WORKSHOP 2: Jonathan Birch:
(Wednesday, June 27, 7:30pm)
In debates about animal sentience, the precautionary principle is often invoked. The idea is that when the evidence of sentience is inconclusive, we should “give the animal the benefit of the doubt” or “err on the side of caution” in formulating animal protection legislation. Yet there remains confusion as to what “applying the precautionary principle” means in practice regarding the burden of proof for animal sentience. I have recently developed a version of the precautionary principle tailored to the question of animal sentience and a practical framework for implementing it. In this talk, I explain and defend the basic proposal, discuss some of its potential applications, and consider some ways of revising and improving it in light of critical responses.
Many of you might have seen this footage of a crayfish taking extraordinary measures to escape the boiling water http://www.dailymail.co.uk/news/article-5800181/The-CRAY-T-escape-Crayfish-COOKED-tears-leg-bid-dodge-dinner.html
ReplyDeleteYes. It highlights the point that death is far from reliably quick with this method of slaughter.
DeleteIt is hard not to see this behavior as evidence of a cognitive motivational trade-off. But I have doubts on how this video was obtained.
DeleteIt is written in the web site that “In the video, the crayfish can be seen clambering out of the hotpot, while one of its claws remained dangling immobile inside, being cooked into lifelessness ».
1) The video starts with the crayfish already out of the soup – positioned backward, thus facing the inside.
2) Already all out of the soup, except for both front claws, it still can’t pull itself all the way out. We can see that all of its legs don’t have enough traction to get out.
3) How can there be enough traction to get out of the soup in the first place, when there is not when the whole body is out?
4) Although I believe in the motivational trade-off observed here, I am sad to come to the conclusion that this was a “movie” setup. I think that the crayfish was intentionally dipped front first in the boiling soup and then placed on the edge to see what he would do - perhaps an attempt to reproduce an anecdotal story, and get "famous" on the web at the same time.
I just hope they did not have to do many “takes” in order to get this final spectacular shot… Am I too cynical ?
About lowering the evidential bar and taking appropriates precautionarys responses, you said that “the response should be proportionate to the plausibility and severity of the threat”. To my question, you respond that evaluating the level of the threat is a question of common sens. In that case, how do you adjudicate at which level the evidential bar should be set precisely? For exemple, you set at two the number of order in which there is at least one credible indicator of sentience as a suffisent evidence of sentience in this taxa. How this criteria is "proportionate" to the threat?
ReplyDeleteThanks for the question. It's the precautionary measure that should be "proportionate" to the threat. The measure should be "proportionate" in the sense that it should not be excessive (it should not do more than is necessary to mitigate the risk of harm) and it should be the most cost-effective of the available non-excessive measures.
DeleteThe evidential bar (at least one credible indicator of sentience in at least one species of the order) is my proposed evidential condition for triggering a precautionary response. The idea is to have a condition that says: when your evidence for sentience clears this bar, some kind of precautionary measure is warranted.
The justification for it is pragmatic; it is intended as a bar that will allow timely responses to new evidence. Requiring separate evidence for every species would be too demanding, and I contend that requiring multiple indicators for each order would also be too demanding (though others such as Mallatt and Woodruff disagree). Setting the bar too high leads is a recipe for inaction: it leads to avoidable delays that we may subsequently regret.
Thanks for your answer.
DeleteI think that the precautionary principle is a good way to get into a question because it seems to be the most rationnal way to figure out a problem, considering the weighting of probabilities. But I think that the way we compute probabilities largely depends of values that a person put in prioritie. It's not necessary to convince people that already defends the welfare of, to take the exemple, decapods, that it's not necessary to hurt an organism when there are others way to proceed. But people wo act in a way that seems to possibly hurt differents types of animals or other creatures seems to have a different hierarchy of values and I think that these people could refuse to bend in front of this idea. Considering this disharmony, I think that it's not possible to get a consensus and that the question about the way we treat the differents kinds of organisms is not a question where we can apply the «common sens».
ReplyDeleteThanks. No precise probabilities are needed to apply the precautionary principle, and I think this is one of its attractions. We don't need to say "there is a 90% probability that decapods are sentient", but only: "there are credible indicators of sentience in decapods". The former claim would be contentious; the latter is much less contentious.
DeleteThe "common sense" idea here is simply that if there is a practice that we agree *would* be inhumane on the assumption that the organism in question is sentient, and if there is some indication that the organism is indeed sentient, then we should regulate the practice now. I think that's common sense.
Some other aspects of the framework are not common sense but try to extend the above common-sense insight in a principled way (the parts concerning orders, phylogenetic inference, etc.).
I agree with you, people engaging in those behavior are the one that need to be convinced that other beings are sentience and can feel pain. "Common sens" can't be a valuable argument in that case. I think the first step to change the belief that some animals can't feel pain would be to informe them.
DeleteIt's not necessarily the case that the people engaging in the practice itself need to be convinced. It's primarily the people who make animal welfare law who need to be convinced.
DeleteBut do you think that if people doing those practices hear/learn about animal sentience it would be easier to change laws? If they realize the pain they're causing, their practice would change and therefor the laws?
DeleteThe bet of Blaise Pascal as to the uncertainty of the existence or not of God, maintains that it is better to believe, a whole life, than to risk to spend eternity in hell. Jonathan Birch's argument with the Precautionary Principle seem to be essentially the same: it is better to believe that animals feel and thus to limit induced suffering (although it is possible that they may not feel), that to believe that they do not feel and that there is no need to worry about any suffering (even if it is possible that they feel!)
ReplyDeleteIt is interesting to note that this argument takes its force from the uncertain nature of the object of which it is the advocate, while the uncertainty as to the presence or not of the animal feeling has always been the weakness of its defense. Even if we can not validate it with certainty, this does not mean that there is no moral duty as to its possibility. Like Dostoïevski that can't be collectively endorsed when he writes, "If God is dead [absent], all is permitted", it is necessary to recognize our responsibility as a moral agent in the face of the mere possibility (not even obvious, but rather rationally justified) that different animal species resent pain.
This argument leads me to reflect on the relevance of metaphysics and dialectical discourse for issues of this kind. The problem of other minds is a problem for science. The presence of a "mind" in other entities than myself is a possibility that dialectical discourse takes into account in its semantic exhaustiveness in addition to being a subject whose metaphysics can rationally deal with. Science must not be the guarantor of all the answers when it comes to a problem the object of which escapes it. The only possibility combined with a rationally justified belief seems sufficient in cases like this.
The main distinction between Pascal's bet and the Precautionary Principle used here is that the object of this bet is not our own welfare and preservation, but rather that of other animal species. The question seems rather to become; Why be sensitive to animal pain? Why choose to avoid it? Is there an interest in doing it, should we even find one?
Thank you for the talk!
DeleteThanks for this interesting comment. There is a resemblance between "Pascal's wager" and the argument for treating potentially sentient animals as if they were sentient: both are pragmatic arguments for, in some sense, erring on the side of caution.
DeleteThere are also significant differences. The difference between finite and infinite payoffs has no role in my argument. The benefits of protecting animals from harm are finite. It's just that, when one realizes that trillions of animals per year are involved, these finite harms are much larger than we often think, and deserve more weight in our decision-making than we normally give them.
Pascal's Wager faces the "many gods" objection: we can't rule out the existence of Pascal's preferred Catholic god, but we also can't rule out many other rival hypotheses, including the existence of gods who deal out infinite disutility to those who believe in a Catholic god and infinite utility to those who don't.
The analogous objection in the case of animal sentience would be something like this: although we can't rule out the hypothesis that some practices (e.g. live boiling) cause massive suffering to animals, we also can't rule out the possibility that alternatives to these practices would cause just as much suffering.
However, unlike in the case of theism, we can use scientific evidence to guide us in assessing and comparing the risks. For example, we know that some methods of slaughter create less risk of suffering than others: all else being equal, fast methods are better than slow ones, and stunning is better than not stunning.
Hi, thanks for the interesting talk.
ReplyDeleteIt seems that establishing what the "credible indicators of sentience" are is essential to having this argument from the precautionaty principle generally accepted. I was wondering if you had an idea about what methodology you would use to determine what counts as a credible indicator (general consensus, scientific reseatch, etc) and if you had a sort of provisional list of credible indicators already sorted out that you could share.
You're right that establishing a clear list of credible indicators is an important challenge.
DeleteI have highlighted motivational trade-offs, conditioned avoidance of places where past injury occurred, self-delivery of analgaesics (such as opioids) when injured, willingness to pay a cost to access analgaesics, and the processing of nociceptive information in integrative brain regions. I think these are all credible indicators, and that's a starting point. We have evidence of all of these in teleost fish, and two of them (the first two) in decapods.
What these indicators have in common is that they give us evidence of the centralized processing of information about noxious stimuli, and the integration of that information with information from other sources, in a way that guides adaptive whole-organism decision-making in response to tissue damage.
It's always conceivable that this processing could happen without any feeling of pain. But it's also plausible to think that this is the general type of processing that gives rise to pain experiences in sentient creatures. That's why I call these "credible indicators" of sentience.
Would the precautionary principle be a moral equivalent of epistemic caution? Both are measures to manage uncertainty in a formalized and systematic way; a way that lawmakers can use with consistency.
ReplyDeleteThe criteria in question should be a transdisciplinary consensus and that in itself is quite a challenge. I don’t think the challenge is of methodological nature, more of philosophical nature. The translation of concepts between different areas of knowledge is problematic (judges talk of autonomy, cognitive scientist talk of processes, activist talk of sentience, etc... This summer school is one of the few places where those silos can be broken and those concepts articulated in a consistent map of categories about the mental life of human and non-human animals.
There was some clues in many of the other conferences of what could constitute possible indicators of sentience [as the capacity to feel]: decision making, contextualization, planing, modulation of energy spending.... Those could be a good start.
THanks for the wonderful talk
Surely merely epistemic "epistemic risk" is not the same as other-minds risks when the epistemic risk is mine and the risk of suffering is that of the other mind, if I misjudge it as sentient...
DeleteJust adding the comment of a question I already asked, for the evaluation for students: how do you apply the precautionnary principle in reality? If you apply that to lobsters, then cows and fishes and all “higher” animals (considered by the public opinion) should all have the concept applied to them. But how do you deal with all of the infrastructure and politics lying around the issues. We also talked about it yesterday during the workshop on pig cognition.
ReplyDeleteAs a first step, Jonathan Birch promotes the precautionary principle toward according invertebrates the legal protections already accorded to vertebrates.
DeleteFor species that we already know to be sentient and capable of suffering, it is not the precautionary principle that implies that we should spare them, but the principle that it is wrong to cause suffering needlessly.
In the developed world, for example, hurting and killing animals for food, fur, fashion or entertainment is unnecessary for our survival and health.
With potentially life-saving biomedical research the problem is greater, because the precautionary principle can be invoked in both directions in this conflict of life/death interests. But the outcome will surely be more merciful once needless suffering -- currently at monstrous levels -- is being brought to an end.
Hi!
ReplyDeleteThanks for the refreshing presentation. It's very encouraging to see that someone has a concrete solution to the problem, even though it still feels a bit abstract.
I have a question regarding one comment you made during the presentation. We were talking about serious risk for applicaton of the PP, and someone asked what about when it's not regarding the extiction of a specie (shrimps). You suggested that we should looked into the number of shrimps - trillions actually. And I was just thinking baout the fact that some people would say 'I agree that shrimps feel pain, but I don't agree that it is an issue'. You said that it should be an issue to everyone and that people should care.
How do you think we can bring people to understand that it is actually a real issue and that people SHOULD care about torturing and killing trillions of shrimps - of decapods?
Remember to distinguish the precautionary principle itself from the principle of not causing suffering needlessly.
DeleteJ’ai vraiment trouvé la présentation de Jonathan Birch forte intéressante! Au sujet de la sensibilité animale, du “ressenti” animal, qu’est-ce qui constitue une preuve? Quels animaux ont ou n’ont pas le ressenti? En tant que société, est-ce une question qui nous intéresse, que nous prenons à coeur? Afin de faire des lois s’appliquant à la protection des animaux, combien de preuves scientifiques démontrant des signes de ressentis devrions-nous avoir avant d’en créer? Nous pensons souvent à beaucoup de problèmes tels que l’environnement, l’économie, l’éducation, la pauvreté, etc., mais rarement à la souffrance animale. Par exemple, 1 à 3 trillions de poissons sont mangés chaque année. Est-ce 1 à 3 trillions de moments de souffrances que l’on pourrait éviter? Le conférencier nous a alors présenté le Principe de précaution qui dit ceci : “En cas de risque de dommages graves ou irréversibles, l'absence de certitude scientifique absolue ne doit pas servir de prétexte pour remettre à plus tard l'adoption de mesures effectives visant à prévenir la dégradation de l'environnement.” Ce principe a été appliqué dans la cause environnementale et également en santé. Birch propose de l’appliquer à la cause animale en le reformulant ainsi : “En cas de risque de dommages graves et négatifs concernant le bien-être animal, l’absence de certitude scientifique en ce qui concerne la sensibilité des animaux en question ne doit pas servir de prétexte pour remettre à plus tard l’adoption de mesures effectives visant à prévenir lesdits dommages.” J’ai trouvé ce principe très intéressant dont la mise en place s’avérerait réaliste et pertinente. Évidemment, convaincre les partis politiques de le mettre en application est une tout autre histoire étant donné, pour ne donner qu’un exemple, la puissance de l’industrie d’alimentation animale...
ReplyDeleteThe precautionary principle is clearly not sufficient for ending human-induced animal suffering and death, since in the case of mammals and birds most people don't doubt that they are sentient. The principle is more for cases where there is bona-fide uncertainty, such as in "lower" invertebrates.
DeleteD'un point de vue de juriste, j'adhère totalement aux explications et motifs présentés par M. Birch, et j'aimerais apporter quelques précisions sur l'application du principe de précaution en droit québécois. Il s'agit d'un principe bien connu en droit qui a été non seulement énoncé dans des dispositions particulières mais qui est aussi appliqué en jurisprudence,et ce dans tous les domaines de droit. Il peut être invoqué à la base des recours généraux en injonction, véhicule selon moi beaucoup plus plausible et potentiellement efficace que celui en habeas corpus. La nouvelle loi sur le bien-être et la sécurité de l'animal pourrait constituer un nouvel appui pour pouvoir invoquer le principe de précaution n'eut été de la largesse de ses champs d'exception d'application, sujet sur lequel je reviendrai à la suite des présentations des juristes à cet effet. Par ailleurs, petit point d'information: la Nouvelle-Zélande, la Suisse et la municipalité de Reggio Emilia en Italie ont déjà expressément interdit dans leur loi la pratique de bouillir vivants les crustacés.
ReplyDeleteUne jurisprudence intéressante de la Cour suprême de cassation d'Italie (district de Florence)portant non pas sur la mise à mort des crustacés mais sur leur traitement préalable: https://fr.sputniknews.com/societe/201706171031871581-homards-italie-souffrance/
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteThank you very much for the talk, it was really interesting.
ReplyDeleteAt the beginning of the conference your mention that sentience, the capacity to feel, is a term that the community brings to science rather that asking scientist to define it. Furthermore, I recall that Mr Harnad said during one of the talks that it is a mistake to try to operationalize sentience (I might be mistaken). Considering that the application of the precautionary principle relies on scientific studies (wich requires operationalization), this seems paradoxical. How can we resolve those two statements to create sensible laws and prevent attributing sentience to living organism that might not be?
This comment has been removed by the author.
DeleteThis comment has been removed by the author.
DeleteThis comment has been removed by the author.
DeleteHi Pascal, I think you're saying: how can we make sure that we don't end up not doing something which turned out not to matter after all (worse case scenario: we all resolve to eating fruits in a world deprived of non-human sentience, turns out it was just an illusion)?
DeleteAs far as I could understand Birch's point, he's simply arguing that we shouldn't wait until we have a way of being completely sure, in fact, we'll never get there (this may be what Harnad meant when you understood that we can't operationalize sentience) and so we just need to set a bar... and be precautionary.
When a governmental program intended to help preschoolers at high risk of developing learning disabilities is put out... there is always this slightly analogous issue of having to set such kind of bar... If bar is set too high... we waste less money, but at a cost: many kids who badly needed the help won't receive it. On the other hand, the lower the bar (or criteria) is set, the higher the risk is that children who did NOT need the help will receive it anyway (wasted public funds), and the reason we may want to take this risk of being wrong is that our (precautionary) priority is to maximize the chance that all the needing children do end up being taken care of (i.e.: minimizing false negatives raises the number of false positives and vice versa and a philosophical question to ask yourself can be: what degree of precaution is ethical).
Birch's BAR is slightly different, of course, but I'm using this example to show that when unsure about something, we don't wait for evidence to pile up forever, we have to set a bar and that inevitably involves a risk of acting needlessly (which you implied we have to avoid). If your available means permit it, the way to go is usually to risk overdoing it and absorb the overcosts. Just because you can't operationalize in a way that will eradicate incertitude has nothing to do with whether or not you're doing proper science. One way or another, Pascal, we’re all making a wager based on the info available and on how much we care to minimize whatever it is we care the most about minimizing. Hoping this can help your reflection, it certainly helped mine... :)
To Jonathan Birch: Do you know Francis Crick Memorial Conference about consciousness in human and in non-human animals that happened in 2012? This meeting gathered around 17 scientists to attest the consciousness of many animal species. Had this event have any relevance, any effective influence in the field of animal cognition ou animal welfare? And Why?
ReplyDeleteLe principe de précaution suppose qu’on devrait intervenir s’il y a un risque de dommage irréversible ou d’impact négatif sur le bien-être des animaux, même si le risque ou le danger n’est pas encore clair dans la littérature scientifique.
ReplyDeleteSuite aux autres conférences sur le droit animal, le problème semble être dans l’application des lois et non comment l’animal est défini selon les lois. Je ne pense donc pas que le principe de précaution ou une application légale de ce principe est la solution idéale pour défendre des animaux avec le ressenti. Mr. Birch. a bien mentionné que certains comités d’éthiques ont bloqué des recherches sur des études animales justement par mesure de précaution, ce qui a tout de même nui à l’avancement des connaissances sur la conscience chez d’autres espèces.