Friday, June 29, 2007

Dualist Explanations

Peter writes:
Chalmers posits that the non-physical mental properties parallel the information processing properties of the system. But if they parallel them perfectly, and thus explain the mind, why not just identify them? ... the dualist explanation posits something more than the materialist version of the same theory does: it must posit additional laws governing a new domain of mental stuff that makes it behave in this way and stick to the right sort of physical systems.

On the other hand, the (type B) materialist theory posits ad hoc 'strong necessities', which we have no independent reason to believe in. (Kripke's "necessary a posteriori" is no help.)

Consider the question: why aren't we non-conscious zombies, mere hunks of matter that exhibit complex behaviour without any "lights" on inside? The materialist answers that zombies are impossible; that consciousness is nothing above and beyond the complex arrangements of matter that our bodies (brains) comprise. But this strikes me as an unsatisfying explanation, that doesn't really do justice to the phenomena.

The dualist can do better. She may acknowledge the depth of the problem -- that consciousness is something new, something that goes beyond merely material properties. She can also acknowledge the modal fact that zombies (non-conscious physical duplicates of ourselves) are possible. So, rather than merely rebuffing the question "why aren't we zombies?" as empty or ill-formed, the dualist takes it seriously, and offers an answer:

The reason we're not zombies is because of the contingent natural laws that govern our universe. There are psycho-physical bridging laws, which ensure that matter gives rise to consciousness. (Note how intuitive this claim is: we think that consciousness emerges from the brain; not that it just is the brain!) The zombie world has no such bridging laws. Its laws are merely physical, so that brains and other matter causally interact without giving rise to genuine consciousness in addition. That's the difference.

Materialists can't explain this difference, because they don't take the zombie intuition seriously. Once the brain matter is there, they think that's all there is to consciousness -- there's nothing further to explain. Most of us think there is something still to be explained, and dualism can achieve this by positing bridging laws that cause 'mind' to emerge from 'matter'.

Even dualists can agree that in our world (i.e. given the actual laws of nature) complex brain states suffice for consciousness. The briding laws make zombies nomologically impossible. And that's all science is concerned with. As philosophers, though, we're interested in a broader sense of possibility, in which we can't just take the natural laws for granted. So, once our familiar psycho-physical bridging laws are taken away, we should ask: does matter alone suffice for consciousness? The zombie world demonstrates that the answer is no. Take away the bridging laws, and our physical stuff might no longer give rise to any conscious experiences.

In summary, it's worth emphasizing three points:

(1) Materialism - perhaps surprisingly - turns out to be theoretically extravagant, due to its modal ambitions. It posits 'strong necessities', which we have no independent reason to grant, and indeed goes against everything else we know in philosophy. Dualism is thus the more philosophically modest theory.

(2) Additional laws are worth positing, to explain why we're conscious rather than zombies. (The unsatisfying alternative is to merely dismiss the question.)

(3) Contrary to popular belief, dualism need not be in tension with science. It only diverges from materialism in its extra-nomological implications -- i.e. matters that concern philosophers, not scientists.


  1. I'm just going to respond from within the context of my project:
    So, a few points
    1- we don't have to explain intutions; intuitions are often wrong. (Although we might want to explain why we have bad intuitions in the context of psychology)
    2- materialism doesn't posit modal laws, materialism doesn't need to say anything about modality at all, just as economics doesn't have to posit modal laws.
    3- your points seem to imply that materialism must answer logical objections; it doesn't in my project, it only needs to answer explanatory objections. It is perfectly OK for materialism to answer: "why can't zombies exist" by appealing to its definition of consciousness.
    4- see also: and

  2. N.B. my linked post on Kripke pre-empts the stock objections to conceivability-possibility inferences that you reference.

    "materialism doesn't need to say anything about modality at all"

    It needs to claim that zombies are metaphysically impossible. Unless you think they're also conceptually incoherent, then this commits the materialist to what Chalmers calls 'strong necessities' -- a bizarre form of the necessary a posteriori that is discussed in more detail in the linked chapter of my thesis.

    It's probably necessary to read those background materials if you want to understand my argument.

  3. And I claim that what I have written, especially in : defeats those arguments. I've read your post on Kripke, and I don't see where it refutes my claims.

  4. Your post makes three objections. Two only concern non-ideal conceivability, and the third appeals to a standard Kripkean case (heat = molecular motion). My post gives a rough account of why those sorts of cases (I discuss 'water = H2O', but the same principles apply) don't really cast any doubt on our a priori grasp of modal space. See also chapter one of my thesis for the full argument.

    Here's the key idea in outline: "Modal rationalism, recall, is the thesis that we have a priori access to the space of possible worlds. The Kripkean counterexamples to CT suggest that there are necessary truths beyond our a priori reach – statements [such as 'heat = molecular motion'] may be true in all possible worlds without our realizing it. But there are two ways to account for this: perhaps we are ignorant of what modal space contains, or perhaps we are merely ignorant of how to describe it. Only the former poses any real threat to modal rationalism; but it is the latter account that we should accept." (section 1.3)

    I go on to explain why. Feel free to comment there if you see any flaws in the argument.

  5. I think you are misreading my argument. Although I use a typical Kripkean example the reasoning behind the argument is anything but Kripkean, since I am opposed to rigid designation in general. So perhaps you should re-read my claim.

    My claim, in simple terms, is that what matters for identity is the possibilities for real heat, not the possibilities for possible heat, and that it is only the possibilities for possible heat, not real heat, that we have conceptual access to.

  6. No, I can freely grant those claims. My point is that the difference between "real heat" and "possible heat" (i.e. other stuff that could have, but doesn't actually, fill the heat role -- let's call it "apparent heat") is mere semantics. There's still a possible world where heat-like stuff (apparent heat) is something other than molecular motion, and that's enough for modal rationalism. Ideal conceivability reveals the qualitative nature of possible worlds; further facts about the "true identities" of things need not concern us.

    Consider: Your line of argument suggests that we can identify the possibilities for "possible" or "apparent consciousness" (i.e. the things that could have satisfied our concept), but not for "real consciousness" (i.e. the actual referent of our concept). So, when we conceive of the zombie world, there's a real possible world there, but the lack of any apparent consciousness in its qualitative description doesn't mean it lacks "real consciousness" (i.e. the stuff that satisfies our concept in the actual world -- brain states of a certain sort).

    The obvious problem here is that consciousness, unlike heat, is conceptually transparent (purely qualitative, or "semantically neutral") -- there's no gap between "seeming pain" and "real pain", for example. Appearances are enough to determine the reality about how things seem!

    And even if you dispute the above conceptual analysis, the fact remains that the zombie world lacks something that our world has -- namely, the superficial, qualitative property of apparent consciousness.

    In short: nothing you've said undermines the modal rationalist's claim that conceivability reveals the qualitative possibilities (abstracting away any deep facts about the identities of things), and that's all we need to refute materialism.

    (But again, I can't really do justice to the argument in a blog comment, and I did write a whole thesis on it, so you're very much encouraged to check that out. See also Chalmers' Does Conceivability Entail Possibility? That fact that you call his position "obviously silly" indicates that you haven't actually read him.)

  7. Well I'm not going to accept the premise that consciousness as a whole is conceptually transparent, given the well known problems with introspection.

    And I don not accept that the zombie world lacks something that our world has, I would claim that the zombie world is impossible, given that what we mean by consciousness is real consciousness. Real consciousness is identical (claim) to some kind of functional property, therefore there are no zombie worlds.

    To argue from the zombie world to possibilities about real consciousness is to assume what you need to prove, that consciousness isn't identical with some functional property, and thus beg the question against materialism.

    And yes, I have actually read Chalmers, and I do think his claims are obviously silly (but not to Chalmer's himself obviously). I think that Chalmers is biting a bullet, one that I don't think it is rationally acceptable to bite.

  8. Actually we can just head this whole thing off at the pass. I don't know how you concieve of consciousness, but I concieve of consciousness naturally as a kind of process occuring in the brain (I was raised very scientifically). This means that I concieve of consciouss as functional property of material objects (I might be wrong, but that is how I concieve of it). Thus I can't concieve of a zombie world. I can entertain the possibility for the sake of argument, but I can't conceive of it. Thus I conclude, by the arguments you endorse, that consciousness is necessarily material. The End. Of course you could argue that I have a faulty conception of conciousness, that my conception of it is distorted by the theories I subscribe to, etc, but I could say the same about your conception of consciousness. Thus we are at an impasse, the very possibility of which reveals that concievability is a poor guide to possibility even in the case of consciousness.

  9. Finally, a note on Chalmers. Even in Chalmers own work he specifically points out that concievability is a guide (at best) to metaphysical possibility. However my argument is that (at least in the case of heat) it is physical possibility that matters when determining what in this world heat is identical to. And I would argue that facts about what consciousness is to be identified with are also to be determined by physical possibility. Hence Chalmers, even if right, completely bypasses my claim.

  10. Introspection is a red herring; the claim is that the qualitative properties by which we identify consciousness are precisely what we mean by 'phenomenal properties'. (By contrast, the properties by which we identify heat are conceptually distinct from heat itself.)

    "I would claim that the zombie world is impossible, given that what we mean by consciousness is real consciousness."

    I was using the name 'zombie world' to denote whatever possible world we were qualitatively conceiving of, which contains apparent zombies. You have no grounds whatsoever for denying that there is indeed such a world, and that our conceivings accurately capture it's qualitative properties. Just as there is a possible heat-vacuum world, where heat-like stuff (not molecular motion) can pass through a vacuum.

    Ah, just saw your new comment: "I conceive of consciousness as a functional property of material objects"

    Okay, you're no longer speaking English, but that's no problem. I want to introduce to you a new concept: phenomenology, or "what it is like"-ness. This is the concept of how things seem, from a first-personal perspective. (That's the concept; nothing explicitly about functionality or material objects. It's conceptually possible for this concept to be instantiated in the absence of any material objects whatsoever!) Hopefully you have an intuitive grasp of what I'm talking about here. If not, then you may reasonably doubt whether this concept of ours corresponds to anything in the actual world. The other 99% of us are more directly acquainted with it than we are with anything else -- but you won't believe us, so further discussion is impossible. But, assuming you're a normal human, you should know what I'm talking about, and have no trouble imagining a world which is physically just like ours, but for which there is nothing it is like for the people living there. My previous arguments will then go through from here.

    Oh, a third comment appeared! Let me address that now. Firstly, you misrepresent Chalmers. He holds that ideal conceivability is a perfect guide to "primary" possibility. This is quite explicit in the linked paper. (Also, he was my thesis supervisor.)

    On your substantive point: conceptual analysis reveals that heat is whatever actually fills the heat role (it is a functional concept). So we can tell a priori that there's a concept/property gap here. This is not the case for phenomenal properties, in contrast, as already explained above.

  11. Is there a reducto ad absurdum?

    How about a philosophy that assumes religion? I.e. it would be a strong necessity to say god doesn't exist, its hugely more satisfying to think that he does, its intuitive that he does therefore we should use philosophy that assumes it is possible.

    Maybe a philosophy that assumes you are god - or that you have lucky talisman?

  12. Hi G., interesting comment! Yablo has attempted some ingenious reductios by appeal to the idea of a 'necessary being' (like God). I discuss these in my linked 'chapter two'.

    Short version: modal rationalists may reasonably insist that necessary beings are, strictly speaking, inconceivable (i.e. it is a priori that all beings are contingent).

    The non-existence of a contingent supernatural being, on the other hand, is true but not metaphysically necessary. (A contingent god - Zeus, say - could have existed, even though no gods exist in the actual world.)

    Strong necessities are such that their denial is conceivable yet impossible. But God is either inconceivable (as a necessary being), or possible (as a contingent being). So, in neither case does denying his existence amount to a strong necessity.

  13. I hate being in the wrong time-zone. I'm going to write a long comment since it takes a long time to get to you guys. (the last comment was at 1:14AM June 30, while my local time is 9:20AM June 29--I wonder if this is even possible given the time-zone separation.)

    Peter says he cannot "conceive of consciousness" other than as a purely material property. What is this conceiving? It sounds like he's saying he can't be "self-conscious" or "conscious of a conscious state" without directing his attention to a materialist origin. In this interpretation, a zombie world is incompossible (from Leibniz: not simply impossible as a generality but impossible given the specific reality of the world we live in). But, it would still be possible, so long as the specifics of our empirical world are left only partially explored; it would thus seem that a pure materialism would explored totally all the possibilities of concepts in this world and effectively dispel alternative possibilities in the process. I don't see such a pure materialism as a possibility itself, so we'll have to say that zombies just aren't there until we can show them empirically.

    Does this make me a materialist or a dualist? I'm only lightly familiar with the debate involving Chalmers et. al., so I'm content to pose the question within the philosophical traditions I'm familiar with. Dualism to me as a student of continental philosophy refers to a more explicit split between mind and body than what Chalmers discusses. Correct me if I'm wrong, but Chalmers is interested in how the physical properties of the brain give occasion to the psychological properties of the mind. From where I'm sitting, the dualist stance is posed as a question of how the mind as a distinct entity is able to interact with a more loosely defined base of materials: words, concepts, ideas in addition to the "stuff" of everyday life. Marx, then, is a materialist because he sees concepts arising from the material behavior of societies; and Kant is a dualist from his distinction between phenomena which are used in the construction of concepts and noumena which are never strictly known but represented by phenomena (this interpretation of Kant is debatable).

    So what, right? Well, the goal of both modern dualists like Kant and modern materialists like Marx was to dispel dogma. I don't know when the term "zombie" began to be used philosophically, but German philosophy became extremely interested in the difference between thinking consciousness and unthinking consciousness--and here is where we may return to the Chalmers-Kripke debate. Can a conscious being lack thought? Can a conscious being know he is writing a response to a blog post and 12 comments but not think that he is doing so? A ("continental") dualist can say I think I am actually sleeping or a brain-in-a-vat while having the conscious sensation of writing a response, and that thinking so raises the serious possibility that I actually am asleep or a BiV. A ("continental") materialist can say that my conscious sensation of writing this response primarily determines the state of my thought, and that if I think I am asleep I must be falling asleep and thus not paying much attention to what I am writing. I don't think (again, correct me) that Chalmers would suggest any such split between thinking and being conscious. But I do think such a distinction is necessary for two reasons: (1) it allows one to discuss opinions as the material of thought and facts as the material of consciousness, thus bypassing certain dogmatic tendencies to posit opinion (Democrats are good) as fact (therefore Democrats are right). (2) It pushes the zombie question off the autopsy table and into the psychoanalysts office, so to speak. "Zombies" are discussed not just as possible or impossible, but as genuine problems: the harmful repetitions of a neurotic, the bizarre automatic responses of a psychotic, or the catatonic behavior of an advanced schizophrenic.

  14. er, looking at point #1 at the end of my post, it doesn't seem I've offered any explanation of the "material of consciousness"--which I guess is the primary focus of this debate. But still...

  15. So
    A) how does the "god"/"not god" case differ from the zombie case?

    B) are there any conclusions one could draw regarding the possible existance of god or the possibility of you being god or the possibility of any of these other things in the same way that you are drawing conclusions from the theoretical existance of zombies?

    Also if we can't say anything concievable doesn't exist then why aren't we left with a meaningless static from this form of inquiry?


  16. G., I'm afraid I don't understand your questions. We can say that both zombies and (contingent) gods are possible but non-actual. The interesting thing about possible zombies is that they demonstrate that material stuff does not suffice for consciousness (since it's possible to have a physical duplicate that nonetheless is not conscious).

    Jared - yeah, New Zealand is GMT+12, which is pretty crazy. I once arrived in LA, after a 12 hour flight, six hours earlier than the departure time! That was one hell of a long day.

    Anyway, you're right that Chalmers' "property dualism" is quite different from the traditional debates you mention. One key distinction here is between "phenomenal consciousness", i.e. what it is like to have an experience, vs. the functional concept of "access consciousness" (i.e. information that is accessible to your brain's "central processor" for thinking).

    Phenomenal zombies are meant to be physically - neurologically - and thus behaviourally indistinguishable from normal agents. They don't suffer any discernible psychological problems. (They'll even claim to be fully conscious, etc., the same as anyone.) But, despite outward appearances, there's not really any "lights on inside". Hope that makes things clearer!

    (The continental analogue of this debate is unfamiliar to me, so I'll need to re-read that part of your comment when I have more time to digest it...)

  17. Ok well then to approach the static point that I made.
    Imagine a set of worlds each missing one "thing" but all identical to the current world in all other ways. That proves no specific things suffice to explain any other specific things. Is not that an odd conclusion?

    Best example is probably consciousness...

    What if we subdivided thought into its sub components so you had Phenomenal concepts just no actual processing of them going on (presumably you'd still be a zombie?) would that not prove that Phenomenal concepts don’t lead to consciousness? Then how do you prove your conscious?

    What if there was processing going on (like a computer) and sensing in a phenomenal input sort of a way but this didn't result in any sort of Ego.

    Now what if there was an Ego but no actual processing using that ego.. (For this example I am thinking if you were the ego you would not actually experience anything.)


  18. "Imagine a set of worlds each missing one "thing" but all identical to the current world in all other ways. That proves no specific things suffice to explain any other specific things. Is not that an odd conclusion?"

    No, sounds good to me -- cf. Hume's maxim, 'no necessary connections between distinct existences'. If they're really two distinct things we're talking about, then one can exist without the other.

    The important part of the thought experiment is working out what things are truly distinct or not. For example, you can't coherently imagine a world that is physically identical to ours but devoid of life. This is because life is merely a matter of physical processes (metabolism, etc.), rather than something distinct.

    "Phenomenal concepts don’t lead to consciousness?"

    According to my latest post ('Why do you think you're conscious?'), phenomenal concepts are partly constituted by phenomenal properties. Hence, they presuppose consciousness. So your proposal here turns out to be inconceivable.

    "sensing in a phenomenal input sort of a way but this didn't result in any sort of Ego."

    That would show that phenomenal consciousness does not suffice for personal identity. I think this is true -- some animals might provide a real-life example. (But, again, others might disagree with us by trying to show that the proposal is actually inconceivable, due to some hidden incoherence.)

  19. ahh I see...

    > So your proposal here turns out to be inconceivable.
    >For example, you can't coherently imagine a world that is physically identical to ours but devoid of life.

    Is not that a function of how you defined the initial set - in this case "a world that is physically identical to ours"?

    What if you very slightly adjust that to say "a world just like ours (except for the animals and plants) just without the animals or plants"
    rather like your one would be "a world just like ours (except without the conciousness) which lacks conciousness".


  20. 'What if you very slightly adjust that to say "a world just like ours (except for the animals and plants) just without the animals or plants"'

    But that's a physical difference -- your new world doesn't have all the same atoms, etc.

  21. (Though it does show that plants and animals aren't metaphysically reducible to the rest of the world outside them. It's possible to have the rest of the world exist without any plants or animals. But there's nothing very philosophically interesting about this fact, since presumably nobody would deny it.)

  22. OK...

    I propose 'no necessary NON-connections between distinct existences'. (which of course leaves the huge middle ground of no necessary anything)

    although I admit Im a way from being as famous as Hume... so I'll not be surprised if you reject it out of hand :)


  23. Richard-
    Yes, that answer to my comment does make things clearer. If one were to ask a continental zombie-philosopher, "are you self-conscious?", the continental zombie would have a crisis. If one were to ask the same question of a analytical zombie, from what you describe, the analytical zombie would say, "yes" although the interviewer would have some indication that the subject is in fact a zombie. I suppose the Turing test and things like that are meant to show that if the answer "yes" is not accompanied by anything that would contra-indicate the self-consciousness of the subject, then one believes the "yes."

    The next question is a bit of a joke, but there's some seriousness to it: What if you are able to get two zombies to quiz each other and to determine the consciousness of one another? Basically, could a zombie find a zombie? Furthermore, could a zombie accept that he is a zombie, or would he deny it or freak out?

  24. If dualism were correct then brain damage wouldn't effect cognitive ability. Simple as that.

  25. Ookla -- no, that actually doesn't follow at all. Try reading the post before commenting. (And note the distinction between necessary and sufficient conditions. The dualist may deny that the brain is sufficient for mentality, whilst still allowing that it is necessary.)


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)