Anandi Hattiangadi: "An Argument Against Semantic Supervenience”

  • Date: –12:00
  • Location: Engelska parken - Eng2-1022
  • Organiser: Department of Philosophy
  • Contact person: Matti Eklund
  • Seminarium

The Higher Seminar in Theoretical Philosophy

Anandi Hattiangadi, Stockholm University: "An Argument Against Semantic Supervenience"

What makes it the case that an arbitrary representation has the semantic properties that it does rather than some other semantic properties or none at all? What makes it the case that ‘Socrates’ refers to the man, Socrates, or that the thought that table is covered with books, is of or about a certain table?

These questions belong to foundational meta-semantics, the investigation of the grounds or determinants of the semantic properties of representations—such properties as meaning, reference, content, truth conditions, and the like. It is almost universally held that the semantic properties and the truths describing how they are distributed in the world are not fundamental; that if there are any semantic properties, then they are in some sense ontologically dependent on some class of non-semantic properties. Though there is no consensus on exactly which class of non-semantic properties the semantic properties ontologically depend on, and there is no consensus on whether the semantic truths are identical to, determined by, grounded in, or merely supervene on the non-semantic truths, there is a widespread collective commitment to the weakest of these claims, the supervenience of the semantic on the non-semantic:

Semantic Supervenience. The semantic truths supervene on some class of non-semantic truths.

I argue in this paper that champions of Semantic Supervenience face a dilemma: if the semantic truths are said to supervene on purely qualitative, non-semantic truths, then Semantic Supervenience is false. On the other hand, if the semantic truths are said to supervene on all truths, both qualitative and non-qualitative, Semantic Supervenience is a trivial consequence of the logic of identity, and is compatible with the view that semantic properties are fundamental.  The dilemma arises on the assumption of haecceitism—roughly, the view that qualitatively indistinguishable possible worlds can differ in non-qualitative respects. First, I argue that a wide a range of influential theories of reference and singular thought entail haecceitism. Second, I present a dilemma for champions of Semantic Supervenience.