Linguistics Department Colloquium:

Peter van Elswyk (Department of Philosophy, UWM)

The semantics and pragmatics of hedging

We often don’t come right out and say it—we hedge. When we do, we use a qualified declarative like Bob retired, I think or Bob probably retired as opposed to the unqualified Bob retired. Hedging is typically glossed as a speaker stating a proposition with diminished strength or commitment. In this talk, I’ll offer an explanation for why epistemic terms like probably and parenthetical I think can have this weakening effect. My explanation will crucially make use of two resources: a multidimensional semantics on which declaratives have both truth-conditional and use-conditional content (Gutzmann 2015), and a Gricean account of scalar implicature. Together, I argue, these resources enable an account of hedging that applies to epistemic terms of multiple categories (e.g. adverb, modal, verb) and that takes the best parts of force-modifier and truth-conditional semantic theories for epistemic terms without the usual problems of either.