Epicurus on Pleasure, Happiness, and Desire

Cornell University Press Chapter Title: Pleasure, Happiness, and Desire Book Title: Epicurus' Ethical Theory Book Subtitle: The Pleasures of Invulnerability Book Author(s): PHILLIP MITSIS Published by: Cornell University Press. (1988) Stable URL: https://www.jstor.org/stable/10.7591/j.cttq45fk.5 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms Cornell University Press is collaborating with JSTOR to digitize, preserve and extend access to Epicurus' Ethical Theory This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms 1 Pleasure, Happiness, and Desire Like most other Greek moralists, 1 Epicurus thinks that the central aims of an ethical theory are to describe the nature of happiness (eudaimonia) and to delineate the methods by which one achieves it (ta poiounta ten eudaimonian; Ad Menoeceum 122). Perhaps the most important and certainly the most controversial feature of his ethical theory is his identification of pleasure (hedone) with our ultimate and final goal (telos), happiness (eudaimonia). By equating pleasure with happiness, Epicurus places his discussion of pleasure not only at the very center of his ethics but also squarely within the tradition of Greek ethical eudaemonism. 2 Many critics, both ancient and modern, have supposed that his entire ethical project stands or falls with his justification of pleasure as our telos. On the whole, this supposition is reasonable, since Epicurus tries to show how the content of morality, including friendship and altruism, can be derived from his analysis of pleasure. He manifestly believes, moreover, that he can justify a life of virtue by showing how it is inextricably linked to a life of pleasure. At the same time, however, even Epicurus' most sympathetic 1. The Cyrenaics are an instructive exception, however (cf. D. L. II.87-88). They claim that happiness is desirable not for its own sake but for the sake of particular individual episodes of pleasure. Thus, happiness is not our final goal (telos eudaimonias diapherein). This claim poses interesting challenges for Epicurean hedonism, which I address in the last section of this chapter. 2. See G. Striker, "Epikur," in Klassiker der Philosophie I, ed. 0. Hoffe (Munich, 1981) pp. 108-14. 11 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory critics have been quick to admit that several obvious inconsistencies afflict his account of pleasure. Because Epicurus' attempts to formulate a coherent ethical system and thereby give plausible answers to central questions about happiness seem crippled from the outset, it is tempting to dismiss his ethical theory as a whole. 3 By sometimes denying that he even needs any arguments for showing that pleasure is the telos of every rational action (De fin. l.3o), Epicurus seems merely to have aggravated and provided additional fuel for his critics' attacks. As Zeller complains, echoing a long tradition of obvious irritation, "No other system troubled itself so little about the foundations on which it rested." 4 If we are to believe his critics, then, Epicurus offers us obviously defective accounts of pleasure and happiness, without even so much as the courtesy of an argument. Although it generally is agreed that Epicurus' claims about pleasure are mistaken, it is not at all clear exactly what conception of pleasure critics mean to ascribe to him. Following Guyau, 5 there has been a widespread tendency to assimilate the Epicurean account of pleasure to hedonist theories in the British empiricist tradition. These comparisons have not always been explicit, but scholars, however consciously, have often relied on this empiricist conception of pleasure in approaching Epicurean hedonism. Such comparisons can be fundamentally misleading, however, and consequently have skewed our picture of Epicurus' general theoretical aims. 3· See G. Bonelli, Aporie etiche in Epicuro (Brussels, 1979), for a recent and extreme statement of such a view. 4· E. Zeller, Stoics, Epicureans, and Sceptics, trans. Reichel (London, 188o), p. 418. 5· J. M. Guyau, La morale d'Epicure et ses rapports avec les doctrines contemporaines (Paris, 1886). Guyau's sense that his study of Epicurean ethics had an important role to play in discussions of the dominant systematic ethical doctrine of his day, utilitarianism, often gives his book an air of intellectual engagement and excitement that subsequent accounts have found difficult to match. Sometimes, though, he is too ready to see correspondences between Epicurus' ethical concerns and those of his contemporaries. I will argue that the divergences between Epicurus and modern hedonists are in many ways more revealing than the similarities. Cf. J. Annas, "Doing without Objective Values: Ancient and Modern Strategies," in The Norms of Nature, ed. M. Schofield and G. Striker (Cambridge, 1986), pp. 3-29, for discussion and a fruitful example of this type of methodological approach. 12 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire Recently, scholars have become increasingly cautious about glossing over or explaining away disparities in the aims and methods of ancient and modern moral philosophers. This interpretive caution has not only made possible impressive gains in our historical understanding but also has begun to clarify some characteristic goals and assumptions that differentiate ancient from contemporary moral theorists. Sometimes this has had a salutary effect on recent discussions of moral topics as well. For example, the recognition that ancient eudaemonism offers an important and distinctive alternative to modern teleological and deontological defenses of morality has reinvigorated contemporary treatments of moral psychology, the virtues, and problems of ethical justification. In marked contrast, ancient discussions of pleasure have had, for the most part, an almost negligible influence on current thinking about pleasure. And this situation will no doubt remain until we begin to gain a better understanding of deeply rooted differences in ancient and modern methods of approaching the problem of pleasure. In any case, without a clearer understanding of these divergences there can be little reason to hope for an adequate appreciation of the distinctive philosophical aims and methods of ancient hedonism. With respect to Epicureanism, we have many reasons to be wary of the anachronism of treating Epicurus as a somewhat crude forerunner of Locke, Bentham, or Sidgwick. Given the tenacity and prevalence of such comparisons, however, a few brief initial caveats may be in order. The British hedonists' view of pleasure depends on a series of interrelated claims in epistemology and the philosophy of mind that cannot confidently be attributed to Epicurus, or indeed to any ancient thinker. One crucial element in Sidgwick's6 account of pleasure, for instance, is the Cartesian assumption that mental happenings are transparent states directly open to introspection. Descartes' particular picture of a private, inner mental life, when 6. In general, when speaking of the doctrines of British hedonism I refer to Sidgwick's formulations, since he gives the clearest account of classic utilitarian doctrine. His discussion is particularly useful because he thinks it is methodologically valuable to consider historically important alternatives to his views. See J. B. Schneewind, Sidgwick's Ethics and Victorian Moral Philosophy (Oxford, 1977), for further discussion. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory combined with the view that pleasure itself is something mental, gives rise to Sidgwick's assumption that pleasure is a special, uniform, internal "feeling" directly open to our introspection. 7 Given these initial assumptions about pleasure, the hedonist's project becomes the fairly straightforward one of discovering which activities tend to maximize this feeling overall. For several reasons, however, the attribution of this type of hedonic project to Epicurus is problematic. To begin with, it is worth noticing how this issue is often prematurely decided by translations that render hedone and voluptas as "the feeling of pleasure." Whereas speakers of English may be encouraged by such expressions as "I feel pleasure when doing x" to conceive of pleasures, at least initially, 8 as falling into a single class of uniform and commensurable feelings, speakers of Greek normally would have more difficulty viewing various pleasures as instances of a particular quality or type of "feeling." 9 This linguistic 7· See J. C. B. Gosling, Pleasure and Desire (Oxford, 1969), p. 52, for contemporary empiricist accounts of pleasure and their roots in the British hedonist tradition. M. F. Burnyeat has shown how dangerous it is to attribute this initial Cartesian claim to Greek thought as a whole ("Idealism and Greek Philosophy: What Descartes Saw and Berkeley Missed," Philcsophical Review 91 [1982], 3-40). See W. Lyons, The Disappearance of Introspection (Cambridge, 1986), for discussion of the role that claims about introspection play in psychological theories and, consequently, moral psychologies. 8. T. Penelhum ("The Logic of Pleasure," Philosophy and Phenomenolcgical Research 17 [1956-57], 488) suggests that we tend to take the noun 'pleasure' as the name of a private episode, analogous to a feeling. This is partly because 'pleasure,' like most nouns, suggests that there is some entity to which it refers. However, one need only think of Ryle's discussions of expressions such as 'enjoy,' 'like,' 'to be amused,' and so on to see how quickly this initial tendency becomes problematic (see G. Ryle, Dilemmas [Cambridge, 1954] pp. 54-67; "Pleasure,'' Aristotelian Society Supplementary Volume 28 [1954], 135-46). 9· Cf. Gosling, Pleasure and Desire, p. 24. I take up this linguistic issue in greater detail below and in my forthcoming commentary on De finibus l, II (ad loc. De fin. !.23). Of the modern translations of Epicurean texts that I have checked, every one uses "pleasure" and "the feeling of pleasure" interchangeably, with no apparent reasons from context. I claim here only that it is unclear that Epicurus considers pleasure a "feeling" in the sense required by hedonists like Sidgwick. Clearly, Epicurus thinks pleasure is a pathos (Ad Men. 129: kai epi tautin [hidonen] katantiimen hiis kanoni toi pathei pan agathon krinontes). But Greek philosophers use pathos for a wider range of states than can be plausibly characterized as "feelings" in Sidgwick's sense; see J. C. B. Gosling and C. C. W. Taylor, The Greeks on Pleasure (Oxford, 1982), p. 347· Cf. J. Brunschwig, "The Cradle Argument in Epicureanism and Stoicism,'' in 14 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire claim is complicated and clearly requires further argument. Nor would I want to argue that linguistic practices necessarily set inflexible limits to philosophical theorizing. But for the moment, by rendering hedone with a sufficiently neutral equivalent like "pleasure," we can perhaps avoid unfairly prejudicing the issue. 10 The importance of this will soon become apparent. At Ad Menoeceum 128-2g, Epicurus insists that pleasure is the arche and the telos, the beginning and the end, of the blessed life (tou makarios zen), because our pursuit of pleasure governs and unifies all of our rational choices and gives a structure to our lives as a whole. A bit earlier in the letter (Ad Men. 128), Epicurus had just claimed that the end (telos) belonging to the blessed life (tou makarios zen) consists in bodily health and tranquillity of mind (ataraxia). We do all things, he explains, to free ourselves from both physical pain and mental disturbance. This identification of hedone with freedom from pain and disturbance is Epicurus' most distinctive, though most problematic, claim about the nature of pleasure. 11 Most scholars have taken his assertion that aponia and ataraxia are the highest possible pleasures to be a clean contradiction or "a simple fraud"; 12 others, perhaps more sympathetically, have found it symptomatic of an ambivalence in Epicurus' commitment to hedonism. 13 Sidgwick, aligning himself with the former group, imThe Norms of Nature, ed. M. Schofield and G. Striker (Cambridge, 1986), p. 115, for a contrasting view about pathe. 10. P. Merlan, Studies in Epicurus and Aristotle (Wiesbaden, 1960), p. 1, claims that the Epicurean use of hedone is sui generis and warns that translations can be misleading. He then suggests, however, that the undisturbed condition of ataraxia is a state to which Epicureans idiosyncratically apply the term hedoni, "the feeling of pleasure" (p. 2). See J. Mewaldt, Epikur, Philosoph der Freude (Stuttgart, 1949), for the claim that the German equivalent of Epicurus' hedoni is Freude, and the defense of Merlan, who argues that Epicurus is not a philosopher of pleasure but a philosopher of joy (p. 15). 11. See Cicero, Tusc. disp. 111.47: "At idem ait non crescere voluptatem dolore detracto summamque esse voluptatem nihil dolore." Cf. U. 419, and for criticism De fin. 11.29-30. 12. Cicero at De fin. 11.29-30 reflects the general reaction: "Quam haec sunt contraria!" See Gosling and Taylor, The Greeks on Pleasure, p. 350. Cf. Plato, Republic 583c-85a. 13. Cf. M. Hossenfelder, "Epicurus-Hedonist malgre lui," in The Norms of Nature, ed. M. Schofield and G. Striker (Cambridge, 1986), pp. 245-63, and his fuller account in Die Philosophie der Antike 3: Stoa, Epikureismus, und Skepsis, vol. 3 of This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory mediately dismissed this "paradox of Epicurus" for its obvious opposition "to common sense and common experience." 14 And indeed, given Sidgwick's conception of pleasure as a feeling that we are to maximize, such a charge would clearly seem justified. Other Epicurean arguments are equally difficult to reconcile with this empiricist account of pleasure. Another conspicuous feature ofEpicurus' theory, for example, is his attempt to demonstrate how pleasure can meet several formal requirements for happiness. Among Greek ethical theorists, disputes tend to arise not so much over there being such formal conditions or requirements, about which there is fairly widespread agreement; rather, disputes generally arise either about the relations among these formal requirements or about the contents that will satisfy them. 15 Accordingly, in order to show how pleasure can meet the formal requirements of a theory of happiness, 16 Epicurus claims that in pursuing pleasure as our final good (see De fin. L2g}, we will be happy and, consequently, invulnerable to chance (AdMen. 131a}, 17 self-sufficient (Ad Men. 130; SV 44, 77}, and in complete possession of all the goods necessary for fully satisfying our natures (AdMen. 131a}. 18 In sharp Geschichte der Philosophie, ed. W. Rod (Munich, 1985). Hossenfelder argues that Epicurus adopts a eudaemonist ethical framework whose principles eventually pressure him into embracing hedonism as "a last resort." Epicurus, he claims, "would have preferred to be a Stoic" (p. 245). While I am sympathetic to Hossenfelder's attempt to show the importance of eudaemonist principles in Epicurean ethics, I doubt that Epicurus is a hedonist malgre lui. I will argue that Epicurus' theory of pleasure, properly understood, offers several plausible answers to the "eudaemonist problems" that Hossenfelder thinks Epicurus must solve. 14. H. Sidgwick, The Methods of Ethics (London, 1907), p. 125. 15. See T. H. Irwin, "Stoic and Aristotelian Conceptions of Happiness," in The Norms of Nature, ed. M. Schofield and G. Striker (Cambridge, 1986), p. 206, for a discussion of these formal conditions. In what follows, I am greatly indebted to his acco·unt. See also Domenico Pesce, Saggio su Epicuro (Bari and Rome, 1974), p. 69, and De fin. !.29: "quod omnium philosophorum sententia tale debet esse ut ad id omnia referri oporteat." 16. Epicurus' theory is not nearly as explicit as Aristotle's about the relations among (or, for that matter, even the necessity of satisfying) these formal criteria. Indeed, he sometimes denies the need for any justification at all of his claims about the telos. Yet these formal conditions appear prominently throughout his discussions of pleasure, and he certainly relies on such considerations when identifying happiness with hidoni. 17. I discuss in greater detail the evidence for this requirement in chapter 3· 18. Aristotle gives an account of these formal criteria at EN 1097a15, except This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire contrast, any attempt to fill this kind of formal inventory would seem distinctly odd to Hobbes, Locke, and their successors. 19 Hedonists of this stripe would argue that individuals attempting to maximize and intensify a particular feeling of pleasure may have good hedonic reasons for rejecting all such formal constraints. In their view, the intensity of particular episodes of pleasure might easily outweigh in overall hedonic benefit the fact that they threaten an individual's self-sufficiency or leave one vulnerable to forces outside of one's own control. Similarly, in applying this hedonic strategy to the conduct of our life as a whole, they might claim that a few years of enjoying such intense episodes could easily outweigh the risks of living either an incomplete or a dependent life overall. 20 If the way that we take our pleasures is strictly a subjective matter, as it evidently appears to be for adherents of this theory, Epicurean attempts to give us a surefire recipe for happiness might seem amusingly and misguidedly pedantic. By claiming to have discovered those needs and desires that are natural for all of us as properly functioning human beings to satisfy (Ad Men. 127; KD 29; SV 21), Epicurus would appear to these hedonists merely to be making an for the requirement that our happiness be up to us. Voluntariness is clearly an important condition for him as well, but the voluntariness of happiness is not strictly an independent formal requirement for Aristotle, since it must be adjusted and made compatible with completeness-that is, some goods necessary for our happiness may not be entirely under our own control. For the Epicurean, in contrast, happiness must be entirely within our control as well as complete (cf. Ad Men. 122). J. Annas examines the importance of completeness as a formal constraint for Epicurean happiness in "Epicurus on Pleasure and Happiness," Philosophical Topics 16 (1987), 5-21. She convincingly demonstrates how this demand for completeness pressures Epicurus in the direction of admitting nonhedonic values into his account of the final good. 19. An interesting exception is Mill, who gets into trouble trying to formulate his principle of utility precisely because he feels the need to account for several of these formal conditions. This point is brought out well by Annas, "Epicurus on Pleasure and Happiness"; for further discussion of the difficulties that these formal requirements present for both Mill and Epicurus, see chapter 3: Friendship, Happiness, and Invulnerability. 20. Such a view is closer to that of the Cyrenaics (D. L. 11.87), who argue that happiness is desired not for its own sake but for the individual pleasures that it contains. The Epicurean, in showing that pleasure can meet the requirements for a theory of happiness, tries to argue that pleasure can serve to structure a whole life rationally. On such a view, pleasure, like happiness, must be stable, permanent, and with fixed limits; cf. Hossenfelder, "Epicurus-Hedonist malgre lui." This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory illegitimate attempt to bolster his idiosyncratic view of pleasure. 21 Thus, the Epicurean appeal to nature 22 by way of the formal conditions of happiness can hold few, if any, attractions for proponents of a Benthamite felicific calculus. Given these manifest differences in aim and method, as well as the high degree of implausibility of central Epicurean doctrines in the context of British empirical hedonism, I want to suggest a more oblique approach to Epicurus' theory of pleasure. If we begin by assuming that Epicurus takes hidone to be a readily identifiable feeling that individuals can measure introspectively and then attempt to maximize, we no doubt will find his theory confused and disappointing. If we try to sort out the distinctive features of Epicurus' theory and examine them within the larger framework of Greek eudaemonism,23 however, we will find that his account of pleasure merits more sympathetic consideration. 24 And since his 21. For attempts to make a similar move without appealing to nature, see J. Griffin's account of 'informed desires' and his objections (Well-Being: Its Meaning, Measurement, and Moral Importance [Oxford, 1986], pp. ll-17). 22. It is instructive to compare Sidgwick's hostility to this procedure: "How then are we to distinguish 'natural impulses'-in the sense in which they are to guide rational choice-from the unnatural? Those who have occupied themselves with this distinction seem generally to have interpreted the Natural to mean either the common as opposed to the rare or exceptional, or the original as opposed to what is later in development; ... But I have never seen any ground for assuming broadly that Nature abhors the exceptional, or prefers the earlier time to the later" (p. 81). For a subtle and important discussion of the interplay of these two claims in Epicurus's theory of pleasure, see Brunschwig, "The Cradle Argument." G. Arrighetti discusses the epistemological dimension of this Epicurean appeal to nature and shows how it is meant to combat skepticism about the telos of action ("Devoir et plaisir chez Epicure," in Proceedings of the Vllth Congress of the International Federation of the Societies of Classical Studies, ed. J. Harmatta [Budapest, 1984], p. 386). 23. Here it is important to remember that within the context of Greek ethical thought, Epicurus cannot merely assume that pleasure and happiness are identical (cf. G. Vlastos, "Happiness and justice in the Republic," in Platonic Studies [Princeton, 1973], p. 111). For British hedonists this connection seems much more obvious, and they can identify happiness and pleasure almost without argument (see R. Brandt, Encyclopedia ofPhilosophy, "Happiness"; Sidgwick, p. 405). See G. Vlastos, "Happiness and Virtue in Socrates' Moral Theory," Proceedings of the Cambridge Philological Society 30 ( 1984), 181, for an important discussion of the respective meanings of 'happiness' and eudaimonia; also, M. Ring, "Aristotle and the Concept of Happiness," in The Greeks and the Good Life, ed. D. Depew (Fullerton, Calif., 1980), pp. 69-71. 24. That is, it merits consideration not only because of his attempt to meet the formal demands of eudaemonism but also because his theory of pleasure offers some plausible answers to questions about the content of happiness. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire analysis of pleasure plays a crucial role in the rest of his ethical theory, it may be possible, as I will argue in later chapters, to attribute to him more nuanced accounts of the virtues, of altruism, and of human action in general. Pleasure, Feelings, and the Satisfaction of Desire At this point, it might be helpful to distinguish two contrasting theoretical approaches to problems of pleasure and of hedonistic explanation. Since it is fairly common to find elements from both theories conflated in discussions of Epicureanism, a brief, though inevitably somewhat schematic, overview of the diverging commitments of these two approaches may prove useful. As a rough preliminary, we might broadly distinguish dispositional accounts of pleasure from theories that treat pleasure as a special type of private episode or feeling. The latter view, held in various forms by the British hedonists, rests on two fundamental assumptions about pleasure. It holds, first, that pleasure is a particular feeling whose presence can be verified by introspection, and second, that pleasant sensations are more or less similar in kind, although they may differ in intensity or duration. As a consequence, pleasures can be ranked on a uniform scale, and our pursuit of pleasure involves the relatively uncomplicated procedure25 of testing various pursuits and activities to discover which produce the greatest levels of intensity and duration of this feeling overall. Since pleasure is separable from the activities that give rise to it, the pleasantness of a feeling can be assessed ultimately only by the individual experiencing the feeling. Agents may make mistakes about the overall hedonic value of a present, occurrent feeling, of course, since when estimating its strength they must compare it with other feelings that they are no longer experiencing. As Sidgwick remarks, "in so far as any estimate of pleasantness involves comparison with feelings only represented in idea, it is liable to be erroneous through imperfections in the representation-still, no 25. That is, theoretically uncomplicated. This procedure may conceal considerable difficulties in practice. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory one is in a position to controvert the preference of the sentient individual, so far as the quality of the present feeling alone is concerned."26 In this view, then, pleasure is essentially subjective and separable from its sources. In contrast to the empiricist view, dispositional theories treat pleasure not as an immediately felt quality but as the realization of a perceived good or the satisfaction of a desire. 27 Proponents of this kind of motivational analysis, familiar from Aristotle, tend to focus more intently on the question of whether a particular desire is satisfied. 28 If I have been hungry and thirsty, but then am able to satisfy my desires for food and drink, I will find my present condition a pleasant one to the extent that I have managed to satisfy these desires. Similarly, if all of my desires have been completely satisfied, I will be in a state of pleasant satisfaction overall. A further moral sometimes is drawn here. If a certain desire cannot be satisfied, either in principle (such as a desire to live in a past century) or as a matter of contingent empirical fact (say, because of the scarcity of a desired good}, then the desire itself is to be viewed as frustrating and painful. Or if each attempt at satisfying a desire-for example, the desire for tobacco or cocaine-merely provokes a stronger craving, then these particular kinds of addictive desires are not for genuinely pleasurable activities. This outlook on desires indicates another difference between these two theories in matters of hedonic strategy. Hedonists pursuing the intensity of a particular feeling might try to strengthen 26. Sidgwick, p. 128. Cf. Gosling and Taylor, pp. 347-48, for the ascription to Epicurus of a similar conception of the incorrigibility of judgments about the pleasantness of occurrent states. See Ryle, Dilemmas (Cambridge, 1954), p. 58, for criticism of this general claim. 27. See Gosling, Pleasure and Desire, chs. 2, 3, and 10, for a much more detailed and nuanced treatment of these issues. I am gready indebted to his discussion in what follows. 28. This is by no means the case for all dispositionalist theories, however. Some, for instance, might focus exclusively on the manner in which a particular desire is being satisfied. I should emphasize that I am giving an account of only one possible version of the dispositionalist theory, because of its special relevance to Epicurean concerns. Nonetheless, I think this contrast between empirical and motivational views of pleasure, however broadly drawn, offers a useful backdrop for examining Epicurus' theory. 20 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire their desires or cravings in order to yield higher degrees of this pleasurable feeling. 29 They might even cultivate desires that are in principle unsatisfiable in order to experience individual intense episodes of a particular feeling. Dispositional theorists, on the other hand, generally avoid claiming that the intensity of a pleasurable feeling can serve as a criterion to rank pleasures. This reluctance is due to their disinclination to think of pleasures as essentially similar in kind or measurable on a uniform scale. Satisfaction, not degrees of intensity, serves as their criterion for assessing the pleasures of different activities. A necessary first step for those pursuing a pleasant life overall will consequently be to foster the types of desires that will be satisfiable. Moreover, agents must structure and coherently order their desires to ensure that none remains unsatisfied and that no particular group of desires will be mutually frustrating. For the dispositionalist, finding procedures for ranking various pleasures becomes more problematic. The empiricist relies on the claim that all activities give rise to a separable, kindred quality of feeling over and above activities themselves. Dispositional theories, however, reject the possibility of separating pleasure from activities in this way. It sounds oddly implausible, they would argue, to suggest to someone who, for example, plays the piano for pleasure that she can get that pleasure in some other way without having to bother with the playing. 30 The dispositionalist argues that doing something for pleasure is doing it for itself; one cannot merely 'take pleasure,' one must take pleasure in something (cf. Aristotle, EN 1175a11, 1175a21-b1). Conceived of in this way, pleasure is not a feeling over and above an activity; it is some further description of the manner in which someone realizes a perceived good, engages in an activity, or perhaps attends to that activity. 31 29. Cf. Aristotle EN 1154b4 for an example of this kind of strategy (autoi goun autois dipsas tines paraskeuazousin). 30. This example is adapted from Penelhum's discussion of Ryle in "The Logic of Pleasure," Philosophy and Phenomenological Research 17 (1956-57), 489. 31. For further discussion, see W. B. Gallie, "Pleasure," Aristotelian Society Supplementary Volume 28 (1954), 147-64; for important criticism, see U. T. Place, "The Concept of Heed," in Essays in Philosophical Psychology, ed. D. Gustafson (New York, 1964). pp. 206 ff. 21 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory It is important to notice, however, that the word 'pleasure' no longer seems to be strictly univocal in the context of this theory. 32 Accordingly, if we are urged to pursue pleasure as our final goal, it becomes difficult to see just how we are meant to follow such advice. Without a separable, measurable criterion, it becomes appreciably harder on the basis of hedonic criteria alone to rank our pleasures or satisfactions. On what purely hedonic grounds, for example, are we to compare the satisfactions of walking, or playing the piano, or reading Homer, or sipping coffee, once we have been deprived of the empiricist's handy yardstick? Not surprisingly, many philosophers who have held dispositional theories of pleasure have not been hedonists, since such theories make it difficult to explain our rational preferences solely on the basis of the hedonic quality of activities. Because we lack a common hedonic measure, they argue, we need to turn to other values and beliefs about the good in search of standards to explain and rationally to ground our actions. One last distinction must be made before we can turn to Epicurus' theory. The claim that pleasure consists in the satisfaction of a desire or the realization of a good may conceal, as Aristotle noticed, three alternative conceptions of the good that is being realized: (a) Pleasure is the attainment of what seems good to x. (b) Pleasure is the attainment of what is good for x, though perhaps not for others. (c) Pleasure is the attainment of what is good, simpliciter. 33 Introspection, for the most part, 34 might determine whether we 32. Gosling, Pleasure and Desire, p. 55· Cf. Aristotle EN 1153ai3-16. 33· Cf. Aristotle EN 1152b26-35 for these distinctions and a defense of his own view at 1175a21-22. 34· The Epicurean conception of 'natural and necessary desires'leads to further ambiguities in this context. It is not always clear that we are consciously aware of whether these 'desires' are satisfied. I may, for instance, be eating too much rich food without realizing it. In such a case, introspection will only tell me when an 'unnecessary' desire is satisfied. Epicurus' notion of individuals' having 'natural and necessary desires' that are not immediately open to introspection (at least for agents with corrupt beliefs) may seem to lead him in a Freudian direction. Thus, some have seen an appeal to unconscious desires and motivations in Lucretius' account of the fear of death. The extent to which Epicureans are prepared to extend such 22 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire have a particular desire and whether it is at present satisfied. 35 But we cannot decide solely from introspection whether we have desires that are good for us or, indeed, desires that are good simpliciter. These two objectivist claims clearly need a further defence not based solely on the evidence of our personal introspection. It is not always clear, however, which, if any, of these alternatives Epicurus' critics think he is committed to defending. We must therefore keep these three possibilities in the foreground to see the role that subjective and objective criteria are playing in Epicurus' arguments. 36 With these preliminary distinctions in mind, we are now in a position to turn to Epicurus' discussion. I begin by examining a few key doctrines that have seemed to offer the most straightforward evidence that Epicurus treats pleasure as a uniform, introspectible feeling. A central feature of the empiricist account, for instance, is the possibility of determining a hedonic calculus that can rank pleasures on a measurable scale of feeling. It is often supposed that when Epicurus urges us to compare the outcomes of various activities and to choose the most pleasurable one overall, he must be relying on similar assumptions about pleasure and the possibility of a determinate hedonic calculus. While it is no doubt true that Epicurus is committed to evaluating and ranking pleasures, 37 evidence for the actual mechanics of his particular calculus seems extraordinarily slim, even by Epicurean standards. We may be somewhat relieved that Epicurus avoids the obsessive categorization and classification of 'pleasure-making feanotions as desire, intention, or wish in a Freudian manner is problematic, however. Such desires may be 'unnoticed,' but it is misleading in an Epicurean context•to call them unconscious, if that is taken to imply that the unconscious corresponds to some special entity with its own explanatory principles. For purposes of the present discussion, I will try to clearly indicate when I am using 'desire' in a subjective or objective sense. 35· I may know by intro_spection whether my desire for tobacco is for the moment satisfied, but I do not know purely on the basis of introspective evidence whether it is, in principle, satisfiable. 36. For a general discussion of these alternatives, see R. Kraut, "Two Conceptions of Happiness," Philosophical Review 88 (1979}, 167-97. 37· When we talk of ranking 'pleasures,' we may mean (a) sensations or (b) pleasurable activities. See chapter 3 for a discussion of the problems for Epicurus' account of friendship caused by conftating these two senses. I will argue that Epicurus thinks that ranking pleasures primarily involves (b), whereas British hedonists appeal to (a). 23 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory tures' that characterize a theory like Bentham's. At the same time, however, Epicurus tends to pass over, or at least downplay in very odd ways, what we might reasonably think are obvious and important candidates for ranking pleasures. Both duration and intensity, for instance, receive treatments that are hard to square with standard empiricist accounts. Epicurus asserts that the duration of pleasure is not ultimately important in any rational assessment of the overall pleasantness of our lives. 58 At Kuriai doxai 19, he claims that "Infinite time and finite time contain equal pleasure [isen hedonen], if one measures its limits by reasoning" (cf. De fin. !.63). If we understand and achieve the rational limits of pleasure, he argues further, there is no reason to suppose that death subtracts anything at all from the best life (ti tou aristou biou; cf. KD 20). Cicero takes Epicurus in these passages to be manifestly, though wrongly, denying that pleasure is increased by duration (voluptatem crescere longinquitate) or rendered more valuable by its continuance (De fin. 11.88). He complains that nothing, in fact, could be more at odds with hedonism than the claim that death involves no loss or deprivation of hedonic goods. If pleasurable states make us happy, he argues, surely we will be happier if we can maintain these states longer. Epicurus' remark about duration at KD 19 certainly has a paradoxical quality, and we might think it merely a weak and ad hoc attempt to defend his rather extreme claim about death's 38. Cf. De fin .• ll.87-88. Gosling and Taylor find this claim about duration puzzling (p. 358) and attempt a solution based on the Philebus doctrine of mixed and unmixed pleasures. In their view, Epicurus can maintain his claim about duration if he means that we cannot compare the pleasantness of "two periods of unmixed pleasure," whatever their differences in length. They conclude that it is only in the "fanciful utopian conditions" of unmixed pleasure that we will not be concerned with duration. "In actual practice," they argue, "it will surely still be true that a wise man will always be concerned with increasing the proportion of pleasure in his-life." It seems to me, however, that not only is there no evidence that Epicurus restricts his denial of the value of duration to "fanciful utopian conditions," but also it begs the question to claim that the wise man will always be concerned with increasing his proportion of pleasure, if by increase we merely mean duration. Conceptions of what an 'increase' in pleasure means depend on a whole range of theoretical commitments, not all of which must take duration as the primary element in increasing pleasure. Moreover, Epicurus' claim about duration is a crucial prop for his argument against the fear of death, which must be eliminated "in actual practice," not just in "utopian conditions." 24 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire inability to harm a pleasant life. 39 We may be tempted to conclude, therefore, that theoretical pressures external to his theory of pleasure are responsible for this odd remark, a remark apparently inconsistent with standard empiricist views of the hedonic calculus. Another way of resisting Cicero's interpretation might be to take Epicurus' claim about equal pleasure (isen hedonen) as a reference not to equal quantities but to equal levels of pleasure. 40 In this view, Epicurus does not say that duration has no bearing at all in assessments of pleasure; rather, he might be asserting merely that we can experience the same level of pleasure in a finite as in an infinite time. No particular complete experience of pleasure, he argues, can ever be intensified beyond certain limits (KD 18), even if it were to be repeated an infinite number of times. No matter how many times we enjoy a particular pleasure, each of our individual experiences will never exceed a certain level of intensity. Nonetheless, on such a reading, calculations of duration within a lifetime will still be important for Epicureans in rationally measuring and assessing pleasures; for example, they will want to maximize their pleasurable experiences and enjoy them for longer rather than for shorter periods of time. There are two difficulties, however, with this attempt to soften the contradictions that Cicero finds in Epicurus' claims about duration. First and most important, it leaves the Epicurean without a leg to stand on in claiming that death in no way diminishes the complete happiness of mortals (KD 20, DRN III.83o). If the duration of pleasure matters within a lifetime, 41 death obviously can harm us by cutting short our pleasures or by robbing us of them entirely; as a consequence, we would have a rational justification 39· For this view, see D. Furley, "Nothing to Us?" in The Norms of Nature, ed. M. Schofield and G. Striker (Cambridge, 1986), p. 81. 40. I am indebted to D. Sedley for suggesting this possible interpretation to me. For further defense see now A. A. Long and D. Sedley, The Hellenistic Philosophers, vol. 1 (Cambridge, 1987). 41. Epicurus tends to treat individual pleasures as well in terms of their completeness, hence giving a qualitative as opposed to a quantitative account of pleasures. By way of comparison, it is interesting to note that Aristotle too, in his discussion of completeness in the Metaphysics, tends to downplay duration in favor of a qualitative account (Meta. 1021b13-14). 25 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory for fearing it. 42 But this conclusion is something that the Epicurean can in no way allow. Second, if in pursuing pleasures we are attempting to maximize a particular uniform feeling, it is hard to see the point ofEpicurus' comparison atKD 19, if he is asserting merely that a comparable quality of feeling arises in a finite or infinite time. If pleasure always consists of the same feeling or phenomenological condition, an individual experience of pleasure will consist of the same feeling whenever it happens to occur. Nor on such a view would it make sense for Epicurus to claim that our experiences of pleasure will feel the same only if we "measure the limits of pleasure by reasoning." If he is just comparing our pleasures in terms of their quality of tone or sensation, there would be no justification for adding this further qualification. There seem to be no very strong reasons, therefore, for doubting Cicero's interpretation of Epicurus' claims about duration. Cicero registers another complaint, moreover, that seems to be justified by the surviving evidence. He objects to an asymmetry in Epicurus' treatment of pleasure and pain (De fin. 11.88), since the Epicurean panacea against pain-namely, that acute pains last only for a short time (KD 4)-must rely on duration as an important criterion in assessing pains (Ad Men. 129: polun chronon hupomeinas&). If Epicurus, in the manner of an empiricist, were committed to treating pleasure and pain as contrary ends of a uniform scale, this would indeed be 42. See Lucretius, DRN Ill. 83o-977, for some attempts to show that duration does not affect the overall pleasure of a life. Perhaps the most interesting is the socalled asymmetry argument at 972-77, where Lucretius raises the problem of the apparent asymmetry in our attitudes toward our death and our prenatal nonexistence. Most of us find it painful to think about our death and its deprivations, but we seem completely unconcerned about our previous nonexistence and its deprivations. Lucretius connects these asymmetrical attitudes to the past and future with our views about the duration of our lives in general, his argument being that if we are indifferent to the possibility of our life extending temporally in the past, it seems irrational to have any special concem about the possibility of persisting into the future. Thus our common attitudes seem to indicate that we have no specific rational attachment to our duration per se; otherwise, we would care just as strongly about our prenatal losses and deprivations. But if we have no concem about duration for its own sake, the Epicurean has an important supporting argument for showing that duration is of no special importance in assessing the overall pleasantness of lives. There are problems with Lucretius' argument, but it demonstrates that Epicureans are concemed with defending the stronger claim about duration that Cicero ascribes to them. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire a serious objection. Epicurus' account of the intensity of pleasure seems to harbor a corresponding difficulty. Epicurus denies that pleasures vary, in one crucial respect, with regard to their intensity. 45 If I am thirsty and may satisfy my thirst either with brackish water or with some more appetizing drink, neither alternative, according to Epicurus, can be more intensely pleasurable, because whenever a pain is removed or a desire is satisfied, the pleasure of the resultant state cannot vary in intensity. We might argue, of course, that Epicurus' account of the alleviation of pain or distress will require some notion of variable intensity. 44 If Epicurus holds that decreasing pain is equivalent to increasing pleasure, we would expect him to distinguish increasing and decreasing levels of intensity. As in the case of duration, however, it is hard to find any explicit evidence for this important canon of the empiricist's view of pleasure. Epicurus mentions that the removal of pain is followed by a state of pleasure (KD 3, 18; De fin. 1.37), but he conspicuously fails to describe this change in terms of increasing and decreasing levels of intensity. 45 Rather, he describes pleasure and pain as two successive, contradictory states, without explicitly claiming that transitions between these two states will consist of variable intensities of a particular feeling. We might think that any plausible conception of a hedonic calculus will require discriminations of intensity and duration, but given such ambiguous indications in the surviving evidence, we should refrain from ascribing these empiricist assumptions to Epicurus without stronger justification. If we look for evidence about the actual operations of the Epicurean calculus, we again are confronted with difficulties. Cicero's 43· At Ad Men. 131 Epicurus says that bread and water give akrotaten hidonin. This is often translated as the "highest or most intense pleasure," and talk about intensity might ostensibly commit Epicurus to an empiricist view. Epicurus is concerned with the limits of pleasure (Ad Men. 133), however, and akrotaten is better rendered with this in mind. Epicurus is claiming that by satisfying our hunger with bread and water, we reach the limit of pleasure, that is, complete satisfaction (D. L. X.121; cf. Bailey, lncert. Ep. Frag. 37, KD 3, 18). 44· Again, I owe this objection to D. Sedley. 45· Cf. Hossenfelder, "Epicurus," p. 255, for a different interpretation. He cites KD 18 for evidence that Epicurus conceives of "one and the same emotion that only varies occasionally in intensity"; this seems to me to go beyond the evidence, however. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory various accounts of Epicurean criteria for ranking pleasures (De fin. !.32-33, 48) are all as disappointingly vague as Epicurus' own statement atAdMenoeceum 130 ff. 46 There Epicurus merely suggests that it is "by means of a comparison [summetreszl and survey [blepsezl of advantages and disadvantages that we must judge [krinein]" matters of pleasure. The very generality of this passage's procedural recommendations, however, not only fails to commit him to anything so specific as the empiricist's theory of pleasure's measurable uniformity47 but perhaps should also alert us to the possibility that Epicurus may be operating with an alternative conception of pleasure. His talk about comparing advantages against disadvantages might fit a dispositionalist view of pleasure just as well, if we take him to be arguing only that we should weigh the advantages of certain desires, pursuits, and activities. The dispositionalist can deny that we are able to measure pleasure as a uniform feeling, yet still think it important to weigh the advantages and disadvantages of cultivating various desires. If, for example, I know that by cultivating a desire for cocaine I will be developing an addictive desire that will grow stronger, I can decide that it would be advantageous for me not to do so. That is, I can compare-without recourse to the empiricists' particular conception of pleasure-the hedonic advantages of having satisfiable desires with the disadvantages of having unsatisfiable desires. 48 The empiricist might claim at this point that such a view still would leave unresolved the problem of ranking various satisfiable 46. This holds true for other doxographic reports as well. Cf. U. 440-45; also, Diogenes of Oenoanda 38.1.8 on the difficulty of comparing mental and bodily pleasures, and his vague statement about how the wise are able to manage such a comparison. 4 7. Nor does KD 9 give evidence for the uniformity of pleasure in this required sense. The present unfulfilled conditional is often used by Epicurus to deny the possibility of what is being asserted. And in any case, this claim would not commit him to an empiricist account; it might easily be taken as a reference to the equal value of katastematic pleasures, not their measurable uniformity on a common scale of feeling (cf. Gosling and Taylor, pp. 378-82, for an opposing view). 48. It perhaps should be noted that someone may have two satisfiable desires that taken together will conflict; or perhaps we might have a desire whose strength varies over time. The hedonic art will consist not just in evaluating particular desires but in ordering one's desires into a coherent whole, taking into account individual relative importance and their mutual relations. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire desires in order to insure that we foster only those that have the greatest hedonic rewards. Even if we can roughly divide satisfiable from unsatisfiable desires without relying on the empiricist's measure, however, it might be argued that there still would be important hedonic differences among the remaining satisfiable desires. Would we not therefore require a more sensitive and fine-grained calculus to make these sorts of rankings? Whatever the plausibility of this empiricist objection, it does not seem to be one that would move Epicurus, and the reasons for his indifference again suggest that he may be approaching the calculus with a different conception of pleasure. Epicurus broadly characterizes our natural and necessary desires as those which are easily satisfiable. Moreover, he claims that our proper attitude toward the various means or processes of satisfying these desires is one of complete indifference. If either white bread or brown bread will satisfy our hunger and both are readily available, Epicurus thinks that we have no hedonic justification49 for choosing one over the other. Empiricists might insist, of course, that it is precisely when we are faced with these kinds of detailed choices that we need the precision of their calculus to help us maximize our pleasure. But, again, that Epicurus does not seem to be interested in such fine-grained and discriminating calculations suggests that his interests in the calculus as well as his view of pleasure may be very different. Gosling and Taylor offer another reason for Epicurus' apparent lack of interest in more intricate hedonic calculation. 50 They claim that he is influenced by what has come to be a perennial objection 49· For this claim, Epicurus relies on his distinction between kinetic and katastematic pleasures. If neither of two choices jeopardizes my katastematic state, it does not matter which I choose, since they are equally good choices. One factqr that cannot help rationally to guide my choice, however, is a preference for the different kinetic pleasures of eating white and brown bread. If I become attached to the pleasures of white bread and slowly learn to despise brown bread, I will be in danger of forgetting that the most complete pleasure consists in not being hungry. If Epicurus allowed kinetic pleasures to be of any rational concern whatsoever, someone might choose a katastematically harmful activity if it were accompanied by great amounts of the requisite kinetic pleasure. But Epicurus denies that we ever could have reason to do so. I take up these issues in more detail below, in "Kinetic and Kastatematic Pleasures." 50. Gosling and Taylor, pp. 359-60. 29 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory against utilitarianism: that constant attention to the details of hedonic reckoning will become a source of painful anxiety. That is, if we continually concentrate on maximizing our pleasures in all the minute dealings of our daily lives, we will quickly acquire the habits and dispositions of somewhat obsessive, anxious, and myopic accountants; our pleasures will slip through our fingers unenjoyed as we nervously busy ourselves over our calculations. The problem with their attempted solution, however, is that Epicurus insists that we must scrutinize every choice (Ad Men. 132) and every desire (SV 71) at all times (KD 25) to make sure that every one (AdMen. 128) leads to our final goal, pleasure. 5 1 Epicurus manifestly recommends the kind of continual attention to our choices and desires that Gosling and Taylor find objectionable. He thinks that paying meticulous attention to our desires not only fails to cause anxiety but will actually reward us with a heightened sense of self-sufficiency and a quiet confidence in our immunity from fortune (Ad Men. 132). Similarly, only by properly monitoring our desires will we make our pleasures complete (U. 417-22). Thus, giving careful attention to our desires not only helps to insure that they will be satisfied but apparently aids us in meeting the formal conditions of happiness as well. At the same time, however, Epicurus' interests in the calculus seem to extend only to a few key criteria. Two of these might be derived from or, perhaps, might be compatible with a dispositionalist theory of pleasure. Like the dispositionalist, Epicurus asks whether a particular desire is satisfied and whether it will remain satisfiable. On the other hand, although not strictly incompatible with a dispositionalist theory, his other central questions about pleasures-"Will they expose me to fortune?" "Will they maintain or threaten my self-sufficiency?" "Are they complete with respect to their satisfactions?" -all would appear to be derived independently from the formal requirements for happiness. 52 Epicurus thus avoids 51. It might be argued that Epicurus means only that we should evaluate every action type, not every individual token in passages such as Ad Men. 132 and SV 71. However, KD 25 and especially Ad Men. 128 suggest a strong concern with every individual choice. 52. Similarly, whereas contemporary critics of hedonism might dispute the possibility of discovering and coherently using such a calculus, the criticisms offered This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire an intricate empiricist calculus, not because he views it as a source of possible anxiety, but because it relies on the wrong criteria to evaluate pleasures. For him, the proper criteria must be derived from these general eudaemonist conditions. Consequently, the Epicurean calculus does not give rise to detailed Benthamite calculations based on intensity, duration, and so forth, although it still will obviously require constant and careful application at this more general level. The empiricist might complain at this point that the Epicurean calculus is not exacting enough to ensure that we will maximize our individual experiences of pleasure. But the Epicurean can answer that any pleasures55 that register and are measurable on the empiricist's calculus alone will not meet the formal requirements for happiness; therefore, such pleasures are a matter of indifference. Given these diverging approaches to the problems of the calculus, I think we now have at least some initial reasons to suspect that Epicurus' theory of pleasure has its own distinctive theoretical preoccupations that can cut across standard empiricist and dispositional theories in important ways. His appeal to formal requirements for happiness, for instance, would appear equally foreign to contemporary empiricists and dispositionalists alike. At the same time, comparisons of his theory to British hedonism clearly seem misguided, since Epicurus has, if anything, much more in common with certain forms of dispositionalism. Like the dispositionalist, he regularly equates pleasure with the satisfaction of desire and pain with the frustration of desires (Ad Men. 128, 13ob; SV 33; see also KD 18 and AdMen. 130-31 for the expression to kat' endeian algoun; Scholion, KD 29; De fin. II.g). Similarly, in assessing pleasures he focuses on activities and desires rather than on accompanying sensations (Ad Men. 132a; SV 78; De fin. !.55; De abstin. 1.51)54 • He also by Epicurus' opponents reflect a concern with formal eudaemonist requirements. Cicero, for instance, argues that if pleasure is our final good our happiness will not be invulnerable (De fin. II.86), complete (De fin. I1.38-44, 86, 87), or self-sufficient (De fin. II.86). Cf. Seneca, De benef. III+ 1. In the last section of this chapter, I take up the question of whether these formal requirements are in any sense hedonic, that is, whether they can be derived solely from an account of pleasure. 53· That is, kinetic pleasures. 54· Here it should be remembered that for Epicurus mental pleasures and pains This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory claims that all the pleasures that a rational agent should pursue presuppose an existing lack or want (Ad Men. 128; KD 18, 33). 55 From this, he concludes that someone with easily satisfiable desires will lead the most pleasant life, and he denies, moreover, that we can gain more pleasure by cultivating more demanding desires. 56 All these doctrines suggest a theory that is incompatible with the view that pleasure is a separable, episodic feeling that admits of varying degrees of intensity and duration. One might object at this point that Epicurus unwittingly vacillates between differing conceptions of pleasure or, perhaps, thinks that they are somehow compatible. Other philosophers have similarly focused on the satisfaction of desires and yet have supposed that such satisfactions are "necessarily accompanied by [a] particular type of sensation of phenomenological condition." 57 Given the popularity of this conflation, why should we think that Epicurus is any more clearheaded about these problems? Could he not, regardless of his views about the calculus, still treat pleasure as a separable, measurable feeling in spite of his many affinities with a dispositionalist account? It might be helpful to turn to a related doctrine that seems to give evidence for this kind of conflation. Notoriously, Epicurus claims that there is no middle state between pleasure and pain. 58 When all of our desires have been satisfied and we can expect them to remain satisfied, then, he argues, we are in a state of the highest pleasure (SV 33). It is widely believed that Epicurus is claiming here fit this same model. The mind may be afflicted with unnecessary desires such as avarice or ambition; they are painful because they are impossible to satisfy. Nor does introspective access have any special role to play in evaluations of these mental pleasures and pains. 55· See below, note 67, for some difficulties. 56. For these connections between pleasure and the satisfaction of desires, see AdMen. 127, 128, 130; KD w, ll, 12, 15; SV68, 69, 71, So, 81; Dejin.l.39; DRN l1.14-36. 57· D. Lyons, In The Interest of the Governed (Oxford, 1973), p. 22. See Sidgwick, pp. 43-56, for a discussion of the source of this confusion. See De fin. !.37 and Ad Men. 131 for the way that Epicureans regularly use the satisfaction and frustration of natural and necessary desires (hunger and thirst) as paradigm examples of pleasure and pain. Cf. KD 30. 58. to de ponou kai hidonis miden einai meson, Plutarch, Adv. Colot. l123a (U. 420). 32 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire that the satisfaction of our desires is accompanied by a special, separable feeling. His particular description and assessment of this accompanying feeling have often been taken to be merely perverse or perhaps just nonsensical. Indeed, many have objected that it seems hardly conceivable that anyone could have made the elementary error of treating a neutral state of sensation as the most pleasant state possible. Not surprisingly, it has been suggested that Epicurus very easily could have refuted for himself this denial of a neutral state by a simple test of introspection. Others have argued that Epicurus did resort to introspection, but because of his innate optimism, mistook what most people59 find neutral or indifferent for not only a pleasurable feeling but indeed the most pleasurable feeling we can experience. 5° If this is the highest state of pleasure Epicurus has to offer us, however, many have wondered why anyone would ever want it. Indeed we might wonder what could have possibly induced Epicureans to structure their whole lives around the so-called pleasures of such desiccated states of feeling and sensation. Amid all this criticism, however, it has not been suspected that Epicurus is referring to anything other than a particular sensation or quality of feeling. We need to look more carefully at the kinds of questions Epicurus is asking and the theoretical commitments that are motivating his questions. Clearly, the denial of a neutral state between pleasure and pain would be somewhat implausible for anyone holding an empiricist view that treats pleasure as a separable feeling. But it is important to recognize that this particular view of our states of consciousness is hardly theoretically neutral. Sidgwick, for instance, argues that if we think of pleasure and 59· Cicero, De fin. 11.77: "quam praeter vos nemo appellat voluptatem." 6o. Cf. Merlan, p. 10: "This, then, Anaxagorean pessimism (and Plato's 'neutralism') is the appropriate contrasting background to the Epicurean doctrine of the katastematic hedone. An organism left to itself alone, an organism just performing its vital functions in an unimpeded way experiences hedone, to be sure, a hedone sui generis. But according to the 'physiological' theory reported by Aristode this organism experiences lupe. ... Radical optimism and radical pessimism clash. And it seems that the choice between them will always be rooted in some personal factor." Merlan goes on to speculate that Epicurus' optimism about the pleasantness of the neutral state stemmed from a heroic defiance and, perhaps, overcompensation in the face of continual sickness. 33 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory pain as being opposite ends of a uniform scale, "we must therefore conceive, as at least ideally possible, a point of transition in consciousness at which we pass from positive to negative. It is not absolutely necessary to assume that this strictly indifferent or neutral feeling ever actually occurs" (my emphasis).5 1 Sidgwick goes on to claim in this passage that we may sometimes experience states that approximate to neutrality, but he holds open to doubt the view that we can ever experience this theoretical 'point of transition' by means of introspection. Few of Epicurus' critics have been willing to show a similar caution. 62 Moreover, arguments for a neutral state of feeling, at least as formulated by Epicurus' critics, are themselves susceptible to troubling regress arguments. The quarrel between Epicurus and his opponents is usually presented as arising from their different evaluations of a neutral sensation or state of consciousness. Epicurus, the argument goes, finds these neutral sensations pleasant and enjoys them, whereas his critics find them merely indifferent.5 5 For the empiricist, however, this notion of finding sensations pleasant or indifferent leads to the following regress: If I ask you whether you find a particular sensation or state of consciousness pleasant, and you answer "Yes, very," on the empiricist view of pleasure this must mean that this sensation or state of consciousness is accompanied by another pleasurable feeling. The same question could then be a!iked about this further feeling, which would give rise in turn to a third-order feeling of pleasure, and so on ad infinitum. 54 61. Sidgwick, p. 1114. 6a. An exception is A. A. Long (Hellenistic Philosophy [Berkeley, Calif., 1974], p. 64), who, although accepting the notion of neutral states of consciousness, thinks that it would be much rarer during our waking hours to describe ourselves as neither happy nor unhappy. Long's observation about happiness is central to Epicurus' denial of a neutral state. I doubt, however, that Epicurus is appealing to introspective experience to defend his claim. I will argue that he is asking the objective question of whether someone is happy, that is, meeting formal conditions; and that question admits of only a positive or negative answer. 63. See Long, Hellenistic Philosophy, p. 63. 64. For this regress argument, see Ryle, "Pleasure," p. 195, and Penelhum, pp. 489-911, who gives a succinct account of another regress described by Ryle in Dilemmas, p. 58:"We can say of any sensation that it is pleasant or unpleasant or neutral. The same sensation might be found pleasant one time, unpleasant at another, and 34 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire At the very least, such difficulties show that we should be more cautious before dismissing Epicurus' position as riddled with outright absurdities. Neither the logical status of claims about a neutral state nor the verdict of introspection suggests that empiricist objections to Epicurus are themselves unassailable. I would argue, moreover, that these kinds of objections to Epicurus' theory are misguided in a more fundamental way. To see why, we must remember the formal eudaemonist framework that motivates his questions about pleasure. At De finibus !.38, Cicero gives the following account of Epicurus' denial of an intermediate condition between pleasure and pain: ltaque non placuit Epicuro medium esse quiddam inter dolorem et voluptatem; illud enim ipsum quod quibusdam medium videtur, cum omni dolore careret, non modo voluptatem esse, verum etiam summam voluptatem. Quisquis enim sentit, quem ad modum sit affectus, eum necesse est aut in voluptate esse aut in dolore. Thus, Epicurus did not accept anything intermediate between pleasure and pain; what some took to be intermediate-a complete lack of pain-was not only pleasure, but also the highest pleasure. For whoever is aware of his condition, must either be in pleasure or in pain. Epicurus thinks that when we have satisfied our necessary and natural desires or needs, we will be in the most pleasant psychological (ataraxia) and bodily (aponia) conditions. Moreover, we also will have attained a condition that satisfies the formal eudaemonist requirements of completeness, invulnerability, and self-sufficiency. Given these particular theoretical interests, it would be very odd to take his denial of a midpoint as referring to states of a particular be neither enjoyed or disliked on a third occasion. If enjoying something consisted in having a sensation at the same time, then it presumably would make sense to ask whether this sensation itself were pleasant or unpleasant or neutral. To answer that it was unpleasant or neutral would produce a contradiction, to answer that it was pleasant would lead to 'a redundancy or worse.' " 35 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory homogeneous 'feeling.'65 In assessing pleasures, Epicurus examines individual desires to see whether they satisfy his formal requirements. Similarly, he wants to know whether we are in a state in which our natural bodily and psychological needs are satisfied overall and whether this state meets the formal conditions of happiness. If pleasure can be equated with the satisfaction of our natural needs and desires, he can claim that we are in a state of pleasure if and only if we meet his formal requirements, and that otherwise, we are in an unpleasant state. It clearly is not necessary that there be some third intervening possibility. In this light, we can see why Epicurus' interests are not in contrary states of a special 'feeling' but in contradictory states that either pass or fail his objective, natural tests. He can argue that certain states of satisfaction meet his requirements, however, without assuming that all satisfactions give rise to a uniform sensation. If by pleasure he means the satisfaction of certain natural needs, moreover, then it is plausible for him to claim that there is no neutral state between pleasure and pain. A natural need is either satisfied or not, just as our natural needs overall are either satisfied or not. Consequently, the notion of neutrality or indifference has no legitimate role to play in his dichotomy. 66 So far, I have used the terms 'natural needs' and 'natural desires' interchangeably, in part mirroring Epicurus' own procedures. 67 65. Here is one place where decisions about translation are crucial. Rackham, for instance, translates medium as "a neutral state of feeling," which immediately decides the issue. A more neutral translation, however, carries no such commitments. It might be claimed that the clause ("Quisquis sentit ... ") surely implies an introspective test of one's feelings. Rackham's translation strongly suggests this: "A man who is conscious of his condition at all must necessarily feel either pleasure or pain." Again, however, it seems to me that a more neutral translation does not necessarily carry a reference to 'feelings' or introspective sensations in the required sense. The Epicurean text that Cicero's account is perhaps modeled on (Stob. Flor. 17.35; U. 422 ), as well as a further report by Diogenes of Oenoanda, mention pathi and aisthisis, but there is no reason to see in this a reference to pleasure as a uniform feeling (see above, note g). 66. Similarly, when applying the formal tests of completeness, invulnerability, and self-sufficiency, Epicurus, however plausibly, avoids allowing gradations in the happiness of sages. One is happy or one is not; there is no iiJtermediate state. 67. For instance, see Ad Men. 127-28 for the way that Epicurus' arguments move easily among epithumian, endeon, deometha, agathon, phusis, and so on. This Epicurean move between 'needs' and 'desires' may account in part for another This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire Cicero complains (De fin. 11.27), however, that equating a 'natural need' (desiderium naturae) with a 'desire' (cupiditatem) begs a number of important questions. To see why, it is useful to invoke Aristotle's tripartite division of objective and subjective goods, mentioned early in this section. We might, for instance, concede to Epicurus the plausibility of linking pleasure with the satisfaction of desires, if he is using 'desire' in the sense required by (a). 68 But his attempt to give his theory a further objective, naturalistic defence by grounding it in 'natural' needs and desires is more problematic. The association of pleasure with the satisfaction of objective, natural needs-either in sense (h) or sense (c)-is far less plausible if it bypasses any reference to our subjective states, intentions, and wants. 69 Take the case of force-feeding a prisoner who is fasting in order to further a political belief. We might agree that an objective 'need' of the prisoner is being met when he is fed forcibly. It seems difficulty in Epicurus' account. Epicurus, at times, identifies hidone with the absence of pain and also of fear. There is evidence that he tends to identify the removal of pain with the satisfaction of a natural need, hence natural desire (see Ad Men. 13031, KD 18 for the expression to kat' endeian algoun). It is harder to see, however, how the removal of fear can be viewed as the satisfaction of a desire, even if it is true that having all of one's desires satisfied is a sufficient condition for the absence of fear. If Epicurus is moving between 'needs' and 'desires' in the way that Cicero suggests, it is somewhat easier to see the grounds for this conflation. We can view fear as the disruption of a natural psychological need-that is, ataraxia-hence also as the frustration of a natural and necessary desire. Similarly, it is plausible to view fear as a second order attitude focused on first-order desires. In chapter 4, I give some reasons for thinking that Epicurus tends to assimilate cases of second order attitudes and second order desires. See B. Inwood, Ethics and Human Action in Early Stoicism (Oxford, 1985) pp. 297-98, for other Hellenistic difficulties with characterizing fear. 68. To go back to an earlier example, the dispositionalist argues that the pleasure one takes from playing the piano is not separable from playing the piano. But this position admits of a subjective or objective interpretation: (a) My pleasure requires a belief that I am playing the piano; (b) My pleasure requires a true belief that I am playing the piano. By itself, (a) might allow me to lead a completely pleasurable life, even though I am continually deceived about the nature of my pursuits, my life, and the world as a whole. Epicurus is committed to defending (b) (SV 54; KD 11, 12). 69. D. Glidden ("Epicurus and the Pleasure Principle," The Greeks and The Good Life, ed. D. Depew [Fullerton, Calif., 1980], pp. 177-97), for instance, argues that Epicurus' theory attempts to bypass the need for intentional explanations altogether by assessing pleasures soleiy on the basis of their atomic configurations. 37 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory considerably more difficult, however, to agree that we can say the prisoner's 'desire' for food is being satisfied. To do so involves an equivocation between objective and subjective senses of 'desire,' since it is no longer clear that the requisite desire is still his in sense (a). Similarly, an equivocation between senses (a) and (c) accounts for the oddity of claiming that the prisoner is in a state of 'pleasure' after having his hunger forcibly satisfied. As Aristotle notices, pleasure seems to be strongly associated with voluntary actions. Just as a desire for food no longer seems to be strictly ascribable to the prisoner, in much the same way the 'pleasure' of satisfaction, in this important sense, no longer seems to be his. Epicurus agrees with dispositional theorists in associating pleasure with the satisfaction of desires. But in moving from (a) to (c) and attempting to give his account this further objective, naturalistic grounding, he might seem to be denying one important feature of dispositionalism, or indeed any plausible account of pleasure: the necessary link between pleasures and our conscious intentions or attitudes toward our satisfactions. 70 Epicurus thinks that, in assessments of our overall pleasure, a central question is whether our objective needs are being met. But this position raises further difficult problems about the role that objective and subjective perspectives play in his evaluation of needs and goods. Consequently, we will have to raise some additional questions for Epicurus' theory. If Epicurus thinks that pleasure satisfies the formal conditions of happiness, are his accounts of both eudaimonia and pleasure excessively objective? That is, can Epicurus show that for individuals to be happy, they must be "in the same state of mind we say people are in when we call them happy"? 71 If he cannot, and if pleasure is associated solely with the satisfaction of our objective needs, will not agents have cause to be anxious if they wrongly assess their present states, even if all their objective needs are being fulfilled? Or, as in the case of the prisoner who is force-fed, may they not 70. If pleasure is a mode of attending to an activity or a further description of the manner in which we are engaging in an activity, Epicurus would claim that these are merely kinetic pleasures for which we can have no rational concern. He is not interested in whether someone eats with great animation or with rapt attention, since these are mere kinetic variations. What matters for him is whether someone is meeting the objective needs of his constitution. 71. Kraut, p. 168. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire fail to find their states of natural satisfaction pleasant? In short, can Epicurus give a naturalistic, objective grounding for pleasure without undercutting the central role that our own judgments must play in evaluating our own pleasures? 72 Epicurus attempts to meet such objections, curiously enough, by relying on his formal criteria for happiness. If our happiness must be in our own power, for instance, then clearly our attitudes toward the satisfaction of our natural needs will play an important role in assessing our eudaimonia overall. Moreover, if we feel content but are deceived about our condition, our feeling of contentment will be insufficient evidence that we are happy. Our pleasures must be veridically grounded or else they will not be up to us. Unless we correctly assess the sources and doxastic status of our pleasures (De fin. I.ss; KD 8), therefore, we will not really be in full control of our happiness. It is another question, of course, whether an account of pleasure can plausibly meet such formal conditions. At the same time, however, we will need to withhold judgment on what is often taken to be a knockdown objection to Epicurus' view. Long, for example, argues that Epicurus fails to notice in his account of pleasure that "one man's meat is another man's poison." 73 If we ascribe to Epicurus either an empiricist view of pleasure or the view that pleasure is merely the satisfaction of whatever desires people may happen to have, then Long's objection would be telling. But Epicurus is keenly aware that people have different desires. He does not have to conclude immediately, however, that eudaimonia consists in whatever states or activities people pursue. We may clearly demand a defense from Epicurus of the desires he thinks we should cultivate, but this sort of objection assumes only the impossibility of an objective defense of happiness and of pleasure. 74 72. The problem here is strictly parallel to the one posed by Kraut, p. 192. Cf. Kraut's discussion and emendation of von Wright's claim that "Whether someone is happy or not depends on his own attitude to his circumstances in life .... To think that it could be otherwise is false objectivism" (The Varieties of Goodness [New York, 1963], pp. 100-101). 73· Long, Hellenistic Philosophy, p. 73· See J. Austin, "Pleasure and Happiness," Philosophy 43 (1968), 51-62. 74· Cf. J. Annas, "Aristotle on Pleasure and Goodness," in Essays on Aristotle's Ethics, ed. A. 0. Rorty (Berkeley, Calif., 1980), p. 296, for some suggestive comments 39 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory Pleasure and Belief We can reformulate Long's objection to raise another serious difficulty for Epicurus' account. We might reasonably claim that our pleasures are causally dependent on our beliefs or prior evaluations. The differing pleasures that the sensualist finds in indulgence and the Epicurean takes in temperance are equally dependent on their other beliefs about the good. Epicureans take pleasure in temperance because they believe it to be good; otherwise they would not find it pleasant. Yet if our beliefs about the good were irreducibly subjective, then it would be misguided to attempt to give an objective account of pleasure. Individuals will take pleasure in whatever they happen to believe is good (cf. Laws 6s8e-6sga.) If the dispute between Epicurus and the sensualist is over beliefs, therefore, it is no longer clear that it could ever be resolved on the basis of hedonic criteria alone. Our beliefs about the good, not the evaluation of our pleasures themselves, will be the more fundamental area of dispute. It is sometimes argued that Epicurus tries to bypass completely the role that our beliefs play in our pleasures. For instance, Cicero75 attributes to Epicurus an argument, reminiscent of Eudoxus,'6 that every living being as soon as it is born pursues pleasure as the good and shuns pain as evil (cf. EN II72b9-15). Diogenes Laertius further explains that this common desire for pleasure, shared by rational and irrational creatures alike, is taken by Epicureans 77 to be natural about the possibility of an objectivist account of pleasure and the ways in which claims about pleasure and the good are mutually related. 75· "Omne animal simul atque natum sit voluptatem appetere eaque gaudere ut summo bono, dolorem aspernari ut summum malum et quantum possit a se repellere; idque facere nondum depravatum ipsa natura incorrupte atque integre iudicante" (De fin. !.30, 11.31-32). Cf. Sextus, PH lll.194, M V.96 (U. 398), and M. Giusta, ed., I dossografi di etica (Turin, 1964), 1:124 for the importance of Epicurean vocabulary in other formulations of the cradle argument; cf. also Brunschwig, pp. 113-14. 76. Cf. Gosling and Taylor, pp. 157 ff., 346-48, for the different status of this claim in Eudoxus and Epicurus. 77· The cradle argument is not found in any surviving text ofEpicurus himself. However, Brunschwig (pp. 116-22) shows why Epicureans might feel the need for this kind of further naturalistic defense of what we as adults have come to recognize (egn.Omen; Ad Men. 129) as our proton kai suggenikon agatlwn. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire and without logos (phusikiis kai chiiris logou; X.137).justas we perceive that fire is hot or honey sweet, so we perceive that pleasure is to be sought after (expetenda; De fin. l.3o). If we take this analogy seriously, pleasure would seem to be an immediately felt sensation just like the sensations of touch or taste. Such a view of pleasure might seem to put in jeopardy many of the claims that I have been making so far about Epicurean hedonism. If the truths of hedonism are immediately given in experience, why would Epicurus care whether pleasure can satisfy any further formal conditions? Furthermore, why should formal tests govern our use of the calculus and our assessments of aponia and ataraxia, if we have such a readily available shortcut to psychological and ethical truths? For Epicureans, moreover, these connections between our sensations and our pursuit of pleasure might provide an additional valuable point of mutual support for their epistemology and their ethics. Epicurus argues that if we fail to take into account the evidence of our senses, everything (panta) will be full of confusion (taraches; KD 22). It is hard to decide from this passage whether Epicurus thinks that the kinds of confusions created by these failures are strictly epistemological, but if we keep in mind the strong ethical connotations of taraches, panta may also include a reference to our psychological states. 78 If this inference is justified, it suggests that Epicurus thinks that an important reason for relying on our senses is a hedonistic one. If we do not rely on our senses, we will be thrown into painful states of tarache. 79 Thus our desire for pleasure offers support for our reliance on perception. The cradle argument, 80 in turn, would seem to suggest that our senses immediately perceive that pleasure is to be sought after. By relying 78. Epicurus elsewhere offers a hedonistic justification of our pursuit of knowledge (e.g., KD 11). KD 22, however, may imply a stronger immediate connection between pleasure and our reliance on our senses. 79· Lucretius is no more forthcoming about the possibility of this kind of hedonistic defense of perception. It is perhaps possible, however, to see the discomfiture of the radical sceptic of DRN IV in this light. On the passage in general, see M. F. Bumyeat, "The Upside-Down-Back-to-Front Sceptic of Lucretius iv 472," Philologus 122 (1978), 197-206. So. Aptly named by Brunschwig, p. 113 (see De fin. V.55: "tamen omnes veteres philosophi, maxime nostri, ad incunabula accedunt"). 41 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory on our senses we perceive the truth of hedonism, whereas the truth of hedonism gives support to our reliance on our senses. 81 Epicurus' argument here would be circular, of course; but for present purposes, it is more important to understand what this relation between pleasure and perception is supposed to show about the nature of pleasure. Does Epicurus think the cradle argument demonstrates that we are genetically programmed to pursue a particular 'feeling' that, independently of any of our beliefs, can serve as an inner meter to gauge the hedonic value we derive from various activities? The cradle argument and Ad Menoeceum 12982 are widely believed to show that each of us has an infallible sensory meter that registers the strength of our pleasures and serves to initiate our actions and choices. Beliefs and deliberation, it is claimed, play the merely negative role of corrupting and confusing our feelings. At Ad Menoeceum 132a, for example, Epicurus recommends the use of reason to drive out those opinions that most trouble our soul. Elsewhere, we find the claim that desires that are neither natural nor necessary arise from vain beliefs, and that vain beliefs are able to distort and lead astray even our necessary desires. It might seem that for Epicurus, and so he is often read, our inner feelings of pleasure and pain could lead us through life unthinkingly, if we could only free ourselves from acquiring opinions. We therefore would be happiest returning to the state of a small child or animal. In support of such a view, Rist argues that the "feelings of pleasure and pain are the criteria of how we should act .... Pleasure, which is appropriate to us, is appropriate in the sense that it indicates courses of action which will maintain us in an untroubled state .... "88 If all beliefs should be eliminated and feelings alone should guide our actions, however, it is not clear why, for instance, we need to know about the true nature of the physical world in order to achieve eudaimonia. 84 Furthermore, Epicurus does not urge us to eliminate 81. I am thankful to C. Shields for suggesting these possible links between Epicurus' ethical and his epistemological doctrines. 82. Cf. J. Rist, Epicurus: An Introduction (Cambridge, 1972), p. 31. However, Brunschwig, p. 115, convincingly shows that Ad Men. 129 is talking about the intellectual recognition (egniimen) of the truths of hedonism; it is the conclusion of rational adults reflecting on their own beliefs and experiences. 83. Rist, p. 31. 84. M. C. Nussbaum, "Therapeutic Arguments: Epicurus and Aristotle," in The 42 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire all our beliefs, only those that are harmful. 85 For Epicurus, the pathe provide a standard of truth and a criterion by which to judge actions. 86 They therefore have the same kind of foundational role to play in our moral life that sensation does in grounding the pursuit of knowledge. Both serve to give us infallible, causal contact with the world. Yet both are alogos (ad Hdt. 38, 82.5) in the sense that they are merely the raw data of our perceptual and moral judgments. They must be sorted out and fitted together by further judgments, or prolepseis, if they are to guide our epistemological or moral judgments reliably. Thus, at Ad Menoeceum 132a, Epicurus argues that reason should drive out troubling opinions, but he does not claim that all opinions are troubling. On the contrary, at Ad Menoeceum 133, we find a list of true beliefs necessary for the pleasant life. 87 Consider the following example: If I find myself before an altar of Zeus, I will experience certain sensory and affective states. If I have mistaken beliefs about Zeus and suppose that he punishes the wicked or rewards the virtuous, I may have feelings of fear, dread, hope, awe, and so on. If I have a correct Epicurean prolepsis of Zeus Norms of Nature, ed. M. Schofield and G. Striker (Cambridge, 1986), pp. 33-34, similarly argues that the Epicurean does not wish to live the life of an untutored child. She stresses perhaps more strongly than I would, however, that the child "is taken to be a reliable and sufficient witness to the end" since it grasps everything that is instrinsically good. However, other claims about the end-the mutual entailment of virtue and pleasure, the attempt to show how pleasure, properly understood, meets the formal requirements of happiness, friendship as an intrinsically valuable part of happiness-suggest that Epicurus sometimes characterizes eudaimonia in ways that would conflict with a child's grasp of the telos; a child's grasp of the end may be uncorrupted, but it is not self-sufficient, autonomous, or defended by knowledge. (Cf. the last section of this chapter.) 85. For negative views about beliefs, see Ad Men. 132a; SV 16, 59; KD 16, 29, 30. 86. Here I follow E. Asmis, Epicurus' Scientific Method (Ithaca, N.Y., 1984), who argues that pathi are standards of truth and of action for Epicurus. For differing views, see Glidden, "Epicurus on Self-Perception," American Philosophical Quarterly 16 (1979), 297-306, who argues that pathi serve as criteria only for action, not for truth; G. Striker (Kritirion tes Aletheias, Nachrichten der Akademie der Wissenschaften in GOttingen, Philologisch-Historische Klasse, 2 [GOttingen, 1974], p. 6o) suggests that Epicurus perhaps treated the pathe as a subset of sensation only in the Kan6n. 87. He argues that we need (a) true beliefs about death based on a grasp of atomic theory; (b) a reasoned account of the telos of life; (c) true beliefs about necessity, chance, and human agency; (d) the right sort of prolepseis about the gods. 43 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory and know that the gods have no concern for human affairs, I will be in a calm and pious state. My pathe, in and of themselves, do not indicate anything about Zeus except that I have come into contact with his altar and it has causally affected me. We can explain my pathe only by ascribing further beliefs to me. Thus the pathe secure infallible causal contact with the world, since every pathos, like every sensation, 88 consists in a real causal event. By itself, however, a pathos can serve only as a sign or as a basis of inference that must be confirmed or disconfirmed by subsequent affective evidence, by prolepseis, and by our knowledge in general. Brunschwig argues that the Epicurean use of the cradle argument sets up "a delicate balance between a summons to intuition and a return to reasoning." 89 The Epicurean insists that pleasure is to be sought after (expetenda), but this normative claim is not directly derived from the evidence of animals and children, since that evidence is, in and of itself, alogon, irrational and needing further confirmation. Rather, as Brunschwig maintains, the cradle argument serves the more negative role of "authenticating the origins" of the Epicurean's normative claim by showing how an adult's beliefs about pleasure are not invalidated when we examine the unadulterated pathe of children. The observation of children is not sufficient, however, to justify the value of pleasure as a criterion. As Epicurus argues (Ad Men. 129), we, as adults, come to recognize pleasure as the end. I would like to postpone further discussion of the relation between pleasures and belief until we have had a chance to examine Epicurus' account of katastematic and kinetic pleasures. A clearer ul!derstanding of this distinction will enable us to see some further connections between pleasure and the formal conditions of hap88. See C. C. W. Taylor," 'All Perceptions Are True,'" in Doubt and Dogmatism, ed. M. Schofield,]. Barnes, and M. Burnyeat (Oxford, 1980), p. 105, for a discussion of Epicurus' causal account of perception; I take Epicurus' account of pathe to be strictly parallel, in the sense that every pathos consists of a real atomic event. 8g. Brunschwig, p. 122. It is interesting that, as in the case of friendship, later Epicureans tend to emphasize different strands of an argument that Epicurus tries to hold in balance. In this case, some Epicureans emphasize the intuitionist element in this argument, while others rely exclusively on discursive argument to show that pleasure is the telos (cf. De fin. I. 31 ). 44 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire piness and, as a consequence, the nature and scope of the ties between pleasure and our beliefs about the good. Kinetic and Katastematic Pleasures Epicurus distinguishes two varieties of pleasure: the kinetic pleasures of motion (satisfying a desire) and the katastematic pleasures of stability (having a satisfied desire). 90 A brief example might help to clarify the chief features of this distinction. We know that Epicurus postulated certain natural and necessary desires. Suppose that I am an Epicurean with a proper conception of the natural limits of my desires and that at present I am faced with the need to satisfy my hunger. Suppose also that I have a good supply of various kinds of bread, all of which will equally satisfy my hunger without any harmful consequences to my constitution. Epicurus suggests that the pleasure of eating, say, brown bread or white bread, and in the process, stilling my hunger is a kinetic pleasure. When my hunger has been satisfied and my natural constitution has been restored to a state of balance, 91 my occurrent state of satisfaction is a katastematic pleasure. Scholars have construed the point of this distinction very differently, depending on whether they think Epicurus is appealing to psychological data or to facts about our constitutions at the atomic level. 92 The priority of atomic explanations in Epicurus' theory of go. For the Aristotelian background of this distinction, see EN 1154b27-32 and the discussion of Merlan, pp. 19-20. 91. Gosling and Taylor, pp. 361-62, argue that because ofEpicurus' physicalism, he construes the value of pleasures of restoration strictly in terms of their contribution to the general physical balance of an organism. One must be careful, however, of conflating atomic and intentional explanations. Gosling and Taylor tend to move between these two explanatory levels without comment and can give the (wrong) impression that Epicurus' interest in a balanced katastima is limited solely to the material effects that certain activities have on our atomic constitutions. See the following two notes for criticism of more explicit defenses of such a view. 92. Rist argues that the distinction between katastematic and kinetic pleasures stems from their differing atomic properties. 'Katastematic' "must mean pleasure deriving from a well-balanced and steady state of the moving atoms in a sensitive organ. And perhaps kinetic pleasures are pleasures deriving from a steady, though 45 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory Ｙ ｾ＠ who argues that pleasure has been defended by David ｇｬｩ､･ｮＬ Epicureans attempt to bypass entirely the need for intentional explanations of pleasure by focusing strictly on the atomic configurations of our psychological states. This claim is misleading, since Epicurus embeds his account of kinetic and katastematic pleasures in macroscopic descriptions of desires, needs, rational preferences, and a knowledge of nature's dictates. Moreover, he derives these from our experience of pleasure and pain at a macroscopic level. Thus, although Epicurus may think that on one level of explanation pleasure is a mechanistic, atomic event, throughout his ethics he appeals to further features about pleasure, which, although ultimately rooted in atomic events, require their own explanatory level. limited and temporary change, in the state of those atoms" (p. 102). He therefore finds Cicero's distinction at De fin. Il.31 between voluptas stans and voluptas movens misleading, since the atoms of katastematic states will also be in motion. However, Rist's own appeal to atomic properties is misleading in this context. Like Aristotle, Epicureans surely might defend this distinction by appealing to the differences between states of stable satisfaction and the states involved in satisfying desires. Cicero's account of the distinction stays securely on a macroscopic explanatory level and appeals to features of our psychological states. 93· D. Glidden argues that Epicurus' "confidence in our ability to detect the feelings, or pathi, of pleasure and pain does not rest on the certainty of a Cartesian self-consciousness, but rather on the material identity of these pathi with atomic motions in our bodies, understanding these psychophysical experiences, with Freud, in mechanical terms" ("Epicurus and the Pleasure Principle," in The Greelcs and the Good Life, ed. D. Depew, p. 184). Glidden is right, I think, to deny that ascriptions of pleasure ultimately rest for Epicurus on the certainty of reports based on introspection. Glidden's general view of Epicurus as an eliminative materialist is misleading, however (see chapter 4 for a more detailed discussion). He claims, for instance, that the distinction between kinetic and katastematic pleasures distinguishes not two kinds of intentional states but "two types of pleasurable atomic episodes" (p. 18g). But even in attempting to formulate this distinction strictly in atomic terms, Glidden must continually rely on descriptions of macroscopic properties and states (see especially p. 18g). Epicurus, I think, appeals to supposedly natural facts about our desires, intentions, and psychological states at the macroscopic level to justify this distinction. Even if he thinks that material explanations of pleasurable states can be given in principle, he does not need to eliminate all reference to intentional states to give objective natural criteria for distinguishing among pleasures, as Glidden supposes; he can rely on his defense of natural and necessary desires to show which satisfactions give rise to ataraxia. A word of caution: Glidden's idiosyncratic use of 'psychological' to refer to atomic explanations and 'moral' to refer to macroscopic or intentional explanations makes his argument sometimes unduly obscure. In discussing his views I use 'psychological' to refer to intentional explanations at the macroscopic level. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire To see the force ofEpicurus' distinction between types of pleasure at the level of our psychologies, we can return to our example. On the empiricist account of pleasure, it is possible for someone to fail to get any pleasure from stilling a desire or satisfying a need. If I am hungry and faced with another dull meal of brown or white bread, I may gain no pleasure from satisfying my hunger, if pleasure is construed as a separable feeling over and above the activity of eating. Perhaps while eating, my mind was elsewhere, or perhaps I was really in the mood for steak and lobster. For the empiricist, the mere stilling of a desire is not enough to qualify as a pleasure; an activity must give rise to the proper accompanying subjective sensation. Similarly, on this view, it is possible to satisfy a desire or need, but then continue to derive further pleasure from an activity: although no longer hungry, I might continue to take pleasure in eating a rich dessert. Such assumptions about pleasure, however, appear to be exactly what Epicurus means to combat with his. distinction between katastematic and kinetic pleasures. He focuses on the question of whether the natural needs of our bodily and psychic constitution are being fulfilled by the things that we desire. Consequently, he argues, desires for goods that satisfy these natural needs are natural and necessary. When the needs of our constitution have been met, we will have achieved a pleasurable state of katastematic balance. Epicurus denies, however, that we could ever satisfy a natural need and fail to gain pleasure. Unlike the empiricist, Epicurus argues that the various kinetic states94 that occur in satisfying genuine needs do not affect our overall satisfaction; they are mere variants95 that can give us no rational grounds for preference. 94· It is perhaps worth asking whether Epicurus must treat painful kinetic states with similar indifference. For instance, given that we want to be in a katastematic state of good dental health, do we have any rational preference for having our teeth extracted with or without anesthetic? By parity of argument, Epicurus must hold that we can try the kinetic variations of Novocaine on a whim, if it is readily available; but our overall experience in the dental office, with or without Novocaine, will be no more or no less pleasant. 95· The view that kinetic pleasures are mere variants has been defended by Rist (pp. 17off.) on the basis of KD 18 and De fin. II.g. Long argues that some kinetic pleasures are a necessary means to katastematic pleasure (Hellenistic Philosophy, p. 65). For present purposes, however, on either view kinetic pleasures cannot add to 47 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory The empiricist might argue that this highly unintuitive doctrine can easily be embarrassed by introspection. Even if we grant Epicurus his claim about the importance of satisfying natural needs, still, surely someone who prefers white to brown bread, and satisfies her hunger with white bread, will have a more pleasant meal overall than if she had to make do with brown bread. Should we not agree with the empiricist that Epicurus' story about the satisfaction of natural needs cannot capture all we need to say about pleasure? Dispositionalists will also object to this element in Epicurus' theory, since an important feature of their view is the claim that pleasure is linked in crucial ways to our manner of engaging in or attending to an activity. Epicurus admittedly does not seem particularly interested in whether someone eats with great animation or with rapt attention, carefully savoring every mouthful. 96 Clearly, Damoxenus (Kock, frag. 2) is merely wrong to suggest that Epicurus thought we should squeeze out the maximum of pleasure by "chewing carefully." If anything, Epicurus' continual focus on the satisfaction of natural needs can give the impression that he is completely indifferent to subjective states of conscious awareness. 97 We might think, then, that for the Epicurean, it merely will be a matter of kinetic indifference whether someone eats with great animation or bored rethe completeness of an overall state of katastematic pleasure. Epicurus apparently feels little pressure to include kinetic pleasures as parts of our final good. The sage will undoubtedly have kinetic experiences but cannot assign them any rational value. I will argue that he feels considerable pressure from his own demand for the completeness of eudaimonia to include both the virtues and friendship as parts of our telos. 96. Such a view is suggested by Diano's interpretation of DRN IV.627-29: "Deinde voluptas est e suco fine palati; I cum vero deorsum per fauces praecipitavit, I nulla voluptas est, dum diditur omnis in artus." (See "Note Epicuree," Studi italiani di filologia classica 12 (1935), 253, and "La psicologia d'Epicuro e Ia teoria delle passioni," Giomale critico della filosofia italiana 20 ( 1939, 1940, 1941, 1942 ). Diano takes this passage as evidence for his general claim that kinetic pleasures supervene on prior katastematic states. In this case, the kinetic pleasures of taste cease (nulla voluptas est) once food passes from the mouth to the rest of the body. But katastematic equilibrium is presupposed throughout this passage, and, consequently, these kinetic pleasures are a matter of indifference. For a differing view of this passage, see Gosling and Taylor, p. 376. 97· For instance, in the Epicurean definition of pleasure as the stable condition of the flesh (sarkos eustathes katastima), healthy function seems to be stressed as opposed to subjective sensation (cf. Gellius IX.5.2; U. 68). This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire luctance, or, indeed, has to be force-fed. 98 What will matter most are the objective questions of whether one's natural needs are being met and whether one's katastematic equilibrium has been achieved. To see the precise force of Epicurus' claims, however, we need to recall his reliance on the formal conditions of happiness. He argues that our final good must be entirely in our own control; therefore, since pleasure is our final good, our pleasures must be completely in our own control as well. Accordingly, by claiming that kinetic pleasures are mere variants and readily substitutable for one another, Epicurus greatly decreases the vulnerability of our pleasures. 99 If I can achieve the same katastematic satisfaction from a wide range of goods, then I greatly increase my ability of avoiding the frustration of my desires. Since some goods may be hard to secure and dependent on chance features of the world, my preferences for the kinetic pleasures associated with such goods would make me more vulnerable. Yet if Epicurus thinks that our pleasures must be in our own power (par' hemas), he must also allow that our rational assessments, beliefs, and conscious attitudes play a crucial role in our happiness. My knowledge that I can be just as happy eating brown bread or 98. We might think that in the latter example the links between katastematic states and psychological explanations of pleasure in terms of desire, wants, and so on have been completely broken. If Epicurus can claim that the force-fed prisoner is in a pleasurable condition of satisfaction, we might suspect that he is describing only an underlying atomic configuration and bypassing the prisoner's intentional states. This sort of extreme case, however, is merely an instance of the general difficulties Epicurus must face in linking pleasure to the satisfaction of natural, objective needs. This example also raises further questions about choice, responsibility, and the importance of autonomy, which I take up in the last chapter. 99· Gosling and Taylor, pp. 373-96, claim that Epicurus has no theoretical motivation for distinguishing katastematic from kinetic pleasures in the way that Cicero suggests. It seems to me, however, that one very clear motivation for making kinetic pleasures mere variants (Diano, Rist) or substitutable means (Long) is Epicurus' attempt to ensure the invulnerability of pleasure. Gosling and Taylor argue that Cicero's account of the distinction between kinetic and katastematic pleasures is unsympathetic and confused, and shows little understanding of his Epicurean sources. I strongly disagree, but I doubt that it would be worthwhile to attempt to meet their many detailed objections here (I try to do so in a forthcoming commentary on De finibus I-II; cf. ad loc. !.37, 39, Il.9, 10, 16). For present purposes, either Long's or Diano's account is compatible with my view that kinetic pleasures and katastematic pleasures are incommensurable and that kinetic pleasures are a matter of rational indifference to our overall happiness. 49 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory white bread is important to my overall happiness and allows me to react to the incursions of chance with minimal risks of pain and frustration. It gives me confidence that my desires will continue to be satisfied. 100 Epicurus thinks, therefore, that when we are eating bread, it is wrong to focus on our subjective awareness of kinetic pleasures, since by concentrating on these we do not give expression to our autonomy and self-sufficiency. Rather, by being aware of our ability to substitute or eliminate such pleasures, we gain a stronger sense of our autonomy and self-sufficiency in the face of the world's sometimes threatening randomness. Thus, Epicurus thinks that his attention to this requirement of happiness will sufficiently account for the role that subjective states play in eudaimonia. We must display the proper rational attitudes and exert the right sort of control over the satisfaction of our objective needs. Otherwise, we will be subject to tuche. Although we can see how this distinction between katastematic and kinetic pleasures meets the demands for the voluntariness of our happiness, it would appear to fail Epicurus' further condition that our happiness be complete. We might reasonably object that our satisfactions will be more complete if, in addition to satisfying our natural and necessary needs, we can also include a wider range of kinetic pleasures in our lives. Similarly, wouldn't our happiness be greater if we could experience a fuller array of goods and their consequent kinetic pleasures? Epicurus' attempt at an answer to this question shows how he must subordinate his requirements for completeness to the demand that our happiness be under our control. He suggests that we can try various kinetic variants as long as we can do so without running risks or developing inflexible attachments to the goods that give rise to them. He denies, however, that richer kinetic experiences can add to our happiness or make it more complete. Epicurus' account of the relations between these two requirements reverses Aristotle's priorities and gives rise to a contrasting evaluation of external goods, the value of practical reasoning, the range of natural capacities, and death. Aristotle subordinates voluntariness to completeness because he thinks that some goods nee100. Cf. Orig. Contra Celsum III.So: to peri tautis piston ... elpisma; cf. U. 68. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire essary for happiness may be vulnerable to loss. Epicurus maintains the invulnerability of happiness, but he lets the scope of satisfactions expand and contract to adjust to individual circumstances. He argues that in such a way we can still achieve complete happiness; but it is difficult to see how he can plausibly defend this claim without in some way reducing the strength of his commitments to invulnerability. Pleasure, Ataraxia, and Aponia Epicurus thinks that there are certain desires whose satisfaction is necessary for happiness (pros eudaimonian; Ad Men. 127). He further claims that someone who satisfies these desires will be in a state of aponia and ataraxia. In identifying these states with eudaimonia, Epicurus relies on the assertion that a necessary condition for happiness is the satisfaction of our natural and necessary desires. It will be useful, however, to raise some objections to his account to see how he can defend this identification of eudaimonia with both aponia and ataraxia. (1) Why should someone impressed with Epicurus' arguments about happiness and pleasure not cultivate a very narrow range of easily satisfiable desires and thereby avoid the dangers of frustration? It is difficult to see how Epicurus can avoid the inference that we should select our desires solely on the basis of how easily they can be satisfied. In the Philebus, Plato argues that the hedonist will have to admit that the life of a contented jellyfish is a happy one (2oc), if pleasure or the satisfaction of desire is the sole good and nothing else can be added to it to make our lives more complete. Can Epicurus give a robust enough account of hedonism to make it more attractive than Plato claims? In other words, can an Epicurean's happiness really be complete? (2) Cicero, on behalf of the Cyrenaics, argues that ataraxia is the summum bonum of a corpse (De fin. 11.22). If we think that pleasure arises from the satisfaction of our desires, should we not try to satisfy as many as possible and as often as we can? If I have satisfied one desire, that will just give rise to another, which also will require This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory satisfaction. Instead of imitating corpses, should we not live a life of restless desire, moving eagerly from satisfaction to satisfaction? (3) Finally, one might raise a Cyrenaic objection. Why should we care at all about the satisfaction of our future desires? If I develop second-order attitudes and desires about future satisfactions, is it really possible to achieve the ataraxia that Epicurus recommends? Will I not rather be continually worrying about my future states and consequently spoiling my present pleasures? It might be helpful to deal with the last two objections first, since Epicurus thinks that (2) will collapse into a version of (3). The view of desires endorsed in (2), familiar from Plato's Gorgias and often attributed to Hobbes, 101 holds that happiness consists in the continual movement from desire to desire and from satisfaction to satisfaction. As soon as we have satisfied one desire, we will immediately feel new urges that we must satisfy. In this view of happiness, if all of an agent's desires were ever satisfied, the resulting state, by definition, could not be a happy one. States of ataraxia and aponia, in this view, would be conditions in which our desires have lost their motive force. Consequently, there could be no future satisfactions or pleasures awaiting us. 102 As Hobbes argues, "The Felicity of this life, consisteth not in the repose of a mind satisfied. For there is no such Finis Ultimus, (utmost ayme,) nor Summum Bonum, (greatest Good,) as is spoken of in the Books of the old Morall Philosophers. Nor can a man anymore live, whose Desires are at an end, than he, whose Senses and Imagination are at a stand. Felicity is a continuall progresse of the desire, from one object to another; the attaining of the former, being still but the way to the later" (Leviathan, pt. I, ch. 11). 101. This conception of desire is explored by T. Irwin in "The Pursuit of Happiness" (unpublished) and in "Coercion and Objectivity in Plato's Dialectic," Revue lnternationale de Philosophie 156-57 (1g86), 65-74. In the following section, I am greatly indebted to his discussions and to conversations with J. Whiting and M. Neuburg. Contrasts between Epicurus and Hobbes are taken up in a rather different manner by J. Nichols, Epicurean Political Philosophy: The" De Rerum Natura" ofLucretius (Ithaca, 1972), p. 183. 102. It is not always clear whether Epicurus thinks of ataraxia as a desireless state, in the sense of a total removal or lack of desire, or as a state in which desires are continually satisfied. Sometimes this ambiguity creates a related confusion about the range of our desires (see Ad Men. 128), since Epicurus is ambiguous about whether he requires a broad range of desires or just the fewest possible for happiness. 52 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire Epicurus takes the proponents of this view of pleasure and desire to be 'sensualists' who claim that one should abandon all restraint and continually develop stronger appetites. In Epicurean terms, they wrongly value kinetic as opposed to katastematic pleasures. Epicurus tries to counter this extreme claim with his arguments about the virtues. Anyone interested in living a life of pleasure will have to be temperate, courageous, wise, and just (Ad Men. 132; De fin. !.42-54). If we value our future states of satisfaction, we may have to restrain some present desires in order to satisfy future ones, or we may have to restrain or eliminate desires whose present satisfaction may have disastrous future results. At the same time, we must be careful about attaching the right value to our future states. Too strong a desire for security or survival, for example, might lead one to develop derivative desires for power, money, or wealth (DRN V.1120-35; KD 7). Epicurus thinks that those who are interested either in maximizing future satisfactions or in acquiring power and wealth must agree with him that concern about future satisfaction is important. Consequently, they should not cultivate desires that are not amenable to restraint. Although sensualists may disagree with him about the best means to future satisfaction, they too will begin from this point of agreement; they will reject the radical Cyrenaic claim that we should not worry about our future states at all. Therefore, they will be just as concerned as the Epicurean with rationally ordering desires to insure future satisfactions. Epicurus portrays his differences with the sensualists as over the means to an agreed-upon determinate end: pleasure (KD 8, 1013). He suggests that the sensualists agree with him about the contents of happiness and its formal requirements but disagree about how best to. achieve this end. He argues that if the sensualists were able to attain ataraxia and aponia, even without a knowledge of phusis, death, and the limits of desire, they would be happy. He denies, however, that this could ever happen and outlines what he takes to be the necessary means for achieving both aponia and ataraxia. Here we need to ask whether the sensualists agree that they really want aponia and ataraxia or whether they disagree with Epicurus about the contents of happiness as well. The sensualist can endorse 53 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory Epicurus' view of the value of future states without agreeing that his is the only plausible conception of happiness linked to this requirement. To justify his identification of eudaimonia with his conception of pleasure, Epicurus must invoke the further formal requirement that our happiness must be entirely in our own control. We should care, he argues, only about those future states that are par' hemas. And the most plausible candidates for satisfying this condition are inner states that are invulnerable to chance and frustration. The controversy between Epicurus and the sensualists is thus over a formal requirement of happiness, not a particular type of feeling. Here again, a comparison with Sidgwick's view of the dispute between philosopher and sensualist might be instructive. Sidgwick claims that their disagreement arises from different assessments of subjective feelings. He further argues that no one can tell whether "the philosopher's constitution is not such as to render the enjoyments of the senses, in his case, comparatively feeble; while on the other hand the sensualist's mind may not be able to attain more than a shadow of the philosopher's delight." 103 Epicurus does not think that his dispute with the sensualists is over their respective capacities for enjoying particular qualities of feeling. Rather, he challenges the sensualists' ability to satisfy the formal requirements of happiness. The sensualist's pleasures, he argues, will be vulnerable to chance and frustration, and they will be incomplete. 104 However, Sidgwick's remarks show why it is unclear that these formal conditions can be justified on the basis of hedonic criteria alone. Epicurus' difficulties over an adequate justification of hedone as our final goal will emerge more clearly if we briefly examine some 103. Sidgwick, p. 148. 104. We might think that the sensualists' pleasures are more complete in that they cover a greater range of satisfactions. Both Mill and Epicurus claim that the sensualist's pleasures will in some way be incomplete. It is important, however, to distinguish Mill's conception of completeness from Epicurus'. Mill thinks that certain capacities and accomplishments are necessary ingredients in complete happiness. For pleasure to be complete, it must encompass a sufficiently rich range of activity. Epicurus regularly takes the completeness of pleasure in a different sense. He does not argue that an unsatisfied Metrodorus is better than a pig satisfied; rather, he takes completeness to be a feature linked to the limits of individual pleasures and desires. I have complete pleasure when I have reached these limits, not when I have satisfied a large range of desires and capacities. 54 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire of the objections raised in (3). The Cyrenaics claim that one should not take the trouble of rationally modifying one's desires in order to secure future states of satisfaction. They similarly deny the possibility of coherently ordering desires for some more ultimate goal. In the context of Greek ethics, this claim is radical, since most Greek moralists assume that without an ultimate structure to relate and adjust our various aims and desires, our individual satisfactions, goals, and so forth will form a disordered heap. Since eudaimonia consists in this correct structuring of desires, the structure as a whole becomes more ultimate and choiceworthy than any of its individual parts. Similarly, prudential calculation to secure future fulfillment becomes a central element in this overall structure. In replying to the Cyrenaic who claims to dissociate himself from his future states and desires, Epicurus again must rely on the formal conditions for happiness. Although this reliance may give him a more compelling view of happiness, it also shows why his account cannot be defended on the basis of his hedonism alone. Epicurus argues against the Cyrenaic that mental pleasures are greater than bodily pleasures (De fin. 1.55). The chief reason for the priority of mental pleasures is that they are temporally extended in a way that pleasures of the body are not. To show that this kind of temporal extension is valuable, Epicurus appeals to a conception of our personal identity, which is similarly extended in time. Rational agents must abstract themselves from their present situations and give equal weight to their future interests and desires. 105 Prudence and practical reasoning are therefore not eliminable if we are concerned about our satisfactions (KD II, I2 ). Nagel, for instance, argues that it would "be wildly peculiar for someone to be unmoved by the possibility of avertable future harm or accessible future benefits." 106 Although Epicurus might disagree with Nagel's 105. Two other Epicurean claims seem to conflict with this demand for prudential calculation of future interests. Epicurus denies that death can harm us by cutting short any of our extended projects, and he denies that duration will increase pleasure. Both these claims are influenced by his strong emphasis on the voluntariness of happiness. Epicurus tries to show how we can take future states seriously by limiting our concern to those that are immune to chance and not enhanced by duration. 106. T. Nagel, The Possibility of Altruism (Oxford, 1970), pp. 37-38. Cf. N. P. White, "Rational Prudence in Plato's Gorgias," in Platonic Investigations, ed. D. J. O'Meara (Washington, D.C., 1985). 55 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory conception of harm and benefit, he can agree about the need for prudence and a concern for future states of an extended self. But can he do so entirely on hedonistic grounds? He and the Cyrenaics disagree about the hedonic benefit of showing concern for future states of ourselves. 107 The Cyrenaic argues that we will gain more pleasure by viewing ourselves as discrete, momentary selves enjoying momentary episodes of pleasure. Epicurus, however, thinks that we will gain more pleasure by regarding ourselves as selfsufficient sages rationally planning for our future. Epicurus can argue that he is explaining more plausibly the ordinary belief that eudaimonia is not just a momentary episode but a stable, persisting condition; similarly, he can claim to be offering a more plausible conception of personal identity, that is, one that requires more than a series of discrete, unrelated, episodic 'selves.' But to make this counterargument, he must appeal to nonhedonic criteria derived from the independent formal conditions of eudaimonia. In much the same way, these considerations will affect our views about the value of developing second-order desires focused on our first-order desires. The Cyrenaic claims that developing secondorder attitudes to our satisfactions is misguided, since we will not be worried about satisfactions that are really ours; our plans will really be for 'other selves' for whom we have no rational concern. Epicurus disagrees, but in some ways his difficulties in justifying concern for our future selves anticipate the problems he faces in justifying concern for others' interests, something he attempts to do in his account of friendship. 108 In each case, he must show how hedonic criteria can give us reasons to take a more expansive view of our present interests. But at the same time, he insists that this more expansive conception must be compatible with the demand 107. We need to distinguish two questions: (a) I get more pleasure from the belief that I am a discrete, momentary self with no concern for the future. (b) I get more pleasure from the true belief that I am a discrete, momentary self with no concern for the future. For present purposes, I am only ascribing (a) to the Cyrenaics. 108. Chapter 3 takes up Epicurus' problems with justifying concern for friends' interests. This analogy between friends and future selves is explored with great subtlety by J. Whiting, "Friends and Future Selves," Philosophical Review 95 (1g86), 547-80. This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Pleasure, Happiness, and Desire for invulnerability. We might object that similar inconsistencies are inherent in his defences of prudence and of friendship. just as he cannot justify a sufficiently rich account of friendship without modifying his demand for invulnerability, so this same formal demand will keep him from justifying a sufficiently wide range of future interests. Epicurus urges us to show concern for our future states only if they are par' himas. But can he defend plausible conceptions of phronesis and eudaimonia without modifying this demand? At this point, we can return to our initial question (1). If we are convinced that happiness must be entirely par' himas, why should we not reduce our chances of frustration by cultivating the barest minimum of desires? Or, if we can be completely happy on the rack, as Epicurus maintains, can the Epicurean give us any rational justification for desiring a life without torture? Epicurus argues that in order to be happy we must be virtuous. We might think that here would be a promising place to look for a justification of a more expansive view of individual development. On this score, however, his conception of virtue will leave us disappointed. Epicurus, as we shall see, attempts to redescribe the virtues so that they will be compatible with the virtuous agent's invulnerability. Epicurean justice consists of a strong noninterference claim: we are just if we can restrain our desires and avoid interfering with others. We might think that, given such a view, we will have even more incentive to pare away our desires to the barest minimum. Surely we will be able to avoid provoking conflicts and thereby increase our chances ofbeingjust if we reduce and eliminate as many of our desires as possible. Epicurus' account of courage and temperance initially seem more promising. He argues that we should display our courage by facing up to strong desires that will have harmful future consequences; similarly, we need to know how to moderate our desires to insure that they are satisfied more advantageously. This argument seems to suggest that we may choose to undergo a fair amount of frustration in pursuit of wider goals. Thus we might think that Epicurus' defense of these virtues would enable him to defend our acquisition of a broader range of desires. Epicurus thinks, however, that he can derive his account of courage and temperance from his account of pleasure. But if he conceives of pleasure as the satisfaction of 57 This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms Epicurus' Ethical Theory desires, a justification of virtues that allow or enjoin the frustration of desires would be problematic. He would need a further evaluative criterion, independent of pleasure, to justify frustration for the sake of the further development of desires. 109 Given his commitment to the voluntariness condition, moreover, it becomes difficult for Epicurus to justify any frustration of desire whatsoever. Thus he seems constrained by this formal requirement to give a somewhat anemic and unattractive account of happiness, but if he attempts to weaken his commitment to this par' himas condition, he can no longer defend his particular conception of happiness or his account of hidone; if he grants that we should develop desires for activities outside of our control, other conceptions of happiness will pass this more relaxed requirement. The sensualists, for instance, can claim that the added risk of pain and frustration from external sources is necessary for pursuing greater hedonistic rewards. The Peripatetic can similarly claim that our final good is composed of several activities, capacities, and goods that are subject to tuche. Epicurus, then, is caught in a dilemma, yet one that poses interesting challenges for the rest of his ethical theory. He relies on a plausible conception of happiness as the satisfaction of our 'natural' aims and desires and outlines a strategy for rationally structuring and satisfying them. If we want to satisfy our desires in certain circumstances, it seems reasonable to adjust and limit them to insure their satisfaction. It is also reasonable, however, to risk the possibility of some frustration in achieving our goals and developing our capacities. Epicurus finds the anxiety of possible loss more painful than the promise of potential, though perhaps vulnerable, pleasures. Consequently, his inability to justify the latter alternative, combined with his unwillingness to modify his par' himas claim, generates widespread difficulties in his ethical theory. To understand the further effects of his commitment to invulnerability on his ethical theory, it is now time to tum in more detail to his arguments about the virtues and friendship. 109. Nor would his account of responsibility offer opportunities for a wider conception of development. Epicurus derives his theory of responsibility from his account of pleasure and concludes that we are responsible only for an inner condition that is invulnerable to chance. (See chapter 4.) This content downloaded from 128.122.149.96 on Thu, 09 Aug 2018 16:00:25 UTC All use subject to https://about.jstor.org/terms