The Semantics of Natural Kind Terms: How to Make Kripke’s Theory Work

This paper clarifies and defends the Kripkean view of na tural kind terms, with some help from Kaplan. Natural kind terms fall into two categories. Vernacular terms such as ‘water’, ‘gold ’ or ‘tiger ’ are directly referential, designate rigidly and are non-descriptive. They contrast in semantic properties with non-rigid descriptive terms such as ‘mammal’, ‘species’, or ‘acid ’ which are clearly defined. Scientific terms are descriptive. A few descriptive terms (‘the smallest prime’, ‘H 2O ’ and perhaps others) may be regarded as rigid de facto. A descriptive term locks the properties of the referent into an analytic, a priori definition, thus entailing change of meaning and referent with each new discovery or theory change. A directly referential term such as ‘water ’ has no descriptive content, thus ensu ring continuity of referent when new properties of that referent are discovered or theory change occurs. [M]y argument implicitly concludes that certain general terms, those for natural kinds, have a greater kinship with prope r names than is generally realised. This conclusion holds for certain for various species names, whether they are coun