Do Animals Have a Right to Liberty?
In TOM REGAN & PETER SINGER (eds.), Animal Rights and Human Obligations, Englewood Cliffs: Prentice-Hall, 1976, pp. 205-223
Philosophers used to talk about "natural" rights, but now we don't hear so much about that subject. Instead, books and articles are written about "human" rights. The change in terminology is thought to be a great improvement; first, because talk about human rights does not bring with it the ontological worries that often attended discussions of natural rights, and second, because the new terminology focuses more precisely on what we are trying to understand: the rights that all human beings have in common. One of my motives in raising the question in my title is to cast doubt on the importance of human rights. I will maintain that human rights are not nearly so interesting or important as philosophers and politicians have thought.
As Richard Wasserstrom puts it, "If any right is a human right, ... it must be possessed by all human beings, as well as only by human beings." What is usually emphasized is that such rights are possessed by all humans; thus the doctrine of human rights has been a formidable weapon against slavery, racism, sexism, and the like. But, as Wasserstrom correctly notes, if any right is a distinctively human right, it is also necessary that it be possessed only by humans. It is this side of the doctrine that I want to emphasize. If it can be made plausible that members of other species also have the rights that are most important to humans—such as the right to liberty—then the whole subject of human rights will come to have much less interest than before; and it will be seen that the differences between humans and other animals are not nearly so important, from a moral point of view, as we have usually assumed.
Some philosophers believe that nonhuman animals (I will sometimes follow the common practice and call them simply "animals," leaving off the qualifier) have no rights at all, because they are not the sorts of beings that can have rights. On their view, it is not logically possible for animals to have rights. Two things need to be done: first, their arguments must be refuted, and second, positive arguments must be advanced to show that animals do have specific rights. I take up the first task in an appendix to be found at the end of this paper. I have relegated this essentially negative material to an appendix so that I can go on here to the second, more positive task.
In arguing that animals do have rights—and in particular that they have a right to liberty—we may use the following method. First we select for discussion a right which we are confident that humans do have. Then we ask whether there is a relevant difference between humans and animals which would justify us in denying that right to animals while at the same time granting it to humans. If not, then the right in question is a right possessed by animals as well as by humans.
This method has a number of virtues. First, it has a clear rationale in the familiar principle of justice that we must treat like cases alike; or, to be more precise, that our moral judgments are unacceptably arbitrary if we judge one way in one case and differently in another case, without there being a relevant difference between the two cases which justifies the difference in our assessments. This principle has been used with great effect in arguing against racism. The assumption there has been that a person's race is not in itself a morally relevant consideration in determining how he is to be treated. Therefore, racist discrimination is unjustified unless some further differences between blacks and whites can be found which would be relevant to justifying the different modes of treatment. But, because there are no such further differences, such discrimination is unjustified. I am going to make the similar assumption that a mere difference in species is not enough, in itself, to justify any difference in how beings are treated. Thus if we want to grant a right to humans but deny it to members of other species, we must be able to point to some relevant difference between them other than the mere fact that the animals are members of another species. A second advantage of the method is that if we follow it closely we will avoid the trap of lumping all nonhuman animals together, as though what we say about one species we must say about all. For it may turn out that, with respect to some particular right, there is no relevant difference between man and one species of animal, but there are differences between man and other species. Finally, I should mention one limitation of this method. The use of this method does not guarantee that we will identify all the rights which animals have, for it is at least logically possible that they have some rights not possessed by humans. If so, then these rights could not be uncovered by my method. However, this is of no concern to me here, for I have no intention of trying to compile a complete list of animal rights.
Now let me give some illustrations of the kinds of results which may be obtained by this method. Article 5 of The United Nations Universal Declaration of Human Rights says that all men have a right not to be subjected to torture. But is this, in fact, a distinctively human right? If members of other species—say, rabbits or pigs or monkeys—are tortured, they also suffer. Of course, there are many impressive differences between men and these animals, but are they relevant here? A man can learn mathematics, and a rabbit can't; but what does that have to do with the business of being tortured? A man has an interest in not being tortured because he has the capacity to suffer pain, and not because he can do mathematics or anything of that sort. But rabbits, pigs, and monkeys also have the capacity to experience pain, and so they have the same basic interest in not being tortured. The right not to be tortured, then, is shared by all animals that suffer pain; it is not a distinctively human right at all. On the other hand, Article 18 of the same Declaration says that all men have the right to worship as they please. This, I think, is a right belonging only to humans, because only humans have religious beliefs and a capacity for worship.
The right not to be tortured, and to freedom of worship, are relatively clear and unproblematic. But what happens when we consider a more puzzling right, such as the right to property? Here we may proceed by asking why it is thought that men have this right—what is the basis of it?—and then, whether the same case can be made in behalf of animals.
Let us consider, for example, Locke's treatment of the right to property. Locke contends that a man has a natural right to his own labor and whatever he produces by it:
The labor of his body and the work of his hands, we may say, are properly his. Whatsoever then he removes out of the state that nature has provided and left it in, he has mixed his labor with, and joined to it something that is his own, and thereby makes it his property.
Locke then illustrates his view with this example:
He that is nourished by the acorns he picked up under an oak, or the apples he gathered from the trees in the wood, has certainly appropriated them to himself. Nobody can deny but the nourishment is his. I ask, then, when did they begin to be his? When he digested or when he ate or when he boiled or when he brought them home? Or when he picked them up? And it is plain, if the first gathering made them not his, nothing else could. That labor put a distinction between them and common; that added something to them more than nature, the common mother of all, had done; and so they became his private right.
If Locke is right, then it follows that animals such as squirrels also have a right to property; for squirrels labor to gather nuts for their own nourishment in exactly the way Locke pictures the man laboring. There is no relevant difference between the man and the squirrel: they both pick up the nuts, take them home, store them away, and then eat them. Therefore there is no justification for saying that the man has a right to the nuts he gathers, but that the squirrel does not.
Now I turn to the right to liberty. The right to liberty has been counted among the most fundamental human rights in all the great liberal manifestos of modern history—the Declaration of Independence of the United States (1776), the French Declaration of the Rights of Man (1789), and the United Nations Universal Declaration of Human Rights 1948), to name three of the most important. Virtually every philosopher who has discussed the subject has followed suit; I have not been able to find any treatment of "human rights" which did not include liberty as a prime example. Considering this, and remembering that some philosophers doubt whether mere animals can have any rights at all, it may not be surprising to find liberty (or freedom, which for present purposes comes to the same thing) being defined by some in such a way that only humans could possibly be free. According to J. R. Lucas, for example,
The central sense of Freedom is that in which a rational agent is free when he is able to act as seems best to him without being subject to external constraints on his actions.
If we start off by conceiving freedom in this way, then the question of whether animals have a right to be free will not even arise, since the notion of a "rational agent" who deliberates about which actions are best is so obviously formulated with only humans in mind. But, just as obviously, this definition won't do as a general definition of freedom, for that concept applies to animals as well as to men. A lion left alone in his natural habitat is free; a lion in a zoo is not. A chicken in a small wire cage is less free than one allowed to roam about a barnyard. And a bird who is released from a cage and allowed to fly away is "set free" in a perfectly plain sense. So, rewriting the definition to eliminate the prejudice in favor of men, we get:
The central sense of Freedom is that in which a being is free when he is able to do as he pleases without being subject to external constraints on his actions.
This expresses well enough the concept of liberty with which I shall be concerned. As before, we may proceed by asking why it is thought that men have this right—what is the basis of it?—and then, whether the same or a very similar case can be made on behalf of members of other species.
One possibility is to take liberty to be, simply and without need of any further justification, good in itself.8 If we take this approach, then we might argue that men have a right to liberty simply because they have a right not to be deprived of any intrinsic goods which they are capable of enjoying. (And here the usual qualifications will be added, to the effect that the right will be only as extensive as is compatible with others having a similar right, that the right may be forfeited or overridden in certain circumstances, etc.) But this line of reasoning will apply equally well to other species of animals. It is parallel to the right not to be tortured, mentioned above. Any animal that has the capacity for suffering pain has a right not to be tortured; and the reason for this is connected with the fact that suffering pain is intrinsically bad. Similarly, if we grant to men a right to liberty simply because we regard liberty as something good in itself which men are capable of enjoying, then we must also grant a right to liberty to any other animal that is capable of desiring to act one way rather than another.
However, not many philosophers would be happy with this approach, because most believe it is possible to provide a rationale for the right to liberty that does not simply stop with calling it an intrinsic good. For example, it may be said that humans have a right to liberty because they have various other interests that will suffer if their freedom is unduly restricted. The right to liberty—the right to be free of external constraints on one's actions—may then be seen as derived from a more basic right not to have one's interests needlessly harmed.
But the interests of many other species are also harmed by a loss of freedom. It is a familiar fact that many wild animals do not fare at all well in captivity: taken from their natural habitats and put in zoos, they are at first frantic and frustrated because they cannot carry on their normal activities; then they become listless and inactive, shadows of their former selves. Some become vicious and destructive. They often will not reproduce in captivity, and when they do, their young often cannot survive; and finally, members of many species will die sooner in captivity than they would in their natural homes.
Dr. Herbert Ratcliffe, a pathologist, conducted a study of the animals in a Philadelphia zoo. He found that the animals were suffering from sharply increased rates of heart disease, cancer, and ulcers. The metabolism of some white-tailed deer had changed to such an extent that their horns became deformed. The zoo's breeding colony of nutria—small, beaverlike animals—had dwindled because the young animals were born dwarfed, failed to breed, and died early. Dr. Ratcliffe attributes all of this to the effects of the artificial, confined environment of the zoo.
Another example is taken from a widely used psychology textbook, which tells the story of a baboon colony in the London zoo. Investigators observed many instances of bloody fighting, brutality, and apparently senseless violence. Some of the females were torn to pieces, and no infant survived to maturity. From these observations, it was concluded that such violence was typical of the "wild" baboons. . . . But later, when baboons were studied under natural conditions in Africa, in the "wild," it was discovered that they lived in well-organized, peaceful groups, in which the only aggressive behavior was directed at predators and intruders.
Once it has been learned that animals can be made to suffer in a certain way, a new field is opened for scientific research. Experiments may then be performed to discover how they will behave when tormented, and exactly what forms their suffering will take. Numerous studies have been made of the effects of confinement on animals. One such series of experiments was reported in 1972. One of the experimenters, Dr. Harry F. Harlow of the University of Wisconsin, is said to have "created" a vertical chamber, which "is basically a stainless steel trough with sides that slope inward to form a rounded bottom." The whole thing measures about four feet by one foot by a few inches. The idea behind the chamber is explained this way:
Depression in humans has been characterized as embodying a state of "helplessness and hopelessness, sunken in a well of despair," and the device was designed on an intuitive basis to reproduce such a well both physically and psychologically for monkey subjects.
Rhesus monkeys were used for the experiments. These animals are often used in such experiments because they are intelligent, sociable creatures that resemble humans in a great many ways. The experiments were conducted by putting six-week-old monkeys into the "well of despair" for a period of forty-five days. The purpose of doing this was said to be to "investigate the chamber's effectiveness in production of psychopathology."
The chamber turned out to be very effective. While confined, the "subjects" were said to "typically spend most of their time huddled in a corner of the chamber." "Huddling" is defined as a "self-enclosed, fetal-like position incorporating any or all patterns of self-clasp, self-embrace, or lowered head." A nine-month period of observation following the confinement indicates that the effects on the animals are permanent:
The results indicated that a 45-day period of vertical chamber confinement early in life produced severe and persistent psychopathological behavior of a depressive nature in the experimental subjects. These monkeys failed to show appreciable changes in home-cage behavioral levels during the 9-month period following removal from the vertical chamber. In comparison to control groups of cage- and peer-reared monkeys, the chambered subjects exhibited abnormally high levels of self-clasp and huddle and abnormally low levels of locomotion and environmental exploration in both the home-cage and playroom situations. Most striking was the virtual absence of social activity among chambered subjects throughout the 8 months of playroom testing.
This new knowledge was obtained with financial assistance from the United States Public Health Service, the National Institutes of Health, and the National Institute of Mental Health.
Any creature that has interests has at least a prima facie right not to have those interests needlessly harmed. Animals that suffer in captivity have an interest in being free, and so at least a prima facie right to liberty. Lucas, immediately after giving the definition of "freedom" (restricted to "rational agents") quoted above, says that "not to be free is to be frustrated, impotent, futile." He is obviously thinking only of men; but the description applies equally well to animals in zoos, and certainly to the monkeys trapped in the well of despair.
Animals raised for food also suffer in confinement. Before being slaughtered cows spend their lives crowded into "feedlots" where they are deprived of any sort of herd life or even adequate exercise. Veal calves are kept in pens so small they cannot even turn around. Peter Singer points out that even the lowly chicken suffers from confinement in the sort of cages used by poultry-farmers:
. . . hens are crowded four or five to a cage with a floor area of twenty inches by eighteen inches, or around the size of a single page of the New York Times. The cages have wire floors, since this reduces cleaning costs, though wire is unsuitable for the hens' feet; the floors slope, since this makes the eggs roll down for easy collection, although this makes it difficult for the hens to rest comfortably. In these conditions all the birds' natural instincts are thwarted: they cannot stretch their wings fully, walk freely, dust-bathe, scratch the ground, or build a nest. Although they have never known other conditions, observers have noticed that the birds vainly try to perform these actions. Frustrated at their inability to do so, they often develop what farmers call "vices," and peck each other to death. To prevent this, the beaks of young birds are often cut off.
Some of these cruelties have to do with the type of confinement rather than with the bare fact that the birds are confined. So, if the cages had flat, solid floors, and perches for the hens, some of the grounds for complaint would be eliminated. But so long as the hens are confined to small cages, their natural desire to scratch the dirt, stretch their wings, build a nest, and so forth, will be frustrated. This is not to say that the interests of chickens can be satisfied only in a state of total freedom: I can see no harm that would be done to their interests if they were kept captive while being allowed freedom to roam a large area, where they could do the things just mentioned. Thus many vegetarians who refuse to buy eggs produced under the conditions described by Singer nevertheless will buy eggs laid by "free-ranging" hens.
So, we need to distinguish two things: first, we need to distinguish the kinds of animals whose interests are harmed by the denial of freedom; and second, we need to distinguish the degree of freedom that is required if the animals' interests are not to be harmed. Lions, but not chickens, may need to be set completely free in their natural habitats in order to thrive; whereas the needs of most insects may be so limited that they have no interest in freedom at all.
At this point the business about man's superior rationality must be re-introduced. For, even if it is a mistake to define freedom in such a way that only rational agents can be free, it may still be said that freedom has a special kind of importance for rational agents which it cannot have for nonrational beings. In one form or another this thought is found in the writings of almost all the philosophers who discuss the "human right" to liberty. I want to make two preliminary remarks about this. The first has only to do with a certain sentiment that I have—so you may want to discount it as an argument—but I will mention it anyway. It is that there is something very sad about a grand animal such as a lion or an elephant being put on exhibit in a zoo, and being reduced to nothing more than a spectacle for people's enjoyment. The reason I mention this here is that, in the past, humans who lacked "rationality" have suffered the same fate. Salt notes that
Two or three generations ago, pauper-lunatics used to be caged where passers-by—nurses perhaps with children in their charge—could see them as they passed, and the spectacle was sometimes enjoyed. (I remember hearing from my mother that such was the case at Shrewsbury. The nurse would say, "Where shall we go to-day, children?" and the cry would be, "Oh, to see the madmen, please!")
Most of us recoil at this, and many reasons may be given why such practices are barbarous: perhaps because they teach children callous attitudes. But of course making a similar spectacle of animals may also have that effect. However, it is hard to believe that our initial reaction has much to do with such considerations. It has to do rather with the sadness and indignity of the spectacle. And the fact that the being on exhibit is not rational hardly matters, either in the case of the lunatic or the lion. The second comment is to express a general doubt about the relevance of rationality to the value of freedom. It may be true, as philosophers have often stressed, that liberty is necessary if we humans are to develop and exercise our powers as rational agents, and to have the kinds of lives we want. But it is also true that liberty is necessary for many nonhuman animals if they are to live the sorts of lives, and thrive, in ways that are natural to them; or, to put things more plainly, if the interests they have, in virtue of the kinds of creatures they are, are to be realized.
So, just what relevance is rationality supposed to have? One very popular view, derived from the teachings of Kant, is that because of his rational powers man alone, among all the animals, is a moral agent, able to form a conception of right and wrong and guide his actions by it. It is even said that, because of this, there are two kinds of freedom that must be distinguished in discussions such as this: the natural freedom of any creature to do as he pleases, and the special, moral freedom of man to guide his actions by his sense of right and wrong. The idea is that, when we talk of man's freedom, it is his moral freedom that is especially important and that must be protected. Thus man's right to liberty has a foundation and importance which the "liberty" of mere animals cannot have.
Although the idea of "moral freedom" is obscure, it is also very powerful, and it is easy to feel the force of it. Here, I do not want to quarrel with the notion, or even to analyze it very closely. Rather I want to suggest that some nonhuman animals have capacities which might very well be called "moral," so that, even if we do hold moral freedom to be an especially important type of liberty, there may still be no reason to deny that these animals have a claim to it. For even though these animals cannot form an intellectual conception of right and wrong, they are nevertheless capable of being motivated to act by desires which, in humans, we would take to be signs of moral goodness. For example, many non-human animals show devotion and love for their offspring, and self-sacrificial loyalty to the other members of their herd or pack or whatever. And if we think it especially important to allow humans the liberty to act on their conceptions of right and wrong, why should it not also be important to allow a nonhuman mother to act from love for her offspring? (Here again the differences between kinds of animals must be kept clearly in mind. Some animals are indifferent or cruel to their young; but others—for example, wild dogs—do show the type of pr°~ tective love to which I am referring.)
But some animals have even more impressive virtues than these familiar ones. Let me describe another series of experiments, this one conducted at the Northwestern University Medical School and reported in the psychological journals for 1964. These experiments were designed to discover whether rhesus monkeys would be deterred from operating a device for securing food if doing so would cause pain to another monkey. One animal (called by the experimenters the "operator" or "O") would be placed in one side of a divided box and taught to secure food by pulling either of two chains. Food would be available only when a light signal was given (a different light for each chain), and the O would be trained to show no special preference for either chain.
Next, another monkey (called the "stimulus animal" or "SA") would be put into the other side of the box. The box was divided by a one-way mirror so that the O could see the SA but not the other way around. The floor on the SA's side of the box was covered by a grid attached to a shock source. The O was given three days to adapt to the presence of the SA, and then the circuit was completed so that whenever he pulled one of the chains to secure food the SA received a severe electrical shock. Pulling the other chain continued to give food, but produced no shock. Now, by turning on only one signal light at a time, in various sequences and at various intervals, the experimenters could determine the extent to which the perception of the SA's distress would influence the O's willingness to pull the shock-producing chain.
The experimenters concluded that "a majority of rhesus monkeys will consistently suffer hunger rather than secure food at the expense of electroshock to a conspecific." In particular, in one series of tests, 6 of 8 animals showed this type of sacrificial behavior; in a second series, 6 of 10; and in a third series, 13 of 15. One of the monkeys refrained from pulling either chain for 12 days, and another for 5 days, after witnessing shock to the SA—which means that they had no food at all and suffered really terrible hunger.
I believe that these experiments show that rhesus monkeys have a capacity for compassion, and by that I mean exactly the same moral virtue which we admire in humans. Of course, these animals are not able to form abstract moral conceptions. But, even for humans, being compassionate does not necessarily involve forming the idea that it is good to be compassionate, or that one is morally required to act compassionately. Being compassionate only requires desiring that others do not suffer, and acting on that desire. Animals may not form abstract conceptions, but they do have desires, and apparently among the desires of the rhesus monkey is that he not cause suffering to others of his own kind. So I conclude that they have compassion. However, in order for this conclusion to be really plausible, we need more information. And the experimenters have provided what we need.
First, the experimenters tested to determine whether the O's reluctance to pull the shock chain was correlated to relative positions on a dominance-submissiveness hierarchy. Relative dominance was determined when the animals "were paired against each other in another apparatus and required to compete for 100 grapes presented one at a time. In most cases dominance was quickly established, the dominant animal getting 90 percent of the grapes." The experiments were then divided into those in which the dominant animal was the SA and the submissive animal was the O; those in which the roles were reversed; and so forth. And it was found that this made no difference to the O's willingness or reluctance to pull the shock chain.
Again, the experimenters were careful to observe whether differences in sex made any difference; that is, whether the O was male and the SA was female; whether they were both male; and so forth. This made no difference either.
These results are important because they are exactly what we would expect if the O's behavior is caused by a generalized compassion for other members of his own species rather than to fear of dominant animals or to some sort of sexual conditioning. The experimenters also rule out "increased noise level" as a possible explanation because "the SA's vocalized infrequently"—although, even if the SA's had cried out often, this would not rule out compassion as the explanation because the cries would obviously be cries of pain. Moreover, the experimenters observe that "the rage and attack mimetics of large male or female SA's during shock proved to be no more effective than those of smaller animals in deterring the feeding responses of persistently indifferent O's or expediting 'altruism' in the others." So still another alternative explanation is ruled out.
Other aspects of the experiments support the hypothesis of compassion in a different way. The experimenters found that animals that had previously been SA's were significantly more reluctant to pull the shock chain when they were made O's than animals who had not been SA's themselves. "This behavior of the shocked O's was not attributable to an acquired aversion to the apparatus itself since they showed no decrement in chain-manipulation during the adaptation sessions immediately following their shock." The explanation suggested by the hypothesis of compassion is that these animals were more reluctant to pull the chain because, having suffered the shocks themselves, they had a more vivid comprehension of what it was like, and so a greater reluctance to see someone else in the same position.
It was also found that O's who had been cage-mates of their SA's were more reluctant to pull the shock chain than O's who had not been cage-mates of their SA's. Again, this is just what we would expect if we take our common knowledge of human beings as our model: we are less willing to harm someone we know than we are to harm strangers.
This, then, is the evidence that rhesus monkeys are compassionate. I can imagine several objections being made to this conclusion. I have already answered the most obvious objection, that animals cannot have moral characteristics because they are not capable of moral thinking. The answer was that having a virtue such as compassion does not require moral conceptions, but only acting on certain sorts of desires. The conclusion that these animals are compassionate is likely to be controversial, so let us consider three other possible objections:
First: "But the monkeys only showed an aversion to causing pain for others of their own kind. Would they do the same for other kinds of animals? The results here are much too limited to justify talk about a virtue of compassion." The answer to this complaint is that even if the compassion of the monkeys is limited to a feeling for others of their own kind, their compassion is no more limited than that of most humans. Most men—even those who have a fine respect for the interests of other humans—are fairly indifferent to the interests of beings not of their own kind. (For example, men give virtually no thought to arranging experiments in which severe electroshocks will be administered to members of other species, or in which they will be reduced to a lifetime of "huddling.") We would not say that a man was completely devoid of compassion merely because he limits his concern to the suffering of other men; and we should not judge other animals more harshly than we judge ourselves.
Second: "But only some of the animals tested showed 'compassionate' behavior. Many did not." Again, this is exactly the result we would get in the case of humans. Human compassion comes in varying degrees and strengths: some of us are quite compassionate; and some are relatively indifferent to the plights of others. When we find similar variations among the monkeys, why should we be surprised? Indeed, the remarkable thing is that, considering the superior ability for moral thinking of which we are so proud, the differences between us and them are not greater than they are.
Third: "This whole way of thinking about animals is unforgivably anthropomorphic. To speak of 'compassion' in nonhuman animals is simply to read human characteristics into their behavior when they are not really there." Now it is certainly true that we should be wary of overly anthropomorphic ways of thinking about animals. We should be careful to notice the differences between them and ourselves, as well as the similarities. But I have been as careful as I can—I hope that the foregoing discussion shows at least some signs of care—and I still think that the animals are compassionate.
We should notice that in many contexts the emphasis is the other way around: the similarities between ourselves and other animals is under-emphasized. Consider, for example, the language used by the experimenters whose work I have just been discussing. The monkeys being tormented are called "stimulus animals" and when they cry out in pain they are said to have "vocalized." The result of this is an "increased noise level"—the language suggests that the animal pulling the chain is not even able to perceive the cries for what they are. When the experimenters speculate as to the significance of the whole phenomenon, they conjecture that "aversion to the perceived pain of conspecifics may be an ethologically innate pattern that serves to preserve the species." Here the animals are pictured as something less than they are. Perhaps we run the risk of going too far in the opposite direction if we speculate that compassion is a quality that "serves to preserve the species"—but I see no reason to think so.
Now the point about "moral freedom" was that beings with moral capacities should be free to exercise those capacities. But why should the point be made by referring exclusively to beings that have intellectual-ized conceptions of right and wrong, rather than by referring simply to beings that have moral virtues? If it is objectionable to deny a man the right to act according to his moral judgments, it is also objectionable, for much the same reason, to place a compassionate animal in a position where he must either cause pain or starve.
The sum of all this is that, whatever rationale is provided for granting humans a right to liberty, it seems that a relevantly similar one is available in the case of at least some other species of animals. The right to liberty, then, is not a "human" right.
As I said at the outset, my motive in arguing the point is to cast doubt on the importance of the concept of human rights. It is not that I think there are no human rights. On the contrary, I think that there are. But they are not rights that we have simply in virtue of being members of a certain species. Rather, they are rights that we have in virtue of possessing other characteristics, which members of other species happen not to have: for example, the right to worship seems to be a distinctively human right, because only humans, among all the animals we know, have any interest in or capacity for worship. But once the reason for this is understood, and once it is seen that such important rights as the right to liberty are not distinctively human, then most of the interest of the notion of "human" rights is, I think, gone. It would be much better to talk about natural rights, or simply rights, and remain alert to the fact that we humans are not the only beings that have them.
Why Some Philosophers Say That It Is Not Even Logically Possible for Animals to Have Rights
In his paper "Rights," H. J. McCloskey says that "important conclusions follow from the question as to whether animals have rights. If they do, ... it would seem an illegitimate invasion of animal rights to kill and eat them, if, as seems to be the case, we can sustain ourselves without killing animals. If animals have rights, the case for vegetarianism is prima facie very strong, and is comparable with the case against cannibalism." However, McCloskey thinks that this unsettling conclusion can be avoided, because it can be shown that nonhuman animals are not even logically possible bearers of rights. He argues:
A right cannot not be possessed by someone; hence, only beings which can possess things can possess rights. My right to life is mine, I possess it. It is as much mine as any of my possessions—indeed more so—for I possess them by virtue of my rights. . . . All these considerations seem to exclude the lower animals in a decisive way, and the higher animals in a less decisive but still fairly conclusive way as possible bearers of rights. (Consider "possess" in its literal use. Can a horse possess anything, e.g. its stable, its rug, in a literal sense of "possess"?)
But clearly this argument does not prove the point. Why can't animals possess things? When a bird gathers twigs and builds himself a nest, isn't it his, and not mine or yours or any other bird's? Would we have the right to take the nest from him, to satisfy some trivial interest of ours, and leave him to build himself another? I think not. The bird's claim to the nest is not merely that he needs it; it is his because he made it by his own labor. (Remember our earlier discussion of the right to property—i.e., Locke's example of the man gathering nuts.) If a larger, stronger bird drives him out and takes over the nest by force, we can recognize an injustice here even though neither animal has or could have an intellectualized conception of justice. All of this presupposes that the bird possesses the nest as a matter of right, once the nest is built; and although saying this does clash with our usual amoral way of regarding animals, I find nothing logically odd about it.
Like McCloskey, D. G. Ritchie believed that if we recognize animal rights we will have to make drastic changes in our ways of treating them. But he also thought that such changes are not required because animals don't have rights. In his book Natural Rights, Ritchie argued that absurd consequences follow from the assumption that they do: for example, if animals have rights, then cats who eat mice violate their rights—but this is absurd.
Now it does sound odd to speak of cats violating mice's rights, but this may be at least partly because we hardly ever think of the matter in this light, and this in turn may be due to the fact that we don't give much thought to the morality of how animals are treated. Some have thought that there is a deeper, conceptual reason for the oddity, but that, once this reason is understood, it provides no grounds for doubting that animals have rights. Plamenatz, for example, suggests that rights be understood as rights against rational beings, so that animals have rights against us but not against one another. Thus, even though the cat cannot violate the mouse's rights, we humans can do so. However, any such explanation as this, which absolutely excludes the possibility of any animal ever violating another's rights, seems questionable to me because I can think of at least some cases where it doesn't seem at all odd to speak of this happening. Consider again the case of the big bird driving the smaller bird out of the nest and taking it for his own use. Here I think that the smaller bird had a right to the nest—it is his, because he built it himself—and that it is not odd to say that the big bird violated his right to it.
There is, however, a somewhat deeper reason for doubting whether animals can have rights. It is plausible to think that moral requirements can exist only where certain conditions of reciprocity are satisfied. The basic idea here is that a person is obligated to respect the interests of others, and acknowledge that they have claims against him, only if the others are willing to respect his interests and acknowledge his claims. This may be thought of as a matter of fairness: if we are to accept inconvenient restrictions on our conduct, in the interests of benefiting or at least not harming others, then it is only fair that the others should accept similar restrictions on their conduct for the sake of our interests.
The requirement of reciprocity is central to contract theories of ethics. Hobbes, for example, conceived of moral rules as rules which rational, self-interested people will agree to obey on condition that others will obey them as well. Each person can be motivated to accept such an arrangement by considering the benefits he will gain if others abide by the rules; and his own compliance with the rules is the fair price he pays to secure the compliance of others. That is the point of the "contract." It is a natural part of such theories that nonhuman animals are not covered by the same moral rules which govern the treatment of humans, for the animals cannot participate in the mutual agreement on which the whole set-up depends.
This implication is made explicit in the most outstanding recent contribution to contract theory, John Rawls's A Theory of Justice. Rawls identifies the principles of justice as those which would be accepted by rational, self-interested people in what he calls "the original position"; that is, a position of ignorance with respect to particular facts about oneself and one's own position in society. The question then arises as to what sorts of beings are owed the guarantees of justice and Rawls's answer is:
We use the characterization of the persons in the original position to single out the kinds of beings to whom the principles chosen apply. After all, the parties are thought of as adopting these criteria to regulate their common institutions and their conduct toward one another; and the description of their nature enters into the reasoning by which these principles are selected. Thus equal justice is owed to those who have the capacity to take part in and to act in accordance with the public understanding of the initial situation.
This, he says, explains why nonhuman animals do not have the "equal basic rights" possessed by humans; "they have some protection certainly but their status is not that of human beings." And of course this result is not surprising: for if rights are determined by agreements of mutual interest, and animals are not able to participate in the agreements, then how can their interests give rise to rights?
The requirement of reciprocity may seem plausible, and I think that it does contain the germ of a plausible idea—I will say more about this in a moment—but nevertheless there are good reasons to reject it. We need to distinguish the conditions necessary for having a moral obligation from the conditions necessary for being the beneficiary of a moral obligation.
For example: normal adult humans have the moral obligation not to torture one another. What characteristics make it possible for a person to have this obligation? For one thing, he must be able to understand what torture is, and he must be capable of recognizing that it is wrong. When someone (a severely retarded person, perhaps) lacks such capacities, we do not think he has such obligations and we do not hold him responsible for what he does. On the other hand, it is a very different question what characteristics qualify someone to be the beneficiary of this obligation. It is wrong to torture a man—he is the beneficiary of our obligation not to torture—not because of his capacity for understanding what torture is, or for recognizing that it is morally wrong, but simply because of his capacity for experiencing pain. Thus a person may lack the characteristics necessary for having a certain obligation, and yet may still possess the characteristics necessary to qualify him as the beneficiary of that obligation. A severely retarded person may not be able to understand what torture is, or see it as wrong, and yet still be able to suffer pain. So we who are not retarded have an obligation not to torture him, even though he cannot have a similar obligation not to torture us.
The requirement of reciprocity says that a person is morally required to accept restrictions on his conduct, in the interests of not harming others, only if the others reciprocate. The example of the retarded person shows this to be false. He is not capable of restricting his conduct in this way; nevertheless we have an obligation to restrict ours. We are in the same position with respect to nonhuman animals: like the retarded person, they lack characteristics necessary for having obligations; but they may still be proper beneficiaries of our obligations. The fact that they cannot reciprocate, then, does not affect our basic obligations to them.
I said that the requirement of reciprocity, although unacceptable, does contain the germ of a plausible idea. What I have in mind is the idea that if a person is capable of acting considerately of our interests, and refuses to do so, then we are released from any similar obligations we might have had to him. This may very well be right. But whether or not this point is accepted makes no difference to our duties to nonhuman animals, since they lack the capacity to "refuse" to recognize obligations to us, just as they are not able to accept such obligations.
These arguments are only concerned with whether it is possible for animals to have rights. My reasons for thinking that they do have rights, and what some of those rights are, are given in the main body of the paper.
 Richard Wasserstrom, "Rights, Human Rights, and Racial Discrimination," The Journal of Philosophy, vol. 61 (1964), p. 631.
 Joel Feinberg defines human rights as those belonging to all humans, but specifically denies that they must be possessed only by humans, "so that a human right held by animals is not excluded by definition." (Social Philosophy, Englewood Cliffs, N.J.: Prentice-Hall, 1973, p. 85.) I prefer Wasserstrom’s definition because, if a right is shared by dogs and cows, there seems little reason to call it human rather than canine or bovine. And, by calling such rights "human," we are directing attention away from the fact that other beings have them, as though all that matters is whether we men have them.
 This point is brought out very powerfully by Peter Singer in "All Animals Are Equal," Philosophic Exchange, vol. 1, no. 5 (Summer 1974), pp. 103-16.
 John Locke, The Second Treatise of Government (1690), chap. 5, par. 27. The next quotation is from the same chapter of the same work, par. 28.
 DJ. R. Lucas, The Principles of Politics (Oxford: Oxford University Press, 1966), p. 144.
 For an account of this type, see Gregory Vlastos, "Justice and Equality," Social Justice, ed. Richard B. Brandt (Englewood Cliffs, N.J.: Prentice-Hall, 1962), p. 51.
 "The Shame of the Naked Cage," Life, November 8, 1968, p. 77.
 Floyd L. Ruch and Philip G. Zimbado, Psychology and Life, 8th ed. (Glenview, 111.: Scott, Foresman and Company, 1967), p. 539.
 Stephen J. Suomi and Harry F. Harlow, "Depressive Behavior in Young Monkeys Subjected to Vertical Chamber Confinement," Journal of Comparative and Physiological Psychology, vol. 80 (1972), pp. 11—18. The quotations that follow are from pp. 11, 12, 13, and 14. For an account of this and related experiments, see Peter Singer, Animal Liberation (New York Review, 1975), chap. II.
 Singer, "All Animals are Equal" op. cit., p. 108.
 "Henry Salt, The Creed of Kinship (New York: Dutton, 1935), pp. 60-61.
 See Edith Watson-Schipper, "Two Concepts of Human Freedom," The Southern Journal of Philosophy, vol. 11 (1973), pp. 309-15. Professor Watson-Schipper provides references to other thinkers who have held this distinction to be important.
 Stanley Wechkin, Jules H. Masserman, and William Terris, Jr., "Shock to a Conspecific as an Aversive Stimulus," Psychonomic Science, vol. 1 (1964), pp. 47—48; " 'Altruistic' Behavior in Rhesus Monkeys," The American Journal of Psychiatry, vol. 121 (1964), pp. 584-85. Because each article is less than two pages long, and both concern the same series of experiments, I will not give page numbers for the quotations that follow.
 Wechkin et al. do not use the term compassion in describing the monkeys' behavior; they speak of 'altruism'. However, they always put the word in single quotes, apparently to indicate reservations about using it. I think that compassion is a slightly more accurate term—although it does not really matter much which word is used—but I see no reason for the single quotes.
 Mrs. Foot has argued for a conception of morality according to which all moral behavior is governed by hypothetical imperatives; that is, imperatives which tell us what is to be done if we have certain desires. The desires which are important for morality are those associated with the moral virtues, such as the desire that others not suffer, associated with the virtue of compassion. This conception of morality is in contrast with others which place importance on acting from a sense of duty. (See Philippa Foot, "Morality as a System of Hypothetical Imperatives," The Philosophical Review, vol. 81 (1972), pp. 305-16.) If she is right about this, then perhaps even animals with no very great powers of abstract thought can be full moral agents. For, while having a "sense of duty" does require fairly Sophisticated thinking, having desires such as the desire that others not suffer, and recognizing simple ways of acting on those desires, does not require any such sophisticated thinking.
 H. J. McCloskey, "Rights," Philosophical Quarterly, vol. 15 (1965), p. 122. The next quotation is from p. 126.
 I say "himself" rather than "itself" deliberately, although either choice may be thought prejudicial: "Words and names are not without their effect upon conduct; and to apply to intelligent beings such terms as brute, beast, live-stock, dumb, etc., or the neuter pronouns it and which, as if they had no sex, is a practical incitement to ill-usage, and certainly a proof of misunderstanding. For example, the Morning Post (September 26th, 1933) thus described a case of cruelty to a cow: "He (the culprit) struck the cow with a milking-stool. It fell to the ground and died." It! One's thoughts turn to the milking-stool, but the allusion was to the cow! Salt, The Creed of Kinship, p. 62.
 D. G. Ritchie, Natural Rights (London: Allen & Unwin, 1894), p. 109.
 John Plamenatz, Consent, Freedom, and Political Obligation, 2nd ed. (Oxford: Oxford University Press, 1968), p. 83.
 Thomas Hobbes, Leviathan (1651), chaps. 13-16.
 John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971). The quotations that follow are from p. 505 of that work.
 A number of people have contributed in one way or another to my understanding of the topics discussed in this paper: E. M. Adams, Edward Erwin, David Marans, Howard Pospesel, Jack Glickman, Eduardo de Marchena, and Melanie Rabin. My greatest debt is to Peter Singer, who has influenced my whole way of thinking about the morality of how we treat nonhuman animals.