5. CONTRACTUALISM AND ANIMALS
In this chapter I shall consider what a contractualist should say about the moral standing of animals. Throughout, I shall make the simplifying assumption that no animals should be counted as rational agents, in the sense that is central to contractualism. The extent to which this assumption is true will be examined in the chapter that follows.
Rawls’s contractualism and animals
According to Rawls, we are to think of morality as the set of rules that would be agreed upon by rational agents choosing from behind a veil of ignorance. While these agents are supposed to have knowledge of all general truths of psychology, economics, and so on, they are to be ignorant of their own particular qualities (their intelligence, physical strength, projects, and desires), as well as the position they will occupy in the resulting society. Their choice of moral principles is to be made in the light of broadly self-interested desires (such as those for happiness, freedom, and power), that the agents know they will possess whatever particular desires and interests they subsequently come to have.
Morality is here pictured as a system of rules to govern the interaction of rational agents within society. It therefore seems inevitable, on the face of it, that only rational agents will be assigned direct rights on this approach. Since it is rational agents who are to choose the system of rules, and choose self-interestedly, it is only rational agents who will have their position protected under the rules. There seems no reason why rights should be assigned to non-rational agents. Animals will therefore have no moral standing under Rawlsian contractualism, in so far as they do not count as rational agents.
It might be suggested that there is, after all, a way in which rights can come to be assigned to animals under contractualism. This is by some of the agents behind the veil of ignorance being detailed to speak on behalf of non-rational agents, their task being to represent the interests of animals in the formulation of the basic contract. Compare the way in which a lawyer may represent the interests of a pet dog in a court of law, in a dispute over the deceased owner’s will. The idea is that behind the veil of ignorance, as in the law court, someone may be detailed to speak for those who are incapable of speaking for themselves.
Notice, however, that even if this extension of Rawls’s theory were acceptable, it would still not yield anything like the common-sense view of animals. On the contrary, it would lead to animals being accorded equal rights with human beings, consistent with their different needs and capacities. (Thus one would not expect that animals could have an equal right to own property, since they are incapable of buying and selling. But they could have an equal right to life, and an equal right not to be made to suffer.) There is no reason why the animal representatives behind the veil of ignorance should settle for anything less. For remember that the people in this position are not supposed to have, as yet, any moral beliefs. So the representatives of animal interests cannot accept as a reason for according animals unequal status, that animals have lower moral importance than human beings. But the idea that animals should be given equal standing with ourselves is a good deal more extreme than we should be prepared to accept, as we have seen over the last two chapters.
Another problem with the above suggestion is this. Once it is allowed that animals may have representatives to speak on their behalf behind the veil of ignorance, there seems no good theoretical reason why other sorts of thing should not have representatives also. Why should there not be people detailed to defend plants and micro-organisms, or indeed mountains and ancient buildings? Moral rights would then become rampant, in a way that would, I think, be acceptable to no one.
The main objection to allowing representatives of animal interests behind the veil of ignorance, however, is that it is arbitrary. It has been done without any independent theoretical rationale, simply to secure the desired result - that animals should have moral standing. Now it might seem that this charge is unfair. For as Rawls stresses, the business of theory-construction in morality is, at least partly, a matter of seeking reflective equilibrium. Although moral beliefs should not be directly mentioned within our theory, it is a constraint on a theory being acceptable that it should deliver a good many, at least, of our firmly held moral convictions. And we do have moral convictions about the appropriate treatment of animals. So it might be said that allowing representatives to speak on behalf of animals behind the veil of ignorance is just the sort of theoretical alteration that we ought to have expected to make all along.
While I have endorsed the method of reflective equilibrium in ethics, I do not think that it can be successful in defending the current proposal. One reason for this is that the proposal does not, in any case, deliver the common-sense view of animals, as I pointed out above. But a more important reason is as follows. As initially presented, the idea of choice of moral principles from a position of ignorance constituted a coherent vision of the nature and source of morality. Moral rules were seen as those that rational agents would agree on to govern their conduct with respect to one another, if they employed only general rational considerations in their choice, not allowing themselves to be influenced by facts about their own particular interests or position within society. But if some of these agents are detailed to represent the interests of animals in the selection of moral rules, this coherence vanishes. It is then no longer clear what morality is. Indeed, it appears that we would have to say - circularly - that morality is the set of rules that would be agreed upon by rational agents who already had a prior belief in the moral standing of animals.
Regan has mounted an argument designed to show that contractualism cannot coherently withhold moral standing from animals, without also withholding it from those human beings who are not rational agents, such as severe mental defectives or senile old people. I shall defer considering the latter part of this charge until later sections, where it will turn out that Regan seriously underestimates the resources available to contractualists, through which they can explain how all human beings should be accorded the same basic moral rights, whatever their mental capacities. Here I shall consider Regan’s argument that Rawlsian contractualism is theoretically arbitrary to the extent that it denies moral standing to animals.
Regan claims that if agents behind the veil of ignorance are to be ignorant of such fundamental matters as their qualities of character, life-plans, and position within society, then there is no good reason why they should not also be ignorant of their species. But if the agents were to be ignorant of the species into which they would subsequently be incarnated, when selecting basic moral principles, then, plainly, they would choose rules protecting the interests of members of all species equally. So Rawls has simply begged the question against the moral standing of animals in the manner in which he sets up the apparatus of the veil of ignorance. Had he arranged the details of that device slightly differently, then animals would have been accorded the same basic rights as human beings within contractualism.
Now, I do not wish to suggest that there is anything sacrosanct about the way in which Rawls characterises the details of the veil of ignorance. (On the contrary, it is plain that a simple extension of his ideas would deliver a strong environmental ethic, that would at the same time be theoretically motivated. We need only suppose that the agents behind the veil of ignorance should have a desire to inhabit a healthy environment, besides desires for the primary goods of happiness, freedom, and power. This would immediately lead to agreement on principles that protect the environment. Yet it would have a theoretical rationale, in that rational agents may surely know that, whatever else they may want, they will also wish to live in an environment that is healthy and pleasant.) Nevertheless, I do not think that the extension suggested by Regan - that agents behind the veil of ignorance should be ignorant of their species - is a coherent one, as I shall now try to explain.
An initial problem is that Regan misinterprets Rawls. He takes him to believe that the veil of ignorance is a genuine metaphysical possibility - that rational agents might really exist in ignorance of their character, desires, strength, sex, and social position, perhaps as disembodied souls - whereas it is, for Rawls, only a device for bracketing unwanted knowledge, whose point is to ensure that we should not, in the construction of moral principles, appeal to knowledge that might undermine the reasonableness of the result. This does not yet dispose of Regan’s argument, however. For it is presumably possible that rational agents should bracket their status as rational agents, even in the very process of rationally constructing a system of rules. If they can forbear from making use of their knowledge of their sex or social status, then presumably they can just as well forbear from making use of their knowledge of their species or, indeed, of the fact that they are rational agents.
The real line of reply to Regan is that his suggestion would destroy the theoretical coherence of Rawlsian contractualism. As Rawls has it, morality is, in fact, a human construction (in the absence, that is, of any other known species of rational agent - a point I shall return to in the next chapter). Morality is viewed as constructed by human beings, in order to facilitate interactions between human beings, and to make possible a life of co-operative community. This is, indeed, an essential part of the governing conception of contractualism. It is crucial to its explanation of how moral notions can arise, avoiding the excesses of intuitionism and strong objectivism. It is also presupposed by contractualist accounts of the source of moral motivation, whether in the Rawlsian version (to make peaceful human community possible in conditions of modernity) or in my own, where the basic contractualist concept (as well as the desire to comply with it) is held to be innate, selected for in evolution because of its value in promoting the survival of our species. To suggest, now, that contractualism should be so construed as to accord equal moral standing to animals would be to lose our grip on where moral notions are supposed to come from, or why we should care about them when they arrive.
It may be objected that this line of reply to Regan implausibly reduces morality to anthropology. But in fact it does no such thing. My claim is not that moral statements are really disguised claims about the conditions for the survival of the species. On the contrary, they are about what rational agents should reasonably accept who share the aim of reaching free and unforced agreement. My claim is only that we have this concept of morality innately, and have an innate desire to justify our actions in terms that others may freely accept, because doing so has promoted the survival of our species in the past. But if the contractualist concept expresses what morality is, for us, then there is no moral stand-point from which it can be criticised, or from which it can be argued that we are morally required to extend that concept so as to accord equal moral standing to animals.
I conclude, therefore, that Regan is mistaken. Rawls is by no means arbitrary in allowing agents behind the veil of ignorance to have knowledge of their species and their status as rational agents. On the contrary, this is crucial to ensure the plausibility of the contractualist governing conception of the source of moral notions and moral motivation.
Scanlon’s contractualism and animals
The points made above in reply to Regan strongly suggest that the exclusion of moral standing from animals is entailed by contractualism as such, rather than by any mere quirk of Rawls’s presentation. As a cross-check on this, let us briefly consider how animals would fare under Scanlon’s version of contractualism, in which the agents concerned are supposed to be real ones, possessing full knowledge of their own idiosyncratic desires and qualities, and their position within the current structure of society. Recall that on Scanlon’s account, moral rules are those that no one could reasonably reject as a basis for free, unforced, agreement, who shares the aim of reaching such agreement. The only idealisations made are that choices and objections will always be rational, and that all concerned will share the aim of reaching such an agreement.
We still have here a coherent vision of the nature of morality - indeed, essentially the same vision as is presented in Rawls’s contractualism. Since Scanlon’s model deals with real agents, however, with their individual desires and concerns, and since many real agents care deeply about the welfare of some or all animals, there is then a genuine question as to whether such people might not reasonably reject rules that give no weight to the interests of animals. It may be that animals will turn out to have moral standing in this version of contractualism, because many of the contracting parties care deeply about them.
What we need to know is, what counts as a reasonable basis on which to reject a proposed rule? It seems clear at least that it will not be reasonable for someone to reject a rule, if others would have an equal basis on which to reject any proposed rule. For in that case we should not be able to satisfy our shared aim of reaching free and unforced general agreement. It cannot be reasonable, therefore, to reject a rule merely because it conflicts with some interest or concern of mine. For every rule (except the entirely trivial) will conflict with someone’s concerns. Perhaps I care deeply about the welfare of animals. But then others care deeply about standards of dress and appearance, modes of sexual activity, and the worship of their God. If I can reasonably reject rules that accord no weight to the interests of animals, then others can equally reasonably reject rules that allow us to dress and make love as we wish, and to worship or not worship as we please. Even rules against killing might be equally reasonably rejected by some people, since they may want very much to kill those who stand in their way.
What can be reasonably rejected, are rules that accord no weight to my interests in general, or rules that allow my privacy to be invaded, or my projects to be interfered with, at the whim of other people. For since I know that others will similarly have reason to reject a rule that allows me to interfere in their lives, yet since I desire that we should agree on some rules to govern our conduct, I shall be happy to give up my right to interfere in the lives of others in order to gain an equal protection against interference in my own life. Indeed, it appears that here, as before, the basic principle that we should agree upon is one of respect for the autonomy of rational agents.
I therefore conclude that Scanlon’s contractualism, like Rawls’s, will fail to give moral standing to animals, in so far as animals do not themselves have the status of rational agents. But let me stress again that this is not to say that a contractualist cannot care very deeply about animals. The point is that not all cares and concerns are moral ones. Just as someone can care deeply about architecture without believing that some buildings have moral standing, or that those buildings have a right to be preserved (not derivative from the fact that people such as themselves care deeply about them); in the same way one can be an animal lover without thinking that animals have rights.
Two varieties of indirect significance
The claim that animals must lack moral standing under contractualism does not necessarily imply that one can, with impunity, do whatever one wishes with respect to any animal. For they may yet have indirect moral significance. This remains to be investigated. The issue is important because we do need to consider whether contractualism can at least approach our common-sense attitude to animals. If contractualism cannot explain any of our ordinary moral judgements in this domain, then that will count against its acceptability as a moral theory, under reflective equilibrium. Two obvious ways in which contractualism might accord indirect moral significance to animals, would be to subsume animals under the rules dealing with private property, or by treating them as a matter of legitimate public interest. Let us consider each of these in turn.
If contractualism would condone a system of property rights, as seems plausible, then it is clear that at least some animals will be protected by those rights. If you have the right against me that I should not, other things being equal, destroy your property, then I am morally obliged not to kill your dog, just as I am obliged not to set light to your car. But notice that it is your rights that I would infringe, not the dog’s. Indeed, the dog would have no rights, any more than the car does. Notice, moreover, that a great many animals, including those in the wild, will not be protected by property rights, since they have no owners (though some may receive legal protection in national parks or game reserves). More importantly, perhaps, even those animals that have owners will receive no protection against their owners. Since I am within my rights to batter or destroy my car, if I choose to do so, I should also be within my rights to batter or destroy my dog, on this approach. It seems that an appeal to property rights cannot take us very far in attempting to reconcile contractualism with common-sense attitudes.
A more plausible approach would be to appeal to the fact that many people care deeply about animals. For this may then make the manner of our treatment of animals a matter of legitimate public interest. Compare the fact that many people care deeply about architecture and the aesthetics of their environment. This may be sufficient to give rise to a moral duty, on the part of the owners of some attractive ancient building, not to destroy or alter it, except for very powerful reasons (such as that the building has, through previous neglect, become a danger to life). The general point is that one might expect contracting rational agents to reject rules that place no side-constraints on the rights of ownership. Where the objects privately owned are nevertheless a matter of legitimate public enjoyment or interest, it may be reasonable that owners’ rights of disposal of their property should be constrained, to some degree.
Similarly, then, in the case of animals. Since many people have concerns for animals, and are deeply distressed at seeing an animal suffer, this may place on us an obligation not to cause suffering to animals, except for powerful reasons. This would not be because needlessly causing such suffering would violate the rights of the animal, any more than someone who defaces a beautiful building violates the rights of the building. On this approach animals, like buildings, would have no direct rights or moral standing. Rather, causing suffering to an animal would violate the right of animal lovers to have their concerns respected and taken seriously.
Such an approach may be able to recover for contractualism a great deal of what common-sense tells us about the moral treatment of animals. In particular, it can explain how it can be true that, while we do have duties towards animals, their lives and interests cannot be weighed against the lives and interests of humans. For the duties in question only arise indirectly, out of respect for those who care about animals. And this duty of respect may surely be overridden where someone’s more fundamental interests or very life are threatened. Thus consider, once again, the example of the ancient building. If the building is the owner’s only residence, and if structural changes are necessary to make it habitable, then it is surely permissible that it should be altered, overriding the interests of the wider public.
Just how strong would the side-constraints on animal suffering be, on this approach, resulting from the legitimate feelings of animal lovers? Plainly, as we have just seen, they would fail to rule out actions causing suffering to animals that are necessary to subserve some important human concern, such as, arguably, the testing of new medicines upon animals. But more importantly, the constraints would only apply to suffering that occurs in a manner that is unavoidably public. So, in particular, painful methods of factory farming, and the testing of detergents on animals, would not be ruled out, even granting that the purposes subserved by such activities (cheaper meat, and new varieties of shampoo) are trivial. For it seems that one can legitimately reply to those who complain of such activities in exactly the way that one would reply to those who are distressed by unusual sexual practices, for example. One can say, ‘If it upsets you, don’t think about it’. While granting that an unusual sexual practice (or the suffering of an animal) should not be flaunted in public, because of the offence this may occasion, it seems there can be no objection to it occurring in private. So while the present suggestion concerning the contractualist attitude to animals can accommodate a good deal of common-sense belief, it gives no support to those who are currently campaigning on behalf of factory farmed and laboratory animals.
The present approach does face difficulties, however, quite apart from its consequences for the controversial practices of factory farming and laboratory testing. For there are two important elements of common-sense moral belief that it cannot accommodate. The first is that duties towards animals can arise equally in the private as in the public domain. While the approach can explain why it may be wrong to severely beat a dog in the street, it is not obvious why it should also be wrong to torture a cat in the privacy of your own home. For those people who would be distressed by the cat’s suffering, were they to observe it, will in fact remain unaware of it. Yet it may be said that such an action would still, intuitively, be very wrong. Secondly, it is also part of common-sense belief that cruelty to an animal is wrong because of what is done to the animal, not because of any suffering caused to sympathetic human observers, as the present approach would suggest.
There is perhaps more that contractualists can say in their defence along these general lines. For example, since animals, unlike most items of property, are capable of independent motion, there is a greater risk that supposedly private actions may become public. If the cat were to manage to escape from me into the public domain while I am torturing her, then other people may yet become distressed at her condition. Such considerations are plainly pretty weak, however, as the following development of the example of Astrid, the astronaut, will make especially clear.
Recall that Astrid has left Earth on a space-rocket, on an irreversible trajectory that will take her out of the solar system and forever out of contact with her fellow human beings. Now in her rocket she carries with her a cat, and a famous work of art of which she is the legitimate owner (the Mona Lisa, say). As the years pass she becomes bored with her books and tapes, and seeks alternative entertainment. Then contrast two cases: in the first case she removes the glass cover from the Mona Lisa and uses the painting as a dart-board; in the second case she ties the cat to the wall and uses it as a dart-board. I think we should feel intuitively that there is a very great difference, morally speaking, between these two cases. This cannot be explained on the hypothesis that our duties towards animals, as towards beautiful objects, arise only out of the likely effect of our actions on the feelings of other people. For both cases are alike in that it can be known that there will be no such effect.
I think I would be prepared to grant that Astrid does nothing wrong in throwing darts at the Mona Lisa. I may regret her philistinism, but cannot claim that she violates any rights or omits any moral duties. For it is in any case true that no one else is ever going to see the painting again. (I might say that Astrid had acted wrongly in taking such a great painting with her in the first place, but that is another matter.) In contrast, it is surely wrong of Astrid to throw darts at the cat out of idle amusement, despite the fact that she may be quite certain that no person will ever become distressed at what she has done, since no one will ever know. I therefore conclude that contractualism cannot accommodate all of what common-sense tells us about the moral treatment of animals, by trying to give animals an indirect moral significance based upon the fact that many people care deeply about them. We face a choice, at this point, of either giving up contractualist approaches to morality, or giving up some common-sense beliefs.
It might be argued that the conflict with ordinary belief on this matter is not a serious problem for contractualism, since it only really arises in connection with imaginary examples. For in any real case of private cruelty to an animal there will be a danger that the public should become aware of it. But there are two replies to this. The first is that imaginary examples cannot be belittled merely because they are imaginary. When we consider the case of Astrid, the astronaut, we feel very strongly that it would be wrong of her to throw darts at her cat. This attitude has as much right as any other to be considered as part of common-sense, even though the example is not real. For the way in which we respond to that example is perfectly real. The second reply is that common-sense does not, in any case, merely tell us that cruelty to an animal is wrong. It also tells us that it is wrong because of what is done to the animal, rather than because of the effects on a likely observer. This has not yet been accounted for.
A problem for reflective equilibrium
As we have just seen, there is a problem for contractualism in attempting to reconcile itself with our common-sense beliefs concerning the appropriate moral treatment of animals. I propose to leave this aside for the moment. I shall return to it in Chapter 7, where I shall show how contractualism can achieve reflective equilibrium on these matters. I shall now pursue a more direct and serious challenge. For contractualism does not just come into apparent conflict with common-sense morality in connection with animals. It also faces difficulties concerning the moral treatment of those human beings who are not, on any account of the matter, rational agents, such as young babies, very senile old people, or severe mental defectives. This is a much more serious difficulty, since the beliefs in question are more centrally embedded within common-sense morality.
If animals are not accorded moral standing under contractualism, on the grounds that they are not rational agents, then it would seem that by the same token all those human beings who are not rational agents will also fail to have moral standing bestowed on them. In which case killing a baby or a senile human being would not violate their rights, since they would have no rights. Such killings would at most violate our duty to respect the feelings of those people who care about babies (or that particular baby) or the senile. This is, to say the least, counter-intuitive.
In the case of babies there may be more that a contractualist can say to explain the wrongness of causing them suffering. For such suffering may be expected to have an effect on the rational agents they will one day become. Our actions may thus directly violate the rights of those future persons, and hence be wrong even when done in private so as to cause no distress to others. In the same way, contractualists may be able to explain the wrongness of killing babies, if they are prepared to accept the principle that it is wrong to prevent a rational agent from coming into existence. (What they say about this will clearly have implications for their attitude towards abortion and contraception.) But they cannot similarly explain the wrongness of killing or causing suffering to mental defectives or senile old people, since such human beings no longer have the potential, in general, to become rational agents.
To make the case as strong as possible, consider again the example of Astrid, the astronaut. Suppose that Astrid has taken her grandfather with her, who becomes increasingly senile as the journey progresses. Would it not be very wrong of her to start using him as a dart-board to relieve her tedium, or to kill him because the sight of his dribbling offends her? Yet on what grounds can such actions be wrong, if only rational agents have moral standing? For no other person will ever be worried or upset at the suffering or death of her grandfather.
It appears that contractualism faces severe difficulties in accommodating our common-sense attitudes towards those living beings who are not rational agents. Since these attitudes are even more deeply entrenched in connection with non-rational human beings than they are in connection with animals, any attempt to brush common-sense beliefs aside, on the grounds of their conflict with the theory of contractualism, will be correspondingly weaker. For example, no one is going to accept the testing of detergents on the senile, or the hunting of mental defectives for sport. If we cannot find some other way of handling these examples within contractualism, it would appear that the latter is doomed as an acceptable moral theory. I shall now consider a number of different ways in which contractualists might respond.
Family lines and the prospect of senility
Rawls himself has a way of securing direct moral rights for all human beings. This is by making the agents behind the veil of ignorance choose on behalf, not just of themselves, but of family lines. The main point of this proposal for Rawls, is to give the same weight to future generations as to the people of the present in the application of the difference principle - arguing, for example, that it would be wrong of us to exhaust the Earth’s natural resources. But the proposal is equally serviceable in according rights to babies, the senile, and mental defectives. For it has the effect of securing the same moral rights, not just for all rational agents, but also for all children of rational agents. Then since every human being - whether baby, senile, or mental defective - is the child of (or, at any rate, is descended from) rational agents, the conclusion will be that all human beings have the same basic moral rights. But since no animals are descended from rational agents, we are supposing, no animals will have direct rights.
The first thing that needs to be asked, is whether the position can be properly theoretically motivated. Or is it theoretically arbitrary, like the suggestion discussed earlier, that some people behind the veil of ignorance might be delegated to represent the interests of animals? Remember that those behind the veil of ignorance are to have knowledge of all general facts about the human condition, as well as about human psychology. They will therefore know that it is highly likely that they will have children, and that they will care very deeply what happens to those children. It then appears entirely reasonable that they should insist on direct rights for all children of rational agents (and hence for all human beings). Rather than being arbitrary, the present proposal flows directly from Rawls’s characterisation of the veil of ignorance.
This is one of those places where the artificiality of Rawls’s construction may matter, however. For it is not nearly so obvious that the above argument can survive translation into other varieties of contractualism. Consider Scanlon’s version, for example. Many real agents know that they will never have children. Others may know that their parents have died before becoming senile. Such persons might, it seems, reasonably reject rules according direct moral rights to babies, mental defectives, or the senile. This is in just the same way that those who know that they are indifferent to art may reasonably reject the proposal put forward by the majority of people who are art-lovers, that works of art should be accorded direct moral rights. If we are to find convincing contractualist arguments for according moral standing to all human beings, it seems we should look elsewhere.
A different suggestion is as follows. As a rational agent I know that it is likely that I shall, one day, slide gradually into a condition of senility. I also know that it is possible for an accident to reduce me to the level of a severe mental defective, or a baby. But, it is claimed, I would surely wish to preserve for myself the same basic moral rights and protections in those circumstances as I now enjoy. If I were actually in a state of senility, of course, I should be in no position rationally to reject a system of rules withholding moral rights from the senile. But I may now rationally reject such rules, on the grounds that they conflict with what I want for myself were I to become senile, and since I can see that all have as much reason as myself to reject such rules.
If the argument above were successful, then we might be able to move on to grant moral standing to all human beings, irrespective of their cognitive powers. For, given that moral standing is to be accorded to those who become senile, or severely subnormal as a result of an accident, it would surely be intolerable that moral standing should be withheld from those who are born so. And then if moral standing is granted to those who, as adults, are congenitally severely subnormal, it would seem that there can be no rational basis for withholding it from young babies, who may enjoy similar levels of cognitive activity.
There are two reasons why this attempt to extend the same basic rights to all human beings must fail, however. The first is relatively simple. It is that not everyone wishes that they should continue to enjoy the same moral protections were they to become senile. It is common for people to say, indeed, (especially as the prospect of senility becomes increasingly real) that they only hope that someone has the nerve to kill them off quickly once they reach that state. The second reason for failure is more deeply metaphysical. It is, that it is doubtful whether personal identity can be preserved through cognitive changes as massive as the slide into senility. Although at the end of such a change the very same human being (or physical body) would still exist, of course, it is highly doubtful whether I would any longer exist. For the resulting human being would have none of the same beliefs, desires, interests, memories, or qualities of character that - arguably - constitute my identity as a person. This, too, accords with ordinary parlance. It is common for people to say of their friends or relatives in such circumstances, such things as ‘It is not really Granny in that hospital ward, any longer’. But if, following the slide into senility, the resulting person is not me, then I cannot now self-interestedly reject rules that affect that person. Yet this is what I would have to be able to do, if the senile are to be accorded moral standing, on this approach.
Slippery slopes and social stability
There is a very different way in which contractualists, of whatever variety, can attempt to secure direct moral rights for all human beings. As with the Rawls’s suggestion, this one, too, will leave animals without moral standing. The strategy depends upon the fact that there are no sharp boundaries between a baby and an adult, between a not-very-intelligent adult and a severe mental defective, or between a normal old person and someone who is severely senile. The argument is then that the attempt to accord direct moral rights only to rational agents (normal adults) would be inherently dangerous and open to abuse.
This is, of course, a version of slippery slope argument. The suggestion is that if we try to deny moral rights to some human beings, on the grounds that they are not rational agents, we shall be launched on a slippery slope which may lead to all kinds of barbarisms against those who are rational agents. It is important to be clear about the level on which this argument is supposed to operate, however. For there is nothing to stop us, at the level of theory, from insisting that only rational agents have rights, leaving a large range of cases in which possession of rights would be indeterminate. Or we could insist that possession of rights itself should be a matter of degree, the killing of a human being becoming more and more serious, in terms of direct infringements of right, as a baby gradually advances into adulthood. There would be nothing incoherent in these theories as such. The claim must be that it is in the application of these theories in the real world that the danger lies. The idea is that such theories would be inherently susceptible to abuse by unscrupulous people, and ought therefore not to be adopted.
In contrast, there really is a sharp boundary between human beings and all other animals. Not necessarily in terms of intelligence or degree of rational agency, of course - a chimpanzee may be more intelligent than a mentally defective human, and a dolphin may be a rational agent to a higher degree than a human baby. But there is not the same practical threat to the welfare of rational agents in the suggestion that all animals should be excluded from the domain of direct moral concern. Someone who argues that since animals do not have rights, therefore babies do not have rights, therefore there can be no moral objection to the extermination of Jews, Gypsies, Gays and other so-called ‘deviants’, is unlikely to be taken very seriously, even by those who share their evil aims.
This argument for according rights to all human beings does seem to have a good chance of success. For rational agents choosing moral principles to govern their behaviour should of course pay attention to the ways in which those principles might be distorted or abused. If the argument has a weakness, however, it lies in its empirical assumption - namely, that a rule according direct rights only to rational agents would be likely to be abused in such a way as to undermine itself. For provided that all understand the theoretical basis of the rules, they will be fore-armed against abuse. Thus suppose it were generally agreed that all rational agents have moral rights, and that those who are not fully rational agents have rights in proportion to the extent of their rational agency. Then the reply is obvious to anyone who tries to argue that since babies do not have direct rights, and since there is no clear boundary between infancy and normal adulthood, therefore there can be no direct moral objection to the holocaust. It is that the gradual transition from infancy to adulthood is at the same time the transition from not bearing moral rights, to having them in full measure.
This attempt to undermine the argument from a slippery slope fails in its turn, however. For one of the facts that rational agents will know, is that most people are not very deeply theoretical. They should therefore select moral principles that will provide a stable and easily understood framework within which ordinary people can debate questions of right and wrong. Seen in these terms, a rule that accorded rights in proportion to degree of rational agency would be wide open to creeping abuse. For to think and speak in terms that withhold moral rights from some human beings is to invite people to try to draw yet further distinctions - for example, withholding rights from those who are sexually or intellectually ‘deviant’, or from those whose intelligence is low. So I conclude that our slippery slope argument is indeed successful in according rights to all human beings.
(It is worth noting the differences between the argument sketched here, and Regan’s superficially similar argument for treating human babies as if they had rights. Recall that for Regan, those who have moral rights are primarily those who are subjects-of-a-life - that is, who have a sense of their own future and their own past. He then realises that on such an account human babies, up to the age of one, at least, will not count as having rights. His reply is that we should, nevertheless, treat such babies as if they had the same rights as everyone else, by way of encouraging a moral climate in which the rights of individuals are taken seriously. The first point to make about Regan’s proposal is that it does not succeed in according rights to human babies. To say that we should treat babies as if they had rights is not the same as saying that they do have rights. Yet it is this stronger conclusion that we were able to deliver by means of the slippery slope argument outlined above. The second point is that it is, in any case, by no means clear how Regan’s argument is supposed to go. That is, it is far from clear how treating those who do not have rights as if they did have them would foster a climate in which the rights of individuals are taken seriously. The only obvious suggestion is that any moral system in which some human beings are denied moral rights is liable, by creeping abuse, to lead to a situation in which some of those who do have moral rights have their rights ignored. This is, in effect, our slippery slope argument, only shorn of its contractualist context. The very argument that for Regan leads to the conclusion that we should treat all humans as if they had rights, for a contractualist leads to the conclusion that they do have rights. This is, I think, to the advantage of the latter.)
Would a slippery slope argument for according direct moral rights to all human beings at the same time rule out abortion? For there is no clear line between foetus and baby, any more than there is a clear line between baby and adult. But in fact the issues here are not the same. For one of the things that contracting rational agents should consider seriously, in framing their rules, are the natural responses of thought and feeling that antecede moral belief. (This point will come to the fore in Chapter 7.) It is natural to be struck by the suffering of senile old people or babies, in a way that both supports and is supported by assigning direct rights to these groups. It is not so natural for us respond similarly towards a foetus, however, especially in the early stages, unless we already have prior moral beliefs about its status. A rule withholding moral rights from foetuses, and hence permitting at least early abortions, may therefore be quite easily defended against abuse. This will become clearer in the chapter after next.
In addition to the argument from a slippery slope outlined above, contractualists have available one further argument for according moral standing to all human beings. This is an argument from social stability. One thing that rational contractors should certainly consider, in framing a basic set of principles, is whether those principles would have the desired effect of facilitating stable, co-operative, community. In this they should have regard, among other things, to the known facts of human psychology. One such fact is that human beings are apt to care as intensely about their offspring as they care about anything, irrespective of age and intelligence. A rule withholding moral standing from those who are very young, very old, or mentally defective is thus likely to produce social instability, in that many people would find themselves psychologically incapable of living in compliance with it.
It might be replied that stability could equally well be achieved by a rule requiring us to respect the legitimate concerns of others. Then all those non-rational humans who are objects of love would receive protection after all, out of respect for the feelings of those who love them. But this is inadequate. It would only accord such humans the same protection as items of property. Just as I am obliged not to damage or destroy your cherished Mercedes, so I should be obliged not to damage or destroy your child. But such obligations may be overridden in cases where more fundamental rights are at stake. Suppose, for example, that your Mercedes blocks the entrance to a mine-shaft in which I have become imprisoned. You have become accustomed to use the entrance as a garage during the week, and I should face a five day wait to get out. Then I may surely destroy the car if this is my only means of escape, no matter how much you may care about it, and even though my life may be in no danger. In these circumstances you would, surely, accept that I had acted reasonably. But no one could bring themselves to accept with equanimity the destruction of their child in a similar situation. The only way of framing rules that we can live with, then, is to accord all human beings the same basic rights - that is to say, moral standing.
A reply from anthropology
In reply to both of the above arguments, it may be objected that there have been many human societies that have not accorded the same basic rights to all human beings, and yet that have been both stable and in other respects civilized - no slippery slope was ever embarked upon. There have been very many human communities in which infanticide has been widely practiced as a means of population control, for example. Yet the members of these communities were in all other ways inclined to show at least as much respect for human life as ourselves, and were tender and loving towards those children who were allowed to survive. So, it may be claimed, it is simply not true that a system of morality that only accords moral standing to those human beings who are rational agents need in any way be self-defeating, or have dire consequences.
This objection fails, however, for a variety of reasons. First, all the communities in which infanticide has been openly practiced have been traditional ones, with the practices in question sanctioned by long-standing social custom, and often by religious belief as well. Such methods of achieving social stability are no longer available to us. In our modern world, moral rules have to be defensible in the face of free and open discussion, without appeal to religious sanction. In order to show that all human beings should be accorded the same basic rights, I do not have to argue that a rule withholding moral standing from some human beings would be disastrous in all circumstances. It is enough that it would be disastrous for us.
Second, almost all the communities in question were teetering on the edge of survival, either existing in particularly harsh environmental conditions, or in areas where productive land was in short supply. Infanticide was thus deemed necessary to prevent general starvation, or to preserve the lives of older children. It is, then, by no means obvious that these communities failed to recognize the moral standing of human infants. Their acts of infanticide may rather be assimilated to legitimate cases of killing in self preservation. Any version of contractualism will surely allow such killing, as the following example will make clear.
Suppose that Doris and Diana are deep sea divers, whose diving bell has come adrift and sunk to the bottom of an ocean trench. They are told by radio that they cannot hope to be rescued for at least twelve hours. They have only six hours of oxygen remaining. So at least one of them must die. Now suppose, additionally, that Diana in any case depends for her survival upon Doris (just as a human infant depends for its survival upon adults) - perhaps she needs Doris to administer a life-saving injection after ten hours, that she cannot administer herself. In these circumstances it is surely permissible that Doris should kill Diana in order that she herself may survive. While it may be admirable of Doris if she were to volunteer to die along with Diana, this cannot be morally required of her. That she should be prepared to kill in these circumstances need not mean denying that Diana has moral standing or, indeed, violating her rights. For rational contractors ought surely to allow that in those rare cases where all will die unless one is killed, it is legitimate to preserve oneself. Then so, too, in the case of infanticidal communities - their actions may be consistent with according full moral standing to infants.
The third reason why the anthropological evidence fails to undermine the slippery slope and social stability arguments outlined earlier, involves a distinction between the general virtues of justice and beneficence, which is in any case central to contractualism, as we shall see in Chapter 7. (Justice has to do with duties of non-interference, beneficence with attachment to the welfare of others.) Suppose that human infants are accorded full moral standing, on a par with normal adults. Then the fact that they are incapable of surviving on their own means that the actions necessary to keep them alive are required by beneficence rather than justice. It need not infringe a person’s rights if we fail to keep them alive, though it may display a serious lack of generosity on our part. But in circumstances where the costs to us would be severe (as they would be in most of the cases we are considering), our failure to keep alive need not even show this. Recall the example of callous Carl from Chapter 2, who failed to save the child drowning in a pond. His action would surely appear in quite a different light if he were rushing his own child to hospital for an emergency operation. That he fails to show beneficence towards the drowning child in such circumstances involves no violation of right, nor need it involve any sort of denial that the child has full moral standing. This example looks entirely parallel to the cases of the infanticidal communities we have been considering.
Since the anthropological objection fails, I conclude that contractualists have at least two successful strategies for according direct moral rights to all human beings. The only problem still facing contractualism, therefore, is to get closer to common-sense attitudes towards animals. For the intuition that Astrid would act wrongly in using her cat as a dart-board is a powerful one. I shall return to the matter in Chapter 7. First, I shall spend a chapter considering the extent to which it is true that animals should not be counted as rational agents, as I have been assuming up to now. For if it were to turn out that most animals are rational agents after all, then reflective equilibrium would in any case have been attained. Since it is clear that only prejudice can stand in the way of assigning the same basic rights to all who are rational agents, we should have found an explanation for why Astrid is wrong to throw darts at her cat.
No version of contractualism will accord moral standing to animals. There may, nevertheless, be indirect duties towards animals, owed out of respect for the legitimate concerns of animal lovers. But the protection thus extended to animals is unlikely to be very great. Nor can this approach explain the common-sense intuition that unmotivated cruelty to an animal is directly wrong. Contractualists also face the challenge of extending direct moral rights to those human beings who are not rational agents. While the first two avenues discussed, through which contractualists might hope to grant such rights, were seen to fail, two others - a slippery slope argument, and an argument from social stability - proved successful.
 See The Case for Animal Rights, ch. 5.4.
 This is made especially clear in ‘Justice as Fairness: Political not Metaphysical'.
 See A Theory of Justice, sect. 22.
 On this, see my Introducing Persons, chs. 7 & 8.
 See The Case for Animal Rights, pp. 319-20.
 See Helga Kuhse and Peter Singer, Should the Baby Live? (Oxford University Press, 1985), ch. 5.