Title:
Risk and the welfare-autonomy conflict
Abstract:
It is a familiar point that the values of welfare and autonomy can conflict. If the theatre is better for me but I prefer the cinema, you better promote my welfare by giving me the theatre ticket, but you better promote my autonomy by giving me the cinema ticket. However, this (banal) kind of value conflict occurs only when, and then obviously because, the moral patient has preferences that are substantively (prudentially) irrational. In the example just given, I have strict preferences that run counter to what is strictly better for me, and that is why you are forced to choose between what is better vs. what I prefer.
Some recent work has suggested that welfare and autonomy conflict in a more subtle and surprising way, and (in particular) conflict even when none of the relevant preferences are substantively irrational, in cases involving risk. However, I will suggest, the existing results along these lines are deficient, for various reasons. In the talk, I will outline a new and better theorem in the same general spirit. I will then discuss what might be the most plausible reactions to the conflict this theorem illustrates, suggesting in particular that in the context of valuing autonomy, the possibility of time-inconsistency should perhaps be taken more seriously than the preceding literature has taken it.