Wilkerson on Natural Kinds

Abstract

While I certainly do not intend to discourage the drawing of inspiration from the works of Aristotle, when one's aim is to illuminate the practice and metaphysics of contemporary science such inspiration must surely be tempered by a passing acquaintance with the deliverances of, say, the last hundred years of scientific inquiry. T. E. Wilkerson's paper, 'Natural Kinds' {Philosophy 63 (1988) 29-42) falls regrettably short of this desideratum. Wilkerson makes a number of points about sortal terms, and proposes a number of distinctions between types of such terms. Much of this seems useful and unobjectionable. My concern here is solely with what he considers to be natural kinds and natural kind terms. Wilkerson's primary concern is with the question what it is to be a natural kind. He does not appear to see any difficulty about the question what are instances of natural kinds. The main point of this note is to claim that if Wilkerson's conception of what it is to be a natural kind is correct, then they are much harder to find than he seems to think. As to the first question, Wilkerson considers a natural kind to be a kind characterized by a real essence, and also one that provides a suitable subject for scientific investigation (p. 29). Presumably this scientific investigation will deliver 'precise' laws, since this is what is said to be impossible for non-natural kinds such as nations and banknotes (p. 30). Finally, and consequently, natural kind predicates are inductively projectible: 'If I know that a lump of stuff is gold, or that the object in front of me is an oak, I am in a position to know what it is likely to do next, and what other things of the same kind are likely to do' (p. 30).