Vagueness, Imprecision and Scale Structure
Chris Kennedy, University of Chicago

The goal of this talk is to establish an empirical and theoretical distinction between vagueness and imprecision, two varieties of semantic indeterminacy that are often treated in the same way. Focusing on empirical differences between relative (tall, expensive, smart) and absolute (impure, straight, full) gradable adjectives, I will argue that vagueness is a matter of semantics, arising from the conventional meanings of certain expressions (their scalar properties), while imprecision is a function of use, reflecting variable tolerance of deviations from the conventional meanings of (crucially) non-vague terms.

I will then address the more general question of why language would contain both vague and imprecise terms, and why we need distinct frameworks for handling these types of phenomena. Here I will argue that at least in the case of gradable predicates, the facts reflect a tension between grammatical principles relating scalar representations to truth conditions and functional principles promoting flexibility of use.