . . . . . . . . . . . . . . . . 2.1.4.3 Cornice 4 . . . . . . . . . . . . . . . .

1

. . . . . . .

11 20 35 36 40 43 44

. . . . .

46 46 47 50 56

. . . . . . .

58 58 59 59 60 63 66

.

78

.

79

. 84 . 86 . 105 . 110

X

Table of contents

2.1.4.4 Cornice 5 . . . . . . . . . . . . . . 2.1.4.5 Cornici 6 and 7 . . . . . . . . . . . 2.1.4.6 Cornice 8 . . . . . . . . . . . . . . 2.2 Other considerations intrinsic to Epicurus’ work . . 2.3 Do P XIV and XV depend on Theophrastus’  λ  ? . . . . . . . . . . . . . . . . . . . 2.3.1 Was Plato the last of the limited pluralists in Theophrastus’  λ  ? . . . . . . 2.3.2 The detail of the arguments against Plato and air monism . . . . . . . . . . . . . . . . 2.3.3 The dating of P XIV and of Theophrastus’  λ   . . . . . . . . . . . . . . . . 2.4 Conclusion . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. 116 . 122 . 125 . 128

. . . 131 . . . 137 . . . 138 . . . 143 . . . 145

Chapter 3 Lucretius’ use of sources in DRN I . . . . . . . . . . . . . . 3.1 The source of DRN I.156–598 and 951–1107 . . . . . 3.2 Did Lucretius change source after line 598 of DRN I? 3.3 The Critique does not derive from the same source as 155 ff. . . . . . . . . . . . . . . . . . . . . . . . . 3.4 The connection between lines 634 and 635 . . . . . . 3.5 Why did Lucretius have the Critique at the centre of book I? . . . . . . . . . . . . . . . . . . . . . . . 3.6 Was Epicurus the source of the Critique? . . . . . . . 3.7 Did Lucretius use a later Epicurean source? . . . . . 3.7.1 The choice of Heraclitus . . . . . . . . . . . . 3.7.2 Lucretius’ use of homoeomeria . . . . . . . . 3.8 Conclusion . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

163 168 171 177 179 181

Chapter 4 Lucretius in the Critique . . . . . . . . . . . . . . . . 4.1 Heraclitus as a general . . . . . . . . . . . . 4.2 Heraclitus’ army . . . . . . . . . . . . . . . . 4.2.1 Stolidi and inanes Graii . . . . . . . . 4.2.2 Sound and truth . . . . . . . . . . . . . 4.2.3 Inversis sub verbis . . . . . . . . . . . 4.3 The theme of the path and the search for truth 4.3.1 Lines 657–59 . . . . . . . . . . . . . . 4.3.2 Lines 690–700 . . . . . . . . . . . . . 4.4 Empedocles and Sicily . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

182 182 185 186 190 199 208 209 211 212

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . 147 . . 147 . . 152 . . 158 . . 160

Table of contents

4.5

4.6 4.7 4.8

4.9 4.10 4.11 4.12

4.4.1 Empedocles’ language: poetry as revelation . . . . 4.4.2 Lucretius’ praise . . . . . . . . . . . . . . . . . . 4.4.3 Etna . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.4 Lucretius’ endorsement of Empedocles’ discoveries 4.4.5 The four elements: Empedocles’ disastrous fall . . Lucretius’ presentation of Anaxagoras’ theory . . . . . . 4.5.1 Lucretius’ transliteration homoeomeria . . . . . . 4.5.2 Parody of Anaxagoras . . . . . . . . . . . . . . . The mortality of Anaxagoras’ primordia . . . . . . . . . Lucretius’ strategy in lines 859–74 . . . . . . . . . . . . The analogy of letters and atoms . . . . . . . . . . . . . 4.8.1 Lines 823–29 . . . . . . . . . . . . . . . . . . . . 4.8.2 Intertextuality . . . . . . . . . . . . . . . . . . . . 4.8.3 Lines 906–14 . . . . . . . . . . . . . . . . . . . . Formularity . . . . . . . . . . . . . . . . . . . . . . . . The parallelism between lines 803–29 and 897–920 . . . The Critique as ‘dialogue’ . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . .

XI 213 216 223 224 231 235 236 238 239 243 245 247 250 251 253 255 257 264

Appendix (A) Two stages of composition? . . . . . . . . . . . . 267 Appendix (B) The format of PHerc. 1148 and PHerc. 1151 . . . . 273 Appendix (C) Do Epicurus’ Ad Herodotum and Ad Pythoclem reflect continuous books of P ? . . . . . . . . . . . . . . . 283 Abbreviations

. . . . . . . . . . . . . . . . . . . . . . . . . . . 289

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291

XII

Table of contents

Introduction

1

Introduction Lucretius’ refutation of the theories of matter of Heraclitus, Empedocles, Anaxagoras and other unnamed thinkers in lines 635–920 of DRN I, which I shall henceforth refer to as the Critique, is a unique piece of literature in that it presents criticism of the views of earlier philosophers in poetic form, rather than prose. Understanding how Lucretius handled his source material in this section – and more generally in book 1 – is essential for judging his achievement, and constitutes the main focus of this study. Like any late Republican Roman poet Lucretius would have considered it natural to reproduce Greek models, and would have been expected to do so. Latin poetry had been heavily dependent on Greek originals from its outset, and Greek influence pervaded all branches of Roman literature, from epic1 to lyric2 and drama.3 The process of appropriation of Greek culture meant that Latin poetry in Lucretius’ day reproduced and was meant to be set against Greek originals, often creating a dialogue with them through allusion. Lucretius’ decision to compose a didactic poem expounding Epicurus’ φ  complicates his relation to the Epicurean source material. Composing such a work meant rejecting the admonition of his venerated ‘guide’ and main philosophical source, Epicurus.4 Although Epicurus’ 1 2

3

4

Ennius famously claimed a connection with the spirit of Homer (DRN I.117–26). Catullus for example translates Sappho in poem 51 and Callimachus in poem 66. In Odes I.1 Horace wishes to be ranked one of the Greek Lyric poets, Odes I.9 and 14 are based on Alcaeus; Porphyrio testifies that Odes I.15 was based on Bacchilides. Caecilius Statius’ Plocium for instance was closely based on Menander’s Tμ  , as is shown by Aulus Gellius II.23. According to Gellius, Statius fell short of the literary qualities of Menander. Roman readers, at least at the time of Gellius, judged the literary merits of comedies by setting them side by side with the original and comparing the poetry of the two. Indeed in the final sentence of II.23 Gellius implies that Statius should not have attempted to rival a model with which he could not compete. This does not mean that Epicurus’ remarks about  φ in language (see pp. 191–92 below) did not exert an influence on the way in which Lucretius wrote his poem. Lucretius may be showing, by providing his verses with the

2

Introduction

precise views on poetry and on the arts more generally are uncertain, the limited available evidence suggests he had at least partial reservations about its value. The earlier belief that Epicurus denied that poetry could produce pleasure – i.e. that it had any value – and therefore rejected it altogether has been dispelled.5 It is now generally agreed that Epicurus rather considered poetry one of the unnecessary pleasures. However Epicurus denied that poetry had any educational value,6 probably on the grounds that the poems known to him invariably spread what was in his view false, and dangerous, belief. The report in Diogenes Laertius X.120 that ó  μ φμ «      « λ  «  ! . P "  # 7 $ Ν   (“only the sage will discuss music and poetry

5

6

7

clarity which Epicurus only thought possible in prose, that poetry could convey a philosophical message clearly (Asmis 1995: 33–34). According to Cabisius (1979: 245), by writing poetry Lucretius was challenging Epicurus’ view and defending conventional   from Epicurus’ attacks. To say that Lucretius was in any way defending conventional   seems inaccurate. Lucretius was rather reinventing and improving upon Empedoclean didactic poetry by adapting it to suit the Epicurean message. See Asmis 1995, Sider 1995, Wigodsky 1995 and Janko 2010: 223. The allegorist Heraclitus (Homeric Problems 4 and 79) criticizes Epicurus for deriving from Homer, despite his wholesale condemnation of all poetry, the notion that pleasure is the ultimate goal by mistakenly taking at face value Odyssey IX.6–7 and 11. According to Asmis (1995: 16–17), if Epicurus cited these lines in his works, it would probably have been to “explain his [. . .] doctrine that we all seek pleasure.” The postulated citation by Epicurus of these Homeric lines seems to show only that he was willing to refer to Homer and therefore presumably did not think that one should refuse to listen to (or read) poetry altogether, and certainly does not entail that he thought Homer’s opinion in any way corroborated his argument. A number of passages in Plutarch suggest that Epicurus was skeptical about the educational value of poetry. In Moralia 1086F Epicurus is said to have spoken in terms of  κ ' and of (O   . Moralia 15C makes reference to the Epicurean advice of filling the ears with wax so as not to hear the poetry of the Sirens. Moralia 1094E provides evidence for Metrodorus claiming that there is nothing shameful about being totally unfamiliar with Homer (Giancotti 1959: 16–17). Further evidence – admittedly partial – comes from Cicero’s De finibus I.72, where the Epicurean Torquatus asks, referring to Epicurus: an ille tempus aut in poetis evolvendis, ut ego et Triarius te hortatore facimus, consumeret, in quibus nulla solida utilitas omnisque puerilis est delectatio . . . ? See further Craca 2000: 7–11 and Beer 2009a: 71–75. Asmis (1995: 22) suggests retaining the alternative reading #), and explaining it as a gloss. According to her, Epicurus, whether the restoration is ac-

Introduction

3

correctly; but he ought not to compose poems in practice”) makes it clear that in Epicurus’ view only the Epicurean sage could discuss poetry correctly,8 but ought not to compose poetry himself. This seems to amount to condemnation of poetry as a medium. It is unclear whether Epicurus in this passage is condemning poetic production outright, or, as Arrighetti argues, he is simply stating that the Epicurean sage, who is inherently interested in philosophical enquiry, ought not compose poetry to communicate that research.9 Epicurus’ view, thus understood, may just clear Philodemus of the charge of being inconsistent with his Epicurean credo,10 but not Lucretius, whose poem is most certainly philosophical. Lucretius’ act of ‘heresy’ seems remarkable, although Armstrong (1995: 218), Dalzell (1996: 45–46) and Arrighetti (1998: 32) seem to have a point in remarking that the derivative and protreptic nature of DRN, a didactic poem that does not present independent philosophical research to some extent lessens Lucretius’ unorthodoxy. It seems noteworthy in this context that, if Lucretius was aware of the doctrinal problem and looked at it in such terms, that might have encouraged him to keep as close as possible to the Epicurean material he found in his philosophical sources. How far Epicureans in Lucretius’ day, some of whom, such as Zeno and Philodemus, not only discussed poetic theory but composed poetry themselves, would have frowned upon attempts at an Epicurean didactic

cepted or not, means “practicing energetically”, so that he would be allowing the wise man to compose poetry ‘as a hobby’. Sider (1995: 35–36) criticizes Asmis’ reasoning about the gloss, but accepts her reading of the fragment. Arrighetti (1998: 16–18) rightly rejects Asmis’ proposal in favour of the more reasonable reading “in practice”, pointing out that the corruption to #) may be the result of the misunderstanding of an abbreviation (Arrighetti 1998: 16, n. 9). 8 The fact that the Epicureans Metrodorus (331–278 BC), Zeno of Sidon (mid 1st century BC), and Demetrius of Laconia (early 1st century BC) discussed poetry is perfectly in line with Epicurus’ statement that only ² φ« can judge poetry. 9 Arrighetti (1998: 18–19) points to three pieces of evidence in order to show Epicurus did not condemn poetry altogether: Epicurus’ own use of poetic quotations, Philodemus’ statement in Pλ $' « that the wise man is able to detach himself from false implications when reading poetry (see Obbink 1995: 189–209), and the positive attitude in Philodemus, Pλ * ’ 6O $ * ' !«. 10 Arrighetti (2003: 142) points out that Philodemus’ choice of the epigram as the form of his poetry is perfectly coherent with the role the Epicureans attributed to poetry.

4

Introduction

poem expounding φ  is debated. Philodemus’ view in Pλ   V. XVII.8–25 (Mangoni 1993: 144–45), which presumably reflects the teachings of his teacher Zeno, has been taken to suggest that contemporary Epicureans followed Epicurus in thinking that poetry was not appropriate for expressing philosophy, nor for scientific discussion. Beer (2009a: 79–83), however, sees the production of DRN as not contradicting the poetic theories of Philodemus and more generally late Epicureans. Even if by embarking on his poem Lucretius was not going against the advice of contemporary Epicureans teaching in Italy, his undertaking was certainly innovative.11 Given the Epicureans’ rejection – if not in principle certainly in practice – of didactic poetry on physics, he had to look for a stylistic model from outside the philosophical school he belonged to. Lucretius found his chief literary model,12 as far as genre is concerned, in Empedocles. At various points of the discussion we shall find evidence that Lucretius wanted his poem to resemble Empedocles’, and thereby invited his reader to compare the two. It is worth pointing out at the outset, however, that although many passages in Lucretius, and especially the prologue to his poem, probably imitate Empedocles, at no point can the content of extended sections of DRN be shown to derive directly from Empedocles.13 Empedocles was not a doctrinal source for DRN. By presenting Epicurean doctrine in Empedoclean guise Lucretius was innovating, and combining the style of one source with the content of an unrelated source. This was a daring and original undertaking. We know little about contemporary attempts at Latin didactic poetry, but it 11 The orator T. Albucius may have composed an Epicurean poem in the second

century BC, since he is referred to as perfectus Epicureus in Cicero Brutus 131, listed among Epicurean authors in De natura deorum I.33 and 93 and called aridus [poeta] in Fronto De eloquentia I.2 (Garbarino 1973: 458–62). During his exile in Athens, Albucius may have been taught by the Epicurean Zeno of Sidon, whose interest in poetry is apparent from Philodemus’ writings (Janko 2000: 9, n. 7). Given that there is no indication that Albucius’ poem was comparable to Lucretius’ poem in theme, scope, genre or tone – indeed Albucius was known for his widespread use of Greek, something which Lucretius studiously avoids – I doubt that Lucretius would have used it as his stylistic model. 12 Lucretius was clearly influenced stylistically by a number of literary models, including Alexandrian ones. The influence exerted on the style of DRN by the epic poets Homer and Ennius is particularly pervasive. On Lucretius’ relation to previous poetic traditions, see further Gale 2009. 13 For a possible exception see n. 676 below.

Introduction

5

seems at least possible that such works could derived both content and style from the same model. Indeed the only certain mentions that we have of Roman didactic poetry contemporary with Lucretius are Cicero’s Aratea, which was heavily dependent on Aratus, and Sallustius’ Empedoclea, which may or may not have taken up both content and form from Empedocles’ poem.14 Let us now turn to contemporary Roman philosophy in prose, in order to establish what the common practice was in using source texts. Comparison with contemporary Roman writers of philosophy suggests that it was the norm to be dependent on Greek models. Roman philosophical authors of the second and first centuries BC were accustomed to follow Greek originals closely, and sometimes simply translated those models.15 The evidence comes mainly from Cicero’s philosophical works, and from his comments on the philosophical writings of earlier Roman authors. Cicero translated from the Greek, not always acknowledging that he had done so.16 Particularly interesting is Cicero’s Timaeus, which was a continuous translation of Plato’s work of the same title. The precise nature and state of composition of Cicero’s Timaeus are uncertain. Powell observes that the work has an introduction similar to the ones Cicero uses in other dialogues. In the few surviving lines Cicero does not mention that what follows is a translation from Plato. Powell (1995: 281) thinks the Timaeus was an abortive effort,17 since there is “no known parallel for such a long piece of direct translation” in Cicero’s other philosophical dialogues. Cicero produced two other (lost) works, his Protagoras and

14 Sedley (1998: 1–2) suggests that this work was presumably a translation or imi-

tation of Empedocles. He convincingly argue that Cicero, by quoting the two works in the same context, implicitly compares Lucretius’ poem to Sallustius’ Empedoclea: Lucreti poemata ut scribis ita sunt, multis luminibus ingeni, multae tamen artis. Sed cum veneris, virum te putabo si Sallusti Empedoclea legeris, hominem non putabo (Ad Quintum fratrem II.9.3). According to Edwards (1989: 106–07), Sallustius’ Empedoclea derived its content from Empedocles’ works, which were popular at the time. That many of Empedocles’ theories would have been of any more than historical interest at Lucretius’ time is far from certain, and it would be dangerous to draw any conclusion as to the content of Sallustius’ poem given the limits of the available evidence. 15 In Tusculanae disputationes I.5 Cicero complains about the fact that philosophy in his age nullum habuit lumen litterarum latinarum. 16 See Powell 1995: 279–80. 17 On the composition of Cicero’s Timaeus see also Lévy 2003: 93–95.

6

Introduction

Oeconomicus, which appear from the fragments to have been direct translations of Plato’s and Aristotle’s works of the same titles. It is interesting that Cicero in De finibus I.7 considers the possibility of translating entire works by Plato and Aristotle,18 if only to reject it in this case. There are indications that Cicero’s first philosophical dialogue, Hortensius, was based on Aristotle’s Protrepticus, although one cannot tell how closely Cicero followed Aristotle in defending philosophy. Cicero at the start of his philosophical career considered it acceptable to simply translate an entire work into Latin, or at least base a whole work on a single Greek model. He was certainly not ashamed of translating passages from Greek originals.19 The extent to which Cicero translated Greek sources in his later works is debated. Books one and two of De officiis have been thought to derive from Panaetius. However, it looks as though Panaetius was the main source which gave Cicero his structure, but not the only one. Cicero clearly followed very closely a Greek Epicurean source in his review of   of the gods in De natura deorum (see pp. 32–33 below). Whether or not we think that Cicero’s famous comment in his letters to Atticus (XII.52) that his works are “$ φ ” is false modesty on his part,20 the background from his philosophical writings suggests that Cicero, and presumably any Roman writing philosophy in Latin in the first century BC, would have found it natural to use Greek sources, and might have done little more than translate his Greek source into Latin. We can get some insight into the work of Roman Epicureans21 who were earlier than, or contemporary with, Lucretius from the remarks

18 Sed id neque feci adhuc nec mihi tamen ne faciam interdictum puto. Locos qui-

dem quosdam, si videbitur, transferam et maxime ab iis quos modo nominavi, cum inciderit ut id apte fieri possit; ut ab Homero Ennius, Afranius a Menandro solet. Cicero explains why there is more point in reading his philosophical works than translations of Greek dramatic texts. He points out that he adds his own iudicium and scribendi ordo. Asmis (1983: 49–50) thinks that Cicero is signalling three contributions: invention, arrangement and style. But one cannot be certain that iudicium means inventio here. Reid (1925: 8) suggested not unreasonably that in this context iudicium means literary taste. 19 Cicero is willing to take over the same material from Plato’s Phaedo 80–81 in Somnium Scipionis 21 and Tusculanae disputationes I.27 and 72. 20 See Rawson 1975: 233 and Obbink 1996: 96–97. 21 For a list of known Roman Epicureans in the late Republic, see Ferguson 1990: 2262.

Introduction

7

about Amafinius and Rabirius in Cicero’s works.22 These two writers are criticised by Varro in Academica I.4.23 The continuation of the passage (Academica I.5–6) shows that Amafinius used the word corpuscula to refer to Epicurus’ atoms, and may suggest that Amafinius discussed how atoms come together to form things. Cassius, in his reply to a letter of Cicero,24 criticizes the same Amafinius, and Catius Insuber. They are referred to as mali verborum interpretes. Catius wrote quattuor libros de rerum natura et de summo bono.25 Cicero reports that he translated Epicurus’ , with the word spectra.26 And Quintilian (Institutio oratoria X.1.24) cites Catius among other Roman prose writers: . . . in Epicureis levis quidem sed non iniucundus tamen auctor est Catius. It is worth mentioning that Pliny the Younger (Epistulae IV.28) shows that Catius was famous enough for his portrait to be hung in the library of Herennius Severus. The picture that emerges is one of Epicureanism spreading rapidly in Rome, probably at first in elite circles, and then increasingly in Roman society at large with the first tentative efforts to give Greek philosophy a Latin voice. Lucretius’ undertaking should be seen within this cultural context. Phaedrus, who was born circa 138 BC, may have gone to Rome in 88 (Ad familiares XIII.12), possibly to teach, but was back in Athens in 79 (Raubitschek 1949: 97–98). Appius and Lucius Saufeius were also known Epicureans who had studied in Athens under Phae-

22 See Tusculanae disputationes IV.6–7 (cum interim illis silentibus C.Amafinius

23

24 25 26

exstitit dicens, cuius libris editis commota multitudo tulit se ad potissimum disciplinam . . . Post Amafinium autem multi eiusdem aemuli rationis multa cum conscripsissent, Italiam totam occupaverunt ); Pro Caelio 40–41; De natura deorum I.8. Howe (1951: 57) considers it probable that Amafinius was a contemporary of Lucretius, and argues that DRN V.336–37 (hanc primus cum primis ipse repertus / nunc ego sum in patrias qui possim vertere voces) should be explained by saying that such writers did not antedate Lucretius. On Amafinius, see further Gambarino 1973: 462–70. Nulla arte adhibita de rebus ante oculos positis vulgari sermone disputant, nihil definiunt nihil partiuntur, nihil apta interrogatione concludunt, nullam denique artem esse nec dicendi nec disserendi . . . It is not easy to see what exactly Varro’s words refer to, but perhaps Cicero would not have made Varro speak in such terms if Amafinius and Rabirius simply translated earlier Greek works. Further criticism of the Latin Epicureans comes in Cicero, Tusculanae disputationes II.7, as Shackleton Bailey (1977: 381) notes. Ad familiares XV.19.1–2; Shackleton Bailey 1977: 62. Porphyrio, Ad Horatii saturas III.4.1. Ad familiares XV.16; Shackleton Bailey 1977: 60.

8

Introduction

drus.27 The production of the works of Rabirius, Amafinius and Catius suggests that Epicureanism was beginning to spread among non-Greekspeaking Romans.28 According to Sedley (2009: 39) these Epicurean text represented the earliest efforts to translate Greek philosophy into Latin language. There was in Italy in Lucretius’ day a lively interest in philosophy (Sedley 1998: 61–65). It is worth noting, however, that Cicero complains about the lack of philosophical production in Latin29 and the fact that some learned Romans refused to read philosophy which was not in Greek.30 It seems fair to say that Epicurean literature in Latin was in its infancy at the time Lucretius embarked on his project and that it was customary to look to Greek texts for models and keep close to such philosophical sources. Both Academica (I.4) and De finibus (I.1) suggest that people who were likely to read philosophical works in Latin also knew Greek. Presumably only part of the eruditi Graecis litteris would refuse to read Latin literature. We have seen above (n. 3) how Gellius compares ‘Roman’ comedies with their Greek counterparts, and likewise some readers of Latin philosophical authors works would have been familiar with the Greek originals such authors were following, and thus able to compare the two.31 It looks, however, as though their primary aim was to be faithful and accurate in reproducing the doctrines in the original rather than elaborating or innovating on the philosophical material. The spirit of emulation / competition that is found in Roman poetry of the period seems absent in philosophical works. To come back to Lucretius’ use of sources. Scholars have questioned whether establishing the nature of his philosophical source-material is important for a ‘literary’ evaluation of DRN. While pointing out that 27 Raubitschek (1949: 103). The possibility that Lucretius himself could have been

taught by Phaedrus in Athens – or Rome? – cannot be discounted. 28 The reference to the Epicurei nostri who Graece fere nesciunt in Tusculanae dis-

putationes V.116 presumably refers to Amafinius’ pupils. 29 Cicero may be exaggerating the extent to which he is breaking new ground, but

Tusculanae disputationes I.6–7 implies that the works of Amafinius were not good enough to be read by the eruditi. It is unlikely that Cicero would attack Amafinius and Rabirius simply because they held Epicurean beliefs. Cicero’s bias does not hinder him from being a close friend of the Epicurean Atticus or from praising Lucretius’ poem. It may be that Amafinius and Rabirius misinterpreted Epicureanism, as well as having a poor style. 30 De finibus I.1–6 and Tusculanae disputationes I.5 31 That DRN would have been compared by readers using the Herculaneum library with Epicurus’ Pλ -« is tentatively suggested by Obbink 2007: 39.

Introduction

9

Lucretius depends heavily on Epicurus for his philosophy, Kenney (1977: 8) remarks that “which work or works he chiefly followed and how much, if any, independence of Epicurus he allowed himself are questions not of primary concern to those interested in the DRN as literature. It is important, however, to grasp the width of the line separating the Epicurean texts that Lucretius had before him from what he made of them.” I am not sure I can agree with this claim. Understanding the difference between Epicurus’ dry prose and Lucretius’ poetry does not seem to be the whole story: additions to content can have a literary aspect too. In a section such as the Critique, where philosophical arguments, rhetoric and poetry become, as we shall see, fused into one, forming an idea of what material Lucretius drew from his philosophical source and what he altered or introduced himself, is relevant to literary analysis of the poetry of DRN. It helps to put his literary merits into perspective and to show where his achievement lies. Although we know that Lucretius followed Epicurus very closely at various points in his poem, and Thucydides for a considerable section of DRN VI, our understanding of Lucretius’ poetry and of how far he understood or altered the philosophical issues in DRN is hampered by our scant knowledge of what kind of source material he was using at various points in his narrative. The indebtedness of this study to David Sedley’s book Lucretius and the Transformation of Greek Wisdom will be apparent to the reader: he has pointed the way. I follow the approach he has adopted and, in many respects, develop, in relation to lines 635–920 of book I, some of the leads he has pointed to, though coming to different conclusions on some issues. My book has two aims. The first is to establish – as far as the available evidence allows – the nature and date of Lucretius’ source material, and the way in which he used and adapted it. The debate on Lucretius’ use of sources is very much open. Two recent studies on the topic take very different views. Schrijvers (1999) thinks of Lucretius as an ‘eclectic late Hellenistic writer’; Sedley (1998), on the other hand, pictures Lucretius as a ‘fundamentalist’ who relied exclusively on Epicurus’ Pλ -« (henceforth P ) as his source. My second aim is to show how Lucretius made the Critique more effective and more likely to grip his audience by adapting, elaborating and adding to his source. There are many sections in the Critique where Lucretius should be thought to be independent of his lost source, and many sections where he can be seen to have heavily elaborated the material he found in his source. Lucretius used his Greek philosophical source comparatively freely, as we shall see in chapter 4.

10

Introduction

Studying the Critique from the point of view of sources is also important in the context of our understanding of the reception of the Presocratics in late Republican Rome and of the history and development of the Epicurean school, which was notoriously prone to refutation of opponents’ theories. Enquiry into Lucretius’ use of sources in the Critique sheds some light on how Epicureans went about producing polemical texts, as well as Lucretius’ place in the development of such a tradition. Discussion of Lucretius’ use of sources for the Critique will involve close analysis of the remains of books XIV and XV of Epicurus’ P , to try and determine their content. Since it has been suggested that these books were Lucretius’ source and this thesis has gone unchallenged in recent Lucretian scholarship,32 I will dedicate considerable space to the relevant papyri from Philodemus’ library in Herculaneum and provide a new edition of some of the fragments. This is the result of autopsy of the papyri in Naples’ Biblioteca Nazionale Vittorio Emanuele III, corroborated by use of the multispectral digital images of the papyri prepared by Steven Booras and David Seely.

32 See Warren (2007: 26), Tatum (2007: 135, n. 13), and Piazzi’s more qualified

endorsement (2005: 8–9, and especially 26–27). It is surprising to read in Piazzi that Epicurus’ use of the term  « in column XL of P XIV (“che può forse riferirsi di nuovo a Eraclito”) can be taken to support the theory of derivation, since in this column, which is part of the methodological section concluding the book, Epicurus is commenting on the methods of an hypothetical thinker. Indeed Heraclitus is not mentioned in any of the surviving fragments of P XIV.

11

Chapter 1 Lucretius drew the Critique from an earlier Epicurean polemic I doubt that Lucretius elaborated his own counter-arguments against their theories of matter of Heraclitus, Empedocles and Anaxagoras having reconstructed their thought either from second-hand information about the three Presocratics or from their original writings.33 Although there are indications that Lucretius knew the works, or at least quotations from the works, of Heraclitus and Anaxagoras, and he was certainly familiar with the work of Empedocles,34 the evidence suggests that it is likely that in DRN I.635–920 Lucretius reproduced the arguments he found in a Greek35 text.36 I do not exclude the possibility that some arguments are Lucretius’ own contribution, but evidence from elsewhere in the poem suggests that he would have depended on a Greek source. DRN VI.1138–286 are extremely close to Thucydides’ description of the Athenian Plague (II.47–52). Lucretius’ use of sources in this passage, however, is complicated by the fact that he includes material from the Hippocratic corpus. Munro (1864b: 394–95) was the first to notice how lines 1184–95 resembled passages in the Hippocratic corpus, comparing e.g. line 1184 with Hippocrates Prorrhet. I. 49, line 1185 with Hippocrates Praenot. Coac. 193 and line 1186 with Hippocrates Progn. 8. Con33 Brown (1989: 150) criticises the suggestion that Lucretius’ knowledge of the

Presocratics does not derive from his own study of the original texts: “this suggestion should be treated with caution in view of Lucretius’ demonstrably wide reading in other areas of Greek literature.” 34 See p. 4 above and pp. 213–15 below. 35 That Lucretius was using a Latin source should, I think, be ruled out, in view of his remarks on the difficulty of translating from the Greek in I.136–39. 36 I do not mean that Lucretius necessarily had a written work in front of him. I intend ‘source’ to be understood in a wide sense. Lucretius is likely to have been taught Epicurean philosophy, and, if so, would have taken notes. If Epicurean education was based on Epicurus’ P , as seems likely, seeing that Philodemus’ library had multiple copies of the work (presumably for handing out to students), this makes it a priori likely that Lucretius was familiar with, and may have taken notes from, Epicurus’ P .

12

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

taminatio is perhaps a better explanation than an intermediary source between Thucydides and Lucretius.37 Whether or not he added the Hippocratic elements himself, it seems a fair remark that Lucretius’ description of the Athenian Plague is little more than a creative adaptation in verse of Thucydides’ account.38 And some philosophical sections of DRN can be shown to reproduce very closely the arguments found in Epicurus’ own works. This is clear when one compares, for example, DRN I.418–48 with Epicurus’ letter Ad Herodotum 39(b)-40,39 or DRN I.483–583 with Ad Herodotum 41, or again DRN IV.53–175 with Ad Herodotum 46–48.40 37 Clay 1983: 290, n. 5. Scholars have suggested that Lucretius used an intermedi-

ate source, possibly a (later) Epicurean text now lost. Ernout and Robin (1928: 351 and 361) postulated a Latin intermediary. This was suggested to them by what were thought mistranslations of Thucydides (Bright 1971: 607, n. 2). Clay, defending Lucretius’ originality, insists that it was Lucretius who fused the two (with the implication that, as Munro thought, Lucretius was familiar with the Hippocratic corpus). That Lucretius used an intermediary source in Greek or, much less probably, in Latin is perhaps unlikely, but cannot be ruled out. That this intermediary could have been an Epicurean text is pure speculation, although Diogenes Laertius X.28 refers to a work Pλ    μ« M by Epicurus (Giancotti 1994: 565). The most probable explanation is that Lucretius depends on Thucydides directly, but on Hippocrates indirectly (Bob Sharples suggested to me that it is not certain that the Hippocratic writings were as prominent in the first century BC as they later became). It is perhaps conceivable that Lucretius used as a source a medical commentary on Thucydides’ description of the Athenian Plague. 38 For Lucretius’ ‘reworking’ of the material in Thucydides, see Commager 1957. Sedley argues that Lucretius would have reworked the account of the Plague had he lived to complete his poem. Sedley’s theory that books IV, V and VI are an unrevised ‘first draft’ is open to a number of objections: see Appendix A below, pp. 271–72. 39 Woltjer 1877: 18–19. 40 Despite the similarities between DRN and Ad Herodotum it looks as though Lucretius did not use it as his primary source, since DRN treats which are omitted altogether in Epicurus’ letter (e.g., the material in II.730–990) and there are divergences in the order of topics. According to Sedley (1998: 141), the fact that the sequence in IV.176–215 reproduces that of P II, and not that of Ad Herodotum, shows that “Lucretius’ debt to On Nature has not been mediated by the use of the Letter to Herodotus.” It is hard to disprove altogether that Lucretius used Ad Herodotum for some sections of his poem, but a hypothesis according to which Lucretius often switched from one Epicurean source to another is rather uneconomical. I take it that Ad Herodotum reproduced, at times, the now lost text by Epicurus which Lucretius used. On whether this text is more likely to have been Epicurus’ P , or some other work, see pp. 148–49 below. On the structure of Ad Herodotum, and how far it reflected the structure of P , see Appendix C below, pp. 283–85.

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

13

Further evidence that Lucretius habitually reproduced the arguments in his Greek sources comes from DRN V and VI. Sections of these books are, as we shall see,41 very close to arguments which appeared in works by Theophrastus. I take it that Lucretius in books V and VI was following a text by Epicurus himself who had incorporated Theophrastan material,42 although it is very hard to prove that Lucretius did not gain access to the Theophrastan information either directly from Theophrastus, or through a later writer who reproduced Theophrastus independently of Epicurus.43 Lucretius may have included material later than Epicurus in DRN V. Schofield (1999: 749, n. 30) suggests that Lucretius cannot be taken 41 See pp. 132–33 below. 42 If Lucretius corresponds to Theophrastus, and Epicurus is not extant, there is no

need to suppose that the intermediary was a later Epicurean text; only if Lucretius and Theophrastus agree against Epicurus, might there be a case for supposing that the intermediary was a later Epicurean one. 43 Runia (1997: 99–101) suggests that Lucretius had “sources of access” to doxographical material other than Epicurus, pointing out that in DRN V.705–50 (a) Lucretius organises his discussion according to the doxographical diaeresis (moon as recipient of light / moon source of its own light) while Epicurus, who refers to this diaeresis in Ad Pythoclem 94, does not integrate it, as Lucretius does, with the questions of the moon’s transformations and its eclipse; (b) Lucretius makes a distinction between bastard (nothus) and own (proprius) light which is not found in Epicurus, nor in Aëtius, but is parallelled in the Philonic text De somniis I.21–32 (I. 23  ξ;    ν  #φ!  φ!« π )« #  $ ) ν ’ 2μ ξ 3   $!, μ ’ # $φ) ³« Ν # 3  λ $  μ« » ;) and in Lucian Icaromenippus; and (c) Lucretius refers explicitly in DRN V.727 to the theory of Berosus that the moon is a rotating sphere, half of which is inflamed. Runia takes it that the astronomical fragments were part of Berosus’ work B '  " which he dates between 290 and 270 BC, too late for Theophrastus and for most of the books of Epicurus’ P . Argument (a) may show no more than that Ad Pythoclem is earlier than the text Lucretius used as his source (if we suppose that once he had made the integration Epicurus could not have failed to do so always). The use of (b) the term nothus certainly links the three texts in question, but can Lucian and Philo be taken as representing the ‘doxographical tradition’ (nothus is not in Aëtius: Runia 1997: 101, n. 43)? As for argument (c) it is not certain – as Sedley (1998: 91–92 and 92, n. 125) notes – that Berosus’ astronomical considerations first appeared in his B '  ". Berosus’ dates make it possible that Epicurus knew of his views. Usener (1887: 384) and Bailey (1947: 1439–40) thought that such a view is cited in Ad Pythoclem 94 ( 4 φκ * 5 «  ). Runia (1997: 101, n. 50) objects that the distinctive feature of being π« is not in Epicurus. Even if Runia is right, it may be that Lucretius is performing contaminatio in this section of DRN V, as he seems to have done in the Plague passage in book VI (see n. 37 above).

14

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

as evidence for Epicurus’ own views on the development of civilisation, although he grants that there is some genuine material from Epicurus in V.925–1157 (1120–30 can be compared with K    7, and 1151–57 with K    35).44 Schofield points out that Lucretius’ phases of softening (1011–23) and exhaustion (1143–50) are at odds with the views of Hermarchus, Epicurus’ friend, philosophical associate and successor as head of the school.45 Assuming this material was not in Epicurus it is unclear whether it is more likely that Lucretius contaminated later material with Epicurus’ treatment, or that he was following a text by an Epicurean author later than Epicurus himself. One may also consider the possibility that Hermarchus’ views were not the same as Epicurus’, because Hermarchus was innovating.46 It has been suggested by Kleve (1978: 67) that the target of the polemic in DRN IV.777–817 is the Academic Carneades, because of the content and style of the passage. Objections comparable to the one in Lucretius are found in Cicero De natura deorum I.108 and Ad familiares XV.16.2 (Bailey 1947: 1274). Cicero may reflect an objection to the explanation we find in Lucretius and Lucretius could be reproducing a response to such a point. But it seems unlikely that Epicurus would not have anticipated what are such obvious objections to his theory of images and explained how there could be a constant supply of mental images. There seems to be no need to ascribe such objections to a specific thinker. As for the style of the passage, this seems wholly Lucretian and indeed comporable to the Critique. That Lucretius introduced ideas later than Epicurus in IV.777–817 is not proven. Kleve defends the theory that DRN I.1052–113 is aimed at the Stoics specifically, pointing out that Diogenes of Oenoanda in fragment 66 (Smith) refutes a similar theory, but while according to the theory in Lucretius the world is spherical and centripetal, in the first version of the geocentric theory Diogenes attacks “the earth is limited by the heavens above, but extends below without limitation.”47 According to Kleve, Lucretius would have deliberately chosen the version of the theory which 44 Cole (1967: 15–46) points to extensive parallels between Vitruvius’, Diodorus’,

Tzetzes’, Posidonius’ and Lucretius’ account of the development of civilisation to show that all of these go back to a Democritean original. Cole (1967: 170) suggests that there may be a Cynic source. 45 On Hermarchus and Epicurus see Longo Auricchio 1988: 25. 46 Vander Waerdt (1988: 90–98) argues that Hermarchus innovated on Epicurus’ views on anthropology. 47 Kleve 1978: 67–68.

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

15

dealt with the Stoics. This argument – if it should be called that – is far from stringent, especially since one cannot be sure that Diogenes was not actually targeting the Stoics.48 The problem with I.1052–113 is difficult because Plato and Aristotle, Epicurus’ usual targets, also adopted a geocentric theory (Plato thought that the elements did not have a natural tendency of their own, while Aristotle thought that fire and air moved upwards). Further, it cannot be determined whether 1083–93 introduce a new theory or not. I am inclined to believe, with Furley (1966: 18), that Lucretius is describing just one theory, according to which earth and water tend towards the centre, as distinct from fire and air, which move away from the middle. Schmidt (1990: 221) puts forward six points to support his claim that the Stoics are the target: (a) Lucretius speaks in terms of tendency towards the middle of all stuff, even light matter: that thinkers other than the Stoics, including Aristotle, held such a view has not been proven; (b) despite the general tendency of all things towards the centre there is a mention of centrifugal movements: a conflict between two such views can only be found within the Stoic school; (c) Lucretius’ opponent believes in void outside the world, but Aristotle did not; (d) Plutarch similarly discusses a tendency towards the centre in cosmology in connection with Stoicism; (e) Aristotle dismissed the theory that the heavenly bodies were close, and (f) the theory of increase which Lucretius treats in connection with the stars is a Stoic theory. One cannot be certain about Schmidt’s points (a) and (b): Lucretius uses omnia in 1053 and 1056, but Furley plausibly suggests that in those lines omnia refers rather to earth and water in the antipodes. Sedley (1998: 79) is right that there is no sign of Lucretius exploiting an inconsistency between centripetalism and centrifugality of air and fire in the position he is attacking, as Plutarch does in SVF II.434, presumably following a Peripatetic source.49 As for (c), it is true that in lines 1074–82 Lucretius assumes void outside the world, but he does not expressly say his opponents subscribed to this belief. Lucretius’ source may simply be arguing in Epicurean terms; Furley may well be right in remarking (1966:

48 Smith (1992b: 511) is unable to rule out that the Stoics are Diogenes’ target, al-

though he prefers to believe that Xenophanes is being refuted. Even if it could be shown that Diogenes was not targeting the Stoics, this seems to have no bearing on the identity of Lucretius’ target in DRN I. 49 Schmidt (1990: 214–15) thinks that Plutarch and Lucretius depended on the same tradition.

16

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

17–18) that there is “no need to think the propositions were held by his opponents.” It is unclear whether (c) makes it less likely that the opponent was Aristotle. Point (d) can be disregarded; Plutarch’s testimony may not be relevant at all to Lucretius. Point (e) seems to cast some doubt on the identification of Aristotle as the target. As for Schmidt’s point (f), this detail seems in fact to speak against thinking that the Stoics are the intended target of Lucretius’ source, since Sedley (1998: 78–79) draws attention to the fact that the Stoics thought that the heavens were nourished not by fire but by moisture.50 One cannot be certain that Lucretius’ source was attacking the Stoics’ geocentric theory. It certainly seems possible, if not likely, that Furley is right in holding that Zeno’s theory reported in SVF I.99 is a response to the arguments we find in Lucretius. Lucretius is probably reflecting Epicurus’ polemic with earlier thinkers. However one cannot take it for granted that the intended target is Aristotle, as this requires Epicurus to have known only the works, or part of the works, where Aristotle had not yet elaborated the fifth element. It would not be the first time that we are unable to identify the quadruple pluralists Epicurus was taking issue with (see pp. 82–83 below). The other contentious passage is V.156–234, where Lucretius refutes creationism. Furley (1966: 27–30) and Sedley (1998: 74, n. 60) argue against the claim that the theory under attack, according to which the world is created and eternal, is Stoic. Lévy (1999: 87–91), on the other hand, argues that Lucretius has formulated the argument so as to attack contemporary Stoics. Lévy points out that the anthropocentrism found in Lucretius is not in Timaeus – the target proposed by Sedley – but was developed by Polemon of Athens, that not all Stoics understood # « as the end of the world (e.g. Chrysippus in SVF II.604 and II.551), that Zeno of Thar50 Lévy (1999: 91–93) argues that an attack on the Academics would not by itself

explain all the details in DRN: the idea of vegetation being fed by fire, which Lucretius appears to develop in I.1092–93 is not found in Plato, but growth-producing fire is connected with plants (and living creatures) in SVF I.120, where Stobaeus is reporting Zeno. According to Lévy, Lucretius is attacking both the Stoics and the Academics in these lines. Unfortunately the text break off at line 1093, which makes it impossible to see exactly how Lucretius’ argument ran. The reference to plants in SVF I.120 is exceedingly brief and there is no indication that Zeno referred to the phenomenon of growth towards the skies, as Lucretius does. It is perhaps conceivable that Lucretius added a Stoic detail in this case, but we are far from proof; see further n. 410 below.

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

17

sus and Diogenes of Babylon had reservations about # « and that Boethus and Panaetius considered the world everlasting. The argument in Lucretius would, according to Lévy, only be representing the more recent version of the Stoic theory which incorporated Academic influences (such ‘Platonising’ tendencies can detected in the Stoic school during the first century BC). It is not easy however to see why Lucretius – if he intended to attack the Stoics – felt the need to stress the eternity of the world, rather than simply target the idea of creation. Since one would assume that Lucretius would have known of the theory of # «, why make the argument apply, at least on the face of it, only to the later Stoics? Certainly the Critique shows that Lucretius was not only concerned with targeting only contemporary theories, indeed quite the contrary. This objection, however, may not apply to the source Lucretius drew from at this point, who may have considered the theory of # « outdated and irrelevant. Smith (2003: 83) points out similarities in argument between DRN V.165–73, especially 168–69, and Diogenes Oenoanda’ argument against the Stoics in NF 126/127 and fragments 20–21, an indication in his view (1998: 145 and 2010: 231–32) that Lucretius has the Stoics in mind in book V. Presumably he means that the argument is directed against the Stoics, and not that Lucretius misunderstands the target of the argument he draws from Epicurus.51 It is difficult to establish how the two texts are related, especially since we do not have Diogenes’ confutation in its entirety (Smith 1976: 290 and 1992: 461). The coincidence in the description of the life of the gods is striking, but it probably derives from Epicurus.52 Assuming that Smith’s reconstruction is correct, both Lucretius and Diogenes develop the argument by asking first (a) whether the gods could have created the world for themselves and then (b) whether they could have created the world for humans. The argument in Diogenes could be the Epicurean reply to the Stoics’ answer to Lucretius’ question (a): i.e. “the gods created the world be51 Gale (2009: 11), seemingly to keep to the theory that Lucretius did not draw

from Neo-Epicurean sources, suggests that “it is possible to argue that the poet implicitly redirects arguments derived from Epicurus’ own writings against new targets – notably Stoicism.” The phrasing suggests that she envisages that Lucretius altered the arguments himself to include the Stoics, although she does not indicate what these alterations would be. It is also possible that Lucretius thought that the argument as formulated by Epicurus dispelled Stoic theories. 52 Smith himself (1998: 145) refers to Epicurus, Ad Menoeceum 123.

18

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

cause they needed neighbours.”53 This idea, which Diogenes explicitly ascribes to the Stoics, is not in DRN. However such a suggestion would conflict with V.146–55 where Lucretius has stated the gods are not part of our world. One cannot rule out that Lucretius – or his source – knew about this suggestion and decided to overlook it. Lucretius’ insistence on the eternity of world is not reflected in Diogenes, where the concept is not mentioned. Indeed this may be further indication that the Stoics are not the target of the polemic on geocentrism in I. 1052–113, where, as we have seen above (p. 16), the theory under attack presented the world as eternal. The emphasis Lucretius places on anthropocentrism in V.152 (hominis causa) is not in Diogenes. Moreover, the idea in Diogenes that the gods are looking after the world as a whole, rather than just humans, is absent from DRN, although Lucretius does stress that the opponents he is targeting considered the world sacred (V.157–59). The fact that Diogenes when attacking the Stoics did not insist on anthropocentrism seems to speak against seeing the focus on anthropocentrism in DRN V as a clue that the Stoics are the target. Lucretius may perhaps have incorporated philosophical material later than Epicurus, but if he did, he did so extremely sparingly. His source was Epicurus. Lucretius’ remarks in DRN III.9–13 and V.52–55 read like a statement to the effect that most, if not all, of his philosophical material is based directly on Epicurus’ words. Lucretius would not have expressed himself in such terms had Epicurus not been his main source.54 It may be objected that in some passages of DRN it cannot be proved, or even inferred, that Lucretius closely followed a Greek source. This seems to be the case for the prologues to each of the six books,55 53 In fragment 20 Diogenes dismisses option (a) by asking why the gods would

want humans as their neighbours and where the gods would have lived before creation (since the Stoics claim that the world is unique), and then option (b), by pointing out that the lack of providence shows that the gods were neither acting in the interest of the world itself, nor in the interest of humans (Smith 1992: 376–77). 54 On the didactic ‘process’ between Epicurus and Lucretius as represented in the poem see Schiesaro 2003: 58–60. He notes that DRN stages two didactic processes: one between Epicurus and Lucretius and one between Lucretius and Memmius. The image of vestigia, which Lucretius uses for Epicurus, illustrates the process of memory: the relationship hinted at is that between father and son. On Lucretius’ didactic relationship with Memmius, see pp. 260–64 below. 55 The proem (DRN I.1–158, or arguably 1–155) is a special case, if Sedley (1998: 23–32) is right in suggesting that Lucretius is following Empedocles’ prologue to his Pλ φ« in structure (see pp. 218–19 below).

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

19

I.398–417,56 I.921–50, the praise of Empedocles and Sicily at I.716–32,57 parts, if not all, of the account of sex and love at IV.1037–287, and possibly the Magna Mater passage in II.600–60.58 Some of these passages certainly contain ‘original composition’, but to suppose that Lucretius – even if he had received philosophical training – independently elaborated the arguments in the Critique seems quite a different matter. None of the passages mentioned contains philosophical arguments,59 let alone detailed counter-arguments such as we find in the Critique.60 The Critique is a challenging passage from a philosophical point of view, and it is unlikely that Lucretius decided to do without a guide in this section of his poem.

56 See n. 392 below. 57 This passage is part of the Critique. Sedley (1989: 15) tentatively suggests that

the passage itself could be “direct imitation of a lost passage of Empedocles.” But Empedocles is reported not to have described Etna in his writings. Lapini (2003: 96–97) points out that Timaeus, as reported by Diogenes Laertius VIII.71, argued that Empedocles could not have committed suicide by throwing himself into Etna because, among other reasons, he never mentions the craters of Etna. On Lucretius’ description of Etna, see pp. 223–24 below. 58 Perret (1935) suggests that Lucretius and Varro, who was the first to treat the topic, depended on a Latin text, probably derived from Phrygian sources, which described the ritual and presented Cybele as mater generosa, while Boyancé (1941: 149) thinks Lucretius’ interest reflects his school’s interest in the exegesis given by the Stoics to the “mythes païens”, pointing out that Zeno and Cleanthes were of oriental origin (Zeno was from Citium on Cyprus, and Cleanthes from Assos) and therefore would have been interested in the cult of Cybele and Galli. Craca (2000: 28–29) refers to the treatment of the cult of Cybele in Philodemus’ Pλ $' « and considers the possibility that Apollodorus of Athens’ Pλ  was Philodemus’ source. According to her (2000: 30), it is possible the critique of the myth comes from the Epicurean school, but it is more likely that Lucretius used a literary source which his audience would know well and appreciate. There is a parallel between the account of Cybele in DRN II and pseudo-Aristotle Mirabilia 162 (Sharples 1985: 133–34). 59 Both the final section of DRN III and IV.1037–287 contain philosophical arguments, but the style and presentation of these lines suggest that Lucretius was not following a philosophical source-text step by step. 60 Some of Lucretius’ characteristically Epicurean counter-arguments are rather complex, e.g. the initial argument against Heraclitus on rarefaction and condensation followed by the ruling out of other means of transmutation of fire (see p. 25 below).

20

Chapter 1. Lucretius drew the Critique from an earlier Epicurean polemic

1.1 Lucretius’ information is second-hand Rösler (1973: 50–53) has shown that the reading of the three Presocratics in the Critique shares misunderstandings introduced by Peripatetic doxography. Diels’ term ‘doxography’, which will occur repeatedly in what follows, is slippery and potentially misleading; I shall therefore define briefly how I understand the word. Distinguishing what is ‘doxography’ from what is ‘use of doxography’ has proved a hard task. Mansfeld (2000: 347) defines doxography as follows: ... a systematic collection of tenets (doxai etc.) and not much more than that, though it may have a critical undertone and even sport explicit criticism.”61 He defines the term more inclusively than Diels, who thought that doxographical texts are those whose content can be traced back to Theophrastus’  λ   (henceforth 7).62 Mansfeld (1990: 3061–62) also emphasises that the tradition was more fluid than Diels had envisaged. The practice of composing lists of   was considerably developed by the Peripatetic school,63 although such collection of tenets most probably originated much earlier. There are indications that some kind of listing of   was practised in the Academy, but also previously (Mansfeld 1986: 3). The nature of the evidence for specific works, and the problems over how we define ‘doxography’ in the first place, make it difficult to deter61 As Mansfeld puts it “doxography proper is the offspring of Aristotle’s dialectical

overviews” and therefore originally considered positive as well as negative aspects of the  . 62 Mansfeld (1990: 358–59 and 1992b: 64–66) argues persuasively that  λ  , rather than    , was the wording of the title. 63 Earlier   are reviewed in Aristotle’s Physics (A.1, 184a10–192b4) and especially in Metaphysics A, where   take up most – twenty-six out of thirtythree and a half pages in the OCT edition – of the book, allowing for some Aristotelian theory within the ‘critical’ section. Theophrastus not only reported views in his lost 7, but is also said to have composed one book Pμ« 8« φ  « (FHS&G 137 and 138) and works devoted to report and, at least in some cases, criticism of individual philosophers’ theories (FHS&G 137, nos. 27–33, 39, 40 and 41), assuming that these were not part of the 7. The Peripatetics were perhaps the first philosophical school to collect information for the sake of collecting. Aristotle held that a consideration of previous views might help in arriving at the truth; therefore understanding and reporting of earlier views was required. In this sense ‘providing information’ was the Peripatetic school’s aim when listing earlier views.

1.1 Lucretius’ information is second-hand

21

mine whether counter-arguments appeared in the doxographical tradition, and what kind of counter-arguments these might have been. Paradoxically, if counter-arguments are ‘use of doxography’, Theophrastus’ 7 may itself turn out not to have been pure, i.e. ‘criticism-free’, doxography. Taurus, as Mansfeld (1990: 3207) mentions, showed that counter-arguments were included in 7 (FHS&G 241A). Baltussen (2000: 242) suggests that the term ‘critical endoxography’ “represents much better the theory and practice of the early peripatetic school”. – Aristotle and Theophrastus collected views on particular topics in the context of dialectic (9 ). It is not wholly clear whether Baltussen thinks Theophrastus’ 7, of which his De sensibus may have been a part, was itself an example of ‘critical endoxography’. Theophrastus’ De sensibus shows that the Peripatetics produced works, or at least sections of works, which listed, in more or less comprehensive fashion, previous opinions and passed judgement on them. The surviving text which comes closest to a ‘criticism-free’ form of doxography is Aëtius’ Placita, as reconstructed by Diels through Pseudo-Plutarch and Stobaeus, although here too criticism appears on rare occasions. Aëtius uses forms of the verb 4 " in reference to Thales in I.2.2, and to Anaximander in I.3.3, Anaximenes in I.3.4 and Anaxagoras and Plato in I.7.7. Criticism is also explicit in I.5.3, against Plato.64 This kind of incidental criticism, however, is very different from Lucretius’ series of counter-arguments. In Lucretius the aspect of reporting positive, as well as unacceptable, aspects of other thinkers’ views is completely lost.65 It can be said, generally speaking, that in their reports of earlier   (a) the Peripatetics found support in earlier views and at times rejected them, to argue that their own views were right; (b) the NeoPyrrhonists and Academic Sceptics argued that no view is right, since for them the very existence of rival views is itself an argument against the 64 Mansfeld (1990: 3206–07) implies that the traces of argument surviving in Aë-

tius derive from Theophrastus, and that the objection in I.7.7 states the Epicurean argument against the Stoics. Baltussen (2000: 242, n. 18) mentions that Aristotle regularly uses 4 " when judging views of others (Nicomachean ethics, passim; Physics 7.6, 213a24; Topics 7.5, 125b20; Metaphysics ;.3, 1090b32; On breathing 7, 474a18). 65 Compare this with Epicurus’ introduction to his polemic against earlier   in column XXIV of P XIV, where Epicurus points out that he discusses such views to free pupils from the "