Final paper

After Virtue, Alasdair MacIntyre

In this text Alasdair MacIntyre dismisses the concept and possibly even the existence of virtues.

CHAPTER 14

THE NATURE OF THE VIRTUES

One response to the history which I have narrated so far might well be to suggest that even within the relatively coherent tradition of thought which I have sketched there are just too many different and incompatible conceptions of a virtue for there to be any real unity to the concept or indeed to the history. Homer, Sophocles, Aristotle, the New Testament and medieval thinkers differ from each other in too many ways. They offer us different and incompatible lists of the virtues; they give a different rank order of importance to different virtues; and they have different and incompatible theories of the virtues. If we were to consider later Western writers on the virtues, the list of differences and incompatibilities would be enlarged still further; and if we extended our enquiry to Japanese, say, or American Indian cultures, the differences would become greater still. It would be all too easy to conclude that there are a number of rival and alternative conceptions of the virtues, but, even within the tradition which I have been delineating, no single core conception.

The case for such a conclusion could not be better constructed than by beginning from a consideration of the very different lists of items which different authors in different times and places have included in their catalogues of virtues. Some of these catalogues—Homer's, Aristotle's and the New Testament's—I have already noticed at greater or lesser length. Let me at the risk of some repetition recall some of their key features and then introduce for further comparison the catalogues of two later Western writers, Benjamin Franklin and Jane Austen.

The first example is that of Homer. At least some of the items in a Homeric list of the aretat would clearly not be counted by most of us nowadays as virtues at all, physical strength being the most obvious example. To this it might be replied that perhaps we ought not to translate the word arete in Homer by our word ‘virtue', but instead by our word ‘excellence'; and perhaps, if we were so to translate it, the apparently surprising difference between Homer and ourselves would at first sight have been removed. For we could allow without any kind of oddity that the possession of physical strength is the possession of an excellence. But in fact we would not have removed, but instead would merely have relocated, the difference between Homer and ourselves. For we would now seem to be saying that Homer's concept of an arete, an excellence, is one thing and that our concept of a virtue is quite another since a particular quality can be an excellence in Homer's eyes, but not a virtue in ours and vice versa.

But of course it is not that Homer's list of virtues differs only from our own; it also notably differs from Aristotle's. And Aristotle's of course also differs from our own. For one thing, as I noticed earlier, some Greek virtue-words are not easily translated into English or rather out of Greek. Moreover consider the importance of friendship as a virtue in Aristotle's list—how different from us! Or the place of phronesis—how different from Homer and from us! The mind receives from Aristotle the kind of tribute which the body receives from Homer. But it is not just the case that the difference between Aristotle and Homer lies in the inclusion of some items and the omission of others in their respective catalogues. It turns out also in the way in which those catalogues are ordered, in which items are ranked as relatively central to human excellence and which marginal.

Moreover the relationship of virtues to the social order has changed. For Homer the paradigm of human excellence is the warrior; for Aristotle it is the Athenian gentleman. Indeed according to Aristotle certain virtues are only available to those of great riches and of high social status; there are virtues which are unavailable to the poor man, even if he is a free man. And those virtues are on Aristotle's view ones central to human life; magnanimity—and once again, any translation of megalopsuchia is unsatisfactory—and munificence are not just virtues, but important virtues within the Aristotelian scheme.

At once it is impossible to delay the remark that the most striking contrast with Aristotle's catalogue is to be found neither in Homer's nor in our own, but in the New Testament's. For the New Testament not only praises virtues of which Aristotle knows nothing—faith, hope and love—and says nothing about virtues such as phronesis which are crucial for Aristotle, but it praises at least one quality as a virtue which Aristotle seems to count as one of the vices relative to magnanimity, namely humility. Moreover since the New Testament quite clearly sees the rich as destined for the pains of Hell, it is clear that the key virtues cannot be available to them; yet they are available to slaves. And the New Testament of course differs from both Homer and Aristotle not only in the items included in its catalogue, but once again in its rank ordering of the virtues.

Turn now to compare all three lists of virtues considered so far—the Homeric, the Aristotelian, and the New Testament's—with two much later lists, one which can be compiled from Jane Austen's novels and the other which Benjamin Franklin constructed for himself. Two features stand out in Jane Austen's list. The first is the importance that she allots to the virtue which she calls ‘constancy', a virtue about which I shall say more in a later chapter. In some ways constancy plays a role in Jane Austen analogous to that of phronesis in Aristotle; it is a virtue the possession of which is a prerequisite for the possession of other virtues. The second is the fact that what Aristotle treats as the virtue of agreeableness (a virtue for which he says there is no name) she treats as only the simulacrum of a genuine virtue—the genuine virtue in question is the one she calls amiability. For the man who practices agreeableness does so from considerations of honor and expediency, according to Aristotle; whereas Jane Austen thought it possible and necessary for the possessor of that virtue to have a certain real affection for people as such. (It matters here that Jane Austen is a Christian.) Remember that Aristotle himself had treated military courage as a simulacrum of true courage. Thus we find here yet another type of disagreement over the virtues; namely, one as to which human qualities are genuine virtues and which mere simulacra.

In Benjamin Franklin's list we find almost all the types of difference from at least one of the catalogues we have considered and one more. Franklin includes virtues which are new to our consideration such as cleanliness, silence and industry; he clearly considers the drive to acquire itself a part of virtue, whereas for most ancient Greeks this is the vice of pleonexia, he treats some virtues which earlier ages had considered minor as major; but he also redefines some familiar virtues. In the list of thirteen virtues which Franklin compiled as part of his system of private moral accounting, he elucidates each virtue by citing a maxim obedience to which is the virtue in question. In the case of chastity the maxim is ‘Rarely use venery but for health or offspring—never to dullness, weakness or the injury of your own or another's peace or reputation'. This is clearly not what earlier writers had meant by ‘chastity'.

We have therefore accumulated a startling number of differences and incompatibilities in the five stated and implied accounts of the virtues. So the question which I raised at the outset becomes more urgent. If different writers in different times and places, but all within the history of Western culture, include such different sets and types of items in their lists, what grounds have we for supposing that they do indeed aspire to list items of one and the same kind, that there is any shared concept at all? A second kind of consideration reinforces the presumption of a negative answer to this question. It is not just that each of these five writers lists different and differing kinds of items; it is also that each of these lists embodies, is the expression of a different theory about what a virtue is. In the Homeric poems a virtue is a quality the manifestation of which enables someone to do exactly what their well-defined social role requires. The primary role is that of the warrior king and that Homer lists those virtues which he does becomes intelligible at once when we recognize that the key virtues therefore must be those which enable a man to excel in combat and in the games. It follows that we cannot identify the Homeric virtues until we have first identified the key social roles in Homeric society and the requirements of each of them. The concept of what anyone filling such-and-such a role ought to do is prior to the concept of a virtue; the latter concept has application only via the former.

On Aristotle's account matters are very different. Even though some virtues are available only to certain types of people, nonetheless virtues attach not to men as inhabiting social roles, but to man as such. It is the telos of man as a species which determines what human qualities are virtues. We need to remember however that although Aristotle treats the acquisition and exercise of the virtues as means to an end, the relationship of means to end is internal and not external. I call a means internal to a given end when the end cannot be adequately characterized independently of a characterization of the means. So it is with the virtues and the telos which is the good life for man on Aristotle's account. The exercise of the virtues is itself a crucial component of the good life for man. This distinction between internal and external means to an end is not drawn by Aristotle himself in the Nicomachean Ethics, as I noticed earlier, but it is an essential distinction to be drawn if we are to understand what Aristotle intended. The distinction is drawn explicitly by Aquinas in the course of his defence of St. Augustine's definition of a virtue, and it is clear that Aquinas understood that in drawing it he was maintaining an Aristotelian point of view.

The New Testament's account of the virtues, even if it differs as much as it does in content from Aristotle's—Aristotle would certainly not have admired Jesus Christ and he would have been horrified by St. Paul—does have the same logical and conceptual structure as Aristotle's account. A virtue is, as with Aristotle, a quality the exercise of which leads to the achievement of the human telos. The good for man is of course a supernatural and not only a natural good, but supernature redeems and completes nature. Moreover the relationship of virtues as means to the end which is human incorporation in the divine kingdom of the age to come is internal and not external, just as it is in Aristotle. It is of course this parallelism which allows Aquinas to synthesize Aristotle and the New Testament. A key feature of this parallelism is the way in which the concept of the good life for man is prior to the concept of a virtue in just the way in which on the Homeric account the concept of a social role was prior. Once again it is the way in which the former concept is applied which determines how the latter is to be applied. In both cases the concept of a virtue is a secondary concept.

The intent of Jane Austen's theory of the virtues is of another kind. C.S. Lewis has rightly emphasized how profoundly Christian her moral vision is and Gilbert Ryle has equally rightly emphasized her inheritance from Shaftesbury and from Aristotle. In fact her views combine elements from Homer as well, since she is concerned with social roles in a way that neither the New Testament nor Aristotle are. She is therefore important for the way in which she finds it possible to combine what are at first sight disparate theorectical [stet] accounts of the virtues. But for the moment any attempt to assess the significance of Jane Austen's systhesis [stet] must be delayed. Instead we must notice the quite different style of theory articulated in Benjamin Franklin's account of the virtues.

Franklin's account, like Aristotle's, is teleological; but unlike Aristotle's, it is utilitarian. According to Franklin in his Autobiography the virtues are means to an end, but he envisages the means-ends relationship as external rather than internal. The end to which the cultivation of the virtues ministers is happiness, but happiness understood as success, prosperity in Philadelphia and ultimately in heaven. The virtues are to be useful and Franklin's account continuously stresses utility as a criterion in individual cases: ‘Make no expence but to do good to others or yourself; i.e waste nothing', ‘Speak not but what may benefit others or yourself. Avoid trifling conversation' and, as we have already seen, ‘Rarely use venery but for health or offspring. . .'. When Franklin was in Paris he was horrified by Parisian architecture: ‘Marble, porcelain and gilt are squandered without utility.'

We thus have at least three very different conceptions of a virtue to confront: a virtue is a quality which enables an individual to discharge his or her social role (Homer); a virtue is a quality which enables an individual to move towards the achievement of the specifically human tefos, whether natural or supernatural (Aristotle, the New Testament and Aquinas); a virtue is a quality which has utility in achieving earthly and heavenly success (Franklin). Are we to take these as three different rival accounts of the same thing? Or are they instead accounts of three different things? Perhaps the moral structures in archaic Greece, in fourth-century Greece, and in eighteenth-century Pennsylvania were so different from each other that we should treat them as embodying quite different concepts, whose difference is initially disguised from us by the historical accident of an inherited vocabulary which misleads us by linguistic resemblance long after conceptual identity and similarity have failed. Our initial question has come back to us with redoubled force.

Yet although I have dwelt upon the prima facie case for holding that the differences and incompatibilities between different accounts at least suggest that there is no single, central, core conception of the virtues which might make a claim for universal allegiance, I ought also to point out that each of the five moral accounts which I have sketched so summarily does embody just such a claim. It is indeed just this feature of those accounts that makes them of more than sociological or antiquarian interest. Every one of these accounts claims not only theoretical, but also an institutional hegemony. For Odysseus the Cyclopes stand condemned because they lack agriculture, an agora and themis. For Aristotle the barbarians stand condemned because they lack the polis and are therefore incapable of politics. For New Testament Christians there is no salvation outside the apostolic church. And we know that Benjamin Franklin found the virtues more at home in Philadelphia than in Paris and that for Jane Austen the touchstone of the virtues is a certain kind of marriage and indeed a certain kind of naval officer (that is, a certain kind of English naval officer).

The question can therefore now be posed directly: are we or are we not able to disentangle from these rival and various claims a unitary core concept of the virtues of which we can give a more compelling account than any of the other accounts so far? I am going to argue that we can in fact discover such a core concept and that it turns out to provide the tradition of which I have written the history with its conceptual unity. It will indeed enable us to distinguish in a clear way those beliefs about the virtues which genuinely belong to the tradition from those which do not. Unsurprisingly perhaps it is a complex concept, different parts of which derive from different stages in the development of the tradition. Thus the concept itself in some sense embodies the history of which it is the outcome.

One of the features of the concept of a virtue which has emerged with some clarity from the argument so far is that it always requires for its application the acceptance for some prior account of certain features of social and moral life in terms of which it has to be defined and explained. So in the Homeric account the concept of a virtue is secondary to that of a social role, in Aristotle's account it is secondary to that of the good life for man conceived as the telos of human action and in Franklin's much later account it is secondary to that of utility. What is it in the account which I am about to give which provides in a similar way the necessary background against which the concept of a virtue has to be made intelligible? It is in answering this question that the complex, historical, multi-layered character of the core concept of virtue becomes clear. For there are no less than three stages in the logical development of the concept which have to be identified in order, if the core conception of a virtue is to be understood, and each of these stages has its own conceptual background. The first stage requires a background account of what I shall call a practice, the second an account of what I have already characterized as the narrative order of a single human life and the third an account a good deal fuller than I have given up to now of what constitutes a moral tradition. Each later stage presupposes the earlier, but not vice versa. Each earlier stage is both modified by and reinterpreted in the light of, but also provides an essential constituent of each later stage. The progress in the development of the concept is closely related to, although it does not recapitulate in any straightforward way, the history of the tradition of which it forms the core.

In the Homeric account of the virtues—and in heroic societies more generally—the exercise of a virtue exhibits qualities which are required for sustaining a social role and for exhibiting excellence in some well-marked area of social practice: to excel is to excel at war or in the games, as Achilles does, in sustaining a household, as Penelope does, in giving counsel in the assembly, as Nestor does, in the telling of a tale, as Homer himself does. When Aristotle speaks of excellence in human activity, he sometimes though not always, refers to some well-defined type of human practice: flute-playing, or war, or geometry. I am going to suggest that this notion of a particular type of practice as providing the arena in which the virtues are exhibited and in terms of which they are to receive their primary, if incomplete, definition is crucial to the whole enterprise of identifying a core concept of the virtues. I hasten to add caveats however.

The first is to point out that my argument will not in any way imply that virtues are only exercised in the course of what I am calling practices. The second is to warn that I shall be using the word ‘practice' in a specially defined way which does not completely agree with current ordinary usage, including my own previous use of that word. What am I going to mean by it?

By a ‘practice' I am going to mean any coherent and complex form of socially established cooperative human activity through which goods internal to that form of activity are realized in the course of trying to achieve those standards of excellence which are appropriate to, and partially definitive of, that form of activity, with the result that human powers to achieve excellence, and human conceptions of the ends and goods involved, are systematically extended. Tic-tac-toe is not an example of a practice in this sense, nor is throwing a football with skill; but the game of football is, and so is chess. Bricklaying is not a practice; architecture is. Planting turnips is not a practice; farming is. So are the enquiries of physics, chemistry and biology, and so is the work of the historian, and so are painting and music. In the ancient and medieval worlds the creation and sustaining of human communities—of households, cities, nations—is generally taken to be a practice in the sense in which I have defined it. Thus the range of practices is wide: arts, sciences, games, politics in the Aristotelian sense, the making and sustaining of family life, all fall under the concept. But the question of the precise range of practices is not at this stage of the first importance. Instead let me explain some of the key terms involved in my definition, beginning with the notion of goods internal to a practice.

Consider the example of a highly intelligent seven-year-old child whom I wish to teach to play chess, although the child has no particular desire to learn the game. The child does however have a very strong desire for candy and little chance of obtaining it. I therefore tell the child that if the child will play chess with me once a week I will give the child 50 cents worth of candy; moreover I tell the child that I will always play in such a way that it will be difficult, but not impossible, for the child to win and that, if the child wins, the child will receive an extra 50 cents worth of candy. Thus motivated the child plays and plays to win. Notice however that, so long as it is the candy alone which provides the child with a good reason for playing chess, the child has no reason not to cheat and every reason to cheat, provided he or she can do so successfully, But, so we may hope, there will come a time when the child will find in those goods specific to chess, in the achievement of a certain highly particular kind of analytical skill, strategic imagination and competitive intensity, a new set of reasons, reasons now not just for winning on a particular occasion, but for trying to excel in whatever way the game of chess demands. Now if the child cheats, he or she will be defeating not me, but himself or herself.

There are thus two kinds of good possibly to be gained by playing chess. On the one hand there are those goods externally and contingently attached to chess-playing and to other practices by the accidents of social circumstance—in the case of the imaginary child candy, in the case of real adults such goods as prestige, status and money. There are always alternative ways for achieving such goods, and their achievement is never to be had only by engaging in some particular kind of practice. On the other hand there are the goods internal to the practice of chess which cannot be had in any way but by playing chess or some other game of that specific kind. We call them internal for two reasons: first, as I have already suggested, because we can only specify them in terms of chess or some other game of that specific kind and by means of examples from such games (otherwise the meagerness of our vocabulary for speaking of such goods forces us into such devices as my own resort to writing of); a certain highly particular kind of); and secondly because they can only be identified and recognized by the experience of participating in the practice in question. Those who lack the relevant experience are incompetent thereby as judges of internal goods.

This is clearly the case with all the major examples of practices: consider for example—even if briefly and inadequately—the practice of portrait painting as it developed in Western Europe from the late middle ages to the eighteenth century. The successful portrait painter is able to achieve many goods which are in the sense just defined external to the practice of portrait painting—fame, wealth, social status, even a measure of power and influence at courts upon occasion. But those external goods are not to be confused with the goods which are internal to the practice. The internal goods are those which result from an extended attempt to show how Wittgenstein's dictum ‘The human body is the best picture of the human soul' (Investigations, p. 178e) might be made to become true by teaching us ‘to regard . . . the picture on our wall as the object itself (the men, landscape and so on) depicted there' (p. 205e) in a quite new way. What is misleading about Wittgenstein's dictum as it stands is its neglect of the truth in George Orwell's thesis ‘At fifty everyone has the face he deserves'. What painters from Giotto to Rembrandt learnt to show was how the face at any age may be revealed as the face that the subject of a portrait deserves.

Originally in medieval paintings of the saints the face was an icon; the question of a resemblance between the depicted face of Christ or St. Peter and the face that Jesus or Peter actually possessed at some particular age did not even arise. The antithesis to this iconography was the relative naturalism of certain fifteenth-century Flemish and German painting. The heavy eyelids, the coifed hair, the lines around the mouth undeniably represent some particular woman, either actual or envisaged. Resemblance has usurped the iconic relationship. But with Rembrandt there is, so to speak, synthesis: the naturalistic portrait is now rendered as an icon, but an icon of a new and hitherto inconceivable kind. Similarly in a very different kind of sequence mythological faces in a certain kind of seventeenth-century French painting become aristocratic faces in the eighteenth century. Within each of these sequences at least two different kinds of good internal to the painting of human faces and bodies are achieved.

There is first of all the excellence of the products, both the excellence in performance by the painters and that of each portrait itself. This excellence—the very verb ‘excel' suggests it—has to be understood historically. The sequences of development find their point and purpose in a progress towards and beyond a variety of types and modes of excellence. There are of course sequences of decline as well as of progress, and progress is rarely to be understood as straightforwardly linear. But it is in participation in the attempts to sustain progress and to respond creatively to problems that the second kind of good internal to the practices of portrait painting is to be found. For what the artist discovers within the pursuit of excellence in portrait painting—and what is true of portrait painting is true of the practice of the fine arts in general—is the good of a certain kind of life. That life may not constitute the whole of life for someone who is a painter by a very long way or it may at least for a period, Gauguin-like, absorb him or her at the expense of almost everything else. But it is the painter's living out of a greater or lesser part of his or her life as a painter that is the second kind of good internal to painting. And judgment upon these goods requires at the very least the kind of competence that is only to be acquired either as a painter or as someone willing to learn systematically what the portrait painter has to teach.

A practice involves standards of excellence and obedience to rules as well as the achievement of goods. To enter into a practice is to accept the authority of those standards and the inadequacy of my own performance as judged by them. It is to subject my own attitudes, choices, preferences and tastes to the standards which currently and partially define the practice. Practices of course, as I have just noticed, have a history: games, sciences and arts all have histories. Thus the standards are not themselves immune from criticism, but nonetheless we cannot be initiated into a practice without accepting the authority of the best standards realized so far. If, on starting to listen to music, I do not accept my own incapacity to judge correctly, I will never learn to hear, let alone to appreciate, Bartok's last quartets. If, on starting to play baseball, I do not accept that others know better than I when to throw a fast ball and when not, I will never learn to appreciate good pitching let alone to pitch. In the realm of practices the authority of both goods and standards operates in such a way as to rule out all subjectivist and emotivist analyses of judgment. De gustibus est disputandum.

We are now in a position to notice an important difference between what I have called internal and what I have called external goods. It is characteristic of what I have called external goods that when achieved they are always some individual's property and possession. Moreover characteristically they are such that the more someone has of them, the less there is for other people. This is sometimes necessarily the case, as with power and fame, and sometimes the case by reason of contingent circumstance as with money. External goods are therefore characteristically objects of competition in which there must be losers as well as winners. Internal goods are indeed the outcome of competition to excel, but it is characteristic of them that their achievement is a good for the whole community who participate in the practice. So when Turner transformed the seascape in painting or W.G. Grace advanced the art of batting in cricket in a quite new way their achievement enriched the whole relevant community.

But what does all or any of this have to do with the concept of the virtues? It turns out that we are now in a position to formulate a first, even if partial and tentative definition of a virtue: A virtue is an acquired human quality the possession and exercise of which tends to enable us to achieve those goods which are internal to practices and the lack of which effectively prevents us from achieving any such goods. Later this definition will need amplification and amendment. But as a first approximation to an adequate definition it already illuminates the place of the virtues in human life. For it is not difficult to show for a whole range of key virtues that without them the goods internal to practices are barred to us, but not just barred to us generally, barred in a very particular way.

It belongs to the concept of a practice as I have outlined it—and as we are all familiar with it already in our actual lives, whether we are painters or physicists or quarterbacks or indeed just lovers of good painting or first-rate experiments or a well-thrown pass—that its goods can only be achieved by subordinating ourselves within the practice in our relationship to other practitioners. We have to learn to recognize what is due to whom; we have to be prepared to take whatever self-endangering risks are demanded along the way; and we have to listen carefully to what we are told about our own inadequacies and to reply with the same carefulness for the facts. In other words we have to accept as necessary components of any practice with internal goods and standards of excellence the virtues of justice, courage and honesty. For not to accept these, to be willing to cheat as our imagined child was willing to cheat in his or her early days at chess, so far bars us from achieving the standards of excellence or the goods internal to the practice that it renders the practice pointless except as a device for achieving external goods.

We can put the same point in another way. Every practice requires a certain kind of relationship between those who participate in it. Now the virtues are those goods by reference to which, whether we like it or not, we define our relationships to those other people with whom we share the kind of purposes and standards which inform practices. Consider an example of how reference to the virtues has to be made in certain kinds of human relationship.

A, B, C, and D are friends in that sense of friendship which Aristotle takes to be primary: they share in the pursuit of certain goods. In my terms they share in a practice. D dies in obscure circumstances, A discovers how D died and tells the truth about it to B while lying to C. C discovers the lie. What A cannot then intelligibly claim is that he stands in the same relationship of friendship to both B and C. By telling the truth to one and lying to the other he has partially defined a difference in the relationship. Of course it is open to A to explain this difference in a number of ways; perhaps he was trying to spare C pain or perhaps he is simply cheating C. But some difference in the relationship now exists as a result of the lie. For their allegiance to each other in the pursuit of common goods has been put in question. Just as, so long as we share the standards and purposes characteristic of practices, we define our relationship to each other, whether we acknowledge it or not, by reference to standards of truthfulness and trust, so we define them too by reference to standards of justice and of courage. If A, a professor, gives B and C the grades that their papers deserve, but grades D because he is attracted by D's blue eyes or is repelled by D's dandruff, he has defined his relationship to D differently from his relationship to the other members of the class, whether he wishes it or not. Justice requires that we treat others in respect of merit or desert according to uniform and impersonal standards; to depart from the standards of justice in some particular instance defines our relationship with the relevant person as in some way special or distinctive.

The case with courage is a little different. We hold courage to be a virtue because the care and concern for individuals, communities and causes which is so crucial to so much in practices requires the existence of such a virtue. If someone says that he cares for some individual, community or cause, but is unwilling to risk harm or danger on his, her or its own behalf, he puts in question the genuineness of his care and concern. Courage, the capacity to risk harm or danger to oneself, has its role in human life because of this connection with care and concern. This is not to say that a man cannot genuinely care and also be a coward. It is in part to say that a man who genuinely cares and has not the capacity for risking harm or danger has to define himself, both to himself and to others, as a coward.

I take it then that from the standpoint of those types of relationship without which practices cannot be sustained truthfulness, justice and courage—and perhaps some others—are genuine excellences are virtues in the light of which we have to characterize ourselves and others, whatever our private moral standpoint or our society's particular codes may be. For this recognition that we cannot escape the definition of our relationships in terms of such goods is perfectly compatible with the acknowledgment that different societies have and have had different codes of truthfulness, justice and courage. Lutheran pietists brought up their children to believe that one ought to tell the truth to everybody at all times, whatever the circumstances or consequences, and Kant was one of their children. Traditional Bantu parents brought up their children not to tell the truth to unknown strangers, since they believed that this could render the family vulnerable to witchcraft. In our culture many of us have been brought up not to tell the truth to elderly great-aunts who invite us to admire their new hats. But each of these codes embodies an acknowledgment of the virtue of truthfulness. So it is also with varying codes of justice and of courage.

Practices then might flourish in societies with very different codes; what they could not do is flourish in societies in which the virtues were not valued, although institutions and technical skills serving unified purposes might well continue to flourish. (I shall have more to say about the contrast between institutions and technical skills mobilized for a unified end, on the one hand, and practices on the other, in a moment.) For the kind of cooperation, the kind of recognition of authority and of achievement, the kind of respect for standards and the kind of risk-taking which are characteristically involved in practices demand for example fairness in judging oneself and others—the kind of fairness absent in my example of the professor, a ruthless truthfulness without which fairness cannot find application—the kind of truthfulness absent in my example of A, B, C, and D—and willingness to trust the judgments of those whose achievement in the practice give them an authority to judge which presupposes fairness and truthfulness in those judgments, and from time to time the taking of self-endangering and even achievement-endangering risks. It is no part of my thesis that great violinists cannot be vicious or great chess-players mean-spirited. Where the virtues are required, the vices also may flourish. It is just that the vicious and mean-spirited necessarily rely on the virtues of others for the practices in which they engage to flourish and also deny themselves the experience of achieving those internal goods which may reward even not very good chess-players and violinists.

To situate the virtues any further within practices it is necessary now to clarify a little further the nature of a practice by drawing two important contrasts. The discussion so far I hope makes it clear that a practice, in the sense intended, is never just a set of technical skills, even when directed towards some unified purpose and even if the exercise of those skills can on occasion be valued or enjoyed for their own sake. What is distinctive in a practice is in part the way in which conceptions of the relevant goods and ends which the technical skills serve—and every practice does require the exercise of technical skills—are transformed and enriched by these extensions of human powers and by that regard for its own internal goods which are partially definitive of each particular practice or type of practice. Practices never have a goal or goals fixed for all time—painting has no such goal nor has physics—but the goals themselves are transmuted by the history of the activity. It therefore turns out not to be accidental that every practice has its own history and a history which is more and other than that of the improvement of the relevant technical skills. This historical dimension is crucial in relation to the virtues.

To enter into a practice is to enter into a relationship not only with its contemporary practitioners, but also with those who have preceded us in the practice, particularly those whose achievements extended the reach of the practice to its present point. It is thus the achievement, and a fortiori the authority, of a tradition which I then confront and from which I have to learn. And for this learning and the relationship to the past which it embodies the virtues of justice, courage and truthfulness are prerequisite in precisely the same way and for precisely the same reasons as they are in sustaining present relationships within practices.

It is not only of course with sets of technical skills that practices ought to be contrasted. Practices must not be confused with institutions. Chess, physics and medicine are practices; chess clubs, laboratories, universities and hospitals are institutions. Institutions are characteristically and necessarily concerned with what I have called external goods. They are involved in acquiring money and other material goods; they are structured in terms of power and status, and they distribute money, power and status as rewards. Nor could they do otherwise if they are to sustain not only themselves, but also the practices of which they are the bearers. For no practices can survive for any length of time unsustained by institutions. Indeed so intimate is the relationship of practices to institutions—and consequently of the goods external to the goods internal to the practices in question—that institutions and practices characteristically form, a single causal order in which the ideals and the creativity of the practice are always vulnerable to the acquisitiveness of the institution, in which the cooperative care for common goods of the practice is always vulnerable to the competitiveness of the institution. In this context the essential function of the virtues is clear. Without them, without justice, courage and truthfulness, practices could not resist the corrupting power of institutions.

Yet if institutions do have corrupting power, the making and sustaining of forms of human community—and therefore of institutions—itself has all the characteristics of a practice, and moreover of a practice which stands in a peculiarly close relationship to the exercise of the virtues in two important ways. The exercise of the virtues is itself apt to require a highly determinate attitude to social and political issues; and it is always within some particular community with its own specific institutional forms that we learn or fail to learn to exercise the virtues. There is of course a crucial difference between the way in which the relationship between moral character and political community is envisaged from the standpoint of liberal individualist modernity and the way in which that relationship was envisaged from the standpoint of the type of ancient and medieval tradition of the virtues which I have sketched. For liberal individualism a community is simply an arena in which individuals each pursue their own self-chosen conception of the good life, and political institutions exist to provide that degree of order which makes such self-determined activity possible. Government and law are, or ought to be, neutral between rival conceptions of the good life for man, and hence, although it is the task of government to promote law-abidingness, it is on the liberal view no part of the legitimate function of government to inculcate any one moral outlook.

By contrast, on the particular ancient and medieval view which I have sketched political community not only requires the exercise of the virtues for its own sustenance, but it is one of the tasks of parental authority to make children grow up so as to be virtuous adults. The classical statement of this analogy is by Socrates in the Crito. It does not of course follow from an acceptance of the Socratic view of political community and political authority that we ought to assign to the modern state the moral function which Socrates assigned to the city and its laws. Indeed the power of the liberal individualist standpoint partly derives from the evident fact that the modern state is indeed totally unfitted to act as moral educator of any community. But the history of how the modern state emerged is of course itself a moral history. If my account of the complex relationship of virtues to practices and to institutions is correct, it follows that we shall be unable to write a true history of practices and institutions unless that history is also one of the virtues and vices. For the ability of a practice to retain its integrity will depend on the way in which the virtues can be and are exercised in sustaining the institutional forms which are the social bearers of the practice. The integrity of a practice causally requires the exercise of the virtues by at least some of the individuals who embody it in their activities; and conversely the corruption of institutions is always in part at least an effect of the vices.

The virtues are of course themselves in turn fostered by certain types of social institution and endangered by others. Thomas Jefferson thought that only in a society of small farmers could the virtues flourish; and Adam Ferguson with a good deal more sophistication saw the institutions of modern commercial society as endangering at least some traditional virtues. It is Ferguson's type of sociology which is the empirical counterpart of the conceptual account of the virtues which I have given, a sociology which aspires to lay bare the empirical, causal connection between virtues, practices and institutions. For this kind of conceptual account has strong empirical implications; it provides an explanatory scheme which can be tested in particular cases. Moreover my thesis has empirical content in another way; it does entail that without the virtues there could be a recognition only of what I have called external goods and not at all of internal goods in the context of practices. And in any society which recognized only external goods competitiveness would be the dominant and even exclusive feature. We have a brilliant portrait of such a society in Hobbes's account of the state of nature; and Professor Turnbull's report of the fate of the Ik suggests that social reality does in the most horrifying way confirm both my thesis and Hobbes's.

Virtues then stand in a different relationship to external and to internal goods. The possession of the virtues—and not only of their semblance and simulacra—is necessary to achieve the latter; yet the possession of the virtues may perfectly well hinder us in achieving external goods. I need to emphasize at this point that external goods genuinely are goods. Not only are they characteristic objects of human desire, whose allocation is what gives point to the virtues of justice and of generosity, but no one can despise them altogether without a certain hypocrisy. Yet notoriously the cultivation of truthfulness, justice and courage will often, the world being what it contingently is, bar us from being rich or famous or powerful. Thus although we may hope that we can not only achieve the standards of excellence and the internal goods of certain practices by possessing the virtues and become rich, famous and powerful, the virtues are always a potential stumbling block to this comfortable ambition. We should therefore expect that, if in a particular society the pursuit of external goods were to become dominant, the concept of the virtues might suffer first attrition and then perhaps something near total effacement, although simulacra might abound.

The time has come to ask the question of how far this partial account of a core conception of the virtues—and I need to emphasize that all that I have offered so far is the first stage of such an account—is faithful to the tradition which I delineated. How far, for example, and in what ways is it Aristotelian? It is—happily—not Aristotelian in two ways in which a good deal of the rest of the tradition also dissents from Aristotle. First, although this account of the virtues is teleological, it does not require any allegiance to Aristotle's metaphysical biology. And secondly, just because of the multiplicity of human practices and the consequent multiplicity of goods in the pursuit of which the virtues may be exercised—goods which will often be contingently incompatible and which will therefore make rival claims upon our allegiance—conflict will not spring solely from flaws in individual character. But it was just on these two matters that Aristotle's account of the virtues seemed most vulnerable; hence if it turns out to be the case that this socially teleological account can support Aristotle's general account of the virtues as well as does his own biologically teleological account, these differences from Aristotle himself may well be regarded as strengthening rather than weakening the case for a generally Aristotelian standpoint.

There are at least three ways in which the account that I have given is clearly Aristotelian. First it requires for its completion a cogent elaboration of just those distinctions and concepts which Aristotle's account requires: voluntariness, the distinction between the intellectual virtues and the virtues of character, the relationship of both to natural abilities and to the passions and the structure of practical reasoning. On every one of these topics something very like Aristotle's view has to be defended, if my own account is to be plausible.

Secondly my account can accommodate an Aristotelian view of pleasure and enjoyment, whereas it is interestingly irreconcilable with any utilitarian view and more particularly with Franklin's account of the virtues. We can approach these questions by considering how to reply to someone who, having considered my account of the differences between goods internal to and goods external to a practice enquired into which class, if either, does pleasure or enjoyment fall? The answer is, ‘Some types of pleasure into one, some into the other.'

Someone who achieves excellence in a practice, who plays chess or football well or who carries through an enquiry in physics or an experimental mode in painting with success, characteristically enjoys his achievement and his activity in achieving. So does someone who, although not breaking the limit of achievement, plays or thinks or acts in a way that leads towards such a breaking of limit. As Aristotle says, the enjoyment of the activity and the enjoyment of achievement are not the ends at which the agent aims, but the enjoyment supervenes upon the successful activity in such a way that the activity achieved and the activity enjoyed are one and the same state. Hence to aim at the one is to aim at the other; and hence also it is easy to confuse the pursuit of excellence with the pursuit of enjoyment in this specific sense. This particular confusion is harmless enough; what is not harmless is the confusion of enjoyment in this specific sense with other forms of pleasure.

For certain kinds of pleasure are of course external goods along with prestige, status, power and money. Not all pleasure is the enjoyment supervening upon achieved activity; some is the pleasure of psychological or physical states independent of all activity. Such states—for example that produced on a normal palate by the closely successive and thereby blended sensations of Colchester oyster, cayenne pepper and Veuve Cliquot—may be sought as external goods, as external rewards which may be purchased by money or received in virtue of prestige. Hence the pleasures are categorized neatly and appropriately by the classification into internal and external goods.

It is just this classification which can find no place within Franklin's account of the virtues which is framed entirely in terms of external relationships and external goods. Thus although by this stage of the argument it is possible to claim that my account does capture a conception of the virtues which is at the core of the particular ancient and medieval tradition which I have delineated, it is equally clear that there is more than one possible conception of the virtues and that Franklin's standpoint and indeed any utilitarian standpoint is such that to accept it will entail rejecting the tradition and vice versa.

One crucial point of incompatibility was noted long ago by D.H. Lawrence. When Franklin asserts, ‘Rarely use venery but for health or offspring . . .', Lawrence replies, ‘Never use venery.' It is of the character of a virtue that in order that it be effective in producing the internal goods which are the rewards of the virtues it should be exercised without regard to consequences. For it turns out to be the case that—and this is in part at least one more empirical factual claim—although the virtues are just those qualities which tend to lead to the achievement of a certain class of goods, nonetheless unless we practice them irrespective of whether in any particular set of contingent circumstances they will produce those goods or not, we cannot possess them at all. We cannot be genuinely courageous or truthful and be so only on occasion. Moreover, as we have seen, cultivation of the virtues always may and often does hinder the achievement of those external goods which are the mark of worldly success. The road to success in Philadelphia and the road to heaven may not coincide after all.

Furthermore we are now able to specify one crucial difficulty for any version of utilitarianism—in addition to those which I noticed earlier. Utilitarianism cannot accommodate the distinction between goods internal to and goods external to a practice. Not only is that distinction marked by none of the classical utilitarians—it cannot be found in Bentham's writings nor in those of either of the Mills or of Sidgwick—but internal goods and external goods are not commensurable with each other. Hence the notion of summing goods—and a fortiori in the light of what I have said about kinds of pleasure and enjoyment the notion of summing happiness—in terms of one single formula or conception of utility, whether it is Franklin's or Bentham's or Mill's, makes no sense. Nonetheless we ought to note that although this distinction is alien to J.S. Mill's thought, it is plausible and in no way patronizing to suppose that something like this is the distinction which he was trying to make in Utilitarianism when he distinguished between ‘higher' and ‘lower' pleasures. At the most we can say ‘something like this'; for J.S. Mill's upbringing had given him a limited view of human life and powers, had unfitted him, for example, for appreciating games just because of the way it had fitted him for appreciating philosophy. Nonetheless the notion that the pursuit of excellence in a way that extends human powers is at the heart of human life is instantly recognizable as at home in not only J.S. Mill's political and social thought, but also in his and Mrs. Taylor's life. Were I to choose human exemplars of certain of the virtues as I understand them, there would of course be many names to name, those of St. Benedict and St. Francis of Assisi and St. Theresa and those of Frederick Engels and Eleanor Marx and Leon Trotsky among them. But that of John Stuart Mill would have to be there as certainly as any other.

Thirdly my account is Aristotelian in that it links evaluation and explanation in a characteristically Aristotelian way. From an Aristotelian standpoint to identify certain actions as manifesting or failing to manifest a virtue or virtues is never only to evaluate; it is also to take the first step towards explaining why those actions rather than some others were performed. Hence for an Aristotelian quite as much as for a Platonist the fate of a city or an individual can be explained by citing the injustice of a tyrant or the courage of its defenders. Indeed without allusion to the place that justice and injustice, courage and cowardice play in human life very little will be genuinely explicable. It follows that many of the explanatory projects of the modern social sciences, a methodological canon of which is the separation of the facts—this conception of the ‘the facts' is the one which I delineated in Chapter 7—from all evaluation, are bound to fail. For the fact that someone was or failed to be courageous or just cannot be recognized as ‘a fact' by those who accept that methodological canon. The account of the virtues which I have given is completely at one with Aristotle's on this point. But now the question may be raised: your account may be in many respects Aristotelian, but is it not in some respects false? Consider the following important objection.

I have defined the virtues partly in terms of their place in practices. But surely, it may be suggested, some practices—that is, some coherent human activities which answer to the description of what I have called a practice—are evil. So in discussions by some moral philosophers of this type of account of the virtues it has been suggested that torture and sado-masochistic sexual activities might be examples of practices. But how can a disposition be a virtue if it is the kind of disposition which sustains practices and some practices issue in evil? My answer to this objection falls into two parts.

First I want to allow that there may be practices—in the sense in which I understand the concept—which simply are evil. I am far from convinced that there are, and I do not in fact believe that either torture or sadomasochistic sexuality answer to the description of a practice which my account of the virtues employs. But I do not want to rest my case on this lack of conviction, especially since it is plain that as a matter of contingent fact many types of practice may on particular occasions be productive of evil. For the range of practices includes the arts, the sciences and certain types of intellectual and athletic game. And it is at once obvious that any of these may under certain conditions be a source of evil: the desire to excel and to win can corrupt, a man may be so engrossed by his painting that he neglects his family, what was initially an honorable resort to war can issue in savage cruelty. But what follows from this?

It certainly is not the case that my account entails either that we ought to excuse or condone such evils or that whatever flows from a virtue is right. I do have to allow that courage sometimes sustains injustice, that loyalty has been known to strengthen a murderous aggressor and that generosity has sometimes weakened the capacity to do good. But to deny this would be to fly in the face of just those empirical facts which I invoked in criticizing Aquinas' account of the unity of the virtues. That the virtues need initially to be defined and explained with reference to the notion of a practice thus in no way entails approval of all practices in all circumstances. That the virtues—as the objection itself presupposed—are defined not in terms of good and right practices, but of practices, does not entail or imply that practices as actually carried through at particular times and places do not stand in need of moral criticism. And the resources for such criticism are not lacking. There is in the first place no inconsistency in appealing to the requirements of a virtue to criticize a practice. Justice may be initially defined as a disposition which in its particular way is necessary to sustain practices; it does not follow that in pursuing the requirements of a practice violations of justice are not to be condemned. Moreover I already pointed out in Chapter 12 that a morality of virtues requires as its counterpart a conception of moral law. Its requirements too have to be met by practices. But, it may be asked, does not all this imply that more needs to be said about the place of practices in some larger moral context? Does not this at least suggest that there is more to the core concept of a virtue than can be spelled out in terms of practices? I have after all emphasized that the scope of any virtue in human life extends beyond the practices in terms of which it is initially defined. What then is the place of the virtues in the larger arenas of human life?

I stressed earlier that any account of the virtues in terms of practices could only be a partial and first account. What is required to complement it? The most notable difference so far between my account and any account that could be called Aristotelian is that although I have in no way restricted the exercise of the virtues to the context of practices, it is in terms of practices that I have located their point and function. Whereas Aristotle locates that point and function in terms of the notion of a type of whole human life which can be called good. And it does seem that the question ‘What would a human being lack who lacked the virtues?' must be given a kind of answer which goes beyond anything which I have said so far. For such an individual would not merely fail in a variety of particular ways in respect of the kind of excellence which can be achieved through participation in practices and in respect of the kind of human relationship required to sustain such excellence. His own life viewed as a whole would perhaps be defective; it would not be the kind of life which someone would describe in trying to answer the question ‘What is the best kind of life for this kind of man or woman to live?' And that question cannot be answered without at least raising Aristotle's own question, ‘What is the good life for man?' Consider three ways in which human life informed only by the conception of the virtues sketched so far would be defective.

It would be pervaded, first of all, by too many conflicts and too much arbitrariness. I argued earlier that it is a merit of an account of the virtues in terms of a multiplicity of goods that it allows for the possibility of tragic conflict in a way in which Aristotle's does not. But it may also produce even in the life of someone who is virtuous and disciplined too many occasions when one allegiance points in one direction, another in another. The claims of one practice may be incompatible with another in such a way that one may find oneself oscillating in an arbitrary way, rather than making rational choices. So it seems to have been with T.E. Lawrence. Commitment to sustaining the kind of community in which the virtues can flourish may be incompatible with the devotion which a particular practice—of the arts, for example—requires. So there may be tensions between the claims of family life and those of the arts—the problem that Gauguin solved or failed to solve by fleeing to Polynesia, or between the claims of politics and those of the arts—the problem that Lenin solved or failed to solve by refusing to listen to Beethoven.

If the life of the virtues is continuously fractured by choices in which one allegiance entails the apparently arbitrary renunciation of another, it may seem that the goods internal to practices do after all derive their authority from our individual choices; for when different goods summon in different and in incompatible directions, ‘I' have to choose between their rival claims. The modern self with its criterion-less choices apparently reappears in the alien context of what was claimed to be an Aristotelian world. This accusation might be rebutted in part by returning to the question of why both goods and virtues do have authority in our lives and repeating what was said earlier in this chapter. But this reply would only be partly successful; the distinctively modern notion of choice would indeed have reappeared, even if with a more limited scope for its exercise than it has usually claimed.

Secondly without an overriding conception of the telos of a whole human life, conceived as a unity, our conception of certain individual virtues has to remain partial and incomplete. Consider two examples. Justice, on an Aristotelian view, is defined in terms of giving each person his or her due or desert. To deserve well is to have contributed in some substantial way to the achievement of those goods, the sharing of which and the common pursuit of which provide foundations for human community. But the goods internal to practices, including the goods internal to the practice of making and sustaining forms of community, need to be ordered and evaluated in some way if we are to assess relative desert. Thus any substantive application of an Aristotelian concept of justice requires an understanding of goods and of the good that goes beyond the multiplicity of goods which inform practices. As with justice, so also with patience. Patience is the virtue of waiting attentively without complaint, but not of waiting thus for anything at all. To treat patience as a virtue presupposes some adequate answer to the question: waiting for what? Within the context of practices a partial, although for many purposes adequate, answer can be given: the patience of a craftsman with refractory material, of a teacher with a slow pupil, of a politician in negotiations, are all species of patience. But what if the material is just too refractory, the pupil too slow, the negotiations too frustrating? Ought we always at a certain point just to give up in the interests of the practice itself? The medieval exponents of the virtue of patience claimed that there are certain types of situation in which the virtue of patience requires that I do not ever give up on some person or task, situations in which, as they would have put it, I am required to embody in my attitude to that person or task something of the patient attitude of God towards his creation. But this could only be so if patience served some overriding good, some telos which warranted putting other goods in a subordinate place. Thus it turns out that the content of the virtue of patience depends upon how we order various goods in a hierarchy and a fortiori on whether we are able rationally so to order these particular goods.

I have suggested so far that unless there is a telos which transcends the limited goods of practices by constituting the good of a whole human life, the good of a human life conceived as a unity, it will both be the case that a certain subversive arbitrariness will invade the moral life and that we shall be unable to specify the context of certain virtues adequately. These two considerations are reinforced by a third: that there is at least one virtue recognized by the tradition which cannot be specified at all except with reference to the wholeness of a human life—the virtue of integrity or constancy. ‘Purity of heart,' said Kierkegaard ‘is to will one thing.' This notion of singleness of purpose in a whole life can have no application unless that of a whole life does.

It is clear therefore that my preliminary account of the virtues in terms of practices captures much, but very far from all, of what the Aristotelian tradition taught about the virtues. It is also clear that to give an account that is at once more fully adequate to the tradition and rationally defensible, it is necessary to raise a question to which the Aristotelian tradition presupposed an answer, an answer so widely shared in the pre-modern world that it never had to be formulated explicitly in any detailed way This question is: is it rationally justifiable to conceive of each human life as a unity, so that we may try to specify each such life as having its good and so that we may understand the virtues as having their function in enabling an individual to make of his or her life one kind of unity rather than another?

CHAPTER 15

THE VIRTUES, THE UNITY OF A HUMAN LIFE AND THE CONCEPT OF A TRADITION

Any contemporary attempt to envisage each human life as a whole, as a unity, whose character provides the virtues with an adequate telos encounters two different kinds of obstacle, one social and one philosophical. The social obstacles derive from the way in which modernity partitions each human life into a variety of segments, each with its own norms and modes of behavior. So work is divided from leisure, private life from public, the corporate from the personal. So both childhood and old age have been wrenched away from the rest of human life and made over into distinct realms. And all these separations have been achieved so that it is the distinctiveness of each and not the unity of the life of the individual who passes through those parts in terms of which we are taught to think and to feel.

The philosophical obstacles derive from two distinct tendencies, one chiefly, though not only, domesticated in analytical philosophy and one at home in both sociological theory and in existentialism. The former is the tendency to think atomistically about human action and to analyze complex actions and transactions in terms of simple components. Hence the recurrence in more than one context of the notion of ‘a basic action'. That particular actions derive their character as parts of larger wholes is a point of view alien to our dominant ways of thinking and yet one which it is necessary at least to consider if we are to begin to understand how a life may be more than a sequence of individual actions and episodes.

Equally the unity of a human life becomes invisible to us when a sharp separation is made either between the individual and the roles that he or she plays—a separation characteristic not only of Sartre's existentialism, but also of the sociological theory of Ralf Dahrendorf—or between the different role and quasi-role-enactments of an individual life so that life comes to appear as nothing but a series of unconnected episodes—a liquidation of the self characteristic, as I noticed earlier, of Goffman's sociological theory. I already also suggested in Chapter 3 that both the Sartrian and the Goffmanesque conceptions of selfhood are highly characteristic of the modes of thought and practice of modernity. It is perhaps therefore unsurprising to realize that the self as thus conceived cannot be envisaged as a bearer of the Aristotelian virtues.

For a self separated from its roles in the Sartrian mode loses that arena of social relationships in which the Aristotelian virtues function if they function at all. The patterns of a virtuous life would fall under those condemnations of conventionality which Sartre put into the mouth of Antoine Roquentin in La Nausee and which he uttered in his own person in L'Etre et le néant. Indeed the self's refusal of the inauthenticity of conventionalized social relationships becomes what integrity is diminished into in Sartre's account.

At the same time the liquidation of the self into a set of demarcated areas of role-playing allows no scope for the exercise of dispositions which could genuinely be accounted virtues in any sense remotely Aristotelian. For a virtue is not a disposition that makes for success only in some one particular type of situation. What are spoken of as the virtues of a good committee man or of a good administrator or of a gambler or a pool hustler are professional skills professionally deployed in those situations where they can be effective, not virtues. Someone who genuinely possesses a virtue can be expected to manifest it in very different types of situation, many of them situations where the practice of a virtue cannot be expected to be effective in the way that we expect a professional skill to be. Hector exhibited one and the same courage in his parting from Andromache and on the battlefield with Achilles; Eleanor Marx exhibited one and the same compassion in her relationship with her father, in her work with trade unionists and in her entanglement with Aveling. And the unity of a virtue in someone's life is intelligible only as a characteristic of a unitary life, a life that can be conceived and evaluated as a whole. Hence just as in the discussion of the changes in and fragmentation of morality which accompanied the rise of modernity in the earlier parts of this book, each stage in the emergence of the characteristically modern views of the moral judgment was accompanied by a corresponding stage in the emergence of the characteristically modern conceptions of selfhood; so now, in defining the particular pre-modern concept of the virtues with which I have been preoccupied, it has become necessary to say something of the concomitant concept of selfhood, a concept of a self whose unity resides in the unity of a narrative which links birth to life to death as narrative beginning to middle to end.

Such a conception of the self is perhaps less unfamiliar than it may appear at first sight. Just because it has played a key part in the cultures which are historically predecessors of our own, it would not be surprising if it turned out to be still an unacknowledged presence in many of our ways of thinking and acting. Hence it is not inappropriate to begin by scrutinizing some of our most taken-for-granted, but clearly correct conceptual insights about human actions and selfhood in order to show how natural it is to think of the self in a narrative mode.

It is a conceptual commonplace, both for philosophers and for ordinary agents, that one and the same segment of human behavior may be correctly characterized in a number of different ways. To the question ‘What is he doing?' the answers may with equal truth and appropriateness be ‘Digging', ‘Gardening', ‘Taking exercise', ‘Preparing for winter' or ‘Pleasing his wife'. Some of these answers will characterize the agent's intentions, other unintended consequences of his actions, and of these unintended consequences some may be such that the agent is aware of them and others not. What is important to notice immediately is that any answer to the questions of how we are to understand or to explain a given segment of behavior will presuppose some prior answer to the question of how these different correct answers to the question ‘What is he doing?' are related to each other. For if someone's primary intention is to put the garden in order before the winter and it is only incidentally the case that in so doing he is taking exercise and pleasing his wife, we have one type of behavior to be explained; but if the agent's primary intention is to please his wife by taking exercise, we have quite another type of behavior to be explained and we will have to look in a different direction for understanding and explanation.

In the first place the episode has been situated in an annual cycle of domestic activity, and the behavior embodies an intention which presupposes a particular type of household-cum-garden setting with the peculiar narrative history of that setting in which this segment of behavior now becomes an episode. In the second instance the episode has been situated in the narrative history of a marriage, a very different, even if related, social setting. We cannot, that is to say, characterize behavior independently of intentions, and we cannot characterize intentions independently of the settings which make those intentions intelligible both to agents themselves and to others.

I use the word ‘setting' here as a relatively inclusive term. A social setting may be an institution, it may be what I have called a practice, or it may be a milieu of some other human kind. But it is central to the notion of a setting as I am going to understand it that a setting has a history, a history within which the histories of individual agents not only are, but have to be, situated, just because without the setting and its changes through time the history of the individual agent and his changes through time will be unintelligible. Of course one and the same piece of behavior may belong to more than one setting. There are at least two different ways in which this may be so.

In my earlier example the agent's activity may be part of the history both of the cycle of household activity and of his marriage, two histories which have happened to intersect. The household may have its own history stretching back through hundreds of years, as do the histories of some European farms, where the farm has had a life of its own, even though different families have in different periods inhabited it; and the marriage will certainly have its own history, a history which itself presupposes that a particular point has been reached in the history of the institution of marriage. If we are to relate some particular segment of behavior in any precise way to an agent's intentions and thus to the settings which that agent inhabits, we shall have to understand in a precise way how the variety of correct characterizations of the agent's behavior relate to each other first by identifying which characteristics refer us to an intention and which do not and then by classifying further the items in both categories.

Where intentions are concerned, we need to know which intention or intentions were primary, that is to say, of which it is the case that, had the agent intended otherwise, he would not have performed that action. Thus if we know that a man is gardening with the self-avowed purposes of healthful exercise and of pleasing his wife, we do not yet know how to understand what he is doing until we know the answer to such questions as whether he would continue gardening if he continued to believe that gardening was healthful exercise, but discovered that his gardening no longer pleased his wife, and whether he would continue gardening, if he ceased to believe that gardening was healthful exercise, but continued to believe that it pleased his wife, and whether he would continue gardening if he changed his beliefs on both points. That is to say, we need to know both what certain of his beliefs are and which of them are causally effective; and, that is to say, we need to know whether certain contrary-to-fact hypothetical statements are true or false. And until we know this, we shall not know how to characterize correctly what the agent is doing.

Consider another equally trivial example of a set of compatibly correct answers to the question ‘What is he doing?' ‘Writing a sentence'; ‘Finishing his book'; ‘Contributing to the debate on the theory of action'; ‘Trying to get tenure'. Here the intentions can be ordered in terms of the stretch of time to which reference is made. Each of the shorter-term intentions is, and can only be made, intelligible by reference to some longer-term intentions; and the characterization of the behavior in terms of the longer-term intentions can only be correct if some of the characterizations in terms of shorter-term intentions are also correct. Hence the behavior is only characterized adequately when we know what the longer and longest-term intentions invoked are and how the shorter-term intentions are related to the longer. Once again we are involved in writing a narrative history.

Intentions thus need to be ordered both causally and temporally and both orderings will make references to settings, references already made obliquely by such elementary terms as ‘gardening', ‘wife', ‘book', and ‘tenure'. Moreover the correct identification of the agent's beliefs will be an essential constituent of this task; failure at this point would mean failure in the whole enterprise. (The conclusion may seem obvious; but it already entails one important consequence. There is no such thing as ‘behavior', to be identified prior to and independently of intentions, beliefs and settings. Hence the project of a science of behavior takes on a mysterious and somewhat outré character. It is not that such a science is impossible; but there is nothing for it to be but a science of uninterpreted physical movement such as B.F. Skinner aspires to. It is no part of my task here to examine Skinner's problems; but it is worth noticing that it is not at all clear what a scientific experiment could be, if one were a Skinnerian; since the conception of an experiment is certainly one of intention- and belief-informed behavior. And what would be utterly doomed to failure would be the project of a science of, say, political behavior, detached from a study of intentions, beliefs and settings. It is perhaps worth noting that when the expression ‘the behavioral sciences' was given its first influential use in a Ford Foundation Report of 1953, the term ‘behavior' was defined so as to include what were called ‘such subjective behavior as attitudes, beliefs, expectations, motivations and aspirations' as well as ‘overt acts'. But what the Report's wording seems to imply is that it is cataloguing two distinct sets of items, available for independent study. If the argument so far is correct, then there is only one set of items.)

Consider what the argument so far implies about the interrelationships of the intentional, the social and the historical. We identify a particular action only by invoking two kinds of context, implicitly if not explicitly. We place the agent's intentions, I have suggested, in causal and temporal order with reference to their role in his or her history; and we also place them with reference to their role in the history of the setting or settings to which they belong. In doing this, in determining what causal efficacy the agent's intentions had in one or more directions, and how his short-term intentions succeeded or failed to be constitutive of long-term intentions, we ourselves write a further part of these histories. Narrative history of a certain kind turns out to be the basic and essential genre for the characterization of human actions. It is important to be clear now different the standpoint presupposed by the argument so far is from that of those analytical philosophers who have constructed accounts of human actions which make central the notion of ‘a' human action. A course of human events is then seen as a complex sequence of individual actions, and a natural question is: How do we individuate human actions? Now there are contexts in which such notions are at home. In the recipes of a cookery book for instance actions are individuated in just the way that some analytical philosophers have supposed to be possible of all actions. ‘Take six eggs. Then break them into a bowl. Add flour, salt, sugar, etc.' But the point about such sequences is that each element in them is intelligible as an action only as a-possible-element-in-a-sequence. Moreover even such a sequence requires a context to be intelligible. If in the middle of my lecture on Kant's ethics I suddenly broke six eggs into a bowl and added flour and sugar, proceeding all the while with my Kantian exegesis, I have not, simply in virtue of the fact that I was following a sequence prescribed by Fanny Farmer, performed an intelligible action.

To this it might be retorted that I certainly performed an action or a set of actions, if not an intelligible action. But to this I want to reply that the concept of an intelligible action is a more fundamental concept than that of an action as such. Unintelligible actions are failed candidates for the status of intelligible action; and to lump unintelligible actions and intelligible actions together in a single class of actions and then to characterize action in terms of what items of both sets have in common is to make the mistake of ignoring this. It is also to neglect the central importance of the concept of intelligibility.

The importance of the concept of intelligibility is closely related to the fact that the most basic distinction of all embedded in our discourse and our practice in this area is that between human beings and other beings. Human beings can be held to account for that of which they are the authors; other beings cannot. To identify an occurrence as an action is in the paradigmatic instances to identify it under a type of description which enables us to see that occurrence as flowing intelligibly from a human agent's intentions, motives, passions and purposes. It is therefore to understand an action as something for which someone is accountable, about which it is always appropriate to ask the agent for an intelligible account. When an occurrence is apparently the intended action of a human agent, but nonetheless we cannot so identify it, we are both intellectually and practically baffled. We do not know how to respond; we do not know how to explain; we do not even know how to characterize minimally as an intelligible action; our distinction between the humanly accountable and the merely natural seems to have broken down. And this kind of bafflement does indeed occur in a number of different kinds of situation; when we enter alien cultures or even alien social structures within our own culture, in our encounters with certain types of neurotic or psychotic patient (it is indeed the unintelligibility of such patients' actions that leads to their being treated as patients; actions unintelligible to the agent as well as to everyone else are understood—rightly—as a kind of suffering), but also in everyday situations. Consider an example.

I am standing waiting for a bus and the young man standing next to me suddenly says, ‘The name of the common wild duck is Histrionicus histrimicus histrionicus.' There is no problem as to the meaning of the sentence he uttered: the problem is, how to answer the question, what was he doing in uttering it? Suppose he just uttered such sentences at random intervals; this would be one possible form of madness. We would render his action of utterance intelligible if one of the following turned out to be true. He has mistaken me for someone who yesterday had approached him in the library and asked: ‘Do you by any chance know the Latin name of the common wild duck?' Or he has just come from a session with his psychotherapist who has urged him to break down his shyness by talking to strangers. ‘But what shall I say?' ‘Oh, anything at all.' Or he is a Soviet spy waiting at a prearranged rendezvous and uttering the ill-chosen code sentence which will identify him to his contact. In each case the act of utterance become intelligible by finding its place in a narrative.

To this it may be replied that the supplying of a narrative is not necessary to make such an act intelligible. All that is required is that we can identify the relevant type of speech act (e.g. ‘He was answering a question) or some purpose served by his utterance (e.g. ‘He was trying to attract your attention'). But speech acts and purposes too can be intelligible or unintelligible. Suppose that the man at the bus stop explains his act of utterance by saying ‘I was answering a question.' I reply: ‘But I never asked you any question to which that could have been the answer.' He says, ‘Oh, I know that.' Once again his action becomes unintelligible. And a parallel example could easily be constructed to show that the mere fact that an action serves some purposes of a recognized type is not sufficient to render an action intelligible. Both purposes and speech-acts require contexts.

The most familiar type of context in and by reference to which speech-acts and purposes are rendered intelligible is the conversation. Conversation is so all-pervasive a feature of the human world that it tends to escape philosophical attention. Yet remove conversation from human life and what would be left? Consider then what is involved in following a conversation and finding it intelligible or unintelligible. (To find a conversation intelligible is not the same as to understand it; for a conversation which I overhear may be intelligible, but I may fail to understand it.) If I listen to a conversation between two other people my ability to grasp the thread of the conversation will involve an ability to bring it under some one out of a set of descriptions in which the degree and kind of coherence in the conversation is brought out: ‘a drunken, rambling quarrel', ‘a serious intellectual disagreement', ‘a tragic misunderstanding of each other', ‘a comic, even farcial miscontrual of each other's motives', ‘a penetrating interchange of views', ‘a struggle to dominate each other', ‘a trivial exchange of gossip.'

The use of words such as ‘tragic', ‘comic', and ‘farcial' is not marginal to such evaluations. We allocate conversations to genres, just as we do literary narratives. Indeed a conversation is a dramatic work, even if a very short one, in which the participants are not only the actors, but also the joint authors, working out in agreement or disagreement the mode of their production. For it is not just that conversations belong to genres in just the way that plays and novels do; but they have beginnings, middles and endings just as do literary works. They embody reversals and recognitions; they move towards and away from climaxes. There may within a longer conversation be digressions and subplots, indeed digressions within digressions and subplots within subplots.

But if this is true of conversations, it is true also mutatis mutandis of battles, chess games, courtships, philosophy seminars, families at the dinner table, businessmen negotiating contracts—that is, of human transactions in general. For conversation, understood widely enough, is the form of human transactions in general. Conversational behavior is not a special sort or aspect of human behavior, even though the forms of language-using and of human life are such that the deeds of others speak for them as much as do their words. For that is possible only because they are the deeds of those who have words.

I am presenting both conversations in particular then and human actions in general as enacted narratives. Narrative is not the work of poets, dramatists and novelists reflecting upon events which had no narrative order before one was imposed by the singer or the writer; narrative form is neither disguise nor decoration. Barbara Hardy has written that ‘we dream in narrative, day-dream in narrative, remember, anticipate, hope, despair, believe, doubt, plan, revise, criticize, construct, gossip, learn, hate and love by narrative' in arguing the same point (Hardy 1968, p. 5).

At the beginning of this chapter I argued that in successfully identifying and understanding what someone else is doing we always move towards placing a particular episode in the context of a set of narrative histories, histories both of the individuals concerned and of the settings in which they act and suffer. It is now becoming clear that we render the actions of others intelligible in this way because action itself has a basically historical character. It is because we all live out narratives in our lives and because we understand our own lives in terms of the narratives that we live out that the form of narrative is appropriate for understanding the actions of others. Stories are lived before they are told—except in the case of fiction.

This has of course been denied in recent debates. Louis O. Mink, quarrelling with Barbara Hardy's view, has asserted: ‘Stories are not lived but told. Life has no beginnings, middles, or ends; there are meetings, but the start of an affair belongs to the story we tell ourselves later, and there are partings, but final partings only in the story. There are hopes, plans, battles and ideas, but only in retrospective stories are hopes unfulfilled, plans miscarried, battles decisive, and ideas seminal. Only in the story is it America which Columbus discovers and only in the story is the kingdom lost for want of a nail' (Mink 1970, pp. 557–8).

What are we to say to this? Certainly we must agree that it is only retrospectively that hopes can be characterized as unfulfilled or battles as decisive and so on. But we so characterize them in life as much as in art. And to someone who says that in life there are no endings, or that final partings take place only in stories, one is tempted to reply, ‘But have you never heard of death?' Homer did not have to tell the tale of Hector before Andromache could lament unfulfilled hope and final parting. There are countless Hectors and countless Andromaches whose lives embodied the form of their Homeric namesakes, but who never came to the attention of any poet. What is true is that in taking an event as a beginning or an ending we bestow a significance upon it which may be debatable. Did the Roman republic end with the death of Julius Caesar, or at Philippi, or with the founding of the principate? The answer is surely that, like Charles II, it was a long time a-dying; but this answer implies the reality of its ending as much as do any of the former. There is a crucial sense in which the principate of Augustus, or the taking of the oath in the tennis court, or the decision to construct an atomic bomb at Los Alamos constitute beginnings; the peace of 404 B.C., the abolition of the Scottish Parliament and the battle of Waterloo equally constitute endings; while there are many events which are both endings and beginnings.

As with beginnings, middles and endings, so also with genres and with the phenomenon of embedding. Consider the question of to what genre the life of Thomas Becket belongs, a question which has to be asked and answered before we can decide how it is to be written. (On Mink's paradoxical view this question could not be asked until after the life had been written.) In some of the medieval versions, Thomas's career is presented in terms of the canons of medieval hagiography. In the Icelandic Thomas Saga he is presented as a saga hero. In Dom David Knowles's modern biography the story is a tragedy, the tragic relationship of Thomas and Henry II, each of whom satisfies Aristode's demand that the hero be a great man with a fatal flaw. Now it clearly makes sense to ask who is right, if anyone: the monk William of Canterbury, the author of the saga, or the Cambridge Regius Professor Emeritus? The answer appears to be clearly the last. The true genre of the life is neither hagiography nor saga, but tragedy. So of such modern narrative subjects as the life of Trotsky or that of Lenin, of the history of the Soviet Communist Party or the American presidency, we may also ask: To what genre does their history belong? And this is the same question as: What type of account of their history will be both true and intelligible?

Or consider again how one narrative may be embedded in another. In both plays and novels there are well-known examples: the play within the play in Hamlet, Wandering Willie's Tale in Redgauntlet, Aeneas' narrative to Dido in book 2 of the Aeneid, and so on. But there are equally well-known examples in real life. Consider again the way in which the career of Becket as archbishop and chancellor is embedded within the reign of Henry II, or the way in which the tragic life of Mary Stuart is embedded in that of Elizabeth I, or the history of the Confederacy within the history of the United States. Someone may discover (or not discover) that he or she is a character in a number of narratives at the same time, some of them embedded in others. Or again, what seemed to be an intelligible narrative in which one was playing a part may be transformed wholly or partly into a story of unintelligible episodes. This last is what happened to Kafka's character K. in both The Trial and The Castle. (It is no accident that Kafka could not end his novels, for the notion of an ending like that of a beginning has its sense only in terms of intelligible narrative.)

I spoke earlier of the agent as not only an actor, but an author. Now I must emphasize that what the agent is able to do and say intelligibly as an actor is deeply affected by the fact that we are never more (and sometimes less) than the co-authors of our own narratives. Only in fantasy do we live what story we please. In life, as both Aristotle and Engels noted, we are always under certain constraints. We enter upon a stage which we did not design and we find ourselves part of an action that was not of our making. Each of us being a main character in his own drama plays subordinate parts in the dramas of others, and each drama constrains the others. In my drama, perhaps, I am Hamlet or Iago or at least the swineherd who may yet become a prince, but to you I am only A Gentleman or at best Second Murderer, while you are my Polonius or my Gravedigger, but your own hero. Each of our dramas exerts constraints on each other's, making the whole different from the parts, but still dramatic.

It is considerations as complex as these which are involved in making the notion of intelligibility the conceptual connecting link between the notion of action and that of narrative. Once we have understood its importance the claim that the concept of an action is secondary to that of an intelligible action will perhaps appear less bizarre and so too will the claim that the notion of ‘an' action, while of the highest practical importance, is always a potentially misleading abstraction. An action is a moment in a possible or actual history or in a number of such histories. The notion of a history is as fundamental a notion as the notion of an action. Each requires the other. But I cannot say this without noticing that it is precisely this that Sartre denies—as indeed his whole theory of the self, which captures so well the spirit of modernity, requires that he should. In La Nausee, Sartre makes Antoine Roquentin argue not just what Mink argues, that narrative is very different from life, but that to present human life in the form of a narrative is always to falsify it. There are not and there cannot be any true stories. Human life is composed of discrete actions which lead nowhere, which have no order; the story-teller imposes on human events retrospectively an order which they did not have while they were lived. Clearly if Sartre/Roquentin is right—I speak of Sartre/Roquentin to distinguish him from such other well-know [sic] characters as Sartre/Heidegger and Sartre/Marx—my central contention must be mistaken. There is nonetheless an important point of agreement between my thesis and that of Sartre/Roquentin. We agree in identifying the intelligibility of an action with its place in a narrative sequence. Only Sartre/Roquentin takes it that human actions are as such unintelligible occurrences: it is to a realization of the metaphysical implications of this that Roquentin is brought in the course of the novel and the practical effect upon him is to bring to an end his own project of writing an historical biography. This project no longer makes sense. Either he will write what is true or he will write an intelligible history, but the one possibility excludes the other. Is Sartre/Roquentin right?

We can discover what is wrong with Sartre's thesis in either of two ways. One is to ask: what would human actions deprived of any falsifying narrative order be like? Sartre himself never answers this question; it is striking that in order to show that there are no true narratives, he himself writes a narrative, albeit a fictional one. But the only picture that I find myself able to form of human nature an-sich, prior to the alleged misinterpretation by narrative, is the kind of dislocated sequence which Dr. Johnson offers us in his notes of his travels in France: ‘There we waited on the ladies—Morville's.—Spain, Country towns all beggars. At Dijon he could not find the way to Orleans.—Cross roads of France very bad.—Five soldiers. Women.—Soldiers escaped.—The Colonel would not lose five men for the sake of one woman.—The magistrate cannot seize a soldier but by the Colonel's permission, etc., etc.' (quoted in Hobsbaum 1973, p. 32). What this suggests is what I take to be true, namely that the characterization of actions allegedly prior to any narrative form being imposed upon them will always turn out to be the presentation of what are plainly the disjointed parts of some possible narrative.

We can also approach the question in another way. What I have called a history is an enacted dramatic narrative in which the characters are also the authors. The characters of course never start literally ab initio; they plunge in medias res, the beginnings of their story already made for them by what and who has gone before. But when Julian Grenfell or Edward Thomas went off to France in the 1914–18 war they no less enacted a narrative than did Menelaus or Odysseus when they went off. The difference between imaginary characters and real ones is not in the narrative form of what they do; it is in the degree of their authorship of that form and of their own deeds. Of course just as they do not begin where they please, they cannot go on exacdy as they please either; each character is constrained by the actions of others and by the social settings presupposed in his and their actions, a point forcibly made by Marx in the classical, if not entirely satisfactory account of human life as enacted dramatic narrative, The Eighteenth Brumaire of Louis Bonaparte.

I call Marx's account less than satisfactory partly because he wishes to present the narrative of human social life in a way that will be compatible with a view of the life as law-governed and predictable in a particular way. But it is crucial that at any given point in an enacted dramatic narrative we do not know what will happen next. The kind of unpredictability for which I argued in Chapter 8 is required by the narrative structure of human life, and the empirical generalizations and explorations which social scientists discover provide a kind of understanding of human life which is perfectly compatible with that structure.

This unpredictability coexists with a second crucial characteristic of all lived narratives, a certain teleological character. We live out our lives, both individually and in our relationships with each other, in the light of certain conceptions of a possible shared future, a future in which certain possibilities beckon us forward and others repel us, some seem already foreclosed and others perhaps inevitable. There is no present which is not informed by some image of some future and an image of the future which always presents itself in the form of a telos—or of a variety of ends or goals—towards which we are either moving or failing to move in the present. Unpredictability and teleology therefore coexist as part of our lives; like characters in a fictional narrative we do not know what will happen next, but nonetheless our lives have a certain form which projects itself towards our future. Thus the narratives which we live out have both an unpredictable and a partially teleological character. If the narrative of our individual and social lives is to continue intelligibly—and either type of narrative may lapse into unintelligibility—it is always both the case that there are constraints on how the story can continue and that within those constraints there are indefinitely many ways that it can continue.

A central thesis then begins to emerge: man is in his actions and practice, as well as in his fictions, essentially a story-telling animal. He is not essentially, but becomes through his history, a teller of stories that aspire to truth. But the key question for men is not about their own authorship; I can only answer the question ‘What am I to do?' if I can answer the prior question ‘Of what story or stories do I find myself a part?' We enter human society, that is, with one or more imputed characters—roles into which we have been drafted—and we have to learn what they are in order to be able to understand how others respond to us and how our responses to them are apt to be construed. It is through hearing stories about wicked stepmothers, lost children, good but misguided kings, wolves that suckle twin boys, youngest sons who receive no inheritance but must make their own way in the world and eldest sons who waste their inheritance on riotous living and go into exile to live with the swine, that children learn or mislearn both what a child and what a parent is, what the cast of characters may be in the drama into which they have been born and what the ways of the world are. Deprive children of stories and you leave them unscripted, anxious stutterers in their actions as in their words. Hence there is no way to give us an understanding of any society, including our own, except through the stock of stories which constitute its initial dramatic resources. Mythology, in its original sense, is at the heart of things. Vico was right and so was Joyce. And so too of course is that moral tradition from heroic society to its medieval heirs according to which the telling of stories has a key part in educating us into the virtues.

I suggested earlier that ‘an' action is always an episode in a possible history: I would now like to make a related suggestion about another concept, that of personal identity. Derek Parfit and others have recently drawn our attention to the contrast between the criteria of strict identity, which is an all-or-nothing matter (either the Tichborne claimant is the last Tichborne heir or he is not; either all the properties of the last heir belong to the claimant or the claimant is not the heir—Leibniz's Law applies) and the psychological continuities of personality which are a matter of more or less. (Am I the same man at fifty as I was at forty in respect of memory, intellectual powers, critical responses? More or less.) But what is crucial to human beings as characters in enacted narratives is that, possessing only the resources of psychological continuity, we have to be able to respond to the imputation of strict identity. I am forever whatever I have been at any time for others—and I may at any time be called upon to answer for it—no matter how changed I may be now. There is no way of founding my identity—or lack of it—on the psychological continuity or discontinuity of the self. The self inhabits a character whose unity is given as the unity of a character. Once again there is a crucial disagreement with empiricist or analytical philosophers on the one hand and with existentialists on the other.

Empiricists, such as Locke or Hume, tried to give an account of personal identity solely in terms of psychological states or events. Analytical philosophers, in so many ways their heirs as well as their critics, have wrestled with the connection between those states and events and strict identity understood in terms of Leibniz's Law. Both have failed to see that a background has been omitted, the lack of which makes the problems insoluble. That background is provided by the concept of a story and of that kind of unity of character which a story requires. Just as a history is not a sequence of actions, but the concept of an action is that of a moment in an actual or possible history abstracted for some purpose from that history, so the characters in a history are not a collection of persons, but the concept of a person is that of a character abstracted from a history.

What the narrative concept of selfhood requires is thus twofold. On the one hand, I am what I may justifiably be taken by others to be in the course of living out a story that runs from my birth to my death; I am the subject of a history that is my own and no one else's, that has its own peculiar meaning. When someone complains—as do some of those who attempt or commit suicide—that his or her life is meaningless, he or she is often and perhaps characteristically complaining that the narrative of their life has become unintelligible to them, that it lacks any point, any movement towards a climax or a telos. Hence the point of doing any one thing rather than another at crucial junctures in their lives seems to such person to have been lost.

To be the subject of a narrative that runs from one's birth to one's death is, I remarked earlier, to be accountable for the actions and experiences which compose a narratable life. It is, that is, to be open to being asked to give a certain kind of account of what one did or what happened to one or what one witnessed at any earlier point in one's life than the time at which the question is posed. Of course someone may have forgotten or suffered brain damage or simply not attended sufficiently at the relevant time to be able to give the relevant account. But to say of someone under some one description (‘The prisoner of the Chateau d'lf') that he is the same person as someone characterized quite differently (‘The Count of Monte Cristo') is precisely to say that it makes sense to ask him to give an intelligible narrative account enabling us to understand how he could at different times and different places be one and the same person and yet be so differently characterized. Thus personal identity is just that identity presupposed by the unity of the character which the unity of a narrative requires. Without such unity there would not be subjects of whom stories could be told.

The other aspect of narrative selfhood is correlative: I am not only accountable, I am one who can always ask others for an account, who can put others to the question. I am part of their story, as they are part of mine. The narrative of any one life is part of an interlocking set of narratives. Moreover this asking for and giving of accounts itself plays an important part in constituting narratives. Asking you what you did and why, saying what I did and why, pondering the differences between your account of what I did and my account of what I did, and vice versa, these are essential constituents of all but the very simplest and barest of narratives. Thus without the accountability of the self those trains of events that constitute all but the simplest and barest of narratives could not occur; and without that same accountability narratives would lack that continuity required to make both them and the actions that constitute them intelligible.

It is important to notice that I am not arguing that the concepts of narrative or of intelligibility or of accountability are more fundamental than that of personal identity. The concepts of narrative, intelligibility and accountability presuppose the applicability of the concept of personal identity, just as it presupposes their applicability and just as indeed each of these three presupposes the applicability of the two others. The relationship is one of mutual presupposition. It does follow of course that all attempts to elucidate the notion of personal identity independently of and in isolation from the notions of narrative, intelligibility and accountability are bound to fail. As all such attempts have.

It is now possible to return to the question from which this enquiry into the nature of human action and identity started: In what does the unity of an individual life consist? The answer is that its unity is the unity of a narrative embodied in a single life. To ask ‘What is the good for me?' is to ask how best I might live out that unity and bring it to completion. To ask ‘What is the good for man?' is to ask what all answers to the former question must have in common. But now it is important to emphasize that it is the systematic asking of these two questions and the attempt to answer them in deed as well as in word which provide the moral life with its unity. The unity of a human life is the unity of a narrative quest. Quests sometimes fail, are frustrated, abandoned or dissipated into distractions; and human lives may in all these ways also fail. But the only criteria for success or failure in a human life as a whole are the criteria of success or failure in a narrated or to-be-narrated quest. A quest for what?

Two key features of the medieval conception of a quest need to be recalled. The first is that without some at least partly determinate conception of the final telos there could not be any beginning to a quest. Some conception of the good for man is required. Whence is such a conception to be drawn? Precisely from those questions which led us to attempt to transcend that limited conception of the virtues which is available in and through practices. It is in looking for a conception of the good which will enable us to order other goods, for a conception of the good which will enable us to extend our understanding of the purpose and content of the virtues, for a conception of the good which will enable us to understand the place of integrity and constancy in life, that we initially define the kind of life which is a quest for the good. But secondly it is clear the medieval conception of a quest is not at all that of a search for something already adequately characterized, as miners search for gold or geologists for oil. It is in the course of the quest and only through encountering and coping with the various particular harms, dangers, temptations and distractions which provide any quest with its episodes and incidents that the goal of the quest is finally to be understood. A quest is always an education both as to the character of that which is sought and in self-knowledge.

The virtues therefore are to be understood as those dispositions which will not only sustain practices and enable us to achieve the goods internal to practices, but which will also sustain us in the relevant kind of quest for the good, by enabling us to overcome the harms, dangers, temptations and distractions which we encounter, and which will furnish us with increasing self-knowledge and increasing knowledge of the good. The catalogue of the virtues will therefore include the virtues required to sustain the kind of households and the kind of political communities in which men and women can seek for the good together and the virtues necessary for philosophical enquiry about the character of the good. We have then arrived at a provisional conclusion about the good life for man: the good life for man is the life spent in seeking for the good life for man, and the virtues necessary for the seeking are those which will enable us to understand what more and what else the good life for man is. We have also completed the second stage in our account of the virtues, by situating them in relation to the good life for man and not only in relation to practices. But our enquiry requires a third stage.

For I am never able to seek for the good or exercise the virtues only qua individual. This is partly because what it is to live the good life concretely varies from circumstance to circumstance even when it is one and the same conception of the good life and one and the same set of virtues which are being embodied in a human life. What the good life is for a fifth-century Athenian general will not be the same as what it was for a medieval nun or a seventeenth-century farmer. But it is not just that different individuals live in different social circumstances; it is also that we all approach our own circumstances as bearers of a particular social identity. I am someone's son or daughter, someone else's cousin or uncle; I am a citizen of this or that city, a member of this or that guild or profession; I belong to this clan, that tribe, this nation. Hence what is good for me has to be the good for one who inhabits these roles. As such, I inherit from the past of my family, my city, my tribe, my nation, a variety of debts, inheritances, rightful expectations and obligations. These constitute the given of my life, my moral starting point. This is in part what gives my life its own moral particularity.

This thought is likely to appear alien and even surprising from the standpoint of modern individualism. From the standpoint of individualism I am what I myself choose to be. I can always, if I wish to, put in question what are taken to be the merely contingent social features of my existence. I may biologically be my father's son; but I cannot be held responsible for what he did unless I choose implicitly or explicitly to assume such responsibility. I may legally be a citizen of a certain country; but I cannot be held responsible for what my country does or has done unless I choose implicitly or explicitly to assume such responsibility. Such individualism is expressed by those modern Americans who deny any responsibility for the effects of slavery upon black Americans, saying ‘I never owned any slaves'. It is more subtly the standpoint of those other modern Americans who accept a nicely calculated responsibility for such effects measured precisely by the benefits they themselves as individuals have indirectly received from slavery. In both cases ‘being an American' is not in itself taken to be part of the moral identity of the individual. And of course there is nothing peculiar to modern Americans in this attitude: the Englishman who says, I never did any wrong to Ireland; why bring up that old history as though it had something to do with me? or the young German who believes that being born after 1945 means that what Nazis did to Jews has no moral relevance to his relationship to his Jewish contemporaries, exhibit the same attitude, that according to which the self is detachable from its social and historical roles and statuses. And the self so detached is of course a self very much at home in either Sartre's or Goffman's perspective, a self that can have no history. The contrast with the narrative view of the self is clear. For the story of my life is always embedded in the story of those communities from which I derive my identity. I am born with a past; and to try to cut myself off from that past, in the individualist mode, is to deform my present relationships. The possession of an historical identity and the possession of a social identity coincide. Notice that rebellion against my identity is always one possible mode of expressing it.

Notice also that the fact that the self has to find its moral identity in and through its membership in communities such as those of the family, the neighborhood, the city and the tribe does not entail that the self has to accept the moral limitations of the particularity of those forms of community. Without those moral particularities to begin from there would never be anywhere to begin; but it is in moving forward from such particularity that the search for the good, for the universal, consists. Yet particularity can never be simply left behind or obliterated. The notion of escaping from it into a realm of entirely universal maxims which belong to man as such, whether in its eighteenth-century Kantian form or in the presentation of some modern analytical moral philosophies, is an illusion and an illusion with painful consequences. When men and women identify what are in fact their partial and particular causes too easily and too completely with the cause of some universal principle, they usually behave worse than they would otherwise do.

What I am, therefore, is in key part what I inherit, a specific past that is present to some degree in my present. I find myself part of a history and that is generally to say, whether I like it or not, whether I recognize it or not, one of the bearers of a tradition. It was important when I characterized the concept of a practice to notice that practices always have histories and that at any given moment what a practice is depends on a mode of understanding it which has been transmitted often through many generations. And thus, insofar as the virtues sustain the relationships required for practices, they have to sustain relationships to the past—and to the future—as well as in the present. But the traditions through which particular practices are transmitted and reshaped never exist in isolation for larger social traditions. What constitutes such traditions?

We are apt to be misled here by the ideological uses to which the concept of a tradition has been put by conservative political theorists. Characteristically such theorists have followed Burke in contrasting tradition with reason and the stability of tradition with conflict. Both contrasts obfuscate.

For all reasoning takes place within the context of some traditional mode of thought, transcending through criticism and invention the limitations of what had hitherto been reasoned in that tradition; this is as true of modern physics as of medieval logic. Moreover when a tradition is in good order it is always partially constituted by an argument about the goods the pursuit of which gives to that tradition its particular point and purpose.

So when an institution—a university, say, or a farm, or a hospital—is the bearer of a tradition of practice or practices, its common life will be partly, but in a centrally important way, constituted by a continuous argument as to what a university is and ought to be or what good farming is or what good medicine is. Traditions, when vital, embody continuities of conflict. Indeed when a tradition becomes Burkean, it is always dying or dead.

The individualism of modernity could of course find no use for the notion of tradition within its own conceptual scheme except as an adversary notion; it therefore all too willingly abandoned it to the Burkeans, who, faithful to Burke's own allegiance, tried to combine adherence in politics to a conception of tradition which would vindicate the oligarchical revolution of property of 1688 and adherence in economics to the doctrine and institutions of the free market. The theoretical incoherence of this mismatch did not deprive it of ideological usefulness. But the outcome has been that modern conservatives are for the most part engaged in conserving only older rather than later versions of liberal individualism. Their own core doctrine is as liberal and as individualist as that of self-avowed liberals.

A living tradition then is an historically extended, socially embodied argument, and an argument precisely in part about the goods which constitute that tradition. Within a tradition the pursuit of goods extends through generations, sometimes through many generations. Hence the individual's search for his or her good is generally and characteristically conducted within a context defined by those traditions of which the individual's life is a part, and this is true both of those goods which are internal to practices and of the goods of a single life. Once again the narrative phenomenon of embedding is crucial: the history of a practice in our time is generally and characteristically embedded in and made intelligible in terms of the larger and longer history of the tradition through which the practice in its present form was conveyed to us; the history of each of our own lives is generally and characteristically embedded in and made intelligible in terms of the larger and longer histories of a number of traditions. I have to say ‘generally and characteristically' rather than ‘always', for traditions decay, disintegrate and disappear. What then sustains and strengthens traditions? What weakens and destroys them?

The answer in key part is: the exercise or the lack of exercise of the relevant virtues. The virtues find their point and purpose not only in sustaining those relationships necessary if the variety of goods internal to practices are to be achieved and not only in sustaining the form of an individual life in which that individual may seek out his or her good as the good of his or her whole life, but also in sustaining those traditions which provide both practices and individual lives with their necessary historical context. Lack of justice, lack of truthfulness, lack of courage, lack of the relevant intellectual virtues—these corrupt traditions, just as they do those institutions and practices which derive their life from the traditions of which they are the contemporary embodiments. To recognize this is of course also to recognize the existence of an additional virtue, one whose importance is perhaps most obvious when it is least present, the virtue of having an adequate sense of the traditions to which one belongs or which confront one. This virtue is not to be confused with any form of conservative antiquarianism, I am not praising those who choose the conventional conservative role of laudator temporis acti. It is rather the case that an adequate sense of tradition manifests itself in a grasp of those future possibilities which the past has made available to the present. Living traditions, just because they continue a not-yet-completed narrative, confront a future whose determinate and determinable character, so far as it possesses any, derives from the past.

In practical reasoning the possession of this virtue is not manifested so much in the knowledge of a set of generalizations or maxims which may provide our practical inferences with major premises; its presence or absence rather appears in the kind of capacity for judgment which the agent possesses in knowing how to select among the relevant stack of maxims and how to apply them in particular situations. Cardinal Pole possessed it, Mary Tudor did not; Montrose possessed it, Charles I did not. What Cardinal Pole and the Marquis of Montrose possessed were in fact those virtues which enable their possessors to pursue both their own good and the good of the tradition of which they are the bearers even in situations defined by the necessity of tragic, dilemmatic choice. Such choices, understood in the context of the tradition of the virtues, are very different from those which face the modern adherents of rival and incommensurable moral premises in the debates about which I wrote in Chapter 2. Wherein does the difference lie?

It has often been suggested—by J. L. Austin, for example—that either we can admit the existence of rival and contingently incompatible goods which make incompatible claims to our practical allegiance or we can believe in some determinate conception of the good life for man, but that these are mutually exclusive alternatives. No one can consistently hold both these views. What this contention is blind to is that there may be better or worse ways for individuals to live through the tragic confrontation of good with good. And that to know what the good life for man is may require knowing what are the better and what are the worse ways of living in and through such situations. Nothing a priori rules out this possibility; and this suggests that within a view such as Austin's there is concealed an unacknowledged empirical premise about the character of tragic situations.

One way in which the choice between rival goods in a tragic situation differs from the modern choice between incommensurable moral premises is that both of the alternative courses of action which confront the individual have to be recognized as leading to some authentic and substantial good. By choosing one I do nothing to diminish or derogate from the claim upon me of the other; and therefore, whatever I do, I shall have left undone what I ought to have done. The tragic protagonist, unlike the moral agent as depicted by Sartre or Hare, is not choosing between allegiance to one moral principle rather than another, nor is he or she deciding upon some principle of priority between moral principles. Hence the ‘ought' involved has a different meaning and force from that of the ‘ought' in moral principles understood in a modern way. For the tragic protagonist cannot do everything that he or she ought to do. This ‘ought', unlike Kant's, does not imply ‘can'. Moreover any attempt to map the logic of such ‘ought' assertions on to some modal calculus so as to produce a version of deontic logic has to fail. (See, from a very different point of view, Bas C. Van Fraasen 1973.)

Yet it is clear that the moral task of the tragic protagonist may be performed better or worse, independently of the choice between alternatives that he or she makes—ex hypothesi he or she has no right choice to make. The tragic protagonist may behave heroically or unheroically, generously or ungenerously, gracefully or gracelessly, prudently or imprudently. To perform his or her task better rather than worse will be to do both what is better for him or her qua individual and qua parent or child or qua citizen or member of a profession, or perhaps qua some or all of these. The existence of tragic dilemmas casts no doubt upon and provides no counterexamples to the thesis that assertions of the form ‘To do this in this way would be better for X and/or for his or her family, city or profession' are susceptible of objective truth and falsity, any more than the existence of alternative and contingently incompatible forms of medical treatment casts doubt on the thesis that assertions of the form ‘To undergo his medical treatment in this way would be better for X and/or his or her family' are susceptible of objective truth and falsity. (See, from a different point of view, the illuminating discussion in Samuel Guttenplan 1979–80, pp. 61–80).

The presupposition of this objectivity is of course that we can understand the notion of ‘good for X' and cognate notions in terms of some conception of the unity of X's life. What is better or worse for X depends upon the character of that intelligible narrative which provides X's life with its unity. Unsurprisingly it is the lack of any such unifying conception of a human life which underlies modern denials of the factual character of moral judgments and more especially of those judgments which ascribe virtues or vices to individuals.

I argued earlier that every moral philosophy has some particular sociology as its counterpart. What I have tried to spell out in this chapter is the kind of understanding of social life which the tradition of the virtues requires, a kind of understanding very different from those dominant in the culture of bureaucratic individualism. Within that culture conceptions of the virtues become marginal and the tradition of the virtues remains central only in the lives of social groups whose existence is on the margins of the central culture. Within the central cultural of liberal or bureaucratic individualism new conceptions of the virtues emerge and the concept of a virtue is itself transformed. To the history of that transformation I therefore now turn; for we shall only understand the tradition of the virtues fully if we understand to what kinds of degeneration it has proved liable.

From Alistaire MacIntyre, After Virtue, 3rd ed., University of Notre Dame Press, © 2007, p. 181–225. Reprinted with permission.