H. P. Grice is justly famous for his work on implicature. He identified a theoretically important phenomenon and provided a theory that is still dominant. Grice's virtual discovery of implicature was a major achievement, a breakthrough in linguistics and philosophy of language. Nevertheless, his theory of implicature, I argue here, is a near-complete failure.

After explaining the concept of implicature and illustrating its theoretical importance, I review Grice's theory. It is distinguished by the claim that conversational implicatures arise from certain principles of cooperative behavior as they apply to conversation. I isolate and analyze four principal elements in Gricean theory: the Theoretical Definition of conversational implicature in terms of the conversational principles; the Calculability Assumption, which holds that conversational implicatures can always be worked out or inferred from the conversational principles; the Generative Assumption, which claims that conversational implicatures exist because of facts involving the conversational principles; and Grice's Razor, which holds that it is more economical to postulate conversational implicatures rather than senses because conversational implicatures can be derived from independently motivated principles. I use the term Gricean Theory broadly to denote not just Grice's own theory, but any principle-based theory with these four or similar elements.

I develop four main lines of argument against Gricean theory. First, the theory generates erroneous predictions as readily as it generates correct ones. Indeed, it is the exception rather than the rule to find the implicatures predicted by Gricean derivations. Second, the conversational principles have insufficient power to enable rigorous derivation of specific implicatures.

Grice (1957) drew an important distinction between what a speaker means or implies and what a sentence or other expression means or implies. The sentence “The airplane is a mile long” means the airplane is 5,280 feet long and implies the plane is over 3,279 feet long. Nevertheless, it would be most unlikely for a speaker uttering the sentence to mean or imply such things unless he were pulling someone's leg. More likely, the speaker would be exaggerating, meaning simply that the airplane is enormous compared with typical planes. A speaker using the sentence in a coded message might mean something completely unrelated to its English meaning.

As a first approximation, word meaning may be characterized as conventional speaker meaning. Whereas we could conceivably use “bachelor” to mean just about anything, we conventionally mean “unmarried male,” which is what the word means. Speaker meaning is determined by the intentions of the individual speaker. The exact intentions required are a matter of controversy. On my view, S means that the airplane is enormous only if S used the sentence to express the belief that the plane is enormous, which entails that S did something in order to provide an indication of that belief. In one typical case, S would have the further intention of producing the belief in the audience. Speaker implication is indirect speaker meaning: meaning one thing by meaning another. A boy who implies he cannot go to the movies by saying, “I have to study” has expressed the belief that he cannot go by expressing the belief that he has to study.

THE PROJECT

For many years now, an interesting conflict has raged within contemporary philosophy. On one side of the conflict are “objectivist” moral and political philosophers, who believe in and accept the existence of distinctive (and irreducible) moral “values” – words that in this conflict have been used to cover a variety of normative notions fundamental to moral and political theories – for example, rights, duties, goods, and reasons for action. These philosophers maintain that moral judgments that involve values can be true or false, and that these moral facts cannot be reduced to the sort of facts recognized by scientific theories. In that sense, they believe there are value-laden, nonreducible moral judgments that are objective. On the other side of this debate are the “naturalists,” who insist that the world, as our best scientific theories portray it, does not and cannot contain values. Since values are the stuff of the theorizing of the moral objectivists, it follows from the naturalists' position that there can be no uniquely moral facts. Hence they deny the possibility that there are value-laden, nonreducible moral judgments that are objective.

In the preceding chapters, I have argued that a theory of instrumental reason of the sort that science requires must be informed by norms that have objective authority, at least some of which concern the structure and content of human good. If these arguments are right, however, they still do not in and of themselves vindicate any objectivist moral theory. The moral skeptic can fall back on the following argument: Alright, I will accept that I have objectively authoritative norms lurking in my theory of rationality, but I do not believe that there are objectively authoritative moral norms. Norms of rationality are real; norms of morality are not.

This response raises an important question: how does one go about showing that any given norm is “real” – that is, a norm that genuinely has objective authority? Moral theorists have worried about this question in a variety of ways, but few outside of moral theory have done so, because, I suspect, they haven't believed that they had to do so.

Although this book seeks to show that the naturalists are wrong to criticize the normativity in moral theory, nonetheless in Part I, I shall be taking the naturalists' side, identifying what it is about the moral objectivists' norms that cannot, in the naturalists' view, pass “scientific muster.” This involves explaining what “scientific muster” is supposed to be and why a theory is commonly thought to be disreputable unless it passes it.

Surprisingly, however, this is a difficult project. In this chapter, I will review a variety of ways that naturalists have tried to show the unscientific nature of objective moral norms, none of which I shall argue is successful. It is remarkable that objectivist moral theorists have been so much on the defensive in recent years, given that the naturalists' arguments against their theories have been incomplete, or imprecise, or have, in various ways, begged the question at issue.

I will then spend the next two chapters developing a more successful argument on behalf of the naturalists. Once we are clear about the unnatural component within objectivist moral theories, we will be in a position to look for it in the theories of the naturalists in Part II.

If there is justification in the naturalists' dismissal of objective moral theories, it is something we still must search for. The goal of the next two chapters is to locate the nonnatural element in such theories. Ultimately I will argue that it is a certain thesis about norms and reasons for action to which naturalists object, but to which moral objectivists are committed.

To identify this thesis, however, I will need to do a lot of philosophical spadework. In Chapter 2, I will show that moral objectivism is characterized by a commitment to the idea that there are moral norms. I will then analyze the concept of a norm, showing how norms give us reasons of all sorts – to act, believe, feel, and decide, among others. Finally, I will discuss the nature of reasons and the various philosophical issues that can be, and often have been, raised to understand, identify, and defend them. I will argue that the most important identifying characteristic of a reason is its “authority” – where this is something quite different from whatever motivational efficacy the reason might have.

Chapter 3 builds on this analysis by developing two theses about the nature of this authority, one of which moral objectivists accept with respect to moral reasons, and which is inconsistent with the commitments of naturalism, the other of which is acceptable to the naturalist, but incompatible with moral objectivism.

Thesis of this book: Naturalist moral skepticism, based on a naturalist theory of reasons, fails. Naturalizing the reasons that the naturalist requires for his own conception of practical reason and scientific methodology fails. The same nonnatural “authority” of moral reasons attends the naturalist's instrumental reasons; the same reflection on the nature of human good that is required in order to live a moral life is also required to live an instrumentally rational life. The naturalist-friendly conception of instrumental reasoning as consequentialist turns out to be inadequate. And if instrumental reasoning must be construed along nonconsequentialist lines in order to understand what we do, the claims by moral theorists that moral reasoning is nonconsequentialist become yet more plausible.

What do we do now? The death of one conception of reasons clears the way for the birth of another. The naturalist conception seemed simple, elegant, and commonsensical, but since it turned out to be none of these things, how do we construct a theory of reasons that is more successful? What are the criteria that we should use? What vestiges of the naturalist program, if any, should we remain wedded to?

There is, in my view, considerable virtue in the naturalist insistence on developing a theory that resists nonsense and flights of metaphysical fancy.

Israel Scheffler once told me that every time he wrote a book, he would swear it would be his last. Any author knows the feeling – books take too long to write, the arguments and the prose never seem good enough, and when one is all done, the enormity of all that is still wrong with the final product hits home. Still, authors are ever hopeful that their contributions will be of some use, and I am no exception. I have aimed to write a book that does not so much attempt to persuade, as to dissuade – I wish to shake readers loose from the grip of a conception of the world that threatens our ability to act both rationally and reasonably.

In the process of constructing a work that attempts to rattle those who read it, I have been the recipient of much help, often from people who are quite opposed to this project. I am very grateful for their generous support and probing criticisms. In particular, I would like to thank Julia Annas, John Broome, Tom Christiano, David Copp, Ron Milo, Ken O'Day, John Pollock, Joseph Raz, John Roemer, Robert Sugden, and Bruno Verbeek. I also owe much to the graduate students attending my seminars at the University of California at Davis and the University of Arizona, in which portions of this book were presented.

Many philosophers and social scientists argue that the only acceptable theory of the nature of practical reason is what is called the “instrumental” theory, which says, roughly, that reason's only practical role is working out and recommending action that best achieves the end of the agent. Such theorists dismiss the idea that reason could ever play a noninstrumental role by dictating or determining ends themselves.

There are two general reasons why philosophers have been troubled by the noninstrumental view. First, it is a conception of reason that seems unacceptable from the standpoint of science. What special “sight” or access to normative reality can we realistically ascribe to human reason, such that it can tell us our ends in life? And how does a scientific worldview permit us to believe that there are unmotivated ends that we are rationally compelled to pursue? Science, after all, does not recognize such objects or properties with inherent prescriptive power. As we noted in Chapter 1, J. L. Mackie calls such objects and properties “queer” – indeed, too queer, given the strictures of science, for us to believe that they obtain. Moreover, no scientific description of human beings has identified a rational capacity within us that can discover these objects.

In the last chapter, I developed a strategy defending EU theory as a representation of a global instrumental theory of reason, which tells us how we ought to satisfy a preference in a situation where we have multiple preferences. Does that strategy succeed? Relevant to this question are counterexamples and experimental evidence purportedly showing that human beings consistently violate the EU axioms. That evidence appears to show that at the very least, EU theory fails as a descriptive account of our global instrumental reasoning. And the violations of some of the EU axioms seem so intuitively reasonable that EU theory also seems to fail as a normative account of our (global) instrumental reasoning. But not only does this evidence indicate that EU theory fails as a theory of instrumental reasoning, more importantly from the standpoint of moral philosophy, I will argue that this evidence shows EU theory fails because it is “too consequentialist” in structure.

The term ‘consequentialism’ was developed in moral philosophy. Indeed, moral philosophy has long been split between those who advocate a consequentialist portrayal of moral justification and those who advocate a nonconsequentialist, or deontological, approach to moral justification. There are a number of different kinds of consequentialist and deontological positions, and a variety of points of disagreement, but one of the most important concerns the nature of moral reasoning.

In Chapter 4, I aimed to show that instrumental rationality, as we standardly conceive of it, is informed by a notion of normativity that includes the idea of objective authority – just the idea that moral skeptics have claimed objective moral theories include, such that they are not acceptable from a naturalist point of view. The aim of that chapter was to begin the “companions in guilt” strategy for defending objectivist moral theory. That strategy is meant to disarm morality's skeptical critics, for if these skeptics presuppose in their conception of rationality the same occult element that they criticize in objectivist moral theories, then their criticisms indirectly defeat their own position.

This chapter continues the “companions in guilt” strategy by examining in more detail the nature of instrumental reasoning. Most theorists have taken it for granted that defining the nature of instrumental reason does not in any way involve developing a conception of the good, because figuring out the nature of instrumental reason is quite separate from the task of figuring out what the good is. This point of view is particularly critical for what I will call the “instrumentalists,” who believe that the only role that reason plays in practical deliberation is an instrumental role.