For and against Descartes

SIMPLE MINDEDNESS. In defense of naive naturalism in the philosophy of mind. By Jennifer Hornsby. 265pp. Harvard University Press. Pounds 23.50. - 0 674 80818 5.

These days there is so much agreement within analytic philosophy about the truth of some sort of materialist theory of the mind that it is hard to believe that just forty years ago the theory was written off as an Australian heresy. U. T. Place, J. J. C. Smart and David Armstrong, once all thought to be so very sadly mistaken, are now widely hailed as the prophets of orthodoxy.

The materialist orthodoxy, like any orthodoxy once generally accepted, is prone to careless formulation and uncritical acceptance. Wherever you find orthodoxies, you should therefore also hope to find heretics of a certain sort, people who are willing and able to play the unenviable task of keeping those who accept the orthodoxy honest. Though she wouldn't like the label "heretic" - she styles herself a defender of common sense - Jennifer Hornsby has played just this role in a series of careful and insightful papers written over the past twenty years. In Simple Mindedness, she does us the great service of collecting twelve of these papers together in a single volume. Nine of the papers have been previously published, three appear in print for the first time. Her overall picture of the mind is filled out in a helpful introduction, and in a series of useful postscripts.

Materialism was set up in opposition to a dualist theory of the sort favoured by Descartes. Descartes thought that people were made up of two very different sorts of substance: a material substance, an extended thing which has a height, a weight, a shape, and so on, and an immaterial substance, a thinking thing which has beliefs and desires and sensations. The material substance was supposed to be the bearer of people's physical properties, the immaterial substance was supposed to be the bearer of their mental properties. Goings-on in the immaterial substance, mental states and events, were supposed causally to interact with goings-on in the material substance, physical states and events.

For their part, materialists agree that someone's wanting to light a candle and believing that they can do so by moving a lighted match towards the wick is the cause of various non-mental events, events such as bodily movements, the subsequent movement of the match towards the wick, and the like. But they think that the very fact that we can in this way redescribe mental states and events as the causes of non-mental states and events gives the lie to Descartes's claim that they are goings-on in an immaterial substance. Rather, according to materialists, these causes are one and all material; indeed, they are neurophysiological.

Hornsby disagrees with both Descartes and the materialists. Against Descartes, she denies that people are composed of a material and an immaterial substance. Instead, she thinks, much as materialists do, that people are unitary objects: naturally occurring biological organisms, the distinctive nature of which is to enjoy both mental and physical properties. Against the materialists, however, she denies that mental properties reduce to physical properties. Instead, she thinks, much as Descartes does, that mental properties are autonomous. Hornsby thus distinguishes herself from Descartes by rejecting substance dualism, and from materialists by embracing attribute dualism (though note that "attribute dualism" is my terminology, not hers; she prefers "naive naturalism").

Hornsby's rejection of substance dualism is welcome and understandable. Her argument for attribute dualism, however, which largely focuses on the failure of arguments for materialism, is harder to evaluate. For example, materialists standardly argue that mental states, which we take to be causes of material states and events, must be identical with material states, on the grounds that to think otherwise, given that a complete story of the causal antecedents of material states and events can be told in material terms, is to suppose (incredibly) that the material states and events thus caused are systematically overdetermined; that is, that they have two distinct causes each of which is causally sufficient. Hornsby's response is that there is nothing in the premisses - nothing in our conception of a mental state, or a physical (or material) state, or a bodily movement, or causation, or completeness, or anything else - to force the conclusion of identity on us. In this way she makes out her case for the autonomy of the mental. Thus, for example, she reminds us that in the theory of persistence there is a plausible tradition according to which the matter that occupies the same spatio-temporal region as that occupied by a statue is not identical with the statue, but rather constitutes it. The implication is that the same might be true of the relationship between the mental and the physical.

Even if she were right about this, however, it is hard to see why this sort of autonomy should worry materialists. Materialism is, at bottom, the view that the mental "supervenes" on the material. In other words, if we imagine making an exact atom-for-atom copy of our own world, then, according to materialists, we wouldn't just create a possible world with the same material nature as our world, but would create a world with the same mental nature as well. In its strong form, attribute dualism is inconsistent with this. It holds that mental properties are autonomous in the sense that they float quite free of the material; possible worlds which are material duplicates may yet differ in their mental nature. Materialists reject this strong form of attribute dualism, on the grounds that it is vulnerable to the overdetermination argument. But whether Hornsby accepts it, I am not sure.

On the one hand, her objections to the standard arguments for materialism are consistent with rejecting it. The claim that the mental is materially constituted, as opposed to being identical with the material, would, after all, suffice to guarantee the truth of the supervenience thesis. Materially constituted mental properties could no more float free of material properties than facts about statues could float free of facts about their constituent matter. It would also suffice to answer the argument from overdetermination. No one thinks that a death is overdetermined simply because it was caused by a falling statue, and hence also caused by falling matter. Weak attribute dualism, a view which merely substitutes talk of constitution for talk of identity, is thus best seen as a form of materialism, a form which incurs the standard materialist burden of explaining how the material constitutes the mental. If Hornsby were a weak attribute dualist, then we might hope to see her helping materialists try to give such explanations. We don't see her doing this, however, and I suspect that is because, on the other hand, she is sceptical about supervenience theses of the sort just described. Indeed, she makes various disparaging remarks about the supervenience of the mental on the material in her essays. The leading idea seems to be that the widespread acceptance of such supervenience is based on a confusion about the nature or scope of the material. If she were right about this, then, of course, materialism would be in deep trouble, eluding precise formulation.

The precise details of Hornsby's own view, and whether or not she is right, is best left for individual readers to judge. What can safely be said here, however, is that materialists who put in the time and effort to make that judgment for themselves will be richly rewarded. There is much that an orthodox materialist can learn from reading the heretical Hornsby.