Philosophy and Doxography in the Imperial Age 8822254740, 9788822254740

Citation preview

In memoriam Jean Pépin

ACCADEMIA TOSCANA DI SCIENZE E LETTERE «LA COLOMBARIA»

«STUDI» CCXXVIΠ

PHILOSOPHY AND DOXOGRAPHY IN THE IMPERIAI. AGE Edited by Awo BκAιvcAccτ

FIRENZE LEO S. OLSCHKI EDITORE MMV

La pubblicazione è stata reali Ζzata con í contributi del Ministero dell'Istruzione, dell'Università e de lla Ricerca (fondi ex 40%) e dell'Università degli Studi di Roma Tor Vergata, Dipartimento di Ricerche Filosofiche ISBN 88 222 5474 0

CONTENTS

Preface DAVm

Pag. VIl Τ. RUNiA, A difficult chapter in Aetius Book 11 on cosmo-

logy

»

1

»

23

»

59

»

79

DΟΜΙΝΙC J. O'MEARA, Plotin "historien" de la philosophie (Enn IV 8etV1)

»

103

1-lANSUELI FLÜCKIGER, The EOEETIEOI in the commentators .

»

113

»

131

BIBLIOGRAPHY

»

157

INDEX OF SOURCES

»

167

INDEX OF NAMES

»

181

JAAP MANSFELD, From Milky Way to Halo. Aristotle's Meteorolοgícα, Aetius, and passages in Seneca and the Scholia on Ara-

tus ALDO BRANCACCι, Stobaeus Anthologium 111 24 CARLOS LÉVY, Deux problèmes doxographiques chez Philon d'Ale-

xandrie: Posidonius et Εnésidème

JEAN PÉPIN, La quaestio De ideis de saint Augustin et la doxogra-

....

phie platonicienne

V

PREFACE This book collects the papers presented at the Third International Colloquium on Philosophy in the Imperial Age, held in Rome at the Consiglio Nazionale de lle Ricerche, 20-22 June 2002. The Colloquium was organized by the Chair of History of Ancient Philosophy of the University of Rome Tor Vergata, thanks to the financial support of the Italian Ministry of Universities and Scientific Research, with the assistance of the University of Rome Tor Vergata. It followed the two previous 1999 and 2000 Colloquia also held in Rome with support from the same Institutions, whose proceedings have already been published.' The subject of the Colloquium was the relationship between philosophy and doxography, one of the main topics dealt with by contemporary philosophical historiography concerning the Hellenistic and Imperial age. The concept of doxography in turn has been placed under close scrutiny in recent decades, thus being enriched and partly altered in its traditional features. This reassessment touched upon the origins of doxography as a genre, its philosophical and literary context, and theories and hypotheses regarding its Sitz im Leben. Within this debate some crucial problems can be identified, ranging from the distinction among sources (a definition of the notion of source and testimony being required too) to the investigation of the doxographers' and anthologists' working methods; from their works' formal features to the relationship linking these works to one another. The echo of these interests and trends in criticism can be seen dearly in the papers published here. Their common element, despite diversity of perspectives and methods, lies in having interpreted the two concepts of doxography and philosophy as being related to each other: and this is also the specific characteristic of the present volume – that, obviously enough, alms only at contributing to this line of investigation. 1 Α. Βzλνcncα (ed.), La filosofia in età impe riale. Le scuole e le tradizioni filosofiche, Atti del Colloquio, Roma 17-19 giugno 1999 («Elenchos» , XXXI), Napo li , Bibliopolis, 2000; Α. Βuu1νCAccI (ed.), Antichi e Moderni nella filosofia di età impe riale, Atti del II Colloquio internazionale, Roma 21-23 settembre 2000 («Elenchos», )XXIV), Napo li , Bibliopolis, 2002.

—VII-

PREFACE

While, until recently, doxography has been studied almost exclusively in order to reconstruct and shed light on the teachings of philosophers and traditions of thought to which it bore witness, nowadays there appears to be a common desire for enrichment of this perspective with a more detailed analysis of the autonomous structure of the doxographic account, of its philosophical background, and of the working method it displays. Above all, the time now seems to be ripe for considering at least some genres of the doxographíc and paradoxographíc tradition as a genre of ph ilosophical writing itself, i.e. as autonomous works on a specific doctrinal topic. In this sense the Colloquium, both in the papers and in the accompanying discussion, has investigated the doxographícal interests of the ph ilosopher, as well as the philosophical perspective of doxography, the latter also being interpreted as a possible feature of philosophical work. I warmly thank the scholars who agreed to deliver a paper, and all PhD students and colleagues who attended the Colloquium; among them I would like to mention at least Francesca Alesse (ILIESI-CNR, Rome), Francesco Aronadio (ILIESI-CNR, Rome), Enrico Berti (University of Padua), Mauro Bonazzi (University of Milan), Franco Ferrari (University of Salerno), Silvia Ferretti (University of Macerata), Anna Maria Ioppolo (University of Rome La Sapienza), Margherita Isnardi Parente (Accademia Nazionale dei Lincei), Walter Lapin (University of Genoa), Marco Ninci (Scuola Normale Superiore di Pisa), Lorenzo Perilli (University of Rome Tor Vergata), Matthias Perkams (Friedrich-Schiller-Universität Jena), Rosa M. Piccione (Friedrich-Schiller-Universität Jena), Giovanna Sillitti (University of Rome La Sapienza), Daniela P. Taormina (University of Rome Tor Vergata), Franco Trabattoní (University of Milan). Besides, I wish to express my gratitude to Francesco Ado rno, President of the Accademia Toscana di Scienze e Lettere La Colombaria, for agreeing to host this book in its series. I thank Antonio Carlini for presenting the book to the Consiglio di Presidenza of the Accademia La Colombaria, and Maria Serena Funghi, who very kindly followed the different steps leading to publication. It is also a pleasure to acknowledge my gratefulness to the Magnifico Rettore of the University of Rome Tor Vergata, Alessandro Fínazzi Agrò, for the interest he has always shown in the series of Colloquia on philosophy in the Imperial age, and for his encouragement. I am also grateful to the Consiglio Nazíonale delle Ricerche, which, following a long-standing tradition, hosted our meetings. Finally, I thank Francesco Aronadio and Lorenzo Perillí, who generously helped me in getting the `manuscript' ready for printing. ALDO BRANCACCI

— VIII —

DΑνm T. RUNIA Queen College, University of Melbourne

A D1ΠΠΙCULT CHAPTER IN AËTIUS BOOK II ON COSMOLOGY

1. INTRODUCTION AND AIM OF PAPER

More than five years ago a small symposium was organized in Leiden to celebrate the publication of volume one of the series of monographs on Aëtius and the doxographical tradition undertaken by Jaap Mansfeld and myself.' We invited two speakers to present papers on our work, Denis O'Brien from the CNRS in Paris and Michael Frede from Oxford. The day was a great success. The quality of the papers presented by the invited speakers can be gauged from the fact that both have subsequently been published in Phronesis. 2 O'Brien's paper was both entertaining and impressive in the manner that we have come to expect from such a thorough and meticulous scholar. In it he concentrated on a brief Aëtian lemma found in Stobaeus only, which records Empedocles' view on the destructibility of the cosmos. He used the lemma to illustrate the debt we all have to Hermann Diels. It was noted that Diels in his Doxographi Graeci retained the emendation of Sturz, but that it was subsequently dropped in his edition of the complete fragments of the Presocratic philosophers. The remainder of the paper described the vagaries of this emendation, which has been retained right up to recent times, even though it is certainly mistaken. O'Brien readily admitted, however, that he did not base his interpretation of the lemma Ι J. MANSFELD and D.T. RurπΑ, Aëtiana: The Method and Intellectual Context of a Doxographer («Philosophia Antiqua», LXXIII), Leiden, Brill, 1996. 2 M. FREDE, Aëtiana, «Phronesis» , XLN, 1999, pp. 135-149; D. O'Bairi, Hermann Diels on the Presocratics: Εmpedocles' Double Destruction of the Cosmos (Aëtius ii 4.8), «Phronesis»,

XLV, 2000, pp. 1-18.

1

DAVID T. RUNIA

on a full reconstruction of the chapter in Aëtius. This would involve him in `complications', as he calls them. He was perfectly willing to leave this task to me, drawing attention to a promise that I made to offer a reconstruction in our study. 3 In this paper I would like to take the opportunity to return to this chapter and fulfil the promise noted by Denis O'Brien. The chapter in which the Empedoclean lemma is found is Book II chapter 4 according to the division offered in Ps. Plutarch's Placita, with the title Eï ~φθαρτος ~~κ~σµος,`whether the cosmos is indestructible'. In this paper I shall analyse the various texts that bear witness to this chapter, then attempt to understand what their relationship is to each other, and on the basis of this investigation offer a reconstruction of what Aëtius' original chapter might have looked like. The result will be a further example of the complete reconstruction of Book Il of Aëtius' compendium which I hope to publish in the not too distant future. 4 It will emerge that the reconstruction of this chapter is rather difficult, and that on certain aspects no final certainty can be reached. The chief reason for this difficulty, as we shall see, is that the theme of this chapter, the destructibility of the cosmos — to which must be added the twin question of the createdness of the cosmos —, is one of the most common of all doxographícal themes, and this has led to considerable contamination in our sources. Nevertheless it will emerge that the exercise is very much worth doing, not only because it is an interesting example of how the Aëtian tradition works, but — more importantly — because it well illustrates a number of techniques used by that enigmatic doxographícal author whom we identify with the name Aëtius. It will be assumed that it is no longer necessary to give a summary of Diels theory of the Placita and how it should be modified in the light of the research carried out by Jaap Mansfeld and myself. This can be consulted in our study. I should, however, perhaps reiterate two points that are germane to the theme of this paper. Firstly, as is noted on the final page of our monograph, we are convinced that the ultimate aim of an analysis of the Aëtian tradition should be a single column of text, not Diels' double columns (even though these have a considerable practical value in showing apJ. MANSFELD and D.T. Ruνιλ, Aëtian, p. 193. See the reconstructions of P 2.25 and 2.20 already undertaken ín: Xenophanes on the 'Moon: a Doxographicum in Aëtius, «Phronesis», XXXI V, 1989, pp. 245-269; Xenophanes or Theophrastus? an Aëtian Doxographicum on the Sun, in W.W. FORTENBAUGH and D. Gurλs (edd.), Theophrastus: his Psychological. Doxographical and Scientific Writings («Rutgers Studies in the Classical Humanities», V), New Brunswick-London, Transaction Publ., 1992, pp. 112-140. 3

4

—2—

A DIFFICULT CHAPTER IN AËTIUS BOOK II ON COSMOLOGY

proximately how the two main witnesses are related to each other). Secondly, a conclusion of the research done so far is that Book II appears to be the best preserved, and this may well give us the best opportunity we have for gaining an accurate idea of what Aëtius' original compendium looked like.' 2. ANALYSIS OF THE EVIDENCE

The three main witnesses to Aëtius' original chapter are Ps. Plutarch, Stobaeus and Theodoret. We start with the last-named. As a rule the bishop of Cyrrhus does not cite, but rather freely paraphrases his material. But his evidence is always interesting, because he understands the method of the Placita and is often able to supplement the evidence of the other two in interesting ways. At 4.15 in the Graecarum affectionum curatio Theodoret begins his summary of the cosmology in Book II. Sections 15-16 summarize the material found in a parallel form in Ps. Plutarch Book ll chapters 1-4. We note how he introduces his material by drawing attention to the διαφων~α of the opinions recorded. This is his Christian adaptation of the theme of the dissensio philosophorum first exploited by the sceptical tradition. 6 The report of our chapter is extremely brief, but as often the Bishop has an eye for essentials (4 16): And some thinkers affirm that the cosmos is generated in thought ( κατ ~π~νο~αν γενητ~ν) but not in time, whereas others assert that it is absolutely ungenerated (~γ~νητον) and without cause ( ~να~τιον). And the one group ( οδτοι) regard it as destructible ( φοaρτ~ν), while the other group (~ κε~νοι) says it is indestructible (~φθαρτον).

It should be noted, firstly, that although the title as we read it in Ps. Plutarch only speaks of the destructibility of the cosmos, Theodoret has no hesitation in also discussing the question of its origin. Indeed, the phrasing of the view that the cosmos is only conceptually and not temporally generAs announced in Aitiana, p. xiv. Attention to this feature has been citawn by J. MANSFELD. See esp. Diaphonie: the Argument of Alexander De Fato chs. 1-2, «Phronesis», III (1988), pp. 181 207; Philosophy in the Service of Scripture: Philo's Exegetical Strategies, in J.M. Dn.LoN and A.A. LONG (edd.), The Question of "Eclecticism", Berkeley, University of California Press, 1988, pp. 70 102; Doxography and Dialectic. the Sitz im Leben of the Placita', in ANRW, II, 36.4, 1990, pp. 3056 3229, esp. 3063. 5

~~

-

-

-

—3—

DAVID T. RUNIA

ated is also found in Stobaeus, as we shall see, and must have been drawn from Aëtius. It might be thought surprising that Theodoret starts with the view that the cosmos is conceptually generated, and not the view of temporal creation. The latter of course is his own view, and he no doubt takes it for granted, citing here only wayward positions. Interesting too is his addition of the adjective αναιτιον in the view of the eternalists. This suggests that the question of the cause of the cosmos was also broached in the chapter. The second half of the report is not so easy to interpret, because of the ambiguity of the opposition ο τοι and ~κεiνοι. In normal Greek usage ovταιΡ should refer to the group holding the view just mentioned, but this would lead to a combination of non-generation and destruction, which, as we shall soon see, is the one option never held by anyone. Theodoret may be giving a separate opposition, without wanting a correlation with the two earlier views. Or he may be suggesting the usual combination of generation/destructibility and non-generation/indestructibility. It is hard to be sure. From brief references elsewhere (1.63, 4.68) we know that he was familiar with the standard views on this question. We turn now to Ps. Plutarch. It should be recalled that the direct and indirect tradition of this text is very complex. Eusebius and the Arabic translation of Qust~~ibn L~g~~basically preserve the original text, Cyril cites it a little more freely, whereas Ps. Galen gives a free paraphrase of what he found in Ps. Plutarch. None of these four authors, however, give us any access to the Aëtian text that is not mediated via the Ps. Plutarchean Placita. P (as this text will now be called) presents us with a compact list of four doxaí. From the systematic point of view this can be presented as: +A

Position 1

—B

Pythagoras & Plato & Stoics

Position 2

+A

+B

Epicurus

Position 3

—A

—B

Xenophanes

B (exceptional)

Position 4 where

Aristotle

A = subject to γ~νεσις, B = subject to φθορ~.

As copious material in the dialectical and doxographical traditions shows, the schema of the first three positions is quite standard, going back ultimately to Aristotle's doxographical analysis in De Caelo 1.10. Perhaps the best known text is found at Philo De aeternitate mundi 7. In this text the systematics of the three positions are first set out in a different and more logical order, i.e. + A + B, – A – B, + A – B. Philo goes on to subdivide the first position by making the distinction between multiple cosmoi and a single cosmos, and adds various name-labels, yielding the following schema: —4—

A DIFFICULT CHAPTER IN AËTIUS BOOK II ON COSMOLOGY many cosmoi single cosmos single cosmos single cosmos

Democritus, Epicurus Stoa Aristotle, Ocellus Plato, Hesiod, Moses.

A striking similarity between P and Ph ilo is that both give as their subject εi ~φθαρτος ~~κ~σµος(P title, Ph ilo Aet. 3), but then go on to discuss both the generation and the destructibility of the cosmos. We will need to return to this question later. P's schema may look straightforward at first glance, but it gives rise to a number of difficulties when the details of his presentation are examined. These amount to at least five. (1) The name-labels in the first doxa are in a state of confusion if we take the entire tradition of P into account: P Eusebius Ps. Galen Cyril Qust~~

Pythagoras and the Stoics Pythagoras and Plato and the Stoics Pythagoras and Plato Pythagoras and the Stoics Pythagoras and the Stoics

Diels, followed by Mau and Lachenaud, thought that Eusebius preserved the right reading here, as far as P (but, note well not Aëtius) is concerned. Given the additional evidence supplied by Stobaeus (which will be examined below), they are probably right, thought the matter cannot be settled beyond all dispute.' (2) The second part of this doxa contains a standard Platonist formula for the question of the eternity of the cosmos derived from Tim. 4 la-b. The attribution to the Stoics, however, should rouse our suspicion. The Stoa stressed the role of divine providence, as we find in this doxa, but not in order to argue the indestructibility of the cosmos. (3) It is also surprising, both from the point of view of logic and when we compare the Philoníc parallel given above, that P should start off with what is basically a compromise position and not the view that the cosmos is unconditionally destructible or not destructible. (4) Moreover, that Epicurus should be the name-label attached the second position is to some de,

7 Cf. H. DizLs, Doxographi Graeci, Berlin, Reimer, 1879, 19764, 11. The grounds for doubt lie in the fact that intervention can so easily occur in the case of this well-known doxa. For P, writing in the 2nd century, Pythagoras can `cover' Plato, while both Eusebius and Ps. Galen were capable of adding the Platonic name-label on recognizing the authentically Platonic view. Note that we are speaking about P here, not Aëtius himself.

—5—

DAVID T. RUNIA

gree unexpected. Áll that the doxa tells us is that the cosmos is destructible, because it is also generated, just like an animal or a plant. We read nothing about the distinctive trademarks of the atomist position, such as the plurality of cosmoi and the distinction between universe and cosmos (this is found in chapter 2.1 Περι κ~σµου).Once again comparison with the Philonic text is instructive. (5) The choice of Xenophanes as representative of the truly eternalist position is also surprising. In other doxographies it is always the Aristotelian view, while Aristotle represents a position which appears to qualify the one he is usually associated with. But these problems raised by P's brief chapter are enormously compounded when the evidence supplied by John Stobaeus in Book I of his Anthology, in which he appropriates so much material from Aëtius' compendium is taken into account. It is quite certain that through Stobaeus' evidence a considerable amount of material can be added to P, but this material first has to be identified and some understanding has to be developed of the methods used by the anthologist in dealing with it before we can comfortably begin to undertake to process of reconstructing the original text that lay before him. Fortunately I can draw on the detailed analysis of Stobaeus' methods which I published in the first volume of Aëtiana $ There I demonstrated that Stobaeus is an artful, purposeful and reasonably consistent arranger of the material that he found in his doxographical sources. For the purposes of our exercise the following six features of his method should be briefly noted. Stobaeus likes both to coalesce material from different chapters of Aëtius into single longer chapters of his own anthology, and to coalesce doxai attributed to the same philosopher in different chapters into little clusters with a single name-label. 9 Further material from other doxographical sources (e.g. Arius Didymus) and non-doxographical sources (e.g. Plato, Neoplatonic authors) is interposed; for the reconstruction of Aëtius this needs to be filtered out. 10 In his composition of chapters a good deal of mental association occurs, both between names of philosophers and doctrines. 1 1 On the whole he is very complete in his recording and arranging of the material in Aëtius. 12 8

J. MANSFELD and D.T. RUνιn, Aëtiana, pp. 196-271. See Aëtiana, p. 218. 1Ο See Aëtiana, p. 225. 11 See Aëtiana, p. 225. 12 See Aëtiana, p. 222, 236.

9

—6—

A DIFFICULT CHAPTER IN AËTIUS BOOK 11 01 COSMOLOGY

(v) He shows the tendency to write out blocks of doxai when he does not wish to apply the processes of coalescence noted abo νe. 13 (ví) He tends to retain lists of name-labels, including very uncommon ones, much more faithfully than P. 1a Bearing these techniques in mind, we now turn to the evidence that Stobaeus furnishes for the chapter we are trying to reconstruct. The first thing to observe is that Stobaeus adds a great amount of material to P, but at the same time makes it very difficult to retain the systematic structure of the chapter such as P has preserved. The material equivalent to P's chapter 2.4 is divided between chapters 20 and 21. The former, Περi γεν~σεως κα φθορ ~ς, is the last of the chapters dealing with general topics taken from Aëtius' Book I (on bodies, colours, movements etc.). As also occurs in 5 15 and 5 17, Stobaeus mixes general doxai from Book I (i.e. P's 1.24) with more specific doxai from Book 2 (i.e. P's 2.4). It thus forms as it were a transition to the following chapters (5 21-27) which deal with cosmology proper. In the first of these, i.e. chapter 21, with the long composite title Περì κ~σ ιου καì ει ~µΨΥχΟςΚα'. προνο~α διοικ ~ΜεΝος και Gov ëχει τ~~ ~ΥεΜονικ~ν καì. π~θεν τρ~φεται, Stobaeus also includes material from P's 2.4. We shall now look more specifically at what Stobaeus has done with the Aëtian lemmata that can reasonably be assumed to have been present in the original chapter. All of P's doxai can be located in Stobaeus. 15 Not only, however, has considerable alteration and adaptation taken place, but also the basic order of the two sources is quite incompatible. P's first lemma appears to be split up between 20.1 ε (170.10 Wachsmuth, Plato), 21.6c (186.14, Pythagoras) and 20.1f (171.20, the Stoics). P's second lemma (Epicurus) is the 7th and final lemma in 20.1f (172.3), but the text of the two sources diverges quite markedly. The third lemma (Xenophanes) can be located in 20.1f (171.11). The fourth Aristotelian lemma reappears as the first lemma in 20.íf (171.9). Stobaeus has a Platonic lemma at 20.1 ε (170.10 W.), but it only treats the subject of φθορ~. It is interposed among the doxai on ΥΥνεσι4 and φθορ~~from P 1.24. Clearly it is placed there in a quite logical fashion See Aitiana, p. 226. See Αëtiana, p. 236. 15 For the text we still have to use the old edition of C. Wλcηsµ urη, Ioannis Stobaei Anthologii libri duo p riores qui inscribi solent Eclogae physicae et ethicae, 2 vols., Berlin, Weidmann, 1884, repr. Zürich 1974. In its treatment of the Placita it was strongly influenced in its presentation by Diels' edition of the Doxographi Graeci, which had appeared five years earlier in 1879. 13

14

—7—

DAVID Τ. RUMA

as a pendant to the Heraclitean view that ουδ~ν µ ~νει, which, as Wachsmuth rightly saw, was added from Plato Cratylus 402 a. The doxa itself is based on the two statements in the Timaeus (27d-28b, 4 la), the former of which was connected in antiquity with the doctrine of Heraclitean flux. We should note, however, that Stobaeus nowhere indicates a Platonic doxa on the question of the cosmos' y ~νεσις, unlike what we found in P, provided of course the Eusebian evidence is accepted. The doxa of Pythagoras is postponed by Stobaeus to § 21.6c (186.14 W.), where it is included as first lemma in a coalesced group of four doxai taken from four separate chapters. The formulation given by Stobaeus – yενητ~ν κατ ' ~π~νοmmν, o~~κατ~~χρ~νον — differs from what we find in P, and is to be preferred, since it is confirmed by Theodoret (4.16), although he drops the name-label. It is important to note that the Pythagorean doxa says nothing on the cosmos' φθορ~. Three further lemmata remain in chapter 21.6f (187.9-12 W.) dealing with the cosmos' genesis. 16 The first provides us with a real puzzle. The doxa attributed to Herac litus is almost identical to the Pythagorean lemma cited just before: v κατà χρ~νον. 21.6c Πυθαγ~ ρας φησí γενητ~ν κατ ' ~πινοrnν τ~ν κ~σµον, ο 21.6f }ρ~κλειτος οÚ κατà χρ~νον ε~ναι yενητ~ν τ~ν κ~σµον,~λλ~~κατ ' ~π~νοιαν.

It is just possible that Stobaeus has messed up an original antithesis. Since the Pythagorean doxa is confirmed by P, Herac litus would then have been given the view that the cosmos came into being κατà χρ~νον, o~~κατ' ~π~νοιαν. This would mean we would have a doxa expressing temporal generation. But this reconstruction, though it could perhaps claim the authority of Aristotle (cf. DC 279b12-16), is unlikely. In P 1.23 it is recorded that Heraclitus affirmed eternal motion for eternal beings, which surely would include the cosmos.'' It is much more likely that in the process of reorganizing the material a single lemma with two name-labels has been split into two. This means we so far have no lemma asserting the unconditional or temporal Υ~νεσις τov κ~σµον. After the Heraclitean doxa we encounter two further views, attributed to the totally obscure Epidícus 1ß and Archelaus. But their subject has 16 At the end of the chapter Diels published three extra lemmata under the heading Hill ~χει τ~~~yελονικ~ν ~~κ~σµος,located in Stobaeus 186.26-187.7. These patently do not belong to

the subject matter of P 2.4 and will be ignored in the present reconstruction. 17 Cf. the Heraclitean doxa at DK 22 Β30. 18 He is only named here in the Placita. Dims, Doxographi Graeci, 100, n. lis non-plussed.

—8—

Α DIFFICULT CHAPTER IN AETIUS BOOK II ON COSMOLOGY

subtly shifted. It is not so much whether the cosmos is γενητ~ ς, but how the cosmos came to exist: it occurred either by nature or by warmth and ensoulment. 19 We may safely assume that a δrnιρεσις between a teleological and mechanistic doctrine of genesis lies behind these doxai.20 In P's first lemma a cause is also given, namely God, but this view is not found in Stobaeus. In order to explain P's conflation, it may be surmised that a third doxa belonged to this group, namely the Stoic view that the cosmos came into being through the intervention of God (i.e. ~iτò Οεov yεγενfjσθαι τòν 21 κ~ σµον), which Stobaeus might have left it out because the Stoa is amply covered earlier on (20.1e, 21.5). Needless to say, however, this explanation is speculative and other explanations of P's alteration can be given. (6) Eight further lemmata are found earlier in chapter 20.1f (171.9172.4 W.). As the heading that Wachsmuth affixes to this block indicates, Diels' treatment of this group is quite remarkable. In his reconstruction he changes the order to a sequence 12-11-13-6-7-8-9-10. The reason for doing this is quite transparent. He wants to retain the order as found in P. But is this justified? As we noted when discussing Stobaeus' methods, he very often records doxai in a block, retaining the presumed order in Aëtius. So we would have to have very good grounds for following the Dielsian reshuffle. These, I believe, cannot be sustained. Certainly it must be conceded that the Aristotelian lemma is difficult. P appears to regard it as an exceptional position, qualifying the doxa of the ~~δι~της of the cosmos conventionally ascribed to Aristotle. In Stobaeus, however, it precedes that lemma. Moreover the connection between the second and third lemma in Stobaeus' block seems secure, since the periodicists qualify the position of cosmic A mistake may be suspected (but the name does occur in P κοττus' list at 155.35 Hεrτπν, so it must have been present in the Stobaean text he read). One might be tempted to think of Epictetus, but this would give rise to chronological problems. Εnι [δι]κ(ονρ)ος must be considered out of the question. One may compare the strange things that happen with the name Dicaearchus in the witnesses at P 4.2. 19 Mawaκa had emended the mss. reading ~λιιυχιας to ~µψυχριας,and this was accepted by DIELS in his Doxographi Graeci and Wac ηsµ uτη. But this is unnecessary and dubious (the word is not found in extant Greek literature). In his Fragmente der Vorsokratiker DIELS reverts to the mss. reading without comment (it is briefly noted by Κanνz in the Nachträge to the sixth edition, 2.421.10). 20 Cf. the lemma on Archelaus' theology at Stobaeus 1.1.29b (34.16 W.): 'Αρχ~λαος ~~ρα καi Ιο~ν TUV θε~ν, ο' λ~ντοι κοοµοπο~~ν TUV νοiν. Note too that the view of genesis through nature and not god is attributed to Straton at Cicero, Luc. 121, Nat. deor. 1.35. But this is a mechanistic view, and in the chapter on ~ρχο~~Strato's are precisely (θερµ óv) κα 'yvχρ~ν (at Stobaeus 1.10.12, 124.18 W; not preserved in P)! 21 Cf. Phílo's doxography at Aet. 8: οi Στωικoι... γεν~σεως δ ' α~τοι [Το~~κ~σµου] θε ~ν aιτιον, ~τι Θε~ν ... ΦEΟΡdς δ~~µηκ

τ

9

DAVID T. RUNIA

eternity, as indicated by the fact that they are introduced as of φ~Μενο~~δY, which implies a contrast. Nevertheless it is difficult to retain the initial position that Stobaeus gives the Aristotelian lemma, because although the term φθορ~~is not used, this seems to be its theme: Aristotle is taken to support the position of a partial destruction confined to the sub-lunary realm.22 We shall return to this lemma later. The sequence of the remaining five lemmata can easily be retained, as Diels has done. If the sequence with the previous two is maintained as well, this means that Aëtius must have discussed the destruction of the cosmos after the eternalist view. The first two of these use the adjective φθαρτ~ς. This was also the case for the Platonist view cited earlier in the chapter. It is plausible, therefore, that this last-mentioned lemma was placed here as well, probably after the Stoic view. It too offers a qualification of the destructionist position. The last three doxai all have a verb φθειρεσΟαt and discuss the process or agent of destruction. The addition of ( Υíνεσθαι κα~> to the Empedoclean lemma, ad proposed by Sturz, is thus clearly misguided from the viewpoint of the structure of Stobaeus' chapter. This additional consideration clearly strengthens O'Brien's attack on the emendati οn. 23 The final doxa is attributed to Epicurus and differs quite strikingly from what is found in P: P Σπ~κονρος φθαρτ~ν, óτι κα~~Υενητ~ν, ~< ζι ον ~ς φντ~1. Stobaeus 'Επ~κονρος πλεíστοtς τρ~ποt< τ~ν κ~σµον φθε~ ρεσθαι• κα~~y~ρ ülς ζüο ν κα~~ ~g φντ~ν κα~~πoλλaχwς It is reminiscent of what we find in P 2.2, 2.13 and 2.22, in which Epicurean doxai are given last and emphasize the fact that there are various possible configurations or explanations possible. One might label this a `modal' view because it emphasizes various possibilities. If a choice has to be made between P and Stobaeus' text for this last lemma, then it would seem easier to explain how the former might have altered the original than the latter. P has `normalized' the Aëtian doxa because he needed a representative for the destructionist view and it was well-known that Epicurus held that position. The real headache is the Aristotelian lemma. Here there is a real dilemma. If the Stobaean block of 8 lemmata is retained, then it is difficult to explain why Aëtius should mention the passibility of part of the cosmos 22 There is an extensive doxographical tradition, found primarily in the Church Fathers, that for Aristotle providence ceases below the supra-lunary realm. Cf. ATHENAGORAS, Legatio 25.2 and the parallel passages listed by MARcονιcκ in his edition, p. 83. 23 In the article cited in n. 1.

—

10 —

A DIFFICULT CHAPTER IN

AËTIUS BOOK 11 01 COSMOLOGY

before he mentions its eternity. It seems likely, therefore, that it was placed later. My suggestion would be after the Platonic lemma on the conditional destructibility of the cosmos. There is a link between the two because of the doctrine of partial providence attributed to Aristotle and already outlined in the previous chapter 2.3 in P. But it remains rather awkward to furnish an explanation as to why Stobaeus put it at the head of this sequence of 8 Aëtian lemmata. It follows reasonably on well from id, which discusses the passibility of matter, but not so well from le, unless it is specifically meant to offer a partial antithesis to its final words τ~ν τ~ν óλων εiς ßv2 µεταβολ ~ ν. There is no more evidence for Aëtius' original chapter than the three witnesses discussed so far. The Isagoge to Aratus of Achilles, which contains some Aëtian material, does not broach the subject of the cosmos' origin in the temporal sense. The so-called Doxographica Pasquali, scholia to Basil's Hexaemeron containing excerpts compiled from material similar to but not wholly the same as Aëtius, offer only the conventional schema found in Ph ilo, without any direct relevance to our chapter. 24 Some further more general help may be forthcoming if we look further afield and examine the huge number of doxographical parallels on this subject in ancient literature. I have collected these and placed them in an Appendix at the end of the paper. The theme goes back at least to the Timaeus and from the reference in Aristotle's Topics we learn that by that time it was already a favourite topic of discussion (i.e. a πρ~βληµα or a 01010. From these parallels the following can be learnt. Although the questions εi ~~κ~σµος γεΝητ~ ς fi ~y~νητος and εï ~~Κ~σµος 4)θαρτ~~ ~~0αρτος are separate subjects, and are sometime cited as such (e.g. in Aristotle DC 1.10 279b4 and the Stoic text at DL 7.132), they are most often treated together, which gives a matrix of three basic positions as found in the Philonic text cited above, since the view that the cosmos is ~y~νητος καì, φθαρτ~ς finds no takers. It is important to observe that in Quintilian and Galen a direct connection is made between the question of the cosmos' beginning and the cause of its coming into being, e.g. God or chance. In Philo the question of the efficient cause is also prominent, as it was originally in Plato (but not in Aristotle's De Caelo). In Ambrose the emphasis shifts to whether the cosmos itself is divine, which in Ph ilo is noted only for the eternalist position (Aet. 10). 24

On these documents see J. MANSFELD and D.T. Ruνιλ, Aëtian, pp. 299-312. —

11 —

DAVID T. RUNIA

(iii) The choice of Xenophanes-Parmenides-Melissus fir the true eternalist position has deep historical roots. One should compare Cicero Acad. 2.118-119, where doxai on the archaiare combined with views on the cosmos, and this view is emphasized for both Xenophanes and Melissus (but not for Parmenides, while Aristotle is added later). Even earlier Theophrastus asserted the same for Xenophanes and Parmenides, in a text that probably goes back to his Physics rather than the Φυσικ~ν δ~ξαι (as Diels surmísed). 25 3.

12EC ΟΝSTRUCTIOΝ OF THE CHAPTER

All the evidence available for the reconstruction of Aëtíus' chapter has been assembled. Although not every issue can be resolved beyond doubt, I would defend the following text as the most likely reconstruction of what Aëtius' original text looked like. δ '. Ei ~φθαρτος ~~κ~σµος Πυθαγ~ρας Ηρ~κλειτος γενητ~ν κατ ' ~π~νοιαν τ~ν κ~σµον, ον κατ òε χρ~νον. Ι ( οι Στωικο~~υπ~~ΘΕο'~~γεγεν~σθαι TUV Κ~σµον). 2 'Επ~δικος υπ~~φ~σεως γεγεν~σθαι τ~ν κ~σµον. ~~κα~~ ~µψυχ ~ ας 3 σνστ~ναι τ~ν κ~σµον. 'Αρχ~λαος υπ~~Θερµο Εενοφ~νης Παρµεν ~ δης Μ~λισσος ~γ~νητον κα~~&~διον κα~~ ~φθαρτον TUV κ~σµ ov. ~~νιον υπ~ρχειν περιοδεντικονς [κα~] 4 ο~~φ~µενοι ~δ~τ~ν διακ~σµησιν α ειναι φασι χρ~νους, καθ ' οûς κατ~~τα~τ~~κα~~ωσα~τως γ~γνεσθαι π~ντα κα~~την αυτην διασυ~ζεσθατ το~~κ~σµου δι ~ταξ~ν τε κα~~διακ~σµησιν. 'Αναξ~µανδρος'Αναξιµ ~ νης 'Αναξαγ~ρας 'Αρχ~λαος ∆ιογ~νης Λε~κιππος φθαρτ~ν τ~ν κ~σµον. κα~ 5 Ο~~Στωικο~~φθαρτ~ν TUV κ~σµον, κατ ' ~κπ~ρωσιν δε. Πλ~των φθαρτ~ν Μ~" τ~ν κ~σµον,~σον ~π~~τ~~φ~σει, α~σθητ~ν γαρ ειναι, δι~τι κα~~σωµατικ ~ ν, ου ~ν φθαρησ~µεν ~ ν γε προνο~~~κα~~συνοχ~~Θεο~ . 'Αριστοτ~λης τ~~υπ~~την σελ~νην τον κ~σµον~ρος µ παθητ~ν,6 ~ν ~~κα~~τ~~ περ~γεια κηρα~νεται.7 'Εµπεδοκλ ~ ς τ~ν κ~σµον φθε ~ρεσθαι8 κατ~~την ~ντεπικρ~τειαν τον νε~κονς κα~~τ~ς φιλ~ας. ~~µε ~ζονος τ~ν µικρ ∆ηµ ~ κριτος φθε~ρεσθαι τ~ν κ~σµον το ~ τερον νικ~ντας. 25 As argued by P. Srετκτν~eτz, Die Physik des Theophrastos von Eresos («Palingenesis», I), Bad Homburg, Gehlen, 1964, pp. 334-351; Diels' position has recently been unsuccessfully defended by L. ZHM UD, Revising Doxography: Hermann Diels and his Critics, «Philologus», CXL.V, 2001, pp. 219-243.

— 12 —

A DIFFICULT CHAPTER IN

AËTIUS BOOK 11 01 COSMOLOGY

13. ' Επ~κουρος πλε~στΟις τρ~πΟις TUV Κ~σµΟν φθ £~ρ£σθαι ' ΚαL γ~ρ ~ς ζ(λ)OV και. ~~ φντ~ν καì, πολλαχwς. 9 ex S: P Cyril Q Πυθαγ~ρας καï ο Στωικοι yενητ~ν vi~~θεο~~ten κ~ο ον, Ε Πυθαγ~ρας κω. Πλ~των κα~~ ii Στωικο~, G Πυθαγ~ραν κα~~Πλ~τωνα. 2

coniecimus ex P. ~µψυχρι aς conj. Meineke, secuti Diets DG et Wachsmuth. 4 sect. Diets, Wachsmuth. sect. Diets. 6 P Ε G Cyril naθητ~ν, Diets nαθΗτικ~Ν, fortasse e P 1.3 876F. 7 Cyril codd. nεραινεται, Q κερ~ννυται ut vid. s (γ~νεσθαι κα~> φθειρεσΟαι conj. Sturz, secuti Diets DG Wachsmuth, non recte. 9 ex S; P Ε Q Cyril post lemma Platons Επ~κο~ρος φθαρτ~ν, ~τι κα~~γενητ~ν, ~ς ζ~ον ~ς φντ~ν. 3

51 —; 22Α10 DK; S 3 —; § 4 60Α14 DK; S 5 21Α37 DK; 28Α36 DK; 30Α9 DK; § 6 cf. SVF 2.597); 57 12Α67 DK; —; 59Α65 DK; 60Α14 DK; 64Α10 DK; 67Α22 DK; § 8 SVF 2.575; § 9 —; § 10 Τ19 Gigon; § 11 31Α52 DK; § 12 68Α84 DK; § 13 fr. 305 Usener.

4.

THE

STRUCTURE OF THE CHAPTER

AS RECONSTRUCTED

If this analysis is on the right track, then Aëtius 's chapter has the following features. (i) He has chosen a symmetrical arrangement, in which first the question of y~νεσις is dealt with (5 1-4), followed by the view of cosmic eternity (5 56), and then finally the question of φθορ~~(5 7-13). This means that he has deliberately rejected the conventional approach to the subject found in Philo and elsewhere. It also means that no attempt is made to relate the treatment of the question to the further complication of single cosmos and multiple or infinite cosmoi, which was broached in an earlier chapter (P 2.1). This contrasts with the treatment found in Aristotle De caelo 1.10 and Philo. (ií) The beginning of the chapter is surprising, because one would expect it to begin with the straightforward view that the cosmos was γενητ~ς, which was always the majority position, as Aristotle and other sources emphasize (e.g. Aristotle 279b12 yεν~.ενον ... ~παντε( ε~να~~φcσι...; Ambrose in Hex. 1.1.4 plurimi). It might be surmised, therefore, that Aëtíus began blandly with of πλεïστοι γενητ~ν TUV κ~σλον vel sim (if there had been a long list of names, Stobaeus would not have passed it by). One important consideration restrains us from including this in the reconstruction, namely that Theodoret does not encourage such a reading but begins with the Pythagorean position, as we find in P. (iíí) In the two lemmata that follow (three if the conjectured Stoic position is added), Aëtius moves from an adjective ( γενητ~ς) to verbs (yεyεν- 13 —

DAVID T. RUNIA

~σθαι, mvστ~ναl). As was noted earlier, in some doxographíes the question of causality is related to the main subject in question. Themes of causality also return in later chapters, e.g. 2.6 and 2.8 in P. Theodoret confirms that he has an eye for the systematics of the doxography by reporting the eternalist position as áyενητον παντελ~ς κα~~ ~να~τιον, i.e. not having a cause as mentioned in the case of the other two positions. (ív) For the imconditional eternalist position Aëtius surprisingly gives only the Eleatic tradition, which is consistent with the position given on y ενεσις and Ψθopá in P 1.23. The strong tradition that Aristotle represents this view par excellence, possibly with Pythagorean predecessors, is ignored. The position of the periodicists, i.e. asserting a δrnκ~σµησις αι~νωως, is precisely the view that Aristotle thought was not entirely incompatible with his own at DC 1.10 279b14-16, 280a11-27, where Empedocles and Herac litus are specifically mentioned. Aëtius here leaves the lemma anonymous, using these names elsewhere (and also that of the Stoics, who later became the standard representatives of this view, as we find in Philo). It is a compromise position, and as such obviously provides a transition to the destructionists. Next follows a rather long list of philosophers who hold that the cosmos is prone to destruction. These, it is worth noting, are all Presocratics. More detailed views, involving explication or qualification, are given for Plato and the Stoa. After these views we have placed the Aristotelian lemma for reasons given above. Then, parallel to what we found earlier, Aëtius moves from the adjective φθαρτ~ς to the verb φθειρεσθαt, i.e. causality comes into view. There is no lemma attributing this to God, since that would be ο~~Θ~µις,as implied in the earlier Platonic doxa. 26 The doxa of Democritus is more mechanistic than that of Empedocles. It is logical for Epicurus to bring up the rear, as we noted above. Finally some words should perhaps be devoted to the title. I have retained the title as given in P. There is little alternative, since Stobaeus gives us no information on that score (he leaves out the subject of the cosmos' genesis or destruction in the titles of both his chapters). There is certainly a discrepancy between the title and the contents of the chapter. As noted earlier, Philo provides an important parallel for a doxography devoted to the question of the cosmos' destructibility which includes discussion of its genesis or eternity. It should be recognized, however, that the parallel is not quite adequate. Ph ilo really does focus primarily on the ques26

Cf. PLATo, Tim., 4ta (cited by Stobaeus at the end of his S

39 43 (perhaps derived from Aristotle's De philosophic). -

— 14 —

20, 1.181.7 W.), PraLl, Aet.,

A DIFFICULT CHAPTER IN AËTIUS BOOK II ON COSMOLOGY

tun of whether the cosmos will be destroyed, whereas in our chapter this is not the case. The question of titles in Aëtius and the witnesses to his work is too complex to deal with adequately here, 27 but the following can be said. Α title fully covering the options would have been somewhat cumbersome, í.e. `whether the cosmos is generated or destructible or everlasting'. Aëtius likes to keep his titles compact and almost never includes theoretical options in the title itself (the only exception in the titles recorded by P is 5.25).28 For example, when discussing the nature of the soul, he gives as title (4.7) Περì ~φ8αρσíας Ψυχ~ς, whereas the option that the soul is destructible plays a prominent role. This is parallel to our chapter, except that the question of the soul's genesis is not touched upon. Turning now, on the basis of our analysis, to the structure of the chapter, we may conclude that from the systematic point of view it yields the following result. Α. cosmos γενητ~ς (missing?) with regard to time a. but not in time (= § 1) with regard to cause by God (= 5 2) by nature (= 5 3) by heat/cold (= § 4)

b

B. cosmos ~ιδtoς unconditionally (= 5 5) conditionally, í.e. periodically (= 5 6) C. cosmos φθαρτ~ς 1. unconditionally (= § 7) 2. conditionally in ~ ΚΠ~Ρωσtς (= 5 8) οΘαρτ~ν/ο~~φθαρησ~µενον (= 5 9) partial γ~νεσt~/φΘορ~~(= 5 10) 3. with regard to cause through νε~κος/φιλ~α (= Sil) through collision (= 5 12) through diverse ways (= 5 13 )

'Ν

It is touched on in Aëtiana, p. 180, but could not be dealt with fully. Note also the full title with four options recorded on in the margin of ms. a as stated in the critical apparatus of G. LACHENAUD'S edition, Plutarque (Euvres morales Tome ΧΤ12: Opinions des Philosophes, Paris, Les Belles Lettres, 1993, 105. 27 28

— 15 —

DAVID T. RUNIA 5. DNERGENCE FROM DIELS' RECONSTRUCTION

Before the structural features of the chapter are further discussed we should first note that our reconstruction diverges rather markedly from that presented by Diels in his Doxographi Graeci. The reason for this is two-fold. Diels uses two columns in his reconstruction and in his left column preserves the order of P. We have attempted to produce a single text and have found that, if we are to do justice to the Stobaean evidence, we cannot retain P's order without change. The difference between Diets' treatment of Stobaeus' evidence and ours is marked. It is best illustrated by means of the following tables: Diets

Runia

Runia

Diets

1 2 3 4 5 6 7 8 9 10 11 12 13

la 9 lb 3 4 7 8 11 12 13 5 10 6

1 2 3 4 5 6 7 8 9 10 11 12 13

1+ 3 —

4 5 11 13 6 7 2 12 8 9 10

As noted earlier, the remainder of the doxai that Diels includes in his reconstruction do not belong in this chapter and can be disregarded. It may be concluded therefore, that this is one of the chapters in which it can be proved that Diets' decision to retain the two columns of P and Stobaeus rather than try to make a single reconstruction was incorrect. 29 It is impossible to keep the order in P if one is to do justice to the evidence in Stobaeus. But this raises an interesting further question. A staunch defender of Diels might argue against my method, for example by claiming that it would be difficult to get from the reconstruction to an epitomized text such we find in P. I do not think this is the case. It is quite easy to see how P might have reached his result if we take into account the fact that he 29

See conclusion reached by J. MANSFELD and myself at Aëtiana, p. 332.

— 16 —

A DIFFICULT CHAPTER IN AETIUS BOOK 11 01 COSMOLOGY

was influenced by the standard doxographícal treatment of the question such as we find in Philo, from which Aëtíus in fact deviates. This involves coalescing the first, eighth and ninth lemmata, as well as other manipulations which we have already discussed. 6. WΗΑT DOES THIS CHAPTER

TELL US ABOUT AËTIUS?

Finally we must not shirk the most difficult question of all. What does this chapter, as we have reconstructed it, tell us about Aëtíus as a doxographer? In many respects the man remains a complete mystery. What do we really know about him? At the end of volume I of Aëtiana Mansfeld and I tried to reach a limited number of reasonably firm conclusions, but even these have been contested. For example, can we really be absolutely certain that the conventional use of the name ' Α τιος, i.e. `eagle-man', is justified? Both Michael Frede and Jan Bremmer have placed question marks against this practice and their arguments are worth considering. 30 But there are limits to scholarly caution and scepticism. Someone wrote this compilation of placita and that person must have done so at about the beginning of the second century C.E. The best thing is simply to look at what he has done, and for that purpose the chapter that has been reconstructed is an excellent example. Firstly, it seems to me that the organized structure of the chapter, as we have reconstructed it, is quite deliberate. This is demonstrated by its well thought out nature and its neatness. It is also reinforced by numerous other examples in the work. Aëtíus will have organized his chapter like this primarily for systematic reasons, I believe, but one might also consider whether mnemonic considerations were an additional factor. A chapter is easier to remember if it is well-organized. Secondly, it is plain that the chief instrument used by Aëtíus to organize his chapter is the diaeresis. As the diagram indicates, the chapter as analysed above consists of seven diaereses nestled within each other at three different levels. The researches of Jaap Mansfeld (and to a lesser degree myself) have shown that the dialectical and sceptical origins of the doxographical method encourage the use of the diaeresis, both in the form of a dichotomy (either A or B, whereby B often can be formulated as not30 See M. FREDE, art. cit. (n. 2), pp. 144-147, J.N. ΒΒΒΜΙ mR, Aëtíus, Arius Didymus and the Transmission of Doxography, «Mnemosyne», LI, 1998, pp. 154-160.

— 17 —

DAVID T. RUNIA

A, e.g. the cosmos is either generated or not generated) or in the form of a list of alternatives (A or B or C etc., e.g. the cosmos is made by God or by nature or by mechanical factors). In the last example the alternatives can hardly be called exhaustive — there are many more possibilities —, but even so it seems as if Aëtius has thought about the question and does cover the various possibilities rather well. Thirdly, it is worth noting that Aëtius loves to make use of systematic positions that bridge alternatives and allow transitions from the one doxa to the next. Comparison with the Philoníc doxography is again instructive. There Ph ilo puts forward two alternatives, the cosmos is either generated and destructible or ungenerated and indestructible, followed by a compromise position, called a µικτ ~~δ~ξα, that it is generated and indestructible, i.e. the Platonic view which Ph ilo himself favours. This is basically a diaphonia, but, because there are two questions involved, i.e. A or — A and B or — B, intermediate views are theoretically possible. These are determined by a technique reminiscent of that used in truth tables, i.e. A and — B or — A and B, in which only the first option is practically viable. When we turn to Aëtius, however, we see that he does not structure his chapter in this way at all. He keeps the two questions well apart, first dealing with the generatedness of the cosmos and later discussing the question of its destructibility. The view that it is neither generated nor destructible, i.e. everlasting, is placed in between these two. One might ask why he has done that. It cannot be meant as a compromise view, because it is opposed to both. It almost appears as if aesthetic considerations have played a role, i.e. the symmetry achieved by having — A — B placed in between + A and + B. But it should also be noted that the conditional form of everlastingness, i.e. that destruction occurs periodically and never-ceasingly, clearly is a compromise position between — B and + B, which is dealt with in the remainder of the chapter. Such compromise views are ubiquitous in Aëtius' doxography as a whole. All in all, these considerations reinforce my thesis that Aëtian chapters are very consciously organized in a quasi-archítectural fashion, involving considerations of deliberate positioning and symmetry. Fourthly, a brief glance should be given at the name-labels used in the chapter. These are quite copious in number, 20 plus one anonymous view. Some of the names are pretty obscure (notably Epidícus, but also some of the Presocratics). Others are unexpected, e.g. the references to Herac litus and Aristotle. Yet others are used to represent standard doxographical positions, e.g. Plato and the Stoics. Without going into further detail, I would argue that the thesis put forward by Mansfeld and myself, namely that the - 18 -

A DIFFICULT CHAPTER IN

AËTIUS BOOK 11 01 COSMOLOGY

names are subordinate to the systematic positions represented, is very much confirmed. Fifthly and finally, how does Aëtius' treatment compare with that furnished by others? In the course of my paper I have referred to parallel treatments such as those found in Cicero, Ph ilo, Quíntílian and Galen. In the appendix a much fuller list of dialectical-doxographical parallels involving nearly 20 authors is given. It goes without saying that these cannot be discussed in the detail they deserve. It is plain, however, that the origins of the treatment is recorded in Plato and Aristotle and on the basis they provided it was subsequently further elaborated.3 Ι Perhaps Theophrastus played a role in this process, but that is by no means certain. It would require a second paper to trace the paths of this development. All I would wish to claim at this point is that none of the parallels given comes anywhere matching Aëtius' chapter, both in terms of method and extent of elaboration. On the basis of this evidence I tend, therefore, to the conclusion that Aëtius as a doxographer was sui generis, with an idiosyncratic style and method which needs to be further investigated. I recognize that to some degree this is an argumentum e silentio. There may have been other similar compendia. As Mansfeld and I have argued, the doxographícal tradition is marked by a fluid and continually shifting character. 32 Ever since Diels, Aëtius has been seen as its central representative. This may well be a mistake. Only from our point of view does Aëtius seem the doxographer par excellence. In an ancient perspective he may well have been a rather unusual but also quite talented compiler and organizer of philosophical views.

31 Of course there must have been earlier debate on the subject. But there is no record of its being systematically reported in a doxographícal form, although this may well have occurred in the writings of a sophist such as Hippias. On pre-platonic doxography cf. J. MAxSFELD, Aristotle, Plato, and the Preplatonic Doxography and Chronography, in G. CAMBIANO (ed.), Stοriοgrafia e dossografia nella filosofia antica, Torino, Tirrenía, 1986, pp. 1-59, reprinted in J. MANSFELD, Studies in the Historiography of Greek Philosophy, Assen, Van Gorcum, 1990, pp. 22-83. 32

Aëtiana, p. xIx.

— 19 —

DAVID Τ. RUNIA

APPENDIX

DIALECTICAL-DOXOGRAPHICAL PARALLELS Plato Tim. 27c4-5, ~µ äς δ~~Το~ς περ' τον παντ~ς λ~γο~ς nοιει~Οα~~n~~ µ ~ λλ~ΝΤας, ε. γεγονεν ~~κα~~ àγεΝ~ς ~στλν; 28b4 7 σκεΠτ~ον δ ' ovΝ πε (zinov [κ~σµο ~ ]... π~τερον ~ν àεí, γεν~σεως ~ρχ~ν ~χων ο~δεµιαν,i' γεγονεν, ~π ' ~ρχ~ς τ~νος ~ρξáµενος.Aristotle Top. 1.11 104b8, ~νια [Τ~ν προβληµ~ των χρ~σΨΨον] δ8 πρ~ς Τ~~εïδενα µ ~νον, στον π~τερον ~~κ~σµοςàιδwwς ~~ο~; 1.14 105a24, φυσικα [προτ~σεις] δ~~ο~ον π~τερον ~~ κ~σµοςà~διος ~~ο~; DC 1.10 279b4, λ~γωµεν... π ~ τερον ~ y~νητος γενητ~ς κα ~φθαρ τος η φθαρτ~ς, &&εξελθ~Ντες nρ~τερον τàς Τ ~ν ~λλων vnoλ~Ψεις... γεν~µενον Μ ~ν ον ~παντες ε~να~~ φασιν, àλλà γεν~ µενονσ µ Y1 à~διΠν, σ δY φθαρτ~ν ~σπερ ~Τww~ν αλλο Τ~Ν φ~σ&& συνισταµ ~ νων, of δ ' ~ναλλ~4 ~τ8 λ~ν ο~τως ~τ8 δ8 αλλως ~χ&&v φθειρ~µενον,κα ΤΟ~ΤΟ ~εì διατελειν ο~τως...; cf. 2.1 283b31, καì, δ~~ 'riις δ~ξης τις Παρ~~ Τ~ν ~λλως λεγ~ντων καì γενν~ντων α~τ~ν. Theophrastus Phys. Dix. fr. 6, 8 Diels (but probably drawn from his Physics). Stoica ap. DL 7.132, καθ ' ~ν ζητε~τα~~ τ ' ο~σια α~το~~ [κ~σµο ~ ]... κα~~ε ενητ~ς 111 ~γ~νητος... κα~~ει φθαρτ~ς 1j ~φθαρτος... Varco ap. Seri. Comm. in Verg. Georg. 2.336, 3.1.248 Th ilo (text at Diels DG 198 n. 1), Varro sutem in satura quae scribitur de salute ait mundum haud natum -

esse neque mori. Plato autem natum at non mo ri . Metrodorus sutem neque natum neque mori. Zenon ex hoc mundo quamvís aliqua íntereant, tarnen ipsum perpetuo m an ere quia inhaereant eí elements e quibus generantur materiae ut dixit crescere quidem sed ad interitum non pervenire manentibus elementis a quibus re νalescant. Cicero Acad. 2.118-9, DID 1.20-21, Tim. 5. Philo Aet. 3, ~ξιον ο ν το~' ζητο~νΤας ε. ~φθαρτος ~~κ~σµος...§ 7, τριττα'1 δ8 περì ΤΟ~~ζΗΤουµ ~νου γε'~νασι δ ~ξα,

Τ~ν µ ~" à~διον 'òv κ~σµο ~~φaµενων,~γενητ~Ν τε καì, ~ν~λεθρον, Τ~ν δ8 ~~ εναντιας γενητ~ν τε Κ&. φθαρτ~w δ ' Οï παρ ' εκατερων ~κλαß~ντες, τ~~ µ8ν γενητ ~ν παρà τ~Ν ~στ~ρων παρà δ8 ~ν προτ~ρων τ~~~~ ΘαρΤον, µυcrfν δ~ξαν àπ~λιπον, γενητ~ν κα ~φθαρτον ο ηΟ~ντες αYΤ~ν ε~ναι; Ebr. 199, Her. 246, Ορif. 54, Abr. 162 3. Quintí1ían Inst. Or. 7.2.2, ut in [quaestionibus] generalibus `an atomorum concursu -

mundus sit effectus, an provídentia regatur, an sit aliquando casurus'. Galen De propr. p/ac. 2 Nutton, igitur díco quod non habeo scientism utrum mundus sit generatus...; De exper. med. 19.3 Walzer; De locis affectis 3.5, 8.159.6 Κ~Εn; De pecc. dign. 3.4 46.23 De Boer; PHP 9.7.9 (list of theoretical questions), οµο ~ ως δ~~κα. ε γεννητ~ς ~γ~Ννητος δδε ~~κ~σµος,~σΠερ 'ε κα ει γεγον~τος α~το~~Θε~ς τις ~Υενετο δηµιουργ ~ς ~~θε~ς µ ~ν ο~δεíς, αιτ~α δ~~τις ~λογ~ς τε κα~~~τεχνος ε~ργ~σατο ΚαΤU τ~χην ο~τως καλ~ν αΥτ~Ι... etc. Tertullían Apo!. 11.5, 47.8. Minucius Felix 34.1 4. Ps. Galen Hist. Phil. 17. Lactantíus Div. Inst. 2.10.17-25, 7.1.6-10. Marius -

Victorinus in Cic. Rhet. 235.27 Háhm, ergo, ut díximus, ex his, quae in opinion suns posits, probab ile colligitur argumentum, sí dícas inferos esse vel non esse, deis esse vel non esse, mundum natum, mundum non esse natum. Ambrose — 20 —

A DIFFICULT CHAPTER IN AËTIUS BOOK II ON COSMOLOGY

Exam. 1.1.4, ipsumque mundum semper fuisse et fore Aristoteles usurpat dícere:

contra autem Plato non semper fuisse et semper fore praesumit adstruere, plurimi vero non fuisse semper nec semper fore scríptis suis testificantur. Augustine c. Acad. 3.23, scio mundum istum nostrum... aut semper fuisse et fore aut coepisse esse minime desiturum aut ortum ex tempore non habere, sed habiturum esse fi nem aut et manere coepisse et non perpetuo esse mansurum; DCD 18.41. Simplicius in Arist. Phys. 1121.5ff. (remarkable doxography which integrates with the question of the duration of the cosmos also the question of single or multiple cosmoi and the question of motion: a. infinitists, cosmoi come into being and pass away (Anaximander, Leucippus, Democritus, later Epicurus); b. unícists, cosmos äιδιος (Plato, Aristotle); c. unícists, but periodically changing (Anaximenes, Heraclitus, Diogenes, later Stoa, special position Empedocles); d. unicists, cosmos had beginning in time (Anaxagoras, Archelaus, Metrodorus); e. special position, Plato as interpretated by Alexander; note especially some of name labels, esp. Arche laus, Diogenes, Metrodorus (not in Aëtíus, but cf. Varro); especially intriguing is his division between early and later philosophers, which might hint at a Theophrastean origin (but also found at Cleomedes 1.5, 27.11 Todd). For further texts see J. Pépin, Théologie cosmique et théologie chrétienne, Paris 1964, 79ff.

—21—

.JAAP MANSFELD Universiteit Utrecht

FROM MILKY WAY TO HALO ARISTOTLE'S METEOROLOGICA, AATIUS, AND PASSAGES IN SENECA AND THE SCHOLIA ON ARATUS

I. INTRODUCTION

One of the aims of the present paper is to find out to what extent a group of `meteorological' chapters, viz. chs. one to six plus eighteen, t in Book ΙII of the Aëtían Placita depend on and are derived from Books I to III of Aristotle's Μeteοrolοgica. 2 Comparison will not be pressed beyond the limits of resemblance. Another aim is to study the macro-structure of this rather large section of the Book. This entails that we look at individual chapters, at significant sections of chapters and at specific individual lemmas in context, rather than at the whole series of chapters in every detail. And I shall not only look at verbal similarities or similarities 1 Thus leaving to one side chapters dealing with other topics also dealt with in Meteorologica Books I to III, such as winds (Placit. 3 7) or earthquakes (Placit. 3 15). For the group of

chapters on the position etc. of the earth (largely dependent on a chapter in another Aristotelian treatise, viz. the De caelo) see next n. The topics of the chapters discussed in the present paper correspond to those treated at AtuSTOT., meteorolog. I 4-8 plus II 9-III 1 plus III 2-6. The numbering of Aëtian chapters and lemmas will be that of Diels. — It is not possible to include in our discussion the meteorological sections in Theophrastus' Metarsiologica, Epicurus' Epistula ad Pythoclem, Lucretius, Achilles' Περι το~~παντ~ς, Arri an ap. Stobaeum, Arius Didymus, etc., or to deal with the commentaries on the Meteorologica of Alexander (genuine, see P. MoaAux, Alexander von Aphrodisias, Der Aristotelismus bei den G riechen, Bd. 3 («Penipatoi», V11/1), repr. Berlín-New York, De Gruyter, 2001, pp. 264-267), Olympiodorus, and Philoponus, though occasional references will prove unavoidable. 2 On Aristotle as a source of Mtius as to methodology and contents see J. MA1SFELD, Physikaí doxaí and problemata physika from Aristotle to Aëtius (and beyond), in W.W. FORTENsnucH-D. Guτλs (edd.), Theophrastus: His Psychological, Doxographical and Scientific Writings («Rutgers Studies in Cl αssical Humanities», V), New Brunswick-London, Transaction Publ., 1992, pp. 63-111, at pp. 67 ff., and at pp. 94-109 («Aristotle and the Placita on the Earth»). See further below, p. 57, complementary note 2.

— 23 —

JAAP MANSFELD

of content, but also and especially inquire into the function of cited positions, or particular lemmas or parts of lemmas, or even chapters, in the context of the dialectical discussion or doxographícal presentation of a particular topic. 3 As to Aristotle as source of, or important parallel for, the contents of specific lemmas of these chapters I owe much to others (as will be duly acknowledged in the footnotes), but as far as I know the influence of Peripatetic dialectic, that is to say the surveying, characterization and discussion of physical tenets,4 upon the Aëtian presentation of such dogmata has not been taken into account by scholars in this case. From time to time I shall also adduce parallels, both as to function and as to content, from Seneca's Quaestiones naturales and the Scholia in Araturn. To be sure, the sequence of the topics that are at issue is not the same in Aëtius, Seneca and Aristotle's treatise. 6 Some of the differences in this respect between Aëtíus and Aristotle will be analyzed in the final section of this paper. Aristotle. We have a problem here, for the explanation of vision by means of optical rays issuing from the eyes found e.g. in the treatment of the rainbow in the Meteorologica is very much different from the explanation of vision as the effect of the transparent medium upon the eye, found esp. in the De anima (2 7). Alexander of Aphrodísias already points this out, and argues that in the Meteorologica Aristotle uses the doctrine of the "mathematicians".' What is more, the complicated geometrical proof of the circular shape of the rainbow at meteorolog. 3 5 is very similar to the demonstration of a theorem by Apolloníus of Perga a century later. 3 See J. MANSFELD, Doxography and dialectic: the Sitz im Leben of the Placita, in ANRW Π 36.4, 1990, pp. 3076-3229; D.T. Ru ντλ, What is doxography?, in Pη. J. VAN DER EijK (ed.), An-

cient Histories of Medicine: Essays in Medical Doxography and Historiography in Classical Antiquity («Studies in Ancient Medicine», )O), Leiden, B ri ll, 1999, pp. 34-55, and his contribution

to the present volume, pp. 1-21. 4 This description hopefully satisfies the severe conditions stipulated by J. BRUNSCHWIG, Dialectique et philosophie chez Aristote, à nouveau, in N.L. CORDERO (ed.), Ontologie et dialogue. Mélanges en hommage à Pierre Aubenque («Tradition de la pensée classique»), Paris, V~n, 2000, pp. 107-130, esp. p. 130. 5 Detailed comparison with Seneca's treatise is impossible here. See further below, pp. 5758, complementary note 5. ~~ For the orders of presentation in a considerable number of sources see the overview at I.G. KID D, Theophrastus' Meteorology, Aristotle and Posidonius, in W.W. FORTENBAUGH-D. GuTAS (edd.), op. cit., pp. 294-306 (an admirable as well as enviably brief paper, which demonstrates that Posidoníus follows Aristotle rather than Theophrastus), at pp. 305-306. See further below, p. 58, complementary note 6. 7 E.g. ALEXAND. APHRODIS., In meteorolog. 141.3-142.20. See P. MORAu χ, Alexander, cit., pp. 291-294.

— 24 —

FROM MILKY WAY TO HALO

Sir Thomas Heath, who discovered this striking similarity, argued that the occurrence of the Apollonian theorem in Aristotle "shows that it was discovered and similarly proved before Aristotle's time". s W.R. Knorr, on the other hand, argued that the co-presence of this geometrical demonstration, the rigour and quality of which (he intimates) are not paralleled by anything of this kind to be found elsewhere in the Aristotelian corpus, with that of the theory of the optical ray emanating from the eye, shows that this section of the treatise was "written more likely by a disciple of Aristotle than by the philosopher himself ". 9 Fortunately, this issue is irrelevant to the argument of the present paper, which as I have said is concerned with the influence of Meteorologica Books I to III on a set of chapters of the Placita and related literature. For the authors of this later literature, as we shall see, the author of the theory of the rainbow as set out in Book III of the Aristotelian treatise is Aristot1e. 10 Accordingly, in the present paper "Aristotle" will be short for "contents of Meteorologica I to III". To return to the Placita: Hermann Diels already pointed out that several chapters of Book IΠ, viz. placit. 3 6 (on shafts and mock suns) and 3 18 (on the halo), as well as a large part of another chapter, 3 5.2-9 (on the rainbow) do not consist of the usual Aëtían collections of various brief lemmas with name-labels. They are limited to a single doctrine to which no namelabel has been attached. One may moreover add that the discussion of the rainbow at placit. 3 5.2-9 is more extensive than is the rule for the treatment of such topics in Aëtius, though there are a few other similarly lengthy passages elsewhere in the epitome. Diels believed that for these passages in Book III Aëtius did not excerpt Placita literature, but a meteorological handbook ("quoddam de meteoris enchiridíon, quod in opinioníbus minus quam in rebus ipsis explicandis versatum est"). 11 Although 8 Τη. HEATH, Mathematics in Aristotle, Oxford, Oxford University Press, 1948, pp. 181190, quotation p. 181. 9 W.R. Kiomut, The Ancient Tradition of Geometrical Problems, Boston etc., Birkhäuser, 1986, pp. 102-108, quotation p. 108. Odd misunderstanding at T.K. JOHANSEN, Aristotle on the Sense organs, Cambridge, Cambridge University Press, 1998, pp. 48-49. Also cf. below, n. 33 ad fInem. In the corpus Arístotelicum references to meteorol. I III are rare, see H. Boixrz, Index Aristotelicus, Berlin, Reimer, 1870, p. 102b49-60. That no name-label is attached to the Aristotelian account of the rainbow at Placit. 3 5 does not have to be explained on the assumption that the doxographer (or his source), like Knorr, was in doubt about its author; see below, section IV. 10 Cf. below, n. 33 (Arius Didymus fr. 14 Dies), text after n. 86 (schol. in Aratum 940), n. 88 and text thereto (Seneca). 11 H. DIELs, op. cit., pp. 60-61, 178. Diels sees this as fraudulent practice ("quis autem -

-

3

— 25 —

JAAP MANSFELD

chapters 3 6 and 3 18 are extant in ps.Plutarch only, they indeed (as Diels argued) clearly belong with chapter 3 5.2-9, extant in both ps. Plutarch and Stobaeus, so are to be attributed to Aëtius Placita. Diels also demonstrated that placit. 3 18 (on the halo), the final chapter of the Book, is out of place. 12 It follows upon the chapters dealing with terrestrial themes, whereas it should have been put among those dealing with the µετ ~ρσια or phenomena situated between the moon and the earth, which are explicitly said to have already been dealt with in what in Diels' reconstruction is the second lemma of ch. 3 8, but in fact is a transitional phrase analogous to the transitional phrase at placit. 3 5.1. 13 Note that the µετ ~ρσια referred to in this conclusion at placit. 3 8.2 also include "winds" (ch. 3 7) and "summer and winter" (ch. 3 8, title plus first lemma) 14 Diels believed that the wrong position of ch. 18 (in ps. Plutarch only, so a check in Stobaeus is precluded) is to be ascribed to Aëtius, but his argument seems to be merely moral, that is to say concerned with Aëtius' in his view scandalous and sloppy attitude to his material. The chapter is inscriptor tanta est pravítate ut mud caput quo íntroducímur quinto loco collocet?") and calls Aëtius a "market-salesman" ("mango"); see J. MANSFELD-D.T. RuiA, Aëtiana: The Method and Intellectual Context of a Doxographer, vol. 1: The Sources («Philosophic Antiqua», LXXVIlI), Leiden, Brill, 1997, pp. 99-100. — Such an epitome is now available, the Arabic translation of a Greek original: Η. DAIDER (ed.), Ein Kompendium derAristotelischen Meteorologie in der Fassung des Hunain ibn Ishaq («Aristoteles Semítico-Latinus - Prolegomena et Parerga», I = Verh. KNAW Afd. Letterk., N.R., LXXXIX), Amsterdam, North Holland Publ., 1975. Note that this deals with Aristotle only (sequence of topics similar to that of Aristotle's original, though those of meteorolog. 14-8 are appended at the end — perhaps because the author of the epitome held the Milky Way and comets to be astronomical phenomena?). 12 Cf. G. LACHENAUD, op. cit., p. 25. 13 περιγεγραµµ ~νων δε µΟιTIJ)'! µεταρσιων,εφοδενθ~σεται κα'~~τα πΡ~σΥεΙf. Cf. Eusebius' paraphrase (found after his abstract from ps. PLUTARCH., placit. 2 32), Praep. ev. 15 54.3, και περι µ gν

τ~ν σ~ρανιων και iεταρσιων τσσα~τa τσ~; δεδηλωµ ~νσις πρ~ς áλλ~λoσ; &&εστασιαστaι• Θ~α δ~~καì τ~~ Περι γ~ς, and note his (standard) emphasis on the díaphonia among the physicists. 14 I intend to discuss placit. chs. 3 7 and 3 8.1 on another occasion, mainly because this paper is already too long. The reality of winds and of summer and winter, moreover, seems not to have been questioned, so the distinction between appearance and reality which, as we shall see, is of crucial importance for the phenomena to be discussed in the present paper never was an issue in relation to these topics (cf. below, text to n. 117). All the same it is possible to establish a few links between the Placita chapter on winds and meteorolog. 1 13 (349a16 sq.) and 2 4-5, although one of our problems is that ch. 3 7.1-3 is extant only in ps. PLUTARCH, 3 7.4 only in Stobaeus (and is moreover, as Diels already saw, a combination of a lemma from Aëtius and a "frustulum" of Arius Didymus). The subject of Plamit. 3 8.1, viz, the causes of summer and winter, is not a topic in the Meteorologica. We only have an obiter dictum at 361a12-14 (their cause is the sun's movement in the ecliptic; cf. also e.g. Gen. anim. 767a6-7) — still, this remark is found in the first substantial chapter on winds in Aristotle's treatise, which may to some extent help to explain why winds are followed by summer and winter in the Placita.

—26—

FROM MILKY WAY TO HALO

deed in the wrong place, but the simple and obvious explanation will be that it was ps. Plutarch who omitted to transcribe the piece about the halo when copying out his epitome of Αëtius, 15 then wanted to make redress for this omission and put the chapter where he had room, viz, at the end of the scroll. In the short preamble of Book III (extant in ps.Plutarch only) Aëtíus says that after his treatment of "the things in the heavens" he now turns to "the things on high, which are located between the sphere of the moon and the place of the earth". 16 Furthermore, it is only in the first lemma, or rather proem (extant in both Stobaeus and ps. Plutarch) of chapter 5, that is to say after his discussion of things on high like the Milky Way, comets and shooting stars etc., thunder and lightning etc., and clouds, rain, snow etc., that he states that "of the things on high some, like rain and hail, have real existence and others come about through appearance, lacking a real existence of their own; [...] now the rainbow is according to appearance". 17 Diels is amazed, and argues that this fundamental distinction between real existence and emphasis, i.e. appearance, image, should have been formulated before, viz, in the preamble of the Book. 18 We should note, furthermore, that the report on "shafts and mock suns" at placit. 3 6 is introduced in a manner recalling the distinction made in the first lemma of chapter 5, for these phenomena are said to come about through "a blend of real existence and appearance". 19 The first section of Book III of the Placita has a clear formal structure which has to do with appearance and reality, but something appears to be wrong or at least puzzling about the passages where this structure is explicitly mentioned. And an original chapter sequence has been disturbed because the chapter on halos (3 18) is found long after the treatment of the µετ ~ρσια proper has been concluded.

15

See below, n. 52 and text thereto.

16

Περ ωδεΥΚ~ùς ~ν τοις ΠΡΟτ~Ροις ~v ~ΠιτΟµ t TUV περι τ&i OvpaViwV λ~γον, σελ~νη δ ' αΥτ~& τ~~Με θ~ριον, τρ~ψοµαι~ν τ τρ1ΤW πρ~ς ΤÓf Ιεταρσια ' τα~τα δ ' ~στι ΤQ ~Π~~τον κYκλον Tl; σελ~νης καθ~κοντα µ ~χρι πρ~ς τ~ν θ~σιν της γ~ς κτλ. 17 ~ν µεταρσ í ων Παθ~ν ΤU ΜYν καθ ' ~στςσν γ~νεται ο~ον διιρρος χ~λαζα, t' δi κατ ' ~µφασιν ιδ~αν ovκ ~χοντα vn~στασιν• [...] ëστιν ο iν κατ ' ~µφαστν~~1ρις. 18 19

H. DIELS, op. cit., pp. 60-61 (see above, n. 11). See further below, notes 30, 91 and 114 and text thereto.

—

27

—

JAAP MANSFELD

II. EXISTENCE AND REFLECTION (APPEARANCE, IMAGE)

Placit. 3 1.220 lists three tenets. This triad derives from Aristotle's Meteorologica,21 and so does the main distinction between the first and second

tenets on the one hand and the third on the other, as will become clear from the following presentation in parallel columns: placit. 3 1.2 {la} τ~ν Πυθαγορε~ων ο~~ µ ~ν ~φασαν α-

meteor. 1 8, 345α14 -18 + 345b9 - 12 {la} τ~ν µ ~ν ονν καλονµ ~ νων Πνθαγορ-

µ ~ν ε~ων φασí τινες ~δ~ν ε~ναι τα~την οι µ ~ν ~π~~τ~ς ~διας ~δρας δι' σ Si ~π~δραµε τ~ν ~κπεσ~ντων τιν~ς ~στ~ρων, κατ~~~ν ~ νην ~πì Φα~θοντος φθορ~ν, // χωρ~ον κνκλοτερ ~ς α~τ~~καταφλ~ξαντος λεγοµ ~• // {lb} ~. δ~~τ~ν ~λιον το~τον τ~ν κ~κλον ~πì το~~ κατ~~Φα~θοντα ~µπρησµο τ~~φασιν (la + lb) ο~ον {lb} ~. δ~~TUI ~λιακ~ν τα~τη φασιΡ κατ' φ~ ρεσθα~~π~ // {2} τιν~ς δ~~ ον διακεκα~σθαι τ~ν τ~πον το~τον ~~τι ~ρχ~ ς γεγον~ναι δρ~µον κατοπτρικ~ν ε~ναι φαντασιαν το~~ ~λιου τοιο~τον ~λλο πεπονθ~ναι π~θος ~π~~τ~ς τ~ς α~γ~ς πρ~ς τ~ν ο~ραν~ν ~νακλ~ντος φορcις α~τ~ν. [...] // {2} ~τι δ' ~στìν τρ~τη (~περ καιΡ ~π~~τ~ς ïριδος καιΡ ~π~~τ~ν νεφ~ν τις ~π~ληψις περιΡ α~το~ • λ~γονσιν γ~ρ 22 τινες ~ν~κλασιν ε~ναι τ~~γ~λα τ~ς ~µετ ~σνµ.βα ~ VΕL), ρας ~ψεως πρ~ς TUV ~λιον, c~~σπερ και τ~ν ~στ~ ρα τ~ν κοµ ~ την. στ~ρος ε ~ναι δι~κανσrv ~κπεσ~ντος

Pythagorei. fr . 37c DK, reference at Oenopid. fr . 9 DK. Verbal parallels underlined. Cf. W. Buaκεετ, Lore and Science in Ancient Pythagoreanism, Cambridge/Ma., Harvard Univ. Press, 1972, pp. 57, n. 26, 321-322 with notes 115 and 117, who is in the first place concerned with Aristotle as a source for Early Pythagoreanísm. He is followed by H. DAIBER, Aetius Arabus. Die Vorsokratiker in arabischer Uberlieferung, Wiesbaden, Steiner, 1980, p. 418, and by G. LACκενλun, op. ca., ad loc., and p. 261, n. 4. 22 Transl. of ps. Plutarch.: "{la} Some of the Pythagoreans say that [the Milky Way] is the [result of the] burning by a star which moved from its proper place, and the region through which it came, this it burned in a circular way, at the time of the conflagration of Phaethon. // {lb) Others say that originally the sun's orbit followed this route. // {2} Some say that it is an appearance, as in a mirror, of the sun, who reflects back its rays against the heaven, just as what happens with the rainbow on the clouds". For (la) Phaethon and the Milky Way cf. Diodor. Sícul. 5 23.2. For (lb), the former path of the sun, cf. ACHILL., Περ' TOU παντ~ς ch. 24 (Περι το~~ycλαξιοο), p. 55.18-21 Μλλss, where this view is attributed to Oenopides of Chius and others (fr. 10 DK, 2nd text) and an explanation derived from myth not found in either Aristotle or Aërius is added. Aristotle does not mention Oenopides but speaks of Pythagoreans; the passage (la + lb) nevertheless is printed as Oenopid. fr . 10 DK, 1st text (cf. H. Sττtoκν, ad loc., op. cit., p. 146). See also W. BURKEBT, op. cit., pp. 321-322 with n. 117. The parallels between the Achilles chapter and the Aëtian chapter (and to a lesser extent between the Achilles chapter and Aristotle's discussion) are sufficiently close and numerous for a shared tradition to be likely (see below, n. 39 and text thereto, on Democritus; for the parallels with Placit. 3 1.1 see n. 86 and text thereto, text ton. 104, and n. 110), but the differences should not be overlooked: much more space devoted to mythical explanations than in Aëtius (and, naturally, Aristotle). On Achilles see now J. MANSFELD-D.T. Ruνιλ, op. cit., pp. 299305. Good discussion of these passages and of some others, esp. the doxography on the Milky 20 =

21

— 28 —

FROM MILKY WAY TO HALO

Naturally Aëtius' language is to some extent different and more explicit, but much of Aristotle's wording has been preserved, sometimes slightly modified. Diels failed to take into account that the opposition between real existence and appearance which, as we have seen in the previous section, is explicitly stated a bit late in the day, viz, at the beginning of placit. 3 5, is thus implicitly but nonetheless clearly present in the above lemma, as well as — as we shall see — in the first lemma of the next chapter. For us it is also important to realize (once again) that the contrast can be expressed in the form of a diaeresis, and that such diaereses of contrasting doctrines as are found in the Meteorologica (we shall encounter more examples) are the result of Aristotle's study of his predecessors in the field. In other words, though he may have contributed to its formulation, the contrast was not invented by him, but derived from the λεγ~µενα,the "things said". The first two tenets (la + lb) reported at placit. 3 1.2 are listed together (and then discussed) by Aristotle, as being largely comparable; to be sure, there is a minor difference between two purported groups of Pythagoreans (or individual Pythagoreans), but these people obviously agree as to the reality of the Milky Way. After these tenets Aristotle cites a view not paralleled at placit. 3 1.2 but a bit later, at placit. 3 1.5-6. This view is attributed by Aristotle to Anaxagoras-cum-Democritus (see below). Conversely, the second tenet (2) of placit. 3 1.2 is paralleled considerably later in Aristotle's chapter, and listed by him as a third assumption (τρ~τη τις 4m520yn~). According to this view the Milky Way is not real, or substantial, but an optical phenomenon. Aëtius (or perhaps rather the tradition on which he depends) 23 has moved this tenet forward to the first lemma of the chapter which contains a plurality of views not only because its purported original author (Hippocrates of Chius, not named by Aristotle) was later Way at MACROB., In somn. Scipion. 1 15.3-7 (= Posid. fr. 130 E.-K.) at I.G. KIDD, Posidonius: II. The Commentary (i), cit., pp. 487-488 (cf. also below, n. 26); minor slip: attribution to Metrodoru s of view lb (in Aëtius and Aristotle), but Metrodorus as reported at Placit. 3 1.3 (= fr. 70Α13 DK) speaks of the path of the sun, not the former path of the sun. H. DIELS, op. cit., pp. 229-230, attributed the whole Macrobius passage to Posidoníus. I.G. KIDD, ibid., p. 488, entertains this as a "possibility" (cf. below, p. 57, complementary note 2, on Diels' and Kidd's views). Macrobius, loc. cit., also reports as the view of Theophrastus (= fr. 166 FHS&G) that the Milky Way is "the strikingly bright junction by which the heavenly sphere is fastened together from the two hemispheres". It is highly dubious that Theophrastus entertained such a strange notion; see R.W. SHARPLES, op. cit., pp. 108-110. For late testimony (fr. 167 FHS&G) attributing to Theophrastus a theory more like Aristotle's, which makes the Milky Way a meteorological fit a celestial phenomenon, see R.W. SHAaPLES, ibid., pp. 110-111. 23 Cf. below, text to n. 61.

— 29 —

JAAP MANSFELD

considered to be a Pythagorean, but also the better to set out the opposition between reality and optical illusion. 24 We should observe that here the main term in Aristotle concerned with optical illusion is ~ν~κλασις ("reflectiοη", the technical term in e.g. Euclid's Catoptrica), while Aëtius in the parallel passage uses a form of the corresponding verb, viz. ~νακλ~ντος. For the triad of cited views in their original Aristotelian sequence see also his summary of "approximately the only views previously put forward by the others" at meteorolog. 345b28 -30. The last view reported at placit. 3 1.2 is presented in a way which as to doctrine is at variance with Aristotle's presentation. Aristotle mentions visual rays which are reflected towards the sun (think of the explanation of the rainbow in Meteorologica Book III, for which see below, section III). Aëtius speaks of the sun's rays which are reflected against the heaven. 25 What he means to say, presumably, is that the heaven functions as a mirror: what we see when we see the Milky Way is in fact a reflection of the sun, just as happens when we see a rainbow, which according to this lemma is a reflection of the sun's rays by clouds. This explanation of the Milky Way, we should notice, is not that of Posidonius 26 but in fact only a garbled version of the tenet cited and refuted by Aristotle. Reality versus optical illusion. The main idea or issue, whether explicitly stated or implicitly present, is the same in both accounts: existence (Aëtius), or substance (Aristotle) on the one hand, versus reflection, or mere optical phenomenon, on the other. The terms used may vary. Aëtius formally opposes ~Π~σΤασtς to ~ Μφασς, but instead of ~Μφασις ( "appearance") also speaks of Qνακλασις ( "reflection"), or κατοπτρικh φαντασ~α ("mirror phenomenon") 27 without explicitly opposing these terms to vπ~στασις ( "existence"). He also uses forms of the corresponding verbs. The formal opposi24 W. BURKERT, op. cit., p. 323, n. 115, referring to schol. in Arat. 1091 (see below, n. 55 and text thereto) argues that the tenet (anonymous in Aristotle) belongs to Hippocrates of Chius, later recruited for Pythagoreanísm because he was a mathematician, and that this is why it was placed in the lemma. Thus already H. DrELS, op. cit., p. 231. The attribution to Hippocrates of what is in Aristotle is certain (cf. Híppocr. + Aeschyl. fr . A6 DK). Also cf. H. Srlaoηµ, ad lac., op. cit., p. 147. 25 Compare the three Presocratíc views on the rainbow at Plait. 3 5.10-12, about the role of the (reflected) light of the sun, which are contrasted with the anonymously cited view of Aristotle (below, section III) that the rainbow is to be explained by the reflection of the visual rays towards the sun. Cf. below, text to n. 85 and to n. 103. 26 Listed in the same chapter (Placit. 3 1.8 = Posidon. fr . 129 E.-K.), and cited at MACROS., In somn. Scipion. 1 15.7 (Posidon. fr. 130 E.-K), cf. above, n. 22. Posidonius believed the Milky Way to be substantial, not an optical illiusion; see I.G. KmD, Theophrastus' Meteorology, cit., p. 297 (also referring to fr. 129 E.-K). 27 Cf. the title of Placit. 4 14: aερì κατοπτρ κ~~~ν ~iΨ~σεωΝ.

—30—

FROM MILKY WAY TO HALO

tun ~Π~στασΙς versus ~Μφασις is also found in the pseudo-Aristotelian De mundo, 28 and elsewhere.29 A compromise category, viz, a blend of existence and appearance, is – as we have seen – formulated at placit. 3 6, and it is paralleled at schol. in Arat. 811. 30 But Aristotle uses Σvστασις (meaning "colnpound(ed) substance" as well as "compacted substance"), 31 not úπ~στασις, and speaks both of ~Μφασις and – rather more often – of ~ν~κλ ασις (or uses forms of the corresponding verbs). Sometimes the wording is very precise, as at meteorolog. 373b30-3 1, on the rainbow: "the process of reflection will give rise to a sort of appearing" (σται 816 τ ν ~ν~κλασιν ~ µφασιςτις). These polar terms are never formally and explicitly opposed in the Meteorologica the way their analogues are in the De mundo, or the Placita, etc. But the distinction itself is of importance in Book I (being a factor in Aristotle's inquiry into phenomena such as the Milky Way) and in Book II (the reference to Clidemus in the final chapter), and of major importance for the discussion of rainbow and mock suns in Book III of the treatise. 32 In the De mundo too, just as in the Placita, the explicit distinction between reality and appearance is not found at the beginning of the account of meteorological phenomena; here it appears not very far from its end, in ~ των τ~~λ~ν ~στι κατ ' 28 [i Rιsτor.], De mund. 395a28-30, σvλλ~Πδην δε τ~~ν ~ν ~~ ρι φαντασµ ~µφασ ~v, τ~~δε καθ ' ~π~στασν κτλ.; `briefly, the phenomena of the air are divided into those which

are mere appearances and those which are realities", tr. FuRu.EY. 29 Cf. G. LACHENAUD, op. cit., p. 265, n. 4 (but the reference to DuG. LAERT. 7 152 for )π~στασ~ς versus ~Μφασς is mistaken). íξει τ~ς vπΡαστ~σως κα~~εΜφ~~εως, and schol. in 39 AËT., Placit. 3 6 (on shafts and mock suns), µ

~~(aiθερος καt) 71% τ& iYv ~στι κατ' Amt. 811, τ~ν γινοµ ~ νων κατ εν τιδ ιετε~ρ ρ συνισταµενων µεταξ

Lν µεν οτον~ρις, äλως,ι u κτ~~δε παρ~λιοι, καθ ' ~µφ a σν, τ~~δ~~u κτ~, τ~~δε καθ ' )π~στασιν• κατ ' ~µφασ ~πΡ~στaσν δε κοι~ται κτλ. Cf. G. LACHENAUD, op. cit., p. 267, n. 5, and above, text to n. 19, below,

n. 91 and text thereto, and text ton. 114. In Aristotle the "chasms, trenches and blood-red colours" in the sky are de facto such a mixture: the phenomena are in part produced by ~ν~κλασς (Meteorolog. 342b6, 11), "while their σ~στaσις only lasts a short time" (342b13-14). Different account at PHILOPoN., In meteorolog. 69.4 sqq. — For the Scholia in Aratum and other Aratea as the remains of in some cases perhaps even quite early commentaries see J. MANSFELD, Prolegomena: Questions to be Settled Before the Study of an Author, or a Text («Philosophía Antiqua», LXI), Leiden, Brill, 1994, pp. 49-52 and 197-198, and the literature there cited. Information of doxographícal provenance is occasionally found in ancient commentaries. For the Scholia in Aratum in relation to Aëtius see also preliminary remarks in J. MANSFELD-D.T. Ru ντA, op. cit., pp. 305-306. 31 E.g. Aιusrοr., Meteor. 344b19, 377b5. 32 Cf. H. STROHM, op. cit., p. 313, followed by G. REALE-A. Bos, Il trattato `Sul cosmo per Alessandro" attributo ad Aristotele («Temi metafisi ci e problemi del pensiero antíco. Studi e testi», XLIl), Milano, Vita e Pensiero, 1995, pp. 139-140. The argument of P. STEINMETZ, Die Physik des Theophrastos von Eresos (Palingenesia, I), Bad Homburg etc., Gehlen, 1964, pp. 197-204 (followed by H. Sn toui, op. cit., p. 312) that the formal distinction ~i~στασς versus ~µφασιςderives from Theophrastus is misleading. At Τanορmtλsτ., metarsiologic. ch. 14.(2)-(13) DAIEER it is not even implicit; I.G. K IDD, Theophrastus' Meteorology, cit., pp. 296 and 297, expresses doubts. — For Clidemus see below, text to n. 74, and n. 75 and text thereto.

— 31 —

JAAP MANSFELD

ch. 4, and introduces the (brief) treatment of the rainbow etc. This position of the general point therefore is comparable with what we found to be the case ín Aëtius. What is more, both the author of the De mundo and Aëtius in this respect are in full agreement with Aristotle's own presentation. In the Meteorologica the strong emphasis on the thesis that certain meteorological phenomena are mere optical appearances is only found in the second chapter of Book III, in the introduction to the treatment of the halo, the rainbow, the mock suns and the shafts, and repeated disertis verbis at the beginning of the account of the rainbow: "all these phenomena are reflections", "the rainbow is a reflection". 33 Compare Aëtius: the formal contrast between reality and appearance is only stated (to Diels' dismay, as we have seen) at the beginning of ch. 5 of Book Ill, that is to say at the beginning of the account of the rainbow. In the earlier passages of the Meteorologica where reflection, i.e. optical illusion, is at issue (one of which we have seen, while others will be cited below),34 heterodox views are being discussed, the contrast being an ingredient of the report, just as in Aëtius. And to anticipate: the Aëtian chapter on the rainbow itself (placit. 3 5, to be discussed in section llI below), the particular stylistic nature of which was already pointed out by Díels, 35 to a large extent reports Aristotle's doctrine and emphasizes that the rainbow is a phenomenon caused by reflection of the visual rays — another instance of agreement, both functional and as to content, with the Meteorologica. Moreover, in the famous proem of Aristotle's treatise the halo, rainbow, mock suns and shafts are not listed, and the all-important distinction between reality and appearance is not found there either. In this respect, then, there is agreement between Aëtius and Aristotle. Seneca too does not set out the opposition between reality and appearance at the beginning of Book I of the Quaestiones naturales. Although in the course of his substantial treatment of the rainbow etc. he now and then states that such phen meteorolog. 372α17 -18, τ δ ' αl τισν τoßτων ~a~ντων ταvτ~~ π~ντα ~ν~Ιλασις τα~τ ' &mmι; ~ν ~στιν ~ν~κλασις ε'ρηται πρ~τερο~. Cf. ARuus DIDYM. fr. 14 DIELS ap. Sτo373a32, ~~δ ' ιρις ~τι µ BAEUM, 1 30.2 ad irait. ~λω δ Και ~ριδας Και nαρ~λισΙ κα'ι ρ~~3δσσς κα ~~τ~λλα -e( κατ~~TUO ~αφ~σεις ~Π~~ µ ~ν τ~ς αYτ~ς αιτ~ας γ~νεσθαu π~ντα y~ρ ε~ναι τα~τα τ~ς ~ψεως ~ν~κλασιν. Different but similar vocabulary at ALEX. APHROD., in meteorolog. 141.30 sqq., µΥλλωνλ~γειν ΙΤερι τε nαρηλιων κιι ~~ ßδων, ~τι δ~~~λω τε Και lριδΟO, ~~π~ντα δι ' ~Μφ~σε ~ ς τε Και κατοΠτρικwς ~ρ~τaι κτλ. (cf. above, n. 7 and text ,

thereto). On the advanced optics and mathematics involved in this part of the Meteorologica see F. SOLMSEN, Aristotle's System of the Physical World: A Comparison with his Predecessors, Ithaca, Cornell Univ. Press, 1960, pp. 418-420, and above, nn. 8 and 9, and text thereto. On Aristotle on optical meteorological phenomena also cf. I.G. KID D, Theophrastus' Meteorology, cit., p. 296. 34 Above, beginning of this section; below, e.g. text to notes 64 and 75. 3s Above, n. 11 and text thereto.

— 32 —

FROM MILKY WAY TO HALO

nomena are deceptive, the explicit distinction is found much later in the Book, viz. immediately before the peroration, when all the phenomena at issue have been gone thrοugh.36 We should however note that the explanation of the phenomenon called ~Μφασις provided by Aëtius at placit. 3 1.2 is not paralleled in the Meteorologica. The doxographer adapts a commonplace comparison found in more or less contemporary literature, referring to our impression, when aboard a ship, that the land is moving. 37 All the lemmas of placit. 3 1, the third item of the first lemma excepted, present tenets which, however varied, agree in representing the Milky Way as real. The view that it is an appearance is an exceptional and maverick assumption. That (together with two others) it is mentioned at the outset of the chapter is also linked to the fact that the presentation of tenets in this chapter (apart from those not paralleled in Aristotle) 38 is much indebted to Aristotle's selections in the Meteorologica. A bit later (before his τρ~τη τις ~π~ληynς) Aristotle briefly lists the view of Anaxagoras plus Democritus, meteorolog. 345a25-31 (Anaxag. fr. Α8O DK, 1st text; only a reference at Democr. fr. Α91 DK): ~ κρtτον φ~ς ειναι τ~~γ~λα λYyουσιν ~στρων τιν~ν τ~ν of 88 περ~~ 'Αναξαγ~ραν κaì ∆ηµ Yν ßúv περισρ~ται y~ρ ~λισν ~Π~~τ~ν )41/ φερ~µενσνo~χ ~ρñv ~ νια τ~ν ~στρων. ~σα µ ~π ' α~το~, το~των ι~ν oY φα~νεσθαι τ~~φ~ς (κωλ~εσΟaι γàρ ~π~~τ~ν το~~~λιου ~κτινων)• ~σοις δ ' ~ντυυρ ~ττει ~~yi' ~στε µ ~~~ ρñσθaι ~π~~το~~~λιου, τ~~το ~ των οικε~ον φ~ς ειναι Ψαοι τ~~γ~λα.

and Anaxagoras posit that the Milky Way is the light of certain stars, for the sun, in its course beneath the earth, does not see [i.e. shine upon] some of the stars. Those (stars) upon which the sun does shine in the round, of these the light is of course not visible, for it is prevented by the rays of the sun. But those which are screened from the sun by the interposed earth so that it does not shine upon them, the light proper to these, they say, is the Milky Way. Democritus

In Aëtius the two parts of Aristotle's sentence have been divided over Anaxagoras (placit. 3 1.5 = ANAXAG. fr. Α8O DK, 2nd text) and Democritus quaestion. natural. 1 15.6, quoted below, n. 66. ~ν ~~~πε~ρΟς κινε~σθaι δοκεi. Cf. CICER., de diplacit. 3 5.2, avτικα Υο~ν πλε~ντωΝ ~µ vinat. 2 120, nam et navigantibus moven videntur ea quae stamt, LUCRET., 4 386-390, SεxΤ. Εµ paRuC., pyrrhon. hypot. 1 107-108 (ín a Pyrrhonist trope). Similar enumeration of trompe 1'oeíl factors in Seneca's account of the rainbow (!), quaestion. natural. 1 3.9-10.9 (among these, naturally, the crooked oar, remus tenui aqua tegitur et fracti speciem reddit; cf. below, text to n. 81). 38 Metrodorus and Parmenides (placit. 3 1.4-5 DIELS) do not occur in the Meteorologica. 36 37

Áß:T.,

Post-Aristotelian persons obviously do not either.

— 33 —

JAAP MANSFELD

(placit. 3 1.6 = DEMOCR. fr. Α91 DK, 2nd text; in Stobaeus the second name-label has dropped out, and the word-order is slightly different). 39 Interestingly enough, the cause of the phenomenon assumed by both physicists according to Aristotle is no longer an issue in the Α tian Democritus lemma, though what remains of the modified or corrected doxa is not contrary to this explanation, placit. 3 1.5-6 (frr. ANAXAG. Α80 DK, 2nd text + DEMOCR. Α91 DK, 2nd text): ~ ρος iσταοθαι τov ο~ρανοv, (5) 'Αναξαγ~ρας ~ν σκ~~ν τißς γ~ς ΚαΤU TiUTO τ~~ µ ~~π~ντα φωτ~ζη. (6) ∆ηµ ~Ν ~ταν úτ~~~ν y~ν ~~~λισς γιν~ΜεΝoς µ ~ κριτος πολλ~ν Καì µικρ καì αυ'εχ~ν ä.στ~ρων συµφωτιζοµ ~νων öλλ~λοις συναυγaσµ ~ ν διà τ~ν ΠvκΝωσΙΝ. (5) Anaxagoras (holds) that the shadow of the earth rests upon this section of the heaven [viz, where the Milky Way is visible] when the sun, having arrived under the earth, no longer illuminates everything. (6) Democritus (holds it is) the combined radiation of numerous and small and contiguous stars giving off light together, because 0f their density.

One assumes that Aëtius, just like Aristotle, presents these tenets as pertaining to something different from mere optical illusions. 40 What is more, one of the sources that were intermediate between Aristotle and Aëtius saw fit to correct Aristotle's account to some extent. This may have been Theophrastus, whom we knοω to have corrected Aristotle on points of detail. 41 But we do not knοω in the present case. The brief lemma with name-label Aristotle (placit. 3 1.7) obviously goes back to the Meteorologica too, viz, to the long exposition at the end of the chapter, 345b31 - 346b10; 42 Aristotle believed the Milky Way to have real 39 For Aristotle as the source see G. LACHENAUD, op. cit., ad loc. and p. 262, n. 7. H. DIELS, op. cit., p. 229, comparing the Placita lemma and the Aristotelina passage, submits that "Aristoteles Anaxagoram et Democritum minus accurate coniunxit"; cf. H. Srxoma, ad loc., cit., pp. 146-

147. Further cf. below, text to n. 70. Because in Aëtius the wording has been altered I have not underlined verbal similarities. The view of Democritus is paralleled at ACmLL., ΙΙερi τov Παντ~ς ch. 24 (cf. above, n. 22): p. 55.24-27 MAnss = DEMOCR. fr. Α91 DK, 3rd text cf. already H. Duns, lic. cit. 4U H. DIELS, op. cit., p. 138, in the synoptic overview of "excerpts de Anaxagora" in his famous "theophrasteorum spud excerptores conspectus", quotes two parallels, viz. HIPPOLYT., refutat. omn. haeres. 1 8.8 (cf. ANAXAG. fr. Α42 DK) and DuoG. LAERT., 2 9 (cf. ANAXAG. fr . Al DK), slightly different from each other, where the Milky Way is said to be a "reflection" ( ~ν~κλααις) of the stars which are not illuminated by the sun. But these references provide only partial parallels, for the Placita lemma on Anaxagoras does not speak of reflection. There is no overview of "excerpts de Empedocle" in Diels' conspectus, but this is by the way. 41 See J. MANSFELD, A ristote et la structure du De sensibus de Théophraste, «Phronesis», 41 (1996), pp. 158 - 188. 42 Inaccurate reference in G. LACHENAUD, op. cit., ad loc. We may further note that in Sto-

- 34 —

FROM MILKY WAY TO HALO

substance. The definition of the Milky Way in the first lemma of the Aëtian chapter (placit. 3 1.1) 43 too states it to be real, not illusory. The emphasis of the chapter accordingly is upon the reality of this phenomenon. To some extent this helps to understand why the distinction between existence and optical illusion is only explicated at the beginning of the chapter (placit. 3 5) which deals with the first phenomenon which should be explained by reflection, viz, the rainbow. We may now turn to the following chapter, placit. 3 2, on comets, shooting stars and "beams" i.e. a kind of meteor (instead of ps.Plutarch's "beams" Stobaeus has "the like"). It lacks a definition at the beginning, which is understandable because a plurality of phenomena are collected. Otherwise the structure of the first part of ch. 2 (placit. 3 2.1-3) is virtually analogous to that of the first section of ch. 1: a díaphonía of Pythagorean (?) views to begin with, then Anaxagoras and Democritus, then Aristotle. These three Dielsian lemmas, all about comets only, are derived from the 44 Meteorologica: for placit. 3 2.1 (not in DK), the two contrasting "Pythagorean" views, cf. meteorolog. 342b329-35 plus 342b35-343a4; for 2.2 (ANAXAG. fr. Α81 DK, second text; not in the Democritus ch. of DK), Anaxagoras plus Democritus cf. the first view cited by Aristotle at 342b27-9; for 2.3, Aristotle himself, cf. the next chapter of the Meteorologica, 344a5-345a10. Just as in the previous chapter, so here too the Aristotle lemma in Stobaeus has been conflated with an abstract from Arius Didymus 45 (irrelevant to the inquiry into the sources of Aëtius, but of value for the study of Stobaeus). A noteworthy difference with the previous Aëtian chapter is that in Aristotle Anaxagoras plus Democritus are placed before the Pythagoreans and "Hippocrates of Chius and his pupil Aeschylus", whereas in Aëtius the sequence is the other way round. Clearly Aëtius, or rather the tradition he baeus the second sentence of the Aëtian lemma has been replaced by an abstract from Arius Didymus, as Diels already suspected (though he printed it as part of the Stobaean lemma of his reconstructed Aëtius!); see D.T. R υκτA, Additional fragments of Arius Didymus on physics, in K.A. ALGRA, P.W. νλν DmR HORST and D.T. RUNIA (edd.), Polyhistor. Studies in the History and Historiography of Ancient Philosophy («Philosophic Antiqua», L3X1), Leiden, Brill, 1966, pp. 363-381, at p. 374 and p. 380 (new fragment 16), and J. MANSFELD-D.T. RuilA, op. pit., p. 252. 43 On this definition see further below, section N. 44 For Placit. 3 2.1 see W. Buaκr tr, op. cit., pρ. 57, n. 26, 321, n. 111, 322, n. 115, followed by H. DAMER, Aetius, cit., p. 419; for 3 2.1-3 G. Lλcι mνλUD, op. cit., ad loc. and pp. 128, n. 4, 262 notes 5 and 6. 4s Cf. above, n. 42, and see D.T. Ruνιλ, Additional fragments, cit., pp. 374-375 and p. 380 (new fragment 17); J. MANSFELD-D.T. RUNIA, op. cit., pp. 252-253.

— 35 —

JAAP MANSFELD

depends on wanted to establish a fairly strict correspondence between the opening sections of these two chapters, so Aristotle's Pythagoreans and Hippocrates plus Aeschylus swapped places with his Anaxagoras and Democritus – who this time share a lemma, whereas in the previous chapter they had been separated. The first views cited in both these Placita chapters are "Pythagorean", as we have seen. The contrast in the first lemma of the Aëtian chapter (placit. 3 2.1) is again one between reality, or existence, and reflection, or optical illusion: "some of the Pythagoreans" believe that the comet is a star which now is visible and now is not, `others" that it is a "reflection of our vision towards the sun, similar to images which appear in mirrors" (placit. 3 2.1, 2nd tenet: ~ν~κλασιν της ηµετ~ ρα ~ψεως πρòς T~V ηλιoν παραπλησιαν Τα~ς κατοπτρικα~ς ~µφ ~ σεmmν). Aristotle attributes to Hippocrates and Aeschylus the view that the comet is a star, but that "the tail does not belong to the comet itself, which acquires it when in its passage through its space it sometimes draws up moisture which reflects our vision towards the sun" (meteorolog. 342b35-343a4; HIPPOCR. + AEsCHYL. fr. 5 DK). This explanation is analogous to that of the anonymous tenet about the Milky Way cited meteorolog. 345b9-12.47 It is surely correct to argue that Hippocrates of Chius, because he is a mathematician, has been incorporated to serve a Pythagoreanizing tradition, but what is equally important is to note that the contrast has been set out in exactly the same way as in the parallel lemma of the previous Aëtian chapter. And note that Hippocrates' view has been modified the better to express the díaphonia: Aristotle states that only the tail of the comet is the result of a reflection of our vision (i.e., visual rays) towards the sun, while the Aëtian lemma posits that this holds for the comet as a whole! Even so, this tenet about the comet has been preserved in a form which conforms a bit better than its sister tenet (about the Milky Way as a reflection) to its Aristotelian original. The tenet of Anaxagoras and Democritus is reported to be that the comet is "a conjunction of two or more stars as to combined illumination" (placit. 3 2.2) or that "cornets are a combined appearance ( σúµφα mιΡν) of planets, when they seem to touch each other because of their closeness" (meteorolog. 342b28-29). It would seem that Aristotle believes this to imply that according to Anaxagoras and Democritus cornets are appearances. For the rare word σ~λφcσσις – limited, it seems, to this passage in Aristotle and , 46

46 47

Cf. below, text to n. 61. Above, n. 24 and text thereto.

—36—

FROM MILKY WAY TO HALO

to the commentators on the Meteorologica – obviously recalls the term Μφασ~ς; it is explained by Alexander ad loc. as "the impression which comes about from all the (stars) that come together, as if it came from a single one".48 The Α tian lemma is less clear in this respect. We shall see, however, that Seneca is explicit about the illusory nature of comets according to Anaxagoras and Dem οcrítus. 49 The sequel of placit. 3 2 is a mixed bag. The first three `Aristotelian' lemmas, as we have noticed, are about comets only. So is the next lemma (placit. 3 2.4), reporting a tenet from Strato, perhaps also 2.6-7, Ερigenes,50 and Boethus (fr. 9, S. V.F. III, p. 267), certainly 2.8, Diogenes. 51 But at 2.5 Heraclides of Pontus (fr. 116 WEHEua) is cited first for his view on comets, and then for his similar explanation of other phenomena: "bearded star, halo, 52 beam, pillar and what is related to these"; this explanation is said to be shared by "all the Peripatetics". At 2.9-10 the tenets of Anaxagoras (fr. Α82 DK) and Metrodorus (fr. Α14 DK) on shooting stars or "sparks" are reported: so Anaxagoras occurs for the second time in one and the same Aëtian chapter, which (combined with the topic, which differs from the lemmas on comets) may be a sign that chapters which originally were separate have been conflated, though one cannot be certain. The Heraclides-cumPeripatetics lemma (2.5), as we saw, already adds "beams" etc. The final lemma (2.11), name-label Xenophanes (fr. Α44 DK), reports that he considered "all these (phenomena) to be concentrations or movements of fiery clouds", so this includes both comets and "beams" and shooting stars, etc. What favours the suggestion that at an earlier stage two doxographical chapters were involved is the fact that we do have two separate sections in the Meteorologica: chs. 1 4 plus 1 5 on meteors etc., chs. 1 6 plus 1 7 on 48 ΑLex. APHIRoD., in meteorolog. 26.16-7 (cf. DEMOCR. fr . Α92 DK, 1st text): σ~ λ' ασww yòρ ~ νην. Cf. PHILOPON., in meteorolog. λ~γει τ~Ν ~E π~ντων Τ Ν αυνελΟ~ΝΤΩΝ ~4 ~~ ~νò4 φα1Τα α1 γενσµ ~Νον φα~νεται. Olympio~αν ~ ργ~ζονται σοµφ a σιν, τοντεατιΝ €ν φ~< εi π~ντων σvµΠεφορηµ 75.29-30, µ don's' explanation (echoed in LSJ and the English translations I have seen) is different but I believe not good: OLYMPIOD., in meterol. 49.19-20, ΣΥΜφααιν ... ~ντì τov σΥΝοδον. Compare Seneca's ~ Ν, said of rendering of the tenet, quoted below, n. 64, and Air., Placit. 3 1.6, the term αννανγααµ the stars that according to the Democritus lemma combine to form the Milky Way (above, text quoted before n. 40). H. STROHM, op. cit., p. 17 translates "Gesamtb il d". a9 Below, n. 63. In Seneca this tenet is anonymous (quidam). 50 Of Byzantium; no lemma in vol. 3 of the Dictionnaire des philosophes antiques. 51 Perhaps uncertain whether of Apo ll onia or of Seleucia, though the tenet has been claimed for the former (DI0GE1. fr . A15 DK = Τ30 LAKs). 52 The word äλω has fallen out in Stobaeus. The fact that it is mentioned in this lemma supports our suggestion that ch. 18 (on the halo) was originally omitted by ps. Plutarch, see above, text to n. 15.

— 37 —

JAAP MANSFELD

comets. 53 Now in meteorolog. 1 4-5, in contrast to 1 6-7, Aristotle omits to cite earlier views. Accordingly even the tenets about meteors etc. in placit. 3 2 with the name-labels of physicists Aristotle was familiar with and may cite elsewhere, do not derive from his account of meteors etc. So this Aëtían chapter, the first section of which, as we have noticed, ultimately derives from the Meteorologica, acquired its tail of further tenets (2.4-11) as it passed through the centuries – and so did the previous chapter, placit. 3 1 as to lemmas 3 1.1, 1.4-5 and 1.8. Whether or not the Presocratic lemmas among these acquisitions ultimately derive from Theophrastus it is impossible to decide. We have no independent information concerning the Physikai doxai on these themes, and the extant abstracts from his Metarsiologica do not contain chapters on the Milky Way and meteors etc. 54 (and even the extant parts do not name physicists). That Theophrastus is involved is not, of course, unlikely, but cannot be proved. To understand the tradition a little better we should now consider the important parallel for the paragraphs on comets of AËτ., placit. 3 2 at schol. in Arat. 1091,55 on comets (only). This passage is in some respects closer to Aristotle's exposition than Aëtius' chapter. Unlike AËτ., placit. 3 2 it moreover includes a paragraph on Posidonius. This reference presumably provides a t.p.q. for the scholiast's account as a whole. It also shows that this scholium depends on an uberior fins similar to that (or to one of the fontes uberiores) on which Aëtius depends. 56 That such an intermediate source for meteorology, or several such sources, did exist is of course demonstrated by the rich doxographíes in Seneca's Quaestiones naturales.57 s3 The chapter divisions in our editions do not have Aristotelian authority, but the original and intentional division into two sections is clear: see the announcements at meteorolog. 341b1 sqq., 342b25 sqq. and 345a11 sqq. (the beginning of the section on the Milky Way). 54 See below, p. 58, complementary note 6. 55 pp . 545.8-546.25 Mnnss — pp. 508.8-510.19 Mimi = Posmoi. fr. 131a E.-K. (who omits to print the paragraph on Hippocrates [not in DK either] — a mistake repeated in the comparison of the scholium with Aristotle's text at I.G. K ID D, Posidonius: Commentary, cit., pp. 491492); largely paralleled in a passage in a Parisian ms. (Paris. gr. 2424, on which see J. MAx τιν (ed.), Scholia in Aratum vetera, Stuttgart, Teubner, 1974, p. xxix) printed as Posmoi. fr. 131b E.-K. (— pp. 511.13-512.12 MARTIN). Only a reference at ANAXAG. fr . Α81 DK. 56 H. Dials, op. cit., pp. 230-231, argues that, though Aristotle is the ultimate source, "proxímus est Posidonius" (cf. below, p. 57, complementary note 2). That an intermediate source (or sources) ís/are involved is dear. But if this scholium goes back to Posidonius, schol. in Arat. 940 (see below, n. 86 and text thereto) should also go back to him. 57 Preliminary suggestions: F.P. WAIBLINGER, Seneca's Naturales Quaestiones: Griechische Wissenschaft and römische Form («Zetemata», LXX), Miinchen, Beds, 1977, pp. 23-29; J. MANSFELD, Sources, in A.A. LONG (ed.), The Cambridge Companion to Early Greek Philosophy, Cambridge etc., Cambridge Univ. Press, 1999, pp. 22-44, esp. pp. 31-32.

—38—

FROM MILKY WAY TO HALO

We may cite two symptoms of the scholium's closeness to Aristotle's text, viz. (1) the precise reference to meteorolog. 343b11-12 (επ'~~yο ~ν τ~~ ~σχιφ τον Κυν~ς 'Αριστοτ~λης φΗσ~~πaρατετηρηκ~Ναι κο ~την), and (2) the identification of the author of the tenet that the tail (and the tail alone! just as ín Aristotle) of the comet is the result of reflection as posited by "Hippocrates the Pythagorean" (sic). But the fact that Hippocrates of Chius is designated a Pythagorean, this time disertis verbis, shows that the tenet has not been abstracted directly from the Meteorologica. 58 Closely parallel to AËr., placit. 3 2.2, the scholium, in sequence, lists the Pythagoreans, Democritus and Anaxagoras, Aristotle – and then Posidonius, lacking, as we have seen, in Aëtius; and it ends with the tenet of Hippocrates which, as a Pythagorean doctrine, is to be found at the beginning in Aëtius. Presumably the sequence from one doxographical account to another was, first, as in the (source of the) scholium, the identification of Hippocrates as a Pythagorean, then, at a second stage, the inclusion of his tenet in a lemma on the Pythagoreans in general. Furthermore the scholium is very explicit about the diaphonia (or even diaphoniai) that is (are) at issue: Hippocrates said that there is only one comet (the others listed in this text speak of comets in the plural), 59 "and while all the others state that they have their own tails he said that it is the result of reflection" ( κα~~ of µεν λΟΙπΟ '~~ ~δ~ας α~τ~~ν τ~~~Κ~~α~~ ~πΟφaιΝΟΝΤα~, 8 δε Κατ~~~ν~κλασιν). And all the tenets cited are criticized, except that of Posidonius: our scholium, or rather its immediate source, is indebted to the tradition of dialectical discussion deriving from the Early Peripatos – a tradition of which but few traces are extant in our Aëtius. b0 The scholium has a díaphoníc structure; and Posidonius himself is included. This would seem to render derivation of this doxographical passage from Posidonius unnecessary. Finally, the fact that the scholium and its parallel in the Parisinus are about comets and not about other phenomena provides some further support for one's impression that our present chapter AËT., placit. 3 2 is a conflation of two different chapters in an earlier source: one on comets, the other on me–

Cf. above, n. 24. According to Amusrοr., meteorolog. 342b29-343a4 both the Pythagoreans and Hippocrates plus Aeschylus (fr. AS DK) hold that there is only one comet. At Α τ., Placit. 3 2.1 too the «Pythagoreans» (i.e. inclusive of the view of Hippocrates and Aeschylus) hold that there is only one comet. The information provided by the scholium is different: another instance of the vagaries of the doxographic traditions. 60 See J. MANsFELD, Physikai doxai, cit., pp. 109-111, pp. 356-358 of the Italian version. In this respect the scholium compares well with the dialectical discussions in Seneca's Quaestiones 58 59

naturales.

—39—

JAAP MANSFELD

teors etc. 61 The alternative, viz, the supposition that the information on comets in the Aratus scholium was carefully abstracted from a doxographical chapter which treated meteors as well as comets, is rather less likely. In Seneca's Quaestiones naturales, which belongs with this tradition, an entire Book is dedicated to comets alone, viz, the present Book VI Ι. In the original Book order of the treatise, Book VII came before the present Book I, 62 which deals with meteors etc. in chs. 1 and 14-15.4 (and in chs. 2-13 with halos, rainbows, shafts and mock suns, which are not at issue in placit. 3 2 but are in 3 18, 3 5 and 3 6). Seneca, just like Aëtius and the source of the doxographical scholia on Aratus, is aware of the important distinction between those who believe that comets have real existence and those who believe that they are optical appearances. 63 But in Seneca's Book on comets neither Hippocrates nor the Pythagoreans are mentioned. For the view that comets are illusory phenomena he cites anonymous authorities at quaestion. natural. 7 12.1 64 (from Aristotle and Aëtius we know these to be Anaxagoras and Democritus), and Zeno (of Citium), ibid. 7 19.1.65 These people are said to hold that comets actually are the combined light of stars in conjunction. Moreover he distinguishes between two views which presuppose that comets are real: as to contents these correspond, repectively, to those of the first group of Pythagoreans and to that of that of Aristotle as reported at AËT., placit. 3 2. la and 3.2.3. As to the distinction between reality and appearance in general, Seneca states near the end of the next Book (our Book I) that no one doubts that meteors etc. are real, whereas it is a matter of discussion whether or not the rainbow and the halo are real or not. His own opinion 61

Cf.

above, text to n. 23. For suggestions about coalesced chapters in Aëtius see J. MAN-

Cosmic distances: Aëtius 2.31 Diels and some related texts, «Phronesis», XLV (2000), pp. 175 204, at pp. 177, 198 199, and J. MA1SFELD, Aëtius, Aristotle and others, cit., pp. 289 290. 62 SENEC., quaestion. natural. 1 15.4, cometas ..., de quibus dictum est. For the original Book order see H.M. HINE (ed.), L. Annaei Senecae Naturalium quaestionum libri, Stuttgart and LeipSFELD,

-

-

-

zig, Teubner, 1996, pp. xxn-xv, also for references to the literature. For Seneca on earlier theories concerning comets cf. P. OLTR vIM{E, op. cit., p. 298 and p. 344, n. 2 to p. 320 (references to AËτ., Placit. 3 2.1 and — a typo — Αmsτοr., Meteorolog. 2 6.2: should be 1 6, 342b30 sqq.); on the present passage P. Ρλιmονι, op. cit., p. 597 ad toc. 63 SENEC., Quaest. nat. 7 19.1 2, ... quidam nullos esse cometas, sed speciem illorum [...]. quidam aiunt esse quidem [...]. Cf. D. Vorrmto, op. cit., p. 700 ad toc., and P. PARRONι, op. cit., p. 600 ad toc. For species cf. below, notes 66 ad finem, and 75. -

64 cum ex stellis errantibus altera se alteri applicuit, confuso in unum duarum 'urine facies longions sideris redditur. Cf. above, n. 48 and text thereto, and P. OLTRAMARE, op. cit., p. 312, n. 3; C. CODONER MERINO, op. cit., vol. 2, p. 142, n. 1, D. VorrERO, op. cit., p. 686 ad lic. 65 Zenon ... congruere iudicat steltas et radios inter se committere: bac socetate luminis existere imaginem stellae longions (= Zaio fr. 122, SVF I, p. 35). Cf. D. VOTTERO, op. cit., p. 686 ad loc.

—40—

FROM MILKY WAY TO HALO

is that rainbow and halo are deceptive optical phenomena, 66 which (though with some exaggeration) corresponds to the point of view of Aristotle, one also found in the Placita. Note however that he says here that no one (!) doubts that comets are real (in the sense that they "show real fire", i.e. have their own light), 67 while a moment ago we have seen that in the previous Book, at 7 12.1 and 19.1, he quoted the views of Anaxagoras, Democritus and the Stoic Zeno that they are appearances (to the extent that they do not have their own light, but are the combination of the lights of a plurality of stars). 68 The long next chapter of Aëtius (placit. 3 3, fifteen lemmas, on thunder, lightning etc.) 69 differs from the two previous chapters in that it fails to report a dissident opinion according to which lightning would be unreal, viz, a phenomenon of reflection. As we shall see Aristotle reports such a view, and this view is also cited by Seneca. On the other hand, such lemmatic echoes of the Meteorologica as it does contain are conspicuously analogous to those found in the two previous chapters: placit. 3 3.4 is on Anaxagoras (fr. Α84 DK, 2nd texct), 3.7 on Empedocles (fr. Α63 DK, 2nd text; Stobaeus only), 3.14 on Aristotle. 70 Empedocles and Anaxagoras (frr. EMPED. Α63, 1 st text, + ANAXAG. Α84, 1 st text) are cited together at Αmsτο'., meteorolog. 369b11-19: κα~τοι τιν~ς λ~yουσιν ~~ν τοi; ν~φεαιν ~γγ~γνεΤcι ßv13• τovτο δ ' 'Ε πεδοκλ~ς Μ Yv ~~αιν ~ιναι 'rd ~ΜπεΡιλαΜβαΝ~ΜεΝον τ~ν τov ~λιο~~äκτινων, 'Αναξαγ~ρας 66 SENEC., quaestion. natural. 1 15.6, de his [shooting stars etc.] nemo dubitat quirt habeant flammam quam ostendunt [H. CoRcoltAN, Seneca: Naturales Quaestiones, vol. 2 (LCL, 450), London etc., Harvard Univer. Press, 1971, p. 81, n. 3 refers to ARIsTOT., meteorolog. 344b15-19]: certa illis substantia est. de prioribus quaeritur (de arc," dico et coronis) deczpiant aciem et mendacio content. an in illis quoque uerum sit quod apparet. nobis non placet in arco aut corona subesse ali. quid corporis certi, sed illam iudicamus speculi esse fallaciam alienum corpus nil amplius quam mentientis. non est enim in speculo quod ostenditur, etc. See I.G. Kmn, Posidonius: Commentary, cit., p. 469. Cf. quaestion. natural. 156.3, tu dicis illum colorer [of the rainbow] esse, ego dr ; 1 6.2, speciem falsi colo ris. For species cf. above, n. 63; below, n. 75. Also cf. D. VOTTERO, op. cit., p. 273 ad loc. 67 nemo dubitat: see previous note.

68 Above, notes 64 and 65, and text thereto.

The title (περì βροντ~ν αστραπ~ν κεραυν~ν πρηστ~ρων 'νε καì [τε καì om. Stobaeus] recalls Aristotle's enumeration in the prier of the Meteorologica, at 339a3 - 4: περì κερ aυνιν πτ~σεως καì τυφ~νων καì πρηστ~ρων — a sequence which does not correspond to the order of treatment in the body of the treatise, but this is by the way. 7° In ps. Plutarch the first part of the final lemma (on Strato) has been omitted (as we see from Stobaeus) and the remainder been coalesced with the Aristotle lemma: a characteristic accident. For the Aristotelian origin of Placit. 3 3.4 and 3.7 see G. LACHENAUD, op. cit., ad loc. and pp. 130, n. 3, 264, n. 5. For the differences between Aristotle's, Posidonius' and Theophrastus' views of thunder and lightning see I.G. KIDD, Theοphrastus Meteorology, cit., pp. 300 - 301. 69

τuΨ~νων)

4

—41—

JAAP MANSFELD

58 το~~~νωθεν αιθ~ρος, 8 δ~~~κε~νσς κaλε~~π ~ρ κατενεχΟ8ν ~νωθεν κ~τω. τ~ν ΜYΝ Ο Ν δι~λαµψιν~ στραπ~Ν ε~ναι 1-00 ΤΟYΤΟΥ ΤΟ~~πuρ~ς, τ~ν δ~~Ψ~Ψον 81==o0βενννµ ~νον και τ~ν σιξιν βροντ~ν. Nevertheless some people say that fire comes to be in the clouds. Empedocles says this is the part of the sun's rays enclosed inside, Anaxagoras that it is a part of the upper `aether' (which as we know is his name for `fire') that is enclosed inside, carried down from above to below. Lightning, he says, is this fire flashing through the clouds, thunder its noise and hissing when quenched.

The Α tian lemma on Anaxagoras describes not only lightning and thunder, but also whirlwinds and firewinds. We may see this as an addition to the original lemma, and need only compare the first section with what is in Aristotle. The Empedodes lemma on the other hand has hardly been expanded: it is about thunder, lightning, and thunderbolt. So here are Mr., placit. 3 3.4 and 3.7 (more or less literal verbal similarities with Aristotle underlined): 'Αναξαγ~ ρας, ~ταν τ~~Θερ ~ν ε~ς τ~~ψvχρ~ν ~µΠ ~σ Ι, το~το 5 ' ~στ~ν α θ~ριον µ ~ ρος εiς ~ερ~δες, Α~~~ µ ~ν ω~φε τ~ν βρσντ~ν ~Ποτελεï'c! 58 ΠαρU τ~ν µελανιαν'rob' νεφ~δο~ς χρ~µατι ~ν τ ~στραπ~ν, Τ(ιΡ 58 πλ~ θεt καì µεγ ~Οει του φωτ~ς τ~v κ~ρανν Ν κτλ. 'Εµπεδοκλ ~ ς ~ΜΠτωσΙν φωτ~ς dλς ν~φος ~ξειργοντος τ~ν ~νθεστ~-τa ~~ρa, ο i τ~ν j.iYv 71 σο~σtν καì τ~ν θραvσιν κτ~» ον ~περy~ζεσοat, τ~ν δ~~λáµψΡιν~στραπ~ν• κερανν~ν 58 τ~ν τ~ ς ~στραx~ς Τ~νοΙ. Anaxaxagoras holds that when the hot falls into the cold, i.e. a part of aether into a part of air, this produces thunder through the noise, and through the colour set off against the blackness of the cloud the lightning, and through the mass and size of the light the thunderbolt, etc. Empedocles speaks of the impact of light in a cloud, light which forces out the resisting air; the quenching and dissolution of this light bring about noise, and the shining lightning; thunderbolt, he says, is the intensity of the lightning.

The tenets of Anaxagoras and Empedodes (frr. ANAXAG. Α84 DK, 2nd text + EMPED. Α63 DK, 2nd text) as cited here are a bit more unlike each other here than in Aristotle, 72 but this is a matter of variatio rather than content. The similarity with the passage from the Meteorologica is clear.

71 At EMPEDOCL. fr. Α63 DK Diels preferred the conjecture σχ~σιν. Kranz and others stuck to the reading of the mss. The Aristotelian origin of the lemma proves that they are right. 72 Cf. Placit. 3 1.5-6, above, text to n. 39. They have also swapped places: Aëtius three times cites Anaxagoras first. For the possibility that Theophrastus is to some extent responsible cf. above, n. 41 and thext thereto.

— 42 —

FROM MILKY WAY TO HALO

The lemma on Aristotle, naturally, briefly reports Aristotle's doctrine as found in the Meteorologica, viz, in the final chapter of Book 1I. 73 What is remarkable, as noted above, is that the dissident view of Clidemus is not mentioned. I suppose that it has dropped out in Α tius, or perhaps even earlier in the history of the tradition to which he belongs, for in a related line, represented by Seneca, Aristotle's reference to Clidemus, inclusive of its illustration, has been repeated. 74 See meteorolog. 370a10-13 (CLIDEM. fr. 1 DK): There are some, e.g. Clidemus, who say that lightning does not exist but is an appearance (οüκ ειναι φασιν äλλà φα~νεσθαι). They compare it to one's being similarly affected when one strikes the sea with a stick; for the water seems to flash at night.

Compare 1 DK):

SENEC., quaestion. natural. 2 55.4

(only a reference at CLIDEM. fr .

Clidemus says that lightning is an empty appearance, not fire; for in this way flashing is caused at night by the movement of the oars. 75

Aristotle charitably adds that this mistaken doctrine is due to the fact that, at the time, people were not yet acquainted with "the [mathematical] doctrines about reflection" (370α16-17, τα~. Περι τ~~ ~νακλáσεως δ~ξαις). 76 I skip placit. 3.4, on clouds etc., where we find little or nothing which recalls Aristotle's account; 77 even 3 4.2, the Anaxagoras lemma (fr. Α85 DK), shows hardly any traces of meteorolog. 348a15-21 plus 348b12-14. This chapter as a whole appears to be on the same level as the sections 73 H. DIELS, op. cit., p. 225, quoting AËT., Placit. 3 3.4 and SENEC., quaestion. natural. 2 12.3 (on Anaxagoras) in parallel columns, refers to the account of Aristotle's view which follows in Seneca, and states: "omnia haec [viz., in Seneca] ex Aristotelis meteorolog. Π 9 excerpts sum", viz, via, as he believes, (a pupil of) Posidoníus (cf. below, p. 57, complementary note 2). Also see P. OLTRAMARE, op. cit., p. 66, n. 5, D. VorraRO, op. cit., p. 307 ad loc., P. ΡARRονι, op. cit., pp. 524-525 and p. 509 ad loc. 74 Cf. P. OLTRAMABE, op. cit., p. 101, n. 1. 75 speciem inanem esse, non ignem — for the oars cf. ibid., 15.6, for species above, notes 63, and 66 ad hnem. In Aristotle the oars are cited for another optical phenomenon, viz, a rainbowlike one, meteorolog. 374a29-32: "the rainbow produced by oars breaking water" etc., ~~δ ' ~Π~~ τ~ν κωπ~ν τ~ν αναφεροµ ~νων ~κ τ~ς Θαλ~tτης ρ ς κτλ. (also cf. 374b5-6). Cf., in Seneca's account of the rainbow, quaestion. natural. 1 5.6, non et aqua rupta fistula sparsa et remo excussa habere

quiddam simile his quos videras in arca coloribus solet7 76 Cf. F. SOLMSEN, op. cit., p. 419. 77

For possible traces of Theophrastus see H. DAIsER, The Meteorology of Theophrastus in Guτλs, op. cit., pp. 166-293, at pp. 275-

Syriac and Arabic translation, in W.W. FORTENSAUGH-D.

277.

— 43 —

JAAP MANSFELD

of the three preceding chapters which have been added to sections ultimately deriving from Aristotle. III. FROM RAINBOW τ0 HALO

We should, briefly, look at the chapter on the rainbow, placit. 3 5. We have seen above that it is introduced at 3 5.1 by the explicit distinction between reality and appearance, and have insisted on the antecedents of this distinction in the Meteorologica. 78 At 3 5.2 the account (preserved in both sources) has been adorned by means of references to Plato, to Homer (one line quoted) and to a mythological story in a way similar to the adornment of placit. 1 6 and 1 7.1-10 (ps.Plutarch only). The pièce de résistance is the detailed account of the rainbow at ch. 5.3-9 (a few words near the end missing in Stobaeus) which for the most part consists of a series of abstracts, more or less modified and differently arranged but still quite faithful, from the relevant section of the Meteorologica.79 It begins with the formula π~ς ovν γ~νεται lρις; (1ρις om. Stobaeus), "what is the origin of the rainbow?", so what we have here is a question that will be resolved. 80 Next (5.3-6a) we are informed in a simplistic didactic way that there are three ways of seeing, viz. via (1) straight; (2) bent (i.e. refracted: the oar under water which seems crooked); and (3) reflected lines, as in mirrors ( τ~~~νακλ~µενα~ς τ~~κατοiτρικ~). 81 Here the abstracts begin: the rainbow is such a phenomenon of reflection (cf. meteorolog. 373a32),82 involving a cloud (cf. 373b20) consisting of tiny water-drops (Οαν~δaς, cf. 374a10, 373b20). The rainbow will be opposite the sun (cf. 373b21-23, the sun will be opposite the rainbow). Of crucial importance for our comparison between Placita and Meteorologica is the statement that our visual rays are reflected by these little drops ( τ~τε ~~δ' ις πρoσπεσovσα ταiς pανιΩσιν ~νακλ~ται) 83 — not, therefore, as other physicists argued, the 78 Cf. e.g. above, notes 17 and 32 and text thereto. 79 See H. DAmsER, Kompendium, cit., pp. 90-91, and H. DAIBER, Aetius, cit., pp. 423-424; G. Lλcanνλtm, op. cit., pp. 265-266, notes 2-5, 133, n. 1. 80 The formula Π~ς γ~νεται vel sim. is rather frequent in Action chapter-headings in ps. Plutarch, much less frequent in lemmas; occasional Stobaean parallels. ~ 1 Aιusτoτ., meteorolog. 373 a32-373b32 also begins his account of the rainbow with an explanation of the phenomenon of reflection. For the bent oar cf. above, n. 37. 82 See above, n. 33 and text thereto. 83 Cf. meteorolog. 372b32-33, ~τι iYv oiν ~ν~κλασις ~~ιρις ~ ς δ'Νεωι πaòc T~v ~λι~ν ~στι, Ψανερ~ν. See further I.G. K ID D, Theophrastus' Meteorology, cit., p. 296: Posidoníus too believed the

—44—

FROM MILKY WAY TO HALO

light of the sun. These raindrops are a bit inelegantly stated to be "not shape(s) of form, but of colour", ε~σ'ι δ ' αi pανιδες o~~σχ~µατος µορφ ~, öλλà χρ~µατος(cf. meteorolog. 373b19, 37232-34). The colour scheme is slightly different from Aristotle's, but the three basic colours are the same as those listed meteorolog. 371b33-372a10 and 374b28-33, viz, red, green and blue, and so are several others. 84 There follow two proofs (5.9, g στιν oúß το~το δοκιµ ~ σαι & ' ρyων) that a reflection from little water-drops is involved, both ultimately deriving from Aristotle. If opposite the sun one spits out water, the water-drops will reflect (scíl., our visual rays) towards the sun and a rainbow will be produced (cf. meteorolog. 374a35-374b5). And people with an eye disease will have this experience when looking at a lamp (cf. 374a19-23). The three lemmas on the rainbow that follow, placit. 3 5.10-12 (ps.Plutarch only!), with name-labels Anaximenes (fr. Α18 DK, 1st text), Anaxagoras (fr. Α86 DK) and Metrodorus (fr. Α17 DK, 1st text) respectively, do not go back to the Meteorologica. 85 But we should note that here the rainbow is explained each time not as a reflection of our visual rays, but as a reflection of the light of the sun, or as the effect of the light of the sun on a cloud. See, for instance, the descriptive formula àν~κλασtν ämò ν~φους ΠΥκνο ~~τ~O ~λmaκ~ς περtφεΥγειας in the Anaxagoras lemma. Accordingly, when we look at the Aëtian chapter as a whole, we find that the contrast is quite clear: Aristotle's view that the rainbow is due to a reflection of our visual rays is the opposite of these Presocratíc views which claim that it is the light of the sun that is involved. See further below, section N. Like the Placita chapter on comets, the present one too may rewardingly be compared with a scholium, or perhaps rather purported set of scholia, on Aratus. 86 Here we have three lemmas: Anaxímenes (fr. Α18 DK, 2nd text), rainbow to be a phenomenon of reflection, but not by little drops but by the whole cloud, see frr. 15 E.-K. (reference to Posidonius' Μετεωρολογικ~~ap. DuG. LAERT. 7 152) and 134 E. -K (SENEc., quaestion. natural. 1 3.14-5.1 + 5.10-14). 84 For the colours of the rainbow according to Aristotle see e.g. P. S τRuvcκΕν, Colour mixtures according to Democritus and Plato, «Mnemosyne», 56 (2003), p. 292. 85 G. LACIENAUD, op. cit., p. 267, n. 3, should not have written that the view reported in the Anaxagoras lemma "est d'autant plus proche de l'explication a ri stotélicienne qu'elle traite en même temps d'un autre phénomène optique, les parhé lies" etc. That a variety of mock suns are included does not entail that the explanation itself is similar to Aristotle's. DXELS -KRANZ, ad Metrod. fr. Α17, 1 st text, hypothesize "nach Theophr.". For Anaxagoras cf. fr. B 19 DK, 'TΙρww δ καλ~οµεΝ Τ~~~ν τtοιν νεΨ~λ σιν ~ντιλ~µπ o ν τo5 ~λíω κτλ. Xenophanes said the rainbow is a cloud, fr. B 32 DK. 86 schol. in Arat. 940, pp. 515.27-516.19 Μλλss, pp. 455.10-456.6 Mimi (better text). Already adduced by H. DIELS, op. cit., pp. 231-232, as an "[e]gregium specimen Posidianae eru-

- 45 —

JAAP MANSFELD

Metrodorus (fr. Α17 DK second text), and Aristotle, and we encounter the same implicit but clear diaphonía as in the Placita chapter. According to the report about Anaximenes the light of the sun is involved ( ~νικα äν ~ταπ~αωσww αi τov ~λιoυ α~ya~). This light is also found in the report on Metrodorus (~Μ απτο~)σΗς τ~ς α~y~ς). But Aristotle is said to have explained the rainbow as a "mirror image" ( κaτoπτρ~κ~ν ~ποι~σατο ~µφασιν),for our "visual rays jump when reflected and refracted by smooth surfaces" like "air and water" ,

(π~φνκε Υ~ρ [το~το] äσσειν τ~ν ινιν προαπ ~ΠΤοΥσα1 τoiς λειoις σ~ λασtΝ äνακλωµ ~νην καì, κατακλω.~νην τοιo~τον 88 εivaι T~v äYρα κα~~ τd ~íδωρ).

We again come across the water-drops of the cloud opposite the sun reflecting the optical rays towards ít. 87 But no account of the colours of the rainbow this time, and no experiential proofs. We are dealing with the abstract of an abstract, once again illustrating the way doxographical presentations may be handled, and to what uses they may eventually be put. Seneca's discussion of the rainbow, quaestion. natural. 1 3-8, is quite substantial; a number of different views are cited, elucidated, criticized. Here I only wish to recall that Seneca, too, attributes the optical rays which in the Meteorologica explain how we come to see a rainbow to Aristotle: "Aristotle is of the same opinion [viz., that reflection is involved when a rainbow is visible]. Our vision, he says, bends its rays back from every smooth surface", 88 etc. Placit. ch. 3 6, entitled "On shafts", is about both shafts and mock suns, this time the variety called ~νθ~λιοι (solar image opposite the sun) not π~ρ~λ~ο~~ (which appear on either side of the sun) as in Aristotle and at placit. 3 5.11 •89 As noticed above, it lacks the set of contrasting views which are the common fare of Placita chapters, and only gives us a single theory, without a name-label. 90 We have also seen that these phenomena are here explained as a blend of reality and appearance, 91 or image. The ditionis" (cf. below, pp. 57-58, complementary note 2) — a view presumably no longer defended (text not in E.-K.), though still found in the Fragmente der Vorsokratiker, ad AiAXIMEi. fr. 18 DK («aus Poseidoníos») and METROD. fr. Α17 DK («Theophr. durch Poseid.»). ß7 Ibid., lemma on Aristotle: kk ~v Τοινυν äντικρνς το~~~ λιου ν~φος συστ~~[σννεστwς] κατ~~pανιδα, ~γγ~νεσθαl τ~ν ó0ιν καθ ' ~κ~στην pavíδα, και τ~ν ~Πι τòν ~λισν ~ν~κλaσιν ïσχειν. 88 quaestion. natural. 1 3.7: A ristoteles idem iudicat. ab Omni, in quit. levitate Aries radios suis replicat, etc. Cf. P. OLTBAIARE, op. cit., p. 21, n. 2, p. 22 notes 3 and 4, and P. PARRoiI, op. cit., ad loc., pp. 35-36, and p. 489. 89 But the difference may be minimal, cf. schol. in Arat. 881 (the fourth scholium on this line, p. 431.6 MARTIN), τ~~iτap~λww ~νθ~λιa α~ν κaλεïτaι. 90 Cf. above, n. 11 and text thereto. 91 Above, text ton. 19, and n. 30 and text thereto; below, text to n. 114. I.G. KIDD, Theo-

-46—

FROM MILKY WAY TO HALO

combined treatment of mock suns and shafts goes back to Aristotle, see

meteorolog. 377a29-378a1 Ι ; both are said to be caused by reflection and

to be purely optical appearances. There are some differences: clouds are involved in the reflection of shafts, dense air or "mist" ( ~χλvς, 377b19) in that of mock suns. The condition of this mist is such that it is "nearly water" (Yyyv' δ ' i~δατος, 377b2O). It is not such a big difference when Aëtius speaks of "clouds" only. But the difference of perspective is more substantial. Aristotle sees these clouds and mists as a kind of mirrors, not as ingredients of the phenomena. We do not know what was Theophrastus' view of mock suns and shafts. 92 Posidonius' view has been transmitted: it is different from Aristotle's and Aëtius'. According to Posidonius at schol. in Arat. 881 (viz., the first scholium on this line) the cloud is illuminated by the sun (οv y~ρ ιδιq φωτì κ~χρηται, ~λλ~~τι τον ~λιOν), 93 so it is not the case that it is our vision which is reflected, as in Aristotle. Ad finem the scholium rather too briefly cites this view of Aristotle: '~ριστοτ~λης 88 τ~~παρ~λi~~φησ~ν ~ΜφασΙν εΤΝα ~, µ ~~~χοντα úiτóστασιν (p. 430.11-12 Martin). We may hypothesize that in the doxographical source of the scholium Aristotle's view that our visual rays are reflected was opposed to Posidonius' that the rays of the sun are reflected: díaphonia again. ly tentative conclusion is that the Aëtian lemma concerned with mock suns represents an interpretation of what we find in Aristotle, which may or may not be the theory of someone else. The remarkable parallelism of the Aëtian account of mock suns as a mixture of reality and appearance with what we find in another scholium on Aratus, 94 and with this scholium only, is a further indication of the doxographical ancestry of some of these meteorological scholia on Aratus. We may conclude with placit. ch. 3 18, the chapter on the halo which as we have noticed has been transmitted in a place where it does not belong.95 The halo is explained as being caused by the refraction, i.e. a sort of reflection, of our visual rays towards the moon or another heavenly body phrastus' Meteorology, cit., p. 297 wonders whether this formula arose from "a Theophrastean

multiple choice method". I think not. 92 Cf. previous note. 93 schol. in Amt. 881 = Posmoi fr. 121 E.-K. For the interpretation see I.G. Km°, Posidonius: Commentary, cit., pp. 467-470. At I.G. KIDn, Theορhrastus Meteorology, cit., p. 297, he is less dear about the difference between Aristotle and Posidonius. 94 For mock suns (only) as a blend see schol. in Arat. 811 as quoted more fully above, n. 30. 9s See above, n. 11 and text thereto.

— 47 —

JAAP MANSFELD

(της ~ψεως κατακλωµ ~νης). This is a version of Aristotle's explanation of the halo as a purely optical phenomenon, meteorolog. 372b12-373a31, for Theophrastus' explanation, fortunately extant, is different. 96 Aristotle speaks of the reflection of our visual rays ( ~ν~κλασις τ~ ς ~ψεως), while Theophrastus speaks of "the rays of the moon" that come up against thick air around the moon, metarsiol. ch. 14 Daíber (presumably this will also apply to the halos around other heavenly bodies). On the other hand the Aëtian formula át) παχùς κα~~~µιχλ ~δης is a shade closer to Theophrastus' "when the air becomes thick and is filled with vapour" than to Aristotle's wording (e.g. 372b31, πúκνωαιν Υδατ~δη), though the difference ad sententiam is small. What precisely is Posidonius' view is not clear; he is said to follow Aristotle as to ~ν~κλαmmς, but no further details are províded. 97

IV. CONCLUSIONS, QUERIES AND ALTERNATIVES

So chapters one to six plus eighteen of Book llI of Aëtius' Placita are to a remarkable extent related to passages and presentations in the Meteorologica. The contrast between (mostly atmospherical) meteorological phenomena which are real and those which are appearances, which is of great importance in Aristotle's treatise, crucially determines the treatment of these phenomena in this section of the doxographícal tract as well. We have moreover seen that the ps. Aristotelian treatise De mundo and several meteorological scholia on Aratus also belong with the tradition which relies on this distinction, and that Seneca too is familiar with it. Though in individual cases modifications, both as to wording and to specific contents, have unsurprisingly occurred, both the doctrines and the wording of lemmas with the name labels Pythagoreans, Anaxagoras, Democritus (or Anaxagoras-cum-Democritus), Empedocles and ... Aristotle in chapters 1, 2, and 3 of Book III of the Placita are ultimately derived from the Meteorologica. What is more, the function of the tenets of these Presocratics in Aëtius is analogous to the function of their antecedents in Aristotle, viz, to stress the difference between realities and appearances. The dependence, to this extent, of the doxographical tract on the great treatise therefore goes much further, and deeper, than a listing of (partial) -

96 See H. DAmER, Aetius, cit., p. 453, and Theophrastus' Meteorology, cit., p. 280; reference to Aristotle also at G. LACHENAUD, op. cit., p. 142, n. 1. 97 See Posmoi. fr. 133 E.-K., with the comments at I.G. KmD, Posidonius: Commentary, cit., pp. 497-498.

— 48 —

FROM MILKY WAY TO HALO

verbal resemblances and similarities of meteorological doctrine would suggest. We have noticed that parallel passages in the scholia on Aratus, and parallel passages (indeed to some extent whole Books) in Seneca's Quaestiones naturales exhibit a similar diaphonic structure, and are in the final analysis equally indebted to the Meteorologica for their dialectical methodology, and for part of their doctrinal contents. One may surmise that both Seneca and these scholia are inter alia dependent on one or more doxographies, or fontes uberiores, that may be characterized as cousins of the more immediate doxographical sources of Aëtius (which are not entirely hypothetical: think for instance of Varro). 98 We have seen that Aëtius' slowness, 99 too, in so to speak putting the all-important contrast between reality and appearance on the agenda (viz, only at placit. 3 5.1, the first lemma of the chapter on the rainbow) — a contrast valid for these meteorological chapters as a whole, may be explained by an appeal to Aristotelian precedent. Aristotle too only emphasizes "appearance" thematically as late as chs. 2 and 3 of Book III of the Meteorologica, where his account of the rainbow and similar phenomena begins. And so, near the conclusion of his compressed account of meteorology (end of ch. 4), does the author of the De mundo. Before arriving at his explicit formulation of the opposition in ch. 51tius, like Aristotle, in the course of his presentation of the issues quotes the divergent views of those who believe that the Milky Way (3 1.2) and comets (3 2.1) are not real but apparent. Aristotle also mentions the dissident view of the little-known physicist Clidemus that lightning is an optical illusion (at the end of Book II). This tenet however is not (or rather no longer) to be found in Aëtius, but it survives in Seneca, where it plays exactly the same part as in Aristotle: precious testimony to the tenacity of the tradition in at least one of its lines. This so to speak temporized unfolding of the distinction between reality and appearance in both the Placita and the Meteorologica may well be one of the most exciting ingredients of our evidence that this section of Aëtius, to a large extent, is ultimately dependent on Aristotle. A Peripatetic source, or rather tradition, is certainly involved here. In the course of time, then, other lemmas were added to the original collection, and the opposition between appearance and reality was formulated in a much more dominant and scholastic way than is the case in Aristotle's treatise. .

98 See already H. DIELs, op. cit., pp. 196-200. Or think of Achilles, see J. MnNvs εεLnD.T. Ruνιn, op. cit., pp. 302-305. 99 Which scandalized Diels, see again above, n. 11.

—49—

JAAP MANSFELD

Such closeness to Aristotle's treatise is also clear from the contents of 3 5.3-9, 3 18 and (to a lesser extent) 3 6: no name-labels; Aristotelian doctrines derived from the Meteorologica. As we have seen Diels believed that these doctrines were excerpted by Aëtius not from a work of (a follower of) Posidonius but from some meteorological handbook or other, which itself for these chapters would naturally be ultimately dependent on Aristotle. 100 To the extent that an Aristotelian derivation of other lemmas of placit. Book ΠΙ argued in the present paper was taken into account by him, he explained this as being mediated by the influence of (such a follower of) Posidοnius. 101 Diels therefore believed that chs. 3 5.3 -9, 3 18 and 3 6 arrived at their present position by a different route from the "genuina Placitorum materia' with its "opíniorum diversitas". He may quite well be right as to the differences between the traditions involved (see below), but we need an explanation which goes further than the belief that identifying, or hypothesizing about, a source is practically sufficient for understanding what is going on. What on earth, we must ask, was Aëtius' motive for interpolating Aristotelian doctrines without name-labels? For Diels' suggestion that he did so because he is a fraud is hardly helpful as an analysis of the σκοπ~ς, or in-

placit.

tentio auctoris.

The anonymous Aristotelians of chs. 5 and 18 have two important aspects in common. Firstly, the phenomena which are described all belong to the class of appearances. Secondly, and as to backdrop, the chapters in Book III of the Meteorologica which discuss this class of phenomena only provide Aristotle's own doctrines and do not cite other views. We may begin by thinking about the reason for the absence of the name-lable Aristotle in these two chapters. Though one cannot be certain, it looks as if Aëtius here wishes to present us, exceptionally, with what he believes to be true or, at the very least, preferable doctrines, not (as he does most of the time) with conflicting statements only. True doctrines do not have to be identified: the addition of name-labels in an Aëtian context presumably would demote them to the status of views which irresolvably conflict with other views. And such exceptional doctrines need not be Aristotelian. One may, for instance, compare the quite substantial chapter placit. 1 4, 102 entitled Π~ ς σuν~στηκεν ~~κ~σµος:no name-label; Atomistic doc100 Above, n. 11 and text thereto. 101 Cf. above, notes 22, 55, 73 and 86; below, p. 57, complementary note 2. 102 H. DIELS, op. cit., p. 58, posited that this chapter, devoted to a subject which according

—50—

FROM MILKY WAY TO HALO

trine; no other views. If this is correct, we should perhaps see the three Presocratic tenets — not deriving, as we have noticed, from the Meteorologica and to our initial surprise extant only in ps.Plutarch, not in Stobaeus — whích are appended to the `Aristotelian account of the rainbow, all three of which posit that it is the light of the sun that is reflected,' 03 as varieties of a less plausible alternative to the `Aristotelian' claim that it is the visual ray which is reflected. To be sure, we have seen above that the main contrast of placit. 3 5 as well as the name-labels (except Anaxagoras') are paralleled in a doxographical scholium on Aratus. 104 But now notice the differences: the scholium, closely resembling a standard Placita chapter, actually sports the name-label "Aristotle"; the tenet with this label is very brief; and it is the last. In comparison the prominent position and unusual length of the `Aristotelian' doctrine of the rainbow in the Aëtian chapter look like deliberate modifications. A more trivial explanation for the absence of name-labels in these two chapters remains possible as well, viz. that Aëtius abstracted his Aristoteleα from an epitome of the Meteorologica which listed Aristotle's name at the beginning only, not with every single item of doctrine. But this still entails that he will have known that the items he incorporated are Aristotelian. Furthermore, this alternative suggestion fails to explain why the (Aristotelian!) account of the rainbow in placit. 3 5 is found where it is, and is as long as it is. One cannot deny, or so I believe, that some amount of preference is involved. We should also look at the first lemma of the first chapter, which provides a definition of the Milky Way not attributed to anyone by name, placit. 3 1.1: κvκλος ~σ~ì [sc. ~~γαλαξ~ας] νεφελοειδ~ς ~ν.~ν τ ~~ρt δt~~παντ~ς φα~ν~µενος, &&~~5è τ~v λενκ~χροιαν γαλαξ~ας ~νοµ a ζ~~νος.

(the Milky Way) is a cloud-like circular band (1) in the air which is visible over its entire length, and (2) it is called Milky Way because of its white colouring.

The substantive λενκ~χροιa ("white skin colour") is a hapax: it only occurs in the sources for Aëtius and in the abstract from ps. Plutarch in the Histo him should belong in Book Π and so is out of place, is not Mëtian because the doctrine is Atomistic: "mnest enim unius Epicuri ex Democríto conformata doctrina". Not a good argument for refusing attribution to Idus. 103 See above, n. 25, and n. 85 and text thereto. 1°4 n. 86 and text thereto.

—51 —

JAAP MANSFELD

toria philosophy attributed to Galen, each time in the definition of the Milky Way. This definition is one of the very few definitions to be found in the Placita beyond Book I, and the only one beyond this Book which is placed right at the beginning of a chapter. In Book I there are no less than six of them, all at the beginning of chapters. 105 Diels has noticed and commented on the definitions in Book I, but overlooked that in Book III ch. 1.106 One can hardly deny that by descriptively defining the Milky Way without attributing this definition to any authority Aëtius states a position which he seems to accept. It is moreover clear that this definition is not his own invention. To some extent it can be paralleled from a weflknown elementary introduction to astronomy containing numerous definitions, viz. Gemínus, element. 5.68:

λοξ~ς 81 εστι κ~κλος ΚαιΡ ö τoï~~ γ~λακτος. [...] συν~στηκε δ~~~Ις ~~ ραχσΜεριας νεφελσειδο~ς κτλ.

The band of the Milky Way is also slanting. [...] It consists of cloud-like small partides.

But for Gemínus and the astronomical traditions he represents the Milky Way is not a phenomenon in the atmosphere. There are also several partial parallels in the Aratea, which seem to be a bit closer. See the following remark from the chapter "On the Milky Way" in Achilles, which really comes quite close to Aëtius because it makes the Milky Way a phenomenon in the atmosphere (it is found after the tenet of Democritus has been cited as being that of `others"), 107 p. 55.28 - 29 MAASS: 105

ps. PLUTARCH., Placit. 1 9-12 and I 14-15 (9 Περ~ ßλης, 10 Περι ~δ~aς, 11 Περ'ι α~τιων, 12 14 Περι σχηµ~ των, 15 Περι χρωµ~ των) all begin with a general definition. These

Περι σωµ ~ των,

chapters form a block interrupted only by ch. 1 13, which lacks a definition at the beginning. Virtually the same definitions are to be found at the beginning of Sτoa., anthol. 1 11 Περι vλης, 1 12 Περι ιδε~ν, 1 13 Περι αιτι~ν, 1 14 Περι σωµ ~ των κτλ., in third position at 1 15 Περι σχηµ ~ των, and again at the beginning of 1 16 Περι χρωµ ~των. — At Placit. 4 19.1 (in both Stobaeus and ps. Plutarch, slight differences in wording) two definitions of φων~, the first καταχρηστικ~ς, the second κυριως, are added by Aëtius or his source to the tenet with name-label Plato with which the chapter begins. 106 H. DiaLs, op. cit., p. 60, speaks of the "disserendí exilitas" typical of Aëtius' proems, which is also to be found "ín brevissímis et scholastíco more conceptís definitionibus c. 9. 10. 11. 12. 14. 15". He fails to speculate about a source. As to the definition at 3 1.1, at p. 60 he states "genuina sum r 1-4", i.e. belong to what he calls the "genuina Placitorum materia" with its ` opiníonum varíetas". There is an opaque note in the apparatus to λ~γεται κτλ. at Placit. 4 19.1 (cf. previous n.) in H. DIELS, ibid., p. 408: "non lam ad Platiner pertinent". On Plac. 4 19.1 see my paper `Illuminating what is Thought'. A Middle Platonist placítum on Voice' in Context, forthcoming in «Mnemosyne», LVIIT (2005). 107 Cf. above, n. 22.

— 52 —

FROM MILKY WAY TO HALO

[...] µηπΟτε~ντοι µ ~Μεwwον αinòΝ λ~γειν ~κ νεφ~~ν 4' πíληΜ~~τt ~~ρος δrnυγ~ς εινα κ~κλου σχ~ µα ~χον.

[...] unless it is better to say that it [sc., the Milky Way] is from clouds, or a transparent compression of air, having the shape of a circular band. Compare schol. in Arai. 462.4 MARTIN, πλ~ν Τov γαλcξíoυ, &r ~στ~~νεφελ~δης ("except the Milky Way, because it is cloud-like"); 469.14-16, ~νος ... ~ψει ~στ~ν εYΚατ~ληπτoς δια τ~ν ~ν ωιτ~~~σπερ νεφελ~δη oúτoς yUp µ πíλησιν ("for this (circle) alone is easily grasped by sight because of its cloud-like contraction"); 253.5-9, about cloud-like formations due to the vicinity of the Milky Way ( νεφελοειδε~ς ... σνστρο q α~~κα~~κονιορτ~δεις, ~τε τov yαλαξιου κ~κλου γειτνι~ντος). 10ß The key terms, in Aëtius and

the Aratea, are "cloud(s)", and "cloud-like". The inventor of the doctrine reflected in Aëtius' definition cannot be identified, so no name-label is available. Posidonius' theory is dífferent. 109 Aristotle's doctrine is also different to the extent that he places the Milky Way in the i)i~κκαυµα,the fiery outermost sphere of inflammable material (the dry exhalation being the material that bursts into flame), not just in the air. But insofar as this ~π~κκανµαis the outermost and potentially as well as actually hottest layer of the air, the Milky Way may be said to be "in the air" nevertheless. Aëtius' definition looks like a typical doxographical construct, viz, in the present case the simplified version of an Aristotelian doctrine which ad litteram (not always ad sententiam) can be paralleled from contemporary and later astronomical and quasi-astronomical literature. A similarly simplified version of Aristotle's complicated exposition served as an argument against the authenticity of Book I of the Meteorologica. It is cited by Olympiodorus; see in meteorolog. 5.16-17, π~λιν 0001 ν~θον Τò παρ~ν βιβλισν, δι~τι TUV Υαλα ~αν πáθος φησ~~ε~ναι τοv ~~ ρος ("another argument that the present book is not genuine is because of his [sc. Aristotle's] affirmation that the Milky Way is a condition of the air"). 110 108 Also cf. schol. in Lucian. veras histor., 16.7-8, γaλσξιου ' ~~κvκλοι ~~~ν τι ο ραν 3 ~Π~~Νεφελ~ν ~ς γ~λα φα~ν~~~ενος, and COSMAS INDICOPLEUST. 1 7.2-3, 'p; y~ρ ëν τq~~ο~ραν νεφελοειδεις ~κατε γαλαξ~αν κτλ. σvστροφ~ς, (1i.% ~νοµα τεθε 109 See I.G. KmD, Posidonius: Π. The Commentary (i), cit., p. 490 (on ÁËT., Placit. 3 1.1).

110 The formula π~θος ... τov ~~ρος (often used in reference to sound, e.g. [ÁRIsrοτ.], problemat. 912b13) to indicate Aristotle's location of the Milky Way is found a number of times in the commentaries of Olympiodorus and Phíloponus on the Meteorologica. These late comment-

ators argue against Aristotle, whereas Alexander of Aphrodísias follows the master. See OLYMPIOD., in meteorolog. 74.17-76.5 (explicitly reporting the argumentation of Ammonius), with inter alia a reference to Ptolemy's argument that the Milky Way cannot be located below the moon, and PmLoPON., in meteorolog. 113.34 - 118.26. See further H. STROHM, op. cit., pp. 147 - 148.

— 53 —

JAAP MANSFELD

Aëtius' apparent acceptance of a doctrine resembling that of Aristotle, which entails that the cloud-like Milky Way located in the atmosphere is not an appearance but real, and the fact that this definition is placed at the very beginning of the first chapter of the meteorology in his Book 1I, further helps to understand that he followed the tradition in putting the distinction between reality and appearance on the map only subsequently. What exceptionally, then, we would have in the Placita chapter on the rainbow is a case of Aëtius taking sides in a dispute, though he does so in a truly and remarkably modest and subdued way. This suggestion, I believe, will also hold for placit. 3 18 (about the halo: `Aristotelian'; no name-label; no alternative dοctrines). 111 Note moreover that in the relevant sections of the Meteorologica Aristotle himself only gives his own view of halos, not those of others. And I think one may also apply it to placit. 14 (cosmogony). There is nothing fraudulent about a combination of Atomistic cosmogony and Aristotelian rainbow-and-halo-theory. Such combinations used to be called eclectic (and they are tolerated, even welcomed, by Epicurus). Now Aristotle seems to have been quite prominent and rather exceptional in claiming and arguing that rainbows and halos are optical phenomena, appearances. In the fields of meteorology Posidonius followed Aristotle rather than Theophrastus, ~ ~ 2 though e.g. his theories of the rainbow and halo are different from Aristotle's as to important details (in the case of the rainbow it we hear that it is the light of the sun that is reflected!). 113 Aëtius (or his source here), who is later than Posidonius (a philosopher who, as we knοω, is often cited in the Placita), apparently wanted to go back beyond Posidonius to Aristotle's theories for the sections of his meteorological exposition dealing with appearances. Is this because for him and others – such as Posidonius and, to some extent, Seneca – Aristotle had again become an important authority in this field? Seneca regularly, and critically, discusses Aristotelian doctrines which are derived from the Meteorologica, attributing them to Aristotle, whereas, as we have seen, Mtius does not always do this. It is on the other hand obvious that Seneca did not consult the original treatise either, but used derivative literature. Diels' suggestion that the Placita chapters on (Aristotelian) meteorological appearances came down by a different route, and that the íntermedi111 Even if such other lemmas have been lost (like the alternative doctrines about the rainbow, lost in Stobaeus only but preserved in ps. Plutarch — but we do not kn οω) the situation as to this chapter need not be different from the one I have suggested in relation to Placit. 3 5. 112 I.G. KIDD, Τheορhrastus Meteorology, cit. 113 Above, n. 83 and text thereto, n. 93 and text thereto.

— 54 —

FROM MILKY WAY TO HALO

ate source could have been a meteorological handbook, may well be on the right track. One does not know. The source may for instance equally well have been a Book of a lost meteorological treatise much resembling a Book of Seneca's Quaestiones naturales. An entirely different but equally uncertain alternative is to assume that the sections, both with name-labels and without, that can be derived from the Meteorologica are the (naturally `revised') descendants of an original epitome of parts of this treatise, viz. chapters 4 to 8 of Book I, ch. 9 of Book II (combined with ch. 1 of Book III), plus chapters 2 to 6 of Book III of the Aristotelian treatise. The rest, viz, the definition in 3 1.1, the explicit scholastic formulations of the distinction between reality and appearance, and those contrasting views which were not cited by Aristotle were added later – presumably not all at the same time. However this may be, the last chapter to be discussed in this paper, viz. placit. 3 6, should not be explained along the lines attempted for chs. 5 and 18. Its account of shafts and mock suns without a name-label is de facto quite close to Aristotle's, but (the formula for) the blending of real existence and appearance constitutes an important difference. In the Meteorologica shafts and mock suns belong with the genus of appearances, and do not form a separate species. The view that shafts and mock suns are blends of real existence and appearance ( µ íξεt τ~ς imοστ~σως κcaì ~Μφ~σεως) - an idea also occurring (and also without name-label) in a scholium on Aratus 1 1a – represents a compromise position. A compromise position is quite often found at the end of an Aëtian chapter, after the presentation of the conflicting opinions. 115 In the present case such a compromise is an ingredient not of the micro-structure of a chapter, or of a pair of chapters, but of the macro-structure of an important section of a Book of the Placita. The absence of a name-label can be explained on the assumption that this dialectically convenient doxa, clearly a current one, was nameless right from the start. In other words, it may have been thought up for the purposes it serves in both Aëtius and the Aratus scholium (σχολικ~~ς οικε πλ~ττεσθαi, to use an expression of Sextus Empiricus). 6 Above, n. 30. See e.g. J. MANSFELD, Doxography and dialectic, cit., pp. 3080 and 3083 sqq., and J. MANSFELD, A ristote et la structure, cit., p. 184, n. 98. 116 Smx .. Eipimc., aduers. mathem. 7 11-13, discusses three conflicting views, the first of which is attributed to the Stoics, the second to Epicurus and Strato, while the third remains anonymous. It is of this third view that he says that "it appears to be a scholastic construct" 114 115

— 55 —

JAAP MANSFELD

What, further, is clear is that main classes of phenomena (viz., real ones and apparent ones) have been put next to each other 117 which, as we have noticed, were discussed in Aristotle's treatise at a considerable distance from each other, viz, real phenomena in Books I 4 - 8 and II 8 - I1 1, appearances in Book III 2 - 6. Á~T., placit. 3 1-3 correspond to ARIsTOT., meteorolog. I 4 - 8, placit. 3 4 to meteorolog. II 8 - ΠΙ 1, and placit. 3 5-6 plus 18 to meteorolog. ΠΙ 2 - 6. Consequently, in the Placita the proper place of the compromise chapter 3 6 concerned with shafts and mock suns as combinations of real existence and appearance indeed is a position at the end of the treatment of meteorological phenomena from Milky Way to halo, that is to say of the treatment of phenomena in relation to which the distinction between real existence and appearance was traditionally believed to make sense. I therefore submit that the original sequence of these chapters was 1-2-3-4-5-18-6. These chapters then were followed by chs. 7 and 8.1, concerned with topics for which the distinction between appearance and reality was (believed to be) írrelevant. 118 The whole `metarsiological' section was (and is) then rounded off by the note at ch. 8 2 that the author will now continue with the πρóσyε~α because the description of the µετ ~ρσια has been completed.*

(or "fiction"). For the expression cf. Dioi Cmtvsοsτοµ., oration. 18.18, σχολικ~~πλ~σΜατα; also ~ των Rh cf. Sextus' definition of πλ~σµα,advers. mathem. 1 263, πλ~σµα ~δ~(sc. ~στιν ~κθεσις) πραγµ γενοµ ~νων µ ~ν οµο ~ ως δ~~τ~~ς γενοµ ~νοις λεγοµ ~νων ("fiction sets out things which are not real, but are similar to real things in the telling"). 117 The macro-structure of Book III as a whole, as compared with esp. Aristotle and Seneca, will be treated another time. 118

See above, n. 14

ad fiit.

* Thanks are due to the members of the Roman corona for observations on the occasion of the oral presentation of an early version of the present paper in June 2002, and to Keímpe Algra, Frederik Bakker, Jan van Ophuijsen, David T. Ruina and Teun Tieleman for critical remarks on drafts in December 2002 and January 2003. As always the mistakes which remain are mine.

—56—

FROM MILKY WAY TO HALO COMPLEMENTARY NOTE 2 Also see the revised Italian version, J. MANSFELD, Physikaí doxai e problemata physics da Aristotele ad Aexio (ed oltre), in A.M. BATTEGAZZORE (ed.), Dimostraziοne, argomentazione dialettica e argomentazione retorica nel pensiero antico, Genova, Sagep, 1993, pp. 311-382, at pp. 313 ff., and at pp. 341-356, "Aristotele e í `placita' relativi alla terra" (here one will also find references to some of my earlier publications). See also J. MANSFELD, Aetius, Aristotle and others on coming to be and passing away, in V. CASTON and D.W. GRAHAM (edd.), Presocratic Philosophy. Essays in Honour of Alexander Mourelatos, Aldershot, Ashgate, 2002, pp. 273-292. Useful notes (as well as references in the apparatus superior) concerning Aristotelian antecedents for the contents of lemmas in placit. Book 111 in G. LACHENAUD (ed.), Plutarque: 6Euvres morales t. Χ112: Opinions des philosophes, Paris, Les Belles Lettres, 1993, pp. 127-142, pp. 261-274 (see also his synopsis, of the "exposés de la météorologie ", ibid., p. 51). Several useful references also in the comments of H. STROH', Aristoteles: Meteorologie. Über die Welt («Aristoteles. Werke in deutscher Übersetzung», XI Ι, 1 -2), Darmstadt, Wiss. Buchges., 3 1984. H. DIELS, in his unrivalled masterpiece Doxographi Graeci, Berlin 1879, pp. 229232, explains passages such as these, which are paralleled in Aristotle, as derived from Posidonius or a follower of Posidonius (cf. above, notes 22, 30, 56, 73, 86). For placit. 3 1 (3 1.8 = Pos τDON. fr. 129 E.-K.) also cf. I.G. KIDD, Posidonius: 11. The Commentary: (í) Testimonia and Fragments 1-149 («Cambridge Classical Texts and Commentaries», X[VA), Cambridge etc., Cambridge Univ. Press, 1988, pp. 487-488, and cf. above, notes 22 and 26. Note that Kidd tends to follow Diels in considering certain doxographícal passages to be, perhaps, Posidonian (cf. above, notes 22 and 55); see the critical remarks of K.A. ALGRA, review of I.G. KIDD, op. cit., «Classical Review», XLΙ (1991), pp. 316-319, at pp. 317318. I prefer to replace the pseudo-precise siglum "Posidonius" by the formula "doxographical tradition(s)". COMPLEMENTARY NOTE 5 The brief introductions and occasional notes in the Budé edition of P. OLTRAMARΕ, Sénèque: Questions naturelles, Paris, Les Belles Lettres, 1929 and later repr., are often useful, e.g. pp. 4-5, on indirect quotations from the Meteorologica of Aristotle in Book I. Useful are also the accounts of the relation of Seneca's treatise to Aristotle's and purported intermediary sources by C. COD ΟΙ ΕR MERINO, L. Annaei Senecae Cuestiones Naturales, vol. 1 («Colección hispánica de autores griegos y latinos»), Madrid, Consejo Sup. Invest. Scient., 1979, pp. xxx-xxxv, by D. VOTTERO, Questioni naturali di Lucio Anneo Seneca («Classici Latini UTET»), Torino, UTET, 1989, pp. 24-39 (in substantial detail), and by P. ΡARROνι, Seneca. Ricerche 5

— 57 —

JAAP MANSFELD

sulla natura («Fondazione Lorenzo Valla»), Milan, Mondadori, 2002, pp. xxi τxxvτ. On the sources of esp. Book Π see further H.M. HONE, An Edition with Commentary of Seneca Natural Questions, Book Two, Salem, New Hampshire, Ayer, 1984, pp. 55-62. Some material is to be found in N. GROSS, Senecas Naturales Quaestiones. Komposition, naturphilosophische Aussagen und ihre Quellen («Palingenesia», XXVII), Stuttgart, Steiner, 1989, a book without an index, and so quasi-impossible to use. Cf. also above, n. 57.

COMPLEMENTARY NOTE ~~ Also cf. G. LACHENAUD, op. cit., p. 51 (see above, p. 57, complementary note 2). Frederik Bakker points out to me that the sequence in (the abstract of) Theophrastus' Metarsiologica roughly resembles that of the Placita: TREOPHR. chs. 1-6 AËr. 3 3 (lightning etc.), ΤκεοΠΗR. chs. 6-12 - AËT. 3 4 (clouds etc.), THEOPER. ch. 13 - AsT. 3.7 (winds), THEOPHR. ch . 14 - AËT. 3 18 (halos: misplaced chapter in Aëtius), TREoPHR. ch. 15 - AlT. 3 15 (earthquakes). But in the Metarsiologica chapters on the Milky Way (- MT. 3 1), comets (- AËT. 3 2), the rainbow (- ΑΙ r. 3 5), shafts and mock suns (- AMT. 3 6), summer and winter (- AËT. 3 8), etc., are lacking. These differences are substantial. Theophrastus may have viewed the Milky Way as an astronomical rather than a meteorological phenomenon (for the doubtful evidence see above, n. 22 ad finem), which would explain its absence. Whether he viewed comets as astronomical rather than meteorological phenomena is uncertain; for suggestions and references see I.G. KDD, Theophrastus' Meteorology, cit., p. 297, and R.W. SHARPLES, Theophrastus of Eresus. Sources for his Life, Writings, Thought and Influence. Commentary, vol. 3.1, Sources on Physics (Texts 137-223) («Phílosophia Antiqua», LXXIX), Leiden etc., Brill, 1998, p. 153. The fact that both the rainbow and the mock suns are missing from the Metarsiologica is at any rate quite remarkable; accordingly, no conclusion about the status of the Milky Way and comets can be drawn from the fact that they are lacking in the extant remains of Theophrastus' treatise.

— 58 —

ALDO BRANCACCI Università degli studi di Roma Tor Vergata Centro Linceo Interdisciplinare B. Segre, Accademia Nazionale dei Linceí

STOBAEUS ANTHOLOGIUM IΠ 24

Ι Chapter 24 of Book III of Stobaeus' Anthologium contains a small, unexpected anthology entitled: Περι συν&&δ~τος, consisting of sixteen passages by various authors all foregroundíng Greek lexemes expressing the idea of consciousness/conscience or closely linked terms.' The chapter occupies a significant position in the book and deserves preliminary highlighting. It is part of a sub group of six chapters, distinguished by their ethical-psychological content from the previous and subsequent ones which have strictly ethical content. After setting out in columns twenty chapters devoted to the same number of moral notions, rigorously subdivided into positive (odd numbered chapters) and negative ones (even numbered chapters) and reaching, with the twentieth chapter the concept of 4471, Stobaeus jumps, in an original, but perfectly coherent fashion, to a chapter entitled Περι τov yν~~θι σαυτ~ν (21), in opposition to one entitled Περι úπερoΨιας (22), both being followed by the pair entitled Περι φιλαυτ~ας (23) and Περι συνειδ~τος (24), and then by Περι µν ~ µης (25)and Περι λ~θης (26). Positioning in the even numbered chapter column immediately warns us that the notion of συνεtδ~ς includes reference to something negative, although, in this case, this shade of meaning is to be understood in a mediated sense, and, as we shall see, a very special one. The term supplying the chapter with its title: τ~~σννειδ~ς is, together with συνεíδησις, one of the Greek nouns designating the consciousness/conscience phenomenon, 1 There is very little bibliography on this chapter, which has been largely ignored by scholars. Recent reference to this part of Stobaeus' work has been made by A. Banνcnccι, Socrate e il tema semantico della coscienza, in G. Gιnννeντονι & M. Nµ tcυ (edd.), Lezioni socratiche, Napoli, Bibliopolis, 1997 («Elenchos», 24), pp. 281-282, and by B. MARzuL.Lo, La «coscienza» di Medea, «Museum Criticum», ΧΧΧΠ-ΧΧΧΝ, 1997 -2000, p. 70, n. 17.

—59—

ALDO BRANCACCI

which is also expressed by related terms, and, above all using a verb rather than a noun. There is ancient, albeit sporadic evidence for both words: the neuter substantívísed participle σννειδ~ς first appears in Dem οsthenes,2 though still lacking the sense of moral conscience, while σννεtδησις made Its first appearance in an ethical fragment of Democritus. 3 It was only from the 1st century B.C. that the two nouns συνεíδησις and σννειδ~ς, together with the Latin noun conscientia, were widely used with the sense of conscience. They appeared both in Jewish-Hellenistic and Roman spheres and in pagan Greek ones. And it can be noted, although it is not easy to establish a different shade of meaning between the two lexemes, that συνε~δησις was more frequently used for self-consciousness in the non-moral sense. 4 Stobaeus set out his chapter on the basis of four different traditions of thought, freely juxtaposing and alternating the poetic, philosophical, rhetorical and sapiential ones. It is worth noting, because it can so easily be forgotten, that the various traditions from which Stobaeus gathered his material correspond to those normally drawn upon for quotation from auctoritates present in any moral treatise dating from the Roman Imperial period, Plutarch's Moralia being an obvious example. The organisation of maxims and excerpta was clearly free, although there are partial references between the entries, though poetical quotations tend to be collected together in the first section, before the group of philosophical maxims. Among the sayings of famous philosophers one is attributed to Socrates, which, following the chapter's logic, should represent his concept of conscience. From this saying we should start our investigation of this almost entirely neglected document regarding the idea of «conscience» in the ancient world: ,

[1] From Socrates [= S.S.R. I C 278] «Socrates, when asked to say who lived untroubled lives, answered: "those who have nothing out of place on their consciences"». 5 2 Cf. DEMOSTH., Orat., XVI Ι, 110: in his speech Demosthenes ignored certain things, since he thought that he could count on the fact that they were well known to his listeners ( ~µοιως παρ'

Vµ ~ν ~ΚUσΤO σννειδ~ς vπ~ρχειν µο ).

Cf. Srοa. IV 52, 40 (= 68 B 297 D.-K.), on which see note 17. Cf. Cu. MAURER, s.v. σ~νοιδα, oννε~δηστς, in G. KITTEL -G. FRIEDRICH (edd.), Theologisches Wörterbuch zum Neuen Testament, Stuttgart, Kohlhammer, 1964, cull. 897-918. The oldest studies of the idea of conscience and self-consciousness include M. KAHI.m t, Das Gewissen. Ethische Untersuchung. Die Entwicklung seiner Namen und seines Begriffes, Halle, Fricke, 1878 [Unveränd. reprograf. Nachdruck Darmstadt,1967]; F. TILLIANI, Zur Geschichte des Begriffs `Gewissen' bis zu den paulinischen Briefen, in Festschrift für S. Merkte, hrsg. von W. SCHELLBERG, Düsseldorf, Schwann, 1922, pp. 336-347; H. OSBORNE, Σννειδησις, «Journal of Theological Studies», )0001,1931, pp. 167-169. 5 STOB. 11 24, 13 (= S.S.R. I C 278): Σωκρ~της ~ρωτηθε 'ις τινες ~ταρ~χως ζ~σιv ε~πεν• "ο~~ 3

4

µηδ ~ν ~αντο1ς ~τοπον σΥΙειδ~τες".

—60—

STOBAEUS ANTHOLOGIUM

111 24

If in this χρε~α conscience and related consciousness appears to be closely linked with the theme of man's happiness or unhappiness, so that Socrates could state that a good conscience was the necessary condition for enjoying a blissful life, there is a further reason for holding that this theme had literary origins and was not a mere invention. This statement can be found neither in Plato nor Xenophon, neither can it be considered an apophthegm resulting from expansion or simplification of a statement by Socrates in these two writers. It is especially distant from treatment of the theme of consciousness in Plato's Apology of Socrates, where it is the marker of an entirely gnoseological theme, the fact that be conscious and so «sharing knowledge» with oneself of something that, concerning the very sphere of one's own knowledge, can only be known by oneself, produces knowledge that possesses the characteristics of absolute certainty, so as to have the same value as the authority of god in Socrates' examination of reason. 6 The way in which the theme of ~ΤΟΠΟΝ appears is equally distant from the Platonic Socrates. In this maxim, it simply designates what is strange or out of place, in the ethical sense, and thus of such a nature as to appear «bad». Here we are a very long way from the positive ~τοπ~α attributed to Socrates by Plato, in the sense of a key to his irony, dissimulation, exceptionality, irreducibility to a formula: in short, from the value of the term for which Socrates said of himself ~τωΠ~ΤαΤ~ς ε 4~1. xαì. πΟuü Τovς ~Νθρ ~ΠσΥς ~πορε~ν, or for which the drunken Alcibiades praised Socrates' ~τοπ~α, or for which the moral truths stated by Socrates seemed ~τοπα to his astonished listeners, including Callicles. 7 In respect of this tradition, which is particularly Platonic, that of which our maxim is an expression necessarily refers back to the peculiarly Socratic branch of our tradition. Here it should be noted that, although it is not to be found in Xenophon, the maxim is certainly close to this writer's usage, i.e. the formula συν&&δ~ναι ~αντ~, where consciousness/conscience is what determines a condition of happiness or unhappiness in man,8 and, in particular, is not far from the meaning taken on by this expression, in Socrates' own words, in a passage in Xenophon's Apology, where the requirement of consciousness/conscience as the Ego's personal 6 Cf. PLAT., Apo!. 21 B 3-7, and 22 C 9-D 2. On these passages cf. A. BRAiCACCI, Socrate e il tema semantici della coscienza, cit., pp. 290 301. 7 Cf. PLAT., Theaet. 149 Α 8-9; Symp. 215 Α 2 -3 (e D 3); Gorg. 494 D 1. But the quotations -

— especially those concerning the theme of Socrates the philosopher who says ~τοπα — could increase in number. 8 See, for example, XEioPH., Anab. 115, 7, on which cf. Cκ. MAURER, art. cit., col. 900.

— 61 —

ALDO BRANCACCI

judgement on its own behaviour is clearly expressed. 9 Following this path, and taking into account the fact that Socrates' saying is immediately followed by a maxim of Diogenes of Sinope, whose content and tone are similar, it is legitimate to posit the hypothesis that the former's remark was extracted from a λ~γος Σωκρατικ~ς of Antisthenes, or that it originated in the branch of the Socratic tradition associated with Antisthenes and the Cynics. This supposition agrees with two certain data, to which it is important to draw attention. Socrates' maxim is evidence of the motif of the good and bad conscience, in the sense of backward looking consciousness of one's own behaviour, which originated in the Sophist sphere and was absent from Plato, and circulated, to some extent, in 4th century ethics. 10 Theoretically, this motif is linked with the idea of a splitting or duality of consciousness/ conscience, which, thanks to this Spaltung carries out a process of self reflection: an idea linked, in turn, with the emergence, not only in the tragedians, as has already been noted, also in the sphere of Socratic literature, Antisthenes in particular, of psychological motifs, linked with the idea of personal consciousness/conscience, such as modesty, guilt, shame and repentance. 11 From this viewpoint Socrates' «conscience» (ín the maxim quoted above) is the expression of an ethical, psychological attitude, which can be legitimately termed «moral conscience», and which well represents the conceptual content summed up in Stobaeus' chapter. From this point of view the latter appears as a significant piece of evidence of the representation of the notion which had already come to the fore in an entirely different theoretical context and cultural and spiritual climate, in late Hellenistic and Imperial literature, in the writings of Ph ilo of Alexandria and the Jewish-Christian tradition, from the first epistle of Peter to Paul of Tarsus. 12 9 Cf. XEiOPH., Apol. Socr. 24: ~ς δè τ~λος εiχεν ~~δ ~κη, εïa~~ν αοτ~ν 'Aλλ', ~~~νδρες, τσYc λ~ν ~Ν ~ΜΟ~~κα~~τοvς πειθΟΜ~Νονς Τοοδιδ~σκοντας ΤΟvς λ~ρΤΥρα( ~ ς (ρ~~~π~ΟρκΟ~ΝΤας Κατaψενδοµαρτυρε τοις ~Ν~γκΤ ~στ~~πΟλλw ëavτο~ς συνειδ~ναι ~σ~Πεmmν κα~~ ~δικ~αν. Cf. also ΧnνοΡΗ., Cyr. IΙ 3,38. 10 On all this cf. F. Zucκmt, Syneidesis-Conscientia. Ein Versuch zur Geschichte des sittlichen Bewusstseins im griechischen und im g riechisch-römisch Altertum, Jena, Fischer, 1928, pp. 7-20, now in ID., Semantica, Rhetorica, Ethica, Berlin, Akademie-Verlag, 1963, pp. 96-117. 11 Restricting ourselves to Antisthenes, it is sufficient to recall the concept of ~δον~~~ιεταµ ~λητος, mentioned by Anmi. XII 513 a (= S.S.R. VA 127); the themes of guilt and atonement mentioned by STOB. r 18, 26 (= S.S.R. VA 124); the motifs emerging from the long self presentation 0f Antisthenes ín XENOPH., Symp. 11 34-44 (= S.S.R. VA 82). On all these passages cf. A. ΒaλνcACCI, Érotique et théorie du plaisir chez Antisthène, in M.-O. GOULET-CAZY et R. GOULET (edd.), Le cynisme ancien et ses prolongements, Paris, Presses Universita ri es de France, 1993, pp. 35-55. 12 On the notion of consciousness/conscience in Philo cf. E. BRYHIER, Les idées philosophiques et religieuses de Philon d'Alexandrie, Paris, Vrin, 1925, pp. 295-310; A. PELLETIER, Deux

expressions de la notion de conscience dans le judaïsme hellénistique et le christianisme naissant, «Revue des Études grecques», LXXX, 1967, pp. 363-371. For New Testament use and Paul's

— 62 —

STOBAEUS ANTHOLOGIUM

111 24

Despite the fact that the idea of «consciousness/conscience» has often been claimed to have been alien to the Greek world, from Hegel to Schwartz, Jaeger, and Stenzel, 13 we know that the ideas of personal and moral consciousness/conscience were perfectly well known to Classical Antiquity. 14 Hegel himself, complicating, although not contradicting, his interpretation of the Ancient Greek world as a world of objectivity still lacking the principle of the inner nature and consciousness of subjectivity, attributed to Socrates the merit of discovering man's inner nature and moral consciousness/conscience. 15 And Wilamowitz defined Socrates as «das wandelnde Gewissen», ~ 6 in accordance with a trait, which is worth noting, which, once again, was to be newly taken up and developed by the Cynics. It has already been mentioned that the idea of consciousness/conscience was not expressed by a single lexeme in Greek, but by a whole group of related ones, and, albeit with different conceptual shades of meaning, also by terms, which, though independent, are, in a way, markers of, or substitutes for the idea. The lexeme that most appropriately expresses the seconcept of συνεíδησις cf. C.A. PIERCE, Conscience in the New Testament, London, SCM Press, 19582 ; J. STELZENBERGER, Syneidesis in neuen Testament, Paderborn, Schönfngh, 1961; Cu. ΜAURER, art. cit. 13 Cf. E. ScηwλRTz, Ethik der Griechen, Stuttgart, Koehler, 1951, pp. 90-91; W. JAEGER, Paideia. Die Formung des griechischen Menschen, Berlin, De Gruyter, 1947, vol. I, p. 32; J. ST ΣNZEL, Platon der Erzieher, Leipzig, Meiner, 1928, p. 278 sq., p. 281. 14 Among the scholars who have made the greatest contribution to documentation of the development of the theories of moral consciousness/conscience in antiquity, as well as the presence of notions like guilt and sin, it is sufficient to mention here W. NESTLE, Vom Mythos zum Logos, Stuttgart, Kröner, 1940; B. SNELL, Die Entstehung des Geistes. Studien zur Entstehung des europäischen Denkens bei den Griechen, Göttingen, Vandenhoeck & Ruprecht, 19754 (Hamburg, Claassen & Goverts, 1946); C. DEL GRANDE, Hybris, colpa e castigo nell'espressione poetica e letteraria degli scrittori della Grecia antica, Napoli, Ricciardi, 1947; E.R. D ΟDDS, The Greeks and the Irrational, Berkeley, University of California Press, 1951 («Sather Classical Lectures», 25); O. SEEL, Zur Vorgeschichte des Gewissens-Begriffes im altgriechischen Denken, in Festschrift Franz Dοrnseiff, Leipzig, Bibliograph. Inst., 1953, pp. 291-319; R. MoNDOLFO, La comprensione del soggetto umano nell'antichità classica, Firenze, La Nuova Italia, 1958 («11 pensiero sto rico», 2); ID., Moralisti greci. La coscienza morale da Omero a Epicuro, Napoli, Ricciardi, 1960. 15 Cf. G.W.F. HEGEL, Vorlesungen. Ausgewälte Nachschriften und Manuskripte, Hamburg, 1983. Bd. 7: Vorlesungen über die Geschichte der Philosophie, Teil 2, Griechische Philosophie. I. Thales bis Kyniker, hrsg. von P. Garnfron u. W. Jäschke, Hamburg, Meiner, 1989, pp. 127-165. Cf. also G.W.F. HEGEL, Vorlesungen über die Philosophie der Weltgeschichte, hrsg. von G. Lasson, III. Band: Die griechische und die römische Welt, Leipzig, Meiner, 1920, pp. 643-647, in part. p. 644: «Sokrates hat die Innerlichkeit des Menschen zu seinem Bewusstsein gebracht, so dass in dem Gewissen das Mass des Rechten und Sittlichen aufgestellt würde». 16 Cf. U. iii WILAMOwiTZ-M0ELLEND0BFF, Platon, vol. I: Leben und Werke, Berlin, Weidmann, 1919, p. 104. It should be recalled that he only used this formula because he glimpsed a characteristic example of inner conflict in Socrates, where he rejected the concept of moral consciousness/conscience in the latter and his followers, since neither wished to distinguish between subjective and objective consciousness of truth and moral values.

—63 —

ALDO BRANCACCI

mantic theme of «consciousness/conscience», in Latin «conscientia» is συappearing for the first time in Dem οcritus 17 — to be linked, in turn, with the verbs συν&&δ~ναι and συνειδ~ναι ~αυτ~, as well as the similar or closely linked expressions avwoia, συwoε~ν ~αντ~, συνιστορε~ν, συγΥιyν~σκεΙν ~αντ~, συνεπιατααθαι ~αυτ~. It can be generally stated that the idea of «consciousness/conscience» was represented, apart from the above mentioned nouns, by the synthesis of a verb expressing a gnoseological activity (εïδ~ναι, νοε~ν, yη'ν~σκεww, iατορε~ν, ~π~ατασθα~) with a preposition indicating concurrence. In opposition to the previous idea which held that this linguistic structure meant a kind of split or duality of consciousness/conscience, inasmuch as it is awareness of oneself, 18 more recent research has shown, by means of general and detailed analyses of these formulae, that the originative meaning was that of "sharing knowledge with few others", and thus, in the end — a situation indicated by the term ~αντ~~- `only with oneself", in the solitude, that is, of one's own «consciousness/conscience», something unknown to others, through which, in respect of a particular content of knowledge, a situation comes about of exclusive possession. 19 This meaning is mirrored in the usage of 5th and 4th century literature, where the use of συνειδ~ναι and συνειδ~va i ~αντ~~is widespread, ranging from Aristophanes, Aeschylus, but above all Sophocles and Euripides, to the Sophists, and some philosophers and orators. The "sharing of knowledge" to which these verbal forms refer νειδησις, a noun

,

17 Cf. STOB. IV 52, 40: "Some men who have no idea of the disintegration to which mortal nature is subject, but have consciousness ( συνειδ~σει) of their bad behaviour in life, are tormented by anxieties and fears throughout their lives, since they fashion in their minds untruthful fables concerning the period after death". On this fragment see W. NESTLE, Bemerkungen zu den Vorsokratikern and Sophisten, «Philologus», LXVII, 1908, p. 548, and pp. 104-107. It should be remembered that the lexeme συνειδησις is of lonían origin. 18 Cf. B. SNELL, review of F. ZUCKER, Syneidesis-Conscientia, cit., «Gnomon», VI, 1930, pp. 21-30, reprinted in Gesammelte Schriften. For detailed criticism of this thesis cf. A. CA νcτuντ, Syneídesis. Il tema semantico della "con-scientia" in Grecia antica, Roma, Ateneo, 1970, pp. 2833. It should, however, be recalled that the idea of a Spaltung of the Ego, which is basic to Snell 's interpretation of verbal forms such as σ νειδ~ναι and συνειδ~ναι ~αvτ45, is actually present, as I mentioned above, though not in their semantic structure, but, at the purely logical level, in two special uses of the theme of «consciousness/conscience», well documented in antiquity and often inter-connected: the distinction between a good and bad conscience, and that of conscience as a tribunal and final judge before which man must account for his actions. The concept of an Aufspaltung of the Ego has been investigated, within a wider analyses, by M. CLASS, Gewissensregungen in der griechischen Tragδdie («Spudasmata», 3), Hildesheim 1964. 19 Cf. A. CλνcRmνι, Syneidesis, cit., pp. 23-24: "In questo caso, non tutti sanno: il sapere privilegio solo di alcuni, o, all'estremo, solo di se stessi. Συνειδ~ναι designa cioè un «con-sapere» privato, ossia un «con-sapere» che, sia nel caso del «con-sapere» con altri, sia nel caso del «consapere» solo con se stessi, pone colui che sa in una posizione privilegiata, perche in condizione di conoscere qualcosa che è nascosto ad altri".

64

STOBAEUS ANTHOLOGIUM 111 24

can concern an individual's act or intimate condition, or actions of others; on the other hand, it can also concern others' acts considered morally bad, or, more generally, designate the evaluation expressed of an individual act or that of others of which "knowledge is shared": which can lead either to highlighting certain, solid knowledge of the truth, i.e. συν&&δ~ναt ~αυτ , or the judgement element, more precisely, moral judgement, pertaining to this evaluation, which makes «consciousness/conscience» the infallible judge of the Ego's mortal life. 20 Actually, the expression σννειδ~ναι ~αντ~~newly appeared with this precise meaning in the following saying attributed to Diogenes of Sinope: [2] From Diogenes [= S.S.R. V B 305]: «Who could fear anything less, and be more courageous than he who has nothing bad on his conscience?». 21 From the Quellenanalyse viewpoint, this saying could be considered either a copy of the previous one or, more probably, an authentic saying, where Diogenes was bearing in mind Socrates' words. This is all the more plausible if the latter originated in Antisthenes' Socrates. There are other examples of similar back references in Diogenes. 22 Support for this line of argument comes from the initial y ~ρ, which I do not believe is there by chance, and which foregrounds the fact that the saying is not a true χρε~α (clearly not having its structure) but that it was extracted from a wider context, most probably a diatribic one, seeing that the interrogative sentences, or series of interrogative sentences are proper to the diatribe. 23 And since the question under examination has clear thematic affinities with the Diogenes extract, reproduced by Stobaeus under the entry ~κ τ~ν ∆ιογYνους δrnτριβ ~ν, 24 the explanation given here appears the most plausible 20 For the latter meaning cf., for example, Alusr οεu., Eq. 183-184; Thesm. 473-479; Vesp. 999-1000; EUmP., Med. 492-495; Or. 395-396. For the meaning of συνειδεναι as private, intimate awareness of an experience concerning others cf., for example, Anscn., Chie. 216-217, and, on these lines, B. SNELL, rev. of F. ZucκΕR, cit., p. 25. ~ λιστα, 21 STOB. III 24, 14 (= S.S.R. V B 305): ∆ωωΥενους• τις 7àp äν r~ ττον φοβο~τ~~τι ~~θαρσοιη µ ~ ν; δmmmς α{τιδ µηδεν συνειδειη κακ 22 Cf. DIOG. LAERT. VI 24 (= S.S.R. V Β 303): συνεχ~ς τε éλεyεν [sc. Diogenes] εις τ~ν βιον παρεσκεν~σθαι δει' λ~γον -4 βρ~χον, which is clearly repetition of the famous ancient saying, which was especially praised by Chrysippus and given by Antisthenes in his Protepticus: cf. PLUTARCH., de Stoic. rep. 14, p. 1039 E-1040 A (= S.S.R. V A 105). 23 There are a great many examples in the diatribes of Tries, for which see the recent edition by P.P. FUENTES G ονΖλι.ΕΖ, Les diatribes de Télès, Paris, Vrin, 1998 («Histoire des doctrines de l'Antíquíté classique», 23). 24 STOB. 1V 9, 46, which I have dealt with in my articles Le orazioni diogeniane di Dione Crisostomo, in G. GIANNANTONI (ed.), Scuole socratiche minori e filosofia ellenistica, Bologna,

— 65 —

ALDO BRANCACCI

one. To this should be added that praise of Θ~ρρος and blame of φ~βος are typical Cynic motifs.25 But the greatest interest of this saying by Diogenes lies in Stobaeus' viewpoint and method of composition. It is now becoming clear that, through his excerpta, Stobaeus wanted to offer a more or less complete overview of the philosophers and philosophical schools that were most interested in the theme of consciousness/conscience. After Socrates, it was the Cynics who invested the theme of moral consciousness/conscience with the greatest importance, and, besides, as shown by the case of Oenomaus of Gadara, also that of self-consciousness, in the strictly theoretical sense. 26 This is confirmed by Stobaeus' third choice: a dictum attributed to Pythagoras: [3] From Pythagoras: «He who commits injustice suffers more from the torment of his conscience than if his body had been struck by whip lashes».27 The authenticity of this saying cannot be accepted without question. It probably dates back to the Hellenistic age or early centuries of the Roman Empire. Its intent is to date the theme of consciousness/conscience as far back as Pythagoras. Here it is placed at a higher level than any punishment from the outer world. This theme certainly belonged to the Pythagorean sphere, in this sense confirming Stobaeus' above mentioned plan of offering a representative conceptual retrospective of his chosen theme. Mondolfo underlined the fact that the origin of the elaboration of the theme of consciousness/conscience was already to be found in ancient Pythagorean philosophy, recalling the practice of examining one's conscience as well as 11 Mulino, 1977, pp. 141-171, and Democrito e la tradizione cinica, in Democrito e l'atomismo antico, Atti del Coni. Intern. (Catania 18-21 apr. 1979), a cura di F. Romano, «Siculorum Gymnasium», n.s. X00 (1980), pp. 411-425: 418-425. 25 On the theme of courage in Antisthenes cf. DuG. LAERT. VI 8 (= S.S.R. V Α 89); for blame of fear cf. STOB. III 8, 14 (= S.S.R. V Α 79); for praise of ~νδρεíα, cf. PLUTARCH., de Alex. M. fort. ant virt. II 3, p. 336 Α (= S.S.R. V A 77); XEiOPH., Symp. II 12. It should also be recalled that Antisthenes wrote a Περì. ~νδρειας: cf. DuG. LAERT. VI 16. These themes were inherited by the Cynics: for the connection of θ~ρσος and absence of Ψ~βος, cf. especially Teles p. 4, 3 f. Hense and the note by P.P. FuεiTεs GONZALEZ, Les diatribes, cit. pp. 109-110. 26 On the concept of self-consciousness, understood as συναισθησυυ και ~ντιληynς αßτο~, in Oenomaus, cf. Eus., Praep. euang. VI 7, 10-12, and my article La polemica antífatalistica di Enomao di Gadara, in A. BRANCACCI (ed.), Antichi e Moderni nella filosofia dieta impe riale, Atti del Colloquio internazionale, Roma, 21-23 settembre 2000, Napoli, Bibliopolis, 2001 («Elenchos», 34), pp. 71-110: 83-89. 27 STOB. III 24, 8: -a~à í ζω π~σχει &&à. τov σΥνειδ~τσς ~~äδικ~ν 3aσανιΟµεΝσς' [~] τοï~~ σ~µατι [κα ~~τα~~ς] πληΥα~ς ΜαστιΥο~Μενος. ,

— 66 —

STOBAEUS ANTHOLOGIUM

111 24

the theme of shame. 28 The latter is mentioned in a passage from the Sacred Discourse, which was also deliberately quoted by Stobaeus to complete his Pythagorean documentation:

From Pythagoras [Am-. carm. v. 12 Nauck]: «Above all feel a sense of shame for yourself». 29 Stobaeus also asked himself about the origins of the notion of moral consciousness/conscience, and, most probably, intended to show how important this theme was by including two maxims attributed to two of the so-called Seven Sages in his chapter: From Bias: «Bias, asked to say what condition of life excluded all fear, answered: "an honest conscience"». 30 "

From Periander: «Periander, asked to say what freedom was, answered: a good conscience "». 31

Maurer argued that the two maxims date from the Christian era and were applied to pagan philosophers, 32 but he ignored the fact that similar maxims to those attributed to Bias and Periander, whose Christian origin has no foundation, are also to be found in other chapters of the Antho ~ο.gΙυm. 33 The theme of the good and bad conscience already existed in the Hellenistic age,34 and arguably these sayings were also of Hellenistic origin, coming quite 28 Cf. R. lINDOLFO, Moralisti greci, cit. On the importance of the idea of conscience in the Pythagorean tradition in the Hellenistic age cf. also Ci. NAVRER, art. cit., col. 904. 29 STOB. ΠI 24, 2: Πνθaγ~ρου• π~ντων δ~~ µ ~ λιοτ α~σχ~νεο σaυτ~ν. Cf. Iamblichi de vita Pytha-

gorica liber ad /idem codicis Florentin recensaitAugustus Nauck, accedit epimetrum De Pythagorae aureo carmine, Petropoli-Lipsiae 1884, p. 205. Cf. anche J.C. Τηοµ,The Pythagorean Golden Verses. With Introduction and Commentary, Leiden-New York-Köln, Brill, 1995 («Religions

in the Graeco-Roman World», 123), p. 96. For the connection between moral consciousness/ conscience and shame ( αiσχ~νη), XEiOPH., Anab. I 3, 10 and the words of the drunken Alcibiades in PLAT., Symp. 216 B should be recalled. 30 S TOB. ΠI 24, 11: Β~ας ~ρωτηθεις τ~~~ν ε{.η τ~ ν κaτ~~TUV β~oν áφοβον, ε1πεν «~ρθτ~~ σ νε~δησις». 31 JD. IΠ 24, 12: Περ~ανδρος ~ρωτηΘεìς τ~~ ~στιν ~λευθερ~α ε1πεΙ «~yαθ~~~ συνε~δησι ». 32 Cf. Cκ. MAURER, art. cit., p. 901. 33 Cf. bannis Stobaei Anthologium, recensuerunt C. Wachsmuth et O. Hense, vol. V, Berlin, Weidmann, 19743 , Index Auctorum, p. 1168 (Bias) and p. 1191 (Periander). 34 For the evidence cf., for example, some maxims of Middle and New Comedy playwrights (to be dealt with later), the fr. from Chrysippus taken from DI0G. LAERT. 11185 (= SVF ΠΙ 178), as well as passages such as CICER., de/in. Π 71; de leg. 140; SENEC., de tranq. an. 1I 4; de ben. IV 21, 5-6; ep. 97, 15-16. On the concept of moral consciousness/conscience in Cicero and Seneca cf. J. DUPONT, Syneidesis aux origines de la notion de conscience morale, «Studia Hellenistica», V, 1948, pp. 119-153; In., Gnosis: La connaissance religieuse dans les épîtres da Saint Paul, ParisLouvain, Nauwelaerts, 1949, pp. 266-268.

— 67 —

ALDO BRANCACCI

possibly from Cynic circles. This is also certainly the reason why Stobaeus did not think it necessary to quote B 297 D.K. from Democritus, which is the oldest surviving use of the lexeme σννε~δησις. Although, despite the σονε~δησις κακοπραΥΜοσúνης link, the term is not particularly incisive in this fragment – and this is in agreement with the fact that Stobaeus quoted it, twice, in the chapters Περ~~τov β~ο~, δτι βραχvς κα~~εvτελ~ς κα~~φροντ~δων ~ναµεστ ~~(N 34, 62) and Περ~~ζωi'ς (N 52, 40) – it is certain that reference to the most ancient sapiential tradition served the purpose of conferring exemplary antiquity and dignity on the theme of good and bad conscience. Furthermore, seeing the content of the maxims, the theme was also invested with the peremptory character, favoured by Stobaeus in this chapter.

II In the 5th and 4th centuries the formula συνειδεναι ~αυτ~~had found special favour among orators and rhetors, in the case of Gorgias, taking on the meaning of certain, solid knowledge of the truth, or, at least, judgement of one's behaviour with defensive or accusatory intentions, in the case of Antiphon of Rhamnus, that of careful examination of man's behaviour, aiming at shedding light on the practical and psychological consequences determined by a good or bad conscience, in the case of Isocrates, illustrating situations from social life, allowing more favourable and peaceful integration of those with nothing to blame themselves for, and, finally, expressing the theme of good and bad conscience, which, in Demosthenes, was coherently included, in various ways, with its characteristic ethical implications, in the argumentative texture of the speeches. 35 But it was only to Isocrates, to the παραιν~σεις, that Stobaeus turned, probably attracted by the former's bland, rational moralism, to provide evidence for the part of ancient tradition concerning the idea of consciousness/conscience. 36 In this situation quotation of the theme 35 For Gorgias cf. Pal. 5, 11 e 15 (= 82 B lia DK); for Antiphon of Rhamnus cf. de caed. Herod. 52, 93; Chor. 1, 5; for Demosthenes cf., for example, de falsa leg. 33, 208; de cor. 263. 36 The fortune of Isocrates under the Empire was already well consolidated in the Second Sophist period, especially with Aelius Aristides, concerning content and subject choice, and with Herod Atticus, concerning the formal imitation of his style. As L. GuλLDO Rοsλ, La fede nel la paideia'. Aspetti della fortuna europea di Isocrate nei secoli XV e XVI, Roma, Pubb licazioni dell 'Istituto Storico Ita lian o per il Medio Evo, 1984 («Studí Sto rici», 140-142), p. 11, noted, Isocrates, together with Lysias and Demosthenes, entered the purest Attic dialect of Pollux and Harpocration and the treatises on style by Aphtoníus and Hermogenes, which were to enjoy great success in Late Antiquity and the Byzantine period. The fortune of Isocrates in collections

— 68 —

STOBAEUS ANTHOLOGIUM 111

24

of the judgement function of moral consciousness/conscience in a non-Isocratic paradigmatic passage from Tο Demonicus was inevitable: From Isocrates, To Demonicus [16]: «you do something bad, never hope to stay hidden. Since, even if you remain hidden from others, you will have consciousness of it within yourself». 37 If the context of this statement, which closely recalls the content of some Democritus fragments, is the listing of precepts to be followed to make progress on the path of virtue, but also to have a good reputation among all others, the statement from Nicocles is similar, only more solemn. This is a fictitious speech by the Cypriot prince to his subjects setting out their duties towards their ruler. 38 The φUβαα theme returns in this passage, and συν&&δ~νςι is simply substituted by εiδ~ναι: From Isocrates, Nicocles or The Cyprians [52] «Do not hide anything from me of what you possess or what you do or intend to do, knowing well that secret actions are surrounded by many fears». 39 Another maxim form Nicocles, placed by Stobaeus at the very end of the chapter, needs to be quoted here: From Isocrates, Nicocles or The Cyprians [59]: «Do not envy those who have very large possessions, but those who have no guilt on their consciences: the most pleasant kind of life can be led in such a s ρirit». 4o Here the saying tones down the stricter climate of the previous maxims with a note of calm eudaemonism, completing Stobaeus' picture and conof moral maxims began with Stobaeus and continued, in the Byzantine period, with Suidas and Johannes Geogides: cf. C. Wncηsµ uτη, Studien zu den griechischen Florilegien, Berlin, Weidmann, 1882, pp. 45-89, and Isocratis opera omnia, recensuít, scholiís, testimonies, apparatu criticu instruxit E. Dremp, vol. I, Lipsiae, Dieterich, 1906, pp. cvn-cxm. ~δ~nοτε µηδ ~ν α~σχρ~ν nΟι~σας ~ λπιζε λ~σειν 37 STOB. III 24, 9 1σοκρ ~τους πρ~ς ∆ηΜ~νικov µ και 1,44 ~ν τOvς ~λλoυO λ~θης, σaυτ~{"" συνειδ~σεις. 3a It should be noted that the choice of Nicocles is evidence of the success of royal speeches and panegyrics during the Empire and Late Antiquity, which has already been linked with the transiatíon from the Principate to the absolutist age. Cf. P. Ηλnoτ, Fürstenspiegel, in Reallexicon für Antike and Ch ristentum, VIΙΙ, Stuttgart, Híersmann, 1972, pp. 574-576.

39 STOB. 1II 24, 10: Τσοκρ~τονς Νικοκλ~ς ~~ΚYiριοι. µηδ ~θ ' ~ν ~ποκρ~πτεσθε ~ µ8' ~ν κ~κτησθε µ ~ των ~ναyκα~~ν ~στι ~να TGry πραγµ ~ν nΟιε~τε 009 ' ~ν µ ~ λλετε nρ~ττε~Ν, ε~δ~τες ~τι περι τ κεκροιµ πολλο~ς φ~ βο~ς γ~γνεσθαι.

~νους, ~λλ~~τοvς µηδ ~ν κακ~ν σφ~σιν α~τ~~ς συ 40 In. III 24, 16: ζελο~τε ~ µ~τοvς nλε~στα κεκτεµ νειδ~τaς µετ ~~1,àß τοιavτη' ψυχ~ς ~διστa ~ν τις δ~ναιτo τ~ν βιου διαγαγεΣν.

— 69 —

ALDO BRANCACCI

tributing stable unity to the chapter. It should also be noted that the second person plural of the verb, placed as it is at the end of the chapter, could even be read as a final invitation to the reader, in the style of a true ethical-parenthetic treatise. At this point it should be noted that the significant presence of Isocrates and Pythagorean sayings in Anthologium III 24, highlighting interest in the rhetorical and Pythagorean traditions, well documented in Neo-Platonic literature, supports arguments in favour of influence of the latter on Stobaeus' sources. 41

r The use, both in comedy and tragedy, of συν&&δ~ναι (with an explicit or understood dative, different from ~αvτg~) and σννειδ~ναι ~αυτ~~to express, often in contexts of particular dramatic pathos, the idea of consciousness/ conscience and the "Momente des Affektes" it conveys, is well known and, together with that of σ~νοrn, has been studied in detail. Once again Stobaeus, though making a selection, did not disappoint, because, with six quotations, he documented, on the one hand, use by Sophocles and Euripides of the notion, providing several very famous lines, and, on the other, use by Middle and New Comedy. It is worth beginning with the quotations from comedy: From Antiphanes [fr. 269 Kock = fr. 267 Kassel-Austin]: «Actually not being conscious in one's life of any injustice is a source of much pleasure». 42 From Diphilus [fr. 92 Kock = fr. 92 Kassel-Austin]: «How could he who is not ashamed of himself, though being conscious of having committed deplorable acts, feel ashamed in the face of someone who knows nothing of it?» 43 41 I

thank D. O'Meara, to whom I owe this suggestion. With regard to this topic see I. Hλ-

Le problème du néoplatonisme alexandrin. Hiéroclès et Simplicius, Paris, Etudes Augustiniennes, 1978; A. ΒaλνCACCι, Rhetorike philosophousa. Díone Crisostomo nella cultura antica e bizantina, Napoli Bibliopolis, 1986 («Elenchos», 11); D.J. O'MEAm διà τ~ς παθητικ~ς ~λκ~ς, iYv τ~~8εωρητικαñ