VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 Colotes and the Epicurean Refutation of Skepticism Paul A. Vander Waerdt M y OBJECTIVE here is to map out the successive stages of de­ velopment in the Epicurean refutation of skepticism. It is widely assumed that Epicurus himself developed in full the orthodox battery of Epicurean anti-skeptical arguments,l but I will try to show that these arguments originated at different stages in the history of Epicureanism to meet the challenge posed by different varieties of skepticism. The common assumption that Epicurus himself undertook to refute the skepticism of Pyrrho or of the Academic Arcesilaus rests on questionable conjecture rather than solid evidence: this is nowhere attested, and Epicurus' only re­ corded anti-skeptical argument, in fact, is directed against his De­ mocritean predecessors. In this case he simply modifies the atomist theory of perception so as to disarm its skeptical implications, employing neither of the strategies (the self-refutation and apraxia arguments) usually attributed to him by modern scholars. Epicurus' strategy against the atomists, however, is not effective against the more powerful version of skepticism practiced by the Academic skeptic, who advances no 06yJ.1(l'tCl (not even his skep­ tical practice of £1tOxit or of suspending assent in all matters) as true in his own name, but merely adopts his dogmatic opponent's prem- 1 These arguments are set out in most detail by Lucretius in DRN 4.469-521. "'It is agreed on all sides that Lucretius in the argument of 469-521 is following Epi­ curus," says M. F. Burnyeat ("'The Upside-Down Back-to-Front Sceptic of Lucre­ tius IV 472," Philologus 122 [1978] 197-206 at 200), but he offers as argument only "'compare 483-499 with D.L. X 31£" (see contra 241 infra). A. Barigazzi, "'Epicure et Ie scepticisme," in Actes du VIlle Congres, Assoc. Guillaume Bude (Paris 1970) 286-93, concludes that "'Epicure eut la constante preoccupation de combattre Ie scepticisme," but only by attributing all Epicurean anti-skeptical arguments to Epicurus himself-most implausibly, since some of these arguments are directed against forms of skepticism not yet current in Epicurus' day. The first part of Mar­ cello Gigante's Scetticismo e Epicureismo (Naples 1981) 25-106 discusses many relevant texts but does not consider the development of the Epicurean refutation of skepticism; summary and discussion in D. Fowler, aSAP 2 (1984) 237-67. 225 VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 226 THE EPICUREAN REFUTATION OF SKEPTICISM ises in order to show that he contradicts himself and thus had bet­ ter, by his own arguments, suspend assent.2 Since this skeptical con­ clusion rests solely on the dogmatist's own premises, there are no skeptical 50YJ.Lo:ta against which he can argue. Thus, to discredit the skeptic, he must attack his argumentative practice, typically (a) by alleging that the skeptic refutes himself by inconsistently advocating his skepticism in contravention of his recommendation that one suspend assent in all matters, or (b) by arguing that the practice of epoche makes human action impossible and so throws life into con­ fusion. It was Epicurus' young contemporary Colotes, I shall argue, who first brought these arguments to bear against skepticism, thus developing an anti-skeptical case which became so much a part of orthodox Epicureanism that its origin in his controversy with the skeptical Academy has been forgotten.3 And once the origin of these arguments is understood, their philosophical significance will emerge more clearly. Since antiquity a monolithic picture of the development of Epi­ curean philosophy has prevailed that seems to be based on little more than the authority and reverence Epicurus commanded with- 2 For a better understanding of classical skepticism, I am much indebted to Michael Frede's pioneering ·Des Skeptikers Meinungen, JJ Neue Hefte fur Philo­ sophie 15-16 (1979) 102-29, and ·The Skeptic's Two Kinds of Assent and the Question of the Possibility of Knowledge, JJ in Philosophy and History, edd. R. Rorty et aL (Cambridge 1984) 255-78, both now available in his Essays in Ancient Philosophy (Minneapolis 1987) 179-222. I have also learned much from recent work on the New Academy: A. M. Ioppolo, Opinione e Scienza: il dibattito tra Stoici e Accademici nel III e nel II secolo a.c. (Naples 1986); J. Annas, ·The Heirs of Socrates," Phronesis 33 (1988) 100-12; G. Striker, ·Sceptical Strategies," in Doubt and Dogmatism, edd. M. Schofield et a1. (Oxford 1980) 54-83; D. N. Sedley, ·The Motivation of Greek Skepticism," in The Skeptical Tradition, ed. M. F. Burnyeat (Berkeley 1983) 9-29; J. Annas and J. Barnes, The Modes of Skepticism (Cam­ bridge 1985); J. Glucker, Antiochus and the Late Academy (Gottingen 1978), with Sedley, -The End of the Academy," Phronesis 26 (1981) 67-75; H. Tarrant, Scep­ ticism or Platonism? The Philosophy of the Fourth Academy (Cambridge 1985). 3 For previous work, discussed below, see P. DeLacy, ·Colotes' First Criticism of Democritus," in Isonomia, edd. J. Mau and E. G. Schmidt (Berlin 1964) 67-77, and the introduction to his Loeb edition of Plutarch, Adversus Colotem, with B. Einar­ son, Moralia XIV (Cambridge 1967) 153-87; also: R. Westman, Plutarch gegen Kolotes (Helsinki 1955) 45-107,293-95, 302f, 310; DeLacy's review, AJP 77 (1956) 433-48; G. Arrighetti, ·Un passo dell'opera 'Sulla Natura' di Epicuro, Democrito e Colote," CronErcol9 (1979) 5-10; Fowler (supra n.l) 256-67. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 227 in his school. 4 Modern scholars habitually assume, often uncritical­ ly, that arguments found in Epicureans as late as Lucretius or Dioge­ nes of Oenoanda derive from the Master himself. This assumption is not, of course, a discovery of philology: as Seneca tells us in con­ trasting Epicurean subservience with Stoic freedom of thought, apud istos quidquid Hermarchus dixit. quidquid Metrodorus. ad unum refertur (Ep. 33.4; cf Numenius ap. Eus. Praep. Evang. 14.5.3f); and the difficulties of attribution philologists face in estab­ lishing the patrimony of specific Epicurean sayings provide striking confirmation of Epicurus' success in teaching his companions to master his doctrines through memorization.s The conventions of philosophical argumentation in antiquity, which generally did not allow the imputation of authority to anyone but the founder of one's school and his own acknowledged authorities,6 no doubt also encouraged Epicurus' followers to clothe their philosophical inno­ vations in well-worn garb. But the assumption of doctrinal conser­ vatism has, unsurprisingly, contributed to a very misleading view of the historical development of Epicurean philosophy. In recent years some progress has been made in recognizing the important innovations introduced by Epicurus' colleagues and successors, but much remains to be done.? The present paper is offered as a small contribution in this attempt to individuate the innovations of early members of the school. The present subject offers, I think, an ex- -4 For the reverence bestowed on the Founding Fathers (oi avop£~ ) of Epicurean­ ism, see F. L. Auricchio, «La scuola di Epicuro," CronErcol8 (1978) 21-37; Colotes was not reckoned among them, but this proved no bar to influence in the later tra­ dition. 5 See D. Clay, Lucretius and Epicurus (Ithaca 1983) 76-81. 6 See D. N. Sedley, «Philosophical Allegiance in the Greco-Roman World," in Philosophia Togata, edd. J. Barnes and M. T. Griffin (Oxford 1989) 97-119. 7 On tradition and innovation in Epicureanism, see F. Longo Auricchio and A. T epedino Guerra, "Aspetti e problemi della dissidenzia epicurea," CronE rcol 11 (1981) 25-40; D. Clay, "Individual and Community in the First Generation of the Epicurean Schoo~" in l:\)~"'tT)O"t~: Studi ... Marcello Gigante I (Naples 1982) 255- 79; Sedley (supra n.6). For an attempt to show that Epicurus' successor Hermar­ chus introduced important innovations into Epicurean anthropology in response to contemporary Stoics, see P. A. Vander Waerdt, "Hermarchus and the Epicurean Genealogy of Morals," TAPA 118 (1988) 87-106. Clay's work is exceptional in its attention to the specific cultural, rhetorical, and philosophical problems faced by later Epicureans: supra n.5; "The Cults of Epicurus," CronErcol 16 (1986) 12-28; «A Lost Epicurean Community" (313-35 infra), and the paper cited in n.66 infra. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 228 THE EPICUREAN REFUTATION OF SKEPTICISM ceptionally clear illustration of the way in which they developed their philosophy in original ways to respond effectively to contem­ porary challenges. I. Colotes and the Skeptical Academy On chronological grounds alone, it would be surprising had Epicurus undertaken to counter Academic skepticism: he died in 271, before Arcesilaus' accession as scholarch during the Olympiad 268/4. In fact, there is no evidence that Epicurus ever took an in­ terest in Arcesilaus. 8 The only version of skepticism he is attested to have combatted is that of his atomist predecessors (on which see 236££ infra). Shortly after Epicurus' death, however, the philosophi­ cal terrain in Athens changed dramatically as Arcesilaus, upon his accession as scholarch, turned the Academy skeptical. Academic skepticism differs radically from its precursors, such as Pyrrho, in that it does not present itself as a OO'YIlCl, a positive thesis the truth of which the skeptic advances in his own name. As Cicero reports (De or. 3.67, Acad. 1.44f; cf Fin. 2.2, Nat.D. 1.11), Arcesilaus rather conceived of his skepticism as the authentic interpretation of Socrates' dialectic: he did not advance any doctrine of his own (Acad. 2.17), but like Socrates he found, upon examination of inter­ locutor after interlocutor, that none possessed the special expertise or knowledge to which he laid claim, even by his own premises and canons of logic.9 Thus Arcesilaus found himself, in case after case, forced into epoche or the suspension of assent (0''U'Y1CClta9Ecnc;). He 8 In the only apparent exception, Adv. Col. 1121 E (=Us. 239), which Usener as well as DeLacy and Einarson take as a testimonium about Epicurus ('t0l) Se 'APKE­ .... E'tpiroC; £Ol.lC£V it &6l;a 1tapaAU1tE'iv), W. Cronert's emendation 'E1t\.KOUPElOV for 'E1tlKOUpOV (Kolotes und Menedemos [Leipzig 1906] 13 n.54), not mentioned by DeLacy, is preferable because it avoids the harsh transi­ tion that would make Colotes the unnamed subject of the next sentence. (Dirk Obbink adds: "And if the emendation seems harsh, cf. Pluto De sera numinis vindicta 548A where the MSS universally corrupt E1t\.KOUPE\.0C; to E1t\'KOUpOC; [see Pohlenz in his 1929 Teubner]; here the un-named individual cannot be Epicurus himself, as he is said to be a contemporary of Plutarch.") Usener counts A dv. Col. 1121 E as a fragment of Epicurus on the unsupportable claim (348) that Lucr. 4.469-521 records Epicurus' reply to Arcesilaus. 9 This interpretation of Arcesilaus requires defense: see 260ff infra; for his inter­ pretation of Socrates, see especially Julia Annas, ·Plato the Sceptic" (forthcoming in aSAP Suppl. 3); also Glucker (supra n.2) 31-47; Frede (1984 [supra n.2D; and section V below. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 229 did not advocate epoche-or any other thesis-as a dogma, but merely drew upon his opponent's premises for purposes of de­ bate. 10 Thus the Academic skeptic is not vulnerable, as the dog­ matic skeptic is, to the charge of self-refutation as characterized above. Now, Arcesilaus' popularity is well-attested (a source of grief for Colotes: Adv. Col. 1121 E, 1124B; cf Strab. 1.2.1, quoting Era­ tosthenes; D.L. 9.44), and so it became particularly important, given the oral and agonistic character of Hellenistic philosophical dis­ cussion, for contemporary Epicureans to develop a new line of argument that took into account the philosophical resources of his version of skepticism. This task was undertaken by Epicurus' young contemporary Colotes, whose book, On the point that it is not possible even to live according to the doctrines oj the other philosophers (1tEpt 'tOU on X:Cl'to. 'to. 'trov CiAArov cptAoaocprov bOyJ.1Cl'tCl OUbE ~i\v tanv, Adv. Col. 1107E), developed a new battery of anti-skeptical strategies for use against Arcesilaus, strategies which he turned against some of his dogmatic opponents as well. Colotes' book so offended Plutarch four hundred years later that he composed two treatises in reply, 11 the first of which, Against Colotes, provides all of our named testi­ mony concerning Colotes' arguments. 12 Plutarch so divorces 10 I am assuming that Arcesilaus did not, as some sources suggest, advocate epoche as a positive dogma, but this is controversial; see section VI infra. 11 Since no Academic prior to Plutarch is attested to have answered the Epi­ curean apraxia argument, it is unlikely that he is relying upon some earlier Aca­ demic source. Plutarch plainly has Colotes' book before him, and I see no reason to doubt his claim that his refutation originated in oral discussion within his school (Adv.Col. 1107D-ll08B, Non posse 1086c-D). For his method of composi­ tion, see esp. Harold Cherniss' remarks in the Introduction to his edition of De Stoic. Rep. in the Loeb Moralia XIII.2 (Cambridge [Mass.] 1976) 396-400; for his habit of keeping "notebooks" (imoJ.lvTtJ.lu'tu, De tranq. 4641'), cf De cohibenda ira 457D with H. Martin, GRBS 10 (1969) 69f. If there had been an earlier Academic reply to the Epicurean apraxia argument, we can expect Plutarch to have recorded it, for he was deeply interested in the history of skepticism: among the nine works on this subject whose titles are preserved in the Lamprias catalogue (collected by DeLacy and Einarson [supra n.3] 187) we find Ei l:btpu1('to~ (,1tcpl 1taV'tCl)V £1tEXCl)V (Lamprias no. 210). 12 Colotes wrote two other books, Against Plato's Lysis and Against Plato's Euthydemus, intended to discredit Academic skerticism. For texts, see Cronert (supra n.8) 5-7, 163-70, with the new readings 0 A. Concolino Mancini, "Sulle opere polemiche di Colote," CronErcol 6 (1976) 61-67; for discussion, see Westman (supra n.3) 31-39, and 257 infra. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 230 THE EPICUREAN REFUTATION OF SKEPTICISM Colotes~ arguments from their original context (his own complaint about Colotes: Adv.Col. 1108D; cf. Non posse 1086D) that it is often difficult to reconstruct the original motivation or meaning of particular arguments; and much of our effort in the present paper will be devoted to this task. It is also difficult to ascertain what Colotes~ motives were in se­ lecting the philosophers and schools he chose to include in his book. Beginning with Democritus, who takes pride of place as the father of atomism, and concluding with certain unnamed contem­ poraries whom Plutarch identifies as the Cyrenaics and the Aca­ demic followers of Arcesilaus, who suspend assent on all matters (oi 7tEpi 7teXv'tcov £7t£XOV'tEC;, Adv.Col. 112(k;-D), Colotes attacks in chronological order Parmenides, Empedocles, Socrates, Melissus, Plato, and StilpO.13 It is often supposed that Colotes attacks philoso­ phers other than Arcesilaus (whom he never explicitly names: lac. cit.) simply because they had been claimed as authorities by the skeptical Academy, but this view cannot stand without substantial q ualification.14 In the first place, the list of philosophers whom Co­ lotes attacks does not correspond exactly to any list of Academic authorities. Secondly, Colotes makes no mention of Pyrrho,15 an omission that is most difficult to explain if, as some contemporaries thought (Ariston of Chios and Timon: D.L. 4.31£), he importantly influenced Arcesilaus' philosophical development. 16 Either Colotes 13 Plutarch avowedly departs from Colotes' order for his own polemical pur­ poses (Adv.Col. IlDE); see DeLacy and Einarson (supra n.3) 155f. 14 So DeLacy and Einarson (supra n.3) 156: -The book is an attack on Arcesi­ laus. The other philosophers are singled out because the skeptics of the Academy regarded them as predecessors." I doubt A. A. Long's similar thesis (ap. Sedley [1983 (supra n.2)] 24 n.27) that Colotes derived his list of opponents from Ar­ cesilaus. Pluto A dv. Col. 1108 B implies nothing about Arcesilaus' own authorities, and 1121E-F is a report of the charges of unnamed contemporary sophists and need not reflect Arcesilaus' own claims at all; in any case, Colotes does not attack Heraclitus, so he is unlikely to have simply taken over his list of opponents from Arcesilaus. As for Cicero, Acad. 2.14 is not a list of Academic authorities (note that Colotes does not attack Anaxagoras or Xenophanes), and the same applies to 2.72-76. Acad. 1.44-46, which does appear to be a list of Socrates' skeptically inclined predecessors, possibly deriving from Arcesilaus, includes Democritus, Anaxagoras, Empedocles, and omnes paene veteres. 15 For the possibility that Colotes is silent about Pyrrho because he did not con­ sider him a skeptic, see 235f infra. 16 Cf n.32 infra. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 231 did not consider Pyrrho a precursor of Academic skepticism, or he is not attacking Arcesilaus' authorities. Thirdly, it is unlikely that Ar­ cesilaus would have claimed Stilpo or Colotes' other contemporary opponents, the Cyrenaics, as authorities to establish a pedigree for his advocacy of epoehe. Finally, since Colotes' purpose is to show (as his title announces) that it is impossible to live according to the doctrines of the other philosophers, he has no reason to restrict his attack to Arcesilaus and his acknowledged authorities. What is striking in the list of Colotes' opponents is that all of them somehow reject the Epicurean tenet that all impressions are true (ef n.37 infra). I suggest that Colotes' purpose is to discredit those opponents who cast doubt upon the plain evidence of the senses, regardless of whether Arcesilaus claimed them as authorities for his skepticism. If this is so, we can easily see why Colotes discusses figures like Stilpo or the Cyrenaics, who cannot plausibly have been claimed as authorities by Arcesilaus, but who did cast doubt on the evidence of the senses, and why he takes no account of the Stoics or Peripatetics, the contemporary schools whose epistemol­ ogy was least likely to be accounted skeptical. It is far more likely that Colotes attacked such comparatively minor figures as Stilpo and the Cyrenaics because Epicurus had already put them on the roster of the school's rivals, 17 than that Colotes sought to discredit them as skeptical authorities. Of course, Colotes may well have considered Arcesilaus his most important opponent: Colotes is re­ ported to have been most annoyed by his reputation (Adv.Col. 1121E), he puts him last in his book, and his attack on Socrates, I shall argue, certainly is intended to discredit Arcesilaus' use of him as an authority for epoehe. 18 17 For Epicurus' criticism of the Megarian Stilpo, see Sen. Ep. 9.1, 8,18 (=frr.173- 75 Us.). It is a puzzle to me why Colotes takes no account of Diodorus Cronus and the Dialektikoi, since Diodorus importantly influenced Arcesilaus' dialectical technique (and was seen to have done so: Ariston of Chios and Timon ap. D.L. 4.32f; cf. Sext. Emp. Pyr. 1.234, Numenius ap. Eus. Praep.Evang. 14.5.11-6.4) and since both Epicurus (On Ambiguities; De Nat. 28 fr.D col. v.6-9) and Metrodorus (Against the Dialecticians: D.L. 10.24) had written books against him; on this see Sedley, "Diodorus Cronus and Hellenistic Philosophy," PCPS N.S. 23 (1977) 74- 120. As for Epicurus' criticism of Cyrenaic doctrines of pleasure, see D.L. 10.136£ (quoting several works by title), frr.449-53, 509 Us. 18 There is ample precedent in Hellenistic philosophical argumentation for Co­ lotes' tactic of responding to a contemporary opponent by refuting his authorities. Thus in his Against Empedocles Hermarchus discredits the use certain unnamed VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 232 THE EPICUREAN REFUTATION OF SKEPTICISM Colotes' principal anti-skeptical strategy is to adapt an argument that featured prominently in Stoic attacks on Arcesilaus, that uni­ versal epoche actually makes it impossible to live. Colotes naturally modifies this apraxia argument (a version of which Epicurus had already used against the ethical determinist in On Nature) to serve Epicurean rather than Stoic epistemology, and against opponent after opponent he argues that to deny the truth of all impressions is to abolish knowledge, and that without knowledge life becomes im­ possible.19 In putting his case, Colotes is given to rather graphic illus­ trations: of Arcesilaus, for instance, he asks, "how is it that the man who suspends judgment does not go running off to a mountain instead of to the bath, or why does he not get up and walk to the wall instead of the door when he wishes to go out to the agora?" (Adv. Col. 1122E; discussed 262ff infra). This colorful polemic well serves Colotes' rhetorical purpose of discrediting his opponents by making them appear ridiculous. The alleged inconsistency between an opponent's views and conduct was a common argumentative strategy in Hellenistic philosophy, one especially useful in silencing him in public debate. Colotes characterized the doctrine of epoche as bait for bold and flighty youth (Adv.Col. 1124B), and no doubt sought to turn the tables against Arcesilaus by discrediting epoche through its inconsistency with living. But his apraxia argument, as we shall see in detail in the next section, does pose a serious chal­ lenge to the Academic to show how universal epoche is consistent with the requirements of life itself. Colotes influenced the subsequent history of Epicureanism much more than has been recognized. His attacks on Academic skepticism quickly became orthodox. The third scholarch, Polystra­ tus, who succeeded Hermarchus in the mid-third century, wrote a pamphlet entitled Against those who unreasonably despise popular beliefs;20 his polemic against contemporary Academic skeptics contemporary rivals-in this case Peripatetics and Stoics-were making of Emped­ odes as an authority for their own views about the natural kinship among human beings; see Vander Waerdt (supra n.7) esp. 89f. 19 For the variants in how Colotes puts this charge, see DeLacy and Einarson (supra n.3) 157 note a. 20 Polistrato: SuI disprezzo irrazionale delle opinioni populari, ed. G. Indelli (=La scuola di Epicuro II [Naples 1978]). VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 233 draws upon Colotes.21 The arguments that Colotes developed to counter Academic skepticism came to feature prominently in such later sources as Lucretius (4.507-10; cf Adv.Col. 1122E) and Dioge­ nes of Oenoanda (fr.6 col. II Chilton), and were even adopted by Sextus (Math. 8.56) in his own attack on Democritus.22 These arguments thus became part of the Epicurean anti-skeptical arsenal. Nor is this the only way in which Colotes' work proved influential: he is responsible for putting Socrates on the map of the Epicurean school's philosophical opponents, and for defining the terms in which he is criticized. Finally, Colotes' extensive attacks on Plato's use of myths led to considerable discussion in antiquity: Colotes had wondered, in the case of Er in the Republic, how a dead man can come back to life;23 and Cicero, according to Macrobius (Com. Somn.Scip. 1.9-2.4), preferred in consequence to have his tale re­ lated by one roused from a dream. Colotes' other objections led to extensive replies by Porphyry and Proclus as well as Macrobius, and this debate over the proper philosophical uses of myth would well repay study. 24 We shall begin with the problem of Ericurus' relation to skepti­ cism; then turn to Colotes' adaptation 0 the apraxia argument for use against his atomist predecessor Democritus, against the princi­ pal authority of his Academic opponents, Socrates, and against Arcesilaus himself; and conclude by considering the effectiveness of Colotes' argument as a refutation of Academic skepticism. 21 For evidence linking Polystratus to Colotes in the anti-skeptical polemic sig­ nalled at De cont. xXlv.2-7, see Indelli (supra n.20) 55-71, who calls attention to the similarity between Colotes' characterization of Arcesilaus (,.UlX£'t~atJlovTtottV tiC; 'tpla 'ta,ha ~Abtttv' 1tpilhov Jltv, 01t0'ia 1tt1tOV AEAoyicr8at ~ie; ~£ 'tOY a,.Uptcr{311'touv'ta 'tou ~Tt Ka'teX , . 'tpo1tOV. This sort of account is self-refuting, and can never prove that every­ thing is of the kind called "'necessitated -; but he debates this very question on the assumption that his opponent is himself responsible for talking nonsense. And even if he goes to infinity saying that this action of his is in turn necessitated, always appealing to arguments, he is not reasoning it empirically so long as he goes on imputing to himself the responsibility for having reasoned correctly and to his op­ ponent that for having reasoned incorrectly.42 Epicurus' claim here is not that the ethical determinist's position is refuted by the content of his thesis; it is, rather, refuted by his conduct in arguing it. He imputes responsibility to his opponent for reasoning badly, and takes credit himself for reasoning correctly: thus his thesis that necessity, rather than he, is responsible for his actions is refuted by the very way in which he advances the thesis. In Epicurean terms, the ethical determinist fails to employ £7tt­ AO'Yt(JJ.16~, the empirical reasoning which, Epicurus explains in the concluding section of On Nature 28, enables one to assess the truth of a view with reference to the advantageousness of the actions based upon it.43 Consistency between an agent's views and his actions thus is required by Epicurean epistemology, and the im­ putation of such inconsistency to an opponent alone suffices to convict him of false reasoning. The argumentative strategy that Epicurus employs against the ethical determinist thus rests on a cardinal tenet of his episte­ mology. Hence it is not surprising that this strategy turns up in a very different context, when Epicurus claims that if an agent fails to refer every action to nature's telos, his actions will not follow upon his views. 44 A similar strategy is evident in Epicurus' criticism of Democritus (De nat. 34.26-30=20C.51-56 L.-S., quoted 253 infra). 42 This is Sedley's text and translation; some modifications are proposed in Laur­ sen (1988 [supra n.33]). 43 See esp. fr.13 VI.I0-XII.20 in Sedley's edition, "Epicurus, On Nature Book XXVIII," CronErcol 3 (1973) 5-83, with commentary 65-79. 44 K D 25: d""1] 1tapa 1to.v'ta lCatpOV £1taVOlOtte; £lCao'tov 'troY 1tpanOllivcov £1ti 'to 'tiAoe; 'tlle; q>UOtCOe;. clA.M 1tpolCa'taUrTlV Ei'tt ~lcol;tV 1tOtOOJ.LtVOe; tic; liUo 'tt, aUlC £oov'tal Oot'tOle; AOyOtt; al1tp6.l;tte; cllCOAou8ot. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 241 In fact, since an Epicurean can always enquire whether an op­ ponent's views accord with £1ttAOYtCJ)lO<;, the tactic of imputing an inconsistency between these views and his conduct could be employed in virtually any case. Now, it has been supposed that Epicurus himself used this tactic in refutation of skepticism: Lucretius' peculiar image of the skeptic as one who has himself stood with his head in his own footprints (4.469-72), it is claimed, must spell out the meaning of Epicurus' term 1tEpuca'tO> 'tPE1tECJSUt (De nat. 34.26-30=20C.25 L.-S.; cf [35J 11 Arrighetti). "Epicurus used 1tEptlCa'tO> 'tPE1tECJSUt to charge the skeptic with self-refutation and Lucretius translated his master's prose into precise poetic detail. "45 Against this, it must be repeated that Epicurus' target in On Nature is not a skeptic, and that he is not attested to have used the self-refutation argument against skep­ ticism. As we saw earlier, there is no evidence of an effort on his part to counter Pyrrho's skepticism or Arcesilaus'. The parallels between DRN 4.469-521 and On Nature are not, in my opinion, close enough to sustain the claim that the refutation of the skeptic exactly parallels that of the ethical determinist;46 and we should not attribute to Epicurus himself an elaborate critique of skepticism merely on account of a verbal parallel that could just as well be explained in other ways.47 For instance, Lucretius' image could be based on the anti-skeptical arguments of one of Epicurus' fol­ lowers, or it could represent his own application of Epicurus' ar­ gument in On Nature to the case of the skeptic. In either case, there is no necessity to attribute Lucretius' entire argument at 469- 521 to Epicurus himself. In fact, this argument may include attacks 45 Burnyeat (supra n.1) 204, endorsed by Sedley (supra nAO) 2M. 46 I depart from Sedley's excellent discussion (supra nAO) in finding no exact parallel between the argument of On Nature 34.26-30 and DRN 40469-521: (a) the targets of these passages are very different, the former being directed against the ethical determinist and the latter against the skeptic; (b) Sedley himself admits (27) that the appeals to preconception in DRN 40473-77 and On Nature 34.38-41 are advanced on different grounds; (c) the criticism of Democritus for being blind to the practical consequences of his actions and Lucretius' arguments (4.500-21) that none of the senses is false seem to me so different as to provide no parallel at all. Lucr. 4.507-10, I suggest, draws on Colotes (see nA8 infra), and this is the only real parallel to Epicurus' criticism of Democritus. 47 I offer this consideration in response to Sedley'S claim (supra nAO: 18) that the source of DRN 40469-521 "can hardly be doubted to be Epicurus." VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 242 THE EPICUREAN REFUTATION OF SKEPTICISM on varieties of skepticism not developed until after Epicurus' death.48 So we should be wary of taking any part of it as evidence for an otherwise unattested attack by Epicurus on skepticism. It is Colotes, I shall now argue, who first introduced the standard Epicurean rejoinder to the Academic skeptic-the apraxia argu­ ment, which challenges him to show how, if he does invariablyprac­ tice epoche, he can actually draw the distinctions among external objects that are necessary to live. We must distinguish this apraxia argument from the self-refutation argument that Epicurus uses against the ethical determinist. In the case of self-refutation, the opponent's thesis is refuted not necessarily by its content (De­ mocritus provides an example: see 247ff infra), but by its failure to accord with epilogismos as evidenced by an inconsistency between thesis and some act that follows from commitment to the thesis. 49 This inconsistency suffices, for Epicurus, to invalidate the thesis. 48 The question of Lucretius' target in this passage is a difficult one. Burnyeat (supra n.l: 204), assuming uncritically that it is taken over directly from Epicurus, argues that the quis of line 469 is Metrodorus of Chios-but Epicurus does not need a self-refutation argument to dispose of his position (see supra nn.38f). Bari­ gazzi (supra n.l: 291f), on the other hand, sees the skeptical Academy as Lucretius' target. Of course, since the Academic skeptic does not assert 'nothing is known' as a claim the truth of which he accepts in his own name, he can escape Lucretius' objection at 469f (denique nil sciri siquis putat, id quoque nescit an sciri possit, quoniam nil scire fatetur) by pointing out that he argues from his interlocutor's premises and canons of logic, but that he is no more committed to these than to the conclusion that nothing is known. But Lucretius, like Colotes before him, might nonetheless choose to treat the skeptic's argument as a dogmatic claim (see 262 infra), in which case we cannot rule out an Academic target. Certainly 507-10 closely parallels Colotes' arguments: the claim vita quoque ipsa concidat extemplo recalls Colotes' attacks on his opponents for abolishing life or throwing it into confusion (e.g. Adv.Col. 11080, F); and Lucretius' claim that the skeptic could not avoid precipicies recalls Colotes' polemic against Arcesilaus (1122E). Since the argument of DRN 4.469-521 appears to apply to a broader range of skeptical positions than Epicurus himself was concerned to refute, I am inclined to see it as a composite of arguments developed originally to counter different varieties of skepticism rather than as a translation of a single text by Epicurus himself (the case against regarding Lucretius as a simple paraphrast of Epicurus is eloquently set out by Clay (supra n.5: 13-53). 49 Burnyeat's failure to take account of this consideration leads him to conclude (supra n.l: 204-06) that Epicurus' self-refutation argument is ·patently invalid" and to postulate that -the Greeks had a wider and more hospitable concept of self­ refutation than any we are familiar with in present-day philosophy." In fact, Epicurus leaves no doubt in On Nature 34 (quoted supra 239f) that the self­ refutation argument relies on appeal to epilogismos. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 243 If the opponent can avoid the charge of self-refutation, the Epicu­ rean can still have recourse to an apraxia argument. This argument invalidates the thesis in a rather different way, by holding that one could not even act on the basis of the thesis: thus why does the skeptic go to the door rather than the wall en route to the market­ place, if he does consistently practice epoche? It is sometimes claimed that this objection does not by itself prove the skeptic's skepticism false, but for the Epicurean epilogismos enables one to assess the truth of a view with reference to the advantageousness of the actions based upon it (see supra 240). Thus, to discredit the skep­ tic's skepticism, it suffices for Colotes to point out the possibility that epoche may kill you. Moreover, the Epicurean can also argue, by an application of .., counterwitnessing'" (clv'ttJ.l(Xp'tuPll(Jt(;), that since we do manifestly act, the skeptic in fact cannot invariably suspend judgment in consistency with his thesis. These points de­ serve emphasis in view of the doubts sometimes expressed about the logical validity of the apraxia argument. What then is the Academic skeptic's task in replying to the apraxia argument? It clearly is not to demonstrate that his thesis is internally consistent or even consistent with other theses he holds,50 since this is not directly under attack. His task is rather to show that he can act without contravening his epoche in all matters. In other words, he needs to show that action is possible without assent. If he can succeed in establishing this, he has not only an­ swered the apraxia argument, but also refuted the premises on which relies the further conclusion that his thesis too is false. We shall explore below (VI) the alternatives available to the Academic skeptic in formulating an account of praxis without recourse to rational assent. It is a difficult problem, to which we must now turn, to deter­ mine what philosophical influences led Colotes to choose the apraxia argument for use against the Academic skeptic. But Colotes found this argumentative strategy itself so useful that he brought it 50 I simply mean theses he upholds for the purposes of debate-of course, the skeptic is not committed to their truth. Many later Academic skeptics in fact hold that one may develop quite elaborate theoretical views, provided that one does not mistake them for theoretical knowledge-on this see especially the fundamental study of Tarrant (supra n.2), and M. Frede, -The Method of the So-Called Methodical School of Medicine," in Science and Speculation, edd. J. Barnes et aL (Cambridge 1982) 1-23, and Galen: Three Treatises on the Nature of Science (Indi­ anapolis 1985) ix-xxxvi. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 244 THE EPICUREAN REFUTATION OF SKEPTICISM to bear against nine different philosophers and schools in making his case that it is impossible actually to live in conformity to the doc­ trines of the other philosophers (cf Adv. Col. I11Sc, 1124c). III. The Origins of the apraxia Argument Colotes did not invent the apraxia argument. It has an antecedent already in Aristotle~s criticism in the Metaphysics of those who ad­ vocate ou mallon and deny the principle of non-contradiction. 51 It is obscure when the apraxia argument was first applied to skepti­ cism: if we are to believe D.L. 9.104f, Timon already had to answer it; and the Stoics early on employed the apraxia argument as a re­ joinder to Arcesilaus' attack on the cognitive impression. Priority in the use of this anti-skeptical strategy, I shall now argue, belongs to the Stoics rather than Colotes, though he could also draw upon the example of Epicurus~ argument against the ethical determinist in adapting this strategy for use against the Academic skeptic. The Stoics first employed the apraxia argument as a rejoinder to Arcesilaus, who had attacked Zeno's definition of the cognitive im­ pression ("an impression stamped and reproduced from something which is, exactly as it is") in order to show that the Stoic sage could not distinguish true from false impressions and hence, by his own principles, had better suspend assent (Cic. Acad. 2.77f).52 Since Arcesilaus advocated no oOYJ.ul'ta of his own, but merely took over his opponent's premises for purposes of argumentation, the Stoics had to discredit him without being able to argue against any positive views to which he was committed. Their strategy, at least as early as Cleanthes, was to allege an inconsistency between Arcesilaus' argu­ ment and his conduct (D.L. 7.171). The purpose of this strategy is to confront the skeptic with the following dilemma: either to accept that impressions do somehow provide a criterion of truth, in which case one cannot suspend assent when faced with a cognitive impres­ sion, or to admit that his action (which according to the Stoics does require rational assent) conflicts with his argument (universal 51 Metaph. 1008b14-19: Aristotle's target is not a skeptic; but the form of his argument, in appealing to the observed conflict between his opponents' views and actions, provides a clear antecedent to the apraxia argument. It is unclear whether the early Hellenistic thinkers drew upon Aristotle in developing their argument. 52 On this controversy, see especially M. Frede, ·Stoics and Skeptics on Clear and Distinct Impressions," in Burnyeat (supra 0.2) 65-93. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 245 epoche), in which case he falls into self-contradiction. 53 Since im­ pressions are the basis for apprehension and knowledge in Stoic epistemology, the Stoics put the apraxia argument in the form of a charge that the skeptic, in denying that there are cognitive impressions to which one may assent with certainty, deprives man of his mind and thereby overturns the very foundations of life. As Cicero puts it (Acad. 2.31£): Those who assert that nothing can be grasped deprive us of these things that are the very tools or equipment of life, or rather actually overthrow the whole of life from its foundations and deprive the animate creature itself of the mind that animates it, so that it is difficult to speak of their rashness entirely as the case requires. In order to answer this charge, the skeptic has to show that action is indeed possible without assent, by following the dSAO'¥OV as criter­ ion, such that one can suspend judgment in all matters and still act; hence one can live happily without knowledge as defined by the Stoics. We shall investigate this line of reply later, when considering how Arcesilaus might have answered Colotes' attack. Now, Colotes seems to have taken over this Stoic argument, adapted it to the terms of Epicurean epistemology, and turned it against those philosophical opponents, including Arcesilaus, who somehow called into question the truth of all impressions. On chronological grounds there is no reason why the Stoics could not have borrowed the apraxia argument from Colotes. My reason for doubting this possibility is that Arcesilaus is never attested to have answered the Epicurean criticism, and that his reply to the apraxia argument is formulated in terms of the eulogon (Sext. Emp. Math. 7.158), clearly an anti-Stoic rejoinder. When Plutarch himself re­ counts the Academic response to the apraxia argument (Adv. Col. 1122A-D), he casts it solely in Stoic terms; and in introducing this re­ sponse, he says that those who wrote at length against the pro­ ponents of epoche,54 "at last brought up like a Gorgon from the 53 For this argument see D.L. 9.107f; Cic. A cad. 2.25f, 31f, 53, 99; Pluto Adv.Col. 1122A-D; and, for the Gorgon that turns men to stone, Cic. Fam. 9.8.1 and Epictet. 1.5.1-3. The classic statement of this objection is that of David Hume, An Essay concerning Human Understanding, section 12. 54 Chrysippus is known to have written a book refuting Arcesilaus' method (D.L. 7.198), and traces of his rejoinder to the early Academic argument at Adv. CoL 1122A-D are preserved by Plutarch at De Stoic. Rep. 1036A-B, 1057 A-B (see VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 246 THE EPICUREAN REFUTATION OF SKEPTICISM Stoa the apraxia argument. '" This seems clear evidence that use of the apraxia argument against Arcesilaus originated in the Stoa. ss Since Plutarch in this passage wishes to defend the proponents of epoche against Colotes, it would be far more effective for him to quote a reply that turned the Epicurean criticism against itself. 56 His failure to do so suggests that no reply from Arcesilaus to the Epicu­ reans was available. Priority in the use of the apraxia argument thus belongs to the Stoics. And it is unlikely that Colotes arrived at it entirely indepen­ dently, for his use of the apraxia argument parallels that of the Stoics very closely. He even uses the Stoic term (J'u'YKV (, KroAOn1l<; EO'cpaA1l XEpt Ai~tV 'tou avopo<;, EV n 57 We have virtually no information concerning the early relations between the Epicureans and Stoics, but see Vander Waerdt (supra n.7) 91 n.19 for the scanty evidence. Colotes does mention Zeno and the Stoa Poikile by name in his work on the Lysis, col. VIII (Cronert [supra n.8] 7). VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 248 THE EPICUREAN REFUTATION OF SKEPTICISM OlOp(~E'tal. JlTt JlUAAoV 'to "OEY"" 'to " Jl110EV" elval., "OEV" JlEV ovoJla~c.ov 'to crooJla, "Jl 11 0 EV" oE 'to leEVOV, cOs leal 'tou'tO\) 'YAuKi>" Kat VOJ.Lq> OU'YlCP1.01.V Kat'ta aUa, "E'teft BE 'to KEVOV Kat'taC; (l.'t0J.L0US" aV'tE1.P11J.LEvOV Kal. XpOOJ.LEVOV ouB' Civ au'tov roc; [liv8pooxoS] EO't1.V i1 eft B1.aVo1181lva1.. In the second of his charges he fails even more to notice that along with Democritus he expels Epicurus from the company of the living. For he says that Democritus' sayings that "color is by convention, sweet by convention," a compound is by convention, and so the rest, "what is real are the void and atoms· are an attack by Democritus on the senses; and that anyone who abides by this reasoning and uses it will not even be able to conceive of himself as [a man] or as alive. In the final line, liv6pOJ7toC; is Pohlenz's supplement to fill the lacuna in MSS. E and B. Colotes alters Democritus (A 9, B9, 117, 125 D.-K.) by reversing the order of atoms and void and by interpo­ lating the Epicurean word aU"(1Cptcnc; into Democritus' list of things 62 On this principle of Epicurean logic, see Sedley, ·On Signs," in Barnes (supra n.50) 263-72. 63 DeLacy (1956 [supra n.3]) 434 and (1964) 74-76 advocates a different inter­ pretation of Colotes' first charge: calling attention to Democritus A38, in which 1l110£V IlUAAoV is used to support the view that atomic shapes are infinitely varied, he points out that the Epicureans rejected this view because it upsets the order of nature (DRN 2.496-521) and suggests that ·Colotes began with the charge that an infinite variety of atomic shapes would throw life into confusion." Of course, Plutarch does not understand the charge in this way, nor does he say anything in his long reply (1109 A-Ill OB) that suggests he considered this question relevant. Certainly Democritus did, pace DeLacy, use the ou mallon argument in the way that Plutarch understood it (see supra n.58). Moreover, I very much doubt that Colotes would have opened his book-which after all was written to persuade the public that the doctrines of non-Epicurean philosophers make life unlivable-with such a minor point of technical atomism. Colotes' second charge is much more general in scope, and his polemic would be much more powerful if it called at­ tention from the start to a fundamental problem with Democritean theory-as, on my interpretation, it does. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 251 tha.t are "~y convention." This interpolation will prove to have a senous pomt. In this case Plutarch cannot accuse Colotes of misconstruing De­ mocritus' text (cf. 1111 A-B). Colotes is quite right to take Democri­ tus' views about the conventional character of perceptible qualities as an attack on the senses: as we saw above (237). Democritus takes these qualities to be simply reducible to atoms and void and hence not to have reality in their own right. Epicurus, on the other hand, considers perceptible qualities to be accidental properties of com­ pound bodies; and these qualities are not reducible to atoms and void, which themselves lack perceptible qualities. This difference may explain Colotes' interpolation of (J'6YKPlO'l~ into the Democri­ tean passage: for Democritus, Colotes may mean to suggest, even the compounds that generate the perceptible qualities to which De­ mocritus refers are merely conventional and without real existence. If this is correct, part of Colotes' criticism is directed against the reductionist character of Democritus' views about perceptible q ualities.64 Democritus' view that perceptible qualities are merely conventional is due to his failure to recognize that they are the emergent properties of compound bodies and have an existence of their own over and above their atomic configurations. On this interpretation, it is not hard to see the force of Colotes' second charge against Democritus. Since he denies any reality to perceptible qualities, Democritus concludes that knowledge is not to be found in them-in direct opposition to the Epicurean view that all impressions are true and that there is no knowledge that does not originate in them. Colotes' charge, therefore, is that by denying that compounds are real, and attacking the senses, Democ­ ritus makes it impossible to live. If Pohlenz's supplement in Adv. Col. 1110F is correct, Colotes, in saying that by his reasoning one cannot conceive of himself as a man, is turning Democritus' claim that "man is what we all know" (B165 D.-K.) against him. In adding "or as alive" Colotes suggests that Democritus' metaphysics does not allow him even to conceive of his own existence. One question that remains concerns the relation between Colo­ tes' first and second charges. On the foregoing interpretation, both 6. Colotes' interpolation provides contemporary evidence, I suggest, that Sedley's interpretation of Epicurean metaphysics (see supra n.33), or something like it, must be right. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 252 THE EPICUREAN REFUTATION OF SKEPTICISM employ the same general argumentative strategy: by denying that any impressions are true, Democritus destroys the capacity for dis­ crimination among perceptible objects that is necessary for our life. DeLacy suggests that the first charge "affects sense-objects, the second ourselves as well"; and he rightly points out that this dis­ tinction can be found in Colotes' criticism of other thinkers.65 On the other hand, the first charge does conclude that Democritus' views about perceptible objects throw life into confusion, so it too concerns us as well as the objects of our perception. I doubt that we are to draw a firm distinction between the two charges: each targets a particular Democritean position and argues to the same conclusion on the basis of the same argumentative strategy. Some support for this view is provided by the adaptation of Colotes' two charges by Diogenes of Oenoanda (fr.6 u.2-12 Chilton): icrcpaAT\ 0' av~~i~ Eau'tou Kat .1T\J.lOlCpl'tOe; 'tae; a'toJ.loue; J.lOvae; lCa't' aATt8nav etmov ;.mapxnv EV 'tOte; O~crl, 'ta of: AOl7ta [V]OJ.llcr'tEl a.7tav'ta. lCa'ta yap 'tOY crov AOYOV, 6> .1T\J.lOlCpl'te, Oux 07t~ 'to aAT\8ee; eupetv, aAA' ouoE Cllv OuVTtcroJ.le8a, J.lTt'tE 'to 1tUP CPUA.a't'tOJ.lEVOl [J.lTt'tE 't ]itv , . . crcpaY'1v .... Democritus also made a mistake unworthy of himself when he said that only the atoms really exist in objects, all the rest merely existing by convention. For according to your argument, Democritus, far from discovering the truth we shall not even be able to live, being un­ able to guard ourselves either against fire or slaughter or .... Diogenes quotes the text to which Colotes' second charge re­ sponds, but draws the conclusion of his first, namely that De­ mocritus' view throws life into confusion, making it impossible even to live. 66 Of course, we do not know whether Diogenes drew directly on Colotes' book or on some intermediate source. But the fact that he combines the two charges suggests that they were not considered fundamentally different in form and intention. Our final question is how Colotes' criticism of Democritus com­ pares with that of Epicurus. In concluding his polemic against the 65 DeLacy and Einarson (supra n.3) 159 with note a. 66 The fact that Diogenes fr.6 Chilton draws upon Colotes' criticism of Democri­ tus (so also Chilton [supra n.22] 46) is no surprise, since he also preserves an exten­ sive criticism of Democritus' (and the Stoics') views on dream interpretation which cannot go back to Epicurus himself; for texts and discussion, see D. Clay, «An Epi­ curean Interpretation of Dreams," AlP 1 01 (1980) 342-65. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 253 ethical determinist in On Nature, Epicurus argues that his Democ­ ritean predecessors had to turn a blind eye to themselves "in order to hold necessity and accident responsible for everything.» He then says of Democritus (34.26-30=20C.51-56 L.-S.): b bTt AOYOS au'tos b 'tou'to bt.baO"lccov Kan:ayvu'to Kat EAav9aVEV 'tOY ayopa 'tOLe; eP'YOle; 1tpOe; 't'T)V o6~av v EPYCOV 'tllS bO~T)S iVEyEivE'tO. ~ E/l1tt/lXAa/lEVov bux 'tTtV \>1tEVavno'tTJ'tCl 'troY EPYCOV Kal ~S b6~TJS. Indeed, the actual account promoting this view came to grief when it left the great man blind to the fact that in his actions he was clashing with his doctrine; and that if it were not that a certain blindness to the doctrine took hold of him while acting he would constantly be perplexing himself; and that wherever the doctrine prevailed he would be falling into desperate calamities, while wherever it did not he would be filled with conflict because of the contradiction between his actions and his doctrine. Epicurus claims that Democritus is saved from desperate calamities by his "blindness,» which causes him to overlook his doctrine that necessity and accident alone are responsible for everything. Since this passage is the conclusion of Epicurus' argument against the ethical determinist, the discrepancy between Democritus' doctrine and his action presumably is offered as evidence that the doctrine itself is false. Colotes, in the charges we have just considered, in­ verts this argumentative strategy. He attempts to show that Democ­ ritus' views about the senses and the nature of atoms and void do in fact entail apraxia, that one who acted on the basis of them could not even live. He does not attribute "blindness" to Democritus in order to save him from the calamity of consistency to his doctrines. V. Colotes versus Socrates Colotes' attack on Socrates holds an especial interest for this period of Hellenistic epistemology. As I shall try to establish, Colo­ tes undertakes to discredit Socrates because he had been claimed by Arcesilaus as an authority for his own practice of epoche. Colotes' polemic provides our fullest, albeit indirect, evidence con­ cerning this skeptical portrait of Socrates, one that provided the VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 254 THE EPICUREAN REFUTATION OF SKEPTICISM Hellenistic Academy with its most illustrious authority until An­ tiochus countered by treating Socrates' profession of ignorance as itself ironical. 67 The conclusion that Colotes is attacking Socrates as portrayed by the skeptical Academy will turn out to have impor­ tant implications for the subsequent history of Epicurean discus­ sion of Socrates. For it is Colotes who puts Socrates on the list of the Epicurean school's philosophical rivals, and the fact that he attacks a skeptical Socrates has an enduring effect on the terms in which he is discussed in the later Epicurean tradition. As we saw above (228), Arcesilaus represents his own practice of epoche as the proper interpretation of Socrates' dialectical practice. There is clear evidence that Colotes is responding specifically to this skeptical portrait of Socrates. Plutarch says that Colotes began by dismissing as entirely "sophistical and vulgar" the story of Chaerephon's return from Delphi with his oracle about Socrates. 68 Plato's story of Socrates' elenchos of the oracle (Ap. 20E-23B) appears to have been one of the central texts in Arcesilaus' portrait of Socrates as a skeptic. 69 In fact, no Socratic text provides a better pedigree for Academic dialectical practice than the conclusion of this passage, in which Socrates states that "even now" he goes around searching and enquiring in obedience to the god, coming to his aid by discrediting others' pretensions to wisdom. Arcesilaus apparently cited this very passage as the origin of Socrates' skepticism, for Colotes replies that it is "ridiculous" to make the Delphic command "know yourself" the starting-point of Socratic 67 Cic. Acad. 2.15. 68 Ad7.1.CoL 1116E-F: Eueu~ ~v 'tOY a KCI)Am'tll~, lCat lhll'Yll­ Oa/lEVo~ on XPllO~V £lC ~eA.£­ cr'tl)Kucn 'tOU ).lTJ 'tOY a [YVEU pv'ta A.6yov EV 'to'i~ 1t£pt 9£rov AEyf.tv rocr't£ Kat ~ 00- KPU'tllV Mll'tPOOOOpo~ EV 'trot npO~ EuSUq>pova 'tOY nM'tovo~ (£11tf.P E1tOn 'tou'to) Ka'ta).l~f.!-[q>r.]'tat "'ti vclote; KOtvO>­ v£tv AOyOte;, roe; IllaV dvat Ox£oov 'tl)V Kat' autov ayO>Yl)v Kat 'tl)V TtJ.letEpaV· ou't£ yap 1tCpt \J1tap~£o>e; 11 aVU1tap~lae; ttvoe; a1tov £1tExn. Kat tEA.oe; IlEv ctvm tl)V £1tOXtlV, n ouvnoEp;(ca9at tl)V atapa~iav TtIlEte; £ttKroe;, £KCtVOe; OE roe; 1tpOe; tl)V v a1to1topav £Aall~avc Oto. 'tile; a1toPllttKlle; d cuvtKrov oOYllato>v, 06~al autov a1tOPllttKoV dvat, tOte; J.lEvtOt yc £uv to. IIAato>voe; 1tapEyxopEtV. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 261 laus was 80YJlCl'ttKO<;. Aenesidemus comes to a similar conclusion in his Pyrrhonian Arguments, claiming that the Academics (especially contemporary ones) sometimes agree with Stoic opinions, so much so that they "appear to be just Stoics in conflict with Stoics" (ap. Photius Bibl. cod. 212, 169b-170a). Aenesidemus and Sextus cer­ tainly were not alone in interpreting the Academic practice of epoche dogmatically,86 even if they have the most obvious motives for doing so. The difficulty with this interpretation, of course, is that it turns Arcesilaus into a negative dogmatist who inconsistently exempts his own ethical recommendation from his practice of epoche. But we have Cicero's testimony (Acad. 1,45) that Ar­ cesilaus departed from Socrates in acknowledging that he does not even know that he knows nothing: Arcesilas negabat esse quid­ quam quod sciri posset, ne illud quidem ipsum, quod Socrates sibi reliquisset. Since this move presumably is intended to forestall the charge that the skeptic claims to know nothing in contravention of epoche, a different interpretation of Arcesilaus' philosophical prac­ tIce seems necessary. In recent years, scholars have favored a view first advanced by Pierre Coussin,87 who holds that all the known arguments of Ar­ cesilaus work from Stoic premises and are ad hominem attempts to show the Stoic that, by his own principles, he had better suspend assent. In the case of epoche, Arcesilaus couples his argument that there is no cognitive impression (see supra 244) with the Stoics' view that the sage will never assent to a non-cognitive impression (Cic. Acad. 2.77), producing the conclusion that, "according to the Stoics," the sage suspends assent (Sext. Emp. Math. 7.150-58). This argument is evidently ad hominem in character, and the same may be said of Arcesifaus' practice of epoche in general. In recommen­ ding that one suspend assent, Arcesilaus most probably is not advancing a dogma in his own name, but merely undertaking "to articulate the views which guide his behavior ... as it were, giving an 86 Numenius ap. Eus. Praep.Evang. 4.15, 7.14; August. Contra Acad. 2.5.11; cf. Cic. A cad. 1.45, 2.28f, 2.59, 77f. For the sense of 06Yfl(l in skeptical texts, cf the evidence assembled by J. Barnes, "The Beliefs of a Pyrrhonist," PCPS N.S. 28 (1982) 1-29, with the discussion of Tarrant (supra n.2) 29-32. 87 "Le Stoicisme de la Nouvelle Academie," Revue d'histoire de La philosophie 3 (1929) 241-76 (English translation in Burnyeat [supra n.2] 31-63); also "L'origine et l'evolution de l'E1toxlj," REG 42 (1929) 373-97. The available alternatives for interpreting Arcesilaus' advocacy of epoche are well canvassed in Long and Sedley (supra n.27) II 446-48. VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 262 THE EPICUREAN REFUTATION OF SKEPTICISM autobiographical report, without taking a stand on the truth of his views. »88 Now, this interpretation of Arcesilaus' practice of epoche is not uncontroversial. In particular, Anna Maria Ioppolo and Julia Annas have recently claimed vigorously that Arcesilaus argues from certain Stoic premises because he accepts them as true, and that, accordingly, he has a positive case to make for epoche. Their case has undeniable attractions, though in my opinion it is not sus­ tained by the evidence. 89 In any case, Colotes does resort to the tactic of branding his Aca­ demic opponents as dogmatists by characterizing their advocacy of epoche as itself an inconsistently held dogma. Of course, the Boy­ ,.uua in Colotes' title by itself implies that he construes Arcesilaus' practice of epoche dogmatically. This strategy enables him to in­ sinuate that Arcesilaus' deeds do not match his words, and he makes his point through a series of cutting puns. Plutarch's com­ ment (Adv.Col. 1121E) that Colotes was grieved by Arcesilaus' reputation (doxa) may preserve a suggestion by Colotes to the effect that one could not have acquired such doxa (to which Co­ lotes alludes elsewhere: 1124 B) without holding doxai-in which case Arcesilaus refutes himself, since he accepts that the sage has no (mere) doxai (Cic. Acad. 2.77). This suggestion is confirmed by the immediately ensuing remark, attributed to Colotes, that Arcesilaus, while he said nothing of his own (a reference no doubt to his ad­ vocacy of epoche), gave the unlettered the assumption and belief (doxa) that he had-thus his words and actions do not match. Plutarch defends Arcesilaus against the charge of being fond of any reputation (doxa) for novelty by citing certain sophists who claimed that Arcesilaus attributed the "doctrines of suspension of judg­ ment» ('ta 'tf\~ E7tOXf\~ BOYlla'ta) to his predecessors: so, Plutarch concludes, thanks are due to those who pass along the pedigree of Academic reasoning (l122A-B). Thus Colotes characterizes Arcesi­ laus' advocacy of epoche as a dogma in order to impute to him an inconsistency between his views and his conduct. Unfortunately, Plutarch preserves only two traces of the polemic Colotes directed against Arcesilaus, and so leaves it unclear why 88 Frede (1984 [supra n.2]) 264. 89 Their case relies principally on two texts, Adv.Col. 1122A-D and Sext. Emp. Math. 7.150-58, both of which seem to me clearly anti-Stoic rejoinders; moreover, I have already argued that we cannot confidently attribute the former passage to Ar­ cesilaus himself (see supra n.54). VANDER WAERDT, PAUL A., Colotes and the Epicurean Refutation of Skepticism , Greek, Roman and Byzantine Studies, 30:2 (1989) p.225 PAUL A. VANDER WAERDT 263 Colotes chose to interpret his practice of epoche as itself an incon­ sistently held dogma. We quoted the first charge earlier (supra 232): Colotes asks why the proponent of epoche does not dash off to a mountain rather than a bath, or walk to the wall rather than the door when he wants to go to the marketplace. This seems to be a straightforward claim 90 that falls well into the pattern of the apraxia arguments we have examined. To judge from Plutarch's reply, Colotes holds that such discrimination requires assent, so that the skeptic either must assent to a proposition about a perceptual object in contravention of epoche, or, if he withholds assent, must admit that he is unable to draw the distinctions among perceptible objects that are necessary to live. Plutarch undertakes to disarm this dilemma, on behalf of Arcesilaus, by restricting the scope of epoche to the realm of opinion: calling attention to the Epicurean doctrine that all impressions are true, he claims that sensation alone suffices for action, and that epoche "eliminates only our doxai, and deals with the others [sensations and irrational affections] as is natural. "91 Plutarch therefore needs to show that the elimination of opinion does not reduce the skeptic to apraxia. There can be no doubt that Colotes rejects such a move. For in his second charge he flatly denies that epoche is possible in all cases: "But it is impossible not to give assent to plain evidence" (aAAcl. aOuvu'tov 'to Jlll OUYKUtUti8E09ut 'to'i~ £VUPYEOt, 1123A-B).92 His 90 DeLacy and Einarson (supra n.3) attempt to find significance in Colotes' choice of examples here, but I find the interpretations proposed far-fetched. 91 Plutarch's response to Colotes' question (1122E-F): 'tou'to £pco't~e; (h:pt~il 'to. aio81]'tTtpux AEYCOV dvat 'lCat 'tae;