Defining “art” has been a central occupation of philosophers of art throughout the history of the subject. This is quite natural. The range of things that have been talked about and treated as art in one context or another at some time or other is bewilderingly vast, including such items as popular films, Greek urns, experimental poetry, cathedrals, novels high and low, gardens, works of fashion, jokes, string quartets, sonnets, shields, ballets, photographs, and on and on. Reputations, modes of appropriate attention and discourse, places in curricula, and prices in various ways turn on what is treated as art how and when. If it is not the business of philosophers of art to analyze the concept of art so as to sort things out a bit, it is hard to see what their business ought to be. Taking up this business, philosophers have proposed definitions that variously treat (mimetic) representational content, (significantly absorbing) form, (well-wrought and shareable) expressiveness, or certification by accredited authorities as a necessary and sufficient condition for art. Nöel Carroll provides a useful survey of such efforts at definition in his recent Philosophy of Art, concluding, alas, that they one and all founder on counterexamples.