Sentience or Species?
Sentience or Species?
Abstract and Keywords
The pro-life and animal rights movements mirror each other: the former makes humanity a sufficient condition for moral consideration; the latter rejects humanity as even a necessary condition, focusing instead on sentience. While good reasons are needed to justify harming any sentient being, conventional morality permits us to favor humans over non-humans with respect to affirmative aid, much in the way that parents may provide aid to their own children that they deny to others.
Competent adult humans are moral agents—that is, we are beings with moral duties to others. But to which others? This chapter argues that we owe moral duties to living, sentient beings—including most animals and late-term fetuses. We consider, but reject, an alternative view that says we owe moral duties to only humans (including fetuses and zygotes). Although we may extend special benefits toward our fellow humans, it does not follow that we may ethically harm and exploit animals. Absent some very strong justification or excuse, we have a duty at least to avoid intentionally inflicting suffering or death on any sentient being, whether human or nonhuman.
As we indicated in the introduction, our strategy for inferring the content of moral duties builds on what we take to be widely shared moral intuitions. Thus, instead of beginning by postulating the general source or nature of moral duty, we begin with concrete examples.
In the 2007 case of Gonzales v. Carhart, the Supreme Court of the United States rejected a constitutional challenge to the federal (p.14) Partial-Birth Abortion Ban Act.1 “Partial-birth abortion” is not a scientific term. It is a category created by abortion opponents to refer to a type of abortion more formally known as intact dilation and evacuation (intact D&E) or dilation and extraction (D&X) that, before the ban’s enactment, had sometimes been used late in pregnancy. Abortion opponents coined the term “partial-birth abortion” to refer to the fact that during a D&X, the fetus is partially delivered before being killed.
The Carhart case involved a number of technical questions, both legal and factual. Did the federal law provide doctors with clearer notice of the prohibition’s scope than a Nebraska partial-birth abortion ban that the Court had invalidated partly on vagueness grounds in an earlier case? How frequently did doctors perform a D&X? For what reasons? How early or late in pregnancy did a D&X typically occur? Were there circumstances in which a D&X could be considered medically necessary? Which side had the burden of proof on that question? What exactly was the government’s goal in banning partial-birth abortion, given that it permitted abortions by other means at the same stage of pregnancy? Much can be, and has been, said about these and other questions, but we mention the case here for a different reason. The majority opinion contains a graphic description of the “partial-birth” procedure, and we think that the description offers an important insight into the moral force of the pro-life position. Here is an excerpt of the opinion, quoting a nurse who testified before a Senate committee about her impression of a D&X performed on a twenty-six-week fetus:
[The doctor] went in with forceps and grabbed the baby’s legs and pulled them down into the birth canal. Then he delivered the baby’s body and the arms—everything but the head. The doctor kept the head right inside the uterus. …
The baby’s little fingers were clasping and unclasping, and his little feet were kicking. Then the doctor stuck the scissors in the back of his head, and the baby’s arms jerked out, like a startle reaction, like a flinch, like a baby does when he thinks he is going to fall.
He cut the umbilical cord and delivered the placenta. He threw the baby in a pan, along with the placenta and the instruments he had just used.2
Only a psychopath can read that description without viscerally reacting with sympathy for the fetus. The clasping and unclasping hands, the kicking feet, and the startled flinch all bespeak something terrible happening to someone. Indeed, that is undoubtedly the point of the narrative. By calling the fetus a “baby” and by juxtaposing the fetus’s apparent fear and pain with the doctor’s casual disrespect for the fetus’s ruined body, the nurse makes the point that the fetus was someone who was unjustly treated as merely something.
What, exactly, makes the fetus someone rather than something? The language quoted here unmistakably points to sentience—to the fact that the fetus is capable of experiencing sensation and experiences terribly painful sensations during an abortion. The fetus does not just react to stimuli in the way that sunflowers bending to the sun’s rays do. He flinches. He is startled. We are meant to infer that he suffers. By calling attention to what the fetus does, the nurse implies that the fetus has mental and emotional states of the sort that a sunflower cannot have. And the nurse assumes that we, the audience, will attach moral significance to the fetus’s suffering and death. When all is said and done, we may still end up pro-choice on abortion, but we cannot in good faith deny that harm to a twenty-six-week-or-later fetus has moral weight.
If fetal sentience underwrites the moral revulsion most of us feel in response to the nurse’s narrative of partial-birth abortion, should we not also feel moral revulsion in response to similar impositions on nonhuman sentient animals? Do we? Let us now consider a routine practice in the egg industry.
(p.16) Through selective breeding, farmers have created two very different sorts of domesticated chickens. So-called broiler birds have been bred to grow large quickly so they can convert as little feed as possible into an enormous amount of meat very rapidly. By contrast, flesh on so-called layer hens is, from an economic standpoint, mostly wasted. These birds are thus relatively small, with feed inputs directed toward the frequent production of eggs. Constant egg laying takes its toll on the hens. They suffer calcium depletion, egg peritonitis, and other maladies. Although the wild cousins of domesticated chickens can live to be ten years old, even hens who avoid succumbing to illness will typically be killed at about two years of age, when their egg production diminishes. Accordingly, farmers must constantly replenish their supply of laying hens.
Hatcheries, meanwhile, cannot determine the sex of chicks until they hatch from their shells. At that point, hatchery employees separate the females from the males. The females are kept for laying, but the males have no economic value. They cannot produce eggs and, because they are from the layer rather than the broiler line, their bodies will not produce meat nearly as efficiently as a broiler’s body will. With no market for male layer chicks, they are killed almost immediately after hatching—regardless of whether the female layer chicks will be sent to cages in “factory” farms or to so-called free-range farms.3
How do farmers kill male chicks of the layer breed? Common industry methods include suffocating them by sealing them in garbage bags, gassing them, and macerating them—that is, grinding chicks to death by feeding them along a conveyor belt into a gigantic high-speed meat grinder. The American Veterinary Medical Association (AVMA), which publishes guidelines for the euthanasia of animals, classifies those last two methods as “humane,” although that classification is questionable.4 It is difficult to imagine anyone thinking that maceration would count as a humane method of euthanasia for an ailing family pet or—for those who think voluntary euthanasia acceptable for humans—for themselves or another terminally ill human. But even if one were to accept the AVMA claims, the truth is that many male chicks are killed by methods that the guidelines acknowledge cause serious distress, such as suffocation. And that is to say nothing of the deprivation (p.17) of life itself suffered by these millions of healthy rooster chicks (a subject to which we will return in chapter 4).
We now want to ask what we hope will seem like a shockingly easy moral question: Is the painful death of a newly hatched male “layer” chick a morally regrettable act? We are not yet asking whether it is an immoral act. Just as you might think that the suffering of fetuses during abortion is morally regrettable, yet you ultimately believe that women should have a right to abortion, so you might think that gassing, suffocating, or grinding up chicks is morally regrettable but nonetheless morally permissible.
Perhaps you think that eating eggs is necessary for human health, that human health is more important than harm to chickens, that in a modern market economy there is no realistic way to produce substantial numbers of chicken eggs without killing newly hatched male chicks, and that therefore what happens to the male chicks is, all things considered, justified. We consider the question of justified harm to animals (and to human fetuses) in the next chapter. Here we ask a more basic question: Even if necessity justifies killing newly hatched male chicks, is it still morally regrettable? If it is justified, is it a justified harm? Or is suffocating a newly hatched chick to death an act devoid of apparent moral significance, akin to cutting your fingernails or smashing a rock into pieces?
We suspect that most people will say that suffocating the baby chick is harmful. If we now ask who is harmed by the suffocating of the chick?, the answer will also be clear: Why, the chick himself, of course. And the reason will be just the same as in the abortion case: because the chick is someone, not something.
The Sentience Criterion
Put more generally, late-term fetuses and newly hatched chicks are beings with interests of their own—including interests in avoiding suffering and death. That is what distinguishes them from inanimate objects and life forms like plants that (so far as we know) lack sentience. A vandal who destroys a building commits a harmful act, but it is not a harm to the building. Instead, the vandal harms the individuals who live in, work in, or otherwise use the building. He (p.18) harms the owner of the building, whose insurance probably will not cover all of his losses. And he harms the shareholders in the insurance company, for that matter. Further, if the building was aesthetically pleasing to the eye, the vandal harms the people who now can no longer derive pleasure from admiring its architecture. But the vandal does not harm the building because he cannot harm the building. The building is not the sort of entity that can experience harm because the building cannot experience anything.
To be sure, we often speak about inanimate objects or artificial entities as though they had interests. For example, we might say that making a larger-than-expected quarterly profit was “good for Google,” even though Google is a corporation that does not itself have experiences. Or you might say that reducing your carbon footprint is good for the environment, even though the environment is not a being with interests. We can best understand such statements as a kind of shorthand. When we say that some event is good for Google, we mean that it is good for shareholders and other people, such as Google employees, who benefit from Google’s profitability. Likewise, when we say that some action is good or bad for the environment, most of us have in mind the many creatures, including humans, who benefit from robust ecosystems, rather than the environment itself.5
There is nothing wrong with talking about artificial entities and inanimate objects as though they had interests. But it is easy to lose sight of the fact that the entity in question is really only a shorthand for the interests of beings that have interests. For example, someone might think that the Holocaust worked a harm to the Jewish people beyond the harm that it did to each of the six million Jews the Nazis exterminated, to the survivors of the ghettos and camps, and to their relatives whose lives were thereby damaged. Indeed, we think something very much like this ourselves (which is hardly surprising, given that we are both Jewish and that one of us is the child of Holocaust survivors). But when you unpack the idea of harm to the Jewish people as such, it too turns out to be a complex kind of shorthand.
We can and do acknowledge that part of what it means to be a person is to identify as a member of a group that has properties greater than the sum of its individual members. Nonetheless, that (p.19) does not mean that the group has its own interests beyond all of the interests that people (and perhaps other animals) have with respect to the group. When people speak about harm to a group, they need not, and generally do not, assume otherwise. For example, when we talk about harm to the Jewish people, we treat group membership as something that individual members of the group (and some others who are not members of the group) highly value. In addition to the harm that the Holocaust caused to its direct victims, it robbed future individuals (including Jews and non-Jews alike) of an important piece of Jewish culture. But all of those harms are harms to people, not harms to any insensate entity.
Below we discuss the question of whether an act can be wrongful even if it does no harm to any being with interests. For now, we focus on our claim that sentient beings—like late-term fetuses and newly hatched chicks—have interests. We think it apparent that, by contrast with inanimate objects and the like, animals and late-term fetuses have interests precisely in virtue of the fact that they are sentient.
The view that sentient animals have interests is uncontroversial. People who are not animal rights activists of any sort claim to hold this view as well. Consider the existence and popularity of legal restrictions on cruelty to animals. These restrictions reflect the widespread sentiment that animals have interests in avoiding harm to them. To be sure, animal welfare laws are grossly under-inclusive and arguably worse than useless.6 For example, the federal Animal Welfare Act7 purports to require the humane treatment of animals used in research and as pets, but it excludes all “farm animals … used or intended for use as food or fiber,” as well as all birds, rats, and mice.8 Nonetheless, the idea behind protecting animal welfare would be incoherent if there were not agreement that animals are beings whose welfare can be considered. And animal welfare laws do more than just say that animals have interests. They restrict human freedom on the ground that it is sometimes objectionable to harm animals. In other words, our laws reflect the view that animals deserve at least some moral consideration.
People will undoubtedly disagree about what consequences follow from giving moral consideration to animals and late-term (p.20) fetuses, but we think the discussion to this point—coupled with what we take to be widely shared moral intuitions—establishes at least the following: People should not harm sentient beings without an adequate reason. We shall call that proposition the sentience criterion for moral consideration. We might equivalently state it in this way: Sentience is a sufficient condition for moral consideration.
Note that we have so far espoused the proposition that sentience is a sufficient condition for moral consideration. We have not said that acts that cause no harm to sentient creatures ipso facto lack moral implications. The late philosopher and legal scholar Ronald Dworkin, in a book in which he espoused a pro-choice view on abortion, developed the position—which he associated with older Catholic doctrine even while attempting to put it on a secular footing—that abortion is a waste of the precious gift of life. Under this view, very early abortion may be immoral even when it works no harm to the interests of the unborn.9 Dworkin himself, however, clearly placed great weight on sentience, as when he wrote “that a fetus cannot have interests of its own before it has a mental life,” even as he allowed that acts could be wrongful without causing a setback to anybody’s interests.10
In distinguishing between what he called “derivative” moral interests (those that derive from the interests of beings capable of having interests) and “detached” moral interests (those that are detached from the interests of any such beings), Dworkin believed he had found the key to unlock the abortion debate. People who oppose abortion may say that it is wrong to abort a zygote because of the harm to the zygote, but, as Dworkin would have had it, what they really think, or what they really would think if they worked it through, is that abortion of a zygote is wrong because it shows disrespect for human life in general, quite apart from any interests or rights a zygote might have.
We are not as confident as Dworkin was that we know the real reasons why people hold the views they hold about abortion (or anything else), and so we will take seriously what people in the pro-life movement actually say in support of their position. And many of them say exactly what Dworkin is at pains to deny that they might actually mean—namely, that abortion is wrong (p.21) because of the harm it does to zygotes, embryos, and fetuses, even well before they are sentient.
Before coming to the pro-life argument as it is usually made in American public debate, we do want to credit Dworkin with an important insight: For many people, the obligation to avoid doing unjustified harm to others is only one sort of moral obligation; there are other grounds for moral obligations as well. Nonconsequentialist Kantian ethics often works in this way, as when Kant says it is wrong to lie even if no one is made the worse for the lie—indeed, even if the lie would do only good.11
Dworkin’s own views about abortion and some related matters seem closer to what is sometimes called “virtue ethics” than to pure Kantianism. In virtue ethics, the rightness or wrongness of particular conduct depends on the character traits it exhibits in the person who has chosen to undertake that conduct. Dworkin expressed a version of virtue ethics when, in the last years of his life, he likened the leading of the right kind of life to a performance. “It is important that we live well,” he said, “not important just to us or to anyone else, but just important.”12
We are ambivalent about virtue ethics. On one hand, we agree with much that Dworkin and others have to say about the inherent value of a life well lived. On the other hand, we are not persuaded that virtue ethics is really an alternative to other approaches to morality, such as utilitarianism and deontology. Most of the hard moral questions that people face are not questions of whether we ought to do our moral duty but of what our moral duty is. An ethical theory that tells us that we ought to act in the way that exemplifies the character traits of a virtuous person appears to depend on some further account of what behavior is virtuous, and we have doubts that such an account can be specified by virtue ethics alone.
To be fair, Dworkin himself distinguished morality (how we treat others) from ethics (how we ought to live)13 and so perhaps we are knocking on an open door. In any event, as we have noted, we are not building a case for or against any comprehensive view of moral obligation. We invoked Dworkin’s views about the inherent value of life simply to acknowledge the possibility that we might have moral obligations that are not obligations to any being with interests (except perhaps ourselves). We do not need or want to contest (p.22) (or defend) this possibility because we mean only to advance the sentience criterion as a sufficient condition for moral consideration. For purposes of much of the argument in this book, we can be agnostic on the question of whether there are moral obligations that are independent of the interests of sentient beings.14
To return to the main current of the argument: even though we sometimes encounter people who expressly reject the sentience criterion, the sentience criterion both appeals to widely shared intuitions and is consistent with the three leading approaches to moral philosophy we have discussed. First, for utilitarians, harming beings who are capable of experiencing harm should be avoided because doing so decreases aggregate utility—unless the harm inflicted leads to compensating increases in utility; that is, unless it is adequately justified. Second, for a deontologist who believes that there are rights and duties that cannot be reduced to any calculation of costs and benefits, no duty seems more basic than a duty to refrain from causing unjustifiable harm.15 And third, to the extent that virtue ethics is, as in Dworkin’s formulation, only about ourselves, it neither affirms nor rejects the sentience criterion, which concerns how we act toward others.
Of course, we do not claim that all or even very many adherents of these various strands of moral philosophy endorse the sentience criterion. Moral philosophers typically begin by asking what obligations we owe to humans, without even pausing to consider whether the arguments they advance for their views should also imply obligations to sentient nonhumans. Even then, however, there are exceptions. As we noted in the introduction, Jeremy Bentham, the father of modern utilitarianism, is a notable example, as is Peter Singer. Likewise, some contemporary deontologists find little difficulty extending moral consideration to nonhumans.16 And the same is true for virtue ethics, as the work of Martha Nussbaum illustrates. Nussbaum is a leading modern proponent of virtue ethics. Although she is more skeptical than we are about the capacity of utilitarianism and deontology to fully embrace sentient animals’ interests as a proper subject of moral concern, she confidently (and, subject to reservations about some of her conclusions, we would also say insightfully) develops an alternative account—which she calls the “capabilities” approach—that builds on the sentience criterion.17
(p.23) Accordingly, we think the sentience criterion can fit comfortably within any of the leading schools of moral philosophy. That should not be surprising because the sentience criterion should be so self-evident as to be axiomatic within almost any defensible moral system. Certainly we hope that most readers will have come along at least so far as the sentience criterion, in light of the partial-birth abortion example, the male chick example, and our discussion to this point. Much hard work remains to be done regarding what does and does not count as adequate justification for harming sentient beings, but the principle that one should not inflict harm on beings capable of experiencing such harm without a very good reason, seems so obvious that we find it almost embarrassing to have to articulate and defend it.
Humanity as an Alternative Criterion
We nonetheless do find it necessary to articulate and defend the sentience criterion because it is routinely denied (or at least downplayed) in public debate over both abortion and the use of animals. The moral force of the description of the partial-birth abortion that we quoted earlier arises from the sentience of the fetus. Yet the pro-life view typically emphasizes something quite different: the humanity of the fetus.
The pro-life movement is diverse, but to the extent that one can find a consistent “official” position, the movement is committed to the proposition that human beings have a right to life from the “moment” of conception.18 For example, in April 2013 Kansas became the eighth American state to enact a law declaring that life begins at fertilization, and although such laws are mostly symbolic in light of federal constitutional limits on abortion prohibitions, what they symbolize is a view that accords moral consideration to human zygotes.19 As a National Right to Life Committee Web page explains: “A new individual human being begins at fertilization, when the sperm and ovum meet to form a single cell.”20 Pro-life doctors, politicians, and advocates frequently say that scientific evidence proves that conception—the fusion of human sperm and human egg—marks the beginning of each new human life. (p.24) In some versions of the argument, the assembly of twenty-three chromosome pairs marks the beginning of a unique human life. In any event, two elements appear to be crucial to the argument: First, that after but not before conception, there is a unique life that can develop into a baby, then child, then adult; and second, that because this unique new life belongs to the human species, it is a proper object of moral concern.
Note that sentience plays no apparent role in this view. Modern pro-lifers do not subscribe to the theory of homunculi or animalcules, according to which each sperm (or in some versions, egg) contains a tiny but fully developed human being, who simply expands to baby-size during pregnancy. They appear to recognize, among other things, that zygotes are not sentient.
To be sure, the “core” of the pro-life movement still seeks to appeal to centrist American public opinion and thus articulates arguments that have broader resonance. We think that explains why the pro-life movement has waged a vigorous campaign against “partial-birth” abortion and in favor of laws regarding fetal pain21—a topic to which we will return in chapters 4 and 5. By focusing on fetuses that are sufficiently developed to feel pain, the movement appeals to people who (regardless of how they would express their views) believe that sentience grounds interests, which in turn makes a sentient fetus the proper object of moral concern.
Fetal sentience, however, is not necessary to the standard pro-life case against abortion. Although there is some controversy over how long it typically takes for a fetus to develop sentience, the answer is certainly going to be measured in weeks or months. Sentience does not emerge in the first seconds, minutes, or days after conception.22 For the pro-life movement, sentience is accordingly not a necessary condition for protecting the unborn.
As we explained, however, one can accept the sentience criterion without also insisting that avoiding harm to sentient beings is the only possible moral imperative. The sentience criterion states only that sentience is a sufficient condition for moral consideration. It is thus possible to favor animal rights on the ground that all sentient beings deserve moral consideration and to oppose all abortion based on moral considerations of the sort that Dworkin (p.25) discussed: considerations that are independent of the interests of particular beings. And in fact, we have friends in the animal rights movement who are also pro-life on abortion. They see both commitments as flowing from a view about the regard they owe to life.
We respect the commitments of our animal rights / pro-life friends, but we also recognize that they constitute a small minority of a minority. Regardless of whether they are pro-choice or pro-life on abortion, most people do not treat animal sentience as sufficient to ground robust animal rights. Moreover, the arguments most commonly advanced by the pro-life community do not rely on sentience as either a necessary or a sufficient condition for moral consideration. They rely instead on the humanity of the zygote, embryo, or fetus, and for that reason, these arguments tend to cut against recognition of animal rights.
Why do most people who oppose abortion think that humanity, rather than sentience, establishes that zygotes, embryos, and fetuses are worthy of moral concern? For some, no doubt the answer is that sentience cannot ground moral concern for zygotes, embryos, and early fetuses because most of them are not sentient. These abortion opponents begin with the conviction that human zygotes deserve moral concern, and they then search for a criterion that justifies that conviction. Humanity suits; sentience does not.
If someone truly holds a settled conviction that human zygotes are proper objects of moral concern, nothing we can say will change that conviction. That is, after all, simply what it means to have a settled conviction. But we note that in public debate over abortion, people who take the pro-life position typically do not rest on their settled convictions. They offer ostensible reasons for why they hold—and why, in their view, others should also hold—the view that zygotes are entitled to moral concern.
The core of one common argument goes like this: From the moment of conception, a zygote has a full complement of his or her unique DNA, the instructions that will enable that zygote to develop into a human baby, then a child and eventually an adult. Therefore, from conception onward, a unique human individual is present.
We think that the foregoing claim trades on a crucial ambiguity in the meaning of the term “human.” A zygote is a human cell, just (p.26) as any cell in a human’s own anatomy is a human cell. No one, however, would suggest that other human cells have rights and interests independent of the living, sentient person to whom those human cells belong. If you decide to pluck a hair out of your arm or face, you thereby terminate the life of some cells, yet no one thinks that you thereby commit a homicide. Indeed, when a surgeon cuts out a cancerous growth from your body, she removes an assortment of human cells, a tumor that in some respects behaves as an organism. Yet again, no one would suggest that you or the surgeon engages in a homicide by removing the tumor—and not only because you are defending yourself from an invading pathogenic process when you do so. The fact that a living cell or organism contains human DNA is simply insufficient to qualify the cell or organism as a “human” being for purposes of allocating rights such as the right not to be killed.
One difference between a human zygote and these other cells is that the human zygote is programmed to grow, eventually, into what we can all agree is a human being. Unlike a cancer cell, if we do nothing to stop the zygote’s further development, then—assuming that conditions are right for its development—the zygote will become a human being. But this fact cannot be sufficient to establish that a zygote already is a human being. If it were, then each of the cells of your body would also already be a distinct human being (rather than simply a microscopic part of the human being that you are). After all, technology that has already been used to clone mammals such as sheep could soon enable doctors to remove the nucleus of a somatic (that is, an ordinary) cell, inject it into a denucleated egg cell, and program it to grow into a human being: a clone. Therefore, all individual human cells are potentially people—the hair cell as much as the zygote.
Furthermore, long before the development of cloning technology, eggs and sperms were potential people too. The differences among the various potential people—between zygotes, sperm cells, eggs, and ordinary body cells—are thus a matter of probabilities that they will become human beings, a matter of degree rather than of kind.23
Do such probabilities matter for moral purposes? Some people think so. For example, in an important 1970 pro-life essay, then (p.27) law professor (now retired federal judge) John Noonan noted that “life itself is a matter of probabilities” and proceeded to offer the following illustration: “If the chance is 200,000,000 to 1 that the movement in the bushes into which you shoot is a man’s, I doubt if many persons would hold you careless in shooting; but if the chances are 4 out of 5 that the movement is a human being’s, few would acquit you of blame.”24 As two of the people who would hold the shooter careless (or worse) for shooting at what could be a bird, a rabbit, or other animal, we cannot help but notice Noon-an’s implicit but obvious disregard for the moral consequences of taking animal life. Nonetheless, we set that concern aside to note that even on its own terms the example is a non sequitur. The question that Noonan was ostensibly addressing was whether the probability that a known biological entity will become a human being in the future is relevant to the morality of killing it in the present. Noonan’s example, however, concerns the morality of killing an unknown entity, given various odds that the entity is currently a human being.
With respect to the zygote, the pertinent probabilities revolve around what will happen in nine months because we already know what it is today. With respect to the movement in the bushes, by contrast, the shooter is attempting to calculate who is now behind those bushes, not what he or she will later become. The question of whether the shooter acts wrongly by firing at the movement in the bushes has no clear relevance to the question of whether it is wrong to kill a zygote (or egg or sperm), even if we know with certainty that the zygote (or egg or sperm) would otherwise become a person. One could know, in other words, that some action destroys a zygote and that absent such an action, the zygote would one day become a human being; yet those probabilities are simply irrelevant to the question of whether the zygote is already a human being, absent some further argument that Noonan fails to provide.
Does anyone offer a better justification for distinguishing between the different kinds of potential to become human beings that we find in zygotes, sex cells, and ordinary cells? Two leading pro-life philosophers have attempted to do so. Here is what Patrick Lee and Robert George say:
(p.28) Human embryos are (just as more mature human beings are) whole human organisms, and, as such, living (albeit immature) members of the species homo sapiens; somatic cells are not. Human embryos have the epigenetic primordia for internally directed maturation as distinct, complete, self-integrating human individuals; somatic cells do not. Thus, the “potential” of somatic cells is nothing remotely like the potential of the embryo. Like sperm and ova, somatic cells, though they themselves are not distinct, self-integrating human organisms (but are rather parts of other, larger human organisms), can contribute constituents to a process that brings into being a new, distinct, self-integrating human organism—a human embryo. By contrast, an embryo—whether brought into being by sexual union or cloning—is already a human being. That human being, given nothing more than an hospitable environment, will actively develop itself from the embryonic through the fetal, infant, and adolescent stages of his or her life and into adulthood with his or her unity and identity fully intact.25
This effort is problematic in at least four respects. First, note that Lee and George describe the hard work that pregnant women do as the mere passive provision of “an hospitable environment.” Yet feminists and others have long argued that abortion prohibitions are so burdensome precisely because pregnancy and childbirth are physically and psychologically demanding impositions.26 To be sure, the burdensomeness of an unwanted pregnancy primarily relates to the question of whether women ought to be entitled to have abortions notwithstanding the harm that abortion does to fetuses. The moral status of the fetus is a separate question. But we nonetheless raise the objection here because in mischaracterizing what pregnancy entails, Lee and George draw a false contrast between the interventions needed to create a human being by cloning and the supposedly automatic process of natural gestation, in which, as they would have it, the active work of pregnancy has disappeared.
Second, that distinction—between ostensibly passive pregnancy and active cloning—breaks down even further if we consider yet another means by which human beings now routinely come into existence: in vitro fertilization (IVF). Just like clones, embryos (p.29) created through IVF require intervention to develop into persons. Yet people who say that a human being’s life begins at conception in fact (and quite logically) regard embryos conceived by IVF as human beings; for that reason many of them regard stem cell research on such embryos as immoral “human experimentation.”27 Pro-life thinkers who call a zygote a person therefore cannot persuasively invoke the fact that a zygote will grow into a person without any deliberate intervention because the criterion of no deliberate intervention would exclude embryos conceived by IVF from the class of human beings.
Third, we note that by distinguishing the potential of a zygote from the potential of a sperm, egg, or somatic cell (in a world with cloning), the pro-life argument tacitly concedes that the zygote itself—as a zygote—lacks present interests. If we have moral duties with respect to zygotes, the pro-life argument acknowledges, that is for one of two reasons: Either because, to use Dworkin’s terms, respect for life as detached from the interests of any individual requires us to treat zygotes as though they were comparable to late-term fetuses, babies, children, and adults, or because the future fetus-then-baby-then-child-then-adult that this zygote could become has interests and, to use a legal phrase, the interests of those future beings “relate back” to the current zygote. We think it clear that by attempting to associate the current zygote with the future sentient being, the pro-life argument offered by Lee and George as well as others unwittingly acknowledges that sentience, not humanity, grounds interests and thus grounds moral consideration.28
Fourth and finally, we note that in our considerable (albeit non-exhaustive) reading of the pro-life literature, we have found virtually no attempt whatsoever to defend the moral relevance of the fact that zygotes, embryos, and presentient fetuses belong to the human species. It is simply assumed that this fact makes a moral difference. For example, Noonan began the conclusion to the essay we quoted earlier as follows: “The most fundamental question involved in the long history of thought on abortion is: How do you determine the humanity of a being?”29 But why? Noonan considered but rejected the objection that “humanity” is merely a secular disguise for the religious notion of ensoulment because he assumed that just about everyone—including nearly all pro-choice secular philosophers— (p.30) would share the premise that human beings have a moral right to avoid being killed simply in virtue of their humanity.30
And Noonan was right about that. For the most part, people who are pro-choice on abortion share the assumption of the pro-life community that only people are entitled to moral consideration. They simply disagree over who counts as a person. As the late pro-choice philosopher Mary Anne Warren contended in response to Noonan, personhood grounds moral entitlements, but not all humans are moral persons. Pro-choice theorists seem to assume, then, that all moral persons on the planet must necessarily be human, even as these theorists reject the assumption that humanity is a sufficient condition for moral personhood.
Interestingly, Warren allowed that extraterrestrial beings might count as persons but apparently not that nonhuman animals here on Earth could.31 More importantly, for purposes of this discussion, her quarrel with Noonan illustrates the general pattern of the pro-life/pro-choice debate among those who think that the moral status of the fetus is dispositive on the question whether abortion is immoral: Most participants in that debate begin with the premise that only humans are worthy of moral consideration and then argue over whether zygotes, embryos, and fetuses are sufficiently like paradigmatic rights-bearing humans to qualify for moral consideration. Perhaps because debates about abortion are debates about members of the human species, participants rarely think it necessary to justify rather than simply stipulate the line that divides humans and nonhumans.
Two notable exceptions are Gary L. Francione and Peter Singer. In chapter 3, we build on Francione’s feminist argument for assigning the abortion decision to individual pregnant women, notwithstanding the fact that doing so results in the abortions of some sentient fetuses, which we and Francione regard as a serious harm.32 Here we consider Singer’s argument because it calls attention to the common ground among utilitarians (like Singer) and others (like ourselves) while also providing an opportunity to see beyond some of the limits of utilitarianism.
With characteristic directness, Singer writes that “whether a being is or is not a member of our species is, in itself, no more relevant to the wrongness of killing it than whether it is or is not (p.31) a member of our race,” adding that “those who protest against abortion but dine regularly on the bodies of chickens, pigs and calves, show only a biased concern for the lives of members of our own species.”33 Singer thus concludes that we ought “to accord the life of a fetus no greater value than the life of a nonhuman animal” with similar capacities.34
We agree with Singer’s main point here. A fetus lacks interests before it attains sentience, and thus abortion cannot be a wrong to a fetus (or embryo or zygote) before the fetus (or embryo or zygote) attains sentience. We likewise agree with Singer that once one recognizes that sentience grounds interests, there is no reason other than species favoritism to deny that sentient nonhumans (like chickens, pigs, and calves) have interests that, for many purposes, are no different from those of sentient fetuses. Accordingly, we share Singer’s conclusion that if abortion is wrong because of the harm it inflicts on sentient fetuses, then killing nonhuman sentient animals is similarly wrong.
We would, however, note an important reservation. In the passage we have quoted, Singer could be read to say that humans are obligated to extend the same consideration to all sentient beings with comparable capacities. We think that view—which appears to implement Singer’s categorical declaration in Animal Liberation that “[a]ll animals are equal”35—fails to capture what we believe to be the permissible place for favoritism both within and between different species. It is in fact possible to recognize the sentience criterion without insisting that we may never favor members of our own species (or of some other species, for that matter).
Which Capacities Matter?
Thinkers who oppose extending moral consideration to animals sometimes point to various gaps between the capacities of humans and of other animals. We note that some of the supposedly unique human capacities—such as the ability to fashion tools, to use language, to show empathy, or to form a conception of oneself (as manifested by being able to use a mirror or otherwise)36—have proved not to be unique when biology showed their presence in (p.32) nonhumans.37 In some respects the efforts of animal rights opponents to identify criteria that distinguish humans from all other animals resemble the shifting rationales for ill-conceived policies, like the decision by President George W. Bush to invade Iraq in 2003—justified at one time or another by a concern about weapons of mass destruction (which proved to be false), ties to al-Qaeda (which also proved to be false), human rights violations (which were real but also occurred when Saddam Hussein was a U.S. client), and the promise of bringing democracy to the Arab world (which proved to be naïve).
Likewise with respect to the capacities that supposedly distinguish humans from other animals. Just as one suspects that Bush began with the determination to invade Iraq, so one suspects that the proponents of the various capacities began with the conviction that humans are categorically superior and then went looking for evidence to defend that conviction. They provisionally settled on some criterion—like language or tool-making—that humans supposedly uniquely possessed, only to shift their ground when the evidence contradicted the uniqueness assumption. But the evidence did not cause them to reexamine their rejection of the sentience criterion because their belief in human uniqueness preceded their search for evidence and was thus not actually founded upon evidence in the first place.
That is perhaps an understandable approach if one begins with the religious conviction that God gave humans dominion over the animals, although even then one could invoke religious tradition in favor of recognizing animal interests and according them moral consideration.38 Still, at least one type of religious view—Biblical literalism—could rationally lead to a search for some crucial factual difference between humans and animals. If humans were specially created by God on the sixth day and (much later) given divine permission to eat other animals, then we would expect to find differences in kind, not just degree, between humans and other animals.
However, anyone who accepts the overwhelming scientific evidence for evolution would expect most human capacities to be found to varying degrees in at least some other animals. And indeed, just as we see obvious morphological similarities between humans (p.33) and other animals, so has science increasingly discovered similarities between the mental lives of humans and other animals.39
Humans find the existence of impressive animal mental capacities fascinating, as evidenced by the popularity of YouTube videos of cats and dogs playing musical instruments, singing, or dialing 911.40 However, none of these capacities is necessary for moral consideration. If one asks why the torture of a human being is wrong, surely the answer has nothing to do with the particular torture victim’s ability to play the piano, to dial 911, or to write in a diary. The intentional infliction of suffering on human beings is presumptively wrong because suffering harms human beings, and it is wrong to cause harm to other beings capable of being harmed, absent adequate justification.41 Likewise, the intentional infliction of suffering on nonhuman animals is presumptively wrong for the same reason.
We quoted Jeremy Bentham in the introduction to note the important contribution that utilitarians have made to the development of our understanding of moral obligations to animals. We quote him again here because his point grounds a moral obligation regardless of whether one accepts utilitarianism. Bentham said: “The question is not ‘Can they reason?’ nor Can they speak?’ but ‘Can they suffer?’”42 That one should generally avoid inflicting unnecessary suffering on beings capable of suffering is so basic a moral principle that it seems to us nearly impossible to argue for it except by pointing to what we and, we think, most readers would regard as the morally disastrous consequences of establishing more stringent criteria for moral consideration. Some of the more stringent criteria—such as language use—would disqualify from moral consideration not only late-term fetuses and many animals but also human infants as well as severely disabled and some elderly humans. And yet widely shared moral intuitions tell us that, so long as such beings are sufficiently developed to have any present experiences at all, we might even regard the infliction of suffering on such individuals as worse, not better, in virtue of their inability to comprehend why they are suffering.
Nonetheless, some philosophers now argue that a being cannot really feel pain unless it understands that it is experiencing pain.43 This strikes us as almost exactly backward: the most searing pains render one incapable of understanding pain or anything else; they (p.34) are raw sensory experiences, and much the worse to endure as a consequence of that fact.
Of course, sometimes anticipation can make suffering worse, as torture victims attest. And some argue that the ability to anticipate the future may bear on the important question of whether the deprivation of life itself—apart from the infliction of suffering—does harm to the being deprived of life. As we explore in chapter 4, it is surprisingly difficult to explain what interest any of us—human or nonhuman—has in our continued existence. But so far as eligibility for any moral consideration at all is concerned, nothing turns on these distinctions. The only capacity relevant to the question whether it is presumptively wrong to inflict suffering on a being is the being’s capacity to suffer.
To be sure, the distinct capacities of different animals, including different humans, are relevant to some moral questions regarding other aspects of our treatment of the various beings. For example, isolation of herd or other social animals (including humans) from others of the same species will typically cause distress, whereas some animals (such as adult male orangutans) rarely come together with others of their species except for mating. A complete list of the varying capacities of different animals would be useful for developing a full theory of what we owe to other animals—what Nussbaum calls a “capabilities approach” for “animal entitlements,” “animal dignity,” and animal “flourishing.”44
Nussbaum’s concern for flourishing raises an important question about our duties to both human and nonhuman animals. Many people think that we have moral obligations to assist just those among our fellow human beings whose limited capacities place them in need of help. Notably, the late moral and political philosopher John Rawls argued for what he called the “difference principle,” according to which inequalities in the distribution of society’s material goods were justifiable only insofar as those inequalities were the result of arrangements calculated to maximize the goods available to the least well off.45 To be sure, no human society has ever implemented Rawls’s difference principle; Nussbaum herself, in discussing the duties we owe to disabled persons (and others) rejects social contractarianism of the sort Rawls embraced. But Rawls remains broadly influential, and even those, (p.35) like Nussbaum, who part company with Rawls on some matters do not deny that human beings have moral duties to the most disadvantaged in virtue of their disadvantage. In other words, incapacity, not just capacity, can give rise to moral entitlements.
But that brings us to the question: Does the widely shared intuition that it is at least permissible and probably laudable to provide affirmative aid to the needy among us entail the proposition that we must provide aid to needy nonhuman animals as well? And if so, does that render the notion that we owe moral consideration to animals a kind of reductio ad absurdum due to the prohibitive cost of providing such aid?
Let us put numbers behind those questions. The average cost of medical care for each American is over eight thousand dollars per year.46 However, those costs are distributed unevenly. Care for the sick and the old costs much more. Although the United States does not have universal health care coverage—even after the enactment of the Patient Protection and Affordable Care Act in 2010—we do provide free medical care for the very poor (Medicaid) and the old (Medicare). Many people (including the two of us) regard such spending as not mere charity but a matter of national responsibility to the poor and the aged. Indeed, it is now common to speak of the government programs that provide health insurance as “entitlement” programs. So now we come to the question: If we believe that it is right for society to spend many thousands of dollars to improve the health of humans, do we have any justification for refusing to spend comparable sums to improve the health of wild animals? Most pointedly, can we defend this and other disparities in treatment without undermining the sentience criterion?
The Difference Between Affirmative and Negative Contributions to Others’ Well-Being
Like many other people in the animal rights movement, we came to veganism through personal experience with animals—in our case, at least initially, our dogs. As we got to know our dogs, we came to understand them as beings with independent interests and (p.36) desires. Eventually, we came to see other animals that way too. Other vegans have shared similar accounts with us. They asked themselves and then they asked others: If you would not eat a dog, why do you eat pigs?47 People who oppose animal rights have begun to develop an answer, and while we think that their answer is unsatisfactory, as we shall show, it contains an insight that helps explain why it is possible to grant animals moral consideration without concluding that we have bottomless duties to animals.
Consider a 2010 essay in the Atlantic magazine by rancher Nicolette Hahn Niman, reacting to the claim that pigs and dogs are morally indistinguishable. We eat what we eat, Niman says, for cultural reasons. Thus, she opines that “[i]t’s no more contradictory to eat a pig but not a dog than it is to eat arugula but not purslane.”48
Niman’s claim confuses a moral question with a sociological question. We vegans acknowledge that cultural factors explain how various eating habits developed, but we are asking whether those habits can be justified. Niman’s comparison of eating some but not other animals to eating some but not other vegetables is comparable to a nineteenth-century slaveholder responding to the question of why he enslaves some but not other human beings by saying that he enjoys some but not other types of paintings or, at the level of a society as a whole, that in some cultures men wear neckties while in other cultures they do not. Niman’s answer and the answer given by our hypothetical slave owner amount to saying “there’s no accounting for taste.” That answer does not respond at all to the claim that enslaving anyone is wrong or that eating any sentient animal is wrong.
At various points in her brief essay, Niman flirts with the notion that there is nothing wrong with eating pigs or dogs. She invokes, without any apparent revulsion, the fact that various human cultures do in fact eat dogs. And she calls the revulsion that most Westerners feel toward eating dogs a mere “taboo.” But Niman also calls the revulsion against eating human flesh a mere taboo, thus suggesting that Niman thinks that only cultural tastes distinguish those societies that have killed and eaten other humans from those societies that forbid such conduct as murder and cannibalism.
We doubt that Niman actually believes that. Believing that taboos against murder and cannibalism reflect nothing more (p.37) than cultural tastes amounts to a radical form of moral relativism that most people reject. If Niman really is a radical moral relativist, then she should say so forthrightly: she should simply admit that she thinks that eating pigs is morally acceptable because she believes that there is no such thing as morality or immorality. But, of course, had Niman openly embraced a radical moral relativist view in that way, Atlantic readers would have properly dismissed her as unqualified to offer any opinions about right and wrong.
The people who ask whether it is possible to draw a moral distinction between pigs and dogs assume that it is not just taboo or icky to eat dogs but that it is wrong to eat dogs. The moral question we vegans pose is this: If it is wrong to eat dogs, why is it not also wrong to eat pigs?
Despite Niman’s confused and confusing flirtation with radical moral relativism, her essay hints at a potential answer to a different question. She states that “[f]or as many as 30,000 years, dogs have literally been indispensible members of the human family. Quite naturally, many humans have qualms about eating a family member.”49 Niman correctly implies that we have special duties to family members that we lack to strangers. Hence, she concludes that in societies in which dogs are regarded as family members, it is wrong to eat dogs for the same sorts of reasons that we treat family members as special. And so long as pigs are not treated as family members, the rest of the argument goes, we need not afford the same solicitude to pigs as we do to dogs, even though pigs have the same basic capacities as dogs. After all, your neighbor’s children might have the same basic capacities as your own children, but you nonetheless may—and in many circumstances must—treat your own children specially.
There is something at least partly right about Niman’s view. Some of our duties do vary based on our relationships. If a custodial parent fails to feed, clothe, or shelter his or her own children, the parent acts immorally and commits the crime of neglect, whereas if that same person fails to feed, clothe, or shelter a random stranger’s children, he or she commits neither a moral wrong nor a legal wrong. To be sure, a person who voluntarily provides aid to the children of strangers acts nobly, but what makes such conduct noble is precisely the fact that it goes beyond the call of (p.38) duty. Moral philosophers refer to such acts that go beyond the call of duty as supererogatory.50
Having in mind the difference between moral duties and supererogatory acts, it becomes easy to see what is wrong with the argument that eating pigs but not dogs is morally permissible because dogs but not pigs are considered family members: Just about all of the agent-relative duties that we owe to family members but not to strangers are duties of affirmative aid rather than duties of nonharm. By contrast, nearly everyone acknowledges that we owe most duties of nonharm to strangers and to family members alike. And more broadly, we may discriminate between kin and non-kin (and those for whom we feel empathy and those for whom we do not) in distributing positive benefits owed to no one, but, in most circumstances, we must equally respect everyone’s right to be free from violence by refraining from violent acts.
The point is painfully obvious in the case of humans. A parent has a duty to care for her own children and for those over whom she has otherwise assumed responsibility, but she has no similar duty to care for the children of strangers. Furthermore, a parent may favor her own children with expensive gifts even as she fails to give impoverished strangers the money to put a roof over their heads. Even people who donate considerable sums to the poor typically provide their own children with many more material benefits than they give to strangers, and they see no moral inconsistency in doing so.
In contrast, every competent person has a duty to refrain from abusing or otherwise harming any child (or adult, for that matter). Although a parent is permitted to buy her child a new iPad rather than to spend her money on the poor, she may not stab or shoot at either her children or a stranger’s children. We applaud the person who performs the supererogatory act of aiding strangers in need, but we would not say that anyone who fails to aid a person in need has thereby wronged that person. Conversely, we most certainly would say that affirmatively harming others—whether or not they are part of our families—is wrong (absent a strong justification).
There is no reason to think that the structure of moral duties differs when the objects of those duties are nonhuman animals. Most “pet owners” would likely say that it is wrong to abandon (p.39) the family dog on the street, even though it is not wrong to fail to adopt every stray dog in the neighborhood.51 Adopting stray dogs would represent a noble act, but it would be supererogatory, whereas pet owners believe (and the law provides) that they have affirmative duties to their own pets.
Just as with duties toward humans, the special duties we owe and the extra kindnesses that we extend to those animals we regard as family members are matters of affirmative aid. And as with our conduct toward humans, so with our conduct toward animals; though we may legitimately show favoritism toward our own family members when it comes to providing affirmative benefits, the duty not to harm is universal. Just as you cannot beat up a stranger’s child even though you need not care for a stranger’s child, so you cannot beat a stranger’s dog or a stray dog even though you need not care for the stranger’s dog or the stray dog.
So far we have been taking Niman’s example literally by asking about special benefits we extend to particular animals whom we keep as family members, but she used the example in a more general way to refer to whole categories of beings. Her suggestion was that societies that treat dogs as family members may have obligations to all dogs in virtue of the family membership enjoyed by some, rather than in virtue of any capacity that dogs and pigs both have. Might Niman’s argument for killing and eating pigs work better at the categorical level than it does at the individual level?
We think the answer is plainly no. Consider the fact that liberal democracies provide various benefits to citizens and to some other residents but not to persons who fall outside of the relevant political community. What justifies the United States in providing Medicaid and Medicare for (respectively) poor and elderly Americans but not for the poor and elderly citizens of the impoverished nations of Africa? Surely it cannot be a matter of need because the typical African has greater need of assistance in obtaining medical care than the typical American does. Instead it is a kind of family distinction at the categorical level. Notwithstanding our ideas of human equality, we permit ourselves to provide greater aid to our countrymen and countrywomen than to foreigners. Indeed, many people think that a developed country has a moral obligation to (p.40) meet the basic health care needs of its citizens. Yet notably even countries with generous social welfare states and generous foreign aid programs spend orders of magnitude more on their own citizens than on foreigners.52
Meanwhile, the important distinction between aiding and not harming continues to apply when we move from talking about the difference between agent-relative duties and agent-neutral duties we owe to specific beings to talking about agent-relative and agent-neutral duties we owe to classes of beings. Thus, even though most people would find nothing objectionable about the United States spending much more to meet the health needs of its own population than it does to meet the health needs of foreigners, nobody thinks that we are therefore entitled to kill and eat foreigners.53 At the categorical level no less than at the individual level, duties and charitable choices to provide affirmative assistance may be specific to our group, but duties of nonharm are universal.
Accordingly, Niman’s suggestion that the reason we do not eat dogs is that they are family members fails to justify the different and far worse treatment we accord pigs and other animals with relevantly similar capacities. Both at the individual and categorical levels, we may legitimately offer affirmative aid to family members that we deny to strangers, but family membership does not bear on the duty to refrain from causing affirmative harm. Niman’s argument no more justifies eating pigs but not dogs than it justifies eating strangers but not family members.
The distinction between family members and others, however, does provide us with an answer to the question with which we ended the last section: Might anything justify our provision of health care and other benefits to some humans but not to animals? The answer is that favoritism—in the sense of treating some but not others, whether human or nonhuman, as members of our categorical family—is a permissible basis for distinguishing between those to whom we provide affirmative benefits and those to whom we do not. There is no contradiction in thinking that all sentient beings are entitled not to be killed and eaten but that we may pick and choose how we wish to distribute various forms of affirmative aid, on bases that include national connections, species, or many other dimensions along which people choose to share their (p.41) wealth. Nicolette Niman may legitimately choose, in other words, to adopt a dog but not a pig, but it in no way follows that she may therefore choose to kill the pig and spare only the dog’s life.
Thus, any puzzles about why we may choose to favor particular categories of animals should not obscure the larger truth. The fact that we understand ourselves as owing special, agent-relative duties of affirmative aid and hold the option of preferentially granting charity to some humans and some select nonhuman animals does not in any way excuse us from our obligations to avoid harming animals absent an adequate justification.
Coda 1: Acts and Omissions
In this chapter, we have argued for the sentience criterion—the very modest claim that sentience is a sufficient condition for moral consideration. As our discussion of family obligations makes clear, acceptance of the sentience criterion does not preclude the possibility that we may choose to provide special affirmative aid to some humans and deny it to other beings, including members of other species. To be more precise, however, we should say that for us the sentience criterion is compatible with favoritism in dispensing affirmative benefits because we are not utilitarians. The notion of tolerating any favoritism does—or at least should—pose a problem for thoroughgoing utilitarians. Utilitarianism has difficulty justifying the distinction between acts and omissions, and that is the distinction we have drawn between the affirmative aid we may provide on an unequal basis to our chosen favorites and the negative injunction against causing harm to any sentient being.
To be sure, it is possible to construct a version of so-called rule-utilitarianism in which acts that affirmatively cause harm are deemed worse than omissions that result in comparable or greater harm.54 But for present purposes it suffices to note that the act/omission distinction is central to deontological moral theories in a way that it is not central to (or necessarily even compatible with) utilitarianism. To advert to an overused example, utilitarians are more likely than deontologists to think that it is morally permissible to throw a large man in the path of a runaway trolley so that (p.42) it kills that one man rather than leaving it on course to kill two or more people.55 Likewise, utilitarians who have not made their peace with the act/omission distinction but who nonetheless think that sentient animals are worthy of moral concern may have difficulty explaining why it is nonetheless morally permissible to do more for humans and dogs than we do for other animals.
Our goal is not to delineate the precise contours of our disagreement with utilitarians like Peter Singer. Instead, we invoke Singer because the deserved widespread influence of Animal Liberation may have led many people to think, mistakenly, that affording animals moral consideration must be grounded in utilitarianism. That mistaken thought might then lead people to worry that favoring some beings for affirmative aid is incompatible with the sentience criterion, and thus to reject the sentience criterion as too demanding. That too would be a mistake, for the sentence criterion is quite modest—although, as we argue in the next chapter, it has important implications for how we should act.
Coda 2: Family Obligations and Abortion
Our discussion of the act/omission distinction and special duties to family members also has potential implications for how to think about abortion. Begin with acts and omissions. Some of the heated disagreement over the moral permissibility of abortion depends on a characterization question: Does a woman who has an abortion harm a fetus—that is, perform an aggressive act—or does she refuse to incubate that fetus with the collateral consequence that the fetus dies, an omission? As we explain in chapter 3, each characterization is at least partly correct.
Yet one might think that abortion is morally impermissible regardless of whether abortion is best characterized as an act or an omission because a pregnant woman aborts her own baby, to whom she has a special duty of affirmative aid. We think that is indeed true, at least presumptively, with respect to a sentient fetus.
To say that a woman has a presumptive moral duty to carry a sentient fetus to term because she is (or will be) its mother is to say only that she has such a duty unless there are sufficient (p.43) countervailing considerations that justify abortion. For sentient human fetuses as for other sentient creatures, that is all that the sentience criterion requires: moral consideration.
In the next chapter, we begin to address the sorts of concerns that might be sufficient to justify deliberate harm—or, in the case of creatures to whom special duties of affirmative assistance are owed, sufficient to justify either harm or a failure to provide such affirmative assistance—to sentient beings. We close this chapter with two final caveats.
First, with respect to both animals and abortion, we have been talking about moral duties, not legal duties. One might well conclude that a woman has a moral duty not to abort a sentient fetus but might still think that the law ought not to forbid her from doing so. Perhaps one worries that criminalizing abortion will simply lead to “back-alley” abortions. Or one might worry about sex equality: After all, laws criminalizing abortion convert moral duties of women into legal duties without converting comparable moral duties of men into legal duties. We have been using the term “pro-life” in this chapter somewhat imprecisely—to refer to moral arguments against abortion—but most people who describe themselves as pro-life believe abortion should be illegal. We have not yet considered the further question of whether, even if abortion in general or some particular category of abortions is immoral, it ought to be legal or illegal.
The second caveat takes us back to our central concern: sentience. Might one think that among the special duties that parents have to their children is a duty not to abort even a presentient fetus, embryo, or zygote? Well, yes, one might think that, but we would regard such thinking as confused. As we indicated in discussing Dworkin’s view, one can regard even very early abortion as problematic because it is a waste of human life, even though the particular abortion does not harm the interests of any being. That is why we have been at pains to argue that the sentience criterion is a sufficient basis for concluding that some particular conduct has moral implications, but we have left open the possibility that other bases for moral obligations may exist as well. However, the view that a woman has a moral duty to her presentient zygote, embryo, or fetus is mistaken because a presentient zygote, embryo, or fetus (p.44) is not the sort of being to whom moral duties can be owed. It is still something, not yet someone.
To be sure, a presentient zygote, embryo, or fetus is a human something, and for that reason there may be reasons to show respect for it—in the same way that one might think it important to show respect for a dead person’s body. But the humanity of a presentient embryo or fetus is not a reason to think that it has interests, or that abortion is a harm to the embryo or fetus, regardless of the fact that the person deciding to have the abortion would be the mother of the sentient baby that the embryo or fetus would become if she carried her pregnancy to term. Sentience, not humanity, gives rise to interests.
(1.) 550 U.S. 124 (2007).
(2.) Id. at 138–39 (quoting testimony before the Senate Judiciary Committee).
(4.) AM. VETERINARY MED. ASS’N, AVMA GUIDELINES FOR THE EUTHANASIA OF ANIMALS (2013), available at https://www.avma.org/KB/Policies/Documents/euthanasia.pdf.
(5.) We say that “most” people think this way in recognition of the “deep ecology” movement, which regards nature itself, or in some articulations, all living beings, including nonsentient organisms like plants, as having interests and value. See DEEP ECOLOGY FOR THE 21ST CENTURY (George Sessions ed., 1995). We share much of the aesthetic sensibility of deep ecologists, but we nonetheless regard sentience as marking an important moral boundary. Nonetheless, nothing in our argument turns on rejecting the possibility that we might have moral duties that extend beyond what we owe to sentient beings.
(6.) Gary Francione argues that animal welfare laws actually cause harm by reassuring the public, thereby inducing complacency. See Gary L. Francione & Robert Garner, THE ANIMAL RIGHTS DEBATE: ABOLITION OR REGULATION? 4, 26, 51 (2010). One of us has argued similarly: see Sherry F. Colb, An Empty Gesture to Soothe the Conscience: Why We Pass Laws Protecting Chimpanzees and Other Animals from Cruelty, FINDLAW (Mar. 4, 2009), http://writ.news.findlaw.com/colb/20090304.html. We return to the possibility that animal welfare laws may be counterproductive to the welfare of animals in chapter 5.
(p.196) (7.) 7 U.S.C. §§ 2131–2159 (2012).
(8.) 7 U.S.C. § 2132(g) (2012).
(9.) Ronald Dworkin, LIFE’S DOMINION: AN ARGUMENT ABOUT ABORTION, EUTHANASIA, AND INDIVIDUAL FREEDOM 30–67 (1994).
(11.) Kant’s view on lying derived from his philosophy that human beings must be treated as ends in themselves, not as means to an end. To lie to another, even to bring about a positive outcome, is to use that person as a means to bring about that outcome. See Immanuel Kant, GROUNDWORK FOR THE METAPHYSICS OF MORALS 47 (Allen W. Wood ed., 2002) (1785).
(12.) Ronald Dworkin, What Is A Good Life?, N.Y. REV. BOOKS, Feb. 10, 2011, at 42 (emphasis original), available at http://www.nybooks.com/articles/archives/2011/feb/10/what-good-life.
(14.) See Jonathan Haidt, THE RIGHTEOUS MIND: WHY GOOD PEOPLE ARE DIVIDED BY POLITICS AND RELIGION 148–49 (2012).
(16.) See, e.g., Julian Franklin, ANIMAL RIGHTS AND MORAL PHILOSOPHY (2007); Tom Regan, The Case for Animal Rights, in ANIMAL RIGHTS AND HUMAN OBLIGATIONS (Tom Regan & Peter Singer eds., 2d ed. 1989). We are aware that Kant himself thought that humans did not owe moral duties to animals, see Immanuel Kant, ANTHROPOLOGY FROM A PRAGMATIC POINT OF VIEW 15 (Robert B. Louden & Manfred Kuehn eds., 2006) (1798), although he nonetheless opposed wanton cruelty to animals on the ground that people who are cruel to animals will tend to act cruelly toward other people. See Immanuel Kant, LECTURES ON ETHICS 212 (Peter Heath trans., J. B. Schneewind ed., 1997).
(17.) Martha C. Nussbaum, FRONTIERS OF JUSTICE: DISABILITY, NATIONALITY, SPECIES MEMBERSHIP 325–407 (2007).
(18.) We place “moment” in quotation marks because conception, depending on how that term is defined, is a process that typically takes between two days and a week. See Elizabeth Spahn & Barbara Andrade, Mis-Conceptions: The Moment of Conception in Religion, Science, and Law, 32 U.S.F. L. REV. 261, 265, 293–94 (1998). For our purposes, nothing turns on this distinction. The resulting zygote lacks sentience at both the beginning and the end of the process and for many weeks thereafter. See Vivette Glover Reader & Nicholas M. Fisk, Fetal Pain: Implications for Research and Practice, 106 BJOG: AN INT’L J. OBSTETRICS & GYNAECOLOGY 881, 885 (2005), available at http://onlinelibrary.wiley.com/doi/10.1111/j.1471-0528.1999.tb08424.x/full; Susan Tawia,When Is the Capacity for Sentience Acquired During Human Fetal Development?, 1 J. MATERNAL-FETAL AND NEONATAL MED. 153 (1992), abstract available at http://informahealthcare.com/doi/abs/10.3109/14767059209161911. (p.197) Thus, for simplicity, we adopt the pro-life convention of assuming a conception moment.
(19.) 2013 Kan. Sess. Laws 779. See Kevin Murphy, Kansas Set to Enact Life-Starts-“At Fertilization” Abortion Law, REUTERS (Apr. 6, 2013, 5:13 PM), http://www.reuters.com/article/2013/04/06/us-usa-kansas-abortion-idUSBRE93501220130406.
(21.) See Audrey White & Becca Aaronson, Anti-Abortion Groups Push a New Round of Rules, N.Y. TIMES, Nov. 23, 2012, at A25A; Sherry F. Colb, Sending Out Partial Birth Announcements: Symbolism and Deception by Pro-Life Legislators, FINDLAW (Jun. 18, 2003), http://writ.lp.findlaw.com/colb/20030618.html. Not all pro-life activists agree with this approach. See Josh Craddock, Why Fetal Pain Hurts (the Pro-Life Cause), LIFESITE-NEWS.COM (Jun. 24, 2013, 4:01 PM), http://www.lifesitenews.com/news/why-fetal-pain-hurts-the-pro-life-cause.
(23.) Peter Singer, PRACTICAL ETHICS 159–63 (2d. ed. 1999).
(24.) John T. Noonan Jr., An Almost Absolute Value in History, in THE MORALITY OF ABORTION 1, 56 (John T. Noonan Jr. ed. 1970).
(25.) Patrick Lee & Robert P. George, The Stubborn Facts of Science: Human Embryos Are Human Beings, NAT’L REVIEW ONLINE (Jul. 30, 2001), http://web.archive.org/web/20121113235110/http://old.nationalreview.com/comment/comment-george073001.shtml.
(26.) See, e.g., Eileen L. McDonagh, BREAKING THE ABORTION DEADLOCK: FROM CHOICE TO CONSENT 181–86 (1996).
(27.) See Elisabeth Rosenthal, Excommunication Is Sought for Stem Cell Researchers, N.Y. TIMES, Jul. 1, 2006, at A3; Sheryl Gay Stolberg, Bush Vetoes Measure on Stem Cell Research, N.Y. TIMES, Jun. 21, 2007, at A21.
(28.) We do not claim, of course, that the pro-life thinkers who oppose animal rights would acknowledge the role that sentience plays in their argument.
(31.) Mary Anne Warren, On the Moral and Legal Status of Abortion, 57 MONIST 43–61 (1973), reprinted in THE ETHICS OF ABORTION 272, 274 (Robert M. Baird & Stuart E. Rosenbaum eds., 3d ed.2001).
(32.) See Gary Francione, Abortion and Animal Rights: Are They Comparable Issues, in ANIMALS AND WOMEN: FEMINIST THEORETICAL EXPLORATIONS (Carol J. Adams & Josephine Donovan eds., 1995).
(35.) Peter Singer, ANIMAL LIBERATION 1 (1975).
(36.) One relatively early champion of extending moral concern to non-humans noted the challenge that animal rights advocates posed to “society’s (p.198) belief that we can distinguish between human and animal moral interests on the basi[s] of rationality, linguistic ability, the human soul, a God-granted dominion over animals, or the facts that humans are unique in being moral agents. …” Andrew N. Rowan, OF MICE, MODELS, & MEN 258 (1984) (emphasis in original).
(37.) See, e.g., Inbal Ben-Ami Bartal et al., Empathy and Pro-Social Behavior in Rats, 334 SCIENCE 1427; Donald M. Broom, Pigs Learn What a Mirror Image Represents and Use it to Retain Information, 78 ANIMAL BEHAVIOUR 1037 (2009); R. Allen Gardner & Beatrice T. Gardner, Teaching Sign Language to a Chimpanzee, 165 SCIENCE 664 (1969); Gavin R. Hunt, Manufacture and Use of Hook-Tools by New Caledonian Crows, 379 NATURE 249 (1996); S. Millot et al., Innovative Behaviour in Fish: Atlantic Cod Can Learn to use an External tag to Manipulate a Self-feeder, 17(3) ANIM. COGN. 779–85 (May 2014).
(38.) See, e.g., Norm Phelps, THE DOMINION OF LOVE: ANIMAL RIGHTS ACCORDING TO THE BIBLE (2002); Matthew Scully, DOMINION: THE POWER OF MAN, THE SUFFERING OF ANIMALS, AND THE CALL TO MERCY (2002).
(39.) See Jonathan Balcombe, SECOND NATURE: THE INNER LIVES OF ANIMALS (2011).
(41.) We say “presumptively” wrong rather than always wrong because there are circumstances in which pain may be necessary. For example, a doctor might perform a lifesaving but painful emergency procedure without anesthesia because there is no time to obtain or administer anesthesia. One can quibble about whether the pain is inflicted “intentionally” but justifiably in such circumstances or whether we should instead characterize it as an unintended-but-foreseen side effect. We address related questions of necessity in chapter 2.
(42.) Jeremy Bentham, AN INTRODUCTION TO THE PRINCIPLES OF MORALS AND LEGISLATION 310 n.1 (Dover Pub. 2007) (1789).
(43.) See Michael J. Murray, NATURE RED IN TOOTH AND CLAW: THE-ISM AND THE PROBLEM OF ANIMAL SUFFERING 43–66 (2011); William Lane Craig, Animal Pain and the Ethical Treatment of Animals, REASONABLE FAITH (Dec. 12, 2011), http://www.reasonablefaith.org/animal-pain-and-the-ethical-treatment-of-animals.
(45.) See John Rawls, A THEORY OF JUSTICE 75 (1971).
(46.) NHE Fact Sheet, CTRS. FOR MEDICARE & MEDICAID SERVS. (Jan. 9, 2013), http://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/NHE-Fact-Sheet.html.
(p.199) (47.) See, e.g., Melanie Joy, WHY WE LOVE DOGS, EAT PIGS AND WEAR COWS: AN INTRODUCTION TO CARNISM (2011).
(48.) Nicolette Hahn Niman, Dogs Aren’t Dinner: The Flaws in an Argument for Veganism, ATLANTIC (Nov. 4, 2010), available at http://www.theatlantic.com/health/archive/2010/11/dogs-arent-dinner-the-flaws-in-an-argument-for-veganism/66095/.
(50.) This concept was first introduced in J. O. Urmson, Saints and Heroes, in ESSAYS IN MORAL PHILOSOPHY 198–216 (A. I. Melden ed. 1958). See David Heyd, Superogation, STANFORD ENCYCLOPEDIA PHIL. (Sep. 27, 2011), http://plato.stanford.edu/archives/win2012/entries/supererogation/.
(51.) We use the phrase “pet owners” because the law currently regards companion animals as the property of the humans who house them. By doing so, however, we certainly do not endorse this classification.
(52.) For example, Norway and Sweden each spent roughly $5 billion on foreign aid in 2012. See Top Ten Donors of Foreign Aid, http://www.mapsofworld.com/world-top-ten/world-top-ten-doners-of-foreigner-aid-map.html. Yet these figures represent only a small fraction of the countries’ total spending: according to the CIA World Factbook, Norway’s estimated 2012 government expenditures totaled $206.7 billion and Sweden’s were $289.3 billion. The World Factbook: Field Listing: Budget, CENT. INTELLIGENCE AGENCY, https://www.cia.gov/library/publications/the-world-factbook/fields/2056.html.
(53.) Many Americans do not object to the killing of foreigners in war, but that is because they regard wartime killing—including wartime killing of innocent civilians as “collateral damage”—as justified. As we have been at pains to note throughout this chapter, for now we are considering only whether moral consideration is owed in the first place, not whether some interest suffices to override that moral consideration.
(54.) See, e.g., Richard B. Brandt, Toward a Credible Form of Utilitarianism, in MORALITY AND THE LANGUAGE OF CONDUCT 107 (Hector-Neri Castañeda & George Nakhnikian eds., 1963).
(55.) The original formulation of the trolley problem arose in an article about the morality of abortion. Philippa Foot, The Problem of Abortion and the Doctrine of the Double Effect, 5 OXFORD REV. 5, 8, available at http://spot.colorado.edu/~heathwoo/phil3100,SP09/foot.pdf. For one take on the ramifications of various philosophical views on one’s approach to the trolley problem, see Judith Jarvis Thomson, The Trolley Problem, 94 YALE L. J. 1395, 1404–14 (1985).