The only feelings we can feel are our own. When it comes to the feelings of others, we can only infer them, based on their behavior — unless they tell us. This is the “other-minds problem.” Within our own species, thanks to language, this problem arises only for states in which people cannot speak (infancy, aphasia, sleep, anaesthesia, coma). Our species also has a uniquely powerful empathic or “mind-reading” capacity: We can (sometimes) perceive from the behavior of others when they are in states like our own. Our inferences have also been systematized and operationalized in biobehavioral science and supplemented by cognitive neuroimagery. Together, these make the other-minds problem within our own species a relatively minor one. But we cohabit the planet with other species, most of them very different from our own, and none of them able to talk. Inferring whether and what they feel is important not only for scientific but also for ethical reasons, because where feelings are felt, they can also be hurt. As animals are at long last beginning to be accorded legal status and protection as sentient beings, our new journal Animal Sentience, will be devoted to exploring in depth what, how and why organisms feel. Individual “target articles” (and sometimes précis of books) addressing different species’ sentient and cognitive capacities will each be accorded “open peer commentary,” consisting of multiple shorter articles, both invited and freely submitted ones, by specialists from many disciplines, each elaborating, applying, supplementing or criticizing the content of the target article, along with responses from the target author(s). The members of the nonhuman species under discussion will not be able to join in the conversation, but their spokesmen and advocates, the specialists who know them best, will. The inaugural issue launches with the all-important question (for fish) of whether fish can feel pain.
Subscribe to:
Post Comments (Atom)
-
Harnad, S. (2003b) Categorical Perception . Encyclopedia of Cognitive Science . Nature Publishing Group. Macmillan. http://www.youtube.co...
-
Searle, John. R. (1980) Minds, brains, and programs . Behavioral and Brain Sciences 3 (3): 417-457 This article can be viewed as an atte...
-
Fodor, J. (1999) " Why, why, does everyone go on so about thebrain? " London Review of Books 21(19) 68-69. I once gave a (perfec...
In this paper, you describe the hard problem as trying to explain how and why survival/reproductive machines would evolve minds; “what is the added causal role and adaptive value of having a mind, over and above the casual role and adaptive value of just having the behavioural capacities themselves, to do whatever needs doing in order to survive and reproduce?” Wouldn’t the added adaptive role of minds be the fact that consciousness increases our ability to survive through the self-awareness it gives rise to and increased ability to not only learn things, but understand the greater purpose of our actions, other than simply reproduction and survival. Our ability to think of ourselves as agents would provide us with the ability to achieve our goals more effectively than animals who have instinct alone.
ReplyDeleteAlexander, it's not for naught that I've since then phased out weasel-words like "mind" (and "consciousness" and "self-awareness" and "agency" and "intentionality" and...). The hard problem is to explain how and why organisms FEEL rather than just DO. Learning and reasoning and planning are things we DO. How/why are they also felt?
DeleteDr. Harnad; there are a few theories as to *why* feeling might be more efficient than a purely behavioural, cause-and-response mechanism. I'm sure you know them far better than I!
DeleteFeelings contain a tremendous amount of information at a relatively low cost to the organism, or so what I've learned in other classes goes. Alex this out as well. This makes sense; maybe we do know, at least conceptually, why we feel things.
I take it what we should be after is the specific mechanism that causes us to feel, and not what role it plays in our survival. It seems in this specific sense to be a *why*.
Or, perhaps, those two things should not be considered exclusive areas of investigation. Your point on the continuum between 'pain' and deprivation from environment in animals makes me think this would be the right approach.
Correction: "Alex points this out"
DeleteJacob, how and why do "Feelings contain a tremendous amount of information at a relatively low cost to the organism"? Easier said than done...
DeleteAlexander, it's not just our experiments on animals that have ethical consequences (for us) and far worse for them...
ReplyDeleteAlexander, please re-post your comment, on which this was my Reply. It was deleted by a bug in blogger whenever you have a conflict of browsers or logins, See: "Why do comments disappear from blogger?"
DeleteI would like to explore the point in which Prof. Harnad states, “our rational, observational and empathic mind-reading skills in general are far more extensive and acute than those of any other species (p. 4).” I do not necessarily disagree with this point, but I believe that humans are at an extreme advantage relative to other species due to our ability to speak complex languages. There is less of a necessity for mind-reading amongst humans because we are able to express most thoughts and feelings verbally. Animals require stronger mind-reading abilities because they do not have the alternative of stringing words together in a sentence to communicate their ideas. Of course, they do have sounds and signals, but they do not match the complexity of human languages.
ReplyDeleteKimberley, you may be right. And you are in good company:
DeleteKiley-Worthington, Marthe (2016)
Nonhuman mind-reading ability. Animal Sentience 1(2)
This comment made me think of antisocial personality disorders, especially psychopathy and sociopathy. Within our species, even with language, we can’t know what (and whether) others feel (OMP). Speaking and expressing our feelings doesn’t automatically give us a guarantee that others feel. The fact that we are able to express most thoughts and feelings verbally might make us think that there is less of a necessity for mind-reading amongst humans but it could all be a deception (humans lie and present themselves in favorable ways). I don't think that the whether there are feelings is obvious and facilitated (that much) by language within humans. Can we really trust language?
DeleteMelis, I think we can do almost as well on the question of whether other humans feel as we can on whether other mammals, birds, reptiles, amphibians, fish and most invertebrates can feel. But language (like heterophenomenology) is a big help in knowing what other humans feel -- and know. There would be a lot more uncertainty without it.
DeleteHi Melis! I definitely see your point about how language can be deceptive and unreliable. However, I believe that if someone is trying to be deceptive, this deception does not end with the language that they use. The deception can continue through gestures, body language, actions, etc. For example, someone may smile at you and hug you, but actually have a strong dislike towards you. The mind-reading abilities that we use to interpret actions may be just as incorrect as the abilities we use to interpret language.
DeleteKimberly, yes, nonverbal deception, including intentional deception is possible in species that lack language (as well as in the species that has it: psychopaths and method-actors use both,).
DeleteI find this discussion very interesting and I agree that complex language is an advantage, but it is also true that deception comes into the equation. On the subject of antisocial personality disorders, where empathy isn't felt, often people with the disorder will be particularly good at mirroring and masking, as in acting like other empathic individuals, either for deception or for self preservation (or both). I would be curious to know the difference in mechanics when it comes to emotional felt pain and physical felt pain in relation to that.
DeleteMathilda, good questions.
ReplyDeleteI. I’d say that our OMP uncertainty about sentience (i.e., whether they feel) is about the same for human adults and infants, nonhuman primates, mammals, birds, reptiles, amphibians and fish.
II. There’s somewhat more uncertainty with invertebrates (bees, spiders, lobsters, octopus), but that is changing quickly as we learn more about them and get to know them as individuals.
III. For the Precautionary Principle, having a nervous system (especially nociceptors) is probably an important marker of sentience.
IV. It becomes increasingly unlikely that sponges, plants, mushrooms or microbes feel.
V. Today there are some who think that all living organisms, right down to single cells, feel. This is a form of vitalism coupled with animism. (“Stevan Says” this is arbitrary nonsense.)
VI. Vitalism/animism blends continuously into panpsychism, according to which all matter (and each of its parts, and combinations of parts), from planets and galaxies to mountains, stones, sand, rivers, waves, atoms and electrons, and, I suppose, cadavers, feel. (“Stevan Says” this is not just arbitrary nonsense but absurd. It’s a symptom of the fact that there can also be OMP uncertainty about what does NOT feel.)
Apart from our OMP theories about what is or isn’t sentient, there is another important consideration, since life consumes life to survive: For the human species, what life-forms is it necessary for them to consume for survival and health?
Although it was different in the evolutionary past of the human species (descended from herbivores and insectivores), today there is no longer any need for humans to consume animals in category I, and most of category II, for survival or health, although some collateral damage to category II (especially insects) is unavoidable.
While we can seek 'conclusive' evidence for sentience in other species (ie bridging the other-minds barrier for other species), this pursuit is likely to be in vain. Indeed, other species are unable to tell us that they feel (ie no language) and sometimes even to show their felt states through observable behavior (ie not the same behavioral correlates of feeling as humans). Our mind-reading abilities are also limited in the case of other species. As such, even if we continue to look for evidence of sentience in other species, I agree with the paper that the benefit of the doubt approach should be applied in situations of uncertainty where there are dire consequences to a misevaluation (ie assert non-feeling when there is feeling). It seems crucial to identify cases where potential consequences largely overpower the degree of uncertainty regarding sentience, and to provisionally grant animals sentience in those cases. Most of the food and beauty/hygiene industry still relies on practices where the hurtful consequences are enormous if animals do feel, hence in the absence of total certainty that animals do not feel, we should assume that they do in these contexts.
ReplyDeleteAmélie, you wrote:
Delete”Most of the food and beauty/hygiene industry still relies on practices where the hurtful consequences are enormous”
I know your answer to the question, Amélie, because I know you are vegan, but, to dispel uncertainty: do you think any industry or practice that hurts or kills sentient organisms needlessly (not to save lives) does not have enormous hurtful consequences?
Loud and clear, no.
DeleteIt seems to me that the only way to fully solve the Other Minds Problem is to solve the Hard Problem. The moment we have a scientific theory explaining how and why (some?) organisms feel, we will automatically obtain a criterion for deciding what organisms can/cannot feel. For example, say we conclude that mechanism X is responsible for causing feelings; then we could infer that any organism lacking X cannot feel, and vice-versa.
ReplyDeleteGabriel, I certainly hope you are wrong that the solution to the OMP must wait for the solution to the HP! Not just for scientific or philosophical reasons, but because we humans are now 8 billion on the planet, and growing, and the amount of suffering we are inflicting on other sentient species, needlessly, is monstrous, incalculable, and greater, always exponentially greater, than it has ever been before, because our own population is growing exponentially.
DeleteGabriel, if you're right and "that mechanism X is responsible for causing feelings" if found, it would mean that there is only one way, a strong equivalence for solving whether and how some organisms feel. I personally don't think that would be very likely since all the beings that have the "possibility" of feeling are so different (this could lead to underrepresentation), a weak equivalence seems more likely.
DeleteThe precautionary principle put forward by Dr. Harnad is very philosophically interesting since it relates to Pascal's Wager.
ReplyDeleteAs explained in class, Pascal's Wager is a "flawed" philosophical argument that can be used to make people become religious (superficially). I.e., it is better to go to Church on Sundays during your lifespan and lose a bit of your time than it is to burn in eternal damnation forever.
However, here the "wager" is not flawed compared to the other one. It is not relying on any metaphysical concepts. It is based on the observable physical world where consequences are instantaneous. It is better not to harm other organisms since the consequences of us being wrong are at a much higher stake than if we were right (i.e., moral/ethical implications, possible future problems).
Alexei, or we could listen to what our mirror neurons are telling us.
DeleteIts interesting that the mirror capacities we have enable us to genuinely feel empathetic emotion, but only insofar as the behaviour of whatever organism we might want to empathize with is observable (virtually or immediately), and understandable. This problem concerns not just other sentient species, but our conspecifics too. It is well known that we do not have the same ability to empathize with abstract, anonymous people such as people suffering across the globe. Being in proximity to these people improves empathetic ability and affects ethical choices. With animals, the problem with self motivated ethical behaviour is not just of perceptibility or proximity: our mirror capacities aren't optimized for us to have compassion with other organisms, especially if they are less similar to us. This is a barrier to empathy, and might explain why we tend to anthropomorphize animals in pushing for having ethical relations to them.
DeleteDarcy, that's exactly what the journal, Animal Sentience is devoted to.
ReplyDeleteRe-posting Darcy's deleted comment:
Delete"“When it comes to other mammals as well as birds, the other-minds problem does not arise for the question of whether they feel. Everyone who has had animals in the family knows that they feel, as surely as human infants feel.” As this line states, we all can mind-read enough to see that species feel, so there is no need to spend a lot of time questioning this aspect of other organisms' feeling capacities. However, because we do not yet understand *what* organisms feel, an appropriate way to approach our interactions with them is to assume that they would feel the same things as us (precautionary principle). This opens another interesting aspect of the discussion that I did not think about before, as differences among species in terms of evolutionary pressures may mean that they feel other things that we cannot mind-read or experience ourselves. Studying the what aspect of other organisms’ feelings can allow us to better inform ourselves on how to better interact with them in an ethical manner."
Something that this paper has caused me to think about has to do with the point made regarding how since animals cannot communicate how they are feeling to us, we must infer it from their behaviour (which is “unclear” in some cases). I think a problem could be that the behaviours that are being searched for are “human-like,” and may not be able for some animals to execute. For example, my dog (who my family claims is “talkative”) whimpers and makes other vocalizations while lifting his paw when he walks on the snow (due to refusing to wear his boots), so we can easily assume his paws are too cold and he feels pain, due to similarities in his behaviour to our own. It would be foolish to assume that other animals (like fish) would execute the same behaviour when experiencing pain (among other states that would deem them sentient beings), so why aren’t we potentially looking for the behaviours they might execute before turning a blind eye and believing they don’t have the ability to feel. Like stated in this paper, it would be more beneficial to assume that all animals are sentient than believing they are not out of ignorance, and our own ease of mind.
ReplyDeleteKarina, there are ethologists as well as activists who devote their lives to getting to know other species and individuals well enough to mind-read their feelings -- both what hurts them and what makes them happy. I think our OM-reading mirror-capacities have a lot more potential than most of us realize or develop and that all species, especially vertebrates, are much more similar, or at least readable, than most people realize.)
Delete“Descartes’s “Cogito.” Descartes did not discover or invent the notion of a mind, but he did give it a concrete, almost operational definition: the “Cogito” (“I think, therefore I am.”)”
ReplyDeleteOne thought I had when reading this sentence that I have seen my entire life was that this logic is very simplistic in itself. It’s true in our personal experience that we know we exist because we think (or feel, to avoid any weasel words). But how does this translate to other organisms? Do they exist even though they don’t know they think? Do they even know they think? What does that make of plants, to whom we don’t attribute sentience, but who exist, and who breathe (if photosynthesis counts) and who do? This might have nothing to do with the class, but although plants might not feel like us, they are sensitive to their environment. They just don’t have explicit self-awareness; they don’t need to know they exist like we humans do. Maybe them and some animals alike simply lack the motivation. Maybe humans are just egocentric and feel only to prove that they exist.
Another note on Descartes–we now know the pineal gland is not metaphysical and certainly not the origin of the soul; rather, it modulates melatonin and other important hormones in the body. It is also shared by many vertebrates and not just humans, which contradicts his view that only humans have sentience.
All in all, the piece eloquently connects a lot of themes we went over throughout the semester. I wonder though; what is the difference between man-eating animals like polar bears, lions and crocodiles and animal-eating humans? It is just that man-eating animals cannot find protein elsewhere?
I apologize, I had posted this comment on the wrong thread.
DeleteProfessor, I just want to make sure I understand the concept of degrees of freedom in relation to the HP. So in statistics, degrees of freedom are essentially the number of INDEPENDENT pieces of information that go into calculating some estimate.. So if we have a set of 3 numbers with an average of 10 (ie. (9 + 10 + 11)/3 = 10) two numbers in the set are FREE but the third would be fixed (ie. (9 + 10 + x)/3 = 10). We can pick any two numbers, but the third has no freedom to be chosen as it must result in an average of 10. Translating this example to cognitive science and the EP/HP, the numbers are actually causal mechanisms reverse engineered by cognitive science. If we use those degrees of freedom (ie. causal explanations) to solve the EP and explain how and why we do what we do, then we have nothing left for the HP… no causal explanations for FEELING (ie. how and why it is we even feel). Is that correct?
Anaïs, that's right. Solving EP leaves feelings (including T4 correlates of feelings) causally unexplained; superfluous.
DeleteI think you're conflating a metaphysical argument with a cognitive one here. Descartes was addressing a completely different issue than we are here. He was questioning whether or not we exist and he used his capacity to think as a causal explanation for his existence (and only his existence). In other words, he would not be able to use his argument as an observer to justify the existence of others, just his own. whereas here, we override any metaphysical questions and we carry the assumption that we exist in addition to animals and plants.
DeleteAlso, you don't have to think to exhibit other mental states. As is mentioned in the reading "It does feel like something to think, but it also feels like something to be in any other mental state…Mental states are felt states, and to have a mind means to have the capacity to feel". Now, the question we're confronting here is whether or not animals can feel (thus are sentient) which is independent of whether they can think.
Injy, or, as Bentham put it: "The question is not, Can they reason? nor, Can they talk? but, Can they suffer?" (Jeremy Bentham, An Introduction to the Principles of Morals and Legislation, 1789, Ch 17. n.122.
DeleteFrom this paper, I understood that the other mind’s problem regarding animals is more complicated because they do not have the valuable skill of language to communicate with us. Indeed, when trying to mind-read another human, we can infer what they feel thanks to communication. Someone can tell me how they feel, giving me relative insight into their mind. With animals, it is different; we can only observe their behavior.
ReplyDeleteHowever, I am not sure what makes our mind-reading capabilities so powerful. As stated in the text, we know animals feel because our “acute mind-reading ability assures us of that.” But where does this specific knowledge come from? One view in the text is that this could be related to mirror neurons, but I only see this applying to observable human behavior. Indeed, with our mirror neurons firing both when observing and doing an action, we might have a better insight into one’s mind. But I am not too sure how this applies to animals. Do our mirror neurons also fire when observing animal behavior? Can they play a part in our excellent faculty of mind-reading non-human species?
Étienne, yes, our mirror-neurons fire when nonhuman animals do the same movement we do. And their mirror-neurons fire when we do the same movement they do. In fact, that's how Rizzolatti discovered mirror neurons: through invasive experiments inside their brains. In humans it was confirmed non-invasively.
DeleteI do not endorse these invasive experiments, and I hope they will no longer be done (especially since we already knew, from behavior such as imitation, that the brain must have mirror-capacities, somewhere, whether in single neurons or distributed in patterns across many.
And we already knew we could mind-read other animals. And that they could do it too.
Reading 11b and this thread seems to suggest that reductionist invasive experiments are getting us nowhere in terms of solving the EP nor HP, nor are they benefitting animal subjects. For example, we could have inferred mirror capacities from behavioural observation alone. In your opinion, Professor Harnad, was animal research necessary in the beginning for cognitive science, or could we have reached the same understanding we're at today without exploiting animals in the past/present?
DeletePolly, cogsci is so far not life-saving.
DeleteHi Polly, I also wonder whether animal research was necessary to pioneer cognitive science. Many publications on animal cognition have led to interesting philosophical standpoints (Baron and Klein looking at insect cognition to establish new principles of consciousness that arise from different brain structures, for example) that might or might not help us further understand feeling. While it is fascinating to learn about how other species interact with each other, I agree that the abuse of animals is in no way justifiable by hypothetical questions or curiosity. Research practices as a whole need to be entirely reevaluated, because the current pressures within academia to publish for the sake of prestige and relevance cause a whole lot of harm to creatures for outcomes that aren't life-saving.
Delete"It's the problem of other species, if they do have minds- if they do indeed feel, yet have the misfortune that our species does not know they feel, or does not believe it."
ReplyDeleteI wanted to ask a question. Although this article was mainly about animal sentience, I wanted to bring up robots because I think it's interesting to consider.
Note: I don't think robots are sentient. But does the notion of "consequence of error" still stand in respect to robots? What if they are sentient, but we don't know or don't believe it like Dr. Harnad points out above?
One time back in high school I helped deliver a robot baby from a robot mother, who blinked and bled and screamed. She literally told me "OUCH that hurts!" when she started having contractions. A girl in my class passed out because the simulation was so realistic. Are we wrong to assume mindlessness here as well? Is Robot Mom more mindless than a fish or a worm?
Hi Teegan, I for one am very curious about this robot labour… sounds freaky!
DeleteTo attempt to answer your question I think there are a few points. For one, I think animals should be the priority in the issue over robots. We are not in a situation where millions of robots are being killed and injured for the sake of human greed. Whereas repeatedly getting the OMP wrong with non-human animals has had dire consequences.
Further, if we have a T3 robot that can do everything we can do, I don’t think there is any reason to assume they won’t feel pain. Assuming they are embodied and have some type of nociceptive receptor, (which they would need to do all the things we do). If anything the robot would have an advantage over the animals, they can speak and communicate like us so can tell us if they're in pain. What reason is there to not believe them? As Stevan has reiterated (and you quoted) the OMP is a problem for the mind we’re trying to read. I think if we get to the point in robot sentience where this becomes an issue, then it is safer to assume they feel the same as we do. BUT I don’t really think this is an issue until we have engineered a T3 robot.
I can't say anything to improve on Sophie's reply.
DeleteTeegan, I think this is really interesting because it might opens up the possibility of something that only passes the Turing Test in certain domains (in this case, giving birth) being able to feel. While I do not think this robot is capable of feeling, there is no logical necessity for a feeling robot to be able to do everything a human does. As such, we should question how far the Precautionary Principle reaches. Of course, we cannot simply treat animals as if they do not feel until we prove that they can. The results would be and are disastrous. However, is it possible to apply the same principle to the birthing simulator?
DeleteElena, the birth simulator is just a toy t0. Turing's Test calls for completely indistinguishable cognitive capacity. All bets are off for toys. It's trivial to make a box that screams of pain whenever you touch it. It could probably make people faint too.
DeleteDr. Harnad, I wanted to go back to edit or rephrase the previous comment. I realized that the way that I wrote it compares the question of whether animals feel pain to something that is much less likely, which trivializes a very important issue. I understand that it is impossible that a t0 toy would feel anything, partially because they are comparatively very simple and follow a very restrictive program with very little ability to accept stimuli or to choose how to react to them.
DeleteI think this is a really interesting and thought-provoking discussion. This discussion lead me to consider the question of at what point should we be applying this notion of ‘benefit of the doubt’ to robots. Sophie brings up applying this idea to a T3 robot, but I wouldn’t entirely agree. We are applying this assumption to animals which lack many of the doing capacities humans have. That raises the question of what are the minimum capabilities necessary for a robot to apply this assumption of sentience.
DeleteThe point Sophie made about believing a robot if they say they are in pain was also interesting to me. There have been experiments in the past where researchers try to teach language to animals. Despite the appearance of success, these studies frequently receive significant criticism on whether the outcome results from operant conditioning vs true understanding. We already put forth the assumption that these animals are sentient, yet there is still significant doubt on whether we can trust what these animals are trying to communicate. Even if robots are not explicitly programmed to communicate with us and are just programmed with the capacity to learn to communicate with us, why would there be any less doubt about whether these robots are conditioned or biased to behave a certain way because of its programming and not because of its actual ability to feel.
Elena, thanks for the correction.
DeleteSophearah, T3/T4 is the test for whether a robot feels. There is no T-Test for nonhuman animals, but our mind-reading mirror capacities work well enough with mammals, birds, almost all vertebrates big enough to see, and, with time spent getting to know them, many invertebrates too.
Re-read the early weeks of this course on what computation is. "Being programmed" is not at all a clear notion.
In this article, Harnad explains how the other minds problem is the problem of other species who are at risk of falling prey to our tendency towards exploitation. The other minds problem is typically understood is the barrier of being able to know if and what other beings are thinking/feeling. Language, and other types of communication, allow us to report our thoughts and feeling and we need not doubt that. For other species however, there is far more at stake when we assume that they do not think/feel because we then assume that they are fair game to kill, eat, and exploit.
ReplyDeleteWell, I had no idea Prof. Harnad was such a cog sci big shot and leading the Animal Sentience journal. Very cool.
I am curious to hear more about your ideas pertaining to capitalism and evolution. You had mentioned once that the current state of global affairs was somewhat inevitable, due to our drive to constantly improve our standard of living that has sort of transformed in to a national greed. Care to elaborate?
Laura, self-interest is programmed into the genes of every species. What lazy evolution could not anticipate was what a monster self-interest could turn into, armed with the nuclear weapon of human language and technology.
DeleteTo expand on the connection to capitalism, I would say that the contexts in which our mind-reading abilities evolved was one where those humans and other organisms affected by our actions were primarily those we actually saw and interacted with on a day-to-day basis. This doesn't mean we would treat every single being with compassion, at least, but it meant we at least had the opportunity to mind-read to some extent; to consider their existence as a sentient being. But the advent of colonialism, its leading to capitalism, and capitalism's consequence of a globalized economic system, means that we don't often get the chance to perform mind-reading with those we affect: the animals either directly killed for food/other industries or who die from the effects of pollution, or those in far substandard conditions who produce goods that we use. Life, including a food and economic system, that is more local and centers the community level (and the local watershed, etc.) rather than our current non-local, hyper-extractive, environmentally-decoupled capitalist model (sorry Kid-Sib) would probably suffer less from a lot of these problems.
DeleteTo clarify, I meant to write "... or those *humans* in far substandard conditions..."
DeleteZahur, yes, capitalism and globalism are like ag-gag. Out of sight, out of mind.
DeleteI was reflecting on last week and the difference between Harnad’s and Dennet’s views on consciousness and the hard problem. The fact that Dennett removed the question of feeling by stating that feelings are only beliefs about feeling, effectively squashes the hard problem of how and why we feel (which I personally disagree with)l. I was wondering how that would impact what Harnad states in this paper on whether animals feel is still as important of a question.
ReplyDeleteI would like to assume that Dennett would take a more consequentialist view in the sense that the feelings (or his “ beliefs about feelings”) still matter in the sense that if we get the answer to the OMP wrong, then there are overwhelmingly disastrous consequences- mass suffering inflicted on individuals by humans without a second thought.
Kayla, DD believes that feelings are just beliefs. He also believes that mattering is just beliefs. So his feelingless, zombie world is indistinguishable from the real world -- except to every feeling creature in it.
DeleteJosie, I agree, but "the OMP is the OM's P" does not mean "Well if you're suffering, that's your problem!" Nor does it mean "Well if you're suffering, it's up to you to prove it to me."
ReplyDeleteBut I'm sure you know that already; this is just in case anyone else makes either of those misinterpretation.
The OM's P is that it's the OM is who's suffering, not the one who doubts it.
Re-posting Josie's comment:
Delete"The “Consequences of error” section helped me to understand what prof Harnad meant last class in saying that “the other minds problem is the problem of the other mind if you misdiagnose the other mind as not feeling when it is.” Before this reading, I thought this meant that the OMP is the problem of the other mind because it is up to them to prove that they can feel. In fact, as the article puts it, “there is something far more important at stake than just the truth.” That is, not only does the other mind have the problem of proving the truth (i.e. that they can feel), but moreover that they can suffer. For example, if I didn’t believe Kayla when she says she can feel, then I might kick her. By refusing to believe that she feels, she is in pain while I come out scot free. Therefore, it is the problem of the feeler to show that they can feel, and what they can feel — especially suffering. I think that the problem needs to be reversed; we need to make it our own problem if a feeler cannot prove that they can feel. As prof Harnad stressed, we need to listen to our mind-reading capacities and believe them."
Your mention of panpsychism is interesting as it's a notion I've been intrigued about for some time now. Something to do with its feeling somewhat intuitive (and its featuring in the His Dark Materials series, which I loved growing up!). But after reading your comment I came to wonder whether panpsychism suffers a "weasel word" issue, not fully defining "consciousness" and ascribing it to matter/the universe without necessarily considering either the implications or the specifics of what is being said - i.e., are panpsychists really saying that everything, including an atom, can "feel" when they are saying that everything has consciousness? I did some reading and it seems that this varies a fair amount across panpsychist theories. Some indeed don't seem to fully define everything; some specifically are "panexperientialists" who maintain that everything, down to an atom, can feel, or at least that experience "is present" throughout the universe; and others take an entirely different definition of consciousness, such as the occurrence or degree of "information integration" (as in the information integration theory).
ReplyDeleteZahur, but the bottom line on panpsychism is that it's just fantasy, and incoherent. And of course it obscures completely the distinction between what feels and what doesn't (which is no favor to chickens, pigs, fish, or vegans!)
Delete(As for Tononi's IIT, it's a promise to solve the OMP and the HP. Have a look and see whether it looks promising...)
This reading clarifies that the other-minds problem (OMP) is a problem for two interrelated reasons: (1) we cannot know that another organism feels without being them and, thus, (2) non-human animals which lack - from our perspective - salient sentience suffer. I've really enjoyed how this course has segwayed somewhat into ethics and ecology because I find it fascinating how cognitive science can shed light on how we can better respect non-human animals, knowledge which is key to achieving environmental sustainability.
ReplyDeleteWith what's going on now, it's hard to imagine teaching (or learning) without connecting it to what really matters.
DeleteWith nonhuman animals it's more a matter of compassion and protection than just respect.
It's interesting to see that the OMP can be applied outside humans, and this further connects to whether we should consume certain life forms and how much is necessary. We assume other humans can feel based on language and our shared biological structures, but this should not be an excuse for us to dismiss animals' feelings, as suggested by the "consequence of errors" from the reading. It is astounding how humans reject or overlook the cruelty and scale of animal consumption, and I think much of this is just based on cognitive dissonance, where we choose to believe animals are insentient or overlook this fact because we want to have the animal products.
ReplyDeleteJenny, robots aren't humans either. But ethical concerns are premature, till cogsci reverse-engineers a T3 (Kayla). Meanwhile, at least 150 billion feeling nonhuman beings are being slaughtered annually by eight billion of us, almost all needlessly.
DeleteIt’s interesting to note that for much of human history, humans have always found ways to justify massive cruelty, even towards each other. Slavery was a common practice around the world for centuries and justified by citing economic value, religious support, tradition, and the belief that those enslaved were born inferior. Nazi propaganda promoted the idea that their targets posed a threat. Millions of ordinary Germans supported or tolerated the crimes of the Holocaust for personal gain (getting jobs, acquiring property, etc). Even in the past, those that promoted human cruelty could not deny that those suffering were human, or at the very least, sentient beings. It might not even matter whether or not we can prove that animals are sentient. From what I understand, it seems like the biggest difference is the ability to communicate. In the past, those who were impacted by acts of cruelty were able to better defend themselves and receive empathy. They are able to learn to speak the same language to get empathy and fight for change. Animals cannot learn to speak but they still have a voice for change through animal rights activists and journals like Animal Sentience.
DeleteI found this reading very interesting. It linked all the important aspects of cognition to Key’s article and argument. I understood how complicated the other minds problem becomes once its applied to other species. Indeed, not knowing how other minds feel is complicated enough but when it’s applied to other species, we lose our two ways of discerning feeling in human: language and intuition coupled with empathy. When talking about other species, there seems to be no way for them to communicate or show their pain to us. I also understood the interests lying behind the question of certain species feeling. Indeed, as the reading points out, there are “financial, personal or even a scientific interest in being able to do whatever we like with them, hence in seeing and treating them more like feelingless machines than as sentient organisms.” Thus, the debate shifts in this reading as acknowledging that other different species from us might have a mind and the capacity to feel at all. I find it interesting that the idea of giving the “benefit of the doubt idea” or the “serious consequences of errors” don’t stand when there are financial interests on the other side.
ReplyDeleteInes, with other species we don't have language but we still have our other means of discerning feeling: our mirror-neurons.
DeleteThis paper placed me in a morally contemplative state. A particular tangent it made me explore was the use of animals as models of human behaviour. Looking back at my education in McGill in psychology, I realized that a shocking majority of the information shared to me was derived from (at least in part) animal analog studies. In these, we say that animals are 'good enough' models to predict human behaviour, and have the data collected from these to apply to our everyday human life. However, this is in massive disconnect with our societies brutal treatment, and permissiveness with animal harm, killing and testing.
ReplyDeleteIt is quite disconnected to use animals to study human pain, but not acknowledge their studied humanity.
Emma, having to cause suffering out of vital necessity is tragic. But causing it without vital necessity is a travesty.
DeleteI am aware that there are a lot of documentaries about "the secret life of pets" or something along the line of trying to understand what animals are doing, thinking, why they behave in a specific way and to what extent are they intelligent. No doubt that they are sentient but seems like the narratives are interpreting them from our POV. For example, this puppy has a guilty look so he must peed on the carpet. But, why? We desperately want to know what animals are thinking, but how are we going to penetrate the barrier of the other-minds? By inferring, sure. In class we mentioned some facial expressions mean different things across species and in fact, we don't share the same mind therefore it takes longer for us to "understand" what the animals are doing.
ReplyDeleteMonica, yes, some facial expressions have different meanings in different species; many are the same. But we are also good at learning them. (E.g., we've all learned what tail-wagging in dogs expresses; and those who know cats know that with them it can mean the opposite.) The canine "guilty face," was probably selected for in co-evolution and domestication. But it could also be a component of mammalian babies visual and vocal distress-signalling to their parents.
DeleteI am torn between the arguments for subjects of behavioural studies. Humans (prisoners/criminals) were used as subjects, then due to serious ethical issues, these studies came to a halt and moved onto animals (mice have been the animal model for over a century). Now, though some genetic studies and cancer treatments are conducted turn out to be beneficial for humans, but lab animals are mistreated in many absurd and gruesome ways, then what should we do next? Replace, reduce and refine all behavioural experiments or only conduct non-invasive research using human-subjects?
ReplyDeleteMonica, RRR while SSS (stop stop stop) all the other things we do to them for FFFT (fur, fun, finance and taste).
DeleteFindings from cognitive science & the like are sufficient to persuade entire countries to recognize animals as sentient beings. Sadly, the main change seems to be that instead of “billions of animals”, it is “billions of sentient beings” that are being slaughtered annually by humans. Cog sci helped enact, legislatively, more than it has proven scientifically. What more can cog sci do in the effort to save these lives?
ReplyDeleteThe “hard problem” is hard for cog sci because you can’t see or directly observe feeling, you can’t even be sure anyone or anything else feels (OMP). For science, this could be an insoluble problem as feelings don’t even seem to have a causal role! But the problem is not as hard for the politics of legal status and protection of animals- it is not to provide an explanation of how and why but an explanation of yes or no. There is a causal role of sentience in politics, and the cause is that legal status is granted. Thus, I think the greatest impact cognitive science could have in this domain is by showing that an explanation of the “doing” capacity of species is not their exhaustive explanation- the OMP restricts an explanation of the rest; suggesting the presence of a mind that cannot be explained and thus cannot be ignored/mistreated/abused/slaughtered.
Sepand, yes, the legislation has not saved many lives or spared much suffering so far. But I don't see how cogsci's HP is likely to help; it's more likely to be invoked as an excuse for not doing anything about the human-inflicted suffering of animals.
DeleteTo add to this, I feel like upon internalising the idea of feeling as something that exists in varying levels, and in many more organisms other than ourselves as humans, this can actually help us progress towards making it a priority to treat all beings with respect, appreciating the diversity in experiences other than what is known to us, rather than viewing the world and treating it as a resource to fulfill our own needs and desires. It's all about being active in treating animals as sentient creatures. And not standing at the sidelines in a "passive way" if that makes sense.
ReplyDeleteNadila, I think what Professor Harnad meant was that a causal explanation for feeling must involve some additional benefit of feeling to our survival. However, after we have answered the why behind our doing capacities, there is nothing to account for. Therefore, it is impossible to find a causal explanation for feeling.
ReplyDeleteThat's it.
Delete"And for every uncertain hypothesis, its opposite is likewise uncertain: (OMP)."
ReplyDeleteWe cannot be cartesian certain whether or not other species are sentient due to the OMP. Just because we cannot confirm whether other species can feel is not sufficient for us to conclude that they do not. When attributing animals' sentience, it seems that we tend to vary our choices based on self-serving interests. For example, it appears that the decision to include fish in one's diet does not depend on moral reasoning as to whether or not one's diet leads to unnecessary suffering. Instead, one's diet choice is based entirely on their liking and preconceived beliefs. The fact that humans are facultative carnivores makes our meat consumption habits unjustifiable. What is more senseless is that our biases in pain assessment in other species also apply to our own. It has long been documented that black Americans are systematically undertreated for pain-related illnesses compared to white Americans (Hoffman et al., 2016). Just like how we fail to attribute sentience to non-human species, many of those undertreatments are rooted in false beliefs about biological differences. Even if biological differences exist, they are insufficient to deny sentience in other species (not to mention in humans) because different causal systems can bring about similar functional outputs and (potentially) the capacity to feel. That being said, it is distressful to see that even members of our own species cannot escape the fatal misjudgement of sentience attribution.
Darcy, your comment reminded me of one of Key's points of view. Key states that our unsupported "anthropomorphic tendencies endow fish with the ability to feel pain simply because they attempt to escape from noxious stimuli." Clearly, Key believes that this form of anthropomorphism is insufficient to address the sentience in other animals. Though we cannot know for certain whether or not other species can feel the same way we can due to the OMP, I think this form of anthropomorphism is the foundation of our empathy for other species and the Precautionary Principle in animal welfare.
ReplyDeleteAdditionally, though Key opposed anthropomorphism in attributing animal sentience, he still utilizes anthropomorphic reasoning by assuming the mammalian pain circuitry to be the stand of pain perception. In other words, on the one hand, Key believes that anthropomorphism leads to misinterpretations of animals' behaviour outputs, which may cause us to assume that they are sentient. On the other hand, he utilizes the same anthropomorphic reasoning to conclude that fish do not feel because they lack the mammalian pain circuitry (i.e., Key knows for sure that he can feel and possesses said circuitry, but fish do not; therefore, fish cannot feel the same way he feels).
Yucen, (1) pain perception is not a doing capacity. Damage-detection (nociception) is a doing capacity. (2) Finding the circuitry that correlates with a feeling is part of T4 (EP). It does not explain how or why the circuitry generates the feeling (HP). Even a heterophenomenology that can identify the T4 (or T5) correlates and correctly predict every single JND of feeling would just be OM-reading. It would not explain how or why feeling organisms feel, so it would not solve HP. Test your understanding by explaining why not.
DeleteI thoroughly enjoyed reading what you wrote, Dr Harnard. When I was reading the previous article for this week (why fish do not feel pain), I could feel nothing but disgust. As you stated many times throughout the article, the other-minds problem despite not being so big within our species, is of extreme importance when it comes to other species that do not have language or have gestures very different from ours. And since I don't think we will have a scientific answer/solution to this any time soon, we need to make our decision based on the pros and cons of believing that other species feel or don't. With that in mind, it is clear that the hazards of believing that other species do not feel are much bigger than believing otherwise.
ReplyDeleteThe Precautionary Principle.
DeleteI am very interested in the connection this article makes with feeling and the other minds problem, and with the neurobiological structures of organisms, since it’s the first real link between the two that I think has been delved into so far in the course. The sentience of organisms may be linked to the complexity of the species’ nervous system, with the idea being that simpler, unicellular organisms likely aren’t capable of “feeling” as we mean in it when we talk about cognition and consciousness. However, there is no way for us to know if these organisms feel, due to the other minds problem, but by studying the structures that are responsible for things like pain and other sensation, we might get a clearer picture.
ReplyDeleteYes, correlates, and T4, can help with the OMP.
DeleteI entirely agree, most other psychology courses that I have taken often have involved learning about unethical studies that are done on monkeys, rats and other animals with the goal to be to try and solve the easy problem of consciousness, and yet these studies have brought us no closer to understanding it. If anything, thought experiments and models like Searle’s Chinese room seem to have brought us closer to the solution of the easy and hard problem, because they offer us a framework to view the very concept of cognition and consciousness through, while picking apart the brain, at the end of the day, does not.
ReplyDeleteOne of the reasons I find the field of cognitive science so fascinating is its approach to scientific topics and questions with such a focus on humanity. Throughout all of the topics of computation, categorization, and consciousness we’ve discussed in the course, it’s affirming to see it all eventually come back to feelings. This section in particular truly drives home the importance of compassion for those around us, human and animal alike.
ReplyDelete...and the importance too, I hope, of taking action on their behalf.
DeleteI enjoyed this reading because it makes individuals think about ethical dilemmas that are presented within cognitive science, psychology, and neuroscience, where there is a lot of research being done on animals without their consent to figure out human behaviours. There is a blatant disregard for the emotional well-being of animals within the field. Still, there is also this hierarchy of how we perceive animals and their ability to feel. For example, during Key’s reading, a continued thought I had was, what if he declared, “dogs do not feel pain”? There would have been uproar, but fish and rats are not given this same respect, despite how we use rats to figure out how mental disorders occur in humans.
ReplyDeleteSee Elena's reply below.
ReplyDeleteHan, yes, the OMP is connected to ethics.
ReplyDeleteBrandon, it's Turing rather than Searle who have pointed the way to solving the EP. (Searle shows what was not the way. (How?)
Maira, yes, we could mind-read other species much better if we got to know them well. Not bothering to do that leads to the self-fulfilling prophecy that they don't feel.
ReplyDelete...until/unless you get to know fish and rats. Then your mirror neurons make you realize (which means feel that you have to help them.
ReplyDeleteHow would mirror capacities acting in a human on human interaction compare with mirror capacities acting in a human and dog interaction? How would they differ if it was a human and fish interaction? I would imagine in the three cases there are different capacities going on. Would it be correct to say that the mirror capacities get weaker as we go from human to dog to fish?
DeleteYes, but not much weaker till you get to very tiny vertebrates or weird invertebrates like jellyfish, sponges or corals. And of course fungi and plants.
DeleteThere are two ways that we can sidestep the OMP when dealing with only humans: language and mirror capacities. Harnad says that even though we only have the latter when dealing with animals, it is enough to show that it is highly probable that animals too can feel. I think it's important to differentiate thinking from feeling at this point, because otherwise one might make the mistake of believing that because animals don't think the same way humans do, that they do not feel either. There are many different felt states which are unrelated to thinking, unfortunately suffering is one of them. This is why we must adhere to the precautionary principle. While we cannot know for certain if animals feel or what kinds of feelings they have, it is too risky to treat them as if they don’t feel. It is actually quite a horrifying thing to think about really, and more horrifying for the animals which we can basically assume do feel. One subject that really fascinates me and I would love to learn more about is mirror capacities between humans and animals.
ReplyDeleteBut it feels like something to think. All thoughts are felt. (Meaning = T3/T4 grounding + feeling.)
Delete"Mirror neurons" were discovered in monkeys, and they were activated when a monkey did a movement and either another monkey or a human did the same action as the monkey.
(This was not a life-saving discovery, so it was monstrous to do that with a monkey's brain. Besides, it can be done non-invasively in humans. I bet it would work in humans, watching a fish being touched, or affectionate.)