Within the discipline of welfare economics, one tradition-the BergsonSamuelson tradition-has been dominant for over a quarter of a century. The most basic concept in this tradition is the social welfare ordering or social welfare function. This represents the judgments of some private individual, public official or ethical observer concerning what is better or worse for society. Certain restrictions are placed on the form such an ordering may take; these represent ethical judgments which, it is presumed, anyone who orders social alternatives would or should accept. These ethical judgments are the Paretian ones. Sometimes a writer argues that his own moral intuitions about the rightness or reasonableness of the Paretian value judgments are so strong that he feels everyone should accept them. (This seems to be Samuelson's (1977) own position.) Other writers argue, although usually without supporting evidence, that, as a matter of fact, almost everyone does have such moral intuitions (e.g. Graaff, 1957, p.9; Mishan, 1960, p.199). In either case the Paretian value judgments are treated as axiomatic for welfare economics. They are not to be derived from anything more fundamental. So long as there is no dispute about the status and force of Paretian principles this procedure is unexceptionable. However it is inadequate if a justification is sought for the principles in question. Usually economists have not given this question much thought. Some (e.g. Graaff, 1957, p.9) have suggested in passing that Paretian principles embody an individualist or liberal ethic, but they have not shown why this is the case. Moreover, recent work has cast doubt on the idea that Paretian social choice theory is compatible with liberalism (e.g. Sen, 1970b; Sugden, 1978). We propose to provide a justification for Paretian principles, and to do this we turn to another tradition in political philosophy and welfare economics, namely social contract theory. In its modern version this theory begins with the idea of an original position, corresponding to the state of nature in classical contract theory, in which all persons are equally free, rational and self-interested, and in which each is located behind a of about his or her place in society (cf. Rawls, 1972, pp.11-17). Ideas of social choice are accordingly understood as those principles that would be agreed by all persons behind the veil of ignorance as rules to govern the terms of their association. To us at least this theory is an appealing device for systematizing those ethical convictions that would otherwise have to be treated as underived axioms within welfare economics. We shall show that the theory provides a justification for Paretian principles. More importantly, we shall argue, it provides a justification for other principles, which some writers have wished to add to the Paretian corpus, but for which no rationale could be found within the framework of the Bergson-Samuelson approach. So, our intention in this paper is to show how social contract theory can form the basis for wel-