Chris Tancredi of Keio University has kindly agreed to visit us this Friday and give a version of the talk that he is giving tomorrow at Princeton, entitled "Domains of quantification, rigid designation and modality: The case for multiple models". The abstract is copied below. The talk will be at 1:30PM in the Syntax/Semantics Lab.
Abstract: In this talk I will argue that natural language interpretation requires the use of multiple models. The argument will be based on three problems. The first is Frege's problem of identity applied to rigid designators: how can "A=A" differ cognitively from "A=B" when both statements are true?. The second is the analysis of identity statements embedded within doxastic modal contexts: why isn't "A may be B and A may not be B" a contradiction as a doxastic modal statement? The third is the problem of explaining the distinct behavior of doxastic modals and metaphysical modals with respect to their interaction with quantified NPs: why can quantifiers take scope over metaphysical modals but not (at least apparently) over doxastic modals? In all three cases it will be argued that a single-model semantics makes unsustainable predictions, but that the phenomena receive a principled explanation once a single-model semantics is abandoned in favor of a multi-model approach.