On the abuse of the necessary a posteriori

We shall proceed as follows. In §2, we draw out the consequences of Kripke’s account of the semantics of natural kinds terms for anyone who wants to claim that the category of the necessary a posteriori is in fact much broader than Kripke suggests. In §3 we turn to Brian Ellis’s scientifi c essentialism (SE), and argue that there are clear cases of Ellisian ‘natural kinds’ that fail to fi t Kripke’s model, and hence cannot be thought to generate the a posteriori necessities that Ellis claims to hold for natural kinds in general. In §4, we assess the consequences for SE, and argue that the lack of the relevant a posteriori necessities undermines Ellis’s own account of de re necessity. Finally, in §5, we offer a brief diagnosis.