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...
Click here to view --> HARNAD VIDEO
Note: Use Safari or Firefox to view;
does not work on Chrome
But... "Claustrum Nostrum"
I've generally tried to play devil's advocate throughout the course, but I think everything you (Stevan) says in this paper is right on the money.
ReplyDeleteDennetts' talk about failures of overlap, beliefs, veridicality, etc. is all besides the point. The fact of the matter is that (some?) organisms have feelings (i.e., they are not zombies), and we want to explain how/why that is so. There's nothing in his so-called "heterophenomenology" method that helps one bit to answer this hard question.
Likewise, I think that Chalmers' talk about conceivability and whether it entails possibility is just unnecessary metaphysical coating. We have feelings, explain why/how that is possible. It's as simple as that (the question, I mean; the solution seems impossible). I also agree that the notion that zombies could have beliefs without being conscious quite absurd. It is indeed like saying that I have unfelt feelings.
As a point of curiosity, I was wandering whether you (Stevan) identify with any of the classical positions on the mind-body problem: substance or property dualism, materialism/physicalism, idealism, interactionism, identity theory, etc.
Gabriel, no, I don’t identify with any of the classical metaphysical positions on the hard problem. We need to provide a causal explanation, not just rename the unsolved problem.
DeleteThis reading clarified a lot of the concepts Dennett used in the 10a paper. Indeed, it made me see that multiple equivocal labels were used for concepts that were therefore too ambiguous to discuss the hard problem as we have defined it, ie how and why organisms feel what they can feel.
ReplyDeleteFor instance, Dennett talks about the notion of belief, but it is now clear that we don’t want to reverse-engineer beliefs or even beliefs about feelings, but just the notion of feeling itself. Similarly, he mentions ‘experience’, which can refer to the experience you had (in the behaviorist way), rather than the felt experience we are interested in.
In my last skywriting, I also used the word ‘consciousness’ as employed in Dennett’s paper, but after this reading I realize that this might not the appropriate word to use, as it implies that the hard problem is concerned with the notion of aboutness, rather than just a causal explanation of feelings and felt states.
Mathilda, "aboutness" (and "intentionality," the version Searle used) are weasel-words. So is "experience" (which can be felt or unfelt).
Delete"Belief," too, is a weasel-word. Believing something -- like understanding something or meaning something -- is a felt state. So the underlying state that needs to be explained in all of these cases is: feeling.
So it's always the same "hard problem" that is at issue with each of these weasel-words for it. The weasel-words just give the impression that there are many different "problems" -- many different things to be explained. But it is always all just one and the same problem (or, if unfelt, not a problem at all).
Dr. Harnad perfectly resumes Dennett’s shortcomings in his sentence: “It is not the aboutness we are after, it is the sentience.” This is precisely the problem with transcripts coming with recorded verbal utterances and raw data. Cognitive science is not in the business of interpreting. If it were we could say nonhumans have beliefs. In the previous skywriting, I talked about qualia. Dr. Harnad elegantly describes why even qualia are somewhat a useless point of discussion since they are just feelings that can be misremembered or misdescribed, that ultimately do not come to help us in understanding what said feelings feel like! And we have time and time again seen that no functional explanation serves to explain the feeling. Can there really ever be much progress in this respect?
DeleteThis response to Dennett’s paper was quite enlightening, as although I find Dennett’s points interesting and rather valid, it is true that they aren’t necessarily relevant to this exact subject. This was a good critique, and specific and witty in its comments.
DeleteI found that it opened my eyes to certain thoughts that I have on these matters that aren’t relevant or are communicated with the wrong words - it gave me somewhat of a self awareness and means to critique my own work, even though Dennett clearly more advanced than me in these matters.
In this realm, it is easy to get carried away into philosophical matters and ‘weasel words’, which are obviously appropriate in certain subjects, however I felt reminded in this reading what one should be focused on (“the how and the why”)…
This critical review, as always, was very insightful in assessing the previous assigned reading, and it really clarified the hard problem and what it is often confused with. The most central takeaway that I extracted from this reading is that “the name of the game is not just inferring and describing feelings, but explaining them”. Dennett’s heterophenomenology attempts to infer people’s feelings by interpreting raw observable data (verbal reports, beliefs, convictions, etc). In his paper, not only does he confuse many concepts (e.g. beliefs and feelings, experience and felt experience, cf Mathilda’s post for more detailed examples), he also circumverts the very problem he is trying to address. Indeed, the hard problem of consciousness attempts to uncover not what people feel, but how and why they feel anything at all. It is irrelevant and beside the question to infer and describe feelings using heterophenomenology, because this doesn’t provide a causal explanation for them.
ReplyDeleteAmélie, "heterophenomenology" is not addressing the hard problem but the easy problem, plus the "other-minds" problem (OMP) of "mind-reading": how can we find out whether -- and what – other people are thinking and feeling?
DeleteWith people, you can ask, and they can tell you, thanks to language. Otherwise, there’s Turing-testing, with T2, T3, or T4 (and with the help of our mirror capacities).
Mind-reading becomes harder with other species, and harder the farther and more unlike us the species is, phylogenetically and behaviorally: primates and other mammals are the closest and most familiar; our human mirror capacities work with them as well (just as they do with preverbal infants).
Other vertebrates are less familiar. Birds are closest, but we can mind-read reptiles, amphibians, and fish too, the more we know them, and about them, especially as individuals.
And lately, more and more invertebrates -- octopus, lobsters, bees, spiders -- are turning out to be mind-readable too. (We’ll discuss them next week, in 10 a,b,c.)
Melis, I think you're right (but I hope you're wrong).
DeleteI had felt (HA!) the exact same way as when Professor Harnad stated, "[a]nd who cares [about] polygraph science." I was also lost in that section of Dennett's article.
ReplyDeleteAs I stated in my comment for 10a, I understand why Dennett is trying to eliminate the 1st person perspective completely. It is too subjective to take part in science. In a perfect world where everything was already pre-determined, utilizing the 3rd person view would be the best available method for explaining natural phenomena. But this is not the pre-determined world. There are many variables at play at all times (one of which is feelings).
Discarding the living being's abilities to "feel" is just a disservice to science and the pursuit of knowledge.
Alexei, what do you mean by “In a perfect world where everything was already pre-determined, utilizing the 3rd person view would be the best available method for explaining natural phenomena”?
DeleteIf the biological trait of sentience -- the capacity to feel -- really exists (and it does) then doesn't cogsci still have to explain how and why, regardless of whether the world is “perfect,” or “predetermined”?
If there were no sentience then there would be no distinction between a “1st-person” and “3rd-person” “point of view” (except for robot-cameras) because there would be no one would be “viewing,” just insentient darwinian T3s (zombies), doing)
Though one does wonder what would be the referent of their word “feeling”: any hunches?
(“Stevan says” there could not be insentient T3 zombies, but you’d have to solve the hard problem to be able to explain why not!)
On your second question about a biological trait of sentience:
DeleteIf for example a gene for sentience exists, the question for how would thus be explained by genetics and natural biology. As for the question of why, couldn't the capacity for sentience then simply be proscribed to enhancing fitness (i.e., darwinian evolution)?
Emma, we can be sure that the capacity to feel must have evolved, must be adaptive, and is coded in our genes, which produce our brains, which produce our capacity to feel. There will no doubt be a protein synthesis story that leads from the genes to every property of our brains, including the the one(s) that produce the capacity to feel.
DeleteWhat that doesn’t tell us is how or why.
Won't you first have to overcome the OMP anyway if you are going to inquire about the adaptive advantage of sentience? I'm not well versed in the methods for this kind of inquiry but I assume we'd have to compare organisms endowed with sentience to those who aren't, and determine how sentience impacts their doing-capacities, since evolutionary advantage is presumably doing capacities that increase fitness. This seems tricky for a number of reasons, including the fact that we'd have to draw some distinction between organisms that feel and those that don't, which takes us to the OMP. Even if we can definitively say of some organisms that they are sentient, and of others that they aren't these organisms are unlikely to be very similar in any other respect. This leads to many, many confounding variables as far as testing causes for doing capacity. The organisms that have the virtue of being more similar to one another except in sentience are likely to be at the very blurry line between sentience and non-sentience, which Is probably where the OMP is most difficult to solve.
DeleteHadrien, the OMP is not the HP.
DeleteDr. Harnad, while Hadrien made a mistake in terminology, I am tempted to say that the idea that explaining why we feel would be answered by looking at the evolutionary disadvantages of not feeling. It is completely possible to do everything that we do without feeling, but perhaps feelings allow us to process things more efficiently or some other advantage that we have not yet discovered.
DeleteMy question was essentially this: how can we even distinguish between sentient and non-sentient organisms without solving the OMP? And it seems that looking at the evolutionary disadvantages of not feeling first requires distinguishing clearly between organisms that feel and those that do not feel. Even if we could say, without solving the OMP, that something like an amoeba does not feel, how are we ever going to compare amoebas to humans or other organisms we definitively put in the sentient camp? I hope I've made my point clearer; I'm not sure where I mixed up the HP and OMP.
DeleteProfessor Harnad: When I said "in a perfect world," I meant to say that in the context of Dennett's message. Dennett believes that the capacity to feel does not exist. Since feelings are subjective experiences, by eliminating them, the only thing left would be objectivity.
DeleteEverything can be pre-determined since there are no more subjective variables left to interpret.
As I understand it, the definition of solving the Other Minds Problem would mean there exists a method to indicate whether others are capable of feeling. I would agree with Elena that observing and comparing the performance of beings with similar capacities for doing, but with different capacities for feeling, could provide insight into the purpose of feeling. However, the first step to achieving this would be to make sure that we are able to accurately identify what is feeling and what is not. Solving the Other Minds Problem would achieve this and potentially provide a tool which could lead to getting closer to solving the hard problem.
DeleteSophearah, we already have a good enough method to give others the benefit of the doubt: our everyday Turing-Testing, together with our OM-reading mirror capacities. Plus ethological observation and understanding.
DeleteWell this was a cathartic read- I'd be interested to see Dr. Dennett's response!
ReplyDeleteWhat really struck me about this reply is that it manages to strike a middle ground between phenomenologists and materialists by taking a step back. This is evident in this passage:
"You both have positive programs that you both believe bear on the "hard" problem -- yours, Dan to debunk it and replace it by the real goods, Dave's to supplement it with some other stuff...."
Pointing out that feelings are underdetermined is a much more useful observation in a way; it allows us to approach them from a perspective that doesn't hinge on what one believes about the ontology of consciousness. These arguments don't do anything besides change a few peoples' intuitions on the subject. What it does allow us to do it to "foreswear the undo-able" and focus on more relevant areas of research that can actually show us the structural properties of feelings!
Jacob, K-S does not understand what "ontology" or "phenomenologism" or "materialism" means, nor what "structural properties of feelings" means.
DeleteK-S does know what "do" means, and what "feel" means, and just wants to know "how" and "why," for both capacities.
By "phenomenologists" and "materialists" I am referring to people who think the contents of our feeling is either the most important data or the least when we try to explain how and why we feel.
DeleteBy the "structural properties of feelings" I meant the thing-ness of our feeling (or lack thereof!) and how that relates to why we feel it. Perhaps it's superfluous, and falling into the trap you point out in this paper!
My apologies to Kid Sib!
"There is no accounting for feelings functionally. Period."
ReplyDeleteThis is an interesting concept because it makes us wonder why feelings even exist. Of course I don't want to get philosophical, and we need to address the "how" as well as the "why" but I wonder if Dennett and others align with the heterophenomenological approach because of a desire to explain feelings...it's odd to reckon with feelings in the way Dr. Harnad does in his paper. But nevertheless, it's true that feelings have no localization, no visible machinery, no structure or organ in the body. This is the very crux of the easy/hard problem because "doing" can be localized, for example the heart pumps blood, but "feeling" is a much more mysterious and like Dr. Harnad says, insoluble.
Teegan, well, “insoluble” is just what “Stevan Says.” (Why?) But feelings do have a location and organ: the brain. So that’s not why the hard problem is hard. (Be sure to read the other comments and replies in 10 a,b,c.)
DeleteFor the most part I agree but not entirely. I recently participated in an experiment at McGill where participants were asked to show where in the body they experience certain feelings. For many of the feelings it was really hard to pick a certain place in the body (besides just the brain). But for a few I really did feel that they took place in multiple locations. For example, I feel anxiety in my chest. Or when you experience heartbreak, it really does feel like your heart is breaking.
DeleteTeegan, the question is not where we feel the location of the cause of our feelings is but where the location of the cause of our feelings is. Just as what matters at the dentist's is not what we feel is wrong with our teeth but what is wrong with our teeth.
Deletetest to see if the comment will stay
Delete[Kayla's comment, continued]
Delete‘Where’ is known as you comment prof Harnad, in the brain. In particular, maybe the claustrum is the seat of consciousness, feeling? I am not fully convinced by the article linked because there is evidence in only one patient. Even if it does happen that this part of the brain directly impacts feeling (consciousness, sentience) I don’t think it will answer the question of how and why we feel, the hard problem."
Kayla, yes, you understand correctly -- though asking for the “purpose” of feeling capacity may be a bit too teleological a demand. The HP is hard enough if we just ask for the same thing we ask for in the case of doing-capacity: How does the brain (or the reverse-engineered T3/T4 robot’s brain) cause feeling-capacity, and what causal function does feeling-capacity provide that doing-capacity alone does not already provide?
DeleteHP is hard, for sure, but “insoluble” is just “Stevan Says.” If there were a a tight heterophenomenological correlation – which there isn’t! -- between human claustrum activity and all and only felt states, and inactivation or absence of a claustrum made a human unable to do anything at all, and if T3 could not be successfully passed without a T4 claustrum (as in the 10a CHALLENGE), perhaps all of this would suggest some sort of a causal clue, but I can’t imagine what it would be).
(That we cannot know with cartesian certainty whether a T4 without a claustrum is a zombie just reminds us of Turing’s point that we cannot do any better than the EP and T-Testing. Ditto if, like an appendix, a human claustrum could be removed with no effect at all on doing-capacity. So OMP is not what makes HP hard. Neither science nor reverse-engineering provides certainty.)
Harnad explicitly yet politely destroys Dennett's argument by addressing its true function as some sort of method to verify reported subjective accounts i.e. some sort of lie detection device to prove whether the subject is really feeling what they are reporting, which we were never wondering in the first place. As Harnad says, this does not provide us with any information on why/how we have feelings and Dennett rather focuses on the prediction/correlation of feelings. And even if we can decipher the causal mechanism for feelings, we still do not know how or why.
ReplyDeleteLaura, if we deny that there is a hard problem of explaining feeling (because feeling is really just "belief about feeling"), the only problem left on the stage is the other-minds problem, and the only thing in those other minds is not feeling but "beliefs about feeling," which we can always suss out with heterophenomenology, which is really just the easy problem!
DeleteTrouble is, feeling really is felt, not just "believed to be felt," Or, otherwise put: it feels like something to believe something -- just as it feels like something to understand something, to mean something, to think something and to be kicked.
And, as Descartes reminds us, you may be wrong about what you feel is wrong with your tooth, but you cannot be wrong that you are feeling what you are feeling.
So there is a world of difference between "feeling is (just) believing" and "believing is felt"...
Fun story, I came to the cognitive science program with the life long goal of “solving consciousness”. I didn't know about the hard problem, or even really what cognitive science was, just that consciousness was this big mysterious thing that somehow needed to be “solved” and I wanted to do it. I gained a crude understanding of the Hard Problem in my PSYC 100 class, and more in depth in following classes, and quickly learned that many experts deemed it unsolvable. Over the past 4 years I have come to agree with Stevan, that the problem of how and why we feel what we feel is likely not going to be solved, or at least not in my lifetime. In this paper he points out criticisms of experts not focussing on the problem itself, rather just spending years rephrasing (using Zombies, Bats, Mary etc.) and rightfully points out that Dennets methodologies will only ever give us correlates, but cannot explain how and why we feel the things we do.
ReplyDeleteSo while I no longer plan on being the one to solve the Hard Problem, may still someone can, but I have yet to be convinced that there is methodology that would allow us to do so.
From what I understood after reading the Mind-body problem, Harnad’s point is that Denett does not answer the question “how/why we feel”. So the whole point of using Heterophenomenology is pointless because never really answers the question that we are seeking to answer. For example, when the methods used in Heterophenomenology are discussed, such as lie detector, what is criticized is the fact that it doesn’t address the causal aspect of feelings. It would just prove that there is something, without answering the question about its cause and function, which relates to the hard problem. And, according to Harnad, the hard problem about the cause and functions of feelings has not been solved and cannot be solved.
ReplyDeleteCharlene, “heterophenomenology” is just about OMP weather-forecasting; the HP is about causal explanation. The HP is hard, but that it is unsolvable is just “Stevan Says.”
Deletethanks fir your reply, it is more clear!
Delete“This is equivocation on "Descartes/Kant" again: We are not interested in whether your toothache was real or psychosomatic, or even if your tooth was hallucinated, nor in the conditions under which these various things may or may not happen or be predicted. We are interested in how/why they feel like anything at all.”
ReplyDeleteThe distinction between different “types” of feelings that Dennett argues need to be “neutralized” are irrelevant to our journey to discover how and why feelings exist as Harnad explains above. Harnad’s refute to this part of Dennett’s argument reminds me about why individual differences in behavior are not necessary to reverse-engineer doing capacities, as we are just looking to reverse-engineer a system that can produce certain behavioral output. Thinking about feeling in this way may provide a means to think about reverse-engineering the doing capacities that give rise to feeling correlates.
Darcy, it’s reassuring to see a sky in which the HP and OMP are not conflated!
DeleteThis paper was satisfying to read as it poked the holes in Dennett's argument that I was mentally noting when I read 10a beforehand. I think it's important for us to learn about unsuccessful cog sci methodologies, so we can learn from them and not fall into the same rabbit holes.
ReplyDeleteAs you drive home in this critique, Professor Harnad, Dennett's paper fails to address HOW his heterophenomenological method will provide any causational explanation for neither the Easy nor Hard Problems, making it nothing more than "functional-correlate science". This reminded me of Jerry Fodor's critique (Week 4), except instead of cognitive science getting sidetracked by functional localization, Dennett is detouring the field toward functional verbalization...?
Although I found the format of this paper intuitive and entertaining, I did not understand the following paragraph. Could you please clarify what you're saying here, Professor Harnad? I'm most tripped up by the last four words… thanks!
"Or, as I prefer to put it [...] 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 [...], but without feeling anything."
Polly, your first (zombie) elision was ok, but the second elision contained the answer to your question:
Delete“Or, as I prefer to put it…: 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 simply doing everything it is so very capable of doing -- behaviorally, physiologically, computationally), but without feeling anything.”
This just points out that “heterophenomenology” (which is really just part of the solution to the easy problem – of causally explaining doing) does not even touch the hard problem of causally explaining feeling.
Remember that Dan Dennett does not think there is any HP at all, just the OMP and the EP. In contrast, Turing saw clearly that only the EP could be solved with his method, and that it left the HP untouched.
To treat the HP as if it were just the OMP is not to solve it.
Dennett writes about how our subjective feelings can be “provably false” when he defines the term “false positive” in his paper, in which he argues that “some beliefs that subjects have about their own conscious states are provably false, and hence what needs explanation in these cases is the etiology of the false belief.” As prof. Harnad argues, we’re not interested in whether the feeling was “real of psychosomatic”, we are interested in how and why they feel. In other words, it’s irrelevant to think about different ways of feeling and whether they’re right or wrong because the question we want to answer is how and why feelings exist, and perhaps this can be done by focusing instead on performance capacity, where if feelings “do have any functional role, [then they will] 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.”
ReplyDeleteThis reading is really satisfying, as it dismantles Dennett’s arguments piece by piece. It points out the irrelevant parts of Dennett’s argument, which did feel like an overcomplication of the problem. Some parts seemed like just setting ground for his argument but deviance from the hard problem at hand: why and how do we feel? Indeed, Dennett discusses beliefs, objective measures of experience and conditions for veridicality. This reading brings every matter discussed in this course with feeling. Indeed, the matter at hand is truthfully felt thoughts, felt experiences and felt learnings. I thought this was an important and interesting nuance to bring, as I had never thought of it that way, but it does encompass the problem perfectly.
ReplyDeleteOne thing that you have to give to Dennett’s theory of heterophenomenology is that it does tackle the Other Mind problem. Indeed, heterophenomenology does give us an access into what the other mind is feeling. However, his theory was intended to bring a solution to the hard problem. For this reason, as skeptical as I am with his theory, rejecting it completely would still leave us with the burden of the other minds problem.
The professor’s critique of Dennet’s paper emphasizes that Dennet’s heterophenomenological method does not address the key questions of causality, how and why, and that instead, it only provides us with correlates of feelings. While this does not solve the hard problem, it does not negate its potential utility in serving as a method for evaluating the reliability of subjective reports.
ReplyDeleteEmma, agreed, see above.
DeleteIn this piece, Prof Harnad identified the difference between feelings and functions that correlate with feelings. Even though heterophenomenology could explain when certain feelings might occur under which circumstances, it cannot explain why and how these feelings arise. One interesting part of the reading was whether unaware detections of stimuli could result in unconscious feelings. The answer turned out to be simple because there isn't such a thing as "unfelt feelings." "What I didn't feel then (and didn't merely forget I felt), I didn't feel, and nothing else felt it either! So neither change-blindness nor backward masking shows that feelings are fallible qua feelings."
ReplyDeleteWhile I do think heterophenomenology can explain when certain feelings might occur under which circumstance, it's really not even a holistic account of it because it still relies on too much functionality (which we know to be insufficient) to be able to provide an all-encompassing account.
DeleteTo address the second part of your commentary, feelings can't be feelings if they aren't made conscious. You can't say and mean "the dog is walking" without feeling or "unconsciously" and you can't be unconsciously sad. When put this way, it's really easy to see how "unconscious feelings" is a purely antithetical term thus not worth even mentioning.
Dear professor, at the very end of the article you mentioned how you are not Team A, because you believe the hard problem has not been solved; and you are not in Team B because you don’t believe in the zombie argument nor believe in the possibility of alternative "science of consciousness”. If there is a team C, what would it be?
ReplyDeleteNadila, it’s simpler to speak about problems than teams. There is an EP, a HP and an OMP.
DeleteI think the EP is solvable (and that it’s cogsci’s job to solve it).
The HP would also be cogsci’s job to solve, if it is solvable. “Stevan Says” the HP is not solvable. (Why?)
And I think (with Turing) that the closest we can come to solving the OMP is to reverse-engineer T3 or T4 capacity
Anaïs, yes, but HP is about how/why organisms feel anything at all. And the HP is not that feeling is unobservable (that's the OMP). And feeling is correlated with doing -- we just don't have a causal explanation of why. And for DD feeling is believing -- so I suppose DD means it's just part of doing, somehow (but how? and why?).
DeleteSince I also mentioned in my 10a skywriting, I had trouble understanding Dennett’s reasoning about the hard problem, so I found these critiques of Harnad very enlightening. Dennett is clearly missing the causal explanation about feeling, and what he is actually addressing is the easy problem rather than the hard problem. I particularly enjoyed this response to heterophenomenology: “you are describing an empirical psychophysical data-gathering paradigm for getting all the measurable correlates of feeling, but the only one who can actually feel the feelings themselves is the 1st person”. Exactly. The feeling is clearly there, but where is the answer to the why and how the person is feeling it? Only the person who is feeling that feeling can know the answer to this, not a 3rd person observer. I think this is similar to the issue we had with mirror neurons: we know we have them and what we can do with them, however how and why we can do the things we can do with them is still not answered. Similarly, the causal explanation is missing in Dennett’s arguments, and I think Harnad addresses each of them very clearly.
ReplyDelete
DeleteHi Alara! I like how you compared heterophenomenology to mirror neurons. Most potential “solutions” to the hard problem that we have covered in this class have the same limitation. They do not provide a causal explanation and are lacking answers to the “how” and “why” questions that are at the heart of cognitive science, as Prof. Harnad repeats many times in the article. He emphasizes the irrelevance of Dennett’s argument, “I don't care if every nook and cranny, every last JND of my feeling life is correlated with and hence detectable and predictable from something you can pick up on your polygraph screen or can infer from my behavior. That's not the question! The question is: How/why does anyone/anything feel at all?”
I agree Dennett’s problem is similar to what we saw with mirror neurons and with neuroimaging earlier in the course: in all instances, the why and how questions are neglected. In response, the main focus of Harnad’s paper is how and why. These two questions are repeatedly posed throughout this paper, which emphasizes their importance regarding the hard problem and why it’s insoluble. Like Pinker&Bloom and Pullum&Scholtz, Dennett brushes over the harder problem (in this case, THE hard problem). He disregards feeling as something that can’t be scientifically determined or even as something that exists, thus failing to recognize what the hard problem is about. Rather than addressing the other minds problem and predicting different feelings, he should be focusing on explaining feeling (why do we have it and what gives rise to it?)
DeleteAlara, I’m afraid you’re still conflating HP and OMP with: “Only the person who is feeling that feeling can know the answer to this” if by “this” you mean the HP rather than the OMP. Can you clarify what you mean?
DeleteKimberly, are you conflating them too? What “how” and “why” questions do you mean?
Josie, I think you might have succeeded in de-conflating them: but be vigilant for relapses!
I actually just saw what I did. I wrote the reply thinking that I was referring to the hard problem when I said "Only the person who is feeling that feeling can know the answer to this.", but just realized that it does not make sense since no one knows how and why we can feel, which is why the hard problem remains unsolved. I have just understood that I have referred to the other minds problem when I was actually targeting the hard problem. Thank you for the reminder!
DeleteHarnad makes it clear that Dennett was not referring to feeling with his zombie hunch when he states that Dennett made a “simple methodological point about the constraints on causal/functional explanation: it works for everything else, but it doesn’t work for feelings…” Heterophenomenology only describes feelings, but it does not provide us with a way to solve the hard problem. It is also evident that one cannot have an unconscious experience, as both consciousness and experience involve feeling something. So observing one’s doing capacity (like the zombie) does not explain their experience, nor does it mean they have an experience. Ultimately making it clear that the hard problem does not have any current methodology to get us closer to solving it.
ReplyDeleteKarina, sounds like you've twigged...
DeleteAlexander, if the HP were hard because of another problem, it would not be because of the OMP but because of the EP (and not because of that problem, but because of its eventual solution): Why?
ReplyDeleteHow and why do feeling organisms feel (HP)?... Everything and anything are permitted by Harnad to be used by Dennett to provide a methodology. The methodology of choice is to gather all psychophysical data of correlates of feelings. I don’t know what DD aims to explain with it, as Harnad vehemently inquires. But nonetheless gathering empirical evidence is a method of choice in many disciplines of human capacities. Dennett although, does not begin with a testable hypothesis because he aims to “forward engineer” consciousness. Effectively, he would be able to predict feelings from their functional correlates. Ie. Mapping the correlation between behavioral/physiological/verbal reports with felt states. I believe this will serve as what DD hopes to be a periscope to bypass the OMP- likely requiring future experiments to test whether subjects produce the same false positives/negatives (biases) that me (a feeler) produces in order to be deemed sentient/conscious/FEELING/thinking. I think the predictability test that Dennett will produce could serve as a great means of assessing sentience (like TT tests doing capacity). The OMP may have a periscope candidate… But does the hard problem? How/why are the subjects sentient (feeling*)? “Because my model predicted it’s feelings” is not a causal explanation.
ReplyDeleteSepand, sounds right, as long as you don't mix up OMP and HP. DD is denying that there is an HP.
DeleteHeterophenomenology is part of the methodology for solving the EP, but there is no periscope on the OMP other than Searle's, and that works only against computationalism.
In this response to Dennett, Harnad demonstrates Dennett's fundamental misunderstanding of the hard problem. Dennett uses the physical correlates of feeling to justify a heterophenomenological approach, but does not realize that the how and why of the basic process or structure that underlies all feelings goes unaddressed. However, it has led me to some questions. For example, are our different feelings modular? That is, does the feeling of pain interact with other feelings? Or is it just that awareness is shifted? Furthermore, at the end of the reply, Harnad states that he believes it to be impossible that zombies exist, but wouldn't this mean that any T4 machine is capable of feeling? Since it is impossible to do what a human does in the way that a human does it without being able to feel, and consciousness most likely does not arrive from the way that we receive sensory or other types of input, any machine that is capable of passing T4 must be able to feel.
ReplyDelete"...are our different feelings modular? That is, does the feeling of pain interact with other feelings? Or is it just that awareness is shifted?"
DeleteThe modularity of feelings is not so relevant to the question of how and why organisms think (HP). In this case, we are interested in the causal explanation (s) behind the capacity to feel pain (or to feel anything at all). Whether our feelings are modular or non-modular, the fact that they are felt is a Cartesian certainty, which is the main concern here. This cartesian certainty precedes your awareness (though prof. Harnad might call it “another weasel word”) in that being aware is a felt state. We now know that there are only felt and unfelt states (i.e., you cannot feel nor be aware of unfelt states), so whether or not the different feelings are the results of shifted awareness does not matter; what matters is that those feelings are felt states. All this is to say that (at least in cog-sci) our energy should be spent on reverse-engineering the capacity responsible for those different felt states.
Yucen, while individual feelings are not the point of the hard problem, I do feel that it may be helpful to know as much as possible about them and especially how they are grouped and whether they are modular. Knowing about patterns and trends in specific feelings may help us better understand the underlying causal mechanisms of feelings in general.
DeleteFrom what I understood in the lecture and readings, debunking weasel words about feeling are crucial to deepening our understanding of feeling and its relevance to our study of cognition. What is the difference between feeling and consciousness? We start by asking ourselves, what are unconscious feelings? What is unfelt consciousness? If you can’t make a concrete distinction between the two, this is an indication that they are referring to the same thing, and hence are weasel words. Conscious states are felt states, mental states. The mind and “mental” are also a weasel words in this case. All it really means is being capable of feeling. You cannot doubt that you are feeling something, but you can doubt what that feeling is telling you, or what the cause of the feeling is.
ReplyDeleteIn the reverse engineering of cognition, unfelt internal states are relevant; and this includes T4 states that are brain states and inferences made about behaviour. Phenomenal states are another weasel word, referring to what it feels like something to be in that state, like when you’re introspecting. The biggest weasel word that we need to move away from is qualia, because all it means is “feeling”. Seeing a color and feeling what it is like to see the color (“qualia”) is the exact same thing as feeling.
It is especially important to identify weasel words that refer to the same thing, because it helps us clarify what we are trying to say in the first place. Going back to consciousness and feeling: if I say someone is conscious, and there is someone else who is feeling, it seems like we’re not really saying anything. Because as we said earlier, they are the exact same thing!
After so many readings on the mind, consciousness and etc., It is so interesting think about what words best describe the HP, this "feeling" that cannot be explained in the same way as any other biological/physiological changes, but it seemed to be so important to us humans. This paper by Harnad went after the point DD made one by one, which made it really clear for me to go through and understand where it went wrong in its arguments.
ReplyDeleteI definitely agree with the professor when he admits that although Chalmers is accounting for consciousness as a part of the human experience, it is not as simple to say that because Turing’s theory accounts for consciousness, that it means that we’ve solved Kant’s question of thought. Many animals are conscious, but we do not necessarily know if they are capable of intelligent thought, much less if they are actually “feeling” the experience of thinking. Feeling is truly what a being must be determined to be capable of if they are to truly be intelligent and to learn in the way that we as humans do, and of course, to determine this, we must solve the easy and hard problems.
ReplyDeleteHi Brandon, I am interested by your last sentence in which you say that the easy and hard problems must be solved to prove intelligence in humans. We cannot know from a heterophenomenological standpoint if animals feel, because we don't have access to displays of feeling or to verbal reports (we can make some assumptions, such as cats pur when they're happy, but not much more than that). However, we know that humans are intelligent without having solved the hard problem. Dennett would definitely not consider solving the hard problem as a necessary step to proving intelligence because he denies its existence altogether, and I believe Prof Harnad would share the same opinion because the hard problem is unsolvable.
DeleteIn the paper, the quote “Eventually I do feel the change. Does that mean I did feel it before but didn't “realize” it? What on earth does that mean?”. This made me recall the reading by Fodor. In a similar manner, it can be seen in the perspective of “why does everyone go on about feelings”.
ReplyDeleteThere are numerous theories that discuss feelings and emotions. Physiological and neurological theories are two examples. The former implies that physiological mechanisms such as hormones direct emotions within the body. The latter implies that certain brain parts and connections are responsible for emotions.
In this regard, while it is claimed that emotions are linked to specific brain areas and physiological mechanisms, is this really important in determining the concept of feeling?
As someone in STEM, the idea of “what even is feeling” never occurred to me. It was always simple, we feel as a result of x, y, and z and the idea of feeling never was put into question. However,This course has made me reconsider previously learned concepts, such as how we can possibly build so many theories about feeling and emotion if we don't even know what feeling is. While it appears that feeling is linked to inner body function and brain mechanisms, there is a need for more emphasis on determining what IS feeling.
At the moment, we know so little, yet we don't take into account this further step. Thus, once again knowledge from this point of view trying to first understand what on earth feeling is can bring much relevance and deeper explorations than just what neuroscience has become today.
In your second paragraph you say that there are “numerous theories that discuss feelings and emotions”. I think it's important to differentiate “feelings” (or the ability to feel) from “emotions” although the two are closely related. There are definitely theories trying to explain emotions and you have pointed out two general frameworks: physiology & neuroscience. There are no such theories though for how we are able to “feel”. I agree with your third paragraph, and my answer to the question you pose would be yes, it is important to study emotions because it gets us closer to solving the easy problem, and completely explaining our doing capacity is probably our only shot at shedding light on the hard problem. I don't think that we should first try and understand what feeling is, as we're not going to get anywhere on that, but should focus on the easy problem.
DeleteFurthermore, upon thinking about this even more, in relation to this quote, “but it is qua feelings that feelings have to be explained (if we are to face the hard question)”.With respect to the hard problem being based on explaining HOW and WHY animals feel, I wanted to understand how this would then apply to cases of Alexithymia.
ReplyDeleteAn aspect of Alexithymia is that an individual has“difficulties recognizing and communicating their own emotions.” In this case, if the aim is to address qualia when facing the hard problem, If a solution were to arise, would it be a general solution that only applied based on the majority of the population?
1. We do not yet understand what exactly feelings are, and
2. If we did, then how can this apply to those with Alexithymia?
Moreso, how does figuring out HOW and WHY animals feel if there are some that appear to not? Where does the gap then provide further insight into what is required to solve if they are unable to experience feelings? Or, more specifically, if a person lacks the ability to recognize and share feelings, can we truly say feelings exist - more or less, just the subjective construct that we as a society have built into our lives to believe it as a concept?
There is this view that values are mind-independent, according to a philosopher named Sharon Street. In relation, the paper mentions, “ 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. If there cannot be such a Zombie, then you need to explain, causally/functionally, exactly how/why there cannot.”
ReplyDeleteIn this regard, would it be plausible to use Street’s argument to explain how/why such a zombie cannot exist?
Street argues that values are mind-dependent. This view grants the awareness of values promoting survival and reproduction. Specifically, in an assignment where I spoke on this argument, I mentioned street’s idea that “values are subjectively valuable to us because previous beings who valued the same things had a better chance of survival and reproduction. Consequently, those who lacked the values we considered valuable for survival and reproduction died out over time.”
Based on this, wouldn't it be reasonable to conclude that such a zombie would not have survived over time, owing to the fact that if they don't have feelings, they would'nt be able to live because they do not have the capacity to know what to value and what not?
Then again, I suppose in this suggestion; we return to the question directed to feelings/intuitions. That is, how do intuitions regarding good and bad feelings arise, when we currently don't even know what constitutes a feeling.
If Sharon Street could explain how and why organisms could not DO what they do (including acting as if they "value" things, like their children's welfare) unless they FEEL, she would have solved the Hard Problem.
DeleteI really enjoyed your critical review on Turing’s paper in the start of course, and enjoyed even more this one. Den clearly leaves out people’s feelings, perhaps because he thinks that “emotional reactions” = “feeling” which clearly is not the case. But I believe the biggest takeaway from this is that we are not trying to ‘record’ and infer someone’s feelings because that's just mind-reading, we are instead trying to understand how someone feels anger and why they feel anger. And obviously this cannot be done by heterophenomenology. Loved this review by the way!!
ReplyDeleteite enjoyed this thorough breakdown of Dennett’s arguments. I, like, many above, found it to be a comprehensive explanation of just why the hard problem matters, and why Dennett cannot write it off as he tries to. It’s helpful to see this constant reiteration of the essential concepts being discussed—of doing, feeling, the easy and hard problem, etc.—when, as Professor has continuously pointed out, so many weasel-words are being used in their place. Feeling and the question of how and why that happens is what is important, and that question cannot be circumvented with the invention of various other words like “belief” in their place.
ReplyDeleteI was deeply critical of Dennett’s paper because his method, heterophenomenology, was centred on treating feelings like any other physiological/biological process and how his approach seemed only to be able to partially answer the easy problem of how/why we do things. The hard problem is asking for causality to explain why we have feelings. The hard problem requires understanding that feelings do not necessarily equal function. Feelings and function are correlated but explaining how one works do not mean we know the causality of the other, as Harnad points out.
ReplyDeleteThis paper was interesting for me because in the 10a reading, I noted Dennett listing Gilbert Harman (of the "representationalist" revival of functionalism) among his A team, and I had just handed in a final paper for another class critiquing Harman. What I had to say was very similar to what is being discussed here. Harman's view is that everything that we are aware of is just a result of *what* we are being aware of (that "intentionality" weasel-word again), which can be captured in the language of functionalism (sensory inputs, etc.) and we don't tend to be aware of any abstract meta-qualities of our experience itself, and so there is nothing that functionalism leaves out. But what I have to say is that it doesn't matter whether or not we are aware of any meta-qualities of our experience over and above what we are experiencing, the important issue is that we are *having an experience* (i.e. *feeling*) in the first place, and Harman's school of thought cannot account for this.
ReplyDeleteZahur, I agree, and "de-weaseling" is always the antidote:
Delete"What I had to say was very similar to what is being discussed here. Harman's view is that everything that we FEEL is just a result of *what* we FEEL (that "FEELING" weasel-word again), which can be captured in the language of functionalism (sensory inputs, etc.) and we don't tend to FEEL any abstract meta-FEELINGS of our FEELING itself, and so there is nothing that functionalism leaves out. But what I have to say is that it doesn't matter whether or not we FEEL any meta-FEELINGs of our FEELING over and above what we are FEELING, the important issue is that we are *having a FEELING* (i.e. *feeling*) in the first place, and Harman's school of thought cannot account for this."
The other f-word (f-ism) is just a fancy word for reverse-engineering (both dynamic and computational) plus T-Testing...