Identity and Reference

Abstract

“Things are identical if and only if they have the same properties.” This definition of identity stems from Leibniz and is nowadays commonly known as Leibniz’s law. Does it state both a necessary and a sufficient condition of identity? It is at least doubtful whether it states a sufficient condition, since it is not obvious that things are logically incapable of being numerically different without differing in any other respect. The question turns in part on what is allowed to count as a property. Clearly if properties like “being identical with me” are admissible, it will follow trivially that no two different things can have all the same properties. No one who is not identical with me can be identical with me. On the other hand, if we consider only general properties, as we must do if the question is to be of any interest, then, as I have argued elsewhere,¹ there are grounds for thinking that the principle of the identity of indiscernibles is not a necessary truth. For instance, it would not be a necessary truth, if we allowed the possibility that things which are not descriptively distinguishable may yet be distinguished demonstratively.