{"id":30,"date":"2018-08-09T02:10:58","date_gmt":"2018-08-09T01:10:58","guid":{"rendered":"http:\/\/generic.wordpress.soton.ac.uk\/skywritings\/?p=30"},"modified":"2018-08-10T16:04:07","modified_gmt":"2018-08-10T15:04:07","slug":"otherminds","status":"publish","type":"post","link":"https:\/\/generic.wordpress.soton.ac.uk\/skywritings\/2018\/08\/09\/otherminds\/","title":{"rendered":"The other-minds problem in other species"},"content":{"rendered":"

Wood-Gush Memorial Lecture, delivered at International Society of Applied Ethology<\/em><\/a> (ISAE)<\/em> 2018\u00a0Charlottetown Prince Edward Island, July 31 2018.\u00a0<\/span><\/p>\n

The other-minds problem in other species<\/strong><\/h1>\n
<\/div>\n
Stevan Harnad<\/strong><\/div>\n
Universit\u00e9 du Qu\u00e9bec \u00e0 Montr\u00e9al & University of Southampton<\/div>\n
<\/div>\n

\u00a0<\/span><\/p>\n

<\/div>\n

Long Summary: \u00a0<\/strong>Both ethology and comparative psychology have long been focused on behavior. The ethogram, like the reinforcement schedule, is based on what organisms <\/i>do rather than what they <\/i>feel, because what they do can be observed and what they feel cannot.<\/i>\u00a0<\/span><\/p>\n

In philosophy, this is called the \u201cOther-Minds Problem\u201d: The only feelings you can feel are your own. The rest is just behavior, and inference.<\/i>\u00a0 <\/span>Our own species does have one special behavior that can penetrate the other-minds barrier directly (if we are to be believed): We can <\/i>say what\u2019s on our mind, including what we feel. No other species has language. Many can communicate non-verbally, but that\u2019s <\/i>showing, not <\/i>telling.<\/i>\u00a0<\/span>\u00a0 \u00a0 <\/i><\/p>\n

In trying to make do with inferences from behavior, the behavioral sciences have been at pains to avoid \u201canthropomorphism,\u201d which is attributing feelings and knowledge to other species by analogy with how <\/i>we would be feeling if we were doing what they were doing under analogous conditions.<\/i>\u00a0 <\/span>Along with Lloyd Morgan\u2019s Canon \u2013 that we should not make a mentalistic inference when a behavioral one will do \u2013 deliberately abstaining from anthropomorphism has further narrowed inroads on the other-minds problem. We seek “operational measures” to serve as proxies for what our own native mammalian mind-reading capacities tell us is love, hunger, fear.<\/i>\u00a0<\/span>\u00a0 \u00a0 <\/i><\/p>\n

Yet our human mind-reading capacities \u2013 biologically evolved for care-giving to our own progeny as well as for social interactions with our kin and kind \u2013 are really quite acute. That (and not just language) is why we do not rely on ethograms or inferences from operational measures in our relations with one another.<\/i>\u00a0<\/span>\u00a0 \u00a0<\/i><\/p>\n

But what about other species? And <\/i>whose problem is the “Other-Minds Problem\u201d? Philosophers think of it as <\/i>our problem, in making inferences about the minds of our own conspecifics. But when it comes to other species, and in particular our interactions with them, surely it is <\/i>their problem if we misinterpret or fail to detect what or whether they are feeling.<\/i>\u00a0<\/span>\u00a0 \u00a0 <\/i><\/p>\n

We no longer think, as Descartes did, that all nonhuman species are insentient zombies, to do with as we please \u2013 at least not in the case of mammals and birds. But, as I will show, we are still not sure in the case of fish, other lower vertebrates, and invertebrates. Debates rage; and if we are wrong, and they do feel, the ones suffering a monumental problem are trillions and trillions of fish (and other marine species). Ethologists have been using ingenious means to demonstrate (operationally) what to any human who has seen a fish struggling on a hook (or a lobster in a boiling pot) is already patently obvious. For such extreme cases some have urged inverting Morgan\u2019s Canon and instead adopting the \u2018Precautionary Principle\u201d: giving the other minds the benefit of the doubt unless there is proof to the contrary.<\/i>\u00a0<\/span>\u00a0 \u00a0 <\/i><\/p>\n

Yet even with mammals, whose sentience is no longer in doubt, there are problematic cases. Applied Ethology is concerned with balancing the needs of other species with the needs of our own. Everyone agrees that sentient species should not be hurt needlessly. \u00a0 \u00a0 <\/i><\/p>\n

Most people today consider using animals for food or science a matter of vital necessity for our species. (It is not clear how right they are in the case of food, at least in developed economies, and we all know that not everything done in the name of science is vitally necessary — but I won’t call into question either of these premises on this occasion.)<\/i>\u00a0<\/span>\u00a0 \u00a0 <\/i><\/p>\n

Entertainment, however, is clearly not a matter of vital necessity. Yet some forms of entertainment entail hurting other species that are known to be sentient. The rodeo is a prominent example. I will close by describing the results of the analysis of a unique and remarkable database gathered in order to test rodeo practices in Quebec in light of a new law that accords animals the status of \u201csentient beings with biological needs\u201d who may no longer be \u201csubjected to abuse or mistreatment that may affect their health\u201d (except in the food industry and in scientific research).<\/i><\/p>\n

I invite applied ethologists and veterinarians who wish to help protect animals from this to contact me at harnad@uqam.ca<\/a> about helping to analyze this unprecedented database of 3 x 45 hours of continuous video evidence on the 20 rodeos in light of the new sentience law.<\/span><\/em><\/p>\n

It is an honor to be here at the University of Prince Edward Island to give the Wood-Gush Memorial Lecture. I did not know David Wood-Gush, but by all accounts he was a remarkable person, for his personal courage, his productivity, and the affection and admiration he inspired in all. The following words — from one of his last writings before his untimely death in 1992 — go to the heart of the problem that I will be talking to you about today:<\/p>\n

\u00a0<\/span>\u201cIn the late 50s and early 60s profound changes took place in the structure of the poultry industry in the USA: … The age of ‘Big is Beautiful’ had arrived… Sociologically it was not pretty and from the animal welfare point of view it was disastrous.\u201d<\/em><\/p>\n

\"\"<\/p>\n

\u00a0<\/span><\/p>\n

Disastrous for them<\/em>, not for us. And it applied and applies to many, many other species besides chickens. I think David Wood-Gush would have been appalled to see what is documented in Dominion<\/em><\/a>, which has just been released, and which I urge all who care about the well-being of the creatures under our dominion to see, and then to proclaim from the rooftops.<\/p>\n

\u00a0<\/span>Ethology prides itself on being a behavioral science: the science that tries to predict and explain what organisms can and do do<\/em>. My own field, cognitive science, is also a behavioral science. But behavior is what goes on outside the organism, and cognition is what goes on inside. This does not mean that cognitive science is just neuroscience \u2013 although there is indeed cognitive neuroscience, and what is going on inside the organism is indeed neural activity. But cognitive science is wider than neuroscience in that it is trying to discover the internal causal mechanisms that generate organisms\u2019 behavior, and their behavioral capacity: what they do, and how they manage to do it.\u00a0<\/span><\/p>\n

\u00a0<\/span>Those causal mechanisms are undoubtedly neural, but that doesn\u2019t mean that by just peeking and poking in the brain, even with today\u2019s remarkable tools for neuroimaging, we will simply be able to read off the causal mechanisms from the neural data, as we were able to do with the other organs of the body, like the heart, the kidneys or the liver. This is because the brain does, and can do, everything that we<\/em> can do. That is how the computer found its way into cognitive science (thanks to Alan Turing<\/a>). We use the computer to design and test causal models to see whether they can indeed do what the brain can do, and if so, how. That is the method underlying what has been dubbed the \u201cTuring Test.\u201d<\/p>\n

\u00a0<\/span>This pursuit of causal explanation has come to be called the \u201ceasy problem<\/a>\u201d of cognitive science, although it is anything but easy, and its solution is still countless years and Nobel Prizes away. —- But if finding the causal mechanisms that generate our behavioral capacity is the \u201ceasy problem\u201d of cognitive science, what is the \u201chard problem\u201d?<\/p>\n

\u00a0<\/span>To answer this we first have to remind ourselves that behaving is not the only thing that organisms do. They also feel<\/em>, they are sentient \u2013 or at least some of them are. We can observe their behavior, but we can only infer their feelings. In fact, determining whether an organism feels \u2013 and if so, what it feels \u2013 is another problem of cognitive science, not the \u201ceasy\u201d problem of discovering the causal mechanisms that generate behavior, but not yet the hard problem either:\u00a0 <\/span>Determining which<\/em> organisms feel, and if so, what<\/em> they feel, is called the \u201cother-minds problem<\/a>.\u201d The other-minds problem is inherited from Philosophy, but it is anything but a merely \u201cphilosophical problem\u201d (whatever that means).\u00a0<\/span><\/p>\n

The other-minds problem is about the only things that really matter in the universe, either materially or morally.\u00a0<\/span><\/p>\n

\u00a0<\/span>That was quite a mouthful.<\/p>\n

\u00a0<\/span>This talk is about the other-minds problem.<\/p>\n

\u00a0<\/span>But first let me dispel the mystery about what is the \u201chard problem<\/a>\u201d \u2013 the problem that is not<\/em> going to be addressed in this talk. If the easy problem of cognitive science is the problem of explaining, causally, how and why organisms are able to do all the things they are able to do, the hard problem is to explain, causally, how and why organisms are able to feel, rather than just do whatever they are able to do. —- I am sure many of you feel that you have a simple and obvious solution to the hard problem. Let me defer that to the question\/answer period after this talk with just the suggestion that it\u2019s not called the hard problem for nothing. So your solution almost certainly does not work. And the reason for that is, in fact, the \u201ceasy problem,\u201d whose solution looks as if it will make feeling causally superfluous. For once we have fully explained the causal mechanism that generates organisms\u2019 behavioral capacities — their capacity to do all the things they can do (move, sense, remember, learn — even speak) there are no causal degrees of freedom left<\/em> to explain how and why organisms feel, rather than just do whatever needs doing (in order to survive and reproduce).\u00a0<\/span><\/p>\n

\u00a0<\/span>Which brings us back to the other-minds problem. Behavior — doing — is observable, measurable. Feeling is not. So how can we know whether and what another organism feels? —-\u00a0<\/span><\/p>\n

\u00a0<\/span>Let\u2019s start with our own species: The obvious answer is that if we want to find out whether<\/em> another person is feeling something, and what<\/em> they are feeling, we ask. But what about all the species that cannot speak? Or even human infants, or handicapped people who cannot speak or understand? The answer is that with our own species humans have a powerful biological capacity, which has lately been dubbed \u201cmind-reading\u201d — a capacity which might or might not be grounded in the activity of what have lately been dubbed \u201cmirror neurons<\/a>.\u201d\u00a0<\/span><\/p>\n

\u00a0<\/span>Mind-reading is the capacity to detect what others are thinking and feeling without having to ask. Mirror neurons are active both when I<\/em> perform a movement and when you<\/em> perform that same movement. The link is between my seeing you do something, and my knowing what it feels like to do that same thing, to the point of being able to imitate it. No one knows yet how mirror neurons manage to do this, but, if you think about it, we already knew that our brains<\/em> could imitate movements, somehow, because we<\/em> can do it.\u00a0<\/span><\/p>\n

So can other animals that imitate, such as songbirds and parrots<\/a>, as well as the apes, elephants, dolphins and magpies<\/a> who can recognize themselves in the mirror.\u00a0<\/span><\/p>\n

\u00a0<\/span>The obvious link between mirroring doings and mirroring feelings is the link between an observable behavior by another body and the internal state which generates that behavior in my own body.\u00a0<\/span><\/p>\n

\u00a0<\/span>Some have even tried to relate the capacity for empathy to mirror neurons. Empathy is not unique to humans either, as we see across many mammalian and avian species, in the parental care of their young, in pair-bonding and in social behavior. It also makes little difference \u2013 insofar as mind-reading capacity is concerned \u2013 whether social perception is inborn or learned. The critical thing is the link between what the observed animal is doing and what the observing animal is feeling (and eventually doing). In the case of prosocial behaviors<\/a>, the link is positive; I want to help<\/em>. In the case of aggression or fear, the link is negative; I want to attack<\/em>, or flee<\/em>.\u00a0 <\/span>Humans are capable of perceiving the emotions underlying the facial expressions, vocalizations and movements of other humans; they are also capable of feigning them, when they\u2019re not feeling them. That<\/em> may be unique to humans. But the ability to detect and respond to the other\u2019s internal state is not unique to humans.<\/p>\n

\u00a0<\/span>At this point the notion of \u201canthropomorphism\u201d of course immediately comes to mind: Both ethologists and comparative psychologists have been firmly taught to refrain from projecting their own mental states onto nonhuman animals: Abide by Lloyd-Morgan\u2019s Canon<\/a>. Never mind the mind-reading; stick to behavior-reading. Forget about whether or what the animal might be feeling or thinking. Just observe and predict what the animal is doing, with the help of the ecological context, the ethogram, and evolutionary theory. In other words, forget about the other-minds problem and treat animals as if they were highly skilled biorobotic zombies<\/a> — even though you know they aren\u2019t.<\/p>\n

\u00a0<\/span>Do<\/em> we know they aren\u2019t? Of course we do. We know it of our own family animals as surely as we know it of our own preverbal infants. No one invokes Lloyd Morgan\u2019s Canon when it comes to their own babies (although it\u2019s not been that long since surgeons [cf.\u00a0Derbyshire<\/a>] dutifully suppressed their anthropomorphism and operated on babies without anesthesia, just as if they had been hewing to Descartes\u2019 assurances, three centuries earlier, that the howls and struggles of the dog tied to the vivisecting table can be safely ignored, because all non-human creatures are just bio-robotic zombies.)<\/p>\n

\u00a0<\/span>I will return to this. But, thankfully, pediatric surgeons have set aside their compunctions about anthropomorphism in favor of their compunctions about hurting sentient babies; the veterinary profession has done likewise when it comes to operating on mammals and birds. But when it comes to \u201clower vertebrates\u201d such as reptiles, amphibians, fish and invertebrates, we \u2013 or rather they (the victims) \u2013 are still living in Cartesian times.<\/p>\n

\u00a0<\/span>Let\u2019s go back to language for a moment. Language is unique to the human species (although communication certainly isn\u2019t). Human language is clearly the most powerful mind-reading tool of all. It\u2019s close to omnipotent. Any thought a human can think \u2013 and just about any object a human can observe or any experience a human can have \u2013 can be verbalized:\u00a0 <\/span>described in words, in any language. Not described exactly<\/em>, because, like a picture, an experience is always worth more than a thousand words; but close enough. Language is grounded in the experiences we all share, and name.\u00a0<\/span><\/p>\n

\u00a0<\/span>Now the important thing to note is that the relation between words and the objects and experiences to which they refer is also a kind of mirror-neuron function. I could never talk to you about what \u201cred\u201d means if we had not both seen and named red things red. Then, when you talk about something that\u2019s red, I know what you have in mind — as surely as I know what you\u2019re doing when you reach for an out-of-reach apple \u2013 or when you scowl at me. And preverbal infants<\/a> know some of these things too. It\u2019s those adept mirror-neurons again. — What about nonhuman species?<\/p>\n

\u00a0<\/span>Some of you may be familiar with the \u201cSally\/Anne<\/a>\u201d \u201cmind-reading\u201d experiments that are now widely used in human developmental psychological research on children.\u00a0<\/span><\/p>\n

\u00a0<\/span>You may be surprised to learn that the Sally\/Anne experiments were inspired by an article written 40 years ago, entitled \u201cDoes the Chimpanzee Have a Theory of Mind?<\/a>\u201d by a former behaviorist called David Premack. Premack\u2019s article was published in 1978 in the inaugural volume of a new journal called Behavioral and Brain Science<\/em><\/a>s<\/em> (or BBS), of which I happened to be the editor at the time.\u00a0<\/span><\/p>\n

\u00a0<\/span>BBS is a rather influential, multidisciplinary journal specializing in so-called \u201ctarget articles\u201d that have broad implications across multiple fields. The target articles are accepted only on condition that the referees and editor judge them to be important enough to be accorded \u201cOpen Peer Commentary,\u201d which means that as many as a dozen or even two dozen specialists from across disciplines and around the world each write a 1000-word commentary \u2013 a mini-article, really \u2013 supplementing, criticizing or elaborating on the content of the target article.\u00a0 <\/span>Premack\u2019s target article drew a good deal of attention across disciplines and around the world. It elicited 43 commentaries followed by the author\u2019s Response. Premack\u2019s clever series of experiments was the first behavioral evidence that apes can and do infer what other apes — as well as humans \u2013 have in mind.\u00a0<\/span><\/p>\n

\u00a0<\/span>So if apes can and do do mind-reading when they are studying one another, or studying us, why would we feel we have to deny and ignore our own mind-reading capacity as \u201cunscientific anthropomorphism\u201d when we are studying them? Owing to our extremely long period of parental dependency, our extreme sociality and social interdependency, and probably also our long period of co-evolution with domesticated animals, our species is probably the mind-reading champion of the biosphere. And not just because of the unique language capacity that we use to mind-read one another verbally. More likely, our language capacity itself is somehow built on the pre-verbal mind-reading capacity and mirror neurons that we share with other species.<\/p>\n

\u00a0<\/span>So this is a good time to ask ourselves \u2013 especially if we are applied ethologists \u2013 why, when it comes to the minds of other species, we self-impose a methodological agnosticism on the dictates of our evolved mind-reading gifts, ignoring them as unscientific anthropomorphism?\u00a0<\/span><\/p>\n

\u00a0<\/span>It\u2019s not that the other-minds problem is not a real problem. There is no way to know for sure whether any other organism on the planet, including my own conspecifics, feels at all. As already mentioned, they could all be insentient biorobotic zombies, doing but not feeling. There\u2019s no way to prove \u2013 either mathematically or experimentally \u2013 that others feel, rather than just do, any more than there is any way to prove, mathematically, that apples fall to earth because of the pull of the inverse square law of gravity (or, for that matter, whether they will keep falling down rather than up \u2013 or even whether apples or the earth exist at all). But in science we don\u2019t demand that unobservable truths should be provable<\/em> as necessarily true, on pain of contradiction, as in mathematics. We demand only that they should be highly probable<\/em> given the available observational and experimental evidence.<\/p>\n

\u00a0<\/span>Each individual does know (with the help of Descartes\u2019s Cogito) that it is not just probable but certain<\/em> that they are feeling, even though I cannot be sure that any other organism feels. I know I am not a zombie. But it would be colossally improbable that I am the only individual on the planet who feels. Hence we accept the dictates of our mind-reading capacity when it comes to our own species.\u00a0<\/span><\/p>\n

\u00a0<\/span>So we know that other people feel too \u2013 and we often have a pretty good idea of what they are feeling. We also know that our mind-reading skills are far from infallible, even if we are the world champions at mind-reading, not only because people can feign feelings nonverbally or lie about their feelings verbally, but because even in the best of times, our mirror neurons can lead us astray about what\u2019s actually going on in another\u2019s mind. And when it\u2019s people we are mind-reading, we certainly can\u2019t hardly blame our mind-reading failures on anthropomorphism, even when we fail! (Think about it.)<\/p>\n

\u00a0<\/span>So with other humans, when words cannot be trusted, our mind-reading intuitions do need to be tested behaviorally, and sometimes even physiologically (think of lie-detectors, and neural imaging), when the question is about what it is, exactly, that another person is feeling.\u00a0<\/span><\/p>\n

\u00a0<\/span>But when it comes to the question of whether other people are feeling at all, rather than just what they are feeling, the only uncertainty is in cases of complete paralysis, coma or general anesthesia, where there is neither behavioral nor verbal evidence to go by. Then we rely on physiological and neurophysiological correlates of feeling, which have themselves been validated against behavior, including verbal behavior.<\/p>\n

\u00a0<\/span>But we are the only verbal species. What about all the other species? As our test case, let\u2019s focus on a feeling that really matters \u2013 not warm\/cool, light\/dark, loud\/soft, smooth\/bumpy, sweet\/sour, itches or tickles, but hurts — in other words: pain. But bear in mind also that what I am about to say about pain also applies to any other feeling. There is a distinction made in pain theory between \u201cnociception\u201d and \u201cpain.\u201d Nociception is defined as detecting and responding to tissue damage, and pain is defined as what is felt when something hurts. The very same distinction can be made about thermoception versus feeling warmth or photodetection versus seeing light. There is a difference between just responding physiologically to stimulation and feeling something. One is doing, the other is feeling.<\/p>\n

\u00a0<\/span>I\u2019ve already noted that when it comes to mammals and birds we are no longer living in Cartesian times. I don\u2019t know whether it\u2019s really true that Descartes insisted it was fine to operate on conscious animals because they were just reflex robots and their struggles did not mean a thing: they were just doing, not feeling. But he did invoke the other-minds problem \u2013 the genuine uncertainty about whether anything other than oneself feels. The fact is that we cannot know whether anyone else feels: we cannot feel their feelings. We can just see their doings. Descartes was ready to allow human language and reasoning to resolve that uncertainty, with the help of some theology and some pseudo-physiology about the pineal gland. So humans were granted the benefit of the doubt.\u00a0<\/span><\/p>\n

\u00a0<\/span>It has taken centuries more to accord the benefit of the doubt to nonhuman animals. No doubt Darwin had a good deal to do with the change, showing that we too are animals, sharing common lines of descent. But there\u2019s no doubt that anthropomorphism \u2013 our powerful, evolved mind-reading capacity \u2013 played a big role too. Other mammals are just too obviously like us to fail to excite the same mirror neurons that our own young do, for example. And the effect applies to other species too, as we see in the many interspecies adoptions<\/a> burgeoning on youtube. Birds seem to fall Within the \u201creceptive fields\u201d Of those same neurons too; and indeed there are many cases of bird\/mammal and mammal\/bird cross-adoptions.<\/p>\n

\u00a0<\/span>But when it comes to the \u201clower vertebrates\u201d \u2013 reptiles and amphibians, and especially fish \u2013 the Cartesian doubt is still there; and it becomes even more pronounced with invertebrates such as insects, shellfish, snails and clams, despite the celebrated anthropomorphic blip in the special case of the octopus.<\/p>\n

\u00a0<\/span>This is the point to ask ourselves: Why does it matter? What difference does it make whether or not fish feel? The answer is instructive. Why does anything at all matter? Would anything matter in an insentient world of only stars and planets and electrons and photons? Matter to whom?\u00a0 <\/span>Or, for that matter, would anything matter in a world of bio-zombies? Darwinian survival machines who only do, but do not feel? Again, matter to whom?\u00a0<\/span><\/p>\n

\u00a0<\/span>(This may sound fluffy to you, But I ask you to think about it. It really is the essence of the other-minds problem in other species — and of welfare \u201cscience\u201d itself.)<\/p>\n

\u00a0<\/span>In a world with people, we know the answer: what matters is what matters to us. And we do feel.<\/p>\n

\u00a0<\/span>But in the biosphere, where we are not the only sentient species, it matters to them<\/em><\/a> \u2013 to the other species, to the other minds<\/em> \u2013 if we think they don\u2019t<\/em> feel, when in reality they do. The reason I didn\u2019t choose the feeling of warmth or light or itch or tickle as the test case but the feeling of pain was to set the right intuition for our mirror neurons, about what matters and why.\u00a0<\/span><\/p>\n

\u00a0<\/span>Where all other species were in Cartesian times, is where fish still are today. Is it because our anthropomorphism, our mirror neurons, have failed us completely in their case? I don\u2019t think so, any more than I think they failed Descartes in the seventeenth century for mammals and birds. Descartes saw animals\u2019 struggles and felt their suffering, just as we do today. But alongside mind-reading and language, our species has another powerful capacity, and that is the capacity to over-ride our feelings with reasons. And that includes philosophers and scientists:\u00a0<\/span><\/p>\n

\u00a0<\/span>In the seventeenth century, Descartes was in the grip of a theory, his own theory, according to which only human beings could reason, and that reasoning was a god-given gift, along with an immaterial, immortal soul, delivered by god via the pineal gland<\/a>. And all of that somehow implied \u2013 for reasons that I personally find rather obscure \u2013 that therefore only human beings could feel and that all other animals were just reflexive robots: zombies, despite appearances. That \u201cdespite appearances\u201d clause was the denial of the dictates of our mirror neurons, in favor of the dictates of \u201creason,\u201d or rather, in this case, Descartes\u2019 reasoning.\u00a0<\/span><\/p>\n

\u00a0<\/span>Reasoning, however, is a two-edged sword: Only in formal mathematics is reasoning logical deduction, with formal proofs verifiable by all. In other fields it is more like the reasoning that goes on in law courts, where there is the question of the burden of proof (which means the burden of evidence). In criminal law, people are assumed innocent until proven guilty; the burden of proof is on those who seek to reject the hypothesis of innocence, not the hypothesis of guilt. We will come back to this.\u00a0<\/span><\/p>\n

\u00a0<\/span>In science, the story is a bit more complicated. In statistics, as most of you will know, when two populations are being compared on some measure, the \u201cnull hypothesis<\/a>\u201d is that any observed difference between the two populations is just due to chance, unless statistical analysis shows that the observed difference is highly improbable on the basis of chance, in which case the null hypothesis can be rejected. The analysis usually consists of comparing the variability within the populations to the variability between the populations. The burden of proof is on those who seek to reject the null hypothesis. (The first thing to reflect on, then, is What is the null hypothesis when it comes to sentience In other species?<\/em> Is it sentience or insentience?<\/em> Remember that it was language and our evolved mirror neurons that answered this question in the case of our own species, and in the case of the species most like us, mammals and birds. We are moving into more uncertain territory now, for our mirror neurons, in the case of fish, and lower. Most people are still sceptical about the tortoise in the upper left corner.<\/p>\n

\u00a0<\/span>In science in general, as opposed to just statistical inference, it is more a matter of the weight of the evidence: If there are two or more hypotheses, the (provisional) winner is the hypothesis with the most evidence supporting it. But rather as in courts of law, hypotheses have advocates, and besides evidence the advocates can also marshal arguments in their favor. And the arguments often concern what evidence counts, and how much it counts. Methodological arguments; arguments about interpretation, and so on. (It can sometimes remind you of the OJ Simpson Trial or Trump\u2019s Supreme Court.)<\/p>\n

\u00a0<\/span>Let me use the case of fish pain as an example, starting on a personal note: In 2016, 38 years after the appearance of David Premack\u2019s 1978 article<\/a> on mind-reading in chimps in Behavioral and Brain Sciences, Brian Key, an Australian neurobiologist wrote an article in the inaugural issue of a new journal called Animal Sentience<\/em><\/a>. In 2003 I had stepped down after a quarter century as editor of Behavioral and Brain Sciences<\/em>, swearing I would never edit a journal again. After a quarter century of umpiring papers on behavioral and brain function \u2013 both bodily doings, by the way \u2013 in animals and humans, I was becoming increasingly concerned about feelings, that is, sentience, especially the feelings of nonhuman animals.\u00a0<\/span><\/p>\n

\u00a0<\/span>I was not alone. The scientific community was becoming increasingly concerned too, and so was the general public. In 2012 the Cambridge Declaration on Consciousness<\/a> and its august signatories solemnly declared (if it needed declaring) that mammals and birds — as well as lower vertebrates, including fish, and even invertebrates such as the octopus — possessed the neural machinery for consciousness (in other words, sentience), even though it differed from ours. In exactly the same week as the Cambridge Declaration, to commemorate the centenary of the birth of Alan Turing, the greatest contributor to the computational approach to solving the \u201ceasy problem,\u201d I organized a summer school in Montreal on the Evolution and Function of Consciousness<\/a>, at which over 50 cognitive scientists, neuroscientists, ethologists, evolutionary biologists and philosophers addressed the \u201ceasy problem,\u201d the \u201chard problem,\u201d as well as the \u201cother-minds problem.\u201d Among the participants was one of the original commentators on Premack\u2019s paper in 1978, Dan Dennett. [some other notable speakers included: Baron-Cohen, Damasio, Searle, Finlay<\/strong>].<\/p>\n

\u00a0<\/span>Two years later, in 2014, the Manifesto \u201cLes animaux ne sont pas des choses\u201d \u201cAnimals are not things<\/a>\u201d was launched in Quebec, and quickly reached over 50,000 signatures, a larger proportion of Quebec\u2019s population than the French petition in France that had successfully induced changing the status of animals to \u201csentient beings\u201d in France, likewise in 2014. And then in 2015 Quebec<\/a> too changed its Civil Code, according animals the status of \u201csentient beings with biological imperatives.\u201d<\/p>\n

\"\"\u00a0<\/span><\/p>\n

What this entails, no one yet knows. Quebec had till then been Canada\u2019s most backward province insofar as animal welfare was concerned. The Quebec Minister of Agriculture announced that the new law had fast-forwarded the province 200 years. I will be closing my talk with what will be this new law\u2019s first test case: Rodeos.<\/p>\n

\u00a0<\/span>But first, let\u2019s note that Quebec\u2019s new law certainly does not afford any protection to fishes, because it explicitly excludes animals from protection, not on the basis of their species, but on the basis of their uses by humans:<\/p>\n

\u00a0<\/span>\u201c<\/em>Sections 5 and 6 do not apply in the case of agricultural activities, veterinary medicine activities, teaching activities or scientific research activities\u2026 Agricultural activities include, in particular, the slaughter or euthanasia of animals and the use of animals for agricultural purposes\u2026\u201d<\/em><\/p>\n

\u00a0<\/span>Other such laws were being adopted in other countries as well.<\/p>\n

\u00a0<\/span>Then in 2016, a lucky 13 years after I had stepped down from the editorship of BBS and resolved never to edit a journal again, I was invited by a long-time BBS Associate and commentator, the veterinarian Andrew Rowan, President of the Human Society International, to edit a new journal on animal sentience<\/a>, and I accepted to do it immediately, pro bono, and, I would add, gratefully, because the human mistreatment of nonhuman animals had become, for me, the thing that mattered most \u2013 the most fundamental moral problem of our age.\u00a0<\/span><\/p>\n

\u00a0<\/span>And in the inaugural volume of Animal Sentience in 2016 there appeared, 38 years after Premack\u2019s seminal target article on mind-reading by chimpanzees, Brian Key\u2019s target article on fish pain<\/a>. Except that Key was arguing that there wasn\u2019t any. And his target article, too, elicited over 40 commentaries. But unlike Premack\u2019s article, on which the vast majority of commentaries were positive, most of the commentaries on Key were negative.<\/p>\n

\u00a0<\/span>Key\u2019s arguments and evidence were simple And easy to summarize: Fish can\u2019t talk, so they can\u2019t tell us whether they feel pain. The nonverbal behavioral tests of pain are unreliable and open to interpretation. So the only way to decide whether fish can feel pain is on the basis of pain\u2019s neuroanatomical correlates —– in humans, as well in other mammals. And those are lacking in fish. They are lacking in birds too, but for a rather less clear reason, Key was willing to agree that birds can feel pain, and that they do it with a different, but homologous neural substrate.<\/p>\n

\u00a0<\/span>The vast majority of the commentators, which included most of the world\u2019s experts on fish behavior, physiology \u2013 and, importantly, on evolution \u2013 Including Don Broom, who is present at this meeting, disagreed with Key, citing many behavioral tests as well pharmacological evidence, such as the effects of anesthetics. Let me just point out a few other commentators [Don Broom Seth Striedter Balu\u0161ka Burghardt Derbyshire Godfrey-Smith Panksepp Sneddon Damasio Devor<\/strong>]<\/p>\n

\u00a0<\/span>Specialists in brain evolution also pointed out that fish do, indeed, have homologues of the neural substrate of pain in mammals and birds, and that it is not at all rare that convergent evolution finds ways to implement functions in structurally different ways in distantly related taxa, especially for a function as fundamental and widespread as pain.\u00a0<\/span><\/p>\n

\u00a0<\/span>Not only was there stout opposition to Key\u2019s thesis in the first, second, and third wave of commentaries, each round eliciting a Response from Key, defending his negative thesis equally stoutly, but there have since been two further target articles from specialists on fish behavior and brain function who had been commentators on Key\u2019s target article, This time both target articles were supporting the positive thesis that fish do feel pain, and eliciting their own commentaries, most of them likewise positive ones. The first in 2017 was by Michael Woodruff<\/a>, The second, by a large number of co-authors, Including Don Broom, was led by Lynne Sneddon<\/a><\/p>\n

\u00a0<\/span>But the issue is not settled; indeed it cannot be settled beyond all possible doubt, because of the other-minds problem. Note especially the name of Jonathan Birch, to whom I will return in a moment.<\/p>\n

\u00a0\"\"<\/span><\/p>\n

The latest of the commentaries from only a week ago, by Terry Walters cites Sneddon\u2019s 2014 paper in which she reviewed the evidence on seventeen criteria for pain perception in seven broad taxa taxa — Mammals, Birds, Amphibians\/reptiles, Fish, Cephalopods, Decapods, Insects \u2013 Note that mammals and birds meet all 17 criteria, and fish already meet all but one of them.<\/p>\n

\u00a0<\/span>But the other taxa have their champions too, notably insects<\/a>.\u00a0 <\/span>Ke<\/strong>y is there, as a critic, of course. And Mallett & Feinberg are writing a book on the evolution of sentience that will be appearing soon. A few weeks ago, at a Summer School on the Other-Minds Problem<\/a> with over 40 speakers<\/a>, the problem was also extended to plants and microbes. [Berns, Birch, Burghardt, Wise, Balcombe, Kona-Boun, Marino, Baluska, Keld\/Mendl, Simmons<\/strong>]<\/p>\n

\u00a0<\/span>But lest you conclude that as the potential scope of sentience widens the welfare implications of the other-minds problem become either indeterminate or intractable —- practical principles are also emerging from the debate: In direct response to Key\u2019s target article, the bioethicist Jonathan Birch<\/a> has proposed adopting a Precautionary Principle, according to which, if one member of a taxon meets the criteria for sentience, the benefit of the doubt should be extended to all members of the taxon, rather than waiting for each species to be tested individually. This is analogous to making the null hypothesis sentience in such cases, which places the burden of proof on those who seek to show that those species are not sentient. There is still much thought to be given to this approach, but Birch has been consulted in the drafting of EU legislation about sentience, especially about extending some of the protections currently accorded to mammals and birds also to fish, as well as to octopus.<\/p>\n

\u00a0<\/span>New non-invasive techniques for mind-reading, especially in mammals, are also emerging. These too are being accorded Open Peer Commentary in Animal Sentience. I especially recommend the approach of Greg Berns<\/a>, who is trying to mind-read dogs, but I also recommend the cautionary remarks of Karen Overall, who is also present at this conference, on the need for many control conditions if we are to mind-read dogs for something as specific as feelings of jealousy. (Note that Peter Singer is, understandably, an enthusiastic commentator at the prospect of non-invasive mind-reading.)<\/p>\n

\u00a0<\/span>I would like to close with a few words about an important ongoing project \u2013 an attempt to put Quebec\u2019s new law<\/a> on animal sentience and biological imperatives to the test in court on a case that is well on the mammalian end of the spectrum, where there is already consensus on both sentience and the capacity for pain: the rodeo.\u00a0<\/span><\/p>\n

\u00a0 <\/span><\/em>Section 5<\/em><\/strong>. The owner or custodian of an animal must ensure that the animal\u2019s welfare and safety are not compromised. An animal\u2019s welfare or safety is presumed to be compromised if the animal does not receive care that is consistent with its biological imperatives. Such care includes but is not limited to ensuring that the animal is not subjected to abuse or mistreatment that may affect its health.<\/em>
\n\u00a0 <\/span>Section 6.<\/em><\/strong> A person may not, by an act or omission, cause an animal to be in distress.<\/em>
\n\u00a0 <\/span><\/em>Section 7.<\/em><\/strong> Sections 5 and 6 do not apply in the case of agricultural activities, veterinary medicine activities, teaching activities or scientific research activities<\/em><\/p>\n

\u00a0<\/span>In 2017, Professor Alan Roy<\/a>, a professor of child and animal law at the University of Montreal, together with 20 law students, drew the (I think) very natural conclusion that for activities that do not fall within the exception noted in Section 7, violations of Section 5 or 6 should mean that the activity is henceforth illegal under Quebec\u2019s new sentience law. Professor Roy accordingly went to court to seek an injunction against an \u201curban rodeo\u201d that had been planned as part of the celebrations of the 375th<\/sup> Anniversary of the City of Montreal. Because of the threat of sizeable financial losses to the Rodeo if the event were to be cancelled, the Rodeo (Festival Western de St-Tite, FWST) and Professor Roy came to a court-certified settlement. In return for not contesting the Montreal centennial rodeo, Professor Roy would be granted the right to appoint three representatives \u2013 a veterinarian, an ethologist and a photographer \u2013 to witness and film not just the 4 Montreal rodeos, but 16 further rodeos by the same producer (FWST), and the evidence would be presented to a Consultative Committee of the Ministry (MAPAQ) in charge of implementing Quebec\u2019s new\u00a0 <\/span>sentience law (AWST) to test the legality of the rodeo under AWST.\u00a0<\/span><\/p>\n

\u00a0<\/span>If the evidence demonstrates that the rodeo contravenes the new sentience law, in whole or in part, and MAPAQ does not act, then MAPAQ will be taken to court. What has resulted so far is an unprecedented database of 3 x 45 hours of close-up rodeo video on the totality of the 20 rodeos. They were analyzed for 3 months at slow motion by Dr. Jean-Jacques Kona-Boun<\/a>, the appointed veterinarian, in a 600-page report<\/a>, which has identified evidence that every one of the rodeo\u2019s 8 categories of events violates section 5 or section 6 of Quebec\u2019s new laws. I was appointed one of the members of the consultative committee, along with a veterinarian and an attorney, to represent the interests of the animals; the rodeo also has three members \u2013 a FWST producer, supplier and veterinarian and the Ministry has three members.\u00a0<\/span><\/p>\n

\u00a0\"\"\u00a0<\/span><\/p>\n

We are now seeking ethologists and veterinarians the world over to examine this unprecedented database and Dr. Kona-Boun\u2019s analysis, to add their judgment to whether the evidence indicates that the rodeo is illegal under Quebec\u2019s new law. It is obvious how the outcome may influence further tests of this kind. Several distinguished experts<\/a> have already reacted to the findings.<\/p>\n

\u00a0<\/span>I will now show a very brief video<\/a> showing a sample of the kinds of evidence Dr. Kona-Boun identified repeatedly in the 3 x 45 hours of video across the 30 rodeos.<\/p>\n