III.2 Referentes
III.2.1 La didáctica en Europa a partir del estudio de la española Andrea
In order to make comparisons of welfare between different individuals, we must have a common scale onto which we can map their ‘units’ of welfare. That is, to be able to say something about how many measured units of welfare for one animal are equivalent to a number of units of welfare for another. Think of temperature measurement – although we have different temperature scales (Fahrenheit and Celsius) we are easily able to convert units in one into units of the other. If there is no such common underlying scale for welfare, or we cannot gain the required information to do the necessary conversions, then we will not be able to make intersubjective comparisons of welfare.
On the surface, it seems we have good reason to be sceptical of this possibility. It is entirely plausible that different individuals could experience vastly different levels of welfare, and that they do not reflect these differences in measurable indicators. We see versions of this in real- world situations – it seems, for example, that people can vary quite a lot with respect to pain thresholds and the degree to which they express pain reactions, and this can make it very difficult to compare pain experience between individuals. It may be the case that some animals
have reduced affect – that their highs are not particularly high nor their lows particularly low. The intensity of all their experiences may be small. Some animals, by contrast, might be capable of reaching far higher heights and far deeper lows – their intensity is just greater overall. If it is possible (which it seems to be) that such individuals could exist without showing different indicator responses, then as the underlying subjective states are private and inaccessible, we might never know whether or when they occur, and this undermines our ability to trust such comparisons. Making intersubjective comparisons of welfare will then require further justification. This problem has been widely discussed in the literature on human welfare, and the approaches taken there will be discussed in detail in Section 6.4.1. Here I will make the problem clearer through use of an example, and discuss the types of comparison the problem may apply to, before turning to my proposed solution.
6.3.1. An example
The problem of intersubjective comparison can be illustrated with this example. I used to work with two otters – Sneezy and Paddy. Imagine that each are given some yabbies, and their behavioural and physiological responses measured – say, the amount of vocalisation, and changes in heart rate. We see that Paddy shows a higher level of response on all measured indicators than Sneezy does. What conclusions can we draw from this? There are a few options: • Paddy enjoys receiving yabbies more than Sneezy does and her indicators reflect
this
• Paddy and Sneezy enjoy receiving yabbies the same amount, but Paddy is more inclined to ‘display’ her pleasure in measured indicators than Sneezy is
• Paddy enjoys receiving yabbies less than Sneezy does, but her reactions are much higher
There is also a possibility that Paddy actually dislikes the yabbies, and her indicators are instead demonstrating the strength of this dislike rather than enjoyment. Here, I will introduce the distinction between valence and intensity. Indicators such as heart rate and level of vocalisation may tell us how strong the reaction is (intensity), while there are other indicators, such as behaviour and type of vocalisation, that tell us whether the reaction is positive or negative (valence). Here, I will take it as fixed that the otters have the same indicators of valence – that is, that they will show the same signs of positive and negative experience. This is plausible, due to their shared evolutionary history and development. Here then, we are only
interested in whether the measured levels of response intensity correspond to the same intensity of experience.
The problem is, that there does not seem to be any evidence here that gives us reason to prefer any of these alternatives over the others. There are multiple conclusions, all equally consistent with the observations and there is not any reason to consider one of the possibilities more likely than the others. This is not just hypothetical – within-species differences in individual behavioural and physiological responses to positive and negative stimuli are common (e.g. Boccia, Laudenslager, & Reite, 1995; Izzo, Bashaw, & Campbell, 2011; Manteca & Deag, 1994), and it is difficult in these cases to determine whether or not results imply a welfare difference.
We cannot escape the problem using tests for the strength of preferences, as they will hold the same problem as other indicators. We could look to see whether Paddy will work harder for her yabbies than Sneezy would, but the results we see only give us comparative information for each individual – how much they value yabbies vs. work – and can’t be used to make intersubjective comparisons without assumptions such as that they both find work equally unpleasant. Sneezy may work less hard, but that won’t tell us whether he values yabbies less, or dislikes work more. He may merely be lazy, and less motivated in general to try and receive his favourite things, despite enjoying them just as much. Again, the information we have gives us no reason to prefer one of these options to the others, and there does not seem to be any further information we could gather that could give us such reason. Similarly, repeating the tests with a larger sample size would not give us further insight. Even if we had a test in which we saw 100 otters showing a higher response and 100 showing a lower response, we would still not have the required information in place to determine whether the higher responses were a result of greater enjoyment of yabbies, or of differences in reactivity.
The problem arises because we have two dimensions along which the animals can vary from one another. The first is in the degree or intensity of their subjective response to stimuli: something like their ‘capacity’ for pleasure or suffering. This includes both the level of their maximum and minimum welfare (scope) – how high their highs and how low their lows – as well as their usual response to stimuli – whether it might be overall more positive or negative than others. The second dimension is the relationship between the level of subjective response and the change in the measured indicators. Some animals may be highly reactive, showing large changes in their measured indicators to only small increases or decreases in their subjective experience. Others may be more circumspect, showing only small external
responses to large subjective changes. We have no way of testing for this possibility, and no a
priori reason to rule it out.
Another example that might be familiar to many is that of the emotional reactions shown by young children. Toddlers are famous for having meltdowns over what we would consider to be very small problems, such as the shape of their sandwiches. When a toddler throws a tantrum, what we don’t know is whether they are genuinely that deeply upset about the issue (i.e. as upset as we would have to be in order to show such a response), or whether they are just being particularly histrionic. Which of these interpretations we take will determine how we should react to the tantrums – should we show genuine concern for a deeply emotionally wounded child, or ignore an excessive display? Without further information to help us decide, either option is equally likely.
The overall problem is that under an observed difference in overall response, we don’t know which of these factors – difference in level of welfare intensity, or in indicator response - is responsible for this, or indeed if both are varying simultaneously. Without such information, we cannot make comparisons.
6.3.2. Types of intersubjective comparisons
There are different types of intersubjective comparisons that can be made, and although the general problem as described above will affect all of them, the degree to which it will become a problem for a situation will differ depending on the type of comparison required. The types of comparisons link to the types of measurement scale discussed in Chapter Four.
The three types of intersubjective comparisons are valence comparisons, level comparisons and unit comparisons (List, 2003). Valence comparisons are the most basic, and simply require an assessment of welfare as positive or negative with reference to some intersubjectively relevant zero point - "i's utility in state x is greater than/equal to/less than a utility level of zero" (List, 2003, p. 5). The zero point represents a ‘neutral’ point of welfare, and if we can assign the same zero point to each individual, then we have a shared basis from which comparison can be made. Different individuals can be compared with regards to where they stand relative to this zero point. We can then make claims such as: both individuals are experiencing positive welfare, or one individual is positive and the other negative.
The problem of intersubjective comparisons does not hold for valence comparisons. For any individual animal, we are able to use indicators to make judgements about the valence of its welfare - whether it is experiencing positive or negative welfare. This is done in reference to a neutral ‘zero’ line, at which welfare is neither positive nor negative. The exact characterisation
of the zero line, and how to identify or measure it, are complex and interesting questions, but will not be explored here. Here I will take what I hope is a somewhat intuitive characterisation of the zero line as the subjective absence of both the characteristic positive and negative mental states (or an equal balance of both), that can be observed through the absence of indicators for either of these. Intersubjective comparisons of this type are made possible by the fact that this zero line is the same conceptual point for all individuals. What it means to be experiencing neither positive nor negative welfare is the same for all individuals – it does not make sense to think of one animal as having a more intense neutral experience than another. So, if we can identify when any individual is in a positive or negative welfare state, we can make comparisons of that to other animals – animal A is experiencing positive welfare while animal B is negative. This allows some basic level comparisons, in that we can assume that all animals experiencing positive welfare are at a higher level than those experiencing neutral or negative welfare.
Level comparisons are more general judgements that the welfare of one individual is higher or lower than that of another individual; of the form: "Person i's utility in state x is at least as great as person j's utility in state y" (List, 2003, p. 1); or more formally: Ui(x) ³ Uj(y). This is
a ranking of welfare states of different individuals from highest to lowest – similar to an ordinal scale of measurement. This type of comparison gives us no information about the valence of or degree of difference between the two levels.
The problem of intersubjective comparisons does not hold for valence comparisons. For any individual animal, we are able to use indicators to make judgements about the valence of its welfare - whether it is experiencing positive or negative welfare. This is done in reference to a neutral ‘zero’ line, at which welfare is neither positive nor negative. The exact characterisation of the zero line, and how to identify or measure it, are complex and interesting questions, but will not be explored here. Here I will take what I hope is a somewhat intuitive characterisation of the zero line as the subjective absence of both the characteristic positive and negative mental states (or an equal balance of both), that can be observed through the absence of indicators for either of these. Intersubjective comparisons of this type are made possible by the fact that this zero line is the same conceptual point for all individuals. What it means to be experiencing neither positive nor negative welfare is the same for all individuals – it does not make sense to think of one animal as having a more intense neutral experience than another. So, if we can identify when any individual is in a positive or negative welfare state, we can make comparisons of that to other animals – animal A is experiencing positive welfare while animal
B is negative. This allows some basic level comparisons, in that we can assume that all animals experiencing positive welfare are at a higher level than those experiencing neutral or negative welfare.
Unit comparisons are the most detailed, being judgements about the amount by which the welfare of one individual is higher or lower than that of another, as well as the relative difference of welfare change that will occur when individuals change conditions. These are of the form: "If we switch from state x to state y, the ratio of person i's utility gain/loss to person j's utility gain/loss is l" (List, 2003, p. 1); or more formally: Ui(x) - Ui(y) / Uj(x) - Uj(y) = l.
Here we can compare the number of units of welfare experienced by different individuals in different situations, not just changes up or down, as in the quantitative (interval and ratio) scales. If it can be made, this type of comparison gives us more information about the magnitude of welfare differences, and is much more useful in many decision-making contexts. Comparisons of this type will require that measurements of welfare can be made to at least an approximate quantification of units; as mentioned in Section 6.1, this seems to be plausible through quantification of indicators.
The three applications of animal welfare comparisons described will generally require unit comparisons, though level comparisons might sometimes be sufficient. In making management decisions trading off current and future welfare, we will need to know the magnitude of the negative and positive experiences in order to compare. For use of animals in welfare science, we will at least need to know if some intervention has made a group of animals better or worse off than another, and often by how much. For management decisions about resource distribution, we want to know how much benefit we will be providing our different animals through provision of resources, in order to determine the optimal distribution.
Both level and unit comparisons, however, do seem to fall to the problem of intersubjective comparisons. In both cases, all we have is information about the conditions the animals are experiencing, and the indicators they are showing, but for the reasons laid out earlier, we cannot simply assume that animals are responding similarly to the conditions, nor that their indicators reflect similar levels of subjective experience. We may have slightly more justification in using level comparisons, at least in cases where the difference in condition or response is quite vast, but even these will rely on some similarity assumptions, which I will address shortly. The assumptions required to justify more detailed unit comparisons will be even more stringent.