Moral realism and antirealist-expressivism are of course incompatible positions. They disagree fundamentally about the nature of moral states of mind, the existence of moral states of affairs and properties, and the nature and role of moral discourse. The central realist view is that a person who has or expresses a moral thought is thereby in, or thereby expresses, a cognitive state of mind; she has or expresses a belief that represents a moral state of affairs in a way that might be accurate or inaccurate. The view of antirealist-expressivism is that such a person is in, or expresses, a conative state of mind, one that consists in a certain kind of attitude or motivational stance toward something, such as an action or a person. Realism holds that moral thoughts have truth conditions and that in some cases these truth conditions are satisfied so that our moral thoughts are true. Antirealist-expressivism holds, to a first approximation, that the distinctive moral content of a moral thought does not have truth conditions.

Anyone who reflects on the way we go about arguing for or against moral claims is likely to be struck by the central importance we give to thinking about cases. Intuitive reactions to cases—real or imagined—are carefully noted, and then appealed to as providing reason to accept (or reject) various claims. When trying on a general moral theory for size, for example, we typically get a feel for its overall plausibility by considering its implications in a range of cases. Similarly, when we try to refine the statement of a principle meant to cover a fairly specific part of morality, we guide ourselves by testing the various possible revisions against a carefully constructed set of cases (often differing only in rather subtle ways). And when arguing against a claim, we take ourselves to have shown something significant if we can find an intuitively compelling counterexample, and such counterexamples almost always take the form of a description of one or another case where the implications of the claim in question seem implausible. Even when we find ourselves faced with a case where we have no immediate and clear reaction, or where we have such a reaction, but others don't share it and we need to persuade them, in what is probably the most common way of trying to make progress we consider various analogies and disanalogies; that is to say, we appeal to still other cases, and by seeing what we want to say there, we discover (or confirm) what it is plausible to say in the original case. In these and other ways, then, the appeal to cases plays a central and ubiquitous role in our moral thinking.

My aim in this essay is to explore the implications of the fact that even our most deeply held moral beliefs have been profoundly affected by our upbringing and experience—that if any of us had had a sufficiently different upbringing and set of experiences, he almost certainly would now have a very different set of moral beliefs and very different habits of moral judgment. This fact, together with the associated proliferation of incompatible moral doctrines, is sometimes invoked in support of liberal policies of toleration and restraint, but the relevance of these considerations to individual moral deliberation has received less attention. In Sections II through V, I shall argue that this combination of contingency and controversy poses a serious challenge to the authority of our moral judgments. In Section VI, I shall explore a promising way of responding to this challenge.

Do moral properties figure in the best explanatory account of the world? According to a popular realist argument, if they do, then they earn their ontological rights, for only properties that figure in the best explanation of experience are real properties. Although this realist strategy has been widely influential—not just in metaethics, but also in philosophy of mind and philosophy of science—no one has actually made the case that moral realism requires: namely, that moral facts really will figure in the best explanatory picture of the world. This issue may have been neglected in part because the influential dialectic on moral explanations between philosophers Gilbert Harman and Nicholas Sturgeon has focused debate on whether moral facts figure in relevant explanations. Yet as others have noted, explanatory relevance is irrelevant when it comes to realism: after all, according to the popular realist argument, it is inference to the best explanation of experience that is supposed to confer ontological rights. I propose to ask, then, the relevant question about moral explanations: should we think that moral properties will figure in the best explanatory account of the world?

This essay emerges from consideration of a question in the epistemology of ethics or morality. This is not the common claim-centered question as to how moral claims are confirmed and whether their mode of confirmation gives us grounds to be confident about the prospects for ethical discourse. Instead, I am concerned with the less frequently posed concept-centered question of where in human experience moral terms or concepts are grounded — that is, where in experience the moral becomes salient to us. This question was central to moral epistemology in the form it took among thinkers such as Locke, Hume, and Kant, and it remains of the first importance today.

How can an agent's desire or will give him reasons for acting? Not long ago, this might have seemed a silly question, since it was widely believed that all reasons for acting are based in the agent's desires. The interesting question, it seemed, was not how what an agent wants could give him reasons, but how anything else could. In recent years, however, this earlier orthodoxy has increasingly appeared wrongheaded as a growing number of philosophers have come to stress the action-guiding role of reasons in deliberation from the agent's point of view. What a deliberating agent has in view is rarely his own will or desires as such, even if taking something as a reason is intimately tied to desire. Someone who wants to escape a burning building does not evaluate her options by considering which is likeliest to realize what she wants or wills. She is focused, rather, on her desire's object: getting out alive. The fact that a successful route would realize something she wants is apt to strike her as beside the point or, at best, as a trivial bonus.

The prospects for moral realism and ethical naturalism have been important parts of recent debates within metaethics. As a first approximation, moral realism is the claim that there are facts or truths about moral matters that are objective in the sense that they obtain independently of the moral beliefs or attitudes of appraisers. Ethical naturalism is the claim that moral properties of people, actions, and institutions are natural, rather than occult or supernatural, features of the world. Though these metaethical debates remain unsettled, several people, myself included, have tried to defend the plausibility of both moral realism and ethical naturalism. I, among others, have appealed to recent work in the philosophy of language—in particular, to so-called theories of “direct reference” —to defend ethical naturalism against a variety of semantic worries, including G. E. Moore's “open question argument.” In response to these arguments, critics have expressed doubts about the compatibility of moral realism and direct reference. In this essay, I explain these doubts, and then sketch the beginnings of an answer—but understanding both the doubts and my answer requires some intellectual background.

The search for moral objectivity has been constant throughout the history of philosophy, although interpretations of the nature and scope of objectivity have varied. One aim of the pursuit of moral objectivity has been the demonstration of what may be termed its epistemological thesis, that is, the claim that the truth of assertions of the goodness or rightness of moral acts is as legitimate, reliable, or valid as the truth of assertions involving other forms of human knowledge, such as common sense, practical expertise, science, or mathematics. Another aim of the quest for moral objectivity may be termed its pragmatic formulation; this refers to the development of a method or procedure that will mediate among conflicting moral views in order to realize a convergence or justified agreement about warranted or true moral conclusions. In the ethical theories of Aristotle, David Hume, and John Dewey, theories that represent three of the four variants of ethical naturalism (defined below) that are surveyed in this essay, the epistemological thesis and the pragmatic formulation are integrated or combined. The distinction between these two elements is significant for the present essay, however, since I want to show that linguistic naturalism, the fourth variant I shall examine, has provided a demonstration of the epistemological thesis about moral knowledge, even if the pragmatic formulation has not been successfully realized.

These days, just about every philosophical debate seems to generate a position labeled internalism. The debate I will be joining in this essay concerns reasons for action and their connection, or lack of connection, to motivation. The internalist position in this debate posits a certain essential connection between reasons and motivation, while the externalist position denies such a connection. This debate about internalism overlaps an older debate between Humeans and Kantians about the exclusive reason-giving power of desires. As we will see, however, while these debates overlap, the new debate is importantly different from the old debate.

In the area of moral epistemology, there is an interesting problem facing the person in my area, ancient philosophy, who hopes to write a historical paper which will engage with our current philosophical concerns. Not only are ancient ethical theories very different in structure and concerns from modern ones (though with the rapid growth of virtue ethics this is becoming less true), but the concerns and emphases of ancient epistemology are very different from those of modern theories of knowledge. Some may think that they are so different that they are useful to our own discussions only by way of contrast. I am more sanguine, but I am quite aware that this essay's contribution to modern debates does not fall within the established modern traditions of discussing moral epistemology.

For a long time, it seemed that Aristotelians and Kantians had little to say to each other. When Kant the moralist was known in the English-speaking world primarily from his Groundwork and his Critique of Practical Reason, Kant's conceptual vocabulary of “duty,” “law,” “maxim,” and “morality” appeared quite foreign to Aristotle's “virtue,” “end,” “good,” and “character.” Yet ever since philosopher Mary Gregor's Laws of Freedom, published in 1963, made Kant's The Metaphysics of Morals central to the interpretation of his ethical thought, it has become clear that such “Aristotelian” terms as virtue, end, good, happiness, and character are also central to Kant. Aristotelians and Kantians now see that they have plenty to say to each other, and they have gone from being adversaries to sharing a sometimes unprincipled urge to merge central aspects of Aristotle's and Kant's ethical thought.

Epistemology, as I understand it, is a branch of philosophy especially concerned with general questions about how we can know various things or at least justify our beliefs about them. It questions what counts as evidence and what are reasonable sources of doubt. Traditionally, episte-mology focuses on pervasive and apparently basic assumptions covering a wide range of claims to knowledge or justified belief rather than very specific, practical puzzles. For example, traditional epistemologists ask “How do we know there are material objects?” and not “How do you know which are the female beetles?” Similarly, moral epistemology, as I understand it, is concerned with general questions about how we can know or justify our beliefs about moral matters. Its focus, again, is on quite general, pervasive, and apparently basic assumptions about what counts as evidence, what are reasonable sources of doubt, and what are the appropriate procedures for justifying particular moral claims.

How many serious mistakes can a brilliant philosopher make in a single paragraph? Many think that Mill answers this question by example—in the third paragraph of Chapter IV of Utilitarianism. Here is the notorious paragraph:

The only proof capable of being given that an object is visible, is that people actually see it. The only proof that a sound is audible, is that people hear it: and so of the other sources of our experience. In like manner, I apprehend, the sole evidence it is possible to produce that anything is desirable, is that people do actually desire it. If the end which the utilitarian doctrine proposes to itself were not, in theory and in practice, acknowledged to be an end, nothing could ever convince any person that it was so. No reason can be given why the general happiness is desirable, except that each person, so far as he believes it to be attainable, desires his own happiness. This, however, being a fact, we have not only all the proof which the case admits of, but all which it is possible to require, that happiness is a good: that each person's happiness is a good to that person, and the general happiness, therefore, a good to the aggregate of all persons. Happiness has made out its title as one of the ends of conduct, and consequently one of the criteria of morality.