CHAPTER 8: POSTSCRIPT TO GREEK ETHICS
THE DIVISION of labor and the differentiation of function in early societies produces a vocabulary in which men are described in terms of the roles they fulfill. The use of evaluative words follows hard upon this, since any role can be filled well or ill, and any customary mode of behavior conformed to or broken away from. But evaluation with a wider scope is only possible when traditional role behavior is seen in contrast with other possibilities, and the necessity of choice between old and new ways becomes a fact of social life. It is not surprising therefore that it is in the transition from the society which was the bearer of the Homeric poems to the society of the fifth-century city-state that good and its cognates acquired a variety of uses, and that it is in the following decades that men reflect self-consciously about those uses. Greek philosophical ethics differs from later moral philosophy in ways that reflect the difference between Greek society and modern society. The concepts of duty and responsibility in the modern sense appear only in germ or marginally; those of goodness, virtue, and prudence are central. The respective roles of these concepts hinge upon a central difference. In general, Greek ethics asks, What am I to do if I am to fare well? Modern ethics asks, What ought I to do if I am to do right? and it asks this question in such a way that doing right is made something quite independent of faring well. A writer thoroughly imbued with the modern spirit in ethics, the Oxford philosopher H. A. Prichard20 could accuse Plato of falling into error simply by attempting to justify justice at all. For to justify justice is to show that it is more profitable than injustice, that it is to our interest to be just. But if we do what is just and right because it is
to our interest, then, so Prichard takes almost for granted, we are not doing it because it is just and right at all. Morality indeed cannot have any justification external to itself; if we do not do what is right for its own sake, and whether it is to our interest or not, then we are not doing what is right.
The assumption made by Prichard is that the notion of what is to our interest, of what is profitable to us, is logically independent of the concept of what it is just and right for us to do. If what is profitable is also just, this, so far as ethics is concerned, is merely coincidental, a happy accident. Doing what we want and getting what we want is one thing; doing what we ought is another. But Prichard here misses the point not only of Plato but of the implications of the Greek moral vocabulary which Plato uses. The Greek moral vocabulary is not so framed that the objects of our desires and our moral aims are necessarily independent. To do well and to fare well are found together in a word like εὐδαίμων. From such purely linguistic considerations, of course, little of substance follows. It still remains to ask whether it is modern ethics which is clarifying a valid distinction that the Greek moral vocabulary fails to observe or Greek ethics which is refusing to make a false and confusing distinction. One way of answering this question would be as follows.
Ethics is concerned with human actions. Human actions are not simply bodily movements. We can identify as instances of the same human action deeds which are executed by means of quite different bodily movements-as the movements involved in shaking a hand and those involved in putting out a flag may both be examples of welcoming somebody. And we can identify as the same bodily movements those which exemplify very different actions-as a movement of the legs may be part of running a race or of fleeing in battle. How, then, do I exhibit a piece of behavior as an action or part of a sequence of actions rather than as mere bodily movement? The answer can only be that it is by showing that it serves a purpose which constitutes part or the whole of the agent’s intention in doing what he does. What is more, the agent’s purpose is only to be made intelligible as the expression of his desires and aims.
Consider now how modern post-Kantian ethics emphasizes the contrast between duty and inclination. If what I do is made intelligible in terms of the pursuit of my desires, if my desires are cited as affording me reasons for doing what I do, it cannot be that in doing what I do I am doing my duty. Hence when I am doing my duty what I do cannot be exhibited as a human action, intelligible in the way that ordinary human actions are. So the pursuit of duty becomes a realm of its own, unconnected with anything else in human life. To this the reply of a writer like Prichard would be that indeed this is so, and that to suppose it could be otherwise would be an error. But we can now more fruitfully approach Prichard’s position in another way, and exhibit its historical roots. If we do so, we shall see a gradual attenuation of the concept of duty and of kindred concepts, in which there is a progress from a notion of a duty as consisting in the requirement to fulfill a specific role, the fulfillment of which serves a purpose which is entirely intelligible as the expression of normal human desires (consider the duties of a father, seaman, or doctor as examples); the next step is perhaps the concept of duty as something to be done by the individual whatever his private desires; finally, we reach the concept of duty as divorced from desire altogether. If we could not explain Prichard’s concept of duty historically, I think we should be very much in the position of anthropologists who come across a new and incomprehensible word, such as, for example, tabu, a word which is puzzling because it appears not simply to mean “prohibited” but to give a reason for the prohibition, without its being clear what reason. So when someone like Prichard says it is our “duty” to do something, he does not just tell us to do it, as though he uttered an injunction “Do that,” but he appears to give us a reason. Consequently, just as we may ask of Polynesians why we should refrain from doing something because it is tabu, so we shall want to ask Prichard why we should do something because it is our duty. And in each case the answer will be similar, and similarly incomprehensible: “Because it is tabu,” “Because it is your duty.” The lack of connection with other aims, purposes, and desires produces in the end unintelligibility. Yet the concept which Prichard elucidates is one in common use. “‘Why ought I to do that?’ ‘You just ought’” is not uncommon as a form of moral dialogue in modern society.
Thus the philosophical elucidation raises interesting problems about the role of the concept in our social life. But rather than pursuing these at this stage, we must instead return to the Greeks. The crucial point for the immediate discussion is that it may now be clearer why we could not use the moral words which express the modern concept of duty in translating Greek moral words; for these retain the connection with the vocabulary of desire in terms of which they can be made intelligible.
The function of evaluative terms in Greek is, then, to grade different possibilities of conduct in terms of our desires; but in terms of which of our desires? Both Plato and Aristotle criticize the simple sophistic picture of human desires. We have to ask not only what we do now in fact happen to want, but what in the long run and fundamentally we want to want. And this implies a picture of man, made explicit, in different ways, in both Plato and Aristotle, in which certain satisfactions are objectively higher than others. The use of the word objectively implies the existence of an impersonal, unchosen criterion. What is it? That there is some such criterion follows from treating the question, What is the good for man? or even just, What is good for man? as an intelligible question at all. For unless there is some criterion by which to judge between possible answers, all possible replies are on a level and the question ceases to have point. It does not, of course, follow that there must be such a criterion (or criteria), but only that the question and the criterion stand or fall together. Plato’s transcendentalism springs in part from his grasp of this point. He believes that there must be a criterion. It cannot be derived from existing social structure and institutions, for we use our evaluative concepts to criticize these. It cannot be derived either from our desires just as they are, for we use our evaluative concepts to criticize and to grade our desires. Hence it is easy to conclude that it must be derived from an order existing apart from human life as it is. Where Plato sees the criterion as transcendent, Aristotle sees it as embedded in one particular sort of practice and social arrangement. Both assume that if the chain of justifications which are constituted by answers to questions about the good for men is to be a chain of rational arguments, there must be essentially only one such chain and there must be one essential point at which it reaches a final conclusion (the vision of the Form of the Good or eudaemonistic contemplation). This is of course a mistake, and it is a mistake which both Plato and Aristotle make because they do not understand the conditions which have to be satisfied for there to be available the kind of criteria the existence of which they take for granted-even if they are sometimes in doubt as to their precise nature.
If I treat any form of inquiry as rational, I presuppose that there is some criterion by which to determine whether the answers to its questions are correct or incorrect. In speaking of a criterion I speak of a standard which the individual is not free to accept or reject as he wills or chooses. He may reject a given criterion on rational grounds, such as that its falsity is entailed by some more fundamental or more generally applicable criterion; or he may find a proposed criterion unintelligible upon closer scrutiny. Consider two quite different kinds of example.
Arithmetic is a rational, because rule-governed, discipline. The rules which govern simple arithmetical operations enable us to determine whether the answer to a given sum is correct or incorrect. Anyone who understands the meaning of the words one, two, plus, equals, and three has no choice over whether to admit the truth of “One plus two equals three.” But a condition of there being an agreed meaning for the words in question is that there should be a socially established practice of counting. We can imagine a tribe lacking number concepts because they lack the practice of counting. I do not mean by this that counting as a social, teachable practice is logically prior to the possession of number concepts, for counting itself in turn presupposes such possession. But I can only appeal to a rule to settle a disputed arithmetical question in a community where number concepts are intelligible, and they will only be intelligible where counting is an established and recognized practice.
Consider now how similar evaluative terms are, in this one respect at least, to arithmetical. We are more accustomed to think of arithmetic as a rational discipline than we are of the criticism of cricket and football, chess and bridge. (This is partly because in our study of Greek culture we usually overrate Plato and underrate the Olympic games about which we could learn from Lucian. Plato thinks of “gymnastic,” of games playing, as merely a means to an end, as part of an educational discipline whose point lies in a final end-product of a quite different kind. This is also the doctrine of the English public schools, for whom the point of games is that they produce “character.”) But a study of the concepts used in the criticism of games for their own sake is philosophically revealing, even about Plato. Consider the concept of a good batsman. The questions of whether a batsman is a good batsman and how good a batsman he is are intelligible because there are established criteria: variety of strokes, ability to improvise, moral stamina in crises. We have these criteria because we have criteria of success or failure in cricket in general, and in the role of a batsman in particular; and the winning of matches is not, of course, the sole criterion. How you win them also enters into it. But these criteria can only be appealed to because there is an established practice of games playing and can only be appealed to by those who share the social life in which this practice is established. Imagine a people who did not share the concept of a game and therefore could not acquire the relevant criteria. All that they would and could grasp would be that the word good was generally being used in contexts where approval of some sort was being indicated. Their philosophers would naturally enough construct theories about the meaning of good to the effect that its use is to express approval. And these theories would necessarily miss a large part of the point.
In Greek ethics something analogous to this imagined situation actually occurs. We begin with a society in which the use of evaluative words is tied to the notion of the fulfillment of a socially established role. We can indeed imagine a society of which this is true in a far stronger sense than it is of the imaginary society depicted in or the real societies reflected in the Homeric poems. Here the nouns and verbs to which evaluative adjectives and adverbs are attached would always invariably be those which name roles and role-fulfilling activities. In consequence, all the uses of good will belong with that class of adjectives whose meaning and force are dependent upon the meaning of the noun or noun-phrase to which they are attached. We can understand this class by contrasting it with the class of adjectives whose meaning and force is not so dependent. Such are, for example, color words. We can render them meaningless by attaching them to a noun or noun-phrase which cannot allow them any sense if it is used with its normal meaning (consider “pink rational number”), but when they are attached to any noun in such a way as to form a meaningful phrase their meaning is independent of that of the noun. Because of this, I can in such cases validly deduce from “This is an XY,” both “This is an X” and “This is a Y.” (So from “This is a red book” it follows that “This is red” and “This is a book.”) But there are also the adjectives of which this is not true, where the force of the adjective is dependent upon the meaning of the particular noun to which it is attached, because the criteria for the correct application of the adjective vary with and are determined by the meaning of the noun. Such is good in those uses which connect with role-fulfillment. The criteria for the correct application of the expressions “good shepherd” and “good general” and “good flautist” are determined by the criteria for the application of the expressions shepherd, general, and flautist. In learning how to describe social life one also learns how to evaluate it. Moreover there are a variety of uses of good where such impersonal and objective criteria can be found: “good at” used of skills and “good for” used of medicines or instruments, of means efficient for given ends, are two examples. In a society restricted to these uses, all evaluation would be a matter of the application of criteria about which the individual was not free to exercise choice.
Such a society’s evaluative usage resembles the usage of those who criticize performances in a game. In both cases there are accepted standards; in both cases to acquire the vocabulary necessary to describe and to understand the game is logically inseparable from acquiring these standards. In both cases the fact that the standards are objective and impersonal is consistent with evaluative disagreements and even with disagreements which are incapable of resolution. This is because there are a number of criteria in terms of which we judge performances or exercises of ability and not just one single criterion for each role or skill. So we may in evaluating a batsman differ in the weight that we give to ability to improvise as against the possession of a particular stroke, and in evaluating a general differ in the weight we give to ability to organize supply lines as against tactical brilliance on the field.
Just as in the case of the criticisms of games we could imagine a social group in which the use of evaluative words was lost, so we can imagine a society in which traditional roles no longer exist and the consequent evaluative criteria are no longer used, but the evaluative words survive. In both cases all that remains of evaluation is the sense of approval attached to the words. The words become used as signs that the individual speaker is indicating his tastes, preferences, and choices. If we conceived of philosophical analysis as an analysis of how different concepts are in fact used in common speech, as a study of the logical features of usage, then we might well fall into an interesting trap at this point. For if we were to insist upon treating as the meaning of the word good only features of its use which were present on every occasion of its use, then we should naturally conclude that the essential meaning of the word is given by laying down its function of commending or expressing approval or indicating choice or preference, and so on; and that its association with criteria of an impersonal and objective kind is secondary, contingent, and accidental. Or we might fall into the opposite trap of supposing that since the word good is in many standard cases only used intelligibly if it is applied in accordance with impersonal and objective criteria, all uses in which it is divorced from such criteria are too unimportant to be taken seriously. In fact however, unless we see these two uses as constituting two successive phases in a historical narrative, we shall miss a large part of the point about the word good. When I speak of a historical narrative I mean one in which the later part is unintelligible until the former is supplied, and in which we have not understood the former until we see that what followed it was a possible sequel to what had gone before. The use of the word good when it is used only or primarily as an expression of approval or choice is unintelligible except as a survival from a period when criteria of an impersonal, unchosen kind governed its use, because it has no distinctive use or function to distinguish it from a simple imperative or expression of approval. What would be being said when something is called good would be no more or other than would be said by someone who said, “Choose one of that kind,” or, “That is the kind I prefer.” This apparent redundancy of good might be explained away by pointing to its propagandistic possibilities. The use of the word good actually says no more than is said by the man who straightforwardly announces his choice or preference, but a man may contrive to give the impression of saying more, and by so doing attach prestige to his announcement by the use of good. Good is a status symbol for expressions of choice, on this view. But this theory, in fact, discloses its own weakness, for why should good have this kind of status? why should it carry this type of prestige? The answer can only be that it carries with it a distinction derived from its past, that it suggests a connection between the speaker’s individual choices and preferences and what anyone would choose, between my choice and the choice which the relevant criteria dictate.
It would be equally mistaken however to suppose that the word good could not become detached from the particular criteria which have governed its use, and still remain intelligible. What gives the word good its generality is partly the fact that a connection with choice and preference is present from the outset. To call something good is to say that anyone who wanted something of that kind would be satisfied with this particular specimen. We bring into the picture more than our own individual choice and preference; we point to more than our own individual choice and preference: we point to a norm for choice. And in a society where traditional roles and the corresponding traditional evaluation of behavior have broken down or disappeared, the sequel to unsuccessful attempts to use the word good as a simple expression of choice or preference may well be an attempt to reestablish norms for choice. And there is no reason at all why good should not acquire new criteria of application.
I have tried to delineate in the argument so far an ideal historical sequence. Such a sequence is useful for two different types of reason. It brings out the connection between historical intelligibility and logical relationships. I cannot understand the logical structure of a given philosophical theory, for example, unless I understand the problems to which it is intended to be a solution.
But in a great many instances I cannot hope to understand what those problems are unless I know what problems were posed by the philosophical predecessors of the theory and how the historical context imposes limits upon solutions to their problems. It is always possible and usually useful to abstract both problem and solution, question and answer, from their particular context and to examine matters of logic without too much reference to their actual history. (This is what idealist philosophers, and particularly R. G. Collingwood, sometimes failed to see; but what they succeeded in seeing and saying on this point is more illuminating than most later writing.) But more than this, the concepts which furnish the materials for philosophical investigation are, as we have already noticed, liable to change. Thus what may appear at first misleadingly as two rival elucidations of the same concept, between which we have to choose, may be envisaged more usefully as two successive analyses of a concept in process of transformation, between which there is no question of a choice. Both are needed and so is their interrelationship, so that we may not lose sight both of the continuity and of the change in the concept.
Moreover, to analyze concepts in terms of ideal historical sequences may be useful for another reason. In abstracting certain characteristics of the sequence, and thus lending it an ideal character, we acquire a method for noting similar sequences embedded in quite different historical processes. And in noting similarity we may also note differences. Consider both the resemblances and the differences between what happened to ἀγαθός in Greek usage and what happened to duty in English. Just as ἀγαθός is originally tied to performance of a role, so is duty. We still talk of the duties of a policeman or a probation officer, and in a society where the moral life is exhaustively conceived in terms of role description, the duties of a father or a king may be as vigorously delimited by custom as those are which are now defined by statute. It is when we detach a man from his roles, but still leave him with the concept of “duty,” that the concept is necessarily transformed. This detachment is a consequence of a sufficiently radical change in established social structure and does not have to occur to a whole society all at once in a once-for-all kind of way. It can occur for part of a society, and it will occur in such a way as to be modified by other moral beliefs. So for part of English society in the eighteenth century the concept of duty became generalized in association with the concept of vocation. Originally we have a society of well-defined occupational roles and functions, hierarchically arranged, and to this arrangement there corresponds a belief in different stations in life to which God is pleased to call men. When the occupational roles become more important, the notion of a calling by God, but not to any one particular “station,” remains. The duties which were tied to a particular office are replaced by the duty one owes to God simply as a man. In such a situation the content of duty will be blurred. This kind of situation provides part of the background to the kind of moral dilemma which is examined in some of Jane Austen’s novels. Her characters cannot simply conceive of morality in terms of the adequate fulfillment of a well-established role. Edmund Bertram in Mansfield Park can be open to Mary Crawford’s criticism of his intention to become a clergyman: he can be forced to ask himself whether it will make him more or less of “a man.” Being a landowner and being “in trade” no longer carry a sense of clearly defined status in the heirarchy of duties. That this is so is highlighted by one remarkable exception. The person of the naval officer provides a touchstone of virtue in Jane Austen precisely because of his professionalized sense of duty. And she is able to speak of duty in this context far more clearly than elsewhere because the link between duty and duties has not been broken.
The history of ἀγαθός in Greek and of duty in English (or German) are of course as different as the history of the breakdown of traditional Greek society is from the history of the transformations of preindustrial England. But in both cases we get a move from the well-defined simplicities of the morality of role fulfillment, where we judge a man as farmer, as king, as father, to the point at which evaluation has become detached, both in the vocabulary and in practice, from roles, and we ask not what it is to be good at or for this or that role or skill, but just what it is to be “a good man”; not what it is to do one’s duty as clergyman or landowner, but as “a man.” The notion of norms for man emerges as the natural sequel to this process, and opens new possibilities and new dangers.
At this point, however, does not the argument involve us in apparent paradox? We can understand why Plato and Aristotle (and why, in the later context, Price and Kant) look for norms independent of the structure of this or that particular social framework. But the cost of doing this is to suggest the truth of exactly the kind of relativism which they were seeking to overcome. If the kind of evaluative question we can raise about ourselves and our actions depends upon the kind of social structure of which we are part and the consequent range of possibilities for the descriptions of ourselves and others, does this not entail that there are no evaluative truths about “men,” about human life as such? Are we not doomed to historical and social relativism?
The answer to this is complex. The first part of the answer has already been suggested in the course of discussing Aristotle. It is that there are certain features of human life which are necessarily or almost inevitably the same in all societies, and that, as a consequence of this, there are certain evaluative truths which cannot be escaped. But, put simply like this, this point can be misleading. We cannot, as I have already argued, conceive of a group of beings who would satisfy the minimal conceptual conditions necessary for us to characterize them correctly as a human group where there was not rule-governed behavior, and where the norms which governed that behavior did not entail a norm of truth telling, a norm of ownership and justice and the like. In any human group some notions of truth and justice necessarily find some foothold. Moreover, as I have also argued, in any human group it is almost inconceivable that certain qualities such as friendliness, courage, and truthfulness will not be valued, simply because the range of ends possible for the activities of those who do not value such qualities is far too restricted. But this kind of argument might be quite wrongly held to provide us with a kind of transcendental deduction of norms for all times and all places; it might be held to provide a guide to conduct for men, irrespective of the nature of the society in which they find themselves. Not only is this a mistaken conclusion, but it derives from a misunderstanding of the import of the premises from which it is derived. Just because human society as such either has to have or will usually have certain norms as part of the ineliminable logical framework of its actions and its discourse, so all choices of different evaluative possibilities arise within this framework and within the context of the norms in question. It follows that these norms cannot provide us with reasons for choosing one out of the set of possibilities rather than another. To put this concretely, human society presupposes language; language presupposes rule following; and such rule following presupposes a norm of truth telling. Lying as a form of human action, it is often pointed out, logically presupposes a norm of truth telling. But although the liar therefore vindicates in his practice the existence of the very norm which his practice violates, he thereby shows that the existence of the norm opens up possibilities both of lying and of truth telling; the existence of the norm entails nothing in the way of guidance on any particular occasion of perplexity as to whether we should lie or whether we should tell the truth. And not only different individual choices but very different codes of honesty lie within the range of possibilities open to us. Thus anyone who claims that the elucidation of the norms governing human activity as such provides a guide to how to live is making a fundamental mistake.
It is in outlining the concrete personal and social alternatives in a particular situation and the possibilities of good or evil inherent in them that we in fact frame practical questions and answers. In this task the alleged alternatives of “historical relativism” and “norms for men as such” scarcely arise. For certainly in asking for criteria to govern my choices I am asking for criteria and not for something else; I am asking for guidance of an impersonal kind, not just for me, but for anyone-anyone, that is, in my situation. But the more that I particularize my stiuation the more I ask for guidance for people who belong specifically to my time and place -or to other times and places of a sufficiently and relevantly similar sort. I am always going to be faced with two dangers. If I abstract sufficiently, I shall be able to characterize my situation in terms quite apart from any specific time and place, but by so doing I shall not solve my problem but relocate it. For the highly general form of problem and solution then has to be translated back into concrete terms, and the real problem becomes how to do this. If I do not abstract sufficiently, I shall always be in danger of making myself the victim of what is taken for granted in a particular situation. I shall be in danger of presenting what is merely the outlook of one social group or part of the conceptual framework for men as such.
Both Plato and Aristotle suppose that from the elucidation of the necessary conceptual framework for human life they can draw practical guidance; and this mistake is both camouflaged and reinforced by their adapting forms of description at once used in and well-suited to characterizing the social life of the Greek πόλις to serve as forms of description for human life as such. This is not only a weakness. Some later writers on moral philosophy have supposed that the problems can be posed in a vocabulary which is somehow independent of any social structure. This supposition is one of the roots of the belief that there are two distinct spheres of life, one for “morals,” the other for “politics.” But, in fact, every set of moral evaluations involves either neutrality toward or assent to or dissent from the social and political structure within which it is made. And insofar as dissent is concerned, the moral evaluations will involve some degree of commitment to an alternative. What is striking about Plato and Aristotle is the unity of morals and politics in their writings. Yet this very unity in the end betrays their ideals.
Both Plato and Aristotle take for granted, naturally enough, the social structure of the πόλις, with the slaves excluded from the political structure, the artisans and farmers coming out at the bottom, a richer class above them, and some kind of elite ruling. Because the questions they pose, and sometimes the concepts they employ, presuppose the πόλις and its social unity, neither of them faces up to the actual decline of the πόλις. Because they are spokesmen for its unity, they ignore or dislike the heterogeneity of Greek society. The concept of a common interest is taken for granted. The conservatism of Aristotle is of course quite different from that of Plato. Plato’s idealization of a πόλις utterly different from that of fourth-century reality means that politics becomes a hit-or-miss affair of the philosopher king happening to turn up at the right time and the right place. Those modern critics of Plato who have castigated him as a fascist have missed the point very badly. For the essence of fascism is that it glorifies and upholds some existing ruling class: the essence of political Platonism was that it excoriated every actual political possibility. Plato’s own political failure at Syracuse, where successive visits met a blank wall in political reality, was grounded not merely in the conditions of Syracuse in particular or in the city-state in general, but in Plato’s own doctrine. Plato may be conceded the title of either reactionary or conservative, but if all reactionaries were Platonists, revolutionaries would have an easy time.
With Aristotle it is different. We are much closer to actual states and constitutions in Aristotle’s empirically based Politics. But in two respects Aristotle faces up to the realities of the πόλις even less than Plato does. The , the mass of ordinary people, appear in Plato as governed by desires which have no room for expression in the just state; in the Republic the desires are to be repressed, in the Laws they are to be remolded. But throughout Plato the natural clash of desires between rulers and ruled figures prominently in the political picture. Both Plato and Aristotle see the desires of the rulers as the characteristic of “man,” those of the ruled as nearer the merely animal. But in the Nicomachean Ethics the baser passions, characteristic of the ruled, appear merely as sources of error and distraction. There is nothing of Plato’s at times nearly hysterical picture of what he takes to be the anarchy of desire. Since all norms belong within the just state, and desire of an untrammeled kind has no place in it, this picture of desire as anarchic is inevitable. But by recognizing that there are desires which cannot be legitimated and allowed expression within his form of ideal state, Plato also recognizes implicitly that those whose desires they are would find in them a criterion for criticizing his state and life in it as “less profitable” than the pursuit of what his state would characterize as injustice. Plato is at times a candid partisan of a ruling class, even if only of an imaginary ruling class.
Aristotle is in this respect uncandid. The Aristotelian ideal of the leisured and perfected life of abstract contemplation is only accessible to an elite; and it presupposes a class structure which excluded the mass of ordinary men both from political power and from the moral idea. But every desire is allowed expression in a form that will either satisfy it or purge it. This explains the difference between Plato and Aristotle on the subject of tragedy. The values of Greek tragedy express the conflicts of Greek society as much as the values of Plato and Aristotle express or attempt to depict Greek society as a unified structure. In the Oresteia tribal and urban values conflict; in Antigone those of the family and those of the state; in The Bacchae those of reason with those of the passions. They present to the mass audience of the πόλις rival allegiances for their desires in an aesthetic mode calculated to arouse passion. Plato rightly sees that they do not present the kind of consistent, single moral ideal he believes in and that they run contrary to his attempt to suppress the desires of the mass audience. Hence his consistent advocacy of censorship and repression. But Aristotle sees that the aesthetic evocation of pity and terror may purge us of them. Far from providing us with motives for action, the drama may evacuate us of otherwise dangerous desires and emotions, and in so doing, it will stabilize the existing social order. Hence Aristotle has none of Plato’s enthusiasm for censorship.
In fact, Aristotle is much more of a quietist in relation to political activity. Provided only that there is room for the contemplative elite, the Nicomachean Ethics does not provide for a condemnation or an endorsement of any social structure; and the Politics uses criteria of stability to judge between types of state which have only this negative connection-of making room for an elite-with the arguments of the Ethics. In fact, by his own practice as the tutor of the young Alexander, and by his advocacy of the life of contemplation, Aristotle, as Kelsen pointed out, sided with the powers which were about to destroy the πόλις as a political entity. For the exaltation of the contemplative life is an exaltation of it as a form of life for those men who have hitherto composed the political elite. It provides a rationale for their withdrawal to the status of citizen, “good citizens” in Aristotle’s sense, but not rulers. And this is just what the absolutism of Macedon, the first of the new large-scale states, required the rulers of what had hitherto been city-states to become. As Kelsen puts it, “the glorification of the contemplative life, which has renounced all activity and more especially all political activity, has at all times constituted a typical element of the political morality set up by the ideologies of absolute monarchy. For the essential tendency of this form of state consists in excluding the subjects from all share in public affairs.”21
The facts of the decline of the πόλις and the rise of the large-scale state have immensely more important consequences for the history of moral philosophy than any gravitational pull that they may have exerted upon Aristotle’s analyses. The milieu of the moral life is transformed; it now becomes a matter not of the evaluations of men living in the forms of immediate community in which the interrelated character of moral and political evaluation is a matter of daily experience, but of the evaluations of men often governed from far off, living private lives in communities which are politically powerless. In Greek society the focus of the moral life was the city-state; in the Hellenistic kingdoms and the Roman empire the sharp antithesis between the individual and the state is inescapable. The question now is not, In what forms of social life can justice express itself? or, What virtues have to be practiced to produce a communal life in which certain ends can be accepted and achieved? but, What must I do to be happy? or, What goods can I achieve as a private person? The human situation is such that the individual finds his moral environment in his place in the universe rather than in any social or political framework. It is salutary to observe that in many ways the universe is a more parochial and narrow moral environment than Athens was. The reason for this is very simple. The individual who is situated in a well-organized and complex community, and who cannot but think of himself in terms of the life of that community, will have a rich stock of descriptions available to characterize himself, his wants, and his deprivations. The individual who asks, What do I desire, as a man, apart from all social ties, in the frame of the universe? is necessarily working with a meager stock of description, with an impoverished view of his own nature, for he has had to strip away from himself all the attributes that belong to his social existence. Consider in this light the doctrines of Stoicism and Epicureanism.
The remote ancestor of both is Socrates, the Socrates who is essentially the critic, the outsider, the private foe of public confusions and hypocrisies. Plato sees that if one asks seriously for answers to the Socratic questions, one necessarily becomes the partisan of one sort of social order against others, and in so doing, one has to abandon the role of the merely private person and critic. But among Socrates’ disciples there were some who retained this mode, who stylized the Socratic way of life and drew their moral code from this style of life rather than from reflection on the character of definition. Independence and self-sufficiency become for them the supreme values; the only way to avoid injury from changing circumstance is to make oneself radically independent of circumstance. Antisthenes, the logician, rejects as goods not merely wealth and honors, but anything that might provide the satisfaction for a desire. Virtue consists in the absence of desire and is sufficient by itself for happiness. The man who is virtuous in the sense of desiring nothing has nothing of which to fear the loss; he is able to bear even slavery without injury. Antisthenes sees conventional politics and conventional religion only as sources of illusion. Not the state, but the universe is the habitation of the virtuous man; not the local gods of the state, but the one good is his god, and the only service of god is the practice of virtue. What independence of this sort could mean is shown in Diogenes’ life in his tub and in his reply to Alexander’s question of whether there was anything that Alexander could do for him: “Yes, get out of the light.” It is Diogenes’ expressed wish to live with the simplicity of an animal and his chosen self-title, “The dog” (ὁ κύων), that won for moralists of this school the title “Cynics.” (The link with the English word cynicism lies in the Cynic claim to see through all conventional values.)
Aristippus of Cyrene begins from the assumption of the identity of the pursuit of virtue and the pursuit of εὐδαιμονία. He identifies εὐδαιμονία with pleasure; but holds that excess of pleasure leads to pain and that the limitation of desires is a condition of their satisfaction. Among Aristippus’ disciples, called Cyrenaics, perhaps the most significant figure was Hegesias, who stressed this latter point to the extent of holding that the absence of pain rather than the promotion of pleasure is the aim of life. What is more, he believed that abstention from actual pleasure was a condition of such absence. When Hegesias lectured atAlexandria, it is said that the effect was such on his hearers that many of them committed suicide, and in the end he was not allowed to give lectures.
Even in the thought of the Cynics and Cyrenaics we can discern a tendency which will be much more strongly exemplified in Stoics and Epicureans. For both Plato and Aristotle, although the relation of virtue to happiness may constitute a problem, that there is a connection between them waiting to be elucidated is a fundamental assumption. Unless virtue somehow leads to happiness, it lacks a τέλος, it becomes pointless; unless happiness is somehow bound up with the practice of virtue, it cannot be happiness for the kind of beings men are, it cannot constitute a satisfaction for a moralized human nature. Happiness and virtue are neither simply identical nor utterly independent of each other. But in the case of both Cynics and Cyrenaics we see the tendency to reduce one to the other, and to in fact operate either with the concept of virtue alone or with that of happiness alone. This separation of virtue and happiness is interestingly accompanied by a large stress upon self-sufficiency, upon avoiding disappointment rather than seeking for positive goods and gratifications, upon independence from contingent bad fortune, and this stress perhaps provides the very clue which we need to understand their separation. The sense one gets in reading the records of post-Socratic philosophy which survive in writers such as Diogenes Laërtius and Cicero is of a disintegrated social world in which there are more puzzled rulers than ever before, in which the lot of the slaves and the propertyless is very much what it was, but in which for many more middle-class people insecurity and an absence of hope are central features of life.
This suggests interestingly that the possibilities of connecting virtue and happiness are dependent not solely upon the features of two concepts which remain unchanged and hence have an unchanging relation, but upon the forms of social life in terms of which these concepts are understood. Let me suggest two extreme models. The first is of a form of community in which the rules which constitute social life and make it possible and the ends which members of the community in question pursue are such that it is relatively easy to both abide by the rules andachieve the ends. A well-integrated traditional form of society will answer to this description. To achieve the personal ideals of the Homeric hero or the feudal knight or the contemplative and to follow the social rules (which themselves invoke a respect for rank and religion) cannot involve fundamental conflict. At the other end of the scale, we might cite as an example the kind of society which still sustains traditional rules of honesty and fairness, but into which the competitive and acquisitive ideals of capitalism have been introduced, so that virtue and success are not easily brought together. Or there may well be intermediate types of society in which for some groups only is it true that their ends and the rules of the society are discrepant. From the vantage point of each of the different kinds of society the relation between virtue and happiness will look very different. At the one extreme we shall find virtue and happiness regarded as so intimately related that the one is at least a partial means to or even constitutive of the latter. At the other extreme we shall find a total divorce, accompanied by injunctions by the would-be moralists to regard virtue rather than happiness, and by the would-be realists (illuminatingly called “cynics” by the moralists) to regard happiness rather than virtue. Even though both words remain, the one will come to be defined in terms of the other. But inevitably in such a situation both the concept of virtue and the concept of happiness will become impoverished and will lose their point to a certain extent. To understand this situation we must look at the relationship between rules and ends, and to do this we must first make clear the distinction between them.
There are rules without which human life recognizable as such could not exist at all, and there are other rules without which it could not be carried on in even a minimally civilized form. These are the rules connected with truth telling, promise keeping, and elementary fairness. Without them there would not be an arena in which distinctively human ends could be pursued, but these rules by themselves in no way provide us with ends. They tell us how to behave in the sense of telling us what not to do, but they provide us with no positive aims. They provide norms to which any action we may perform is required to conform, but they do not tell us which actions to perform. Which actions we shouldperform depends upon what ends we pursue, what our goods are. In general happiness is a rubric relating to ends, virtue one dominating rules. It would be a mistake to suppose that in identifying this distinction between rules and ends we are also demarcating the public and the private domains in morality. For while it is true that ends may admit of private choices in a way that rules do not, it is also true that there are societies in which there are publicly established and agreed or imposed ends, as well as societies which leave alternative ends open to a great degree to individual preference. Moreover, there may be private innovations in the realm of rules as well as in that of ends. What does remain true, however, is that the dissociation of rules and ends will inevitably have repercussions on the relationship between private and public life. For where the observance of rules has no or relatively little connection at all with the achievement of ends, the observance of rules will become either pointless or an end in itself. If it becomes an end in itself, then the observance of rules may become a private ideal for the individual as well as a requirement of social morality. If the achievement of ends is in the same type of situation, as it will be, relatively independent of the observance of rules, then ends become dissociated from the requirements of the public domain. They provide other and rival private ideals. It will be natural in this situation to conceive of the pursuit of pleasure and the pursuit of virtue as mutually exclusive alternatives. Moreover, in each case, long-term projects, which tend to depend upon the possibility of relying on a widespread public congruence of rules and ends, will appear far less viable than short-term. Moral advice will most naturally be either of the “Gather ye rosebuds while ye may” kind or of the “Do what is right regardless of the consequences” kind. “Fiat justitia, ruat coelum” is a slogan that is pointless rhetoric except when it seems quite possible that the heavens will crumble. We can see these alternatives embodied in private moralities by the Cynics and the Cyrenaics. They rise to the level of universal codes in Stoicism and Epicureanism.
For the successive founders and refounders of Stoicism, Zeno, Cleanthes, and Chrysippus, morals become unintelligible apart from cosmology. The universe is at once material and divine. Theprimary material of the universe, fire, is transmuted into various physical states by the activity of a universal rational principle, the Logos, which is the deity. In the transmutation of the universe a regular cycle recurs, returning again and again to a cosmic conflagration in which the original fire brings to an end one period and begins another. Each of these cyclical periods is identical, and every event in the universe therefore recurs indefinitely. Since man is an integral part of the universe, this eternal recurrence is also true of human history. Indefinitely often in the past and indefinitely often in the future I have written and shall write these words, and you have read and will read them, just as you do at this present moment.
Since human nature is part of cosmic nature, the law which governs the cosmos, that of the divine Logos, provides the law to which human action ought to be conformed. At once an obvious question arises. Since human life proceeds eternally through an eternally predetermined cycle, how can human beings fail to conform to the cosmic law? What alternatives have they? The Stoic answer is that men as rational beings can become conscious of the laws to which they necessarily conform, and that virtue consists in conscious assent to, vice in dissent from, the inevitable order of things. What this answer means can be better understood by considering the Stoic answer to the problem of evil.
Since everything is formed by the action of the divine principle, and that principle is entirely and unquestionably good, it follows that no evil can occur in the world. But evil does occur. How so? The Stoic rejoinder is, in effect, that evil does not really occur. A variety of arguments, which later on are to reappear in Christian theology, take the stage for the first time in Stoic costume. Chrysippus argued that of a pair of contraries, neither could be conceived to exist without the other, so that good and evil each require the existence of the other. Evil, being therefore a necessary condition for the occurrence of good, is in terms of a larger scheme not really evil at all. From this, Chrysippus deduces the impossibility of pleasure without pain and of virtue without vice. Courage could not occur did not cowardice; justice, did not injustice. Indeed we call actions cowardly or unjust not with reference to the act itself, but with reference to the agent’sintention. The same action, in the sense of the same physical behavior, can be cowardly if done with one intention (the agent aims only to save himself) and courageous if done with another (the agent aims to prevent a struggle, even at the cost of his own reputation for courage).
We can now understand why the Stoics think it possible to combine determinism with a belief that men can either assent to or dissent from the divine law. What is determined is the entire physical world, including human beings insofar as they are part of that world; what apparently escapes determination is human assent or dissent to the course of things expressed in the form of intention. Even if I dissent from and rebel against the predetermined course of nature, my physical behavior will still conform to it. “Ducunt volentem fata, nolentem trahunt,” wrote Seneca later on.
In what form does the divine law to which my assent is invited present itself? As the law of nature and of reason. Nature now becomes a term quite other than what it was in either Plato or Aristotle. It refers to the cosmic status of the moral law; as such, it still contrasts with convention in the sense of what is merely established for local observance. But somehow the moral law and the physical universe now share a source, a prefiguring again of Christianity. What nature and reason invite us to is the observance of the four traditional virtues, prudence, courage, temperance, and justice. But one cannot, for the Stoics, possess one of these without possessing all. Virtue is single and indivisible. One cannot possess it in part; either one is virtuous, or one is not. There is a single dividing line among men. Above all, virtue is to be sought only for its own sake. “Virtue,” as Diogenes Laërtius regards it, “is a rational disposition, to be desired in and for itself and not for the sake of any hope, fear, or ulterior motive.”22 Pleasure, by contrast, is not to be sought at all. Cleanthes thought that it was positively to be shunned; most of the Stoics that it was merely to be disregarded. Desire, hope and fear, pleasure and pain are against reason and nature; one should cultivate a passionless absence of desire and disregard of pleasure and pain. This the Stoics called apathy.
What then does one do? How does one actually behave? Onedisregards all attractions of external goods; one is therefore not exposed to the pain of their loss. Peace of mind is thereby secured. (Hence the later use of the adjective stoical.) In the world at large, one disregards those differences between men which are merely a consequence of externals. There is one divine universe, one rational human nature, and therefore one appropriate attitude to all men. The Stoic is a citizen of the κόσμος, not of the πόλις It we turn not to Epicureanism expecting a sharp contrast, we find that what is striking about Epicureanism is in the end not the contrast with, but the resemblance to Stoicism. Superficially the differences are what stand out. Morality exists in a universe which is alien to it, and not, as with the Stoics, in a universe of which it is the highest expression. The atomism which Epicurus inherits from Democritus and bequeaths to Lucretius is a theory of blind physical determination. The moral consequences of atomism are negative; the gods do not control or interest themselves in human life. They dwell apart and indifferent, and natural phenomena have physical, not theological explanations. Plagues are not punishments, and thunderbolts are not warnings. Morality is concerned with the pursuit of pleasure, and not, as with the Stoics, with the pursuit of virtue independently of pleasure. Indeed, for Epicurus, virtue is simply the art of pleasure. But Epicurus then proceeds to argue that many pleasures, if heedlessly pursued, bring great pains in their wake, while some pains are worth tolerating for the ensuing or accompanying pleasures. He argues further, as the Cynics did, that the absence of pain is a greater good than positive pleasures; he argues, moreover, that a moderation in external goods is the only guarantee of not being pained by their loss; and he argues finally that freedom from intense desire is a condition of pleasure. All the conventional virtues are reinstated as means to pleasure and the gulf between Stoic apathy and Epicurean tranquillity (ἀταραƷιἀ), verbally wide, is practically narrow. Epicurus’ practical atheism makes him less pompous than the Stoics, and his high valuation of friendship makes him attractive as a person, but the regard for a quiet life, and detachment of the individual from the Platonic-Aristotelian morality of social life is as complete as it is in the Stoics.
Both Epicureanism and Stoicism are convenient and consoling doctrines for private citizens of the large impersonal kingdoms and empires of the Hellenistic and Roman worlds. Stoicism provides a better rationale for participation in public life, Epicureanism for withdrawal from it. Both place the individual in the context of a cosmos, not of a local community. Both have a function in a world in which pain is to be avoided rather than pleasure sought. In the Roman world especially, each has a function which is left unfulfilled by Roman religion. Roman religion is essentially an integrative cult in which the gods of the hearth, the gods of the formerly independent nations, and the gods of the empire express by their unity the single hierarchy of familial and imperial deities. The earliest Roman rulers speak from within their roles as fathers and consuls; if they use a religion to manipulate the plebeians, it is at least a religion which they share. But relatively early this ceases to be so. Polybius could write that “it is the very thing which among other peoples is an object of reproach, I mean superstition, which maintains the cohesion of the Roman state. These matters are clothed in such pomp and introduced to such an extent into their public and private life that nothing could exceed it, a fact which will surprise many. My own view at least is that they have adopted this course for the sake of the common people. It is a course which perhaps would not have been necessary had it been possible to form a state composed of wise men, but as every multitude is fickle, full of lawless desires, unreasoned passion, and violent anger, the multitude must be held in by invisible terrors and suchlike pageantry.”23
Where religion is thus manipulative, the members of the middle and upper classes become unable to share the religion which they use for political purposes. They need beliefs which are rational by their own standards and will justify what Romanitas itself once justified or which will justify the withdrawal from public duty. These needs were admirably met by Stoicism and Epicureanism. Seneca and Marcus Aurelius exemplify the public side of Stoicism; Lucretius the liberating qualities of Epicureanism.
The doctrines of the Roman upper classes are, however, vulnerable in one crucial respect. The doctrines of apathy and ataraxia are useless as advice to those who already are propertyless and in no position to become hedonists. Exposed to poverty, disease, death, and to the will of those who are their rulers and often enough their owners, they still question how they are to live and what virtue and what happiness might be in their case. For some of these the mystery religions provided an answer. For even more an answer was to be given with the coming of Christianity.