Speaker: Amir Anvari (MIT)
Title: On the subject matter of formal semantics
Abstract: Formal semantics entered the scene with more than a whiff of what is sometimes referred to as externalism, the claim that a theory of meaning is a theory of how mental representations relate to the world, in contradistinction to internalism, the claim that a theory of meaning is a theory of how representations relate to each other. One may ask whether externalism provides an adequate conception of what formal semanticists actually do. I will argue that it does not, and neither does internalism. The central role of “entailment” and “contradiction” in semantic theory, often implicit, suggests that formal semantics should be understood as a theory of entailment for natural language. If a logic is a language plus a relation of entailment between sentences of that language then formal semantics is a theory of natural logic. If externalism is correct then formal semantics is not a theory of meaning. If internalism is correct then formal semantics is that part of the theory of meaning that pertains to a particular mode of relation between complex representations. Either way, the project of formal semantics is orthogonal to the internalist/externalist debate concepts. A relation of entailment between sentences implies a computational system that operates on linguistic structures, a transformational syntax in the original sense of the term. But a relation as such can be fruitfully studied in abstraction from how it is computed. This, I will argue, is the de facto route taken by formal semanticists and their use of terms such as “truth”, “reference”, etc. should be understood accordingly. The consequences of this view for linguistic theory as well as the theory of mind (e.g. “language of thought”) will be discussed along the way.