Animal sentience workshop

[This video was filmed during a workshop in front of an audience. The speaker stands at a speaker’s podium and talks into a microphone. Sometimes, she looks or points to a screen with slides containing information to support her talk. The slides aren't shown in the video.]

Ngaio Beausoleil: I think I win the prize for the vaguest and most generic title. And I've been quite deliberate in that so I can say whatever I want now. So over the next 14 minutes and 59 seconds I'd like to make four key points.

The first point is that in acknowledging sentience I think what we're doing is steering ourselves towards the affective state orientation to understanding and assessing animal welfare. The second thing I want to talk about is the importance of using specific descriptions of the unpleasant or negative experiences that animals have, if we want to accurately assess animal welfare. The third point seems relatively obvious but we have actually had quite a bit of discussion about this already today, this is the idea that the absence of unpleasant or negative experiences, we now think is probably not sufficient for animals to have a good welfare status - we need more than that. And then finally I want to finish up by talking about what you guys have just been working on, the ways in which using the five domains model is valuable because it emphasises the affective states that animals have and encourages users to be both specific and comprehensive when they're undertaking welfare assessments.

So I'm not going to define sentience we've had some discussion about that, I want to make two main points and I think that these are assumptions that underpin the rest of what I'm going to say. The first is that sentience is really considered to be a prerequisite for animals having welfare to be considered at all. The second is that because sentience relates to the animal's capacity for mental experiences, our explicit acknowledgement, that at least some animals are sentient in our law, really steers us towards this affective state orientation to understanding animal welfare. And what I mean by the affective state orientation is a way of thinking or characterising welfare as being a state within the animal itself, that arises from the integration of all the different kinds of mental experiences both positive and negative that animals have at any particular point in time.

And I want to just contrast this with other ways that people commonly think about animal welfare. So another way that people think about welfare is in terms of the biological functioning orientation. And so this is where we think about animal welfare relating to the health status and the productivity of the animal, if it's healthy and producing well then probably we're okay with its welfare state. In contrast, another way of thinking about welfare is using the naturalness or the natural living orientation. And according to this orientation animal welfare relates to the naturalness of the environment that we provide and the degree to which animals can express at least most of the repertoire of their natural behaviours. So if we just look at the difference between these orientations the affective state orientation obviously as the name suggests has to do with what the animals feeling, what matters to it, how it experiences its world. The other ones really are about the physical state of the animal or the inputs that we provide so the environment we provide for example and they don't necessarily take us to the point where we're thinking about what the animal is experiencing. So the second point - if we agree that sentient animals are those that can have at least unpleasant experiences, I want to now talk a little bit about the importance of using specific terminology to both describe and in our assessments of unpleasant affective experiences.

And we've heard these words coming up today. So if we think about the way the animal welfare has been assessed in the past, and actually really commonly even today, those evaluations often centre on the absence of unpleasant things, and we hear the use of these terms all the time – the absence of pain and distress, the absence of pain and suffering, the absence of fear and distress. So we have these generic kind of composite terms, often pain and distress or suffering, and I would argue that these terms are really commonly used because we consider them to be useful for covering all our bases. We want to make sure that we catch all the unpleasant experiences that animals might have that would impact negatively on their welfare. But I would argue that in reality that using these kind of catch-all terms, like suffering and distress, actually can limit our accurate assessment of animal welfare. And there's a number of reasons for this belief and I want to just run through those quickly.

So, the use of generic terms like suffering and distress I think can be problematic because it doesn't help us understand how those unpleasant experiences arose. If you look in the literature you'll see that terms like the stress and suffering are rarely defined and that's because it's really hard to do so, we don't actually know what they mean. If we do find a definition both those terms, distress and suffering, tend to be used to indicate the badness of a primary unpleasant experience. So either it goes on for a very long time, it's a prolonged badness or it's a very intense badness. When something is prolonged or intensely bad we start to talk about distress or suffering. The problem is it doesn't actually tell us what kind of badness is prolonged or intense. So related to that first point, if we use terms like suffering and distress that doesn't help us to develop and to implement targeted solutions. In other words if we don't know what the badness is we can't fix it. We have analgesics to deal with pain. We have antiemetics to deal with feelings of nausea. We don't have any anti-distress pill. There's no such thing as an anti-suffering elixir. So we don't know how to fix it. The third problem with using terms, composite terms particularly, like pain and distress or pain and suffering is that it leads us to under emphasise the importance of other experiences. So unpleasant experiences that are not pain, they're qualitatively dissimilar from pain, but they are unpleasant and impact on animal welfare nonetheless. So things like breathlessness, nausea, sickness, fatigue and maybe if you believe in them, boredom and frustration for animals.

So we overwhelmingly put the emphasis on pain in many contexts, and we've heard that today, we do it naturally all of us do, and we see this not only in animal contexts but also when we're talking about other people. So I gave this talk a while ago and I had someone come up to me afterwards and say, this has happened to me, I had an elderly relative who was in a palliative care situation and we were repeatedly reassured by the doctors that this person was not in any pain. Well the person had a respiratory disorder and was probably not in pain, but the problem was that the main unpleasant symptom they were experiencing was probably respiratory discomfort or breathlessness but nobody was reassuring the people about that, the focus was on pain. And this potentially misguided focus can lead us to actually fail to systematically look for those other unpleasant experiences, and in fact a failure to recognise evidence that they're occurring even if they are right in front of us. So we just don't see what we're not looking for.

And I just want to quickly summarise this issue I call this the no pain no problem fallacy, but one of my students came up with a much better way of describing it which is the 'pain et al.' problem. So for anybody who's ever published a paper or knows the value of publishing a paper, we start off with a paper that was written by Beausoleil, Mellor and Johnson. The first time that paper is cited, its cited as such Beausoleil, Mellor and Johnson all get a mention. The next time it's cited it becomes Beausoleil et al. Now we have pain and distress - pain and its friends; Beausoleil et al. The third thing that happens is that I go to a conference and I talk about this fantastic paper that we wrote and suddenly everybody's talking about did you hear about Beausoleil's fantastic paper. So we've now gone from Beausoleil, Mellor and Johnson to Beausoleil. We've gone from pain, breathlessness, nausea, fatigue to pain. We look for evidence of pain, we don't find evidence of pain, no pain - no welfare problem. But we know that that's not the case. So moving on to the third point, and this is one that we've talked about quite a bit today which I think is really fantastic. This is simply the idea that if we want to provide acceptable or good welfare for animals or quality of life or life worth living, simply avoiding or mitigating unpleasant experiences is not sufficient. We have to now think about providing opportunities for positive experiences as well. And Ian's really led to this really nicely in terms of our thinking about responsibilities for those who are in charge of animals. What are those responsibilities in terms of providing opportunities for positive experiences? And I think that this changing paradigm, a changing way of thinking about providing appropriate animal welfare, is going to throw up a huge number of challenges, some of which have been alluded to already.

So I just jotted down some of the things that came to my mind from a scientific perspective. And the first one was about indicators. So we have indicators of some, at least some, unpleasant experiences that are relatively well understood and we feel that we can recognise those experiences pretty well, particularly the ones where if you don't do something about it the animal will die - the survival-critical ones. But we're still relatively early in our scientific understanding of indicators of various positive experiences, and in fact the problem is that even if you don't give the animals positive experiences they tend to survive and sometimes they're still really productive. The other things that came to my mind is I think there's gonna be a lot of debate, and this is probably what we've talked about already today, how much positive experience or how many positive experiences are required to meet the threshold for good welfare? And that becomes a legal issue which is really important. And another question might be, can we actually offset some unavoidable negative experiences by providing opportunities for some positives?

So the final point I want to make, is about the five domains model which you've all had an opportunity to have a play around with, and this is the point really that the five domains model I think can help emphasise the importance of sentience when we want to assess animal welfare. So for those of you who were not previously familiar with the model, I just want to give you a quick run-through of our latest version and our latest thinking about how the model might be applied. So as the name suggests the model has five domains, there are four physical or functional domains, and you can see these listed on the left hand side here, and we have one affective experience domain which is about mental experiences. So the four physical or functional domains  - domain one's about nutrition and hydration, domain two is about the environment physical and sensory environment, domain three is about health and functional status and the fourth domain we refer to is behaviour and I sometimes talk about this one as being interaction both interaction with the environment itself and interaction with other animals including people.

So just to add to that within each of those physical functional domains, as you guys have just done, we talk about restrictions and problems. These can then aligned to specific unpleasant or negative mental experiences in the fifth domain. In each of the four physical domains we also have opportunities or solutions and those again would align with specific positive mental experiences in the fifth domain. So what we do using the model is we use scientific indicators to evaluate the occurrence of those physical restrictions and problems, or opportunities and solutions. And the really key point I want you to take away, if you want to apply this model, is that we put things that we can observe, measure, quantify - the tangible things - go in the physical and functional domains. Behavioural, physiological those kinds of indicators go in the first four domains. Some of those will be animal based indicators, so things that animals do in response to what we provide, and some of them might be resource or management based. So there is a bucket of water there for example. We use that body of information, which we've collated in the first four physical functional domains, to then carefully interpret what those indicators are telling us in terms of the animal's affective experiences. So we cannot measure, we cannot quantify mental experiences, they are by definition internal, subjective, intangible, only the animal knows what they feel like. So we can never measure those. All we can do is use the information from the things that we can observe and measure to make careful scientific inferences about what the animal might be experiencing. And that's a really key point about the five domains, you have to go to the fifth domain to understand what all your evidence means to the animal itself.

So, I would propose that the five domains model is really valuable because of its structure. And the reason I say that is because the structure of the model encourages users to do what I've just said. To actually go from what we can observe and measure and quantify, and then carefully, cautiously, using the available science, interpret what it means to the animal itself. What it means in terms of the animal's mental or affective experiences. And some of those other orientations to animal welfare like the naturalness one and the physical state, the biological functioning orientations, don't encourage moving that next step forward - they stop in the physical functional domains.

So just to quickly show you some examples, I don't know if you can read those, but the point I want to make here is just this is some examples from the third domain about health and functional status. Injury is important because it causes an unpleasant experience, it causes pain to the animal and we use the behavioural and physiological indicators to infer that the animal is experiencing pain. Likewise if the animal breathes harder and faster because it has a respiratory disease, it matters because the animal is experiencing some kind of unpleasant breathlessness.

The model is also useful because its structure encourages users to look for evidence, or evidence of absence, of a wide range of specific negative experiences. And again this is just an example if you can't read it really all I want you to see is that there are a whole lot of red examples in domain five. These are various negative experiences, they are specific, we're not looking for evidence of distress or suffering because what are they and how do recognise them? We're looking for evidence of specific things like thirst, breathlessness, pain, boredom and thermal discomfort. So the Model acts as what David calls an aide memoire and basically it encourages us to look to see which of these are actually evidenced by what the animal is doing and what we're providing. Likewise the structure of the model actually encourages us to look for evidence for or against the presence of a wide range of positive experiences. And it's also valuable because it can help suggest ways that we might provide opportunities for animals to have positive experiences that we wouldn't otherwise have thought of.

And again this is just showing the wide range of positives, and this is not comprehensive, just reminding us what to look for. So just to conclude before I get pulled off the stage, my first point was that the affective state orientation to animal welfare is consistent with our new explicit emphasis on sentience because both of them focus on the animals mental experiences. Animal welfare assessments should interpret the observable indicators in terms of the specific affective states, in other words what matters to the animal itself, and we should look for evidence of a wide range of both the negative and positive experiences. And because of its structure the use of the five domains model supports the affective state orientation and it encourages users to be both specific and comprehensive when evaluating animal welfare. Thank you

[Video ends]