Most prominent models of economic justice (and especially those proposed by Harsanyi and Rawls) are based on the assumption that impartiality is required for making moral decisions. However, although Harsanyi and Rawls agree on that, and furthermore agree on the fact that impartiality can be obtained under appropriate conditions of ignorance, they strongly disagree on the consequences of these assumptions. According to Harsanyi, they provide a justification for the utilitarian doctrine, whereas Rawls considers that they imply egalitarianism. We propose here an extension of Harsanyi's Impartial Observer Theorem, that is based on the representation of ignorance as the set of all possible probability distributions. We obtain a characterization of the observer's preferences that, under our most restrictive conditions, is a linear combination of Harsanyi's and Rawls' criteria. Furthermore, this representation is ethically meaningful, in the sense that individuals' utilities are cardinally measurable and unit comparable. This allows us to conclude that the impartiality requirement cannot be used to decide between Rawls' and Harsanyi's positions. Finally, we defend the view that a (strict) combination of Harsanyi's and Rawls' criteria provides a reasonable rule for social decisions.
