The discussion of John Rawls's work over the last twenty years has made contractarianism a familiar term in political theory.' But although Rawls's work was rooted in the familiar social contract theories of Locke, Rousseau and Kant and although other theorists have made some use of the approach, the general discussion of the concept of contractarian argument in the modem literature has remained focused on Rawls's particular conception of it. Criticisms of Rawls's theory have been assumed to be criticisms of contractarianism in general, and the limitations in Rawls's use of the "social contract" idea have been assumed to be limitations that any contractarian would encounter. In part, this is because Rawls himself has contributed so much in his more recent writings to the critical discussion of the methods used in A Theory of Justice. We think it is time, however, for the discussion of contractarianism to be detached from the analysis of that particular work, so that it can be considered on its merits as a general approach to political and legal evaluation.

Our aim in this article is to present contractarianism in an attractive light. But we do not assume that there is just one contractarian method. There are many, and a consideration of contractarianism as a general theoretical approach involves a consideration of the differences among these various methods and the choices that must be made in their design. In this article, we consider some of those choices-between negative and positive contractarianism; hypothetical, tacit and actual consent; holistic and particular evaluations; rational choice and Rousseauian assumptions about the motivation of the contracting parties, and so on. These choices raise issues with which any particular contractarian conception must deal.