Ethics: II. Moral Epistemology

views updated


Moral epistemology is the systematic and critical study of morality as a body of knowledge. It is concerned with such issues as how or whether moral claims can be rationally justified, whether there are objective moral facts, whether moral statements strictly admit of truth or falsity, and whether moral claims are universally valid or relative to historically particular belief systems, conceptual schemes, social practices, or cultures.

The subdiscipline of moral epistemology is hardly a recent arrival on the philosophical scene. Plato's Republic, Aristotle's Nicomachean Ethics, Hume's Treatise on Human Nature, Kant's Critique of Practical Reason, and Hegel's Phenomenology of Spirit all grapple with moral-epistemological themes and issues. However, the lion's share of explicit, selfconscious reflection on moral-epistemological problems has taken place in the twentieth century, reflecting Western philosophy's more general preoccupation with the problem of knowledge since the time of Kant. This entry describes and critically evaluates some of the major options in moral epistemology taken during that period.


When one describes a person as "good," or when one says of an action that it is "the right thing to do" under the circumstances, is one pointing out an objective feature of the person or action, or is one expressing one's own subjective reaction? Is one stating something that could be either true or false? Is one making a claim that could be supported by reasons or evidence, and that would warrant the assent of any rational human being? Or is one merely giving voice to one's own attitudes or feelings? Much of the contemporary debate in moral epistemology turns on the answer to these questions.

Intuitionists, chief among whom were G. E. Moore and W. D. Ross, insist that moral terms such as "good" and "right" name objective properties, refer to real aspects of real things, events, activities, and persons, and claim that we have access to these properties by a form of direct insight or perception. Because of this, moral statements are genuine propositions capable of being assigned a truth value of "true" or "false." To use a technical, philosophical term, morality is "cognitive." Intuitionists, while drawing an analogy between sensory intuition and moral intuition, also generally insist that moral intuition is different in kind from sense perception. While sense perception acquaints us with objective facts, moral intuition acquaints us with equally objective values.

According to G. E. Moore's Principia Ethica (1903), "good" is a simple, unanalyzable concept. Like the property concept "yellow," "good" cannot be defined except by pointing out instances of the concept, which enables one to grasp its unitary meaning. Unlike "yellow," which denotes a property intuited by our ordinary sensory apparatus, "good" names a nonnatural property, which, despite the fact that it is not empirically given, is nonetheless just as objective and real as is the property "yellow." W. D. Ross, in The Right and the Good, expands Moore's table of simple, objective moral properties to include "duty," or "rightness," and the degrees of rightness that attach to conflicting prima facie duties in different circumstances.

Intuitionists like Moore do not deny that there is moral knowledge; in fact, they affirm it emphatically. But for both Moore and Ross, our knowledge of what is ultimately good or right is not inferred or deduced but immediately given; we do not need to define, rationalize, or justify it. Thus a physician, deciding to remove an irreversibly brain-dead patient from a respirator, might give reasons for her decision by citing the beneficial consequences (e.g., an end to the patient's fruitless suffering) that might be achieved, or by insisting that the duty to preserve life is trumped by the higher duty to preserve a patient's dignity. But as to why these consequences are good, or why these putative duties are duties, the intuitionist physician can rightfully appeal only to her perception of the basic quality of goodness or rightness in them. Look and you too shall see.

The very immediacy of moral knowledge poses a serious problem for the intuitionist, namely, how moral argument and moral disagreement are possible. According to Moore, one either "sees" that something is good or one doesn't, and if one doesn't, there's little to be done except to look again. But what if two or more competent moral agents persistently "see" different values in the same circumstances? Who is "seeing" what is really there, and who is "seeing" a moral illusion? The intuitionist faces the difficulty of accounting for genuine moral disagreement—disagreement not about the empirical, factual issues of how to bring about the greatest good or do one's duty, but the evaluative issue of what sorts of things are genuine, intrinsic goods or actual obligations. This faculty of moral intuition is therefore curious. It is supposed to yield insight into objective properties of things, outcomes, deeds, and institutions, yet it lacks any public criterion against which claims like "X IS good" or "Y is the morally right thing to do" might be checked and rationally validated.


A number of thinkers influenced by logical positivism, most notably A. J. Ayer and Charles L. Stevenson, rejected intuitionism and with it the conviction that moral discourse was objective and cognitive. The resulting theory, emotivism, denied that "good" or "right" named any sort of objective, intuitable property. Rather, to say of something that it is "good" or "evil," "right" or "wrong," is to express a subjective attitude or emotional response toward it. For example, the proposition, "You ought not to have lied to that patient," asserts nothing more than "you lied to that patient"; the "ought" merely notes an attitude of disapproval on the part of the speaker. Emotivists emphasize the imperative quality of moral utterance. To say lying is wrong is, in effect, to issue the command, "Do not lie." To place ethical discourse in a recognizable context, the effort on the part of agents is to influence the behavior of others and to persuade them to adopt different beliefs. If emotivists like Ayer and Stevenson are right about the meaning of moral statements, the demand to account for "moral knowledge" is senseless, since all moral discourse is inherently noncognitive, nonrational, and subjective.

Perhaps this is an acceptable price to pay to make the phenomenon of moral disagreement intelligible. An intuitionist would be vexed by disagreements such as the following:

  • (1a) Active, involuntary euthanasia is morally acceptable under certain conditions, versus
  • (1b) Active, involuntary euthanasia is always immoral, under any and all conditions.

Yet what for the intuitionist is an epistemological dilemma, for the emotivist is not a dilemma at all. The proponent of (1a) is "commending" the permissibility of involuntary, active euthanasia under certain conditions rather than asserting a true-or-false proposition; she is expressing a "pro-attitude" toward (1a), and trying to persuade others to do so as well. The proponent of (1b) is doing precisely the same thing, expressing an "anti-attitude." The disagreement is one of subjective attitude and feeling and does not concern anything objective; there is no deep, moral truth under dispute.

But perhaps it might be premature to claim that the ability to make sense of moral disagreement thereby vindicates emotivism. One serious difficulty with emotivism is that it narrows the human significance of moral discourse by flatly denying that whenever one makes a moral claim, one places oneself in the position of having to back up that claim by citing what one takes to be good reasons in its behalf.

Universal Prescriptivism

Universal prescriptivism is a compromise between emotivism and the commonsense conviction that morality is a rational enterprise. Its chief exponent, R. M. Hare, argues in The Language of Morals (1952) that moral imperatives carry certain inexorable rational constraints. If I make the moral judgment, "Active, involuntary euthanasia is wrong," I am in effect declaring that one ought not to perform active, involuntary euthanasia on someone, and thus commanding, "Do not perform active, involuntary euthanasia," where the ought command is issued to anyone in the relevant situation, including me, the speaker. So while moral judgments have an imperative or prescriptive component—like Moore, Hare rejects naturalism—they exhibit a universality that binds the speaker's deeds to her claims, and enables the speaker to use reason to draw further moral conclusions on the basis of prescriptions that function as premises in deductive arguments.

In affirming the role of deductive reason in ethics, Hare's universal prescriptivism challenges the emotivist's assumption that only indicative premises are beyond suspicion in valid argumentation. For surely the following argument is a valid deduction:

  • (2a) I ought not to lie to my patients and thus intentionally mislead them.
  • (2b) My patient Bill asked me to tell him about his medical condition.
  • (2c) I ought not to lie to Bill about his condition.

All its premises are meaningful, and since the major premise is prescriptive, the taboo against deducing an "ought" from an "is" is not violated. Furthermore, (2a) itself could be justified by being a valid conclusion drawn from more general prescriptions:

  • (2d) I ought not to be unjust.
  • (2e) To lie to one's patients and thus intentionally mislead them is unjust.
  • (2f) I ought not to lie to my patients and thus intentionally mislead them.

However, there cannot be an infinite hierarchy of such deductions. For the prescriptivist, one's ultimate prescriptive or evaluative premises are chosen rather than deduced: One cannot ground one's moral convictions in premises more basic. The foundations for moral reasoning cannot themselves have a foundation; they reflect one's basic stance or attitude toward persons and things. No "ought" can be derived from an "is." One's moral first principles, being prescriptions, cannot be rooted in indicative soil.

This might lead one to wonder whether universal prescriptivism is more a refinement of emotivism than a genuine advance on it. It seems to push the point where ethical discourse is a matter of attitude and criterionless choice back to the most general evaluation the agent wishes to make. For example, substitute the following premise for (2a) above:

  • (2a1) I ought not to lie to my patients and thus intentionally mislead them unless I have ample reason to judge that doing so will confer some psychological or medical benefit to them.

If a physician were to judge that some such benefit were to be obtained from intentional deception, then the conclusion that one may intentionally deceive a patient will follow, in direct contradiction to (2c) and (2f). Given the initial moral orientation, certain principles for action are validated, but the original moral orientation cannot itself be validated; it can only be accepted, endorsed, chosen. Since this nonrational, inaugural choice provides the basis for all subsequent moral reasoning, the content of an agent's morality appears to be ultimately arbitrary, even if it is not arbitrary in all its detail.

Hare disagrees. In Freedom and Reason (1963) he argues that universal prescriptivism sets limits on the kinds of fundamental moral choices an agent can make. Consider the following:

  • (3a) Certain people ought to be persecuted because, and only because, their skin is black.

If moral imperatives using "ought" are, as Hare claims, universal prescriptions, then the agent uttering these words is, or ought to be, committing himself to the proposition that if his skin were black he, too, ought to be persecuted. It is clear that few individuals who make such assertions, apart from those Hare dubs "fanatics," would assent to the latter claim. Yet it is entailed by the universal prescription (3a); hence, the morality of any agent who asserts (3a) and refuses to extend it to cover himself is, for that very reason, rationally inadequate.

Of course, there is no possibility of genuine argument with a genuine "fanatic": The fanatic's assertion of ultimate principles or fundamental commitments, however odious or bizarre they may be, can only be met with counterassertion and not counterreasoning. Hare seems willing to accept this lack of logical resources against fanaticism. Nevertheless it seems reasonable to ask universal prescriptivists such as Hare whether, by cutting off rational argument at fundamental principles, they are granting too much to fanatics by ruling out any way in which their convictions can be criticized, rather than their unpleasant characters. The fanatic may be vile and depraved, but by universal prescriptivist standards, he is not necessarily defective in reason.


Intuitionists, emotivists, and prescriptivists all agree that "facts" are distinct from "values"—that an "ought" cannot be deduced from an "is." G. E. Moore coined the term, "the naturalistic fallacy," to describe the frequent attempts on the part of philosophers to define "the good" by deducing it from some matter of fact about human beings and their desires. A number of philosophers have challenged this no-ought-from-an-is doctrine by providing counterexamples to it, in effect denying that the naturalistic fallacy is a fallacy.

Philippa Foot (1959), for example, has cited "rude" and "courageous" as concepts whose evaluative meaning cannot be pried from their descriptive meaning. The criteria for identifying someone as "rude" or "courageous" are factual. If someone fits a given description, one has warrant for saying that he or she is rude or courageous; thus, the proposition "She is rude/courageous" is cognitive. But to describe someone as rude is to evaluate that person negatively. Consider the absurdity of saying: "You're rude, cowardly, and abusive, but that isn't meant as a put-down." So, according to Foot, valid moral arguments can draw evaluative conclusions from factual premises.

Peter Geach (1956) makes an analogous point in his analyses of "good." To say that a thing is good is to say something concerning the kind of thing it is. "Good" does not mean precisely the same thing in the following sentences: "That car is good"; "that watch is a good watch"; and "Mohandas Gandhi was good." To say of each one of these that it is good is to employ criteria determined by the kind of thing being evaluated. But this is to say, again against the emotivist and the prescriptivist, that the criteria that fix the meaning of evaluative terms such as "good" are not ultimately matters of choice, but rather matters of fact. To know a good watch, one needs to know what a watch is and what it is for; to know a good person, one, likewise, must know what a human being is and those ends at which humans aim in their actions.

Finally, John Searle (1964) accuses noncognitivists of harboring an arbitrarily constricted notion of what constitutes a "fact." Human institutions are part of what is the case, and these "institutional facts" can appear in descriptive premises in valid deductive arguments that generate evaluative conclusions. For example, to acknowledge the institution of promising is to grant that under certain circumstances, when one utters the words, "I promise to do X," one places oneself under an obligation to do X, and therefore is obliged to do X, and therefore one ought to do X. Because institutional facts are determined by the rules guiding the aims and actions of participants, one can deduce values from them.

Naturalists sketch a picture of moral language in which moral concepts are understood by deriving them from nonmoral, "naturalistic" ones, upon which moral knowledge rests. A robust naturalism in bioethics, then, would show no qualms about defining "the good" or "the right" in a medical context by appealing to certain key facts about human beings (e.g., their pain, dignity, mortality, etc.) and about the social and institutional setting for these facts.

At this point, however, the prescriptivist can offer a rebuttal that is difficult to answer on the naturalist's own terms without begging an important question. The prescriptivist concedes that moral language necessarily has a factual or descriptive component, but insists that it also makes ineliminable reference to the agent's desires, aims, and wishes. These can be more or less rational depending on whether their satisfaction interferes with or complements other sets of desires, aims, and wishes, but no desire can be judged rational or irrational per se. These basic desires and attitudes might differ from person to person; there is no escaping the fundamental choice behind all evaluations and prescriptions. So when the naturalist claims to have deduced an "ought" from an "is," either the major premise harbors an implicit prescription (e.g., "One ought to honor institutions like promise-keeping") or the argument is not a strict deduction.

Naturalists might reply that the "natural" premises to which they appeal and that ground moral judgment and description are rooted not in the desires or aims of individuals but in general facts about human nature of which it is the philosopher's job to remind us. For example, Aristotle understood eudaimonia, or "human flourishing," to be the good for a human being, because it was a result of acting in accord with one's rational human nature; Thomas Aquinas defined the good in terms of human creatures' reestablishing a right relation to God; and John Stuart Mill's psychological theories stand behind his definition of the good as pleasure seeking and pain avoidance. Aristotle, Thomas Aquinas, and Mill all pursued ethics in the context of what might be called "philosophical anthropology." Yet this simply elevates the naturalist's dispute with the prescriptivist to a higher level of abstraction. The prescriptivist could deny that there is any fact of the matter that might constrain the choice between philosophical anthropologies, while the naturalist could just as adamantly insist upon it. Thus naturalism might provide a coherent, consistent alternative to prescriptivism, but only by accepting philosophical stalemate at a higher level.


One possible avenue around the prescriptivist/naturalist impasse would be to repudiate the naturalistic fallacy, yet insist that moral principles are justified by examining the nature of rationality itself. This sort of moral epistemology owes much to Kant. A number of notable philosophers, inspired by Kant yet eager to avoid his dubious treatment of the self, have endeavored to ground moral knowledge in the reflective exercise of reason by actual human agents.

The most ambitious of these attempts is clearly that of Alan Gewirth, who in Reason and Morality tries to prove the fundamental principle of morality by analyzing the bare concept of rational agency. Every rational agent, Gewirth argues, must presuppose certain generic goods—namely, freedom and a degree of well-being—that make the exercise of his or her agency possible. If the agent must claim these generic goods as necessary, he or she must also claim them as rights. But since these goods flow from the generic features of agency, he or she must also concede that all other agents must claim them as rights, and that there is a corresponding obligation to acknowledge and respect them. Hence, the Principle of Generic Consistency (PGC)—"Act in accord with the generic rights of your recipients as well as yourself" (1978, p. 135)—is the fundamental, categorical principle of morality, from which all other concrete moral norms and precepts can be derived, and which can be denied only on pain of logical self-contradiction.

Many of Gewirth's critics (e.g., Nielsen; MacIntyre, 1984; Arrington) have questioned a crucial move in his dialectical "proof" of the PGC: Acknowledging that there exist necessary goods of rational agency need not entail recognizing them as one's rights. If these critics are correct, Gewirth's foundational moral principle is not necessarily true. If it is only contingently true, Gewirth's claim to a proof of the one fundamental principle of morality has not been vindicated.

In contrast to Gewirth's "hard" rationalism, other moral rationalists adopt a "soft" rationalism that proceeds not from unassailable premises about rational agency, but from contingent truths about what all rational agents would, in fact, choose under ideal conditions. For example, John Rawls, in A Theory of Justice (1971), maintains that in a hypothetical "original position," where the specific identities, desires, and advantages of rational agents are deliberately obscured behind a perspective of impartiality—a "veil of ignorance"—rational agreement would be secured regarding two specific principles of justice, equal liberty and equal distribution of goods, except in those cases where an unequal distribution of goods would work to the benefit of the worst-off social group.

"Soft" rationalism proceeds from assumptions about the rational choices individuals would make in imagined, empirical situations; thus it lends itself well to concrete application in such fields as legal, business, and medical ethics. For example, Robert M. Veatch, in A Theory of Medical Ethics (1981), argues that the responsibilities of medical professionals are set in an implicit "triple contract" involving those professionals, their patients, and society at large; specifically, medical rights and obligations are fixed by determining what sorts of agreements would be rational for all three interested parties to agree upon.

There are serious difficulties with these "soft" forms of moral rationalism. Rawls's "original position" suggests that individuals could and should be able to abstract themselves from their specific, contingent identities when formulating and justifying the principles of justice. But, as Michael Sandel (1982) and Charles Taylor (1985) have argued, this project faces formidable epistemological difficulties. It presupposes that "the self" is prior to its ends, that one's identity as a pure, rational chooser is separable from and more basic than one's identity as, say, an American, a Christian, a physician, and so on—and that it can and must draw upon rational resources that are neutral with respect to the ends and desires connected with these identities. Yet it is questionable whether such an "unencumbered" self would have any rational resources upon which to draw or any concrete intentions upon which to act; whether, indeed, the contracting chooser in the "original position" could ever be more than a philosophical fiction. Thus it seems as if moral rationalism—if it is to remain on epistemologically solid ground—must compromise its purity by admitting that the contingencies of time, place, and personal identity do make at least some difference in determining which choices and which sets of moral beliefs will be accepted as rational.

Realism and Antirealism

Another way to get around the prescriptivist/naturalist standoff would be to insist with the naturalist that there are objective moral truths, but to question whether such truths can be deduced from more basic facts concerning human nature or human institutions. On this "realist" account of moral knowledge (so called because it affirms objective moral realities independent of the knowing subject), moral discourse is less a matter of reason than of careful perception and insight, of developing the capacity to discriminate moral facts and to describe them accurately and adequately. To the extent that moral knowledge rests on "seeing" moral properties, moral realism suggests Moore's intuitionism. Yet, unlike Moore, moral realists claim no special faculty of moral intuition, insist that moral properties are observable in precisely the same way as are empirical properties, and hold that moral judgments and observations are fallible and revisable.

This renewed form of moral realism has been advanced by a number of British philosophers (Platts; McDowell, 1979; Lovibond) influenced by Donald Davidson's theory of meaning and Ludwig Wittgenstein's critique of reductionism in the philosophy of language. From Davidson they have borrowed the idea that to know what any sentence means is to be able to specify the conditions under which it is true. From Wittgenstein they have taken the conviction that there is no way to establish a ground for language that is independent of and cognitively superior to actual language in use. Taken together, these Davidsonian and Wittgensteinian commonplaces work to deflate all forms of noncognitivism.

The noncognitivist needs to rely on a contrast between two kinds of utterances—those that carry truth values and those that do not—and thus insists on two kinds of "meaning" and two kinds of discourse. One kind of discourse can accurately represent facts (usually assumed to be science), and the other does not represent facts, but expresses attitudes and imposes those attitudes on a world plastic enough to accept them (art, poetry, morality). But since determining the meaning of any linguistic statement is inseparable from determining whether that which it asserts is true or false, the noncognitivist cannot plausibly draw the required contrast between first-rate, fact-picturing discourse and second-rate, value-projecting discourse. To know what any expression means is to know what would make it true, and this ability neither demands nor supports any assumptions about the superior cognitive reliability of any one form of discourse (scientific) over any other (commonsense, literary, or moral).

The moral realist argues that there are moral facts just as there are scientific facts, and does not expect moral facts to be reducible to or deducible from any other kind of fact. Moral properties are "supervenient" upon nonmoral properties. One discerns a moral property by enumerating a number of nonmoral properties standing in relation to each other, from which the moral property "emerges" without being strictly entailed by them. "Supervenience" becomes clearer when one turns from examining "thin," abstract moral concepts ("good," "right," "duty") to "thick" moral concepts (concrete, specific concepts, like "courage," "loyalty," or "mercifulness"). To know, for example, that a physician's treatment of an end-stage cancer patient with larger than usual doses of painkillers was merciful involves knowing a great number of facts concerning cancer, pain, the special needs of the terminally ill in general and of this patient in particular, and so on. While one does not infer the moral property of being merciful from these nonmoral facts, the property is a function of them; one perceives the moral fact that this act is merciful in and through perceiving the aforementioned nonmoral facts.

"Seeing" the moral facts in the associated nonmoral facts is a complex skill, demanding discipline, practice, and attentiveness to matters of minute detail. For the moral realist, becoming a morally competent bioethicist is largely a matter of acquiring and honing a certain sensibility, akin to that of understanding a work of art or literature, whereby one comes to notice the moral goods and obligations in the context of medical practice, and to disclose and explicate them in descriptive speech.

A number of moral epistemologists (e.g., Mackie) have complained that the realists' account of supervenience is incoherent. If the supervenient moral properties of a person change (for example, if someone ceases to be courageous or just), it is necessary that other, nonmoral properties also have changed (fleeing from every danger; ceasing to give others their due). Yet if that person possesses all the nonmoral dispositions associated with a moral property (steadfastness in the face of danger; a consistent willingness to keep promises), it cannot be inferred that he or she necessarily possesses the associated moral properties (the person might not be courageous or just, "despite appearances"). Supervenience is supposedly a logical relation between properties, yet because it cannot be interpreted as a form of inference, it becomes an inexplicable fact.

John Mackie subscribes to a form of moral antirealism or "projectivism" that allows for cognitive expressions in moral discourse—that is, the truth or falsity of moral beliefs, the validity or invalidity of moral arguments—yet understands them in an equivocal sense, as a disguised, second-level reflection upon first-level moral judgments and attitudes. The moral idiom forces us to speak as if there were moral facts, but such "facts" are ultimately projections of our attitudes. To insist that moral judgments are more than expressions of attitude would be to reintroduce supervenience, with all its difficulties. Moral antirealists would not exactly deny, then, that moral knowledge is a result of coming to "see things" and describe them in a certain way; they would, however, deny that such descriptions bear more than an instrumental function. The physician who "sees" that a particular act toward a patient is merciful is indeed "seeing" something, but that something is a function of the physician's subjective attitude projected outward toward the patient.

This may not be cause for genuine worry on the realist's part. He or she could, of course, stand firm and endorse the reality of objective moral facts—the instantiation of "thick," descriptive moral properties such as "courage," "patience," and "mercifulness"—in the face of the logically peculiar notion of supervenience. Perhaps supervenience is an inexplicable logical and epistemological fact. So what? Supervenience is a feature of ordinary moral discursive practice, one that morally competent speakers can handle without much trouble. The difficulties that antirealist moral epistemologists claim to have uncovered are more a matter of their a priori prejudices (perhaps their epistemological "scientism") than their discovery of a defect in moral language and moral practice.

The realist, like Wittgenstein, confidently affirms that ordinary moral language is in good working order as it is.

The antirealist, of course, can reply that such "folk" moral philosophy is untidy, plagued with logical ambiguities and desperately in need of philosophical reinterpretation. Thus the clashes between moral realists and moral antirealists recapitulate the earlier standoff between prescriptivists and naturalists. What is at issue is not whether values can be derived from facts, but whether it even makes sense to speak of emergent "moral facts" alongside nonmoral ones.

Against Epistemology

Virtually all the various schools of moral epistemology considered seem to employ an ahistorical approach to moral discourse, argument, and judgment. Both prescriptivists and naturalists confidently speak of "the language of morals," presupposing that "morality" has a singular essence lurking under all the various "moralities" of human history. Their dispute only concerns what this "essence" might be. Rationalists, realists, and antirealists also claim their particular moral epistemologies for morality per se, as opposed to the morality characteristic of a particular time, place, or community; these epistemologies are seen as perennial options for anyone who wishes to think about ethics.

The assumption that "epistemology" studies the invariant universal structures of human knowledge, entitling it to "legislate" over all knowledge claims, has been the target of sustained philosophical attack in the latter half of the twentieth century by Ludwig Wittgenstein, Martin Heidegger, and John Dewey, among others. Richard Rorty's landmark Philosophy and the Mirror of Nature (1979) was one of the first works to point out the affinities between the projects of Wittgenstein, Heidegger, and Dewey. Rorty showed that all three undermined the pretense of "epistemologically oriented philosophy" to have attained a timeless, ahistorical, necessary vantage point in its judgments about knowledge by pointing out, in different ways, how knowledge claims are situated and justified in shared practical and social contexts and are unintelligible apart from such contexts. From Rorty's perspective, the different approaches of moral epistemologists are less important than their common goal of discovering the foundations of moral reason and showing how these foundations might (or might not) be "justified" to any rational person. But Rorty insists that the epistemological assumptions undergirding their "common goals" are baseless. Among those assumptions are the idea that there are moral truths available to human rationality as such, or that "morality," like "knowledge" and "being," is a concept with a unique, stable core meaning. Rorty's Wittgensteinian, Heideggerian, and Deweyan case against foundationalist philosophy thus makes a new, antifoundationalist and self-consciously historical approach to moral knowledge all the more appealing.

Relativism and the Feminist Critique of Objectivity

Antifoundationalism in moral philosophy has taken a number of different forms. One of them, relativism, has once again emerged as a serious option in moral epistemology. The doctrine associated with the ancient Sophists—that objectivity, truth, and knowledge are matters of adhering to sociocultural convention rather than of attaining insight into nature—has been revived and expressed in more sophisticated ways by Gilbert Harman (1975), Bernard Williams (1985), Joseph Margolis (1991), and David Wong (1984). Wong, for example, maintains that the concept of "an adequate moral system" is relative to particular places and times: There is no single, universally valid moral system available, even as an unattainable ideal. Within each extant system, there are resources available for evaluating and criticizing rival systems binding on all who share its standpoint. Wong is neither a subjectivist nor a noncognitivist. There is, however, no standpoint outside all such systems from which judgment could be passed upon each of them indifferently. For Wong, the collapse of epistemological foundationalism, and the acknowledgment that our "moral systems" are not the deliverances of pure, universal human reason but are products of historical contingencies, supports a form of relativism that is less concerned about specific judgments of right or wrong than with the assessment of moral systems or cultures on the widest scale.

Many critics of contemporary relativism have argued that it retains most of the self-referential inconsistencies that plagued its earlier incarnations. Can the relativist maintain that the relativistic thesis is "true" or "reasonable" without begging the question? (See Putnam.) Other critics argue that the historical contingency of moral beliefs and their lack of necessary epistemic foundations does not imply relativism, since it does not preclude the possibility of one moral system being more rationally adequate than its competitors (see Stout).

Yet this response elicits a further question: Whose conception of "rationality" is being employed when someone judges a moral system superior or inferior? Several important feminist philosophers have responded to this question by noting that, generally, the "rationality" employed and championed by moral philosophers has been "rationality" as understood and defined by men, who are ideologically biased by their place in a patriarchal social system and who tend to exclude the experiences and judgments of women (Tong; Code; Tuana). The idea that reason and objectivity could be "gendered" concepts has led some feminists to conclude that men and women evince different kinds of moral knowing, and to champion a feminine "ethic of care" as against a masculine "ethic of principles" (Gilligan), just as it has led others to reject those very "feminine virtues" as yet another aspect of women's oppression by men (Bartky; Puka). Whatever the ultimate outcome of these debates, contemporary feminism has done much to reinforce the antifoundationalist and historicist critique of "objectivity" and "rationality" as universal, unproblematic features of human thought and discourse. But does that critique undermine the idea of "moral knowledge" as such?

Historicism, Virtue, and Tradition

One systematic moral philosopher who disagrees with that sentiment, and who has used the insights of historicism and antifoundationalism in rethinking and recovering a workable notion of "moral knowledge," is Alasdair MacIntyre. After Virtue (1984) begins by noting both the interminable and arbitrary character of contemporary moral arguments and the vehemence with which they are conducted, and asks what might account for the powerlessness of contemporary moral philosophy to resolve moral conflict and secure agreement. MacIntyre attempts to answer this question by pursuing a historical inquiry into the succession of moral theories and the social contexts in which they arose. MacIntyre maintains that the intractability of moral disagreement is one aspect of the "emotivist culture" of late modernity that provides no solid basis for making shared, rational moral judgments and thus renders the idea of genuine moral knowledge unintelligible.

Most modern moral theory and practice has dispensed with the Aristotelian idea of a human telos, an "end" proper to human beings as such. Modern social and political orders have ceased to define their mission as that of articulating a shared vision of the good life and communally pursuing it, since it is assumed that there is no good-defining end to seek. Then what can moderns claim to "know" when they make ethical assertions, decisions, and judgments? MacIntyre dubs the standard modern response to this question "the Enlightenment project": the task of finding the universal rules or standards that guide conduct yet swing free from any substantive conception of a good life, and are justifiable by appealing to rationality.

All attempts to fulfill the ambitions of the Enlightenment project have failed, according to MacIntyre, by their own standards of success. Kantians, Utilitarians, Humeans, Intuitionists, and so on, all presuppose that there is something universally known or grasped (the Categorical Imperative, the principle of utility, the sentiment of benevolence, the self-validating property of goodness or rightness) that provides an adequate ground for moral judgment and action. Upon closer inspection, however, both the prescriptive force and the specific content of such moral foundations seem arbitrary and local rather than necessary and universal. If this is so, the epistemological universalism of the Enlightenment project functions as a mask, concealing the manipulative, will-driven ambitions of its disciples under guise of the objectivity of universal principle. Friedrich Nietzsche thus stands as both the fruition and the ruin of the Enlightenment project. His achievement is to have revealed that behind the rhetoric of objective, universal rational foundations, the morality of the modern West is yet another arbitrary upsurge of "will to power," and its impending collapse is testament to its own timid denial of this hard truth.

While MacIntyre insists that Nietzsche is certainly right about modern moral theory and practice, he has not thereby shown that all morality falls victim to the same disease. If the history of moral beliefs and moral theories can reveal the bankruptcy of the Enlightenment project and the moral nihilism of Nietzsche's "genealogical" critique of morality, it can also show how the moral philosophies they displaced can succeed where they themselves failed. MacIntyre contends that contemporary Aristotelians can draw upon epistemological resources that both Enlightenment rationalists and Nietzschean skeptics lack.

First, Aristotelians begin thinking about morality with a systematic conception of the virtues, a set of character traits that enable human agents to perfect their natures and thus realize, however imperfectly, their ultimate end. Duties and obligations—what one ought to do—begin to make sense only against the background of belief about what one ought to be. Since virtue is intrinsically connected to a conception of well-being or human flourishing shared by members of a moral community, one can establish a sound, rational motive for being moral, without reducing what one ought to prefer or desire, in light of one's true end, to what one empirically happens to prefer or desire.

Second, by understanding moral behavior as action that proceeds from a character perfected by these virtues, one eliminates the need for thinking of morality as exclusively, or even primarily, a matter of conscientious rule-following. Hence one evades the difficulty afflicting most forms of moral rationalism, that of specifying substantial moral principles, rather than empty generalities, to putatively compel the rational assent of anyone whosoever. For Aristotelians, as MacIntyre understands them, there is no moral knowledge apart from moral education and training, education not so much in assimilating precepts and norms, but in acquiring the skilled moral wisdom (phronesis) to express the proper responses and sentiments in the proper way at the proper times.

Finally, Aristotelianism, for MacIntyre, can make sense of the ways in which traditions of rational inquiry and communal practice can sustain a conception of the virtues while subjecting it to both internal scrutiny and external challenge. Most moral epistemologists make the false assumption that morality names a universal phenomenon rooted in universal human reason. If MacIntyre is right, there is no morality except as rooted in particular communities with their own particular traditions concerning the nature of the virtues and their role in promoting human well-being. This might seem to lend comfort to those moral and political conservatives who take reason and tradition to be polar opposites, and who denigrate the former and deify the latter. Yet only by participating in the common life and practices of a tradition can we come to recognize moral reasons as reasons. By dialectically examining and testing these reasons against those of rival traditions of thought and practice, we can confirm or deny their adequacy and provisionally justify our confidence in them. Traditions are the primary bearers of moral reasons; the internal evolution of traditions and the conflicts between alternative traditions indicates the way in which moral knowledge is embodied in time and history, and how moral knowers can yet transcend historical limitations.


The virtue-centered historicism exemplified by MacIntyre might seem, at first, to be yet another item on the menu of moral epistemologies, yet another intellectual position for ethicists to choose and then defend. But it would be a mistake to view it in this way. Moral epistemology, as a historicist like MacIntyre conceives of it, differs from moral epistemology as most moral epistemologists have conceived of it. MacIntyre denies the ability to transcend all traditional allegiances and to spell out the conditions for moral knowledge in general and as such. As MacIntyre suggests, the moral system it would be rational to adopt depends on who one is and how one understands oneself; there is no moral system that is rational without qualification (1988). This is certainly not to suggest a radical moral relativism, since one's initial loyalties, convictions, and self-understandings are precisely what are to be tested by inquiry and comparative criticism. One must begin inquiring somewhere, however, and the only available starting points are within the assumptions and ways of life of the specific traditions one happens to inhabit.

Thus, for historicists like MacIntyre, Rorty, Stanley Hauerwas, and Jeffrey Stout, moral epistemology can no more escape the gravitational pull of human practice and human history than can any other form of inquiry. Since they cannot be detached from the changing, finite traditions that give them rational legitimacy, it may be more accurate to speak of moral epistemologies in the plural rather than a singular moral epistemology.

The implications of historicism for bioethics are, if anything, even more profound. Since claims to moral knowledge are always made within specific traditions of thought and practice, the claims made by bioethicists about informed consent, active and passive euthanasia, paternalism and autonomy will inevitably reflect these particular traditions and will preclude appeal to any neutral ground transcending these traditions to bioethics as such. "Bioethics as such," like "rationality as such," is a post-Enlightenment fiction. Each moral tradition—whether Christian, Jewish, Islamic, or secular—will provide resources for bioethical reflection, but the individual bioethicist cannot escape reflecting and theorizing as a member of his or her tradition, as opposed to being a disengaged, impersonal spectator on "universal values." From the vantage point of historicism, bioethical inquiry and debate need to be reconfigured as conflict among and reconciliation between these traditions, which give moral thought and action their lease on life.

michael j. quirk (1995)

bibliography revised

SEE ALSO: Authority in Religious Traditions; Autonomy; Care; Communitarianism and Bioethics; Conscience; Conscience, Rights of; Consensus, Role and Authority of; Feminism; Medicine, Art of; Natural Law; Principalism; Profession and Professional Ethics; Utilitarianism;Virtue and Character; and other Ethics subentries


Arrington, Robert L. 1989. Rationalism, Realism, and Relativism: Perspectives in Contemporary Moral Epistemology. Ithaca, NY: Cornell University Press.

Ayer, A. J. 1946. Language, Truth, and Logic. New York: Dover.

Audi,-Robert. 1997. Moral Knowledge and Ethical Character. New York: Oxford University Press.

Bartky, Sandra Lee. 1990. "Feeding Egos and Tending Wounds: Deference and Disaffection in Women's Emotional Labor." In Femininity and Domination: Studies in the Phenomenology of Oppression, pp. 99–119. London: Routledge.

Bernstein, Richard J. 1983. Beyond Objectivism and Relativism: Science, Hermeneutics, and Praxis. Philadelphia: University of Pennsylvania Press.

Blackburn, Simon. 1971. "Moral Realism." In Morality and Moral Reasoning, ed. John Casey. London: Methuen.

Bloomfield, Paul. 2000. "Virtue Epistemology and the Epistemology of Virtue."

Bloomfield, Paul. 2001. Moral Reality. New York: Oxford University Press.

Code, Lorraine. 1991. What Can She Know? Feminist Theory and the Construction of Knowledge. Ithaca, NY: Cornell University Press.

Dancy, Jonathan. 1991. "Intuitionism." In A Companion to Ethics, pp. 411–414, ed. Peter Singer. Oxford: Basil Blackwell.

Fins, Joseph J.; Miller, Franklin G.; and Bacchetta, Matthew D. 1998. "Clinical Pragmatism: Bridging Theory and Practice." Kennedy Institute of Ethics Journal 8(1): 37–42.

Foot, Philippa. 1959. "Moral Beliefs." Proceedings of the Aristotelian Society 59: 83–104.

Foot, Philippa. 1978. Virtues and Vices: And Other Essays in Moral Philosophy. Berkeley: University of California Press.

Geach, Peter T. 1956. "Good and Evil." Analysis 17(2): 33–42.

Gewirth, Alan. 1978. Reason and Morality. Chicago: University of Chicago Press.

Gewirth, Alan. 1982. Human Rights: Essays on Justification and Applications. Chicago: University of Chicago Press.

Gilligan, Carol. 1982. In a Different Voice: Psychological Theory and Women's Development. Cambridge, MA: Harvard University Press.

Goldman, Alan H. 1988. Moral Knowledge. London: Routledg & Kegan Paul.

Grimshaw, Jean. 1991. "The Idea of a Female Ethic." In A Companion to Ethics, pp. 491–499, ed. Peter Singer. Oxford: Basil Blackwell.

Hare, R. M. 1952. The Language of Morals. Oxford: Oxford University Press.

Hare, R. M. 1957. "Geach: Good and Evil." Analysis 17, no. 5:103–11.

Hare, R. M. 1963. Freedom and Reason. Oxford: Clarendon Press.

Hare, R. M. 1964. "The Promising Game." Revue Internationale de Philosophie 70: 398–412.

Harman, Gilbert. 1975. "Moral Relativism Defended." Philosophical Review 84(1): 3–22.

Hauerwas, Stanley. 1979. The Peaceable Kingdom: A Primer in Christian Ethics. Notre Dame, IN: University of Notre Dame Press.

Hauerwas, Stanley. 1986. Suffering Presence: Theological Reflections on Medicine, the Mentally Handicapped, and the Church. Notre Dame, IN: University of Notre Dame Press.

Lovibond, Sabina. 1983. Realism and Imagination in Ethics. Minneapolis, MN: University of Minnesota Press.

Jaggar, Alison M. 2000. "Ethics Naturalized: Feminism's Contribution to Moral Epistemology." Metaphilosophy 31(5): 452–468.

MacIntyre, Alasdair C. 1966. A Short History of Ethics. New York: Macmillan.

MacIntyre, Alasdair C. 1984. After Virtue: A Study in Moral Theory. 2d ed. Notre Dame, Ind.: University of Notre Dame Press.

MacIntyre, Alasdair C. 1988. Whose Justice? Which Rationality? Notre Dame, IN: University of Notre Dame Press.

MacIntyre, Alasdair C. 1990. Three Rival Versions of Moral Enquiry: Encyclopaedia, Genealogy, and Tradition. Notre Dame, IN: University of Notre Dame Press.

Mackie, John L. 1977. Ethics: Inventing Right and Wrong. New York: Penguin.

Margolis, Joseph. 1991. The Truth About Relativism. Oxford: Basil Blackwell.

McDowell, John. 1979. "Virtue and Reason." Monist 62(3): 331–50.

McDowell, John. 1988. "Values and Secondary Qualities." In Essays on Moral Realism, ed. Geoffrey Sayre-McCord. Ithaca, NY: Cornell University Press.

McNaughton, David. 1988. Moral Vision: An Introduction to Ethics. Oxford: Basil Blackwell.

Moore, G. E. 1903. Principia Ethica. Cambridge: Cambridge University Press.

Nielsen, Kai. 1984. "Against Ethical Rationalism." In Gewirth's Ethical Rationalism: Critical Essays with a Reply by Alan Gewirth, pp. 59–82, ed. Edward Regis, Jr. Chicago: University of Chicago Press.

Pigden, Charles R. 1991. "Naturalism." In A Companion to Ethics, pp. 421–431, ed. Peter Singer. Oxford: Basil Blackwell.

Platts, Mark de Bretton. 1979. Ways of Meaning: An Introduction to a Philosophy of Language. London: Routledge & Kegan Paul.

Puka, Bill. 1990. "The Liberation of Caring: A Different Voice for Gilligan's 'Different Voice.'" Hypatia 5(1): 58–82.

Putnam, Hilary. 1981. Reason, Truth, and History. Cambridge: Cambridge University Press.

Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press.

Rawls, John. 1985. "Justice as Fairness: Political not Metaphysical." Philosophy and Public Affairs 14(3): 223–251.

Regis, Edward, Jr., ed. 1984. Gewirth's Ethical Rationalism. Chicago: University of Chicago Press.

Rorty, Richard. 1979. Philosophy and the Mirror of Nature. Princeton, NJ: Princeton University Press.

Ross, W. D. 1930. The Right and the Good. Oxford: Clarendon Press.

Sandel, Michael. 1982. Liberalism and the Limits of Justice. Cambridge, Eng.: Cambridge University Press.

Sayre-McCord, Geoffrey, ed. 1988. Essays on Moral Realism. Ithaca, NY: Cornell University Press.

Searle, John R. 1964. "How to Derive 'Ought' from 'Is.'" Philosophical Review 73(1): 43–58.

Sinnott-Armstrong, Walter, and Timmons, Mark, eds. 1996. Moral Knowledge?: New Readings in Moral Epistemology. New York: Oxford University Press.

Stevenson, Charles L. 1944. Ethics and Language. New Haven, CT: Yale University Press.

Stevenson, Charles L. 1963. Facts and Values: Studies in Ethical Analysis. New Haven, CT: Yale University Press.

Stout, Jeffrey. 1989. Ethics After Babel: The Languages of Morals and Their Discontents. Boston: Beacon Press.

Taylor, Charles. 1985. "The Nature and Scope of Distributive Justice." In Philosophy and the Human Sciences. vol. 2 of

Philosophical Papers. Cambridge, Eng.: Cambridge University Press. Tong, Rosemarie. 1989. Feminist Thought: A Comprehensive Introduction. Boulder, CO: Westview.

Tuana, Nancy. 1992. Woman and the History of Philosophy. New York: Paragon House.

Veatch, Robert M. 1981. A Theory of Medical Ethics. New York: Basic Books.

Williams, Bernard. 1985. Ethics and the Limits of Philosophy. Cambridge, MA: Harvard University Press.

Wong, David B. 1984. Moral Relativity. Berkeley: University of California Press.

Wong, David B. 1991. "Relativism." In A Companion to Ethics, pp. 442–450, ed. Peter Singer. Oxford: Basil Blackwell.