Jump to ContentJump to Main Navigation
The Quality of Life$

Martha Nussbaum and Amartya Sen

Print publication date: 1993

Print ISBN-13: 9780198287971

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0198287976.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 05 August 2021

Non‐Relative Virtues: An Aristotelian Approach

Non‐Relative Virtues: An Aristotelian Approach

(p.242) Non‐Relative Virtues: An Aristotelian Approach
The Quality of Life

Martha Nussbaum (Contributor Webpage)

Oxford University Press

Abstract and Keywords

All Greeks used to go around armed with swords.

Thucydides, History of the Peloponnesian War

The customs of former times might be said to be too simple and barbaric. For Greeks used to go around armed with swords; and they used to buy wives from one another; and there are surely other ancient customs that are extremely stupid. (For example, in Cyme there is a law about homicide, that if a man prosecuting a charge can produce a certain number of witnesses from among his own relations, the defendant will automatically be convicted of murder.) In general, all human beings seek not the way of their ancestors, but the good....

All Greeks used to go around armed with swords.

Thucydides, History of the Peloponnesian War

The customs of former times might be said to be too simple and barbaric. For Greeks used to go around armed with swords; and they used to buy wives from one another; and there are surely other ancient customs that are extremely stupid. (For example, in Cyme there is a law about homicide, that if a man prosecuting a charge can produce a certain number of witnesses from among his own relations, the defendant will automatically be convicted of murder.) In general, all human beings seek not the way of their ancestors, but the good.

Aristotle, Politics, 1268a39 ff.

One may also observe in one's travels to distant countries the feelings of recognition and affiliation that link every human being to every other human being.

Aristotle, Nicomachean Ethics, 1155a21–2


The virtues are attracting increasing interest in contemporary philosophical debate. From many different sides one hears of a dissatisfaction with ethical theories that are remote from concrete human experience. Whether this remoteness results from the utilitarian's interest in arriving at a universal calculus of satisfactions or from a Kantian concern with universal principles of broad generality, in which the names of particular contexts, histories, and persons do not occur, remoteness is now being seen by an increasing number of moral philosophers as a defect in an approach to ethical questions. In the search for an alternative approach, the concept of virtue is playing a prominent role. So, too, is the work of Aristotle, the greatest defender of an ethical approach based on the concept of virtue. For Aristotle's work seems, appealingly, to combine rigour with concreteness, theoretical power with sensitivity to the actual circumstances (p.243) of human life and choice in all their multiplicity, variety, and mutability.

But on one central point there is a striking divergence between Aristotle and contemporary virtue theory. To many current defenders of an ethical approach based on the virtues, the return to the virtues is connected with a turn towards relativism—towards, that is, the view that the only appropriate criteria of ethical goodness are local ones, internal to the traditions and practices of each local society or group that asks itself questions about the good. The rejection of general algorithms and abstract rules in favour of an account of the good life based on specific modes of virtuous action is taken, by writers as otherwise diverse as Alasdair MacIntyre, Bernard Williams, and Philippa Foot,1 to be connected with the abandonment of the project of rationally justifying a single norm of flourishing life for all human beings and a reliance, instead, on norms that are local both in origin and in application.

The position of all these writers, where relativism is concerned, is complex; none unequivocally endorses a relativist view. But all connect virtue ethics with a relativist denial that ethics, correctly understood, offers any transcultural norms, justifiable by reference to reasons of universal human validity, by reference to which we may appropriately criticize different local conceptions of the good. And all suggest that the insights we gain by pursuing ethical questions in the Aristotelian virtue‐based way lend support to relativism.

For this reason it is easy for those who are interested in supporting the rational criticism of local traditions and in articulating an idea of ethical progress to feel that the ethics of virtue can give them little help. If the position of women, as established by local traditions in many parts of the world, is to be improved, if traditions of slave‐holding and racial inequality, religious intolerance, aggressive and warlike conceptions of manliness, and unequal norms of material distribution are to be criticized in the name of practical reason, this criticizing (one might easily suppose) will have to be done from a Kantian or utilitarian viewpoint, not through the Aristotelian approach.

This is an odd result, as far as Aristotle is concerned. For it is obvious that he was not only the defender of an ethical theory based on the virtues, but also the defender of a single objective account of the human good, or human flourishing. This account is supposed to be objective in the sense that it is justifiable by reference to reasons that do not derive merely from local traditions and practices, but rather from features of humanness that lie beneath all local traditions and are there to be seen whether or not they are in fact recognized in local traditions. And one of Aristotle's most obvious concerns was the criticism of existing moral traditions, in his own city and in others, as unjust or repressive, or in other ways incompatible with human flourishing. He uses his account of the virtues as a basis for this criticism of local traditions: prominently, for example, in Book II of the Politics, where he frequently argues against existing social forms by (p.244) pointing to ways in which they neglect or hinder the development at some important human virtue.2 Aristotle evidently believed that there is no incompatibility between basing an ethical theory on the virtues and defending the singleness and objectivity of the human good. Indeed, he seems to have believed that these two aims are mutually supportive.

Now the fact that Aristotle believed something does not make it true (though I have sometimes been accused of holding that position!). But it does, on the whole, make that something a plausible candidate for the truth, one deserving our most serious scrutiny. In this case, it would be odd indeed if he had connected two elements in ethical thought that are self‐evidently incompatible, or in favour of whose connectedness and compatibility there is nothing interesting to be said. The purpose of this paper is to establish that Aristotle did indeed have an interesting way of connecting the virtues with a search for ethical objectivity and with the criticism of existing local norms, a way that deserves our serious consideration as we work on these questions. Having described the general shape of the Aristotelian approach, we can then begin to understand some of the objections that might be brought against such a non‐relative account of the virtues, and to imagine how the Aristotelian could respond to those objections.


The relativist, looking at different societies, is impressed by the variety and the apparent non‐comparability in the lists of virtues she encounters. Examining the different lists, and observing the complex connections between each list and a concrete form of life and a concrete history, she may well feel that any list of virtues must be simply a reflection of local traditions and values, and that, virtues being (unlike Kantian principles or utilitarian algorithms) concrete and closely tied to forms of life, there can in fact be no list of virtues that will serve as normative for all these varied societies. It is not only that the specific forms of behaviour recommended in connection with the virtues differ greatly over time and place, it is also that the very areas that are singled out as spheres of virtue, and the manner in which they are individuated from other areas, vary so greatly. For someone who thinks this way, it is easy to feel that Aristotle's own list, despite its pretensions to universality and objectivity, must be similarly restricted, merely a reflection of one particular society's perceptions of salience and ways of distinguishing. At this point, relativist writers are likely to quote Aristotle's description of the ‘great‐souled’ person, the megalopsuchos, which certainly contains many concrete local features and sounds very much like the portrait of a certain sort of Greek gentleman, in order to show that Aristotle's list is just as culture‐bound as any other.3

But if we probe further into the way in which Aristotle in fact enumerates and (p.245) individuates the virtues, we begin to notice things that cast doubt upon the suggestion that he simply described what was admired in his own society. First of all, we notice that a rather large number of virtues and vices (vices especially) are nameless, and that, among the ones that are not nameless, a good many are given, by Aristotle's own account, names that are somewhat arbitrarily chosen by Aristotle, and do not perfectly fit the behaviour he is trying to describe.4 Of such modes of conduct he writes, ‘Most of these are nameless, but we must try . . . to give them names in order to make our account clear and easy to follow’ (EN 1108a 6–19). This does not sound like the procedure of someone who is simply studying local traditions and singling out the virtue‐names that figure most prominently in those traditions.

What is going on becomes clearer when we examine the way in which he does, in fact, introduce his list. For he does so, in the Nicomachean Ethics5 by a device whose very straight‐forwardness and simplicity has caused it to escape the notice of most writers on this topic. What he does, in each case, is to isolate a sphere of human experience that figures in more or less any human life, and in which more or less any human being will have to make some choices rather than others, and act in some way rather than some other. The introductory chapter enumerating the virtues and vices begins with an enumeration of these spheres (EN II. 7); and each chapter on a virtue in the more detailed account that follows begins with ‘Concerning X . . . ’, or words to this effect, where X names a sphere of life with which all human beings regularly and more or less necessarily have dealings.6 Aristotle then asks, what is it to choose and respond well within that sphere? And what is it to choose defectively? The ‘thin account’ of each virtue is that it is whatever being stably disposed to act appropriately in that sphere consists in. There may be, and usually are, various competing specifications of what acting well, in each case, in fact comes to. Aristotle goes on to defend in each case some concrete specification, producing, at the end, a full or ‘thick’ definition of the virtue.

Here are the most important spheres of experience recognized by Aristotle, along with the names of their corresponding virtues:7




1. Fear of important damages, esp. death


2. Bodily appetites and their pleasures


3. Distribution of limited resources


4. Management of one's personal property, where others are concerned


5. Management of personal property, where hospitality is concerned

Expansive hospitality

6. Attitudes and actions with respect to one's own worth

Greatness of soul

7. Attitude to slights and damages

Mildness of temper

8. ‘Association and living together and the fellowship of words and actions’

a. Truthfulness in speech


b. Social association of a playful kind

Easy grace (contrasted with coarseness, rudeness, insensitivity)

c. Social association more generally

Nameless, but a kind of friendliness (contrasted with irritability and grumpiness)

9. Attitude to the good and ill fortune of others

Proper judgement (contrasted with enviousness, spitefulness, etc.)

10. Intellectual life

The various intellectual virtues, such as perceptiveness, knowledge, etc.

11. The planning of one's life and conduct

Practical wisdom

There is, of course, much more to be said about this list, its specific members, and the names Aristotle chose for the virtue in each case, some of which are indeed culture‐bound. What I want to insist on here, however, is the care with which Aristotle articulates his general approach, beginning from a characterization of a sphere of universal experience and choice, and introducing the virtue‐name as the name (as yet undefined) of whatever it is to choose appropriately in (p.247) that area of experience. On this approach, it does not seem possible to say, as the relativist wishes to, that a given society does not contain anything that corresponds to a given virtue. Nor does it seem to be an open question, in the case of a particular agent, whether a certain virtue should or should not be included in his or her life—except in the sense that she can always choose to pursue the corresponding deficiency instead. The point is that everyone makes some choices and acts somehow or other in these spheres: if not properly, then improperly. Everyone has some attitude, and corresponding behaviour, towards her own death; her bodily appetites and their management; her property and its use; the distribution of social goods; telling the truth; being kind to others; cultivating a sense of play and delight, and so on. No matter where one lives one cannot escape these questions, so long as one is living a human life. But then this means that one's behaviour falls, willy‐nilly, within the sphere of the Aristotelian virtue, in each case. If it is not appropriate, it is inappropriate; it cannot be off the map altogether. People will of course disagree about what the appropriate ways of acting and reacting in fact are. But in that case, as Aristotle has set things up, they are arguing about the same thing, and advancing competing specifications of the same virtue. The reference of the virtue term in each case is fixed by the sphere of experience—by what we shall from now on call the ‘grounding experiences’. The thin or ‘nominal’ definition of the virtue will be, in each case, that it is whatever being disposed to choose and respond well consists in, in that sphere. The job of ethical theory will be to search for the best further specification corresponding to this nominal definition, and to produce a full definition.


We have begun to introduce considerations from the philosophy of language. We can now make the direction of the Aristotelian account clearer by considering his own account of linguistic indicating (referring) and defining, which guides his treatment of both scientific and ethical terms, and of the idea of progress in both areas.8

Aristotle's general picture is as follows. We begin with some experiences—not necessarily our own, but those of members of our linguistic community, broadly construed.9 On the basis of these experiences, a word enters the language of the group, indicating (referring to) whatever it is that is the content of those experiences. Aristotle gives the example of thunder.10 People hear a noise in the clouds, and they then refer to it, using the word ‘thunder’. At this point, it may be that nobody has any concrete account of the noise or any idea about what it really is. But the experience fixes a subject for further inquiry. From now on, (p.248) we can refer to thunder, ask ‘What is thunder?’, and advance and assess competing theories. The thin or, we might say, ‘nominal’ definition of thunder is ‘That noise in the clouds, whatever it is’. The competing explanatory theories are rival candidates for correct full or thick definition. So the explanatory story citing Zeus' activities in the clouds is a false account of the very same thing of which the best scientific explanation is a true account. There is just one debate here, with a single subject.

So too, Aristotle suggests, with our ethical terms. Heraclitus, long before him, already had the essential idea, saying, ‘They would not have known the name of justice, if these things did not take place.’11 ‘These things’, our source for the fragment informs us, are experiences of injustice–presumably of harm, deprivation, inequality. These experiences fix the reference of the corresponding virtue word. Aristotle proceeds along similar lines. In the Politics he insists that only human beings, and not either animals or gods, will have our basic ethical terms and concepts (such as just and unjust, noble and base, good and bad), because the beasts are unable to form the concepts, and the gods lack the experiences of limit and finitude that give a concept such as justice its points.12 In the enumeration of the virtues in the Nicomachean Ethics, he carries the line of thought further, suggesting that the reference of the virtue terms is fixed by spheres of choice, frequently connected with our finitude and limitation, that we encounter in virtue of shared conditions of human existence.13 The question about virtue usually arises in areas in which human choice is both non‐optional and somewhat problematic. (Thus, he stresses, there is no virtue involving the regulation of listening to attractive sounds, or seeing pleasing sights.) Each family of virtue and vice or deficiency words attaches to some such sphere. And we can understand progress in ethics, like progress in scientific understanding, to be progress in finding the correct fuller specification of a virtue, isolated by its thin or nominal definition. This progress is aided by a perspicuous mapping of the sphere of the grounding experiences. When we understand more precisely what problems human beings encounter in their lives with one another, what circumstances they face in which choice of some sort is required, we will have a way of assessing competing responses to those problems, and we will begin to understand what it might be to act well in the face of them.

Aristotle's ethical and political writings provide many examples of how such progress (or, more generally, such a rational debate) might go. We find argument against Platonic asceticism, as the proper specification of moderation (appropriate (p.249) choice and response vis à vis the bodily appetites), and in favour of a more generous role for appetitive activity in human life. We find argument against the intense concern for public status and reputation, and the consequent proneness to anger over slights, that was prevalent in Greek ideals of maleness and in Greek behaviour, together with a defence of a more limited and controlled expression of anger, as the proper specification of the virtue that Aristotle calls ‘mildness of temper’. (Here Aristotle evinces some discomfort with the virtue term he has chosen, and he is right to do so, since it certainly loads the dice heavily in favour of his concrete specification and against the traditional one.14) And so on for all the virtues.

In an important section of Politics II, part of which forms one of the epigraphs to this paper, Aristotle defends the proposition that laws should be revisable and not fixed, by pointing to evidence that there is progress towards greater correctness in our ethical conceptions, as also in the arts and sciences. Greeks used to think that courage was a matter of waving swords around; now they have (the Ethics informs us) a more inward and a more civic and communally attuned understanding of proper behaviour towards the possibility of death. Women used to be regarded as property, to be bought and sold; now this would be thought barbaric. And in the case of justice as well we have, the Politics passage claims, advanced towards a more adequate understanding of what is fair and appropriate. Aristotle gives the example of an existing homicide law that convicts the defendant automatically on the evidence of the prosecutor's relatives (whether they actually witnessed anything or not, apparently). This, Aristotle says, is clearly a stupid and unjust law; and yet it once seemed appropriate–and, to a tradition‐bound community, must still be so. To hold tradition fixed is then to prevent ethical progress. What human beings want and seek is not conformity with the past, it is the good. So our systems of law should make it possible for them to progress beyond the past, when they have agreed that a change is good. (They should not, however, make change too easy, since it is no easy matter to see one's way to the good, and tradition is frequently a sounder guide than current fashion.)

In keeping with these ideas, the Politics as a whole presents the beliefs of the many different societies it investigates not as unrelated local norms, but as competing answers to questions about justice and courage (and so on) with which all societies are (being human) concerned, and in response to which they all try to find what is good. Aristotle's analysis of the virtues gives him an appropriate framework for these comparisons, which seem perfectly appropriate inquiries into the ways in which different societies have solved common human problems.

In the Aristotelian approach it is obviously of the first importance to distinguish two stages of the inquiry: the initial demarcation of the sphere of choice, (p.250) of the ‘grounding experiences’ that fix the reference of the virtue term; and the ensuing more concrete inquiry into what the appropriate choice, in that sphere, is. Aristotle does not always do this carefully; and the language he has to work with is often not helpful to him. We do not have much difficulty with terms like ‘moderation’ and ‘justice’ and even ‘courage’, which seem vaguely normative, but relatively empty, so far, of concrete moral content. As the approach requires, they can serve as extension‐fixing labels under which many competing specifications may be investigated. But we have already noticed the problem with ‘mildness of temper’, which seems to rule out by fiat a prominent contender for the appropriate disposition concerning anger. And much the same thing certainly seems to be true of the relativists' favourite target, megalopsuchia, which implies in its very name an attitude to one's own worth that is more Greek than universal. (A Christian, for example, will feel that the proper attitude to one's own worth requires an understanding of one's lowness, frailty, and sinfulness. The virtue of humility requires considering oneself small, not great.) What we ought to get at this point in the inquiry is a word for the proper attitude towards anger and offence, and for the proper attitude towards one's worth, that are more truly neutral among the competing specifications, referring only to the sphere of experience within which we wish to determine what is appropriate. Then we could regard the competing conceptions as rival accounts of one and the same thing, so that, for example, Christian humility would be a rival specification of the same virtue whose Greek specification is given in Aristotle's account of megalopsuchia, namely, the proper attitude towards the question of one's own worth.

In fact, oddly enough, if one examines the evolution in the use of this word from Aristotle through the Stoics to the Christian fathers, one can see that this is more or less what happened, as ‘greatness of soul’ became associated, first, with the Stoic emphasis on the supremacy of virtue and the worthlessness of externals, including the body, and, through this, with the Christian denial of the body and of the worth of earthly life.15 So even in this apparently unpromising case, history shows that the Aristotelian approach not only provided the materials for a single debate but actually succeeded in organizing such a debate, across enormous differences of both place and time.

Here, then, is a sketch for an objective human morality based upon the idea of virtuous action–that is, of appropriate functioning in each human sphere. The Aristotelian claim is that, further developed, it will retain the grounding in actual human experiences that is the strong point of virtue ethics, while gaining the ability to criticize local and traditional moralities in the name of a more inclusive account of the circumstances of human life, and of the needs for human functioning that these circumstances call forth.

(p.251) [4]

The proposal will encounter many objections. The concluding sections of this paper will present three of the most serious and will sketch the lines along which the Aristotelian might proceed in formulating a reply. To a great extent these objections were not imagined or confronted by Aristotle himself, but his position seems capable of confronting them.

The first objection concerns the relationship between singleness of problem and singleness of solution. Let us grant for the moment that the Aristotelian approach has succeeded in coherently isolating and describing areas of human experience and choice that form, so to speak, the terrain of the virtues, and in giving thin definitions of each of the virtues as whatever it is that choosing and responding well within that sphere consists in. Let us suppose that the approach succeeds in doing this in a way that embraces many times and places, bringing disparate cultures together into a single debate about the good human being and the good human life. Different cultural accounts of good choice within the sphere in question in each case are now seen not as untranslatably different, but as competing answers to a single general question about a set of shared human experiences. Still, it might be argued, what has been achieved is, at best, a single discourse or debate about virtue. It has not been shown that this debate will have, as Aristotle believes, a single answer. Indeed, it has not even been shown that the discourse we have set up will have the form of a debate at all–rather than a plurality of culturally specific narratives, each giving the thick definition of a virtue that corresponds to the experience and traditions of a particular group. There is an important disanalogy with the case of thunder, on which the Aristotelian so much relies in arguing that our questions will have a single answer. For in that case what is given in experience is the definiendum itself, so that experience establishes a rough extension, to which any good definition must respond. In the case of the virtues, things are more indirect. What is given in experience across groups is only the ground of virtuous action, the circumstances of life to which virtuous action is an appropriate response. Even if these grounding experiences are shared, that does not tell us that there will be a shared appropriate response.

In the case of thunder, furthermore, the conflicting theories are clearly put forward as competing candidates for the truth; the behaviour of those involved in the discourse about virtue suggests that they are indeed, as Aristotle says, searching ‘not for the way of their ancestors, but for the good’. And it seems reasonable in that case for them to do so. It is far less clear, where the virtues are concerned (the objector continues), that a unified practical solution is either sought by the actual participants or a desideratum for them. The Aristotelian proposal makes it possible to conceive of a way in which the virtues might be non‐relative. It does not, by itself, answer the question of relativism.

The second objection goes deeper. For it questions the notion of spheres of (p.252) shared human experience that lies at the heart of the Aristotelian approach. The approach, says this objector, seems to treat the experiences that ground the virtues as in some way primitive, given, and free from the cultural variation that we find in the plurality of normative conceptions of virtue. Ideas of proper courage may vary, but the fear of death is shared by all human beings. Ideas of moderation may vary, but the experiences of hunger, thirst, and sexual desire are (so the Aristotelian seems to claim) invariant. Normative conceptions introduce an element of cultural interpretation that is not present in the grounding experiences, which are, for that very reason, the Aristotelian's starting point.

But, the objector continues, such assumptions are naïve. They will not stand up either to our best account of experience or to a close examination of the ways in which these so‐called grounding experiences are in fact differently constructed by different cultures. In general, first of all, our best accounts of the nature of experience, even perceptual experience, inform us that there is no such thing as an ‘innocent eye’ that receives an uninterpreted ‘given’. Even sense‐perception is interpretative, heavily influenced by belief, teaching, language, and in general by social and contextual features. There is a very real sense in which members of different societies do not see the same sun and stars, encounter the same plants and animals, hear the same thunder.

But if this seems to be true of human experience of nature, which was the allegedly unproblematic starting point for Aristotle's account of naming, it is all the more plainly true, the objector claims, in the area of the human good. Here it is only a very naïve and historically insensitive moral philosopher who would say that the experience of the fear of death, or of bodily appetites, is a human constant. Recent anthropological work on the social construction of the emotions,16 for example, has shown to what extent the experience of fear has learned and culturally variant elements. When we add that the object of the fear in which the Aristotelian takes an interest is death, which has been so variously interpreted and understood by human beings at different times and in different places the conclusion that the ‘grounding experience’ is an irreducible plurality of experiences, highly various and in each case deeply infused with cultural interpretation, becomes even more inescapable.

Nor is the case different with the apparently less complicated experience of the bodily appetites. Most philosophers who have written about the appetites have treated hunger, thirst, and sexual desire as human universals, stemming from our shared animal nature. Aristotle himself was already more sophisticated, since he insisted that the object of appetite is ‘the apparent good’ and that appetite is therefore something interpretative and selective, a kind of intentional awareness.17 But he does not seem to have reflected much about the ways in which historical and cultural differences could shape that awareness. The Hellenistic philosophers who immediately followed him did so reflect, arguing that the experience (p.253) of sexual desire and of many forms of the desire for food and drink are, at least in part, social constructs, built up over time on the basis of a social teaching about value; this is external to start with, but it enters so deeply into the perceptions of the individual that it actually forms and transforms the experience of desire.18 Let us take two Epicurean examples. People are taught that to be well fed they require luxurious fish and meat, that a simple vegetarian diet is not enough. Over time, the combination of teaching with habit produces an appetite for meat, shaping the individual's perceptions of the objects before him. Again, people are taught that what sexual relations are all about is a romantic union or fusion with an object who is seen as exalted in value, or even as perfect. Over time, this teaching shapes sexual behaviour and the experience of desire, so that sexual arousal itself responds to this culturally learned scenario.19.

This work of social criticism has recently been carried further by Michel Foucault, in his History of Sexuality.20 This work has certain gaps as a history of Greek thought on this topic, but it does succeed in establishing that the Greeks saw the problem of the appetites and their management in an extremely different way from that of twentieth‐century Westerners. To summarize two salient conclusions of his complex argument: first, the Greeks did not single out the sexual appetite for special treatment; they treated it alongside hunger and thirst, as a drive that needed to be mastered and kept within bounds. Their central concern was with self‐mastery, and they saw the appetites in the light of this concern. Furthermore, where the sexual appetite is concerned, they did not regard the gender of the partner as particularly important in assessing the moral value of the act. Nor did they as morally salient a stable disposition to prefer partners of one sex rather than the other. Instead, they focused on the general issue of activity and passivity, connecting it in complex ways with the issue of self‐mastery.

Work like Foucault's — and there is a lot of it in various areas, some of it very good — shows very convincingly that the experience of bodily desire, and of the body itself, has elements that vary with cultural and historical change. The names that people call their desires and themselves as subjects of desire, the fabric of belief and discourse into which they integrate their ideas of desiring: all this influences, it is clear, not only their reflection about desire, but also their experience of desire itself. Thus, for example, it is naïve to treat our modern debates about homosexuality as continuations of the very same debate about sexual activity that went on in the Greek world.21 In a very real sense there was no ‘homosexual experience’ in a culture that did not contain our emphasis on the gender of the partner, the subjectivity of inclination, and the permanence of (p.254) appetitive disposition, nor our particular ways of problematizing certain forms of behaviour.

If we suppose that we can get underneath this variety and this constructive power of social discourse in at least one case — namely, with the universal experience of bodily pain as a bad thing — even here we find subtle arguments against us. For the experience of pain seems to be embedded in a cultural discourse as surely as the closely related experiences of the appetites, and significant variations can be alleged here as well. The Stoics had already made this claim against the Aristotelian virtues. In order to establish that bodily pain is not bad by its very nature, but only by cultural tradition, the Stoics had to provide some explanation for the ubiquity of the belief that pain is bad and of the tendency to shun it. This explanation would have to show that the reaction is learned rather than natural, and to explain why, in the light of this fact, it is learned so widely. This they did by pointing to certain features in the very early treatment of infants. As soon as an infant is born, it cries. Adults, assuming that the crying is a response to its pain at the unaccustomed coldness and harshness of the place where it finds itself, hasten to comfort it. This behaviour, often repeated, teaches the infant to regard its pain as a bad thing — or, better, teaches it the concept of pain, which includes the notion of badness, and teaches it the forms of life its society shares concerning pain. It is all social teaching, they claimed, though this usually escapes our notice because of the early and non‐linguistic nature of the teaching.22

These and related arguments, the objector concludes, show that the Aristotelian idea that there can be a single, non‐relative discourse about human experiences such as mortality or desire is a naïve one. There is no such bedrock of shared experience, and thus no single sphere of choice within which the virtue is the disposition to choose well. So the Aristotelian project cannot even get off the ground.

Now the Aristotelian confronts a third objector, who attacks from a rather different direction. Like the second, she charges that the Aristotelian has taken for a universal and necessary feature of human life an experience that is contingent on certain non‐necessary historical conditions. Like the second, she argues that human experience is much more profoundly shaped by non‐necessary social features than the Aristotelian has allowed. But her purpose is not simply, like the second objector's, to point to the great variety of ways in which the ‘grounding experiences’ corresponding to the virtues are actually understood and lived by human beings. It is more radical still. It is to point out that we could imagine a form of human life that does not contain these experiences — or some of them — at all, in any form. Thus the virtue that consists in acting well in that sphere need not be included in an account of the human good. In some cases, the experience may even be a sign of bad human life, and the corresponding virtue therefore no (p.255) better than a form of non‐ideal adaptation to a bad state of affairs. The really good human life, in such a case, would contain neither the grounding deficiency nor the remedial virtue.

This point is forcefully raised by some of Aristotle's own remarks about the virtue of generosity. One of his arguments against societies that eliminate private ownership is that they thereby do away with the opportunity for generous action, which requires having possessions of one's own to give to others.23 This sort of remark is tailor‐made for the objector, who will immediately say that generosity, if it really rests upon the experience of private possession, is a dubious candidate indeed for inclusion in a purportedly non‐relative account of the human virtues. If it rests upon a grounding experience that is non‐necessary and is capable of being evaluated in different ways, and of being either included or eliminated in accordance with that evaluation, then it is not the universal the Aristotelian said it was.

Some objectors of the third kind will stop at this point, or use such observations to support the second objector's relativism. But in another prominent form this argument takes a non‐relativist direction. It asks us to assess the grounding experiences against an account of human flourishing, produced in some independent manner. If we do so, the objector urges, we will discover that some of the experiences are remediable deficiencies. The objection to Aristotelian virtue ethics will then be that it limits our social aspirations, encouraging us to regard as permanent and necessary what we might in fact improve to the benefit of all human life. This is the direction in which the third objection to the virtues was pressed by Karl Marx, its most famous proponent.24 According to Marx's argument, a number of the leading bourgeois virtues are responses to defective relations of production. Bourgeois justice, generosity, etc., presuppose conditions and structures that are not ideal and that will be eliminated when communism is achieved. And it is not only the current specification of these virtues that will be superseded with the removal of the deficiency. It is the virtues themselves. It is in this sense that communism leads human beings beyond ethics.

The Aristotelian is thus urged to inquire into the basic structures of human life with the daring of a radical political imagination. It is claimed that when she does so she will see that human life contains more possibilities than are dreamed of in her list of virtues.


Each of these objections is profound. To answer any one of them adequately would require a treatise. But we can still do something at this point to map out (p.256) an Aristotelian response to each one, pointing the direction in which a fuller reply might go.

The first objector is right to insist on the distinction between singleness of framework and singleness of answer, and right, again, to stress that in constructing a debate about the virtues based on the demarcation of certain spheres of experience we have not yet answered any of the ‘What is X?’ questions that this debate will confront. We have not even said much about the structure of the debate itself, beyond its beginnings — about how it will both use and criticize traditional beliefs, how it will deal with conflicting beliefs, how it will move critically from the ‘way of one's ancestors’ to the ‘good’ — in short, about whose judgements it will trust. I have addressed some of these issues, again with reference to Aristotle, in two other papers;25 but much more remains to be done. At this point, however, we can make four observations to indicate how the Aristotelian might deal with some of the objector's concerns here. First, the Aristotelian position that I wish to defend need not insist, in every case, on a single answer to the request for a specification of a virtue. The answer might well turn out to be a disjunction. The process of comparative and critical debate will, I imagine, eliminate numerous contenders — for example, the view of justice that prevailed in Cyme. But what remains might well be a (probably small) plurality of acceptable accounts. These accounts may or may not be capable of being subsumed under a single account of greater generality. Success in the eliminative task will still be no trivial accomplishment. If we should succeed in ruling out conceptions of the proper attitude to one's own human worth that are based on a notion of original sin, for example, this would be moral work of enormous significance, even if we got no further than that in specifying the positive account.

Second, the general answer to a ‘What is X?’ question in any sphere may well be susceptible of several or even of many concrete specifications, in connection with other local practices and local conditions. The normative account where friendship and hospitality are concerned, for example, is likely to be extremely general, admitting of many concrete ‘fillings’. Friends in England will have different customs, where regular social visiting is concerned, from friends in ancient Athens. Yet both sets of customs can count as further specifications of a general account of friendship that mentions, for example, the Aristotelian criteria of mutual benefit and well‐wishing, mutual enjoyment, mutual awareness, a shared conception of the good, and some form of ‘living together’.26 Sometimes we may want to view such concrete accounts as optional alternative specifications, to be chosen by a society on the basis of reasons of ease and convenience. Sometimes, on the other hand, we may want to insist that a particular account gives the only legitimate specification of the virtue in question for that concrete context; in that case, the concrete account could be viewed as a part of a longer or fuller version (p.257) of the single normative account. The decision between these two ways of regarding it will depend upon our assessment of its degree of non‐arbitrariness for its context (both physical and historical), its relationship to other non‐arbitrary features of the moral conception of that context, and so forth.

Third, whether we have one or several general accounts of a virtue, and whether this/these accounts do or do not admit of more concrete specifications relative to ongoing cultural contexts, the particular choices that the virtuous person, under this conception, makes will always be a matter of being keenly responsive to the local features of his or her concrete context. So in this respect, again, the instructions the Aristotelian will give to the person of virtue do not differ from part of what a relativist would recommend. The Aristotelian virtues involve a delicate balancing between general rules and a keen awareness of particulars, in which process, as Aristotle stresses, the perception of the particular takes priority. It takes priority in the sense that a good rule is a good summary of wise particular choices, and not a court of last resort. Like rules in medicine and navigation, ethical rules should be held open to modification in the light of new circumstances; and the good agent must therefore cultivate the ability to perceive and correctly describe his or her situation finely and truly, including in this perceptual grasp even those features of the situation that are not covered under the existing rule.

I have written a good deal elsewhere on this idea of the ‘priority of the particular’, exactly what it does and does not imply, in exactly what ways that particular perception is and is not prior to the general rule. Those who want clarification on this central topic will have to turn to those writings.27

What I want to stress here is that Aristotelian particularism is fully compatible with Aristotelian objectivity. The fact that a good and virtuous decision is context‐sensitive does not imply that it is right only relative to, or inside, a limited context, any more than the fact that a good navigational judgement is sensitive to particular weather conditions shows that it is correct only in a local or relational sense. It is right absolutely, objectively, anywhere in the human world, to attend to the particular features of one's context; and the person who so attends and who chooses accordingly is making, according to Aristotle, the humanly correct decision, period. If another situation should ever arise with all the same ethically relevant features, including contextual features, the same decision would again be absolutely right.28

It should be stressed that the value of contextual responsiveness and the value of getting it right are seen by the Aristotelian as mutually supportive here, rather than in tension. For the claim is that only when we have duly responded to the complexities of the context, seeing it for the very historical situation it is, will (p.258) we have any hope of making the right decision. Short of that, the importation of plausible general values, however well intentioned, may do no good at all, and may actually make things worse. Nor, the Aristotelian argues, have we been sufficiently responsive to the context before us if we do not see the humanity in it: do not, that is, respond to the claims of human need, the strivings towards the good, the frustrations of human capability, that this situation displays to the reflective person. To study it with detached scientific interest, as an interesting set of local traditions, is not to respond sufficiently to the concrete situation it is; for whatever it is, it is concretely human.

An example from the development context will illustrate this mutual support. In A Quiet Revolution, an eloquent study of women's education in rural Bangladesh,29 Martha Chen describes the efforts of a government development group, the Bangladesh Rural Advancement Committee, to increase the rate of female literacy in certain rural areas. The project began from a conviction that literacy is an important ingredient in the development of these women towards greater capability to live well. It was seen as closely linked with other important values, such as economic flourishing, autonomy, and self‐respect. This conviction did not derive from the local traditions of the villages, where women had in fact little autonomy and no experience of education; it derived from the experiences and reflections of the development workers, who were themselves from many different backgrounds and two nationalities. (Chen herself is an American with a Ph.D. in Sanskrit.) The group as a whole lacked experience of the concrete ways of life of rural women, and thus had, as Chen says, ‘no specific concepts or strategies’30 for working with them. In the first phase of the programme, then, the development workers went directly to the rural villages with their ideas of literacy and its importance, offering adult literacy materials borrowed from another national programme, and trying to motivate the women of the communities they entered to take them on.

But their lack of contextual knowledge made it impossible for them to succeed, in this first phase. Women found the borrowed literacy materials boring and irrelevant to their lives. They did not see how literacy would help them; even the accompanying vocational training was resisted, since it focused on skills for which there was little demand in that area. Thus failure led the agency to rethink their approach. On the one hand, they never abandoned their basic conviction that literacy was important for these women; their conclusion, based on wide experience and on their picture of what the women's lives might be, still seemed sound. On the other hand, they recognized that far more attention to the lives and thoughts of the women involved would be necessary if they were going to come up with an understanding of what literacy might do and be for them. They began to substitute for the old approach a more participatory one, in which local co‐operative groups brought together development workers with local women, (p.259) whose experience and sense of life were regarded as crucial. This concept of the co‐operative group led to a much more complex understanding of the situation, as the development workers grasped the network of relationships within which the women had to function and the specific dimensions of their poverty and constraints, and as the women grasped the alternative possibilities and began to define for themselves a set of aspirations for change. The result, which continues, has been a slow and complex evolution in the role of women in the villages. A visiting journalist wrote, some years later:

I saw the seeds of the quiet revolution starting in village women's lives. At the meeting houses BRAC has built, the wives, young and old, are learning to read and write. Forbidden from doing marketing, they now at least can keep the accounts . . . In one fishing village, the women have even become the bankers, saving over $2000 and lending it to their men to buy better equipment. It started in the simplest way—they collected a handful of rice a week from each family, stored it, and sold it in the market. About 50 villages from each area have thriving women's cooperatives, investing in new power‐pumps or seed, and winning respect for their members.31

This is how the Aristotelian approach works—hanging on to a general (and open‐ended) picture of human life, its needs and possibilities, but at every stage immersing itself in the concrete circumstances of history and culture. Chen's detailed narrative—which in its very style manifests a combination of Aristotelian commitment to the human good and Aristotelian contextual sensitivity—shows that the two elements go, and must go, together. If the development workers had approached these women as alien beings whose ways could not be compared with others and considered with a view to the human good, no change would have taken place—and the narrative convinces the reader that these changes have been good. On the other hand, general talk of education and self‐respect did nothing at all until it came from within a concrete historical reality. Immersion made it possible to get the choice that was humanly right.

Thus the Aristotelian virtue‐based morality can capture a great deal of what the relativist is after, and still make a claim to objectivity, in the sense we have described. In fact, we might say that the Aristotelian virtues do better than the relativist virtues in explaining what people are actually doing when they scrutinize the features of their context carefully, looking at both the shared and the non‐shared features with an eye to what is best. For, as Aristotle says, people who do this are usually searching for the good, not just for the way of their ancestors. They are prepared to defend their decisions as good or right, and to think of those who advocate a different course as disagreeing about what is right, not just narrating a different tradition.

Finally, we should point out that the Aristotelian virtues, and the deliberations they guide, unlike some systems of moral rules, remain always open to revision (p.260) in the light of new circumstances and new evidence. In this way, again, they contain the flexibility to local conditions that the relativist would desire—but, again, without sacrificing objectivity. Sometimes the new circumstances may simply give rise to a new concrete specification of the virtue as previously defined; in some cases it may cause us to change our view about what the virtue itself is. All general accounts are held provisionally, as summaries of correct decisions and as guides to new ones. This flexibility, built into the Aristotelian procedure, will again help the Aristotelian account to answer the questions of the relativist, without relativism.


We must now turn to the second objection. Here, I believe, is the really serious threat to the Aristotelian position. Past writers on virtue, including Aristotle himself, have lacked sensitivity to the ways in which different traditions of discourse, different conceptual schemes, articulate the world, and also to the profound connections between the structure of discourse and the structure of experience itself. Any contemporary defence of the Aristotelian position must display this sensitivity, responding somehow to the data that the relativist historian or anthropologist brings forward.

The Aristotelian should begin, it seems to me, by granting that with respect to any complex matter of deep human importance there is no ‘innocent eye’, no way of seeing the world that is entirely neutral and free of cultural shaping. The work of philosophers such as Putnam, Goodman, and Davidson32—following, one must point out, from the arguments of Kant and, I believe, from those of Aristotle himself33—have shown convincingly that even where sense‐perception is concerned, the human mind is an active and interpretative instrument, and that its interpretations are a function of its history and its concepts, as well as of its innate structure. The Aristotelian should also grant, it seems to me, that the nature of human world interpretations is holistic and that the criticism of them must, equally, be holistic. Conceptual schemes, like languages, hang together as whole structures, and we should realize, too, that a change in any single element is likely to have implications for the system as a whole.

But these two facts do not imply, as some relativists in literary theory and in anthropology tend to assume, that all world interpretations are equally valid and altogether non‐comparable, that there are no good standards of assessment and ‘anything goes’. The rejection of the idea of ethical truth as correspondence to an altogether uninterpreted reality does not imply that the whole idea of searching for the truth is an old‐fashioned error. Certain ways in which people see the world can still be criticized exactly as Aristotle (p.261) criticized them: as stupid, pernicious, and false. The standards used in such criticisms must come from inside human life. (Frequently they will come from the society in question itself, from its own rationalist and critical traditions.) And the inquirer must attempt, prior to criticism, to develop an inclusive understanding of the conceptual scheme being criticized, seeing what motivates each of its parts and how they hang together. But there is so far no reason to think that the critic will not be able to reject the institution of slavery, or the homicide law of Cyme, as out of line with the conception of virtue that emerges from reflection on the variety of different ways in which human cultures have had the experiences that ground the virtues.

The grounding experiences will not, the Aristotelian should concede, provide precisely a single, language‐neutral bedrock on which an account of virtue can be straightforwardly and unproblematically based. The description and assessment of the ways in which different cultures have constructed these experiences will become one of the central tasks of Aristotelian philosophical criticism. But the relativist has, so far, shown no reason why we could not, at the end of the day, say that certain ways of conceptualizing death are more in keeping with the totality of our evidence and the totality of our wishes for flourishing life than others; that certain ways of experiencing appetitive desire are for similar reasons more promising than others.

Relativists tend, furthermore, to understate the amount of attunement, recognition, and overlap that actually obtains across cultures, particularly in the areas of the grounding experiences. The Aristotelian, in developing her conception in a culturally sensitive way, should insist, as Aristotle himself does, upon the evidence of such attunement and recognition. Despite the evident differences in the specific cultural shaping of the grounding experiences, we do recognize the experiences of people in other cultures as similar to our own. We do converse with them about matters of deep importance, understand them, allow ourselves to be moved by them. When we read Sophocles' Antigone, we see a good deal that seems strange to us; and we have not read the play well if we do not notice how far its conceptions of death, womanhood, and so on differ from our own. But it is still possible for us to be moved by the drama, to care about its people, to regard their debates as reflections upon virtue that speak to our own experience, and their choices as choices in spheres of conduct in which we too must choose. Again, when one sits down at a table with people from other parts of the world and debates with them concerning hunger, or just distribution, or in general the quality of human life, one does find, in spite of evident conceptual differences, that it is possible to proceed as if we were all talking about the same human problem; and it is usually only in a context in which one or more of the parties is intellectually committed to a theoretical relativist position that this discourse proves impossible to sustain. This sense of community and overlap seems to be especially strong in the areas that we have called the areas of the grounding experiences. And this, it seems, supports the Aristotelian claim that those (p.262) experiences can be a good starting point for ethical debate.

Furthermore, it is necessary to stress that hardly any cultural group today is as focused upon its own internal traditions and as isolated from other cultures as the relativist argument presupposes. Cross‐cultural communication and debate are ubiquitous facts of contemporary life, and our experience of cultural interaction indicates that in general the inhabitants of different conceptual schemes do tend to view their interaction in the Aristotelian and not the relativist way. A traditional society, confronted with new technologies and sciences, and the conceptions that go with them, does not in fact simply fail to understand them, or regard them as totally alien incursions upon a hermetically sealed way of life. Instead, it assesses the new item as a possible contributor to flourishing life, making it comprehensible to itself, and incorporating elements that promise to solve problems of flourishing. Examples of such assimilation, and the debate that surrounds it,34 suggest that the parties do in fact recognize common problems and that the traditional society is perfectly capable of viewing an external innovation as a device to solve a problem that it shares with the innovating society. The village woman of Chen's narrative, for example, did not insist on remaining illiterate because they had always been so. Instead, they willingly entered into dialogue with the international group, viewing co‐operative discussion as a resource towards a better life. The parties do in fact search for the good, not the way of their ancestors; only traditionalist anthropologists insist, nostalgically, on the absolute preservation of the ancestral.

And this is so even when cross‐cultural discourse reveals a difference at the level of the conceptualization of the grounding experiences. Frequently the effect of work like Foucault's, which reminds us of the non‐necessary and non‐universal character of one's own ways of seeing in some such area, is precisely to prompt a critical debate in search of the human good. It is difficult, for example, to read Foucault's observations about the history of our sexual ideas without coming to feel that certain ways in which the Western contemporary debate on these matters has been organized, as a result of some combination of Christian moralism with nineteenth‐century pseudoscience, are especially silly, arbitrary, and limiting, inimical to a human search for flourishing. Foucault's moving account of Greek culture, as he himself insists in a preface,35 provides not only a sign that someone once thought differently, but also evidence that it is possible for us to think differently. (Indeed, this was the whole purpose of genealogy as Nietzsche, Foucault's precursor here, introduced it: to destroy idols once deemed necessary, and to clear the way for new possibilities of creation.) Foucault announced that the purpose of his book was to ‘free thought’ so that it could think differently, imagining new and more fruitful possibilities. And close analysis of spheres of cultural (p.263) discourse, which stresses cultural differences in the spheres of the grounding experiences, is being combined, increasingly, in current debates about sexuality and related matters, with a critique of existing social arrangements and attitudes, and an elaboration of new norms of human flourishing. There is no reason to think this combination incoherent.36

As we pursue these possibilities, the basic spheres of experience identified in the Aristotelian approach will no longer, we have said, be seen as spheres of uninterpreted experience. But we have also insisted that there is much family relatedness and much overlap among societies. And certain areas of relatively greater universality can be specified here, on which we should insist as we proceed to areas that are more varied in their cultural expression. Not without a sensitive awareness that we are speaking of something that is experienced differently in different contexts, we can none the less identify certain features of our common humanity, closely related to Aristotle's original list, from which our debate might proceed.

  1. 1. Mortality. No matter how death is understood, all human beings face it and (after a certain age) know that they face it. This fact shapes every aspect of more or less every human life.

  2. 2. The body. Prior to any concrete cultural shaping, we are born with human bodies, whose possibilities and vulnerabilities do not as such belong to any culture rather than any other. Any given human being might have belonged to any culture. The experience of the body is culturally influenced; but the body itself, prior to such experience, provides limits and parameters that ensure a great deal of overlap in what is going to be experienced, where hunger, thirst, desire, and the five senses are concerned. It is all very well to point to the cultural component in these experiences. But when one spends time considering issues of hunger and scarcity, and in general of human misery, such differences appear relatively small and refined, and one cannot fail to acknowledge that ‘there are no known ethnic differences in human physiology with respect to metabolism of nutrients. Africans and Asians do not burn their dietary calories or use their dietary protein any differently from Europeans and Americans. It follows then that dietary requirements cannot vary widely as between different races.’37 This and similar facts should surely be focal points for debate about appropriate human behaviour in this sphere. And by beginning with the body, rather than with the subjective experience of desire, we get, furthermore, an opportunity to criticize the situation of people who are so persistently deprived that their desire for good things has actually decreased. This is a further advantage of the Aristotelian approach, when contrasted (p.264) with approaches to choice that stop with subjective expressions of preference.

  3. 3. Pleasure and pain. In every culture, there is a conception of pain; and these conceptions, which overlap very largely with one another, can plausible be seen as grounded in universal and pre‐cultural experience. The Stoic story of infant development is highly implausible; the negative response to bodily pain is surely primitive and universal, rather than learned and optional, however much its specific ‘grammar’ may be shaped by later learning.

  4. 4. Cognitive capability. Aristotle's famous claim that ‘all human beings by nature reach out for understanding’38 seems to stand up to the most refined anthropological analysis. It points to an element in our common humanity that is plausibly seen, again, as grounded independently of particular acculturation, however much it is later shaped by acculturation.

  5. 5. Practical reason. All human beings, whatever their culture, participate (or try to) in the planning and managing of their lives, asking and answering questions about how one should live and act. This capability expresses itself differently in different societies, but a being who altogether lacked it would not be likely to be acknowledged as a human being, in any culture.39

  6. 6. Early infant development. Prior to the greater part of specific cultural shaping, though perhaps not free from all shaping, are certain areas of human experience and development that are broadly shared and of great importance for the Aristotelian virtues: experiences of desire, pleasure, loss, one's own finitude, perhaps also of envy, grief, and gratitude. One may argue about the merits of one or another psychoanalytical account of infancy. But it seems difficult to deny that the work of Freud on infant desire and of Klein on grief, loss, and other more complex emotional attitudes has identified spheres of human experience that are to a large extent common to all humans, regardless of their particular society. All humans begin as hungry babies, perceiving their own helplessness, their alternating closeness to and distance from those on whom they depend, and so forth. Melanie Klein records a conversation with an anthropologist in which an event that at first looked (to Western eyes) bizarre was interpreted by Klein as the expression of a universal pattern of mourning. The anthropologist accepted her interpretation.40

  7. 7. Affiliation. Aristotle's claim that human beings as such feel a sense of fellowship with other human beings, and that we are by nature social animals, is an empirical claim; but it seems to be a sound one. However varied our specific conceptions of friendship and love are, there is a great point in seeing them as overlapping expressions of the same family of shared human needs and desires.

  8. 8. Humour. There is nothing more culturally varied than humour; and yet, as Aristotle insists, some space for humour and play seems to be a need of (p.265) any human life. The human being is not called the ‘laughing animal’ for nothing; it is certainly one of our salient differences from almost all other animals, and (in some form or other) a shared feature, I somewhat boldly assert, of any life that is going to be counted as fully human.

This is just a list of suggestions, closely related to Aristotle's list of common experiences. One could subtract some of these items and/or add others.41 But it seems plausible to claim that in all these areas we have a basis for further work on the human good. We do not have a bedrock of completely uninterpreted ‘given’ data, but we do have nuclei of experience around which the constructions of different societies proceed. There is no Archimedean point here, no pure access to unsullied ‘nature’—even, here, human nature—as it is in and of itself. There is just human life as it is lived. But in life as it is lived, we do find a family of experiences, clustering around certain focuses, which can provide reasonable starting points for cross‐cultural reflection.

This paper forms part of a larger project. The role of the preliminary list proposed in this section can be better understood if I briefly set it in the context of this more comprehensive enterprise, showing its links with other arguments. In a paper entitled ‘Nature, Function, and Capability: Aristotle on Political Distribution’,42 I discuss an Aristotelian conception of the proper function of government, according to which its task is to make available to each and every member of the community the basic necessary conditions of the capability to choose and live a fully good human life, with respect to each of the major human functions included in that fully good life. I examine sympathetically Aristotle's argument that, for this reason, the task of government cannot be well performed, or its aims well understood, without an understanding of these functionings. A closely connected study, ‘Aristotelian Social Democracy’,43 shows a way of moving from a general understanding of the circumstances and abilities of human beings (such as this list provides) to an account of the most important human functions that it will be government's job to make possible. It shows how this understanding of the human being and the political task can yield a conception of social democracy that is a plausible alternative to liberal conceptions.

Meanwhile, in a third paper, ‘Aristotle on Human Nature and the Foundations of Ethics’,44 I focus on the special role of two of the human capabilities (p.266) recognized in this list: affiliation (or sociability) and practical reason. I argue that these two play an architectonic role in human life, suffusing and also organizing all the other functions — which will count as truly human functions only in so far as they are done with some degree of guidance from both of these. Most of the paper is devoted to an examination of Aristotle's arguments for saying that these two elements are parts of ‘human nature’. I argue that this is not an attempt to base human ethics on a neutral bedrock of scientific fact outside of human experience and interpretation. I claim that Aristotle seeks, instead, to discover, among the experiences of groups in many times and places, certain elements that are especially broadly and deeply shared. And I argue that the arguments justifying the claims of these two to be broad and deep in this way have a self‐validating structure: that is, anyone who participates in the first place in the inquiry that supports them affirms, by that very fact, her own recognition of their salience. This is an important continuation of the project undertaken in this paper, since it shows exactly how Aristotle's ‘foundation’ for ethics can remain inside human history and self‐interpretation, and yet still claim to be a foundation.


The third objection raises, at bottom, a profound conceptual question: What is it to inquire about the human good? What circumstances of existence go to define what it is to live the life of a human being, and not some other life? Aristotle likes to point out that an inquiry into the human good cannot, on pain of incoherence, end up describing the good of some other being, say a god—a good that, on account of our circumstances, it is impossible for us to attain.45 What circumstances then? The virtues are defined relatively to certain problems and limitations, and also to certain endowments. Which ones are sufficiently central that their removal would make us into different beings, and open up a wholly new and different debate about the good? This question is itself part of the ethical debate we propose. For there is no way to answer it but to ask ourselves which elements of our experience seem to us so important that they count, for us, as part of who we are. I discuss Aristotle's attitude to this question elsewhere, and I shall simply summarize here.46 It seems clear, first of all, that our mortality is an essential feature of our circumstances as human beings. An immortal being would have such a different form of life, and such different values and virtues, that it does not seem to make sense to regard that being as part of the same search for good. Essential, too, will be our dependence upon the world outside us: some sort of need for food, drink, the help of others. On the side of abilities, we would want to include cognitive (p.267) functioning and the activity of practical reasoning as elements of any life that we would regard as human. Aristotle argues, plausibly, that we would want to include sociability as well, some sensitivity to the needs of and pleasure in the company of other beings similar to ourselves.

But it seems to me that the Marxist question remains, as a deep question about human forms of life and the search for the human good. For one certainly can imagine forms of human life that do not contain the holding of private property—not, therefore, those virtues that have to do with its proper management. And this means that it remains an open question whether these virtues ought to be regarded as virtues, and kept upon our list. Marx wished to go much further, arguing that communism would remove the need for justice, courage, and most of the bourgeois virtues. I think we might be sceptical here. Aristotle's general attitude to such transformations of life is to suggest that they usually have a tragic dimension. If we remove one sort of problem—say, by removing private property—we frequently do so by introducing another—say, the absence of a certain sort of freedom of choice, the freedom that makes it possible to do fine and generous actions for others. If things are complex even in the case of generosity, where we can easily imagine the transformation that removes the virtue, they are surely far more so in the cases of justice and courage. And we would need a far more detailed description than Marx ever gives us of the form of life under communism, before we could even begin to see whether it would in fact transform things where these virtues are concerned, and whether it would or would not introduce new problems and limitations in their place.

In general it seems that all forms of life, including the imagined life of a god, contain boundaries and limits.47 All structures, even that of putative limitlessness, are closed to something, cut off from something—say, in that case, from the specific value and beauty inherent in the struggle against limitation. Thus it does not appear that we will so easily get beyond the virtues. Nor does it seem to be so clearly a good thing for human life that we should.


The best conclusion to this sketch of an Aristotelian programme for virtue ethics was written by Aristotle himself, at the end of his discussion of human nature in Nicomachean Ethics I:

So much for our outline sketch for the good. For it looks as if we have to draw an outline first, and fill it in later. It would seem to be open to anyone to take things further and to articulate the good parts of the sketch. And time is a good discoverer or ally in such things. That's how the sciences have progressed as well: it is open to anyone to supply what is lacking. (EN 1098a20–6)

(p.268) Bibliography

Bibliography references:

Abelove, H. (1987). ‘Is Gay History Possible?’ Paper (unpublished) delivered at the Conference on Homosexuality in History and Culture, Brown University, Feb. 1987.

Abeysekera, C. (1986). Address (unpublished) to the WIDER Conference on Value and Technology, summer 1986.

Chen, M. (1986). A Quiet Revolution: Women in Transition in Rural Bangladesh. Cambridge, Mass.: Schenkman Publishing Company.

Davidson, D. (1984). Inquiries into Truth and Interpretation. Oxford: Clarendon Press.

Foot, P. (1978). Virtues and Vices. Berkeley: University of California Press.

Foucault, M. (1984). Histoire de la sexualité, ii, iii. Paris: Les Belles Lettres.

Goodman, N. (1968). Languages of Art. Indianapolis: Hackett.

—— (1978). Ways of World‐Making. Indianapolis: Hackett.

Gopalan, C. (1992). ‘Undernutrition: Measurement and Implications’, In S. Osmani (ed.), Nutrition and Poverty. Oxford: Clarendon Press.

Halperin, D. (1990). One Hundred Years of Homosexuality and Other Essays on Greek Love. New York: Routledge, Chapman, and Hall.

——, Winkler, J., and Zeitlin, F. (1990). Before Sexuality. Princeton: Princeton University Press.

Hampshire, S. (1983). Morality and Conflict. Cambridge, Mass.: Harvard University Press.

Harré, R. (ed.) (1986). The Social Construction of Emotions. Oxford: Basil Blackwell.

Klein, M. (1984). ‘Our Adult World and its Roots in Infancy’, in Envy, Gratitude and Other Works 1946–1963. London: Hogarth Press, 247–63.

Lukes, S. (1987). Marxism and Morality. Oxford: Clarendon Press.

Lutz, C. (1988). Unnatural Emotions. Chicago: University of Chicago Press.

Macintyre, A. (1981). After Virtue. Notre Dame: Notre Dame University Press.

—— (1988). Whose Justice? Which Rationality? Notre Dame: Notre Dame University Press.

Nussbaum, M. (1972). ‘Psuche in Heraclitus’, Phronesis, 17, 1–17, 153–70.

—— (1978). Aristotle's De Motu Animalium. Princeton: Princeton University Press.

—— (1985). ‘The Discernment of Perception: An Aristotelian Model of Public and Private Rationality’, Proceedings of the Boston Area Colloquium for Ancient Philosophy, 1, 151–201. Also in Nussbaum (1990).

—— (1986a). The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy. Cambridge: Cambridge University Press.

—— (1986b). ‘Therapeutic Arguments: Epicurus and Aristotle’, in M. Schofield and G. Striker (eds.), The Norms of Nature. Cambridge: Cambridge University Press, 31–74.

—— (1987a). ‘The Stoics on the Extirpation of the Passions', Apeiron, 20, 129–77.

—— (1987b). “Finely Aware and Richly Responsible”: Literature and the Moral Imagination’, in A. Cascardi (ed.), Literature and the Question of Philosophy. Baltimore: Johns Hopkins University Press, 169–91.

—— (1988a). ‘Nature, Function, and Capability: Aristotle on Political Distribution’, Oxford Studies in Ancient Philosophy, suppl. vol., 145–84.

—— (1988b). ‘Narrative Emotions: Beckett's Genealogy of Love’, Ethics, 98, 225–54.

(p.269) —— (1989). ‘Beyond Obsession and Disgust: Lucretius' Genealogy of Love’, Apeiron, 22, 1–59.

—— (1990a). ‘Mortal Immortals: Lucretius on Death and the Voice of Nature’, Philosophy and Phenomenological Research, 50, 305–51.

—— (1990b). Love's Knowledge: Essays on Philosophy and Literature. Oxford: Oxford University Press.

—— (1990c). ‘Aristotelian Social Democracy’, in R. B. Douglass, G. Mara, and H. Richardson (eds.). Liberalism and the Good. New York: Routledge, 203–52.

—— (forthcoming a). ‘Aristotle on Human Nature and the Foundations of Ethics’, in a volume on the philosophy of Bernard Williams, ed. J. Altham and R. Harrison. Cambridge: Cambridge University Press.

—— (forthcoming b). The Therapy of Desire: Theory and Practice in Hellenistic Ethics. The Martin Classical Lectures, 1986.

—— and Sen, A. (1989). ‘Internal Criticism and Indian Rationalist Traditions’, in M. Krausz (ed.), Relativism. Notre Dame: Notre Dame University Press.

Procope, J. (forthcoming). ‘Hochherzigkeit’, Reallexikon für Antike und Christentum, 14.

Putnam, H. (1979). Meaning and the Moral Sciences. London: Routledge and Kegan Paul.

—— (1981). Reason, Truth, and History. Cambridge: Cambridge University Press.

—— (1988). The Many Faces of Realism: The Carus Lectures. La Salle, Ill. Open Court.

Sturgeon, N. (1984). Review of Foot, Journal of Philosophy, 81, 326–33.

Walzer, M. (1983). Spheres of Justice. Oxford: Basil Blackwell.

—— (1987). Interpretation and Social Criticism. Cambridge, Mass.: Harvard University Press.

Williams, B. (1984). ‘Philosophy’, in The Legacy of Greece, ed. M. I. Finley. Oxford: Oxford University Press.

—— (1985). Ethics and the Limits of Philosophy. Cambridge, Mass.: Harvard University Press.

Winkler, J. (1990). The Constraints of Desire. New York: Routledge, Chapman, and Hall.


This paper was originally motivated by questions discussed at the WIDER Conference on Value and Technology, summer 1986, Helsinki. I would like to thank Steve and Frédérique Marglin for provoking some of these arguments, with hardly any of which they will agree. I would also like to thank Dan Brock for his helpful comments, Amartya Sen for many discussions of the issues, and the participants in the WIDER conference for their helpful questions and comments. Earlier versions of the paper were presented at the University of New Hampshire and at Villanova University; I am grateful to the audiences on those occasions for stimulating discussion. An earlier version of the paper was published in Midwest Studies in Philosophy, 1988.

(1) See MacIntyre (1981), and by contrast MacIntyre (1988); Foot (1978); Williams (1984, 1985); Walzer (1983, 1987).

(2) For examples of this, see Nussbaum (1988a).

(3) See, for example, Williams (1985: 34–6); Hampshire (1983: 150ff.).

(4) For ‘nameless’ virtues and vice, see EN (Nicomachean Ethics) 1107b1–2, 1107b7–8, 1107b30–1, 1108a17, 1119a10–11, 1126b20, 1127a12, 1127a14; for recognition of the unsatisfactoriness of names given, see 1107b8, 1108a5–6, 1108a20 ff. The two categories are largely overlapping, on account of the general principle enunciated at 1108a16–19, that where there is no name a name should be given, satisfactory or not.

(5) It should be noted that this emphasis on spheres of experience is not present in the Eudemian Ethics, which begins its discussion with a list of virtues and vices. This seems to me a sign that that treatise expresses a more primitive stage of Aristotle's thought on the virtues—whether earlier or not.

(6) For statements with peri (‘concerning’) connecting virtues with spheres of life, see EN 1115a6–7, 1117a29–30, 1117b25 and 27, 1119b23, 1122a19, 1122b34, 1125b26, 1126b13—and EN II. 7 throughout. See also the related usages at 1126b11, 1127b32.

(7) My list here inserts justice in a place of prominence. (In the EN it is treated separately, after all the other virtues, and the introductory list defers it for that later examination.) I have also added at the end of the list categories corresponding to the various intellectual virtues discussed in EN VI, and also to phronesis, or practical wisdom, discussed in EN VI as well. Otherwise the order and wording of my list closely follows II. 7, which gives the programme for the more detailed analyses from III. 5 to IV.

(8) For a longer account of this, with references to the literature and to related philosophical discussions, see Nussbaum (1986a: ch. 8).

(9) Aristotle does not worry about questions of translation in articulating this idea; for some worries about this, and an Aristotelian response, see below, sections 4 and 6.

(10) Posterior Analytics II. 8, 93a21ff.; see Nussbaum (1986a: ch. 8).

(11) Heraclitus, fragment Diels‐Kranz B23; see Nussbaum (1972).

(12) See Politics I. 2, 1253a1–18; that discussion does not explicitly deny virtues to the gods, but this denial is explicit in EN 1145a25–7 and 1178b10ff.

(13) Aristotle does not make the connection with his account of language explicit, but his project is one of defining the virtues, and we would expect him to keep his general view of defining in mind in this context. A similar idea about the virtues, and about the way in which a certain sort of experience can serve as a plausible basis for a non‐relative account, is developed (without reference to Aristotle) in Sturgeon (1984).

(14) See EN 1107a5, where Aristotle says that the virtues and the corresponding person are ‘pretty much nameless’, and says, ‘Let us call . . . ’ when he introduces the names. See also 1125b29, 1126a3–4.

(15) See Procope (forthcoming).

(16) See, for example, Harré (1986); Lutz (1988).

(17) See Nussbaum (1978: notes on ch. 6), and Nussbaum (1986a: ch. 9).

(18) A detailed study of the treatment of these ideas in the three major Hellenistic schools is presented in Nussbaum (forthcoming b); portions are published in Nussbaum (1986b, 1987a, 1989, 1990a); see also Nussbaum (1988b).

(19) The relevant texts are discussed in Nussbaum (forthcoming b); see also Nussbaum (1986b, 1989, 1990a).

(20) Foucault (1984).

(21) See also Halperin (1990); Winkler (1990); Halperin, Winkler, and Zeitlin (1990).

(22) The evidence for this part of the Stoic view is discussed in Nussbaum (forthcoming b); for a general account of the Stoic account of the passions, see Nussbaum (1987a).

(23) Politics 1263b11ff.

(24) For discussion of the relevant passages in Marx, see Lukes (1987). For an acute discussion of these issues I am indebted to an exchange between Alan Ryan and Stephen Lukes at the Oxford Philosophical Society, March 1987.

(25) Nussbaum (1986a; ch. 8), and Nussbaum and Sen (1989).

(26) See Nussbaum (1986a: ch. 12).

(27) Nussbaum (1986a: ch. 10); Nussbaum (1985, 1987b).

(28) I believe, however, that some morally relevant features, in the Aristotelian view, may be features that are not, even in principle, replicable in another context. See Nussbaum (1986a: ch. 10) and Nussbaum (1985, 1990b).

(29) Chen (1986).

(30) Chen (1986: p. ix).

(31) Cited in Chen (1986: 4–5). Chen stresses that one important factor in this later success was that the group had no dogmatic adherence to an abstract theory of development, but had a flexible and situation‐guided approach.

(32) See Putnam (1979, 1981, 1988); Goodman (1968, 1978); Davidson (1984).

(33) On his debt to Kant, see Putnam (1988); on Aristotle's relationship to ‘internal realism’, see Nussbaum (1986a: ch. 8).

(34) Abeysekera (1986).

(35) Foucault (1984: ii, preface).

(36) This paragraph expands remarks made in a commentary on papers by D. Halperin and J. Winkler at the conference on ‘Homosexuality in History and Culture’ at Brown University, February 1987; Halperin's paper is now in Halperin (1990) and Winkler's in Halperin, Winkler, and Zeitlin (1990). The proposed combination of historically sensitive analysis with cultural criticism was forcefully developed, at the same conference, by Abelove (1987).

(37) Gopalan (forthcoming).

(38) Metaphysics I. 1.

(39) See Nussbaum (1988), where this Aristotelian view is compared with Marx's views on truly human functioning.

(40) Klein (1984): 247–63.

(41) See Nussbaum (1990c) for a slightly longer list, including discussion of our relationship to other species and to the world of nature. It is very interesting to notice that three other lists in this volume, prepared independently, contain almost the same items as this one: Dan Brock's list of basic human functions used in quality of life measures in medical ethics; Erik Allardt's enumeration of the functions observed by Finnish social scientists; and Robert Erikson's list of functions measured by the Swedish group. Only the last two may show mutual influence. So much independent convergence testifies to the ubiquity of these concerns and to their importance.

(42) Nussbaum (1988a).

(43) Nussbaum (1990c).

(44) Nussbaum (forthcoming a).

(45) Cl. EN 1159a 10–12, 1166a 18–23.

(46) Nussbaum (forthcoming a).

(47) See Nussbaum (1986b: ch. 11).