General Names and Particulars

Abstract

I shall disregard the metaphysical implications of Locke’s theory of essences, historically interesting though these are, and consider it simply as a theory of language. Locke’s view is that an essence, whether or not the properties which compose it have ‘an union in nature’, is a simple, unstructured list of properties. This is most certainly not a view of merely historical interest: indeed, it all but exhausts presently available wisdom on the subject. Very many contemporary philosophers write and talk as if to know the meaning of a general name (to possess a sortal concept in Locke’s sense of ‘sortal’) were to know a list of characteristic properties which things of that sort possess. The main difference between Locke’s account and present-day discussion is that most contemporary philosophers following — or, rather, assigning — a certain interpretation to Wittgenstein’s remarks on ‘family resemblances’ would deny that a thing need exhibit all of the properties corresponding to the name of a given sort, in order to be a thing of that sort. The concept of a game, for example, it is commonly said, covers a very large number of different sorts of activity. There need be no common property which all these activities possess and in virtue of possessing which they are all games, provided that there is a list of properties some large selection of which a thing must possess to be a game.