Constitutional Choice: Rawls vs. Harsanyi

Abstract

In recent years there has been considerable exploration of two approaches to the problem of the justification of principles for the ordering of social alternatives. One, associated with Rawls, but anticipation of which is to be found in Harsanyi, turns on the notion of what the representative person would choose given a state of (radical) uncertainty.¹ The other, to which Harsanyi has also contributed, but which is familiar to most through the work of Arrow, concerns the delimitation of a social welfare function which takes a set of individual rankings (or utilities) into a social ranking (or index of social utility). The notion, in this case, is to explore reasonable constraints which can be placed upon such a function, and to determine whether, given a suitable set of such constraints, the function can be uniquely defined.²

This paper was prepared with support from the National Science Foundation, Grant SOC76-03915.