Abstract

Kripke's argument for the rigid designation of natural kind terms is fallacious because he does not distinguish natural kinds from second-order functional properties; by clarifying the concepts of natural kind and functional property, we can show that natural kind terms do designate their referents rigidly, but that functional property terms are not rigid designators. My discussions of functional property will also help dispel the worry about the alleged cases of contingent identity with regard to theoretical statements in science. There is no contingent identity even in the form of second-order logic: Property identity is also a necessary identity. The principle of necessary identity rules relentlessly