THE AMBIGUITY OF THE EMBRYO: ETHICAL INCONSISTENCY IN THE HUMAN EMBRYONIC STEM CELL DEBATE

The Ambiguity of the Embryo

The embryo is a deeply, perhaps irretrievably ambiguous entity, one that defies classification and slips seamlessly between moral and biological categories. While many features of this ambiguity have been evident for millennia, discussed by Aristotle (Barnes 1984) and in many religious traditions, it is only really with the advent of modern embryology, genome analysis, and stem cell science that the truly radical features of the power of cells, to differentiate and specialise, have challenged many of the myths concerning the embryo, and its moral status.¹ If we start by examining some of the ways in which the ambiguity of the embryo is at its most dramatic, the problems created by this dubious status will become clearer.

Embryo Splitting

When identical twins occur in nature they result from the splitting of the early embryo in utero, and the resulting twins, true clones, have identical genomes. This process can be mimicked in the laboratory, and in vitro embryos can be deliberately split, creating matching siblings, which can be used for reproduction, genetic testing, or scientific research. This process itself has a number of ethically puzzling if not problematic features (Harris 1998, 2004; see also Marquis, McMahon, and Sagan and Singer in this collection).

Embryo splitting can be done at various stages of embryonic development. When performed in an embryo consisting of 32 to 150 cells, the technique is referred to as “blastocyst division.” The blastocyst is split in half, and the two halves, when implanted into a uterus, can develop as identical “twins.” Embryo splitting at a very early stage of embryonic development (<32 cells) is called “blastomere separation.” In an embryo consisting of four cells, all cells (blastomeres) are still “totipotent” (that is, where all cells could become any part of the resulting individual or, indeed, could develop into a whole new individual). Consequently, if you take a four-cell-stage embryo and split it into four cells, each one of these cells constitutes a new embryo, which could be implanted with the potential for successful development into adulthood (as we discuss below, there is a dramatic wastage rate of embryos in all human reproduction). Each cell is the clone or identical “twin” of any of the others and comes into being not through conception but because of the division of the early cell mass. Moreover, these four cells can be recombined into one embryo again. This creates a situation where, without the destruction of a single human cell, one human life, if that is what it is, can be split into four and can be recombined again into one.

Did “life” in such a case begin as an individual, become four individuals, and then turn into a single embryo again? We should note that whatever our answer to this question, all this occurs without the creation of extra matter and without the destruction of a single cell. Those who think that ensoulment takes place at conception have an interesting problem in having to account for the splitting of one soul into four, and for the destruction of three souls when the four embryos are recombined into one, and to account for (and resolve the ethics of) the killing of three individuals, without a single human cell being removed or destroyed. These possibilities should perhaps give us pause in attributing the beginning of morally important life to a point like “conception,” which, moreover, is not a point but a process.²

Embryo splitting allows the use of genetic and other screening by embryo biopsy. In embryo biopsy usually one or two cells are detached from an early embryo for genetic testing. If the biopsied cell is totipotent, it is effectively an embryo (or would be considered so by many). Embryo biopsy would then involve the testing of one embryo to ascertain the health and genetic status of the remaining clone or clones (since the rest of the embryo may be further split to multiply cloned siblings).

Embryo splitting may also be used to create human embryonic stem cell (hESC) lines. Recently, a research team from the biotechnology firm Advanced Cell Technology in Massachusetts succeeded in producing an hESC line from a biopsied blastomere obtained from an embryo in the eight-cell stage (Klimanskaya et al. 2006). If the biopsied cell and the seven remaining cells could still develop into an adult human being without being aggregated or combined with other embryos, they are totipotent. It is not certain whether or not a blastomere from an eight-cell-stage embryo has become sufficiently differentiated to lose its totipotency. If the biopsied blastomere does have the capacity to develop into a full-grown human being, then it would be an embryo, and those who care about embryos as “one of us” would find the destruction of blastomeres for stem cell production immoral.

Splitting one embryo into two or more embryos could also be of great benefit for infertile couples by increasing the number of embryos available for transfer in a single in vitro fertilisation (IVF) cycle (Wood 2001). Not only would this reduce the number of egg retrieval cycles, thus relieving the physical burdens and costs of IVF treatment for infertility, it could also increase the pregnancy rate. If a couple can only produce one or two embryos, their chance of having a baby would be increased by splitting each of the embryos into two or four. This implies that a child could be produced that may otherwise not exist. An obvious question is whether it would be morally defensible, by outlawing the creation of such “clones” by twinning, to deny a woman the chance to have the child she desperately seeks. If this procedure would enable a woman to create a sufficient number of embryos to give her a reasonable chance of a successful pregnancy, then the objections would have to be weighty and would have to distinguish this sort of embryo loss from that which naturally and inevitably occurs in sexual reproduction (see below).

We should note a tension between how we think of the ethics of the destruction of an individual embryo involving cell loss, on the one hand, and destruction of an individual embryo without cell loss, on the other. As we have seen, the process of embryo splitting also allows for recombination. If the embryos are recombined following embryo splitting, a number of individual twins will have been “destroyed” or “killed” without the destruction of a single cell. Is such a process more ethically problematic or less ethically problematic than destroying a biopsied totipotent cell or a clump of cells that has the capacity to implant and grow to term? If, as seems likely, the reason why it is thought objectionable to recombine such clones is the loss of potential human beings, then perhaps it would be considered unethical not to split any embryo into as many twins as possible. By so doing we would after all maximise just that potential, the loss of which, supposedly, inhibits recombination. If all this has a dizzying effect, it is perhaps because the language that we use misleads us.

We are often misled by terminology. To call these early cells or clumps of cells “twins” tempts us to think of them as “persons.” But, as we have pointed out, if an in vitro embryo in which all cells are at the totipotent stage were to be split into four cells, you will have created four (new?) embryos, quadruplets. Take three away and destroy them or recombine all four into one and you are in a sense back where you started, having done exactly the same, namely, created a single potentially viable embryo with a particular genome. You have wasted potential experimental material or potentially viable embryos or even killed three human individuals. Yet this waste arguably also occurs whenever a cell mass that could viably be divided is left undivided. If the recombined embryo, or the surviving quadruplet, is implanted, comes to birth, and grows to maturity, it will have the same genome it would have had if the division and recombination had never taken place or if its siblings had never been created and disappeared. Will it be the same person it would have been? Does it have the same identity it had in its former incarnation? Certainly its life story is different.

It is difficult to analyse the ethics of the possibilities we have just described. In the recombination scenario not a single human cell has been destroyed. In the case of embryo splitting no new cells or matter have been created, and yet three individuals come and go. Have the interests of any individuals been harmed? If these embryos may be said to have an interest in actualising their potential, then perhaps there may be a sense in which they have been wronged if not harmed.³ We must look at potentiality more closely.

Potentiality

The idea of potentiality is central in discussing the ethics of using embryos for research and therapy. One feature of human embryos that members of other species do not share is their particular potential, not simply to be born and to be human but to become the sort of complex, intelligent, self-conscious, multifaceted creatures typical of the human species. There seem to be at least two problems with potentiality interpreted as the idea that human embryos or fetuses are morally important beings in virtue of their potential or their having a protectable interest in actualising that potential.⁴

The first objection to protecting individuals because of their potential is logical: acorns are not oak trees, nor are eggs omelettes. It does not follow from the fact that something has potential to become something different that we must treat it always as if it had achieved that potential. Unless and until we achieve the possibility of immortality, all of us share one important and inexorable potentiality—we are all potentially dead, but it does not follow that we must be treated as if we were dead now.

The second difficulty with the potentiality argument involves the scope of the potential for personhood. If the human zygote has the potential to become an adult human being and is supposedly morally important in virtue of that potential, then what of the potential to become a zygote? Something has the potential to become a zygote, and whatever has the potential to become the zygote has whatever potential the zygote has. It follows that the unfertilized egg and the sperm also have the potential to become fully functioning adult humans. In addition, it is possible to stimulate eggs, including human eggs, to divide and develop without fertilization (parthenogenesis). As yet it has not been possible to continue the development process artificially beyond the blastocyst stage, but if it ever does become possible, then the single unfertilized egg, without need of sperm or cloning, would itself have the potential of the zygote. Cloning by nuclear transfer, which involves deleting the nucleus of an unfertilized egg, inserting the nucleus taken from any body cell, and electrically stimulating the resulting newly created egg to develop, can in theory produce a new human. This was the method used to produce the first mammal cloned from an adult body cell, Dolly the sheep, in 1997. This means that any cell from a human body has the potential to become a new “twin” of that individual. All that is needed is an appropriate environment and appropriate stimulation. The techniques of parthenogenesis and cloning by nuclear transfer mean that conception or “fertilization” is not a necessary precursor process for the creation of human beings. (See also Sagan and Singer in this collection.)

The account of potentiality given here may be thought to have misrepresented the argument from potential. John Finnis, for example, has argued that “[a]n organic capacity for developing eye-sight is not ‘the bare fact that something will become’ sighted; it is an existing reality, a thoroughly unitary ensemble of dynamically inter-related primordia of, bases and structures for, development” (Finnis 1995, 50). He concludes that “there is no sense whatever in which the unfertilized ovum and the sperm constitute one organism, a dynamic unity, identity, whole” (50). This sort of potentiality is often referred to as “active” or “inherent” potentiality: an internal state that gives an entity, say A, the capacity to change itself, under appropriate conditions, and that determines that A as a potential Z will become Z rather than something else (Persson 2003). A has an “inherent dynamic” causing A to realise its potential if it is not hindered in its natural development.

However, it is surely the case that A has the potential for Z if, when a certain number of things do and do not happen to A (or to A plus N), then A (or A plus N) will become Z. For even a “unitary ensemble of dynamically inter-related primordia of, bases and structures for, development” must have a certain number of things happen to it and a certain number of things that do not happen to it if its potential is to be actualized. If A is a zygote, it must implant, be nourished, and not be exposed to dangerous substances in the womb, and it must have a genetic constitution compatible with survival to term and beyond. Defining potentiality as an all-or-nothing matter solely dependent on an entity's inherent dynamic to become a human person ignores the immense importance of diverse external factors that play a role in the actualisation of this potential as well as the substantial differences in potential at various stages of development (Devolder 2005). Moreover, insistence on a “unitary ensemble,” on “one organism,” seems also to apply to cloning by nuclear substitution, surely an embarrassing fact. In any body cell there is a complete single human genome; if treated appropriately, that genome present in the cell nucleus might be cloned. Thus this method of cloning allows for the “existing reality” of a complete genome which exhibits the “dynamic unity, identity, whole[ness]” that the Finnis analysis requires, and we can therefore now ascribe potentiality in the Finnis sense to the nucleus of every cell in every body.

Some, however, would object that only the entity resulting from nuclear transfer (after the somatic cell has been fused with the enucleated egg) has the same potential as an embryo and, therefore, should be equally protected. Only then would a new cell exist that would have the inherent dynamic necessary to become a human person. But even if there would indeed be a difference in potential between a somatic cell before and after nuclear transfer, we still need a strong argument why a particular “sort” or “amount” of potential would give us a reason to protect an entity as an adult human being. A particular “inherent dynamic” or “inherent potential” may be a necessary condition to become a human person and thus to accord value to an entity, but it is far from obvious why this should also be a sufficient condition.

The moral importance of drawing attention to the potentiality of something suggests that it is actualising a particular potential that matters. Our moral concern with what it is that has the potential to become an adult human being would be inexplicable if persons or adult humans did not matter. We are interested in the potentiality argument because we are interested in the potential to become a particular, and particularly valuable, sort of thing. If the zygote is important because it has the potential for personhood, and if that is what makes it a matter of importance to protect and actualize its potential, then whatever has the potential to become that zygote must also be morally significant for the same reason. Those who value potentiality for personhood surely do so not because the potential is contained within “one organism” but because it is the potential to become something the actualisation of which has moral importance.

Moreover, it is not required that potentiality be contained within one organism in order to be preserved or actualised, as has been demonstrated by recent developments in stem cell research (see Devolder and Ward in this collection). ESCs are obtained from embryos at the blastocyst stage. A blastocyst consists of two distinct cell types: inner cell mass (ICM) cells, which will become the “embryo proper,” the fetus, and later the adult human being, and trophoblast cells, which will contribute to the placental support system necessary for the development of the fetus in the uterus. ESCs are derived from the isolated ICM. So far, it has been generally accepted that hESCs have no significant moral status because, just like ICM cells, they are “merely” pluripotent, which means they can form all embryonic but only some extra-embryonic tissues. A totipotent cell (i.e., an embryo) can produce the extra-embryonic tissues as well and can thus result in a whole new individual. This moral division between pluripotent and totipotent cells, however, may not be as sound a criterion as it seems to be.

Scientific evidence suggests that human ICM cells as well as hESCs can develop into a whole new person. In the mouse it has been proved that the isolated ICM as well as mESCs derived from it maintain their capacity to form an adult mammal. When they are aggregated with tetraploid embryos—two-cell-stage zygotes that have been fused and have twice the normal number of chromosomes—they develop into normal mice (Li et al. 2005a, 2005b; Nagy et al. 1990, 1993). These mice consist only of the ICM cells or the ESCs and not of the tetraploid embryos, which only provide a surrogate trophectoderm. (The cells of the trophectoderm give rise to extra-embryonic tissues and do not incorporate into what is referred to as the “embryo proper,” which will eventually form the fetus and adult organism.) Moral issues prevent these aggregation experiments from being carried out in humans, but there is no reason why such procedures would not work with human ICM cells or hESCs. Because the use of tetraploid human embryos as surrogate trophectoderm could raise moral issues, as it involves the instrumental use of embryos, tetraploid embryos could be replaced by trophectodermal cells derived from hESCs (Gerami-Naini et al. 2004).

The aggregation process has not been done yet, but it is theoretically possible.⁵ This highlights a further ambiguity with the embryo (see Devolder and Ward in this collection). If these possibilities are proved, then there will be elasticity between pluripotent and totipotent cells, and it will not be possible to claim that pluripotent cells have permanently lost the capacity to form an embryo and a fetus because they cannot make the required extra-embryonic tissue and membrane. If hESCs can do everything a totipotent cell or an embryo can, then those who accord full moral status to the embryo should treat hESCs, and also ICM cells (which have the same potential), as moral equals of the embryo and thus of the adult human being.

The conclusions we can derive from the fact that something has the potential to become a human person are few. It may be a necessary condition to accord moral status to an entity, but saying that it is also a sufficient condition, and that therefore we need to protect an embryo as a full human being, lacks every rational basis. If there is a protectable “interest” in actualising potential or even a powerful moral interest in doing so, then the consequence is a very demanding ethic and one that would surely require us to actualise all human potential whenever we have an opportunity to do so. Particularly given the present state of technology, this would be a very demanding ethic indeed for women. That is not, of course, a decisive argument against acceptance of the ethic of always attempting to actualise valuable potential. However, those who use this argument as a reason for protecting embryos must, in order to be consistent, protect whatever has the potential to become a human being in the same way and to the same extent. This, among many other things, would entail an ethic of maximal procreation, or never knowingly missing an opportunity to create and protect embryos.

Rights

Failing to protect embryos does not involve any violation of the rights, at least of the embryos, although the progenitors or others in lawful possession of embryos may have rights at stake. Perhaps a word or two of explanation for this claim is appropriate. There are two main theories of rights: choice theory and interest theory (Cohen 1995, 68; Dworkin 1977 and 1994, 210–16; Steiner 1994; Waldron 1988). Choice theory sees rights as securing “the protection and promotion of autonomy or liberty,” and interest theory sees rights as serving to further individual well-being or welfare. Clearly, on choice theory embryos cannot possess rights because embryos are not autonomous and so their rights cannot be analysed in terms of choices. Even according to interest theory embryo rights are problematic, not least because, arguably, embryos cannot experience welfare and therefore have no welfare interests that can be served. Joseph Raz, for example, suggests that an individual is capable of possessing rights “if and only if…his well-being is of ultimate importance” (1986, 166). Embryos, however, have no well-being, for well-being is a state of being experienced as good by the subject of the relevant experiences. Certainly all early embryos, such as those in the pre-implantation stage that we have been discussing, lacking as they do both a central nervous system and indeed a brain, are incapable of experiences of any sort (Sumner 1987, 47).

Most legal systems and international human-rights law find no room for attributing rights, particularly the right to life, to embryos or fetuses. Two recent cases decided by the European Court of Human Rights have added to a formidable set of precedents confirming this status as far as the law of most jurisdictions is concerned.⁶

Humans Are Reckless of Embryonic Human Life

A further feature of current attitudes to the embryo which reveal not only the ambiguous status of the embryo but also the ambivalence of most humans towards the precursor forms of themselves concerns the extremely high rate of embryo loss and abnormality in human reproduction. It is doubtful that sexual reproduction, with its risk of sexually transmitted disease, its high abnormality rate in the resulting children, and its gross inefficiency in terms of the death and destruction of embryos, would ever have been approved by regulatory bodies if it had been invented as a reproductive technology rather than simply been “found” as part of our evolved biology.

Given the moral importance attached to embryos and the fact that embryos are regarded by many as sharing the same moral status with the rest of humankind, it is the tolerated rate of embryo loss that is particularly interesting. Embryo loss in normal sexual reproduction, including unprotected intercourse not directly intended to result in conception, is certainly very high. Robert Winston gave the figure of five embryos lost for every live birth some years ago in a personal communication (to John Harris). Anecdotal evidence from a number of sources confirms this high figure, but the literature is rather more conservative, making more probable a figure of three embryos lost for every live birth (Boklage 1990; Leridon 1977). Again, in a personal communication (to John Harris) Henri Leridon confirmed that a figure of three lost embryos for every live birth is a reasonable conservative figure. Ron Green (2001, note 185) has suggested (to John Harris) that between two-thirds and three-quarters of all embryos do not implant.⁷ Additional embryo loss occurs as a result of various means of contraception that are widely accepted. The combined oral contraceptive pill has a number of modes of operation, one of which prevents implantation of the embryo at between five and eight days' development. The so-called morning-after contraceptive pill also prevents implantation, as does the intra-uterine device, or coil. Even those who use so-called natural means of contraception, such as the rhythm method, intentionally decrease the chances of the embryo to implant and thus to actualise its potential (Bovens 2006). The combined effects of these various contraceptive methods increase the tally of embryo loss as a “side effect” of human sexuality, but it is impossible to arrive at reliable estimates as to the total numbers of embryos involved. Interestingly, most of this embryo loss involves the death of embryos at precisely the stage of development at which stem cells are usually harvested for hESC research, namely, between five and eight days' development. Those who attempt to have children in the light of these facts and indeed those who have unprotected intercourse and those who use contraceptive methods that risk embryo loss are all accepting that what they are doing or trying to do justifies the creation and sacrifice of embryos.

In the case of attempts to procreate using sexual reproduction, one obvious and inescapable conclusion is that God or nature has ordained that “spare” embryos be produced for almost every pregnancy, and that most of these will have to die in order that a sibling embryo can come to birth. Thus the wilful creation and sacrifice of embryos is an inescapable and inevitable (and presumably acceptable, or at least tolerable?) part of the process of procreation, and is not unique to assisted reproduction technologies (ART). Both natural procreation and ART involve a process in which embryos, additional to those that will actually become children, are created only to die. If either of these processes is justified it is because the objective of producing a live healthy child is judged worth this particular cost. It follows that no one who regards it as acceptable to try to have children has any principled objection to the creation and destruction of embryos in a good cause. The only question is how good the cause must be to justify such deliberate embryo destruction. Those who accept such destruction as part of a procreative project accept that the creation of new life is a cause good enough to justify such a course of action. Since most people believe that the saving of existing life takes priority over the creation of new life, research directed towards life-saving therapies or the production of those therapies must justify embryo loss if reproduction does (Harris 2002, 2003a).

Embryo-Sparing Assisted Reproduction Technologies

It might be said that there is a difference between justifying the embryo loss entailed in reproduction and justifying the embryo loss entailed in life-saving therapies. Researchers who engage in assisted reproduction create and destroy an unnecessarily high number of embryos and do not do everything they can to actualise the potential of each embryo created, whereas in sexual reproduction the sacrifice is not intentional, and it happens despite us doing everything we can to realise the potential of the embryo. Embryo loss may indeed not be intentional sacrifice, and it may not attend every pregnancy. However, the loss of many embryos, including “healthy” ones that would have grown to term in other circumstances (e.g., another womb or other conditions in the womb), is the inevitable consequence of the vast majority of (perhaps all) pregnancies. If creating a single embryo by IVF became a reliable technique for procreation, it would be interesting to know whether those from a rather inappropriately termed “pro-life” position would feel obliged to use this method rather than sexual reproduction because of its embryo-sparing advantages (Harris 2003a). We say inappropriately termed “pro-life” because those who regard themselves as “pro-life” so often support positions that can only be thought of as anti-life and that moreover are profligate of human life and safety (Harris 2003b). IVF could also be combined with the possibility to grow embryos to term in artificial wombs, which, when perfected, will offer a safer environment for the embryo than a woman's womb. Furthermore, the use of artificial wombs could save the lives of embryos left over from IVF treatments (Kaczor 2005) or from totipotent cells obtained through embryo splitting in order to maximise human procreation. Would “pro-lifers” support these technologies? It looks as though there would indeed be a strong moral obligation to abandon sexual procreation and use only embryo-sparing ART. If such an improvement in ART occurred, this development would seem to make using it mandatory for those who believe that the embryo is one of us. It is interesting that so-called pro-lifers are not investing heavily in technologies to this end in the hope that sexual reproduction could eventually be entirely replaced by an embryo-sparing method of reproduction.

What follows from all this? It is difficult to see how most people could live lives that are today accepted as normal while maintaining a strict “pro-life” position or by acting consistently to protect embryos. The alternatives seem clear. We, humankind, must accept that human embryos are deeply ambiguous and problematic entities of a kind whose lives or “dignity” simply cannot be protected in ways consistent with other values that we hold. Does this mean we should accord a moral status to early embryos that is compatible with their use for purposes considered morally at least as important as creating new life, such as saving lives through stem cell research?

Ethical Inconsistency in the Embryonic Stem Cell Debate

The hESC debate has been characterized by a search for compromise positions that seek to find a middle ground between the position that embryos can be used for morally important purposes and the position that embryos should be protected as one of us and, consequently, their use for stem cells should be opposed (see Tännsjö in this collection). Most countries do not want to forego the potential benefits of hESC research and have adopted regulations based on one of the main compromise positions. Some countries allow the use of hESCs but not the derivation process, as the latter involves killing embryos. There are several variations of this compromise, such as restricting the use of hESCs to those derived before a set date, with private money, or abroad.⁸ Other countries are a bit more liberal, allowing the use of embryos left over from infertility treatments and no longer used in a parental project but rejecting the creation of embryos solely for the purpose of stem cell research.⁹

We have seen, however, that the view that embryos have a moral status incompatible with their instrumental use as stem cell source is not supported by strong argumentation and cannot be maintained consistently with other values or with how most people live their lives. Why, then, should we look for compromises that try to satisfy this view and, by doing so, slow down or block hESC research?

Many people may have some respect and care for some kind of protection of the embryo, but these feelings can change and often depend on people's intentions, in particular on whether the embryo is included in a parental project. Moreover, there are forms of respect and deference that are less absolute and that admit of gradations. The respect one has for an entity does not exclude it, provided that a meaningful argument is presented, from being used as a resource for a goal that is believed to be important (which is comparable with research on cadavers). A way to show respect to early embryos is by ensuring that they are used with care in research that incorporates substantive values, such as the alleviation of human suffering, which is in accordance with the widely accepted principles of beneficence, non-maleficence, and proportionality (see Manninen in this collection). Well-regulated hESC research can be consistent with these widely accepted principles. Of course, disagreement can still exist on the scope of the principles. However, arguments for establishing scope, such as stating that the protection of the embryos falls within the scope of the principle of non-maleficence, like the principles themselves, have to meet the standards of adequacy and validity required for arguments to hold true (Harris 2005a).

We have shown that no strong argumentation that meets these standards has been provided. First, the embryo is an ambiguous entity, which casts serious doubt on the arguments claiming its full protection. Secondly, those who claim the embryo should be protected as if it was a person are committed to a position even they do not uphold in their practices. No society treats the embryo as one of us or has ever done so. Third, views that defend the protection of the embryo in virtue of its potentiality to become a person fail, and, finally, the embryo does not have any rights or interests to be protected. Moreover, the validity of not considering the embryo as one of us is corroborated by scientific evidence and analysis of the (changing) and ambiguous characteristics of an early human embryo. New technical possibilities to manipulate embryos and assemble and reassemble them will continue to challenge our views about the embryo and what constitutes its value. Even if these views cannot conclusively be shown to be fallacious, they can at least be shown to be inconsistent, erroneous, or suspect.

Compromising on the moral status of embryos is impossible. Once one accepts certain uses of embryos it will always be hard, if not impossible, to justify prohibiting other uses for morally equivalent purposes. This is the main reason why the compromise positions in the hESC debate cannot be sustained (Devolder 2006b). Given that many are willing to treat the embryo as a means in other practices, and that hESC research holds great potential to benefit many people, one cannot but conclude that hESC research is permissible and, given its immense promise for alleviating human suffering, even obligatory (Devolder and Savulescu 2006).

Moral diversity on deeply held beliefs about the embryo must be respected. But powerful moral reasons to pursue research should not be drowned by the powerful reasons we have to respect people's fundamental views (Harris 2005b). Respect for morally diverse views does not require that we as a society prohibit or severely restrict hESC research. Pursuing and supporting hESC research is in the interests of all people, now and in the future. The compromise positions that restrict hESC research cannot be sustained and therefore cannot offer a legitimate moral basis for stem cell policy.

The way we respond to practical dilemmas can differ from our professed beliefs about abstract problems. These revealed beliefs should influence our ideas about what is good or bad. They say something about the morality we accept. (Many practices that are now routine and uncontroversial, such as blood transfusion and organ transplantation, were once considered unethical by many). This morality should be projected onto the principles we profess. If our reactions in real-life situations are inconsistent with the beliefs we profess, then these principles and/or their scope may have to be reviewed and modified in the light of the morality we accept or wish to retain. It is important to realise that there is a two-way interaction between the principles we accept and our practices. There is nothing fundamentally wrong with temporal inconsistency, as it is part of moral development. There is something seriously wrong, however, with knowingly being inconsistent. Just as we cannot force people to act morally, we cannot force them to apply their principles consistently. What we can do, however, is point out inconsistencies through critical analysis of the arguments and by properly informing people through an honest and open debate. Justifying a prohibitive or restrictive stem cell policy by citing a moral position that is not based on solid arguments and by employing principles that are not applied consistently in practice is hypocritical and also, in a sense, “instrumentalises” ethics in order to find a purely political compromise.¹⁰