| OCR Text |
Show Rationality and the Structure of the Self, Volume I: The Humean Conception 113 end that conflicts with or obstructs others she deems more important. For example, Myrtle may find it very difficult to live simply, get things done efficiently, and hold down a full-time job. Or, to take another example, an agent might assign the greatest weight to an end with the lowest probability, such as winning the lottery, and subordinate all of her other endeavors to that one - thereby depriving herself of the resources necessary to achieve any of them, as the gambling addict does. So it would seem that the coherence set interpretation of (U) is not susceptible to the reproach of vacuous universality. 3.2. Nonvacuity But now let us examine each of these features, i.e. universality and nonvacuity, at greater length. First assume universality and consider nonvacuity. The claim is that an agent's coherence set as just described is nonvacuous in the sense that it is possible for him to violate it, by failing to order all of his ends in the requisite way. But is it possible for him to thus fail to order his actual ends? Suppose that, having arrived at such a coherence set, he now proceeds to pursue an end not contained in the set, that conflicts with its ordering. Is he maximizing utility anyway, or is he not? There are at least three possible answers to this question. A first is that he is not; that he is then acting irrationally, since he is, by hypothesis, not advancing all the ends contained in the set. This is the answer the utility theorist should give. Nevertheless it ignores the criterion according to which we were originally supposed to identify irrationality, namely failure to adjust all of one's ends so as to produce a coherence set. The case is one in which the agent is guilty of no such failure. He simply pursues an end not contained in the set he successfully ordered. It is tempting to respond that if he pursues this end, then it is his end that then must be ordered relative to the set. But this does not follow. It is not difficult to imagine a case in which an agent pursues an end that is not his own. For example, wives traditionally have been expected to pursue their husbands' ends, regardless of what they thought about those ends, and have done so sometimes despite their own severe reservations or opposition to them. If an agent can pursue an end that is not his own, then an agent can pursue an end not contained in the coherence set of all his ends. A second answer might be that if the agent pursues such an end, then if it is his end, it conflicts with the coherence set he has ordered. He has, therefore, violated that set, has thereby failed to maximize utility, and so has acted irrationally. But the fact that he is pursuing an end of his that conflicts with the set he ordered is evidence - indeed, for the revealed preference theorist, 19 conclusive evidence - that he has reorganized his priorities, reordered the set to incorporate the seemingly delinquent end, and indeed has ascribed to it 19 I address revealed preference theory at length in Chapter IV, Sections 2 and 3, below. © Adrian Piper Research Archive Foundation Berlin |
