Production Determines Category: An Ontology of Art

1. Are There Singular Artworks?

It is a mainstream view within the ontology of art that there are singular as well as multiple artworks, but it is also a view that is contested. In what follows, I will investigate whether the singular/multiple distinction can be sustained and will argue for a new way to determine the category to which an artwork belongs. Though this seems to be an enterprise of rather theoretical concern, it has a direct practical dimension to it as well. Whether we are right to distinguish between singular and multiple artworks, and on what grounds such a distinction could be made, is of course affecting our treatment of artworks and our whole practice of exhibiting, selling, and appreciating works of art because it is based on the assumption that certain works are unique (and thus exhibited in museums and sold for considerable prices), whereas others exist in several instances (which make them widely available and not to be identified with a single object of appreciation). I hope to show that our practice of categorizing art is mostly appropriate but for different reasons than it has been suggested before.

The first question to consider is, What defines a singular artwork? The most obvious characteristic is that it is singular—that is, it exists in only one instance. There is only one Mona Lisa, and there cannot be multiple instances of the Mona Lisa, whereas there can be multiple instances of Ulysses. But a little more precision is advisable. The criterion to decide whether a given work is singular cannot be if there actually is just one instance of the work.¹ There might be a multiple artwork, say the novel x, of which there is only one instance—the manuscript. But this does not make x a singular artwork because there could be many copies of x, and they would all be instances of x if they fulfilled the necessary identity criteria between type and token.

A copy of the Mona Lisa, on the other hand, would not—according to the common understanding of "singular artworks"—be an instance of the painting but just that: a copy of the Mona Lisa. Thus, a singular artwork can only exist in one exemplar, and it is impossible that there is more than one. There can only be copies, reproductions, posters, forgeries, facsimiles, or pictures of the Mona Lisa but not an exemplar of the Mona Lisa other than the original. Talking of one exemplar or instance of the Mona Lisa might even be slightly misleading, as it is not supposed to suggest that there is some abstract entity Mona Lisa of which Leonardo's painting is the one and only possible instance. The claim is rather that in the case of singular artworks, there is no type the painting would be a token of at all but only the original physical object. The original is the work.

The view that paintings and sculptures are singular artworks that are to be identified with physical objects is held by many ontologists of art, including Richard Wollheim, Jerrold Levinson, and Nicholas Wolterstorff. The view is not unchallenged, however, and some people think that what would normally be considered as singular artworks can—at least in principle—exist in several instances as well. In other words, there are no singular artworks at all. And if there could be several instances of the Mona Lisa, this not only means that the Mona Lisa is not a singular work of art but also that it is not a physical object; rather, it is an abstract object like a literary or musical work.² This view is defended by, amongst others, Gregory Currie, Maria Reicher, and Peter Strawson.

But let us first look at some reasons why one might want to insist that some artworks are concrete particulars. To begin with, this is how we deal with and think about paintings and sculptures. The assumption that, say, paintings are singular artworks is implied in our social practices concerning paintings. We put the originals in museums and not copies of them. We buy and...