On Wednesday, Christine Korsgaard delivered the annual Dewey Lecture in Law and Philosophy. This year’s Dewey Lecture was one of many events planned for Animal Welfare Week at the law school. Korsgaard, unhappy with utilitarian approaches to animal welfare for reasons set out below, sought to provide a Kantian account of our obligations to other animals. Throughout the lecture, Professor Korsgaard used the words “other animals” and not merely “animals,” I imagine, so as to not let us forget that we are all animals. I will emulate her here because I think that it is particularly important to keep this in mind when tackling questions of how other animals fit into our conceptions of morality. There are clearly differences between humans and other animals, just as there are differences between dogs and cats; however, not all of these differences are morally relevant. It will not do us service in our investigation to use language that implies a denial of the animal nature of humans.
Update: Audio and video of the lecture are now available after the jump.
Korsgaard began by describing the features that all animals share. We are all homeostatic organic systems that must react to, and interact with, our environments in order to survive. We have evolved to perceive sensory stimuli and this evolutionary process presupposes that we have motivational states that drive us to action. The ability to see and hear would have no evolutionary effect if sounds and sights did not produce a reaction in us that made us want to go towards or away from their sources. These attractive and aversive motivationally states lead their possessors to see the stimuli that cause them as good and bad, respectively. The claim that other animals feel is uncontroversial, at least today. Descartes would deny this, but, as far as I know, his view that the other animals are merely complex organic machines is idiosyncratic. I have never met a person who denies the sentience of non-human animals. For this reason, it is generally held that animal cruelty with no purpose is wrong. However, many people believe that any human interest or desire, except the pure desire to cause pain, is enough to override this. There are a lot of people who would not characterize their views this way, but attention to widely practiced human behavior, such as the testing of cosmetics on animals, suggests more agreement than we might like to admit.
There are two types of arguments, laid out by Korsgaard, that will help us evaluate the claim that any human interest is more valuable than any other animal interest. Both of these types of arguments must reference the differences between humans and other animals; however, it will not be enough to simply state a difference. A moral justification must rest on morally relevant differences. For example, it makes no sense to argue that we are justified in torturing certain animals, but not humans, because they have tails. On the other hand, it is perfectly reasonable to say that English speaking people are required to say “please” and “thank you” when requesting and receiving things, but cats and dogs are not. The first approach to evaluation is related to the good and bad that are pursued and avoided by humans and other animals. Behavior that discounts the pleasure and pain of the other animals might be justified if it can be shown that the way that they experience good and bad is different and inferior to the way that humans experience good and bad. The second approach is related to rights and obligations and how these arise.
The arguments based on the natures of human and non-human animal goods are utilitarian. Sentience, to varying degrees, is one of many qualities shared by all animals. To the extent that a creature can feel pain and pleasure these feelings add, or detract, from the total utility that utilitarians seek to maximize. In this sense, there is no relevant difference between humans and other animals. We should not cause them to suffer, just like we should not cause human suffering. However, it is often said that animals lack the mental capacities, shared by most humans, to recognize their own continuity through time. They cannot plan, hope, fear, etc., in a forward-directed manner. Peter Singer, who delivered the Dewey Lecture in 2004, has argued that, for this reason, there is no disutility cause by painlessly killing a non-human animal. Nothing is lost. The utility that would have been created by the animal’s future pleasure can be added by breeding another animal to take its place. For Singer, this is not the case with humans because due to our ability, and tendency, to plan, etc., more is lost when one of our lives ends. When we die something (morally relevant) is taken from the world that cannot be replaced.
Korsgaard believes that Singer’s view is mistaken. In order to explain this, she must explain why he holds it. For Singer, the value of an experience is derived entirely from its character, and not from its relation to the person or animal who experiences it. Sentient beings are merely locations where pleasure and pain happen. The boundaries between subjects are not morally relevant for utilitarianism. If good and bad are seen as tethered to the creatures that experience them, then it would make little sense to talk about maximizing and trading off utilities. This is what allows the utilitarian to do his utility calculus. Korsgaard explained that according to a utilitarian, what’s bad for you plus what’s bad for me is worse than each separately even though the union is not in any sense bad for anyone. This, Korsgaard argues, creates a major problem for Singer’s claim that there is something special about human death. He is not able to explain why the feelings of anticipation and planning, experienced only by humans, cannot be replaced by creating a new human. If accepted, the best his view can do is explain why it might be better to kill a non-human animal than a human one. But he gives no explanation as to why it is not morally acceptable to kill a person painlessly. Singer might reply that because a person has expectations and plans, awareness of the imminence of death causes more disutility for humans than for other animals. This cannot save his theory, however, because he can still not explain why it is wrong to kill a person quickly, painlessly and by surprise. Korsgaard’s proposed solution is the claim that things might be good or bad for humans in a way that is meaningfully different from how they are good and bad for the other animals. This, however, is an enormous variation. She is not claiming that Singer can be saved by arguing that there is a hierarchy of pleasures, those of humans being placed on top (a view that might be endorsed by John Stuart Mill). He would, instead, have to argue something very different, that human pain and pleasure is not replaceable in the same way as other pain and pleasure, a view that is entirely incompatible with utilitarianism. Korsgaard points out that under this view, value cannot be added across, or traded off between, subjects. Tiger death could not be meaningfully compared to human death because, if the character of an experience is tied to the subject, then a statement that one is worse than the other could only be deconstructed to mean that human death is worse for tigers than tiger death, or tiger death is worse for humans than human death; these claims are unintelligible.
After rejecting Singer’s argument, Korsgaard put forth her own view. Since her approach is Kantian, or deontological, she is not focused on maximizing the good in the world, but on interactions between individuals. This is important because it allows her to get away from comparisons between the values of human goods and non-human animal goods. However, it is not intuitive how a Kantian account can explain why we are obligated to interact with other animals in ways that are mutually beneficial and fair, her ultimate conclusion. For Kant himself, humans cannot have obligations to animals. One essential concept for Kantian ethics is that people, in virtue of their capacity for rationality, are ends in and of themselves and must always be treated as such. They can be means to other ends as well, but it is immoral to interact with a person in a way that denies that he has intrinsic value. For example, it is not immoral to take a taxi, even though the driver is being used as a means to an end, as long as you do not act in a way that denies the humanity of the driver. This idea is translated into prescriptions for action through the categorical imperative: One must act in such a way that his maxims, or the principles that guide his action, could be willed as universal law. Kantians do not recognize a hierarchy of value, so acting in a way that implies that I am more valuable than other people is the same as denying that they have value at all. Because all of these claims are tied directly to the possession of a rational will, which only humans (and angels) have, he did not believe that humans could be obligated to non-human animals. For Kant, humans can treat the other animals as means. In fact, Korsgaard read a quote from his Conjectures on the Beginning of Human History, that stated this quite clearly: When [man] first said to the sheep, ‘the pelt which you wear was given to you by nature not for your own use, but for mine’ and took it from the sheep to wear it himself, he became aware of a prerogative which . . .he enjoyed over all the animals; and he now no longer regarded them as fellow creatures, but as means and instruments to be used at will for the attainment of whatever ends he pleased.”
Kant’s view of obligation must be explained further before diving into Korsgaard’s central thesis. Obligation can only exist between two people if they are under the authority of shared laws, in the name of which they can each claim what the other owes them as a matter of right. These laws that guide behavior must be self legislated in every case. For Kant, a law that is not self legislated has no authority. Rational beings who are reasoning correctly through the use of the categorical imperative should come to the same conclusions; it is in this sense that the laws are shared. Korsgaard briefly sketched out this process of self legislation. It starts with some kind of natural drive towards what can be called a natural good. The agent must then determine whether or not to pursue this natural good. If the agent pursues the good, the act of pursuit confers absolute and normative value on the good that is pursued. The absolute value can be traced to fact that rational beings will only pursue things that are good for themselves or from their point of view; it would be irrational to do otherwise. The normative value can be traced to the categorical imperative. A rational being cannot pursue something that he would not will that everyone pursue. During this process, the moral agent has two aspects, the legislator and the subject. The subject has the initial drive towards the good and the legislator determines whether pursuit of this good could be willed as universal law. For Kant, it is the legislative aspect, the rational will, that has absolute worth. Non-human animals are like subjects without legislators. They are not rational, so for Kant, they do not have absolute worth; they are means, but not ends.
Korsgaard accepts Kant’s view that animals are not rational agents. She explained, in response to a question, that intelligence should not be confused with rationality, which she defined as the capacity for normative self government. Normative self government requires awareness of one’s grounds for action, a capacity that she does not believe that even the most intelligent animals possess. Notwithstanding this agreement, she is still able to use his theory to explain why humans can be obligated to animals. She argues that there is a difference sense in which we can talk about obligation. One is not only obligated to those with whom she has shared laws, rational beings, but also to the source of interests that the law she is under was made to protect. Korsgaard notes that this formulation might sound incompatible with the Kantian account, because it sounds like value is being assessed independently from the moral law. However, this is not the case; the law is the source of value. And if we look at the way that the legislative process confers value, we will see that the question that the legislative aspect must answer is whether the natural good is good for the subject, not the legislator.
This leads directly to Korsgaard’s thesis, which is formulated in a strong and weak version. The weak version is that some of the interests of our animal nature, that are given value by our rational nature, are shared by other animals and we must protect these. The strong thesis is that in order for the Kantian account to make sense, our animal nature must have absolute worth because otherwise we are left with a theory that holds that the source of human value is the ability to legislate regarding what is good for something that only has relative value. That is like saying that the absolute worth of people lies in our ability to take care of our children. Obviously, the analogy is not perfect because the subject cannot become a legislator, but the main idea remains. In order for the claim to make sense, children must have absolute worth as well. It follows that animals, even though they do not have a human nature, are ends in themselves.
Korsgaard’s conclusion may be too hastily rejected by some, independent of its wisdom, for what they see as pragmatic reasons. She believes that this is a misconception. Recognizing the intrinsic value of all animals does not require that we maximize their happiness. That concept is incoherent to a Kantian. The idea is that we have an obligation to interact with all animals, human or otherwise, in ways that are mutually beneficial and fair. For example, we are not required to stop predation altogether in the animal kingdom even though, for Korsgaard, we may be required to eliminate it from our own lives. This rests on the fact that while humans can survive without meat, certain other animals cannot.
Korsgaard’s thesis suggests several questions about the specific obligations of humans. In response to a question, Korsgaard stated that we might be obligated to restore the balance of nature in situations where we have created an unsuitable habitat, even though we would not be required to fix the problem if it had occurred naturally. She acknowledges, however, that this is a difficult question and the answer in a particular case will most likely depend on the details. Another interesting question is whether, under this theory, it would be permissible to breed, for food, an animal-like organism that did not have interests. The question was asked by an audience member who expressed a clear distaste for the idea. Korsgaard, however, responded that this is not so nightmarish if we re-frame the organism as a protein producing plant. She does not see a problem here as long we act in accord with our obligations to the other animals with which we must interact in order to create this new organism.
The lecture ended with a question about how this theory can be used by a government that follows John Rawls in not grounding political principles in values that cannot be accepted by all reasonable citizens. Korsgaard’s response was that we must first look for overlapping consensus. In this case, that might be that senseless cruelty to non-human animals is not tolerable. Investigation then must be done into the basis for the legal rule that everything is either a person or a thing, which, given generally held ideas of animal cruelty, might not actually be the product of overlapping consensus itself. It is unlikely that this inquiry would lead to government prescribed vegetarianism; however, the creation of a third legal category would certainly produce some positive change in the way that animals are treated.