Presuppositions and Pedigrees for Discourse Markers

In this paper I defend the claim that certain discourse markers (DMs) are presupposition triggers. In itself, this claim is not new but its exact content and consequences are rarely analyzed in detail. However, such an analysis is required for at least three reasons, which correspond to the three main contributions of the paper. (1) The claim that DMs are presuppositional contrasts with two other approaches in the literature. According to Bach (1999) and Potts (2003), DMs like but contribute information on the same level as what is asserted and do not trigger any implicature or presupposition. According to Grice (1989), DMs trigger conventional implicatures. In section 2, I show that certain DMs trigger presuppositions (2.1), that Bach’s and Potts’ view must be rejected (2.2) and that presuppositions are a special kind of conventional implicature, characterized by their epistemic dynamic behavior (2.3., 2.4, 2.5). (2) In section 3, I offer a DRT representation of the presuppositional pattern described in section 2. (3) DMs express relations between discourse segments (discourse relations) and the interplay between their presuppositional behavior and their connective import raises substantial problems for a presuppositional analysis. It turns out that certain consequence DMs are sensitive to the semantic information or pedigree carried by the presupposed proposition. In section 4, I explore the modal/attitudinal status of pedigrees and show how they can integrated in a DRT framework.