192 Marzia Soavi
- For artifacts, there is no regularity whatsoever in the type of properties that can be copied
nor in the degree of similarity between original and copied features. That is, it is not
possible to distinguish between inheritable and noninheritable characters. - It is not always the case that items of the same artifact kind are copies of one another.
The holding of a copy relation is just a contingent historical fact that cannot be taken
to be a necessary condition for artifacts to be artifacts of the same kind. - While for biological entities the arity of the relation is widely dependent on the kind of
entities being considered (kinds with sexual or asexual reproduction), for artifacts it is
not even possible to decide how many arguments the relation has. It is possible to design
a new type of glass drawing inspiration from one, two, three, four, and so on, different
types of previous glasses.
I am not denying that sometimes a sort of weak copy relation can hold between artifacts
of the same kind due to the fact that designers draw their inspiration from previous types
of artifacts, but it is doubtful that this relation can be compared to the biological copy
relation and it is doubtful that it can play the same causal role.
Hence my conclusion that, on the one hand, if we take artifact kinds to be general kinds
like those of the glasses in the example, then the inference to the existence of a copy rela-
tion holding between artifacts is not straightforward, because either we can account for
their similarity and historical development without appealing to any copying process or
we must appeal to the designer’s intentions.
On the other hand, if we admit the existence of a copy relation that is suffi ciently weak
to cover both the case of biological entities and artifacts, we simply run the risk of using
the same term with two different meanings. For biological entities, the copy relation is a
well-defi ned relation, warranting counterfactuals and based on the existence of well-
known copying processes; for artifacts, the copy relation would not be a well-defi ned
relation, will be unable to warrant counterfactuals, and will not be grounded in any clear
copying process. Ultimately certain observed similarities among items of the same kind
seem to be the only common features shared by biological kinds and artifact kinds.
11.4 Second Option
The second criterion proposed is
- o is an object of an artifact kind K iff o is used for F.
Unlike in (1), the attribution of function according to use seems to rely mainly on the
common pretheoretic habit of speaking of artifact function in this way. The main problem
is that it is not clear if this practice perceives the attribution of functions as an attribution
to single objects or to types of objects. Both possibilities must thus be taken into account.
Henceforth I use o for single objects and O for types.