Thursday, March 5, 2009

Searle and property dualism

David Lewis once wrote that “philosophical theories are never refuted conclusively. (Or hardly ever, Gödel and Gettier may have done it.)” To this list should be added John R. Searle, who has in my estimation conclusively refuted the computationalist theory of mind – not so much with his famous “Chinese Room” argument, but with the less well-known but far more devastating arguments presented in his paper “Is the brain a digital computer?” and in chapter 9 of his brilliant book The Rediscovery of the Mind. When (not if, but when) the philosophers and psychologists of the future look back at the bizarre fad for characterizing the brain as a kind of computer and the mind as software, and ask “So what the hell was that all about?”, Searle will be remembered as the man who did more than any other philosopher to break the spell of this illusion.

Searle is also an effective critic of other materialist theories of the mind. But though he rejects all extant forms of materialism, Searle also famously denies being any kind of dualist. Still, his critics regularly insist that his views nevertheless entail dualism whether he realizes it or not, and that this suffices to show that they are mistaken. In short, Searle says: “My arguments are correct, and they do not entail dualism,” while his critics say: “Searle’s arguments do entail dualism, and therefore they are incorrect.” In my view both sides are partly right and partly wrong: Searle’s arguments are correct, and they do entail dualism.

As my longtime readers know, the version of dualism I think one ought to accept is Aristotelian-Thomistic hylemorphic dualism. As it happens, Searle’s views have been compared by some commentators to Aristotle’s (see e.g. Alan Code’s essay in Lepore and van Gulick’s John Searle and his Critics). But Searle rejects any such interpretation. (At a conference at which Searle and I were both presenters, I gave a paper the first part of which put forward a diagnosis and critique of naturalistic theories of the mind, and the second part of which proposed a return to hylemorphism as a remedy. Searle called the first part “brilliant” and the second part “crazy.” Coming from a man whose work I admire so much, that was good enough for me. The first part, incidentally, would go on to become “Hayek the cognitive scientist and philosopher of mind,” in The Cambridge Companion to Hayek. The second part was merely a sketch of ideas that I have developed in more detail elsewhere, most fully in The Last Superstition and in the forthcoming Aquinas.)

Indeed, when Searle has worked his positive views out more fully, the version of dualism they end up resembling most is property dualism. But Searle rejects this interpretation as well, arguing against it at length in his article “Why I am not a property dualist.” Victor Reppert kindly linked yesterday to a paper of mine, “Why Searle is a property dualist,” which replies to this article of Searle’s. Since Searle’s essay has just been reprinted in his new anthology Philosophy in a New Century, I thought I would post a link of my own to my reply, for what it is worth.

(Bonus link: Here is an interview Steven Postrel and I did with Searle for Reason magazine some years back. Among other things, it gives a good sense of Searle’s political views, which aren’t quite the sort you’d expect from a UC Berkeley professor.)


  1. Dr. Feser:

    Are you aware of any contact Searle has had with Thomists? This post suggests such contact.

    Also, could you please talk about McGinn's views of consciousness and naturalism?

  2. Oops. LINK:

  3. Very interesting, Cogitator. No, I had never heard such a thing. I gather from the paper Shea links to that the Thomists in question were Fadok and perhaps some other Thomists at the Dominican school in Berkeley. Perhaps Fadok sought Searle out for comments on his M.A. thesis? Anyway, the thesis itself seems interesting and I will take a look at it.

    Time permitting, perhaps I'll post something on McGinn at some point.

  4. One question I have, as a philosophy of mind novice...

    I now and then hear about 'emergence' positions in PoM. Which I take to mean that, at certain thresholds of complexity/order in a system, distinct new traits appear that you could not get (or perhaps not even predict) from the lesser, reduced state.

    My thought has been, though, that 'emergence' sounds an awful lot like formal causality. Since it stipulates 'You get these powers/traits in a system only at or above these thresholds/forms'. Am I on the right track there, or completely off?

  5. Hello Anonymous,

    I can see why you'd think there might be a similarity, but I think the term "emergent" is misleading in this context and should be avoided. The point has nothing to do with complexity, nor with new forms "emerging" out of old ones through some unusual natural process. From an Aristotelian-Thomistic POV, rational animals cannot even in principle "emerge" from non-rational ones. The reason is not that human brains have crossed some threshold of complexity, but because intellectual activity simply differs in kind, and not degree, from the sensation and imagination that lower animals are capable of.

    A longer answer would have to explain the difference between a hylemorphic vs. a mechanistic understanding of the natural world (for which explanation, see The Last Superstition). "Emergence" theories essentially take for granted a mechanistic view, which is why they are incommensurable with the Aristotelian-Thomistic approach.

  6. Thanks for the response! I'll reread TLS and try to get a better grasp of this. I think I see what you mean about mechanistic versus AT at this point, but it's still tough.

    Also, I just read your response. Again, as a novice, it backs up the first impression I had when reading about Searle on this topic. I came away thinking 'He seems to want to deny dualism by admitting to the existence of things/states dualists assert, but just calling them physical anyway.' It seems like he may as well just argue that property dualists aren't really dualists.

  7. Dr. Feser,

    I recently bought "Minds, Brains, and Science" at my local used book store ( for 2.95!). I noticed that in the second chapter he defends the view that computers cannot think because they only have syntax, not semantics. This left me scratching my head, because it seemed as if he was simply making the subjective/objective property distinction that dualist make. How can one have semantics if we do not have minds?! I also found his analogy between water and mental states to be strained at best. I just finished reading your essay "Why Searle is a Dualist" and I found myself agreeing. You put it in better terms than me, but I had the same sort of questions.


  8. Ed

    Steven Horst is also a brilliant critic of the Computational theory of Mind as well -

  9. Given that the 'mental' features that Thomists claim cannot 'emerge' from anything else, perhaps one could rephrase the question as a question about those 'mental' features that Thomists agree that human beings share with non-human animals. These features (perception, desire, consciousness, intentionality) are usually thought to be problematic for materialists -- and so either eliminated, 'reduced' in some strong sense, or taken on in arguments against materialism. The response to the emergence question about those features that make an animal rational aren't available in this case -- so what does the Thomist want to say?

    One problem with 'emergence' is that there doesn't seem to be any clear picture of what it means. It seems wise, though, to distinguish (as our previous anonymous did not) between epistemological concerns (we could not predict that certain things A, B, and C, when arranged in the requisite fashion, would have features X, Y, and Z) and more strictly ontological issues (e.g. it is in virtue of the structural organization of A, B, and C that the entity constituted by them has features X, Y, and Z). The truth of the epistemological claim in any case has no bearing on the truth of the ontological claim. It certainly seems to me that at least some versions of emergentism come close to re-stating a basically Aristotelian view of the issue. If 'mechanism' is supposed to involve commitments to eliminativism or strong reductionism about all higher-level properties, mental or not, and to the rejection of any form of final causality (and not simply to a weird sort of vitalist teleology), then it's hard to see how one could possibly be a mechanist and an emergentist.

  10. I think there's a non-obvious flaw in these "syntax is not semantics" arguments against materialism/monism/whatever.

    What is semantics? Suppose I'm not willing to commit to a definition of semantics, and I'm not willing to say precisely what it means for a mind to fix the meaning of a syntax.

    In that case, I would be saying that we minds have some "semantic ability" that I haven't defined, but that I somehow know I have. I would be saying that intentionality is some supernatural (=inexplicable) property.

    Now, as soon as I fail to make my commitment to define intentionality, it is a foregone conclusion that I will forever fail to find (undefined/undefinable) intentionality in physical systems. Clearly, I had better step up and define what semantics is really all about or else I'll just be begging the question in every argument I make against naturalism. I don't see Searle defining what intentionality/semantics actually is. He just leaves it floating around out there as if we all know perfectly well what he's reasoning about when we don't (and, presumably, neither does he).

    I don't think it's difficult to devise a physical or scientific model of intentionality that will fix syntax to semantics (for a particular mind). I think one should start out by asking, "How do I know when my own (human) thought about X is really about X?"

  11. You write very clearly.

    Something I have never understood about the "The Homunculus Fallacy" (to quote Searle) is why this is thought to be a fallacy afflicting those who oppose materialism rather than vice versa.

    We all observe that we have minds that are empirically Cartesian (as described in Descartes' "Meditations"), if not analytically Cartesian, so if materialism cannot account for our empirical minds without the introduction of a non-existent homunculus then materialism is false.

    See Materialists should read this first

  12. Speaking of John Searle, you might find this post from physicalism is dead to be interesting: