The Moral Attractions of Relativism:

A Critique

Introduction

Historicism holds that moral reasoning consists in the interpretation and elaboration of the moral beliefs by which a group of people live.[1] For the historicist, there are no reasons for choosing between the moral traditions of different groups of people. Nor is such choice usually necessary. Rather, moral reasoning operates and can only operate from within some set of moral concepts that constitute a tradition. Since drastically new moral beliefs may be advocated by some members of a group of people, the possibility of substantial, indeed revolutionary, moral innovation does exist. But where these new moral views are radically different from the ones they are meant to replace, there will be no common framework within which to evaluate both the old and new moral conceptions. As a result, it is possible that no rational basis for choosing the new moral views over old ones can be adduced.

From Callicles to Nietzsche, historicism has been presented as, and widely considered to be, a doctrine that undermines moral thought. Historicists exposed the pretensions of moral philosophy and the uncertain status of our moral beliefs. They were often the enemies of conventional political views. Whether historicists intended this result or not, their doctrines seemed to weaken the commitment that people had to their own-or indeed any-moral beliefs. Those who hoped to defend particular moral beliefs and the very practice of moral reasoning looked at historicism as a challenge to be overcome. The historicist stood to the moral philosopher as the skeptic stood to the epistemologist

In recent years, however, historicism has been defended by those eager to vindicate moral reasoning. And, most strikingly, political philosophers such as Michael Walzer (1983, 1987), Richard Rorty (1982), Bernard Williams (1972, 1981, 1985), and Michael Sandel (1982) have argued that historicism, or a view much like it, is a morally attractive doctrine. Rather than challenging moral reasoning and conventional ideals, these historicists defend them. And where their moral conclusions differ from conventional conceptions, these theorists hold that historicism provides support for attractive moral beliefs. Or, perhaps more commonly, they argue that historicism undermines dangerous moral views. Some of these theorists-most notably Michael Walzer-argued for the moral attractiveness of not just historicism but relativism as well. Walzer has held that we should recognize that what is just and unjust or good and bad is relative to the moral traditions of people in different times and places. What is just here, may be unjust somewhere else.

That historicism and relativism are morally attractive doctrines sounds paradoxical. What, we might wonder, could be the basis of this moral evaluation? The historicist or relativist can consistently answer, however, that our judgments of historicism and relativism are themselves based upon our own moral beliefs. Historicism and relativism are attractive because they provide support for moral beliefs we hold dear.

My aim in this paper is to evaluate the claims that historicism and relativism are morally attractive. My argument is directed at both claims. I will focus primarily on the more striking claims of relativism however, occasionally pointing out the implications of my argument for a moral philosophy that is historicist but not relativist in nature.

There are two ways one could argue against the moral attractiveness of historicism and relativism. One might demonstrate that some form of non-historicist and non-relativism (or, as I shall say, rationalist and universalist) moral reasoning is possible and then continue to show that there is good reason to hold moral beliefs contrary to those held by historicists and relativists. I try to make this first argument in other work but will not pursue this tack here (1989, chs. 2, 5 - 7). A second way in which to undermine the attractiveness of moral relativism is to show that, given our own moral beliefs, the consequences of historicism and relativism are not salutary. This is the approach I will pursue. In doing so, I will, to some extent, adopt the historicist perspective or, more accurately, that of some contemporary historicists who wish to defend liberal political principles. I, like these liberal historicists and relativists, hope to defend moral principles that undergird civil liberty and democracy. From this shared point of view, then, our evaluations of historicism and relativism turns on the following questions: Are these principles safer with an historicist or relativist defense or would we be better off with a rationalist argument for them? Can one provide good arguments for these principles if one is a historicist or relativist? And, what is not exactly the same thing, are people more or less likely to believe and act on these principles if they also are historicists or relativists? I hope to show that the answers that historicists and relativists give to these questions are not satisfactory.[2]

In considering the virtues and vices of historicism and relativism, I will begin, in the next section, by setting out in more detail what I mean by historicism and relativism and how I think these two doctrines are related. Most of the rest of the paper is devoted to examining the views of Michael Walzer and, more briefly, Richard Rorty. I focus on Walzer's arguments because his great achievement has been to do more than provide a programmatic account and defense of moral relativism. Spheres of Justice presents a comprehensive conception of justice from a point of view that is largely-though, as we shall see, not entirely-historicist and relativist in nature. Examining Walzer's arguments about the distribution of some social goods will allow us to consider some limits of relativist moral philosophy in concrete detail. In addition, although other defenders of historicism, such as Rorty, have been more concerned with the critique of rationalist moral theories or the philosophical positions on which they rest, Walzer's approach is distinct. In both Spheres of Justice and his most meta-ethically self-conscious work, Interpretation and Social Criticism, Walzer has explicitly argued for the moral attractiveness of relativism and historicism. Relativism, for Walzer is not a doctrine that we must reluctantly accept but comes close to being a moral imperative itself. And historicism is, for Walzer, the most plausible approach to moral reasoning. Thus in considering his argument in some detail, we can grapple with the central question I wish to raise: should we be satisfied with relativism and historicism?[3]

Historicism and relativism are contested terms which are used in a variety of ways. And the relationship between these different doctrines is problematic. To avoid confusion, let me make some initial remarks about how I will use these terms and how historicism and relativism, as I understand them, can be related.

Historicism holds that moral reasoning must take place within the confines of a particular moral tradition. It is a conception of how moral reasoning must be conducted. As such, it is opposed to what I will call a robust or a rationalist conception of political and moral philosophy, which holds that, in one way or another, we can find rational grounds for evaluating different traditions and, in this way, transcend our own political and moral views. Moral relativism, as I will use the term, is not a view of the nature of moral reasoning but, rather, a substantive, if quite general, moral doctrine. It holds that what is just or unjust, good or bad, is relative to the views found in a particular polity and society. Thus for relativists, what is just here, in our polity and society, might be unjust somewhere else. Relativism is opposed to moral universalism, the claim that there is one basic, universal standard of morality that should ultimately guide everyone, in all times and places. Universalism need not hold that precisely the same actions or policies will be just or unjust everywhere. A moral universalist can recognize that universal moral principles must be applied to particular cases and that, in different times and places, these universal moral principles will lead us to different conclusions about particular actions or policies. But universalists do hold that the same moral principles underlie these different particular judgments. Relativists, on the other hand, hold that basic moral principles vary from one time and place to another and thus, as a result, what is moral or not varies as well. Actions and policies must, for relativists, be evaluated in terms of the moral principles found in a polity and society.

Relativists usually defend their views, in part, by supporting the historicist claim that most, if not all, moral reasoning about our actions and political and social institutions can only take place through the elaboration and refinement of our moral tradition. Relativist philosophers sometimes argue that, given the variety of moral views found in different polities and societies, we can have no grounds on which to criticize or condemn the members of another society. That we should evaluate others in their terms does not, however, follow from historicism. Bernard Williams has shown that the argument from historicism to relativism is mistaken (1972, 20-26). The notion that we have no business criticizing anyone else's moral views-that is, moral relativism-is itself a substantive moral claim. And to take that claim to be a moral requirement binding on everyone, whatever their own moral traditions, surely contradicts the proposition that all moral beliefs are justified only in terms of some tradition. Thus one cannot sensibly derive the moral imperative for relativism from the existence of different moralities alone.

If relativism does not follow from the historicist account of moral reasoning, what substantive moral principles can be derived from this approach? The most plausible answer has been given by Richard Rorty, who says that historicists should be ethnocentric. Rorty is an historicist about all forms of thought and belief. As part of a more general critique of philosophical rationalism, Rorty has argued that both our scientific and moral ideas are only defensible in our own terms (1979, 1982). Against foundationalist conceptions of the nature of knowledge, Rorty has held that there is no "neutral matrix" in which all claims to knowledge can be grounded. In particular, we have no unmediated perception of nature with which we can test our scientific theories. Instead, we judge our particular claims to knowledge by reference to the beliefs we already hold. These beliefs are an inseparable amalgam of both observational and theoretical notions. Similarly, in morality, claims about the rightness or goodness of some action, policy and institution can only be made in terms of the moral discourse in which our live are situated.

Rorty acknowledges that the terms in which members of different societies and cultures advance claims to scientific or moral knowledge are diverse and incompatible. He concludes that this situation should not lead us to adopt the kind of relativism that says that "anything goes." Rather, it should lead us to be ethnocentric. Rorty rejects the notion that one view as good as another in either science or morality. We can always evaluate different claims to knowledge in terms of our own beliefs or, to use the pragmatist's phrase, ways of coping with the world. And this is the only way in which scientific or moral claims can be adjudicated. For, given the ways in which we live, other ways of coping with the world are not real options for us. Rorty, like Bernard Williams, argues that we do not need reasons to prefer our scientific or moral views beyond that they are our own (Williams, 1981; Rorty, 1985).

Rorty does not think that we should dogmatically defend all or even most of our scientific and moral views. He does argue, however, that the only grounds upon which we can challenge some of our beliefs is provided by our other beliefs. Philosophers since Socrates have held that transcendence of conventional beliefs and values requires breaking with or stepping back from what we share with the members of our society. Rorty rejects this conception entirely. His pragmatic understanding of the nature of knowledge calls for solidarity with our fellow members of society rather than transcendence of them.

So, for Rorty, the acceptance of historicism, by itself, does not and cannot lead us to challenge, revise or reject our own moral views. Historicism leaves us with the moral principles and claims we already accept. Indeed, by denying that there can be any rational grounds for rejecting what we already believe, historicism provides us with a good defense of our own views.

Rorty's account of the implications of historicism is quite right, as far as it goes. He is right to hold that the precise moral implications of historicism depend upon the content of our moral traditions and beliefs. Now it would seem that if historicism should lead us to be ethnocentric, then historicists should not be relativists. In many cases, this will be true. If our moral principles and beliefs are universalist in nature, if we take them to apply to everyone everywhere, then historicism leads not to relativism but to a universalist morality. The moral stance of relativism can, however, be defended in terms of an historicist view of moral reasoning provided that our own political and moral ideas commit us to the moral doctrine of relativism. That is to say, this historicist form of relativism holds that it is central to our moral tradition to think that that what is just and unjust, or good and bad is relative to the views found in a particular polity and society. This defense of moral relativism rests not on some rationalist moral claim but on our own moral views. In this case, one of our own moral views, our relativism, lead us to recognize the our other moral views have only a limited application. On this approach, our own traditions views of such issues as civil liberty and distributive justice tells us what is right and wrong here, but not what is right and wrong everywhere.

So an historicist account of moral reasoning can lead to both universalist and a relativist moral claims. Similarly, a rationalist account of moral reasoning can lead to both universalism and relativism. Typically, rationalists are universalists. They hold that reason leads us to accept certain moral principles and that these principles are binding on everyone. But a rationalist can claim that the highest moral principle is relativism. We might call this the principle of moral self-determination. On this principle, even if the members of some society were completely wrong about justice or the human good, they might still have the moral right to political and social institutions that are in accord with their (mistaken) beliefs. Or we could go further and apply the principle of self-determination domestically. We could argue that it would be wrong for some of us to act on our rationalist moral beliefs unless and until they become the conventional wisdom of our society.[4] Thus, applied domestically, the principle of self-determination aims at limiting the political authority of any rationalist moral reasoning. There is nothing paradoxical about these arguments. They parallel an argument long familiar to defenders of democracy, the so-called "paradox of democracy (Wollheim, 1962)." This argument that holds that people have the right to govern themselves though they make morally wrong choices in doing so.

Thus, a rationalist view of moral reasoning can lead to moral relativism. Moreover, the rationalist version of moral relativism is provides a stronger defense of relativism than that available from an historicist account of moral reasoning. As historicists who believe in moral relativism, we have grounds for respecting the right of other people to live in accord with their own moral views. But we have no argument to convince people in other moral traditions to accept this relativist view. However, if we have a rationalist defense of relativism, then we can argue that everyone should obey the principles of moral self-determination and moral relativism regardless of whether these principles are their own.

So the disputes between rationalists and historicists, on the one hand, and universalists and relativists, on the other, are largely independent of one another. Rationalists can be universalists or relativists. And so can historicists. I have tried to summarize these various distinctions in table 1. (This table also includes a further distinction between generalism and particularism which I discuss below.) With the distinctions between historicism and rationalism and universalism and relativism in mind, let us turn to Michael Walzer's arguments for the moral attraction of relativism.

The relativist conception of political and moral philosophy is attractive to Walzer because he believes it rules out philosophical arguments that could justify one or another form of tyranny. The moral advantages of relativism flow from the kinds of moral arguments they exclude. That one can rationally justify some moral principles in terms that transcends the conventional beliefs and values of one's own or any other society is, for Walzer, potentially dangerous. For any such argument can justify the domination of those who do not have this knowledge by those who do. Someone who, if only in thought, stands outside his society and claims moral insight unavailable from within it can legitimate the rule of other outsiders.

According to Walzer, the central moral implication of relativism, and the source of it's attractiveness, is the view that "a given society is just if its substantive life is lived . . .in a way faithful to the shared understandings of the members (1987, 313)." Like all historicists, Walzer argues that moral judgments of our actions or political and social institutions and policies must begin with our own fundamental moral conceptions. As a relativist, Walzer further claims that our judgment of actions and institutions in another society must be based upon the moral beliefs of those who we wish to evaluate. The dominant political implication of this form of relativism, then, is that people should live their lives in accord with their own political and moral beliefs and ideals. The denial of self-determination, whether by foreign aggressors or domestic usurpers, is the essence of tyranny.

To make self-determination the starting point of political and moral philosophy is not, however, to favor any particular form of political and social life. Where the moral beliefs of men and women assume the equality of man, then self-determination is usually to be exercised through institutions that allow all human beings some say in their political and social arrangements. Where the moral beliefs of the members of a society have another starting point, hierarchical and inegalitarian political and social institutions might be justified. Relativism does not lead to the conclusion that whatever political and social institutions exist in some society are just. For these institutions might be at variance with the moral beliefs of the members of that society.[5] Yet, one can only assess the justice of these institutions in terms of the moral conceptions found in the society in question.

It is important to note that Walzer is concerned with a number of different kinds of tyranny and his arguments against these tyrannies are quite different. For Walzer, universalist moral philosophy can justify both foreign and domestic forms of tyranny. First, it can lead us to criticize the institutions and practices of another society. In different works Walzer has warned against the "philosophical arrogance" of such judgment (1981, 380). Far worse than the philosophical arrogance that condemns, say, the unjust policies of ancient Babylonians, are the contemporary political practices it justifies: imperialism and (unjustified) foreign intervention. Here Walzer's defense of the principle of self-determination, is linked to his criticism of foreign military intervention in Just and Unjust Wars (1977, 86-108). If we accept that moral reasoning can operate only within some tradition of moral discourse and that different societies may legitimately have different moral traditions, we will not be so quick to engage in foreign adventures that aim to-or at least are justified by our claim to-defend and apply our own moral beliefs.

The domestic tyrannies which concern Walzer are more diverse. Mild forms of tyranny can result from liberal political and moral thought. For many liberals, the ends of politics are not discovered through actual political life but are the object of philosophical discovery or construction. Such a conception of political theory, however, is not entirely compatible with a democratic society. Liberal principles, of both the deontological and utilitarian variety, are not based upon and may even override the particular ends and moral claims of men and women in any one society. Thus political action to realize these principles is likely to lead to conflict with the actual moral beliefs and ideals of men and women. Moreover, the effort to enforce these principles can interfere with the ability of men and women to take part in the democratic life most contemporary liberals otherwise defend.[6] For these reasons, Walzer has persuasively argued that when judges in the United States seek to enforce deontological rights-beyond those which make for a democratic community-democracy itself is compromised.[7] The utilitarian penchant to vest power in bureaucracies can lead to similar results. When bureaucrats are delegated broad powers to secure the greatest happiness of the greatest number, the content, if not the form, of democratic rule is eviscerated and the particular moral beliefs of men and women are ignored.

The more extreme form of tyranny that concerns Walzer is that created by revolutionary elites acting to realize some transcendent moral or political truth. Leninist or fascist totalitarians exemplify this danger of non-relativist morality when they run roughshod over the moral beliefs-not to mention the lives-of millions of people.

To undermine the moral attractiveness of relativism, I will first have to try to minimize the fear of moral philosophy that is both rationalist and universalist. To understand this fear, however, we have to grasp a second strand of Spheres of Justice. Walzer combines two arguments in his attempt to undermine the various forms of philosophical arrogance. As we have seen, he argues for a relativist conception of justice in which what is just depends upon the moral understandings of the members of any one society. However, he also presents a particularist conception of justice. Moral arguments about justice, on this second view, should not rely upon abstract and general moral principles. Instead of giving us abstract and general theories of how to distribute abstract and general goods, the moral philosopher should tell us how the various goods in some society can be distributed in a way that respects their particular social meanings. In a society like our own, where there are many different types of goods, they should be distributed in particular ways, depending upon our conception of each one. In such a society, egalitarian arguments, and equality itself, will be complex rather than simple. Instead of being concerned with elucidating principles to redistribute one or a few especially dominant goods, they will have to set out, in detail, the various ways many different goods should be distributed.

This second, particularist strand, provides a powerful argument against the kinds of moral philosophies that breed tyranny. Some moral ideals are so distant from the common moral notions of the members of a society that they can only be realized through tyranny. Particularism, like relativism, can be used to undermine these moral beliefs that breed tyranny.

It is Walzer's particularism and not his historicism and relativism that provides the basis for his criticism of John Rawls's moral procedure and two principles of justice.[8] In recent years, Rawls has construed his work in historicist terms (1980, 1994). He wants the original position and the two principles of justice to help us elucidate the moral principles that underlie our liberal democratic political institutions. Indeed, given that Rawls's sometimes argues that his two principles of justice are only appropriate for certain kinds of polities and societies, it may well be that Rawls too has come to accept some form of relativism. Yet the argument of A Theory of Justice is inadequate, according to Walzer, because Rawls's procedure and his principles abstract from the particular interests and concerns of the human beings in any society. Walzer has argued, rightly in my view, that this feature of Rawls's work lends itself to anti-democratic conclusions (1981, 390-393).[9] In Spheres of Justice, however, his main focus is on the failures of Rawls's theory of distributive justice.

The main difficulty with Rawls's theory is that any account of distributive justice that abstracts from the particular moral life of a society is likely to disregard important moral distinctions and claims. A theory of distributive justice of this type can only tell us how to distribute abstract and general goods. However, men and women pursue particular goods. Many philosophers have held that Rawls's notion of primary goods is inadequate because it is not completely neutral between different conceptions of the good (Schwartz 1973). Walzer's argument is partly in the same vein. If there is no neutral set of goods which political and social institutions should distribute, moral argument cannot be based solely on general principles. To try to carry on moral argument in this way is both (self-) deceptive and unfair. Walzer's argument, however, goes further than this. Even if there were some neutral set of primary goods, the actual purposes of men and women and the particular goods they pursue will likely be far more important to them then these primary goods. For example, what would, according to Rawls, be a fair distribution of income and wealth might be of no avail for those who need expensive medical care. To be fair to actual rather than abstract people one must distribute the goods they actually pursue fairly (1983, 79, 82).

Rawls holds that the actual goods we pursue are not relevant from a moral point of view. He is driven to this conclusion because he believes that there is no way to order the importance of the immense variety of goods pursued in any one society. Thus there is no basis upon which to argue that these goods should be distributed in particular ways or that the distribution of some goods should take priority over that of others. In response to Walzer's criticism, Rawls might ask why political and social policies should be especially concerned with distributing one good such as medical care over another such as fast cars.

This conclusion is ultimately rooted in a particular conception of human ends and action, which Rawls shares with most other liberals. On this Cartesian or subjectivist view, our final ends or desires are essentially given to each of us as individuals (1971, §64).[10] They can only be placed in some order of importance through an individual's exercise of instrumental and deliberative reasoning that takes their subjective importance as given. This view of human ends leads Rawls to conclude for moral philosophy to concern itself with particular rather than general and neutral goods is to adopt utilitarianism, which advocates the distribution of particular goods in a way that maximizes the greatest happiness. Rawls argument is plausible in that utilitarians do defend their doctrine by rejecting the possibility of qualitative distinctions in the character of goods.

Walzer's retort to this line of thought is that generality and abstraction are not the hallmarks of the moral point of view. Utilitarianism is not the only plausible moral theory which is concerned with the particular ends human beings pursue. For human desires are not unordered to begin with. Our desires are not discovered through introspection. Rather, they and the goods they aim at are in large part socially constituted (1983, 610). And in every society there is some conception of the meaning and the relative importance of different types of goods. Here Walzer's claims are tied to the interpretative conception of human ends that has been offered by contemporary philosophers who criticize the Cartesian or subjectivist conception underlying both A Theory of Justice and utilitarianism (Taylor 1985; Sandel 1982).

Walzer seems to think that the historicist, relativist and particularist strands of his views are woven together. Although he recognizes that general moral principles can be defended in either a rationalist or historicist manner, he does assume that rationalists must be generalists who search for or defend general moral principles that leads them to discount the particular moral conceptions of the members of any one society. That is, for Walzer, to be opposed to historicism and relativism is to opt for general and abstract moral principles. It seems to me, however, that this connection is by no means necessary. It seems to me that the debate between generalists and particularists is independent of both that between historicists and rationalists and that between relativists and universalists. There may be ways to defend rationalist or robust forms of moral philosophy that are also universalist in nature without being tempted to discard or discount all the particular moral conceptions of the members of a society. If this is the case, then much of the fear of a rationalist or universalist -relativist moral philosophy will go by the boards, at least in so far as what I called Walzer's domestic concerns are stake. To better see this, we need to consider two forms that rationalist and universalist  moral philosophies might take.

My suggestion is that Walzer's case for the attractiveness of relativist moral philosophy rests on an incomplete account of what a rationalist political and moral philosophy could be. For there are types of rationalist and universalist thought that do not lead us to discount the particular ends and moral conceptions of the members of a society. I cannot begin to present a full account of different varieties of rationalist moral thought. Instead, let me briefly sketch two ideal-typical accounts of how a rationalist and universalist moral philosophy could deal with the relationship between general principles and the particular desires and moral claims of the members of a society.

These two approaches to moral philosophy differ on three critical issues: how rationalist moral claims are validated; whether the process of rationally adjudicating these claims is closed or open ended; and how rationalist moral claims are applied in a particular society.

The validity of the first type of rationalist morality-which with some trepidation I will call the Platonic-Kantian perspective-rest on certain knowledge, whether gained through noesis of intelligible essences, introspection of human desires, synthetic a priori reasoning, or divine revelation.[11] This knowledge consists in principles that admit of no exception. The rational process by which we come to have moral knowledge of this type is closed ended: once one has grasped the basic principles upon which this knowledge rests and carried out the necessary logical deductions, general moral inquiry can end. Finally, the application of this moral knowledge is straightforward, provided one has the necessary expertise. We come to know what a moral action, policy or institution is by applying our certain moral principles to particular instances. Although the process of applying moral principles requires some subtlety of technique or expertise, on this way of doing moral philosophy it is, at least in principle, no less certain than the process by which we gain knowledge of our moral principles. The conclusions reached by this process are as singular as the moral principles with which it begins.

A second way of doing rationalist moral philosophy-which I will dub Aristotelian-takes a different stance on all three issues. On any view, some deductive reasoning plays a role in finding or constructing general moral principles. For the Aristotelian, however, these principles can be validated only by the way in which they account for our experience in the world. The role of moral principles varies from one philosopher to another: they may synthesize common opinion; explain the nature of human fulfillment while accounting for the variety of human desires and actions; or call for institutions that serve the variety of our most central goals. On this account, moral principles hold for the most part but not always. Given the vagaries of human experience, exceptions to moral principles will always be found. Philosophers of the Aristotelian type will have different views about what their moral principles are meant to explain. In each case, however, the way in which we come to justify these principles will not be closed ended. A particular moral philosopher who takes this approach will surely argue for his own conception of valid moral principles. But, given the nature of his procedures, he will recognize the possibility that another philosopher, with new human experiences to examines or superior insight, might arrive at different and perhaps superior conclusions.

When it comes to using general principles to guide particular moral decisions, philosophers of this second type expect even less certainty than that found in the pursuit of first principles. The use of general moral principles to resolve particular moral dilemmas will not necessarily require any special expertise. But it will be all the more difficult for that. For the application of Aristotelian moral principles require that one make essentially debatable judgments about particulars. The goal here is not so much to apply principles but to modify one's actions, policies and institutions to realize, as far as possible, the ideal given by one's moral principles. This requires not some technical skill or a body of knowledge that can be applied in algorithmic fashion, but practical wisdom. Practical wisdom requires subtle and intricate application itself. And it is gained as much through experience as from instruction.

I have drawn two ideal types, here. But that is all the more reason to say that philosophers fitting each of them should be familiar to all students of the history of political thought. With the usual caveat about no single philosopher fitting any category in all respects, at least as they are conventionally interpreted, Plato, Hobbes, most liberals, including Bentham and Kant, and Marxism-Leninism fall in the first category while Aristotle, Aquinas, Hume and what I like to call Marxism-Luxemburgianism fall into the second.[12]

With an outline of these two approaches in political and moral philosophy, I think we can see that Walzer's account of the dangers of rationalist and universalist moral thought is well-founded only when it comes to the first, Platonic-Kantian approach. And clearly Walzer's targets in Spheres of Justice are moral philosophies that fall into the first category, certain forms of liberalism and the totalitarian doctrines of Leninism and fascism (1983, 317). While Walzer's particularist criticism of these forms of rationalist and universalist political and moral thought is a welcome attack on what are, respectively, mildly and extremely dangerous moral doctrines, the second, Aristotelian type of non-historicist moral thought is not subject to similar criticisms.

A rationalist and universalist moral philosophy of the Aristotelian type would not pose the same threats as liberalism, Leninism or fascism. Particular examples of a rationalist and universalist theory of the second type, such as Aristotle's, would not place self-determination on the same plane as Walzer does. And a moral theory of the Aristotelian type is likely to have a model of the ideal political and social order that transcends any particular polity and society. But, it would not entirely disregard the particular moral claims made and goods pursued in a society. For theorists of the Aristotelian type typically recognize that the particular conditions in which people find themselves determine the extent to which one should aim at attaining any ideal model or some second best yet practicable modification of it.[13] Moreover, the theoretical principles offered by these philosophers are usually incomplete guides to political and social practice. They do not demand total political and social renovation through the technical application of certain moral principles. Rather, they require completion through the modification of an on-going form of political and social life. Due to the vagaries of the particular form of life in which people live, the ways in which an ideal polity and society might be realized are likely to be diverse. That practical wisdom or prudence is required to apply the moral principles of an Aristotelian theory also means that their use is not likely to be the work of one person, acting alone. No one person will have all the detailed knowledge and the political skill with which to attain any ideal. And since the relevant qualities one needs in the political leaders attempting to realize this ideal do not end (or perhaps even begin) with technical expertise, those qualities will not be so hard for the other members of a society to recognize. Finally, because the delineation and justification of any political and social ideal is not finished once and for all, the moral claims of philosophers will necessarily be somewhat tentative and exploratory. Extreme sacrifices of the present to the future could not be justified by moral claims of this sort.

The upshot of these features of an Aristotelian rationalism and universalism in political and moral theory is this: The political actor guided by a theory of this type must be a participant in the politics of his society rather than someone standing above it. Usually she will be a reformer, not a revolutionary. Her actions, and those of the members of a society who listen to her, will differ from those guided solely by the pre-existing moral beliefs of her society. The realization of the moral ideal given by these conventional beliefs and institutions will not be her aim. But neither will she try to override these conventional beliefs and practices. Instead her goal is to modify them, at first a little, then perhaps substantially, so that they better articulate the moral principles and standards justified by her non-historicist political philosophy. She would try to use the ideas current in her society as far as possible. Then, where necessary, she would modify them in what would seem to be a natural way.[14] Reforms of both ideas and institutions would not be the result of an alien intrusion into the life of a political community but the outcome of a philosophically guided engagement with that life.

A political theory of this type would not necessarily be democratic. Nor would all the moral claims defended by relativists. However, an attempt to realize the political ideals justified by a theory of the Aristotelian type would not support the kinds of intrusions in the on going life of a democratic society justified by some liberal political theories. The realization of political ideals in a democracy would especially require participation in democratic politics rather than action outside it. The outcome of debates about where authority should lie might still give much power to judges and bureaucrats. Yet people who held these positions could not make expansive claims based on their expertise in applying fixed political principles. And, given the importance of prudence, experience, and a detailed knowledge of the conditions of political and social life in realizing a political ideal, some presumption in favor of the political power of legislators probably would be warranted.

This precis of the political role of a rationalist and universalist moral theory of the second, Aristotelian, type is, I suppose, an idealization of my exemplars of such a theory. At any rate, it is this type of political and moral theory that I would defend. I hope these suggestions show why a rationalist moral theory of this second, Aristotelian type is not likely to serve as the intellectual justification of either the mild or extreme forms of tyranny which concern Walzer. If I am right, then, Walzer's particularism can be separated from his historicism and relativism and the insights of the former position be accepted while dissenting from the latter. Indeed, given the extent of Walzer's critique of Rawls's theory, even on it's historicist interpretation, clearly his real target is not so much rationalist and universalist moral theories but the generality and abstraction that can be found in moral theories that can be historicist or rationalist, relativist or universalist. Walzer's critique of moral philosophies that disregard the particular aims and moral conceptions of men and women is a brilliant analysis of the dangers of this form of thought. Even if we largely accept this part of Walzer's argument, however, we need not reject all forms of rationalist and universalist moral philosophy. We can have a rationalist morality that not only has regard for the moral beliefs and practices of particular men and women but is completed through them.

Even the Aristotelian sort of rationalism and universalism rejects Walzer's claim that it is the understanding of particulars that should play the major role in moral criticism.[15] But, if these dangers can be substantially reduced, then we might be willing to accept the risks of rationalist and universalist moral theory. For, as I shall try to show in the next few sections, there are risks in historicism and relativism as well.

Though his approach to moral philosophy rejects many fundamental premises of liberal thought, Walzer's argument reflects something of what we might call liberalism's epistemological strategy. Liberalism undermined the claims of philosophers to know the human good and priests to know the path to salvation in order to defend the right to civil liberty. Walzer's argument for moral relativism is designed to deflect the claims of philosophers to know moral truths in order to make room for self-determination in international affairs and civil liberty and democracy in (our) domestic affairs.[16] Indeed, his argument turns the liberal intellectual and moral strategy against the philosophical pretensions of liberalism itself. While relativism lends support to civil liberty and democracy in our society, it does not lead us to criticize the absence or promote the existence of civil liberty and democracy where these are not valued by the members of other societies.

This parallel to the liberal intellectual and moral strategy adds a great deal to the initial plausibility of Walzer's argument. In the rest of this paper I want to suggest some difficulties in the connection between relativism and the defense of self-determination abroad and civil liberty and democracy at home. To grasp these difficulties, we need to ask how Walzer's particularist relativism is itself justified. As we have seen, there are two possibilities here. On the one hand, Walzer could be putting forth a ethnocentric historicist defense of relativism. That is, he could be arguing that moral relativism is one of our moral principles. On the other hand, Walzer could be making a rationalist argument for relativism. That is, he could be claiming that relativism should be adopted not because it is one of our moral beliefs but because we have a good reason to think that everyone should adopt relativism. I think it is fairly clear that, ultimately, Walzer adopts the latter view. But, let us first see whether Walzer's relativism can be defended in historicist terms.

To defend Walzer's arguments against imperialism and intervention and for civil liberty and democracy from an ethnocentric historicist perspective, one would have to think that these ideas successfully articulate our conception of morality. And this is dubious, at least in foreign affairs. Are we in the United States or the West relativists about civil liberty and democracy? We certainly believe that human beings have the right to the various civil liberties and democratic government. For an ethnocentric relativist, to believe this is to hold that everyone has these rights whatever their own opinion. Do we then take moral relativism to be an overriding moral principle that justifies our looking away when violations of the rights of others are in accord with their own moral beliefs? Again, I think the answer is no. The members of Western societies are seriously divided on this issue. This casts doubt on whether the arguments for moral relativism in Spheres of Justice (and against intervention in Just and Unjust Wars) capture our moral beliefs. Many of us have doubts about the costs or the efficacy of American intervention to create civil liberty and democracy in other parts of the world. Yet, I would be surprised if many of us would not agree that intervention in support of liberty and democracy is morally desirable where the costs are low and the benefits likely. Where we do oppose American intervention in the internal affairs of other countries it is not always because self-determination is our first principle or because we have doubts about the moral status of civil liberty and democracy. Rather, we believe that in these cases intervention cannot attain its goals. Along with a broader defense of self-determination, Michael Walzer presented this narrower case for rejecting foreign intervention in Just and Unjust Wars (1977, 86-91 and 96-101). When the members of a society are not themselves committed to civil liberty and democracy we reject intervention because we believe that these aims cannot be attained. And we fear that the violations of the human right to life arising from foreign intervention would be extreme. Where we do not find these strictures, we have fewer doubts about intervention.

If my observations are correct, Walzer's moral relativism is not an accurate interpretation of our moral beliefs. Moreover, whether or not the ethnocentric defense of relativism provides ground for us to reject imperialism and intervention, it would not give the members of other societies, who do not share any of our scruples against intervention, reason to do so. If we have reason to follow our moral beliefs, so do the Leninists and Islamic fundamentalists. They may have reasons to think as they do. But this does not mean that, for us, their view is right. Yet, on the ethnocentric way of understanding moral argument, we have few resources with which to argue against those who see virtue in what we take to be imperialism.

If we turn from foreign to domestic concerns, we find that ethnocentrism provides some defense against violations of our principles of civil liberty and democracy. But note that, to the extent that an ethnocentric historicist conception of political and moral philosophy allows us to defend civil liberty and democracy, it can do so whether we are moral relativists or not. Moreover, a complaint can be made about the limited resources an ethnocentric relativist could bring to debates about liberty and democracy. Rorty argues that we can only defend civil liberty and democracy by reasserting our own reasons for standing behind these moral imperatives. The maintenance of freedom of speech is, for Rorty, one of the central projects of Western civilization (1982, 172). Ultimately, however, we would have no means by which to undermine the beliefs of someone, say a domestic Leninist who rejects these commitments. Of course, the Leninist would have no arguments to raise against us either. By denying the possibility of non-relativist moral thought, we eliminate the possibility of our being rationally convinced to reject civil liberty and democracy and take up Leninism.

If our metaethical preferences are based upon the extent to which our most central moral commitments are safe, the key question at this point is whether liberty and democracy are more secure if we have no rational defense of our values or our opponents have no rational arguments for theirs[17] In political and moral philosophy, is the best offense a good defense?

Perhaps Rorty and other ethnocentric relativists are right to think that we could have no way to justify civil liberties and democracy to someone who lived among us but shared few of our moral beliefs. Perhaps they are also right to imply that it makes no difference whether we do so or not. It might be that even good arguments at this level are not often effective.[18] But if we could find rationalist and universalist arguments of the right kind in favor of civil liberty, democracy and self-determination, it is hard to believe that these principles would not be more secure.[19]

If I was right to argue above that rationalist and universalist moral reasoning need not always be so abstract and general as to undermine the claims of democracy, then I think we can conclude that the putative attractions of moral relativism are not real, at least on the ethnocentric version. Michael Walzer's relativist defense of self-determination is probably not a good interpretation of our moral beliefs. And it is not held by the members of many societies who we might wish would read, and be convinced by, Spheres of Justice. If rationalist moral reasoning need not itself threaten the principles of democracy and civil liberty, a rationalist argument for these principles is all to the good.

The alternative to ethnocentric historicism is a moral relativism based upon universalist, and possibly rationalist, moral principles. It seems to me that this is the argument Michael Walzer offers.[20] This version of moral relativism is based upon a universalist claim that what I will call the right of moral self-determination should takes precedence over any other moral principles in foreign affairs. And, applied in domestic affairs, it limits the authority of other rationalist and universalist moral claims. This version of relativism asserts that the right to moral self-determination ought to be respected by all the members of all societies, regardless of whether their own moral beliefs would lead them to accept this principle. Moral self-determination is the rule, except where it would compromise the self-determination of others. And thus this type of moral relativism presents a rationalist argument which holds that within any polity and society, moral choices must be guided by historicist moral reasoning.

This rationalist and universalist principle of self-determination is consistent. It also provides a stronger basis of support for what is morally attractive about relativism than does ethnocentrism. For it is an example of the first, Platonic-Kantian kind of moral principle. As such it is morally binding upon those whose own moral views might otherwise lead them to reject self-determination at home or abroad. Yet there are questionable features of this form of relativism as well.

If we are to adopt the non-historicist principle of self-determination as our first moral standard, we need know why it is itself justified. In his more recent works Nation and Universe and Thick and Thin, Walzer acknowledges that his must provide a non-historicist and non-relativist defense of relativism (1990 1994). Such a defense is found in what Walzer calls a minimal and thus universal morality. The principle of respecting the moral beliefs and ideals of the members of different societies is grounded in the recognition that just as we have a history and make moral claims, so do others. Walzer argues that once we acknowledge that others have their own history and moral life, we must recognize that our own claims have "no special standing" beyond that they are our own (527). The recognition of others leads, then, to a acknowledgment of first, the limits of our own moral claims and then, second, the right of others to make their lives as they see fit. Thus, respect for moral self-determination, comes from the recognition that moral making is reiterated in different times and places.

Two important questions can be raised about this argument. First, what is the nature of the rationalist and universalist moral principle Walzer is providing here? In Nation and Universe, Walzer suggests that there are two ways of defending a universal moral principle. One he calls "covering-law universalism." Covering-law universalism is the kind of rationalism and universalism we are most familiar with. The proponent of this kind of morality seeks general principles that can be discovered on the basis of reasoned argument that, potentially, transcends the reasons found within any particular moral tradition. Such principles are then said to be found in and the basis for all particular moral traditions, in so far as they are true at all. Those who work within particular moral traditions apply and elaborate these general principles as appropriate for their own time and place.

The second kind of universalism is what Walzer calls "reiterative universalism." Reiterative universalism does not rest on some abstract argument that transcends any particular time and place. Rather it is abstracted from a large number of particular moral traditions. Walzer's claim is that a few basic principles or precepts can be found in all or nearly all moral traditions. These precepts make up what Walzer calls a "minimal morality." They include prohibitions against murder, deceit, torture and oppression (Thick and Thin, p. 10). There is no universal expression of this minimal morality. Rather, the minimal morality can only be expressed in the idiom of a particular moral tradition where it is typically part and parcel of a more maximal morality. Those of us in each tradition can, however, draw from our own tradition and that of others, some minimal precepts that serve us on the particular occasions in which we find ourselves drawn into conflicts in other places. Because this minimal morality is reiterated in most or all moral traditions, we can recognize slavery, torture and murder elsewhere. Then we can take part, vicariously or through moral or armed support, in the struggles of other people to resist them.

For Walzer, the principle of moral self-determination is part of this minimal morality. That is, Walzer believes that a universal principle of moral self-determination can be defended without appealing to a rationalist account of morality. Rather these universal principles can be shown to be a part of all or most moral traditions. Thus the adherents of each moral traditional recognize or can be brought to recognize that other people have different moral beliefs. And, once they recognize this, they will conclude that their own moral tradition has no authority in other times and places. The universal principle of moral self-determination, then, is not defended through appeal to some argument that transcends all particular moral traditions but, rather, from within each particular moral tradition, one after another.

Walzer's reiterative universalism is ingenious. But does it work as a defense of the principle of moral self-determination? Do most moral traditions do not recognize or include such a principle? I have already argued that our own moral tradition does not do so. Even if I am wrong about this, however, how plausible is it to think that most other moral traditions can accept such a claim. Do fundamentalist Christians or Moslems accept the principle of moral self-determination? Do imperialists? (Did John Stuart Mill, who justified imperialist rule over "barbarians" accept such a principle?) Do Leninists or Fascists recognize the principle of moral self-determination? How can we say that the the principle of moral self-determination is part of a minimal morality when so many moral philosophies reject it?

Not only is it unlikely that the principle of moral self-determination is part of a minimal morality, it is not all that clear that any reiterative minimal morality exists at all. Take, for example, the question of slavery. The morality of slavery was defended by the Ancient Greeks and Romans and by Americans in the South. And, what about terrorism? Terrorism was morally justified by the world historical utilitarianism of Lenin and Trotsky, was held to be a necessary part of national liberation struggles by Frantz Fanon and is repeatedly though, in most cases unwittingly, justified by those who defend the dropping of the A-bomb on Hiroshima and Nagasaki.

Walzer would certainly respond to these examples by saying that those who are oppressed by slavery or terrorism do not share in the moral codes that justify these abhorrent practices. It is true that the moral opposition to slavery and terrorism is often reiterated among those who struggle against them. But, in both Ancient and Modern times, newly enslaved people have formerly held slaves and those who suffered from terrorism have also practiced it. Moreover, even if that were not true, Walzer must explain to us why it matters whether the members of an oppressed group shares in or rejects the moral tradition that justifies their oppression. When we are looking for some reiterated minimal morality, why should we not take the moral traditions of oppressors as seriously as moral traditions of the oppressed? After all, from the point of view of the oppressors, it typically does not matter whether the oppressed accept their own oppression. What is important to them is that all of the oppressors accept that their actions are just. Walzer would undoubtedly argue that we can discount the moral traditions of oppressors when formulating a minimal morality because it is illegitimate to rule over a group people in the name of a moral tradition that these people do not accept. Or he might say that a moral tradition is only legitimate if it is accepted by all who, one way or another, live under it. But this is only to re-state, not defend the principle of moral self-determination.

The basic problem for Walzer's reiterated universal morality, then, is this: When looking for universal moral principles that are reiterated in many times and places Walzer always focuses on the moral codes of the oppressed not the oppressors. This speaks well for Walzer as a human being. But it raises questions about his philosophical acuity. I fully share Walzer's inclination to think that we are more likely to find true expressions of morality among the underdogs of history. But I do not see how this preference can be justified on the basis of a minimal morality that is found in all moral traditions. It can perhaps be based on a minimal morality found in most of the moral traditions of the underdogs. But the principle of selection by which Walzer decides which moral traditions to take seriously and which to reject is by no means part of a reiterated universal morality. Rather, it can only be defended in terms of an ethnocentric endorsement of our own moral tradition or in terms of some moral principle that is not only universal but rationalist in nature. The ethnocentric argument would allow us to discount moral traditions that justify slavery and terrorism. But, as we saw in the last section, it would not lead us to give absolute priority to the principle of moral self-determination.

So, if we are to defend the principle of moral self-determination, we must have a rationalist defense of it as a universal moral claim. So far, Walzer has not provided such a defense. It is possible that Walzer's account of how different groups of people make their own moralities might be part of such an argument. But more is obviously needed before the argument is complete. It is certainly not a logical mistake to recognize others as making a morality of their own while denying that they have made it as well as we (or some of us) have.[21] So a sound defense of a rationalist principle of self-determination remains to be found.[22] But suppose we were to find such a defense. Should this be the only rationalist and universalist principle we look for or accept?

Even, perhaps especially, if we could find a justification for a rationalist and universalist principle of self-determination, why we should stop there? Why not look for other rationalist and universalist moral principles as well? If we accept the right to moral self-determination as the foremost moral principle then we accrue the advantages of relativist moral thought, which I have discussed above. But one need not be a thorough going historicist at all to accrue these advantages. Or so I shall argue.

At different places in his recent work, Walzer acknowledges that rationalist moral philosophy might be possible. And well he should. For all Walzer needs to do to make the case for his own conception of justice is to undermine the claim of rationalist and universalist moral philosophy to political authority. If we accept the primacy of the principle of moral self-determination, no one could use a rationalist moral theory to justify ruling against the moral beliefs of the members of a society. This would rule out intervention in foreign affairs. And, in domestic affairs, it would limit the political authority of rationalist moral thought. Judges and bureaucrats, let alone vanguards, could not justify their actions in terms of rationalist and universalist moral claims not widely accepted in their society. Rationalist and universalist moral theory would then be relegated to a subsidiary, though still extremely important role. For it would presumably play an important part in the political debates that shape and transform the moral beliefs of the members of a particular society.

Thus, if we take the principle of moral self-determination to be an overriding moral principle, there is no reason to find historicism or relativism attractive doctrines. For we can get all the benefits of historicism and relativism from a rationalist principle of moral self-determination. Yet, at times, Walzer seems to want to go further and discredit all rationalist and universalist moral philosophy. He wants to argue that this type of moral philosophy is, if not impossible, then both unnecessary and dangerous. Unnecessary, because we can engage in moral thought without rationalist and universalist moral principles. And dangerous, because the strong claims of rationalist and universalist moral philosophy might encourage people to forget the primacy of self-determination. But, as we have seen, Walzer is mistaken at least about the danger of rationalist and universalist moral philosophy. We need not fear all forms of rationalist and universalist moral philosophy. For a theory of the Aristotelian type would not lead to tyranny in the name of philosophy.

If we accept the principle of self-determination then, we get the supposed moral benefits of relativism without having to actually accept the doctrine of relativism. Indeed, the primacy of this principle rests on a rationalist defense of it. This is not my view. For reasons I cannot give here, I think we have good reasons to refrain from interfering in the political and moral life of other societies without raising self-determination to the status of an absolute moral principle. But in some cases, it is possible to aid people in the defense of their rights to civil liberty and democratic government without inflicting greater harm to both life and liberty. Then we have the right, though not the obligation, to do so. Similarly, in domestic affairs politicians might be justified in acting against even the central moral conceptions of members of their own society. This might be necessary, for example, to protect civil liberty and democratic government.

For my present purposes, however, these disagreements with Walzer's views are not what is most important. However one stands on these issues, there is no reason to adopt either moral relativism or moral historicism to avoid the supposed dangers of universalist and rationalist moral philosophy. The putative attractions of Walzer's moral relativism have little to do with relativism or historicism. They result in large part from the principle of moral self-determination, which can, I think, only be defended in a rationalist fashion. They also follow from the criticism of undue generality and abstraction in moral philosophy. But this criticism is properly directed not at rationalist moral philosophy but at a certain type of universalist and general moral thought found in both rationalist and historicist varieties.

The attractions of moral relativism are, then, less than meets the eye. But is a rationalist and universalist political and moral philosophy necessary or desirable? I have already given one reason to think it is: a rationalist political theory could provide a stronger defense of our rights to civil liberty and democracy than an ethnocentric defense of these values. In other work I hope to suggest further reasons to think that moral philosophy, and our political and social lives, would be impoverished if we were to reject the possibility of rationalist and universalist moral philosophy.

Berlin, Isaiah. 1986. The Hedgehog and the Fox. New York: Simon & Schuster.
Dworkin, Ronald. 1983a. "To Each His Own", New York Review of Books, April 14, 1983, pp. 3-4.
Dworkin, Ronald. 1983b. "Response to Walzer," New York Review of Books, July 21, 1983, pp. 43-46.
Dworkin, Ronald. 1986. Law's Empire. Cambridge: The Belknap Press.
Gadamer, Hans-George. 1975. Truth and Method. New York: Seabury Press.
Galston, William.
Nozick, Robert. 1981. Philosophical Explanations. Cambridge: The Belknap Press of Harvard University Press.
Popper, Karl R. 1964. The Poverty of Historicism. New York: Harper.
Rawls, John. 1971. A Theory of Justice.
Rawls, John. 1975. "Fairness to Goodness." Philosophical Review 84:536-65.
Rawls, John. 1977. "Kantian Constructivism in Moral Theory." Journal of Philosophy 77:515-72.
Rorty, Richard. 1979. Philosophy and the Mirror of Nature. Princeton: Princeton University Press.
Rorty, Richard. 1982. Consequences of Pragmatism. Minneapolis: University of Minnesota Press
Rorty, Richard. 1985. "Solidarity or Objectivity." In Post-Analytic Philosophy, ed. John Rajchman and Cornell West. New York: Columbia University Press.
Rorty, Richard. 1989. Contingency, irony and solidarity. Cambridge: Cambridge University Press.
Sandel, Michael. 1982. Liberalism and The Limits of Justice Cambridge: Cambridge University Press.
Schwartz, Adina. 1973. "Moral Neutrality and Primary Goods." Ethics, 83:294-307.
Stier, Marc. 1989. Politics and Reason: The Structure of Political and Moral Theory. Ph.D. Dissertation, Harvard University.
Stier, Marc. Forthcoming. "Making Historicism Safe for Democracy or The Best Defense is a Good Offense: Notes on Rorty's Philosophy and Politics."
Taylor, Charles. 1985. Philosophy and the Human Sciences, Philosophical Papers 2. Cambridge: Cambridge University Press.
Walzer, Michael. 1977. Just and Unjust Wars. New York: Basic Books.
Walzer, Michael. 1981 "Philosophy and Democracy." Political Theory 9: no. 3.
Walzer, Michael. 1983. Spheres of Justice. New York: Basic Books.
Walzer, Michael. 1983b. "Reply to Dworkin," New York Review of Books, July 21, 1983, pp. 43-46.
Walzer, Michael. 1987. Interpretation and Social Criticism. Cambridge: Harvard University Press.
Walzer, Michael. 1990. Nation and Universe. The Tanner Lectures Vol. XI. Salt Lake City: University of Utah Press.
Walzer, Michael. Thick and Thin 1994. Notre Dame: The University of Notre Dame Press
Williams, Bernard. 1972. Morality. New York: Harper, 1972.
Williams, Bernard. 1981. "The Truth in Relativism" in Moral Luck. Cambridge: Cambridge University Press.
Williams, Bernard. 1985. Ethics and the Limits of Philosophy. Cambridge: Harvard University Press.
Wollheim, Richard. 1962. "A Paradox in the Theory of Democracy" in Philosophy, Politics and Society (Second Series) ed. Peter Laslett and W. G. Runciman. Oxford: Basil Blackwell.