How Not to Imagine Zombies

Greeting from Tucson! I am on my way out to hear a talk by Bernie Bars and Wolf Singer, so I haven’t much time. I hope to get to the comments that are building up. 

I have recently been mounting an offensive against the back-from-the-dead Zombie argument against materialism. My most recent attempt was to offer a parody of the zombie argument to the effect that dualism is false (since I can conceive of non-physical zombies). This is, in my opinion, enough to show that the zombie argument against materialism is hopeless and misguided. Richard Chappell disagrees

The zombie argument begins by providing an undisputed specification of the “physical respects” of the world. It then asks whether phenomenal consciousness logically follows from the specification. Our answer is ‘no’. That’s why physicalism is false.

This is of course nothing more than more of the usual question begging. Does phenomenal conscious follow from a complete physical description of the world that we live in? That depends on whether materialism is true or not. If it is, then OF COURSE it logically follows that there is phenomenal consciousness; if it is not then OF COURSE it doesn’t logically follow. The point is that this cannot be known a priori. To imagine otherwise is as absurd as saying that one can know a priori whether the caloric fluid theory of heat is true or not. 

Chappell goes on to say

A proper analogy, then, would require building up the “non-physical zombie” world from an undisputed non-physical specification, just as we earlier built up a physical zombie world from an undisputed physical specification. But of course RB cannot do this. So that’s why the zombie argument cannot be turned against dualism in this way.

I, of course, cannot do that because there is no ‘undisputed non-physical specification’ of ANYTHING. So the fact that I cannot build up such a description is irrelevant. What is relevant is that I can imagine a creature just like me in all non-physical respects; therefore dualism is false.

OK, so I am running late…gotta go!!!

Having thought about this a bit, I am at a loss to understand why RC thinks it is so important that we ‘build up’ rather than ‘subtract’ when we do this conceiving. What’s the relevant difference?

21 thoughts on “How Not to Imagine Zombies

  1. I don’t follow, honestly, why Richard is so committed to the claim that the zombie case is an argument at all. It’s clearly an intuition pump, rather like the direction of fit metaphor over in the philosophy of action/reasons for action literature, or Frankfurt-style examples. If you have the intuition, you’ll find the case congenial; if you don’t, you’ll find it silly. What matters, surely, is whether there is some independent reason to suppose zombies are actual, or at least genuinely possible (rather than logically possible). If it turned out, say, that cats were zombies, and we could somehow prove this, then we’d have an argument. As is, the zombie notion, all on its own, seems to do no work. What has to do the work is something like the question-begging assumption, which you note, that the physical world cannot fully account for consciousness.

  2. I fully agree with what ADHR said here. The whole “building” spiel is just a hand-wave over the fact that they can’t give any evidence for the existence of “zombies”. What they are simply saing is, my imagination is correct and your is false. They’re only satisfying their will to power, not focusing on reality!

  3. Well, there may not be a contradiction; that depends on what we find out. Put it this way. To me, we seem to be in the same position as someone in 1700 wondering if it is possible that *that stuff* (pointing at some water) is made out of H20 or some other stuff. It seems possible that *that stuff* could be made out of all sorts of other stuff besides H2O, but since what is actually being demonstrated is in fact H2O, those seemingly possible scenerios are not in fact really possible. When we try to imagin a creature that is physically identical to me in every way except that it lacks consciousness we may be in the same position. My point is that this is an empirical question not an a priori one in the fullest sense that RC wants. So there may be a couple of routes to the contradiction (some I mention in the post The Inconceivability of Zombies)

  4. Hm. I see what you’re saying, but I’m unclear as to why you hold the nomological background fixed. If physical laws could have been different — and it seems that they could have — then that stuff could be phenomenologically identical to water, and yet be XYZ rather than H2O. Which, I think, is the point the zombie-advocates are starting from: if the world were different than conceived under some set of laws (i.e., the ones accepted by materialism), then there could be creatures that have all the physical and observable attributes, and yet lack the mental ones. I’m not endorsing this, of course, but I think that’s the kind of possibility they’re playing with.

    The issue is still empirical, though. After all, what’s going to determine for us whether our world is a world of physical laws or of some other set of laws is going to be which set of laws (and theories, etc, etc — I’m using “laws” imprecisely) is, at least, empirically adequate.

  5. Yes, I agree; there could be things which look just like us which lack phenomenal consciousness. But that is unproblematic for just the reasons given. From the fact that something could look just like water and not be H2O it doesn’t follow that water isn’t H2O. So too from the fact that there could be creatures that look just like us and lack phenomenal consciousness it doesn’t follow that phenomenal consciousness is not physical. What you need to show is that those creature not only look just like us but are physically identical to us. This has not been shown, as the waterr H2O example nicely illustrates. The thing that you are imagining is not something that is physically identical to water and isn’t H2O…

  6. Hang on; I think you’ve changed subjects on me. The issue was whether the existence of zombies is logically possible, not whether they are actually possible. I accept that the latter has to be demonstrated (empirically): we’d need to find an uncontroversially physically identical (to us) creature that we could somehow determine lacked phenomenal consciousness. I’m on board with that.

    What I’m wondering about, though, is why go so far as to deny the logical possibility of zombies. They certainly seem logically possible. It also seems to me that it’s logically possible that water was not H2O. In some other world with different physical laws, something that had all the properties we associate with water could, nonetheless, turn out to chemically be XYZ rather than H2O. That is not the actual world, ex hypothesi, but I don’t see why it isn’t a possible world. We can repeat this for zombies fairly easily.

    I suspect the reason that the XYZ/H2O case seems to not be logically possible is that the referent of “water” has already been fixed in the actual world as being H2O. So, when we play the possible worlds game, there’s probably a covert switch of reference-fixing going on. That is, rather than “water is XYZ”, what we actually imagine is something like “a substance qualitatively identical to water is chemically identical to XYZ”.

    In the case of zombies, though, whether and how the referents of terms such as “mind” (and “belief” and “desire” and all that good stuff) are fixed is exactly the issue. So, while I agree that the zombie case itself does no argumentative work, I’m not seeing why we have any basis to deny its logical possibility. It looks to me like, pending a successful completion of a reductionist or eliminativist or emergentist program — or any other such materialist relation between mind and body/brain — we don’t have any good reasons to rule out the possibility that a dualist story is true, i.e., that we don’t live in a materialist world.

  7. I didn’t change the subject. I agree that the issue is whether they are logically possible; I was giving an argument that they are not. The reason is that things which seem to us conceivable sometimes turn out to inconceivable. What you are conceiving when you think of Twin Earth is not a world where water is not H2O, itis a world where there is stuff that looks like water and is not H2O. So ocould there be a world where it seems like people are in pain when they are not; sure! In fact we live in that world! Can there be a world where there are creatures that look like us and do not have conscious pains? Sure! Could there be a world where there were creatures physically identical to us and lack consciousness? Well, that is a whole ‘nother story! (compare, no one thinks that there could be a world physically just like this one where water wasn’t H2O, why not the same in the zombie case?)

  8. I think I addressed that; I suppose I didn’t do a very good job!

    Suppose we didn’t know what water was, chemically. There’s this stuff in rivers, we drink it, wash in it, and so on — it does what water does. Until we do the chemical analysis, it’s an open question what actual chemical fixes the referent of the term “water”, right? It could be H2O, it could be XYZ. Once we do the chemical analysis and determine water = H2O, then what follows is that any possible world like Twin Earth is not a world where water = XYZ but a world where something very like water = XYZ.

    But, moving to the zombie case, we’re in an epistemic situation comparable to that of those poor unfortunates who existed before the chemical analysis of water was completed. We don’t really know what the referents of terms like “mind”, “belief”, “desire” and so on are, except in terms of similarly-loaded intentional vocabulary (e.g., a belief is a certain kind of propositional attitude such that…). So, until we either finally do the full physical analysis of consciousness, or find out no such analysis is possible, zombies are still logically possible. The world where there are physically identical creatures to us that lack phenomenal consciousness could be this one. As I’ve said elsewhere, since all the zombie case does is point out this is a possibility, it’s really not that interesting. Lots of things could be; I’m more interested in what actually is, which requires either doing the afore-mentioned analysis or showing it can’t happen. But, at the least, I think we need to concede that, right now, zombies are still logically possible.

    (This assumes, of course, that mental terms rigidly designate, which I’m not sure about. (Unlike “water”.) But that’s a whole ‘nother issue.)

  9. I think what is tripping us up here is the use of the term ‘logically possible’. For something to logically possible is for it to something that is free from contradiction. So, in the water case it is not the case that before we discovered the chemical nature of water it was logically possible that water be xyz. What was logically possible was the ‘looks like water is xyz’ world. This is why I said earlier that what is logically possible depends on what is actual and since we are in large art ignornat about what is actual we will be in large part ignorant about what is logically possible.

    So too in the zombie case.

    You say “The world where there are physically identical creatures to us that lack phenomenal consciousness could be this one.” I think it IS this one, if by ‘phenomenal consciousness’ you mean ‘epiphenomenal non-physical things’…I of course disagree if you just mean ‘conscious pain, tickles, etc’. But I really don’t buy the claim that this world might be the zombie world. I agree with Galen Strawson that phenomenal consciousness is the one thing who exisytence we are absolutely sure of (it is physical objects that are in trouble). But even if it could, that sense of possibility is the same as above. It is an ‘for all we know right now’ kind of possibility, not logical possibility as I understand it.

  10. I would agree that we’re getting tripped up on what constitutes a logical possibility. As I’m reading it, something is logically possible if we don’t know enough to find a contradiction. So, “water is XYZ” is not logically possible because we know enough about chemistry to say that the thing we mean when we say “water” is the very same thing we mean when we say “H2O”. However, “consciousness is not a physical phenomenon” is, I would say, still logically possible because we don’t know enough to determine that it’s a contradiction. That is, we’re not sufficiently clear on what the “physical phenomenon” is (nor, really, on what consciousness is) to determine that we’re not in a zombie world.

    The issue with the zombie world, of course, is not to deny that phenomenal consciousness — by which I do just mean the subjective experience of “stuff going on in the head” — exists, but that it doesn’t follow from a certain kind of physical construction. So, the existence of zombies wouldn’t imply there was no consciousness, just that some creatures, physically identical to conscious creatures, nonetheless lacked consciousness.

  11. Yeah that’s what I thought. When I say ‘logically possible’ I mean ‘has no implicit contradiction’, so what you call logical possibility I was calling epistemic possibility…in that case we do not actually disagree. I completely agree that we seem to be able to imagine worlds where there are creatures that are physically identical to us and which lack consciousness. I just deny that our seeming to be able to do this means that we are really doing it.

    As for your second point, I never meant to imply that the zombie argument means that there is no phemonenal consciousness here! What I meant is that I think this is a world where there is no non-physical consciousness…so we are the zombies that Chalmers’ imagines…it is just that we have physical consciousness…

  12. That depends on whether materialism is true or not. If it is, then OF COURSE it logically follows that there is phenomenal consciousness; if it is not then OF COURSE it doesn’t logically follow

    This doesn’t seem right. The question is whether physicalism is true. So we ask, what would the world be like were physicalism true? You seem to want to answer: (i) it would be just like it in fact is and (ii) physicalism would entail consciousness. But the Zombie argument gives us evidence that, yes, (ii) would be true, but not (i). The evidence they offer is that we can ideally conceive (or perhaps something a bit weaker) a world with (i) true and (ii) false.

    So, your claim above that ‘if materialism is true then of course it entails that there is phenomenmal consciousness’ is just the question at issue. That the relation is something short of entailment is evinced by the conceivability of the relevant worlds.

    It might help to compare the relation between (a) ‘x is H2O’ and (b) ‘x is water’. As in the Zombie case, the relation is not one of entailment since there are conceivable worlds in which (a) is true and (b) is false. That’s true despite the property-identity being metaphysically necessary, as the story goes.

    Anyway, it looks like your argument is that we cannot know a priori that (i) and (ii) are related by metaphysical identity. True, but no one denies that initially. The initial argument is that they’re not conceptually related and so not logically related. The tricky part of the the Zombie argument comes when trying to reach the metaphyscal conclusion. But you don’t seem focused on that part of the argument.

  13. Hi Mike, nice to hear from you again and thanks for the comment.

    I wonder if we are talking past each other a bit. I am claiming that if physicalism is true of our world then physicalism entails consciousness, since we know already that there is consciousness here at our world. I am not saying that at any given possible world where physicalism is true it entails consciousness (though I do think that, that is not the premise I was using).

    I claim that the zombie world is not ideally conceivable, or more accurately, that we are not in a position to say whether it is or not and that is the reason that the zombie argument is bad. We are in the position of someone who has not learned what the square root of 58,965. So before we learn that it is 242.83 (rounded up to the nearest hundreth) it seems as though we can conceive of it being a whole number but when we find out that it is a rational number we realize that it is not really ideally conceivable that it be a whole number. So too, if we find out that physicalism is true of our world we will find out that it is not really ideally conceivable that there be zombies.

    “As in the Zombie case, the relation is not one of entailment since there are conceivable worlds in which [x is H20] is true and [x is water] is false. “

    I find this somewhat shocking! Which world do you have in mind?

  14. “As in the Zombie case, the relation is not one of entailment since there are conceivable worlds in which [x is H20] is true and [x is water] is false. “ I find this somewhat shocking! Which world do you have in mind?

    The claim that water = H2O is a posteriori necessary, yes? Do you know of a logical truth that is a posteriori and necessary? I don’t. The contingency of the claim that H2O = water entails that there are worlds w such that, from the point of view of w, it is false that H2O is water. There’s no quarrel about that that I know of. So, indeed, there are such worlds. These are worlds w which are such that, considered as actual, it is false at w that water = H2O. Certanly, there are NO worlds w such that, considered as counterfactual, it is true at w that water is H2O. There is in short a difference in assessing the idenity claim that water = H2O that varies depending on whether the worlds of evaluation are considered as actual or as counterfactual. This is central to Chalmers 2D assessment of such claims.

    Maybe you don’t think much of the contribution of Chalmer’s 2-dimensionalism here? Or maybe you have offered some criticism of 2D that I’m just not familiar with. I’m obviously pretty late to this discussion.

  15. “Do you know of a logical truth that is a posteriori and necessary? I don’t. “

    Well, I suspect that you will not like my answer, but I claim that all logical truths are a posterori and necessary since they are empirical truths like any other (well, they are especially well confimerd empircal truths).

    I do have a problem with 2-dism, but I will address that in a seperate post…

Leave a Reply to Mike Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s