| OCR Text |
Show Rationality and the Structure of the Self, Volume II: A Kantian Conception 111 Under these circumstances it may well seem that the only remaining alternative is the standard move: to acknowledge the empirical reality of the cyclical ranking, then rule it out normatively by imposing the axioms - which secures its de jure irrationality while preserving it de re as a seeming logical possibility nevertheless. But this is not the only remaining alternative, and it is not a good one. I suggested in Volume I, Chapter IV.2 - 3 that the main effect of imposing axiom conditions so as to restrict the scope of application of (U) to the narrowly normative is that "consistency" is preserved at the expense of universality of application, whereas removing the conditions preserves universality - to the point of vacuity - at the expense of "consistency". The problem is exactly analogous to that which I then examined in Volume I, Chapters IX - XI, encountered by Humean metaethical views that impose normative conditions on instrumentally rational choosers: the more such conditions are imposed, the more the outcome is restricted to the normatively moral at the expense of its objective (or intersubjective) validity; whereas lifting the conditions increases the objective validity of the choosers' choice at the expense of its normative morality. The reason the failure takes the same form in both cases is that the relation between the two terms in each choice is exclusive rather than implicative: Just as normative morality and objective validity are mutually exclusive in an instrumentalist justificatory scheme, so, too, are normative consistency and universality in a utilitymaximizing model of rationality more generally. Indeed, the former is an instance of the latter. Excluding cyclical rankings through the imposition of normative conditions alone drives an unnecessary wedge between our conceptions of what rationality requires, what logic demands, and what reality permits. The scope of the logically possible has an authority backed by what reality permits. That authority is not superseded by the more restricted scope of the rationally admissible that the utility theorist stipulates in order to exclude the reality of cyclical rankings from its normative purview. On the contrary: that reality undermines the authority of the normative purview its exclusion helps to define. Because the normative-descriptive distinction does not solve the problem of how to parse the cyclical "inconsistency" of (1), it is tempting to draw a cruder distinction, between classical logic and decision theory simpliciter; i.e. to contend that decision theory applies to a radically different "realm" - perhaps the "realm" of the free will - from that of classical logic, in the same ways - and perhaps for the same reasons - that practical reason is inherently different from theoretical reason and action is inherently different from thought. I reject the distinction between practical and theoretical reason explicitly in Chapter V below. I call these pseudo-Kantian distinctions because I do not believe this was Kant's view, although it is often attributed to him. None of these pseudo-Kantian distinctions is ultimately convincing. Action © Adrian Piper Research Archive Foundation Berlin |
