3. UTILITARIANISM AND ANIMAL SUFFERING
In this chapter I shall begin considering what a utilitarian should say about the moral standing of animals. I shall confine my attention to the question of the moral standing of animal experience (particularly pleasure and pain), reserving to Chapter 4 discussion of utilitarian approaches to the value of animal life.
Racism, sexism, and speciesism
Peter Singer has been prominent in arguing for the moral standing of animals and animal suffering, through such books as Animal Liberation and Practical Ethics. He does not, in fact, explicitly premise his argument on any version of utilitarianism. For he wants that argument to be acceptable to all, whatever their theoretical stand-point. His strategy here is a good one. Any moral argument will be the stronger for being able to survive translation between ethical theories, being equally stateable in a variety of them. I shall show later that Singer’s argument is only really acceptable from a utilitarian stand-point, however - in particular, that it has no force against a contractualist.
Singer’s argument starts from a principle of equal consideration of interests. This holds that in any situation the interests of all those affected should be considered equally, which may sound, on the face of it, like a utilitarian principle. But in fact Singer is correct that, suitably interpreted, it should be equally acceptable to contractualists. For contracting agents may reasonably reject any rule that gives their interests no weight, or that treats those interests as having less significance than those of other agents, as we shall see in Chapter 5. It seems plain, in contrast, that no one could reasonably reject a rule requiring us to treat everyone’s interests as of equal weight - which is the principle of equal consideration of interests.
To say that everyone’s interests should be given equal consideration is not to say, of course, (for a contractualist at least) that everyone’s interests should be met equally. Much will depend upon the circumstances. If some of us have freely agreed to pay money into a lottery, for example, then only those who have contributed may be considered for a prize. This does not conflict with the principle of equal consideration, since the others are not discriminated against - they were free to take part if they wished. What would genuinely conflict with the principle of equal consideration would be, for example, a policy of only considering for prizes those with white skins, whether or not they have contributed to the lottery.
Singer’s explanation of the immorality of both racism and sexism is that these practices violate the principle of equal consideration. For example, the policies of the South African government through most of the decades of this century have counted white interests as worth a lot, and the interests of blacks and coloureds as being comparatively much less significant. Similarly, the policies of many governments and individuals around the globe treat the interests of women as worth less than those of men. These policies are wrong, and count as violating the principle of equal consideration, because the characteristics of skin colour and sex are morally irrelevant ones. While the fact of having contributed money to a lottery is morally relevant when it comes to the distribution of prizes, the fact of having white skin, or being male, is plainly not. Indeed, there are hardly any important contexts in which these latter features could be morally relevant. (Exceptions might be the distribution of creams to prevent skin cancer in whites, and screening for testicular cancer in men.)
Singer’s main argument is then that speciesism, like racism and sexism, is wrong because it discriminates on the basis of morally irrelevant characteristics. To give lesser, or no, weight to the interests of animals violates the principle of equal consideration because such a policy must be grounded in either the mere differences of species between animals and ourselves, or in the differences of appearance, or in the differences of intelligence. Yet none of these characteristics is morally relevant, Singer argues. Let us discuss each of them in turn.
It seems plain that species membership is a morally irrelevant characteristic. Two examples will suffice to make the point. First, suppose that the experiments attempting to teach language to chimpanzees had been successful beyond their originators wildest dreams. The apes in question gained a complete mastery of English within a few years, were able to attend school and later university, and made many close friendships with human beings. In these imaginary circumstances it would plainly be absurd to claim that the apes lacked moral standing, or had a moral importance that was lower than our own. At any rate, to make these claims would plainly be morally objectionable in exactly the way that similar claims made about members of different races are morally objectionable. So species membership, in itself, cannot be a morally relevant characteristic, serving to justify differential treatment of animals.
Consider a second example to make the same point. It is well known that about ten per cent of human couples are infertile. Then suppose it had been discovered that the reason for this is that human beings in fact consist of two distinct species, otherwise hardly distinguishable from one another, the members of which cannot inter-breed. In these circumstances it would plainly be objectionable for the members of the majority species to attempt to withhold moral rights from the members of the minority, on the mere ground of difference of species. This, too, would be obvious speciesism.
Now consider two examples to make the point that differences of appearance between humans and animals cannot serve as a moral ground for differential treatment of the latter. First, consider human beings who have been victims of the drug Thalidomide, taken by their mothers during early pregnancy. These people often appear very different from normal humans, with perhaps no legs and just a few fingers sprouting from a shoulder. But that is plainly no ground for treating their interests as being lesser than our own. Second, suppose that there were to be radiation victims following a nuclear accident who, while otherwise normal human beings, were born covered with thick dark fur like monkeys. Again these people’s difference in appearance from ourselves would plainly be no ground for refusing to count them equally.
Finally, consider two examples to make essentially the same point in connection with differences of intelligence between animals and ourselves. Suppose that doctors with a limited supply of kidney dialysis machines were to begin intelligence-testing their kidney patients, only offering treatment to those scoring above a certain level. Would there not be an immediate moral out-cry, and rightly so? It is plainly morally repugnant to make life-or-death decisions on the basis of intelligence. Similarly, suppose that a cosmetics company were to begin testing their products in a home for severely retarded children, using the same painful experiments currently employed on animals to ensure that their products are safe. Again the out-cry would be immediate. The sufferings of those children could not be ignored merely on the grounds that the children are of lower intelligence than ourselves.
The conclusion Singer draws from considerations of this sort is that it is morally indefensible to exclude animals from the scope of the principle of equal consideration of interests. Since the various characteristics that distinguish us from animals - that is, species, appearance, and intelligence - are morally irrelevant, animal interests should be counted equally with our own. Pain is pain, no matter who feels it, and is just as morally significant. I shall shortly discuss one of the presuppositions of Singer’s position - that animals do genuinely have interests to be considered - and what the practical consequences of his conclusion would be. But first I shall dig a little deeper into the notion of a morally relevant characteristic.
The relativity of relevance
My first thesis is a general one, that relevance is always relative to a point of view. Imagine that Tania and Teresa are watching a tennis game, and are asked whether or not it is relevant who wins. Tania might reply that it is not, since she just likes to watch good tennis. Teresa, on the other hand, might reply that it is, since she has a substantial bet riding on the outcome of the game. What is relevant to the one is irrelevant to the other, in virtue of the different perspectives that they take towards the game - the different kinds of interest that they have in it. So when it is claimed that species membership is a morally irrelevant characteristic, we need to know the point of view being taken, in order to assess the claim. That is, we need to know how the moral point of view is being characterised. Once the matter is viewed in this light, it can be seen that Singer’s argument is, in fact, only sound from the stand-point of a utilitarian conception of morality.
Given a contractualist conception of the moral point of view, intelligence - or at least a certain kind and level of intelligence - is not morally irrelevant, as we shall see in Chapter 5. On the contrary, for a contractualist it will turn out to be a sufficient condition for a creature to have moral standing that it should be a rational agent, and this is, broadly, a matter of its intelligence. This explains the appeal of many of the examples used above, such as the example of the English-speaking apes, and the examples of the Thalidomide and radiation victims. For in each case it is plain that the individuals in question are rational agents. To say that rational agency is morally relevant under contractualism, however, is not to say that differences of intelligence amongst rational agents must also have moral relevance. On the contrary, one would expect rational contractors to outlaw discrimination on the basis of such differences, of the kind involved in the example of the kidney dialysis machines discussed above. For those who are of lower intelligence might surely be reasonable in rejecting any rules that allow their interests to be discounted, or counted for less.
It is also the case that species membership, together with the similarities of appearance and patterns of behaviour with which it is associated, is not morally irrelevant under contractualism - at least if the arguments to be presented in Chapters 5 and 7 are sound. There I shall argue that rational contractors should extend direct moral rights to all members of the human species, in order to avoid the dangers of a slippery slope and to preserve social stability, and in order not to undermine our natural reactions of sympathy for human suffering. Since these arguments do not support the extension of direct moral rights to members of other species, it will turn out that species membership is a morally relevant characteristic from the perspective of contractualism.
From this discussion it emerges that what really drives Singer’s argument is a particular conception of the moral point of view, which identifies it with the stand-point of an impartial benevolent observer, who is equally sympathetic to the interests of all who are affected by a given action or situation. It also emerges that there is nothing really driving Regan’s argument in The Case for Animal Rights, which makes similar claims about the moral irrelevance of species membership and intelligence. For as we saw in Chapter 1, Regan fails altogether to provide us with a characterisation of the moral point of view, relying entirely, as he does, on his - restricted - understanding of reflective equilibrium to establish his views. Since we lack such a characterisation, we lack the means necessary to assess his claims about moral relevance.
The stand-point of the impartial benevolent observer is, you will recall from the last chapter, the governing conception of utilitarianism, which regards moral concerns as arising out of rationalised (impartial) sympathy. It is certainly true that there can be no reason why an impartial observer should count animal interests as of lower importance than our own. It is also true that rational agency and species membership are both morally irrelevant, from the stand-point of such an observer. The only relevant features are the capacity for pain and pleasure, as well as the capacity for desire. Under utilitarianism the boundaries of moral concern are co-extensive with the boundaries of sentience - that is to say, with the capacity for experience. If an animal can suffer, then, plainly, it may be said to have an interest in avoiding suffering. (Even if you think that to have an interest in something implies, strictly speaking, a desire for that thing, this point will still hold. For the very idea of pain seems conceptually tied to the desire for its avoidance. If an animal can suffer at all then it must have at least this one desire.) The principle of equal consideration of interests will then require us to show equal respect for the sufferings and frustrations of every sentient creature.
Singer’s argument for extending the principle of equal consideration of interests to animals is therefore less powerful than he would have liked. In particular, that argument can have no force against someone who is already a convinced contractualist. It is, indeed, an argument from the stand-point of utilitarianism. What we are therefore investigating is what a utilitarian should say on the subject of the moral standing of animal experience.
Do animals have interests?
What a utilitarian (or, indeed, anyone else) should say on this subject is partly dependent upon the facts. I have been assuming, thus far, that animals do have experiences and at least some desires. In which case, the only issue arising when we consider whether the principle of equal consideration of interests should be applied to animals is the moral one. But many have denied this. Many philosophers and psychologists have held that animals are biological automata, lacking mental lives, and engaging in their characteristic behaviours out of acquired habit or innately determined action sequences, rather than because of anything resembling genuine cognition.
There is no doubt that some animals are, in the relevant sense, automata, despite their apparent sentience. When caterpillars hatch from their cocoons, for example, they will climb to the tops of trees to eat the leaves at the tips of the branches. But this apparently purposive behaviour is, in fact, a tropism - a mechanical feedback process of a very simple sort. The caterpillars have two eyes, symmetrically positioned on their heads. When the same amount of light enters each of the eyes the caterpillars move straight ahead. But when more light enters one of the eyes the legs on that side of the body move more slowly. The result is that the caterpillars move towards the light. In experiments where the trees were artificially lit from below, the caterpillars moved to the bottom of the trees, where they remained even when starving. When a caterpillar was blinded in one eye it would move endlessly in a circle, again to the point of starving.
Caterpillars will also wriggle vigourously when impaled on the end of a pin. It seems highly likely that this, too, is a simple tropism. Although it might appear to a human observer that the caterpillar is in pain, and is wriggling in an attempt to avoid the source of the pain, in reality it is likely that the nerves sensitive to the presence of the pin feed directly into the muscles responsible for the subsequent movement, without any intervening cognition. Compare this case: you are present at a medical examination of your daughter Patricia, and you observe the doctor tap her on the knee with a hammer, upon which Patricia kicks out her leg. An untutored observer might conclude that Patricia had tried to kick the doctor because she felt pain at the blow. But you would know that it was, in fact, a mere reflex. So, too, I suggest, in the case of the caterpillar.
It seems unlikely that insects are genuinely sentient, in the sense of having mental lives that would include sensations and desires. It is worth noting that this is already to undermine one aspect of common-sense belief. Children who pull the wings off flies, or the legs off ants, are told that it is cruel, and to desist. In most cases their actions are presumably believed to be instances of what Regan calls brutal cruelty (as opposed to sadistic cruelty) - that is, actions that display indifference to the suffering caused to others. Yet these beliefs are false, if insects are not genuinely sentient. Once we realise that insects feel no pain, the only remaining motive for discouraging the children is that their activities are a sort of play-acting for real cruelty. But in fact it would be equally effective to teach children to distinguish between sentient and non-sentient creatures - supposing, at least, that the distinction were known.
What, then, are the boundaries of sentience? What sorts of creature are genuinely capable of having pains and other experiences, and of having beliefs and desires? I shall concentrate on the capacity for experience here, returning to the question of animal beliefs and desires in Chapter 6. It seems safe to assume that all mammals, at least, are genuinely sentient, given the variety and flexibility of mammalian behaviours, and given the close similarities in brain structure and function between even the lower mammals and ourselves. A variety of types of evidence also suggest that birds should be classed together with mammals in respect of levels and degree of cognitive organisation, and in contrast with lower vertebrates such as fish, amphibians, and reptiles. At any rate, this is what I propose to assume - that all mammals and birds are genuinely sentient, but that no insects are. For present purposes I shall remain agnostic about the mental lives of lower vertebrates, since we have already done enough to settle most of the cases relevant to the morally contentious practices of hunting, farming, and animal experimentation. I shall assume that in all such cases involving mammals and birds the animals can experience pain, and thus do genuinely have interests to be considered.
Minds and brains
Many of those who have maintained that human beings are unique in the animal kingdom in possessing mental lives have done so, in part, because they have denied that mental events and (some) brain events are one and the same. These people have either held that humans have non-physical souls, which form the true locus of their thoughts and feelings, or they have believed, at any rate, that mental events are non-physical ones, partly caused by and partly causing brain activity. Such people will then be unmoved by the manifest similarities of structure and function between the brains of humans and higher vertebrates.
It is not strictly necessary that we should reject the doctrine of mental immaterialism in order to maintain that higher vertebrates have genuine mental lives. For we could hold that the considerable similarities in behaviour between such animals and human beings would warrant ascribing non-material mental events to both. But our case will be that much stronger if we can also argue, as mental materialists, that the similarities between our brains suggest corresponding similarities in cognition. It is, therefore, worth indicating briefly why I hold that the thesis of mental immaterialism is false.
There are two main arguments for maintaining that our mental lives must consist of physical events in our brains. Both are premised on the common-sense belief that mental events and brain events interact causally with one another. We believe, for example, that retinal stimulation causes certain brain events that cause in us visual sensations in turn, and that mental events such as decisions cause bodily movements that have certain brain events as their immediate causes.
The first argument against mental immaterialism is that, if mental events are non-physical, then we shall be required to recognize a whole new species of causality, hitherto unknown to science. All the types of causal relationship that science has given us reason to believe in - chemical, electrical, mechanical, and so on - relate different classes of physical event. Indeed, it can be said that the distinctive feature of scientific progress over the last three centuries has been the assumption that there will be some sort of physical mechanism underlying any causal relationship. Science really began to make advances when people ceased to offer explanations in terms of causation by spirits and other non-physical forces, and began to hunt for physical mechanisms underlying the observed regularities in nature. The immense success of science then gives us reason to extend this policy into the domain of the mind until proved otherwise. But in fact, far from there being any proof of mental immaterialism, the arguments for it are relatively weak.
The second main argument against mental immaterialism, and in favour of identifying mental events with physical ones, is that enough is already known about the brain for us to be confident that each brain event will have a sufficient physical cause. We know that the human brain is made up of nerve cells, and quite a lot is known about the causes of nerve cell activity. All of these causes are physical ones, including chemical changes in the blood stream, as well as the physical activation of connected cells. Given that this is so, there is then no room for mental events (such as decisions) to cause brain events (in this case the immediate causes in the brain of bodily movements) unless mental events can be identified with the relevant brain events. If decisions (and other mental events) can be causes of bodily movements, then decisions must be brain events. For we know that bodily movements are caused by brain events, each of which, in turn, has a sufficient physical cause.
It is worth emphasising in conclusion of this section that mental materialism has nothing whatever to do with materialism as a system of values. Nothing in what has been said here commits us to the claim that the only things worth caring about are wealth, power, and physical comfort. Nor need there be anything here that is inconsistent with theological beliefs in life after death. For it is possible to believe in resurrection of the body, as many Christians have, in fact, done. Indeed, there might also be other forms of after-life available for a mental materialist to believe in.
Comparisons of interest
I have been arguing that members of many other species of animal, at least, should be counted as having interests, because they are capable of experiencing pain. But what does the principle of equal consideration of interests amount to, where the interests to be considered may cross species boundaries? Are such comparisons of interest even so much as possible? Similar worries to these can, in fact, be raised in connection with comparisons of human interests. For it is possible to doubt whether we can ever know the extent to which other people are really suffering, or even whether they are suffering at all. But this is just the philosophical problem of knowledge of other minds, raised for the particular case of knowledge of other people’s experiences. Although this problem may be a theoretically interesting one, few doubt that it must have a solution. That is to say, few are genuinely sceptical of our ability to have knowledge of the mental states of other persons, based on our observations of their behaviour.
If this is so, then it would seem that essentially the same basis must exist for knowledge of the mental states of animals. It is true that in connection with the mental states of human beings we have one additional source of evidence, namely those people’s descriptions of the qualities and intensities of their experiences. But it is important to see that this is just one further piece of behavioural evidence, having no special authority. For we still need to make assumptions about the speakers’ sincerity and, more importantly, about what they mean by the words that they use. These, in turn, can only be known by means of an inference to the best explanation of the observed patterns in those speakers’ behaviour.
There are two different sorts of basis for our judgements of the extent of an animal’s suffering. First, we can judge the intensity of that suffering from direct observations of the animal’s behaviour - that is, from the degree of the animal’s reaction (screaming or howling, for example), and how desperately it tries to avoid the source of the pain. Second, we may judge the intensity of the suffering by making reasonable hypotheses, on the basis of our observation of other similar cases, about what the animal would be prepared to do to avoid the suffering in question. Would it be prepared to endure a painful stimulus like that in order to get food when it is extremely hungry, for example?
Physiological differences between different animal species, of course, and between animals and ourselves, rule out any simple comparisons of suffering. To use an example of Singer’s, it is doubtful whether a hard slap to a horse would cause it as much pain as would a similarly hard slap to a human baby. This is because the horse’s skin is much thicker than that of the baby. But as Singer correctly says, there must be some degree of stimulus that would cause as much pain to the horse as does the slap to the baby. And we would judge this by seeing how the horse reacts - how hard it tries to get away, for example - and by seeing how much the horse would put up with to get something that it really wants, such as water when it is dehydrated.
When considering comparisons of suffering, Singer concedes that levels of intelligence do in fact turn out to be important. In particular, the greater intelligence of most humans gives vastly increased opportunities for suffering. For example, imagine the varied sorts of suffering that would be involved in a policy of random seizures of people off the streets, for use in painful eye-tests of cosmetics. In the first place, of course, there would be the immediate pain of the tests themselves, presumably roughly comparable to the levels of pain experienced by laboratory rabbits. But then in addition, there would be the fear beforehand, when people know that these seizures are taking place, and when they know exactly what is going to happen to them once they are seized. There would also be the memories remaining afterwards, perhaps with an attendant destruction of the self-esteem of the person in question. For these reasons Singer concedes that if these experiments have to be done at all, then it is better that they should be performed on rabbits than on human beings, since the suffering produced will be less. (He assumes, here, that most animals will lack the sorts of higher mental processes of thought and feeling that give rise to the kinds of additional sufferings mentioned above. I shall grant this assumption for the moment, returning to discuss it in some detail in Chapter 6.) This concession of Singer’s is consistent with the principle of equal consideration of interests, since it is only equal suffering that should be considered equally.
As we have seen, Singer allows that cross-species comparisons of interests are very difficult to conduct with any degree of accuracy, and that the greater intelligence of most humans vastly increases their capacity for suffering. But he also claims that we only have to be able to make the very roughest of comparisons in order to have an enormous impact on current practices in the treatment of animals. Let us consider in turn the four main categories of hunting, factory farming, cosmetic testing, and medical testing.
People who go hunting often consume the meat of the animals that they kill, and they often wear or sell their skins. But it is arguable that these benefits should not enter into the moral equation at all, since animal pain is not really necessary to provide them. In our modern world meat and furs can be obtained by normal farming, which need involve no suffering for the animals. Such animals can in principle be kept in pleasant conditions throughout their lives before being slaughtered painlessly and unexpectedly. (Recall that the question of the moral standing of animal lives is being deferred to Chapter 4.) So the only relevant advantages to humans from hunting are the pleasures of the hunt itself - tracking, stalking, or chasing an animal, and then trying to kill it. It is in the nature of this activity that it cannot be carried out without frequently causing pain to the animals involved.
Now although the pleasures of hunting, for some people, may be considerable, they are surely trivial by comparison with the painful death frequently endured by the animal. If the pain and terror caused by a fatal wound in a deer or rabbit is even roughly comparable to what a human would feel in similar circumstances, then it is obvious that they greatly outweigh any pleasure felt by the hunter. For consider: would even the most dedicated of hunters pursue their sport, if they themselves had to endure suffering comparable to that of each animal they failed to kill cleanly? It seems plain that they would not. In which case, applying the principle of equal consideration of interests here will show hunting to be wrong.
Now consider the practice of factory farming. Here again the suffering caused to the animals, through being kept in extremely cramped and unnatural conditions, is considerable. Yet the only gain to humans is that we should enjoy cheaper (and in some cases, perhaps, tastier) meat. So for each animal that suffers, the proportion of its suffering that is caused by factory farming, throughout the course of its life, has to be set against the marginal pleasures of the dozens of humans who eventually have a share in consuming its flesh. If that animal had not been factory farmed, then the only loss to those people is that they would each have had rather less money to spend on other things.
In this case, too, the application of the principle of equal consideration of interests seems easy. The gains to human beings - even when totalled up - appear trivial when compared with the extensive suffering of the animals. In which case factory farming will be wrong from the stand-point of a utilitarian. It is important to note, however, that this does not yet justify moral vegetarianism (as opposed, for example, to vegetarianism adopted for reasons of health). For some utilitarians (Singer included) hold the view that while it is wrong to cause animals suffering, it is permissible to kill them painlessly. So farming methods where the animals are kept in enjoyable conditions throughout their lives before being painlessly killed for their meat, may turn out to be morally unobjectionable. What utilitarians should say about this will depend on their views on killing in general, and on the value of animal life in particular, which we shall discuss in the next chapter.
The case against cosmetics testing on animals seems equally clear-cut. For the tests are such that the animals in question suffer very severe pain, whereas the gains to humans of being able to use a new cosmetic are marginal. Now it might be replied that in an age of mass-production, even fairly minor pleasures brought to millions of people might easily outweigh the intense sufferings of a few hundred animals. But the pleasures in question are very marginal indeed. For there are already a wide range of cosmetics products in existence. The only cost of banning testing now, would be the loss of the pleasure that some people feel on being able to try something completely new.
There is, however, a point about those who are employed in the cosmetics industry, many of whom might lose their jobs if testing of new products were banned. (This issue does not arise in connection with factory farming, since traditional farming methods are more, rather than less, labour-intensive.) Here the issue is in danger of merging into complex questions of economics and social policy. But as a corrective to it, try to take seriously the application of the principle of equal consideration of interests to animals. For if it were young children rather than animals who were involved, for example, then how many people could seriously place their possession of a job higher than the option of avoiding the suffering that their jobs produce?
Finally, consider the use of animals in painful scientific experiments, particularly those connected with the development and testing of new medicines. This is in many ways the most difficult case, because of the very considerable gains that may result from such experiments, through reducing or avoiding the incidence of painful illnesses in both humans and animals. Might not these benefits outweigh the sufferings of the animals used in the tests? In some cases, surely, they might. But in order to justify any particular series of experiments, we would need to have an assurance that the probability of such benefits accruing is quite high. A mere chance of great benefit will not be enough, when set against the certainty of the suffering caused to the animals involved. In any case, what Singer suggests here as a useful rule of thumb, is that such tests are morally acceptable only if it would be equally acceptable to perform them on mentally retarded orphan humans. (They should be orphans to rule out the question of vicarious suffering caused to parents and relatives.) If this would not be acceptable (as presumably it would not, for most people), then from a utilitarian stand-point it can only be unacceptable speciesism to allow the tests in question to be conducted on animals of similar intelligence.
Is reflective equilibrium possible?
Utilitarianism is clearly committed to making substantial revisions in the common-sense moral beliefs held by most people. It entails (when taken together with reasonable assumptions about the reality of animal experience) that hunting, factory farming, cosmetics testing, and many of the uses of animals in medical experiments are all seriously wrong, and should be stopped. For from the stand-point of the governing conception of utilitarianism - that of an impartial benevolent observer - there can be no reason why the interests of animals should be discounted or outweighed where they conflict with those of human beings. If these consequences are to be acceptable under reflective equilibrium, then we need somehow to explain away the almost universal human belief in their contraries - for example, the belief that the interests of an animal count for practically nothing when set against the suffering of a human being.
Utilitarians have a reply to this difficulty, explaining how it is that most people have had, until now, false beliefs about the extent of the moral significance of animals. For the impartial perspective is by no means an easy one to attain. Indeed, moral progress can be characterised, for a utilitarian, as a constant struggle against our own natural partiality. We are all naturally partial to those who are closest to us, linked to us by ties of blood or affection. Hence the most primitive form of morality is the morality of the clan, withholding moral standing from all outsiders. But reason can gradually modify this partiality, forcing us to recognize that there is no rational basis for counting the interests of those close to us above the interests of other people. In addition, a utilitarian can point out that there have been long periods of human history when an argument for equal consideration of the interests of slaves and slave owners would have struck most people as equally counter-intuitive. So our initial intuitive reaction to the claim that the interests of animals should be counted equally with our own might be claimed to be no more than a product of our natural (but unreasonable) partiality towards members of our own species.
This reply might have been considered adequate, if there had been no other theoretical alternatives available. That is, if we had been faced with a choice between having no coherent theory of morality at all, and one which entailed equal moral standing for animals, it might - perhaps - have been more reasonable to opt for the latter. For similar reasons, the envisaged reply might also have been acceptable if utilitarianism had enjoyed huge theoretical advantages over all alternative moral theories. But these are not in fact the choices before us. For we do have an alternative theory, namely contractualism. This is equally, if not more, theoretically defensible, and can explain the duties towards animals postulated by common-sense morality without granting moral standing to animals, as we shall see in Chapter 7. Given this situation, I propose that the way to achieve reflective equilibrium is to reject utilitarianism altogether, and embrace contractualism instead.
It is worth emphasising that the prohibitions against hunting, factory farming, and laboratory testing on animals are by no means the most counter-intuitive consequences of the utilitarian approach to this issue. Indeed, many ordinary people may hardly find them counter-intuitive at all. Rather, the hardest thing to accept is that the suffering of an animal should have equal moral standing with the (equally severe) suffering of a human being. An imaginary example will make the point vivid. Suppose that Saul is a very powerful and evil sadist. You have discovered that in the dungeons of his castle he keeps a number of creatures, including a human being, in conditions of perpetual torture. Now imagine yourself on a rescue mission to his castle. You have somehow discovered a way in, that can only be used once, and you know that you will only have time to release just one of those who are undergoing torture, before the alarm bells start to ring and you are captured yourself.
What should you do? Utilitarianism, being committed to the extension of the principle of equal consideration of interests to animals, entails that, other things being equal, there is nothing to choose - morally speaking, you are free to release any one of the imprisoned creatures at random. Indeed, if one of those creatures were to have a natural life-expectancy greater than that of the human - perhaps an elephant or giant turtle - then a utilitarian might have to claim that you are morally obliged to rescue the animal. These consequences are hugely counter-intuitive. I think most of us would feel that you are under a strong moral obligation to liberate the human being, and that you would, normally, do something very wrong indeed if you chose to save a dog, or an elephant, or a monkey instead.
It is important not to become distracted by irrelevancies at this point. It should be supposed, for example, that you have evidence that is as good as it can be that the degree of suffering of the imprisoned human is no greater than that of the animals. (You may perhaps have had the opportunity to study in detail videotapes of the torture in progress.) Moreover, it should be supposed that you know that the torture of each creature will continue until its natural death, and is of such a severity that it leaves no space in consciousness for any further thought. This is to side-step the point about the additional suffering that the human may later undergo, by re-living the torture in memory, and also the point that the human may suffer additionally through hopelessness and fear of suffering further. Suppose, too, that the human in question is quite old, with a life expectancy no greater than any of the dogs, cats, or monkeys involved. So there is nothing to be gained in the way of additional future happiness by saving the human being. Despite all this, the intuition remains that it would be unforgivable to do anything other than rescue the human being. My view is that this belief is so deeply and firmly held, by most of us, that any moral theory that requires its rejection ought, in turn, to be rejected under reflective equilibrium.
In reply, it may be said that many people have in fact found it quite easy to lose this intuition, and have embraced with enthusiasm the thesis of the equal moral standing of animal suffering, but without especially adopting a utilitarian stand-point. This is true. But so too, and in the same sense, have people managed to lose their belief in the physical world. In both cases the basic form of the argument is sceptical. Those who have lost their belief in physical reality have done so because they doubted whether there is anything that justifies belief in a world of physical objects, given that it is possible for our experiences to be a gigantic hallucination, or to be caused by an evil demon working directly on our minds. Similarly, many of those who have lost their belief in the differential moral standing of human and animal suffering have done so because they doubted whether there is anything that justifies belief in the difference. But, in common with many other philosophers, I believe that scepticism about physical reality is answerable. And it will be the task of Chapters 5 and 7 to answer scepticism about the unequal moral standing of animal suffering. In both cases the sceptical argument is initially attractive (not to say seductive), hard to answer, but ultimately unsound.
Most utilitarians will probably concede that they are in conflict with a key aspect of common-sense moral belief on the issue of animal suffering. But, they may claim, the forces of progress are on their side, in such a way that future generations will judge them, in retrospect, to have been correct. Our current attitudes towards animals, on this analysis, are similar to eighteenth century attitudes towards slavery and members of ‘inferior’ races. Indeed, many utilitarians are fond of pointing out that there have been numerous periods of human history in which the extension of the principle of equal consideration of interests to members of other races would have seemed equally counter-intuitive to most ordinary people. Yet we now judge that those people were wrong, and that the minority who protested against such practices as slavery were right.
In fact, however, the two cases are importantly different. For there has never really been a theoretically respectable moral theory that could justify a system of slavery, at least in any of the forms that have actually been practiced. (Recall from Chapter 2 that utilitarianism itself may imply that in certain - hypothetical - circumstances an institution of slavery would be justified.) In particular, contractualism, too, entails that slavery was, and is, seriously wrong. Indeed, what more obvious breach could there be of the central contractualist principle of respect for autonomy? All that really sustained common-sense beliefs about the permissibility of slavery were false - and probably self-deceived, certainly self-interested - beliefs about the inferior cognitive powers of members of other races. Once these beliefs were overturned, justifications for slavery collapsed without the need for any further theoretical argument.
There is, in contrast, a genuine theoretical dispute about the moral standing of animals. For, as we shall see in Chapters 5 and 7, contractualism entails that such standing should be withheld from animals, while it at the same time accommodates almost all the elements of common-sense moral belief. At no point does this theoretical case rely upon false beliefs about the cognitive powers of animals. Indeed, it is just as much a part of common-sense that animals have mental lives in many respects like our own, as it is that their sufferings and interests cannot be counted equally with ours. So in the case of animals, in contrast with the case of opposition to slavery in the past, utilitarianism is urging a substantial moral change upon us that is insufficiently motivated. Since it will turn out that there is, in fact, a moral theory that would preserve the status quo while being equally (at least) as theoretically attractive as utilitarianism, it must be unreasonable for us to accept such a change.
Higher and lower pleasures
I can think of just one way in which a utilitarian might hope to avoid the consequence that there is no obligation, in an example like that of Saul, the sadist, to opt for the prevention of human suffering above the prevention of animal suffering. This is by appealing to a distinction between higher and lower pleasures, that is in any case sometimes defended by utilitarians. It is sometimes said that there are pleasures, such as those of listening to a Schubert piano sonata, that are higher than others, such as the pleasures of eating or of masturbation. These (broadly intellectual) pleasures are said by some utilitarians to be intrinsically more valuable, counting for more in any calculation of overall utility. This distinction between higher and lower pleasures will loom large in the chapter that follows, but it is worth briefly considering, now, how it might be deployed at this point in the argument.
We might wonder, on the face of it, how the distinction between different orders of pleasure could apply to the example of Saul, the sadist, at all. For in that example no pleasure is in question, only physical pain. And this we are presuming to be the same in both quality and intensity in both human and animal. But this is to forget that one of the characteristic effects of pain, particularly if it is intense, is to interfere with other enjoyments, especially those that are cognitive in nature. Those who doubt this should try making love while suffering from a migraine, or listening to a Schubert sonata with one. It may then be said that while the pain undergone by a human being is intrinsically no more morally significant than the suffering of an animal, the human case is distinctive in that the pain also prevents enjoyment of pleasures that are higher. So the suggestion is that we may explain our intuition in the example of Saul, the sadist, as follows. The human being definitely ought to be the one to be rescued, because if that human were not undergoing torture, then at least some of their time would be occupied with pleasures that are higher than those that would occupy the time of a dog or a monkey if they were not being tortured.
As I say, the distinction between higher and lower pleasures will be considered in the next chapter, where I shall argue that it is of doubtful coherence. But it does not, in any case, really get a utilitarian out of the present difficulty, as can easily be seen. For suppose you know that the human who is undergoing torture is a convinced hedonist, of a sort who would in fact devote their entire time to the pursuit of lower pleasures if rescued. Or, if it is thought that experience of torture may be sufficient to cure anyone of hedonism, suppose that the human being in question is mentally retarded, so that they are constitutionally incapable of enjoying any but the lower pleasures. These possibilities make not the slightest difference to my intuition that it would be wrong to do anything other than rescue the human being. In which case, it cannot be the characteristic human capacity for intellectual pleasures that underlies the intuition.
Quality-of-character-utilitarians can perhaps reply to this argument. They can claim that our habit of taking human suffering more seriously than the comparable suffering of an animal has been formed in circumstances in which human suffering normally does (but animal suffering does not) interfere with the pursuit of higher pleasures. Then the intuition that I take so seriously in the case of Saul, the sadist - that it would be wrong to save the dog before the human being - may merely reflect this habitual way of thinking, which itself has a utilitarian justification.
There are two points to be made about this reply. The first is that it is a double edged weapon. For it is plain that most ordinary people do not seriously rate animal suffering at all, in comparison to the sufferings of human beings. A quality-of-character-utilitarian might then be expected to urge, as a corrective to this, that we should try to develop in ourselves a disposition to take animal sufferings more seriously than the sufferings of humans. This is on the sound Aristotelian principle, that if you are trying to mold a quality of character that is difficult for us to attain, you should aim, initially at least, to exaggerate it - by overshooting the mark, you may hit the target. So the quality of character that is manifested in our judgement that it would be wrong to rescue the dog first is arguably lacking in utilitarian justification in any case, even supposing that the distinction between higher and lower pleasures could be made out. For we ought, on utilitarian grounds, to be trying to become the sort of people who take animal suffering more seriously than we do.
The second point that should be made against the above reply is to note that reflective equilibrium, in its widest sense, must essentially involve comparison between moral theories, as well as mutual adjustment of theoretical detail and ordinary belief within a given theoretical approach. For I take it that our common-sense intuition in the case of Saul, the sadist, is not simply that it would be wrong to rescue an animal before the human (which might perhaps be explained by appeal to the distinction between higher and lower pleasures), but that it is wrong to weight the suffering of an animal equally with the equal suffering of a human being. So the plausibility of suggesting that our intuition in this case should be rejected needs to be contrasted with the relevant theoretical alternatives. As we shall see in Chapter 7, contractualism can explain all the main elements of common-sense moral belief here without having to give up this intuition. It should therefore, other things being equal, be preferred - especially given the strength of our feeling on the issue.
There is an argument for saying that speciesism is just as morally objectionable as racism or sexism. This argument would, if accepted, have important implications for our practices that cause suffering to animals, such as hunting and factory farming, since there are good reasons for believing that higher vertebrates, at least, have interests. But in fact the argument presupposes that the moral stand-point may be equated with that of an impartial sympathetic observer, which is the governing conception of utilitarianism. Moreover, the fact that utilitarianism has such a consequence renders it reflectively unstable, since the conclusion is at odds with apparently fundamental features of our moral thought. Even an appeal to the distinction between higher and lower pleasures cannot really help. The utilitarian approach to animal suffering is therefore inadequate, and should be rejected.
 Jonathan Cape, 1975; 2nd edition, 1990.
 See Practical Ethics, ch. 3.
 For Singer's commitment to this characterisation of the moral point of view, see Practical Ethics, ch. 1.
 See The Case for Animal Rights, ch. 5 & p. 261.
 See H. Rachlin, Behaviour and Learning (Freeman, 1976), pp. 125-6.
 See The Case for Animal Rights, p. 197.
 See Stephen Walker, Animal Thought (Routledge, 1983), chs. 4 & 5.
 See Walker, Animal Thought, ch. 6.
 For a recent example, see Peter Harrison ‘Do Animals Feel Pain?' Philosophy 66 (1991).
 See my Introducing Persons (Routledge, 1986), chs. 2-3 & 5, and also Peter Smith and O. R. Jones, The Philosophy of Mind (Cambridge University Press, 1986), part one.
 See my Introducing Persons, ch. 7.
 For example, see my Introducing Persons, chs. 1 & 4-6.
 See Practical Ethics, p. 52.
 See Practical Ethics, p. 59.
 See my Human Knowledge and Human Nature, chs. 11-12.
 Most of those who employ the higher/lower distinction would put the pleasures of sex in general on the ‘lower' side of the divide. This is a definite mistake. Normal human sex has an irreducibly intellectual component. For I do not just enjoy my own sensations (as I do when masturbating) - I am also aware that my partner is enjoying hers, and that she is similarly aware of, and takes further pleasure in, the fact that I am enjoying mine; and I enjoy that too. In fact there is nothing animal about human sex; it is a matter of mutual enjoyment in the fullest sense. In this much I follow Thomas Nagel, ‘Sexual Perversion', in his Mortal Questions (Cambridge University Press, 1979).
 See Aristotle, Ethics (c.330 BC), bk. 2, final sect.