Abstract

This article is divided into two main sections. In section 1, I highlight some of the most significant results of Parfit's discussion of self-defeating theories in Part I of Reasons and Persons. I then argue, against Parfit, that, depending on the nature of the good, the structure of consequentialist, or agent-neutral, theories does not preclude the possibility that such theories may be self-defeating. In section 2, I discuss Parfit's ingenious argument against the self-interest theory, to the effect that as a “hybrid” theory, which is “partial” with respect to people but “neutral” with respect to time, the self-interest theory occupies a dangerous middle ground that is open to attack from two sides: on one side, by a “pure” theory that is partial with respect to both people and times, and on the other side, by a “pure” theory that is neutral with respect to both people and times. I then raise doubts, prompted by the form of Parfit's argument, as to whether there is a general requirement that we treat people, places and times the same, except, perhaps, in cases involving special relations. Specifically, I suggest that there may be reason to treat space differently than time, and that there are compelling reasons to treat persons differently than time.