In this paper, I argue that ‘art’, though an open concept, is not undefinable. I propose a particular kind of definition, a disjunctive definition, which comprises extant theories of art. I co-opt arguments from the philosophy of science, likening the concept ‘art’ to the concept ‘species’, to argue that we ought to be theoretical pluralists about art. That is, there are a number of legitimate, perhaps incompatible, criteria for a theory of art. In this paper, I consider three: functionalist definitions, (...) procedural definitions, and an intentional-historical definition. The motivation for this pluralism comes from an analysis of practice, because the term is of apparent value to practitioners. However, a closer analysis of the concept reveals that, while disjunctive definitions help us to understand how we use certain terms, they lack ontological import. In sum, I attempt to glean lessons from the philosophy of science about the philosophy of art. If my analysis is correct, we ought to be eliminative pluralists about art as a concept. (shrink)