A Semantics for Temporally-dependent Referring Expressions

In this paper, we sketch a new approach to the problem of the temporal interpretation of nominal predicates. The approach differs from previous analyses (e.g., Enç (1986), Musan (1999), Tonhauser (2002)) in that it is grounded in a different ontology, namely one that regards individuals as spatiotemporal entities (a.k.a. ‘spatio-temporal worms’) (Heller (1990), Sider (1997), inter alia): roughly, the idea is that individuals have temporal parts (or stages) —and are indeed like events, for that matter — in the same way they have spatial parts. As we will show, this view, known in the philosophy literature as four-dimensionalism, has interesting repercussions for formal semantics in general, and for the formal treatment of temporal NPs in particular. 1 Background on temporal NPs 1.1 The problem The task of computing the temporal interpretation for a natural language utterance is often taken to be a fairly simple matter. That is, computing the temporal interpretation for an utterance would basically boils down to giving a temporal interpretation to its verbal predicate, or, to be more precise, to its inflected projection. A direct consequence of this view is that all the other predicates present in a sentence, including nominal predicates (but also, as we shall discuss adjectival and prepositiona