An experimental investigation of (non-)exhaustivity in es-clefts

Abstract

We present an empirical study on exhaustivity inferences in German es-clefts compared to definite descriptions (pseudoclefts with an identity statement), exclusives, and focus constructions. Our study uses a novel mouse-driven picture-verification task in which the incremental updating of the context allows one to determine at which point participants take exhaustivity into consideration. Our results are compatible with a parallel analysis of clefts and definite pseudoclefts (see Percus 1997; Büring and Križ 2013; cf. DeVeaugh-Geiss et al. 2015), in so far as both structures are indeed interpreted on a par. In striking contrast to these analyses, however, we found that clefts do not systematically receive exhaustive interpretations, nor do definite pseudoclefts. We conclude that exhaustivity is not conventionally coded in either clefts nor definite pseudoclefts but rather arises through the anaphoric existence presupposition in both constructions together with a number-based implicature.