Animal Sentience

What would it take to convince you that a crab is conscious? You can ask that question, of course, for any animal you like, but I think a crab is a particularly good case to focus on because people’s intuitions on this can go either way. Some people think about a crab, and they think: how could that possibly be a conscious creature? It only has a hundred thousand neurons in its nervous system compared to about a hundred billion in the human nervous system; its nerves are distributed throughout its body; it just seems so simple, so neurally simple compared to a human. But then some people go completely the other way and say: well, look at it! Look at this animal: it’s a complex mobile animal navigating the challenges of a complex environment, behaving in subtle and sophisticated ways. why would we not think that that animal is conscious? And then there’s a third group still, who say, well, maybe we can never know, maybe we can never know the answers to these questions about which animals have conscious experiences and which don’t.

I think that that third group is too sceptical: I don’t believe it. I don’t think we can ever know, and also, I think it’s very important that we try and find ways of answering these questions because really a lot hangs on it. Policymakers and governments have to make animal welfare legislation, and they have to decide the scope of that legislation; they have to decide which animals are going to be included and which aren’t. It’s a very widespread idea that it’s those animals that have some basic kind of conscious experience of the world, what’s often called sentience (you have to be sentient in the sense of having a kind of subjective perspective on the world, being capable of feeling, particularly capable of feeling things like pleasure and pain), it’s those animals that we should be trying to include within the scope of animal welfare legislation, it’s those animals that have welfare in the relevant sense, whereas animals that aren’t sentient, that aren’t conscious don’t have welfare in the relevant sense in the same way that it’s plausible to think that vegetables don’t have welfare and shouldn’t be legally protected.

So when it comes to a case like the crab, that’s a difficult case. You’ve got to decide whether that species should fall within the scope of animal welfare legislation or not. It’s not good enough to just say: oh, we’ll never know. Now you have to actually find practical ways of trying to address that question of whether crabs feel anything, of whether they’re in any basic sense conscious.

So what, then, is the evidence? What does science, the science of animal sentience, have to say about this question? First of all, it helps to fix ideas a little bit and be clear about what we’re talking about. What we’re talking about here is sentience, consciousness in a very basic sense that doesn’t require a lot of the very sophisticated capacities that humans have, so it doesn’t require the capacity to have a kind of inner monologue in which you have a conscious train of thought. Crab very probably doesn’t have that; crabs are not thinking to themselves as they scuttle along on the beach.

The question is whether there’s any feeling there, any subjective feeling, and of particular ethical relevance and particular legal relevance is the question of whether there’s any pain or pleasure, any feelings that have what psychologists call valence, sensations that are attractive or aversive, that have this kind of bad or good quality to them.

So, how do we address that question in the case of something like crabs? How do we tell whether they feel any pain or any pleasure? This has not really been studied very much at all, actually, but there are some experiments that have recently studied this question, experiments carried out by Bob Elwood and colleagues, in particular at Queen’s University Belfast. I’ll give you an example of one experiment Elwood and colleagues have carried out.

We’re looking here for evidence of pain in its most basic, evolutionarily ancient sense, just a kind of negative feeling, a negative response, an aversive response to some sort of noxious stimulus or injury. Elwood thought, how can we test for that sort of thing? How can we see whether the animal is centrally integrating the information about noxious stimuli that it’s received with other information it has about its environment and about the things it needs and wants? And they thought: well if you have evidence of that kind of integration of information from different sources in the brain, that could be evidence of pain.

He studied hermit crabs. Hermit crabs are this type of crab that lives in shells they find on the seafloor, so every hermit crab has its shell, and the crabs have preferences for certain types of shell over others. So there are certain types of shell that they really like and other types that they really don’t like, and in the wild, what they’ll do is they’ll swap a less preferred shell for a more preferred shell, so if they’re in one they don’t really prefer and they come across one they prefer more, they’ll swap into that shell. What Elwood did is he drilled holes in the shells of the hermit crabs in his lab and put electrodes through the holes to administer shocks to the hermit crabs in their shells: not a very pleasant procedure, but one that was really important in order to try and gauge how it is that crabs actually respond when you give them a noxious stimulus.

What he found was that when the shock got too severe, the crab would leave the shell: that’s not particularly surprising. What was more surprising and more striking and potentially more evidence of pain is that the crabs traded off the intensity of the shock they got in the shell against the quality of that shell in terms of their preference ordering over types of shell.

So if they were in a really good-quality shell, they were really reluctant to leave and they’d take a higher shock in that type of shell before leaving than if they were in a less preferred shell: in the less preferred shell they’d leave with a lower intensity of shock.

That’s really striking, and it’s the kind of evidence that really seems to give us some traction on this question because it’s evidence that it’s not just a reflex, not just an entirely unconscious reflex response to the electric shock, but rather the shock is detected, and then the crab integrates information about that shock with information about other things that are going on, even information about how good-quality the shell is that it’s actually in at the moment. That’s really quite sophisticated and impressive.

It’s an example of what I call a credible indicator of sentience. It’s not conclusive evidence: no one would say that this is deductively proving that the crab is sentient, that it’s feeling pain, but it’s a credible indicator in the sense that it ties in with a plausible picture of what pain is and what pain does for animals that feel it, namely that pain is this way of trading off how serious an injury is against other things that you need. Pain is a guide to decision-making; pain is something that tells the animal: look, how bad is this situation? Is this situation so bad I need to get out of the shell, or is it actually not so bad, and even though I’m receiving this shock, I should actually stay in the shell because it’s a really valuable shell? So it’s a credible picture here of what pain does tie in with credible behavioural evidence of the crab actually behaving in that way.

That brings us back to the question of what to do in the face of this kind of evidence. Is this evidence enough to conclude that the crab is sentient? Is it enough to conclude that it should be brought within the scope of animal welfare legislation? Currently, crabs, lobsters, and other decapod crustaceans, as they’re called, are generally not included in reality welfare legislation; they are usually considered non-sentient. So we face this question that has a philosophical aspect to it: is this evidence enough? Is this credible indicator enough to warrant us bringing crabs and lobsters within the scope of animal welfare law? We need a framework to think about this kind of question.

What I’ve done in recent work is propose such a framework. The framework is based on the precautionary principle, which is quite an influential idea in environmental law where the thought is that when you have suggestive but not conclusive evidence, when there’s a lot of uncertainty, you have some evidence, but it’s not conclusive, but potentially a very serious threat if you fail to act on this evidence, then you should err on the side of caution, and you should act on the evidence anyway, and you should take appropriate precautions to mitigate the threat. This is an idea that’s very influential in the context of climate change. People say that even if the evidence for anthropogenic climate change is not conclusive, it doesn’t have to be conclusive because the threat is so serious that we should act on this evidence that we have now and take appropriate precautions to mitigate potential climate change.

I think this same principle, this precautionary principle, can actually be usefully applied in the context of animal sentience. When you have a case like the crab or the lobster, and there’s some evidence of sentience, but it’s not conclusive, but you know that if you carry on treating this animal as if it wasn’t sentient, if you carry on treating it in the way crabs and lobsters are treated today, often having their limbs removed alive, often being cooked alive, often being treated very very poorly because they’re not included within animal welfare law, when you have this situation, you can say, if we carry on treating this animal as if it’s not sentient, and it is, the consequences of that will be really terrible, the pain inflicted on the crabs will be enormous on a huge scale. On the other hand, if we err on the side of caution and give the animal the benefit of the doubt, then if it turns out that it’s not actually sentient and we were wrong, then we’ve protected it unnecessarily, perhaps we’ve regulated the way these animals are processed and cooked and so on unnecessarily, but that cost doesn’t seem so bad compared to the cost if we make an error in the other direction.

So I’ve argued that we should err on the side of caution here and we should be willing to attribute sentience to species like crabs and lobsters even when the evidence isn’t conclusive because the risks of not attributing sentience are so severe.

That’s the basic framework that I’ve proposed. There are a lot of details to be thrashed out there, details about the precise level of burden of proof that’s required before we say the evidence is now sufficient. I’ve suggested that the evidence we have at the moment for crabs and lobsters, just a few credible indicators of sentience, is actually enough, and if that’s right, then our whole approach to the way we treat crabs and lobsters should change.

But there’s room for debate here: others may argue that the burden of proof should be set a bit higher. I think the main thing is that we have this discussion; we don’t just think that animal sentience is something which scientific evidence can never bear. There is actually clear evidence that does bear on the question. It doesn’t conclusively resolve it, but it is relevant, and moreover, I think it’s enough for us to actually change the law and change the way we think about invertebrates like crabs, lobsters, octopus, squid, and cuttlefish. They’re really a lot more intelligent, a lot more sophisticated than we first thought.

Become a Patron!