If natural kinds have microstructural essences they have them independently of rules for the application of kind terms. This suggests that what those rules are should make no difference to the essences being discoverable. I present two thought-experiments that suggest otherwise, however. Each shows an authority’s application of rules creates the appearance of there being kind essences; absent those rules, the appearance vanishes. This suggests natural kind essences are not independent of authority-sanctioned rules.

David Chalmers and Frank Jackson have promoted a strong program of conceptual analysis, which accords a significant philosophical role to the a priori analysis of concepts. They found this methodological program on an account of concepts using two-dimensional semantics. This paper argues that Chalmers and Jackson’s account of concepts, and the related approach by David Braddon-Mitchell, is inadequate for natural kind concepts as found in biology. Two-dimensional semantics is metaphysically faulty as an account of the nature of concepts and concept (...) possession. It is also methodologically flawed as a guideline for how to study scientific concepts. Proponents of two-dimensional semantics are criticized for not taking seriously semantic variation between persons and for failing to adequately account for the rationality of semantic change. I suggest a more pragmatic approach to natural kind term meaning, arguing that the epistemic goal pursued by a term’s use is an additional semantic property. (shrink)