Jonathan Birch on the Edge of Sentience
28:03 Share

This is Philosophy Bites with me, David Edmonds. And me, Nigel Warburton. Philosophy Bites is available at www.philosophybites.com. Nobody doubts that dogs and cats feel pain. But what about goldfish? Or squid? Or bees? The question of which animals are sentient is of enormous importance. Not least because it should surely influence how we treat them.

Jonathan Birch, a philosopher at the London School of Economics, has been leading a project investigating animal sentience and the ethical status of animals. Jonathan Birch, welcome to Philosophy Bites. Hi Nigel, thanks for inviting me. The topic we're going to focus on is the edge of sentience. Just before we get to the edge, what is sentience? It's good to start by thinking about pain and pleasure. Ask yourself questions like...

can an octopus feel pain can a crab can an insect what about a spider in fact i think it's good to have a concept that is a bit broader than just pain and pleasure a concept that captures any experience that feels bad or feels good any experience with a positive or negative quality that includes pain and pleasure

but it also includes things like joy, excitement, stress, boredom, comfort, discomfort. And so when I'm talking about sentience, I'm talking about the capacity to have any positively or negatively valenced feeling. So some people talk about consciousness. Why did you choose sentience rather than consciousness here? I think they're both words that get used in a variety of ways.

I find sentience a little bit clearer because I think what it does is it draws our attention to what is, to me, the most basic, elemental, evolutionarily ancient base layer of human consciousness. Sometimes when we use the word consciousness, we're using it to refer to things that are laid over the top of that raw sensation and feeling.

Things like our ability to reflect on what we're experiencing and our ability to have an inner monologue about our lives and how it's going. And these are things that are on top of that much more basic thing that is just feeling. Feeling things like pleasure and pain. You use the word valence. That's not obvious what that means. Could you just gloss that for us? You think of...

Pain, well, it has a negative quality. So does boredom. So does fear. So does discomfort. There's this whole family of feelings that have this negative quality. And so psychologists and philosophers call these negatively valenced feelings. And then there's this other family, pleasure, comfort, the satiety of having eaten and feeling full up, warmth.

These are all positively valenced. They've all got a positive quality. So your interest in this isn't just a kind of intellectual interest, I wonder which animals are sentient. It's much more than that.

I think in all of these cases at the edge of sentience, really important practical questions hinge on whether we regard these systems as being sentient beings or not. If you think about animal welfare laws, for example, decisions often have to be made about the scope of those laws. Which animals are we going to protect?

And there, sentience is the threshold, I think, an appropriate threshold. We should be trying to protect those beings that are sentient, that can suffer. As Jeremy Bentham put it,

The question is not can they reason or can they talk, but can they suffer? I think that's absolutely right. So sentience ends up having this huge ethical and policy significance. But presumably sentience isn't all or nothing. It tends to come by degree. So you can say some things are more sentient than others.

I don't personally think one can talk of one animal being more sentient than another. I think what we've got is tremendous variation across the animal kingdom in the sentience profile. Sensory abilities vary enormously. Lots of other things vary as well, including the intelligence and cognitive sophistication of these animals.

But these are all dimensions of variation among sentient beings. It doesn't mean that if an animal is less intelligent, it's less sentient. In fact, I don't think we should be talking about more or less sentient as if there were a single sliding scale that would allow us to make sense of this massive variation. Does that mean that you think there is a cut-off point where we can say these animals are sentient, these are not, or these living beings are sentient, these ones aren't?

I think it's a big open question about whether there is a sharp boundary or whether there are borderline cases. If you think about a concept like tall or old, there's not a sharp cutoff when you become old.

It's a vague matter and there are borderline cases. And some have argued that sentience might be like this and the edge might be somewhat blurry. An alternative view is that, well, that makes sense for tallness and oldness, but it doesn't make sense for there being something it feels like to be you. That really is something you either have or you don't have.

And I think either of those views could be right, and it's not quite clear which. But what is clear, I think, is that often in practical contexts, like when you're formulating a law, when you have to make a decision, in practice, it is quite a sharp thing. You are either going to treat these animals as having some moral significance, or you're going to ignore them. And in a way, that is a binary thing. We've decided to focus on the edge of sentience. What is the edge for you?

I think we can talk about edges in three different senses, actually. There's the edge that is in the world, the sharp or perhaps slightly fuzzy boundary that is really out there, the boundary of the class of sentient beings. But we can also talk about an edge in our knowledge.

This threshold at which doubt really starts to creep in and we start to seriously entertain doubts about whether a being is sentient or not. And for different people, that edge is in different places. Everyone has some point at which they start to have doubts. For some people, it's around fishes. For other people, it's octopuses, insects.

But everyone at some point, as you get more and more distant from humans in evolutionary terms, doubt starts to creep in. And then I think we can think about edges of a practical kind as well, which is all of these contexts where we have to make decisions, whether it's setting the scope of a law, deciding how we're going to treat animals of certain kinds or other potentially sentient systems. We have to draw lines.

in laws, in decisions, and these practical edges have particularly fascinated me over the last few years. So let's take that useful tripartite approach then. You've got the edge where in reality there is an edge. There are sentient beings and there are non-sentient beings and there's either a fuzzy area or a sharp cut-off, we don't know, but there is a truth of the matter whether a stick insect is sentient or not. So a particular stick insect even, not just a species.

Then you've got the question of the edge of our understanding, as it were, the point at which I, for example, will start to have concerns and doubts about whether what I'm dealing with is or is not sentient.

And then there's a third sense of edge, which is also very important that you described, which is where practically we have to make a decision and draw a line. There's that kind of edge. Let's go with the first one first. Where is the edge of sentience in terms of reality? Where do you believe we can draw the line and why? We're currently not in a position to really know where the edge in the world is. And I think it's important to have a lot of humility about this.

And in fact what I urge is shifting from the question of "Is the system sentient?" to a slightly different question, which is "Is the system a sentience candidate?" Where being a sentience candidate is about there being a realistic possibility of sentience that has been established by at least one credible theory of what sentience is. So we're in this position of great uncertainty, we're faced with this zone of reasonable disagreement,

where lots of people have very different views, but we can talk about a range of realistic possibilities that have some substantial evidence behind them. And then we can define a sentience candidate as a system that is sentient according to at least one of those realistic possibilities. And there, when you shift the question in that way, I think it becomes a lot more tractable because we've defined our concepts in this way, we now have a question that we can answer using current evidence.

And I think we've also got a very broad category where, because we're not talking about knowledge, we're not talking about certainty, but rather about realistic possibility, I think we have to immediately see that all adult vertebrate animals are sentience candidates. And moreover, that many invertebrates are too, including octopuses, including crabs, lobsters, including even insects.

Back in April, I was involved in launching this statement called the New York Declaration on Consciousness. We did use the term consciousness in that instance rather than sentience, even though I prefer the term sentience myself. And we were trying to send a signal in that statement that there's a realistic possibility of sentience in many, many animals, including insects.

And of course, that is a very hard idea to get one's head around. It's an idea that might potentially have quite significant ethical implications. But I think that's where the evidence is leading us. And you talk about evidence. You're a philosopher, but you work very closely with scientists. Yes, my Foundations of Animal Sentience project at the LSE has been about working closely with scientists on these questions of sentience.

for years. Part of it has been about gathering new evidence, particularly concerning bees, working with Lars Chitka. Part of it also has been about synthesising the existing evidence, coming up with frameworks for doing that. So in 2021 I led a report for the UK government that was about octopuses, crabs, lobsters, shrimps,

We weren't gathering new evidence about those animals, but we were bringing together evidence from over 300 scientific studies and evaluating the overall strength of the case for thinking of these animals as sentience candidates. And we ended up concluding that it is quite a strong case. I'm really interested in bees in the sense that they're not really isolated individuals. I can see an octopus is quite a solitary being for most of its life.

And there are very plausible reasons why we can think of them as sentient. There have been a number of important popular discussions of this through Peter Godfrey Smith's work, but also through film with My Octopus Teacher. And there's a sense in which the public is ready to accept that an octopus, although it's very different from us, is very likely sentient.

But with bees, you're not dealing with individuals. Most bees don't survive outside of a colony. And are you looking at the individuals as sentient or somehow the combination of individuals as sentient? It is the individuals. I think you're totally right. Firstly, that it's octopuses that often bring people around to the idea that an invertebrate might be sentient despite being separated from us in evolutionary terms by over 500 million years.

They seem to have evolved their own route to some kind of sentience. But then once you've got to octopuses, I think it's very natural to start thinking about other complex, active, mobile invertebrates like crabs, lobsters, insects. And you're also right that in the case of bees, the hive behavior is incredibly impressive. The decisions they make as a collective are very impressive. But that can sometimes lead us to neglect what the individual bee is capable of.

And in fact, a lot of Lars Chitka's work and then some of our collaborations in recent years have been about looking at what the individual bees are capable of. And in fact, the view that an individual bee is just a kind of

reflex machine. Everything it does is just pure reflex. It's been around a long time, but I think it's just been completely smashed by the evidence. We now see that individual bees are incredibly capable learners, and they can learn to succeed in tasks that

neither they nor their ancestors would ever have encountered in the wild. Things like pulling a string to access a reward, rolling a ball into a hole to get a reward. And from then realizing they're so good at learning, people have then over the last 10 years or so started to ask, well, what about their emotional lives? Could there also be emotion there as well as cognition?

The evidence seems to be strongly pointing towards that being the case. For example, there was a study that took inspiration from that experiment where the bees roll balls into holes to get reward and just took out the reward and was studying will the bees still roll the balls for no reward just because they seem to be finding it fun and showed that they do. They seem to display a kind of play-like behaviour.

And then once you've got to that, I think it's very natural to start asking questions about pain. Again, the idea of pain in insects has been widely dismissed a lot of the time, but not really for any strong reasons. And when Matilda Gibbons and other colleagues associated with my project really started to look at this, well, if you touch a bee's antenna with a heat probe,

they then seem to groom that antenna. They display self-protective behaviour, tending to their injuries, just like many, many other animals would. So this is not conclusive evidence that they're feeling pain, but it's shaking some of those old assumptions, and it's starting to move towards establishing a realistic possibility that they have not just impressive intelligence, but really sophisticated emotional lives as well.

I mean, that's hard to stomach. And that's really moving on to the second sense of edge that we discussed. The question of where I personally or any individual sets the limit for their capacity, as it were, to recognise sentience. I might recognise that

Mammals, fish, octopuses, lobsters are sentient to some degree and in some ways, but I might draw the line that stick insects and bees. Somebody else will draw the line differently. How do you approach that as a philosopher? Is it just people being irrational not to immediately conclude that bees have emotional lives? What do you say about that kind of edge? I think there's real uncertainty. And so in a way, when people say, well, I'm not

persuaded by this evidence, they're not wrong because if the bar was certainty or knowledge even, we're still up against the problem of other minds and we can still conceive of all this going on in the bee without any conscious experience. But what I do think we're doing is establishing a realistic possibility where

Getting enough evidence here that it becomes irresponsible to ignore that evidence, if, for example, you're a beekeeper or an insect farmer, you're someone working with these animals on a day-to-day basis, it's really rather negligent to treat them as though they felt nothing when there is all this evidence. And so I think we need to err on the side of caution in these cases.

Recognising a realistic possibility of sentience whenever the evidence is pushing us to do so, and we then need to be taking proportionate steps to try to respect that. And then of course there's going to be a lot of debate around what proportionality requires. But that's good, that debate needs to happen. And I'm strongly in favour of broad, inclusive, democratic debate.

about what we should do in these cases. I love the way that you're not certain. You talked about humility earlier. There's a sense in which where there isn't certainty, the philosopher shouldn't be bluffing and saying, we're absolutely certain that bees feel pain. And what you proposed in those circumstances is a kind of precautionary principle that goes on to a third case in a sense about what you do about drawing the line with morals and law.

We need to err on the side of caution. We need to, in some sense, give the animals the benefit of the doubt. And we need to push against this tendency to be overconfident about the absence of sentience that I think we see over and over again. And try to take precautions even in cases where we're not sure. And then, of course, it's important to go beyond that slogan. Erring on the side of caution could mean anything from the tiniest gesture of

to the most dramatic change to our way of life. And so I think what we need is this concept of proportionality, that our precautions need to be proportionate to the identified risks. They need to do enough without being excessive. And we need to establish processes. And I'm quite a fan of processes like citizens' assemblies, where people are brought into the discussion to have a debate about what would actually be proportionate to these risks.

It was interesting, human beings are pretty good at being cruel to one another. Some people will defend that, all kinds of political and nationalistic defences can be mustered against the need to kind of take firm physical action against people you object to. Obviously we've got a kind of anthropocentrism where we put our own species above others on the whole. Given that,

What hope is there of swaying people into thinking that we need to be proportionally sensitive to the possibility of sentience and say, even fish, even fish? I think a lot of the precautions we can take are quite low cost. And I think people can see that maybe

Some changes to how we do things are justified by the risks of causing terrible suffering. So if you think about the case of dropping lobsters into pans of boiling water, when producing our review, we reviewed evidence that unfortunately the animal does not die quickly, it takes around two minutes usually. Two minutes is characterized by a huge storm of nervous system activity, just as you'd expect to see in any other animal.

And on the face of it, this is not a humane method and the risk of causing terrible suffering is so obvious. And there's also low-cost ways you can do things better, for example, stunning the animal before you put it into the water. And this is not excessive in any way and I think people can see that. If we have this shared value that we want to relate to animals in a positive way, treat them humanely, care about their welfare,

Sometimes we will have to take some precautions, make some sacrifices maybe to actually live by that value. And I do think people can see that, particularly in their more reflective lives.

which is why I like things like citizens' assemblies that try to bring people together, inform them about the issues, allow them to deliberate and reach a reflective, considered view. And I think in the case of intensive farming, often the problem is that people are actually not informed. That when people are informed about the conditions in which the animals they eat are kept,

they immediately see the problem and they say, I don't want farming to be like that. I do want stronger welfare protections. But you have to give them accurate information to allow them to actually be in a position to express that value. To me, it's not that there's a moral failing, really, in a way. People do care about animal welfare genuinely, and you see that in how they relate to their companion animals like cats and dogs. The moral value is there.

But what's not there is the information about how badly society falls short of that value in its treatment of farm animals. So we've been talking about sentience and we both agreed, of course, that animal sentience is very relevant to our moral and political behaviour towards other species and how we approach our diet, how we approach how we treat pets and wild animals. But one thing that

people differ a lot on is once an animal has started to exist and is sentient, is there some kind of moral obligation to keep it alive? If it could be killed painlessly without any knowledge of its imminent demise, do you do anything wrong by killing it painlessly in the way that you would with a human being? It's a huge question. It's sometimes suggested, for example, in relation to fish, and Martha Nussbaum suggests this in her Justice for Animals book,

that because they don't have any sense of the future, she claims, it becomes then permissible to kill them. And that the problem with killing animals as such, if you treat them incredibly well up to that point, only arises for those animals who have some sense of their own future, some plans that are then interrupted. I'm not sure if that's right to begin with. I think there are various relationships we might have

If you think about our own case, even if we had no plans for the future at all, the relationships we have to the people around us, the people we love, they create a kind of implicit pull towards the future. We have something to lose if we suddenly exit those relationships. And I think that may well be true of a wide range of animals as well. I'm also not sure that that claim about fish is really correct.

For me it's just a huge unknown which animals have this subjective sense of the future, plans for the future. We've got positive evidence for animals like corvids and crows, jays, that they do plan for the future. We don't have that kind of evidence for fish. But I think humility is needed in the cases where we have evidence gaps. There's just a huge amount we don't know about the minds of other animals generally.

And we need to err on the side of caution in lots of ways. Sentience is obviously absolutely the heart of this, but is it the only relevant aspect of animals when considering how we treat them? I do think sentience is extremely important. It means there are some morally significant interests there. And then you get the question, well, is there a second tier of moral status, as it were, that you might get by being rational or being intelligent?

And personally, I tend to think not. I don't think there's a VIP tier in the club of individuals that matter morally. What I do think is that when you have sophisticated forms of cognition, well, sometimes it can lead to new ways in which you can suffer.

or enjoy your life. If you think of the fact that we can remember events from our distant past that might have been traumatic, this creates a new way in which we can suffer. The fact that we can think about our own mortality and dread events in the future, this creates a new way in which we can suffer. Could argue then there are ways in which we can suffer that perhaps other animals can't. At the same time, I think our intelligence also brings with it ways of managing our suffering.

the fact that we can have hope for the future, that we're in pain now but we've taken this painkiller and so we can hope that in half an hour's time we'll feel better. And animals that don't have that sense of themselves as beings in time will not be able to form that hope. So I think life can be both better and worse when you're more intelligent but I don't think it gives you any kind of special level of moral standing.

I think it's fair to say that you and your team have actually changed the way that lobsters are treated in Britain and Europe. Is that true? Well, we produced this report in 2021 that informed the UK government's Animal Welfare Sentience Act, and we advised them

to change the bill they'd put to Parliament which only extended to vertebrate animals so it included fish and recognised them as sentient but it said nothing about invertebrates and we advised them to amend it and include octopuses, crabs, lobsters, crayfish

and they did. I think that was a really positive step. What it does is it creates a duty on policymakers in the UK to consider the animal welfare impacts of their actions and to include octopuses, crabs and lobsters in that. Now it's a step, but our report actually recommended many other steps as well. We really think that

Once you've got that basic recognition of sentience, you need to do a lot of other things that follow from that, including amending lots of other pieces of legislation around...

slaughter regulations, for example, as well. And we've not seen that yet. So we've not seen that following through on the consequences, which of course we're still pushing for and hoping for. So one last question. Is there some kind of animal that you think isn't getting treated well in the sense that its sentience is not being adequately recognised at the moment? I think we're seeing huge investment in aquaculture, in intensive farming of fish, in

We're seeing a lot of investment in farming invertebrates like insects, and there's this narrative around it being a sustainable source of protein. We're even seeing an attempt to farm octopuses that has been very controversial, I think rightly so. I think animal welfare gets neglected in all of these discussions. It's a really, really neglected topic where even very simple things you can do

Like if you're farming shrimps, people will often cut the eye stalks off the breeding females to get them to produce eggs quicker. They're things that just show absolutely no regard for the animals as sentient beings and would be so easy to change. And so I really want to see more attention going into these issues, more pushing for these low-cost, simple changes that could make a huge difference to the lives of trillions of animals. Jonathan Birch, thank you very much. Thanks, Nigel.

For more Philosophy Bites, go to www.philosophybites.com. You can also find details there of Philosophy Bites books and how to support us.