Abstract

Mereological Universalism is the thesis that, for any disjoint Xs, the Xs automatically compose something. In his book, Material Beings, Peter van Inwagen provides an argument against Universalism that relies on the following crucial premiss: (F) If Universalism is true, then the Xs cannot ever compose two objects, either simultaneously or successively.1 I have argued elsewhere (Rea 1998) that van Inwagen’s defence of (F) fails because it relies on the false assumption that Universalism is incompatible with the view that, for some Xs, what the Xs compose depends upon how the Xs are arranged. However, Matthew McGrath (1998) has recently provided a new – and in his opinion, better – formulation of van Inwagen’s argument for (F). Furthermore, he claims (contrary to what van Inwagen himself apparently thinks) that four of the ten assumptions listed at the outset of Material Beings are ‘jointly sufficient for the falsity of Universalism’. (1998: 121) Those assumptions, as they appear on page 121 of McGrath 1998, are as follows.