Jump to ContentJump to Main Navigation
Creation EthicsReproduction, Genetics, and Quality of Life$

David DeGrazia

Print publication date: 2012

Print ISBN-13: 9780195389630

Published to Oxford Scholarship Online: September 2012

DOI: 10.1093/acprof:oso/9780195389630.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 16 January 2019

Prenatal Moral Status and Ethics

Prenatal Moral Status and Ethics

Chapter:
(p.16) 2 Prenatal Moral Status and Ethics
Source:
Creation Ethics
Author(s):

David DeGrazia

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780195389630.003.0002

Abstract and Keywords

This chapter addresses the question of how we should understand the moral status of the prenatal human being and the attendant ethical issues of abortion and embryo research. The first section defends a framework for understanding prenatal moral status, a framework that supports liberal views about abortion and embryo research. The next section rebuts the three strongest arguments in favor of a pro-life approach. It is argued in the next section, perhaps surprisingly, that one might reasonably doubt the author’s framework. Hence a sort of pluralism regarding prenatal moral status. In view of this stalemate, the discussion is redirected to the level of political philosophy and social policy; a liberal approach to policy is defended. The final section sketches and defends such an approach to abortion and embryo research.

Keywords:   prenatal moral status, abortion, embryo research, pluralism, pro-life, pro-choice, political philosophy, social policy

Most human beings are created naturally through sexual reproduction. Some are created less naturally, with the assistance of reproductive technologies prior to implantation in a woman’s uterus. And some are created in vitro for scientific purposes without any reproductive intent. Today, no ethical issue is more contentious than abortion, which involves the killing of an embryo or fetus. Not much less contentious are embryonic stem-cell research, research cloning, and other types of embryo research that involve the destruction of embryos. All of these issues provoke the question of how we should understand the moral status of prenatal human beings.

Let us use the term “prenatal human being” broadly to apply to any living but unborn member of our species. That would include the one-cell product of conception (fertilization)—the zygote—as well as what develops from the zygote, often referred to as the embryo in the early weeks after conception and later, when organs become apparent, as the fetus (although sometimes “fetus” is used to refer to the developing human organism throughout gestation). “Prenatal human beings” in our broad sense also includes living human organisms that are not expected to be born or even to enter a woman’s uterus—in particular, embryos created artificially for research purposes.

How are we to understand the moral status of prenatal human beings? Do they matter morally in their own right, independently of their usefulness or their relations to people like you or me (postnatal human beings whose moral status and personhood are uncontested)? If they matter morally in their own (p.17) right, how much do they matter? Do they have a right to life such that it is impermissible to kill them? If they lack such a right to life, does their status nevertheless ground a moral obligation not to destroy them for trivial purposes? To answer these questions, we need to know what underlies moral status. Why do those who are uncontroversially persons enjoy moral status—or, if the latter admits of degrees, full moral status? The answer will permit us to determine whether some or all prenatal human beings share this status. If they do, and if it is wrong to kill anyone with such status, this would appear to vindicate a “pro-life” or conservative approach to the ethics of abortion and embryo research. If they do not, that would presumably open the door to liberal approaches to these issues.

It is worth noting early on that the ethical issues of abortion and embryo research involve moral dimensions in addition to prenatal moral status. Abortion, for starters, involves the termination of unwanted pregnancies, which occur in women’s bodies; and people have extensive rights to determine what happens to and within their bodies—at least as far as other people’s actions and social policies (as opposed to natural forces like diseases) are concerned. Indeed, according to one school of thought we will consider, a woman’s rights to bodily integrity and liberty are of such paramount importance that the fetus’s moral status proves largely irrelevant to the ethics of abortion. This issue may also be understood to involve broader social issues such as gender roles, patriarchy, and freedom of conscience in a secular, pluralistic society. Meanwhile, the ethics of embryo research implicates issues of how taxpayers’ money ought to be spent, embryo research being heavily dependent on public funding, as well as concerns about the possible commodification of prenatal human life and threats to traditional understandings of human procreation. The purpose here is not to provide an exhaustive catalogue of issues other than moral status that can and do come into play in discussions of abortion and ethical research. The purpose, rather, is to note some of them in order to avoid an oversimplified picture according to which prenatal moral status is all that matters in these discussions. At the same time, because it is widely believed that full moral status entails a right to life, and that the intentional destruction of an embryo or fetus would violate such a right (if it exists), it is no wonder that so many ethical discussions of these topics have centered on moral status. We will focus on prenatal moral status for much of the chapter before turning to other considerations that underlie some of the most powerful arguments in the ethical debates over abortion and embryo research.

The chapter’s first major section will defend a tripartite framework for understanding prenatal moral status. This framework consists of (1) a view about our numerical identity, essence, and origins; (2) an account of the relevance of sentience to moral status; and (3) a version of the “time-relative interests account” (p.18) of the harm of death. (All technical terms will be explained later.) As we will see, this framework supports relatively liberal views about abortion and embryo research. In the next section, I rebut what I take to be the three strongest arguments in favor of a pro-life approach. I also address what many consider the strongest argument for a liberal view of abortion—the Good Samaritan Argument—asking whether it clinches the case for a liberal position, and arguing that it does not. In the section that follows, I argue, perhaps surprisingly, that one might reasonably doubt the framework I have defended, that there are considerations that could lead a reasonable, well-informed person thinking entirely in secular terms to maintain a pro-life view. Thus, I argue for a sort of pluralism regarding prenatal moral status. In view of what I regard as a stalemate at the level of ontology (in particular, the issues of our essence and origins) and ethics, I redirect the discussion to the level of political philosophy and social policy. I argue that while a pro-life approach is reasonable, it rests on three assumptions: one about our essence and origins, another about the constancy of moral status throughout one’s lifetime, and a third about the relationship between full moral status and the ethics of killing prenatal human beings. Because each of these assumptions is highly debatable, as demonstrated by the preceding discussion—and in view of women’s interests in liberty and bodily integrity as well as biomedical researchers’ interest in freedom of inquiry—I argue that ontological and ethical pluralism supports some sort of liberal approach to policy. In the final section, I sketch and defend such an approach to abortion and embryo research.

A Framework for Understanding Prenatal Moral Status

Those who are indisputably persons have full moral status. This status incorporates a right to life—by which I mean a nearly absolute moral protection against being intentionally or negligently killed that generally resists appeals to utility as justifications for killing. If zygotes, embryos, and fetuses share this moral status, that is presumably because (1) they are beings of our kind from the time of conception, and (2) a being of our kind has moral status for the entire duration of its existence. Let us first consider the issue underlying claim (1). When do beings like you and me come into existence?

Our Essence, Numerical Identity, and Origins

The question of when we come into being, or originate, is conceptually tied to the question of our essence: What are we human persons, most fundamentally? Which of our characteristics are so fundamental that their loss would entail that we literally go out of existence? Answering this question will tell us which (p.19) properties have to be present for one of us to exist; one of us originates as soon as those essential properties appear. A related question is that of our numerical identity: Once we have come into being, what are the criteria for our continued existence, amid change, over time? What are our persistence conditions? (The concept of numerical identity is distinct from that of narrative identity, though scholars routinely confuse them in the bioethics and philosophical literature.1 A person’s narrative identity is something particular to her as an individual—namely, her self-concept or how she views herself as the subject of her implicit autobiography.2)

Psychological Views and their Difficulties

One view of our essence, which we may call person essentialism, holds that we human persons are essentially persons and therefore could not exist at any time without being persons at that time.3 Here persons are understood, roughly, as beings with the capacity for relatively complex forms of consciousness such as reasoning, intentional action, self-awareness, linguistic thought, and the like. (Hereafter, I will use the term “person” in this way. Because I do not assume that only persons have full moral status, this usage will not beg questions about the moral status of those classified as nonpersons.) According to person essentialism, you would not survive in an irreversible coma; the human organism that continues to breathe would not be you.

Elsewhere I have argued at length that this view is thoroughly implausible.4 For one thing, it implies that you were never a newborn. Because newborns lack the capacity (as opposed to potential) for the sorts of consciousness associated with personhood, they are not persons, from which it follows—for the person essentialist—that no human person is numerically identical with any newborn of the past. I take it, however, that each of us was once a newborn. Another major problem with person essentialism is its difficulty accounting adequately for the relationship between any person and the human organism or animal associated with her. Here is one aspect of this problem: Since a person, on this view, has a time of origination and persistence conditions that differ from those of the associated human organism—the organism existing for the duration of biological life, the person existing only when the capacity for sufficiently complex forms of consciousness exists—the person, strictly speaking, is not an animal. This implies, contrary to scientifically informed common sense, that you are not an animal.5

More plausible than person essentialism is another psychological view of our essence, mind essentialism, which claims that we human persons are essentially minds: beings with the capacity for consciousness (that is, for at least some conscious states even if not those characteristic of persons).6 Late fetuses and newborns are clearly sentient and therefore have states of consciousness, however (p.20) primitive. So mind essentialism does not imply, implausibly, that we did not exist as newborns. It does imply, however, that we were never pre-sentient fetuses. While I find that dubious, many people do not, so this implication may not threaten the view. I suggest, though, that like person essentialism, mind essentialism lacks an adequate characterization of the person/human animal relationship; among its implications are again that, strictly speaking, you are not an animal. This in turn apparently implies that, as you read these words, there are two conscious beings sitting in your chair: you, the person or mind, and the associated animal, who (possessing a well-functioning brain) is certainly conscious. As far as conscious beings go, this seems one too many.

One who defends mind essentialism might claim that you are part of the animal associated with you—namely the brain (more precisely, the portions of the brain associated with consciousness).7 But the brain seems capable of surviving death, whereas, on this view, you are not. Are you, then, a functioning brain, which goes out of existence at the irreversible loss of the capacity for consciousness? But it seems odd to identify the functioning brain—as distinct from the brain—as you. How could you be some organ only when it functions? Presumably you are a substance in the philosophical sense—something that bears properties—not a particular substance only when it has certain properties such as those of the brain that make consciousness possible. A proponent of mind essentialism might reply that the functioning brain is itself a substance, a substance distinct from the brain, but that claim, too, strains credibility. Might you instead be not the brain, but the mind understood as the conscious properties of the brain (or perhaps the physical properties of the brain that make consciousness possible)? That would imply that you are a set of properties, rather than a substance, a wildly counterintuitive thesis.

When the argumentative dust settles, I find mind essentialism less plausible than the view I defend, the biological view. But the biological view also faces challenges.8 The balance of arguments is quite close, so mind essentialism, unlike person essentialism, should be considered a contender among views of our essence, numerical identity, and origins.

The Biological Approach

According to the biological view, we are essentially human organisms, or animals, so that we come into being whenever the relevant organism does, and we persist for the duration of biological life. When does the human organism that characteristically develops into a person originate? Many defenders of the pro-life view assume that the answer must be: at conception, the moment when sperm and egg unite, forming a new entity—the zygote—endowed with a full complement of human genes, the program that over time drives the emergence and growth of a human person.9 I argue that this seemingly innocent assumption (p.21) is highly questionable. True, at conception a new entity comes into being, and this being appears at first glance to persist through all the stages of gestation and post-natal life, but there are powerful reasons to doubt this appearance.10

For approximately two weeks after conception, the embryo can divide into two or more parts that go on to develop into separate human beings, and it can fuse with another embryo to constitute a new organism. Call this the case of division monozygotic twinning (ignoring the possibilities of triplets, quadruplets, etc.). The resulting embryos, from which identical twins derive, have virtually identical DNA. By contrast, “fraternal” twins, resulting from the fertilization of two eggs in one cycle, feature only the genetic similarity typical of ordinary siblings. Let us use the term “fusion” to refer to the unusual occurrence in which two embryos, fraternal twins, merge into one, a chimera. The chimera has two complete sets of DNA, which together determine (along with prenatal and postnatal environment) the individual’s phenotype—observable characteristics such as eye color, height, and talents.

So, until about two weeks after conception, an embryo can divide into two and, in cases of fraternal twinning, two embryos can fuse into one. Arguably, then, the single-cell zygote is not really a human organism of the sort that we are—that is, one of us at an early stage. Perhaps it is more like living building material out of which one (or more) of us may emerge.

Consider any pair of adult identical twins and any adult chimera. The twins derived from the same zygote. If each human organism originated as a single-cell zygote, then each of twins came into being as the very same zygote. But this is an incoherent result. After all, the twins are numerically distinct from each other, so both cannot be identical to a single earlier zygote. (I make the standard assumption that identity is transitive: If A = B, and B = C, then A = C.) Meanwhile, no chimera could have originated from a single zygote because two such beings were needed to constitute her genome. Surely, she did not originate until fusion occurred; and the twins did not come into being until twinning took place. Neither chimeras nor identical twins could have originated at conception. (As we will see, one contending model of twinning somewhat complicates this claim.)

Now, it does not follow from these points that human beings who are neither identical twins nor chimeras did not originate at conception. Maybe different human beings originate at different stages of development. This is possible, but there is substantial reason to think that no human being comes into existence at conception.

In order to see why, it will be helpful to consider two models of how twinning occurs.11 According to the division model, the more traditional understanding of twinning, a single embryo divides into two embryos roughly equal in size. According to the budding model, a blastomere (cell) leaves the (p.22) original embryo and develops independently. We currently do not know which model is correct; perhaps each applies in some instances. If we originate at conception, what happens to the original human organism in twinning cases? If division occurs, then the original organism exists very briefly before vanishing at the time of twinning; whenever an embryo divides, a human organism of our kind goes out of existence. This coheres with the thesis that, while most human beings originate at conception, identical twins originate at the time of twinning. If budding occurs, on the other hand, the original human organism continues to exist despite losing a cell; that cell, which goes on to develop, would constitute a fresh human organism. So budding entails that one twin originates at conception whereas the other comes into being a bit later. These are implications of the thesis that we—that is, most of us—originate at conception.

Some will find it odd that at least some twins and the rest of us, who are of the same basic kind, should originate at different stages of prenatal development. More important than the possible sense of oddity is a related, deeper concern: that the early embryo is, arguably, less like a single, unified organism than like a colony of organisms loosely conjoined by the zona pellucida (the membrane that keeps the cells together). To develop this idea, we need to bear in mind what an organism is. Organisms are characterized by internal complexity featuring interdependent subsystems and the drawing of energy from the environment to maintain internal organization and resist entropy; and no organism (unlike, say, an organ) is part of a larger biological entity. Organisms can consist of a single cell because even a cell features great internal complexity and interdependent subsystems. Amoebas are such organisms. So is the single-cell zygote. Furthermore, the zygote is human as opposed to, say, feline or canine. The question is whether it is a human organism of the same kind that you and I instantiate.

For the first few cell divisions, according to the present school of thought, the embryo functions less like a single integrated, energy-using unit of the sort we call an organism than like a collection of single-cell organisms contingently stuck together. That is precisely why twinning and fusion remain possible. After the sperm penetrates the egg, their respective sets of chromosomes remain separate for about a day. Contrary to popular belief, fertilization is completed at the two-cell stage.12 Two further divisions in the next couple of days result in an eight-cell embryo. Significantly, each of the eight cells retains the potential, if separated, to produce a human organism like you or me. Thus far there is no specialization of cells to perform distinct tasks; nor is there significant interaction or integration among them. In that sense, they are not functioning as a single, integrated organism. Rather, they are tantamount to a colony of eight contingently joined zygotes.

(p.23) Cell differentiation begins, the argument continues, at the 16-cell stage, as the outer cells begin to transform into what will become the placenta. Division continues as the embryo travels through the fallopian tube, entering the uterus about five days after (the beginning of) fertilization. Two or three days later, it penetrates the uterine wall, establishing connections with the mother’s blood supply. But the middle cells have not differentiated and the embryo could still divide spontaneously into two viable organisms. On day 14 or 15, some middle cells differentiate, and it is now determined which cells will form part of the placenta and which will become part of the fetus. Within a day, in the fetal portion, a column of cells differentiates into the “primitive streak,” the precursor to the spinal cord. Spontaneous twinning is now impossible. There undeniably exists a human organism that functions as a single integrated unit. From the standpoint of the biological view of our essence and identity, this organism is a being of our kind. In other words, it is a being that can grow into—become—what is indisputably a human person.

So, once spontaneous twinning is precluded and every portion of the embryo is differentiated at about two weeks after conception, one of us exists. But maybe we should judge that we emerge earlier, when differentiation begins at the 16-cell stage. Rather than drawing a line marking our time of origination, I tentatively advance two theses: (1) None of us existed, prior to the 16-cell stage, as a completely undifferentiated embryo (a portion of which later differentiates into the placenta); and (2) By the time twinning is precluded and all embryonic parts are differentiated, one of us has originated. Possibly we come into existence somewhat gradually over several days rather than at a single moment.

Having presented this understanding of our origins, from the backdrop of a biological view of our essence and identity, I must acknowledge the respectability of another biologically based understanding of our origins. This alternative receives motivation from the thesis that there is significant integration very soon, perhaps even immediately, after conception. Suggesting this possibility is a report of several experiments involving mouse embryos, the upshot of which is the appearance in mice of cell differentiation as early as the two-cell stage.13 Might human embryos also feature such early, or even earlier, integration and differentiation? Some have argued affirmatively, citing recent scientific sources. One team of authors, for example, has contended that “there must be some kind of exchange of information both within the single-cell zygote and within the multiple-cell zygote….”14 This possibility should not be ignored.

At the same time, even if there proves to be some integration of functioning among various parts of the zygote, it seems fair to ask, “How much integration and differentiation are sufficient for the zygote to be appropriately classified as an organism that can grow into one of us (entailing numerical identity between the zygote and later person) as opposed to an organism that merely precedes (p.24) and furnishes the building blocks for one of us?” After all, there are insufficient integration and differentiation of the zygote/early embryo to preclude spontaneous twinning and fusion. And much of this organism will go on to form the placenta rather than the later embryo. So the facts of embryology alone may not settle the issue of when we originate. Some degree of interpretation may be necessary.

We conclude this discussion of our essence, identity, and origins with the first prong of a tripartite framework for understanding prenatal moral status—and some uncertainty about one end of this prong. I have recommended, and briefly defended, a biological view according to which we are essentially human organisms or animals such that our persistence conditions are biological: the continuation of a single life. For those who accept the biological view, there is uncertainty about our origins. We come into being as pre-sentient embryos, but when exactly? The view I suggest places our origins somewhere between the 16-cell stage and around two weeks post-gestation when twinning is precluded and differentiation is fully established. Another respectable view is that we originate much closer to, and perhaps at, the time of conception.

Sentience, Interests, and Moral Status

On the biological view, we originate within the first two weeks or so of gestation. Do we have full moral status as soon as we originate? We do if we have this status throughout the entirety of our existence. But we cannot simply assume that this is the case, so we need to consider the basis of our moral status. What is it about us, or about any beings with moral status, that confers it? We will begin by considering moral status in general before introducing the possibility of degrees of moral status.

The Appeal to Species

In light of moral tradition and current practices, it may seem natural to judge that the basis of moral status is being human in the sense of species membership. On this view, all and only members of Homo sapiens have moral status—and they have it precisely because they are members of this species. However natural it may seem to think along these lines, this position proves remarkably implausible. For one thing, it is hardly credible that a biological category such as species could furnish the criterion of moral status. If species itself—as opposed to characteristics normally associated with a species—is the basis of moral status, then a person who was not Homo sapiens (say, a member of a now-extinct hominid species, an extraordinary animal, or a space alien) would not qualify. This is hardly plausible. Why should our species be so special? (p.25) Moreover, if membership in a biological grouping is supposed to make all the difference, why species rather than a more general category such as genus, family, or order? Why only Homo sapiens and not any other hominid species, or other primates, and so forth? Or why not a biological grouping more specific than species? For example, one might claim that only male members of our species have moral status. It would be useless to reply that women and girls are just as intelligent, self-aware, and sensitive as men and boys, because the working assumption is that it is a biological grouping, not some associated characteristic, that underlies moral status. And, unlike distinctions of race, distinctions of sex—male and female—have about as much claim to being biologically real as species distinctions do.15 There is no good reason to believe that membership in our species is the basis of moral status.16

The Personhood View

A more plausible thesis is that some characteristic or a set of characteristics associated with our species provides the basis of moral status. The paradigm bearers of moral status, after all, are persons, where persons are understood as beings with the capacity for sufficiently complex forms of consciousness; normal post-infancy human beings are generally assumed to make the grade. Persons, moreover, can be accountable to each other, each person having obligations to other persons. If all and only persons have moral status, then there is a neat congruence between (1) beings who have moral status and (2) beings who bear moral responsibility—moral agents or persons. For these reasons, it is understandable that many people might believe personhood to be the basis for moral status.

Nevertheless, like the species-based view, the personhood view proves deeply problematic. For one thing, it has difficulties with the moral status of human nonpersons. Consider so-called nonparadigm humans—human beings who, due to genetic anomaly or environmental insult, do not have (even potentially) the mental capacities that constitute personhood. No matter how personhood is defined, so long as it is defined in terms of mental capacities that are supposed to distinguish us from cats and dogs, some humans—including the most severely retarded and the most severely demented—will not qualify as persons. Yet it is incredible to think that they, at least those who remain sentient (capable of having feelings), lack moral status; surely they are not fair game for dangerous, nontherapeutic experiments, forced labor, or consumption by cannibals. There must be something wrong, therefore, with any view whose criteria for moral status imply that these individuals fall on the wrong side of the line. Moreover, it is not so clear that the personhood view adequately handles the moral status of ordinary infants. Newborns are not yet persons in the relevant sense, but it seems pretty obvious that they have moral status. To be sure, they (p.26) are potential persons—meaning that in due course they will develop into persons—but so are embryos (at least after twinning is precluded) and fetuses. Only those who hold that these prenatal human beings have the moral status of persons can assert, on the basis of potential, that infants do as well.17 I suggest that the personhood view’s difficulties with nonparadigm humans, and perhaps ordinary infants, reflects an untenable criterion for moral status.

A second major problem for the personhood view is that it seems unable to account adequately for the wrongness of cruelty to animals. Horses are not persons, yet it is wrong to abuse them. Can the wrongness of cruelty to horses (or other sentient animals) be fully explained in terms of human interests? Many horses are legally owned by particular persons, so harming the horse damages someone’s property—which is generally wrong. And many people are upset by cruelty to horses, so to abuse these creatures is to offend these people’s sensibilities. But these points about human interests do not get to the heart of the matter. To abuse a horse is to harm her, greatly, for no good reason—and, intuitively, that fact by itself seems sufficient to explain the wrong. Accordingly, to abuse an animal who was no one’s property, and about whom few if any people cared, would seem wrong, a point that is not accounted for if only persons have moral status. There is compelling reason to believe that at least some nonpersons—some human, some nonhuman—have moral status.

Do all Living Things have Moral Status?

One might conjecture that all living things have moral status. This thesis would avoid the implausible narrowness of the personhood criterion, but at the expense of excessive breadth. In addition to covering persons and clearly sentient animals, it would cover the most primitive animals, whose nervous systems seem incapable of sustaining (conscious) sensations, as well as plants, fungi, and bacteria. Why should anyone believe these insentient creatures have moral status? Given their lack of sentience, in what sense can they be harmed? It is a plausible thesis that only sentient creatures—beings who can experience feelings—can be harmed or benefited and therefore have interests.18

Some argue, however, that all living creatures have interests grounded in biological needs—for whatever is necessary to survive, maintain health, and perhaps reproduce—and that to thwart these interests by damaging or killing the life in question is to harm it.19 The notion of biological needs helps us to understand how something might have interests despite never experiencing feelings such as pain and pleasure. We may say that a flower needs water, sunlight, and soil conducive to growth even though the flower has no mind and therefore does not care about these things. After all, if the biological needs are not met, the flower will wilt and die. “Conditions necessary to sustain life and health” is a tolerably clear standard for determining the interests of a living being.

(p.27) But is this standard plausible? Why should we judge that a lifeform has an interest in remaining alive if that being does not care about its life? It certainly does not take an interest in its life, or anything for which the life is a precondition, because, being permanently insentient, it cannot take an interest in anything. For this reason, it seems that the mindless lifeform has no stake in its own life or anything else.

One might reply that a flower’s life in some sense matters to the flower, because, if it languishes or dies, it loses its physical integrity. But why should that matter—in any sense that can be unpacked in terms of the flower’s interests as opposed to the interests, say, of people who like flowers? If we assert that a flower needs to stay alive to retain its physical integrity, we should also judge that a lawn mower needs to be oiled to maintain its integrity and not break down, and a house needs to be well maintained for the same reason. Yet lawn mowers and houses do not have interests.

One might reply that living things are natural entities whereas lawnmowers and houses are artificial, and that only natural entities can have interests. But this reply is not persuasive. Surely an advanced artificial-intelligence system that (or who) achieved consciousness and feelings would have interests despite being artificial. Moreover, many natural but nonliving systems such as waterfalls, planets, and galaxies have structures that could be considered the bases of their integrity, yet no one would claim that they therefore have interests. Physical integrity or the “need” to maintain structure is not sufficient to ground interests.

Is there another possible basis for saying that plants and insentient animals, as living things, have interests and therefore moral status? One might claim that all and only living things have a naturally given end—to grow and live in a characteristic way—and that this end gives rise to interests. But one might as well say, absurdly, that rivers have a naturally given end to keep flowing, that black holes have a naturally given end to absorb and compress everything within their reach into singularities, and that rivers and black holes therefore have interests in whatever conditions are necessary for them to fulfill their respective ends. These reflections suggest that there is no credible basis for claiming that interests and therefore moral status are grounded in biological needs, independently of conscious experience.

Consider, then, a different possible basis for the claim that all living things have moral status: the view that all life commands our respect, irrespective of whether it has interests and can be benefitted or harmed. Personally, I tend to agree that we should have an attitude of awe or reverence for living things, but I think that this attitude should extend to all of nature. As we just saw, living things are not distinguished from the rest of nature by possession of interests, because many living creatures lack interests. It seems, then, that whatever respect is appropriately directed to living things as such, or to living things as (p.28) part of nature, must be due to some value-based factor (perhaps aesthetic) other than moral status.

The Sentience View

Our reflections on moral status suggest that (1) not only persons and human beings have moral status, and (2) not all living things have it. Some nonhuman nonpersons have moral status because they can be affected by the actions of moral agents in ways that they, the nonpersons, experience and presumably care about insofar as their quality of life is affected. In that way, they can be benefited or harmed. The reason why not all living things have moral status is that many of them have no subjective experience and therefore no quality of life and no interests. These observations strongly suggest a basis for moral status that has been embraced by many theorists and that I have defended at length in another place: the sentience view.20 According to this view, sentient beings—who by definition are capable of having feelings—are precisely those beings who have interests and can be harmed or benefited (in any sense that might be morally important); all and only such beings are of direct moral concern. That I help a person affects her interests and, for this reason, matters morally. That I abuse a horse thwarts her interest in avoiding suffering and, for this reason, matters morally. That I kill bacteria with an antibiotic or smash a rock does not affect the interests of the bacteria or rock, which lack interests, and does not matter morally—unless my action importantly affects the interests of persons or other sentient beings.

So the sentience view has several advantages. It has no difficulty explaining why not only human persons, but also (sentient) human nonpersons have moral status and should not be abused or exploited. It straightforwardly explains the wrongness of cruelty to animals who are capable of suffering from cruelty. (One cannot really be cruel to an amoeba or plant.) More generally, it squares very well with our intuitions about morally important harms and benefits. And, if there is something right about the personhood view of moral status, the sentience view can partly accommodate it: One can hold both that all sentient beings have moral status and that persons have special moral status.

In virtue of its conceptual tie to interests, the sentience view connects plausibly with two leading types of ethical theory. First, it connects well with consequentialism, the general approach according to which the rightness or wrongness of our actions is a function of their consequences. Such theories need an account of individual well-being to cash out how the consequences of actions affect us in ways that matter. The concept of an interest—as a component of well-being—serves this role well. Second, the sentience approach fits well with rights theories, according to which individuals have certain rights, understood as moral protections or claims that are not (or only rarely) to be (p.29) violated even in pursuit of good ends. To what do individuals have rights? This a matter of great dispute, of course, but one safe claim is that A can have a right to X only if A has an interest in X. We might add that it is our most vital, central, or important interests that provide the basis for rights. In any case, interests provide the content or conceptual material for a well-developed account of rights just as they provide the content for a well-developed account of individual well-being.21

Thus, the second prong of the recommended three-part framework for understanding prenatal moral status is this thesis: Beings with moral status are precisely those beings who have interests, which are closely connected with sentience. But what is the precise connection between sentience and interests? Some who adopt the sentience view hold that sentience is necessary and sufficient for having interests and therefore for having moral status,22 but I’m not so sure. Based on our reflections, we may confidently assert that sentience is sufficient for interests and moral status (a claim that implies that late fetuses have moral status). But is sentience necessary? One might assume so, considering such cases as paintings, planets, plants, and protozoa. But what if a non-sentient being has the potential to grow into a sentient being? Might it have an interest in realizing this potential, despite being presently unable to appreciate it? We must consider this possibility.

Potential Sentience and Time-Relative Interests

Persons have moral status and so do other sentient beings. What about beings that are potentially sentient beings or persons? Obviously, it does not follow from the fact that A is potentially X, which has a certain status, that A already has that status. For example, it does not follow from the fact that I am potentially Professor Emeritus that I already have this status, on the basis of which I can claim certain privileges such as commandeering an office without doing any work. In discussions of moral status, liberal scholars often make much of this logical point, but doing so provides little help in understanding the real issue concerning potential.

The real issue concerns not potential in general, but the prenatal human being’s potential—for sentience, personhood, or both. Taking our cue from the sentience view, and accepting that a fetus ordinarily becomes sentient late in pregnancy (probably no earlier than the beginning of the third trimester23), we can reformulate our question as follows: Does the pre-sentient human being already have interests on the basis of what it can become? Can it have interests before it has a nervous system capable of producing states of consciousness?

Many thinkers, especially liberals, are likely to answer in the negative and to endorse: (p.30)

The View that Sentience is Necessary. Pre-sentient fetuses cannot have interests. To have an interest is to have a stake in something; it is to be susceptible of being made better off or worse off. But to have a stake or to be so susceptible requires that one care about certain things, have preferences, or at least have sensations that are experienced as pleasant or unpleasant—in other words, a quality of life. True, we sometimes have interests in things that we don’t care about, prefer, or even know about—for example, an interest in learning that one has an inheritance from a distant relative. But in such cases, our interest is ultimately a function of things we do care about, prefer, etc. such as having enough money and obtaining what is legally ours. Having no conscious life at all, a pre-sentient fetus has no carings or concerns that can be the basis of such interests.

This is a powerful argument. I think it may be correct. But I am not at all sure, because there is another way of looking at the matter that strikes me as equally cogent:

The View that Potential is Sufficient. Imagine that a baby has just been born to loving parents who are well-positioned to care for her. The baby has one striking anomaly but is otherwise perfectly healthy: Due to a problem with a portion of her brainstem, she has never been conscious. She can gain consciousness only if a simple medical procedure is performed. With this procedure, the baby can acquire a mental life that will very likely develop normally. Undoubtedly, it is in this baby’s interest to have the procedure performed so that she can become conscious. Some believe that any conscious life—or sentience24—is intrinsically valuable (assuming the subject isn’t terribly miserable). Others believe that, while sentience automatically gives rise to interests, it is only the conscious life of a person that is intrinsically valuable (assuming, again, the quality of life isn’t too low). Either way, this baby has the relevant potential. It would be a terrible loss for her not to have the procedure and thereby lose the riches of the life she can have. In the same way, a fetus with the same potential has an interest in remaining alive. Whatever it is about human lives (of tolerable quality) that makes them so valuable for their subjects, a fetus stands to gain such a treasure if she is able to live. It is irrelevant that the fetus, like the baby, cannot now appreciate this possible future of value. It is relevant that the fetus is a creature whose endowment naturally inclines it to develop in such a way that it will eventually achieve sentience and later the mental life of a person.

This, too, is a powerful argument—about as powerful, I think, as the View that Sentience is Necessary.

(p.31) Political liberals tend to overlook or underestimate arguments along the lines of the View that Potential is Sufficient. One reason, surely, is a desire not to threaten a woman’s presumed right to abortion. Another reason, I think, is a tendency to trivialize the notion of potential. Critics of pro-life arguments often claim that appeals to potential are illegitimate because they would absurdly legitimate claims of moral status on behalf of sperm, ova, and somatic cells (which, in principle, can be cloned).25 But the relevant potential is not mere logical possibility or even factual possibility. Rather, it is natural potentiality, which is possessed by members of a kind of thing whose natural development involves realizing the relevant potential. Another error is to think that for A to be potentially a B, it is enough that A can become a piece of a B. I have heard it said that a piece of dust has the potential to become a statue because the former could end up being combined with other particles in the construction of a statue. Not much more plausible is the claim that a sperm cell could become one of us. No, a sperm cell can combine with an ovum to constitute a new entity with a full complement of human genes; depending on the correct view of our origins as discussed earlier, either this entity can become one of us or it can provide the material from which one of us emerges when there is sufficient integration and differentiation. To do justice to pro-life thinking, we need to appreciate that its proponents have a special sense of “potential” in mind when they appeal to a fetus’ potential. I do not mean that such appeals should be accepted uncritically, just that they should be understood on the terms of those who advance them rather than misconstrued into straw targets that are easily knocked over. Properly understood, the appeal to a fetus’ potential is part of a powerful argument, as displayed in the View that Potential is Sufficient.

Does a pre-sentient fetus have an interest in staying alive, an interest based on its potential to enjoy a valuable sort of existence? I just do not know. If the answer were clearly No, then prenatal status would be greatly clarified: Fetuses would lack moral status until they achieved sentience. This result would greatly simplify the ethics of abortion and embryo research. But suppose, now, that the answer is Yes: The pre-sentient fetus has an interest in remaining alive. For the sake of argument, I will assume that this view is correct. I will argue that, even on this pro-life-friendly assumption, there are powerful grounds for denying that this fetal interest supports a fetal right to life.

These grounds stem from the time-relative interest account (TRIA) of the harm of death.26 This account is motivated by observations about what we believe about the harm of death in different cases. How much is one harmed by death? It might seem logical to assume that the magnitude of the harm of death should be determined solely in terms of how much good life one loses (where “how much good life” takes into account both quality and quantity of life). From this whole-lifetime perspective, the younger one is, other things—quality (p.32) of life, expected lifespan—being equal, the more prudential value or good one loses from death. This implies that a pre-sentient fetus is harmed a bit more by death than is an infant, who is harmed more by death than a 10- or 25-year-old. But precisely the opposite seems correct: Other things equal, death seems to harm a 10- or 25-year-old more than it harms an infant, and an infant at least a bit more than a pre-sentient fetus. We may well regard the death of the pre-sentient fetus as unfortunate for it; people’s intuitions seem to vary on this point (as reflected in the clash between the View that Sentience is Necessary and the View that Potential is Sufficient). Most of us would probably consider the infant’s death a major misfortune for him. By contrast, we regard the death of a 10- or 25-year-old as utterly tragic for the victim—and in a way that fetal and infantile death cannot be for its victim. Note: The claim is not that others are likely to mourn an infant’s death less than that of a 10- or 25-year-old, a claim that can be explained by the family’s having invested more, emotionally and otherwise, in the older victim. The present claim concerns the harm of death to the one who dies.

So the view that the harm of death is simply a function of total lost prudential good generates implausible implications about the comparative harm of death in different cases.27 Our reflections suggest that the harm of death is a function not only of lost opportunities for valuable life, but of another factor as well. Simply stated, the second factor according to the TRIA is the extent to which one is psychologically connected with one’s possible future life.

The TRIA discounts the harm of death to its victim, at the time of death, for any weakness in the psychological unity that would have connected the victim at that time with herself in the future. In this way, it denies that numerical identity—being one and the same individual over time—is the only relation that an individual has to herself over time that matters prudentially; psychological unity over time also matters. The degree of psychological unity in a life, or over a stretch of time, is a function of (1) the amount of internal reference between earlier and later mental states (e.g., memories of past experiences, anticipations of future experiences), (2) the proportion of the subject’s mental life that is sustained over the stretch of time in question, and (3) the richness of the mental life.28 When the psychological unity that would have bound an individual at the time of death to herself in the future, had she lived, is weak, death matters less prudentially—for that individual—at that time.

The TRIA steers a plausible middle course between two polar positions, as we can see by considering the death of an infant. On a desire-satisfaction account of the harm of death, one is harmed by death only if one desires to remain alive, or not to die, but such a desire seems impossible within the primitive conceptual repertoire of infants. So the desire-satisfaction account implies, implausibly, that normal human infants are not harmed by death. Contrast the (p.33) view that construes the harm of death in terms of lost opportunities for valuable future experiences, while assuming a whole-lifetime perspective. This view implies that death typically harms an infant more than a child, adolescent, or young adult. This, at least to my mind, seems highly implausible. Steering between the implications of these two views, the TRIA suggests that death significantly harms the infant but not nearly as much as it harms the child, adolescent, or young adult. In addition to delivering this plausible verdict, the TRIA explains precisely what factor justifies discounting the disvalue of death, in cases like the infant’s, as it would be understood from a whole-lifetime perspective: degree of psychological unity over time. The desire-satisfaction view is correct that caring about or appreciating (and therefore desiring) one’s future is relevant to the harm of death, but wrong that one who does not appreciate or desire one’s future life loses nothing from having that future snatched away. The whole-lifetime approach is right that appreciating one’s future is not necessary for having a stake in that future, but incorrect in thinking that such appreciation is irrelevant to the harm of death. The TRIA, meanwhile, gets right what the polar views get right while avoiding their errors.

So the third prong of my tripartite framework for understanding prenatal moral status is the TRIA. How does it help us? Recall that we encountered a stalemate between the View that Sentience is Necessary and the View that Potential is Sufficient. For the sake of argument, we assumed that the latter was correct: The pre-sentient fetus has an interest in remaining alive. The significance of the TRIA is to show that any such interest must be very weak in comparison with a person’s interest in remaining alive. Persons, in anything like ordinary circumstances, have a profound stake in staying alive, continued life being a foundation for the experiences, relationships, accomplishments, and so on that make one’s life valuable for one. Persons also, more or less by definition, have deep psychological unity over time. Infants have much less, sentient fetuses precious little. To put it starkly, pre-sentient fetuses have no psychological unity with future stages of their lives because, prior to sentience, they have no psychology. Thus, even if we grant that the pre-sentient fetus has an interest—a prudential stake—in remaining alive, the TRIA implies that this interest is very weak. What the pre-sentient fetus stands to lose is enormous, but its lack of psychological life and therefore the lack of psychological unity between itself now and possible future stages of itself entail a weak time-relative interest in remaining alive. (A less generous interpretation, from the standpoint of the TRIA, would be that the pre-sentient fetus’ lack of psychological life implies that this creature cannot have any time-relative interest in its future; in mathematical terms, where one of the factors is 0, the product must be 0.) So death is not a great harm for the pre-sentient fetus, suggesting little basis for attributing a right to life to it. This assumes, naturally, that a right to life depends on a (p.34) strong interest in remaining alive, but that assumption seems entirely reasonable. Rights are grounded in strong interests.

This completes my defense of a tripartite framework for understanding prenatal moral status: (1) a biological view of our numerical identity, essence, and origins; (2) a view about the connections between interests, sentience, and moral status; and (3) the TRIA, which bears importantly on the moral status of pre-sentient fetuses. If this framework is substantially correct, it has implications for the ethics of abortion and embryo research that are at least moderately liberal, for it would imply that embryos and presentient fetuses have very little (if any) moral status in comparison with persons. But it is one thing to defend a framework, especially when, as here, the defense is relatively brief; it is another thing to make a compelling case for the framework. In an effort to make my case more compelling, I will turn to what I believe to be the strongest arguments in favor of a pro-life position and attempt to show that they are unpersuasive.

Replying to the Strongest Pro-Life Arguments

The Future-Like-Ours Argument

The Future-Like-Ours Argument (FLOA) has become the most celebrated anti-abortion argument in the English-speaking philosophical literature. Its celebrity is due, I think, partly to its quality, but also to the fact that its chief proponent, Don Marquis, is a politically liberal atheist who presumably lacks a conservative or religious agenda.29 As we will see, the FLOA has much in common with the View that Potential is Sufficient.

The FLOA proceeds as follows. Why is it wrong, in ordinary circumstances, to kill paradigm persons like you or me? The fundamental reason is that killing us would deprive of us valuable futures, which would include all our personal projects, enjoyments, meaningful activities, and other valued experiences that are plausibly believed to make human life valuable in ordinary circumstances. This account of the wrongness of killing persons explains why we regard killing as such a terrible crime; it also accommodates our belief that death ordinarily harms the individual who dies. Meanwhile, it avoids certain difficulties of other leading accounts of the wrongness of killing persons, such as the desire-satisfaction account. Further, this approach plausibly implies that, since killing infants would (ordinarily) deprive them of valuable futures—futures like ours—infanticide is (at least normally) wrong. By contrast, accounts that base the wrongness of killing persons on their special moral status as persons—beings with the capacity for relatively complex forms of consciousness—struggle to explain the commonsense judgment that infanticide is normally wrong.

(p.35) Now, the FLOA continues, consider fetuses (using the term broadly to refer to prenatal human organisms at any stage of development). A human fetus is an individual that can normally, if permitted to live, develop into a person who has the sorts of experiences we value so highly. So the fetus has a future like ours, making abortion comparable to the killing of paradigm persons: wrong in at least ordinary circumstances. This is a substantial moral argument against abortion that is free of religious assumptions, equivocation on the moral and descriptive senses of such terms as “human being” and “person,” and question-begging appeals to potential.

When exactly does the FLOA first apply? In asserting that the fetus has a future like ours, the argument assumes that the fetus can develop into a paradigm person and, conversely, that each of us was once a fetus. So the FLOA applies whenever a human organism of our kind comes into existence. Our earlier discussion of our origins tentatively suggested that we come into existence at least a few days and no later than about two weeks after conception; it also acknowledged the respectability of the view that we come into existence as early as conception. If the FLOA is sound (whenever it first applies), it has far-reaching implications for prenatal moral status—and, if the latter is the primary consideration in the ethics of abortion, then for that issue as well. (Depending on when it first applies and what sort of embryo research is in question, it may or may not have far-reaching implications for the ethics of embryo research.)

Acceptance of the TRIA of the harm of death suggests that the FLOA is unsound. For the latter assumes that numerical identity is the sole basis for prudential concern for what most matters in our continued existence, suggesting that evaluation of a fetus’ future should be made from a whole-lifetime perspective. From this perspective, of course, abortion (ordinarily) entails an enormous loss, the loss of a future like ours. On this basis, the FLOA infers that the fundamental consideration underlying the wrongness of killing paradigm persons also applies in the case of abortion. But in the previous section it was argued that the TRIA was a better account of the harm of death than the view presupposed by the FLOA. Fetuses, especially pre-sentient fetuses—the vast majority of fetuses aborted—have only weak (if any) time-relative interests in remaining alive. Thus, even if fetuses have a future like ours, the FLOA does not show that killing them is comparable to killing a paradigm person, or that fetuses have a right to life.

The Appeal to the Practical Necessity of Early Moral Protection

A second argument, which I call the Appeal to the Practical Necessity of Early Moral Protection, contends that we have full moral status for the full duration of our existence.30 Coupled with the biological view of our essence and identity, (p.36) it entails that the prenatal human being has full moral status from the moment of its origination (which, for the sake of simplicity, we will here assume to be conception). With the help of a third assumption—that a fetus’ having full moral status would entail that it may not be intentionally killed—the Appeal to the Practical Necessity of Early Moral Protection supplies a substantial moral argument against abortion and (on the conception view of our origins) against embryo research.

The argument begins by noting that each of us who is able to consider this issue rightly holds herself to have full moral status. It then argues that it would be irrational for any of us—any person—to deny that she had full moral status at some earlier point in her existence. Why would this be irrational? As Alfonso Gomez-Lobo explains the idea, “Inviolability [full moral status] now without inviolability at previous stages entails that I would risk not being alive now. It would have been morally permissible to destroy me in the past so that my later inviolability would be worthless.”31 The idea is that the claim of present inviolability and the denial of earlier inviolability are somehow contradictory, and that, to avoid contradiction, we must maintain that every person is inviolable for the duration of her existence. But what kind of contradiction is supposedly involved? There is no logical contradiction, because the conjunction of “I now have full moral status” and “At some earlier time during my existence I did not have full moral status” cannot be resolved into the form of P and not P, as any logical contradiction can. Rather, the alleged contradiction, or absurdity, is practical: It would be absurd from the standpoint of practical reasoning to advance this conjunction of claims. Why? It is absurd, allegedly, to say that it would in the past have been permissible to destroy a being who later, if permitted to live, would have a moral status that makes such destruction impermissible. Yet, when I focus on this supposedly absurd statement, I perceive no absurdity. It may help in this context to highlight the distinction between what is permissible from an ethical standpoint and what is desirable from a prudential standpoint.

Consider an analogy. Marcie is glad to be alive, despite a rocky childhood. Her childhood was rocky because her parents lacked any good sense about how to raise children. In fact, Marcie can honestly say that her parents were far too immature to marry responsibly. They should not have married and certainly should not have had children. Yet, if they had not done these things, Marcie would not exist. So she judges that her parents did something wrong in marrying and having children, yet she is glad, self-interestedly, that they did so because she is glad she’s alive. Is that contradictory or absurd? Not at all. From an impartial, ethical standpoint, Marcie judges that her parents failed to meet appropriate criteria for when people ought to feel free to marry and have children. From a partial, self-regarding standpoint, she values their actions as (p.37) means to her current existence, which she values. While her position is likely to be psychologically awkward, it is certainly coherent and free from absurdity.

Returning to the central issue, can I consider myself presently inviolable and deny that I was inviolable in the earliest stages of existence? Yes, I can. My present full moral status entails that it would be wrong—extraordinary circumstances aside—for another to kill me. If I lacked this status as a fetus, then it would have been permissible for my mother to have an abortion, which would have killed me. I am glad that she did not because, like Marcie and most other people, I am glad to be alive. But I do not deny that it might have been permissible for me to have been aborted—say, if my mother had realistically judged that going to term and having another child would pose an overwhelming burden on her or the family. Partially and self-interestedly, I am glad that I was permitted to develop into a being with full moral status. But gladness does not entail entitlement. Impartially, I grant the possibility that this being could have been permissibly destroyed prior to acquiring full moral status. Moreover, present entitlement not to be killed does not entail past entitlement to enjoy conditions necessary for me to enjoy my present entitlement. Stated differently, present entitlement not to be killed does not entail present entitlement not to have been killed. So there is no contradiction or absurdity. Perhaps any feeling of a practical contradiction here trades on a conflation of ethical and prudential standpoints, which are distinct perspectives whose appropriate criteria of evaluation sometimes diverge.

The Appeal to Our Essence and Kind Membership

The third of what I take to be the three most powerful pro-life argument appeals to our membership in a kind defined by our essence, and the natural potentiality of members of this kind. The argument can be reconstructed as follows:

  1. 1. You and I have full moral status.

  2. 2. We have this status because of our essence, which determines our kind.

  3. 3. We are essentially human organisms—a kind endowed (essentially) with a rational nature.

  4. 4. Human organisms come into existence at conception.

Therefore,

  1. 5. We have full moral status from the time of conception.32

Like the Appeal to the Practical Necessity of Early Moral Protection, the present argument begins with the uncontroversial assumption that you and I, and (p.38) paradigm persons generally, have full moral status. It further assumes a version of the biological view of our essence and identity, but adds the claim that human organisms are essentially rational. As for our origins, it assumes that we come into existence at conception—an assumption I will grant here, again, for simplicity’s sake. The main vulnerabilities in this argument are two assumptions: the second part of 3 (call it 3b), the claim that our kind is essentially, therefore necessarily, rational; and premise 2, that our moral status is grounded in our essence and kind membership.

These two assumptions are importantly connected. Suppose the argument dropped 3b, asserting that we are essentially human organisms without claiming that we are essentially rational. One who, like me, accepted this assertion could hold that we are only characteristically rational, allowing that some human beings are not rational. Moreover, one could claim that this characteristic rationality is a trait that develops, gradually, after birth. It is certainly not true, one might argue, that we are rational at the time of conception—when we do not even have brains.

The Appeal to Our Essence and Kind Membership needs the assumption that all human organisms are rational by nature, and therefore rational regardless of how cognitively abnormal or damaged, and rational from the moment of conception. Without this assumption, given the traditional understanding (which pro-life thinkers accept) that personhood is necessary for full moral status, there would be no basis for saying that all members of our species—including the unborn and the severely cognitively disabled—enjoy this status. Persons are rational animals, the thinking goes, revealing its Aristotelean flavor. They are part of nature but their endowment, which includes rationality, distinguishes them from the rest of nature.

While it will strike many people as obvious that fetuses are not rational, since they evidently lack any capacity to reason, proponents of the present argument claim that fetuses actually have the capacity, in the form of natural potentiality, to reason. This capacity is part of their nature and simply needs to unfold with time. As Patrick Lee puts it, “… there is a sense in which human embryos and fetuses also have the capacity for higher mental functions…. They are members of a natural kind—a biological species—whose members, if not prevented by some extrinsic cause, in due course develop the immediately exercisable capacity for such mental functions.”33 So potentiality is involved. It is the potentiality to develop a certain kind of mental life that prominently includes rationality.34

Many will remain skeptical that fetuses, infants, or those human beings whose cognitive impairments prevent them from manifesting the mental life characteristic of persons are nevertheless rational. Doubts may be most acute in the case of the latter group because they cannot develop in such a way as to be (p.39) able to reason, even if they are of a natural kind whose members characteristically develop this ability. But let us set aside the dubious claim that human organisms are essentially rational.35 Let us focus instead on premise 2, that our moral status is determined by our essence and kind.

Our moral status must be based on something. Premise 2 states that the basis is membership in a kind determined by our essence. But other possible bases have been suggested as well. For example, it was argued earlier that the basis for moral status is the possession of interests. Now, a pro-life position can accept that having interests is necessary for moral status, because it can maintain that fetuses have an interest in remaining alive, being healthy, and so forth. But, as we have seen, the (standard, unmodified) sentience view asserts that only sentient beings have interests, potential sentience being insufficient. If this is correct, then a human being’s moral status is not determined by kind membership but by a particular property: the (present) capacity to experience feelings. Another view, of course, is that personhood—in the sense requiring an already developed capacity for complex forms of consciousness—is necessary and sufficient for moral status, or perhaps full moral status. So it is far from self-evident that one’s moral status is determined by one’s essential kind. As far as I can see, the Appeal to Our Essence and Kind Membership lacks any strong grounds for its assumption about the basis of moral status. My sense is that this third important pro-life argument is driven by (1) uncritical acceptance of the idea that personhood is necessary and sufficient for moral status—an idea with roots in the Biblical notion of God creating all nonhuman creatures for disposal by humanity—and (2) conflation of personhood with membership in Homo sapiens. The conflation may be due to the discredited idea that natural kinds are susceptible to classical definitions—in this instance, that humankind can be defined as “rational animal” (see note 35). Whether or not my speculations are correct, the Appeal to Our Essence and Kind Membership cannot be considered sound because it has provided no compelling reason to accept its second premise about the basis of moral status. And again, its assumption that rationality is essential to our species is highly dubious.

Does the Good Samaritan Argument Clinch the Case?

None of the three strongest pro-life arguments has proved persuasive. This strengthens the case for the earlier-defended framework for understanding prenatal moral status, which can withstand the challenges of these arguments. At this point, with an eye on the abortion issue as opposed to embryo research, I will consider whether the so-called Good Samaritan Argument (GSA) clinches the case for the moral permissibility of abortion in at least a broad range of cases. What is most striking and novel about this argument is that it grants to (p.40) the pro-life position the seemingly pivotal assumption that the fetus is a “person” in the moral sense: a being with full moral status and a right to life.36

So let us assume, for the sake of argument, that the fetus has a right to life in the same sense that you and I do. It may appear to follow from this assumption that abortion is, perhaps with some exceptions, morally impermissible. The GSA, however, contends that abortion is permissible in a wide range of cases (not just a few exceptional ones) even if the fetus has a right to life. Introduced by Judith Thomson, the GSA has recently been developed and refined by David Boonin, whose discussion I take to be representative.37

In the renowned thought-experiment, you awake and find yourself in a hospital, hooked up to a violinist who can survive his kidney ailment only if you, who alone have the right blood type, remain in the hospital bed for nine months with the violinist attached to you. He unquestionably has a right to life. To unplug him in less than nine months will guarantee his death. Nevertheless, it seems obvious that you have no obligation to undergo such a great burden to save his life. While being such a good Samaritan would be highly praiseworthy, going to such lengths to assist another person (whom one did not consent to assist) is well beyond the call of duty. Importantly, it is not that your rights to liberty and bodily integrity trump his right to life. Rather, there is no conflict of rights because his right to life does not encompass a right to use your body. The GSA contends that, even if the fetus has a right to life, unwanted pregnancy is relevantly similar to the situation involving you and the violinist. Terminating pregnancy with abortion, like disconnecting the violinist, is permissible.38

Is this argument successful? Surely it is permissible to disconnect the violinist, so critics of the GSA must argue that pregnancy and the imagined hospital scenario are dissimilar in at least one morally important respect. I will argue that several considerations—responsibility, the parent-child relationship, and the killing/letting die distinction—are, in combination, sufficient to cast significant doubt on the GSA. I argue not that the GSA fails, but that it is uncertain that it succeeds, except in a small range of cases.

Let us begin by considering the matter of responsibility. In non-rape cases—that is, where a woman has had intercourse voluntarily and becomes pregnant—she (along with the biological father) is responsible for the situation in which the fetus needs life support. Boonin helpfully distinguishes two senses of responsibility: (1) A is responsible for the fact that B exists, and (2) A is responsible for the fact that, given that B exists (anyway), B needs A’s help. In the violinist case, you are not responsible in either sense; similarly with pregnancy due to rape. In non-rape cases, a woman is responsible in sense (1)—she helped to create the fetus—but not in sense (2), because the fetus cannot exist independently of the woman’s assistance.

(p.41) Boonin next presents a case in which A is responsible for B’s existence in sense (1) in a different way—by extending rather than starting a life—but not in sense (2).39 Imagine that you, B’s physician, saved his life seven years ago in the only possible way: by administering a medication that cured his disease but would predictably cause kidney failure seven years later. Now, as expected, you alone can rescue him from his present ailment by giving him the use of your kidneys for nine months. Surely you are not obliged to accept this burden. You are responsible for B’s existence, because you saved him, but not for his neediness given that he exists, since he could not have lived to the present day in a non-needy state. According to Boonin, “where you are responsible in sense (1) and not in sense (2) for the fact that another now stands in need of your assistance … the individual in need has not acquired the right to your assistance.”40 Appeals to responsibility, he argues, do not prevail in non-rape cases and, of course, they do not apply in rape cases. They therefore do not refute the GSA.

But let us consider responsibility further, together with the killing/letting die distinction and the parent-child relationship. In discussing the GSA, McMahan helpfully states that “it is very hard to believe that it is permissible to kill one’s own child in order to avoid the burden of providing the aid that one has caused it to need.”41 Again, we assume for the sake of argument that the fetus, like a child, has a right to life. And, except for hysterectomy and hysterotomy—rare procedures that pose special risks to the pregnant woman—current methods of abortion involve killing the fetus. By contrast, unplugging the violinist seems to be a case of letting die rather than killing. Moreover, though the pregnant woman is not responsible for the fetus’ neediness in sense (2)—it is false that the fetus would have existed even if she had not caused its neediness—she and the biological father have through voluntary action caused the fetus to exist in a needy state, whether or not they conscientiously attempted to avoid this result by using birth control. Let us focus on a sense of responsibility (one compatible with either of Boonin’s two senses) that may help us get to the heart of the matter: A is responsible for the fact that B exists in a state in which B needs A’s assistance. In non-rape cases, one is responsible in this sense, which arguably grounds a right to assistance—or, more minimally, a right not to be killed—at least when B (who is assumed to have a right to life) is one’s own child.

Suppose that you, the physician in Boonin’s case, had seven years ago saved the patient, who now, as expected, cannot survive without the use of your kidneys for nine months. Assume that the burden to you is roughly comparable, overall, to the burden of a typical pregnancy in terms of nausea and other discomforts, impact on mobility, and so on: very substantial but less than confinement to bed the entire time. Finally, assume that the patient is your own child. In Boonin’s presentation of the case, which indicates no special relationship beyond that of doctor to patient, you do not seem obligated to give the patient (p.42) use of your kidneys. But you may well have such an obligation if the patient is your child. A fortiori, it is dubious that you may kill your child in order to avoid the burden, comparable to a full-term pregnancy, of providing assistance that you have caused her to need.

But there is an ambiguity in the phrase “your child.” It might refer to someone who is socially your child: You are raising him or, if he is grown up, have already raised him. The present challenge to the GSA is strongest if one imagines a parent-child relationship of this sort. But “your child” might refer to someone who is only biologically your child. Perhaps you gave him up for adoption immediately after he was born, or maybe you just furnished gametes for his conception. The biological sense of “your child” seems more applicable in the thought-experiment because, when abortions are sought, a biological mother does not want to become a social mother to the individual in question. But it is not clear that you have any special responsibility at all to someone who is merely your biological child. Let us assume that the individual in question is your biological child, whom you gave up for adoption and with whom you never had any relationship—except for serving as his one-time doctor and saving his life seven years ago. May you let him die in order to avoid a pregnancy-like burden that saving him again would impose on you? I am fairly confident that you may. May you kill him if doing so is the only possible way to avoid this burden (admittedly a very hard circumstance to picture)? It is not at all clear to me that you may. True, you did not want biological parenthood, and you refused social parenthood, but maybe just bringing the kid into the world (even accidentally) through voluntary action is enough to entail that you may not kill him in order to avoid burdens comparable to pregnancy.42 Here I have in mind burdens that are typical of pregnancy. If the burden would include, say, your death (as with some pregnancies), I would judge that you may, in self-defense, kill the individual, even if he is innocent of any intention to kill you.

Here, therefore, is a reply to the GSA based on appeals to responsibility and one or two further considerations. If allowing one’s (merely) biological child to die in order to avoid the burden one has caused her to need is sometimes permissible when killing would not be, as I tend to believe, then this reply to the GSA invokes both the (biological) parent-child relationship and the killing/letting die distinction. If so, it will not apply to methods of abortion that do not involve killing. And, again, there can be no appeal to responsibility in rape cases. So a successful reply to the GSA along these lines would mean that the GSA could justify abortion only in (1) rape cases, (2) any other cases in which a girl or mentally compromised woman is not responsible for being pregnant (as discussed later), (3) cases in which the mother’s life is at stake (killing being permissible in self-defense even if the threat is innocent), and, if I’m correct, (4) cases of abortion that do not involve killing. Because these types of cases constitute such (p.43) a small proportion of abortions,43 this result would largely defeat what is commonly believed to be a successful argument for the permissibility of abortion.

My cautious conclusion is not that the GSA clearly fails to justify abortion in a wide range of cases, but that there are very substantial reasons to doubt that it succeeds. This suggests that the GSA does not clinch the case for the permissibility of abortion. So I will not rely on the GSA in countering the pro-life position. The tripartite framework, especially the TRIA, remains crucial to my case for the permissibility of abortion.

Reasonable Doubts and Pluralism Regarding Prenatal Moral Status

On the basis of my tripartite framework and critique of leading pro-life arguments, I have defended the following basic picture of prenatal moral status. First, because sentience is sufficient for having interests and moral status, the sentient fetus has moral status. Again, sentience is likely to emerge no earlier than the beginning of the third trimester (which is later than the vast majority of abortions are sought and, of course, long after embryo research is possible). Whether the sentient fetus has full, or only partial, moral status depends on whether sentience is sufficient for full moral status. Regarding the beginning of our existence, I have somewhat tentatively argued that we originate somewhere between the 16-cell stage and about two weeks of gestation. If so, then the zygote and early embryo (by which I mean an embryo prior to the origination of one of us) has no significant claim to moral status. Such beings have no intrinsic properties such as sentience or complex forms of consciousness that lay claim to moral status. Nor can they develop into—become—such beings because they are not numerically identical to any sentient being or person of the future. What about late (post-origination) embryos and pre-sentient fetuses? I have argued that these human organisms have little or no moral status. The argumentative road therefore may appear smoothly paved for liberal views, and no other views, on abortion and embryo research.

Actually, I think not. I have defended an outline of what I believe to be the most defensible position on prenatal moral status. But I may have erred in some of my reasoning. Or, even if I have not erred—even if I’m right that the position I have outlined is the most defensible—there may be another position or several other positions that are nearly as defensible. I am very confident that the position I have sketched is a reasonable position. I am not confident that no alternative position is reasonable. In fact, I am pretty sure that this is not the case. Unlike nearly every philosopher I know who holds a liberal view on prenatal moral status and/or the ethics of abortion and embryo research, I believe that a broadly pro-life approach remains standing as a reasonable option.

(p.44) To be sure, there almost certainly will have to be some exceptions to the asserted impermissibility of abortion for the pro-life approach to remain reasonable. For example, based on the plausible claim that one may intentionally kill a person, even an innocent person, as a last resort to save one’s own life, I suggest that an exception must be made for cases in which the woman’s life is at stake. Moreover, because someone impregnated through rape bears no responsibility for her situation, an abortion in rape cases seems justified—assuming one may kill the violinist, as painlessly as possible, if there is no other safe way to free oneself from him. But a reasonable pro-life view need not admit many exceptions. It might, for example, embrace the conception view of our origination, which would suggest that even zygotes and early embryos have a right to life. Then again, it might not; it might claim a later time for our origination and for that reason prove more permissive about destroying zygotes and early embryos. In any case, I maintain that a broadly pro-life view is a reasonable option.

Let me elaborate on why I think so, notwithstanding my defense of the tripartite framework. Certainly, the biological view of our essence, identity, and origins does not block a pro-life view—which, indeed, presupposes some version of the biological view, even if there is room to quibble about details. Nor does the view that sentience is closely tied to moral status preclude a pro-life view. Nearly all fetuses are potential sentient beings and potential persons. Finding the View that Potential is Sufficient to be about as compelling as the View that Sentience is Necessary, we left open the question of whether having the potential for sentience—or personhood—entailed an interest in surviving so that one could become sentient or realize the life of a person. Such an interest could be the basis of a right to life. For this reason, the TRIA of the harm of death was a crucial part of my framework: Without the TRIA, pre-sentient fetuses could be judged to have full moral status based on their potential (with exceptions, naturally, for those lacking the potential).

Although I believe that the TRIA makes more sense than competing accounts of the harm of death, I also believe there is room for reasonable disagreement on this point. The TRIA is motivated by various intuitions about the comparative magnitude of the harm of death in different cases. These intuitions are not shared by everyone. Even if they were—or should be—shared, perhaps there is a way of accounting for these intuitions that is consistent with a pro-life view. Moreover, the TRIA faces its own challenges. For example, some believe that it has unacceptable implications regarding the ethics of infanticide.44 At least one scholar has challenged the coherence of the concept of a time-relative interest.45 Rather than address these and other challenges here, I simply note that there are such challenges and acknowledge that one might responsibly doubt that any version of the TRIA will adequately address them. We have also found reason to doubt that the GSA establishes the moral permissibility of abortion (except in a (p.45) small range of cases). Such doubts, I submit, leave a broadly pro-life approach standing as a reasonable view about prenatal moral status.

How should we conceptualize this plurality of reasonable views? We might judge that there are several views that it is reasonable to hold only because we have yet to identify an argument or set of arguments that will settle the issue of prenatal moral status once and for all—at least in the minds of adequately informed, reasonable people. This suggests the possibility that some such body of reasoning is “out there” if only we could discover it. If that is the case, then perhaps continued intellectual labor will enable us to identify the correct view as the correct view. On the other hand, even if there is a uniquely correct view and some body of reasoning that could in principle show that this view is uniquely correct, perhaps—due both to our intellectual limitations and to the complexity of the abortion issue—we will never be in a strong position to say, with confidence, that we know that a particular view is the correct one. We may never be in a position to grasp that body of reasoning or, even if we can, we may not be able to ascertain that it is definitive—that it settles the debate. That seems quite possible to me. It would be a problem of moral knowledge.

Another possibility, which I take seriously, is that there is no uniquely correct view on prenatal moral status; rather, there is a range of reasonable views. If so, then no amount of information, insight, logical reasoning, or other resources that can improve our moral thinking could direct us, in practice or in principle, to a uniquely correct position. My taking this metaethical possibility seriously should not be understood to mean that I embrace moral relativism, a view that I find completely implausible. For this metaethical possibility requires no such sweeping denial of the objectivity of ethics. All it requires is the belief that ethics may be partly indeterminate: that there may be some ethical issues on which there is, at the end of the day, a plurality of more or less equally defensible yet incompatible views. A metaphor may help to illuminate this idea of partial indeterminacy in ethics: a hunk of Swiss cheese. Perhaps ethics is objective in the sense of having uniquely correct answers to most ethical questions—picture the solid parts of the cheese—while also having holes of indeterminacy. If so, that would explain why on some issues (abortion, embryo research, perhaps animals’ moral status, and/or the finer points of distributive justice) consensus seems especially elusive, without implying, insanely, that there are never correct answers to ethical questions.

From Moral to Political Philosophy: Responding to Reasonable Pluralism

I have argued that there is an impasse, a plurality of reasonable views, on the issue of prenatal moral status. Nor is the range of reasonable views narrow, as they would be if, say, they were all variations on a liberal approach; the range, (p.46) as we have seen, is very wide. How, then, can we proceed in a discussion of the ethics of abortion and embryo research? My suggestion is first to recall that these ethical issues arise at two levels and then to move our analysis to the level we have yet to address. The ethical issues arise at the level of individual choice (e.g., “Is it morally permissible for this woman in these circumstances to have an abortion?”) and at the level of public policy (e.g., “Is it morally defensible for the State to prohibit women from having abortions in these sorts of circumstances?”). Our discussion so far has taken place at the level of individual choice. Insofar as what is right at this level may not translate smoothly into what our laws and other public policies ought to be—for example, there is no law, nor should there be one, against the unethical behavior of speaking rudely to one’s relations—we should now ask about defensible public policies regarding abortion and embryo research. I submit that the best policy responses to reasonable pluralism about prenatal moral status will be fairly liberal.

Consider that conservative policies on the issues in question necessarily involve significant government-imposed limitations on liberty—on a woman’s liberty with respect to her own body and major life choices if abortion is prohibited or severely restricted, and on the scientific community’s freedom of inquiry if embryo research is banned or severely curtailed. Any government, of course, must impose some restrictions on liberty through the mechanism of enforceable law. But, a government appropriately imposes significant restrictions on liberty on the basis of moral values only if there is an overwhelming moral case—at the level of individual morality—for the restrictions.46 Such is the case in the legal prohibitions of rape, murder, and child abuse. As far as abortion and embryo research go, there is reasonable pluralism and no overwhelming moral case. This warrants some sort of liberal policy response.47

Consider the matter this way. For the pro-life view to be correct, several controversial assumptions must all be correct. For the sake of clarity in exposition, I will articulate the most conservative version of the pro-life view, noting along the way that certain amendments or qualifications are possible within a broadly pro-life approach. The first assumption concerns our origins:

  1. 1. We human beings come into existence at conception.

This assumption can be modified by replacing the last two words with “within the first two weeks after conception” or the like to reflect an alternative biological understanding of our origins. The second assumption concerns our possession of moral status:

  1. 2. We have full moral status, including a right to life, throughout our existence.

(p.47) From these two premises it follows that we have full moral status, including a right to life, from the moment we come into existence at conception. The third pivotal assumption concerns the relationship between full moral status and the morality of abortion and embryo research:

  1. 3. If we have full moral status from the time of conception, then abortion and embryo research are impermissible.

This assumption takes the possession of full moral status to be decisive to the ethics of killing embryos and fetuses. From these three premises, the conservative conclusion follows:

  1. (C) Abortion and embryo research are impermissible.

I have argued that it is reasonable to believe these premises—with the qualification that (3) and therefore (C) need to be modified to permit a few exceptions to the claim of impermissibility. Now I want to emphasize that it is reasonable to disbelieve one or more of these premises. Yet all three must be correct for the conclusion to be well-supported.48

The first premise can be reasonably doubted. I have defended a modified version of this premise on the basis of a biological view of our essence and identity; one could accept this modification and remain broadly pro-life. But, as noted earlier, mind essentialism—the view that we are essentially minds—should be regarded as a contender among views of our essence and identity. One might reasonably embrace mind essentialism, which implies that beings of our kind do not originate until sentience emerges in fetal development, sometime in the third trimester. If this is true, then the pro-life view cannot be correct.

As for (2), that we have full moral status throughout our existence, I have defended and deployed the TRIA in challenging this premise. But even one who doubted the TRIA might reasonably doubt (2): It is not self-evidently true, and the major arguments advanced in its favor proved less than compelling. One might reply that, while (2) is not self-evident in a strict sense, it is nevertheless highly plausible. I am sure that many people regard it as highly plausible. Yet many liberals do not—and sometimes even plausible theses can be responsibly rejected on the basis of a countervailing theoretical consideration (e.g., the TRIA) if the latter is sufficiently well supported.

Like the first two premises, the third—that full moral status on the part of fetuses would entail the impermissibility of abortion and embryo research—is also open to reasonable doubt. The GSA constitutes a powerful reason to question (3). Our examination of the GSA raised significant doubts about this argument, but we certainly did not refute it.

(p.48) The upshot is that the significant deprivations of liberty demanded by a conservative position on abortion and embryo research rest on three debatable premises. Laws and policies that significantly curtail liberty generally should not rest on moral positions about which consensus among reasonable people is so sorely lacking. In effect, women should have the prerogatives of deciding the contested matter of prenatal moral status for themselves and of making highly personal decisions regarding whether to have an abortion. Similarly, scientists should have the prerogative of deciding the moral status of embryos for themselves and pursuing embryo research if they think it appropriate. Whether the federal government should fund such research, however, raises distinct issues such as (1) whether partial deference to the pro-life sensibilities of a sizable portion of the public justifies withholding public funds from embryo research, and (2) whether such money is well spent on this investment in view of a variety of factors (e.g., alternatives to embryo research, competing claims for public funds). We will consider such issues in the final section of this chapter. What is clear at this juncture is that embryo research should not be prohibited.

Reflections on the Basic Shape of Justified Policies

Abortion

What would be the general shape of justified policies on abortion? I will take current American abortion laws as the point of departure for my analysis.

The U.S. Supreme Court’s decision in Roe v. Wade (1973) had the effect, for all practical purposes, of legalizing “abortion on request.”49 According to the Court, it was unconstitutional for a state to have laws prohibiting the abortion of a pre-viable fetus—that is, a fetus prior to the time at which fetuses can be born and kept alive with medical technologies (which, at that time, was around the end of the second trimester). A woman, the Court contended, has a constitutionally guaranteed right to privacy, on the basis of which she has a right to terminate a pregnancy prior to viability—although a state, for reasons relating to its interest in maternal health, may restrict the manner and circumstances in which abortions are performed after the first trimester. A woman’s nonabsolute right to terminate a pregnancy gives way, or may give way, to the state interest in fetal life at the point of viability: “For the stage subsequent to viability the State, in promoting its interest in the potentiality of human life, may, if it chooses, regulate, and even proscribe, abortion except where it is necessary, in appropriate medical judgment, for the preservation of the life or health of the mother.”50

In its decision in Casey (1992), the Supreme Court did not overturn Roe v. Wade, as many expected, but reaffirmed its “essential holding” that a woman (p.49) has a constitutional right to have an abortion prior to the point of viability.51 Moreover, the Court preserved Roe’s rule that states choosing to regulate or proscribe abortions post-viability must allow exceptions in cases in which the woman’s life or health is at risk. However, the Court abandoned the trimester framework of Roe in light of medical advances that better protected maternal health and promised to push viability to earlier stages of pregnancy. In a significant development, the Court adopted an “undue burden” framework for determining the constitutionality of certain procedural requirements for obtaining abortions—for example, 24-hour waiting periods and other requirements that are supposed to promote an informed decision by the mother, parental notification requirements, and spousal notification requirements. According to the Court, the informed consent and parental notification provisions in question passed legal muster whereas the requirement of spousal notification did not.

Much of the reasoning and wording of these two decisions have been rightly criticized. For example, the use of the terms “potential life” and “potentiality of human life” is conceptually monstrous. And what is most at issue for women is presumably less a concern of privacy than one of liberty with respect to a woman’s own body and life plan. No matter. I will consider whether the basic legal framework established for abortion in the United States is more or less sound as understood from the perspective of justified public policy. Thus, my exploration is not a legal analysis, which would address legal precedents and the content of the Constitution, but a moral analysis at the policy level. Before turning to this analysis, let us note two further components of the legal status quo.

In Harris v. McRae (1980), the Supreme Court addressed the constitutionality of the so-called Hyde amendment, which had passed through Congress with strong pro-life support. The Hyde amendment restricted federal Medicaid funding to cases in which the pregnant woman’s life is at risk and cases of rape or incest. Upholding this policy, the Court judged that a woman’s right to abortion does not encompass a right to have society pay for it.

Another part of the legal status quo concerns the use of a rare, late-term abortion procedure known medically as intact dilation and extraction (intact D & X) and popularly, at least among opponents, as “partial-birth abortions.” Sometimes used for late second-trimester abortions and more often for third-trimester abortions, intact D & X, in its most common form, involves the partial, feet-first delivery of a fetus followed by extraction of the brain in order to collapse the skull, so the head can pass through the cervix. In Stenberg v. Carhart (2000), the Supreme Court struck down Nebraska’s ban on intact D & X for, among other reasons, its ambiguous formulation. Congress later passed the Partial Birth Abortion Ban Act of 2003, which the Court (perhaps reflecting (p.50) a change in membership) upheld in Gonzalez v. Carhart (2007). According to the Court, because intact D & X is not the only possible late-term abortion procedure and is never necessary to protect a woman’s life or health, the Partial Birth Abortion Ban Act is not unconstitutional for failing to include such an exception.

Bringing these elements of the legal status quo together, we find this picture:

  1. 1. Abortion is entirely permissible prior to viability;

  2. 2. States may restrict or prohibit post-viability abortion except where necessary to protect the woman’s life or health;

  3. 3. Procedural requirements such as parent notification and a 24-hour waiting period are permitted if and only if they do not constitute an undue burden to the pregnant woman;

  4. 4. Federal funds will not be spent on the provision of abortions (with a few exceptions);

  5. 5. Intact dilation and extraction is prohibited.

Let us briefly consider these elements in turn.

The conjunction of elements (1) and (2) is, I think, very appropriate policy. Its liberality is justified by reasonable pluralism regarding the pre-sentient fetus. The cut-off point of viability, at which time states may set restrictions on abortion, is surely debatable, but I believe it is optimal for public policy purposes. Fetuses are thought to be viable near the end of six months’ gestation, which is around the earliest point at which fetuses may become sentient. Sentient beings have moral status; whether or not they have full moral status, the same as a person, is reasonably disputed. Many people believe that a late-term, sentient fetus has full moral status including a right to life. Nearly everyone agrees that a sentient infant has full moral status, and it is probably at least implicit in many people’s thinking that a viable fetus could already be a newborn infant and therefore should have the same moral status. In view of these points, and out of respect for the sensibilities of those who are pro-life, I think the viability criterion is optimal. Only if the time of viability moved much earlier in gestation (perhaps a couple of months earlier) would I think that this criterion should be abandoned, in the interest of ensuring that women have ample time in which to secure abortions. Some states may prohibit or substantially restrict post-viability abortions. Requiring exceptions where necessary to protect the woman’s life or health is appropriate given a woman’s right to self-defense: One may, as a last resort, kill to save one’s life or even to protect one from a grave threat to one’s health, even if the threat is entirely innocent of bad intent.

As for the third element of the legal status quo, the undue-burdens criterion for procedural requirements also strikes me as reasonable. Although having to (p.51) wait a day to have an abortion is inconvenient, there can be no denying that the decision is momentous and that time for reflection can help to promote genuinely informed consent (without being unduly burdensome). By contrast, requiring wives to notify their spouses of their decision to abort would be an undue burden due to the likelihood that many women who were pregnant “illicitly” would be subject to spousal abuse.

Turning now to the status quo’s fourth element, I, unlike many liberals, believe it is acceptable to prohibit the provision of federal funds for abortion services (preferably with exceptions, as discussed in the next paragraph) so long as that is the choice of duly elected representatives. That women have the right to terminate pregnancy does not mean that the public has to pay for abortion. The pro-life community is a substantial minority of persons whose views are not unreasonable and who have to tolerate very permissive policies on abortion. From the standpoint of their sensibilities, it likely would seem to add insult to moral injury to have to pay for abortions with public funds. To some extent, this sensibility is reflected by acts of Congress such as the Hyde amendment. At the same time, if Congress chooses to abandon this funding restriction based on a different weighing of the values at stake—more emphasis, say, on pluralism and economic justice for the poor—that, too, would be reasonable.

Where a society’s duly elected representatives decide to prohibit public funding for abortion, I believe that it would be best to make exceptions in two kinds of cases: where a mother’s life or health is at stake and in cases of what I call “no responsibility.” Some individuals are not responsible for their pregnancy in the sense that their taking part in an activity (sex) that carried the risk of pregnancy was either involuntary and/or uninformed in a relevant way. Rape victims are obviously not responsible. Nor, in some cases, are minors who are impregnated by other minors if they did not really understand the risks they were taking and, for the purposes of their decision to have sex, not responsible in the way a competent adult would be. The same may be said for some severely mentally impaired adults, who may sometimes have pregnancy-causing sex—perhaps with others who are similarly impaired—without being coerced yet without “knowing what they were doing.” So the relevant category justifying an exception is “no responsibility,” with rape being one instance. At the same time, I reject incest as a ground for a categorical exception. Probably, most cases of incest that lead to pregnancy involve rape of a minor by an adult family member. But where incestuous sex is voluntary and informed, among adults, I see no special reason why a society should feel compelled to make an exception and fund an abortion (although disgust alone may motivate an exception).

Moving on to the court decision about “partial-birth abortion,” my reaction is more critical. Contrary to the law that was upheld, it is not the procedure itself—with its partial delivery of the fetus—that matters. What matters is the (p.52) stage of fetal development. So a more justified policy would be to permit this procedure in those rare instances when it is performed pre-viability. And, as with post-viability abortion in general, it should be up to states to determine whether to restrict or ban this abortion procedure after the point of viability. So the law about intact D & X is redundant where it is justified and overreaching where it is not. Moreover, the court’s claim that the procedure is never necessary for the woman’s health is disputed among medical experts and represents an inappropriate intrusion into a pregnant woman’s prerogative to defend herself. The Partial Birth Abortion Ban Act was upheld by an irresponsible decision on the part of the Supreme Court.

Embryo Research

The kind of embryo research that has received the most academic and public attention in recent years is embryonic stem-cell research (ESCR). Another type of embryo research that has attracted considerable attention is research cloning (prematurely called “therapeutic cloning” by many proponents). There are other types of embryo research as well, but what is common to all of them—or at least all that are the subject of this discussion—is that the research entails the destruction of embryos.52 The issues I will consider are (1) whether the research in question should be legally permitted, (2) whether creation of embryos for the purpose of using them in such research should be legally permitted, and (3) whether public funds should be dedicated to the research. The major factors to be taken into consideration in forging justified policy are reasonable pluralism about embryos’ moral status, the value of scientific freedom of inquiry, the medical promise of embryo research, the moral sensibilities of the pro-life community, the possibility of viable alternatives to embryo research, and—in connection with the funding issue—alternative ways to use public funds.

Let us begin with ESCR, which is thought to hold the greatest medical promise among the varieties of embryo research. First, some background. A human embryo reaches the blastocyst stage of its development about five days after fertilization. At this stage, the embryo contains an inner cell mass that normally would develop into the fetus and an outer layer of cells that normally would develop into the placenta. The cells of the inner cell mass—embryonic stem cells—can be extracted to develop into virtually any type of human cell or tissue. They are thus described as pluripotent. (By contrast, the cells of an embryo through the eight-cell stage are totipotent—capable of forming a new embryo). Embryonic stem cells are thought to have tremendous therapeutic potential, especially in regenerative medicine, in view of their versatility. It is hoped that ESCR can lead to therapies for many diseases and impairments, including Parkinson’s disease, Alzheimer’s, diabetes, heart disease, and spinal (p.53) cord injury, to name just a few. Importantly, deriving stem cells from an embryo at the blastocyst stage requires destruction of the embryo. Due to disagreement over the embryo’s moral status, the aspect of embryo destruction generates the most ethical controversy surrounding ESCR.

The legal status quo of ESCR—indeed, of all embryo research—is somewhat up in the air. At the end of the George W. Bush administration, it was roughly the following: (1) Federal law prohibited federal funding of research that involved the destruction of embryos except for studies using embryonic stem cells derived from the small number of cell lines created prior to Bush’s August 2001 executive order on the subject, (2) federal law prohibited federal funding for any research that involved the creation of embryos for purposes that would involve their destruction, and (3) there was no significant federal regulation of embryo research conducted in the private sector. President Barack Obama reversed Bush’s restriction on federal funding for ESCR and clinical trials were approved, apparently auguring a new era of publicly supported research. However, a federal judge stunned the research community with an August 2010 ruling that Obama’s more permissive funding policy actually violated the law against federal funding of research that involves embryo destruction. This decision was reversed in April 2011 by the D.C. Court of Appeals and, at the time of this writing, the legal door has opened to federal funding for ESCR.

As for the issue of morally defensible policy, three factors are paramount: reasonable pluralism about embryos’ moral status, the value of scientific freedom of inquiry (which is generally agreed to establish a rebuttable presumption against government interference), and the extraordinary promise of this research. The sensibilities of the minority who oppose ESCR (reasonably) on the basis of their view of embryos’ moral status must be taken into account, but how to do so is not obvious. This factor does not justify prohibiting such research. Might it justify a prohibition on public funding? I would argue that it does not—biomedical research being a fundamentally public enterprise—unless and until it is demonstrated that alternatives to ESCR have comparable medical promise.

Some scientists believe that “adult” stem cells (stem cells derived from full-grown human beings) have such promise, meaning there is no medical advantage unique to ESCR. This argument gained momentum in the wake of several breakthroughs in which differentiated “adult” cells were genetically modified into a stem-cell-like state, resulting in induced pluripotent stem (iPS) cells.53 Much remains uncertain about the potential of iPS cells, however, and it seems very possible that ESCR has unique long-term therapeutic potential, especially in light of some recent findings.54 The reasonable course at this time, it seems, is to provide substantial public funds both for ESCR and for alternatives—especially iPS cell research—that avoid embryo destruction. If it becomes apparent on the basis of solid scientific evidence that iPS cell research is just as promising as ESCR, then it (p.54) would be appropriate for our elected government officials to consider withholding public funds from ESCR out of deference to the minority’s reasonable sensibilities. Until such a time, the great promise and possibly unique value of ESCR leave no doubt that public money is well invested in this research enterprise.

Is it ethical to create embryos for the express purpose of using them in research that will entail their destruction? According to a prominent school of thought, it is morally preferable to use embryos left over from fertility clinics, which would otherwise be discarded, than to create new ones for research. This thinking is consistent with Obama’s executive order, which left in place the prohibition of federal funds for research involving the creation of embryos that would be destroyed. I do not find this reasoning cogent.55 The embryo’s moral status—whether it is full, partial, or no moral status—is the same whether or not the embryo is going to be discarded. If embryos had full moral status (and this were the only reasonable view), then either they should not be produced for fertility purposes, or they should be produced in smaller numbers such that none is ever discarded unless it died of natural causes; accordingly, they should not be sacrificed for medical purposes. Analogously, the fact that someone is terminally ill, or waiting on death row to be executed, does not justify killing him even for important medical purposes; it does not matter that the individual is “going to die anyway.” But, in fact, there is reasonable pluralism about embryos’ moral status, in view of which it is acceptable to kill embryos for research purposes. It is also acceptable, on account of pluralism, to create embryos for fertility purposes. So it is hard to see why it would not also be acceptable to create them for research purposes.

One might object, however, that to create embryos solely for research purposes is to commodify them. To create embryos for fertility purposes, by contrast, is to afford them a significant chance to be implanted and come to enjoy the life of full-grown human being, which is hardly a way of commodifying them. To allow spare embryos from fertility clinics to be used in research, the argument continues, permits them to be dedicated to a noble research enterprise while avoiding the crassness of creating them for this purpose. But this reasoning makes sense only if we assume that embryos have partial moral status: enough so that concerns about commodification can have a foothold, but not so much that we must prohibit the turning over of spare embryos to research laboratories instead of waiting for them to be “adopted.” Thus, this reasoning does not do justice to the plurality of reasonable views that includes the view that embryos lack moral status. Finally, if embryos have only partial moral status, as this reasoning assumes, it is entirely unclear why we should not create embryos in support of the noble enterprise of biomedical research. So the creation of embryos for ESCR, or any other type of embryo research, seems no less justified than the conduct of such research.

(p.55) Let us finally consider research cloning, which is frequently asserted to promise unique medical opportunities. Stem cells extracted from cloned human embryos, produced specifically for research through somatic cell nuclear transfer—which involves removing the nucleus of a somatic cell, transferring it to an egg whose nucleus has been removed, and stimulating the egg into a state of totipotency—may prove uniquely valuable in studying genetic diseases and developing novel therapies. One major prospect for using cloning to derive embryonic stem cells would be to derive the stem cells used to treat a particular patient from her own cloned somatic cell, thereby avoiding the tissue mismatches that cause immune rejection. (This advantage of avoiding immune rejection has also been claimed on behalf of induced pluripotent stem cells derived from a patient’s somatic cells.) Another important possibility would be to clone embryonic stem cells from people with particular diseases to generate a limitless source of cells that can be used to study these diseases without having to extract tissue samples from patients.56

Does the fact that the stem cells derive from an embryo produced through cloning make any moral difference? I submit that it does not matter. There is nothing inherently wrong with cloning to produce an embryo. There are special concerns (taken up in Chapter 6) about reproductive cloning, the use of somatic cell nuclear transfer to create an embryo with the intention of implanting it and bringing the fetus to term, but research cloning is not reproductive cloning; and research cloning can be safely and appropriately circumscribed by making it illegal to implant a cloned embryo into a woman’s uterus. Research cloning as part of ESCR poses no novel difficulties. It is simply a way of creating embryos for research, which, we found, poses no moral difficulties beyond those involved in embryo research per se.

Notes

(1) Human Identity and Bioethics (Cambridge: Cambridge University Press, 2005)

(2) Marya Schechtman, The Constitution of Selves (Ithaca, NY: Cornell University Press, 1996)

(3) Lynne Rudder Baker, Persons and Bodies (Cambridge: Cambridge University Press, 2000)

(4) . Human Identity and Bioethics, chap. 2.

(5) Eric Olson, The Human Animal (New York: Oxford University Press, 1997)

(6) embodiedJeff McMahan, The Ethics of Killing (New York: Oxford University Press, 2002)

(7) Ibid.

(p.56) (8) Human Identity and BioethicsMcMahan, The Ethics of Killing, chap. 1

(9) Patrick Lee, “The Pro-Life Argument from Substantial Identity: A Defence,” Bioethics 18 (2004): 249–63

(10) “Moral Status, Human Identity, and Early Embryos: A Critique of the President’s Approach,” Journal of Law, Medicine & Ethics 34 (2006), pp. 50–54

(11) . Thanks to Alfonso Gomez-Lobo for making me aware of these competing models.

(12) Lee Silver, Remaking Eden (New York: Avon, 1997), p. 45

(13) Helen Pearson, “Your Destiny from Day One,” Nature 418 (July 4, 2002): 14–15

(14) . Gregor Damschen, Alfonso Gomez-Lobo, and Dieter Schoenecker, “Sixteen Days? A Reply to B. Smith and B. Brogaard on the Beginning of Human Individuals,” Journal of Medicine and Philosophy 31 (2006), p. 171. See also Alfonso Gomez-Lobo, “Individuality and Human Beginnings: A Reply to David DeGrazia,” Journal of Law, Medicine & Ethics 35 (2007): 457–62. Both articles cite S. F. Gilbert, Developmental Biology, 6th ed. (Sunderland, MA: Sinauer, 2006) and Pearson, “Your Destiny from Day One.”

(15) . I take it that distinctions of sex and of species are largely biological, partly conventional. By contrast, distinctions of race seem to be largely conventional, partly biological.

(16) Taking Animals Seriously: Mental Life and Moral Status (Cambridge: Cambridge University Press, 1996), pp. 56–61

(17) Elizabeth Harman, “The Potentiality Problem,” Philosophical Studies 114 (2003): 173–98

(18) . Feelings include not only sensations but also emotions and moods. Thus, an individual who lost all sensory capacity but was still capable of feeling upset, for example, would remain sentient and have interests.

(19) . See Gary Varner, In Nature’s Interests? (New York: Oxford University Press, 1998). Cf. Albert Schweitzer, “The Ethic of Reverence for Life,” in Tom Regan and Peter Singer (eds.), Animal Rights and Human Obligations (Englewood Cliffs, NJ: Prentice-Hall, 1976): 133–38.

(20) . See, e.g., Joel Feinberg, Harm to Others (Oxford: Oxford University Press, 1984); Peter Singer, Animal Liberation, 2nd ed. (New York: Avon, 1990); and Bonnie Steinbock, Life Before Birth (New York: Oxford University Press, 1992). For my development of this approach, see Taking Animals Seriously.

(p.57) (21) Joseph Raz, “On the Nature of Rights,” Mind 93 (1985): 194–214

(22) . See, e.g., Singer, Animal Liberation and Steinbock, Life Before Birth.

(23) Susan Lee et al., “Fetal Pain: A Systematic Multidimensional Review of the Evidence,” JAMA 294 (2005): 947–54

(24) . Strictly speaking, the capacity for consciousness is not equivalent to sentience—the capacity for feelings—because, in principle, a being might be capable of having conscious states such as thoughts but incapable of having feelings. In the world as we know it, however, all creatures capable of having conscious states are sentient.

(25) . See, e.g., Mary Mahowald, “Respect for Embryos and the Potentiality Argument,” Theoretical Medicine and Bioethics 25 (2004): 209–14 and Eugene Mills, “The Egg and I: Conception, Identity, and Abortion,” Philosophical Review 117 (2008): 323–48. For helpful replies to such arguments, see Alfonso Gomez-Lobo, “On Potentiality and Respect for Embryos: A Reply to Mary Mahowald,” Theoretical Medicine and Bioethics 26 (2005): 105–10 and Gomez-Lobo, “Does Respect for Embryos Entail Respect for Gametes?” Theoretical Medicine 25 (2004): 199–208.

(26) . This theory has roots in Derek Parfit’s reasoning about prudential value (Reasons and Persons [Oxford: Clarendon, 1984], Part 3) and was explicitly formulated in McMahan, The Ethics of Killing. Benefitting from McMahan’s ideas, I have developed a version of the TRIA and applied it to issues connected with prenatal moral status (see, e.g., “The Harm of Death, Time-Relative Interests, and Abortion,” Philosophical Forum 38 [2007]: 57–80).

(27) . The question of who is harmed more by death must be distinguished from another question: Who is typically better off at the time of death? The second question concerns not only the harm of death but also the amount of good one enjoyed while alive. Clearly, in typical cases the older victims of death have enjoyed more good in life than have younger victims. Thus, one might claim that the 25-year-old who dies is better off, overall, than the infant who dies: The young man is tragically harmed by death but had enjoyed lots of good before that time whereas the infant is substantially harmed by death but had enjoyed almost no good before that time. So one can acknowledge that the young man who dies is better off, overall, than the infant who dies and judge that death harms the young man more than the infant. Now, the 10-year-old who dies may seem roughly as badly off as the infant who dies. This coheres with the judgment that, while the child had enjoyed considerably more good than the infant had, death harms the child much more than the infant, making their overall prudential situations comparable.

(28) McMahan, The Ethics of Killing, p. 75

(29) “Why Abortion is Immoral,” Journal of Philosophy 86 (1989): 183–202

(30) “Must We Have Full Moral Status Throughout Our Existence? A Reply to Alfonso Gomez-Lobo” Kennedy Institute of Ethics Journal 17 (2008): 297–310

(31) “Does Respect for Embryos Entail Respect for Gametes,” p. 79

(p.58) (32) . This paraphrases an argument presented in Lee, “The Pro-Life Argument from Substantial Identity,” p. 250. Essentially the same argument is developed in Christopher Kaczor, The Ethics of Abortion (New York: Routledge, 2011), possibly the strongest book-length defense of the pro-life position.

(33) “The Pro-Life Argument from Substantial Identity,” pp. 252–53

(34) . Ibid, p. 262. See also Gomez-Lobo, “On Potentiality and Respect for Embryos: A Reply to Mary Mahowald,” p. 109.

(35) . My sense is that the present argument wrongly attributes to us two distinct essences. I agree that we are essentially human animals, and I can accept (setting aside some doubts) that human animal is a natural kind. The problem, in my judgment, is that natural kinds are generally not susceptible to classical definitions. As Saul Kripke has argued, our original concepts for natural kinds typically come by way of ostension: in the case of a species, “this kind of thing” (Naming and Necessity [Cambridge, MA: Harvard University Press, 1970]). So a natural kind such as our species would be “defined” by ostension, not by means of a description. Aristotle thought, by contrast, that the human being was both a biological kind and a kind susceptible to classical definition, namely “rational animal.” To believe this today, in my opinion, is philosophically naïve and reflects a failure to understand leading contemporary work on natural kinds.

(36) . Our use of person, by contrast, has been purely descriptive—applying to beings that meet a certain description—rather than implying anything about moral status.

(37) . Judith Jarvis Thomson, “A Defense of Abortion,” Philosophy and Public Affairs 1 (1971): 47–66 and David Boonin, A Defense of Abortion (Cambridge: Cambridge University Press, 2002). Boonin draws not only from Thomson but also significantly from Frances Kamm’s work. See especially Kamm, Creation and Abortion (New York: Oxford University Press, 1992).

(38) Boonin, A Defense of Abortion, pp. 135–39

(39) Ibid, pp. 172–75

(40) Ibid, p. 174

(41) The Ethics of Killing, p. 398

(42) . I agree with Kamm that intentionally killing an innocent person is sometimes justified, but requires more justification than letting an innocent person die (Creation and Abortion, p. 31). Accordingly, in order to justify killing, the burden one would face if the individual is not killed would have to be a lot higher than the burden that would be sufficient to justify allowing to die.

(43) . By “abortions,” I really mean those that involve a medical procedure—what people generally have in mind when using the term. I do not mean to include the use of birth control methods that work (at least sometimes) by preventing implantation of early embryos, thereby allowing them to die. If we included these instances of birth control as abortions, then the category of abortions that do not involve killing would be considerably larger.

(44) . See, e.g., Matthew Liao, “Time-Relative Interests and Abortion,” Journal of Moral Philosophy 4 (2007): 242–56. I reply to this charge in “The Harm of Death, Time-Relative Interests, and Abortion,” pp. 76–80.

(p.59) (45) Ben Bradley, “The Worst Time to Die,” Ethics 118 (2008): 291–314

(46) . One might think that this principle implies either that infanticide should be legal or that there is an overwhelming moral case against infanticide at the level of individual morality. I disagree, however, because it seems clear to me that prohibiting people from killing infants is not a significant restriction of personal liberty.

(47) John Rawls, Political Liberalism (New York: Columbia University Press, 1993)

(48) . I assume that no pro-life argument that did not make these assumptions would be promising. In view of how consistently these assumptions are made by pro-life thinkers, including the academics among them, I consider my assumption fairly safe.

(49) United States Supreme Court, Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705 (1973)

(50) Ibid.

(51) Planned Parenthood of Southeastern Pennsylvania v. Casey, Governor of Pennsylvania, United States Supreme Court, 112 S. Ct. 2791 (1992)

(52) . A statistical analysis of data involving embryos, for example, could be considered embryo research, but it would fall outside the present discussion.

(53) . See Rob Stein, “Scientists Report Advance in Stem Cell Alternative,” Washington Post (9/26/08): A17 and Rob Stein, “Cell Technique Works Without Embryos,” Washington Post (10/1/10): A2.

(54) Andy Coghlan, “Stem Cell Bust as Mice Reject Own Tissue,” New Scientist (May 21, 2011): 10

(55) Dan Brock, “Creating Embryos for Use in Stem Cell Research,” Journal of Law, Medicine & Ethics 38 (2010): 229–37

(56) Andy Coghlan, “UK Cloners Target Diabetes Cure,” New Scientist 183 (August 21, 2004): 8–9

Notes:

(1) Human Identity and Bioethics (Cambridge: Cambridge University Press, 2005)

(2) Marya Schechtman, The Constitution of Selves (Ithaca, NY: Cornell University Press, 1996)

(3) Lynne Rudder Baker, Persons and Bodies (Cambridge: Cambridge University Press, 2000)

(4) . Human Identity and Bioethics, chap. 2.

(5) Eric Olson, The Human Animal (New York: Oxford University Press, 1997)

(6) embodiedJeff McMahan, The Ethics of Killing (New York: Oxford University Press, 2002)

(7) Ibid.

(p.56) (8) Human Identity and BioethicsMcMahan, The Ethics of Killing, chap. 1

(9) Patrick Lee, “The Pro-Life Argument from Substantial Identity: A Defence,” Bioethics 18 (2004): 249–63

(10) “Moral Status, Human Identity, and Early Embryos: A Critique of the President’s Approach,” Journal of Law, Medicine & Ethics 34 (2006), pp. 50–54

(11) . Thanks to Alfonso Gomez-Lobo for making me aware of these competing models.

(12) Lee Silver, Remaking Eden (New York: Avon, 1997), p. 45

(13) Helen Pearson, “Your Destiny from Day One,” Nature 418 (July 4, 2002): 14–15

(14) . Gregor Damschen, Alfonso Gomez-Lobo, and Dieter Schoenecker, “Sixteen Days? A Reply to B. Smith and B. Brogaard on the Beginning of Human Individuals,” Journal of Medicine and Philosophy 31 (2006), p. 171. See also Alfonso Gomez-Lobo, “Individuality and Human Beginnings: A Reply to David DeGrazia,” Journal of Law, Medicine & Ethics 35 (2007): 457–62. Both articles cite S. F. Gilbert, Developmental Biology, 6th ed. (Sunderland, MA: Sinauer, 2006) and Pearson, “Your Destiny from Day One.”

(15) . I take it that distinctions of sex and of species are largely biological, partly conventional. By contrast, distinctions of race seem to be largely conventional, partly biological.

(16) Taking Animals Seriously: Mental Life and Moral Status (Cambridge: Cambridge University Press, 1996), pp. 56–61

(17) Elizabeth Harman, “The Potentiality Problem,” Philosophical Studies 114 (2003): 173–98

(18) . Feelings include not only sensations but also emotions and moods. Thus, an individual who lost all sensory capacity but was still capable of feeling upset, for example, would remain sentient and have interests.

(19) . See Gary Varner, In Nature’s Interests? (New York: Oxford University Press, 1998). Cf. Albert Schweitzer, “The Ethic of Reverence for Life,” in Tom Regan and Peter Singer (eds.), Animal Rights and Human Obligations (Englewood Cliffs, NJ: Prentice-Hall, 1976): 133–38.

(20) . See, e.g., Joel Feinberg, Harm to Others (Oxford: Oxford University Press, 1984); Peter Singer, Animal Liberation, 2nd ed. (New York: Avon, 1990); and Bonnie Steinbock, Life Before Birth (New York: Oxford University Press, 1992). For my development of this approach, see Taking Animals Seriously.

(p.57) (21) Joseph Raz, “On the Nature of Rights,” Mind 93 (1985): 194–214

(22) . See, e.g., Singer, Animal Liberation and Steinbock, Life Before Birth.

(23) Susan Lee et al., “Fetal Pain: A Systematic Multidimensional Review of the Evidence,” JAMA 294 (2005): 947–54

(24) . Strictly speaking, the capacity for consciousness is not equivalent to sentience—the capacity for feelings—because, in principle, a being might be capable of having conscious states such as thoughts but incapable of having feelings. In the world as we know it, however, all creatures capable of having conscious states are sentient.

(25) . See, e.g., Mary Mahowald, “Respect for Embryos and the Potentiality Argument,” Theoretical Medicine and Bioethics 25 (2004): 209–14 and Eugene Mills, “The Egg and I: Conception, Identity, and Abortion,” Philosophical Review 117 (2008): 323–48. For helpful replies to such arguments, see Alfonso Gomez-Lobo, “On Potentiality and Respect for Embryos: A Reply to Mary Mahowald,” Theoretical Medicine and Bioethics 26 (2005): 105–10 and Gomez-Lobo, “Does Respect for Embryos Entail Respect for Gametes?” Theoretical Medicine 25 (2004): 199–208.

(26) . This theory has roots in Derek Parfit’s reasoning about prudential value (Reasons and Persons [Oxford: Clarendon, 1984], Part 3) and was explicitly formulated in McMahan, The Ethics of Killing. Benefitting from McMahan’s ideas, I have developed a version of the TRIA and applied it to issues connected with prenatal moral status (see, e.g., “The Harm of Death, Time-Relative Interests, and Abortion,” Philosophical Forum 38 [2007]: 57–80).

(27) . The question of who is harmed more by death must be distinguished from another question: Who is typically better off at the time of death? The second question concerns not only the harm of death but also the amount of good one enjoyed while alive. Clearly, in typical cases the older victims of death have enjoyed more good in life than have younger victims. Thus, one might claim that the 25-year-old who dies is better off, overall, than the infant who dies: The young man is tragically harmed by death but had enjoyed lots of good before that time whereas the infant is substantially harmed by death but had enjoyed almost no good before that time. So one can acknowledge that the young man who dies is better off, overall, than the infant who dies and judge that death harms the young man more than the infant. Now, the 10-year-old who dies may seem roughly as badly off as the infant who dies. This coheres with the judgment that, while the child had enjoyed considerably more good than the infant had, death harms the child much more than the infant, making their overall prudential situations comparable.

(28) McMahan, The Ethics of Killing, p. 75

(29) “Why Abortion is Immoral,” Journal of Philosophy 86 (1989): 183–202

(30) “Must We Have Full Moral Status Throughout Our Existence? A Reply to Alfonso Gomez-Lobo” Kennedy Institute of Ethics Journal 17 (2008): 297–310

(31) “Does Respect for Embryos Entail Respect for Gametes,” p. 79

(p.58) (32) . This paraphrases an argument presented in Lee, “The Pro-Life Argument from Substantial Identity,” p. 250. Essentially the same argument is developed in Christopher Kaczor, The Ethics of Abortion (New York: Routledge, 2011), possibly the strongest book-length defense of the pro-life position.

(33) “The Pro-Life Argument from Substantial Identity,” pp. 252–53

(34) . Ibid, p. 262. See also Gomez-Lobo, “On Potentiality and Respect for Embryos: A Reply to Mary Mahowald,” p. 109.

(35) . My sense is that the present argument wrongly attributes to us two distinct essences. I agree that we are essentially human animals, and I can accept (setting aside some doubts) that human animal is a natural kind. The problem, in my judgment, is that natural kinds are generally not susceptible to classical definitions. As Saul Kripke has argued, our original concepts for natural kinds typically come by way of ostension: in the case of a species, “this kind of thing” (Naming and Necessity [Cambridge, MA: Harvard University Press, 1970]). So a natural kind such as our species would be “defined” by ostension, not by means of a description. Aristotle thought, by contrast, that the human being was both a biological kind and a kind susceptible to classical definition, namely “rational animal.” To believe this today, in my opinion, is philosophically naïve and reflects a failure to understand leading contemporary work on natural kinds.

(36) . Our use of person, by contrast, has been purely descriptive—applying to beings that meet a certain description—rather than implying anything about moral status.

(37) . Judith Jarvis Thomson, “A Defense of Abortion,” Philosophy and Public Affairs 1 (1971): 47–66 and David Boonin, A Defense of Abortion (Cambridge: Cambridge University Press, 2002). Boonin draws not only from Thomson but also significantly from Frances Kamm’s work. See especially Kamm, Creation and Abortion (New York: Oxford University Press, 1992).

(38) Boonin, A Defense of Abortion, pp. 135–39

(39) Ibid, pp. 172–75

(40) Ibid, p. 174

(41) The Ethics of Killing, p. 398

(42) . I agree with Kamm that intentionally killing an innocent person is sometimes justified, but requires more justification than letting an innocent person die (Creation and Abortion, p. 31). Accordingly, in order to justify killing, the burden one would face if the individual is not killed would have to be a lot higher than the burden that would be sufficient to justify allowing to die.

(43) . By “abortions,” I really mean those that involve a medical procedure—what people generally have in mind when using the term. I do not mean to include the use of birth control methods that work (at least sometimes) by preventing implantation of early embryos, thereby allowing them to die. If we included these instances of birth control as abortions, then the category of abortions that do not involve killing would be considerably larger.

(44) . See, e.g., Matthew Liao, “Time-Relative Interests and Abortion,” Journal of Moral Philosophy 4 (2007): 242–56. I reply to this charge in “The Harm of Death, Time-Relative Interests, and Abortion,” pp. 76–80.

(p.59) (45) Ben Bradley, “The Worst Time to Die,” Ethics 118 (2008): 291–314

(46) . One might think that this principle implies either that infanticide should be legal or that there is an overwhelming moral case against infanticide at the level of individual morality. I disagree, however, because it seems clear to me that prohibiting people from killing infants is not a significant restriction of personal liberty.

(47) John Rawls, Political Liberalism (New York: Columbia University Press, 1993)

(48) . I assume that no pro-life argument that did not make these assumptions would be promising. In view of how consistently these assumptions are made by pro-life thinkers, including the academics among them, I consider my assumption fairly safe.

(49) United States Supreme Court, Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705 (1973)

(50) Ibid.

(51) Planned Parenthood of Southeastern Pennsylvania v. Casey, Governor of Pennsylvania, United States Supreme Court, 112 S. Ct. 2791 (1992)

(52) . A statistical analysis of data involving embryos, for example, could be considered embryo research, but it would fall outside the present discussion.

(53) . See Rob Stein, “Scientists Report Advance in Stem Cell Alternative,” Washington Post (9/26/08): A17 and Rob Stein, “Cell Technique Works Without Embryos,” Washington Post (10/1/10): A2.

(54) Andy Coghlan, “Stem Cell Bust as Mice Reject Own Tissue,” New Scientist (May 21, 2011): 10

(55) Dan Brock, “Creating Embryos for Use in Stem Cell Research,” Journal of Law, Medicine & Ethics 38 (2010): 229–37

(56) Andy Coghlan, “UK Cloners Target Diabetes Cure,” New Scientist 183 (August 21, 2004): 8–9