INTRODUCTORY COMMENTS

In “Adjusting Utility for Justice” I proposed an axiology according to which the intrinsic value of each pleasure and each pain is adjusted by consideration of the extent to which the recipient deserves that pleasure or pain. My idea was that by thus adjusting the values of pleasures and pains for desert, I could make the axiology sensitive to justice. If this succeeds, we can maintain that our moral obligation is always to make the world as good as we can make it. We will not be open to objections based on the notion that utilitarianism ignores considerations of justice. For on the proposed view, considerations of justice play a central role in determining which world is best.

But it often happens that the suggested solution to one problem serves only to introduce further problems. In this case, the further problem concerns the nature of desert. Under what circumstances does a person deserve some pleasure or some pain? What factors might affect a person's desert of some good or evil?

Essay 9 does not contain anything like a thoroughgoing analysis of the concept of desert. My aims here are modest. I merely try to show that two deeply entrenched assumptions about desert are false.

The first assumption concerns the way in which desert interacts with responsibility. It has been assumed that desert is always based on actions for which the deserving party is responsible.

INTRODUCTORY COMMENTS

In “A Problem for Utilitarianism,” Castaneda pointed out a difficulty for traditional formulations of act utilitarianism. A careless reader might think that Castañeda's problem is just a small one of interest primarily to deontic logicians. But that would be a mistake. The problem Castañeda discovered is deep and serious. He showed that act utilitarianism requires a thor-oughgoing reconception.

On traditional formulations, act utilitarianism assumes that there are “concrete acts,” and that most of these are never performed. They remain mere possibilities. Yet these concrete acts are supposed to be the fundamental bearers of normative status. Once we accept this metaphysical assumption, it is quite natural to go on to accept the idea that some of these concrete acts are complex and have simpler acts as parts. The act of tying my shoes this morning had as its parts the act of tying my right shoe and the act of tying my left shoe.

Castañeda discovered that traditional forms of act utilitarianism run counter to deep-seated moral intuitions about the normative status of complex acts and their parts. Suppose that the act of tying my shoes this morning had higher utility than any time-identical alternative. Nothing else I could have done during the same time period would have produced better results. Then act utilitarianism of traditional forms concludes that it was obligatory.

INTRODUCTORY COMMENTS

Each of the hedonistic axiologies that I described in “Mill, Moore, and the Consistency of Qualified Hedonism” is a form of “totalism.” So is the desert-adjusted axiology of “Adjusting Utility for Justice.” In the present context, we may understand totalism to be the view that the value of a whole possible world is the sum of the values received by the inhabitants of that world.

Totalism has great intuitive appeal. It may seem obvious that all the value of a world must be distributed among those who receive value in that world. How could the value of the world be different from the sum of the values received by those in the world who receive value?

However, a number of philosophers have recognized that problems arise when we compare worlds with populations of dramatically different sizes. If there are very many recipients of value in a world, then the sum of their receipts may be quite large, even if each recipient receives just a tiny share. The sum is large not because each lives such a good life, but because there are so many of them. Totalism declares such a world to be very good, but it may seem less good than a less populous world in which each recipient gets much more but the total is smaller. This would seem to show that totalism is false.

INTRODUCTORY COMMENTS

Traditional forms of act utilitarianism were subjected to a variety of moral objections. Among these, two of the most important concerned examples involving promises and punishment. Objectors pointed out that if act utilitarianism were true, then it would be morally permissible to break a solemn promise whenever doing so would have even a tiny bit more utility than doing otherwise. They also pointed out that if act utilitarianism were true, then it would be morally permissible to punish an innocent victim whenever doing so would maximize utility. Particularly lurid and counterintuitive examples blossomed in the literature. These results were deeply disturbing.

Some utilitarians were moved by objections such as these to abandon act utilitarianism in favor of rule utilitarianism. They suggested that we go wrong in focusing on the utilities of individual acts and their individual alternatives. Instead, we should focus on the results of general or universal performance of acts relevantly like the ones in the original alternative set.

It was claimed that morally permissible acts are ones that have something like this feature: General performance of acts relevantly like them would have higher utility than general performance of acts relevantly like their alternatives. Such views were classified as forms of rule utilitarianism. Their advocates (including such luminaries as John Rawls in “Two Concepts of Rules”) claimed that these theories did not have the same offensive results as act utilitarianism.

INTRODUCTORY COMMENTS

Since time immemorial, moral philosophers have been drawn to the idea that morality is justified by its social benefits. The idea surfaces in many forms. Some, for example, have suggested that a society's morality is a system of rules adopted by the society for the purpose of coordinating the behavior of the members; if the members obey the rules, the society as a whole will be best off. Such views are versions of the Principle of Moral Harmony, or “PMH.” My aim in this essay is to show that such principles are false.

PMH has special relevance to utilitarianism because an advocate of act utilitarianism might appeal to PMH in an effort to justify act utilitarianism. The reasoning might go like this: If everyone in a group were to act in accord with the demands of act utilitarianism, the group would be best off. No other normative theory can in this way guarantee to maximize group benefits. In other words, act utilitarianism is the only normative theory that generates obligations sure to conform with PMH. Therefore, there is a reason to advocate and believe in act utilitarianism above all other normative theories.

However popular such thinking might be, I am convinced that it is deeply misguided. The greatest mistake, as I see it, is the assumption that a normative theory needs or could be given this sort of justification.

The deepest disputes in epistemology focus on concepts that are quite obviously ethical and often are borrowed directly from theoretical moral discourse. We frequently find references to epistemic duty or epistemic responsibility, to the fact that we ought to form beliefs in one way rather than another, to the fact that one way of believing is good, or at least better than some other, and more recently to the idea of intellectual virtue. But these concepts are often used with little reflection, and rarely with any concern for the fact that they may be borrowed from a particular type of moral theory. Any problems in the theory may adversely affect the epistemological inquiry. On the other hand, the theory's advantages may be advantages for epistemology as well.

Almost all epistemological theories are modeled on act-based moral theories. When their model is deontological ethics, that is usually readily apparent. Less obvious is the fact that the popular theory of reliabilism is structurally parallel to consequentialism. To my knowledge, no epistemological theory is closely modeled on a pure virtue theory. The idea of intellectual virtue was introduced into the epistemological literature by Ernest Sosa, but Sosa does no more than mention an association with virtue ethics, and subsequently “virtue epistemology” has been used as another name for reliabilism. The works of Lorraine Code and James Montmarquet come closer to linking epistemology with virtue ethics, but neither one derives the concept of epistemic virtue from a background aretaic ethics or pushes the similarities between intellectual virtue and moral virtue very far.

The nature of knowledge is arguably the central concern of epistemology and unarguably one of the major interests of philosophy from its beginning. Ever since Plato and no doubt long before, knowledge has been held in high regard. Plato called knowledge the most important element in life (Protagoras 352d) and said that the only thing truly evil is to be deprived of it (Protagoras 345b). Even today, few deny that it is the chief cognitive state to which we aspire, and some claim that it is the chief state of any kind to which we aspire. The possession of knowledge is one of life's great joys – or, at least, one of its benefits. In short, knowledge is valuable.

The valuational aspect of knowledge and of the related states of justified, rational, or warranted belief has led to numerous parallels between moral and epistemic discourse. As Roderick Chisholm observed years ago, “many of the characteristics philosophers have thought peculiar to ethical statements also hold of epistemic statements” (1969, p. 4). Since then epistemologists have routinely referred to epistemic duty and responsibility, to epistemic norms and values, and to intellectual virtue. On occasion they also use forms of argument that parallel arguments in ethics. In some cases this is done consciously, but in other cases it appears to be unnoticed, and the epistemological discussion is carried on without attention to the fact that the corresponding discussions in ethics have by now become more advanced.

Part I surveyed the use of ethical theory in epistemology, showed the advantages of a pure virtue theory, and distinguished this approach from that typically used in contemporary epistemology and from other forms of “virtue epistemology” Part II developed a theory of virtue and vice broad enough to handle epistemic as well as moral evaluation and showed how the concepts of justified belief and epistemic duty can be defined within the theory Now we turn to an investigation of the most critical concern of epistemology: the analysis of knowledge. We have already seen that the motivation to know is the most basic constituent of every intellectual virtue, and each intellectual virtue is constructed in such a way as to regularly lead to its end of attaining knowledge. Since intellectual virtues are forms of moral virtue, it follows that knowledge is intimately bound up with moral concepts, although it has not always been treated that way. In this, the final, part of the book, I begin by locating the concept of knowledge within the domain of ethics. I then propose a definition of knowledge and show how it is immune to Gettier problems. The theory is externalist according to the most common definition of externalism but has a stronger internalist aspect than the more common externalist theories. theory – and answer anticipated objections to my theory.

We now begin the task of focusing epistemology on the concept of a virtue. In section 1, I distinguish several types of virtue theory by the ways they relate the fundamental moral concepts of a virtue, the good, and a right act. A pure virtue theory makes the concept of a right act derivative from the concept of a virtue, although there is more than one way such a theory can relate virtue to the good. Much of what I will do in Part II is compatible with many forms of aretaic ethics, but I am particularly interested in two forms of pure virtue theory: happiness-based and motivation-based theories. One of my aims in the rest of this book is to demonstrate that both forms of pure virtue theory, including the more radical, motivation-based theory, can be developed in ways that adequately handle epistemic evaluation. Incidentally, I expect moral philosophers will find something of interest in this part of the book, whether or not they have any interest in epistemology.

Section 2 will be devoted to a careful account of the nature of a virtue, distinguishing it from feeling states, from natural capacities, and from skills. I identify both a motivational component and a reliability component in virtue. In section 3 I turn to an investigation of the intellectual virtues. I argue that intellectual virtues are forms of moral virtue and that the many logical and causal connections among the moral and intellectual virtues make it important for a virtue theory to be broad enough in scope to account for the entire range of intellectual and moral virtues in a single theory.