| OCR Text |
Show Chapter III. The Utility-Maximizing Model of Rationality: Informal Interpretations 118 an end about which further rational deliberation is possible. Of course this does not mean it cannot function as a methodological criterion for identifying and evaluating rational action from the third-person perspective. But it does mean that it cannot be an ascribable meta-end that any utility-maximizing agent might rationally and deliberatively choose to adopt. It is either an absolute, final, and rationally arbitrary meta-end within the rationality model of utility-maximization, or it is not an end at all. It would be unfortunate indeed if this were the reasoning that motivated some economists to deny that (U) is an end. But an absolute and final meta-end that is by definition incapable of entering into a utility-maximizing agent's cost-benefit analysis cannot be ranked relative to her other final ends, not even as superior to all of them. For to ascribe to it this superior ranking implies that, all things considered, it has lower opportunity costs than any alternatives. And this presupposes the contingent dispensability of (U) we have just seen is excluded by stipulation of its universality. Hence utility-maximization is not just a rationally arbitrary, absolute, final meta-end; it is a conceptually inconsistent one. For it both is and is not superior in ranking to all other ends, so both is and is not absolute, and so both is and is not universal. Therefore it is not a meta-end that any utility-maximizing agent could consistently intend, even minimally, to adhere. It seems, then - here as in the single-end interpretation, that assuming (U) to be both nonvacuous and universal implies that it is inconsistent. (U) can be made consistent only if it is either vacuous or limited in its scope of application. I take it that these conclusions give us some reason to rethink the claim that (U) is universal. For of course people do sometimes carry out their intentions to accomplish things efficiently, and to maximize utility in all of their projects. The argument has not been that utility-maximization is an inherently inconsistent end. Rather, it is inherently inconsistent when conceived as an absolute final end. Utility-maximization as an overriding value could not be both universal and nonvacuous in its application, for in that case it would be conceptually inconsistent. As soon as we acknowledge that utilitymaximizing considerations might be compared with other contingently valued meta-ends according to completely different and extrinsic rationality criteria and ranked or rejected accordingly, the inconsistency disappears. If this strikes you as a reason to reject its claim to universality, then you must view the value of rational consistency as overriding it. This is the extrinsic rationality criterion I shall try to defend at length in Volume II. 4. Three Interpretations of "Utility" So far I have argued that, in order to avoid the Scylla of vacuity and the Charybdis of inconsistency, (U) must be understood as contingent and restricted in its scope of instantiation. In discussing the single end and © Adrian Piper Research Archive Foundation Berlin |
