This paper offers an interpretation of Plato's Cratylus 427d1-431c3 that supports a reading of the dialogue as a whole as concluding in favour of a conventionalist account of naming. While many previous interpretations note the value of this passage as evidence for Platonic investigations of false propositions, this paper argues that its demonstration that there can be false (or incorrect) naming in turn refutes the naturalist account of naming; that is, it shows that a natural relation between name and nominatum (...) is neither a necessary nor a sufficient condition for reference. Socrates secures this outcome by using demonstratives and their concomitants to show how any putative natural imitative link between name and object may be overridden. Furthermore, Socrates' employment of demonstratives and context-dependent statements in his case-studies of false naming speaks in favour of a reading of this passage as primarily focussing on naming rather than on propositions in general. (shrink)

In Plato's Cratylus, Hermogenes contends that the correctness of names is conventional. Appealing though this claim sounds to modern ears, it does not meet with approval in the Cratylus. Why? I argue that the conventionalism promoted by Hermogenes is discredited by unacceptable relativist implications because it incorporates the mistaken assumption that correct names are individuated exclusively by their phonetic composition.