I have two aims in this book. First, to support the thesis of the externality of mind: mind must be located outside, in exchanges among people, rather than inside, in the internal flux of representations. Second, to comprehend the difference between these two responses… (p. 2).
The Mind's Provisions is not the easiest read, even for the philosopher of mind. I do not know, because I do not read French, whether the issue is with Descombes himself or with the inevitable difficulties of translating a work of philosophy. There were passages where I was quite certain I was not following the line of argument; and I am quite certain there were other passages where what I thought I was following was not what the author (or his translator) intended.
With those caveats, I am broadly sympathetic to if not indeed in agreement with the thrust of the work, which I take to be a passionate argument against a certain naïve but widely embraced understanding of physicalism: namely, one that attempts to equate mental states with physical states, so that "two subjects are necessarily, at a given instant, in the same mental state if they are in the same physical state" (p. 229). Descombes is quite right that one can be a physicalist in good conscience (rejecting any form of substance dualism or pluralism) and still take mental states and physical states to be quite different kinds of things: not properly both called "states" at all. On the naïve view, a caveman, say, who is struck by lightning so that his neurons are scrambled into the appropriate arrangement could have such a thought as "I need to go to the bank" even though banks, and the whole cultural context in which banks are embedded, have not yet been invented. Descombes has a problem with this -- as does someone like Ruth Millikan; as do I.
Descombes' target is a school of philosophy often known as internalism or cogntivism, which takes mind as something that can be treated quite separately from world (perhaps, in an extreme case, detached from it altogether). The mind interacts not with the world directly but with representations of the world, within the independent world of the mind. Information is fed in through the senses; it is processed within the inner world (not unlike the way a computer is often described as processing information from its environment); the result produces actions within the mental world, which are then translated through motor responses into the "actual", external world. The role of the philosopher then is to understand how inner and outer worlds do or do not correspond.
In contrast, Descombes advocates a strong form of externalism, albeit it one that is more like the content externalism of Hilary Putnam than the active externalism advocated in recent years by Andy Clark and Dave Chalmers with their extended mind hypothesis. I, myself, am cool to much if not most of what gets labeled either internalist or externalist because I am inclined to believe that either position tends to lend too much ontological reality to the boundary between inner and outer, then paying attention only to its side of the line. The boundary between mind and world, it seems to me, is one that we draw pragmatically and are forever negotiating, in a way that some recent writings of Clark suggest. It is a functional not a physical boundary, and taking both to be members of one all-embracing category of "boundary" is, as I think Descombes would enthusiastically agree, a mistake.
Mental and physical are not two opposing kinds of substances, as the substance dualists would hold; nor are they two opposing kinds of properties, as per the property dualists. Rather, they are two opposing kinds of perspectives, and they lend themselves to different sorts of explanations. The physical is decompositional in a way that the mental is not: that is to say, the mental is unavoidably holist. My hand, the physical entity, forms into a fist; but I as an intentional whole -- and not my hand, nor my brain, nor any other identifiable sub-part of me -- chooses to throw the punch. (Likewise, "one cannot say: this brain professes the Platonic doctrine but has been refuted by this other brain" (p. 6).)
As fundamental as this point is, it is often missed. As Descombes notes, "the definitions often given of materialism leave aside the most important point: before deciding whether mental abilities are rightly attributed to an immaterial or to a material part of a person, one must first decide whether it is a mistake to attribute the abilities of the whole to a part of that whole, whether material or not" (p. 69).
The real strength and the delight of Descombes' (and Schwartz's) book -- once one struggles through the more difficult passages -- is the treatment he offers to some of the most influential ideas (Jerry Fodor's language of thought) and thought experiments (Putnam's twin Earth, Alan Turing's imitation game, John Searle's Chinese room) in the recent history of philosophy of mind.
Turing's imitation game is about a computer trying to fool a person into thinking she is conversing with another person (and not a computer). Given how badly the imitation game is typically misread -- often as a litmus test for intelligence -- I find Descombes' reading of it particularly insightful: "the problem is therefore one of knowing whether the questioner, having learned that he has been speaking with a machine and not a human, would lose all desire to continue conversing with such a partner or whether, on the contrary, he would continue to take an interest in his partner's psychology, knowing full well that it is a machine's psychology" (p. 111). I can almost hear Turing breathing a sigh of relief that someone, finally, got the point!
2010 Joel Parthemore
Joel Parthemore is finishing his thesis write-up as a visiting research student of the University of Lund in Sweden, and hopes to submit this summer. His primary affiliation is as a member of the Philosophy of AI and Cognitive Science research group of the School of Informatics, University of Sussex, UK. His thesis concerns enactive philosophy and theories of concepts, in particular Peter Gärdenfors' conceptual spaces theory. In his spare time, he plays with Linux systems.