Key, Brian (2016) Why fish do not feel pain. Animal Sentience 3(1) (after reading the article, read the abstracts of some of the commentaries too, for contrary vuews).
Only humans can report feeling pain. In contrast, pain in animals is typically inferred on the basis of nonverbal behaviour. Unfortunately, these behavioural data can be problematic when the reliability and validity of the behavioural tests are questionable. The thesis proposed here is based on the bioengineering principle that structure determines function. Basic functional homologies can be mapped to structural homologies across a broad spectrum of vertebrate species. For example, olfaction depends on olfactory glomeruli in the olfactory bulbs of the forebrain, visual orientation responses depend on the laminated optic tectum in the midbrain, and locomotion depends on pattern generators in the spinal cord throughout vertebrate phylogeny, from fish to humans. Here I delineate the region of the human brain that is directly responsible for feeling painful stimuli. The principal structural features of this region are identified and then used as biomarkers to infer whether fish are, at least, anatomically capable of feeling pain. Using this strategy, I conclude that fish lack the necessary neurocytoarchitecture, microcircuitry, and structural connectivity for the neural processing required for feeling pain.
Only humans can report feeling pain. In contrast, pain in animals is typically inferred on the basis of nonverbal behaviour. Unfortunately, these behavioural data can be problematic when the reliability and validity of the behavioural tests are questionable. The thesis proposed here is based on the bioengineering principle that structure determines function. Basic functional homologies can be mapped to structural homologies across a broad spectrum of vertebrate species. For example, olfaction depends on olfactory glomeruli in the olfactory bulbs of the forebrain, visual orientation responses depend on the laminated optic tectum in the midbrain, and locomotion depends on pattern generators in the spinal cord throughout vertebrate phylogeny, from fish to humans. Here I delineate the region of the human brain that is directly responsible for feeling painful stimuli. The principal structural features of this region are identified and then used as biomarkers to infer whether fish are, at least, anatomically capable of feeling pain. Using this strategy, I conclude that fish lack the necessary neurocytoarchitecture, microcircuitry, and structural connectivity for the neural processing required for feeling pain.
In his text, Brian argues that fish do not feel pain. He demonstrates that fish respond to noxious stimuli (ie: an electric shock) by executing some motor program (ie: swim away), because of some innate non-conscious mechanisms in the brain. For example, when an electric shock is administered to the fish, it will try to swim away, and may show avoidance towards the electric stimuli. However, Brian argues that the fish does so without feeling any pain because it lacks multiple cortical structures which have been shown to be associated with the feeling of pain in humans and other animals.
ReplyDeleteAnaïs, please don't just read the target article but also at least the abstracts of the commentaries (they are like skywritings). There you will see that specialists from all over the world strongly disagree with the target article. Now what is your conclusion?
DeleteReading through the commentaries to Key’s article, it is clear that there is substantial evidence that fish have the capacity to feel pain, and do not solely react to stimuli based on evolutionary response mechanisms. This evidence not only disproves Key’s argument, but also demonstrates that fish feeling pain is not simply accepted at “face value” as Key suggests, but is supported with actual scientific evidence. Furthermore, Key states that people exhibit “anthropomorphic tendencies that bias interpretations of behavioural observations” in so far as seeing a fish react to a stimuli in a manner that we equate with our own experience of pain causes us to assume the fish, too, has experienced pain. However, I believe Key’s argument too carries bias because it assumes that the human experience of pain is the only way to experience pain, thus subtly asserting that human experience is the most advanced and valid experience of any animal. This perspective has ramifications in particular for our treatment of animals, because it allows human beings to write off animal suffering as less fulsome or meaningful than our own. In turn, this allows us to justify horrendous acts which we would never support if done to human beings.
DeleteEmma, the hypothesis that there is an élan vital (vital force) is called "vitalism." It has turned out that not only has no observable evidence been found for an élan vital in any biological organism, but there is no biological observation that cannot be fully explained without hypothesizing that there is an élan vital.
ReplyDeleteIn contrast, every human has direct ("first-person") evidence that they themselves feel, so there is no basis for dismissing the likelihood that other humans feel too: Yet there is no causal explanation of how or why (HP): In fact, there is no reasonto doubt that biology and cognitive science will eventually be able to explain all their observable evidence (EP) -- except the direct evidence of feeling.
So the need to explain feeling cannot be dismissed the way the élan vital could be dismissed: for lack of anything that needs to be explained. Hence there is an HP for sentience, but not for life.
("Stevan Says" that part or all of the belief in vitalism was probably based on animism, which is not a "vital" force but a "mental" force, probably rooted in the fact of feeling, and our sensitive -- but not infallible -- mind-reading capacities.)
But just as it is obvious (to the mirror-neurons of a feeling person) that other people feel too, it is obvious that many other species likewise feel -- certainly mammals, birds, fish, reptiles and all but the simplest of invertebrates. Serious doubt only sets in with microbes, fungi and plants -- and nonliving things, like robots. And the main correlate and predictor of feeling, apart from behavioral capacity (T2, T3), is a nervous system (T4, T5).
But what if I were to say I feel this vital force... Anyone could say (hopefully), yes I feel that I am alive. And because of this, there is such a thing as a vital force. Then, although there is no need for this vital force to be felt, and no need for it to exist to explain all of biology, would be be forced to reconsider this. I know this might be a weaselly answer because I am now talking about "feeling a vital force". Could feeling a vital force be enough to say it exists, is what I am asking, I guess
ReplyDeleteBrian insists that fish lack the necessary cortical structures to feel pain. Overall, what I understood is that, according to Brian, fish don't have a brain that is as complex as ours : it is not subdivided enough, not interconnected enough etc... The articles mentioned in the end however, mostly argue that we misinterpret the "complexity" of the brains of animals. The article on chickens seems particularly relevant: it suggests that although chickens are part of a family of birds that has been found to have "complex neural structures" as Brian would say, they have not been re-evaluated as such. As both fish and chicken are a large source of nutrition, we can see how our wish to view them as non-feeling and simpler organisms is problematic.
ReplyDeleteEmma, that's it. Wecome to "cognitive dissonance".
DeleteI find cognitive dissonance super interesting and I love these comments! I think we can assess feelings of animals by scientific method however and by just observing them quite simply. Yes, we would never know how they feel (HP) but we know that they do. The movie my octopus teacher proves it quite well in my opinion. The octopus clearly has feelings and the man and the octopus both start having feelings for each other. Now we may not know what the octopus felt but how is that actually relevant?
DeleteI agree with your point Marine. Just as underdetermination describes, we cannot be 100% certain about scientific theories--which in this specific case the feeling of pain in fish, because there will always be a gap between probability and certainty. As mentioned in other replies, the OMP is really not a problem for us, but for the 'other mind.' The OMP emphasizes underdetermination in that we cannot know if our friends and family actually feel, but we will give them the benefit of the doubt because our mirror neurons activate empathy, and through linguistic/ gestural communication, they make their feelings explicit to us. The only reason we would use our skepticism about whether any organism other than ourselves has feelings (although significantly less for our own), is to exploit others for our own benefit. We may not know what the octopus feels and so what? We also do not actually know what our friends feel either, just that they both do.
DeleteIn Key’s article a central argument to his claim that fish cannot feel pain is that the measures used to identify behaviours that indicate a fish is experiencing pain are erroneous, and that the behavioural assays that are used do not accurately measure whether fish feel pain. He suggests that measures should focus on disruption of normal behaviours, rather than reflexive behaviours (such as escape reflexes) I struggled with this point, as what is a disruption of a normal behaviour in humans may look far different for fish. Further, in Braithwaite and Droege’s (2016) response they take up this point, and note that evidence exists that fish that are exposed to painful stimuli do not behave normally. I also think that measures which study reflex behaviours to noxious stimuli (such as escape behaviours) should not be discounted, as they indicate that there is a detection of a noxious stimulus, something I think Key incorrectly disregards.
ReplyDeleteShona, that's right (and an understatement...)
DeleteHi Shona, that really intrigued me too! I was so baffled to how much Key disregards when he was boldly claiming that fish don't feel pain, especially when he literally contradicts his introduction to the paper stating that feeling is basically subjective and we won't really know with certainty if it happens or not. It's a bit ironic to say they don't feel pain, and then turn a blind eye to the reflexive reactions to noxious stimuli!
DeleteOut of all the papers, I agree the most with the paper "No evidence that pain is painful neural process" by Riccardo Manzotti. He states that one cannot assume that every organism requires the same neural structures to feel pain and also notes that pain is a complex phenomenon that involves not just neural structures, but also the body, the environment, stimuli, tissue damage, past and future behavior, and social interactions. He also suggests that we need to take a more comprehensive approach to understanding the physical substrate of pain, one that encompasses a broader network of causal processes, including the body and the environment.
ReplyDeleteMarie-Elise, yes, there are many ways a nervous system can generate pain. But what help is that holistic social story to a fish struggling on a hook, or gasping from suffocation -- or, for that matter, the unanesthetized dog shrieking as it's being cut open on Malebranche's vivisection table...? Is there really that much more we need to find out "scientifically" before we can draw the obvious conclusion? OMP is a challenge with robotic modelling, but with human-inflicted animal torment, it is an excuse.
DeleteI'd like to emphasize that my agreement with Manzotti's perspective in "No evidence that pain is painful neural process" is not intended to advocate for a purely holistic approach, but rather an acknowledgment of the complexity of pain. Manzotti's argument suggests that pain arises from a variety of sources including neural structures, the body, environmental interactions, and social contexts. This multifaceted view aligns with the concept of OMP. However, I agree with you that this broader understanding should not detract from the immediate and palpable suffering of animals in distressing situations. Recognizing the complexity of pain, as suggested by OMP, should lead us to a more compassionate approach to dealing with pain in all beings, balancing our scientific inquiries with ethical imperatives.
DeleteI have went through the abstract part of a few commentaries to Key's research: they question on whether a framework of consciousness is necessary to comprehend pain in humans and non-human animals. I think it prove that feeling is a matter, but HP (why and how human feel) is not only unsolvable but also unnecessary to be solved. While feeling is part of EP and takes important status in solving T4 and T5, feeling is not the motivation for our action and behaviors. Consciousness seems do not engage in every cognitive action. For this reason, what we need to know is as human, we could feel: it is an abstract trait for human and currently no need to expand it.
ReplyDeleteAlso, it reminds me a bit about the Baldwinian evolution, as it claims that evolution involves learning. An organism's interactions with its surroundings can cause changes in its traits, which are then progressively integrated into its evolving genetic and epigenetic repertoire. It seems like suggests that we got certain functions first then affects on our structure. (Please tell me if I misunderstood this - this statement is too over-simplified for some reason)
But Evelyn, this Week is not about humans and the human Other Minds Problem (OMP) with members of our own species, or our re verse-engineered robots, but about human OMP with members of other species. If other species really do feel, and we are feeling that we don't feel sure enough that they do, whose problem is that? What is the default OMP hypothesis when it comes to the human potential infliction of suffering on other species?
DeleteEvelyn, OMP is about probability and evidence; mathematics is about proof.
DeleteAll empirical theories are underdetermined.
Distinguish between prioritizing vital interests and prioritizing interests.
I believe the difference is that vital interests are ones that promote survival. Interests such as pleasure and comfort are not vital. Pain promotes vital interest as it informs an animal of injury that threatens its survival. Eating meat is not a vital interest for non-carnivores.
Delete"The salient element of a quale is that the experience is conscious, and it is hence dependent on conscious neural processing" (Key, 1)
ReplyDelete-> "We feel things because we feel things, and it is hence dependent on... feelings."?
(not my official skywriting)
Bravo Ohrie!
DeleteEmma, do you really want to take this direction? Does feeling a "spiritual" force mean that gods exist? Mine or yours? And what if I feel yours don'r? Does feeling a levitated state mean you are levitating? The only thing that the existence (and feeling) of feelings proves is that feeling exists. But your felt toothache only proves that you're feeling a toothache, not that there's something wrong with your tooth, or even that you have teeth.
ReplyDeleteThe paper argues that fish do not feel pain because they lack the neural machinery that allows for one to experience what pain ‘feels like’. Key first talks about the human brain and how conscious processing in the brain is dependent on signal amplification (increase in intensity of a signal) and global integration (sharing neural information) across the surface of the cortex. One part I found especially interesting was when Key talks about gamma oscillations and how ‘it is the synchronization of gamma oscillations between specific and distal cortical regions via recurrent connections that uniquely defines the quality of the feelings correlated with them’ (12). So, when different areas are sharing information they’re in sync, and this is necessary for consciousness because different areas of the cortex are involved in the conscious processing of stimuli. If this connectivity is diminished, under anesthesia for example, awareness is lost.
ReplyDeleteKey comes to his conclusion (fish don’t feel pain) by comparing conscious processing in humans to the supposed lack of conscious processing in fish. We are very different organisms so is his conclusion valid? I think not, because as he even mentions himself, we do not currently have the tools to adequately test whether a fish does or does not feel pain.
Fiona, read at least the abstracts of some of the 50+ commentaries ("skywriting") in the journal, experts from all over the world, and then come back and tell me whether you think Key's conclusion is valid (and why, or why not).
DeleteBrathwaite mentions how there is already evidence that fish treated with noxious stimuli lose the ability to perform normal behaviours. She also mentions that it is not sufficient to conclude that only human neural structures can process conscious pain. Fish have very different anatomy from us, and therefore lack the neural machinery to experience consciousness in the way we do, but that does not mean that they don’t feel. Their fish bodies experience the world very differently from us and so we cannot say whether or not they feel (OMP), so we should give them the benefit of the doubt since we cannot prove otherwise.
DeleteFiona, that's it, but de-weasel your reply to see what it ends up saying...
DeleteIf I understand correctly, another way to put the argument against Key without weasel words is as follows: we cannot say that human brain mechanisms are needed to feel pain. Fish bodies are very different from human bodies, so they may not feel things in the same way that we do, but that doesn’t mean they don’t feel—here especially, don’t feel pain—at all. Moreover, we can’t even be certain of when other humans are feeling pain, only I can be certain of my own feelings, due to the OMP. But just as I have every good reason to believe that another human being feels pain when they show behaviour that I recognize as a response to pain, so too do I have every good reason to believe that fish and other species are feeling pain when it looks like they are.
DeleteThere is already evidence that fish treated with noxious stimuli (stimuli that could be potentially painful, damaging, disruptful, etc...) lose the ability to perform normal behaviours. Fish have very different anatomy from us, and therefore do not experience the world in the way we do, but that does not mean that they don’t feel. We cannot say whether or not they feel (OMP), so we should give them the benefit of the doubt since we cannot prove otherwise.
DeleteThe commentaries on this paper point out the many weaknesses of Key’s argument. One that I was particularly convinced by was Braithwaite’s paper, where she points out that there’s no reason to conclude that only human neural structures can give rise to the feeling of pain. This relates to the idea of weak equivalence, where the same function can be accomplished by different functional structures. Moreover, we also should acknowledge that we still don’t understand the causal mechanism for the feeling of human pain, as this would involve solving the HP. Thus, it’s not even valid to say that we know why/how humans feel pain, and then conclude that since fish don’t have our same “pain” mechanisms that they don’t feel pain. Lastly, while Key may have been well-intentioned in coming up with these arguments, the ethical implications of his arguments are significant, as it lessens the implied consequences of inhumane fishing practices.
ReplyDeleteJessica, you got most of it. The main point is that evolution produced the same capacity in different ways in species with different brains. Nonsense to say it has to be implemented the human way in fish (or birds, or reptiles, or cephalopods... Evolution is too lazy for "Strong Equivalence".
ReplyDeleteBut it's not true that we have to solve the HP to identify the brain basis of pain. Even the OMP does not need to be solved either. Behavioral and brain correlates (together with ethology and observation are fine. The challenge is one of probability, not certainty (in our own species and the others too). And, where it's a close call (which is not until well below the fish) I'd be for giving the benefit of the doubt to the victim, not the tormentor...
In this reading, Key presents some evidence from the field of neuroscience to claim that fish do not possess the ability to feel pain, disregarding the fact that we have very limited knowledge on the pathways of human pain and how painful feelings arise, and even less on that of other species. Indeed, Key’s assertion of this claim is rooted in the incorrect assumption that the causal mechanisms behind animals feeling pain, such as the neural pain pathways, and the structural and functional aspects of processing noxious stimuli must be identical to that of humans, although some similarities are present. Furthermore, even if we could reverse-engineer nociperception, its causal explanation does not provide an explanation of pain perception. Put differently, solving the easy problem would not solve the hard problem! As such, in my opinion, claiming that animals lack the ability to feel pain and attempting to poorly justify this claim is simply a way to reduce one’s guilt about harming other species.
ReplyDeleteMelika, correct, except you don't mean the HP but the OMP. (Someone please make that point explicit.)
DeleteHi Melika,
DeleteI just want to make it explicit that you are referring to OMP and not HP. Solving HP in this context requires us to explain why pain "hurts" (i.e., why we have the *feeling* of pain), which may not be solved once we reverse-engineer nociperception. In Key's reading, he claims that fish do not feel pain; but we cannot know whether they feel pain or not because of the OMP; OMP is complicated enough in humans, it gets even more complicated when it's cross-species as we cannot know what it's like to be a fish. On a personal note, I believe that feeling (and sentience) can manifest in different ways across species and it is presumptuous of us to think that the world would mirror us; we cannot possibly be the *only* blueprint for sentience and it is wrong for us to behave as if we were.
Aashiha, thanks, you can be my TA for next year's Psyc 538.
DeleteOne thing you can be almost-sure of despite OMP and evolutionary variation and divergence: In all species, pain and fear are aversive and unpleasant -- in other words, negative, not neutral or positive. The HP prevents us from being able to explain the causal function of FEELING, rather than just DOING the right thing, zombily, but it would be absurd if that function were the opposite of its behavioral correlate's causal function.
I believe that Key's argument that fish do not experience or feel pain is flawed. Indeed, it relies on the premise that fish lack the necessary anatomical structures found in humans. He argues that since fish possess different brain structures, they must be incapable of feeling pain. However, this argument sets the human brain's pain inducing/processing structures as the standard for all living beings to feel pain, neglecting the possibility that fish may have alternate mechanisms for experiencing pain. Contrary to Key's position, Jonathan Balcombe's abstract challenges the notion that the absence of a cortex is sufficient grounds to dismiss the capacity for pain in fish. Balcombe suggests that if we accept this criterion, we must also deny the ability to feel pain to all non-mammals, including birds, which contradicts the established scientific consensus. I believe that Key forgets to consider that different species such as humans and fish evolved in different environments and under different conditions, which surely had an effect on the structure of the brain. Lastly, I think that structure is not enough to explain whether an organism feels pain or not. By making a link with the easy and hard problems of cognition, by looking at the structure we can probably answer the easy problem of how and why we can do what we can do, but what about the hard problem? Structure does not fully explain how and why an organism feels what it feels.
ReplyDeleteValentina, good points but make sure you understand that they are about OMP, not HP...
DeleteI went through a number of the responses to Key’s article, “Why fish do not feel pain” and found a few that echoed my concerns with the paper. As others have pointed out, Key, and those of us who eat seafood, have a vested interest in believing that fish don’t feel pain. Professor Brown’s article expands on this point, pointing out that since fishing is a massive industry, to popularize the idea that fish feel pain could lead to major economic costs to our fisheries. Brown further argues that Key’s structure-function framework for feeling pain, which he based on human anatomy, is improperly applied to fish. I had a similar feeling while reviewing Key’s paper. I have a decent background in human neuroscience and while the analogous structures between humans, rats, and apes have been well proven, fish don’t appear to be all that similar to us in structure. Unlike Key, I take this to mean that it would be wrong to apply knowledge about human neuroanatomy to fish, not that their lack of certain structures humans have indicates their lack of correlated functions. Professor Dinet’s response to Key echoes this argument. Key lists a number of rodent studies to support his argument, and then attempts to leap to fish without bridging the explanatory gap between rodents and fish, as Dinet points out. I particularly liked this section from Dinet’s response analogizing Key’s error: “crickets are deaf because they don’t have human-like ears complete with eardrums; birds can’t eat solid food because they lack teeth; and fish cannot move because they don’t have legs” (Dinet 1). Based on the articles I have to disagree with Dinet’s conclusion that fish don’t feel pain.
ReplyDelete^ Edit on the last sentence: I have to disagree with Key's conclusion.
DeleteLillian, good synthesis. Is there anything more to be said about that "vested interest"?
DeleteBy vested interest I mean that people who eat fish have a personal stake in wanting to believe fish don't feel pain so they can feel better about eating them. One of your other replies referenced "cognitive dissonance" which in this case comes from a belief that fish do feel pain, which is incongruous with the action of continuing to eat them.
Delete(Upon re-reading my post it seems like I'm accusing Key of having such a vested interest, though I don't know for sure that he isn't a vegetarian/vegan).
Hi Lilian, similarly as you, I thought about my neuroscience background when behaviors and pain where brought up in the reading. Indeed, many experiments have been conducted on rats about pain, stress and depression. These are relevant to humans as findings on rats can be generalized (thanks to anatomy parallels). Avoidance is often used in those studies - should be interesting on fish to measure their pain. When it comes to fish, we cannot use the same principles and make conclusions from them (even though they don't apply anymore). - It's not because a tree doesn't cry when I try to cut it down that it does not feel that I am cutting it down - .I also think that the "vested interest" argument cannot be made without further investigating the author's background.
DeleteLillian, people who do invasive research on fish have a personal stake in wanting to believe fish don't feel pain.
DeleteGarance, the OMP cuts both ways: one can believe that a rat is a zombie and a tree (or a stone) is not.
Key's assertion that fish do not experience pain reflects an anthropocentric perspective, relying on the absence of neuroanatomical structures in fish that correlate with human pain sensation. Critics and some skies argue that he oversimplifies the processing of pain, asserting that neural structures unique to humans are not the exclusive means through which pain can be felt. Given the vast evolutionary differences between humans and fish, it suggests that fish likely perceive the world differently than us.
ReplyDeleteI found Diggles' commentary interesting--aligning with Derbyshire's skepticism about using the term "pain" in a phylogenetic context. Derbyshire emphasizes the potential misconception arising from expecting equivalence between the human experience of being hooked and the fish's experience of being hooked.
One flaw in Key's argument arises as variability in neural structures across species implies that disruptions to normal behaviors, which may indicate pain in humans, manifest differently in fish. While I don't advocate justifying torment based on the arguments surrounding whether fish feel pain, I find it interesting that the disparities between human and fish experiences raise questions about the applicability of our human-centric understanding of pain to fish.
Elizabeth, no doubt what it feels like to humans to see color differs from what it feels like to birds or to insects (for whom the feeling is far richer than ours). But does anyone want to conclude from that that injury feels any less aversive to fish? Evolution is lazy, but it's not perverse: The unsolved HP means we can't explain why pain hurts, but it surely is not to attract rather than repel us.
DeleteKey argues that fish lack the neural architecture required for feeling pain. It emphasizes that conscious neural processes of pain necessitate specific organizational principles in the nervous system, which fish do not possess. The fish nervous system lacks the parcellation and laminated organization found in mammals, crucial for pain-related computations. The fish pallium, analogous to the mammalian amygdala, is suggested to be involved in nociceptive defensive behaviors rather than pain processing. The visual system in fish is described as primarily operating through feedforward pathways, lacking the strong reciprocal and networked circuitry essential for conscious neural processing. After I read this part, I thought “But how can we rule out the idea that there might be another structure that might enable feeling pain?”, and that is what Key mentions next. The possibility that fish might have evolved alternative brain structures for pain-related processing is considered, but the lack of evidence supporting the emergence of such regions in fish is highlighted. The argument against fish feeling "human pain" is presented, suggesting that fish might experience a different kind of pain or raw experience, but the absence of the necessary neural architecture questions this hypothesis.
ReplyDeleteThe conclusion asserts that there is a lack of evidence for fish feeling pain, and Key suggests reevaluating behavioral assays, proposing the analysis of loss of normal behavior rather than reflex behaviors as a more indicative measure of pain. The argument concludes by stating that, based on evolutionary biology and neuroscience principles, it is logically and confidently inferred that fish do not feel noxious stimuli, even though definitive proof is currently unavailable.
Selin, please read some of the critical commentaries following his target article in the journal.
DeleteDistinguishing "nociception" from pain is, at best,. It means "injury-perception". But if injury feels like anything at all, it is unlikely that it feels pleasant...
Why do fish not feel pain is a reading that goes over the neuroanatomical reasons as to why fish cannot feel pain. In fact, he argues that consciousness, or the feeling of pain is embedded in neural processes and arise in the cortex. It concurs with Searle’s view that consciousness is a neurobiological process and that to study consciousness, we can infer causation from lesion studies. While Key denies that there is a “fish consciousness” that would argue that our experience of feelings differs from that of fish, as well as the existence of a different neural pathway form ours that would allow fish to feel pain, it seems that Key wants to project the way we (humans) work and feel pain specifically onto fish. This entire article is based on the assumption that “structure determines function”: we should share similar structures that should lead to the same function, but it isn’t true that pain relies on the same neural processes in humans and in fish, and Key’s presupposes that feeling pain is a state of explicit knowing (Derbyshire).
ReplyDeleteMitia, evolution is full of examples where the same function has been produced by different means in different species. A good exercise is to look at the multiple ways in which vision has been implemented in molluscs.
DeleteAnd as for Derbyshire, try de-weaseling his abstract and see whether any of his verbal distinctions have any substance.
I think one of the main issues that resulted in Key's fallacious argument is that Key is too stuck in the human/mammalian perspective with regards to the OMP. The "solution" to the OMP for humans is that since I know that I feel, I can reasonably assume that others feel, but Key takes it too far in asserting that since he can study and confirm pain in humans and other mammals via brain imaging and lesion studies, then because he did not find the same results in fish, then fish must not feel. Many of studies in response have illustrated how this is fallacious, in particular Braithwaite, who shows that fish do indeed display behaviors that indicate pain, and Manzotti, who shows the insufficiency of relying on the mammalian brain to exclusively explain pain.
ReplyDeleteOmar, comparative and evolutionary biologists know this effect (related to underdetermination) -- that there is more than one way to produce a biological function, such as vision or locomotion -- very well. Philosophers call it "multiple realizability," but they often conflate it with implementation-independence and Strong vs. Weak Equivalence in computation -- and then they dub it all "functionalism." But Key does not appear to understand any of it. Can you explain it to kid-sib?
DeleteThis is interesting. Strong AI, also computationalism, is just symbol manipulation. Proponents of strong AI believed that the mind is just a computer and the programs being run in the mind could be run on any device / brain (implementation-independence). Based on Dr. Harnad’s comment, functionalism/multiple realizability is a theory that there are many different ways that the same function (e.g., pain) can be produced with different mechanisms in different species, or within species. These concepts are similar since they both assume that the way in which a outcome (e.g., computer algorithm) or function (e.g., vision) is attained doesn’t matter, just the final desired product. However, the critical difference is that computationalism and implementation-independence has to do with robots (likely zombies), but functionalism applies to FEELING machines (i.e. humans). Computationalism does not have to deal with the OMP or the HP.
DeleteKristi, good points, but wasn't Searle's Periscope about the OMP? How did Searle know he understands English and not Chinese?
DeleteSearle’s periscope does revolve around the OMP and the fact that feeling is an important part of understanding something. He concludes that he does not understand Chinese even though he is able to successfully complete the computations given by the experiment. This is because it feels like something to understand and Searle did not experience this feeling while producing answers in Chinese. He concluded from this that cognition was not computation at all, but this takes the issue too far in the opposite direction to computationalists. In fact, we see that cognition is partly computation but not all. So, Searle’s work on the Chinese Room Experiment was able to present the idea that we are unable to know if a machine understands what it is doing because we are not able to reliably know that it feels.
DeleteSearle bypasses the OMP by making himself the computer, who is in charge of computation. From successfully communicating with Chinese pen pals by following Chinese rules, Searle concludes that it feels different when executing Chinese characters and when communicating in English. He bypasses the OMP as he is now no longer wondering about the another machine, instead he himself is the machine. Therefore, from his periscope, we can say that since Searle executes computation perfectly as another machine, but still does not have the feeling of understanding, that computation alone is not enough for cognition.
DeleteJenny & Tina, good points, but wrong thread: should be in 3a and 3b!
DeleteOne of the key points leading Brian to his conclusion is the lack of connections and neural clusters in fishes’ brains to feel pain. Yes, pain integration takes place in the brain, but as mentioned by Braithwaite and Droege (2016), we can’t rely on the neuroanatomical knowledge we have about human pain to conclude that other systems can’t feel what we feel because they have a different brain structure. Withdrawal reflex can be seen in many vertebrates and invertebrates with even just a glimpse of a nervous system, such as worms, and the constant stimulation can affect the animal’s ability to produce a certain behavior. On the other hand, some researchers such as Broom (2016) demonstrated that the neuroanatomy of fish is very close to the mammals’, at least when we speak about the functions assessed by each brain area, suggesting that the pain circuit might be very close to what can be observed in humans.
ReplyDeleteI find this article especially pertinent when we think about it through the lense of the hard problem. What does it mean to feel something, why is it that we feel and how can we make sense out of it. This field of research goes even further by asking whether other species are able to feel and don’t simply have reflex circuits dictating their behaviors in specific situations. I really liked the caution of some researchers like Balcombe (2016) or Haikonen (2016) putting forward the benefit of the doubt about the uncertain conclusions that were drawn from very limited datas, and about an unsolved problem, even in humans.
DeleteOther skywritings have also mentioned that HP prevents us from understanding whether fish experience pain. Initially, I was also puzzled about why the HP, in addition to the OMP, wouldn't be applicable to the problem of knowing whether fish feel pain. It seemed logical that understanding how feelings arise in humans would enable us to investigate if fish possess these necessary systems for feelings to emerge. However, even if we were to solve the HP in the context of humans, we have no way of knowing whether the HP is implementation-independent; even if we were to unravel it for humans, we cannot assume that the same principles apply to fish, and whether they would experience feelings if they possessed the same mechanisms. We can’t even be sure that fish experience feelings in the first place, and so whether there even is a HP to solve for fish.
DeleteTherefore, we circle back to the fundamental issue of the OMP. We are unable to know whether fish possess the prerequisites for a HP to explain, and even if we were certain of their capacity to feel, we cannot determine if the underlying mechanisms are the same in humans and fish. Therefore, the obstacle is more rooted in the OMP than the HP itself.
Natasha, all your points are about OMP, not HP. Please be sure you understand the difference. No one has a clue of a clue about HP, for humans or any other species.
DeleteMy points are about the OMP because I was trying to explain why the problem with knowing whether fish feel pain revolves around the OMP not the HP.
DeleteWhile I was trying to formulate the distinction between those 2 concepts as I understood it, it made it more clear to me what are the topics covered by each one of those problems. Indeed, after reading my notes, I can see a more clear distinction between the hard problem and the other-mind problem. The first one addresses the link between a phenomena and the experience that we derive from it as well as why do all the neural connections elicit this specific sense of feeling. The other-mind problem on the other hand, concerns our ability to recognize that other individuals are capable or not of living the same feelings as we do. To try to answer your question, I would say that those experiments are indeed considering the mechanisms leading to the ability to feel pain which refers to the OMP, but if we were able to demonstrate that the nervous system of fishes allows them to feel, then we would have a base to address the hard problem, but we still wouldn’t be able to solve it.
Delete"Solving" the OMP (does entity X feel?) does not solve the HP (how and why does any entity feel?)
DeleteAs others have pointed out, Keys' argument relies on the assumption that for an animal to feel pain it must have the same neural hardware as humans, which, as Braithwaite and Droege point out, doesn't square with the disruption of normal behaviours in response to noxious stimuli seen in fish. He also makes the claim that our assumption of pain in fish has lead to restrictions in fishing that may impact food supply and nutrition, which feels like a leap beyond the scope of the paper. Gonçalves-de-Freitas argues that the paper reads like a selective argument against fish welfare more than a fair look at the issue, which is what I also felt. I would argue against Keys that it is valuable to assume animals feel pain even if we do not have concrete evidence, because I would rather limit cruel experiments and unneccessary animal suffering and be mistaken, than go ahead on the assumption that they do not suffer and be mistaken.
ReplyDeleteAdrienne, good points. See:
DeleteBirch, J. (2017). Animal sentience and the precautionary principle. Animal Sentience, 2(16), 1.
Key distinguishes between the feeling of pain and the behavioral reaction to a painful stimulus. I agree with Key on this point. However, as Broom pointed out in response to Key, the similarity in the behavioral responses of fish and mammals to pain suggests that fish might also experience pain (though this is not conclusive proof, only a probability). On the other hand, Chella argued in "Robot fish do not need sentience" that simulating the behavior of a fish does not require the capacity to feel pain, bringing us back to the HP (why not just zombies?).
ReplyDeleteKey's argument is straightforward, but the conclusion is flawed. The accurate conclusion would be that, because fish lack the human neural structures responsible for experiencing pain, we can only assert that fish do not feel 'human pain.' Key acknowledged this counterargument but dismissed it without providing a response. Therefore, we cannot exclude the possibility that 'fish pain' relies on other neural structures, which is the most common critique to Key's article.
Lastly, I don’t understand whether the suppression of species-specific behaviors in fish after craniotomy informs us about their feeling of pain. The change in behavior could be induced by the perception that there is something painful without requiring an associated feeling.
Joann, yes, FEELing is different from DOing (behavior).
Delete"Conclusive proof" is for maths. Science is underdetermined; it is about probability and evidence.
See other replies in Week 11 about why there is no animal TT.
The HP is not at issue here; it is just as unsolved for fish as for humans.
"Human pain" vs "fish pain" is moot. HP aside, it is unlikely, based on its evolutionary correlates, that "fish pain" feels any less aversive than "human pain".
The most frequent critique of Key is about parallel and convergent evolution ("multiple realizability")
"the perception that there is something painful without requiring an associated feeling" needs de-weaselling lest you tumble into "unfelt feelings."
I find Key’s overall method of argumentation fascinating, however I think it contains a major flaw. He states at the beginning of the paper that he assumes that structure determines function. On this basis, he pinpoints structures in the human brain he believes are responsible for pain perception in humans. Then, he argues that if these structures are missing in fish, fish must not be able to feel pain. Upon failing to find these particular structures in fish, he claims fish cannot feel pain. However, I think it is possible that a single function (like experiencing pain) could theoretically be determined by different neurological structures.
ReplyDeleteBy showing that fish do not share the neurological prerequisites of conscious pain perception in humans, Key does not rule out the possibility that a fish could still feel pain. I think it is conceivable that the conscious experience of pain can be generated by different organizations of the nervous system. Key shows that fish brains do not generate a conscious experience of pain in the way human brains do, but he doesn’t rule out the possibility that fish brains generate the conscious experience of pain in a different way.
Dani, please read the other comments and responses. Your (valid) point is about parallel vs. convergent evolution ("multiple realizability") and has been made repeatedly.
DeleteYour last sentence has to be de-weaseled..
I found Key's extreme skepticism about the feeling of pain in other species worrying not just for fish welfare, but for human welfare as well. The argument that reacting to avoid noxious stimuli is not sufficient evidence for the idea that an organism actually experiences pain may be right (Key's example is of human nocifensive reflexes). But taken to an extreme, it could lead to the solipsist solution to the OMP. We call the OMP a "problem" because we're not sure how to justify our tacit assumption that other people have feelings, which we infer based on their behaviours. But if you believe that behaviour is not evidence for feeling, then you can reject the idea that anyone else feels outright. Now you have no obligation to avoid doing harm to any person OR animal, because only you experience feelings. Indeed, Key seems to veer dangerously close to this idea with his discussion of hydranencephalic children. He argues that, although these cases have been referenced to show that feeling is not dependent on the cortex, their ability to perform functions like fixation and saccadic tracking doesn't necessarily mean that they feel, as these functions may be vegetative. Fine, but I think Key needs to reexamine what it means to make this argument in a paper where he has also said that our supposed illusions of fish pain are holding us back economically. I think you do need to give the "benefit of the doubt" argument a broad application or you could end up in some dark ethical places.
ReplyDeleteAya, wise words. Have a look at the replies about the history of human neonate anesthesia. The moral of this story is that blinkered "operationalist" researchers holding out scientistically for "proof" of what is obvious to our mirror-neurons are not the ones to trust on OMP judgments.
DeleteAsk GPT about (1) operationalism, (2) Occam's Razor, (3) Lloyd Morgan's Canon, (4) psychologists' fears of being accused of "anthropomorphism," (5) the "null hypothesis" on sentience, (6) signal detection theory, (7) cognitive dissonance and (8) the Precautionary Principle.
Yes, mirror-neurons are far from infallible; and they have a false-positive bias. But they also have millions of years of "tuning" by evolution, plus a lifetime of fine-tuning by close observation from naturalists and ethologists (including "whisperers" like Marthe Kiley-Worthington (11b)) who have no conflict of interest, either commercial or scientistic (sic).
Here's GPT4's reply. It's OK, if very superficial, on most of the questions )the usual C+/B-), but lots of weasel-words; it scrambled the null hypothesis, missed the potential relevance of signal detection analysis (false negatives, false positives, and biases), and dropped the sentience thread repeatedly, because of the weasel-words in its Big Gulp. But it gives you the idea in each case -- but needs to be followed up with some Google-Scholaring in the real publications.
DeleteIn the context of the other-minds problem of inferring whether other species are sentient, what can you tell me about about (1) operationalism, (2) Occam's Razor, (3) Lloyd Morgan's Canon, (4) psychologists' fears of being accused of "anthropomorphism," (5) the "null hypothesis" on sentience, (6) signal detection theory, (7) cognitive dissonance and (8) the Precautionary Principle.
DeleteChatGPT-4:
DeleteIn the context of the other-minds problem, particularly in inferring whether other species are sentient, several conceptual and methodological issues arise. Here's a detailed exploration of each of the topics you mentioned:
1. Operationalism: Operationalism in psychology is the idea that concepts should be defined in terms of observable operations, procedures, or experiments. This principle can complicate the study of sentience in other species since sentience, as an internal experience, is inherently non-observable. Researchers must rely on external indicators (behavior, physiological responses) as proxies for internal states, which may not fully capture the essence of sentience.
2. Occam's Razor: This principle suggests choosing the simplest explanation that fits the data. In the context of animal sentience, it may be invoked to argue against assuming complex mental states in animals if simpler explanations (like basic reflexes or conditioning) suffice. However, this principle can be contentious, as what seems 'simple' or 'complex' can depend on one's perspective and understanding of cognition.
3. Lloyd Morgan's Canon: This is a principle in comparative psychology suggesting that animal behavior should not be interpreted as arising from a higher mental process if it can be explained in terms of a lower one (e.g., simple learning mechanisms rather than complex reasoning). It's a precaution against anthropomorphism but may also lead to underestimating animal cognitive abilities.
4. Fears of Anthropomorphism: Psychologists often fear being accused of anthropomorphism, the attribution of human characteristics to non-human entities or beings. This concern can lead to overly cautious interpretations of animal behavior, potentially ignoring or downplaying evidence of complex mental processes in other species.
5. The "Null Hypothesis" on Sentience: In scientific research, the null hypothesis often represents a default position that there is no effect or difference. In studies of animal sentience, the null hypothesis might be that a particular species does not experience sentience. This stance can be problematic, as proving a negative is inherently difficult, and the ethical implications of incorrectly assuming a lack of sentience can be significant.
6. Signal Detection Theory: This theory, used in psychology and neuroscience, involves discerning between information-bearing patterns (signal) and random patterns that distract from the information (noise). In studying animal sentience, this theory might help in determining whether a particular behavior is a meaningful indicator of internal states or just 'noise.'
7. Cognitive Dissonance: This is the psychological discomfort experienced when holding two or more contradictory beliefs, values, or ideas. In the context of animal sentience, researchers might experience cognitive dissonance if their scientific understanding (e.g., recognizing signs of sentience in animals) clashes with cultural or personal beliefs (e.g., regarding the use of animals in research).
8. The Precautionary Principle: This principle suggests that if an action or policy has a suspected risk of causing harm, in the absence of scientific consensus, the burden of proof falls on those taking the action. In debates about animal sentience, this principle might advocate erring on the side of assuming sentience and thus ensuring humane treatment, even in the absence of definitive evidence.
Each of these concepts plays a role in shaping the scientific and ethical landscape of studying animal sentience and cognition. Balancing these considerations is crucial for advancing our understanding while respecting the ethical implications of our assumptions and conclusions.
I found GPTs comments about how to apply Occam’s Razor in the context of animal sentience to be quite interesting. It claims that Occam’s Razor in the context of animal sentience can become contentious since its unclear which explanations actually require the least number of assumptions. Keys seems to think that the route of least assumptions is to believe that non-humans like fish must possess the same neural correlates for feelings like pain due to shared evolutionary history, and if these correlates do not exist, then non-human animals must not have feelings. Though, Dinets may argue that the explanation that requires the least number of assumptions is that non-human animals must have also evolved feelings, especially given the behavioural evidence, though there is no reason to make the additional assumption that they must have been implemented by evolution the exact same way.
DeleteThe debate on whether fish feel pain revolves around differing interpretations of their neural structures and consciousness. Key argues that fish do not feel pain due to the absence of neural structures necessary for pain perception in humans, such as parcellation and laminated, columnar organization in the nervous system. He also denies that fish have consciousness or other structures for pain perception. Contrary views include Braithwaite's argument that pain understanding in animals should be framed within consciousness and that fish show behavioral changes when exposed to harmful stimuli. Hart discusses the overestimation of consciousness in human behavior, the assumption that complex behavior implies consciousness, and the misuse of terms like "pain." Manzotti argues that there's no conclusive evidence that mammalian neural structures are solely responsible for pain perception. Derbyshire points out that the necessity of the cortex in pain perception is clear only when considering pain as a consciously felt experience, making it absurd to attribute pain to fish under this definition.
ReplyDeleteAimée, it will reveal a lot if your paraphrase of Derbyshire's text is de-weaseled:
Delete"Derbyshire points out that the necessity of the cortex in pain perception is clear only when considering pain as a consciously felt experience, making it absurd to attribute pain to fish under this definition."
_"Derbyshire points out that the necessity of the cortex in feeling pain is clear only when considering pain as a felt felt feeling, making it absurd to attribute pain to fish under this definition."_
Or just de-weasel Derbyshire's abstract:
Subcortical circuits in the fish might support an immediate, raw, “pain” experience. The necessity of the cortex only becomes obvious when considering pain as an explicitly felt subjective experience. Attributing pain to fish only seems absurd when pain is considered as a state of explicit knowing.
_Subcortical circuits in the fish might support an immediate, felt, “pain” feeling. The necessity of the cortex only becomes obvious when considering pain as an explicitly felt felt feeling. Attributing pain to fish only seems absurd when pain is considered as a state of felt feeling._
If I were an advocate for the protection of newborn human babies, this kind of explanation would give me pause before consulting Dr. Derbyshire. I'd sooner trust his neural imaging than his insight. And if I were an advocate for fish, I would ignore him completely.
I found it ironic that Key himself states that “absence of evidence for pain does not mean evidence of absence of pain.” The irony goes even further with his paper defending himself to the commentaries that disagreed with him titled “Going beyond just-so stories”. Drawing conclusion on the absence of evidence and spinning just-so stories is, to me, exactly what he is doing. He is precisely using what we know to explain what we don’t know: He concludes that in the absence of neural architecture like that of the one associated with the feeling of pain in humans, fish cannot feel pain. Key’s argument for the inability for fish to feel pain is on the neural basis of pain in humans, to which he devised three steps: (1) identify which brain region is directly responsible for the feeling of pain in humans; (2) define the neuroanatomical features of this region that allow the neural processing that generates felt pain; and (3) assess whether fish have similar structural features that can then be used to make inferences about whether fish can, or cannot, feel pain.
ReplyDelete“Cognitive evidence of fish sentience” by Balcombe demonstrate the fundamental problem of using human anatomy, specifically the lack of a cortex, as grounds to understand pain in animals as this would unravel a host of problems for sentience for all non-mammals. The problem of underdetermination arises; just because we know that humans feel and have a certain anatomy does not mean that species with a different anatomy cannot feel. Braithwaite and Droege’s paper “Why human pain can’t tell us whether fish feel pain” builds around this argument and further demonstrate examples where fish treated with noxious stimuli lose the ability to perform normal behaviour, which serve as evidence for the ability to feel pain in fish. Broom’s paper “Fish brains and behaviour indicate capacity for feeling pain” links back to previous readings in decentralising the focus on anatomical and neural structures and instead on functionality. This, however, still does not solve the how/why problem of feeling as we only deal with its correlates rather than the feelings themselves. Key himself addresses this in his example of gamma oscillations: “It is not yet known how these self-regulatory oscillations may generate the pain sensation; however, strong functional interconnectivity between discrete cortical regions is essential for pain.” The hard problem remains unanswered!
Jocelyn, excellent synthesis -- but none of this is about the HP; it is about the OMP.
DeleteMy comment on the HP was specifically in regards to Broom’s and Key’s attempt to focus on functionality over anatomy. The basis of Key’s paper is indeed about the OMP, demonstrating how the extent of the problem exacerbates with difference. In this case, Key addresses a different species that we lack the ability to mind-read and communicate.
DeleteThis reading provides valuable insights into how the human brain processes pain. However, I also disagree with Key’s conclusion of fish not being able to experience pain due to the absence of a neural architecture analogous to humans. Our understanding of the feeling of pain may be anthropocentric, limiting our ability to perceive pain in non-human entities, like plants and fish. Tomato and tobacco plants emit unique sounds when cut or dehydrated, and fish exhibit flailing and struggling behaviors when hooked. These responses parallel human reactions to pain. Understanding how non-human entities, such as fish, experience things poses a continuing challenge. What about other feelings such as excitement, fear, and anger? Can fish and animals experience them, and how is it different from how humans experience them?
ReplyDeleteRosalie, you need to reflect more on the contrast between the behavioral and neural data on fish sentience and the "behavioral" and "neural" data on plant sentience. Do you think they are equally strong?
DeleteNo I definitely don’t think behavioral and neural data on fish sentience and the “ones” one plant sentience are equally strong. Besides behavior, another thing that predicts feeling is the nervous system. We know that plant roots grow in soil and that plants do photosynthesis, but they lack a nervous system that humans and other beings like fish have, so it is very likely that they do not have feelings.
DeleteRosalie, but is that because of the lack of a nervous system or because the "behavior" (growth and photosynthesis) does not need either sentience or a nervous system?
DeleteThere are many specialists who disagree with Key's conclusion for many reasons. I would also like to question his explanation that fish "exhibit innate escape responses involving non-conscious sub-pallial pathways" rather than felt pain. How can we say that innate responses are not felt? It feels like something to me when my hands reflexively cover my ears upon hearing a loud painful noise. The mere fact that fish exhibit behavior that prove they can detect noxious stimuli, combined with our knowledge of Darwinian and Baldwinian evolution, strongly suggests that fish feel negatively towards noxious stimuli. Also, when key describes the "defining properties of conscious processing" (page 11), which sounds like the HP (conscious processing = feeling, or felt states) he provides correlative evidence, which is not the HP solution and it is incorrect. Especially when we talk about other species.
ReplyDeleteCsenge, good points (but again conflating HP with OMP). Make sure to clarify them in your mind: OMP is about whether and when organisms FEEL. HP is about how and why organisms who FEEL, FEEL rather than just DO.
DeleteKey’s argument could be summed up as saying, “Because fish do not have the neural structures that are necessary in humans for humans to feel pain, fish do not feel pain.” And, though it’s clear that fish respond to noxious stimuli and display escape behaviors, Key attempts to claim that there’s nothing more to it — that the escape behaviors are reflexive motor outputs, and that fish do not feel anything in response to noxious stimulus. He rejects the precautionary principle (which, in this application, posits that we should give fish the “benefit of the doubt”) on grounds that it can lead to “unsupported anthropomorphic conclusions”, but then shrugs off the (valid) consideration that pain and other feelings could emerge in fish via systems different than ours. His view of fish is strangely robotic and makes them out to be little more than input-output zombies. It seems to me that his argument takes advantage of the difficulty of the OMP in order to draw a conclusion with severely negative implications for the fish towards whom he argues we should drop all empathy.
ReplyDeleteAdam, excellent synthesis: spot-on.
DeleteKey’s article breaks down the neurobiology as to how humans feel pain, and then goes on to list 4 main neuroanatomical features that allow for vertebrates to feel pain. These 4 features also seem to mostly rely on the conception of human cortex, something I have that seems disingenuous when comparing to radically different specimens. I also believe that downplaying the role of reflexive pain behaviours in fish is a major weak point in his argument. Yes we can’t directly infer that fish feel pain from these behaviours alone, however, doesn’t the existence of these behaviours imply in some way that the fish are attempting to remove a stimuli that if left uninterrupted would harm their chances of survival. The reason we have reflexive behaviours like immediately moving our hand away from the stove before our brain can even process pain is to stop our body from being harmed. It feels safe to assume that fish would exhibit these behaviours for similar reasons, and that reason being noxious stimuli over any other stimuli. While reading through the abstracts, the paper by Derbyshire resonated with me most. He states that subcortical circuits in fish may “support an immediate, raw, ‘pain’ experience”. Derbyshire takes the pragmatic approach that yes fish most likely feel pain, and although key’s argument avoids the abstractness of the nature and subjectivity of pain, there are inherent flaws in said argument. The biggest flaw in Key’s argument being that we need to blindly accept that the fish nervous system cannot generate pain as it is unlike our own. To Derbyshire, the cortex is only obvious in relation to pain for the explicitly felt experience of pain. Fish can experience pain, but whether or not they know they experience pain seems to be actual problem at hand here.
ReplyDeleteEthan, you grasp it well -- except when you are seduced by the W-Ws that make no-content verbal distinctions that lead to either empty redundancy or self-contradiction.
DeleteStarting from "The reason..." your words become hopelessly weaselled. De-weasel them and you'll see that what is left is what Adam says above.
And remember that no one has a clue of a clue about the HP (why do we FEEL, rather then detect and DO?). So this is all about the OMP, not the HP.
Josie, it's all about behavioral and brain correlates, mind-reading, and the OMP (which is part of the EP). How do you think a closer look at the brain could solve the HP?
DeleteI find it funny that Key first takes issue with the assumption that fish feel pain, claiming that we are biased and anthropomorphizing fish, but then he decides that the best way to see if fish feel pain is if their nervous systems and brain have similar structures to humans. He mentions Nagel, and the OMP, but dismisses it quickly, saying that it doesn't really apply because fish are so different from humans.
ReplyDeleteBecause pain is a feeling, we can only study correlates of it, and correlates of pain aren't pain themselves, and in fish these correlates may look very different!
Key speaks of rats and mentions that despite our common ancestor with them being distant, we share cortical structures and similar 'pain' patterns, and it is agreed upon that rats feel pain(there is a whole grimace scale to study pain in rats and mice and this makes me incredibly sad). However, I don't follow the logic that it has to be this structure and this pattern that is alone responsible for pain(which is a feeling!). It is the same problem with reverse engineering, which is that we may be able to reverse engineering a trait, or feeling, but we don't know that that is the way it occurs(there could be a number of different ways to reverse engineer it, all with the same outcome). I found "On the sentience of fish"(Haikonen) to be excellent supplementary reading, as it brought up the important point that we don't know which parts of activation in the brain can be attributed to feeling, and detection of noxious stimuli in the brain. If we could separate the two and then study it we might have better hope of this(implications for somatic symptom disorders and CIPA here).
Hey Josie, it's super telling that in the section where he defends the cortex as the hub for the feeling of pain, there is no mention of the translatability of this to fish. If you Ctrl + f the article for the word "fish", you'll see the spot I'm talking about. I also think that in this section he too readily bridges the translational gap between other mammals and humans.
DeleteI'd love to comment on what Prof Harnad said regarding it being obvious that "many other species likewise feel". Dogs, like many vertebrate species, exhibit clear bodily expressions such as wagging their tails to show happiness, exposing their teeth in aggression, and flattening their ears and tails when fearful or submissive. Even an octopus, despite being an invertebrate, shows similar remarkable expressive behaviours. They possess specialized cells called chromatophores in their skin, which allow them to alter their appearance by expanding or contracting these cells. Their ability to change colour is primarily used for camouflage, but it also serves as a means to express emotions. For example, when threatened or agitated, an octopus might turn a darker colour whereas they typically have a lighter coloration when relaxed. This cross-species similarity in expressing emotions through bodily changes points to how different organisms have evolved convergent ways to communicate their internal states, despite major biological differences. It's quite fascinating to think about how these expressions of felt states are so similar across species...
ReplyDeleteI wanted to return to this paper and something I said in my previous skywriting, that being the big flaw of Key’s paper. Derbyshire’s critique (and one that very closely resembles my own) is that Key’s argument rests on the fact that a fish’s nervous system cannot generate pain as it lacks the neocortex we have and the structures our brain relies on to feel pain. I wanted to further critique Key on this point, mostly on his use of the term “Fish”. Key mentions the term “phylogeny” a couple times throughout this article. And in bringing this up I want to expand on why I believe it was disingenuous to compare humans to fish. Fish are a paraphyletic group, which essentially means the term includes almost everything in the ocean. Monkeys would be another paraphyletic group, while humans monophyletic (something much less broad and all encompassing). Although everything here is a gross simplification of phylogenetics, comparing humans, to the broad category of fish, where neuroanatomical structures may have minor to large differences between the species seems like a large oversight and oversimplification for the sake of his argument.
ReplyDeleteEthan, good point, and part of Birch's argument for the Precautionary Principle.
DeleteI really find it increasingly fascinating how one can simply assume that a specific organism can or can’t feel emotion or pain. On numerous different occasions in this course, I have been brought back to this dilemma of, how can I know if anyone else but me can feel. It is the same question I have for others as I have for certain animals. How is it that we can be so dismissive of other organisms' ability to feel when we just assume on a daily basis. Meaning, I just assume every human I interact with around me has the same feeling capabilities as I do, no more so than I assume that every dog I come into contact with also has the ability to feel. It’s truly a remarkable thought experiment that I find myself coming back time and time again throughout the duration of this course. Personally, I find it really difficult to draw a line. In the example of the fish for discussion sake, I find it ridiculous to dictate that fish cannot feel pain, when you have certain fish species mourn after the death of their offspring. It has been observed, for example, that Koi fish show signs of distress after the death of a fellow fish in a given pond or tank. Although not considered fish technically, dolphins and whales are seen to weep and mourn for passed offspring, relatives, or companions. I think it is important to try and find a means of measuring the ability to feel, but I believe it has much more to do than just simply dictating neural capabilities.
ReplyDeleteStefan, welcome to the OMP, which is about probability, not certainty:
Delete1. There is OMP with normal human children and adults. Hardly any with preverbal infants either. But there is an OMP with very early stages of the embryo, comatose patients and patients in chronic vegetative states.
2. No OMP with mammals (including whales and dolphins) nor birds nor fish nor amphibians nor reptiles, nor with all but the simplest invertebrates.
3. OMP becomes substantive (and feeling becomes improbable) only when we get to organisms without nervous systems (very primitive invertebrates, microbes, fungi, and plants).
4. And with TT candidates, from T2 - T5.
5. The correlates from which you can infer sentience (feeling) are both behavioral and neural.
6. You can also use your mind-reading capacities (mirror-neurons), but don't be gullible: elephants, but not fish, mourn (grieve) for lost young. the feeling of mourning is much harder to infer from behavior than the feeling of fear or pain. Direct, informed anthropomorphism helps, but uncritical hearsay does not.
Like Adrienne, one of the comments that I read was Gonçalves de Freitas'. Key's statement about how the discovery that fish feel pain might lead to "inappropriate approaches to fish welfare", "serious negative implications for native subsistence fishing", "human nutrition and food supply" seems ridiculous to me. Key does not deny the fact that animals feel pain. However, the fact that farm animals feel pain has resulted in no such consequences - people still eat burgers, and meat is still a large part of many peoples' diets. Why would it be any different for fish...?
ReplyDeleteI read the Braithwaite and Droege review as well. Though others seem to agree in general with their commentary, one thing that I find myself disagreeing with is their statement that "the question is not
whether fish have a cortex or an insula [this part I agree with], but whether they have the functional capacity for global
integration and attentional amplification" [this part I'm not sure if I agree with.] I feel that this would simply be another form of mapping consciousness to neuroanatomical function which, as we've seen in this course, is not sufficient to explain the HP.
Ohrie, it's good to be critical about anthropocentric neural correlates, but not about the importance of neural correlates to complement behavioral correlates.
DeleteThis article was an excellent example of how science can be used as a subjective tool to fool the masses and convince the gullible public, especially as it fuels consumerism and anthropocentrism, making the unsuspecting reader feel not only superior as a species, but also morally good about their choice to eat fish. How convenient for humans. Certainly, throwing around a bunch of neuroscientific jargon words and going in-depth about (human) neuroanatomy might be more than enough to convince the public that Key must know what he’s talking about, but a deeper analysis reveals a great deal of hypocrisy –especially as others have pointed out, Key writes in his introduction that “absence of evidence for pain does not mean evidence for the absence of pain” but then spends the rest of the paper proving his arguments on that very basis.
ReplyDeleteIn fact, science being used as a tool for consumerism and capitalism is nothing new, and this article reminded me in some ways of how the tobacco industry historically swayed what scientific information about the dangers of smoking was available to the public, and even to this day, the way that the world’s most powerful companies and oil industries strategically downplay climate change, such that the mindless consumer continues to consume and the economy continues to grow.
The moral of the story is: As humans and consumers, it is important for us to be mindful of our own limited perspective and retain the ability to question when something is presented to us as a scientific fact, even if it bursts our comfortable bubble of mindless consumerism.
The connection to consumerism is very interesting!!
DeleteKey’s argument, centring on the absence of certain neural structures in fish, opens a Pandora's box in cognitive science and ethics. The crux of the debate isn't just about neuroanatomy but about understanding consciousness and pain beyond our human-centric perspectives. If we acknowledge that evolutionary paths can diverge wildly, then why confine the capacity for pain to structures familiar to us? This nudges us toward a broader, more inclusive understanding of sentience, urging us to consider that the absence of human-like structures doesn't necessarily imply the absence of experience. Ultimately, in my opinion, Key's hypothesis, while intriguing, may be too narrow a lens to view life's varied experiences.
ReplyDeleteI agree with your comment on Key's argument. This debate in cognitive science and ethics points out the difficulties we face when trying to understand consciousness and pain outside of human-centric perspectives. The fact that evolutionary paths can diverge significantly made me question the constraints of limiting the ability to feel pain to structures that we are accustomed to. We should be able to see that diverse species may possess unique neural architectures that enable experiences beyond our immediate comprehension. It's essential to remain open to the possibility that different structures can still support experiences. We should look beyond seeing things only from a human perspective and explore the diverse ways consciousness might exist in various living beings.
DeleteBrian Key's point that fish do not feel pain due to differing brain structures seems overly simplistic, particularly when contrasted with views like Brown's evolutionary approach and Striedter's argument on neocortex absence. These perspectives suggest that differing brain structures do not necessarily equate to an absence of pain. This discussion ties back to our ongoing examination of the Hard Problem of cognition. Pain perception in fish, or its denial, seems reflective of broader cognitive and evolutionary mechanisms, questioning our assumptions based on human neuroanatomy. It highlights the need for a more nuanced understanding of pain across diverse species.
ReplyDeleteI agree to your observation but one, that the emphasis is more on the Other Minds Problem (OMP) than the HP, since the question of interest (or criticisms to Brian's paper) is how much we can be sure of others' feelings compared to our own. In contrast, the argument still did not address the HP in how we feel the pain, only assuming human's neurocircuitry causes feeling of pain. You've mentioned a good point that differing brain structures do not necessarily equate to an absence of pain, specifically, fishes may just be feeling "pain" which does not feel the same as us feeling pain.
DeleteKristie, neither the human nor the nonhuman animal OMP helps with the HP: it only makes it harder.
DeleteMiriam, expressing behavioral intentions is something many species do, but the intention need not be felt. (just as the camouflage coloration need not be felt). Yet another example of why the HP is so hard.
ReplyDeleteKey argues that some vertebrates, such as fish, are unable to feel pain due to the lack of certain neural structures, and that their behavioural response to pain is not because of their feelings and need to escape the pain but instead ‘the circuitry has evolved to ensure survival’. I do get the temptation of using humans and our perception of the world as the standard of what every other living being is supposed to be like since we’re most familiar with our own selves, but it is simply unfair to solely base all the ‘evidence’ on that comparison and make a ground-breaking assumption of what and how they feel/can’t feel.
ReplyDeleteI found one of the commentaries by Ng interesting: he uses the term ‘flexible’, in which he explains that feelings affect the species’ ability to make flexible choices(not have all their behaviour hard-wired genetically) and thus determines whether they survive evolutionary competition.
Marie, yes: the costs (in suffering) of false negatives greatly outweigh the costs of false positives (according to our mirror-neurons, which are not our money-neurons).
ReplyDeleteKey argues that fish can’t feel pain based on their lack of frontal cortex, a region implicated in pain sensation in humans. I find it strange that Key argues that we must identify pain-sensitive regions based on correlation in humans, but then goes on to say that we have to causally manipulate in other animals. Since his argument is predicated on reports from humans, this seems like a major oversight. In the responses, the one that stuck out the most to me was that from Haikonen, as he seems to hit the nail on the head with where Key’s argument falls short. The idea that, because we lack more direct methods for causally testing the neural basis of pain, we must agree that correlation is sufficient, is not enough. Moreover, to selectively identify activation in one region and link that to pain, when there is in fact neural modulation occurring globally when an animal feels pain, seems biased. I also take issue with the idea that a single brain region can be fully implicated in the sensation of pain. Increasingly, neuroscientific research is moving away from a circuit-specific view of the brain, and instead appreciating how the interconnected nature of the brain makes it nearly impossible to have functional and activational segregation, even in highly conserved regions like motor and visual cortex. To hinge the entire argument of fish feeling (or not) pain on the presumption that a single region or population can be implicated is spurious, and undermines both Key’s argument and some of the responses (Dinets, Panskepp, Burghardt).
ReplyDeleteKey argues that fish cannot feel pain because they lack the cortical structure. However, even in his paper there are clues that this may not be the case. He mentions the hypothesis that fish feel “fish pain” not “human pain”, meaning they may very well experience negative emotions equivalent to ours but involving a different neural pathway and experience. There was also a case of cortical lesioning that did not remove pain, opening the possibility that pain is caused in the brainstem (which fish possess). Furthermore, in his article, Broom lists the mental representations and avoidance behaviours that fish exhibit. Such behaviour is most logically explained by a pain-like experience. We can not say with absolute certainty that fish feel pain, but the evidence (and our mirror neurons) is significant. I believe that (at the very least) we are not in a position to conclude that fish do not feel pain.
ReplyDeleteAdditionally, my own thoughts are that pain has an enormous evolutionary advantage to living creatures. Humans who cannot feel pain struggle significantly. Pain is essential for knowing when to rest and heal oneself, and knowing when a situation is putting you at risk of death. A creature that cannot feel pain seems more likely not to survive natural selection. Therefore, it seems quite likely to me that all animals should feel some form of pain. (The same function has previously been found to have evolved in different ways for different species: plant photoreceptors vs. eyes; one chambered hearts vs. four chambered; and skeletons vs. exoskeletons. This may be applicable to fish pain vs human pain).
ReplyDeleteHi Nico, I think you might be conflating pain with something more like states of distress when you say that pain has an evolutionary advantage (but even calling it a "state" of distress might be me anthropomorphizing). Distress is useful and necessary for learning what is an aversive stimuli in animals and plants, but I think it would be hard, but maybe not impossible, to call that "pain" in plants, because it's not accompanied by a subjective experience, it's just an aversion to something noxious. In humans, aversion to noxious stimuli is important too. Maybe it's the case that subjective experience of pain is necessary to learn that aversion, but it doesn't seem necessary in plants. Aaand back again to the HP -- why is an experience of pain necessary?
DeleteElliot, good points, but the HP is much harder than you think. Detecting noxious stimulation and reacting accordingly) is essential, but feeling it? Answer that and you've solve the HP.
DeleteBrian Key's article theorized that fish cannot feel pain because they lack necessary cortical structures that would make them feel pain. Brian talks about Qualia to explain pain in the beginning of his paper. As we have seen in class, Qualia is just a weasel word to say sentience. Does Brian believe that fish can be sentient at all? Although he defends his arguments rather well, he also tends to underestimate the Other Mind Problem and not really take it into account as valid criticism. One thing that is for sure is that Brian does not bring us closer to solving the Hard Problem. He says that some structures give rise to pain (due to the cortex amplifying signals). But HOW is it that signal amplification MAKES us feel pain. How are we able to FEEL pain? This is the hard problem. A problem that he doesn't consider really in his piece.
ReplyDeleteHi Lili, to answer your question regarding whether Brian believes that fish can be sentient at all, I believe that Key doesn't completely rule out the possibility of fish not being sentient. I read Key's article a couple of days ago, however as far as I remember, he doesn't rule out the possibility that there might be other cortical structures that enable fish to feel pain. He just highlights that there is a lack of evidence for this, so I would say that he does believe that fish can be sentient.
DeleteKey waffles on the question of whether fish are sentient (i.e., whether they feel anything at all). In view of the importance of detecting noxious stimuli, it is hard to imagine that they feel, but don't feel pain.
DeleteBut, no, the question of whether fish feel -- pain or anything -- is the OMP, and cannot demand solving the HP if it does not demand it in other species, including ours.
Simply put: Key argues that fish lack the cortical architecture necessary and sufficient to create the feeling of pain in humans and so they must not feel pain. I believe his arguments are thoroughly dismantled in the series of responses for which we read the abstracts and further dismantled in these very skywritings. The largest issue that I took with his article has also been explored thoroughly here: that he fails to provide good reasons for discounting the idea that fish may feel pain differently than humans. Because of the other minds problem, what it feels like to be a fish feeling pain is obviously unfathomable to us (as is imagining what it's like when anyone other than myself feels pain) so if by pain you mean "the feeling I feel when I'm in pain" then no, fish don't feel that, nor do any other humans for that matter (not necessarily anyway). However, Key goes one step further by saying that: "If 'fish pain' existed, it would need to be consciously processed using at least the minimal neural architecture described above". I find this problematic since (as far as I'm aware) we don't know the neural correlates of "consciousness" (feeling), at least we don't understand them well enough to deny feeling to those entities that lack those handpicked neural correlates yet seem to display every behaviour inherent to the feeling in question (pain).
ReplyDeleteIn Animal Sentience: The Other Minds Problem, Harnad states that “thanks to Descartes, I cannot be sceptical about the fact that I feel, but I can certainly still be sceptical about whether anyone else feels”, and this is the Other Minds Problem manifested. However, Harnad stresses that, while we can’t be absolutely sure whether other animals feel, we can certainly be sure enough. This sense of sure enough should be sufficient to give other species the benefit of the doubt, and to thus ascribe to them the ability to feel, for “there is incomparably more at stake for others if we assume that they don’t feel, yet they do, than if we assume that they do feel, yet they don’t.” This paper by Brian Key is a perfect demonstration of the point that Harnad is making. Key concludes that “fish clearly respond to noxious stimuli” but “these behaviors… are executed in the absence of feeling pain.” He claims that “the evidence best supports the idea that they [noxious stimuli] don’t feel like anything to a fish.” Sure, the evidence may “best support this conclusion”, but in the (even unlikely) event that they do feel pain even though we assume they don’t, the experiments Key cited and/or conducted in his paper become very disturbing. Key’s approach to supporting his claim that fish, and other non-mammalian species, don’t feel pain or anything at all, is a perfect example of what not to do when experimenting on other species. Even though we assume that they don’t feel, it is impossible to be certain of this due to the Other Minds Problem, and so it does them an immense kindness to grant them the benefit of the doubt and assume that they do feel (however unlikely).
ReplyDeleteWhat caught my attention in this paper is the irony between Key's claim that fish do not feel pain and his position on qualia; they're very conflicting. It makes sense to say, based on the empirical evidence that fish can't feel pain. After all they do not possess the neuroanatomical structures necessary for experiencing pain in the same way that mammals and birds do, he argues that they cannot feel pain. So what do they feel? Key considers alternative explanations for fish behavior that may be mistaken as pain responses, where he suggests that certain behaviors exhibited by fish in response to harmful stimuli may have evolved as protective mechanisms rather than indicators of pain. The very sheer irony in this claim is that how do we know? I remember when I was first introduced to qualia, it was described as what we feel, our unique perspectives to the experiences that surround us. It is the raw, subjective feeling or sensation that accompanies our mental states, such as the sensation of seeing a color, feeling pain, or experiencing a taste. We know that qualia are often considered to be inherently private and cannot be directly accessed or shared with others. How can we confidently say that fish cannot feel pain based on what WE observe externally? I feel like a lack of the neural mechanism that is required to feel pain isn't a condition to completely omit the subjectivity of the perception of pain
ReplyDeleteHey Mal,
DeleteThis is a great point that I didn't consider in my first read – the fact that he goes out of his way to define qualia but doesn’t link it to his argument is quite strange. I suppose it makes sense, given that it seems contrary to his argument, that he wouldn’t point this out, but if he agrees that qualia exist (at least for humans) and that other animals (vertebrates) experience pain, it seems a rather arbitrary line to draw at neuroanatomy as the delineating factor of the animals for which pain does and doesn’t occur. I also find it surprising that he would not address the possibility of a gradient of pain perception – it seems like he conceptualizes it as a binary capacity that one does or doesn’t possess, due to the presence or lack of nociceptors. But the capacity for nociception evolved in vertebrates, so it seems to be like there’s no reason not to believe that there is some neuroanatomical equipment that fish have that could supplement a pain that, while qualitatively different than our own, still exists. It seems like the OMP taken to its very extreme.
Hi Madeleine
DeleteExactly!! I feel like it completely dismisses that since we haven't been able to objectively research how and why us humans feel ourselves, it is not fair to assume that other species don't possess different mechanisms that bring up their sentience. It really points out the flaw in introspection, reiterating that this is exactly why we cannot employ it to solve the HP!!
It wasn't until 1987 that the American Academy of Pediatrics formally declared it unethical to operate on newborns without anesthetics. The mainstream belief was that in babies the expression of pain was reflexive and, owing to the immaturity of the infant brain, the pain could not really matter. A major change in practice was brought about by events surrounding one operation. Infant Jeffrey Lawson underwent open heart surgery in 1985. His mother subsequently discovered that he had been operated on without any anesthesia, other than a muscle relaxant ; he obviously didn’t survive for long. My main point is that it shows how easy it is to take advantage of our lack of solution to the OMP to inflict harm to other individuals on the ground of underdetermined scientific assumptions. Key's argument relies on this assumption that fish do not feel pain due to their lack of specific neural structures that are essential for pain perception in humans. While it is evident that fish react to harmful stimuli and exhibit behaviors aimed at avoiding such stimuli, Key argues that these responses are merely reflexive (automatic motor responses) and do not imply any feeling of pain. This story sounds familiar to the one mentioned above…Going back to week 4, the sole process of neuroplasticity showed us that strong equivalence is not the best way to go about the brain ; different inputs might lead to the same output (especially for different species like fish).
ReplyDeleteHaving taken PSYC302 last semester - which went on extensively about the experiences of pain in humans - we learned that even in the field of neurobiology we are not confident in the HOW and WHY of pain (especially chronic pain). I know this is more of a HP issue than an OMP one but I think that’s what we would need to focus on most right now instead of inferring rushed and underdetermined conclusions about other species based upon our own machinery.
As mentioned earlier, I also agree that Key’s argument about fish comes across as robotic, and one of the responses in Antonio 2016 did in fact mention such robotic fish in support of Key’s argument. The idea presented is that you can create a robot fish for which sentience is not necessary as they are able to do all the functions of a real fish. They are showing that it remains difficult to explain the causal role of any form of fish sentience in the robotic capacities. All this really is however is a reiteration of the ability to solve the easy problem with a turing machine but not the hard problem. Yet even Searle who similarly argues that such a machine does not feel, does not argue that humans cannot feel. Thus the argument presented by Antonio is not relevant nor supportive of an argument that fish do not feel, but rather demonstrates that we do not know how or why they feel. Hart on the other hand presents another view in that we do not know what the state of feeling pain is exactly like for others (OMP), but we have enough similarity and language abilities with other humans to what theistates are like. Building off of this however, simply establishing fish sentience may not be sufficient to prevent unethical treatment of animals. Humans sometimes display a lack of empathy to fellow humans from characteristics such as inflexibility or a lack of perspective taking and high self centeredness. Thus for the goal of animal welfare it could be more effective to look into how to enhance such abilities and qualities in humans.
ReplyDeleteYes, you can make a "T3" fish robot, but that would no more solve the OMP in fish than it does in humans. Does F3 or F4 give probability enough? It would if it were Turing Equivalent as well as Turing Indistinguishable from a fish, but who's to judge? Other fish? They can be fooled with just a fish-lure...
DeleteAs some have mentioned, it's widely acknowledged that there's not enough neurobiological evidence to definitively state that fish do not experience pain. Given the lack of proof, I believe in a fair approach, suggesting that, until extensively proven otherwise, it's reasonable to assume that fish do feel pain. However, Key says that this is problematic as ‘accepting such an assumption at “face value” in biology can lead to devastating consequences.’ He then gives rather specific situations to illustrate his point, ‘It could also lead to legislative restrictions on fish-related activities with potentially serious negative implications for native subsistence fishing (Capistrano and Charles, 2012), human nutrition and food supply (Tacon and Metian, 2013), and economic development (Tveteras et al., 2012).’ I believe Key overlooks the fact that indigenous subsistence fishing and other survival-driven fishing methods shouldn't be a basis for dismissing the idea that fish can feel pain. I think it's essential not to rationalize actions that cause harm to animals by outright denying their capacity for pain.
ReplyDeleteAnd subsistance cultures are exempt because for them consuming fish is vitally necessary..
DeleteIn “Why fish do not feel pain” (2016) Key specifically argues in this paper that for structural reasons, fish are not capable of feeling pain. Through this, he offers an examination of fish neural structure and escape behaviors, fuelling more discussion in regards to the “Other Minds Problem” (OMP). In particular, he emphasizes that in comparison to human neural anatomy, the neural circuitry of fish has limitations. In other words, they lack the organization required for complex neural processing— for example, there is an absence of layering in the pallium and relevant neural areas are only broadly connected. As well, he states that fish do not have a frontal cortex, a key region for feeling.
ReplyDeleteKey also highlights the escape behavior experiments that are commonly studied in neuroscience labs. It is typical for scientists to associate certain behaviors they observe to feelings— when a fish is shocked in the presence of light, it learns to swim away whenever it sees the light; researchers then form the assumption that this means the fish feels fear. As a result, Key argues that fish do not demonstrate this behavior out of “feeling.” Instead, in his words, it is because “the circuitry has evolved to ensure survival” (Key, 2016).
Several research papers have argued against Key, including Vladimir Dinets’ paper “No cortex, no cry,” which I found had some valuable insights on Key’s argument and topics related to the OMP. Specifically, Dinets criticizes Key’s approach of a comparative analysis between structures in humans and fish (Dinets, 2016). He comments on the disadvantages of using model organisms in scientific research— what is the probability of feeling what a fish feels, and how can we assume this simply based on differences in anatomy? At what point do comparisons not work? Similar questions could also be investigated in relation to other model organisms in science; research has heavily relied on other living organisms, such as birds, rats, and drosophila, as a way to study and link what they observe to humans. Comparing across many varying species adds more complex variables to studies. Therefore, to what extent is this information reliable, given the OMP? What are the limitations in regards to this form of research?
Michelle, see the many other skywritings and replies on these questions.
DeleteThis text highlights that people tend to mistakenly use animal reflex behaviours as their signs of pain, or some of their unintended moves are assumed they are feeling pain. Exploring animals' moods and emotions might still be complex since we study their behaviours based on our current knowledge and experience. For example, when feeling pain, humans tend to yell or escape. So when fish shake their tails or some painful-look moves, we conclude they are feeling pain, which is not convincing yet. This reminds me of an experimental video I watched a long time ago that showed that when rats fight, they might freeze for ten minutes in the middle of the fight. Few experimenters discussed that those rats might communicate in a shallow voice that humans can't hear or realize. And when they fight, even if it is very intense, some do not even show any painful-look moves, so does it mean they are not feeling pain?
ReplyDeleteHuman combatants (and athletes) often do not feel pain during combat and competition.
DeleteThere are many issues within Key’s argument. The flaws entail his idea of certain animals not feeling pain; however, he doesn't argue that fish don’t feel anything at all, just specifically not pain. I found this very interesting because pain is the one thing I thought was fundamental for most creatures to experience. He attempts to argue that since it is not the same as the way that the human brain perceives pain, they must not do it, but that doesn't make sense as each species is so different. For comparison, just because we may not have the same vocal chords or speech production as other animals doesn’t mean they don't communicate, it just means they do it in a different way than humans. Evolution may have caused the birds and fish to evolve in a way different than humans that results in them feeling pain differently. While his argument flow seems convincing at first glance, it is a constant reminder to always inspect and criticize an argument.This also brings to light that we still do not know the true mechanism and ways that we as humans feel or as other animals feels in its entirety and it is important as Key does incorrectly, to continue to speculate on the hard problem of finding a causal explanation to feeling.
ReplyDeleteKey argues that fish lack the neural structures that lead to feelings of pain in other mammals, and because of this, they are unable to feel pain. I found Manzotti’s commentary very insightful. He explains that because fish and humans are so different, it is not productive to use correlational data when comparing them. There is not enough evidence to prove that neural activity alone is sufficient to generate pain even in mammals, so how are we able to apply that confidently to fish? While neural activity is shown to be involved, there are other components involved as well, such as the environment. Because of this, it is not logical to jump to the conclusion that fish cannot feel pain at all. The very mechanisms that lead to the feeling of pain in other species are still not fully understood, so why is Key so sure fish don’t feel it?
ReplyDeleteMegan, it is highly unlikely that things outside the body are part of the mechanism that produces the feeling of pain, but they are certainly part of what cases the bodily damage that activates the mechanism that produces the feeling of pain. On the other hand, it is likely that the mechanism that produces the feeling of pain is not (just) local but distributed (within the central and peripheral nervous system). And of course correlations -- both behavioral and neural -- are the only evidence we have; what has to be taken into account is that both can differ across species.
DeleteThe best refutation I read was "No cortex no cry" illustrating beautifully how the cortex supremacy put in the central article is futile, illustrating it with examples such as manta ray behaviour, passing the mirror test, etc.
ReplyDeleteAlthough I wonder how else we can draw lines for feeling throughout the tree of life, octopodes for one seem to exibit clear capacities for feeling, even with complex behaviors such as grudges against specific humans, and vertebrate's common ancestor with them is a primitive flatworm 750 million years ago! did feeling evolve multiple times? or was the flatworm itself already feeling? Then Many gut parasites would be feeling creatures.
If "Dr. Fish" were to author a research paper in the field of "Anthropology" with the assertion, "Unlike us fish, humans are unable to swim due to the fact that human offspring fail to utilize their tails for propulsion in a laboratory setting, leading me to believe they lack the fundamental aspect of swimming. Moreover, their respiratory systems are not adapted to function underwater," we might find such a statement amusing coming from "Dr. Fish".
ReplyDelete
DeleteI perceive certain parallels between the works of "Dr. Fish" and Dr. Key. One of the challenges in cross-species research is accurately measuring and interpreting data across different organisms. Dr. Key attempted to disregard extraneous factors, focusing solely on the stimulus-response relationship within a controlled laboratory setting. However, he was unable to obtain conclusive evidence to refute the existence of feelings beyond simple stimulus-response. Therefore, in order to definitively disprove the notion that fish are incapable of feeling, we must address and refute each aspect of their experience, rather than just the basic stimulus-response dynamicm, which remains as a mystery and hard to quantify for sure.
Jiajun, good points, but the fact that swimming (DOing) is observable and pain (FEELing) is not makes the OMP harder than the rest of the EP. (But that is still not the HP.)
DeleteI found that Ng, Yew-Kwang (2016) work called ‘Could fish feel pain? A wider perspective’ summarized a major issue with Key’s conclusion with this sentence, “True, the feeling of pain in humans may involve the several factors and the corresponding neuroanatomical features discussed by Key that fish do not possess. However, we cannot conclude that a jet plane cannot fly because it does not have a propeller”(Ng, 1). When considering Key’s conclusion from this perspective it is clear how ridiculous it sounds. Ng goes on to explain other oversights that Key made, making this paper an excellent summary and criticism of Key’s work
ReplyDeleteMiriam, I think your comment is very interesting and I thought exactly of that as well. Yes, dogs wag their tails when they feel happy. And that is a feeling. Prof, I do not fully understand what you mean by saying that the intention need to be felt? Again, if dogs flatten their ears because they are scared, they do feel they are scared. They wouldn't do it if there was no feelings involved.
ReplyDeleteMarine, some comments were erased by blogger bugs, so I don't know the context of the reply about intention. No doubt that organisms can communicate intentionally. But the intention (volition) would need to be felt, otherwise it could be an automated reflex in an insentient organism.
DeleteI quite enjoyed Derbyshire’s refutation “Fish lack the brains and the psychology for pain”. He argues that “pain” itself occurs in subcortical areas, it is the “conscious” experience that occurs in the cortex. To feel pain (“ouch”) and to feel as though one is feeling pain (“I feel pain”) are a subtle distinction. Ultimately, the former matters more as it is the primary negative experience. We cannot know for sure if fish experience the latter (they likely do not as this occurs in the cortex), but it makes no ethical difference as it does not stop them from feeling pain. It may matter for the HP, but both the former and the latter are feelings, so either is relevant for the HP.
ReplyDeleteNico, Derbyshire is weaselling: There is no such thing as "unconscious pain experience." Un-weaselled, that's "unfelt pain-feeling." The hoary philosophical canard from which this kind of thinking originates is the distinction between "feeling pain" vs, "being aware of feeling pain": De-weaselled that becomes "feeling pain" vs, "feeling feeling pain"...
DeleteI think this paper is simply an attempt to justify what we as humans inflict on animals. There is a confirmation bias in this and a lot of scientific papers that what we do is okay because they are simpler, “less advanced” living beings. Mr. Brian here falsely operates on the assumption that we know what brain structures and mechanisms that are necessary to feel pain, or feel in general, but really this is the HP. I think this same cognitive dissonance is what allows people to “discipline” animals, ignore what they are eating, deliberately run over animals like squirrels or harm them in other ways because we have as a society designated them as “below us” . My personal belief is that they are living beings who also require autonomy to flourish (and that means not being born to die at the hands of industrial agriculture or the other forces of evil).
ReplyDeleteKaitlin, I agree with you. It shouldn't be the case where we compare an animals's brain anatomy to ours and conclude that what they lack means that they do not feel pain or feel other things that we humans do. The paper, while convincing, seemed rushed if that makes sense. The research didn't seem fish based but rather human based. That makes the introduction of the paper misleading. I'm glad to see that we agree :)
DeleteOnly humans can report feeling pain. *** wrong**. In contrast, pain in animals is typically inferred on the basis of nonverbal behavior ** also wrong, animals cry out** . Unfortunately, these behavioral data can be problematic when the reliability and validity of the behavioral tests are questionable** also wrong, we use the same sort of behavioral data to examine human pain, also arguably human verbal pain questionnaires can lead to more bias** . The thesis proposed here is based on the bioengineering principle that structure determines function. Basic functional homologies can be mapped to structural homologies across a broad spectrum of vertebrate species. For example, olfaction depends on olfactory glomeruli in the olfactory bulbs of the forebrain, visual orientation responses depend on the laminated optic tectum in the midbrain, and locomotion depends on pattern generators in the spinal cord throughout vertebrate phylogeny, from fish to humans.** different systems have different goals and they're built to reflect that** Here I delineate the region of the human brain that is directly responsible for feeling painful stimuli ** the one region? Pain is represented all throughout the central and peripheral nervous system, and circumstantially it is represented in different areas**. The principal structural features of this region are identified and then used as biomarkers to infer whether fish are, at least, anatomically capable of feeling pain. Using this strategy, I conclude that fish lack the necessary neuro cytoarchitecture, microcircuitry, and structural connectivity for the neural processing required for feeling pain.
ReplyDeleteKate, it's a little hard to discern what you are meaning to affirm and deny here, but I assume you are not endorsing the concluding sentence.
DeleteThe way we wouldn’t use research on a fish’s brain to apply absolutely to humans, it’s silly to try to apply human brain research absolutely to a fish. Just because some people believe humans are the hierarchically superior species, it’s more realistic to understand that it isn’t an evolutionary hierarchy but a difference in evolutionary paths. Key’s belief that fish have survived this long in evolution without the ability to feel pain feels like a just-so story. If you tell yourself that certain species evolved just for the consumption and use by “more superior” species you can trick yourself into feeling like you’re correct (like Prof Harnad commented earlier-> cognitive dissonance!) After reading more comments I see other researchers call it the parallel vs. convergent evolution ("multiple realizability") argument.
ReplyDeleteAfter reading the ChatGPT comment, I really like the Precautionary Principle and rethinking the null hypothesis on animal sentience. If we assume sentience, we take more care about what we choose to do with these animals and will lead to less “learning the hard way” in the future.
“Key (2016) claims that fish do not feel pain because they lack the neural structures that have a contingent causal role in generating and feeling pain in mammals. I counterargue that no conclusive evidence supports the sufficiency of any mammalian neural structure to produce pain. We cannot move from contingent necessity in mammals to necessity in every organism.” (Manzotti, Riccardo (2016))
ReplyDeleteI will refrain from summarizing Key’s points as my colleagues have already done so with elegance. From all the counterarguments I have read, Manzotti seems to sum up the main fallacy of Key’s argument most concisely. He emphasizes the logical fallacy in extending a principle observed in mammals (contingent necessity) to all organisms, suggesting that just because something is necessary for a particular function in one group (like mammals), it doesn't mean it is necessary for the same function in a different group (like fish). This idea is simple yet bulletproof as a hole-maker in Key’s claims. As it stands, the reasoning seems not only to be based on the limitations of our function neurobiological understandings, but also, of course, the OMP.
Gabe, "necessity" is ambiguous here because of the HP (nobody has a clue as to how and why feeling is necessary at all). "Contingent necessity" is a bit of an oxymoron too: "accidental necessity"?
DeleteNothing is "bulletproof" here; not just because of OMP but also HP (and multiple realizability).
OMP in between humans is a convolution already. I can understand the feeling of not wishing to apply it to other species, whether for personal interests or to lessen the guilt when performing ruthless experiments on them. However, if successfully proven that animals like fish do not have feelings, I only see little help in using fish as experimental targets to try solving the HP. Because if one species does not have feelings, studying its cognition would be very different from studying human cognition, not that it cannot evolve into its own field, just not very useful when reverse-engineering human cognition.
ReplyDeleteTina, the issue here is not HP but OMP, and ethics.
DeleteBefore reading this text, in my opinion, to identify the pain from fish, can be explored from these directions: anatomy, neuroscience, physiology, and pharmacology, which are all structural applied sciences. However, after reading the text, it started to get a connection with philosophy and cognitive science. We can't directly experience a fish's inner state, but the convergence of evidence from multiple disciplines supports the conclusion that fish are capable of experiencing pain and distress. Their responses serve an adaptive purpose, and engaging analgesic mechanisms provide indirect validation. This leads to the study of subjective experience.
ReplyDeleteOne argument that Key made in his 2016 paper arguing against the possibility that fish can feel pain is that it cannot be proven that fish are conscious (or feeling, if we're de-weaseling). But this argument is flimsy at best because you cannot prove that another person is feeling. This is the OMP, which applies to other human beings as much as it applies to fish. I am surprised that using a lack of evidence for "consciousness" in fish as grounds that they cannot feel pain didn't generate any kind of cognitive dissonance here. The reason why we do not hurt other people is because we are giving them the "benefit of the doubt", the same benefit of the doubt Key accuses people of giving fish. Edelman's argument for evolutionary convergence, along with Damasio's argument that the brain is not the sole structure responsible for the subjective feeling of pain in humans were the most compelling to me, in the replies to the original (intentionally?) incendiary paper.
ReplyDeleteThe paper was a frustrating journey, claiming that fish do not feel pain since they do not have the similar nervous system organization that we do. This clearly shows the OMP in the context of nonhuman animals, which allows the constant chain of animal suffering that humans impose, with the help cognitive dissonance we have when thinking about animals, pain, and suffering. The justifications provided in the paper just reminds of Descartes’ “animal machines”. As a vegan, it is very disappointing to see this cycle in daily life. People tend to perpetuate this idea for insects as well: for example, lots of people do not think twice when killing a spider in their house, often backed up with the statement “They don’t feel pain”. However, personally, even the act of a spider running away to avoid getting hurt is sufficient for me to avoid killing any living organism. Pain, its implications, and its evolutionary mechanisms (or means) might be different for fish or spiders, but ultimately suffering from pain is the same for all animals. In addition, (I apologize if this is a bit off-topic) I had a seminar about the ethics of animal research, and one of the readings argued that pain caused to animals might be experienced even worse compared to humans, since we usually have an idea about the cause of the pain and might know when it will end. Yet, with animals (both in terms of research and daily exploitation for food), they don’t know when the inflicted pain might end. Overall, Key’s arguments show how the OMP might fuel people’s cognitive dissonance about animal suffering, providing justification for the constant pain that humans impose on animals.
ReplyDeleteThanks to the authors for neurobiologically proving the point that fish can't feel human pain, rather than stating it as a likely fact without any scientific evidence to support it. As the first sentence of this article says "only humans can report feeling pain", another problem with the idea at this point is not between individual humans, but between species and species. Since most people will tend to anthropomorphize when thinking about other minds of other species, does this mean that people will take a similar approach, i.e., self-anthropomorphizing, when thinking about other individuals' OMP problems?
ReplyDeleteHi Jinyu, I have trouble understanding what you mean by self-anthropomorphizing. In terms of whether or not people use an anthropomorphic approach when thinking about the OMP for other individuals, I think it comes down to our responsibility to decenter human perspectives. Especially since we have a more difficult time empathizing with other species, we must be careful not to discount their feelings. The challenge of the OMP may lead individuals to project their own feelings onto others (maybe this is what you were implying about self-anthropomorphizing) because what we feel is the only point of certainty that guides how we feel like we know (understand) about others.
DeleteWriting this prompted a separate thought for me though. Although anthropomorphism can be harmful, sometimes personifying others helps us to empathize with other species more. For instance, if you name a chicken George, there is an immediate emotional attachment to him (even though chickens do not give each other names).
The idea that different animals may express feelings such as pain differently is a very refreshing and introspective one. Previously, I had always naively assumed that mammalian sensory/emotional expression was universal, like universal grammar, innate and experienced the same feelings. This article made me realize that fish pain may be accompanied by different expressions or coping mechanisms, and that it was me who had unconsciously anthropomorphized fish pain before. Since pain is a default possible universal feeling, I wonder if the expression that chimps possess for pain and the coping mechanisms are actually very different from humans.
ReplyDeleteKey’s paper's main argument is that, unlike humans, fish lack the specific neural structures in their brains that are essential for the conscious experience of pain. He draws this from a comparison of the structural and functional aspects of the human brain with that of fish, focusing on the cerebral cortex's role in pain perception. I can’t find myself agreeing with Key’s on this point as it seems to make intuitive sense that different species may have developed different mechanisms to do similar things. “Lack of neocortex does not imply fish cannot feel pain" by Georg Striedter supports my intuition by refuting Key’s claim by arguing that different species have evolved various neural mechanisms to achieve similar behaviours. Also disputing Key's argument that non-mammalian vertebrates do not perform behaviors driven by expected rewards or punishments. Pointing out that classical and instrumental conditioning, including learned avoidance behaviors, are observed in fish, suggesting they might experience pain.
ReplyDeleteI also agreed with Striedter's commentary against Key! He explained that simply saying that fish do not feel pain because they do not have a cortex is not a fair conclusion to come to. He proposes that there is a lot of evidence that all points towards the conclusion that just because fish do not have a cortex, it could be very possible that they have evolved a different set of neural circuits that could still produce the feeling of pain. An important point he brought up is that scientists who either believe that fish feel pain or scientists who do are looking at the same data. This goes to show how prior beliefs can influence the interpretation of data, therefore allowing biassed conclusions to be drawn.
DeleteIn this text by Key, he argues that fish do not have the necessary structures to feel pain as humans do. They display a variety of other behaviors that may seem like they feel pain but Key explains that these behaviors are simply reflexive in nature and that there is no feeling of pain for the fish. The main connection that this article relates to our study of cognitive science is in the Other Minds Problem. So far, we have only addressed the OMP in other humans, but in this case, the OMP can also be applied across species to fish. Where Key makes the argument that fish do not feel pain, the real question is related to how we could possibly know whether fish feel pain or not. It’s about knowing whether they can or can’t feel pain. Obviously, since we are not fish, we have no way of knowing whether fish are able to feel pain or not. That is the main question posed here and it is related solely to the OMP.
ReplyDeleteHyeokbin, "knowing" is a weasel-word here: If it means certainty, we don't know whether other humans feel pain either. OMP. Moreover, humans cannot Turing-Test fish, whether F2, F3, or F4. And fish cannot T-Test one another either. (Why?)
DeleteBefore going into the reading my first thought was “Oh then fish just don’t have nociceptors, cool.” Then I started reading and saw that there is much MUCH more to feeling pain than just nociceptors and it was very humbling. While the conclusion of fish not feeling pain like humans is logical when followed from the fact that they do not have the required brain architecture, the craniotomy example isn’t as convincing to me. Mostly because I think there’s more to social and physical functioning diminishment than just the post-surgery pain. It seems to have a psychological level as well. I don’t think that fish being able to continue their mating and swimming doesn’t necessarily prove that they do not feel pain.
ReplyDeleteI also read Fish Pain: A Painful Topic by Carl Safina. Safina’s argument about defense mechanisms for survival suggesting that fish might be feeling a different kind of pain and to defend themselves against hurtful stimuli they have such defense mechanisms makes sense to me. Of course, this paper only relies on logic and adaptive function so it would need to be scientifically proven first. But I do think that it is a good starting point for research on the subject.
In the text, Key explains that fish do not feel pain, since, “fish lack the parcellation of the nervous system into distinct regions with architectures capable of performing pain-related computations”. Key also mentions the ‘fish-pain alternative hypothesis’, where he states that if fishes were to have ‘fish-pain’ they would need a minimal neural structure like humans as described above. With such evidences, Key concludes that since fish lack such architecture, they cannot feel pain. Although, in our point of view, this biological neural mechanism explanation seems to make sense, I cannot fully agree with Key that fish do not feel pain because they lack the same neural architecture that humans have. As others have mentioned in their comments above, I do think that this links back to the OMP. We can certainly observe the behaviors and neural mechanisms of fishes, but this does not directly show their minds or feelings. In our human point of view, their lack of neural architecture and their lack of pain-showing behavior may seem like they are not experiencing pain. But in truth, we do not have the ability to think or feel like a fish. This kind of reminds me of Thomas Nagel’s book, “What is it like to be a bat?”. In the book Nagel argues that even if we learn or know the objective facts about bats, that does not mean we can know what it is like to be a bat, since the bat’s conscious experiences are subjective, which can only be accessed subjectively by being a bat. I think this argument can be applied to fishes too. Since we are humans, who do not have access to fishes’ subjective perspective, it is hard to know if they really feel pain or not.
ReplyDeleteEunsung, fine except for a few weasel-words (which ones?) -- including Nagel's "is like" instead of "feels like".
DeleteI have the feeling that OMP should not be considered as a barrier when considering the sentient animals, or creatures in general. It is true that because of OMP, we cannot absolutely sure that sentient creatures have exactly the same feelings as we humans do, but cannot be the excuse to deny that we do share the feeling to certain degree. For example, you do not know if the orange in Canada is exactly the same as the orange in China, but they should tastes similar to certain degree, and that is why it is also called “orange” – some characteristics from outside is shared, so does the tastes (feeling). Through evolution, sentient creatures may not share a strong equivalence to humans, in terms of every feeling, but there is no need to. Even structural differences exist, but the same or similar outputs could be realized though multiple realizability with different implementations. There should be no question about whether animals do feel as we humans do, and the question could be what could be the boundary to roughly cut the line between feel and non-feel, with more inclusive evidence needed to be considered.
ReplyDeleteWhether or not the neural correlates for fish pain are similar or not to other organisms feelings of pain (ie humans) is besides the point ultimately I feel. Just as computationalism touts implementation independence so to is that point relevant here. Of course there is the point in the paper that they lack the parcellation of the nervous system into distinct regions with architectures capable of performing pain-related computations but as is clear in humans, we do not know why we FEEL pain. We cannot attribute it only to the specific architecture of the nervous system, there is possibility for multiple realizability in bringing about the FEELING of pain. Further, this argument is predicated on the perceived need to justify the mass fish farming that our world is home to at this point, and we are very good at convincing ourselves of things that we want to believe so it is important to actively work against that cognitive bias.
ReplyDeleteJellyfishes do not have a central nervous system but a distributed nervous system. They do not have a brain, therefore they “lack” the different brain regions we understand as essential, for instance, for feeling pain and fear. However, jellyfishes still learn to avoid obstacles and have a flight system. We therefore interpret them as capable of showing feelings, challenging the notion that feelings are expressed the same way in humans and various other animals.
ReplyDeleteMiriam, good point, and maybe jellyfish should be added to the other three substantially uncertain cases (microbes, fungi and plants). But they not only seem to have the behavioral capacities used to infer sentience in other species. Moreover, although they do not have a centralized nervous system, they do have a nervous system.
DeleteI may be misunderstanding the main argument in this article, but I find this unconvincing:
ReplyDelete1. cortical neurons are one of the best proxies of ability for pain, in **mammals**
2. fish do not have similar structures, brain regions, or neurons, such that they don’t have the precise mechanism for pain that humans do
3. therefore, fish don’t feel pain..?
It seems likely, to me, that feeling has survival advantages, in that the computational process involved is described more simply by involving feeling than by doing without (though without would be theoretically possible). In this sense, it seems likely that the outside behaviour fish display when they are injured (I am not assuming, using this word, that they “feel” pain) lead to them experiencing pain, despite the pain mechanism being very different from the one whose most common proxies are somewhat reliable in mammals.
Thomas, you say "the computational process involved is described more simply by involving feeling than by doing without ". Explain that and you've solved the HP: why is detecting not enough, without feeling? And isn't the feeling an added complicationsthat still needs a causal explanation? That's the HP...
DeleteI believe there is a flaw in the initial logic of the argument presented in this text. Key focuses on examining the neural correlates of pain in humans to compare them with those in fish. The claim is that if fish possess subsystems functionally equivalent to those that enable humans to feel pain, we would have grounds to assert that they do indeed experience pain. While I agree with the positive version of this argument, I do not agree with the negative alternative. The failure to identify subsystems in fish functionally equivalent to the pain systems in humans is not sufficient grounds to claim that fish do not feel pain. It is possible that fishes employ an entirely different mechanism to experience pain than humans, but we may not be aware of it because fish cannot communicate that pain to us.
ReplyDelete
ReplyDeleteI was vegan for 7 years and throughout that time, the argument “animals don’t feel” never made sense to me. How could that even make sense? Key argues that by examining the neuroanatomical features of the human brain responsible for feeling pain, we can conclude that fish lack the necessary neurocytoarchitecture, microcircuitry, and structural connectivity for the neural processing required for feeling pain. This, to me, was as good of an argument as “ you need to eat dead animals to get your proteins”. As Jones argues in “Fish sentience and the precautionary principle”, Key's reliance on the "bioengineering principle" that structure determines function is problematic and naive. He points out that different structures can carry out the same function, and that pain in fish may be realizable through multiple different means. Therefore, the absence of certain neuroanatomical structures in fish does not necessarily mean they cannot experience pain. Plus, Key's criticism of anthropomorphism is flawed because it conflates two different senses of anthropomorphism. Jones explains that while unnecessary anthropomorphism involves attributing human traits to beings or objects where it is not necessary, biocentric anthropomorphism involves using human terms to explain animals' phenomenal states. Jones asserts that attributing pain states to fish based on behavioral evidence is a reasonable interpretation, and not an unsupported anthropomorphic conclusion.
As Jones stated, it would just make more sense that given the lack of scientific consensus on fish sentience, it is more reasonable and prudent to apply the precautionary principle and consider the possibility that fish do feel pain.
Amélie, good points and syntheses.
Delete