Local satisfaction guaranteed: A presupposition theory and its problems

Abstract

After an intermission that lasted for almost a decade, the past few years have witnessed a renewed interest in the notion of presupposition. The recent work in this area shows at least two major trends, one of which is fairly new, while the other has its roots in some of the earliest contributions on the subject. The new trend is the exploration of the links between presupposition and anaphora. This trend has received much of its impetus from van der Sandt's thesis that presuppositional and anaphoric expressions are in one and the same natural category, and are therefore amenable to a unified treatment. I shall come back to van der Sandt's theory towards the end of this paper, but in the following I am primarily concerned with the second trend, which is not necessarily consistent with the first. I am referring to the revival of what I propose to call the 'satisfaction theory' of presupposition. The ideas underlying the satisfaction theory go back to Stalnaker (1973, 1974) and Karttunen (1974). They were shaped into an explicit theory by Heim (1983), and have recently been taken up, e.g., by Chierchia and McConnell-Ginet (1990), van Eijck (1991, 1993), Beaver (1992, 1993), Heim (1992), Krahmer (1993), and Krifka (1993). In the following I shall criticize the satisfaction theory, mainly on the grounds that the predictions it delivers are too weak. I feel justified in referring to the satisfaction theory of presupposition because my objections apply to all varieties that I know of. In fact, they apply to many other theories besides, such as Karttunen and Peters' (1979) and some logical presupposition theories