Though the revised edition of A Theory of Justice, published in 1999, is the definitive statement of Rawls's view, so much of the extensive literature on Rawls's theory refers to the first edition. This reissue makes the first edition once again available for scholars and serious students of Rawls's work.

The paradox of desire is an objection to desire-satisfaction, or preferentist, theories of welfare. In a nutshell, the objection goes like this. I can certainly desire that I be badly off. But if a desire-satisfaction theory of welfare is true, then—under certain assumptions—the hypothesis that I desire that I be badly off entails a contradiction. So much the worse for desire-satisfaction theories of welfare.

Owing to a genral dissatisfaction with hedonistic theories of value, a number of recent discussions have sought to identify an agent's selfinterest, individual utility, or personal welfare with what the agent most wants to do, all things considered. Two features of these accounts merit special attention for the argument in this paper. First, on such accounts any desire or aversion which persists in the face of complete information is logically relevant to the determination of an agent's self interest. This includes (...) apparently altruistic desires like a desire for another's happiness as well as more narrow and self-regarding desires like a desire to eat a good meal. Second, when the agent finds himself in a situation where he must choose between incompatible desires, his informed preference determines the act which is most in his self-interest. (shrink)

It seems to be a widely shared view that any defensible desire-fulfillment theory of welfare must be framed not in terms of what an agent, in fact, desires but rather in terms of what an agent would desire under hypothetical conditions that include improved information. Unfortunately, though, such accounts are subject to serious criticisms. In this paper I show that in the face of these criticisms the best response is to jettison any appeal to idealized information conditions: the considerations put (...) forward in support of the appeal to what would be desired in hypothetical circumstances of improved information do not, in fact, give adequate reason to make that appeal. (shrink)

When we compare contemporary moral philosophy with the well-known moral systems of earlier centuries, we should be struck by the fact that a certain assumption about human well being that is now widely taken for granted was universally rejected in the past. The contemporary moral climate predisposes us to be pluralistic about the human good, whereas earlier systems of ethics embraced a conception of well being that we would now call narrow and restrictive. One way to convey the sort of (...) contrast I have in mind is to note that according to Plato and Aristotle, there is one kind of life, that of the philosopher, that represents the summit of human flourishing, and all other lives are worth leading to the extent that they approximate this ideal. Certain other ethical theories of the past were in a way more narrow than this, for whereas Plato and Aristotle maintained that many things are in themselves worthwhile, others argued that there is only one intrinsic good—pleasure according to the Epicureans, virtue according to the Stoics. By contrast, it is now widely assumed that all such approaches are too exclusive, that not only are there many types of intrinsic goods but there is no one specific kind of life—whether it is that of a philosopher or a poet or anyone else—that is the single human ideal. Even hedonism, a conception of the good that had a powerful influence in the modern period, has few contemporary proponents. A consensus has arisen in our time that there is no single ultimate good that provides the measure by which the worth of all other goods must be assessed. (shrink)

The desire-satisfaction theory of welfare says, roughly, that one's life goes well to the extent that one's desires are satisfied. On standard 'actualist' versions of the theory, it doesn't matter what you desire. So long as you are getting what you actually want – whatever it is – things are going well for you. There is widespread agreement that these standard versions are incorrect, because we can desire things that are bad for us -– in other words, because there are (...) 'defective desires'. The aim of this paper is to defend the actualist desire-satisfaction theory against the problem of defective desires. I aim to show how the theory can accommodate the obvious fact that we can desire things that are bad for us. Admittedly, there are kinds of allegedly defective desire the theory cannot accommodate, but these desires, I argue, turn out not to be defective in the relevant way. (shrink)

According to the argument from self-sacrifice, standard, unrestricted desire-based theories of welfare fail because they have the absurd implication that self-sacrifice is conceptually impossible. I attempt to show that, in fact, the simplest imaginable, completely unrestricted desire-based theory of well-being is perfectly compatible with the phenomenon of self-sacrifice – so long as the theory takes the right form. I go on to consider a new argument from self-sacrifice against this simple theory, which, I argue, also fails. I conclude that, contrary (...) to popular opinion, considerations of self-sacrifice do not pose a problem for preferentist theories of welfare. (shrink)

In what follows I consider whether the idea of a person's interest or good might be better understood through that of care or concern for that person for her sake, rather than conversely, as is ordinarily assumed. Contrary to desire-satisfaction theories of interest, such an account can explain why not everything a person rationally desires is part of her good, since what a person sensibly wants is not necessarily what we would sensibly want, insofar as we care about her. First, (...) however, a tale: There was no other explanation which seemed reasonable. … [W]as it not reasonable to assume that he meant never to claim his birth-right? If this were so, what right had he, William Cecil Clayton, to thwart the wishes, to balk the self-sacrifice of this strange man? If Tarzan of the Apes could do this thing to save Jane Porter from unhappiness, why should he, to whose care she was intrusting her whole future, do aught to jeopardize her interests? (shrink)

The article explicates a notion of prudence according to which an agent acts prudently if he acts so as to satisfy not only his present preferences, but his past and future preferences as well. A simplified decision-theoretic framework is developed within which three analyses of prudence are presented and compared. That analysis is defended which can best handle cases in which an agent's present act will affect his future preferences.

Sometimes people desire that their lives go badly, take pleasure in their lives going badly, or believe that their lives are going badly. As a result, some popular theories of welfare are paradoxical. I show that no attempt to defend those theories from the paradox fully succeeds.