Back to Animals Issue contents




In this chapter I shall confront the problem left over from Chapter 5, arguing that there is a way in which contractualism can accommodate duties towards animals that is independent of the question of offence caused to animal lovers. I shall then investigate just how extensive these duties may be, on the resulting account.


Judging by character

The general thesis I want to defend in this section from a common-sense perspective, is that some actions are seriously wrong, not because they cause any harm or violate any rights, but simply because of what they reveal about the character of the agent. I shall later go on to argue that this thesis is not only correct, but fully explicable within contractualism. It will then turn out that some ways of treating animals are morally wrong, just as common-sense tells us, but only because of what those actions may show us about the moral character of the agent. This will then be a form of indirect moral significance for animals that is independent of the fact that many rational agents care about animals, and hate to see them suffer.

Consider the example of Astrid, the astronaut, once again. Suppose, as before, that she has set her craft irreversibly to carry her out of the solar system, and that she is travelling with her cat, and her grandfather. Now at a certain point in the journey the grandfather dies. Out of boredom, Astrid idly cuts his corpse into bite-size pieces and feeds him to the cat. Is her action not morally wrong? It seems to me intuitively obvious that it is. But why? Plainly no harm is caused to the grandfather, nor is there anyone else who is to know of what she has done and be offended. Nor need the action violate any rights. For even if we allow that the dead have rights, such as might be infringed by ignoring someone’s will, the grandfather may have waived all relevant rights in this case. Astrid may have heard him say many times, when he was still in full possession of his faculties, that he did not care in the least what happened to his corpse after his death. Even so, it seems to me that Astrid’s action is morally wrong.

What Astrid does is wrong because of what it shows about her. Her action is bad because it manifests and expresses a bad quality of character, and it is an aspect of her character that is bad in the first instance. While there is perhaps no precise name for the defect of character that her action reveals, it might variously be described as ‘disrespectful’ or ‘inhuman’ - though each of these terms is really too broad for what is wanted. That she can act in the way that she does, shows either a perverse hatred of her grandfather in particular, or a lack of attachment to humanity generally.

It seems to be a universal feature of human nature that the treatment of corpses reflects something of our attitude towards the living. Certainly there are no human cultures that fail to have ceremonies of various sorts for honouring and disposing of the dead. Exactly what kinds of activities will count as honouring the dead is, of course, a highly conventional matter. In some cultures the proper way to dispose of a corpse is to bury it, in some to burn it, in some to embalm it, and in some to eat it. There might even be a culture in which the correct treatment of a corpse would be to cut it in pieces and feed it to a cat - though I am supposing that this is not true of Astrid’s case. But in no cultures are corpses simply cast aside, as some might throw away a dead rabbit or a dead rose bush.

I propose that the manner in which we treat our dead is best understood symbolically, the corpse being an embodied image of the person who has died, and perhaps also an image of persons generally. If this is right, then an attack on a corpse would universally be interpreted as a symbolic attack on the dead person. It shows something about one’s attitude to the individual person, and perhaps towards humanity generally, that one is prepared to attack their concrete image - that is to say, their corpse. Then if the attitudes revealed are bad ones, the actions that manifest and sustain them may be morally condemned as well.

Once we have realised that it is part of common-sense morality that an action can be criticised for what it shows about the character of the agent, we may begin to see that such judgements are really very common. For example, suppose that Jane is a doctor attending a medical conference, who happens to be relaxing in the hotel bar in the evening with many of the other doctors present. The room is subdivided into a number of separate cubicles, in such a way that, while she can see the central area in front of the bar itself (which is currently empty), she cannot see any of the doctors seated in the other cubicles, although she knows that they are there. Suppose that she then notices someone walking across the central area, who collapses with what appears to be a heart attack. Out of laziness, Jane does not move to assist him. This is surely very wrong of her. But why?

Suppose that Jane’s inaction does not in fact do any harm, since some of the other doctors are soon there to help. She had every reason to think that this would be the case, moreover, given that she knows there are many well-qualified and well-motivated people as close to the person as she is. Nor does she violate any rights by not going to the sick man’s assistance. For while he may have a right to assistance from doctors in general, he has no special claim against her in particular. The only explanation is that Jane’s inaction is wrong because of what it reveals about her character. It manifests a lack of humanity, more specifically a failure of beneficence. (Beneficence being the virtue that attaches us to the welfare of others.)

This is not to say that behaviour such as Jane’s must always display lack of beneficence. If she happens to have a migraine at the time, or a twisted ankle, then her lack of action may easily be excusable. (It would be a different matter, of course, if Jane were to believe herself to be the only doctor present. Then even a migraine would be no excuse for inactivity.) There is a general truth here, that whether or not a given action manifests a particular defect of character will depend crucially on the circumstances, and the motives from which it is performed. Suppose that instead of being in a space-rocket, for example, Astrid had been adrift with her grandfather on a life-raft in the Atlantic. As before, the grandfather dies, and as before she cuts his corpse into small pieces. But now her motive is to use the pieces as bait to catch fish for her to eat. This makes a crucial difference. In such circumstances her treatment of his corpse need display no disrespect or inhumanity. For her own survival is at stake.


A contractualist rationale

I have presented an intuitive case that actions may not only be judged for the harm that they cause or the rights that they infringe, but also for what they reveal about the character of the agent. Now recall from Chapter 2 how I argued that a utilitarian should take a serious interest in character - arguing, indeed, that qualities of character should be the primary object of utilitarian assessment. I also suggested in that chapter that contractualists should believe in a duty to develop in themselves a disposition towards beneficence - a point I shall return to shortly. But we have yet to achieve any general theoretical understanding of the way in which contractualists should regard qualities of character.

Why should rational agents who are trying to agree principles to govern their interaction with one another take any interest in character? Part of the answer lies in a realistic assessment of the springs of human action. While we are rational agents, in that we are capable of planning and evaluating alternative courses of action, most of our actions are by no means calculative. Some are routine, having passed the point where conscious deliberation is any longer necessary. Many others are done on the spur of the moment, prompted by circumstances in a way that pre-empts careful reasoning. Here general features of character, such as fair-mindedness or honesty, may make a great deal of difference to what we do. And even when there is time for deliberation it may require courage, for example, in order for us to pause for thought. (To take a fanciful case, if you find yourself in a room with a time-bomb, knowing that you have five minutes remaining until the explosion, courage may be displayed in pausing to reason that it would be better to take the bomb out into the garden, than to run into the garden yourself, leaving the bomb to destroy the house.) Indeed, the very readiness to take thought at all is itself a general feature of character, possessed by some people but not others.

In so far, then, as contracting rational agents are interested in the principles that are to govern their behaviour, they should also be interested in the general dispositions of thought and feeling that may make appropriate action more likely. So they should at least require people to try to develop those virtues that are sometimes described as ‘enabling’. These include the virtues of courage, self-control, and thoughtfulness, that may be useful whatever it is that you are trying to do, but also if you are trying to comply with moral rules. This is then one reason why rational agents should agree, not just to accept certain rules, but also to try to develop certain features of character.

But what reason would rational contractors have for taking an interest in the specifically moral virtues, such as those of generosity, loyalty, friendliness, and honesty? Some of these are easy, in so far as they fall under the general category of justice. One would expect that rules requiring honest and open dealing and speaking, for example, would certainly be amongst those agreed upon. Then rational contractors, taking a realistic view of the springs of human action, should also require agents to develop a general love of, and disposition towards, honest action, rather than mere calculated compliance with the rule.

The reasons why contractualists should regard themselves as required to develop virtues of beneficence, such as generosity and loyalty, are more theoretically interesting. They arise out of the fact that rational agents should surely wish to agree on more than merely rules of non-interference. For they may be certain that they shall, at some point in their lives, require help from others. Most of us require assistance from others almost every day, indeed. This may take many different forms. It may be material, such as gifts or loans for those who find themselves temporarily without funds. Or it may be practical, in the form of physical assistance, for example, as when another pair of hands is necessary to lift something into place. Or it may be psychological, in the form of advice, friendship, sympathy, or support. A society in which rules of non-interference were respected, but in which no positive assistance was ever given, would not only be cold and cheerless, but many of our desires would remain unfulfilled as a result, and many of our most cherished projects would be incomplete.

Given that rational contractors should provide for duties of assistance, as well as those of non-interference, how should the former be instituted? Plainly such agents cannot agree that everyone has a duty to help each of those who are in need, in the way that they should agree that everyone has a duty to respect the autonomy of others. For the result would be incoherent. Suppose that my wallet has been stolen, for example, and that I need the bus-fare home. If everyone were to give me the fare, then I should be a millionaire! (It is no real reply to this, to say that as soon as one person has helped me, everyone else’s duty to pay me the bus-fare lapses. For there will be many cases in which we are required to act in ignorance of what others may or may not have done.) Yet we cannot require only that everyone should give an appropriate proportion of what is needed, since many kinds of assistance are not divisible - as when the battery of my car is flat and I need someone to help me jump-start it.

For similar reasons, we cannot agree that those in need should have the right to assistance from others. For all rights imply a correlative duty. Sometimes, as in the case of the right not to be interfered with, the correlative duty is owed by all other agents - everyone is obliged not to infringe my autonomy. But then this would return us to the position we have just been discussing. Other rights imply duties on the part of particular persons or groups of persons. My right to fulfilment of a promise only implies a duty on the part of the promisor, for example, and my right to treatment of a minor illness only implies a duty on the part of the group of doctors with whom I am registered. But the problem with applying this model to the supposed right to assistance, is to find some appropriate person or group of persons who owe the correlative duty. For example, just who would be supposed to owe me the duty of jump-starting my car? (Note that we cannot answer ‘The first person who can help me, who is aware of my need’. For then agents would not be able to know, in general, whether or not they owe the requisite duty. How are you supposed to be able to tell whether or not you are the first person to have become aware of my need who could help, without engaging in extensive investigation first?)

The obvious and only feasible solution, is that rational agents should agree to develop a general disposition to help those in need, to be exercised when the opportunity arises to do so at no comparable cost to themselves. What they should agree to develop is a general attachment to the good of others, and a preparedness to act on their behalf. For if all have such an attachment, then almost everyone will get the help they need, when they need it (given certain obvious conditions of normality). Note that, in general, there will be no particular person whom I am obliged to assist. This will depend upon the circumstances, and on the other projects that I may be pursuing at the time. But each failure to help when the situation arises will count towards showing that I am not the sort of person that I ought to be. However, those situations in which I am aware that I am the only person who can help, as in the example of callous Carl from Chapter 2, constitute a special case. Here failures to assist may encounter direct criticism - that Carl can walk on past the child in such circumstances is sufficient to show that he has not developed the right sorts of attachments, no matter whether he is late for work or has a horror of water.

Rational agents should thus agree to try to develop virtues of beneficence, because of their knowledge that all would wish to live in a community of a certain sort. If human beings are to flourish, they need the support and sympathy of those around them when they are struck down by illness, or poverty, or grief. They also need a sense of community with others, that will require a degree of loyalty towards those who are close to them, and a general friendliness in their dealings with others. Rational agents should therefore agree, not just to abide by certain rules and principles (not to kill, not to steal, not to cheat), but also to develop certain positive attachments and dispositions of feeling. They should agree that they may be criticised, not just for infringing one another’s rights, but also if they fail to show compassion, and are not ready to help those who are in need.

It is important to stress a significant difference between the sort of contractualist treatment of character that has just been outlined, and the utilitarian approach to character presented in Chapter 2. In the context of utilitarianism, the value of virtues of character lies entirely in their consequences, leading to greater utility overall. Under contractualism, on the other hand, the consequentialist values of virtues of character - in facilitating right action or in contributing to a certain sort of society - enter in only at the stage when rational contractors are considering what sorts of persons they should try to become. Thereafter the rightness or wrongness of possessing, or failing to possess, a virtue of character is independent of such consequences. Thus in our example above, Astrid may be criticised for the failing of character displayed towards her grandfather’s corpse even though, in the nature of the case, her faults will never again have any effect on her treatment of another human being. Rather, the criticism is that she has failed to do her fair share, in the moral sphere - like anyone else, she was obliged to try to create in herself the sort of moral character that would (in the right circumstances) contribute to the form of society that all would wish for. Utilitarians, in contrast, must deny that Astrid does anything wrong, since no harm of any sort will result.


Animals and character

We can now explain, from a contractualist perspective, why it may be wrong of Astrid to use her cat as a dart-board, even though no other person will ever know or be distressed. Such actions are wrong because they are cruel. They betray an indifference to suffering that may manifest itself (or, in Astrid’s case, that might have manifested itself) in that person’s dealings with other rational agents. So although the action may not infringe any rights (cats will still lack direct rights under contractualism), it remains wrong independently of its effect upon any animal lover. Animals thus get accorded indirect moral significance, by virtue of the qualities of character that they may, or may not, evoke in us.

It is important to stress that right action with respect to animals, on this character-expressive account, will generally be non-calculative. People who act out of sympathy for the suffering of an animal do not do so because they calculate that they will become better persons as a result. Rather, their actions manifest an immediate sympathetic response, and are undertaken for the sake of the animals in question. For this is what having the right kind of sympathetic virtue consists in. This immediacy of response, however, is entirely consistent with the view that the moral value of the virtue, in so far as it manifests itself in our treatment of animals, derives from its connection with our treatment of human beings. Contracting rational agents should agree to try to develop a ready sympathy for one another’s suffering, and sympathy for animal suffering is, on the current proposal, merely a side-effect of this general attitude.

I shall investigate in a later section just how powerful a constraint is placed on our treatment of animals by this account. But it seems plain, at least, that actions that cause suffering to animals will be wrong whenever they are performed for no reason, or for trivial reasons (on the grounds that they manifest brutish cruelty), or whenever they are performed for their own sake (since they will then manifest sadistic cruelty). So the hit-and-run driver to whom it never occurs to stop, in order to help the dog left howling in pain at the side of the road, as well as the one who drives on because late for an appointment at the hair-dresser (as well as, of course, the driver who runs down the dog for fun in the first place), will each count as having acted wrongly, on the present account. For in each case the action will show the agent to be cruel.

Will this character-expressive account extend also to actions that cause the (painless) death of an animal? If so, then those who hunt animals for sport, or who kill them (or arrange to have them killed) for the pleasure of eating their flesh, will stand morally condemned. For if it were true that sympathy should be extended for the death of an animal as well as for its suffering, then it is plain that such actions would count as brutishly cruel, since the pleasures for which they are undertaken are trivial ones. In fact, however, it is doubtful whether we manifest cruelty by such killings, as I shall now try to explain.

It is obvious that beneficence towards human beings normally encompasses actions necessary to preserve life, as well as those necessary to prevent suffering. Thus callous Carl would surely constitute a paradigm of heartlessness, even if the process of drowning, undergone by the child he fails to rescue, were not itself a painful one. But this may only be so (at least in the first instance) where the life in question is the life of a rational agent. (Arguments similar to those deployed in Chapter 5 may then be used to widen the extent of the required attitude to include all human beings.) For recall that we are now viewing beneficence within the context of contractualism, and that rational contractors may be expected to value their own rational agency above all else. Moreover, recall from Chapter 3 that our reasons for fearing death derive from the fact that we have forward-looking desires that presuppose continued life. We would then expect rational contractors to agree to develop a general attachment to one another’s lives. They will then be prepared, not only to avoid killing one another (a requirement of justice), but also to act positively to try to prevent death where possible, grounded in a sympathetic appreciation of the motives that rational agents have for going on living.

It counts in favour of the contractualist approach to these issues, that when we enter sympathetically into the death of another, trying to see what their death may have meant from their point of view, we do seem naturally to focus on those plans and projects that have now been cut short. For this would explain the fact noted in Chapter 4, that many people feel less sympathy - from the perspective of the one who dies - for the death of a baby or an old person who has let go of their hold on life. For in such cases no forward-looking motives for survival may exist. But this sort of sympathy is only possible in respect of the death of a rational agent, since only such an agent has long-term projects, or the desire for continued life.

What emerges is this. While the fact that the death of an animal may bring to an end a worthwhile existence and prevent future satisfactions of desire may acquire moral importance in the context of utilitarianism, it will lack such significance in the context of the present character-expressive account. That someone fails to be moved by the painless death of an animal need not display any cruelty. For there is no such thing, here, as entering sympathetically into the reasons that the animal had for going on living. Of course we could, if we wished, enter sympathetically into the future pleasures and satisfactions of the animal, that have now been lost through death. And no doubt if we were utilitarians we should be obliged to do so, as we saw in Chapter 4. But given that this is not what sympathy for the death of a rational agent normally amounts to, the fact that we fail to have such feelings in connection with the death of an animal need not show that there is anything amiss with our moral character.


Reflective equilibrium attained

I believe that the account now sketched of our duties towards animals is sufficiently plausible to enable us to achieve reflective equilibrium overall. First, it can explain our common-sense belief that it is wrong to cause unnecessary suffering to an animal, where ‘unnecessary’ means either ‘for no reason’, ‘for trivial reasons’, or ‘for its own sake’. (In the next section I shall consider what the implications of the account may be for practices that are more controversial, such as hunting, factory farming, and animal experimentation.) Second, the present approach also retains our intuitive belief that there can be no question of weighing animal suffering against the suffering of a human being. Since animals are still denied moral standing, on this contractualist account, they make no direct moral claims upon us. There is therefore nothing to be weighed against the claims of a human being. Finally, the account can retain the intuition shared by many people (including some champions of animals like Singer, as we saw in Chapter 4), that there need be nothing wrong with causing the painless death of an animal. Since the sort of sympathy that we should feel for the loss of a human life is only appropriate, in the first instance, in connection with the death of a rational agent, such actions may fail to manifest any degree of cruelty. (Some killings may, however, be wasteful, in the same sense that the motiveless cutting down of an oak tree may be.)

A further advantage of our account, is that it can explain how people so easily come to be under the illusion, when they engage in theoretical reflection, that animal suffering has moral standing, mattering for its own sake. For those who have the right moral dispositions in this area will act for the sake of the animal when prompted by feelings of sympathy. Since right action requires that you act for the sake of the animal, it is then easy to see how one might slip into believing that the animal itself has moral standing. But this would be to miss the point that there may be a variety of different levels to moral thinking.[1] On the one hand there is the level of thought that manifests our settled moral dispositions and attitudes (this is where sympathy for animal suffering belongs), but on the other hand there is the level of theoretical reflection upon those dispositions and attitudes, asking how they may be justified by an acceptable moral theory. It is at this level that we come to realise, as contractualists, that animals are without moral standing.

For similar reasons, the proposed character-expressive account of our duties towards animals is able to avoid the charge of absurdity often levelled at Kant’s somewhat similar treatment of the issue.[2] Kant is sometimes represented - unfairly - as claiming that those who perform acts of kindness towards animals are merely practicing for kindness towards humans. As if anyone ever helped an animal with such an intention! But in fact he is best interpreted as presenting an account along the lines of that above, which distinguishes between the motives of those who act out of the sort of beneficent state of character they ought to have, and the theoretical explanation of the moral value that state possesses. It is only at the latter level that we may see the value of a sympathetic character as deriving from the way in which it manifests itself in our treatment of human beings.

It therefore looks as if the present proposal can account for every aspect of common-sense. The only apparent difficulty remaining, is that it denies that animal suffering has moral standing. However, this is not, properly, part of common-sense itself, but is rather a theoretical construction upon it. Here the account can explain how we come to be under the illusion of direct significance. The contractualist treatment of animals thus has all the hall-marks of a powerful moral theory, acceptable under reflective equilibrium in the absence of any more plausible proposal. It remains to investigate the consequences of this approach for the controversial practices of hunting, factory farming, and laboratory experimentation upon animals.


Controversial consequences

How powerful a constraint does contractualism place on our behaviour towards animals, on the present account? That is to say, under what circumstances is it wrong to cause suffering to an animal, on the grounds that to do so would display cruelty or some other defect of character? Here our earlier observation becomes important, that whether or not a given act displays a defect may depend on the surrounding circumstances and the motive from which the action is performed. In the case where Astrid was adrift at sea, it was clear that no disrespect was shown to her grandfather by cutting up his corpse for bait. But it is worth noting that a similar act may be excusable when a great deal less than human life is at stake.

Suppose that Candy lives with her grandfather in a cabin in a particularly bleak part of Canada. For two months every winter they are completely snowed in, with drifts even covering the main windows. The only source of ventilation remaining is a small window under the eaves. Now suppose as before that the grandfather dies, and that, as is the way of all flesh, his body begins to decompose. In order to avoid the nauseating smell, Candy cuts up his corpse into pieces small enough to throw out through the only functioning window. It seems to me that she no more lays herself open to criticism than did Astrid on her life-raft, although a great deal less than survival is at stake.

I think that all our judgements in such cases make substantial psychological assumptions, concerning what actions and attitudes are psychologically connected with, or separable from, others. We judge that Candy’s action is permissible because we think that, taken together with its motives and circumstances, it may easily co-exist with a deep love of her grandfather, and with a respect for humanity of whatever strength we deem appropriate. In contrast, we think that lazy Jane’s lack of preparedness to go to the aid of the man who collapses in front of her shows a lack of compassion that may manifest itself in other, more serious, circumstances.

When we apply these ideas to actions that cause suffering to animals, it turns out that almost any legitimate, non-trivial, motive is sufficient to make the action separable from a generally cruel or insensitive disposition. For example, consider technicians working in laboratories that use animals for the testing of detergents, causing them much suffering in the process. That they can become desensitized to animal suffering in such a context provides little reason for thinking that they will be any less sympathetic and generous persons outside it. Consider, also, farm-hands working in conditions that cause considerable suffering to the animals under their care. Again there seems no reason to think that they will thereby be more likely to be cruel or unsympathetic when it comes to dealing with other human beings in their society. Note that in both cases the motives from which the people in question are acting are by no means trivial, since they are earning a livelihood.

It is important to stress that the only basis for direct moral criticism of actions such as these that cause suffering to animals, has to do with the qualities of character manifested by the individual actors. There is thus no scope, here, for criticising the overall practices of factory farming and animal experimentation. (We shall return in a later section to the argument from the legitimate concerns of animal lovers.) This point is important because even if the reasons why we have such practices are trivial - cheaper meat and new varieties of cosmetic - the motives of those who engage in them are not. There is then no reason to claim that those people are cruel in what they do.

It may be objected that a significant difference between Candy, the Canadian, and the laboratory technicians is that Candy’s action is ‘one off’, whereas the actions of the technicians are continually being repeated. It may therefore be felt that, although the latter actions do not in themselves display a cruel character, they may be apt to cause such a character, through desensitization to suffering, and may thus be morally condemned on that basis. But I think that human beings are more discriminating than this argument suggests. That someone can become desensitized to the suffering of an animal need not in any way mean that they have become similarly desensitized to the sufferings of human beings - the two things are, surely, psychologically separable.

Hunting may present a rather different case. For those who hunt animals for sport, rather than to feed themselves or to earn a living, do so from motives that must certainly count as trivial in comparison to the suffering that they cause. While the pleasures of the hunt need not be directly sadistic - it need not be the suffering of the animal that is the object of enjoyment - they are inseparably bound up with the enjoyment of power, and of violent domination. (If the challenge of creeping up close to an animal through the woods were the only pleasure, then one could just as well hunt with a camera as a gun.) It does seem plausible that those who indulge such pleasures may be reinforcing aspects of their characters that may unfit them, in various ways, for their moral dealings with human beings.

Part of the explanation for the psychological separability that I have claimed to exist between attitudes to animal and human suffering, lies in the obvious differences of physical form between animals and humans. Because most animals look and behave very differently from humans, it is easy to make and maintain a psychological distinction between one’s attitudes to pain in the two cases. The most brutal butcher can nevertheless be the most loving parent and sympathetic friend. For this very reason, indeed, it seems to me (from a contractualist perspective, remember) to be a conventional, culturally determined, matter that one’s attitudes towards animals should show anything at all about one’s moral character, as I shall try to explain in the next section.

It may be objected that people’s attitudes to the sufferings of different classes of human being are equally psychologically separable. White racists who are indifferent to the sufferings of blacks, for example, may nevertheless behave impeccably with respect to other whites. But this is not to the point. For such people are drawing distinctions amongst those who have moral standing, and who are therefore entitled to equal concern and consideration. The separability between attitudes to human and animal suffering, in contrast, is grounded in a distinction between those who have, and those who lack, moral standing. There is therefore no direct moral objection to those who are able to keep their attitudes to pain separate in the two cases.


Animals and culture

There are a number of different respects in which our particular form of society encourages a connection between our attitudes towards animals, on the one hand, and human beings on the other. The first is that many of us keep animals as pets. Now people do keep some strange pets, including alligators, spiders, and stick-insects. But in general they keep the sorts of animals that are most human, particularly in their response to affection. In fact, we model our relationships with our pets on our relations with other human beings, and these relationships serve many of the same purposes of companionship and the enjoyment of shared experience. Since we treat pets as honorary humans, as it were, it follows that if someone can be cruel to a pet then this is fairly direct evidence of a generally cruel disposition.

The second point about our society, a correlative of the first, is that for most of us our only direct contact with animals is with pets. This is largely a product of increasing urbanisation. It also makes contemporary Western culture unique in all human history. In all other cultures the majority of people would also have had extensive contact with animals in the course of their work, whether hunting, farming, or through other forms of labour such as towing barges and lifting weights. It cannot be an accident that our society has, in consequence, recently seen an explosion of sentimentality towards animals.

The final point about our society is that we frequently use animals as moral exemplars in the training of the young. (This may be connected with a phenomenon we noted in Chapter 6 - namely, the extent of anthropomorphism now present in children’s literature.) It may be true of many children in our society that their first introduction to moral notions is to be told that it is cruel to pull the whiskers out of the cat. So, again, if someone is cruel to an animal, then this is evidence that something may have gone drastically wrong with their moral education.

These features of our society are highly contingent. There may be (indeed, there are) many other societies in which animals are not accorded these roles. In such a society a dog may be slowly strangled to death because this is believed to make the meat taste better, while it never occurs to the people involved that there is any connection between what they are doing and their attitudes to human beings - indeed, there may in fact be no such connection. While such an action performed by someone in our society would manifest cruelty, when done by them it may not.

It therefore seems to me that, while contractualism can find a place for the indirect moral significance of animals, and for duties towards them, it is a fairly minimal and culturally determined place. Given certain facts about our society, it may be true that some behaviour towards animals is wrong because of what is shown about the character of the agent. But what is shown may not be very much, in many circumstances. And there may be other social conditions in which nothing of moral significance would be shown at all. While contractualism is thus vindicated, in that it can explain how there is a large element of truth in our common-sense attitudes towards animals, at the same time little or no comfort is given to those who would wish to extend greater moral protection to animals.

One question remains at this point: to what extent is the role of animals in our particular society morally desirable? That people need pets at all is, arguably, a product of the social alienation felt by many people in societies as fluid and fragmentary as ours, and we could surely engage in successful moral education of the young without using animals as exemplars. We therefore need to look at the question whether current attitudes towards animals may not be getting in the way of other, more fundamental, moral concerns. I shall return to this issue in the final section of the chapter.


Non-rational humans re-visited

The position reached above, concerning the limitations of our obligations towards animals, would of course carry little conviction if contractualists were forced to say similar things about our treatment of those human beings who are not rational agents - namely, young babies, severe mental defectives, and the very senile. For no one is going to accept that babies may be factory farmed for their meat, or that aggressive mental defectives may be killed in the way that one might put down a vicious dog. Now in the final sections of Chapter 5 I presented arguments from the danger of a slippery slope, and from social stability, for the conclusion that all categories of human being should be accorded the same basic direct rights. Those arguments can now be strengthened by points arising from our present discussion of attitudes to suffering.

No doubt human babies, mental defectives, and senile old people may enjoy similar levels of mental activity to animals - frequently lower, in fact. But in other respects they will have a moral salience that is quite different from that of animals. The crucial point is that they share human form, and many human patterns of behaviour, with those who are rational agents. It is no mere accident of culture or upbringing that a crying baby, or a senile old woman moaning with the pain of terminal cancer, can evoke our sympathy. For what is presented to our senses in these cases differs only in slight degree from the suffering of a child or normal adult. We should therefore expect sensitivity to the one form of suffering to be closely psychologically connected with sensitivity to the sufferings of those human beings who are rational agents. Someone who behaves in such a way as to be indifferent to the suffering of a baby or a senile old lady is therefore very wrong, because of what their behaviour reveals about their character, quite apart from any question of infringements of rights. That they can act in such a way is almost certain to manifest cruelty.

This is not to say, of course, that we need be psychologically incapable of drawing distinctions within the category of human beings, and of arranging our moral attitudes accordingly. On the contrary, it is plain that many people in the course of human history have done just this. Some of these distinctions, for example on racial or sexual grounds, mark divisions within the class of rational agents, and may therefore be directly condemned on grounds of justice. But the general point is that it is highly dangerous to attempt to draw distinctions within the category of human beings at all. Given the immense similarities of appearance and behaviour that exist amongst all human beings, whatever their intellectual status, attempts to ground attitudes to suffering on distinctions between them are likely to undermine attitudes to suffering elsewhere. Those who begin by rationalising their indifference to the sufferings of the senile may end by so warping their attitudes and moral imagination that they become insensitive to the sufferings of some who are, indisputably, rational agents.

Rational contractors who are trying to agree on the rules that will assign basic rights and duties should therefore be aware that any attempt to draw distinctions within the category of human beings may have psychological effects that would prove morally disastrous. They should then agree to assign basic moral rights to all human beings, irrespective of their status as rational agents. For suppose that they were to agree on a rule excluding mental defectives from possessing moral standing, and were thus to allow that there is no direct moral objection to killing or hurting such a being. This rule would clash powerfully with our natural impulse of sympathy for the sufferings of all who share human form, and may cause the latter to be undermined. If so, then our duties towards rational agents would also be endangered.

In contrast, no similar dangers attend the exclusion of animals from possession of moral standing. (Nor, arguably, are there such dangers in the exclusion of human foetuses, in their early stages of development. So abortion may remain a moral option.) For there is a large gulf, both of physical form and modes of behaviour, between human beings and even their closest animal cousins. A dividing line drawn here, being clear-cut, and appealing to features that are strikingly salient, may therefore be a stable one. For it will then be easy to create and maintain a psychological distinction between one’s attitude towards suffering in the two cases.


Indirect arguments again

Let us review our conclusions thus far. Since animals are not rational agents, they will not, in the first instance at least, be accorded direct moral rights under contractualism. But then nor are there slippery slope or social stability arguments for granting them such rights. Animals have indirect moral significance nevertheless, in virtue of the qualities of moral character they may evoke in us. Actions involving animals that are expressive of a bad moral character are thereby wrong. Because attitudes to animal and human suffering may be readily psychologically separated, however, the constraints so far placed on our treatment of animals are minimal ones. All that follows is that it is wrong (in our culture) to cause suffering to an animal for trivial reasons, or to obtain sadistic pleasure.

It seems that nothing bad need be displayed in the moral character of someone whose job involves testing detergents on animals, or in farm-hands whose practices cause suffering to the animals under their care (provided, at least, that they try to minimise the pain that they cause, in so far as this can be done without great cost to themselves). Nor need there be anything amiss with the characters of those who employ such people, since their motive will generally be to retain the profitability and competitiveness of their business, which is certainly not trivial. So in all such cases, there is neither a direct moral objection (no rights are infringed), nor is there any indirect moral objection arising out of a judgement on the character of the agent. Even in the case of activities, such as hunting, where there is a moral objection on grounds of character, it would hardly be permissible to intervene, as do hunt saboteurs. For we do not think it right, in general, that we should try to force people to change their characters, in advance of them performing actions that constitute infringements of right. For example, even if there were reliable psychological tests for aggressiveness, we surely would not think it right that those who score highly in such tests should be required to undergo treatment, in advance of evidence of actual violent behaviour towards others.

There remains, however, the question of the likely offence to animal lovers, discussed in Chapter 5. While this was rejected at the time as an adequate basis on which to ground all moral duties towards animals, it might still be re-introduced, at this point, as an argument for forbidding hunting, factory farming, and many forms of animal experimentation. For it now emerges that there is a major difference between this sort of offence, and the offence caused to prudish people by the thought of unusual sexual practices. This is because distress at the thought of an animal suffering is a response that we may want people to have, if the argument of this chapter has been sound. It is a response that both manifests and reinforces an admirably sympathetic character. Sexual prudery, in contrast, has no particular moral worth. So while the retort ‘If it upsets you, don’t think about it’ may be appropriate for the latter group, it is not appropriate for the former. The concern of animal lovers for the sufferings of animals is not only legitimate, but expressive of a morally admirable state of character. We may wonder, then, whether this is sufficient to render such practices as hunting and factory farming morally unacceptable - not because of any infringement of animal rights, but because those practices are insufficiently respectful of the concerns of animal lovers.

Some might be puzzled at how I can claim, on the one hand, that sympathy for animal suffering is expressive of an admirable state of character and yet claim, on the other, that those who become desensitized to such suffering in the course of their work need not thereby display any weakness in their character. How can I have it both ways? The answer is that the contexts are different in the two cases. Those who become distressed when they think of the animals suffering in our factory farms and experimental laboratories do so, as it were, in the abstract - in independence from any further morally significant purpose. Those who become desensitized to that very same suffering, on the other hand, do so in the context of earning a living. Roughly speaking, the position to emerge from this chapter is that sensitivity to animal suffering is admirable when, and only when, it fails to interfere with purposes that are morally significant in a more direct sense.

But now we have a difficulty. For the proposal that factory farming and animal experimentation should be forbidden because they encroach on the sensibilities of animal lovers, means that those feelings would be interfering with morally significant purposes, namely the purposes of earning a living and of maintaining a viable business. In this case it would seem that the latter must take priority. It is too much to demand that people should forgo employment out of respect for the feelings of animal lovers, just as it is too much to demand that the owner of an ancient building should do without a habitable residence out of respect for the feelings of those who would not wish to see the building altered. If the legitimate feelings of animal lovers are to have any important place in this debate, it will not be as a ground for criticising the individual practitioners, but rather as a basis for criticism of the practice as a whole. What might be claimed is that, out of respect for these feelings, the organisation of our society should be altered so that it has no need of practices that cause suffering to animals on a regular basis - where the changes should include compensation for those who might lose employment or income as a result.

We are now brought back to the question of public policy we left open earlier - namely, whether we want to encourage and reinforce the psychological connection that already exists, in our culture, between attitudes to animal and human suffering. Or is this connection already too strong, so that there is a moral case for trying to weaken it? As we noted earlier, there are forces in our culture that are responsible for this connection, and arguably these forces are increasing. With increasing wealth, and yet increasing social alienation, more and more people are keeping pets. And young children are increasingly exposed to forms of entertainment in which anthropomorphic treatments of animals are rife. Yet there are no particular moral gains in this tendency. On the contrary. To restrict current patterns of treatment of animals out of respect for the sensibilities of animal lovers would only reinforce a trend that has considerable moral costs.

There would be economic and social costs of placing further restrictions on our treatment of animals, particularly if factory farming and scientific experiments on animals were forbidden. But I do not wish to focus especially on these. More important, is that the cost of increasing concern with animal welfare is to distract attention from the needs of those who certainly do have moral standing - namely, human beings. We live on a planet where millions of our fellow humans starve, or are near starving, and where many millions more are under-nourished. In addition, the twin perils of pollution and exhaustion of natural resources threaten the futures of ourselves and our descendants. It is here that moral attention should be focussed. Concern with animal welfare, while expressive of states of character that are admirable, is an irrelevance to be opposed rather than encouraged. Our response to animal lovers should not be ‘If it upsets you, don’t think about it’, but rather ‘If it upsets you, think about something more important’.

It may be objected that it is always possible to think about both. It might be claimed, indeed, that increased concern for animals will help to foster the attitudes of general sympathy and respect for the environment that will be necessary in tackling the world’s wider problems. But in fact, much of the moral energy currently spent in defence of animals has been diverted from other domains. Amongst those who campaign actively on behalf of animals, indeed, the feelings of sympathy that motivate their actions have ceased to be morally admirable, precisely because those feelings have been allowed to get in the way of concerns that are more directly morally significant. Moreover, there is no way in which we can, as contractualists, tell ourselves a plausible story in which increased concern for animals will be morally beneficial. For we ought to be able to see clearly that it is only the sufferings of humans that have moral standing, and that have direct moral significance irrespective of facts about character. In which case, increased feelings of sympathy for animals can only serve to undermine our judgements of relative importance, having the same moral effect as decreased concern for humans. So if contractualism provides us with the best framework for moral theory, as I have argued that it does, then we should wish to roll back the tide of current popular concern with animal welfare.



Contractualism withholds direct moral rights from animals, while at the same time granting them to all human beings. Yet contractualism can explain our common-sense belief that animals should not be caused to suffer for trivial reasons, since causing such suffering is expressive of a cruel character. This position is sufficiently plausible to be acceptable under reflective equilibrium. But the constraints thus justified are minimal. Contractualism certainly provides no support for those who would wish to extend still further the moral protection already available to animals.


[1] This point is emphasised, from a utilitarian stand-point, by Hare in Moral Thinking.

[2] See the lecture on duties towards animals in Kant's Lectures on Ethics (1775-80), published in translation by Methuen.