ABSTRACT
If I am to comment upon future directions for "applied social choice," the term itself must be better defined than it has been in the papers of this conference. The meaning which I shall use is intentionally broader than that which is implicit in the uses of the papers, but it captures, I believe, the essence of the phenomenon in which we are interested. The definition contains inherently two elements: a certain kind of phenomenon and a certain orientation to the phenomenon. The orientation is one of rational action: we assume that individuals act rationally to gain their ends. But the phenomenon of interest is not the individual behavior, it is the social or collective result of that behavior. Yet it is not the collective result in the particularly simple and straightforward aggregation of individual action that is found in private goods markets where there are no externalities. The phenomena of interest are social outcomes unintended by any of the actors in a more profound sense than that implied by Adam Smith's invisible hand. Indeed, the outcomes are not only unintended, but in some of the most prominent instances of social choice, are perverse to individual intentions.
