In his critique of rational choice theory, Amartya Sen claims that committed agents do not (or not exclusively) pursue their own goals. This claim appears to be nonsensical since even strongly heteronomous or altruistic agents cannot pursue other people's goals without making them their own. It seems that self-goal choice is constitutive of any kind of agency.

In this paper, Sen's radical claim is defended. It is argued that the objection raised against Sen's claim holds only with respect to individual goals. Not all goals, however, are individual goals; there are shared goals, too. Shared goals are irreducible to individual goals, as the argument from we-derivativeness and the argument from normativity show. It is further claimed that an adequate account of committed action defies both internalism and externalism about practical reason.

Gary Becker and others have done important work to broaden the content of self interest, but have not departed from seeing rationality in terms of the exclusive pursuit of self-interest. One reason why committed behavior is important is that a person can have good reason to pursue objectives other than self interest maximization (no matter how broadly it is construed). Indeed, one can also follow rules of behavior that go beyond the pursuit of one's own goals, even if the goals include non-self-interested concerns. By living in a society, one develops possible reasons for considering other people's goals as well, which takes one beyond an exclusive concentration on one's own goals, not to mention the single-minded pursuit of one's own self interest. The recognition of other people's goals may be a part of rational thought. If rational behavior may depart from the relentless pursuit of one's own goals, commitment has to be important in a theory of rationality. Furthermore, seeing the role of commitment in human behavior can have explanatory importance in allowing us to understand behavior patterns that are hard to fit into the narrow format of contemporary rational choice theory. Commitment is, thus, important both for practical reason and for causal explanation.

Why does Sen maintain that people are capable of putting their own goals offline and deliberating and acting out of sheer commitment to others? How can he endorse such a rejection of the belief-desire model of agency? The paper canvasses three explanations and favors one that ascribes an unusual position to Sen: the belief that so far as agents remain in the belief-desire mould, they cannot deliberate on the basis of reasons other than those that derive from standing goals that form an integrated system. What he thinks of as deliberation on the basis of commitment is just the sort of deliberation that involves the formation of a novel, perhaps occasion-specific goal in which the good of another is prioritised.

While very much in Sen's camp in rejecting revealed preference theory and emphasizing the complexity, incompleteness, and context dependence of preference and the intellectual costs of supposing that all the factors influencing choice can be captured by a single notion of preference, this essay contests his view that economists should recognize multiple notions of preference. It argues that Sen's concerns are better served by embracing a single conception of preference and insisting on the need for analysis of the multiple factors that determine ‘preference’ so conceived.

In the current debate on economic rationality, Amartya Sen's analysis of the structure of commitment plays a uniquely important role (Weirich 2004: 387ff.). However, Sen is not alone in pitting committed action against the standard model of rational behavior. Before turning to Sen's analysis in section 2 of this paper, I shall start with an observation concerning some of the other relevant accounts.

In general, capitalists deserve profits and losses for their contribution to the general welfare. Market imperfections and the range of permissible prices (at least within the boundaries of exploitation) prevent the alignment from being a direct one, but the connection generally holds. In the context of the market, this thesis preserves the central place of moral responsibility in moral desert. It also satisfies the fittingness and proportionality conditions of moral desert and provides a backward-looking and pre-institutional ground of it. In addition, the focus on contribution unifies several different types of act-based desert, specifically deserved profits and losses, deserved punishment, and deserved wages. Hence, to the extent that desert-satisfaction is relevant in the selection of an economic system, this result strengthens the case for capitalism.

A central question for assessing the merits of Amartya Sen's capability approach as a potential answer to the “distribution of what”? question concerns the exact role and nature of freedom in that approach. Sen holds that a person's capability identifies that person's effective freedom to achieve valuable states of beings and doings, or functionings, and that freedom so understood, rather than achieved functionings themselves, is the primary evaluative space. Sen's emphasis on freedom has been criticised by G. A. Cohen, according to whom the capability approach either uses too expansive a definition of freedom or rests on an implausibly active, indeed “athletic,” view of well-being. This paper defends the capability approach from this criticism. It argues that we can view the capability approach to be underpinned by an account of well-being which takes the endorsement of valuable functionings as constitutive of well-being, and by a particular view of the way in which endorsement relates to force and choice.

Axiomatic utility theory plays a foundational role in some accounts of normative principles. In this context, it is sometimes argued that transitivity of “better than” is a logical truth. Larry Temkin and Stuart Rachels use various examples to argue that “better than” is non–transitive, and that transitivity is not a logical truth. These examples typically involve some sort of “discontinuity.” In his discussion of one of these examples, John Broome suggests that we should reject the claim which involves “discontinuity.” We can, I suggest, make sense of the examples which Temkin uses while sacrificing neither transitivity nor “discontinuity.” This response to Temkin's examples involves developing and modifying James Griffin's account of “discontinuity.” If the account of “discontinuity” seems implausible, that is because of a failure to allow for vagueness. A similar argument can be made in the context of the well-known “repugnant conclusion.”