PHIL 202; Fall 2011 Greek Ethics David O. Brink Handout #10: Happiness and Virtue in Aristotle's Nicomachean Ethics

Transcription

1 Draft of PHIL 202; Fall 2011 Greek Ethics David O. Brink Handout #10: Happiness and Virtue in Aristotle's Nicomachean Ethics Though three works on ethics are often attributed to Aristotle - - the Nicomachean Ethics (NE), Eudemian Ethics (EE), and the Magna Moralia (MM) - - the NE appears to be the most comprehensive and mature, as well as the most clearly authentic of the works. 1 We shall focus on the NE, for the most part, relying on the EE and MM only when they shed light on issues in the NE. As NE book i makes clear, Aristotle shares the eudaimonism of his predecessors; his conception of ethics is based upon a conception of what sort of life will secure the agent's own eudaimonia or happiness. Like Socrates and Plato, Aristotle believes that virtue (arete) is a state of character whose exercise promotes the agent's happiness. THE FINAL GOOD Aristotle initially describes his inquiry as a form of political science: all sciences, actions, decisions aim at some good; political science is a superordinate science that aims at the final good for an individual or, better still, for a city (1094a1-3, 1094b5-8). Most of the NE is concerned with the nature of the final good for an individual, although, this good involves the good of others as well. Why a final good? He begins the NE by arguing that... every action and decision seems to aim at some good; hence the good has been well described as that at which everything aims [1094a2-3]. However, it doesn't seem to follow from the fact that we always aim at a good that there is one good at which we always aim. We might have a number of distinct ultimate aims. Understood as a straightforward psychological claim, eudaimonism may seem somewhat dubious. But we should notice two things. First, Aristotle insists that our ultimate aim can be a complex whole whose parts we desire both for their own sakes and for their role in a good life (1097a31- b6). If so, psychological eudaimonism need not assume all our desires are instrumental to some unitary end. We can recognize a number of distinct things as intrinsically valuable. Second, Aristotle's initial concern is normative, not simply descriptive. He thinks that a rational person will impose structure on the goods she seeks. On these points we must first notice that everyone who is capable of living in accordance with his own decision sets up some goal of living finely - - either honor or 1 Though all three works are Aristotelian in nature and provide valuable evidence about Aristotle's ethical theory, only the NE appears to be uncontroversially authentic. It is very uncertain whether the MM is a work of Aristotle's or one of his students, though that needn t prevent the MM from providing reliable information about Aristotle s ethical views. Though it is more common to view the EE as authentic, some scholars dispute this. Three books (NE v- vii and EE iv- vi) are common books. Those that regard the EE as authentic often suppose that the EE is earlier, though this is not undisputed.

2 2 reputation or wealth or education with reference to which he will do all his actions; for not having one's life organized with reference to some end is a sign of much folly [EE 1216b6-11]. An overall structure of goods is arguably part of rational action. We think it a mark of irrationality if someone consciously prefers and so pursues good x, rather than y, even though y is more important to her. We would think it irrational to pursue x unless it is true that x does not conflict with some greater available good. This requires a synchronic structuring of goods. But similar considerations suggest the need for a diachronic structuring of goods. Because I am a temporally extended being and this fact conditions my present desires in the sense that my present desires all have future objects, it would be irrational for me to pursue my present desires and ends at the expense of the desires and ends that I will have. So it will be rational for me to pursue some present good x just in case it contributes to a greater overall good than alternatives. The demand for a well- integrated, diachronic conception of the good is the demand for a final good. The final good is that for whose sake we pursue other things but which - - alone - - we pursue for its own sake (1094a18-19). If it's true that action is rational insofar as it aims at one's final good and we conceive of ourselves as rational agents, then it may be appropriate to see our all actions as implicitly and imperfectly directed at a final good. If one fails to promote one's own final good, this will be due to mistakes of fact (e.g. about the appropriate means to the good), mistaken beliefs about the final good (e.g. about its constituents), or weakness of will. EUDAIMONIA Everyone agrees that eudaimonia is the final good; they disagree in their conceptions of eudaimonia (1095a17-21, 1097b22-3). Here, as elsewhere, his method is dialectical: he begins by examining the appearances (phainomena), especially common and respected beliefs (endoxa), including those of his predecessors. As in other cases we must set out the appearances, and first of all go through the puzzles. In this way we must prove the common beliefs... ideally, all of the common beliefs, but if not all, then most of them and the most important. For if the objections are solved, and the common beliefs are left, it will be an adequate proof [NE vii b4-7]. Employing this method, Aristotle begins (in I 5) by examining three common conceptions of eudaimonia. He notes that there are three lives they should examine the life of enjoyment, the life of political activity, and the life of contemplation because these lives reflect different conceptions of happiness (1095b15-19). He goes on to discuss three conceptions of happiness, though he indicates that they will need to examine the life of contemplation and its conception of happiness later, presumably at x 7-8 (1096a5). pleasure honor virtue

3 3 Presumably, pleasure is the conception of eudaimonia embodied in the life of gratification and both honor and virtue could be associated with a life of political activity or practical reason. Though each conception provides part of a plausible conception, none is adequate by itself (1095b a4). a. The life of (pure) pleasure is fit for grazing animals, not for humans; b. The life of honor is too passive and is too much outside agent's control; and c. The life of virtue is close to being right, but is not complete because it may lack of various goods that are not within our control, viz. external goods (e- goods). These criticisms reflect three assumptions about eudaimonia. 1. The final good must be complete. 2. The final good must be an appropriate life for human beings. 3. The final good should be relatively stable and within our control. This discussion is not supposed to be self- contained or decisive. These assumptions and their implications are elaborated in the rest of NE i and x. COMPLETENESS, SELF- SUFFICIENCY, AND CHOICEWORTHINESS Aristotle claims that eudaimonia is complete (teleios), self- sufficient (autarkes), and choiceworthy (hairetos) (1097a27- b22). Incomplete goods are not chosen for their own sakes; they are mere instrumental goods. By contrast, complete goods are chosen for their own sakes; they are intrinsic goods. Unconditionally complete goods are chosen for their own sakes and not chosen for the sake of anything else. Eudaimonia is the only unconditionally complete good. This means that merely complete goods are good in themselves but are also chosen for the sake of eudaimonia, perhaps as parts are chosen for the sakes of the wholes of which they are parts. It follows that there are two different ways in which something may be chosen for the sake of something else. Sometimes when x is chosen for the sake of y, x has no value itself and is a mere instrumental means to producing something else that is valuable. Here, x is an incomplete good. However, in other cases, x is chosen for the sake of y, where x is valuable as constituent of y. Here it has contributory value and is good in itself. This idea that something might be good both in itself and for the sake of the larger whole of which it is a part is not unfamiliar. My philosophy articles have structure; they defend larger aims by a series of arguments and so have constituent sub- aims. When I work on a particular sub- argument, I want to get that argument right both for its own sake and for or because of its constituent role in my larger argument. Aristotle will claim that the virtues are complete goods. In doing so he makes explicit the sort of assumptions Plato must make about the relationship between justice and eudaimonia in Republic ii.

4 4 So the only unconditionally complete good is eudaimonia. Aristotle also tells us that it is self- sufficient and most choiceworthy. 2 We might suppose that a self- sufficient good will be one which one can have all on one's own. But Aristotle denies this, insisting that we are political animals and that our happiness depends upon family, friends, and fellow citizens (1097b10-11). Instead, he means that a complete good contains everything within itself; it must be lacking in nothing (MM 1184a7-13). But then a complete good must be most choiceworthy. A good that could be improved by adding other elements would not be complete and would not be most choiceworthy. One does wonder if there is or should be any realism constraint on the possible additions that might improve a life. The Big Rock Candy Mountain describes a hobo s paradise in which, among other things, the handouts grown on trees, all the cops have wooden legs and the bulldogs all have rubber teeth, and there s a lake of stew and of whiskey too. If we take literally the idea that a complete good should be lacking in nothing that could improve it, are we committed to thinking that we can only be happy on The Big Rock Candy Mountain? That might seem to be an absurd sort of utopianism. Moreover, elsewhere, Aristotle explicitly eschews utopianism. For instance, in discussing the appropriate sort of political constitution, Aristotle's chief concern is a special kind of non- ideal theory that aims at the best feasible constitution. We have now to inquire what is the best constitution for most states, and the best life for most men, neither assuming a standard of excellence which is above ordinary persons, nor an education which is exceptionally favored by nature and circumstances, nor yet an ideal state which is an aspiration only, but having regard to the life in which the majority are able to share, and to the form of government which states in general can attain [1295a25-32]. But in his discussion of completeness in the NE, there is no explicit mention of such a realism constraint. However, there might be a different way to deal with the worry about utopianism. For Aristotle, eudaimonia and completeness are both maximal concepts. One is only eudaimon or has a complete good if one is as happy as can be or one s happiness is perfectly complete. So far, this may just seem to restate the problem. But even though few are happy because few have lives lacking in nothing (imaginable), one person can be more happy than another and also, presumably, happy enough. This would allow Aristotle to say that the completeness of eudaimonia requires that it literally be lacking in nothing good, so that one is only truly happy on the Big Rock Candy Mountain, but that nonetheless people can be happy enough in less utopian circumstances. I don t know if this silences the utopianism worry about completeness, but it may take some of the sting out of it. THE HUMAN FUNCTION Aristotle believes that a proper account of our final good depends upon an account 2 It is tempting to equate completeness, self- sufficiency, and choiceworthiness. But notice that while particular goods are complete or incomplete, it seems that only the unconditionally complete good (eudaimonia) is self- sufficient and most choiceworthy.

5 5 of what sort of beings we are and what our function is (1097b a16). The function of craftsmen and artifacts consists in their performing their characteristic activities, and if appeal to our function is to ground an account of our final good, it looks as if Aristotle must claim that performing these activities well is good for the thing for whom these activities are its function. Do people have functions? Artifacts (e.g. knives) have functions (e.g. cutting well), because they are designed to perform these activities (e.g. to cut). Aristotle thinks that natural organisms can be ascribed functions even if they are not the result of intelligent design. His general account of teleology implies that an organism or a part of an organism has a function just in case it is a goal directed system in which the achievement of certain goals or states explains the behavior of the system. An account of x's good must reflect what x's function is or what x essentially is. De Anima gives an account of the functions (souls) of different organisms and beings. These functions depend on different capacities. a. Capacities for self- movement (413a20-35), including growth, reproduction, and nutrition. b. Capacities for sensation (413b1-25), including perception, imagination (?), pleasure and pain, desire, and action. c. Capacities for reflection (413b25-34), including practical and theoretical reasoning (which depend upon imagination). The later functions typically involve or presuppose the earlier one(s) (414a30- b15), and different sorts of souls correspond to these three functions. Thus plants are those organisms that exercise only (a)- type functions; animals are those who also exercise (b)- type functions; and rational animals are those who also exercise (c)- type functions; and gods are disembodied beings who exercise theoretical reason only. It's worth noticing that imagination involves the capacity to represent situations, actual or merely possible, that are not part of one's immediate perception. Imagination plays some role in animal sensation, as when Rusty digs for the bone she believes buried in the yard. But imagination plays a larger role in theoretical and practical reasoning, inasmuch as this often involves working out the consequences of alternative possible actions and hypotheses. The peculiar function of humans is not the single activity that they and they alone perform, but the organized cluster of activities that they and they alone perform. For instance, as Tom Nagel notes, only in this way can we understand the peculiar function of a Swiss Army knife. 3 But neither should we understand the functions of various sorts of organisms as a mere conjunction of independent capacities and activities. Higher functions do not just add more capacities; the higher capacities modify or transform the operation of the lower capacities. Humans are self- movers whose movement reflects reasoning about which perceptual representations to endorse, which pleasures to bring about and pains to avoid, and how to reproduce, grow, and get nourishment. Our doing well, Aristotle believes, consists in the exercise of this ordered set of capacities (1098a12-16). This last inference may seem suspicious. Does it follow that x's performing its 3 Thomas Nagel, Aristotle on Eudaimonia reprinted in Essays on Aristotle s Ethics, ed. A. Rorty.

6 6 function well is good for x? The function of a purebred retriever may be, in part, to present a certain physical appearance, including a long waist, but realizing this function may not be good for the dog if it causes hip dysplasia. The function of the king's food- taster may be to thwart attempts to poison the king, but performing this function well may not be good for the food- taster. A prostitute's function may be to solicit and perform kinky sex as often as possible, but being this sort of sex machine is not obviously good for the person who is a prostitute. 4 x's performing her function F may benefit her qua F, but may not benefit her as a person or human. But this worry won't arise about the benefits of performing human functions, for this principle assures us that the person (human) will benefit from perform- ing her (human) function. This may seem merely to push the problem back a step. Why should someone associate his doing well with his benefit as a human being, rather than his benefit as a prostitute or food- taster? But Aristotle can argue that I am essentially a human being (F) and not essentially a prostitute or food- taster (G); I can exist and could have existed without G, but not without F. I must be concerned about the good of the sort of being I am essentially. If so, what I am essentially will benefit by performing the human function well. 5 EXTERNAL GOODS Aristotle claims that the function argument implies that happiness consists primar- ily in virtue, where virtues are conceived of as activities, rather than mere states (1098b a6, 1176a33- b1). And this fits a common classification of goods into external goods, which are matters of fortune and not fully within our control, and goods of the soul and the common view that goods of the soul are the most important (1098b13 15). Aristotle wants to claim that such a weighted, pluralist conception of the final good satisfies the sort of formal constraints he has identified - - not only is it a life suitable for humans but also that it is complete and stable. Even if virtue is a very important good, we know that it cannot be a complete good; a life of virtue is not lacking in nothing. There are externals that would make a virtuous life even better, and there are e evils that make a virtuous life worse than it otherwise would be. Further, deprivation of certain [externals] e.g. good birth, good children, beauty mars our blessedness; for we do not altogether have the character of happiness if we look utterly repulsive or are ill born, solitary or childless, and have it even less, presumably, if our children or friends are totally bad, or were good but have died [1099b2 5]. You don't have to agree with all of the things Aristotle recognizes as external goods or evils 4 The example is used in Jennifer Whiting's useful article, "Aristotle's Function Argument: A Defense" Ancient Philosophy 8 (1989). 5 Here I leave to one side worries that I am essentially a (particular) person, rather than a (particular) human being. If I am essentially a person and only contingently a human being - - if, for instance, I could survive disembodied or in a nonhuman body provided my rational agency was preserved - - then my ultimate concern would be to realize the function of a person and I would be concerned to realize the function of a human only insofar as this realized the function of a person.

7 7 or with the importance he attaches to different ones to agree that there are some things - - such as, family, friends, and health - - that are outside the agent's control yet affect the completeness of her happiness. However, many events are matters of fortune, and some are smaller, some greater. Hence, while small strokes of good or ill fortune clearly will not influence his life, many great strokes of good fortune will make it more blessed, since in themselves they naturally add adornment to it, and his use of them proves to be fine and excellent. Conversely, if they are great misfortunes, they oppress and spoil his blessedness, since they involve pain and impede many activities [1100b23 30]. Only a philosopher concerned to defend his theory would deny the value of e goods (1096a1 3, 1153b16-20). Presumably, Aristotle has Socrates in mind here and his belief that virtue is sufficient for happiness - - that is, a complete good. That is why the happy person needs to have goods of the body and external goods added, and needs fortune also, so that he will not be impeded in these ways. Some maintain, on the contrary, that we are happy when we are broken on the rack, or fall into terrible misfortunes, provided that we are good [virtuous]. Whether they mean to or not, these people are talking nonsense [vii b17-21]. But it is important to be clear about different ways that externals affect happiness. Aristotle seems to assign two kinds of value to externals. Against Socrates, he claims that some externals contribute directly to one s happiness by embellishing or marring it (1099b1-7, 1100a6-8, 1100b27). But he also thinks that externals sometimes function as necessary conditions for the realization of some virtues that are, apparently, constitutive of happiness, as, for example, both generosity and magnificence require resources for redistribution (1099a31 33). 6 UNRECOGNIZED AND POSTHUMOUS GOODS Aristotle thinks that we are subject to goods and evils of which we may not be aware (1100a20). He doesn't suggest what these might be, but it's not hard to imagine. If I have already lost contact with a loved one, and then unbeknownst to me she dies, it seems I have suffered a loss, whether I realize it or not. Or suppose my friend or spouse is secretly unfaithful to me. Isn't this a misfortune that a complete life would not contain? These are examples of what we might call contemporaneous unrecognized goods and evils. But their recognition and the completeness constraint suggest the possibility of posthumous goods and harms. Solon tells us that we should not pronounce on a person's happiness until after he is dead for only then will he be beyond misfortune (1100a10 19). But the possibility of posthumous goods and misfortunes suggests that pronouncements even at the graveside may be premature. My children and projects may succeed or fail, and why shouldn't these successes or failures, like contemporaneous 6 The dependence of some virtues, such as magnifence, on a level of external goods not required by the other virtues, may raise a question about whether Aristotle can defend either the unity or inseparability of the virtues.

8 8 unrecognized harms, add or detract from the significance and value of my life? Aristotle treats the issues surrounding recognition of posthumous goods and harms as raising a puzzle, and he warns against making a person's fate depend too much on such goods (1100a23-30). But he thinks we ought to recognize such goods, provided we don't assign them too much importance. Posthumous benefits and harms may be of comparatively less significance and presumably will diminish in proportion to the strength of their connection with the activities and relationships engaged in during the agent's life, but they are real benefits and harms (1100a19 23,30 1, 1101a23 30). THE DOMINANCE OF VIRTUE So completeness requires recognition of external goods as both necessary conditions and parts of happiness. But Aristotle does not think of externals as being as important to happiness as virtue. Severe misfortunes may rob a virtuous person of happiness, but (a) they cannot make a virtuous person miserable, and (b) a person is always better off choosing virtue over any amount of e goods (1098b15 16, 1099b17 27, 1100b1 11, 1100b31 34). But surely it is quite wrong to be guided [in our judgments of happiness] by someone's misfortunes. For his doing well or badly does not rest on them; though a human life, as we said, needs these added, it is the activities expressing virtue that control happiness, and the contrary activities that control its contrary [1100b8 10]. Aristotle's claim here about the dominance of virtue invites comparison with Socrates and Plato. Though he rejects Socrates's belief that virtue is sufficient for happiness, because he recognizes some externals as components of happiness, he agrees with Plato's comparative claim in the Republic that one is always better off being virtuous than not. Virtue may exct a price in some circumstances, if only in the form of the opportunity costs of virtue, but it is always a price worth paying. Of course, it s one thing to assert that virtue is the dominant componenet of happiness; it s another thing to defend that claim. A proper defense is possible only in light of Aristotle s discussion of friendship (see below). But we can see how Aristotle might think that the dominance claim is supported by his formal constraints on eudaimonia. The dominance of virtue can appeal to the second and third formal conditions or constraints on happiness. The function argument implies that happiness consists largely in the life of activity expressing one's rational capacities. The life of virtue is such a life, whereas a life filled with externals is not (1099b25 8). This implies and is reinforced by the condition that happiness must be largely within our power and not easily taken from us, because being virtuous, Aristotle believes, is largely up to us, whereas external goods are not (1099b18 20, 1100b1 7,12 15; EE 1215a8-18). The dominance of virtue highlights a kind of tension in Aristotle's formal criteria of happiness between completeness, on the one hand, and stability and control on the other. The life that is fully within our control will not be complete, and the life that is most complete will not be fully within our control. There is no inconsistency here provided Aristotle does not assume that happiness is completely within our control, and he does not. This allows him to identify the most important part of happiness with that part that is

9 9 largely within our control, viz. virtue, while recognizing that various externals, though largely out of our control, are genuine goods. The function argument may actually harmonize these constraints in a certain way. For Aristotle thinks that human life is essentially embodied and social in nature, but these are the two principal sources and uses of externals (e.g. health, physical appearance, social station, friends, and community). It is a condition of exercising some of the virtues that one have certain externals (e.g. health for any virtue, wealth for generosity and magnificence, and other people for friendship and political community), and many of the virtues concern the proper use of externals (e.g. generosity and magnificence). If so, then a life expressing well the human function will be a life that is virtuous and is appropriately supplied with and makes appropriate use of externals. Then why not say that the happy person is the one who expresses complete virtue in his activities, with an adequate supply of external goods, not just for any time but for a complete life? [1101a14 15] COMPREHENSIVE AND INTELLECTUALIST CONCEPTIONS OF HAPPINESS The picture of eudaimonia in book i appears to be comprehensive in the sense that it includes external goods as well as virtue and both intellectual virtues and practical virtues or virtues of character (1103a5 6). Aristotle elaborates the distinction between intellectual and practical virtues in NE vi. The intellectual virtues exercise capacities of theoretical reason, reasoning about what is necessary and not up to us (e.g. in contemplat- ing God and celestial bodies). They include contemplation or study (theoria), scientific knowledge (episteme), and understanding (nous). By contrast, the practical virtues (e.g. temperance, justice, friendship) involve reasoning about what is up to us. The practical virtues involve intelligence (phronesis) displayed in correct deliberation and decision (prohairesis). Intelligence can be exercised in both crafts (techne), which are undertaken for the sake of something else, and actions (praxis), which are undertaken for their own sakes. The virtues of character exercise capacities of practical reason concerning how to live and act. As we will see, the virtues of character involve agreement between the rational and non- rational parts of the soul. So practical virtues require more than proper deliberation and decision. They also require proper training and habituation of the appetites, encouraging agents to take pleasure in the right things (ii 1-5). 7 Most of the rest of the NE is devoted to a study of the virtues of character, either to the nature and conditions of such virtues in general (e.g. NE ii iii, vii) or of specific such virtues - - assorted virtues in iii and iv (e.g. bravery and temperance in iii; generosity, magnificence, and magnanimity in iv), justice in v, and friendship in viii ix. Both the completeness of eudaimonia and the function argument support such a comprehensive conception, and NE ii ix confirms this expectation. This apparently overwhelming commitment to a comprehensive conception of eudaimonia explains why many commentators wish that Aristotle had never written NE x 7 8. In x 6 Aristotle 7 It s unclear to me whether the first principles of ethical science are necessary truths that might be the object of contemplation, rather than deliberation. If so, this might be one explanation of Aristotle s famous claim that deliberation, which is about what is not necessary and is up to us, concerns not ends, but what promotes ends (112b12-14, 1112b 32-35).

10 10 returns to a general discussion of happiness, apparently resuming the discussion from book i. In x 7 8 he appears to claim that theoretical reason or contemplation is not only the most, but indeed the exclusive good. This conception of eudaimonia is sometimes called strict intellectualism. Hence the activity of the gods that is superior in blessedness will be an activity of study. And so the human activity that it most akin to the god's will, more than any others, have the character of happiness. A sign of this is the fact that other animals have no share in happiness, being completely deprived of this activity of study. For the whole life of the gods is blessed to the extent that it has something resembling this sort of activity; but none of the other animals is happy, because none of them shares in study at all. Hence happiness extends just as far as study extends, and the more someone studies, the happier he is, not coincidentally but insofar as he studies, since study is valuable in itself. And so happiness will be some kind of study [1178b22-31]. Why is Aristotle attracted to strict intellectualism? At one point, he appeals to the divinity of contemplation (1177a13-18, 1177b a10, 1178b8-31). 1. Happiness should be identified with the best or most divine elements. 2. Contemplation is most godlike, because it is the activity characteristic of gods. 3. Hence, our happiness should be identified with contemplation. But this appeal seems to run afoul of the function argument. After all, we are humans or rational animals, not gods. Indeed, we are essentially rational animals; were we to be transformed into gods, this would be a substantial change. But that means it wouldn't be good for us to undergo such a transformation, even if gods have better lives than rational animals. Hence there is this puzzle: do friends really wish their friend to have the greatest good, e.g. to be a god? For [if he becomes a god], he will no longer have friends, and hence no longer have goods, since friends are good. If, then, we have been right to say that one friend wishes good things to the other for the sake of the other himself, the other must remain whatever sort of being he is. Hence it is to the other as a human being that a friend will wish the greatest goods... [1159a6-12]. Aristotle seems to recognize this same point in NE x 8 where, after apparently endorsing contemplation, he suggests that contemplation is not fit for humans. However, the happy person is a human being, and so will need external prosperity also; for his nature is not self sufficient for study, but he needs a healthy body, and needs to have food and other services provided [1178b33 5; cf. 1178a19 21, 1178b5 8]. A similar appeal to the human function argues for recognition of the practical virtues as

11 11 part of happiness. Aristotle also appeals to our control over contemplation in defense of strict intellectualism (1177a27-35, 1178a24-8). 1. The life of contemplation is most within our control. 2. Happiness must be within our control. 3. Hence, happiness must consist in contemplation alone. Aristotle puts this point by claiming that contemplation is a more "self sufficient" good. This is unfortunate language, because self- sufficiency was understood in book i (1097b7 15) to indicate that happiness must be lacking in nothing, not that it must be within our control. And this is the main problem with Aristotle's appeal to control. Even if the life of contemplation is more within our control than a life including the virtues of character, 8 Aristotle does not believe that happiness need be fully within our control. Indeed, as we have seen, if happiness is to be self- sufficient or complete, it cannot plausibly be thought to be fully within our control. And a life of pure contemplation seems not to be a complete good (cf. MM 1184a37-38). Aristotle responds that contemplation is not incomplete, because the virtues of character are only appropriate for embodied intellects or psychophysical compounds, which the gods are not (1178b9 17). This would perhaps be a good defense of the completeness of contemplation for beings, such as the gods, who are essentially only intel- lect, but it fails for beings, such as humans, who are essentially embodied. Because of this fact about us, our happiness will be more complete when we possess various practical virtues, including friendship, and the right external goods. Must we conclude that Aristotle is just inconsistent in his claims about eudaimonia and that his later intellectualist claims are simply lame? The most common verdict in the secondary literature reluctantly accepts this as unavoidable inconsistency in the text. 9 In Aristotle on the Human Good Richard Kraut has proposed a heroic interpretation of Aristotle as a consistent intellectualist. He reads the function argument as an argument for identifying happiness with the exercise of the most supreme element in human nature (1098a16 19) and claims that practical virtues and externals are chosen for the sake of contemplation. Though heroic, this view seems unattractive. First, 1098a16-19 claims that "if there are more virtues than one, the good will express the best and most complete virtue". This does not imply that the good is to be identified with contemplation. For we've just been told that a complete good must be lacking in nothing, and Aristotle clearly thinks completeness requires a complex or plural good. But then the most complete virtue for humans will not select a proper part of virtue but rather the whole of virtue. Second, this reading of the function argument does not respect Aristotle's claim that the human good consists in the exercise of essential human capacities and conflicts with the pluralistic implications of the completeness requirement. Third, this reading of the value of the 8 Though Aristotle does think that contemplation is more within our control than the virtues of character, he admits that even contemplation requires externals; the contemplator needs her health, intellectual training (i.e. education), and leisure, and her contemplation is enhanced when she has colleagues with whom to talk (1177a28 b1, 1177b5-, 1178a24 29). 9 A representative verdict of inconsistency is John Cooper, Reason and Human Good in Aristotle.

12 12 virtues of character and externals does not appear to fit Aristotle's claims. Though "for the sake of" does not always refer to an instrumental relation, presumably it would have to on this intellectualist view. In order for x to be chosen for the sake of y and for x to be intrinsically valuable, x must be a constituent part of y, where y is intrinsically valuable. But even if externals or practical virtues are necessary conditions or causal means to contemplation, it's very unclear how either could be a part of contemplation. So if happiness consists in contemplation, as the intellectualist interpretation claims, these things must be only instrumentally valuable and only when they are means or necessary conditions for contemplation. But Aristotle treats both as valuable for their own sakes (externals: 1099b1-7, 1100a6-8, 1100b27; virtue: 1105a32, 1144a19), and he gives no sign of restricting their value to contexts in which they are productive of contemplation. An alternative strategy for a consistent reading of Aristotle s eudaimonist claims would try to reconcile the apparent intellectualism of book x with the comprehensive conception that seems to be a commitment of the rest of the NE. Jennifer Whiting and Terry Irwin have explored this possibility. 10 They suggest that we can reconcile intellectualist and comprehensive claims if the former assert only comparative or conditional theses. 1. Contemplation is the single activity that best fits the criteria for happiness. 2. If happiness must be some single activity, contemplation is the best candidate. 3. If happiness includes more than one activity, contemplation will be the most important. 4. Happiness consists in contemplation alone. The comprehensive conception could perhaps endorse (1)- (3) but not (4). Some endorsements of contemplation can be understood in comparative or conditional terms. However, other endorsements of contemplation seem to defy such analysis (e.g. 1178b25-29). Hence the activity of the gods that is superior in blessedness will be an activity of study. And so the human activity that it most akin to the god's will, more than any others, have the character of happiness. A sign of this is the fact that other animals have no share in happiness, being completely deprived of this activity of study. For the whole life of the gods is blessed to the extent that it has something resembling this sort of activity; but none of the other animals is happy, because none of them shares in study at all. Hence happiness extends just as far as study extends, and the more someone studies, the happier he is, not coincidentally but insofar as he studies, since study is valuable in itself. And so happiness will be some kind of study [1178b22-31, italics added]. It s hard to see how to square the italicized phrases in this passage with the comprehensive reading. For this reason, I think it s hard to avoid the conclusion that Aristotle s claims about happiness are inconsistent. However, this does not prevent us from concluding that 10 See Jennifer Whiting, Human Nature and Intellectualism in Aristotle Archiv für Geschicte der Philosophie 68(1986) and Irwin s Notes to the NE, p. 397.

13 13 his predominant and best commitment is to the comprehensive conception. That conception is supported not only by the crucial claims in book i but also by the body of the NE in books ii- ix. Strict intellectualism is flirted with only briefly in book x and the arguments there on its behalf are seriously inadequate and inconsistent with the rest of the NE. EUDAIMONIC AND MORAL VIRTUE Much of the NE is concerned with the various virtues. Given Aristotle's eudai- monism and, especially the function argument, we might expect him to identify, individu- ate, and explain the various virtues by reference to the agent's good or to the exercise of the agent's essential capacities. We might call such virtues eudaimonic virtues. But, for the most part, this is not his procedure. His concern is with fairly familiar virtues, such as bravery, temperance, generosity, and justice. Such virtues are fairly standard entries on lists of moral virtues, and, for Aristotle, they form the principal subject matter of political science. In fact, Aristotle identifies, individuates, and explains these virtues by the fact that they are states of character that are fine (kalon) and praiseworthy (1103a10, 1120a24, 1122b6 8). We might call such virtues moral virtues. It is commonly thought that eudaimonic and moral conceptions of the virtues are quite different, perhaps incompatible, much as psychic and conventional virtues may seem irreconcilable in Plato's ethical theory. Because Aristotle begins the NE with a eudaimonic conception, it is sometimes assumed either that his conception of ethics is quite different from our conception of morality or that if he links these two conceptions he does so without proper justification. Later, I will return to the relationship between eudaimonic and moral virtue and argue that Aristotle's discussion of friendship can and should be seen as way of connecting eudaimonic and moral virtues. Here I only note that his discussion of virtues of character stresses the moral, rather than the eudaimonic, aspects of these virtues. VIRTUE, REASON, AND DESIRE Aristotle initiates his general discussion of virtue as a discussion of eudaimonic virtue. The human function is activity of the soul in accordance with reason. The virtues of character concern both the part of the soul that has reason and the part that obeys reason (1098a3 5, 1102b a3). Virtue is the condition in which the nonrational part of the soul that can obey reason does so and harmonizes with rational choice (1102b25 8). The virtues (of character) require correct deliberation and decision. But this is not enough. They also require proper training and habituation of the appetites, in which agents learn to take pleasure in the right things (ii 1-5). Unlike crafts, which are undertaken for ends beyond themselves, virtuous actions are undertaken for their own sakes (1140b7-8), and they must be performed from a stable state of character. The agent must not merely conform to the behavioral demands of virtue, but choose virtuous action for its own sake (1144a14-20). Rather, the agent must be in the right state when he does them. First, he must know [that he is doing virtuous actions second, he must decide on them, and decide on them for themselves; and, third, he must do them from a firm and unchanging state [1105a30-35].

14 14 There are four possible relations between these two parts of the soul that are important to Aristotle's moral psychology (1102b14-28 an vii 1-10, esp. 1145a15- b2, 1148a a10, 1151a a35). 1. Virtue: the rational and nonrational parts agree in pursuing the right ends. 2. Continence: the rational and nonrational parts disagree; the rational part chooses the right ends; the nonrational part chooses the wrong ends; and the rational part wins. 3. Incontinence: the rational and nonrational parts disagree; the rational part chooses the right ends; the nonrational part chooses the wrong ends; and the nonrational part wins. 4. Vice: the rational and nonrational parts of the soul agree in pursuing the wrong ends. This taxonomy avoids the familiar, but oversimple, two- fold classification of characters into virtuous or vicious. 11 Note different patterns within this taxonomy. (a) Virtue and vice both involve agreement between the rational and nonrational parts of the soul, whereas continence and incontinence both involve conflicts between these two parts. (b) In the case of virtue and continence, the agent acts as he should, whereas in the case of incontinence and vice he does not. (c) In the case of virtue, continence, and incontinence the agent makes the right judgment, whether or not he acts on it, whereas only the vicious person forms the wrong judgment. Can the virtuous person experience conflict? This taxonomy, especially the distinction between virtue and mere continence, may suggest that the virtuous person experiences no conflict or regret. On one interpretation, the truly virtuous person sees no conflict between the demands of virtue and other options. Virtue, on this view, does not outweigh or override other reasons; it silences them altogether. 12 Considerations that would otherwise provide reasons for action altogether lack practical significance when pitted against the demands of virtue. In matters of temperance and bravery, actions are not enough. [W]e must take as a sign of someone's state his pleasure or pain in consequence of his action. For if someone who abstains from bodily pleasures enjoys the abstinence itself, then he is temperate, but if he is grieved by it, he is intemperate. Again, if he stands firm against terrifying situations and enjoys it, or at least does not find it painful, then he is brave, and if he finds it painful, he is cowardly [1104b4-7]. This silencing interpretation need not make Aristotle fully Socratic. For while it implies, as Socrates claims, that virtue has no price, it can allow, as Socrates apparently cannot, that e- 11 Sometimes, Aristotle mentions a fifth state of bestiality, in which the agent is controlled by appetite and passion and either lacks deliberation or has it corrupted beyond repair (1145a24-32) 12 McDowell develops this idea in "Virtue and Reason" The Monist 52 (1979) and The Role of Eudaimonia in Aristotle s Ethics reprinted in Essays on Aristotle s Ethics, ed. A. Rorty.

15 15 goods have value when not in competition with virtue. If so, this interpretation can allow, as Socrates cannot, that virtue is an incomplete good. Nonetheless, the silencing interpretation is not compelling. It's quite clear that Aristotle does not think that virtue is a complete good. But if e- goods have value independently of virtue, why should their value be limited by virtue so that they have no value whatsoever when they conflict with virtue? The common view is that virtue can have a price, even if it turns out to be a price well worth paying. Indeed, this seems to be Aristotle's view. The brave person will find death and wounds [potential costs of bravery] painful, and suffer them unwillingly, but he will stand firm against them because that is fine or because failure is shameful. Indeed, the more he has every virtue and the happier he is, the more pain he will feel at the prospect of death. For this sort of person, more than anyone, finds it worthwhile to be alive, and is knowingly deprived of the greatest goods, and this is painful. But he is no less brave for all that; presumably, indeed, he is all the braver, because he chooses what is fine in war at the cost of all these goods [1117b8-15; cf. EE 1229a7, MM 1191a26-29]. The only reason for supposing that Aristotle ought to deny this common view is the assumption that he must to make out his distinction between the virtuous and merely continent. But that distinction requires only that the virtuous person have a sufficiently steady and unwavering commitment to act as he judges best - - that he not be seriously tempted to act otherwise. But one can be steadfast in one's commitment to the virtuous course of action without insisting that virtue has no cost. One need only see that the cost is well worth paying, that is, as Aristotle says, that virtue is the controlling ingredient in eudaimonia. THE DOCTRINE OF THE MEAN Virtue is a mean between excess and deficiency of action or feeling (ii 6). The doctrine of the mean is not a decision procedure; the right decision determines the mean, not the other way around (1106b b4, 1109a24-5). There is no mean with respect to base actions, such as adultery or murder, or base feelings, such as envy or childishness (1107a9-25; MM 1186a36- b3); and where there is a mean it is not always the mid- point (cf. 1106a33 b6). So what is the significance of the doctrine of the mean? In contrast with a Socratic account, it underscores the role of affective states in virtue. For example, magnanimity requires the proper mix of concern for self and the opinion of others; it requires not only curbing self- aggrandizement but also a healthy concern for oneself and pride in one's accomplishments. Virtue requires proper habituation and training of the nonrational parts of the soul. It also contrasts with a Kantian account of virtue or the good will. [T]here are many persons who are so sympathetically constituted, that without any further motive of vanity or self interest, they find an inner joy around them and can rejoice in the satisfaction of others as their own work. But I maintain that in such a case an action of this kind, however dutiful and amiable it might be, has nevertheless no true moral worth.... [It] accords with duty and is thus honorable,

16 16 deserves praise and encouragement, but not esteem... Further still, if nature has put little sympathy in this or that man's heart, if (while being an honest man in other respects) he is by temperament cold and indifferent to the sufferings of others... if such a man (who would truly not be nature's worst product) had not been exactly fashioned by her a philanthropist, would he not yet find in himself a source from which he might give himself a worth far higher than any that a good natured temperament might have? By all means, because just here does the worth of the character come out; this worth is moral and incomparably the highest of all, viz. that he is beneficent, not from inclination, but from duty [Groundwork ]. Whereas Kant's grudging moralist may display a good will, he would appear to display Aristotelian continence, rather than virtue. AKRASIA After an examination of various familiar and less familiar (nameless) virtues and the ways in which they constitute means, Aristotle returns in book vii to more general issues about virtue, continence, incontinence, and vice. The virtuous person has no base appetites; by upbringing and effort her nonrational appetites have been shaped to harmonize with her right judgments. In the vicious person, there is also agreement between her rational and non- rational parts, but her judgments have been adapted to harmonize with inappro- priate appetites. By contrast, there is disagreement in the souls of both continent and incontinent people (1151a6-14). Both have right judgments and base appetites. Whereas the actions of the continent person conform to her right judgments, the actions of the incontinent person conform to her base appetites. These assumptions lead Aristotle to examine the puzzles created by different common or respected beliefs about the possibility and nature of akrasia. On the one hand, Aristotle thinks that the Socratic denial of akrasia is hard to believe (1145b23-9) and seems to endorse the Platonic view that the incontinent person is overcome by appetites and pleasure (1147b23-34, 1149b26). On the other hand, Aristotle does concede the Socratic claim that the (putative) akrates lacks a kind of knowledge (1147b14-20). The details and even the outline of Aristotle's own account are hard to understand. On the one hand, he thinks that the Socratic denial of akrasia is hard to believe because it denies the manifest appearance that akrasia is possible (1145b23-29) and that the incontinent person is overcome by appetites or pleasure (1147b23-34, 1149b26), which appear to be noncognitive states. On the other hand, Aristotle's own account seems to concede the Socratic claim that the incontinent person is lacking a certain kind of knowledge (1147b14-20). Aristotle's account is interesting because of the way it tries to combine affective and cognitive explanations of the phenomenon in question. In general outline, Aristotle's account of incontinence is something like this. In one way, the incontinent shares the temperate person's judgments about what is best. But the incontinent person's appetites have not been trained properly so as to harmonize with her judgments. Because of this, her judgments are unstable. When the pleasures associated with certain appetites are especially strong, this undermines her confidence in her beliefs about what is best. Though she still understands the sort of thing that is best to do and will say that this is a case of that sort, the pleasure of acting intemperately prevents her from believing or knowing in the requisite way that this is a case of that sort (1147a17-23). This