The following papers were presented at the annual conference of the Australasian Association for the History, Philosophy and Social Studies of Science held in Melbourne in June 2001. David Armstrong raises what he sees as being the two most serious difficulties for essentialism. He argues, first, that essentialism requires the existence of properties whose identities depend on the relations that their bearers must have to the possibly non-existent events that would constitute their displays. He does not claim that such properties cannot exist. But he does argue that if essentialism requires that there be such properties, then this is a serious problem for essentialists. This is Armstrong's Meinongian objection. The second difficulty is that presented by Richard Swinburne with his infinite regress argument. This is not, as Armstrong concedes, a problem for the version of essentialism defended in my two books on essentialism. But then, says Armstrong, my way of avoiding the problem leads to other problems about the nature of categorical properties. The discussion of these problems is taken up in my reply.

Two problems for essentialism

D. M. Armstrong

In his book Scientific Essentialism (2001), and in the draft of his new book The Philosophy of Nature: A Guide to the New Essentialism, Brian Ellis has told us a lot about the advantages, as he sees them, of his essentialist view. That is fine as far as it goes, but like most large projects in metaphysics there are serious difficulties to be addressed. I find this even with my own views. I will take up what I think are the two most serious problems for Ellis’s position.