TWO KINDS OF CONSEQUENTIALISM

In a famous footnote in Anarchy, State, and Utopia, Robert Nozick wonders whether " a side-constraint view can be put in the form of the goal-without-side-constraint view " (Nozick 1974, 29). Put in more familiar terms, Nozick wants to know whether it is possible to represent side-constraints in a broadly consequentialist framework. For those who think that the arguments in favour of some sort of consequentialism are overwhelming, this question is of the first importance. Nozick himself thinks that the issue is a tricky one, especially if the consequentialist is allowed to distinguish between two goals that an agent might have: the goal of there being no violation of one person's rights by another—where I assume he means us to understand this in terms of a list of prohibited actions—and the goal of the there being no violation of anyone's rights by the agent himself. Indeed, the consequentialist might also need to distinguish between the goal of the agent himself not violating anyone's rights at any time and the goal of the agent himself not violating anyone's rights at the time at which he acts. But even if he is allowed to distinguish between such goals, Nozick thinks that the representation would still be objectionable because " gimmicky " (Nozick 1974, 29). 1 In recent years several theorists have addressed a more general version of Nozick's question Can all substantive moral theories, and perhaps all theories of reasons for action, both moral and non-moral, be represented in consequentialist terms? The arguments for and against the possibility of such a representation are complicated, but even if they could, Nozick's residual objection that such representations are gimmicky remains. So let's confront that question head-on. Are such representations gimmicky? My own view is that they are not. Such representations of moral theories, and of theories of reasons for action more generally, are compulsory on