February 3, 2006
In 55 BC, the Roman leader Pompey staged a combat between humans and elephants. Surrounded in the arena, the animals perceived that they had no hope of escape. According to Pliny, they then "entreated the crowd, trying to win its compassion with indescribable gestures, bewailing their plight with a sort of lamentation." The audience, moved to pity and anger by their plight, rose to curse Pompey — feeling, wrote Cicero, that the elephants had a relation of commonality (societas) with the human race.
In 2000 AD, the High Court of Kerala, in India, addressed the plight of circus animals "housed in cramped cages, subjected to fear, hunger, pain, not to mention the undignified way of life they have to live." It found those animals "beings entitled to dignified existence" within the meaning of Article 21 of the Indian Constitution, which protects the right to life with dignity. "If humans are entitled to fundamental rights, why not animals?" the court asked.
We humans share a world and its scarce resources with other intelligent creatures. As the court said, those creatures are capable of dignified existence. It is difficult to know precisely what that means, but it is rather clear what it does not mean: the conditions of the circus animals beaten and housed in filthy cramped cages, the even more horrific conditions endured by chickens, calves, and pigs raised for food in factory farming, and many other comparable conditions of deprivation, suffering, and indignity. The fact that humans act in ways that deny other animals a dignified existence appears to be an issue of justice, and an urgent one.
Indeed, there is no obvious reason why notions of basic justice, entitlement, and law cannot be extended across the species barrier, as the Indian court boldly did.
In some ways, our imaginative sympathy with the suffering of nonhuman animals must be our guide as we try to define a just relation between humans and animals. Sympathy, however, is malleable. It can all too easily be corrupted by our interest in protecting the comforts of a way of life that includes the use of other animals as objects for our own gain and pleasure. That is why we typically need philosophy and its theories of justice. Theories help us to get the best out of our own ethical intuitions, preventing self-serving distortions of our thought. They also help us extend our ethical commitments to new, less familiar cases. It seems plausible to think that we will not approach the question of justice for nonhuman animals well if we do not ask, first, what theory or theories might give us the best guidance.
In my new book, Frontiers of Justice: Disability, Nationality, Species Membership, I consider three urgent problems of justice involving large asymmetries of power: justice for people with disabilities, justice across national boundaries, and justice for nonhuman animals. During the past 35 years, theories of justice have been elaborated and refined with great subtlety and insight, stimulated by John Rawls's great books, which built, in turn, on the classical doctrine of the social contract in Locke, Kant, and Rousseau. The social-contract tradition has enormous strength in thinking about justice. Devised in the first instance to help us reflect on the irrelevance of class, inherited wealth, and religion to just social arrangements, its theories have been successfully extended, in recent years, to deal with inequalities based on race and gender. The three issues that are my theme, however, have not been successfully addressed by such theories, for reasons inherent in their very structure — or so I argue.
In each case, a "capabilities approach" I have developed provides theoretical guidance. It begins from the question, "What are people actually able to do and to be?" It holds that each person is entitled to a decent level of opportunity in 10 areas of particular centrality, such as life, health, bodily integrity, affiliation, and practical reason.
On the question of animal entitlements, the approach gives better results than existing Kantian theories — which hold that respect should be given to rational beings — or Utilitarian approaches — which hold that the best choice is to maximize the pleasure or satisfaction of preferences. A capabilities approach can recognize a wide range of types of animal dignity, and of what animals need in order to flourish, restoring to Western debate some of the complexity the issue had in the time of Cicero, which it has subsequently lost.
As Richard Sorabji argues in his excellent book Animal Minds and Human Morals: The Origins of the Western Debate (Cornell University Press, 1993), the ancient Greek and Roman world contained a wide range of views that held promise for thinking about the moral status of animals. However, Stoicism, with its emphasis on the capacity of humans for virtue and ethical choice, exercised far more widespread influence than any other philosophical school in a world of war and uncertainty — but it had a very unappealing view of animals, denying them all capacity for intelligent reaction to the world, and denying, in consequence, that we could have any moral duties to them. Because of the attractiveness of Stoicism's view of human virtue and choice, that picture of animals became widespread. I think we need to add to Sorabji's account the fact that Stoic views of animals fit better than others with the Judeo-Christian idea that human beings have been given dominion over animals. Although that idea has been interpreted in a variety of ways, it has standardly been understood to give humans license to do whatever they like to nonhuman species and to use them for human purposes.
Kant argues that all duties to animals are merely indirect duties to human beings: Cruel or kind treatment of animals strengthens tendencies to behave in similar fashion to our fellow humans. So animals matter only because of us. Kant cannot imagine that beings who (as he believes) lack self-consciousness and the capacity for ethical choice can possibly have dignity, or be objects of direct ethical duties. The fact that all Kantian views ground moral concern in our rational and moral capacities makes it difficult to treat animals as beings to whom justice is due.
Classical Utilitarianism has no such problem. It begins, admirably, with a focus on suffering. Its great theoretical pioneers, Jeremy Bentham and John Stuart Mill, had intense concern for the well-being of animals. Bentham famously argued that the species to which a creature belongs is as irrelevant, for ethical purposes, as race: It does not supply a valid reason to deprive a sentient being of a decent life. If, as Utilitarianism holds, the best choice is the one that maximizes total (or, in some versions, average) utility, understood as pleasure and/or the absence of pain, good choice would lead to radical change in our treatment of animals. Peter Singer's courageous work on animal suffering today follows the Utilitarian paradigm. Singer argues that the right question to ask, when we think about our conduct toward animals, is, What choice will maximize the satisfaction of the preferences of all sentient beings? That calculation, he believes, would put most of our current pain-inflicting use of animals off limits.
Nevertheless, valuable though Utilitarian work on animal suffering has been, it has some serious difficulties. One notorious problem concerns the Utilitarian commitment to aggregation: that is, to summing together all pleasures and pains. The choice maker is instructed to produce the largest total (or average) pleasure. That can allow results in which a small number of creatures have very miserable lives, so long as their miseries are compensated for by a great deal of pleasure elsewhere. Even slavery is ruled out — if it is — only by fragile empirical calculations urging its ultimate inefficiency. It remains unclear whether such a view can really rule out the cruel treatment of at least some animals, which undoubtedly causes great pleasure to a very large number of meat eaters, or the infliction of pain on small numbers of animals in laboratory testing in order to provide benefits for many humans. (Here Kantian views about humans offer a good corrective, insisting that even the well-being of society as a whole does not justify egregious harms and indignities to any individual.)
Another sort of aggregation also causes difficulty: Utilitarians consider together diverse aspects of lives, reducing them all to experienced pain and pleasure. But we might think that a good life, for an animal as for a human, has many different aspects: movement, affection, health, community, dignity, bodily integrity, as well as the avoidance of pain. Some valuable aspects of animal lives might not even lead to pain when withheld. Animals, like humans, often don't miss what they don't know, and it is hard to believe that animals cramped in small cages all their lives can dream of the free movement that is denied them. Nonetheless, it remains valuable as a part of their flourishing, and not just because its absence is fraught with pain. Even a comfortable immobility would be wrong for a horse, an elephant, or a gorilla. Those creatures characteristically live a life full of movement, space, and complex social interaction. To deprive them of those things is to give them a distorted and impoverished existence.
Finally, all Utilitarian views are highly vulnerable on the question of numbers. The meat industry brings countless animals into the world who would never have existed otherwise. For Utilitarians, that is not a bad thing. Indeed, we can expect new births to add to the total of social utility, from which we could then subtract the pain the animals suffer. Wherever that calculation might come out, such a view would countenance the production of large numbers of creatures with lives only marginally worth living. So Utilitarianism has great merits, but also significant problems.
My capabilities approach, as so far developed, starts from the notion of human dignity and a life worthy of it. But it can be extended to provide a more adequate basis for animal entitlements than the other two theories under consideration. It seems wrong to think that only human life has dignity. As the Indian court said, the idea of a life commensurate with a creature's dignity has clear implications for assessing the lives we all too often make animals live.
The basic moral intuition behind my approach concerns the dignity of a form of life that possesses both deep needs and abilities. Its basic goal is to take into account the rich plurality of activities that sentient beings need — all those that are required for a life with dignity. With Aristotle and the young Marx, I argue that it is a waste and a tragedy when a living creature has an innate capability for some functions that are evaluated as important and good, but never gets the opportunity to perform those functions. Failures to educate women, failures to promote adequate health care, failures to extend the freedoms of speech and conscience to all citizens — all those are treated as causing a kind of premature death, the death of a form of flourishing that has been judged to be essential for a life with dignity. Political principles concerning basic entitlements are to be framed with those ideas in view.
The species standard is evaluative. It does not simply read off norms from the way nature actually is. Once we have judged, however, that a central human power is one of the good ones, one of the ones whose flourishing is essential for the creature to have a life with dignity, we have a very strong moral reason for promoting it and removing obstacles to its development.
The same attitude to natural powers that guides the approach in the case of human beings guides it in the case of nonhuman animals: Each form of life is worthy of respect, and it is a problem of justice when a creature does not have the opportunity to unfold its (valuable) power, to flourish in its own way, and to lead a life with dignity. The fact that so many animals never get to move around, enjoy the air, exchange affection with other members of their kind — all that is a waste and a tragedy, and it is not a life in keeping with the dignity of such creatures.
So the capabilities approach is well placed, intuitively, to go beyond both Kantian and Utilitarian views. It goes beyond Kant in seeing our ethical duties to animals as direct, not indirect, and also in its starting point, a basic concern for sentient life, not just rational life (though there is surely far more rationality in animal lives than Kant would have acknowledged). It goes beyond the intuitive starting point of Utilitarianism because it takes an interest not just in pleasure and pain, but in complex forms of life. It wants to see each living thing flourish as the sort of thing it is, and wants political principles to protect, for all sentient beings, a set of basic opportunities for flourishing.
Does justice focus on the individual, or on the species? It seems that here, as in the human case, the focus should be the individual creature. The capabilities approach attaches no ethical importance to increased numbers as such; its focus is on the well-being of existing creatures, and the harm that is done to them when their powers are blighted. Consequently the survival of a species may have weight as a scientific or aesthetic issue, but it is not an ethical issue, and certainly not an issue of justice — apart from the harms to existing creatures that are usually involved in the extinction of a species. When elephants are deprived of a congenial habitat and hunted for their tusks, harm is done to individual creatures, and it is that harm that should be our primary focus when justice is our concern, even while we may for other reasons seek the preservation of elephant species.
Almost all ethical views of animal entitlements hold that there are morally relevant distinctions among forms of life. Killing a mouse seems to be different from killing a chimpanzee. But what sort of difference is relevant for basic justice? Singer, following Bentham, puts the issue in terms of sentience. Animals of many kinds can suffer bodily pain, and it is always bad to cause pain to a sentient being. If there are animals that do not feel pain — and it appears that crustaceans and mollusks, as well as sponges and the other creatures Aristotle called "stationary animals," fall in that category — there is either no harm or only a trivial harm done in killing them. Among the sentient creatures, moreover, some can suffer additional harms through their cognitive capacity: A few animals can foresee and mind their own death, and others will have conscious interest in continuing to live. The painless killing of an animal that does not foresee its own death or take a conscious interest in the continuation of its life is, for Singer and Bentham, not bad, for all badness consists in the frustration of interests, understood as forms of conscious awareness. Singer is not, then, saying that some animals are inherently more worthy of esteem than others; he is simply saying that, if we agree with him that all harms reside in sentience, the creature's form of life limits the conditions under which it can actually suffer harm.
Similarly, James Rachels, whose view does not focus on sentience alone, holds that the level of complexity of a creature affects what can be a harm for it. What is relevant to the harm of pain is sentience; what is relevant to the harm of a specific type of pain is a specific type of sentience (for example, the ability to imagine one's own death). What is relevant to the harm of diminishing freedom, Rachels goes on, is a being's capacity for freedom or autonomy. It would make no sense to complain that a worm is being deprived of autonomy, or a rabbit of the right to vote. My capabilities view follows Rachels, denying that there is a natural ranking of forms of life, but holding that the level of complexity of a creature affects what can be considered to be a harm to it.
Like Bentham, however, I do think of sentience as a minimum necessary condition for moral status. Does species membership matter when we consider the form of life that is good for a creature? For Utilitarians, and for Rachels, the species to which a creature belongs has no moral relevance. What matters are the capacities of the individual creature: in Rachels's words, "moral individualism." Utilitarian writers are fond of comparing apes to young children and to mentally disabled humans, suggesting that the ethical questions we should consider are the same in all those cases. The capabilities approach, by contrast, with its talk of characteristic functioning and forms of life, seems to attach some significance to species membership as such.
What type of significance is that? There is much to be learned from reflection on the continuum of life. Capacities do crisscross and overlap: A chimpanzee may have more capacity for empathy and perspectival thinking than a very young child, or than an older child with autism. And capacities that humans sometimes arrogantly claim for themselves alone are found very widely in nature. But it seems wrong to conclude from such facts that species membership is morally and politically irrelevant. A child with mental disabilities is actually very different from a chimpanzee, though in certain respects some of her capacities may be comparable. Such a child's life is difficult in a way that the life of a chimpanzee is not difficult: She is cut off from forms of flourishing that, but for the disability, she might have had. There is something blighted and disharmonious in her life, whereas the life of a chimpanzee may be perfectly flourishing. Her social and political functioning, her friendships, her ability to have a familyall may be threatened by her disabilities, in a way that the normal functioning of a chimpanzee in the community of chimpanzees is not threatened by its cognitive endowment.
That is relevant when we consider issues of basic justice. For children born with Down syndrome, it is crucial that the political culture in which they live make a big effort to extend to them the fullest benefits of citizenship they can attain, through health benefits, education, and re-education of public culture. That is so because they can flourish only as human beings. They have no option of flourishing as happy chimpanzees. For a chimpanzee, on the other hand, it seems to me that expensive efforts to teach language, while interesting and revealing for human scientists, are not matters of basic justice. A chimpanzee flourishes in its own way, communicating with its own community in a perfectly adequate manner that has gone on for ages.
In short, the species norm (duly evaluated) tells us what the appropriate benchmark is for judging whether a given creature has decent opportunities for flourishing.
There is a danger in any theory that alludes to the characteristic flourishing and form of life of a species: the danger of romanticizing "Nature," or seeing nature as a direct source of ethical norms. Nature is not particularly ethical or good. It should not be used as a direct source of norms. In the human case, therefore, my capabilities view does not attempt to extract norms directly from some facts about human nature. We must begin by evaluating the innate powers of human beings, asking which ones are central to the notion of a life with dignity. Thus not only evaluation but also ethical evaluation is put into the approach from the start. Many things that are found in human life, like the capacities for cruelty, despair, or self-destruction, are not on the capabilities list.
In the case of nonhuman animals, however, we need to remember that we are relatively ignorant of what a good life for each sort of animal is and strongly biased in favor of our own power interests. Thus our attempts to evaluate the capacities of animals, saying that some are good and others not so good, may easily go wrong. Moreover, while we can expect a potentially violent human (as all humans are) to learn to restrain her or his capacity for violence, we cannot expect so much learning and control from many animal species. Thus to deny a tiger the exercise of its predatory capacities may inflict significant suffering, whereas we require a human to learn to live at peace with others (or we should!).
Here the capabilities view may, however, distinguish two aspects of the capability in question. A tiger's capability to kill small animals, defined as such, does not have intrinsic ethical value, and political principles can omit it (and even inhibit it in some cases). But a tiger's capability to exercise its predatory nature so as to avoid the pain of frustration may well have value, if the pain of frustration is considerable. Zoos have learned how to make that distinction. Noticing that they were giving predatory animals insufficient exercise for their predatory capacities, they have had to face the question of the harm done to smaller animals by allowing such capabilities to be exercised. Should they give a tiger a tender gazelle to crunch on? The Bronx Zoo has found that it can give the tiger a large ball on a rope, whose resistance and weight symbolize the gazelle. The tiger seems satisfied. Wherever predatory animals are living under direct human support and control, such solutions seem the most ethically sound.
Much more remains to be done to ground this approach philosophically and to articulate its results, which I try to do in Frontiers. What, however, should the practical upshot be?
In general the capabilities approach suggests that it is appropriate for each nation to include in its constitution or other founding statement of principle a commitment to regarding nonhuman animals as subjects of political justice and to treating them in accordance with their dignity. The constitution might also spell out some of the very general principles suggested by the capabilities approach, and judicial interpretation can make the ideas more concrete. The High Court of Kerala made a good beginning, thinking about what the idea of "life with dignity" implies for the circus animals in the case. The rest of the work of protecting animal entitlements might be done by suitable legislation and by court cases demanding the enforcement of laws, where they are not enforced. At the same time, many of the issues covered by this approach cannot be dealt with by nations taken in isolation, but can be treated only through international cooperation. So we also need international accords committing the world community to the protection of animal habitats and the eradication of cruel practices.
It has been obvious for a long time that the pursuit of global justice requires the inclusion of many people and groups not previously included as fully equal subjects of justice: the poor; members of religious, ethnic, and racial minorities; and more recently women, the disabled, and inhabitants of poor nations distant from one's own. But a truly global justice requires not simply looking across the world for fellow species members who are entitled to a decent life.
It also requires looking around the world at the other sentient beings with whose lives our own are inextricably and complexly intertwined. Kant's approach does not confront these questions as questions of justice. Probably a strict Kantian could not so confront them, not without considerably modifying Kant's own view about rationality as the basis of moral respect. Utilitarian approaches boldly confront the wrongs animals suffer, and they deserve high praise. But in the end, I have argued, Utilitarianism is too homogenizing — both across lives and with respect to the heterogeneous constituents of each life — to provide us with a fully adequate theory of animal justice. The capabilities approach, which begins from an ethically attuned concern for each form of animal life, offers a model that does justice to the complexity of animal lives and their strivings for flourishing. Such a model seems an important part of a fully global theory of justice.
Martha C. Nussbaum is a professor in the philosophy department, law school, divinity school, and the college at the University of Chicago. Her book Frontiers of Justice: Disability, Nationality, Species Membership has just been published by Harvard University Press.