Repeatable Artworks as Created Types For Neil

objects have accidental intrinsic properties as long as the properties’ supervenience base is contingently configured too. What does the type-creationist take to be the supervenience base for repeatable artworks? The creative acts of artists are obviously part of it, but also relevant are the type’s embodiments, including the patterns of (re)production and modification. But given that these creative acts and the configuration of embodiments are contingent, the intrinsic properties of types can also be contingent. This is just to say that the truth of statements concerning the intrinsic properties of created types depends on the truth of statements concerning its 18 For sets the associated concrete objects are its members, whereas for types they are its embodiments. 12 embodiments and artists, but not on the truth of statements about other concrete objects. And this is consistent with (14). Of course, not all abstract objects which have a contingently configured supervenience base, such as {Eiffel Tower}, have accidental intrinsic properties. But from this it does not follow that no contingently existing abstracta have accidental intrinsic properties. The intrinsic properties of a type, unlike the intrinsic properties of a set, are sensitive to the properties of its supervenience base, and so there can be variation in the intrinsic properties of types, given the possibility of variation in its supervenience base. It is in this way that repeatable artworks differ from sets. What Hazlett would have to do to reinstate his argument is to show that the intrinsic properties of created types are not sensitive in this way to variation in their supervenience base. But this he cannot do. So Hazlett’s argument for the modal inflexibility of intrinsic properties does not work for created types. And this is not a surprise as (15) does not seem compelling: we knew all along that entities whose intrinsic properties supervene on the properties of other entities can have accidental intrinsic properties: simply take any macroscopic concrete object or mental life. Hazlett’s argument fails because he implicitly assumes that the intrinsic properties of an abstract object are insensitive to the properties of its supervenience base, as is the case for sets. But this generalization will not be granted by the type-creationist. Hazlett, then, is guilty of making the kind of overgeneralizing move that we noted in §2 is illegitimate.