Art works, ontology of

1. Works as physical objects

As in other parts of metaphysics, a theory of what art works are should seek to provide as simple and economical an account as possible, consistent with robust intuition about sameness and difference between works. We shall begin with maximum simplicity, introducing complications as required by consideration of particular cases, some of them real and some of them highly artificial constructs. By the end we shall have arrived at a position of some complexity, but one which has the virtue of treating all works of the kind we shall consider here in a uniform manner.

It seems relatively easy to say what a painting or a sculpture is: a canvas on a gallery wall, a piece of stone shaped in a certain way. In that case, these works are just physical objects. Three objections to this proposal have been made. The first is that art works have properties that a physical object could not have: aesthetic, expressive and representational properties, for example. This objection relies on the assumption that a physical object can have only physical properties. But a physical object might become, say, valued without ceasing to be physical, and being valued is not a physical property.

The second objection is this. If Michaelangelo’s statue David, D, is identical with a certain block of stone, B, then D and B must have the same properties. But B existed prior to D, which came into existence only when Michaelangelo set to work. So ‘existing prior to D’ is a property B possesses and D lacks. One response to this objection is to say that physical objects are four-dimensional, extended in time as well as in space, and therefore possessed of temporal parts that are themselves physical objects. In that case we can say that D is identical with a physical object B*, which consists of that temporal part of B that began when David was fashioned from the stone and ends when the degraded condition of the stone no longer warrants our saying that David exists. But now the distinct-properties argument can be used again to show that D and B* are different. It is true of D that, if Michaelangelo had not lived, it would not have existed; the same is not true of B*. Substantial deformation would destroy D but not B*.

An alternative and less ambitious proposal would be that art works (of the kinds we are currently considering) are not identical with physical objects; instead they are embodied in or constituted by physical objects. I shall not pursue this line of thought further here. Many of the arguments about identity and constitution apply equally to art works and to other kinds of things (for example, tools), and so belong to general metaphysics, as do the arguments we have considered so far. The only specifically aesthetic argument that has been developed in connection with this is an argument against both the idea that art works are physical objects and the idea that art works are constituted by physical objects. To this argument we now turn.

The claim that certain works are physical objects or are constituted by them depends on an intuitive contrast between singular and multiple works. Among works of the former kind are paintings and sculptures, where the object fashioned by the artist (the ‘authentic’ object) seems to have a unique status – a proper appreciation of the work requires that the viewer sees that object rather than any copy of it, however good. With novels, plays and poems, on the other hand, the original inscription of the work by the author (the autograph) has no special significance for appreciation of the work; any word-for-word copy of the autograph will do. But the significance placed on authenticity in painting and sculpture is due to the fact that aesthetically adequate copies of these works are very hard to achieve. Every visible feature of the work is potentially relevant to the proper appreciation of it, and so an aesthetically adequate copy of the work would have to look exactly like the original. With literature, mere sameness of spelling with the autograph is all that is required of a copy for it to allow us to appreciate the work fully. But this difference between painting and literature is a merely technical one and cannot be the basis for treating these two forms as fundamentally distinct. It is possible (though by no means easy) to produce copies of paintings and sculptures indistinguishable from their originals by the modes of perceptual access appropriate for those works. If this were frequently done, the aura of indispensability that surrounds originals would dissipate. They would be regarded as no more essential to the existence of the work than autographs of novels currently are. (Originals might continue to command high prices on grounds of their personal and historical interest, much as autographs of novels and poems do.) In that case, the claim that in painting and sculpture the work is the authentic object is only as plausible as the comparable claim that the novel is the autograph copy, and this is untenable, as we shall now see.