Aristotle's Legacy to Stoic Ethics

zyxw zyxwvu ARISTOTLE’S LEGACY TO STOIC ETHICS by A . A. Long I zy zyxwvu zyxwv In his life of Zen0 Diogenes Laertius (vii, 2; 25) makes the founder of Stoicism a pupil of Crates the Cynic, Stilpo the Megarian, Xenocrates and Polemo of the Academy, and Diodorus Cronus. The same teachers, except Diodorus, are mentioned by Numenius (Stoicorum Veterum Fragmenta = SVF i, 11) and Strabo and Cicero also cite Polemo (ibid. 10 and 13). Tradition does not credit Chrysippus with such a varied formal education, but he did apparently go beyond the Stoa to hear Arcesilaus at the Academy (Diog. Laert. vii, 183-4). No ancient authority mentions the Peripatetics, Theophrastus, Strato and Lyco, as having any direct influence on the early Stoics. Plutarch (Comm. not. 1069e) asserts that Zen0 agreed with Aristotle and Theophrastus, a s well a s Polemo and Xenocrates, in taking q6uis and ~b K ~ T &p6uiv a s the ‘elements of happiness’. This enigmatic remark, which I will attempt to explain later, is the nearest Plutarch comes to suggesting a Peripatetic influence - and a shared one at that - on the Stoics, though Aristotle i s mentioned several times in his antistoic treatises. Cicero, on the other hand, cites Carneades (De fin. iii, 41) for the view that only terminology distinguished Stoic ethics from that taught in the Lyceum, and Piso, the spokesman for Antiochus (ibid. v, 74), claims essential agreement between the ‘Old Academy’ and the Stoics, after expounding a system allegedly based on Aristotle and Theophrastus (ibid. 9-13). But the polemic of the sceptic and the over-simplification of the eclectic have been sufficient grounds for discrediting these statements, though it remains to ask why they could have been made at all. What then do we say about the antecedents of Stoicism? For Zeller, Socrates and the Cynics had the primary claim to influence Stoic ethical theory. Aristotle inspired much in logic and physics, but his influence on ethics i s “restricted to the formal treatment of the material. . . and the psychological analysis of individual moral faculties” .2 We should look rather to Polemo and Xenocrates. Professor Brink has recently given somewhat similar instruction^.^ Unfortunately, the ethical theories of these Academics are desperately elusive. In most cases they are cited by eclectic sources not for independent moral positions but for positions which they shared with the Peripatetics. Von Fritz’s attempts to explain away the references in such passages to Aristotle and give Polemo the credit are not c ~ n v i n c i n g . ~Strangely enough, he omits Cicero De fin. iv, 45, Polemone. . .a quo quae essent principia naturae acceperat (sc. Zeno). On the basis of this evidence Philippson and Brink reasonably concluded that Polemo influenced the Stoic concept of -rrpGjTa K ~ T &q 6 0 1 v . ~ Not even so much can be said safely about Xenocrates. zyxwvuts zyxwvutsrqp 12 zyxw zyx Partly perhaps because of this paucity of evidence a racialist theory about some of the influences on Stoicism was born. I do not know who was the first to build hypotheses on the semitic origins of the early Stoics, but Grant in h i s edition of Aristotle’s Ethics (1874, p. 307) speaks of them a s “established”, and Pohlenz built on the theory.6 It is not clear that anything in Stoicism requires exotic sources, nor that such sources actually explain anything, and to-day they are probably quite discredited. Now one can read in Edelstein that Stoicism belongs within the tradition of Greek philosophy, and it is gratifying to find him opposing Bevan’s extravagant claim that “ i t was a system put together hastily, violently, to meet a bewildered ~ o r l d ” . ~ But Edelstein has not developed his points, and like many scholars finds the differences from Aristotle more marked than the resemblances. I am persuaded that the direct influence of Aristotle on Stoic ethics has been greatly underrated. It cannot be proved by documentary evidence, and if proved it would not rule out the important influence of Heraclitus (and perhaps Empedocles), Socrates and the Academy, and the Cynics. But I believe it can be shown that Stoic ethics owes more than some of its formal treatment and terminology to Aristotle; that the Stoics borrowed many concepts, sometimes altering their language, from the Peripatetics, and consciously developed or diverged from others. The subject i s too large to be treated fully here, but I hope to prove the general point by discussing a selection of concepts in detail. This is not merely a matter of Queffenforschung. If the thesis i s sound, Stoic ethics cannot be completely understood without reference to Aristotle, and this might suggest i t a s a worthier subject than modern English scholarship has been inclined to admit. zyxw zy zyx zyxwvu zyx In seeking to establish Aristotle’s influence on Stoic ethics I recognize that a place should a l s o be assigned, perhaps a large one, to Theophrastus. Indeed, those scholars like von Arnim 8 who have looked to the Lyceum for some interpretation of Stoicism, have concentrated on him. T h i s , I think, is a mistaken approach. It i s probable that the Stoics knew Aristotle both through his own works and a l s o through Theophrastus, but we have no sound means for establishing Theophrastus’ independent contribution to ethic^.^ If Arius Didymus’ Epitome and the Magna Morafia (abbreviated a s A M hereafter) derive ultimately from Theophrastus, the form in which 10 we p o s s e s s these works shows unmistakable traces of Stoic influence and Stoic criticism. Both present, a s ostensibly Peripatetic, doctrines which are not paralleled in the Nichomachean restriction of &PET( to moral virtue and the theory of a &DOS p‘los or Eudemian ethics (e.g. ,\!XI’S in Arius) but they cannot be safely fathered on Theophrastus, and are relevant more to the middle than the early Stoa.” In the present state of knowledge there i s little reason to think that Theophrastus was a noteworthy innovator in ethics. He seems to have conceded more weight to externals than Aristotle, and thereby earned criticism from Antiochus.12 But to establish a positive Peripatetic influence on early Stoic ethics we must go to the text of Aristotle. A possible further objection may a l s o be mentioned now. It could be argued that since so much of our information about the Stoics derives from eclectic sources who were influenced by Peripatetic terminology, correspondences between the two systems may merely s e e m close. By the time of Antiochus both Stoa and Lyceum had influenced the other, and controversy was limited to topics such as the a h h p u i a of &PET{ which even eclecticism did not reconcile. Cleanthes’ Hymn to Z e u s , perhaps the longest piece of unadulterated Stoic writing which survives, looks so different from the summaries of ethics in writers like Stobaeus and Clement of Alexandria that we hardly seem to be dealing with the same system. But that hymn looks equally strange when s e t against some fragments of Chrysippus. It displays a religious feeling which i s not recalled before Epictetus and Marcus, and some of its concepts seem scarcely compatible with evidence for other early Stoics. The correspondences with Aristotle which I will seek to show are major concepts which do not rely for their demonstration on purely verbal parallels. 13 zyxw zyxwvutsrq zyx zyxwvuts We may begin with some fundamental (or apparently fundamental) differences in order to s e t the discussion in perspective. Stoic ethics i s based upon understanding cp6ai~, a category both factual and moral which presents itself as the objective standard for moral action: the Stoic ought to grow into a good man. Aristotle’s ethics prescribes the application of q p 6 v r p i ~ to those situations which provoke pleasure and pain: the Peripatetic learns to be a good man by habituating himself to be pleased and pained by those situations which a perfect man’s practical reason prescribes. The Stoic concept of cpdois, which drew inspiration from the Cynics, backed a moral theory valid for all men in any social environment. Aristotle wrote for civilized Greeks, and this limitation i s seen in his account of the particular virtues, though i t does not, I think, affect the conceptual framework of his moral theory in general. T o be irrational in Stoicism i s to a c t viciously, contrary to reason. For Aristotle the moral virtues are dispositions of the irrational, emotional self. There are other major differences. But now we must s e t against them some evidence which makes the antithesis l e s s sharp. I1 E3baipouia and external goods One of the fundamental differences between the Aristotelian and Stoic accounts of E36aipovia i s thought to lie in their different assessments of external goods. The Stoics prided themselves on the a3TdrpKEia of their &PET(, while Aristotle found the moral life l e s s adequate a s a source of happiness than the contemplative life, partly because of i t s greater dependence on externals. When however we look closely a t the two systems the differences, though persistent, appear l e s s clear-cut, and i t becomes reasonable to ask whether the Stoics have been partly prompted in their different thesis by Aristotle. zyxw The passage in Aristotle most sharply distinguished from the Stoic appears in the Rhetoric (i, 1360b 14-29). There Aristotle offers, almost casually, a number of candidates for E\jbaipov\a and finds that a l l require as their parts such external goods as ECyh1a, Trohucpih’la, I T ~ O ~ T O iSt c . , as well a s bodily goods. Now comes the interesting remark: “hence a man would be a3-rapdoTa-roS if he possessed both internal and external goods.’’ The argument of Nichomachean Ethics (abbreviated a s EN hereafter) x (1177a27 ff.) about the superiority of BEwpla to moral action shows the same attitude. Claimants to E36aipouia must satisfy, a s far as possible, the condition of a h h p u i a . Aristotle of course fails in h i s quest for absolute a3T&ptcEia since even the &WPTlTiKbS pi05 i s not entirely self-sufficient. He recognizes an a s p e c t of 6o6hcia in human c p 6 u 1 s . ~ ~The Stoics succeed, because they insist on finding a solution. But that does not show, a s some writers tend to imply, that a\TIT&puia i s a mainly Cynic-Stoic ~ 0 n c e p t . l ~It i s implicit in Plato and explicit in Aristotle. zyxw The b a s i s of this argument r e s t s on what the Stoics call T& KaTh q6oiu and what Aristotle c a l l s T& CKT& &yae& or T& Cr.rrhBs &yaedr.15 Whatever the influence of the Cynics may have been,.it i s essential to remember that Aristo, the most cynic of Stoics, was labelled a heretic j u s t because he insisted on no concept of value outside PET^ and K a K i a .16 For orthodox Stoics T& KaTh cp6oiv include and develop from the objects of instinctive ch0ice.l’ Their value i s natural, but not moral, and to p o s s e s s them i s no guarantee of E36aipov)Ia. They comprise health, wealth, mental and physical advantages, and indeed form the & p x i of KaefiKovTa, misKaefiKovTa are actions involving love of self, parents, fellowmen etc., called ‘duties’.’ based on natural and biological affiliation.” Now KaI3fiKovTa are not in the strict s e n s e ‘moral actions’; but their performance, on the promptings of a consistent, rational disposition, i s moral action.20 And the Stoics pointed the distinction between the two types of action by making 74 zyxwvutsrqp zyxwvutsrqp zyxwvutsrq zyxwvutsrqp i b Kai& p h i v the Ghq, not drpx:, of the latter (see note 18). Indeed, later definitions of the TEAOS are formalized in terms of “rational selection of ~h KaT& qdaiv and rejection of their opposites”. 21 I say ‘later definitions’. But it may be argued that these definitions are a product of the criticism which the Stoics encountered in the second century B.C. I have attempted to show elsewhere that “selection of ~h KaT& q h v ” was implicit, if not explicit, in Chrysippus’ definition of the T ~ O .22 S And it i s not a concept which requires reference to later, external influences on the Stoa. We have Plutarch’s testimony, a s I mentioned earlier (p. 72), for Zen0 making 9 6 ~ 1 s and ~b K a T h q6uiv the moip’iu i i j ~ EOGaipovinS, a position he i s said to have shared with the Academics and Peripatetics. The meaning of this statement i s not found in Plutarch (it occurs j u s t after his quotation from Chrysippus about the drpxi and Ghq of KuefiKovia and &(PET{), but it is surely illustrated by Aristotle’s discussion in EIV i , 9. zyx zyxwvu Aristotle i s considering the relation between T& E K T ~ S dryuedr and E36aipovia. Some, he says, are necessary a s instruments (Kuf3hmp 61’ 6pydrvwv) such a s friends, wealth, political power; others, such a s noble birth, disfigure happiness if they are absent (1099a31-1099 b 8). Again we seem to be a long way from the Stoic position. But a s he proceeds Aristotle widens the gap between E36uipoviu and E3iuxia. Living E; or KaKcSS does not depend upon the favours of fortune but on €vipyEiai K U T ’ & P E T ~ V(1100b8 ff.). The gifts of fortune are rather cosmetics which naturally add to or detract from ~36aipoviu(1100b25ff.). “If”, s a y s Aristotle, “it i s actions which determine the quality of a life, the truly good or wise man will never become unhappy, but will always make the best use of the materials available (T& 6ndrpXovTa) like the good general and the good cobbler” (1100b33-1101a8). Here Aristotle is distinguishing between the goods which are necessary a5 instruments for performing good actions (friends, wealth etc.) and the gifts of fortune (noble birth, beauty erc.). He seems to argue that a man deprived of both cannot be happy, but changes of i6xq in later life will not easily or not necessarily affect the good man’s Elj6uipovia. This i s a different concept from that offered in the Rhetoric, and a different one again i s given in the Politics (vii. 1323b 21 ff.). There E36aipovia and E3TuXia are said to be “necessarily different”. A man’s E36atpovia i s a product of his qp6vqaiS and -witness God, who without any external goods i s happy 61 ’ i .rroi& T i 5 6 v a i T ~ 9V 6 a i v . The best life i s one furnished sufficiently to a h b v aljibs ~ d T+ provide the means for virtuous action ( b MET’ d r p ~ ~ ~i j~~x o p q y p k vEqn‘~i T O U O ~ T O VGUTE~ E T & E I V TGV KUT’ drpETjV TTphSEWV, 1323b40-1324a2). 23 zyxwvutsr Taking EN Book I and Politics V I I together Aristotle’s position seems to be this: there i s a necessary connexion between E36aipov’la and T& E K T ~ S dryaedr since the latter are materials or instruments for the exercise of d r p ~ i i ; but there i s no necessary connexion between Ed6aipoviu and the gifts of fortune since one can be unhappy with them and partly, if not completely, happy without them. zyxwvutsr Is there a necessary connexion between i h KuT& q h v (so far as these approximate to ~h ~ K T dryue6) and d r p ~ in ~ i Stoic theory? The answer, I think, i s an unequivocal ‘yes’. T o perform a KuTbpeopa, a morally right a c t , entails exercising qpbvqais in the sphere of ‘natural advantages’, i.e. knowing when to s e l e c t l r e j e c t them e t ~ In . this ~ ~ s e n s e i h KaT& ~ ~ U might I V be called the d v 03, &JEW of moral action, a s indeed they are by Clement ( S V F iii, 114), but they are required only as Ghq, not a s things which the agent needs to p o s s e s s for himself (as, so to say, gifts of fortune).25 According to Cicero (De f i n . iii, 60) it is because T& K U T q~ h v form the materiu 75 ~ S zyxwvutsrqp zyxwvutsrq zyxwv scipieritiae that the decision to live / d i e must be determined by reference to them. A man would only die if his E36aipovia were endangered, and the good man’s ~ 6 6 a i u o v i acould only be endangered if his exercise of &PET$ were uncertain - if he ran out of materials. We should a l s o note that the Stoics did not define happiness a s an Evlpyaia but a s a concomitant of the virtuous disposition.26 Two conclusions emerge from all this, and both relate to Aristotle and ah&pKaia. For t h e a h & p w i a of E68aiVovia to be guaranteed it i s necessary to show that i t cannot be assailed by anything. The Stoics satisfied this requirement by making E36aipovia an internal state. But aO6aivovia must be defined in terms of something and the Stoics followed Plato, Aristotle and the Cynics in defining it in terms of drp~-ri, Stoic PET^^ i s purely moral; it requires a sphere of action. Aristotle found this in the situations which provoke pleasure and pain; the Stoics found i t in the objects of our natural affiliation and aversion - in concrete terms, health, wealth, friendship etc. and their opposites. Aristotle’s good man requires these in order to exercise his &PET{. The Stoic does not require any one of them for himself, but he too cannot exercise h i s &PET{ in isolation from them. And this fact led Alexander of Aphrodisias (SVF iii, 64) to comment, not entirely unfairly, that “Stoic &PET{ i s not self-sufficient since i t requires a sphere of action external to itself”. But the Stoics tried to avoid this problem by making a distinction between ‘possession’ ( T U Y X & V E ~ V ) and ‘selection or adoption’ ( ~ K ~ ~ Y E I V , ha@&vEiv). The good man aims not a t possession of natural advantages, but to display d r p ~ ~ i in selecting them; and to avoid confusing E3Tuxia and ai16aipovia the value of ~ b KaTh : q)hv i s sharply distinguished from the value of a moral action. It i s hard not to s e e the Stoics going consciously beyond Aristotle here, and this i s equally true of their treating a36aiuovia a s a concomitant of the virtuous 6ibeEaiS rather than an Cvbpyaia. Even the good man cannot be active all the time. Both are moves to safeguard ah&pwia, but the conceptual framework of the two theories i s very close. When Aristotle s a y s Giah&pmi ~b Kah6v in situations where a good man bears many great misfortunes lightly (EN i, 1100b 30 ff.) and a s s e r t s in the Politics (vii, 1323b 18-20) that T& ~ K T and ~ S bodily goods ~ t j sywxtjs i‘velav T ~ ~ I U K E Va ‘ i p d , he i s ta!king zyx zy in different contexts but in Stoic terms. Aristotle found himself unable to formulate an entirely consistent position with regard to a66aiclovia and T& E K T ~ S &yaeh, and his common sense stopped him short of the Stoic standpoint. But he prefigures much in theirs, and one can s e e why some ancient critics saw superficially little difference between the schools on this issue. me$KovTa and KaTopedpara compared with Aristotle’s distinction between actions leading to and those proceeding from a virtuous disposition zyxw zyx The terms Kae{KovTa and KaTopedpaTa are Stoic. Neither in itself reflects exactly a conce9t found in Aristotle, but the distinction between the terms, with respect to judgments about moral action, corresponds very closely to something in EN i. I noted a moment ago that zyxwvuts T& K ~ T &q)iraiv form the &px{ of KaeiKovTa and the iihq of &PET{. i s manifested in right actions, which are KaTopOhpara. We a l s o saw that Aristotle makes external goods the instruments of right action, though he does not base h i s argument upon a concept of biological affiliation, like the Stoics. Since the performance of certain KaefiKov-ra i s natural to man (and indeed to animals and even plants) irrespective of his moral s t a t u s , it follows that they denote a category of action wider than ~ a ~ o p e h p a ~AaKaefiKov .~~ is an action “in conformity with life’’ ( ~ & bK ~ ~ O U ~ iOv VT?J [wq,Diog. Laert. vii, 107); something which “when performed can be reasonably defended” Cd rrpax8kv ~GhoyovI q a i &-rrohoywp6v, ibid.), though not necessarily (we may add) by the agent himself.28 Thus we can say that i t i s reasonable in general that men should look after their health, their children, their parents &PET$ 76 zyxwv zyxwvutsrq zyxw zyx e t c . , but they are not to be commended in a strictly moral sense merely for doing so. What then is a moral action? It i s an action performed by a virtuous agent. That, however, does not get u s very far, and the Stoics elucidated the concept by reference to K a e f i K o v T a . Right actions are perfect or complete(d) KaefiKovTa ( T i k i a or TEkiw8ivTa), KaefiKovTa “possessing all numbers”.29 What perfects a K a e q K O V is its performance 6.rrb qpovjmos or sapienter. 30 We may say, I think, that the good man performs K a O i K o v T a not because, or not merely because, he has an instinct or training to d o so, but because his q p 6 v q u i S tells him so to act. Unlike the man who merely performs K a O G K o v T a successfully, the good man performs them with his ~~ eye on a U K O T ~ ~ Sa, fixed goal or plan of life, and in accordance with a K a e q K O V h 6 ~ o v .Clement, who states this, goes on to show that i t involves possessing yvcjuis and knowing how one should a c t throughout the whole of life. The wise man’s actions are good not because they have a certain look about them or because they achieve certain external results (whether they do SO i s quite contingent), but because they follow from a consistently rational disposition ( h b Ka‘i 6 i a e h o s ~ O h o y i u ~ o v ) .They ~ ~ are called T E ~ I K && y a e d c because they are parts of PET^ and ends in t h e m ~ e l v e s . ~ ~ 8 0 t h KaefiKovTa then and Ka-ropehpaTa are natural actions, but the latter are peculiarly, though l e s s universally, human, since they are the product of the peculiarly human quality, q p b v q u i ~ . By the evolution of his A6yo~the Stoic intuits, or extrapolates, from T& KaTh q h v , which he instinctively pursued a s K a e i K o v T a , an ordo and concordia rerum agendarum, a notio boni, which i s the standard and basis of moral action.34 His task i s to develop such a harmony within himself and thus ‘flow smoothly’ through life. Cicero and others imply clearly that the performance of K a e j K o v T a i s a necessary antecedent to the performance of ~ a ~ o p ~ d pIndeed, a ~ a . ~ ~ the man who i s about to earn the title ‘good’ performs all K a e i K o v T a , but h i s disposition still lacks the firmness and infallibility required of the strictly moral agent.36 It appears then that the performance in Stoicism of natural, but morally neutral, actions should lead to the performance of natural moral actions if the former lead the agent to grasp the principle which makes them appropriate. The morally neutral actions are primarily instinctive or taught, and judged by external standards. They can be specified. The moral actions are thought out and fit a definite plan of life. They are judged by reference to the agent’s disposition and intention. We have no detailed lists of specific K a - r o p O h V a T a . zyxwvut z zyx I turn now to Aristotle. At EIV ii, 110Sa17 ff., Aristotle faces an aporia. If men become j u s t by performing j u s t actions must they not already be j u s t (sc. in order to perform those actions)? No. A man might get something in grammar right by accident or assistance, but this would not entitle him to be called y p a p p a T i K 6 S . His grammatical ability i s something internal ( K a T h T ~ V i v a k + ) . And this points the difference between &pE.rai and T k X V a i . Works of art contain their excellence in themselves. We do not make inquiries about the artist’s disposition in making judgments about them. We are concerned exclusively with what the works of art are like (vos ZXovTa). In the c a s e of moral judgment we proceed differently. Here it i s the agent’s disposition (Cbv 6 T T ~ ~ T T WT&S V Zxov T P ~ ~ T T ~not J ) the look of the action which determines its justice etc. Aristotle now (1105a31 ff.) stipulates three conditions of moral action which explain his s t r e s s on the agent’s disposition: 1. A man must a c t E X ~ SBy . ‘knowingly’ Aristotle means ‘practical knowledge’ ( q p b v q u i ~ ) of the right goals and how they are to be achieved. The wicked man does not know what he ought to do (1110b28), and the involuntary wrong-doer gets the wrong view of a particular situation (1110a33 ff.). Making moral mistakes in planning one’s life (.rrpoor’lpEuis) produces p o X B q p i a (1110b 32). We have Plutarch’s testimony ( S V F i , 201) for Zeno equating h i m f i p q , in terms of which the Stoics defined PET^, with q p 6 v q u i S . 77 zyxwvuts zyxwvutsrqpon zyxwv zyxwvutsrq zy 2. A man must a c t Trpoaipo6yvo5, Ka‘i Trpoaipo6pEvos 6 i ’ a h & . The Stoics did not draw Aristotle’s careful distinction between wishing the end and deliberating the means, but here Aristotle probably means that moral actions must be chosen for themselves because it is only by such actions that E66aipovia, the goal of all action, can be actualized. For the Stoics too virtuous a c t s are T E A ~ K &&ya0& (Diog. Laert. vii, 96) since they are the activities of a virtuous disposition and manifestations of E 6 G a i p o v i a . A Stoic would a c t deliberately for the sake of the action because all h i s actions are the result of a settled plan. 3. A man must a c t kpaiws Ka‘i & w T a K i v q - r o s 2xxwv. This condition i s entirely Stoic. Both Aristotle’s and the Stoics’ good man must p o s s e s s an inflexible moral character. p‘&xios i s a common Stoic term, and along with & ~ E T & T ~ T W T OitS characterizes the virtuous 6uXkaiS (Plut. D e virt. mor. 441c). &~ET&TTTWTOS i s equivalent to Aristotle’s & p E T a K i v q T o S . The Stoics then accept all these conditions of the virtuous agent, though their precise s e n s e in Stoicism must take account of the all-embracing concept of q h s . Aristotle goes on to show that men become j u s t by repeatedly performing just a c t s . Actions are culled just when they are the sort of actions performed by the j u s t man (1105b 5 ff.). But a man i s just not in virtue of performing them, but only if he performs them in the way that the good man does. Aristotle here has told u s briefly what it i s to be a virtuous agent, but h i s detailed discussion of q p 6 v q a i S i s reserved for Book VI. In the light of his claim that a man must perform j u s t actions in order to become just it i s logically correct for Aristotle to consider what such actions are before establishing the intellectual element which i s their basis. We s e e then that the Stoics agree with Aristotle that men become virtuous by performing certain actions. But they do not claim that these preliminary actions are even homonymously virtuous. Actions performed by non-virtuous agents are necessarily not-good since the particular disposition (nos i x o v ) of the agent i s the sole criterion for judging the moral status of h i s actions.37 What i s required of the good man i s not infallible performance of Ka0iKovTa, but an infallibly virtuous disposition. To attain this, Ka0iKovTa are necessary since their consistent performance can lead to the grasp of bpohoyia, the principle of moral action. Aristotle does not, I think, make explicit how consistent performance of virtuous a c t s leads to the qualities required of the qpbvipos. Repetition and developing the right feelings are the means, but it i s not clear (to me) how they produce the necessary intellectual conditions. The Stoics, by their concept of an evolving logos, show how the O K O I T ~ Srequired by both moral systems comes to be grasped.38 What the Stoic qp6vipos takes a s his U K O T T ~ Si s not of course ~b &v0pdTrivov &ya06v, conceived a s a mean between two extremes, but q6ois in the largest s e n s e , and therefore what a Stoic actually does i s fundamentally different. But qp6vquis in both systems consists in knowing and consistently doing what one ought to do, and i t i s acquired by consistently behaving in a certain way. The differences between Aristotle and the Stoics derive from different concepts of E66aipovia, p.?ai~,PET^, not from different ways of analysing moral action. (Having the right feelings, which I have only barely mentioned, i s a l s o common to both accounts, a s I will shortly show.) The gulf between theory and practice in Stoicism, though strictly nil, i s wider in fact than in Aristotle because the theory i s so difficult to fulfil. When later Stoics tried to bridge the gap their modifications brought them closer still to Aristotle. Both Panaetius and Posidonius may have conceded some positive value to external goods.39 Both again accepted the existence of an irrational faculty in the human yuxi. One may surmise that part of the rigidityof early Stoicism stems from the adaptation of an Aristotelian framework to un-Aristotelian premises. zyxwvut zyx zyxwvutsr zy To conclude this section, it i s important to notice that the distinction expressed by K a 0 f i K O V T U and KaTop0dpaTa could not be explained in Platonic terms a s a distinction between true belief and knowledge. The performer of mf3iKovTa does not perform the same actions a s the good man, 78 zyxwvut but fall short of the latter’s ability to account for his moral principles. hloral actions in Stoicism are distinguished by the agent’s disposition; hence only external appearance could strictly be shared by a KaeqKov and a KaT6pewpa. S6<a in Stoicism cannot be right or true because only True belief then, a s John Rist has observed, would be indistinguishable k ~ r i u ~ grasps j~q from knowledge in Stoicism, and S6ca i s actually defined a s weak or false assent.41 Bad men are probably not in a permanent state of S 6 t a since they can perceive truly by KaTrihqyiS, and they can a l s o make true statement^.^^ But their doing so does not entitle them to any share in &hj€kia, which i s firmly restricted to the good man.43 zyx zyxwv zyxw zyxwv zyxwvu zyx zyxwvutsr IV Emotion and Virtue Stoic ethics was calculated to promote &TrdreEia, but this term does not mean freedom from all feeling, or insensibility; i t means suppression of judgments based on false assessments of pleasure and pain (or good and bad). In a recent article, R. P. Haynes has examined the early Stoic theory of pleasure, and he concludes that “the Stoics followed Aristotle in maintaining that the virtuous man i s the man who not only performs acts of virtue but performs them a s the good man would, i.e. with pleasure”.44 Haynes did not make any detailed comparison of Stoic and Aristotelian attitudes on this question, but he was certainly right to point out the connexion between the two schools. I will conclude with some observations on the relation between emotion and goodness in the two systems. After establishing that moral virtue i s concerned with TrpEcis Aristotle a s s e r t s that the pleasure or pain which accompanies our actions must serve a s an index of dispositions (EN ii, 1104b 2 ff.). Pleasure and pain are c a u s e s of bad actions and the moral agent needs to be trained to find pleasure and pain in the right things. Since every ~ r p E c i sor Tr&3oS i s accompanied by pleasure or pain, virtue takes these a s her material. Aristotle’s insistence that the moral agent must have the right feelings about his actions i s rightly related by David Furley to his discussion of are motivated by the psychology of action in the’De anima and De motu a n i m a l i ~ m . Actions ~~ desire ( 6 p t i ~ or ) choice (TrpoaipEuiS) because some change has occurred through perception or i m a g i n a t i ~ n . ~The ~ perception or imagination of painful or pleasing objects causes changes in the animal’s temperature and a mental reaction to pursue or avoid the object imagined or perceived. 41 Now it i s clearly part of Aristotle’s doctrine that what a man finds painful or pleasant depends upon himself and how he has been brought up. The man of self-control finds pleasure in abstaining from physical pleasures; the courageous man f e e l s pleasure, or no pain, in facing dangers (EN ii, 1104b 5-8). But pleasures and pains are not merely accompaniments of action. They are also, qua pleasing or painful objects, necessary factors in moral choice. 6 p ~ ( i s~ a pleasurable response to the consciousness of something good.48 It i s by pursuing and avoiding pleasures and pains, either the wron ones, or a t the wrong time, or in the wrong way, that men become morally corrupt (ibid. 1104 21-24).49 A little later Aristotle distinguishes three factors which determine moral choice, Kah6v, U U ~ & I Oand V , jS6, but he makes it clear that the first two are not necessarily distinct from the last, t Koivj TE yhp a h q (sc. jSov$ TOIS (+air, Ka‘i ~ E u Ti O ? ~ h b T ~ a‘ipauiv V TrapaKohoueE? Ka‘i yhp ~b Kahbv Ka‘i ~b uupq6pov ?Sb qaiveTai. (1104b34-05a1).50 - In some sense all choice i s concerned with pleasure (or the pleasant), and the morally good and expedient manifest themselves a s j S 6 . The judgment that i t is good to abstain i s pleasurable to the udppov and rouses the dp~{is to suppress his appetites; the pleasure of indulgence prompts the actions of the drK6haUTOS. The psychological distinction between the two types of action 79 zyxwvu zy zyxwvutsr zyxwvutsr zyxwvutsrq zyxw zyxwv zyxwvutsrq r e s t s not on the presence or absence of $6, but on the distinction between rpoaipEaiS and h 8 u p i a or 8 ~ p 6 5 . ~ ~ O ~ Q P O V reaches an object of desire by po6heuoiS which i s Kah6v The and pleasing to him. The ErKbhaaToS pursues the pleasure prompted by h i s appetites and passions. For Aristotle then ‘ 6 0 ~ 4i s essential to r p & < i s and the truly pleasurable is what seems to be s o to the good man!2 Since pleasure and pain are the general standards by which men regulate their conduct (Kauovi[opEv T&S T T ~ ~ { E I ~it) i s fundamental that the moral agent should find virtuous activity pleasurable. Aristotle does not, I think, ever imply that the moral agent chooses to a c t virtuously in order to acquire pleasure. He holds rather that moral actions are inherently desirable, and the virtuous man finds the thought and generally the a c t of performing them pleasurable. When we compare this thesis with the Stoic attitude to emotions and moral action it seems a t first to be totally different. Stoic writers say nothing about pleasure and pain forming the material of virtue. In so far a s they designate rdteq, jS0.4 and h 6 r q are always & h o y o ~contrary , to right reason.54 They are a l s o both described a s 66{ai, a term applied to the emotions in general. i s the false belief that a good i s present, and h 6 q i s the f a l s e belief that an evil i s pr2:z?5 Since the good man knows and never opines, it follows that he never h a s beliefs which can be called jtiovai or ~ O T T C X I .As ~ ~ terms describing irrational impulses pleasures and pains are the signs of KcxKia, and the Stoics could only say that virtue i s concerned with them in the s e n s e that i t is concerned to remove them. j S o u i in itself, however, is not classified by the Stoics a s bad, but i n d i f f e ~ e n t . ~It~often denotes a physical feeling, and as such i t i s a sensation common to a l l creatures. Aulus Gellius ( X A xii, 5 , 7) makes omnibus corporis sui cornmodis gaudere a primary instinct, and Stobaeus ( S V F iii, 136) makes bodily pleasure neither preferred nor rejected. We can probably take i t then that jSov4 in Stoicism i s of no moral significance when it merely denotes bodily sensations, but acquires bad moral significance when it denotes the mental state of those who regard such sensations a s good. Chalcidius, in a long discussion of the Stoic theory of moral corruption ( S V F iii, 229), notes as primary the belief that pleasures are good and pains bad. Many other authorities confirm this. But i t i s not the whole story. jSoufi is a l s o described by Diogenes Laertius in his life of Zen0 a s an imyivvqpa: hriykuuqpa yhp qaaiu, ~i6pa b i u , j S o u j u dual ii-rau a C T j Kae’airTju q6uiS im<qTfiuaaa T& kvapu&ov.ra T ~ JC W U T ~ D E&rohhpq ~ * Bv T ~ ~ T T O ErqiAap6vETai T& @a Ka‘i 8hhhEi T& ~JT&. fi V (vii, 86) This passage naturally prompts comparison with Aristotle’s description of jS0.4 a s something which crowns or perfects activities (EN x, 1174b 31-33), but how is it to be understood in Stoicism? A little later (94) Diogenes Laertius observes that “the good can be defined as the natural perfection of a rational being qua rational, and under this category we may put E r p ~ ~ faii , T T ~ & < E I SKUT’ & P E T ~ ~ Voi , crro6Saioi, and such kriyEvvipaTa a s Xaph and ECqpoa6vq”. It seems therefore that at l e a s t in a derivative s e n s e jSou4 (equals xcrph) i s a good, and this i s certainly the view of Epictetus.’* He distinguishes between physical pleasure and a mental pleasure which i s €6h o y o ~?napaiS and an accompaniment of ~b &ya86v (iii, 7 , 7). Let u s c a l l this accompanying j S o u i Xaph, and consider i t s relation to virtue. zyxwvu U by Stobaeus (SVF iii, 106), i s one of the three so-called Xaph, named a s a T E ~ I K ~&ya86v peculiar to the good man. 5 9 The others are po6hqai~and EGhhpEia, and all three are defined a s mental s t a t e s which are ~ d h o y o s ,though only Xaph i s an iniykvvqpa. ECdd3Eiai 80 zyxwvu zyxwvuts zy zyxwvu zyxwvu zyxwvu zy Diogenes Laertius s t a t e s explicitly that Xaph i s not a permanent feeling (vii, 98), and Stobaeus adds that i t is not a necessary good ( S V F iii, 113). If they are right (and the evidence elsewhere suggests they may not be) the Stoics would differ from Aristotle in not making the good man’s feeling joy a necessary indication of his disposition, though a man’s experiencing the nhf3os of j S o v i would guarantee his badness. But poOAqai~,another ‘good feeling’, i s defined a s ~ G h o y o ~ bp~cig,and 6pEtis forms an aspect of b p p i which i s a prerequisite for all action.60 Galen, on the basis of Chrysippus, defines b p ~ t as ~ sb p p i ~ O Y I K ~ T I 6aov x p i ?$ow ( S V F iii, 463), “rational impulse towards something pleasing, whatever it ought to be”. I am not quite clear how fl6ov i s to be interpreted here, but we get some help from Seneca: ad summa pervenit qui scit quo gaudeat ( E p . tnor. xxiii, 2). This gaudium i s explained a s an internal condition which comes from bona zyxwvutsrq conscientia, honesta consilia, rectae actiones, contemptus fortuitibus, placidus vitae et continuus tenor (7).61 Putting the two passages together we may say that the good man i s one who knows good conscience etc. to be pleasing and desires on the basis of his logos to attain them. This i s not to say that he desires them because they please; he desires what he ought to find pleasing. The Stoic concept of AoyiKi b p p i i s closely analogous to Aristotle’s $ O U ~ E U T ~ K $6pEtiS (EN iii, 1113a9 ff.). I suggested that this 6 p ~ t 1 sin Aristotle i s a pleasurable response to something Kah6v, reached by the good man via /3oOhEuaiS. But the Stoics, I believe, have slightly modified Aristotle here. They subsume /3oOhqaiS under bpEEiS, thus relating ‘wish’ to the particular object desired now, not the long-term objective. And bpEciS ( b p p 4 ) is itself subsumed under auyKad3EaiS, ‘assent’.62 Desiring or being motivated is an activity of the reason which follows from assent.63 The Stoic does not deliberate. He a s s e n t s to or withholds a s s e n t from presentations (qav-raaiai) and propositions (6EidpaTa). His actions, like those of Aristotle’s moral agent, are prompted by CjpEciS, but the Stoic’s bpEcis i s not an irrational faculty under the control of reason, but the logos itself in i t s appetitive aspect.64 Assent in Stoicism entails (perhaps simultaneously) the i s s u e of an impulse or imperative. Aristotle too argues that in moral choice the bpE& must pursue what the h 6 y o ~a s s e r t s (EN vi, 1139a23-8), but his bpEcis, though necessarily associated with judgment, i s not an aspect of h 6 y o ~itself.65 It i s the faculty which responds with a feeling of pleasure to ~b ~ ~ E K the T ~ good V , or apparent good.66 Again, in both systems 6pEtiS (or bppi) i s the means by which the mind gets the body (if we may use this misleading distinction) to execute its decisions; but in Stoicism, which wants to exclude the irrational at all costs, 6pEtiS is not a response to a pleasing stimulus but a movement, consciously articulated, towards the object of a ~ s e n t . ~ ’Since the objects of the good man’s a s s e n t s are always valid and entail morally right action xapci, the h i y i v v q p a of & p ~ ~ fisi , associated with them. It i s this perhaps which makes the objects of his bpE<is ‘whatever ought to be pleasing’. If he i s a bad man his a s s e n t s are weak or false, and i s s u e impulses towards the wrong goals. His disposition expresses itself in TrbOq, s t a t e s of pleasure and pain, or false beliefs that he enjoys something good, or experiences something bad. Enjoyment in this sense, or suffering, are never conditions experienced by the good man; but he generally feels joy in acting virtuously, and he knows the objects which promote this condition, though his actions are never prompted by feelings. The Stoics then agree with Aristotle that the feelings of the good man are utterly different from those of the bad man. And they might agree that some feeling is a normal accompaniment of any action. But it i s not certain that xapb i s a necessary condition of PET^. Seneca a s s e r t s that the good man always f e e l s gaudium (Ep. mor. lix), but Diogenes Laertius denies the permanence of Xapdr. One may be wrong, or the Stoics may have differed on the point. At least i t seems clear that they have given it a different emphasis from Aristotle, though their position i s very close to his. And the Stoics would certainly deny that the presence of xapb perfects happiness. It i s something like an aesthetic quality which the good man normally displays. Nor again does pleasure or the pleasant enter into the Stoic’s moral choice. Having assented to (i.e. chosen) the morally good the Stoic commands himself to pursue this pleasing object. Unlike Aristotle’s moral agent the Stoic does not need the objects of his moral choice to present 81 zyxwvutsr zyxwv zyxwv zy zyxwvu ttiettiselves U S pleasing, but he does, a s a result of his xapb, find what he actually pursues a source of joy. The Stoics have excluded all factors except 6pObs h6yos from moral choice, but they permit the agent to derive some emotional satisfaction from his actions. Since the Stoics differ fundamentally from Aristotle in regarding moral virtues a s dispositions of a rational soul, not an irrational faculty, these modifications of his doctrine of pleasure are necessary and intelligible. What i s remarkable i s the closeness of the two accounts. The Stoics have eliminated pleasure a s a factor in moral choice, but they have retained it a s an accompaniment and description of the virtuous actions which are the objects of the good man’s Bpeg15. To avoid confusion they have a l s o distinguished this pleasure from i t s pathological counterpart by different terms, but they accept that how a man feels i s a normal, if not necessary, indication of his moral status. V No philosophical system is a creation ex nihilo, and Stoicism i s more derivative than many. The indebtedness of Stoic ethics to pre-Aristotelian influences i s generally acknowledged, but i t does not rule out the direct influence of Aristotle himself. Time and again in Stoicism we meet the technical vocabulary which he established. In itself this does not prove that Zen0 and his successors had read particular works of Aristotle, though we can hardly suppose that in Athens of a l l places they were not generally acquainted with his views. I have tried to show that the Peripatetic influence on Stoic ethics goes much deeper than this. My treatment has not been exhaustive, but it h a s , I hope, been sufficient to establish a close rapport on three basic moral doctrines. Considering the unsatisfactory nature of our sources for Stoicism it i s surprising now precisely we can focus this relationship. The Stoics accept certain basic distinctions, first clearly drawn by Aristotle, and in some places they appear to be consciously correcting or modifying his arguments. These points of agreement or divergence belong to the core of both systems. They show that Stoic ethics belongs in important respects :o the mainstream of Greek philosophy. zyx University College London NOTES zyxwvutsr zyxwvut zyxwvu 1 This is an annotated version of a paper delivered to the Southern Society for Ancient Philosophy at Cambridge, September 1967. 2 Philosophie der Griechen (revised Wellman, Leipzig 1909) iii, 1, pp. 368-9. Some of these formal correspondences are noted by Dyroff, Die Ethik der alten Stoa ( B e r l . Stud. NF 2, 1898), especially pp. 16 f . , 88 ff., 249 ff. 3 “Theophrastus and Zen0 on Nature in moral theory”, Phronesis 1 (1956) 142-45. 4 RE S.V. Polemon, xxi. 2, cols. 2524-2529. 5 Philippson, “Das erste Naturgemtisse”, Philologus 87 (1932) 447-50; Brink, o p . cit., 143. 6 Die Stoa3 (Gottingen 1959). especially pp. 107, 134 f., 164 f . For a criticism of Pohlenz in this respect cf. the reviews of Edelstein, AJPh 72 (1951) 427 f.; Sandbach, J H S 71 (1951) 262. 7 Edelstein, The Meaning of Stoicism (Camb. Mass. 1966), p. 18; Bevan, Stoics and S c e p t i c s (Oxford 19131, p . 3 2 . 82 zyxwvutsrqpon zyxwvuts zyxwv zyxwvutsr zyxwv 8 From h i s examination of Arius Didymus’ Epitome of Peripatetic Ethics ( S i t z b . Wien, 1926), pp. 157 f f . , von Arnim concluded that many major Stoic moral principles derived from Theophrastus. His t h e s i s h a s been forcefully rejected in general by Regenbogen (RE S . V . Theophrast, Suppl. vii, 1940, col. 1492) who showed that it r e s t s on the improper assumption that a l l major divergences from Aristotle in Arius are due to Theophrastus. T h e epitome i s an eclectic work which combines Peripatetic and Stoic teaching, cf. Walzer, Magna Moralia und A r i s t . Ethik (Neue Phil. U n t e r s . vii, Berlin 1929). Brink, Phronesis 1 (1956) 123-45,h a s recently rejected von Arnim’s attribution of Stoic o i K E i o u i S to Theophrastus. zyx 9 As sources for Theophrastus’ e t h i c s Regenbogen ( o p . cit.) c i t e s Arius Didymus, the MM and Cic. De fin. v . It s e e m s almost certain that Arius’ source used the MM (cf. Dirlmeier, RhM 88 (1939) 214-43). and the MM i s probably considerably later than Theophrastus (cf. the arguments and authorities cited by D. J . Allan, J H S 77 (1957) 7-11 and now a l s o Gnomon 38 (1966) 142-144,criticizing Dirlmeier’s l a t e s t claims for i t s Aristotelian authorship). The spokesman for Antiochus (De fin. v, 12) claims to follow Theophrastus on most points, but h i s assertion cannot be used a s a reliable b a s i s for deriving Theophrastus’ e t h i c s from Cicero, cf. Madvig’s De finibus, Excursus vii; Luck, Der Akademiker Antiochus (Bern 19511, pp. 31 f., 55 ff. 10 e.g. in Arius (130,21 Arnim), criticism of the Stoic distinction between U K O T T ~and ~ T ~ O S ; rejection of the view that the wise man’s happiness i s &vaTr6pAqTov (133, 11); the u s e of the terms K a B i K o v T a and KaTopedpaTa (145, 6-10). For Stoic technical terms in MM cf. Dirlmeier, Rh.V 88 (1939) 218-228. Allan ( J H S o p . c i t . p. 7) argues that the MM offers a selective version of Peripatetic ethics, perhaps designed to make converts from Stoicism. For Dirlmeier’s new views s e e h i s Aristotelis Magna Moralia (Berlin 19581, 118-46. 1 1 cf. von Arnim, p. 145; on MM’s doctrine of moral virtue s e e Allan, 8-11. I would venture t o suggest against Dirlmeier and others that the dppi doctrine of MM reflects some Stoic influence. zyx 12 c f . C i c . De f i n . v, 12; A c . i, 33; ii, 134; T u s c . d i s p . v, 85. 13 Met. A, 982b29-34; EN x, 1178b33 f f . cf. Schaerer in “ L a Politique d’Aristote” (Entretiens de la Fondation Hardt XI, Geneva 19641, pp. 91 f . 14 J. LConard, Le bonheur c h e z Aristote (Brussels 1948). p. 54, distinguishes two broad s e n s e s of d ~ h p K E i a : 1, sufficiency and independence (Cynic and Stoic); 2, mere sufficiency (Aristotle). Of some significance here i s the anecdote about Zen0 recorded by Stobaeus ( S V F i, 62); Zen0 approved of Crates for finding in a poor cobbler more of the necessary requirements TrpbS ~b q i h o u o q i j u a r than Aristotle had found in the Cypriot king t o whom he wrote the Protreptikos. For a valuable discussion of the pre-Platonic influences on Aristotle’s theory of E66aipovia s e e G. Muller, “Probleme der aristot. Eudaimonielehre”, Vus. Helv. 17 (1960) 121-143. 15 On Stoic ‘natural advantages’ cf. I. G. Kidd, “The relation of Stoic Intermediates to the Summum Bonum” CQ NS 5 (1955) 181-94; A. A. Long, “Carneades and the Stoic T e l o s ” , Phronesis 12 (1967) 59-90. 16 S V F i, 361; 364-5. 17 cf. Stobaeus, SVF iii, 140-1; Aulus Gellius, ibid. 181 zyxwvutsr 19 Cicero, De fin. iii, 22, cum vero illa, quae officia e s s e dixi, proficiscantur a b initiis naturae, n e c e s s e e s t e a ad haec referri. . . They can of course be rationally justified, s e e p.76. 20 cf. Cic. De f i n . iii, 59; Clement, SVF iii, 515; Sextus, ibid. 516; Stob. ibid. 510. 21 Diogenes of Babylon, SVF iii, 44-46; Antipater, ibid. 57-8. 22 Phronesis 12 (1967) 63-69. pt 23 EN x, 1178a25 f f . appears to lay more s t r e s s on the quantity of goods required, Lkonard, op. c i t . p. 47, warns against taking the remarks in too material a s e n s e , cf. ibid. v, 1120 7 ff., ob y&p )Ev -r+ TrhjeEl TGV 616opkvov ~b ikUekp1ov, &M’ bv T? TOG 6itji)v~os‘&I, a i i ~ ~ 6’El ~mixT ~ oiruiav V 6i6ouiv. 24 cf. Cic. De f i n . iii, 31; Alex. Aphr. SVF iii, 766. zyxwvu zyxwvut zyxwvutsrqpo zyxwvutsrq zyxwvutsrqpon zyxwvutsrq 25 cf. Plut. Comm. not. 1071a: the -rkhoS i s not obtaining -rb: KaT& Q ~ U I V , but these S u m p Ghq T i $ h b w n a i -rjv ~ K ~ E K T &$av ~ K ~ V %xovoa, cf. Phronesis 12 (1967) 69-71. Acc. Arius Didymus (129, 19) the &v O ~ ~K V E Uare not U V ~ T T ~ T - ~ C O T I r?j~ K & dGaipovias, but they uvvcpyEiv cis -rb rhhos. 26 cf. Diog. Laert. vii, 89; Stob. S V F iii, 16. Contrast Arist. EN 1098b31-1099a7. 27 cf. Diog. Laert. vii, 107; Cic. De fin. iii, 5 8 , id officium nec in bonis ponamus, nec in malis; Philo, S V F iii, 512, 6 q&Ao~ ;via 6p@TGV K C S ~ ~ K ~ V T O VThe . fact that bad men can perform morally neutral ~ a 0 j ~ o v produces ra an apparent contradiction. Professor Sandbach, in a paper delivered a t Cambridge (see n. 11, suggests that “an appropriate action viewed in isolation from the agent i s neither good nor bad, but seen as the action of an agent acquires that agent’s character”. 28 cf. Stobaeus, S V F iii, 494. 29 Stobaeus, S V F iii, 499-500. to Mzrcus Aurelius, iii, 1. I am not here concerned with Panaetius’ treatment of ~ a B { ~ o v - r a . zyxwvut zyxwvuts On the queer phrase Ka%KovroS hpi0poi s e e Farquharson’s Commentary 30 Sextus, A d v . Math. xi, 200, 06 y&p r b inip&‘ia0ai yovhwv K a i 6 h h q r i p & yovEIs TOG arrou6aiov bu-riv ?pyov, &Ah&anov6aiov i b hrrb qpovfimws TOGTO TTOIEIV. cf. Cic. De fin. iii, 32. 31 S V F iii, 515; cf. Marcus Aurelius, ii 7; 16. 32 Philo, S V F iii, 512; cf. Stobaeus, ibid. 510. 33 Cic. De fin. iii. 5 5 ; cf. Diog. Laert. vii, 96. 34 cf. Cic. De fin. iii, 21; 33. 35 ibid. 20; Stobaeus, S V F iii, 510; Epictet. Ench. xxx. 36 Stobaeus. loc. c i t . 37 cf. Plut. S V F iii, 459; Stob. S V F iii, 560; 563; Galen, S V F iii, 471a. b 38 See n. 31, and Arist. EN vi, 1138 21 ff.; of the a~orrb5. 1144a23 f f . See further n. 27. At EE 1227b24 f. there i s said to be no Ahyo5 39 The matter is controversial, cf. Phronesis 12 (19671, n. 75, 89-90. and qtAaptmorhhqS (Ind. Here. col. 61). Panaetius was called piho.rrh&-rwv 40 cf. Sextus, A d v . Math. vii, 42. 41 Eros and P s y c h e (Toronto 19641, p. 164; cf. Sextus ibid. 151. 42 cf. Sextus, cited n.40 and 41. 43 Sextus, Hyp. Pyrrh. ii, 83. 44 “The Theory of Pleasure of the Old Stoa”, A J P h 83 (1962) 412-9. 45 Two Studies in the Greek A t o m i s t s (Princeton 19671, pp. 216-23. 46 De motu an. 701a4-6, cf. De an. iii, 10. 47 De motu an. 701b13-22; 33-702a1. cf. also P h y s . vii, 247a7 ff. 48 cf. De an. iii, 431a8-14. Hence the importance of its being controlled by A6yo5 (EN i, 1095a10) so that it becomes &p% (1139a24). 49 That i s to say, not by pursuing pleasure a s such, cf. vii 1153b 30 f., 0 6 6 ’ ~ $ 0 d p GI&KOUUI T T ~ N T E S {dovjv , &VTOI rrhv-rcs; De h i s t . an. 589a8 f . 50 cf. iii, 1110b9-ll. For the necessary connexim of Kah& and {Sha, cf. ibid. 1113a31 ff. 84 T$V &r$v zyxwvutsrqpo zy zyxwvuts zyxwvutsr zyxwvutsrqp zyxwvuts 51 EN iii, l l l l b 1 0 f f . 52 c f . EN x, 1176a15 ff., e s p . 25-28. At iii, 1117b 15 f. Aristotle observes that in the c a s e of certain &pa-ra\ (such a s courage) the aspect of (66 i s associated with the T ~ O Srather than the kwkpyEia. 53 c f . E N i , 1099a11-21. Some modern commentators, in the s t r e s s they place on EN vi, 2, seem to me to underrate the place of -rb (66 in Aristotle’s scheme. Confusion of course arises because (Sowfi and (66 are frequently used to denote physical pleasure, which i s to be suppressed in moral choice. But it i s Aristotle’s doctrine that genuine pleasures stimulate the actions of a particular q h S (EN vii, 1154b20). A just appraisal of Aristotle’s views in this respect is given by Jelf in his commentary to EN (Oxford 18561, pp. 59, 153. 54 cf. Cic. De fin. iii, 35; Stob. SVF iii, 394; Andronicus, ibid. 391; Plut. ibid. 459. 55 Andronicus loc. c i t . ; Galen S V F iii, 463. 56 cf. Diog. Laert. vii, 121; Cic. Tusc. d i s p . iii, 19. 57 Diog. Laert. ibid. 102; Stob. S V F iii, 70. 58 cf. Bonhoffer, Epictet und d i e Stoa (Stuttgart 18901, pp. 293 f . 59 Diog. Laert. vii, 116. See in general Bonhoffer, op. c i t . , pp.284-98. 60 cf. Stob. SVF iii, 169. 61 In a later letter (lix), after observing that gaudium in the Stoic s e n s e is peculiar to the good man, Seneca defines it a s “animi elatio s u i s bonis verisque fidentis”. 62 cf. Stob. SVF iii, 171; Plut. i b i d . 177. What applies to 6ppG in these passages applies a l s o to bpaei~, See in general Dyroff (cited in n. 2), pp. 16-24. zy 63 For further discussion s e e “The Stoic concept of Evil”, an article forthcoming in the Philosophical Quarterly, October 1968. 64 cf. Alex. Aphr. SVF ii, 823; Iamblichus, ibid. 826. Pohlenz “Zenon und Chrysipp”, NGG 2 (1938) 181-99, argues that this “monistic psychology” was an invention by Chrysippus who “rationalized” Zeno’s teaching on the TrdrOq. I am doubtful if the evidence gives Pohlenz his firm conclusion, but even if he i s right there is no reason to think that human 6ppG a s such was irrational for Zeno. For the different theses of Panaetius and Posidonius cf. Galen, P l a c . 457, 2-8; Cic. De off. i, 101; ii, 18. 65 cf. De a n . iii, 431a8-17; 434a11-12; EN i, 1102b29-31 b 66 cf. De an. ii, 413 21-4; iii, 433a18-29; De h i s t . an. 589a8 f . zyx 67 cf. Chrysippus’ definition or description of 6ppfi a s Abyos TrpomaKT1KbS aim+ ( s c . &wep&Try) 105 TroiE?w (Plut. S V F iii, 175). Clement attributes to the Stoics the description of bpaei~a s hoy1~1) Kivquis ( S V F iii, 442). 85