Presupposition Projection: Two Theories of Local Contexts Part II

How do complex sentences inherit the presuppositions of their parts? This is the problem of presupposition projection. An old idea is that the presupposition of an elementary expression must be entailed by the context in which it is evaluated; the relevant notion of context is, as a first approximation, what is 'common ground' between the speech act participants. We survey two theories of presupposition projection which are crucially based on the idea that there are more contexts than meets the eye. In addition to the 'global context', both theories posit that an expression has a 'local context' which aggregates information provided by earlier parts of the discourse together with the global context. The key idea is that the presupposition of an elementary expression must be entailed by its local context. But how are local contexts computed? Heim's dynamic semantics departs from the standard view of meaning as truth conditions, and takes the very meaning of words to be instructions to change the context (Heim 1983). This framework makes it possible to define empirically adequate lexical entries for a variety of operators, but it has often been considered to be insufficiently explanatory. An alternative is to stick to the standard view of meaning as truth conditions, and to reconstruct local contexts on the basis of a classical (bivalent, nondynamic) semantics (Schlenker 2009, 2010). We discuss conceptual and empirical issues that might distinguish between the two frameworks, as well as some data that are problematic for both. A final section offers a comparison between approaches based on local contexts, and an influential alternative, Discourse Representation Theory (DRT) (Geurts 1999; van der Sandt 1992). (The article is divided into two parts: Part I introduces the problem of presupposition projection and presents Heim's theory. Part II lays out the non-dynamic reconstruction of local contexts, and sketches a comparison with DRT.)