Christine Korsgaard argues in this book that we humans have obligations to other animals because life can go well or badly for them: they are the subjects of their lives, and they therefore have an interest in how those lives go. Those facts about them give them a moral claim on us. Korsgaard, as the Kantian she is, makes this claim by saying that we ought to treat animals—all of them, severally and collectively—as ends in themselves and never merely as means to our ends. The burden of Fellow Creatures is to explain the underpinnings of this thesis, and to show some of its practical implications.

Philosophers have devoted considerable attention to these questions during the past forty years or so. Peter Singer’s Animal Liberation, which makes a case similar to Korsgaard’s though from a utilitarian rather than a Kantian point of view, was published in 1975 and has been the subject of passionate debate ever since—it’s the kind of book that causes people to change their lives. There have also been distinguished contributions by Wittgensteinians such as Stanley Cavell. Tom Regan’s Case for Animal Rights dates from 1983, and went into a second edition in 2004. Lori Gruen, Clare Palmer, Jeff McMahon, and many others have made substantial and serious contributions to the developing anglophone discussion. Korsgaard herself, perhaps the most distinguished Kantian of her generation, began to write on the topic about fifteen years ago, and this book organizes and develops positions sketched in various essays published during that time.

Novelists, too, notably J. M. Coetzee, especially in Disgrace (1999) and Elizabeth Costello (2003), have shown, grimly and unforgettably, what the relations between human and non-human animals look like when they’re ordered around industrial-scale slaughter, and what our apparent insouciance about that arrangement suggests about us. Margaret Atwood, in her dystopian MaddAddam trilogy (2003–2013), dramatizes our genetic manipulation of non-human animals and how this contributes to the descent of our world into chaos. Anthropologists, ethnologists, and primatologists, notably Frans de Waal, have explored similarities between us and other primates, with an eye to reordering our sense of ourselves. There’s a massive literature from almost every angle on the question of human exceptionalism—what, if anything, is unique to our species, and does it, whatever it may be, justify the way we relate to members of other species? There are activists, lobbyists, and political pressure groups attempting to transform law, policy, and public opinion on matters having to do with the relations between human and non-human animals. And, perhaps most pressing at the moment, it’s increasingly evident that our current methods of producing and consuming animal protein are, in addition to the slaughter they involve, contributing to the rapidity of climate change and to the concomitant extinction of species. These are matters that exercise many people, and questions about how we humans should treat non-human animals are by now unavoidably implicated with them.

And it’s not only people who write books and make laws and engage in political activism. Everyone, in one way or another, has to do with non-human animals, whether by eating them, hunting them, farming them, living with them as pets, killing them when they appear where they’re not wanted (a mosquito on your skin, a cockroach in your kitchen). Among the middle classes, reliably lively dinner-table conversations can be had about meat-eating, factory farming, pet-keeping, and so on. And among the things Christians seem to care most deeply about, at least in Europe and the United States, is the eternal destiny of their pets. That’s a feature of our time and place and mode of social organization.

Korsgaard’s topic, then, has purchase and relevance. And her book is a very good one. It can be instructive to professionals in the field, and to anyone prepared to give it time and patience even without being knowledgeable about the interpretation of Kant and Aristotle. She writes, mostly, in such a way as to make each chapter capable of being read as an independent essay, and when this isn’t the case she clearly indicates where to turn in the book for further elucidation. And she writes, above all, with clarity and care. As every philosopher knows, to get your readers to see what your distinctions come to and what work they do is just about the whole game. Korsgaard is very good at that. She also shows enough of herself to readers to make them, or at least this one, like her: she dedicates her book to the five cats (named) with whom she’s lived during the past thirty-five years; and she includes, surprisingly for the kind of philosopher she is, occasional confessions of moral failure and moral compromise on her part.

For Korsgaard, to be an animal, human or not, is to be sentient. If you are sentient, that means, roughly, that it seems like something to you to be you. A capacity for pleasure and pain is a minimal version of this. More elaborate versions include a sense of having a life extended in time, and of being related to a world external to yourself in ways more complex than simple pleasure or pain. But the minimal version of sentience is sufficient for being an animal, and also sufficient for having interests and a point of view. To have those is to have a self, to be the subject of a life, and that, in turn, is to have moral standing as the animal you are. All this is, in short, what it is to have animal life: you have an interest in what’s good for you, and you act in accord with that interest.

All this is true, writes Korsgaard, of both humans and non-human animals. It isn’t, however, true of all living creatures. She thinks it isn’t true of plants, for instance, and that it may not be true of some living things we ordinarily classify as animals. It’s clear, though, that these are empirical questions for her: if experimental evidence showed that some living things we ordinarily classify as plants are sentient in this sense, she’d say of them what she says of animals. And if it turns out that, say, planaria or oysters aren’t sentient in this sense, she would not apply the argument of this book to them. Although she isn’t especially interested in worries about whether particular living things are sentient or not, it is clear from by-the-way discussions and examples that she thinks sentience belongs to many (most?) simple animals. The book includes an affecting discussion of arachnids, including dust mites, which, she allows, are possibly, even probably, sentient.

All selves are sentient and all sentient creatures are selves. For Korsgaard, however, not all selves are persons. To be one of those, you need to be rational, and that means that you need to have, or be capable of having, or be the kind of creature that could have (all these subjunctives indicate the delicacy of the position) what she calls a “normative self-conception”: the capacity to ask, and sometimes to answer, the question of whether the reasons for your actions are good ones, and to order your life in accord with answers to that question. So far, this is just Kant. Like Kant, Korsgaard thinks that human animals are rational in this sense, and that, probably, no other animals are.

So Korsgaard is an exceptionalist about humans: she thinks we have properties and capacities that non-human animals lack. But she doesn’t do what most human exceptionalists do, which is to link her exceptionalism to some objective scale of value. Quite the contrary. All value, for her, is “tethered,” by which she means that all value is value for some person or some self. The universal value is that all selves should get whatever is good for them, but the particulars of what’s good for one kind of creature won’t, ordinarily, be the same as the particulars of what’s good for another. Humans are not, in this view, objectively more important or better than spiders or tigers; we’re different from them, as they are from one another, and so what’s good for us is also different from what’s good for them. What’s uniquely good for us is to develop and live in accord with a normative self-understanding; what’s uniquely good for a spider is success at web-spinning and fly-trapping. The only way humans could be more important than all other creatures, Korsgaard writes, would be if what’s good for us were what the world was for. And she considers that idea massively implausible, part of what she calls an antiquated teleological (she might also have said “theological”) view of things.

Kant thought that humans have no reason to treat non-human animals as ends and may therefore treat them as means. Korsgaard differs from him on this, and thinks that had he seen more clearly the implications of his own position he would have agreed with her. She deploys the resources of Kantian theory to develop it. What this development amounts to is the claim that sentient non-human animals are valuable precisely because they are selves. It isn’t, as utilitarians like Singer would have it, just that their experiences are valuable, and that our task is to minimize the pain we bring them and maximize the pleasure; they’re not exchangeable receptacles for pain and pleasure. No, it’s them, the sentient creatures themselves in all their particularity and variety, that place a moral claim on us, and the content of that claim is that we may not treat them merely as means. We must treat them as ends. We must understand that each of them has a life as important to it as your life is to you: for each animal self, her life “contains absolutely everything of value,” as yours does for you. And we must act in accord with this understanding.

Korsgaard has much of interest to say about what it means to act thus, more than I can review here. First, she addresses several responses to the question of the relations between human and non-human animals with which she differs. She disagrees with those who argue that we should, if we can, abolish animal predation, whether by altering predators genetically or by removing them in some other way (she likens this view to gentrification). She disagrees with those who advocate a radical separation of human animals from all others, with the aim of having only “wild” animals in the world. She disagrees with those who advocate the domestication of all non-human animals, and with those who advocate the extinction of the human species in favor of non-human animals. The utopian nature of these projects (other than the last) should be obvious enough.

Korsgaard also takes positions on narrower questions with possibly non-utopian answers. She thinks that we should not eat animals, whether or not they’re raised and killed humanely. That’s because we can get on quite well without doing so, and to involve ourselves in an economy that takes sentient creatures’ lives so that we may eat them is to treat these creatures as means rather than ends. Similarly, and for the same reason, we should never do painful or lethal experiments on them for our benefit. (She includes in her analysis of this question an utterly convincing discussion of the place of animal experimentation in the discovery of insulin.) She thinks that in certain very limited cases we may put non-human animals to work for us—as guides, for rescue, for some kinds of police and military work, and so on. But this may be done only when it is arguably good for the animal in question, which will most often be because of a long process of coevolution that makes work with and for humans part of what seems good to the animal. We should never, she thinks, put undomesticated animals to work in this way—as, for example, the U.S. military has tried to do with dolphins. She thinks we may live with certain kinds of animals as companions, on the double ground that by so doing we may give them a life that is good for them, and that sharing our lives with them may be good for us. She is, on this, self-revealing about the compromises involved in living with and caring for obligate carnivores like cats, who must be fed meat in order to live.

I find myself in agreement with much of this, and instructed by all of it. I’m convinced that it’s possible to derive these positions from Kantian axioms, and that it’s a good thing to do so. I’m also convinced—and now I write as a Catholic theologian—that there’s much here that Catholic Christians ought to embrace, both at the level of conclusion, and at the level of distinction and argument. Still, I want to raise some doubts about the book, and to suggest some extensions to and applications of its conclusions that I think Korsgaard would be unlikely to accept.

First, there’s the question of how Korsgaard’s position applies to the question of abortion—that is, to the question of the conditions under which it’s proper to take the life of those things (there’s no uncontroversial label for them) that come into being via conception in a woman’s womb and ordinarily grow there for nine months or so until they’re born into the world.

With respect to the question of what the fetus (that’s what she calls it) is, Korsgaard’s overall position seems to allow four possibilities. One is that the fetus doesn’t live at all: it’s inanimate. The second is that it does, but that it’s not sentient, being in this way more like a houseplant than an arachnid or a cat. The third is that it lives and is sentient, being therefore a self and the subject of a life. And the fourth is that it lives and is a person, being in this like Korsgaard and you and me. If the fetus is, at any stage of its life in the womb, a creature of the third or fourth kind, then for Korsgaard it would, at that stage, have the moral status common to all such creatures: it may not be treated by us as a means rather than an end, which also means that we offend against our duties to it when we take its life.

Korsgaard’s relaxed and generous approach to the question of when we should consider a creature sentient makes the first two possibilities implausible. We can dismiss the thought that the fetus is simply inanimate at once. And the thought that it’s animate but not sentient would require someone following a Korsgaardian line to assimilate it to plants and (perhaps) bacteria and the like, which would look like special pleading, to put it mildly. Both the third and fourth possibilities are a natural fit, however; and both entail the position that killing a fetus is at least as wrong as killing a cat, and perhaps, if the fourth is followed, as wrong as killing you or me.

This isn’t, however, what Korsgaard seems to think. She offers nothing like a full-dress treatment of abortion, and when she does discuss it in passing (in a thousand words or so) she’s not trying to convince anyone of anything about what a fetus is, or about when and how and if it becomes a person. She raises the issue only to point out that taking the moral rights of creatures to be atemporal—that is, believing it’s possible to do something that damages a creature that doesn’t yet exist (I can in this sense damage my great-grandchildren)—doesn’t entail anything about when a creature thus damaged comes to be. And about that she’s abundantly right.

But the lexicon and syntax of her discussion of abortion stand in significant tension with the deepest and most systematically worked-out aspects of her own position. She mentions, and at least entertains, the view that there’s a right to abortion during the earlier stages of pregnancy. She’s attracted by the view that a decision about when to treat fetuses as persons is like the decision about the voting age—that is, the establishment of an arbitrary bright line that doesn’t reflect any biological or metaphysical reality. But she doesn’t take that line about other forms of life. With respect to those (cats, dolphins, mosquitoes, oak trees), she thinks that there are truths in the order of being about whether they’re sentient and whether they’re persons, and that those truths are capable of being arrived at by empirical study. She says none of this about abortion and, as far as I can tell, that is because she makes a special case of human fetal life without offering any reason to do so.

A position on abortion consistent with her assumptions would be that the moral claim of a fetus depends on its sentience, that it’s an empirical question when that sentience begins, and that our criteria for assessing and acting upon the moral claims of fetuses ought to be no different than those for assessing and acting upon the moral claims of any other creature.

Second, there’s a problem about tragedy and its proper accompaniment, lament. Korsgaard, unlike many Kantians and arguably unlike Kant, rejects the thought that ought implies can. That is, she does not agree that having a duty to do something—in this case, to treat sentient creatures as ends rather than means—entails that it is possible for us to perform that duty. This means she embraces a tragic sense of life, a sense that sees life as confronting us with duties that are beyond our capacity to perform. And she’s quite right to do so; her position requires it. We could do some of the things she thinks we should: we could stop eating sentient creatures, and killing them for food; we could stop lethal or painful experiments upon non-human animals; and we could, perhaps, stop living collectively in such a way that the habitats needed for survival by some animals cease to exist. I agree with her that we should do all those things—and that we could, even if it’s deeply unlikely that we will. But we also kill small sentient creatures just by walking about and breathing and gardening, and this we cannot stop. We cannot remove ourselves from nature’s charnel house, and we cannot stop adding to the piles of corpses in it. Even Jain monks, who employ whisks to remove small things from the paths they tread and masks to prevent inhaling small things, can’t succeed in what they aspire to, which is to refrain from taking any sentient life. This is tragic, and Korsgaard goes far toward acknowledging this.

But she doesn’t get as far as she should, given her own understandings. Her work is, by and large, free from the tone of lament that would be appropriate to its argument. Having a moral duty of the kind she sketches and being unable to perform it ought to be, for a Kantian, a matter for wailing and gnashing of teeth. It means that we can’t act in accord with the goods we perceive and legislate for ourselves, and that begins to call into question the very foundations of the system: What good, we might ask, is rationality, if it demands the impossible and thus leads us to despair? I wish there were in Korsgaard’s work a greater degree of responsiveness to this question, and some engagement with what the rejection of ought-implies-can suggests for the fabric of a Kantian moral life.

Christians acknowledge, or should, that in a fallen world, there are many things we ought to do that we can’t, many evils—including death of all kinds—as yet irremediable. We have an eschatology and an understanding of creation and fall that make sense of this parlous state of affairs, no part of which is available to Kantians. With Augustine, we lament our necessities; and we might, though we rarely do, include in our table-graces lament for whatever has died in order that we might eat (including plants). But these are features that distinguish Christianity from Kantianism. Moral lament is a guest ill at ease among Kantians. Korsgaard’s positions require that it be invited in, but she does vastly too little to make it at home. Her tone is too optimistic.

Third, there’s a problem about God and hierarchies of value. This isn’t a problem internal to Korsgaard’s system. She’s right that, in the absence of a set of theological and teleological assumptions she doesn’t share and doesn’t seriously entertain, it isn’t possible to find sense in the idea that the flourishing of humans is in some objective sense more important than the flourishing of non-human sentient creatures. And she’s right in saying that all value is “tethered” as distinguished above. A counter-view would have to claim that the good of human creatures is also the good of all other creatures, that they find what is good for them exactly in what is good for us, because the whole cosmos centers upon what is good for humans.

This view seems, on its face, implausible, and I agree with Korsgaard that it is false. But it has a significant lineage: on some readings of Christianity, it is just what Christianity claims. The triune Lord of Christian confession, the one who called everything into existence out of nothing, arguably established everything in just that way. Among the many possible evidences that can be marshaled in support of this view, there’s the fact that this Lord is incarnate precisely as a human being; and that all non-human sentient life is presented in Scripture as ordered to, and named by, human creatures. Why not then say that all creaturely good—what’s good for all non-human creatures (except the angels)—is their rightly ordered relation to us?

Korsgaard briefly comments on the existence of views like this. In a note, she mentions that the Christian philosopher Linda Zagzebski once asked her whether it would make a difference to her anti-hierarchicalism if human creatures were more important to God than non-human sentient ones. Korsgaard responds that, absent further argument, there’s no reason to think that being created for some purpose by a third party (God) entails that the purpose in question should be important to those created for it. If an evil demon had brought humans into existence as crocodile fodder, that wouldn’t make being eaten by a crocodile good for us, and it wouldn’t remotely suggest that it ought to seem good to us. This response is right if God is thought of as a third party, as a being like others with a point of view and interests and powers—a being, that is, who exists in essentially the same way that we do. But it’s not right if God is understood as Christians understand the Lord—as the one in whom all creatures participate according to their kinds, and who gives being to them by fiat, out of nothing. That God, the Lord, isn’t a third party; that God’s interests aren’t extrinsic to what creatures are, but are constitutive of them. There is, on this view, no divergence between what’s good for a particular creature and what it’s made by the Lord to be, because there are no truths about creatures other than those given them by the Lord’s creative act. That’s the fully Christian view, and Korsgaard’s critique doesn’t touch it.

She is nevertheless right, I think, that what’s good for non-human sentient creatures isn’t explicable only in terms of their relations to us. Rather, Christianity is better construed as saying that the created order has goods proper to it, as a whole and in its parts, that belong to what it is independently of any relations it has to us. This is perhaps easier to see in the case of so-called wild sentient creatures than in the case of so-called domesticated ones. The whale, the elephant, and the python are what they are independently of us, and glorify the Lord, therefore, by existing and seeking what is good for them independently of any relations they might bear to us. This is compatible with a hierarchical view in the order of being: it can still be said that we are more intimate with the Lord than they are, and that we are made in the image and likeness of God in ways that they aren’t. But this doesn’t mean they exist only for us.

Christians differ with one another about this, and the view summarized in the preceding paragraph is a minority report. It is in significant ways compatible with what Korsgaard has to say—though there remain deep incompatibilities.

Finally, should Christians do as Korsgaard does, and renounce both the eating of sentient creatures and the performance of experiments on them? Yes, we should, to the extent that we can, and Korsgaard is very helpful in getting us to see why. Sentient creatures have lives and interests and concerns; their lives and interests and concerns are as important to them as ours are to us; and what is good for them is not reducible to what is good for us. When we can—and we often can—we should refrain from killing them and eating them and experimenting on them; and when we can, we should seek and advocate what, as it seems to us, is good for them. So refraining and so seeking would glorify the Lord. And even when we can’t do these things, we should try to remain aware of our unavoidable failures by lamenting them.

Fellow Creatures
Our Obligations to the Other Animals
Christine M. Korsgaard
Oxford University Press, $24.95, 272 pp.