Saturday 2 January 2016

10b. Harnad, S. (unpublished) On Dennett on Consciousness: The Mind/Body Problem is the Feeling/Function Problem

Harnad, S. (unpublished) On Dennett on Consciousness: The Mind/Body Problem is the Feeling/Function Problem

The mind/body problem is the feeling/function problem (Harnad 2001). The only way to "solve" it is to provide a causal/functional explanation of how and why we feel...








53 comments:

  1. In this article, Steven Harnad presents the method of heterophenomenology which is a third-person view that is used to better understand the view and perceptions of the subject. This method is described as being a highly objective method that allows to objectively understand what the subject is thinking and perceiving. However, based on the description of this approach, this method seems to be inherently subjective.
    There is a lot of interpretation that is needed to get something sensible out of the results obtained from studies using the heterophenomenology approach. In one of the examples given, the subject is presented with a yellow blob on the screen and then asked many questions by the experimenter. Without some sort of interpretation, this study cannot give anything useful to better understand cognition. Furthermore, some of the sentences used by the experimenter, such as ‘’Does the yellow blob really move?’’ or ‘’Does it really move or do you just judge that is moves? ‘’ (Harnad, p. 10) can bias the subject to certain responses. The way these questions are asked, as a subject, I would be really tempted to tell the experimenter that I do not think that the blob moves as this seems to be the right answer the way these questions are structured. I know that this is one example of many others, however, it still clearly shows the biases that can be created by the experimenter. This example also shows the amount of interpretation that is necessary to give some sensible to the field of cognitive science. How does the fact that the subject sees or does not see the blob lead to a better understanding of how the subject thinks? Whether the subject sees or does not see the blob requires a considerable amount of interpretation in order to try to understand the way the subject is thinking. Interpretation of these results can add a strong bias.

    ReplyDelete
    Replies
    1. Yes, T4/heterophenomenology probably needs some interpretation in order to predict what subjects are feeling; but that does not change the question at hand: Are the feelings that are being predicted from the T4 correlations really being felt, or are they just doings?

      Delete
  2. Is Dennett really trying to answer how/why we feel? The commentary seems to be fighting him on not answering that question but did he set out to answer that question or is that the question you’ve given him and tried to fit his article into? It’s fine to say his question from the beginning is the wrong question but he set out to explain a certain question in his article and to then just frame the entire article by a new question and critique it doesn’t seem fair. Although, I do share the frustration because I really couldn’t figure out what question he was asking and what he was answering with his remarks

    “Defenders of the first-person point of view are not entitled to this complaint against heterophenomenology, since by their own lights, they should prefer heterophenomenologys treatment of the primary data to any other. Why? Because it does justice to both possible sources of non-overlap.

    Who are the defenders of the 1st person view? what is that view? and what is being defended?”

    This quote reiterates my frustration, there were many parts in his original article where he seems to have created a debate without giving it much context or explaining what it is.

    “The hard part will be deciding (on what grounds?) which features of which states to declare to be qualia and why.”

    If we are to assume he should be trying to solve the hard problem with his heterophenomenology then he could be making an interesting point here. It seems to me that in this quote he may be trying to break down “feeling” (although he uses “qualia”) into smaller parts of definable features which could be measured. While I have no idea how that could practically work, I find it an interesting thought experiment in trying to understand feelings better.

    “ If there can indeed be such a Zombie, the how/why difference under discussion would be that difference between actually feeling and merely functioning-as-if-feeling. If there cannot be such a Zombie, then you need to explain, causally/functionally, exactly how/why there cannot.”

    I do agree with your critique on his zombie questions, because I too felt it odd that he said they are impossible without explaining how or why. That to me exemplifies the misdirection from discussing how/why feelings arise if in an exact molecular duplicate he doesn’t suggest why one can feel and the other cannot. This I suppose does offer evidence that he really was trying to explain heterophenomenology as a way to approach the hard problem

    “I am not A, because I definitely believe, and argue (vigorously and rigorously) that the "hard problem" (perhaps a misnomer too -- but for me, the problem of giving a causal/functional explanation of feeling: explaining how and why we feel) has not been solved, and is insoluble.”

    Regardless of what Dennett is proposing, I have a hard time understanding the point of view where you claim what he is proposing is useless because you yourself believe giving a causal/functional explanation of feel is not possible. Of course it makes sense that if something is unsolvable then attempts at trying to solve it are a waste of time, but I’m not convinced with the evidence so far that the problem is in fact insolvable. Assuming other beings are conscious, feeling creatures and there are things that are not feeling then there must be a line drawn or a graded scale with finite portions of what does feel and what doesn’t feel. And if there is a line to be drawn then there is a feature that will distinguish between them and I don’t see how that can’t be solved. If the problem is just that we can never know whether or not we have solved it then I see that as no reason to diminish Dennetts efforts in progressing methods in trying to find possible solutions. But often when I hear that the problem can’t be solved I regress into assuming the opposing party is suggesting nothing less than a magical explanation for how it occurs.

    ReplyDelete
    Replies
    1. Hi Jordan,

      I also share your frustrations. Dennett’s approach is a little behviourist-esque to me, which is at some point mentioned in Dennett’s own paper, which makes me inclined to ask the same question that you expressed in your commentary. I'm not sure if these two papers are even trying to address the same things. Thus, what I’m having trouble understanding and what I didn’t get from Harnad’s response is what Dennett’s actual stance is on the how/why issue of feeling. Did Dennett miss it entirely because he thinks it is irrelevant or does his model just not happen to account for it? In other words, does Dennett not address the hard problem of feeling because its beyond what he wants to address and he chose to exclude it, or because he doesn’t think its important at all and doesn't feel it should be addressed whatsoever? I’m trying to gauge academic attitudes towards the hard problem through these readings. It doesn't seem that anyone (apart from Fodor) are willing to explicitly state that a field of research is not worth what we are putting into it. I would love to hear an argument from someone who has deemed the hard problem completely unimportant, though I’m not sure how close we’ve gotten to that in the class, and I'm not even sure it exists at all.

      Delete
    2. Dan Dennett was trying to show there is no hard problem because there are no feelings, just "beliefs about feelings."

      My rebuttal was mostly addressed to that:. Yes, there are feelings. Yes, there is a hard problem of explaining how/why organisms feel. No, feelings are not just "beliefs about feelings." (But the part about the hard problem being insoluble is just "Stevan Says.")

      Delete
    3. I found it helpful when you framed the hard problem for kid sib, as “Unlike you, Dan, I stand ready to admit that neither I nor anyone else has even a clue-of-a-clue about how one could cash in that “somehow” functionally.

      I like the how and why questions for the Turing-equivalent frog as well – the Zombie hunch is laid out in Dennett’s paper to be sufficient to claim that we cannot understand qualia for example or any first person science, and that instead we should rely on third person science. I think asking WHY we feel and recognizing that there is a hard problem legitimizes first person experience in a really helpful way.

      Delete
  3. “What position? What proposition is being traded for what proposition? One loses sight of what is at issue...”

    What IS the issue was my issue too. Professor’s Harnad’s response to Dennett parallels a lot of the questions that I had about Dennett’s outlook, but I think I’m now left a little more confused. Dennett, if I’m interpreting Harnad correctly, is focusing on the stated results of consciousness (learning, misperception, etc.) but is missing the mark when it comes to actually studying the felt experience of being conscious and doing all of the aforementioned things. It seems like he’s going for this measurement of meta-cognitive results. I was very confused by the inclusion of Kant’s question, as it seemed like it didn’t really equate to what we’ve been discussing in class, and now I know why. But I (as I mentioned in Jordan’s commentary) am perplexed. It feels like Harnad and Dennett are arguing past each other.

    But, if I’m just analyzing this rebuttal/critique, it seems to me that Dennett is arguing against something that doesn’t really exist. The way he describes the “first person perspective” doesn’t really encapsulate the B team’s analysis at all. It sounds like Dennett is constructing the B team as believers of introspection, as in, only the person’s own feelings are relevant, and because these feelings cannot be experienced by others (other minds problem) then the ONLY thing the B team has in their arsenal is introspection. He seems to dismiss that there is a middle ground (the C Team, as noted at the end of the article) in which both A and B methods are utilized. How many people area even arguing that introspection is a viable method of research?

    I don’t think, however, that Dennett’s methodology is not useful. I think what he is trying to do could lead us to advances in cognitive science. But I don’t think he’s studying consciousness. This seems to be somewhat of a theme in this field of research, with researchers and scientists throwing darts at the dart board of cognitive science and always missing the bull’s-eye that is consciousness and the how/why problem, only to hit the slightly off mark but still important results of our conscious experience (like perception, change blindness, etc.). On top of that everyone is arguing about what the bull’s-eye actually is and what it means, and how to hit it. And some people think that we can never hit that bulls-eye. Lots of synonyms are used, causing more confusion. A confusing game of darts is had by all.

    ReplyDelete
    Replies
    1. Heterophenomenology is just T4. Both the "A Team" and the "B Team" endorse T4. The real difference between A and B is whether they agree that there is feeling, and hence a hard problem of explaining how and why organisms feel, rather than just do.

      There is no C team on this question, just perhaps nuances on the B-team as to how hard the hard problem really is. (I am probably in a minority in thinking it's insoluble: Chalmers doesn't thinks so -- but he gives no hint of how or why...)

      Delete
  4. A few thoughts related to this paper:

    1. This article articulated a lot of the issues I had with Dennett’s paper. I think this passage sums it up best:

    D: Notice that Chalmers allows that zombies have internal states with contents, which the zombie can report (sincerely, one presumes, believing them to be the truth);”
    H: This is equivocal. A sentence in a book can be true, but it cannot be "sincere" [on the part of the book] and it cannot be "believed" [by the book]. Same is true for dynamic, on-line books. Same is true for Zombies: No feelings of sincerity, or credence, or anything...

    Dennett seems to have not paid proper attention to feelings again. Heterophenomenology is interesting in plenty of ways, but it can’t answer the hard problem.

    2. “Causal/functional explanations do not explain feeling."

    This struck me as a little odd but perhaps I’m misunderstanding. I understand that functional explanations won’t be enough. But I thought we were looking for a causal mechanism? Or is that just for the easy problems - if so, why won’t searching for a causal mechanism work for feeling? Also, what kind of explanation could work, if not a causal one?

    3. “Frankly, this pseudo-puzzle looks like it's just a consequence of the highly counterfactual premise: To suppose that something that is molecule-for-molecule identical to me could fail to have feelings sounds about as sensible as to suppose that something that was molecule-for-molecule identical to the moon could fail to have gravity.”

    There’s an interesting parallel between gravity and feeling. The moon has gravity but it was a mystery for a long time why it did (ie how and why any physical matter does). I don’t know a lot about physics but I know that part of that issue was recently elucidated with the discovery of the Higgs Boson particle, which gives mass to other particles. (Kid sib: sorry, that’s all I know).

    Could there be something analogous for feeling? Could there be some property of matter (I’m tempted to say organic matter arranged in a very specific way but I don’t know enough to restrict it to that a priori) that gives rise to nonmaterial felt states? I’m being careful not to use the word “emerge” but I feel like that’s where my idea is going. Either way I’m not explaining much - just giving a reason to be optimistic in response to Prof. Harnad’s “Bid welcome to a pessimist”.

    We know that we feel. We know that particles have mass (or we have very, very good reason to think so). But the how and why for the existence of mass in particles was not understood at all. I’m sure there were plenty of people who said it was an intractable question. The “how” is starting to be elucidated.

    And as to why it exists, I can’t give a solid answer for feeling because any evolutionarily adaptive process I can think of could still be accomplished without feeling. Maybe it’s some kind of accident of evolution in the same way that there is inefficiency in photosynthesis - it’s not adaptive, but it’s still there.

    Also, the “why” is a disanalogy because the moon doesn’t need to “survive” in the same way we do (and certainly wouldn’t need to have mass for that) so that kind of explanation may not work.

    ReplyDelete
    Replies
    1. Overdetermination and Psychokinetic Dualism

      The reason the hard problem is hard (and, "Stevan Says," insoluble) is that once you've solved the easy problem, i.e., once you've explained how and why organisms can do everything they can do, you've used up all the degrees of freedom of causal explanation. You've explained everything observable that is explainable (T4), including all the observable correlates of feeling (heterophenomenology).

      Now you're faced with the problem of explaining how and why doings are felt doings (i.e., not zombic, no just done doings). You have to explain how and why vision is felt seeing rather than just optomotor processing, how and why pain is felt , rather than nociceptive processing, how and why people mean something when they speak, rather than just doing grounded symbol processing.

      You've already got a T3/T4 explanation of everything but that (the easy problem's solution). The problem is explaining what is still left over to explain, rather than just denying its existence, as Dan does.

      It would have been so easy if it had turned out that there is a 5th force in the universe -- aside from the 4 usual forces: electromagnetic, gravitational, strong and weak atomic. If there were a psychokinetic force -- mind over matter -- that was necessary in order to do the things we do voluntarily -- cognitively -- as opposed to vegetatively. That would solve the problem.

      Trouble is there's no psychokinetic force. Only dualists (including believers in the supernatural explanations of religion) believe in psychokinetic forces. The rest of us are left with just the usual four forces (along with computational algorithms). And we have already used them all up explaining all doings with T3 and T4.

      So in every attempted explanation of feeling, feeling will keep turning out to be superfluous: causally redundant.

      But that's just "Stevan Says."

      Have a go!

      (The problem this time is overdetermination rather than underdetermination.)

      Delete
  5. Why, oh why do we keep coming back to the feeling question?
    I understand why Harnad always evokes the how and the why there is a correlation between physical brain states and feelings. He argues that the important problem is all about feeling. How it is that brain activities give raise to correlated feelings? And Harnad is right, but I wonder; it seems impossible and even illogical to ever reach an explanation to this question. Isn’t it what’s beautiful about feeling? The fact that it exists and that it might be the only thing we will never find an answer to? Let me add something odd which is kind of related to the topic. There is a theory which states that if ever anyone discovers exactly what cognition is and why it is here, the Universe will instantly disappear and be replaced by something even more bizarre and inexplicable. (I actually didn’t come out with the idea on my own. Refer to the Hitchhiker’s Guide to the Galaxy) Then that would be a reason for nature or god to have made cognition unsolvable. Or maybe nature created feelings so that we could lose our time wondering why and how they are possible! My point is yes, feelings are the important explanation cogscientists need to solve; but they might never even get close to it! I feel there is something intrinsic to cognition that doesn’t allow us to reach it; so in the meantime, can’t we focus on other aspects of cognitive science? Harnad says he thinks the hard problem is unsolvable. I would like to know then, why do we keep coming back to the same question then? Is it because all the other proposed explanations still don’t explain anything?

    ReplyDelete
    Replies
    1. Actually, we are not always coming back to the hard problem. The only (brief) appearance it made in the first 9 weeks was in Searle's Chinese Room Argument. Apart from that, it was all just the easy problem.

      It is now, in week 10, that we will devote one session to the hard problem. You can say (feel) that it's just a mystery and we should just sit back and enjoy the mystery. But if cognitive science is a science, then that's not an option: Explaining how and why organisms feel rather than just do seems a natural part of cogsci's explanatory task, don't you think?

      Delete
    2. Hi Roxanne,

      I can understand why you're frustrated by discussing the hard problem but then again that's why it's the 'hard' problem. Don't you think that this is the real crux of figuring so much out in cog sci? I don't see how considering the answer to the hard problem is illogical because feeling is attached to all that we do as humans. Just because something is probably insoluble that doesn't mean that we wouldn't want to explore it, right? Don't get me wrong, I agree that it's frustrating but I also think that because it's so intrinsic to our human experience, feeling will always be something that we're trying to better understand.

      Delete
  6. Writing this article may have been frustrating for Dr. Harnad due to the need to repeat himself over and over again, but I found it quite comical for exactly the same reason. While I was reading Dennett’s paper, I found myself constantly thinking, “okay sure, but you’re still not getting at the how/why problem here.” Dr. Harnad has clearly been training us well, and it was enjoyable to see so many of the problems I had with Dennett’s paper included in this systematic breakdown of exactly why he is missing the point. However, this article did leave me struggling with a couple different concepts.

    The (hypothetical) Zombie does not "fervently" anything, because he does not feel! He only behaves in a way that is interpretable (by us) as if he felt. If there can indeed be such a Zombie, the how/why difference under discussion would be that difference between actually feeling and merely functioning-as-if-feeling. (page 14)

    First, could a zombie like this ever exist? And if it could, wouldn’t it seem like just another person to us rather than an unfeeling zombie? Assuming T3-indistinguishability is enough for reverse engineering cognition (and several different quotes scattered throughout the article make it seem like this is indeed the case), then wouldn’t the difference between actually feeling and merely functioning-as-if-feeling be nonexistent? Or at the very least, irrelevant? We can never know if the zombie is actually feeling, we can only deduce that it is based on the way it acts. And if it functions-as-if-feeling, then it must be feeling, just like I conclude that everyone around me is feeling even though I can never be certain. So a zombie like this could never exist, just like an unfeeling person could never exist, because to function-as-if-feeling, one must be feeling. I guess this question would be insoluble, because how could we ever know one way or the other?

    A simple methodological (hence also epistemic) point about the constraints on causal/functional explanation: It works for everything else, but it doesn't work for feelings... (page 15)

    Second, why not? What’s so special about feelings? Towards the end of the article, Dr. Harnad starts to hint at an answer: “but feeling is not doing” (page 20). Feeling is not a function, so a functional explanation could not include everything we need to know. But what are feelings then? Are they in a class of their own? Is this the reason why the hard problem is insoluble? We can never discover a causal/functional mechanism for feeling because feeling is, by definition, not a function? This must be the fundamental difference between Dr. Harnad and Chalmers: Chalmers still believes we can find a causal/functional explanation for feeling, while Dr. Harnad has completely given up on this idea. And if we will never be able to reach this causal/functional explanation, where do we go from here? Do we just pack it up and go home? What other methods can we use to get at the problem of how/why we feel?

    Finally, on a completely unrelated note, quick question about how the publishing world works. Does this article remain unpublished because Dennett’s article is still unpublished? Would you be able to publish this article even if Dennett does not publish his? Or can you only publish a response to an article once that article itself has been published?

    ReplyDelete
    Replies
    1. "if it functions-as-if-feeling, then it must be feeling" -- maybe: but the hard problem is explaining how and why. Is it obvious? On the contrary, when it comes to causal explanation, feeling seems causally superfluous rather than obvious, even if it feels obvious to us master mind-readers that it is there...

      Chalmers still believes we can find a causal/functional explanation for feeling, while Dr. Harnad has completely given up on this idea. And if we will never be able to reach this causal/functional explanation, where do we go from here? Do we just pack it up and go home? What other methods can we use to get at the problem of how/why we feel?

      Good questions. What do you think. It could be just "Stevan Says" that the hard problem is insoluble. But then what might a solution look like (since psychokinetic dualism is a nonstarter)?

      But don't forget that before we can pack up and go home, there's still the "easy problem" to solve...

      (The article remains unpublished because Dennett's article is unpublished. But in today's open access world that doesn't prevent people from reading (and even citing) it! Nor does publishing it ensure that they will...)

      Delete
    2. I think you present the right idea in your next article: the best way to approach the problem is by cracking the T3 challenge. That's how we'll get the answer. But if the hard problem is indeed insoluble, wouldn't that imply a solution to the T3 challenge is impossible? If T3 explains the hard problem, and the hard problem can't be solved, then T3 must be insoluble too.

      Delete
    3. Do you really think that the hard problem is insoluble though? Of course with the insight we currently have into brain mechanisms/causes of cognizing/etc, the hard problem seems insoluble. But do you think there could never be something that solves why we feel? I suppose that would have to be something that would then explain our existence in the universe because I see feeling as enveloping everything we do, but then that would be related to feeling giving meaning and purpose to our lives etc etc which still doesn't present any casual direction for why/how we feel. Sorry if this rambles, I just have many thoughts on this topic.

      Delete
    4. T3 doesn't explain the hard problem, T3 provides a causal mechanism for the easy problem. The assumption is that if/once T3 is conceived, we will a) not be able to tell (for sure) if T3 is a zombie or is a sincere in talking about 'feeling' and b) should assume that T3 does in fact feel sincerely and does not merely act as though it does because feeling is causally superfluous (any doing can be done without added feeling) and a robot that has been built based not a causal mechanism meant to explain/perform anything we (as 'feeling' beings) are able to do should assumingly feel as well, independently of whether it is made of the same kind of organic 'stuff' as we are (though that is an assumption, not a proof)

      Delete
  7. “Chalmers allows that zombies have internal states with contents, which the zombie can report (sincerely, one presumes, believing them to be the truth);
    This is equivocal. A sentence in a book can be true, but it cannot be "sincere" [on the part of the book] and it cannot be "believed" [by the book]. Same is true for dynamic, on-line books. Same is true for Zombies: No feelings of sincerity, or credence, or anything...”

    I find it interesting that of all the “Zombie talk” that takes place in this paper, that the zombie is never taken to be a surrogate for our more familiar T3 robots (or possibly T4). Like a Zombie, T3 robots do not feel. They can act and speak like normal human beings but they are cannot feel. Consequently, they also cannot “seem” to have beliefs, because as Harnad points out to “seem” to have a belief is to “feel” as though one has a belief. I think if a parallel were drawn between T3 robots and zombies this argument would be a little more familiar, and a little more clear.

    For example, a T3 robot can be programmed to “report” their internal states to make them even more indistinguishable from real humans. People in the class find comments of Renuka and Riona’s to be sincere when they report that they are feeling sick, or tired. A Renuka or Riona can also be programmed to report seeing a moving yellow blob when presented with a particular visual array, but again, they are not feeling what it is like to see the blob the same way the real humans in the class do.

    One aspect of the Dennett article that isn’t hashed our enough for me is the piece about a Zombie being constructed identically molecule by molecule. It seems quite unbelievable to me that two things (a person and a zombie) could be considered “identical” if one has blood pulsing through its veins and oxygen in its lungs, and the other doesn’t. Harnad calls this “empty sci-fi” and this agrees with my intuition but, I’m not sure exactly how to justify it.

    ReplyDelete
    Replies
    1. If we follow the zombie hunch that something (e.g. our robot) could pass T3 without feeling, then we could make the parallel between T3 robots and zombies. However, given that we are not able to explain “how or why we are not Zombies,” or for that matter how to reverse engineer a robot to be able to do everything we can do (easy problem), I think there still is a possibility that a T3 robot can feel. Therefore, if we assume that our robot passes T3 and as a result feels, then we could not use the zombie as a surrogate for our T3 robot.

      I agree that Chalmers description of his “zombie twin” seems unbelievable, and does not seem to answer anything (it does not show the causal mechanism). But I think he was going for a more T5 molecular function description non-conscious ‘Zombie’ thing/organism (rather than the Zombie we typically think of i.e. fictional undead being/human corpse without blood pulsing, etc.). But, feel free to correct me if I’m wrong.

      Delete
  8. Before reading this article, I found Dennet’s commentary to be very interesting and a hopeful path to understanding consciousness. However, after reading this commentary, it seems that Dennet had distracted me from the core issue of why and how we have feeling. While reading Dennet’s article, I had confused the ability to predict with having a cause explanation of how consciousness would work, but I understand now that these two are not the same thing.

    However, I am still having trouble grasping one notion.
    "Am I appealing to the Zombic hunch when I ask why a frog should feel something when you shock its foot, rather than just going through the familiar functional (Turing) nociceptive story?” (2001).

    After taking the Psychology of Pain, I had understood Pain to be a conscious experience that proved adaptive in evolution to ensure organisms would avoid similar damaging situations in the future. I understand that Pain is a step further than nociception, but I don’t understand why we can’t propose this as a possible causal notion. I believe it comes down to being able to test this and since we can’t reliably produce it in one thing and not in another, then we have no way of testing its function. So perhaps this is why it is insoluble then? Or is the main flaw that nociception can explain everything without referring to the higher order function of pain?

    I understand that feeling and function have a 100% correlation in us, so 100% predictability is possible. Essentially, we should be able to mind read in the future if this is possible, but the act of mind reading doesn’t tell us why we feel something in the first place. If this is true, I think the only thing to do is abandon the search for a solution for the hard problem. The only conclusions I have now are that it is either impossible to test the hard problem or we do not have enough information or technology to allow us to test the issue as of now.

    I am very happy that Dr. Harnad tackled the issues of “unfelt feelings.” I never understand what people mean by unconscious feelings in change blindness and even in the system’s reply. I think this is a paradox in and of itself and I don’t understand why it keeps being brought up in the Hard Problem.

    ReplyDelete
    Replies
    1. Hey William!

      I like the point you raised about a potential causal mechanism/ adaptive purpose of feeling because it sort of sparked my imagination.

      "I had understood Pain to be a conscious experience that proved adaptive in evolution to ensure organisms would avoid similar damaging situations in the future."

      This seems to be the most intuitive response when we're trying to explain why nociceptive signaling is felt pain. It seems almost obvious that felt pain is more salient... or in some way enhances our behavioural response, but upon closer examination we can see why this shouldn't be the case. The neural mechanisms of nociception are fairly well understood, and there's no causal reason that a nociceptive signal should be felt at all. The behaviour that results from a nociceptive stimulus is fully explained by our understanding of nociceptive signaling and leaves no room (no degrees of freedom) for felt pain.

      This 'enhancement' by feeling is one of those just-so stories, useful in bending the putative adaptive function of feeling to fit our arbitrary whims. It is almost reminiscent of the sort of hypothesizing that we discussed in the evolutionary psychology.

      But as you say, all this of course begs the question of how to test the importance of felt cognition.

      "I believe it comes down to being able to test this and since we can’t reliably produce it in one thing and not in another, then we have no way of testing its function."

      It seems feeling is inextricably linked to behaviour. When we are awake (sometimes sleeping) and behaving we are feeling. To test hypotheses about the adaptive value of feeling, it would be necessary to separate feeling from behaviour... to have behaviour in the absence of feeling - to observe a zombie and see how they fare without feeling.

      Wouldn't it be interesting if we could...
      I've sat here for tens of minutes trying to come up with some sort of experiment that could surgically separate feeling, but without luck.

      Oh! What if we hypothesize that feelings originate because of sensory modalities? What if we temporarily removed the brain of a subject, severing their sensory connection to the outside world, but so that the brain's function could somehow be maintained for a short period. If the brain is active, what is it doing in the absence of sensory input? Does it feel like nothing to be deprived of your senses? Like a non-existence (like being asleep/dead)? Or does the brain generate its own feelings because there's no sensory input? What would this person say when you put them back in their body? Did they have feeling in the absence of any behaviour?

      So many questions!
      Then again this is all fairly baseless speculation, but interesting to talk about nonetheless.

      Delete
  9. “Only felt feelings count. If I didn't feel it at the time, I didn't feel it.”
    - From this quote, does that mean that you believe feeling is a discrete mental state? Is feeling even a mental state or is feeling itself a state, as in kinds of feelings are kinds of feeling states and not of mental states or of both? Is feeling an action?

    - Just like how Harnad paraphrased the hard problem as “How and why we are not Zombies,” I think of it as what is the purpose of feelings at all? If the answer is to differentiate between us and zombies/Turing-indistinguishable machine (if even possible), that doesn’t explain what’s wrong with being a zombie because then it’d be due to the fact that they don’t feel, and would thus be the only reason behind feelings but would be using the same term to explain itself, that the purpose of feeling is to differentiate us from things that are not feeling, and would not be meaningful. Though it’s pointless/meaningless to paraphrase the hard problem like this, for me it helps give a different kind of perspective on the problem as to using the term “why.”

    ReplyDelete
    Replies
    1. Hi Oliver,

      “‘Only felt feelings count. If I didn't feel it at the time, I didn't feel it.’

      - From this quote, does that mean that you believe feeling is a discrete mental state? Is feeling even a mental state or is feeling itself a state, as in kinds of feelings are kinds of feeling states and not of mental states or of both? Is feeling an action?”


      I think that all of this language that you suggest (or at least ask about) would only muddy the waters if we used it. Let’s just call feeling “feeling.” A feeling is merely something which is felt. If it is not felt, it is not a feeling. When people refer to “mental states,” I think that they are usually simply referring to felt states. They might be referring to the functional state of an organism when they have a given felt state, but to wrap the two up into the term “mental” just makes it easier to gloss over the Hard Problem by treating the correlation of felt and functional states as a causal explanation for these felt states. To call feeling an action would also be to conflate doing and feeling; we think that doing leads to feeling (i.e. a T3 robot could feel), but we have no idea how or why.

      Delete
  10. “Well, that doesn't sound like our data (we being the E's, the 3rd person's) but S's data (the 1st person). But who cares? Include them if you like! Call S's toothache, the feeling, a part of your data-set too, if you like. We know that sometimes at least, toothache is indeed correlated with S's behavior. So what do we lose if we just suppose that it's always true. The correlation is perfect, 100%, and you, Dan, the hetero E, "own" both, the behavior and the feelings. They're both your data, in the hetero batch.”

    “Levine seems to want to deny you the right to count the feelings themselves as part of your hybrid data-set. I'm allowing you. It won't help...”

    I’m not sure I quite understand why including 1st person data is not useful. By using 1st person data with third person data for a set of subjects, we could come closer to finding a pattern between mental activity and subjective feelings. This may not help solve the problem that we “feel” differently, but to answer the question of how we feel, comparing first and third person data is a good starting point. Am I misunderstanding the problem here?

    “You've lost me. I don't for a minute doubt that eventually we will be able to do 100% mind-reading via functional correlates of feeling. So surely that's not at issue either. What will be left unexplained by this perfect predictability of feelings from their functional correlates is how/why there are feelings at all. The explanation, not the prediction! Back to square one. Let us not waste our time on veridicality, either S's or E's...”

    If we are talking about why we feel, shouldn’t we be taking a more evolutionary perspective? Perhaps looking at the functional correlates can help us better understand how those functional correlates came to be. For example, let’s say we found DNA methylation marks in mothers undergoing a stressful time while pregnant, and then noticed the progeny having the same methylation and different physiological and psychological characteristics showing increased levels of anxiety. The progeny with the increased stress levels would “feel” more anxious. First person data could try to help us compare how they feel to how others do. We would have the start of a causal explanation for why these individuals “feel” things in a more anxious way than others. This may be a huge over-simplification, but epigenetic studies, like the example given above, in my opinion, are the start of a new understanding for why we feel differently and where that stems from. Epigenetics use the functional correlates of aspects that influence our feeling, like stress.

    ReplyDelete
  11. This is an exhaustive review of how Dennett’s commentary does fail to address the how/why we feel argument. I am convinced that Dennett has not proposed a solution to understand all of cognition, but I am perturbed by a pessimistic response to all of this. If we decide that there is no way of solving the hard problem, or at lease no way to study its causal mechanisms, then doesn’t this mean that cognitive science should cease to exist as a discipline? Or is it rather that cognitive science should disregard the problem and focus on how and why we can do what we do? If so is this even possible to do without consultation of feeling? Rather than take on a wholly pessimistic approach, I believe it should insetad be noted that at this point there is no appropriate methodology available to causally answer the hard problem (which I think is what Harnad meant with the use of the word “pessimist”). Taken further, this should not stop our attempts at unearthing this methodology, because to do so, in my opinion, is to stop studying cognitive science.

    “"as they put it"? But why pick equivocal cases where we really aren't quite sure whether the Ss are feeling anything at all? Why not pick an open-and-shut case like feeling or not feeling warm-now? Not thermosensitivity or thermoregulation or thermolucution. Not being warm now (that's function); not being ready to say an instant later "well, maybe I didn't feel warm then after all, just a little tense" etc. None of these qualitative variants matters a whit. It is that anything is being felt at all that is at issue here. Exotic data on priming and implicit processing don't have any bearing on this at all!”

    This complete rejection of experimentation on whether someone actually feels something or not, goes a bit far. To discover the how/why of feeling it would be useful, in my opinion, to compare the difference between when something is felt and when it is not felt. I have no clue what this difference would or could be and I am not suggesting that Dennett was anywhere close to deciphering this problem. I just believe that in order to understand, causally, how we feel, it would be useful to observe “something” when we feel and when we don’t. This doesn’t push us any closer to an answer but I believe outright rejecting this type of methodology may be unwarranted.

    ReplyDelete
  12. This article clarified a lot of questions I had about Dennett’s article.

    According to Dennett, if a robot is able to pass the Turing Test heterophonmenologically, it is proof that the robot can feel. Furthermore, Dennett believes that heterophenomenological methods will allow us to understand any aspect of one’s consciousness. However, this isn’t true since, the Turing Test cannot help us solve the Hard Problem. This is because as of right now, the only one that can ‘feel the feelings themselves is the part of the first part, the feeler’. In other words, we are still left with no way to objectively measure feeling since we do not know ho feelings happen. It seems that every attempt at solving the Hard Problem ends up being blocked by the other minds problem. 

    ReplyDelete
    Replies
    1. I completely agree with what you said. I would even add that more than being able to measure feelings, the questions of why there are feelings at all is the one that remains unanswered. Dennett said that heterophenomenological metholds will allow to account for feelings and predict them, which even if possible/realistic, still does not answer the question of why/how feelings arise in the first place. And him saying that these feelings can be observed still does not guarantee that they are actually going on within supposed “feeler”, only that person can say whether it’s the case or not. And even if it were the case, as you said, the problem of other minds is the ultimate barrier to anything.

      Delete
  13. I was reading the comments after the one Harnad wrote in the claustrum article and many were discussing fetal ability to feel. Most people said they don’t believe fetuses have conscious thought or awareness. I read some interesting articles like this one http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1440624/ that try and deduce whether or not fetuses could feel pain. This one says no, even though they have the proper neurological circuitry at 26 weeks for nociception. They don’t have concessions and without consciousness you can’t have pain. The reasons they give for the lack of consciousness are really interesting and I suggest reading the article but basically the mind doesn’t develop until birth. If they are right, then the fetus is like a zombie and maybe their development into a feeling human being can give us more insight into the hard problem.

    Also, what about sleep? I think it’s pretty well agreed that you are not conscious during sleep but can’t you feel? I mean you can have intense dreams that definitely feel like something when you’re in them – or is it only once you’ve woken up that this “feeling” gets attached to the sleep state? You definitely can’t “do” anything in sleep so if you can indeed feel maybe this could be an example of separating the doing from the feeling.

    ReplyDelete
    Replies
    1. Hi Ailish,

      Thanks for sharing! I found that quite an interesting article to read. It brings up many questions about the distinction between the states of consciousness and how we develop consciousness. The fact that the fetus in development is almost always ‘asleep’ and ‘unconscious’ (which is partially due to endogenous sedation mechanisms) does counter the idea that fetuses feel pain.

      The article mentions: “When a primary caregiver points to a spot on the body and asks “does that hurt?” he or she is providing content and enabling an internal discrimination and with it experience. This type of interaction provides content and symbols that allow infants to locate and anchor emotions and sensations. It is in this way that infants can arrive at a particular state of being within their own mind. Although pain experience is individual, it is created by a process that extends beyond the individual.”

      I feel like this is a bit of stretch. I do not think that pain “is created by a process that extends beyond the individual.” When thinking of an extremely premature newborn (less than 24-weeks gestational age) that required resuscitation and mechanical ventilation to breath, there are noticeable signs that the infant may be in pain regardless of the fact that their eyes are closed and they are unable to fully interact/communicate with the world. The infant may clench their fists, arch their back, etc. In this sense, I feel that a fetus within the womb at the same gestational age may also feel pain (thereby, also have the capacity for consciousness).

      The questions you brought up about sleep are also ones I was wondering about too. As consciousness = feeling, I believe that we only have access to the inputs of the dreams when we are conscious and thus we are feeling everything associated to the dreams in this conscious state. Again, I think this is similar to the offline data that goes on in our brains that provides inputs for us to feel/think about the dream.



      Delete
  14. “From recorded verbal utterances, we get transcripts (e.g., in English or French, or whatever), from which in turn we devise interpretations of the subjects speech acts, which we thus get to treat as (apparent) expressions of their beliefs, on all topics. Thus using the intentional stance, we construct therefrom the subjects’ heterophenomenological world. We move, that is, from raw data to interpreted data: a catalogue of the subjects convinctions, beliefs, attitudes, emotional reactions, … (together with much detail regarding the circumstances in which these intentional states are situated)” p.5.

    Few issues with this statement:
    For one, Dennett is saying that all these things (convinctions, beliefs, attitudes, emotional reactions, etc) are somehow correlated with feelings, and I am not quite sure I understand what’s his basis for saying this… How and why? What are “beliefs, attitudes, emotional reactions”? How are they correlated with feeling? But also, why do we care? Our question is not about beliefs of feeling, but feeling itself!
    And let’s say we give him the benefit of the doubt, let’s assume that he’s right and these things do in fact correlate with feelings and he finds a causal explanation linking the two, what more does this explain? It may allow us to predict but it does not allow us to causally explain how and why we feel at all!

    Further in the paper, Dennett discusses how a study by Ramachandran & Gregory “anticipated how individuals would react to the motion capture stimulus based on their knowledge of the brains visual system” and were able to give a causal explanation for the correlation between objective and subjective conditions in the brain (p.9). But this still doesn’t remotely explain the feeling issue…
    Let’s say they replicated this study using zombies as participants (these zombies can do everything we do, however they don’t feel)– these zombies would also say they saw the same thing (and Ramachandran & Gregory’s prediction would still be correct); however these Zombies don’t feel anything … As such, heterophenomenology doesn’t touch upon the explaining of how and why we feel – it “explains” something else…
    Also, I can’t help but thing, if Zombies do exist, why don’t they feel? And why and how are we not Zombies?

    Lastly, is it just me who gets the sense that Dennett is more like a behaviorist? In the sense that he thinks the brain is solely some sort of mechanism generating doing, without needing to further explain anything else?

    ReplyDelete
    Replies
    1. This comment has been removed by the author.

      Delete
  15. Harnad's paper did a nice job of outlining some concerns I had with Dennett's piece, however, I was left thinking - ok, but now what? If heterophenomenology is just reliant on correlation between the functional/behavioural/cognitive aspects and feeling, it does not give us causation and we aren't able to answer the how/why of feeling. Harnad does great job of proving this fact but we are not left with any particular direction of how to address this hard problem? It seems as though we will always remain from the outside looking in - whether it is 3rd person or even 1st person science - and that we are only looking into a black-box - we'll never know for sure, only speculation!

    I agree with what Harnad said here:
    If Turing had instead said:
    *TURING: "I don't think we can make any functional inroads on feelings, so let's forget about them and focus on performance capacity, trusting that, if they do have any functional role, feelings will kick in at some point in the performance capacity hierarchy, and if they don't they won't, but we can't hope to be any the wiser either way."


    We can continue to find neural correlates of feeling but that will not explain it. Is it possible that, like UG, there is some sort of innate code we are born with, some special ingredient, that makes us feel the way we do when we interact with the environment around us? As in, it is a combination of this hard-wired code and our somatosensory experiences that "tweak the parameter settings" of our feelings, causing us to have particular feelings when presented with different stimuli in the future? You can think of the example of a hashing algorithm creating a unique user ID. The inputted information contains the person's name, birth date, age, and other relevant info and is then combined with a special ingredient - only beknown to the system that created it - to create a unique string of bits that identify that particular person. We all see how people react differently to different scenarios, they have different feelings as well as varying degrees of a certain feeling someone else may be feeling. So what I mean to say is that we could all have this unique innate code that we are born with that will set out the foundation for our feelings.

    Of course and as usual, this doesn't answer the how/why question. I just thought it would be interesting to consider. It could explain why some people experience emotional numbness or seem to feel less than others. It's not that they are heartless zombies, but that their level of "feeling" activity is different because their initial (and unique) foundation was tweaked by their somatosensory/physical/mental experiences in life.

    ReplyDelete
  16. "You both have what you consider to be positive empirical programs for solving the hard problem -- you Dan, by showing that it is a nonproblem, and David, by showing that there are ways to study the laws of consciousness despite the hard constraints. Whereas I just call a methodological spade the spade it is: There is no accounting for feelings functionally. Period. Back to Turing modeling and heterophenomenological weather-forecasting..."


    Stevan is clearly unsatisfied by the way Dan and David respond to his Stevan calls the Hard Problem. Dan spends so many useless words trying to point how this can’t be a problem for science. His arguments kind of seem like a super contemporary version of Behaviorism. Ignoring “the box” is appropriate because it simply can’t be studied? If it’s a hard problem, then it’s probably not a problem? Yeah, I can imagine Stevan’s distaste for Dan’s position.

    David, on the other hand, seems to miss the gravity of the Hard Problem, by suggesting it may not be that hard. Despite what may be called “hard restraints” we can still forge methods to study “consciousness” thereby making the Hard Problem… less hard. Stevan says we can’t make the Problem less hard just by using a bunch of fancy words to disguise what we’re trying to get at: how/why feeling. It’s not about transcendental-belief-faith-experience-consciousness-intentionality-units. Those words actually makes the Problem harder to solve.

    I think the best way to sum up this paper here is by capitalizing the first letters in the phrase “Hard Problem.” Stevan has to repeat himself over and over again merely defend the nature of the problem at hand. Its (controversially) impossible to solve, and we must treat this as a problem to be confronted and accounted for. Furthermore, anything that purports to be an answer to the Hard Problem is really just a re-formulation of the question, not a solution to how/why feeling.

    ReplyDelete
  17. “Dear Dan, you keep giving examples of successful prediction of functions from functions, and then an overall causal/functional explanation of the correlation. But when feeling rather than function is what is being predicted, all progress stops with the prediction.”

    Indeed it seems as though Dennett is only interested in finding the correlatons between function and feeling in order to generate predictions. For argument’s sake let’s assume that this is true, since he hasn’t provided any direction toward finding the answer to the how/why of feeling. So if heterophenomenology and the how/why question aren’t conflated, do the two parts (heterophenomenology and the study of how/why we feel) not each have value as areas of study? I can see tremendous advantages that heterophenomenology would bring to fields like social psychology, even if it doesn’t help us move forward after that prediction, as Stevan mentions above.

    That being said, I do agree that the end Dennett is trying to reach by the means of heterophenomenology is problematic. His argument reminds me of behaviourism in the sense of simply describing what is happening, not the causal mechanisms, and seems “just-so.” Worst of all, he doesn’t acknoweldge the separation between the easy and the hard problems:

    “Turing showed us how we could trade in the first-person perspective of Descartes and Kant for the third-person perspective of the natural sciences and answer all the questions–without philosophically significant residue.”

    Really, though, Turing’s question is just the prompt to build a T3 robot is it not? Even if we were able to replace the building of the robot with the 3rd person study of other people via heterophenomenology (which clearly we could not since the results would seem to be painfully homuncular as they were for introspection and the easy problem), we would still only have the answer to the easy problem and wouldn’t have even touched the hard one.

    ReplyDelete
  18. Reading this critique of Dennett helped my realize something that had been bothering me all throughout Dennett's original paper. As I mentioned in a previous comment, obviously heterophenomenology is fatally flawed with respect to the hard problem, as all it does is 'weather forecasting', to use Stevan's term.

    But what's really troubling about the whole notion is that Dennett presents it as this sound, predictive, rigorous science, when in reality it doesn't even hold a candle to something like meteorology, as even weather forecasters have an established causal mechanism behind their predictions. It's sort of baffling to me that someone like Dennett could make this kind of proposal as a be-all-or-end-all of answering cognitive science's greatest questions, while completely refusing to acknowledge that, as the description stands, heterophenomenology would is hardly a science at all, just a series of correlations and some educated guesses. Maybe that's a bit harsh - clearly Dennett is an intelligent and successful figure in the field and knows considerably more about the topic than I do - but I have to admit that I find the entire thing just silly, even after accounting for its insufficiencies re: tackling feeling.

    ReplyDelete
  19. “Ex hypothesi, you don't even suspect you have them--if you did, you could verbally express those suspicions. So heterophenomenology's list of primary data doesn't leave out any conscious experiences you know of, or even have any first-person inklings about. On the other hand, unless you claim not just reliability but outright infallibility, you should admit that some--just some--of your beliefs (or verbal judgments) about your conscious experiences might be wrong. In all such cases, however rare they are, what has to be explained by theory is not the conscious experience, but your belief in it (or your sincere verbal judgment, etc). So heterophenomenology doesn't include any spurious "primary data" either, but plays it safe in a way you should approve.
    You've lost me. I don't for a minute doubt that eventually we will be able to do 100% mind-reading via functional correlates of feeling. So surely that's not at issue either. What will be left unexplained by this perfect predictability of feelings from their functional correlates is how/why there are feelings at all. The explanation, not the prediction! Back to square one. Let us not waste our time on veridicality, either S's or E's...”

    I feel like this is quite a good example of the vibe of this paper. Dennett is making his circular, repetitive arguments, while Harnad points out the confusion, beating-around-the-bush, and overall unanswered questions and unaddressed issues. The entire issue of Dennett’s paper is about being able to account for first-person (conscious) experiences, within the mind. However as Harnad puts it, Dennett is not answering the question. He is merely saying that we can (or soon might be able to) predict what someone should be feeling based on their observable features. He is not answering the fundamental question as to why and how there are feelings at all. And should one be able to predict the feeling based on observable features, there would still be no guarantee that those feelings are actually being realized within the person/robot/thing. They could just say “yes, you’re correct”, and there still would be no way of knowing for sure as they might just be giving a pre-programmed semi-false answer. So in his entire paper, while Dennett postulates ideas regarding evidence of feelings, he just formulates this idea in many different ways and doesn’t answer the questions of how feelings themselves arise.

    ReplyDelete
  20. Can you explain what you mean when you use the term "performance hierarchy"? I'm confused about how feeling would somehow come out of a hierarchy if it's supposed to be all or nothing but maybe I'm seeing this in the wrong way.

    ReplyDelete
  21. “An emotion is just a synonym for a certain kind of feeling”

    I found it interesting to look up words we’ve being hearing a lot in a simple thesaurus.

    Synonyms: Emotion: Feeling, sentiment; Feeling: sensation, consciousness; Consciousness: Aware, sensible

    Emotion and Feeling seem to be direct synonyms of each other. However, consciousness is not in line in a thesauras with feeling and emotion. Sensible means “appreciable, displaying prudence, chosen in accordance with wisdom or prudence, like to be of benefit”. If consciousness implies feeling, why is this no where in the definition or as a synonym?


    Have cognitive scientists ever used hypnosis to study the hard problem? Let’s think about this for a second. I’m thinking of the times I’ve seen 10 people up on a stage and a hypnotist hypnotizes them and can make them do whatever he wants, say whatever he wants, literally under a spell. When the hypnotist un-hypnotizes them, for lack of a better word, none of the 10 people remember ANYTHING that happened. So I pose the question, did they feel anything? Or were they essentially zombies, or better yet, robots being programmed and forced by the hypnotist to do what he wants them to do? This entire 20 minutes is an experience of all 10 people on the stage. But maybe not felt experience. So when Stephen says in the paper: “You are leaving out feeling… the relevant bit is felt experience – not just “had” experience, or “real-time-past” experience, or “functioned through” experience. Would this be an example of an experience that is not felt? Is there something that Cognitive Science can learn from hypnosis?

    Here is an example of hypnotist type situation that I am referring to: https://www.youtube.com/watch?v=shOouPsjaxw

    They don’t seem to have any feeling at all, they are just doing!

    ReplyDelete
  22. When I first started reading this article, I began thinking back to the Turing test and I wondered if feeling is actually required for cognition. How do we know that we can't create a robot that doesn't feel and can still pass the Turing test? Maybe we just haven't gotten there yet. How do we know feeling is the missing puzzle piece? Intuitively it feels like it is, but what if it's not necessary? But then I got to this passage,

    "....my "Zombic hunch" is not that there could be a Turing-equivalent frog that did not feel! My Zombic "hunch" is that I know a how/why explanation when I see/hear one, and there's none in sight for how and why the frog either isn't or cannot be a Zombie. It may very well be the case that it cannot be. But I want to know how/why (and not just "that," or "just-so")!"

    To have a robot that feels as we do, we have to have a robot that can feel and react to sensations such as pain. And we need a how/why explanation for why this is so, not just a just-so explanation (which is no explanation at all!), i.e. we need a solution to the hard problem (or we need to at least attempt to address it when we talk about cognition).

    Harnad insists that heterophenomenology avoids the questions of the how/why of feeling. Dennett does ask why do we have certain beliefs about how we feel but this is just a detour from the direct question we should be asking which is why do we feel. Even if we are able to answer Dennett's question about beliefs, this doesn't get us any closer to solving the hard problem. If anything it is a collosal waste of time.

    ReplyDelete
  23. This paper is quite refreshing as it juxtaposes "The Fantasy of First-Person Science"'s at times jargon-y jumble in which Dennett says heterophenomenology is used to measure possibly predict one's feelings based on behaviour and the straightforward arguments in which Harnad points out Dennett does not address the core issue of the how and why of consciousness.

    Articles 10a, 10 b, as well as just general class discussions etc, have pretty much convinced me that yes, the hard problem is relevant, and yes, it probably is impossible to solve, and even if we were able to solve it for ourselves, the other minds problem stands in the way of further progress. So I guess, leaving this class the question I have is where do we go from here ? Perhaps the hard question is the ontological barrier in which science cannot penetrate, and even with the possibility of Team C, is the hard question the precise limit of scientific explanation?

    ReplyDelete
  24. I quite enjoyed Professor Harnad's rebuttal of Dennett and his rejection of feelings and the hard problem in this article. I unfortunately was away in New York for a few days and had to miss class, which I'm pretty bummed about because the topics of these papers is one I would have loved to debate with my fellow classmates about! So furthermore, forgive me if I do say something that is misunderstood or that has already been explained or decided upon in class that I missed out on.

    ""I believe that X"

    is no different from a sentence on paper or on screen, or implemented dynamically as a computer state -- unless there is something it feels-like to have that belief. If you don't feel, "you" [I hesitate to use the animate 2nd person to refer to a Zombie, I should really say "it"] don't have beliefs, "you" merely have (meaningfully interpretable) internal sentences (or internal states that are interpretable as sentences are)."

    I think this is a great summary of the obvious absurdity of saying that we don't have feelings whatsoever and that we only believe we do. It makes it seem as though we all have these empty delusions in our heads about most of our life experiences. Saying to someone in excruciating pain who just had a limb cut off that their qualitative experience is void and that they just believe that they have those feelings sounds pretty ludicrous. To me, I suppose Dennett is simply taking these stances because, as Ailish mentioned in her comment on the previous article, the notion of the hard problem, as specially when you first encounter it, does seem kind of silly almost. However, the more that you flesh out the argument and really understand the hard problem, the "harder" it really reveals itself to be. A quick, get-out-of-jail-free card is to simply deny the existence of all those pesky feelings that are causing this problem in the first place. Yet, to do so seems to deny the very corners stone of the human experience and seems rather counter intuitive.

    ReplyDelete
  25. A few random questions:

    Would proving the impossibility of a zombie be equivalent to showing that thinking must include feeling? Yes, but we can’t prove such a thing either, because its not falsifiable so we’re in the same boat in the end.

    Would we get any sort of interesting data out of a comparison between an unconscious experience to a conscious one. Ie looking at what kinds of cognition have to be done consciously or unconsciously? Perhaps, but I don’t know how to go about making the comparison pairs.

    ReplyDelete
  26. Having read Dennett before reading this Harnad paper, I can notice my fatal flaw: I was not questioning the question in Dennett’s paper, I was questioning the answer. You cannot question the answer however until you have the right question, and do know if you have the right question, you must question the question. Harnad is right, neither the Descartes or Kant are question so Turing can’t really be the answer. Feeling did not enter the questions, but feeling is the problem we desperately want to solve.

    “I can tell you that until you explain why and how a pinch hurts, the game's not won. (That is does hurt, and that that hurting correlates perfectly with some functional story, is not the how/why explanation we were seeking...)” Feeling vs. function. We know we feel, the function pinch + arm = hurt is what we know but as Harnad plainly states, and as many others over look, is that this function narrative is not enough. Why does pinch + arm = hurt? How does pinch + arm = hurt? When Chalmers is talking about consciousness, and Turing not being able to solve this part, I think he does mean this feeling of thought, however consciousness also means performance capacity, which Turing can solve, and I think that is where Dennett is misunderstanding.

    ReplyDelete
  27. “So the hard problem (of how/why we are not zombies) is the "feeling/function" problem, or, even more directly, the "feeling/doing" problem: How/why does it feel like something to have (or to be!) certain functional powers? Although that sounds superficially like asking "How/why does gravity pull?" it isn't, because pulling is gravity, but feeling is not doing.”
    Does it follow, that perhaps feeling and doing might be so closely inter-related, in the case of humans, that parsing apart a set mechanism might be impossible? I think about the number of circumstances or conditions, some of which we are aware of and others we are likely blind to, that effect our feeling of situations or circumstances that differ by only the slightest of degrees. And that, in order to create a casual mechanism for feeling, we would have to factor in likely an infinite number of permutations to ever accurately resolve a specific feeling.

    “you will be unable to give even a hint of a hint as to how and why a mechanism with precisely the structural and functional properties you have correctly and completely divined in your successful causal mechanism, should be feeling at all (rather than simple doing everything it is so very capable of doing -- behaviorally, physiologically, computationally), but without feeling anything.”
    Is this the argument we take when discussing T4 equivalence?

    ReplyDelete
  28. I honestly have no remarks except that I agree 100% and I wish I had read this before writing my response to Dennett’s paper (10a) since it says what I wanted to say but better.

    I have been curious why you say “feeling” and not “experience”, and this article references it (“experience is equivocal”). But I think “feeling” is equivocal too. The proof of this is literally the first class this term, when someone thought that by feeling, you meant “emotions”.

    ReplyDelete
  29. Ok, I get it. I can’t tell you how and why we feel rather than not feel. But neither can you, and a proof to that effect, should it exist, would not satisfy your thirst for feeling, either.
    I know what feeling is, and I know why you’re irked. But why should we dwell on feeling when we have no way of doing any reasoning or science about it?
    I think feeling, until further notice, should be considered as “knowledge/awareness of some (change of?) internal state.” Whether the “knowledge/awareness” is felt or not is irrelevant for the same reason that whether or not T3 thinks is irrelevant, so long as it can do.

    ReplyDelete
  30. "Frankly, this pseudo-puzzle looks like it's just a consequence of the highly counterfactual premise: To suppose that something that is molecule-for-molecule identical to me could fail to have feelings sounds about as sensible as to suppose that something that was molecule-for-molecule identical to the moon could fail to have gravity."

    I'm willing to concede that heterophenomenology (HP) is basically fancy "weather-forecasting" and doesn't actually access the important causal mechanisms of how, nor the adaptive value of why we feel, but I'd like to play devil's advocate using the above analogy to get a better understanding of why we are rejecting HP.

    Let's compare heterophenomenologists and physicists. A HPist is someone who takes as their data set an individual's described feelings, and their neural and behavioural correlates. In an ideal world a HPist would be able to use this data to predict with confidence the feelings which result from given neural states. But are they able to explain how these neural states conjure up feelings?

    Now let's imagine a physicist who is attempting to explain the interaction between the earth and the moon. They take as data the movements and properties of both the earth and the moon. In an ideal world a physicist would be able to use this data to predict with confidence the relative motion and positions of the earth and moon on the basis of their physical properties. Luckily for us (and astronauts) physicists are able to do this and have devised an equation describing gravity which accounts for their predictions. But have they explained how these physical properties actually conjure up the gravitational force which determines their motion and position?

    In both of these cases it would seem that a comprehensive 'equation' could be used to accurately predict an outcome (feeling/motion & position) on the basis of some input (neural states/planetary properties). Yet in the case of feeling, we say that we haven't explained how neural states CAUSE feelings. But since we have a 'law of gravity' we say that physicists have explained how properties of planetary bodies (mass, distance) CAUSE gravitational forces between them.

    How are these different if both represent 'equations' which successfully predict phenomenon which were previously unpredictable.

    Intuitively, I can already feel a few things might be wrong with this analogy...
    The laws of gravitation are well defined, whereas our hypothetical 'feeling equation' seems like the sort of thing that would be unique to an individual, and not generalizable to the feeling capacities of every individual.
    Furthermore, I'm fairly certain that the laws of gravitation have some sort of causal basis in the very nature of subatomic particles, which would suggest that the laws of gravitation are more concrete, arising naturally from the most basic forces in the universe, whereas it feels that 'feeling' has not really been explained but relegated to some statistically functional equation.

    I feel that my counterarguments are somewhat ill-formed but I am grasping at something. I guess I'm reeling at the bigger question of what constitutes 'causal explanation'. With gravitation, there's always a 'why'. 'Why is there gravity?' 'Because of the nature of subatomic particles!' 'Why are subatomic particles like that?' 'String theory (or something)!' 'Why...' and so on.
    And so I imagine it is with feeling. Why is predictive power not enough?

    ReplyDelete
  31. 1. “ ‘I guess I should take some of the blame for the misapprehension, in some quarters, that heterophenomenology restricts itself to verbal reports. […] But all other such data, all behavioral reactions, visceral reactions, hormonal reactions, and other changes in physically detectable state are included within heterophenomenology. I thought that went without saying, but apparently these additional data are often conveniently overlooked by critics of heterophenomenology.’
    Dear Dan. All bodily and molecular motions are there and available. And obviously 100% coupled with the feelings. The question (I hate to be repetitious) is how and why?”

    By “coupled” what is meant is that bodily/molecular motions and feeling are correlated: each feeling has its characteristic pattern of bodily/molecular motions? So any data added to the verbal reports concerning what goes on at the physical level is only a description of the “what” (what is going on) and not of the why any of it occurs.

    2. “This is equivocal. A sentence in a book can be true, but it cannot be "sincere" [on the part of the book] and it cannot be "believed" [by the book]. Same is true for dynamic, on-line books. Same is true for Zombies: No feelings of sincerity, or credence, or anything...”

    “I would add only two other things. First, that "belief" is a weasel-word. This is controversial and (based on the
    resistance I have encountered over the years) probably original with me: "I believe that X"
    is no different from a sentence on paper or on screen, or implemented dynamically as a computer state -- unless there is something it feels-like to have that belief”

    Definition question: for a belief to be different from a sentence in a book anything that is “belief” needs to be felt, the feeling that some statement/theory/idea is true?

    ReplyDelete
  32. I had purposely written my response to 10A before looking at 10B to see the change in opinion – the commentary did clarify many points to which I was initially confused. I have created this response as a response to your response on the Dennett's article.


    “If they are unfelt, they are not feelings and hence not relevant to any of this.”
    What about things that are felt, but we are not attending to them such that we are unaware that they are felt? Experiments will show that they were processed and thus felt at least somewhat. For example, someone could be completing a task on a computer, while being shown subliminal messages. In the next task, these messages will have an effect on how you respond. They were still felt, even if not within awareness.

    Pain is too simple of a feeling to use. We know there is functionality to pain: to prevent damage physical being, this being true for creatures from a nematode to a great whale. Why do we feel like we understand English for monolingual English speakers and not Chinese? That type of feeling is more difficult to solve in terms of functionality. This relates to comments above considering the level of consciousness between the complexity of different animals. While I believe all animals can feel pain, we can develop a metric for which animals are capable of more complex thinking (or feeling) and which are not. For example, if a dolphin sees its reflection, the dolphin understands that it is seeing itself, not another animal. Whereas simpler fish, like a beta fish, will try to battle its own reflection since it cannot comprehend the difference. So, while I agree that whether or not you can feel pain is an all-or-none phenomenon, I think levels of consciousness lies on a continuum.

    I am at risk of completely missing the point in this statement but I’ll give it a go regardless. I think the reason there is disagreement between Dennett and Harnad is that Dennett is not trying to answer the Hard Problem, “why” there are feelings. That is why he doesn't answer it in his initial article, and why Harnad is unimpressed with Dennett's article. Dennett is admitting that the Hard Problem isn’t something he is can answer, and is instead focusing on the how question. Sure, the Hard Problem may be the more difficult question to answer, but if we’ve all agreed that it is “insoluable”, then let’s move on and start tackling the “how”.

    ReplyDelete
  33. “"Thoughts" is 100% equivocal. If it just means "internal goings-on that generate certain outputs in response to certain inputs," then no problem (and no problem solved!). But if "thoughts" means "felt thoughts," then you might as well call them "feelings" (what it feels-like to think and reason is just one instance of the multiqualitative world of feelings; there's also what it feels-like to see, touch, want, will, etc.):”

    When we talk about felt thoughts and feelings, are they not also internal going-ons that have the capacity to generate certain outputs in response to certain inputs? Certain times when we feel a certain way, does it not influence the way we think/cognize?



    I believe that most of this article highlights the question of how and why of feeling. Almost every point Dan brings up is countered with questions regarding feeling and “how and why”.



    “The "certain experiences" can be any feeling!”

    What is the relation between experiences and feelings? Are they same or is one part of the other? Or does one cause the other? To me, it seems that they could go either way but it would be great to see what everyone else thinks.

    ReplyDelete
  34. In this article, Harnad addressed some of my frustrations with Dennett's paper. The contention between the two authors is that Dennett doesn't believe that the how/why questions of feelings are important. He argues that if a robot passes the TT through heterophenomenological methods, this is the proof of feeling. However, even if heterophenomenology can predict the existence/content of feelings, there is no causal (how/why) explanation of feeling, so Dennett misses the mark of the hard problem.

    ReplyDelete