¹ For the modern discussion, see Philippson, R., ‘Die Rechtsphilosophie bei Epikureer’, AGP 23 (1910), 289–337, 433–46;Google ScholarStrauss, L., Natural Right and History(Chicago, 1953), 109–15;Google ScholarChroust, A.-H., ‘The Philosophy of Law of the Epicureans’, Thomist 16 (1953), 82–117, 217–67;CrossRefGoogle ScholarMuller, R., ‘Sur le concept de Physis dans la philosophic epicurienne du droit’, in Actes du VHIe Congres(Association Guillaumc Bude [Paris, 1969]), 305–318, and Die epikureische Gesellschaftstheorie(Berlin, 1972);Google ScholarClay, D., ‘Epicurus’ Kyria Doxa XVII′, GRBS 13 (1972), 59–66;Google ScholarBollack, J., Lapensee duplaisir(Paris, 1975), 353–92;Google ScholarGoldschmidt, V., La doctrine d′Epicure et le droit(Paris, 1977) and ‘La theorie epicurienne du droit’ in Science and Speculation,ed. Barnes, J.et al.(Cambridge, 1982), 304–26 (subsequent citations refer to Goldschmidt′s book);Google ScholarLaks, A., Gnomon 53 (1981), 1–4;Google ScholarVoelke, A.-J., ‘Droit de la nature et nature du droit: Callicles, Epicure, Carneade’, Revue Philosophique 172 (1982), 267–275;Google ScholarMuller, R., ‘Konstituierung und Verbindlichkeit der Rechtsnormen bei Epikur’, in : Studi...Gigante(Naples, 1982), 153–183;Google ScholarDenyer, N., ‘The Origins of Justice’, 133–52; and now the work of Long, A. A. (‘Pleasure and Social Utility - the virtues of being Epicurean’, Entretiens sur Tantiquite classiqw32 [1986], 283–329) and of P. Mitsis (see infra,n. 80), which was not available to me at the time of writing. Long does not consider the problem of the wise man′s justice, but his wide-ranging discussion of the relation between Epicurus′ doctrine of pleasure and his social philosophy is largely compatible with, and provides additional support for, the interpretation advanced herein (for our most important difference, see infra,n. 54). The main texts for Epicurus′ teaching on justice are KD5, 17, 31–8; Ep. ad Men.132; SV7, 51, 70 and F519, F53O–4 (fragments are cited according to H. Usener, Epicurea[Leipzig, 1887]), which should be supplemented by Hermarchus′ account of the Epicurean genealogy of morals ap.Porphyry, De Abst.1.7–12 ( = F24 inGoogle ScholarKrohn, K., Der Epikureer Hermarchos[Berlin, 1921]); Colotes ap.Plutarch, Adv. Col.1124d; Philodemus, Rhet.1.249–59 (inGoogle ScholarSudhaus, S., Philodemi volumina rhetoricai [Leipzig, 1882]); Diogenes of Oenoanda NF21 (inGoogle ScholarSmith, M. F., Thirteen New Fragments of Diogenes of Oenoanda[Vienna, 1974], 21–5); Cic. De Fin.1.50–4, 57; Lucr. 5.958–9, 1011–1160; Hor. Sat.1.3.99–114. D.L. 10.28 reports that Epicurus wrote Tltpl Sixaicmpayiasand Fltpl hiKaioainnqs KO\ rwv aWwv aperwv,but nothing is known of their contents.Google Scholar

² It is a characteristic feature of Cicero's argumentative strategy to appeal to cases like this which call into question the Epicurean claim that hedonism is compatible with traditional notions of virtue, as Dr Inwood argues in an unpublished paper, ‘Rhetorica Disputatio:The Strategy of De Finibus 2’.Google Scholar

³ Cf. Cic. De Leg.1.42–3, De Fin.3.62–8; D.L. 7.85–6; Plut. De Stoic. Repugn.1039b–e, De Soil. An.962a–b, De Amore Prolis495b–c; Porph. De Abst.3.19; Hierocles coll. 6.22–11.21Google Scholar (in Arnim, H. von, Hierokles, Ethische Elementarlehre (Papyrus 9780)[Berliner Klassikertexte IV, 1906]); and the Anonymous Commentator on Plato′s Theaetetus,coll. 5.36–6.35 (inGoogle ScholarDiels, H. and Schubart, W., Anonymer Kommentar zu Platans Theaetet (Papyrus 9782)[Berliner Klassikertexte II, 1905]). For the Stoic derivation of justice from olxflaiois,seeGoogle ScholarPembroke, S. G., ‘Oikeiosis’in Problems in Stoicism,ed. Long, A. A. (London, 1971), 122–32. Epicurus′ successor Hermarchus may have sought to forestall the Stoics‘ criticism by integrating into his genealogy of morals on a purely utilitarian basis; seeGoogle ScholarWaerdt, P. A. Vander, ‘Hermarchus and the Epicurean Genealogy of Morals’, forthcoming in TAPA 118 (1988).Google Scholar

⁴ Cf. Plut. Non Posse1090c, 1104b, Adv. Col.1127d–e; Sen. Ep.97.15; Epict. Diss.3.7.8–18; Clem. Strom.4.22 [= F582]; Atticus ap.Eusebius, Praep. Ev.15.799b–c [= F532]. Also: Epict. Diss.2.20.6–20; C. Cassius Longinus to Cicero, Ep.15.19.2 (on Epicurean political activity see A. Momigliano, JRS31 [1941], 151–7). Cicero is drawing on an orthodox Stoic account since he ties his criticism to oliaiwois,but his more general charge (the one most commonly brought against Epicurus), that Epicurean hedonism is incompatible with justice, need not depend specifically upon Stoic doctrine.Google Scholar

⁵ Horace here is contradicting the Stoic view according to which nee solum ius el iniuria natura dividicatur, sed omnino omnia honesta et turpia(Cic. De Leg.1.44); for his use of Epicurean doctrine in this Satiresee Goldschmidt (supra[n. 1], 150–65).Google Scholar

⁶ For Epicurus′ response to Glaucon′s challenge, see infra,p. 418; for his revision of social contract theory, see infra,pp. 420–1. Miiller (supra[n. 1 (1972)], 55–61, 83–7) discusses the influence of Democritus.Google Scholar

⁷ So e.g. Goldschmidt (supra[n. 1], 72–3, 78–9) attempts to read KD33 as a polemic against Plato, but the connexions adduced seem to me too general to serve as proof of deliberate polemic.Google Scholar

⁸ The evidence for Zeno′s Republicis collected in SKFi.259–71 and in Philodemus′ De Stoiciis,now re-edited with commentary byDorandi, T., CErc 12 (1982), 91–133. For discussion seeGoogle ScholarBaldry, H. C., ‘Zeno′s Ideal State’, JHS 79 (1959), 3–15;CrossRefGoogle ScholarRist, J. M., Stoic Philosophy (Cambridge, 1969), 64–72; J.Ferguson, Utopias of the Classical World(Ithaca, 1975), 111–21;Google ScholarMiiller, R., ‘Zur Staatsauffassung der friihen Stoa’ in Proceedings of the Vllth Congress...of Classical Studies,ed. Harmata, J. (Budapest, 1984), i.303–11; and infra,n. 12.Google Scholar

⁹ See also Goldschmidt (supra[n. 1], 17–18, 142–50), and infra,section IV. I take Plutarch′s to refer not to all men (Baldry [supra[n. 8], 12–13];Google ScholarMoles, J. L., JHS 103 [1983], 115), but only to all wise men (O. Murray, CR80 [1966], 369).CrossRefGoogle Scholar

¹⁰ Summarily: (i) Zeno holds that Eros is a god of friendship, concord and even liberty (Athenaeus 561c [= SVFi.263];cf. D.L. 7.129), and advocates the community of wives among wise men (D.L. 7.33, 131); Epicurus that the wise man will not fall in love, and that intercourse never has profited anyone (D.L. 10.118). (ii) Zeno prohibits the building of temples, law—courts and gymnasia (D.L. 7.33 [= SVFi.264–7, with further testimonial), Epicurus' wise man will dedicate statues (D.L. 10.121) in adhering to traditional religious practices (cf. infra,n. 24), and will take a suit to court (D.L. 10.120). (iii) Zeno rejects the introduction of money (D.L. 7.33); the Epicurean accepts it (D. L. 10.120). (iv) Zeno′s wise man will marry and father children (D. L. 7.121); Epicurus′ will not (D. L. 10.119). (v) Zeno says that the wise man will participate in public affairs(Sen. De Otio3.2 [= SVF i.271];cf. D. L. 7.121–2), Epicurus that he will not (D. L. 10.119).

¹¹ This section cites several works by Epicurus including the (10.119), in which he considered the wise man′s justice (see infra,section II); since this work considered a series of problems concerning the wise man, D. L. 10.117–21 may partly summarize its contents, perhaps drawing upon the epitome of Epicurus′ moral teaching by Diogenes of Tarsus, which is cited (10.118; cf. 10.26) on a question apparently considered in the Google ScholarGiusta, M., ‘Passi dossografici di morale epicurea nel X libro di Diogene Laerzio’, AAT 97 (1962–1963), 120–74 supposes this dgoxography to be of Stoic origin: if so, it might exaggerate the extent to which Epicurus was concerned to oppose the Stoics, although the fact remains that it is based closely on Epicurus′ writings. For the notorious orthodoxy of the Epicurean school, see D. Clay, ‘Individual and Community in the First Generation of the Epicurean School’ in (supra,n. 1), 255–79.Google Scholar

¹² In reconstructing this controversy, we must bear in mind that the Stoic theory underwent important changes between the time of Zeno and that of Cicero. Zeno wrote the Republicwhile still a student of the Cynic Crates (D. L. 7.4), and many of his doctrines were influenced by Antisthenes and Diogenes (see particularly Rist [supra(n. 8), 54–80], ‘Zeno and Stoic Consistency’, Phronesis22 [1977], 167–74;Google ScholarFisch, M. H., ‘Alexander and the Stoics’ AJP 58 [1937], 132–4). Although Chrysippus defended some of these Cynic tenets (cannibalism: SVFiii.747–53 [cf. Cleanthes, i.584]; incest: iii.734–6,753; community of women and children: iii.728, 744–5), later Stoics were so shocked to find incest (SKFi.256) and cannibalism (i.254) in Zeno's work that they rejected the Republicas spurious, or excused it on account of Zeno′s youth (cf. Philod. De Stoiciis,col. 9.1–19; D.L. 7.32–4). As expounded by Cicero, the Stoic theory of natural law bears little trace of its Cynic origins: not only is natural law compatible with civil society, it is even embodied in its ideal form in the Roman constitution (De Rep.1.70, 2.22–3, 3.33; cf. De Fin.3.65–8). (For Cicero′s position see Strauss [supra,n. 1], 153–6; J. E. Holton, ‘Marcus Tullius Cicero’ in History of Political Thought,edd. L. Strauss and J. Cropsey [Chicago, 1972], 130–50;Google ScholarNicgorski, W., ‘Cicero and the Rebirth of Political Philosophy’, Political Science Reviewer 8 [1978], 93–4.) Such developments in the Stoic theory are likely to have influenced their line of attack against Epicurean justice, although the most fundamental point of contention seems not to have changed: Cicero no less than Zeno maintains that one law is valid for all nations according to nature (cf. Plut. De Alex. Vin.329a–b with Cic. De Rep.3.33), whereas Epicureans hold that justice has no natural basis apart from a compact. The Stoic theory of natural law and its development is a neglected subject, and I am presently engaged in a comprehensive study. For the current state of the discussion seeGoogle ScholarKoester, H., ‘The Concept of Natural Law in Greek Thought’ in Religions in Antiquity,ed. Neusner, J. (Leiden, 1968), 521–41;Google ScholarWatson, G., ‘Natural Law and Stoicism’ in Problems in Stoicism,ed. A. A. Long (London, 1971), 216–38; R. A. Horsley, ‘The Law of Nature in Philo and Cicero’, HThR71 (1978), 35G. –59; G. Striker, 'The Origins of Natural Law', Proceedings of the Boston Area Colloquium in Ancient Philosophy 2(1986), 79–94 (with a reply by B. Inwood, 95–101).Google Scholar

¹³ Cf. my article cited supra,n. 3.Google Scholar

¹⁴ So e.g. Taylor, A. E., Epicurus(London, 1911), 94; C. Bailey, The Greek Atomists and Epicurus(Oxford, 1928), 510–14; J. M. Rist, Epicurus(Cambridge, 1972), 116–17, 122–3.Google Scholar

¹⁵ So e.g. Denyer (supra,n. 1), 145–6, who claims that ‘doing wrong has such benefits intrinsic to it that a sage would never agree to refrain from wronging others unless he got something very good in return' - 'an assurance that they will not harm him’ss. Sometimes it is recognized that fear of punishment alone does not motivate just action (e.g. C. Bailey, Epicurus: The Extant Remains[Oxford, 1926], 370; Miiller [supra(n. 1), 1982], 159–60), but no one has developed the point.Google Scholar

¹⁶ Strauss (supra[n. 1], 109–11). Strauss' view seems to be influenced by Cicero's argument (De Fin.2.50, 71–2) that the Epicurean wise man cannot live pleasantly without the approval of public opinion, although Strauss does not cite either passage. This view is not supported by any Epicurean text and is mistaken: what the Epicureans mean by 'justly' is not what the many approve but rather is what is socially advantageous (cf. KD37–8; Hermarchus 1.8.4). Moreover, the distinction Strauss draws between justice and the other virtues (which he thinks have a salutary effect in themselves) is explicitly contradicted by Torquatus, who says that justice admits of nearly the same treatment as the other virtues, and who emphasizes its tranquillizing effect on the mind quite apart from others' recognition of it (De Fin.1.50); see infra,section III.Google Scholar

¹⁷ See M. Packer, Cicero's Presentation of Epicurean Ethics(New York, 1938), 32–4, 91–3. The most promising suggestion I have seen is the undeveloped insight of Miiller (supra[n. 1 (1982)], 155): 'Weil der Weise sich auf die natiirlichen und notwendigen Bediirfnisse beschranke, werde es fur ihn hochstwahrschenlich gar keine Versuchung geben, Rechtsbrecher zu werden. Weil der Weise die Tugenden... als Instrument fur die Gewinnung des hochsten Gutes, der Lust, einsetze, werde er gerecht handlen.'Google Scholar

¹⁸ Its interpretation is complicated by a textual problem: Goldschmidt, De Lacy and Einarson accept the MSS reading (a hapax),while C. Diano, Epicuri Elhica(Florence, 1946), 147 and R. Westman, Plutarch gegen Kolotes(Acta Philosophia Fennica 7 [1955]), 186, following Usener, accept Stephanus′ emendation (which Diano explains: ‘ vox technica est atque id significat quod de quaque re, ut eius rei sit proprium, ’); see now A. Angeli′s review of the question, CErc11 (1981), 84–6. The sense of is unclear: it is usually taken to mean ‘predicate’, although it can also mean ‘accusation’ (so LSJ s.v., rejecting this as a falsa lectio).In my opinion the evidence is too slender to recover the precise logical sense Epicurus intended, but either reading seems to preclude an unqualified answer - which suggests, I take it, that Epicurus is unwilling to claim that the wise man will neverviolate the laws (see infra,pp. 4118).Google Scholar

¹⁹ Of the question raised in Plutarch, E. Zeller, Die Philosophie der Griechen1(Leipzig, 1923), iii(l).463 n. 4 says 'wollte sich Epikur nicht einlassen', and he is apparently followed in this approach by Diano and Westman (supra,n. 18), by Miiller (supra[n. 1 (1982)], 154, 160) and by Goldschmidt (supra[n. 1], 121), whose attempt to cast Epicurus' problem in the form of an Aristotelian syllogism does not inspire confidence and whose discussion (118–23) arrives at no clear view of Epicurus' position. I doubt that Epicurus would have posed this question only to deny its premiss, especially since the Epicurean wise man will not feel fear in any circumstance (cf. Cic. De Fin.2.57). Even if tenable, however, this explanation would mislead by implying that fear alone restrains injustice.Google Scholar

²⁰ Philippson (supra[n. 1], 3023); cf. E. Bignone, Epicuro(Bari, 1920), 1634 n. 2 and G. Arrighetti, Epicuro: Opere2(Turin, 1973), 573. This explanation leaves unclear (i) how to defend Philodemus' unparalleled suggestion on the basis of Epicurus' theory; (ii) why Epicurus would have considered the question problematic, were this his answer; and (iii) what relevance Philippson's explanation has to the condition specified - immunity from detection.Google Scholar

²¹ If De Lacy and Einarson ad loc,note b, were right in assuming that fear of punishment alone motivates just action, why would the wise man violate only disadvantageous laws when immune from detection? Denver (supra[n. 1], 1456) has a variant on their view, and it is open to similar objections.Google Scholar

²² It is puzzling that scholars have overlooked this fact: pleasure and pain supply the motives of conduct (e.g. Cic. De Fin.1.42), and hence it is the wise man's disposition toward pleasure which must determine whether he will commit injustice.Google Scholar

²³ Epicurus' position may have been influenced by Democritus, who clearly recognizes the insufficiency of law to restrain against injustice in secret (B181):Google Scholar

²⁴ See A. E. Raubitschek, 'Phaidros and his Roman Pupils', Hesperia18 (1949), 96103; the evidence and literature cited by H. Cherniss arfPlut. De Stoic. Rep.1034c, note d; C. J. Castner, 'Epicurean Hetairai as Dedicants to Healing Deities?', GRBS23 (1982), 517; B. FrischerThe Sculpted Word(Berkeley, 1982) with D. Clay, AJP105 (1984), 4849; D. Obbink, ' POxy.215 and Epicurean Religious dtcupia'in Atti del XVII Congresso Internazionale di Papirologia(Naples, 1984), ii.60719; D. Clay, 'The Cults of Epicurus', CErc16 (1986), 11–28.Google Scholar

²⁵ Plut. Adv. Col. Google Scholar

²⁶ Epicurus adopts a very strong view of the relation of virtue to eudaimonia:he claims not only that the pleasant life is inseparable from the virtues (Ep. ad Men.132; cf. Sen. Ep.85.18 [= F508]), but that virtue aloneis inseparable from pleasure (D.L. 10.138). The first claim commits Epicurus to the position that virtuous conduct is necessary for eudaimonia;the second that virtue is sufficient for it. The source of virtue is or ‘sober reasoning’ (vr),which searches out the motives for all choice and avoidance and thereby generates the pleasant life (Ep. ad Men.132). Epicurus thus conceives virtue as a cognitive state which provides one with the disposition toward possible objects of choice conducive to, and sufficient for, eudaimonia.Given this general conception of virtue, it is hardly surprising that justice provides not merely a guarantee of security against harm, but also the psychic harmony necessary for eudaimonia(see infra,section III).Google Scholar

²⁷ Epicurus does not explicitly consider what the wise man should do when faced with a disadvantageous law. Although he formulates criteria whereby to decide whether or not a law is just (KD37–8), no Epicurean text appeals from positive law to what is naturally advantageous. The explanation presumably is that the Epicurean, generally speaking, is not interested in the goods obtainable through political life (apart from security), and so does not seek its reform.Google Scholar

²⁸ Cf. Lucr. 3.1013–23, 5.1154–60, 1218^0; Plut. Non Posse1104a-b; Cic. De Rep.3.26, Tusc. Disp.3.32–3.Google Scholar

²⁹ Cf. Clay (supra[n. 1], 59–66), who shows that Epicurus modelled KD17 on Solon F l l (Diehl).Google Scholar

³⁰ According to Hermarchus (1.9.3–5), the ancient lawgivers introduced fearbeyond punishment as an additional means of civilizing irrational savagery.Google Scholar

³¹ Cf. Cic. De Fin.2.51–9, 70–1; De Off.3.38–9; De Leg.1.40–3; and the other passages cited supra,n. 4.Google Scholar

³² Cf. Goldschmidt (supra[n. 1], 110–11).Google Scholar

³³ So Cic. De Leg.1.40–1; De Fin.2.28, 53–9; De Off.3.35–41, 77; Sen. Ep.97.15; Plut. Non Posse1090c, 1104b; Epict. Diss.3.7.18; Clem. Strom.4.22 (= F582).Google Scholar

³⁴ Usener's attribution of F533 to Epicurus is conjectural, but well accords with Epicurean doctrine. Hermarchus sharply distinguishes between the lawgivers, who understand the advantageous and act according to it without the law's compulsion, and the many, who are ignorant of it and must be restrained from mutual homicide by law - fear of penalty is their (jxipnaKov(1.8.3). 'If all men', he continues, 'were equally able to discern and remember the advantageous, they would have no need of laws' (1.8.4) - a doctrine confirmed by Diogenes of Oenoanda's account of the 'Golden Age' which will come when all mankind has been saved by Epicurus' philosophy: then 'all things will be full of justice and mutual love, and there will come to be no need of defensive walls or laws and all the things we contrive on account of one another' (NF21 Smith; cf. F25 col. 2.3–11 Chilton).Google Scholar

³⁵ Goldschmidt (supra[n. 1], 118) denies that De Fin.2.28 is a report of this discussion, but he offers no arguments.Google Scholar

³⁶ On katastematic pleasure see Rist (supra[n. 14], 102–22, 170–2). There is a challenging and thoughtful discussion in J. Gosling and C. Taylor, The Greeks on Pleasure(Oxford, 1982), 365–96, but I am not persuaded by their attempt to banish the distinction between katastematic and kinetic pleasure from Epicurus' thought: they do not seem to me justified in rejecting Cicero's unequivocal evidence or the traditional interpretation of D. L. 10.136. My argument in this essay does not require that Cicero's understanding of katastematic pleasure be correct, although it does require that De Fin.2.28 be an accurate report of Epicurus' line of argument.Google Scholar

³⁷ For the Epicurean account of how freedom from pain constitutes pleasure, see De Fin.1.37–9 and, for some of the difficulties in this account, M. Hossenfelder, ' Epicurus - hedonist malgre lui'in The Norms of Nature,edd. M. Schofield and G. Striker (Cambridge, 1986), 245–63.Google Scholar

³⁸ Epicurus holds that the wealth demanded by vain opinion stretches to infinity (KD15; cf. SV25, 67–9; Ep. ad Men.130; D.L. 10.11; F181, F469; Lucr. 5.1117–19), that political power and fame are unable to provide security from the many (cf. KD 1with SV82, F552–4, F556 and Lucretius' denunciation oiambitio,5.1120–35), and that the craving for immortality prevents one from enjoying life (Ep. ad Men.124; cf. KD19–20; Lucr. 3.830–1094). Yet he urges those who by nature are lovers of honour and glory to pursue politics, because they would suffer a greater disturbance from anpayfioavv-q(F555; cf. F549).Google Scholar

³⁹ Cf. Ep. ad Men.122; SV54, 6 4; F220, F224, F471; Hermarchus 1.8.3; Cic. De Fin.1.42, 59; Diog. Oen. F2 Chilton.Google Scholar

⁴⁰ For Epicurean dtpanreia,see M. Gigante, ' Philosophia medicansin Filodemo', CErc5 (1975), 53–61 and M. Nussbaum, 'Therapeutic Arguments: Epicurus and Aristotle', in Schofield and Striker (supra[n. 37], 31–74). Nussbaum's central claim, that ' E p i c u r u s ' use of the analogy [between philosophy and medicine] is developed in reaction to Aristotle's ethical-medical analogies, both continuing and criticizing them', is not supported by solid evidence. F423 may conceivably be an attack on Peripatetic 'chatter about the good', but it does not mention the analogy, and the first sentence shows clearly that this is not its referent: Epicurus' objection to Aristotle (if he is indeed the target) is that he does not understand 'the nature of the good' as just denned, and there is no hint or implication that Otpa-neia.of the passions is at issue here. Hence Nussbaum's attempt to stitch this passage together with F211 is unfounded, especially since there is no need to refer the latter specifically to Aristotle. As to the arguments which she advances on pp. 65–6: (i) states as fact what needs to be proved; (ii) Epicurus' use of the ordinary term larptia.certainly does not prove that he was familiar with the Aristotelian use of the medical analogy; (iii) the documentation adduced does not support the claim that Philodemus 'seems to show detailed knowledge of Aristotle's use of the medical analogy', but even if correct this would not justify Nussbaum's central claim - Philodemus very often records later developments in the school, and hence one must have corroborating evidence to trace a point like this back to Epicurus. The fact that later Epicureans may have debated with Peripatetics on the analogy between medicine and philosophy proves nothing about the genesisof Epicurus' own views. Finally, the question of Epicurus' knowledge of Aristotle's ethical writing is far more problematic than Nussbaum allows, and her claim that 'we can safely invite ourselves to look for a relationship between the two thinkers on an ethical topic, even in the absence of more concrete evidence of a connexion', itself methodologically unsound, leads in this case to an unfounded account of Epicurus' relation to Aristotle.Google Scholar

⁴¹ Bailey (supra [n. 14], 493 n. 3) argues on the basis of F456 that should be classified as unnatural and unnecessary desires; see next note.

⁴² The wise man will gratify natural but unnecessary desires only when they do not harm his ataraxia. Concerning these desires, Cicero reports that Epicurus secundum autem genus cupiditatum nee ad potiendum difficile est censet nee vero ad carendum (Tusc. Disp. 5.93), and Epicurus himself grants that 7roAuTeAeta will sometimes be enjoyed if free of disagreeable consequences (cf. F181, Ep. ad Men. 131). Natural but unnecessary desires cannot remove pain, but merely' vary' kinetic pleasure (schol. adTO29;cf. KD 18, F417); and, since kinetic pleasure cannot increase katastematic pleasure, it cannot improve the wise man's state of ataraxia (cf. Rist [supra (n. 14), 106–8] and D. L. 10.136, where Epicurus clearly restricts ataraxia to katastematic pleasure). (For the possibility that natural but unnecessary desires are choiceworthy not as a class, but individually, see M. Wigodsky, AJP 107 [1986], 402–5.) Consequently the wise man will gratify natural but unnecessary desires as only when they are compatible with his ataraxia; see further infra, n. 50.

⁴³ For Cicero's objections see De Fin. 2.26–30, and for the Stoic response to t he doctrine that pleasure is animals' 'first impulse', D.L. 7.85–6 and Cic. De Fin. 3.17.

⁴⁴ For a possible qualification, see infra n. 59 with the corresponding text. Natural and necessary desires of course are easily satisfied (e.g. KD 21). Hence the wise man eschews the pleasures of profligates, which engender vain opinions and so trouble both body and soul (KD 10; SV 68–9, 81; Ep. ad Men. 131); rather, he seeks to attain a reasoned understanding of the end and limits of the flesh and thereby to remove the pain due to want (KD 20–1).

⁴⁵ Cf. Rist (supra [n. 14], 104–5, 108–9).

⁴⁶ 48 For this controversial doctrine, see KD 18, 20; Cic. De Fin. 1.41, 55–6; 2.89, 106–7; and generally F429–39.

⁴⁷ Cf. Cic. De Fin. 1.55–6, Tusc. Disp. 5.96; D.L. 10.137; Diog. Oen. F37–8 Chilton, NF20 Smith.

⁴⁸ Cf. Torquatus' account of the wise man's state of happiness (De Fin. 1.62; cf. 1.40–1, F68).

⁴⁹ Cf. SK 25, 67–9; F469, F548; Cic. Tusc. Disp. 5.90–1; Lucr. 5.1117–19.

⁵⁰ It might be objected that the wise man, since he will sometimes gratify natural but unnecessary desires, may commit injustice in order to vary his pleasure. The whole notion of variation is unclear, as Cicero complains (De Fin. 2.10), and it is also unclear why the wise man would undertake to gratify desires which do not contribute to his ataraxia (cf. supra [n. 42]). De Fin. 2.28 seems to rule out injustice merely for the sake of variation, since the argument that nothing can enhance the pleasure of freedom from pain is used to avoid the conclusion that Epicurus would do injustice for pleasure's sake. In any event, I doubt that variation, however understood, could serve as a significant motivation for injustice: the wise man will gratify only natural desires compatible with his ataraxia, and this restriction severely limits his interest in external goods.

⁵¹ Thus the Epicurean wise man will avoid all acts of what Aristotle terms particular injustice and traces to or 'the pleasure of gain' (EN 113Oa32–b5).

⁵² Cf. supra, n. 26.

⁵³ Carneades confirms this interpretation by reporting that, for the Epicureans, nothing obtainable by injustice can offset the penalty of fear which burdens the unjust man: nullum autem emolumentum esse, nullum iniustitia par turn praemium tantum, semper ut timeas, semper ut adesse, semper ut impendere aliquam poenam pules, damna (Cic. De Rep. 3.26).

⁵⁴ This conclusion presents serious difficulties for the view of Long (supra [n. 1], 301–5, 323) that for Epicurus friendship 'has a positive value and constitutive connexion with happiness, which needs to be clearly distinguished from that of mere justice. No pleasurable sentiment or intrinsic value pertains to just conduct.' Epicurus does consider friendship the greatest possession in securing blessedness (KD 27; cf. SV 52, 78), but Long's attempt to divorce just conduct from pleasurable sentiment is contradicted not only by Ep. ad Men. 132, according to which 'sober reasoning' or prudence actually generates the pleasant life, with which the virtues are naturally bound up (cf. KD 5; Cic. De Fin. 1.57; D.L. 10.138; F70, F509, F512; and supra, n. 26), but also by Torquatus' extended argument that justice is desirable precisely because it is productive of pleasure: ilaque ne iustitiam quidem recte quis dixerit per se ipsam optabilem, sed quia iucunditatis velplurimum afferat (Cic. De Fin. 1.53; cf. 1.42). Long seems to assume (p. 302) that the Epicurean is just merely because of the utility of the social contract, but, as I have argued in detail, his disposition to satisfy only natural desires conducive to ataraxia provides him with a motivation to act justly quite independently of whether or not those around him adhere to the social contract.

⁵⁵ 55 Why does Cicero fail to take into account Epicurus' positive argument for the wise man's justice when attacking his theory in De Fin. 2.51–9? Presumably because he rejects the tenets about pleasure on which Epicurus bases this argument: Cicero believes that Epicurus, for all his protestations, really is a vulgar hedonist (2.204), that his three–fold classification of desires is confused and untenable (2.26–30), that, in consequence, his wise man will indeed be attracted to great wealth and its pleasures (2.56–7), and that he cannot justify the distinction between katastematic and kinetic pleasure he uses in defending the wise man's justice (2.28).

⁵⁶ There is a challenging discussion of Epicurus' theory of friendship by P. Mitsis,' Epicurus on Friendship and Altruism', forthcoming in OSAP 5 (1987), which usefully brings out some competing tensions in Epicurus' position. But I cannot accept his conclusion that Epicurus recognizes an altruistic basis for friendship which is inconsistent with his hedonism because it sets up another end or criterion of choice than pleasure. It is evident from Cicero's account of the three different ways in which Epicureans treat friendship (De Fin. 1.66–70) that there was considerable uncertainty within the school as to how to explain it on egoistic grounds; probably n6 clear explanation was available in Epicurus' own writings. Hence it is hardly surprising that his position seems unclear in certain respects. But the evidence Mitsis adduces to show that Epicurus recognizes a non–egoistic basis for friendship is very weak: his argument depends crucially on accepting Usener's emendation of SV 32 Usener , the MSS reading of which is perfectly defensible (so e.g. Long [supra (n. 1), 305]), since De Fin. 1.68 does not purport to represent Epicurus' own view and moreover is an elaboration of an argument that starts (De Fin. 1.66) from the premiss that our friend's pleasures are not to be desired to the same degree as our own. (Mitsis misrepresents Cicero throughout by assuming that De Fin. 1.66–8 presents Epicurus' own view: in fact, Cicero represents it only as one of the three treatments of the subject current in the school, and De Fin. 2.82 does not suffice to assign the whole argument to Epicurus.) This evidence seems to me far too slender to justify Mitsis' claim that Epicurus recognizes an altruistic basis for friendship and therewith another end of action than pleasure, no trace of which may be found in Epicurus' surviving accounts of the criterion of choice. Clearly, Epicurus' followers found it difficult to construct a theory of friendship on an entirely egoistic foundation; but this difficulty does not justify revision of Epicurus' constantly repeated doctrine that pleasure is the sole criterion of choice.

⁵⁷ This point is effectively made by Long (supra [n. 1], 305 n. 22 [on p. 306]). Since the Epicurean has no fear of death, dying for a friend could be defended on egoistic grounds as an appropriate way of avoiding pain.

⁵⁸ Presumably it is a case like this that leads Epicurus, in the passage from the Atawopuu discussed earlier, to decline to give an unqualified answer to the question of the wise man's justice. To avoid the conclusion that the wise man might commit injustice in an extreme case, one would have to deny the premiss that the wise man ever could be immune from detection (cf. supra, n. 19), or argue that the wise man, since he will do what is advantageous whether or not there are laws (supra, n. 34), will violate them only to act according to what is naturally advantageous (this option restricts the view of De Lacy and Einarson [supra, n. 21 ] to an extreme case in which the wise man must violate the law to satisfy a natural and necessary desire). In the absence of discussion in our sources, any attempt to cash out these alternatives must remain conjectural. The important point is that natural desires are easily gratified without wrong–doing (cf. KD 21, Ep. ad Men. 130–1, 133, F469; Cic. De Fin. 1.53, 2.90–1, Tusc. Disp. 5.93), and the wise man's disposition toward pleasure narrows the circumstances in which he might violate the law to the extreme case in which it thwarts some natural and necessary desire (clearly not what Epicurus' critics had in mind). If one believes that he would commit injustice to vary his pleasure, these circumstances would be expanded, but not significantly (see supra, n. 50).

⁵⁹ I owe this point to unpublished work by Dr Mitsis.

⁶⁰ For Epicurean arguments concerning death, see D. J. Furley,' Nothing to Us?' in Schofield and Striker (supra [n. 37], 75–91).

⁶¹ So also if confronted with great pains: Epicurus holds that these will quickly put an end to life, while pains of long duration are not severe (F447; cf. F448; KD 4; SV 4).

⁶² See supra, n. 58.

⁶³ See Strauss (supra [n. 1], 156–63).

⁶⁴ See Inwood, B., 'Goal and Target in Stoicism', Journal of Philosophy 88 (1986), 553–4.Google Scholar

⁶⁵ This is the view of Philippson (supra [n. 1], 295, 298–9), rightly rejected by Miiller (supra [n. 1 (1969)]) and Goldschmidt (supra [n. 1], 170–86).

⁶⁶ See Goldschmidt (supra [n. 1], 223–4O).

⁶⁷ Note the similar role played by the differences in the abodes of various nations in Epicurus' account of the origins of language (Ep. ad Her. 75–6).

⁶⁸ For commentary on this text, see Goldschmidt (supra [n. 1], 125–38).

⁶⁹ On the problem of their identity, see Goldschmidt (supra [n. 1], 176–9).

⁷⁰ Cf. supra, n. 3.

⁷¹ For other considerations, see Strauss (supra [n. 1], 108).

⁷² Aristotle's understanding of man as by nature a political animal who possesses a common function which can only be properly realized in a certain kind of political community (HA 487b33–488al4; Pol. 1253a2–18, 1278bl7–30; EE 1242al9–28; EN 1097bll, 1162al6–21, 1169b 16–22) has its counterpart in the Stoic doctrine that man is a (cf. SVFiii.262, 314; Cic. De Fin. 3.62–3; Hierocles col. 11.13–21; and Pembroke [supra (n. 3), 125–7 with 144–5 nn. 61–3]).

⁷³ According to Sextus, the sceptic suspends judgment concerning the natural existence of anything good or bad because of the diversity of custom, and follows undogmatically the ordinary way of life (PH 3.235–8); for the sceptic's notion of belief, see now Frede, M., 'The Sceptic's Two Kinds of Assent and the Question of the Possibility of Knowledge', in Philosophy and History, ed. Rorty, R. et al. (Cambridge, 1984), 255–78. The diversity of custom is of course wholly ineffective as an argument against natural right (see particularly Strauss [supra (n. 1), 9–10, 97–101, 124–6]), but it is also ineffective against Epicurus, whose view that what is just varies according to a community's particular circumstances is merely confirmed by the manifest diversity of opinion concerning the just (note Carneades' use of Epicurean views in his attack on natural right, De Rep. 3.26). Since Epicurus' teaching on justice is parasitic on his doctrine of the human good, the acute sceptic presumably would attack via the latter. Polystratus' On Irrational Contempt (ed. G. Indelli [Naples, 1978]) is an interesting response to sceptical arguments from the Epicurean standpoint; see Goldschmidt (supra [n. 1], 180–6).CrossRefGoogle Scholar

⁷⁴ This paragraph applies only to Chrysippus and later Stoics who make the basis for their teaching on justice. Zeno accepted the principia naturae in his controversy with Polemo and so laid the basis for personal but it is not clear that he developed a theory of social s (cf. Porph. De Abst. 3.19; Cic. De Fin. 4.45; D. L. 7.87), and hence we cannot assume that Zeno would have argued against Epicurus' theory in the same way later Stoics did.

⁷⁵ The locus classicus is Laws 10.889e–890a; cf. 891c–892c, 966c–968a; Aristotle, Soph. Elen. 173a7–18;HeracIitusB102.

⁷⁶ See Kerferd, G. B. survey, The Sophistic Movement (Cambridge, 1981), 111–30, 139–62. One thinks at once not only of Glaucon's canonical exposition of the conventionalist thesis, but also of Thrasymachus (Rep. 338c), Callicles (Gorg. 482c–486d), Hippias (Prot. 337c–e), Antiphon (who seems to be engaged in a critique of popular views about justice rather than a positive exposition: D. J. Furley, 'Antiphon's Case Against Justice', in The Sophists and their Legacy, ed. G. B. Kerferd [Hermes Einzelschriften 44 = Wiesbaden, 1981], 81–91, developing an idea of Kerferd's) and the Critias fragment (D. K. 88B25), now assigned by A. Dihle (Hermes 105 [1977], 127–42; see contra D. F. Sutton, CQ N.S. 31 [1981], 33–8) to Euripides. But the doctrine of social contract as originally formulated seems to have been politically neutral (so Strauss [supra (n. 1), 119] and, independently, C. H. Kahn, 'The Origins of Social Contract Theory', in Kerferd, op. cit., 92–108), and it was also employed by such thinkers as Protagoras (Plato, Prot. 320c–328d), and the authors of the Anonymus Iamblichi and Ps.–Demosthenes XXV, to defend the vofioi against the claims of nature as a necessary and positive constituent of social life.Google Scholar

⁷⁷ An exception is Strauss (supra [n. 1], 114–15), who properly distinguishes vulgar from philosophic conventionalism. More typical are the views of Bailey (supra [n. 14], 511) and of Denver (supra [n. 1], 144–7), who wrongly assimilate Epicurus' theory to that of the sophists.

⁷⁸ This is the clear implication of F53O, quoted supra, p. 410.

⁷⁹ For the Epicurean claim that hedonism is compatible with traditional notions of virtue, see especially Cic. De Fin. 2.49–51; F513.

⁸⁰ After this essay had been accepted for publication, Dr Phillip Mitsis of Cornell University kindly sent me the chapter on 'Epicurean Virtues' in his book (in preparation) on Hellenistic ethics. Since this work is not yet available to the general public, I shall not discuss it in detail here. But Mitsis has independently recognized the dependence of Epicurus' theory of justice on his doctrine of the human good (though we approach the question in different ways), and his original and stimulating discussion of such related topics as Epicurus' cognitive conception of virtue and the precise form of his social contract teaching deserves the careful attention of scholars.