The sorites argument against descriptivism

Abstract

In the theory of reference, descriptivism about proper names is a family of closely related theories which broadly hold that a proper name refers to an object by virtue of being associated, in the idiolect of a speaker, with a description or descriptions. The unique object, if any, that satisfies the description or sufficiently many of the descriptions is then said to be the referent of the name. It is now widely believed that descriptivism—or, at any rate, one influential version of it—was soundly refuted by Kripke's arguments against it in Naming and Necessity. According to this influential version, the so-called "cluster" theory of names, a name is typically associated with a vague cluster of descriptions. But if the cluster theory is somehow vague, then one would expect a sorites argument corresponding to this vagueness—for vagueness is typically accompanied by soriticality. In this thesis, I begin by offering a slightly more precise formulation of the cluster theory, in light of which I argue that it is the most general version of descriptivism possible. I then present the promised sorites argument against descriptivism, which reveals the descriptivist's commitment to the idea that names are vague predicates. I conclude by suggesting that vagueness is not an accidental feature of the cluster theory but rather an essential—if not the essential—feature of descriptivism itself.