The Ghost In The Machine
David Chalmers has attained a degree of celebrity and earned the adoration of theists, with his philosophical argument against physicalism. The argument is based on the conceivability of philosophical zombies (or p-zombies). Before I get into the argument itself, I should explain what a p-zombie is. This is not the fictional creature of movies that has returned from the dead, but rather a philosophical concept of something that is physically and behaviorally identical in every respect to a person, but that nevertheless lacks any conscious experience. A p-zombie can't be distinguished from an ordinary person, because it behaves the same, reacts the same, and gives the same answers to any questions. It would recoil from pain and say "ouch", for example, but not actually experience the feeling pain. Another way of saying this is that the p-zombie has no subjective or first-person experience.
Now that we understand the concept of a p-zombie, we can turn to Chalmers' argument, which can be stated this way:
1. According to physicalism, all that exists in our world (including consciousness) is physical.I won't argue against the validity of the concept stated in this argument that the conceivability of a zombie world would imply that physicalism must be false. But I will argue that such a zombie world isn't conceivable in any coherent manner. In fact, Chalmers' view of what is conceivable is incredibly naive. Allow me to explain.
2. Thus, if physicalism is true, a metaphysically possible world in which all physical facts are the same as those of the actual world must contain everything that exists in our actual world. In particular, conscious experience must exist in such a possible world.
3. In fact we can conceive of a world physically indistinguishable from our world but in which there is no consciousness (a zombie world). From this it follows that such a world is metaphysically possible.
4. Therefore, physicalism is false - wiki
The theistic (or non-materialist) view is that our bodies are essentially machines. They give us the ability to interact with the world on a physical level, but they do not have any experience of the world, nor are they in control of the machine. But there is a ghost in the machine that has the experience of our sensations, thinks about it, and decides what we will do in response. When our eyes see something red, that sensation is (somehow) passed on to the ghost, which feels the quality of redness, forms thoughts about what we are seeing, and makes a plan of action that will guide our response. If we contrast this view with the p-zombie, there is essentially no difference, except that the zombie lacks the ghost. In both cases, the body lacks experience of the world, and lacks conscious thoughts.
But there is a significant disparity that points to a fundamental flaw in the theistic view. Remember that the theist says that the notion of a zombie world that is physically indistinguishable from the actual world is conceivable. In other words, it is metaphysically and logically possible - it is a sensible notion. But the zombie, despite its lack of consciousness, still has some kind of mechanism built into it that is able to assess the situation and make a decision on what action it should take - an action that would be fully consistent with that of a conscious person. Which leads us to ask, if the zombie, without a mind, has all the physical causal mechanisms to produce behavior identical to that of a person with a mind, then what difference does it make to have a mind? They are both physically identical, and so they both have identical physical causal mechanisms, and they both behave in the exact same way, so it would appear that the mind adds nothing to the equation. But that is in direct contradiction to the theistic notion that it is the ghost in the machine that must be in control of the person's response to a given situation.
Do you see the problem here? If the theist wants to insist that physicalism is false, he is saying that there is a non-physical entity (the mind, or ghost) that controls the actions of the physical body. But with regard to p-zombies, he is faced with an unpalatable choice, because the zombie must have some kind of equivalent controlling mechanism that is not a mind. Either that mechanism is physical (which implies that mind itself is effectively inert), or it is not physical (which implies that it's the same thing as a mind. The latter choice is not consistent with the definition of a p-zombie, and the former choice is an admission that the ghost in the machine is utterly superfluous.
If they bothered to think it through, they would realize that the non-materialist conception of a p-zombie leads to the understanding that human behavior is completely physically determined without the involvement of any mind, which is inconsistent with their own view of free will and the necessity of some non-material entity that controls the body, or alternatively, that a zombie is not really a zombie, since it essentially does have a mind. Either way, it is an incoherent conception.
Of course, there is another choice. Just dump your incoherent theistic beliefs, and admit that there is conscious experience in our world without the need for any ghost in the machine. Not only is this a less convoluted view of the reality of our world, but it is consistent with science, and everything we observe.
Well, first off, it may not be all that important but David Chalmers is not a theist, and he leans to some pan-psychic or proto-panpsychic explanation of consciousness, iirc...and I'm not even sure where he stands on free will.... but certainly there are more positions in philosophy of mind than only theistic dualism vs. atheistic physicalism.
ReplyDeletehttps://blogs.scientificamerican.com/cross-check/david-chalmers-thinks-the-hard-problem-is-really-hard/
DeleteOK. He says he's not. I have corrected it. I was going by this statement:
Delete"And it can even be used to argue against materialism. If there is a possible world which is just like this one except that it contains zombies, then that seems to imply that the existence of consciousness is a further, nonphysical fact about our world. To put it metaphorically, even after determining the physical facts about our world, God had to "do more work" to ensure that we weren’t zombies." - Chalmers.
If he's not a theist, his views are compatible with theism, and that's why theists keep citing him as an authoritative reference.
https://m.youtube.com/watch?v=0p0BjA8mvU4
ReplyDeleteNot all advocates of non physicalism are theists. Non physicalism is obviously more likely to be found among theists than atheists, but among philosophers the difference isn't as great relative to the non philosophers.
Herman Philipse, author of "God in the Age of Science", a now semi popular atheist arguing against theism, is very unambiguously an atheist and a defender of non physicalism. You might end up being surprised at the non theistic reasons to defend non physicalist positions among philosophers such as in the case of Quine and his favoring of mathematical realism.
I'm not surprised by many of the things people believe.
DeleteSee new update.
DeleteEven if Chalmers advocated for physicalism, his views would still be logically compatible with theism. The difference is that physicalism about the mind is more probable given various non theistic views than theistic views, and non physicalism is essentially a necessary condition for most religions to be true.
ReplyDeletePeter van Inwagen is an example of a Christian physicalist. He thinks consciousness in organisms is fully material in nature, and he had wrote books on how this is compatible with specific Christian doctrines such as the resurrection of the dead.
If theists cite Chalmers, it's because they either aren't brave enough to take van Inwagen's route, or in a way they might be more honest that physicalism about the mind is something they cannot reasonably reconcile with their religious beliefs.
Nope, physicalism isn't incompatible with theism. Why would it be? - as van Inwagen points out.
DeleteA lot of positions like these - theistic dualisms opposed to atheistic physicalisms - just tend to get grouped together for (probably) mostly historical reasons, but logically they seem (to me, at least) basically independent of each other. That's what I was saying to skep on Joe's blog: an absolute proof of physicalism wouln't be the final defeater for theism anyway.
Well, natural physicalism, I meant, since God is by def not a part of nature.....
DeleteWhy would physicalism be incompatible with theism? Because most theists believe in a soul that is the essence of the person and that lives beyond the life of the body. That is an essential part of their belief.
DeleteBut as van Inwagen also points out, the more Biblical Xian belief is in resurrection of the BODY!
DeleteSomeone who doesn't water down his belief in the face in scientific knowledge. I like that.
DeleteWell, I'm not familiar with van Inwagens theology (since I only remember reading his contributions to the Platinga vs Davis debates on modal theory at one point) so I can't say much about it, commenting on how "fundamentalist" his ideas are. But the basic concept is that the core texts of Xianity have quite favourable views on the body and the bodily, at least compared to some other traditions...& are perhaps not so ethereal as some of the later ecf's readings of those texts which may have brought in a more definitively idealistic & dualistic Hellenic tradition ...is well known.
Delete"an absolute proof of physicalism wouln't be the final defeater for theism anyway."
ReplyDeleteWell, an absolute proof of mental physicalism surely wouldn't be a defeater for theism, though whether it's evidence against theism is a related matter. But even if mental physicalism was evidence against theism, if it's the best position then I'd suggest any theist adopt it so long as theism all else considered is more probable than not.
It's worth noting that many (if not most) theists claim that their belief is not at odds with science. But this is one area where they unquestionably diverge.
DeleteI think the "qualia gap" a la Chalmers is too wide for physicalism to be the best position & irregardless of anyone's theological views.
DeleteQualia are simply the subjective aspects of our perception. The only "gap" is that it is not possible to describe a subjective feeling in an objective way (other than comparing it to other things). This is not an issue for physicalism. After all, there has to be some feeling we have on our perception, but that is true whether it is physical or not. The whole issue is a red herring that theists toss out there, but it doesn't prove anything at all about belief in non-physical mind.
DeleteI'm not sure it's so simple. Chalmers makes pretty goof arguments that "what red looks like to me" IS a thing - ie objective--and not a "feeling" that can be waved away as merely subjective (or in any way related to any kind of desire). So "what red looks like you" is a thing too, and one that exists only within your unique perceptions. But, within current understandings of "matter" at least, no amount of measuring our electro-neurological and bio-chemical responses within our perceptual apparatus, no amount of that kind of information right down to the molecular or atomic level, could ever seem to definitely answer the question of whether we see the same thing (ie perceive the same color) or not.
DeleteI'd submit that a complete theory of consciousness would have to include an answer, however, to that exact same, age-old Phil 101 question - "when you and I see a color, do we see it the same?" Otherwise important phenomena does seem to have been been lost or overlooked, as Joe says....
But, within current understandings of "matter" at least, no amount of measuring our electro-neurological and bio-chemical responses within our perceptual apparatus, no amount of that kind of information right down to the molecular or atomic level, could ever seem to definitely answer the question of whether we see the same thing (ie perceive the same color) or not.
Delete- That's precisely because it is SUBJECTIVE. Chalmers is making a mountain out of a molehill. He says it's a hard problem because that's what he wants it to be.
Well, let's not argue that, but whether you want to call them "subjective" or not, perceptions are still a significant part of what consciousness is...basically the gist of the whole thing, in fact, as C's p-zombie argument is designed to show.... and they should be accounted for.
DeleteOtherwise, there's something left out...
So let's talk about that. What exactly is the zombie lacking if it is physically and functionally identical to a human? A human sees something red and says "that's red". A zombie does the same thing. What am I missing here?
DeleteWell, that's an elimitivist concept and slightly different than what you've said in comments before, but okay....
DeleteI think we assume consciousness from certain functional aspects, yeh, but that's only from reflection of our own interior experiences and similarities in reactions, etc, and it's those inner things that really are the big deal.... Consider, if you found out for certain that no one else in the world but you had any inner existence, would you just go on the way you're going, or would everything be changed for you...?
I'm not suggesting anything eliminativist, or anything different from what I have said. What I'm suggesting is that the concept of a p-zombie is incoherent, if you think about what it implies.
DeleteBy definition, such a being would be indistinguishable from a human. And that means he could see something red and say "That's red." Just like the rest of us. But how does the zombie know it's red if he can't experience the quality of redness? He must perceive something about it that distinguishes it from other color qualities. Otherwise, he wouldn't be able to say "That's red." Whatever perception he has, whether you want to call it qualia or not, it has the same effect. It is his own experience of color. If you deny that he has this experience, then you are implying that he cannot possibly be functionally equivalent to a human. So one way or the other, there cannot be something that meets the definition of a p-zombie. If you say it is conceivable, it is only because you haven't given the concept enough thought.
https://www.youtube.com/watch?v=71-r6TENTgw
DeleteWell, we can make machines that can discern colors, but does it make sense to ask what red "looks like" to such a machine?
DeleteSeems questionable....a "conscious" machine, or one we suspect of consciousness, might be possible, but can it be broken down into a list of properties and abilities? Or would we suspect a machine was actually conscious exactly when we thought that something else had "emerged", something that transcended the whole of its programmable, pragmatic functioning?
Well, we can make machines that can discern colors, but does it make sense to ask what red "looks like" to such a machine?
Delete- Good point. All the machine can do is describe red by comparing it to other things that are red. But you can do better than that. You can give us a first-hand description of what it actually looks like. So go ahead.
Or would we suspect a machine was actually conscious exactly when we thought that something else had "emerged", something that transcended the whole of its programmable, pragmatic functioning?
- In the case of a zombie, whatever it is that "emerges" must be exactly identical to a human mind. But I didn't really expect you to get the point.
Yes, I see what you're getting at. I don't recall what Chalmers had to say there, but as I understood it, the p-zombie's claim would just be taken something like a lie or a mistake? (Remember, the p-zombie does not have mental states at all, so it cannot have any kind of inner states by definition, so it cannot have " intention" either, and we can't tell one of those from the other.)
DeleteMy other point, well, the machine I was thinking of just recognized the frequency range of colors and wasn't capable of comparing different things of the same color or trying to make descriptions of its perceptions, and one that was capable of all that might be coming close to "consciousness" and/or "seeing" anyway, for all we know. When does a machine stop being just a "machine?"
DeleteRemember, the p-zombie does not have mental states at all ...
Delete- I still thing that what I'm saying hasn't made the journey from my brain to yours. The zombie has something inside that causes him to behave and respond in EXACTLY the same way the rest of us do. But by definition, the zombie is made the same way we are - he is physically identical. So what is this thing in him that produces his behavior? You might say, well, he's like a computer running some kind of software. But if so, it's the same software we all have in us, because he's physically identical. So whatever makes him behave the way he does MUST BE exactly the same thing that makes us all behave the way we do. If we have something extra that he lacks (like mental states), then those mental states are meaningless, because we are already under the causal effects of the same hardware that the zombie has. Don't you see how stupid the whole concept of zombies is? It is nothing but a fantasy that fails to stand up any logical analysis.
P-zombies exist in a POSSIBLE world, so they only need to be conceivable, no matter how highly unlikely. Is it inconceivable that a "thing" could pass for a conscious being?
DeleteMore than that, isn't that "things" are what conscious beings really are the real concept at the core of physicalism anyway?
DeleteWell, right you are. We do have a thing in us that is responsible for our behavior. We call it a brain. And what that brain does is produce thoughts and memories and perceptions of color. The absurd fantasy of a zombie is the idea that he is in ANY WAY different from the rest of us.
DeleteYet, for all we know with certainty (as jdhuey pointed out) everyone else really is a p-zombie, and has no inner experience like or unlike our own?
DeleteIs that the same possible world as the more commonly assumed one in which others also have inner experiences?
I think when Chalmers said "we can conceive of" something in some possible world, what he meant is that it is supposedly logically coherent. You can imagine all kinds of things, but they may not be truly conceivable (in that sense). Which is really the point of my article.
DeleteWell, if you wanna argue the point robustly, Chalmers has a huge, complicated, 50 page essay available on the internet about conceivability
Deletehttp://consc.net/papers/conceivability.html
PDF - http://selfpace.uconn.edu/class/ccs/Chalmers2002Conceivability.pdf
DeleteIt is not necessary to go through this tome to understand what is meant by a common term.
DeleteThe traditional answer is that what is logically impossible is a 'contradiction in terms'--something that is 'inconceivable'. - here
In the case of Chalmers' argument, the p=zombie might seem conceivable on first look, but that quickly vanishes when you look under the hood.
Well, maybe, but Chalmers means what he means by his own usage of the term. And I think you are going against a philosophical tradition that has always considered - eg - that the non-existence of other minds was conceivable, that no necessary inner phenomena was ever inferable from outer behaviour, and Chalmers point is just about that same thing....
DeleteChalmers goes beyond that. He specifies that a zombie is physically identical to a human (in a world that is physically identical to ours). So it's not just a matter of inference based on "outer behavior" alone. He is telling us that they have the exact same physical functional mechanisms that people have.
DeleteBut does that really matter to the gist of his point...?
DeleteHis point, as I understand it, it that because p-zombies are conceivable, then consciousness must not be physical. I think I am addressing that point directly by pointing out that p-zombies, as he defines them, are logically incoherent, and so his argument is not sound.
DeleteBy definition, such a being would be indistinguishable from a human. And that means he could see something red and say "That's red." Just like the rest of us. But how does the zombie know it's red if he can't experience the quality of redness? He must perceive something about it that distinguishes it from other color qualities. Otherwise, he wouldn't be able to say "That's red." Whatever perception he has, whether you want to call it qualia or not, it has the same effect. It is his own experience of color. If you deny that he has this experience, then you are implying that he cannot possibly be functionally equivalent to a human. So one way or the other, there cannot be something that meets the definition of a p-zombie. If you say it is conceivable, it is only because you haven't given the concept enough thought.
DeleteThe mind/brain can process information and cause avoidant behavior before you consciously know what's happening. Think of blindsight, or pulling your hand from a hot stove, or flinching when an object is unexpectedly flying at your head. So the idea is that a p=zombie could process the visual information associated with redness and then be caused to say "I see red" without any experience of red. The problem comes when you ask my zombie twin, "Are you conscious?" He'd have to say "Yes" to be fnctionally indistinguishable from me, but that would mean that his behavior in saying "Yes" isn;t caused by his actually being conscious, so that consciousness in that case would be causally inert.
"That's true. But the problem is that the zombie is aware of redness. It's not just an unconscious recognition. He must be aware of the color he sees. Otherwise he wouldn't be able to say what color it is.
DeleteIf you have a well equipped psych lab and are handed two test subjects, one a p-zombie and the other an identical person but not a p-zombie, how would one even go about determining which is which? Given the definition of a p-zombie there is no detectable difference. This open the possibility that everyone in the world (except for yourself, of course) is a p-zombie. Or, everyone including yourself is a p-zombie but we just don't realize it.
ReplyDeleteThe last thing is impossible, in spite of Daniel Dennett's assertions that that is actually the case.
DeleteP-zombies as strictly defined in the example could well be impossible tho. There might well be limitations as to how far a nonconscious thing could ever imitate a conscious one without requiring some actual inner experience to fall back on. Tho, if so, I don't see why that wouldn't be a blow to the physicalist's position in the end rather than a point in his or her favor....
I wouldn't say P-zombies actually do exist in a possible world. But anyway, I'd characterize Chalmer's argument like this:
ReplyDelete1. If mental physicalism is true, then our physical makeup is sufficient for mental states.
2. If our physical make up is sufficient for mental states, then mental states exist in every physically possible world where our physical makeup exists.
3. If mental states exist in every metaphysically possible world where our physical makeup exists, then it is not the case that there exists a metaphysically possible world where our physical make up exists but mental states do not exist.
4. If P-zombies are metaphysically possible, then there exists a metaphysically possible world where our physical make up exists but mental states do not exist.
5. If P-zombies are conceivable, then P-zombies are metaphysically possible.
6. P-zombies are conceivable.
7. Therefore, P-zombies are metaphysically possible.
8. Therefore, there exists a metaphysically possible world where our physical make up exists but mental states do not exist.
9. Therefore, it is the case that there exists a metaphysically possible world where our physical make up exists but mental states do not exist.
10. Therefore, it is not the case that mental states exist in every metaphysically possible world where our physical makeup exists.
11. Therefore, it is not the case that our physical makeup is sufficient for mental states.
12. Therefore, it is not the case that mental physicalism is true.
Where do people attack that argument? Often, premise 5 is attacked. Chalmers has defended premise 5, but I don't think it's plausible. At best I think we can conclude something is epistemically possible given that it is conceivable AND we don't know of anything that rules it out, but that doesn't help Chalmers.
If we accepted that conceivability is sufficient for metaphysical possibility, then we'd probably concede too much (for suppose it's conceivable that souls exist but mental states do not exist!).
That sounds like a better formulation of the argument than the one I used. I did not feel comfortable attacking the notion that conceivability implies metaphysical possibility. However, I am aware that many people regard the concept of p-zombies as incoherent (even if not specifically in response to this argument). Dennett is one of them. Here is another.
DeleteI think Chalmers p-zombie example points out effectively how consciousness is not even derivable from known physical laws so a p-zombie world seems quite sensible from that perspective. The world as we know it would seem to be able to exist just fine without inner states; they seem functionally superfluous. But there might also be "laws" we don't know anything about yet. (Chalmers does exclude a pan-protopyschic scheme from being "physical" at the end of his long Conceivability essay too, for reasons that I find unclear, since it seems arguable that such a scheme could well be an expansion of the "physical" attributes of matter, or at least arguably so, since even our understanding of what "physical" means is currently fairly limited....)
DeleteBut you must realize that's nothing but an argument from ignorance. The fact is that we know enough from science to be quite confident that consciousness is physical.
DeleteIn what sense of "physical"?
DeleteI mean no ghost in the machine. No spirit. No soul. The brain is entirely responsible for consciousness.
DeleteThere are a lot of ideas other than consciousness arising from the bio--chemical-electrical principles we know now, such as pan-protopsychism or consciousness being quantum. Are hypotheses like those, which expand the definition of "physical", included in your brand of "physicalism" then? ,
DeletePan-psychism? Quantum consciousness? Those things are wacky ideas that aren't supported by evidence or any coherent scientific view (just like souls). No, they do not fit in my idea of physicalism.
DeleteDerivability argument - The existence of consciousness cannot be soundly derived from the present known physical facts.
DeleteMy take: I don't think this is very interesting.
First, we don't know the extent to how many physical facts we're missing, so we cannot justifiably say that consciousness cannot be derived from the complete set of physical facts since we don't know if we have that.
Second, whether something is derivable from a system can be controversial. For all we know, might consciousness be physical, but our methods of derivation are not sufficiently strong enough to derive such facts from our set of physical facts? I don't know.
I don't know either, but not knowing doesn't make someone a physicalist, which is a positive assertion. The point would be about whether those new facts about physicality and/or new derivations would fall comfortably in line with what is now called "physical reality" - as we now understand it - or would have to stray so radically from our current paradigm in order to answer such thorny questions that it would (at least) invoke "a new dimension in physicalist theorizing."
DeleteSome philosophers, maybe McGrinn, have said a new laws of physics would be needed to make consciousness physical in any sense that we use now.
DeleteI think what would be needed is supervening laws that work at a different level of analysis. Not laws that change existing physical laws. It would be like the relationship between biology and physics. Biology is a different level of analysis, and is useful for explaining phenomena that are not easily understood in terms of masses and forces. But nothing in biology negates fundamental physical laws.
DeleteWhere do people attack that argument? Often, premise 5 is attacked. Chalmers has defended premise 5, but I don't think it's plausible. At best I think we can conclude something is epistemically possible given that it is conceivable AND we don't know of anything that rules it out, but that doesn't help Chalmers.
ReplyDeleteYou'dknow more about this than I would, but he relies on the idea that consciousness isn't water, in that it doesn;t seem to make sense to say that there is "consciousnessy stuff" the way there could be "watery stuff." Consciousness is sui generis, but I forget how that gets you from conceivability to possibility.
Where do people attack that argument? Often, premise 5 is attacked. Chalmers has defended premise 5, but I don't think it's plausible. At best I think we can conclude something is epistemically possible given that it is conceivable AND we don't know of anything that rules it out, but that doesn't help Chalmers.
ReplyDeleteBut how could we, or I, be mistaken about being conscious? It's conceivable that everyone other than me is a p-zombie, but then how does their behavior when they speak about being conscious come about? I think the point was precisely that we can not know who is or is not a p-zombie other than via a first person perspective. If we could detect a difference form a third person perspective, then that criterion could not apply to p-zombies.