Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Facing Up to the Problems of Facing Up to the Problems of Consciousness

  1. Dec 21, 2004 #1


    User Avatar
    Gold Member

    Facing Up to the Problems of "Facing Up to the Problems of Consciousness"

    These are actually just problems I have which may be resolved with some clarification- I just couldn't resist that title. (Yes, I know, I'm SO original. :rolleyes: ) BTW I ended up asking more questions than I intended to, I really want to understand this, I greatly appreciate everyone's help, and I certainly don't mind working for the answers.

    Quotes in indigo are from Chalmers' "http://jamaica.u.arizona.edu/~chalmers/papers/facing.html [Broken]."

    I'm rather new to the easy v. hard distinction, but, from what I gather, the hard question is answered by 1) settling on some system of "methodological and philosophical rules and assumptions"- or defining what qualifies as an acceptable answer- then 2) stating the necessary and sufficient conditions from which experience follows or which give rise to consciousness.
    The first step may need some explaining, but presumably the hard question is only hard under certain combinations of rules and assumptions, since they define and are used to evaluate the second step, the "necessary and sufficient conditions from which experience follows". No? Have I made that more difficult than necessary? I need a statement in the form, "The hard question is answered by...".

    "I suggest that a theory of consciousness should take experience as fundamental. We know that a theory of consciousness requires the addition of something fundamental to our ontology, as everything in physical theory is compatible with the absence of consciousness."

    (I guess I should say that I'm not yet convinced that experience cannot be explained as an emergent property so you know where I'm coming from.) Though it may follow from his assumptions- and I'm not sure it does- my main problem is that this seems like such a cop-out, as he seems to admit and excuse:

    "Of course, by taking experience as fundamental, there is a sense in which this approach does not tell us why there is experience in the first place. But this is the same for any fundamental theory. Nothing in physics tells us why there is matter in the first place, but we do not count this against theories of matter. Certain features of the world need to be taken as fundamental by any scientific theory. A theory of matter can still explain all sorts of facts about matter, by showing how they are consequences of the basic laws. The same goes for a theory of experience."

    Since when was the hard question asking "why there is experience in the first place"? Hasn't he changed the question? Granted, it may end up being swallowed by the "creation of the universe" question, but isn't that a different question? That is, is his "naturalistic dualism" a possible answer to the hard question, or is it a claim that the hard question cannot be answered? It seems that defining experience as fundamental makes it an easy problem: "The easy problems are easy precisely because they concern the explanation of cognitive abilities and functions." Isn't the problem hard only when experience is assumed to not be fundamental?
    (I'm new to the following also but indulge me.) Is he doing something analogous to completing an incomplete system by adding G to its axiom set? If this process is acceptable, (when) does it ever end? And how does he know this very process is not the mechanism which gives rise to experience? He has argued in another http://jamaica.u.arizona.edu/~chalmers/papers/penrose.html [Broken], published the same year as "Facing Up...", that "the assumption that we know we are sound leads to a contradiction." Now, I'm on very thin ice here for several reasons, but it seems like he is assuming his combination of rules and assumptions (CRA) are true, he knows he experiences, yet he knows he cannot demonstrate that he can experience from his current set of rules and assumptions. He says, "Experience may arise from the physical, but it is not entailed by the physical." So he takes experience as fundamental- that is, he adds it to his CRA. Ugh, I don't know enough to make an argument with this, and I'm tempted to delete this part, but I really want to know what is going on here, and I suspect it is relevant to his argument.
    (Please go easy on me- those are all honest questions. :wink: )

    "For any physical process we specify there will be an unanswered question: Why should this process give rise to experience? Given any such process, it is conceptually coherent that it could be instantiated in the absence of experience. It follows that no mere account of the physical process will tell us why experience arises. The emergence of experience goes beyond what can be derived from physical theory."

    I don't understand what he means by "conceptually coherent". Does he mean "it is possible that..."? Is he evaluating it from inside or outside of his CRA?
    He does seem to admit that his CRA, or "physical theory", is incomplete. But I don't know if physical theory- in itself- meets the requirements for Gödel's incompleteness theorems to apply.
    I also don't understand the difference between entailment and implication though I have tried to find consistent definitions (I understand implication). I also don't understand the difference between proof and demonstration. Mathworld defines "proof" as "a rigorous mathematical argument which unequivocally demonstrates the truth of a given proposition." It has no definition for "demonstration" and begins the definition of "rigorous" with "A proof or demonstration is said to be rigorous..." :confused:
    My main problem arises because I don't understand how he concludes that experience is not entailed by the physical. Is the problem with forming or deciding the statement? Or something else?

    Finally, I started this thread mainly to clarify the meanings of intrinsic, extrinsic, and emergent properties as they are used in the context of consciousness. And I'd like to understand the implications of defining these three terms in different ways, as they relate to each other.
    For even a little help with a single part of this, thank you.
    Last edited by a moderator: Apr 21, 2017 at 11:06 AM
  2. jcsd
  3. Dec 22, 2004 #2


    User Avatar
    Staff Emeritus
    Gold Member

    His student Rosenberg defines entailment as "B is entailed by A if it cannot be the case that A is true and B is false." Chalmers, and Rosenberg, claim that the non-existence of experience is not inconsistent with physicalism, and so experience is not entailed by physicalism. The biggest problem I can see with this argument is that there is just no way for us to know whether or not the non-existence of experience is consistent with physicalism. Chalmers uses a rather simple thought experiment, envisioning a universe in which experience does not exist. He envisions the existence of zombies, which behave exactly like humans, but have no subjective experience. The issue here is that we do not know whether or not a zombie is really possible, even though we can imagine them to be. The simple fact is, we don't know enough about a human to know that a being consistent with all the physical aspects of humanity would be possible without experience. Assuming that it is possible is begging the question, as he is already assuming that the physical facts of being human do not entail experience, which just happens to be exactly what he is attempting to prove with his zombie argument!

    I think Rosenberg's argument is a little stronger, because as he says, he doesn't rely on an argument from lack of knowledge or conceivability, both of which are hopelessly flawed. But I'll leave discussion of Rosenberg to the discussion threads for his book.

    One of the big issues here is brought up by Steve Esser in the "Is the Hard Problem Silly?" thread. He points out that Chalmers' argument do not resonate with people that have Dennett-like intuitions. That is simply because Chalmers' argument rely on just that, intuition. He says that he can or cannot imagine a certain set of facts, and so what he can or cannot imagine must be true. Any argument from intuition such as that is bound to not convince a good number of people, because not everybody is going to have the same intuition. Some of us can imagine what Chalmers cannot.
  4. Dec 22, 2004 #3
    I think the zombie illustration is making more a point of epistomology than it is ontology. "Entailment" is obviously dependent on our knowledge of the relationships between A and B. So if we don't have the knowledge then our conclusion on entailment with A and B could be wrong. No one doubts this. This illustration just makes the point very clearly that currently there is no entailment. We have no basis for making the claim that consciousness is entailed by brains. Other arguments are then made to show that this entailment cannot happen. But I don't believe that is the sole purpose of the zombie illustration. Of course one can then begin to try to imagine how one could ever determine whether a person was a zombie or not but that's not the sole purpose of the illustration.

    The interesting thing is that no one else has imagined anything either. Dennetts views amount to simply re-defining consciousness so that there's nothing to explain.
  5. Dec 22, 2004 #4


    User Avatar
    Staff Emeritus
    Gold Member

    I kind of have the same feeling that Dennett just cops out in the third part of Consciousness Explained, but that doesn't make him wrong. As far as what has or has not been imagined, I pointed out elsewhere that there still exists no theoretical framework of the neuronal basis of cognition. Without that, there is no paradigm within which to imagine anything. Expecting an explanation without a framework in which to make an explanation is like expecting an alchemist to explain acid neutralization. Because of this fact, I'll hold off making any judgements until such a paradigm exists.

    Note: Dennett does try to establish such a theoretical framework, his "multiple drafts" model, in which subjective experience amounts to an emergent property of recursive parallel processes. I'll admit that the explanation feels hollow, but I've never seen anyone address his explanation on its own terms. Every argument against it that I've seen argues from pre-existing, intuitive paradigms. No one has ever argued against the internal consistency of the explanation within his framework.

    Addendum: An analogy would be arguing that time must be absolute because Newtonian mechanics (which is the system more easily derived from intuition) dictates that this be the case. In fact, many of the old posters in the TD forum did just that, without realizing it. I suspect that many of the arguments against Dennett-like explanations are much the same.
    Last edited: Dec 22, 2004
  6. Dec 22, 2004 #5
    I'm sure there are some arguments against it somewhere but I don't know how good they are. Personally, I just assumed that since the framework didn't address the problem to begin with then why bother debating it?
  7. Dec 23, 2004 #6


    User Avatar
    Gold Member

    So there's no difference between entailment and implication?
  8. Dec 23, 2004 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Listing the necessary and sufficient conditions for consciousness is not enough to answer the hard problem. If it were, we wouldn't have a hard problem to begin with, because the arguement is certainly not that science (in principle) will not be able to provide us with those necessary and sufficient conditions.

    An answer to the hard problem would not just list the conditions under which subjective experience occurs. It would allow us to understand how it is that, when those conditions occur, subjective experience also occurs. It would unify our understanding of nature with what we know of our immediate experience, such that the existence of the latter would no longer appear mysterious or contingent, but would rather find a natural, logical fit in our theoretical understanding of the world. (NB: This does not necessarily imply that such an answer would appeal to intuition.)

    As we do not have an answer to the hard problem, we are perplexed by questions like, "How is it that when the brain does such-and-such, I end up experiencing this? Going one step further, why is that when my brain does such-and-such, I experience this instead of this? In fact, how is it that all that stuff going on in my brain has anything to do with experience at all?" If we had a complete answer to the hard problem, we would be able to answer these questions. Our answers would not have the flavor of brute correlation, but rather would have the flavor of logical consequence. We would say, "Well, you see, we believe the world is such-and-such a way, and if this is right, then it logically follows that those facts about experience obtain in these conditions. Because the world is the way it is, it [edit: the facts about experience] could not have been any other way."

    From the paper:

    It is a proposal for a possible answer. Verifying such a metaphysical assumption is another question altogether; it would be difficult, if not impossible, to do. The best support for such a framework, at least initially, would not come from empirical verification, but rather from its ability to coherently answer the question.

    Assuming experience to be fundamental would only make the hard problem 'easy' insofar as it might be a way to make it solvable in the first place. It would not, however, amount to reducing the hard problem to one of the 'easy problems,' as Chalmers uses the term. The easy problems, by definition, are those that are solvable in principle using only structural and functional concepts. The whole point of assuming experience to be fundamental is to find a way to place it in a theoretical framework without trying to derive it from structure and function alone.

    By proposing that we take experience to be fundamental, Chalmers is indeed doing something like completing an incomplete system by adding an axiom. The justification for this is that a certain phenomenon cannot be logically derived from an old axiom set, so the only route available to accomodate the phenomenon in the theoretical framework is to add some new axioms.

    This process is acceptable, and has been done before in standard physical theory (see Chalmers' example of Maxwell and electromagnetic charge). Presumably the process of adding axioms ends when some axiom set is arrived upon which is sufficiently powerful to accomodate all known natural phenomena.

    I'm not clear on what you mean here.

    "Conceptually coherent" means the same thing as "logically consistent." If two propositions are logically consistent, then there is no contradiction in assuming that they are both true; that is, it is logically possible that they could both be true.

    In the text you cited here, Chalmers is evaluating the CRA of standard physicalism/materialism and seeing what its implications are. He is essentially stating that, given the axioms and rules of physicalism, it is logically possible that we could have a normally functioning human brain but no subjective experience. There is nothing in the CRA of physicalism that, when given a human brain as 'input,' forces us to conclude that subjective experience exists as a consequence of the brain's activity.

    It is important to realize that this is a claim about a theoretical framework and what it can entail, not a claim about actual reality and what really occurs. Chalmers is not claiming that a human brain could function normally but have no subjective experience; he is claiming that a human brain, as described by the CRA of physicalism, could function normally but have no subjective experience. If this is right, the implication is that something is wrong with physicalism, because it does not account for everything that actually occurs in nature. If a universe Z existed that followed the CRA of physicalism perfectly, then such a universe could have normally functioning brains without attendant subjective experience (thus, we would say that 'zombies' are metaphysically possible); but Chalmers does not suppose that our universe is identical with Z, and thus in this universe, it is impossible to have a normally functioning brain that does not experience (thus, we would say that zombies are nomologically impossible).

    Entailment and implication mean the same thing. I don't know what instance of the word 'demonstration' you're referring to, but it's probably safe to assume that proof and demonstration mean the same thing as well.

    The general form of the argument is:

    1. Facts about structure and function can only entail further facts about structure and function.
    2. Physicalism only deals with structural and functional facts.
    3. Therefore, physicalism can only entail facts about structure and function. (1 & 2)
    4. Not all the facts about p-consciousness are facts about structure and function.
    5. Therefore, physicalism cannot entail all the facts about p-consciousness. (3 & 4)

    I have gone into more depth about this in the thread Is the "Hard Problem" Just Silly?.

    I've seen various uses of the terms 'intrinsic' and 'extrinsic,' but as it concerns consciousness, these terms relate to the dichotomy between properties that can be understood as relationships (roughly 'extrinsic') and those that cannot (roughly 'intrinsic'). Structural and functional facts fall under the 'relational' category, as structural and functional facts are just elaborations on types of relationships.

    If a property is not relational, then it cannot be understood entirely in terms of 'what it does,' but rather has some aspect above and beyond 'what it does,' which we might call 'what it is, in-and-of itself.' Certain aspects of experience, such as phenomenal redness, appear to be intrinsic. One can think about redness entirely in the absence of structure and function (e.g. a uniform, unchanging visual field of redness) and still have something to think about. In contrast, if one thinks about a cognitive process in the absence of structure and function, one is not left with anything to think about at all, since all cognitive processes are defined in terms of structure and function.

    As for emergence, please see this paper: http://jamaica.u.arizona.edu/~chalmers/papers/granada.html [Broken].
    Last edited by a moderator: Apr 21, 2017 at 11:08 AM
  9. Dec 23, 2004 #8


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I don't think this analogy flies. Here we would say, "The way we subjectively experience time is not indicative of the true nature of time." Regardless of concerns about intuition, this claim must be taken to at least be plausible. But in the case of consciousness, this becomes "The way we subjectively experience is not indicative of the true nature of subjective experience." Regardless of concerns about intuition, it is not clear that this claim is even logically plausible.
  10. Dec 23, 2004 #9


    User Avatar
    Gold Member

    Thanks for the clarifications.
    Nevermind, I don't know.

    Why not take "zombies are nomologically impossible" as an axiom?

    Yes, I've been following, or trying to follow, that thread. I wish I could respond to his and your ideas, but I don't yet see how p-consciousness, or experience or whatever, is determined to be non-functional. I'll keep searching and reading.
  11. Dec 23, 2004 #10


    User Avatar
    Staff Emeritus
    Gold Member

    I'm beginning to think I should start a thread in the logic forum on what constitutes a proper analogy. I think my analogies have been mostly sound, because I haven't been using metaphorical analogies, but rather formal analogies. If the argument form is retained judiciously and the amended argument is either inconclusive or gives a conclusion known to be false, then any argument of that form is either inconclusive or invalid.

    But to defend this specific analogy, all you can say against it is that you don't think it flies, because it is not clear that your counteranalogy is logically plausible. Well, I think it is logically plausible for what we think about subjective experience to not be indicative of the true nature of subjective experience. This is exactly my qualm with such plausibility arguments. What two people find to be plausible is not always the same.

    There is no difference formally between entailment and material implication.

    There's the rub right there. You can essentially replace "facts about structure and function" in 4 with "physical facts" due to the formulation and end up with this:

    1. Not all the facts about p-consciousness are physical facts.
    2. Therefore, physicalism cannot entail all the facts about p-consciousness.

    We're left with an argument that looks awfully circular.

    Anyway, I finally got around to reading the entire paper earlier this morning, and I'm currently working on the follow-up paper linked to from this paper. I'll get into a more detailed analysis when I've finished. I'll point out this one little bit here:

    • The facts about experience cannot be an automatic consequence of any physical account, as it is conceptually coherent that any given process could exist without experience.
    • Given the extremely plausible assumption that changes in experience correspond to changes in processing, we are led to the conclusion that the original hypothesis is impossible, and that any two functionally isomorphic systems must have the same sort of experience. To put it in technical terms, the philosophical hypotheses of "absent qualia" and "inverted qualia", while logically possible, are empirically and nomologically impossible.

    How this little hiccup could go unnoticed I don't know. He doesn't technically contradict himself, as in both cases, he says that physical functions without qualia are logically coherent. But there is a conflict here nonetheless. He first says that a physical account cannot be complete because of the logical coherence of function without experience, but later says that function without experience, while logically possible, it impossible in reality.

    This again highlights my qualm with arguments of this type. You cannot falsify an empirical hypothesis by appealing to logical possibility. You falsify an empirical hypothesis by appealing to empirical facts, and he explicitly states that the evidence he uses to falsify the physicalist hypothesis is empirically impossible. By analogy (I know, another analogy), one can say that evolution is falsified because Last Thursdayism is logically coherent, even though no empirical facts exist that either contradict evolution or suggest Last Thursdayism.
  12. Jan 7, 2005 #11
    This statement:

    "You cannot falsify an empirical hypothesis by appealing to logical possibility. You falsify an empirical hypothesis by appealing to empirical facts"

    ..... is where I have a problem with your argument. To me, it seems you are doing here exactly what you are claiming Chalmers and friends and doing. You have a built in assumption that puts an empircal hypothesis at the top of the food chain even though that's the metaphysical position in question. With this view, it would be impossible to disprove an empirical hypothesis, even if empiricism(using the physicalists definition) is not the most accurate model of reality(which is exactly what Chalmers is saying). This is the exact same issue I was pointing out in the other thread about this particular situation being trickier than any other issue.

    I could have misunderstood your position but this view looks like it's ok to allow a judge to sentence himself. It's similar to the question I've asked before. Should an argument against materialism be held to materialists standards? Not to me.
  13. Jan 8, 2005 #12


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Your analogy treats phenomena-as-representations and phenomena-as-themselves as equivalent, which is where it falters. Although the outer structure of the argument about time is identical, the facts about the our epistemic access to the terms involved are not. We know about time in an indirect way; we know about experience directly.

    Even if you are uncomfortable with accepting qualia as straightforward intrinsic properties, we should we be able to agree that we still have a case of beliefs about external phenomena vs. beliefs about internal (mental) phenomena. The seperate issues of epistemic access involved here are still enough to make direct analogies between the two dangerous.

    For instance, if I note that my intuitions about the subjective pain of another can easily be mistaken (e.g. in the case where a person is only acting as if he is in pain), it does not follow that my intuitions about my own pains can be mistaken. Sure, I can be mistaken about what the experiential pain represents (I may take a pain in my foot to be indicative of broken bones when in fact there is no damage to the bones); but can I really be mistaken about the sensation of pain itself?

    Even if you decide that one can be mistaken about one's own subjective experiences, you must admit it is a much murkier issue than being mistaken about external, physical circumstances like the nature of time or the state of the body. There are significant issues of epistemic access here that should be acknowledged in a way that your analogy about time does not address.

    That's not a circular argument, that's just straightforward logic. Any logical deduction from a set of premises could be made to look trivial in the way you are making this argument appear trivial. For instance,

    1. A is an X.
    2. All Xs are Bs.
    3. Therefore, A is a B.

    We can replace X in 1 with B, and wind up with

    1. A is a B.
    2. Therefore, A is a B.

    There is, of course, a sense in which the conclusion "A is a B" is already 'assumed,' in that we assume the premises are true, and if the premises are true then the conclusion must be true. But that's just what logical deduction is; if it were not the case that the truth of the conclusion is already 'contained' in the premises, we could not derive a logical deduction in the first place. The reason this argument is not circular is because it requires us to make a logical connection between two distinct premises: A is an X, and All Xs are Bs. The conclusion stands on more than one leg; knock any of those legs down, and the conclusion falls with it. In a circular argument, the conclusion can only be felled by denying the singular premise which happens to already assume the conclusion; circular arguments only stand on one leg.

    The argument about subjective experience is not circular, for the same reason the toy example above is not circular. It assumes the truth of a set of distinct premises, none of which is identical to the conclusion, and from these premises, it derives the conclusion.

    I know your main grievance is with the premise, "Not all the facts about p-consciousness are facts about structure and function." We can certainly question the veracity of this premise, and I will grant that there is some sense in which the only support for this premise is something like intuition (although I don't think 'intuition' is necessarily the best word, and as I alluded to above, 'intuition' about subjective experience is not as flimsy as intuition about the external world). But while this premise may be controversial, it certainly does not make for a circular argument. Let's list the argument form again:

    1. Facts about structure and function can only entail further facts about structure and function.
    2. Physicalism only deals with structural and functional facts.
    3. Therefore, physicalism can only entail facts about structure and function. (1 & 2)
    4. Not all the facts about p-consciousness are facts about structure and function.
    5. Therefore, physicalism cannot entail all the facts about p-consciousness. (3 & 4)

    Suppose proposition 2 and 4 are true, but that proposition 1 is false. Then it straightforwardly follows that 3 and 5 are false as well. If this argument were circular because the truth of the conclusion is already assumed in propsition 4, then it should be logically impossible for 4 and 5 to have different truth values. But I have just demonstrated a logically consistent case where they do have different truth values, because the truth of 5 critically depends upon the truth of propositions 1 and 2, in addition to 4. Therefore, this argument is not circular.

    This isn't a conflict. He is merely saying that the theory (physicalism) cannot be true, because the theory allows kinds of events that cannot happen in reality. If there were some physical theory that allowed massive bodies to travel up to speeds of 2c, then although this theory might be logically (mathematically) coherent, it would be false, because it is impossible in practice for massive bodies to travel at speeds exceeding c. Likewise, Chalmers' argument is that the theory of physicalism is false because it allows for theoretical events (normal human brain function without qualia) that cannot occur in nature.

    Physicalism is falsified because the set of all possible worlds consistent with physicalism contains worlds that are inconsistent with the way our world actually works. For example, physicalism is consistent with zombie worlds, where qualia do not exist. As such, physicalism does not give an adequate account of qualia (otherwise it would not be consistent with zombie worlds). Therefore, physicalism does not give an adequate account of all the existent phenomena in nature. Therefore, physicalism is false.

    Your analogy of Last Thursdayism is disanalogous because both the theory of evolution and Last Thursdayism account for all the phenomena (in this case, just limited to the existence of the various life forms on Earth) in need of explanation. At best, Last Thursdayism casts some measure of doubt that evolution is true, but it does not falsify it. Physicalism is said to be falsified not because there are competing theories that explain the same phenomena as it does. It is falsified because it is supposed to be a theory that accounts for all the phenomena that actually exist in nature, and it ultimately fails to do so in the case of subjective experience.
    Last edited: Jan 8, 2005
  14. Jan 8, 2005 #13


    User Avatar
    Staff Emeritus
    Gold Member

    But you can disprove empirical hypotheses. The hypothesis that all corporeal objects behave according to Newtonian mechanics was disproven by finding instances in which there were corporeal objects that did not.

    I really no idea about an argument against materialism, but that isn't what this is. This is an argument against the general hypothesis that consciousness has a physical explanation. This hypothesis itself is very difficult to disprove, maybe even impossible, simply because I don't see any way to exhaustively test every possible physical explanation. That's why researchers don't work to invalidate wide encompassing theoryesque hypotheses such as those. They seek to falsify specific hypotheses that postulate one physical explanation.

    I will add here that I didn't mean to imply that empirical hypotheses are impossible to rule out without testing. Hypotheses that actually are logically impossible can be. What you cannot do is claim that a hypothesis is logically impossible because a competing hypothesis is logically possible. Otherwise, you run into arguments such as the argument against evolution that I presented:

    It is logically coherent that all of the facts about life can be without evolution.
    Therefore, evolution is logically impossible.

    That is essentially the form of the specific argument of Chalmer's that I was criticizing, and is clearly not a valid argument form.
  15. Jan 8, 2005 #14


    User Avatar
    Staff Emeritus
    Gold Member

    You know what? I suppose you're right about that. But you still just admitted that the argument does not prove anything. In order to be sound, the argument requires that we assume facts about structure and function can only entail further facts about structure and function, and that facts about consciousness are not facts about structure and function. Both assumptions are key, because they are really what it seems to me that people like Chalmers are attempting to prove, which, of course, they do not.

    There you go. That's where the argument fails. Physicalism does not allow a possible world where all of the functional and structural facts of human nervous systems obtain without consciousness. I'm not even sure why anyone would think that. In fact, physicalist theories of consciousness specifically state that consciousness is the product of exactly these structural and functional qualities and that any time we have these qualities, we have consciousness. Beings with all of the nervous qualities of humans that are not conscious are inconsistent with physicalism.
  16. Jan 9, 2005 #15


    User Avatar
    Gold Member

    Okay, I'm trying to understand. I think you guys are arguing about two problems I'm just reading about in Daniel Stoljar's "Physicalism": "The epiphenomenal ectoplasm problem" and "The modal status problem". I can't really sum them up so here they are:

    "The epiphenomenal ectoplasm problem

    (Cf. Horgan 1983, Lewis 1983.) Imagine a possible world W that is exactly like our world in respect of the distribution of physical and mental properties, but for one difference: it contains some pure experience which does not interact causally with anything else in the world -- epiphenomenal ectoplasm, to give it a name. The problem this possibility presents for (1) [Physicalism is true at a possible world w iff any world which is a physical duplicate of w is a duplicate of w simpliciter.] is that, if (1) provides the correct definition of physicalism, and if physicalism is true at the actual world, then there is no possible world of the kind we just described, i.e., W does not exist. The reason is that W is by assumption a physical duplicate of our world; but then, if physicalism is true at our world, W should be a duplicate simpliciter of our world. But W is patently not a duplicate of our world: it contains some epiphenomenal ectoplasm that our world lacks. On the other hand, it seems quite wrong to say that W is an impossibility -- at any rate, physicalism should not entail that it is impossible.

    In order to solve the epiphenomenal ectoplasm problem, we need to adjust (1) so that it does not have the truth of physicalism ruling out W as a possible world. While there are a number of different proposals about how to do this, the simplest is due to Frank Jackson (cf. Jackson 1993. For earlier proposals and further discussion, see Horgan 1983 and Lewis 1983.) He proposes replacing (1) with:

    (2) Physicalism is true at a possible world w iff any world which is a minimal physical duplicate of w is a duplicate of w simpliciter

    By ‘minimal physical duplicate’, Jackson means a possible world that is identical in all physical respects to the actual world, but which does not contain anything else; in particular, it does not contain any epiphenomenal ectoplasm. Unlike (1), (2) does not have physicalism ruling out W, and so (2) is preferable to (1), as a statement of physicalism, and it is (2) with which we shall work in this entry."


    "The modal status problem

    Some philosophers (e.g. Davidson 1970) have thought of physicalism as a conceptual or necessary truth, if it is true at all. But most have thought of it as contingent, a truth about our world which might have been otherwise. The statement of physicalism encoded in (2) allows a way in which this might be so. (2) tells us that physicalism is true at a world just in case the world in question conforms to certain conditions. But it leaves it open whether or not the actual world conforms to those conditions as a matter of fact. Perhaps it is not true of our world that a physical duplicate of it would be a psychological duplicate. If so, physicalism would not be true at our world.

    But for some it is puzzling that physicalism is stated using modal notions (i.e. notions such as possible worlds) and nonetheless is contingent. To see the problem, notice first that, supervenience physicalism tells us that the minimal physical truths of the world entail all the truths; hence

    (3) The minimal physical truths entail all the truths.

    Now suppose that S is a statement which specifies the minimal physical nature of the actual world and S* is a statement which specifies the total nature of the world. (It might be that neither S nor S* are expressible in languages we can understand, but let us set this aside.) If supervenience physicalism is true, it will then be true that:

    (4) S entails S*

    On the other hand, (4) is clearly a necessary truth. However, if (4) is a necessary truth, how can physicalism be contingent? After all, (4) seems equivalent to physicalism. But if the two are equivalent, how can one be necessary and the other contingent?

    But the response to this problem is straightforward. (4) is necessary, but it is not equivalent to physicalism. Rather, (4) follows from physicalism given various contingent assumptions, in particular the assumptions that S and S [sic] are the statements we say they are -- it is contingent fact, for example that S* summarizes the total nature of the world. On other hand, (3) is equivalent to physicalism but it is not necessary. (It is important to bear in mind here that not all entailment claims are necessary. Consider ‘my aunt's favorite statement entails my uncle's favorite’ -- that statement is contingent even though it is most naturally thought of as an entailment claim.)"
  17. Jan 9, 2005 #16


    User Avatar
    Staff Emeritus
    Gold Member

    I like this, but don't even think it's pertinent to this discussion. I'm not personally a physicalist because I hold out the possibility that some phenomenon, such as epiphenomenal ectoplasm or anything else that is not physically explainable, does exist (simply because there is really no way to ever know - you cannot prove that something doesn't exist). I just think that any argument I've seen attempting to prove that consciousness is such a phenomenon fails. Any argument that holds as a premise that physicalism is somehow consistent with worlds in which properly functioning human nervous systems exist without consciousness is patently absurd.
  18. Jan 9, 2005 #17
    What is there to face up to, when the answer is to be had within consciousness itself? Certainly there would be no means by which to comtemplate any of this if we weren't conscious were we? So why do we attempt to look outside of the fact that we're conscious to find the answer? If consciousness tells us everything about the world we know (which it does), why don't we just learn to listen to ourselves?
    Last edited by a moderator: Apr 21, 2017 at 11:34 AM
  19. Jan 9, 2005 #18
    In the history of things, I think have to take some responsibility for Chalmers' sometimes stating his argument in these terms. When we were in grad school together I was pretty relentless in forcing this on him as the fundamental problem (something he acknowledges in an endnote in his 1996 book). So I feel an obligation to say that at one point long ago I had the same qualm you state here, and I set out to settle it.

    Taking the two premises one at a time: It turns out that if you take some simplified model like the Life world, it is pretty easy to prove something essentially like the first premise using mathematical induction. The base case is trivial. The inductive case isn't much harder: Assume that a Life world with n number of cells only exhibits properties of structure and function (and a few other related properties that are easy to identify, like historical properties and location based properties), then what properties can a world of n+1 cells exhibit?

    Anyway, I didn't reproduce the proof in my book because it is tedious reading and I imagine Dave feels the same way. Most people can see the intuitive point without needing a formal mathematical proof. But it is reassuring to walk through the proof for oneself, and I encourage you to try it out yourself.

    As for the claim that there are facts about consciousness that are not facts about structure and function, that's pretty explicitly an empirical belief based on observational evidence. I argue as much in my book. Perhaps the observations are wrong? It is possible, but the observation seems highly replicable across people, cultures and time. Even people who disagree with the premise (including Dennett himself!) often say that their own observation of their own consciousness seems to deliver similar observational evidence, but they choose to be skeptical of their own observations on theoretical grounds: it conflicts with what they think they know about the brain and they also think there is no other reasonable theoretical position.

    This is a basic choice point and people can and do go either way. Dennett and some others choose to disregard the first-person observational evidence as delusion ("a user illusion"). I choose to respect the observational evidence, given its high degree of replicability. Given the empirical evidence, I am skeptical of the theory. Given my skepticism of the theory, I have tried my best to produce an alternative framework that is well-motivated and does a better job of aligning with all the observational evidence.

  20. Jan 9, 2005 #19


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    This has certainly raised the discussion to higher level, and I certainly hope we can maintain it. Since I pretty much adopt the Dennett position, I note the massive experimental psychology evidence he adduces in Consciousness Explained directed at undermining our naive assumptions about the reliabilty of our intuitions based on casual questions of people in various cultures.

    There is no doubt at all that we are all, !Kung or Swede, in the same boat regarding our relation to our consciousness, so multiplying evidence of that kind doesn't add anything solid to the discussion. Or so a believe; I am prepared to be convinced otherwise.

    On another note, I am skeptical of mathematical induction arguments on simple cellular automata. I suspect they miss the autocatalytic properties of complex systems which can produce emergent phenomena (in the weak sense).
  21. Jan 9, 2005 #20
    Dennett is very explicit in his writings that his heterophenomenology is a kind of 3rd-person absolutism, and the clinical phenomena he discusses are interpreted from that point of view. Given his starting point, his arguments are actually for a conditional, "If 3rd-person absolutism is true, our first person observations should be rejected." I agree with that conditional. I would even go further and suggest that there are a whole host of more fundamental ways that what seems to be true of consciousness seems inconsistent with our physical image of the world.

    From Dennett's starting point, what needs explanation primarily are not facts about consciousness itself (from the heterophenomenological starting point, we're not sure consciousness exists). Our primary targets of explanation are instead facts about why we make the reports that we do about consciousness. If those reports are explained, then our primary explanatory target is explained, and it may just turn out that many of the reports themselves are delusions (i.e., Dennett's 'user illusions'). This is exactly what Dennett argues.

    Stepping back, it is clear that Dennett's work is actually a two step argument. In step one, he delivers a detailed demonstration that there really is a problem of other minds, where "mind" is understood in its traditional reference to subjectivity. It's fun to see it done in such detail, but it is not startling work. In step two, he generalizes from this to the conclusion that no one has a mind in its traditional sense.

    Dennett never directly argues for step two, and instead makes the choice inevitable by the way he constructs his methodology. It's built into his starting point.

    There is a more neutral starting point that tries to give more equal weight to the first person observations. It is this: The first-person observations present consciousness itself as an explanatory target. The third-person observations present the associated cognition as an explanatory target. From this starting point his arguments lose their force, and it leads to a quite different project from Dennett's, one that tries to respect both sets of observations as presenting explanatory targets.

    Given a choice of starting points, I choose the more neutral starting point partly because I find Dennett's starting point to be question begging, and partly because when I try to accept Dennett's suggestion that the apparent intrinsic quality to my subjective existence is a kind of user illusion, the cognitive dissonance is overwhelming. I find myself immediately swimming in a thick sea of contrary evidence that I have to continuously reject as delusionary. Essentially, every aspect of my experienced world contradicts the theoretical conclusion, and the only life raft the theory gives me is its own internally generated prediction that I will feel that way. 3rd-person absolutism begins to seem too ideological ("Who are you going to believe, me or your own eyes?") For me, rejecting Dennett's methodological stance becomes a question of intellectual integrity.

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Facing Up to the Problems of Facing Up to the Problems of Consciousness
  1. Face transplants (Replies: 3)

  2. My face (Replies: 51)

  3. Face Analyzer (Replies: 19)

  4. Orbs with faces (Replies: 41)