Supervenience Argument

Many philosophers have worried that physical causation may “exclude” mental causation—physical and mental causes “compete” for efficacy, and because of some principle (the causal closure of the physical domain, for example), physical causes inevitably “win”. This picturesque language is common, but explicit arguments, which identify prima facie plausible principles from which epiphenomenalism about the mental would logically follow, are less so. Jaegwon Kim’s “supervenience argument” represents perhaps the most influential attempt to construct such an argument. In Kim’s case, the argument is presented as a reductio ad absurdum of nonreductive physicalism. The present paper is an attempt to get clear on just what Kim’s supervenience argument is, and how, or whether, it works. We will only discuss Kim’s most recent formulation of the supervenience argument, or arguments—Kim gives us two “versions” of the argument (“Completion 1” and “Completion 2”). This formulation is found in ch. 2 of Kim’s most recent book, Physicalism, or Something Near Enough (Kim 2005; henceforth “PSNE”), in which, according to the chapter title, “The Supervenience Argument [is] Motivated, Clarified, and Defended”. We will attempt to give a reconstruction of the arguments that renders them deductively valid, so that every assumption on which we believe Kim relies is made explicit. We will show that the premises of Kim’s arguments are not nearly as innocent as they seem, and that one assumption in particular, which concerns the identity conditions of events, can only be viewed as begging the question against nonreductive physicalism, as that position has traditionally been conceived. Whatever else they may be, Kim’s arguments then are not reductios of nonreductive physicalism. Nor do they succeed in posing a problem about the possibility of mental causation by showing that we must either accept epiphenomenalism or reject one or another of a set of prima facie plausible metaphysical principles. Kim takes himself to have shown that the claims he calls his “substantive premises”, which are indeed prima facie plausible—at least relative to prevailing assumptions about causation, on which we will comment in the final section of this paper—are not consistent with the claim that mental events have causal efficacy. But what he has in fact shown is the inconsistency of a larger set of claims, some of which have no particular prima facie plausibility, even relative to the prevailing assumptions. We think the aforementioned assumption about the identity conditions of events is the least plausible among these.

Kim thinks his arguments show that the following claims cannot all be true: