A Journal of Philosophy, Applied to the Real World

The Claims of Animals and the Needs of Strangers: Two Cases of Imperfect Right

Harvard University


This paper argues for a conception of the natural rights of non-human animals grounded in Kant’s explanation of the foundation of human rights. The rights in question are rights that are in the first instance held against humanity collectively speaking—against our species conceived as an organized body capable of collective action. The argument proceeds by first developing a similar case for the right of every human individual who is in need of aid to get it, and then showing why the situation of animals is similar.

I first review some of the reasons why people are resistant to the idea that animals might have rights. I then explain Kant’s conception of natural rights. I challenge the idea that duties of aid and duties of kindness to animals fit the traditional category of “imperfect duties” and argue that they are instead cases of “imperfect right.” I explain how you can hold a right against a group, and why it is legitimate to conceive of humanity as such a group. I then argue that Kant’s account of the foundation of property rights is grounded in a conception of the common possession of the Earth that grounds a right to aid and the rights of animals to be treated in ways that are consistent with their good. Finally, I return to the objections to the idea that animals have rights and offer some responses to them.

KEYWORDS: animals, duties of aid, humanity, Immanuel Kant, perfect and imperfect duties, rights.



Some people use the term “animal rights” to refer simply to the moral claims of non-human animals, whatever those might be. In this paper I use the term “rights” in its more specific sense, to designate a particular kind of moral claim. Roughly speaking, a “right” is a claim that may be, and ought to be, legally enforced. The enforcement of rights can be claimed as a matter of law and justice: it is not merely a matter of charity.

Most people think that non-human animals (hereinafter “animals”) have what philosophers call “moral standing.” That is, they believe that animals are the appropriate objects of moral concern, and they think that we have moral duties to treat animals in certain ways that are good for them for the sake of the animals themselves, and not just, say, for the sake of their owners or of keeping them profitable. Admittedly, many people think that these duties are of a rather weak kind: they think we ought to treat animals “as humanely as possible” given the ways that we use them. For example, according to some animal welfare laws, animals used in scientific research in ways that might be painful to them must be given anesthetics if it does not interfere with the purpose of the research. The duty to spare the animal pain is not taken to be a reason against doing the research. But most people admit at least a duty to prevent animals from experiencing “unnecessary” pain.

Most people also think that at least some of our duties to animals should be enforced by legal sanctions. That is why there are animal welfare protection laws on the books. Yet, for reasons I will describe below, many people think it is absurd to suppose that animals either do or should have rights, in the specific sense of rights I am concerned with here. This combination of views may look contradictory on the surface, but in fact, animal welfare laws do not usually grant the animals themselves any rights, at least if we think of a right as something that the right-holder can claim. For example, at least in the United States as things stand now, the legal representatives of animals cannot sue for the enforcement of animal welfare laws in the name of the animals themselves. Only human beings who can claim to have some interest of their own in the enforcement of animal welfare laws can sue to have them enforced (Sunstein 2004).

In this paper I argue that animals do have rights, but that these rights have a distinctive structure. I say that animals “do have rights” rather than merely that they “should have rights,” because the argument I will give is based on Immanuel Kant’s conception of rights, and is therefore in the natural rights tradition, according to which rights are grounded in morality, and can in a sense exist prior to, or independently of, any positive laws that are actually on the books. I will discuss the idea of a natural right further in section three. The distinctive structure of animals’ rights is this: Many of our rights are held against individuals, at least in the first instance: either against every individual in a group, as one’s civil rights are, or against some specific individuals, as when two people are bound by a contract or a promise. Animals, I argue, have rights that are in the first instance rights against humanity collectively speaking, humanity as a group, to be treated by people in ways that are consistent with what is good for them. The corresponding duties of individuals are derived from the duties we have collectively. I believe that as individuals we also have duties to treat animals well, and in particular to avoid cruelty. But I think that the duties of humanity collectively speaking are in a distinctive way the ground of rights that ought to have legal force.

I will try to demonstrate that there are such rights—rights against humanity collectively—by arguing for another important case of this kind of right. I believe that every human being who is in need of aid in order to survive in a reasonable condition has a right with exactly this structure—a right against humanity, collectively, to ensure the provision of that aid. Both that particular human right and animal rights are traceable to the same fact, namely the fact that humanity, collectively speaking, is in a position to exercise extensive control over the fate of all of the inhabitants of the planet with whom we interact.


In this section, I sketch some of the reasons why people are resistant to the idea that animals either do or should have rights. After I have offered my account of why animals have rights I will return to these objections and discuss how we might respond to them.

2.1. Animals are property, and property cannot have rights.

Following the tradition of Roman law, legal systems generally divide the world into persons and property, treating human beings as persons, and pretty much everything else, including non-human animals, as property. Persons are the subjects of both rights and obligations, including the right to own property, while objects of property, being by their very nature for the use of persons, have no rights at all. Animals, of course, are traditionally classified as property—domestic animals as the property of individuals or organizations, and wild animals as the property of the state. For obvious reasons, there can seem to be a kind of incoherence in granting rights to property, especially rights against the owners of the property. For this reason, some animal rights advocates have proposed that animals, or that animals with a certain degree of cognitive sophistication—primates and cetaceans perhaps—should be reclassified as legal persons.

To anticipate, the view I advocate in this paper does not call for this kind of reclassification. I think we should reject the legal bifurcation instead, and acknowledge the existence of a third legally relevant category, whose members can have some kinds of rights. Though I do not have space to discuss the issue in depth here, I want to mention one reason why I think this matters. Although people use the term “person” in different ways, I think that we can identify one fairly widely accepted concept of a person that is morally and legally important, and that non-human animals do not fit that concept. Speaking roughly, the concept of a person is the concept of a normatively self-governing being. By that I mean a being who is able to govern her choice of ends and actions by applying standards of rightness and wrongness, and/or of goodness and badness. Only human beings and certain organizations, such as political states, fit that concept.1 As the ideas I will discuss in connection with the third objection suggest, many of many of our most important rights—those associated with autonomy—spring from the fact that we are persons in this sense. Animals do not share in these rights, so reclassifying animals as persons blurs a distinction that we have good moral and legal reasons to make.

2.2. Rights require a kind of reciprocity of which animals are incapable.

Many people think of the realm of rights as being in some distinctive way a realm of reciprocity. Rights are something we accord reciprocally to each other. To say you have a right implies that you also have obligations, in particular obligations to respect the rights of others.

It can be hard to pin this thought down in exactly the right way. For instance, we might be tempted to think of it as something like a bargain based on mutual self-interest—I will keep my hands off of your property or your liberties if you will keep your hands off of mine. David Hume argued that animals do not have rights with something like this conception of reciprocity in mind. Hume makes the argument in order to prove that the duties of justice are grounded in considerations of self-interest and utility. We expect people to conform to the duties of justice only under certain conditions, Hume argues, and those conditions are exactly the ones in which conforming to the duties of justice is useful to all concerned. One of these conditions is an approximate equality of power between the parties to the agreement, which renders it in the interest of all parties to make and maintain the agreement. If you had enough power to completely control someone else, it would not be in your interest to make any concessions to him, and Hume thinks you therefore would not owe him anything. On these grounds, Hume argues that we do not have duties of justice to the other animals:

Were there a species of creatures intermingled with men, which, though rational, were possessed of such inferior strength, both of body and mind, that they were incapable of all resistance, and could never, upon the highest provocation, make us feel the effects of their resentment; the necessary consequence, I think, is that we should be bound by the laws of humanity to give gentle usage to these creatures, but should not, properly speaking, lie under any restraint of justice with regard to them, nor could they possess any right or property, exclusive of such arbitrary lords. Our intercourse with them could not be called society, which supposes a degree of equality; but absolute command on the one side, and servile obedience on the other. Whatever we covet, they must instantly resign: Our permission is the only tenure, by which they hold their possessions: Our compassion and kindness the only check, by which they curb our lawless will: And as no inconvenience ever results from the exercise of a power, so firmly established in nature, the restraints of justice and property, being totally useless, would never have place in so unequal a confederacy.

This is plainly the situation of men, with regard to animals; and how far these may be said to possess reason, I leave it to others to determine (Hume 1975, pp. 190-91).

But bargains based on self-interest are subject to the free-rider problem: you lose your reason to comply with them when it is in your interest to renege. So this kind of reciprocity cannot ground any strong commitment to upholding the rights of others.

Later I will propose a conception of reciprocity that I think is more to the purpose. But however exactly we understand the nature of the reciprocity involved, we are inclined to view it as essential to rights. Having rights goes along with having responsibilities or obligations; you cannot have one without the other. But non-human animals are not moral beings, and therefore we might suppose they cannot have moral or legal obligations. So we might conclude that they cannot have rights.

2.3 The function of rights is to protect autonomy, and animals are not autonomous.

A third and related problem, for some thinkers, is that they believe that the special point of rights, as opposed to other kinds of moral claims, is to secure the liberty and autonomy of those who hold the rights. This is particularly true of thinkers in the natural rights tradition, and of those who uphold the “will theory” rather than the “interest theory” of rights. The interest theory of rights holds that rights exist to protect an individual’s important interests; while the will theory holds that rights define a sphere over which an individual has normative control. That sphere consists of those actions she is free to do if she chooses, and the objects of property she may use as she likes. The control is “normative” because others count as wronging her if they interfere with her actions or use her property without her permission. Immanuel Kant, who represents the natural rights tradition, thought of rights as coercively enforceable prescriptions that are essential to maintaining the (equal) liberty of everyone. The use of force or coercion against free rational beings, he argued, is wrong, except when you are using force to protect someone’s freedom—to hinder a hindrance to freedom, as he put it (Kant 1996, 6:231).2 So we may use force or coercion—that is, we may use the power of law—in order to protect equal liberty, but not to promote other ends. John Rawls, to take another example, thought of rights in a liberal society as aimed at securing to each citizen, as far as possible, the ability to pursue her own conception of the good—that is, her own conception of what is worth doing and caring about in a human life (Rawls 1971). According to such views, rights ensure that we are not bound by restrictions grounded merely in other people’s ideas about what is worth doing and caring about. We are bound only by restrictions that spring from the requirement that everyone’s liberty or autonomous pursuit of her own conception of the good should be equally protected.

But there is no point in trying to secure political liberty or autonomy in this sense to the other animals. They live according to their natures, not according to their values or their free choices or their personal conceptions of what is good. Nor is there much to be said for granting them what might seem the nearest analog to that kind of liberty—freedom of action in the simplest sense, allowing them to go where they please and do what they want. When animals, either wild or domestic, live within the confines of human society, it is essential both for their safety and ours that they live under a certain degree of human control. So on this showing rights do not seem to be the sort of thing we could intelligibly grant to animals. Rights protect a kind of liberty that the other animals do not and could not possibly have.

2.4. The idea that animals have rights has counterintuitive implications about the actions of our ancestors and those now living in the developing world.

There are also worries, perhaps of a more vague kind, about the implications of the idea that animals have rights for humanity’s past relations with animals, and, for similar reasons, for livestock farmers in the developing world now. It is natural to think that our duties of beneficence or kindness—duties that we do not usually think of as grounded in rights—are relative to our own resources and necessities. We cannot be required to give more than we can afford. That implies that these duties can change with changing conditions. When we think of our duties to animals as mere duties of kindness or humanity, we can say things that sound sensible about the actions of our ancestors. We no longer need to use real furs to keep from freezing, for example, because now we have indoor heating and artificial fabrics that work just as well. But our ancestors, especially in northern places, could hardly have refrained from wearing furs. We do not need to eat meat nowadays, when we can ship vegetables from wherever they are now growing to places where they are out of season, or keep them frozen until times of need. But this was not always true. We can now afford to be more humane than we once could; and we can make that change without any implied criticism of those who came before us. A similar point can be made about the impoverished owners of small holdings of livestock in the developing world. They cannot be expected just to give their livestock up. Kindness to animals is a luxury that they cannot afford.

But it is harder to think of a right as something that can be balanced against the costs to those who are supposed to respect it. Slave owners, most of us now think, were always wrong, regardless of cultural and economic conditions. No one has the right to treat another human being as property, and no one ever did. Our ancestors had to use furs, eat meat, hunt, and use animals for labor. That is how the human species evolved and came to live all over the planet—by using animals. So if animals have rights, were our ancestors simply wrong to do these things? If that seems counterintuitive, doesn’t that show that even if animals have claims on our kindness, those claims do not spring from anything like animal rights?

2.5. The idea that animals have rights has outrageous practical implications.

Most importantly, however, the idea of animal rights seems worrisome because of its practical implications. Are all animals going to be accorded a basic right to life, and, if so, are meat-eating and hunting going to be outlawed? Would experiments on animals have to be brought to an end, or limited to those that do not harm the animals? And what about those of our own activities that inevitably hurt or kill animals, even when the purpose of those activities is benign? We kill mice and voles when we drive tractors through the fields on which we grow our vegetables. We kill and injure birds when we put up wind turbines—or for that matter, when we put up windows and power lines. Whenever we use pesticides, we kill not only their targets but other animals as well. Should all such activities be illegal, or legally regulated in some way?

These are controversial questions, and I think that under any reasonable conception of animal rights, the answer to some of them will be “yes,” even if those implications do seem outrageous to many people now. But the questions I have just raised lead naturally to further questions that begin to make the idea of animal rights seem not just controversial but silly. If animals have a right to life, for instance, might people be prosecuted for killing vermin that are infesting their homes or their food supplies or for swatting mosquitoes that are trying to bite them? And if animals have rights to life, or rights not to have suffering inflicted upon them, ought we to interfere between predators and their prey, in the name of protecting the rights of prey animals to escape this form of suffering and death? And must we then find adequate meat substitutes for the predators? Or should we actually try to eliminate predator species altogether—if only we knew how—as Jeff McMahan suggested in a New York Times editorial a few years ago? (McMahan 2010).

There are obvious reasons, based on past experience, to worry about the human competence for tinkering with nature on this kind of scale. But even leaving those worries aside, many people find the idea of our managing nature in this way deeply distasteful, although the reasons are a little obscure. Some of us do not like the idea of living in a completely domesticated world. Some people feel that it is not part of humanity’s business to try moralize nature, or even that we do not have the right to. These worries give rise to some rather deep questions about what the human place in nature is, and how we should think about and respond to the amoral character of nature itself.

For all of these reasons, many people think that even if we have duties of kindness and humanity to animals, these duties do not arise from any rights on the part of the animals themselves, and should not be the grounds for giving animals legal rights.


Before I challenge that claim, I need to clarify something about the conception of rights I am working with here. So far, most of what I have said has been ambiguous between two claims: that animals should have rights and that they do have rights. There is a reason for this ambiguity, for there is a problem about what is going on when someone makes the case for a legal right.

Saying that I am arguing that animals “should have rights” has the disadvantage of making it sound as if all I am saying is that there is something to be said in favor of their having legal rights, some reasons that would support the policy of giving them rights. But that may not seem like the correct way to argue for a right, since a right ordinarily functions as a trump and a trump requires something stronger than some considerations in its favor. If I have a right to something, call it X, then you have no right to deprive me of X. My right is supposed to be a decisive consideration against your depriving me of X, however good your reasons for depriving me of X would be if I did not have the right. So to say that I have a right to X is not just to say that there is a very strong reason for me to have X: it is to say something about the relations in which I stand to those against whom I claim the right. However good others’ reasons are for depriving me of X, they will be wronging me if they do so. That includes my relations to society collectively speaking. But if my right is a trump even against society collectively speaking, how can society collectively speaking be in a position to grant me the right? When someone claims that she has a right, she is claiming precisely that no one is in a position to deny her that to which she has a right. But if no one is in a position to deny her the right, then it seems as if no one is in a position to grant her that to which she has the right either. What she is saying is precisely that this is not the sort of thing that others may withhold or to grant, however good their reasons. Consider, for instance, the idea that a nation might give its slaves a right to their freedom. Is another human being’s right to her freedom something that it is ours (all of us? the rest of us?) to give? How can society give someone his freedom, if it was already his own by right?

Some philosophers propose to deal with this problem by invoking the idea of a “moral right” and saying that moral rights are the grounds on which we should establish legal rights. That enables them to split the difference—the moral rights do already exist in the nature of things, even if the legal ones do not. Then we can say that what society does when it enacts laws protecting people’s rights is not granting them rights they did not already have, but protecting their moral rights by making them legal and so coercively enforceable.

That can sound sensible until we remind ourselves what exactly a right is. A right, at least according to Kant and others in the natural rights tradition, is—by definition—a claim that may legitimately be coercively enforced. You have a right when you have a claim on others to act in a certain way and it is morally legitimate for you (or for society on your behalf) to defend yourself with the use of force against violations of that claim. Not all moral claims, we believe, may be coercively enforced. I cannot sue you for hurting my feelings or being rude to me or have you thrown into prison for breaking my heart, though you should not do these things. I cannot have you arrested if you fail to open a door for me when my arms are full of packages or to help me change a tire by the side of the road. How do we draw the distinction? Some philosophers would argue that the distinction should be drawn on pragmatic or consequentialist grounds: on whether the costs of coercive enforcement are worth preventing wrongs of this kind. Kant, however, believed that the distinction is based on principle. We have already seen what the principle in question is: we may coercively enforce a claim only when we are “hindering a hindrance” to freedom. According to Kant, I am free when I can pursue my own ends and in doing so I am not subject to the wills of other people, that is, they may not legitimately just decide to interfere with my actions or prevent me from pursuing my ends. I am not made subject to your will when you try to break my heart, for I am perfectly free not to care. I am not made subject to your will when you fail to open a door for me, for that does not stop me from going through the door. But I am made subject to your will when you enslave me or make use of my person or my property without my consent. So my claims against these actions are coercively enforceable—that is, they are rights.

This account of what makes a moral claim one of right makes trouble for the proposed use of the distinction between moral and legal rights. It follows from it that if there are any rights, there is a sense in which they already have the status of law: that is, they may legitimately be coercively enforced. This, after all, is why we think it can sometimes be morally legitimate for people to fight even their own governments for their freedom: because they have a coercively enforceable right to that freedom even if there is no positive law upholding it. On this view, natural right is underwritten by “natural law.” So the state cannot be seen as making it possible to coercively enforce a claim that is already there, since the claim was not only already there, but already coercively enforceable too.

Now this may not seem like a big problem. For of course there is still a question about the relation between law in this natural sense and the positive statutes that are actually passed by some political society. So why shouldn’t we say that a state that makes a law establishing a right is simply acknowledging a natural right that is already there, by making its own laws match the natural laws?

But there’s a problem with this too, which was brought out first by Thomas Hobbes, and then, following him, Kant. They pointed out that there is a sense in which rights do not exist even morally until laws upholding them are enacted by political society. After all, to say that you have a moral claim of right is not only to imply that you are entitled to defend your claim with force. It is also to imply that people have a moral obligation to respect your claim. But Kant and Hobbes argued that no one can be morally obligated to respect my rights until he has some guarantee that I will respect his rights. For if I force you to respect my rights without giving you a guarantee that I will respect yours, then I am putting you in a position where you are subject to my will and so are unfree. As Hobbes put it, a person who respects the rights of others when they do not respect his “would but make himself a prey to others, and procure his own ruin.” Kant agrees: “No one is bound to refrain from encroaching on what another possesses if the other gives him no equal assurance that he will observe the same restraint towards him.” (Hobbes 1994, Part 1, chapter 15, p. 9. Kant 1996, 6:307). Hobbes and Kant argued that it follows that no one has a duty to respect anyone’s rights until some mechanism of enforcing everyone’s rights is in place. Since a right implies a duty on the part of others to uphold that right, and others cannot have that duty unless their rights are upheld as well, rights occupy what we might call interpersonal space—my rights and yours can only be realized together. Notice that this explains one of the intuitions about rights that I mentioned earlier: that there is something essentially reciprocal about rights. Rights are reciprocal in the sense that the full realization of my rights depends upon the full realization of the rights of others. It follows that whenever you claim a right, you commit yourself to respecting the rights of others.

Kant argued that it is only the political state that can provide guarantees of the enforcement of everyone’s rights. So if I say, “I have a right to X,” I make a demand on others that I am not in a position to make unless we live together in a political state: claims of right presuppose the existence of the political state, that is, it presupposes our membership in a collective body committed to upholding the rights of all. Claims of right presuppose this even if we are in a pre-political “state of nature” and the political state exists only in idea. So when I claim a right in the state of nature, I commit myself to supporting the existence of a political state. According to Kant, this means that we have a duty to live in the political state. Our rights in the state of nature, are, as Kant put it, “provisional.” They exist in the sense that we have the right to defend them, but not in the sense that anyone else has a duty to respect them. It is only when the state is actually formed that they become, again as Kant put it, “conclusive.”

Kant’s distinction between provisional and conclusive rights explains the status of natural rights better than distinction between moral and legal rights does. Provisional rights are in one sense already “legal” rights, since the right-holder is morally entitled to coercively enforce them. In another sense, however, they are not yet quite moral rights, since no one else is obligated to respect them. What society does when it legalizes a right is neither to grant the right holder something that is already his own and not society’s to give, nor to acknowledge a merely moral right that is already there by making it enforceable. What society does instead is to realize a right whose existence is essentially incomplete or imperfect in the state of nature.


With that conception of natural rights in view, I now want to return to the question of what our obligations to the other animals are. What exactly is the difference between saying that human beings owe a moral duty of humanity or kindness to the other animals, and saying that animals have a right—that is to say, a natural right, in the sense I have just defined—to be treated in ways that are consistent with their good? The tradition of moral philosophy supplies us with one fairly standard way of thinking about this difference: in terms of the distinction between perfect and imperfect duties. As the distinction is normally understood, it goes something like this: A perfect duty is a duty to do some particular action. It is what is nowadays called a directed duty, meaning that it is owed to someone in particular, someone who is wronged by its omission. A standard example of a perfect duty is the kind of obligation we incur through an act of our own, like making a promise. If I make you a promise, I owe you the act I promised to perform, and I wrong you if I fail to perform it. The duties of justice are generally considered perfect duties. Here what is owed is often omission rather than action. If you violate one of my rights, you have wronged me by performing that act, which you ought to have omitted.

We have an imperfect duty, on the other hand, when we are obligated to do some general kind of action, but the duty is not specific about exactly which actions we should do, and it is not owed to anyone in particular. Many of us think of our duties of charity as taking roughly this form. We ought, we feel, to do something for those who are less well off than ourselves, but it is entirely up to us what form the giving takes and exactly to whom we will give. You can give volunteer time if you do not have much money, or money if you do not have much time. You can give to Oxfam or the Red Cross or you can micro-finance would-be entrepreneurs in third world countries. Or you can give spare change to homeless people sleeping in doorways on the street. For that matter you can give your spare change to one such person and give nothing to the next. All of that is entirely up to you. But no homeless person or charitable organization has a particular claim on you. No such individual is entitled to feel wronged or to drag you to court if you choose not to give to him. Imperfect duties, in short, do not correspond to rights.

The picture that results is a familiar one. There is a realm of justice in which we can make claims on one another. If those claims are not met, someone has been wronged. The duties of justice serve mainly to protect our freedom or autonomy and the rights in which it is embodied. But a good person, a well-motivated person, will go beyond meeting the demands of justice. She will want to promote the good of others as well as to respect their rights, and so will engage in acts of charity, kindness, and beneficence. These actions cannot be required of us by law, but a good person will want to do them. It is into this category, according to this traditional story, that the morality of both kindness to animals and of charity to strangers falls.3 This is the picture I wish to challenge.


Nowadays many people would call this story into question. One worry concerns the duties of rescue. If you are driving down a deserted road and encounter someone who has been injured in a car accident, your duty to help him seems to be a duty of beneficence. You are concerned with promoting his good, not with protecting his freedom. But you do not seem to have the discretion supposedly associated with the imperfect duties of beneficence. You cannot just choose either to help or not as you please in the way you can (supposedly) choose to give money to a particular homeless person on the street or not as you please. Of course if stopping would put you in some sort of danger, or if you are even now rushing someone else to the hospital, then we might think you do nothing wrong in not stopping to help. But short of that, it seems as if you owe it to the injured person to stop, and as if you have wronged him if you just drive on by. So there seems to be a required action and a specific person who will be wronged if you do not perform it—the characteristics of a perfect duty.

We might try to argue that the duties of rescue, unlike ordinary duties of charity, simply are perfect duties. But Peter Singer has famously argued that our duties of charity to distant strangers often have the same features as the duties of rescue (Singer 1972 and 1997). There are people in third world countries in conditions of famine or extreme poverty who are dying right now. Assuming that there is simply not enough charity already on offer to cover all the need, a small donation, not costly to you, could make the difference of life and death to one of these people. It is not clear which one, of course, so there is no particular person who can accuse you of wronging him if you do not make the donation. In that respect, the duty looks imperfect. But in other ways, it looks just like the more local case of rescue—someone will die or be harmed if you do not act—and that seems to make it a perfect duty.

This suggests that there is something wrong with the distinction between perfect and imperfect duties. Imagine a reasonably affluent society in which most people are entirely well motivated—good, genuinely charitable people. If it were really true that it is entirely a matter of discretion to whom each person gives, and exactly when and how, then someone could starve or die of an injury in their midst, without anyone doing anything wrong, simply because no one happened to choose him as the object of their charity.

Faced with this sort of problem, I think we should take a look at a different possible way of understanding the duties of beneficence—not in terms of the idea that benefactors have an imperfect duty, but in terms of the idea that those in need of aid have an imperfect right. The early modern philosophers who introduced these categories into our moral thinking often write as if an imperfect right were just the right correlative to an imperfect duty. But in fact it matters which of these ideas we take to come first.

Suppose we take the fundamental moral fact here to be that each of us, individually, has an imperfect duty to help others. It is not owed to anyone in particular. Is it owed to anyone at all? There seem to be two possible answers. First, it is not owed to anyone, but is an undirected duty. Or perhaps it is owed in some vague way to “the needy.” Apparently, then, it is the needy, taken collectively, who hold the correlative imperfect right, and the needy, taken collectively, who are wronged if it is not met. Either way, the analysis seems inadequate to the case: it is individuals, not just this group, who are wronged if their need is ignored.

But suppose instead that we take the fundamental moral fact to be that of an imperfect right. Then we can say that every person has a right to, say, adequate sustenance, and to care when sick or injured. But the right is “imperfect” because it is not clear who among us owes it to him, or exactly how the need is to be met. Who then holds the correlative imperfect duty? The parallel answer is that it is all of us— the members of the group taken collectively—who have a duty to ensure that everyone among us gets help when he is in need.

I am not the first to point out that assigning the duty of aid collectively to a group explains some other intuitions that many of us have about this kind of duty (Murphy 2000). In particular, it explains the intuition that for each of us who does have enough to help others, there is something like a “fair share” of charitable contributions, a proportion of the total need that she should be expected to meet, given her level of wealth and income. If others in the group do not give their fair share and you do, then not only those who remain in need but you also are in a sense wronged by them. It also implies that the duty of each of us is not, at least in the first instance, to give enough to charity to support all of the unmet need that we possibly can. Instead, our most immediate individual duty is to try to organize the group in such a way that all unmet need will be met. Speaking roughly, this means that taking political action to realize better conditions, where possible, may be just as good as or better than trying to solve the problem of unmet need through extensive charitable giving.

I believe that the idea that each of us has an imperfect right to aid, should we come to need it, makes much better sense than the idea that each of us has an imperfect duty to give such care to some unspecified people. It implies that if people among us suffer or perish for want of aid, those individuals have been wronged. The “imperfection” of the right rests in the fact that it is hard to pick out any particular individual who is responsible for the wrong or for correcting it. But it seems perfectly reasonable to say that society, or humanity taken collectively, has failed to uphold the right of this needy person to assistance, and so has done wrong.

That is, it seems reasonable to say this if is possible to have a right as against a group or collective in the first instance, a right that does not arise from one’s rights against the individuals in the group. We can only have a right against a collective if the collective can have the correlative duty. So before we can ask whether there really is an imperfect right to aid we must address some questions about how a collective can have a duty.


For some entity to have a duty—to be required to do something—that entity must be an agent. On most accounts of morality, and certainly on Kant’s, it must be a rational agent, one who is capable of thinking and making decisions based on reasons. In particular, the rational agent must be capable of recognizing that it has duties and responding to the fact that by doing them. How can a collective or group of people have a duty, on this account?

Like many philosophers, I think there is no problem in recognizing that groups of people can form collective agents in this way. Groups can think when their members can think, and when they can also talk and debate, which amounts to thinking together. They can act—not just individually but as a group—when they have procedures for making and enacting decisions that makes those decisions and actions count as the decisions and actions of the group. Political states are collective agents in exactly that sense. A political state thinks through the thoughts of its citizens and legislators when they debate public issues, makes its decisions by voting and passing laws, and enacts its decisions by enforcing those laws. The enforcement of the law counts as the action of the state. It also acts as a collective agent when its duly elected representatives, or otherwise legitimate leaders, perform actions in their official roles that the laws entitle them to do. That’s what happens when one country declares war on another, for instance. That is something that the country, considered as a collective agent, does.

So it is possible for a group to hold a duty corresponding to a right. But of course, humanity, taken collectively, is not at present such an organized group. We have no mechanism for making decisions that makes those decisions count as the decisions of humanity, collectively speaking. But Kant’s conception of natural right shows us that a right can exist provisionally against a group that as yet exists only in idea, so long as the group is one to whose existence those who claim the rights are committed. Suppose all human beings must claim rights, perhaps including rights to enough of the world’s resources to survive in reasonable conditions. And suppose that, as Kant thinks, this commits each of us to the existence of a collective body dedicated to upholding the rights of everyone. Then that collective body, by virtue of the commitments of its own members, exists provisionally itself. In that case it is possible for individuals to have an imperfect right against that collective body itself. So we each could have an imperfect right to aid, owed to us by humanity collectively speaking.

But do we? In the next few sections I will argue that we once we understand the implications of the grounds of property rights, at least as Kant understands them, we will see why we cannot have rights to our own property without also having a duty to share it with others when they are in need.


Kant, as I mentioned earlier, thought that the function of rights is to protect the freedom of everyone, and he thinks freedom itself is something to which every individual has a right. Among the rights he thought necessary for the protection of freedom is the right to claim things as our property: we can claim bits of the land and its products as our own. His argument for this comes in steps.

First, imagine you are living in the state of nature and you pick an apple off an unowned tree. If anyone attempted to prevent you from doing this, he would have to interfere with you physically—grab you by the wrist, perhaps, and tear the apple from your hand. This would be inconsistent with your freedom—he would be using force to try to prevent you from acting, and you would therefore have the right to resist. You would be hindering a hindrance to your freedom. It follows that you have a right to eat the unowned apple, to make it your own by consuming it, simply because no one has a right to prevent you from doing so.

But that only shows we have a right to otherwise unowned things when they are in our physical possession. Kant thought that it must also be possible to have rights to objects even when they are not in our physical possession—that is, to own them. If it were not possible to do that, we could not make free and effective use of any natural resources that we cannot immediately consume or keep at all times in our physical control. The point is perhaps most obvious when we think about the use of the land, the primary object of “property” in Kant’s theory. I cannot grow beans on a piece of land if you are free to rip them out and plant wheat there whenever I am not actually in physical control of the land we both wish to use. So without property rights, we could not use natural resources effectively without being subject to the wills of others who might compete with us for the use of those resources. In effect it would render the use of natural resources for anything except immediate consumption impossible. Kant thought this would be inconsistent with our freedom. Therefore Kant concluded that we must all agree that it is possible for an individual to claim ownership in the land and its products, where ownership is a normative relation that goes beyond physical possession. It is a right, a kind of extension of our original right to freedom ( Kant 1996, 6:246-255).

As we saw in section three, however, so long as we remain in the state of nature, these rights are necessarily “provisional.” They cannot be made “conclusive” until we come together into a political state, which will guarantee that the rights of all are upheld. This would be true as a conceptual matter even if the natural world offered itself to us in neatly bundled lots with fences already around them, but of course it strengthens the argument when we recall that there will always be disputes about the exact contours of our rights. Do the oranges falling from your tree into my yard belong to you or to me? May I put on a dam on the river as it flows through my land without regard for the effects on the water supply when the river reaches you? Such disputes can only be settled legitimately by laws we make together; otherwise one of us will be unilaterally imposing his will on the other: our relations will be relations of power rather than right.

Kant took the argument to imply that those who live on adjacent territory have a duty to organize themselves into a political state. It is people who live in proximity who make claims of right against each other, and who therefore form the group against whom an individual right is held. In more modern terms, participation in a shared economic system may be a more relevant form of proximity than shared territory. But Kant was also keenly aware that ultimately, all territory is adjacent, and that applies to economic systems as well. The Earth is round, as Kant liked to emphasize, and human technology conquers the mountains and the deserts and the sea, breaking down the barriers between us (Kant 1996, 6:311, 6:262). So, ultimately, a claim of property right is a claim against everyone else who can claim rights: a claim at least against every other human being. And that means that ultimately, whenever we claim a right, we presuppose the organization of the whole human race into an organized body dedicated to upholding and protecting the rights of everyone, and commit ourselves to membership in that group.4


Before I can explain why this leads to an imperfect right to aid, it will be useful to revert to the question of animals for a moment. I have explained why Kant thinks it must be possible for us to acquire rights in the land and its products—why we must be able to own property. In the past, philosophers have taken this kind of story also to show that we are able to claim the other animals as property. But why exactly should that be?

In the traditional doctrines of rights developed in the seventeenth and eighteenth centuries, especially in the theory of John Locke, it is perfectly clear what makes it possible for people to claim property in the other animals. It is a view, derived from Genesis, to the effect that God gave the Earth and everything that is in it to humanity to hold in common, explicitly including the animals (Locke 1980, p. 18; Genesis 1:29-30 and 9:3). Each of us can claim parts of this commonly held possession, provided that he does so in a way that respects the fact that the whole is owned or possessed in common with others. Locke thought that when you take property from the commons, you have to leave “enough and as good for others.” (Locke 1980, p. 21). Kant, as we have seen, thought that when you claim rights to property, you commit yourself to upholding the rights of others. Both insisted that the assumption that human beings hold the world in common plays an essential role in their explanations of why we can have rights to property. The role of the assumption is to answer an obvious question: How could human agreements to divide the Earth up into objects of private property be legitimate, if we had no right to the Earth in the first place?

But rather than grounding the idea of common ownership in revelation, Kant characterizes it as an idea that we necessarily presuppose when we claim our rights. He describes it, rather surprisingly, as the right of every human being to be where he or she is. He says (1996, 6:262):

All human beings are originally…in a possession of land that is in conformity with right, that is, they have a right to be wherever nature or chance…has placed them. …The possession by all human beings on the Earth which precedes any act of theirs that would establish rights…is an original possession in common…, the concept of which is not empirical…Original possession is, rather, a practical rational concept which contains a priori the…only principle in accordance with which people can rightly use a place on the Earth.

Whenever we claim a right to the land or its resources, as individuals, we presuppose that humanity in general is in rightful possession of the Earth. We have to presuppose it, because if we lacked this right, our right to freedom would be pointless, since it would not include a right to use any of the resources of action. There would be nowhere we had a right to go and nothing we had a right to use or consume. What makes humanity the group that holds the right to the Earth, according to Kant, is that each human being has a right to be “wherever nature or chance has placed them.”

The argument for this last point can be reconstructed in a way that brings out a parallel with the argument by which Kant shows we may claim ownership in previously unowned consumable goods. Prior to any individual claims of right, you are not doing anything wrong just by being here—you are not, so to speak, trespassing. It follows that if someone attempted to use coercion to remove you, he would be wronging you, by violating your freedom. Just as you have the right to resist someone who tries to snatch the apple from your hand as long as it has no individual owner, you have the right to resist someone who tries to remove you from a place on the Earth so long as that place has no individual owner. And it follows from that that you have a right to be here. Every human being has the right to freedom, and so every human being must be regarded as one of the group whose rights you must acknowledge when you claim your own rights. You claim to be part of this group simply because you are here, and so must accord that standing to others who are here as well. Characterizing this most basic of all rights as a right to be where you are gives us a pleasingly literal way of interpreting the idea of moral standing: if you find yourself standing somewhere on the planet, then you are a being who can claim rights in the land and its resources.


The idea that we originally owned the world in common may sound fanciful without its theological grounding, but I think there is a way of thinking of it that brings out an important point. Obviously, the world is not literally owned by anyone, individually or in common, until human beings come along and set up our systems of property. So prior to any system of rights there would be nothing wrong in my establishing myself on some piece of land and defending it against all comers: that is, I would not be violating anyone else’s rights in doing so. But if I am to think of this act as establishing that I have a right against all comers, as something that they have an obligation to respect, then more is needed. Or to put the same point another way, if I am to conceive of my relations with others in moral terms, in terms of right and wrong, and not merely in terms of power—then more is needed. For whenever I claim a right, as we have seen, I presuppose the idea of humanity as an organized body committed to enforcing the rights of all the inhabitants of the Earth. And if humanity did form such an organized body, then we would all have to agree, through our legislative process, about who had a right to what: about how we are going to divide things out (I will say more in defense of this claim in the next section). And—finally now to come to the point about the right to aid—we would not all agree to a division of things that effectively denies someone the resources he needs in order to live. That, after all, would amount to evicting him from the planet, as if he had no right to be here. So my rights are by their very nature limited to a reasonable share of the Earth’s resources. They are limited for the simple reason that others also have a right to a reasonable share.

That means that if property is now divided up in such a way that some of us can only live because of the charity of others, then there must be injustice in the situation, even if no one has done anything wrong to bring the injustice about. Although in the Groundwork Kant categorizes beneficence as an imperfect duty, elsewhere he puts forward exactly this way of looking at it. In the Metaphysics of Morals, after describing the duty of beneficence, he remarks:

Having the resources to practice such beneficence as depends on the goods of fortune is, for the most part, a result of certain human beings being favored through the injustice of the government, which introduces an inequality of wealth that makes others need their beneficence. Under such circumstances, does a rich man’s help to the needy, on which he so readily prides himself as something meritorious, really deserve to be called beneficence at all? (Kant 1996, 6:454).

In his course lectures Kant was even more forthright:

One may take a share in the general injustice, even though one does nobody any wrong by civil laws and practices. So if we now do a kindness to an unfortunate, we have not made a free gift to him, but repaid him what we were helping to take away through a general injustice. For if none might appropriate more of this world’s goods than his neighbour, there would be no rich folk, but also no poor. Thus even acts of kindness are acts of duty and indebtedness, arising from the rights of others (Kant 1997, 27:416).

Kant’s point here is that if some people are rich enough to give and some so poor that they need to be given to, then there must already be injustice, even if it is not the result of any particular individual doing anything wrong. The world is not divided up into shares of resources in the way it would be if humanity were actually organized into a group and deciding on this division together. This need not be anyone’s fault: it is the result of the historical development of economic conditions over a period of time when no one (no representative of humanity collectively) was in charge of world’s economy as a whole. It was no one’s job to make sure that things did not work out so that some people are rich and others poor, especially across national borders. Indeed, this kind of story illustrates why there should be such a thing as an “imperfect” right. There are imperfect rights because there are provisional rights which have not yet been made conclusive, even if this is through nobody’s fault.

It is important to see that there is injustice here, not just an unfortunate situation. In the first of the two passages I just quoted, Kant draws our attention to the fact that this kind of inequality makes the poor need the charity of the rich. Plainly, if someone else is in a position to determine whether you are able, say, to get enough to eat or not, through discretionary acts of his own, then you are subject to that person’s will. And that means that by Kant’s criterion, you are not free. The traditional picture’s distinction between a realm of justice concerned with the protection of freedom and a realm of beneficence concerned with the promotion of the good does not hold up. Without rights to a fair share of the world’s resources, the poor are a subject population. They are subject to the wills of the rich. So, as Kant urges here, we should regard giving aid not as a discretionary act of virtue, but as the correction of a wrong—as justice rather than as charity.

Following Kant, I have argued that the world’s needy have an imperfect right to the aid they need to live in reasonable conditions. This right is a provisional right, held most immediately against those on adjacent territory or in a shared economic system, but ultimately against humanity collectively speaking. Correlative to this right is a duty of humanity collectively speaking to provide the needed assistance. The right is imperfect in the sense that it is unclear who exactly should provide the assistance. Yet we have a duty to make this right perfect, by organizing ourselves in such a way that the need will be met, for example, by collecting taxes to be used in support of the poor. We have this duty because such an organization is presupposed by our own claims of right. So here we have a right with a particular structure: an imperfect right held as against humanity collectively speaking. That, I will argue is the kind of right that the other animals have against us.


Before I can explain why a similar argument shows that the other animals have rights against humanity, I must make explicit something that I have already assumed in making the argument above. In making his argument for property rights, Kant emphasizes one consequence of legitimate possession: when you are in legitimate possession of something, anyone who uses it without your permission wrongs you. But it is a more general consequence of the legitimate possession of a piece of territory that you are the one who determines what may and may not happen there: in other words, you may make the laws for a territory that is yours.

Earlier I argued that when we claim a right, we claim to be acting in the name of a political state dedicated to upholding the rights of everyone. The result of property rights having these two aspects—a right to the use of the territory and a right to determine what happens on it—is that the sovereign of the political state functions both as the legitimate ruler, and as what Kant calls “supreme proprietor” of the land. As such, the sovereign has the right both to levy taxes and to make laws that apply to the territory in general. This view has consequences both for the way we think of welfare rights and the way we think of the right to a vote, or more generally to a say in what happens.

First, consider welfare rights. Kant claims that, on his conception, “the determination of the particular property of each is in accordance with the …principle of division, instead of with principles of aggregation.” (Kant 1996, 6:323-324). In other words, it remains true within a particular state (and not just of the Earth originally) that the land is regarded as held collectively by citizens, and divided out by the sovereign among the people for their use. What Kant means when he speaks of “division” rather than “aggregation,” is that the social contract should not be conceived as made by people each of whom brings his private land into the state with him, and with the land then being aggregated into the nation’s territory, as Locke sometimes suggests (Locke 1980, Chapter IX, section 120, p. 64). Rather, from the start the territory is seen as possessed in common by all the citizens, and divided out among individuals by the sovereign. Indeed, Kant tells us that:

…the real definition [of a right] would go like this: a right to a thing is a right to the private use of a thing of which I am in (original or instituted) possession in common with others…Hence, speaking strictly and literally, there is also no (direct) right to a thing. What is called a right to a thing is only that right someone has against a person who is in possession of it in common with all others (in the civil condition) (Kant 1996, 5:261).

This means that no individual really owns the land (or the means of production more generally), a permanent thing which belongs to the people of the state collectively, and more broadly to humanity. Instead private “owners” have a sort of life interest in such property.

This is one of the main reasons why, despite his emphasis on freedom and private ownership, Kant is no libertarian. Indeed the very first conclusion he draws from the fact that the sovereign must be regarded as the supreme proprietor of the land is that the government has the right to tax the people for the support of the poor (Kant 1996, 6:325-327).

Now consider the right to a vote. If the right to territory includes a right to determine what may or may not happen on it, and we all own the territory in common, then we all share the right to determine its laws. This grounds something like a right to democracy, a right to a vote about what the laws of the state will be. To be one of the possessors of the world is to be legislative for it, to have a right to a share of control in what happens in it. This aspect of rights matters when we turn to the question of the rights of animals, because human beings have taken complete control over the world that we share with them, and their fate is determined, to varying degrees, by our laws.


I have argued that, on Kant’s view, making a claim of right commits you to the endorsement of humanity as a collective agent, a body committed to upholding the rights of everyone. If there were such an agent, the world’s resources would be divided out fairly—not necessarily equally, that would require further argument—but certainly in such a way that no one was dependent on the charity of others, and therefore subject to their individual wills. People therefore have an imperfect right, as against that collective body, to a sufficient share, and your duty to respect that right commits you to giving others aid when they are in need. This is a duty you have not simply as a particular individual, but as a representative of humanity collectively speaking. I have also claimed that people’s right to a share of the world’s resources includes a share in the determination of what happens in the world, a right to participate in the making of the laws that govern the world. I believe that, for similar reasons, animals have a right that we take into account their interests when we make our laws.

In Genesis, God gives the world to humanity, explicitly including the other animals as part of the gift. Early natural rights theorists like Locke believed that shows that we can have individual rights over animals, as we can over the land and the objects that we own. But even if that argument were sufficient to show that we can have the right to do what we like with the animals, many of us believe that nobody gave the Earth to humanity, with or without the other animals; instead, we just took it. Our relation to the other animals as things stand now is not one of right; it is one of raw power, as Hume emphasized in the passage I quoted in section 2.2. But animals are not mere objects: animals, at least the ones I am concerned with here, are sentient beings with lives and interests of their own. So we should conceive our relationships to the other animals in moral terms, not merely in terms of power. We should see them as fellow creatures, not just as objects that may be owned. To conceive of our relation to animals in moral terms, however, is think of the other animals, like ourselves, as having a right to be where they are—that is, as among those who are in rightful possession of the Earth. After all, if a human being’s claim to a rightful share of the resources of the earth and a share of control over what happens in the world is not grounded in revelation, then it is grounded, as Kant claims, simply in the fact that we are here, and are doing nothing wrong by being here. But that is true of the other animals as well. If we regard the other animals not merely as property, but as fellow creatures with a stake in what happens in this world, then their subjection to the human species raises issues of justice, not merely of kindness.

Someone might reply that in the Kantian story I am relying on here, all rights are ultimately grounded in a right to freedom, and that the other animals can have no such right, because they are not free in the relevant sense: they are governed by instinct. But at this point it is important to see that once we focus on the question of animals, two aspects of freedom that in the human case go together can come apart. The right to freedom, in Kant’s theory, involves having the right to act autonomously—in accordance with one’s own values—but it also involves having the right not to be subject to the wills of others. Each of us as an individual is, of course, subject to the collective will of the citizens of whatever political community we live in, and ultimately to the collective will of humanity, but our right to freedom includes the right not to be subject to the arbitrary will of other individuals and private organizations. Animals are not autonomous, in the Kantian sense, because they act according to their natures rather than their values, but they are agents, and they may certainly be subject to the private wills of others, and in particular, of course, to the wills of human beings. So even in Kant’s theory, issues about the other animals’ freedom—and therefore issues of right and justice—can arise.

Do animals have a right to that freedom? In spelling out his conception of right, Kant does not make it clear why he thinks that human beings have a right to their freedom, and there is room for controversy over the point. One possibility is that he intends to derive that right from the moral value he believes that autonomy gives human beings as ends in themselves. Elsewhere I have argued that on Kant’s conception, non-human animals must also be seen as ends in themselves, because they have a good of their own, and a kind of autonomy, because their instincts teach to act in ways that promote that good (Korsgaard 2018).5 On that conception, the argument would be that it is inconsistent with their status as ends in themselves to subject them to the wills of others if that means that those others may force them to act in ways that are contrary to their good. However, Kant seems to have believed that the arguments in his account of right could be made independently of his moral theory, and it is possible to read him simply as working with a conception of where the burden of proof lies. On that showing, if a creature, human or animal, is a free agent, you may not subject him to your will unless you have the right to treat him as him your property. This is because unless you have that right, he is doing nothing wrong if he resists, and the fact that he is doing nothing wrong if he resists means that he has the right to resist. The burden of proof rests on the one who wants to treat another free agent as property. If we think of things that way, in the absence of something like the Genesis story, animals are not property by their nature, and have the same right to be free of arbitrary control as human beings have.

But now that human beings taken control of most of the world’s territory and undertaken to determine what can and cannot happen on it by making our laws, we, collectively speaking, have made animals subject to human wills. We have taken over the control of the animals themselves and of a world that is as much theirs as it is ours. This gives them a two-fold right against us. First, they should not be made subject to the arbitrary wills of individuals and private organizations, a right which is violated by their status as property; and second, their interests must be taken into account when we make our laws.

Of course that way of putting it brings out a difference: it can only be the interests of the other animals, not their wills, that help to determine what sorts of laws we should have. I have suggested that, as joint possessors of the world, human beings can claim a right to something like democratic government—a right to participate in making the laws that determine what will happen in the world, so far as what happens is up to human beings. But we cannot form a collective agent or a democratic body that includes all of sentient life, and if we do presuppose such an collective, it can only be in idea. The other animals cannot join with us in setting up a set of rules to govern the world that we share with them. They cannot join with us in shared deliberations about these matters, or in making the laws that inevitably affect their own lives and welfare. Insofar as human beings govern the world, animals are necessarily a subject population. They are what Kant called “passive citizens,” citizens who cannot vote (Kant 1996, 6:314; see Korsgaard 2018, 125-126). The best we can do is try to represent their interests in making our laws.

Humanity—increasingly as time goes on—controls what happens on this planet, and the other animals, or at least some of the other animals, are—also increasingly as time goes on—subject to our wills. Animals—maybe not all of the insects and certainly not all of the even tinier creatures that are everywhere around us—but middle-sized animals of the sort that are at stake in these discussions—have gradually become a subject population, dominated by the collective power of the human species. If we regard ourselves as making laws not merely about how we should treat the other animals, but on their behalf, because like us they are among the rightful inhabitants of the world, then they have a right that we should make laws that are consistent with what is good for them. They have that right for the same reason that each of us does, because they are here, and are doing nothing wrong by being here.

The rich control the resources of the planet that we ought to share with the poor, so at least until we have greater justice, we owe them aid as a matter of right. The more humanity becomes actually organized, and therefore actually able to take collective action, the more outrageous it becomes that we do not organize ourselves in such a way that this need is met. Humanity determines the fates of both individual people and individual animals through the laws we collectively make determining what individuals may do to one another. Unlike the power imbalance between the rich and the poor, the power imbalance between human beings and the other animals is not something that will ever go away. But to the extent that animals live under the control of human laws, they are owed the protection of human laws. Animals have a right to that protection, in the same way that the poor have a right to our aid. Both groups have a right to that protection because they are a population effectively subjected to humanity’s collective control over the Earth.


At the beginning of this paper I described some of the objections that people make to the idea of animal rights: In conclusion, I will briefly sketch some responses to these worries that follow from the account I have defended. This will also give me an opportunity to bring out some of the implications of the view, although obviously that is a large and complex topic to which I cannot do justice here.

11.1. Animals are property, and property cannot have rights.

Animals are not things, and should not be property. They do not have to be, because we do not need to choose between treating them as property and treating them as persons. Since they are not autonomous, they do not have all the same rights as people, but that doesn’t mean that they have none. If private individuals and organizations are going to keep domestic animals as companions, make them work, and use their products, those individuals and organizations should serve as their guardians, and be answerable to the law for their treatment. Obviously, which kinds of interactions should be allowable should depend on what is compatible with their good, and this is a contested issue. Some people believe that death is not a bad thing for non-human animals, so long as they are treated humanely during their lives; to such people the idea that farmers might be the guardians of animals being raised for food will make more sense than it does to those of us who do not believe early death is consistent with an animal’s good.6 Wild animals and commensal animals who live in niches created by human beings and wild animals who live on territory owned by the state are currently regarded as the property of the state; instead, the state should be their guardians, at least to the extent the control of the state affects their lives. Of course there will still be hard questions when those animals become dangerous to human beings or in other ways impossible for us to live with. But once we realize that it is not true that only human beings have rights, we should not assume that human interests should always prevail in such conflicts.

11.2. Rights require a kind of reciprocity of which animals are incapable.

I have argued that the kind of reciprocity inherent in the idea of a right is this: natural rights are provisional, and can only be made “conclusive” together. For human beings, that means that when we claim a right, we commit ourselves to upholding the rights of others, including, when we can, the rights of the other animals.7 But we also commit ourselves to the existence of an organized body that will enforce the rights of all coercively. In understanding the implications of this for animal rights, it helps to know that Kant distinguished two kinds of obligation: ethical and juridical (Kant 1996, 6:219). Ethical obligation is the obligation to be motivated by respect for the moral law in general and, in this area, by respect for the rights of others (Kant 1996, 6:391). You are juridically obligated when you may legitimately be forced to conform to a law. Animals cannot be ethically obligated, but domestic animals at least can be juridically obligated, that is, forced to conform to human laws, even though punishment for disobedience would be out of place. Domestic animals are in fact juridically obligated not to harm people or each other, or to keep away from areas where they would be dangerous to people or obstruct our projects, for example. Obviously, things are a trickier when we think of the commensal animals who live in the human world, and I will not try to take up that difficult topic here.8 In any case, if we respected the rights of animals, forcing them to conform to our laws and practices would be legitimate, and not, as it is now, an exercise of raw power.

11.3 The function of rights is to protect autonomy, and animals are not autonomous.

It is true that animal rights of the sort I have envisioned are concerned with their good, not with protecting their liberty. I think it is important that we should preserve a category of rights that are specially concerned with the protection of human autonomy, and these will only pertain to human beings. But as I have tried to bring out, there is nevertheless an issue about the freedom of animals, for even in Kant’s eyes freedom is not just the capacity to live autonomously, but the absence of domination by the arbitrary wills of others. And so long as animals do not have legal rights, the law fails to acknowledge their right not to be subject to the arbitrary wills of people and organizations. Laws that are intended to protect animal welfare without acknowledging that the animals have a right not to be subjected to people’s arbitrary wills are inadequate because they are not answerable to any standard of natural right, and we therefore feel free to shape those laws to our own convenience. For example, the federal animal welfare act in the United States simply exempts laboratory rats and mice and farm animals from any protections. Laws designed to uphold the rights of animals could not, at least in principle, be arbitrary in this way.

11.4. The idea that animals have rights has counterintuitive implications about the actions of our ancestors and those now living in the developing world.

Cruelty to animals has always been wrong. But on the view I am proposing, the natural rights of animals against humanity collectively speaking arise from a circumstance that has developed gradually: the human takeover of the world. In my discussion of the duty of aid, I have tried to emphasize that we can believe the failure to aid is a violation of right, without thinking that the gradual rise of global economic inequality is the fault of anyone in particular, because it dates from a time when no one was in charge of the economic arrangements of the world. Once we see the wrong, and we are more globally organized, and we have the institutions needed to address inequality, of course, things are different—then we are at fault if we do not. I think something similar is true in the case of our relations to animals. Certainly it is true of the loss of habitat for wild animals: the spread of the human population to nearly every part of the world happened gradually, and when no one was in charge of the division of the world, or ever thought about leaving space for wild animals. We could do something about it now, if it is not too late. 9 In a similar way, the practices of domestication arose from individuals and groups simply trying to cope with human necessities, at a time human beings were not sufficiently organized to find ways to protect animals while addressing those necessities. But at least in the developed world, this is no longer true. We can correct it now, without implying that our ancestors and those in less favorable conditions are at fault.

11.5. The idea that animals have rights has outrageous practical implications.

The rights that I have attempted to defend in this paper are rights that animals hold against humanity collectively, rights that arise from our collective control of their world. They are not rights that animals hold against each other, and do not imply that, for instance, a wildebeest has a right against a lion not to be eaten.10 I also believe that they are, as it were, proportioned to the kind of control that we exercise over the lives of animals. That kind of control is different in the case of those wild animals who live in “the wild,” or anyway away from human beings; the wild animals of commensal species who live with us, and domestic animals.11 I think their rights against us would be different accordingly, though in each case they have a right that we exercise that control as far as possible in ways that are consistent with what is good for them.

But unquestionably, acknowledging the rights of animals would have some consequences that some would regard as outrageous. Some of our practices, certainly factory farming, and invasive scientific research on animals, cannot be regarded as anything but naked exercises of power and should be given up. My aim in this paper has not been to show that the idea of animal rights can be made uncontroversial. It has been to show that there is a philosophical basis in Kant’s theory of right for the idea.

We may suppose that animals have rights against humanity for the same reason that the needy have claims to our aid: the world belongs to us all. To regard the inhabitants of the Earth morally rather than in terms of power is to regard the Earth itself as the collective possession of all its inhabitants, who all have a right to share in its bounty and to a share in the determination of what happens, including who may do what to whom. In fact, whenever we ourselves make claims of individual claims of right, whenever we claim to live as we choose or do what we like with our own property, we are asserting a right whose basis is simply that we find ourselves resident on the planet, and are doing nothing wrong by being here. So a commitment to dealing fairly with our fellow inhabitants of the planet is implicit in our claims.


Donaldson, Sue and Kymlicka, Will. (2011). Zoopolis: A Political Theory of Animal Rights. Oxford: Oxford University Press.

Hobbes, Thomas. Leviathan. (1994). Edited by Edwin Curley. Indianapolis: Hackett Publishing Company.

Hume, David. Enquiry Concerning the Principles of Morals. (1975). In David Hume: Enquiries Concerning Human Understanding and Concerning the Principles of Morals. Third Edition edited by L.A. Selby-Bigge and revised by P.H. Nidditch. Oxford: Oxford University Press.

Kant, Immanuel. Lectures on Ethics. (1997). Translated by Peter Heath. Cambridge: Cambridge University Press.

——— The Metaphysics of Morals. (1996). Translated by Mary Gregor. Cambridge: Cambridge University Press.

Korsgaard, Christine M. (2018). Fellow Creatures: Our Obligations to the Other Animals. Oxford: Oxford University Press.

——— (2009). Self-Constitution: Agency, Identity, and Integrity. Oxford: Oxford University Press.

McMahan, Jeff. “The Meat Eaters.” (2010). Editorial, The New York Times, September 19, 2010.

Locke, John. Second Treatise of Government. (1980). Edited by C. B. MacPherson. Indianapolis: Hackett Publishing Company.

Murphy, Liam. Moral Demands in Nonideal Theory. (2000). Oxford: Oxford University Press.

Rawls, John. A Theory of Justice. (1971). Cambridge, MA: Harvard University Press.

Singer, Peter. (1972) .“Famine, Affluence, and Morality.” Philosophy and Public Affairs, vol. 1, no.1., pp. 229-243.

——— (1997). “The Drowning Child and the Expanding Circle.” The New Internationalist, April 1997, pp. 28-30.

Sunstein, Cass R. (2004). “Can Animals Sue?” in Cass R. Sunstein and Martha C. Nussbaum, Animal Rights: Current Debates and New Directions. Oxford: Oxford University Press.


1.For further discussion see Korsgaard 2018, Section 3.2.3.

2.References to Kant will be given in the usual way by the page numbers of the relevant volume of Kants gesammelte Schriften, which appear in the margins of most translations.

3.You might think that the duty of kindness to animals doesn’t fit the category of imperfect duty because cruelty to an animal plainly wrongs that animal. But this ignores the fact that on this traditional conception, hurting or killing an animal just isn’t counted as cruelty in the sense that violates duty if people deem it “necessary.”

4.What I’ve just said may seem to imply that everyone who claims a right is committed to the existence of a world government. I think that it does imply that, but Kant, notoriously, held back from drawing that conclusion, for reasons that are not perfectly clear. He did, however, think that claims of right commit us to membership in some sort of collective body dedicated to upholding the rights of everyone—he thought it could be something like a non-coercive League of Nations. I leave the question to what sort of collective exactly we are committed to one side here.

5.For the view that animal action exhibits a kind of autonomy, see Korsgaard 2009, chapter 5.

6.For further discussion see Korsgaard 2018, 4.3, pp. 59-67; and 12.3, pp. 220-226.

7.Although I have argued for the Kantian conception of natural rights from the idea that rights are trumps, I do not assume that there can be no genuine conflicts of right once the rights of animals are recognized. For further discussion see Korsgaard 2018, chapters 10 and 12 generally, especially section 12.1.

8.Many proponents of animal rights believe that human beings should leave all wild animals alone. Donaldson and Kymlicka, 2011, argue forcefully this position ignores the case of commensal wild animals with whom we cannot help but interact.

9.In this essay I have argued that groups can have duties and individuals can have therefore rights against them, but I have not argued that groups can have rights and we can owe duties to them, as these remarks suggest. For further discussion see Korsgaard 2018, chapter 11.

10.This does not settle the question whether the antelope might have a right against us that we should protect him from the lion, of course. For further discussion, although not, unfortunately, a settling of the question, see Korsgaard 2018, Chapter 10, and section 11.7.

11.Donaldson and Kymlicka 2011 argue that domestic animals should be citizens, that the status of what they call “liminal” animals—commensal wild animals who depend on human beings—is comparable to that of immigrants, and wild animals who live away from human beings are like the citizens of other nations. I think of animals simply as members of a population that has become subjected to humanity, but I agree with them that the rights we hold against these groups are different.