Abstract
In 1976, Hilary Putnam set forth his model-theoretic argument, claiming that it showed that the semantic realist’s program1 was ‘unintelligible’, since it implied, contra the realist view, that reference is radically indeterminate. Although I find the conclusion that reference is indeterminate unattractive, I argue that the descriptivist position needs to be supplemented with a premise about the sorts of kinds or structure that our world includes. The need for this premise gives a counterintuitive result: the descriptivist account of reference makes the very possibility of determinate reference contingent upon the kinds that happen to make up the world. The need for this premise paves the way towards a new kind of skepticism about what we may and may not assume about the nature of the world.