Disability, Enhancement and the Harm -Benefit Continuum more

Ch_03.qxd 11/21/2005 3:46 PM Page 31 3 Disability, Enhancement and the Harm–Benefit Continuum∗ LISA BORTOLOTTI AND JOHN HARRIS Introduction Suppose that you are soon to be a parent and you learn that there are some simple measures that you can take to make sure that your child will be healthy. In particular, suppose that by following the doctor’s advice, you can prevent your child from having a disability, you can make your child immune from a number of dangerous diseases and you can even enhance its future intelligence. All that is required for this to happen is that you (or your partner) comply with lifestyle and dietary requirements. Do you and your partner have any moral reasons (or moral obligations) to follow the doctor’s advice? Would it make a difference if, instead of following some simple dietary requirements, you consented to genetic engineering to make sure that your child was free from disabilities, healthy and with aboveaverage intelligence? In this paper we develop a framework for dealing with these questions and we suggest some directions the answers might take. First, we shall contrast our own ‘harmed condition’ account of disability1 with the social conception of disability put forward by Reindal2 and Koch3 by revisiting the distinction between conditions which are ‘socially disabling’ and conditions which are disabling irrespective of any particular social environment. In our framework, it makes sense to claim that certain conditions of the person can be ∗ The authors acknowledge the stimulus and support of the ‘European project on delimiting the research concept and research activities (EURECA)’ sponsored by the European Commission, DGResearch as part of the Science and Society research programme–6th Framework in the preparation of this paper. 1 J Harris, Wonderwoman and Superman: Ethics and Human Biotechnology (Oxford, OUP, 1992) 84–97 and J Harris, ‘Is Gene Therapy a Form of Eugenics?’ (1993) 7 Bioethics 178, 180. 2 SM Reindal, ‘Disability, Gene Therapy and Eugenics—A Challenge to John Harris’ (2000) 26 Journal of Medical Ethics 89–94. 3 T Koch, ‘Disability and Difference: Balancing Social and Physical Constructions’ (2001) 27 Journal of Medical Ethics 370–76. Ch_03.qxd 11/21/2005 3:46 PM Page 32 32 Lisa Bortolotti and John Harris regarded as disabling given the physical or social environment in which the person is embedded when they are harmful conditions that a rational person would have a preference not to be in. Our account is preferable to the social conception of disability, because it can explain how certain conditions of the person remain disabling even after issues of discrimination are addressed and society is free from prejudice which harms the subjects. Moreover, the ‘harmed condition’ account of disability does not rely on the problematic assumption that disability is a deviation from normal species functioning or from species typical functioning. From the perspective of a ‘harmed condition’ account of disability, we shall then proceed to explore the relation between disability and enhancement, given our commitment to the harm–benefit continuum. The harm–benefit continuum is the idea that: ‘[T]he reasons we have not to harm others or creating others who will be unnecessarily harmed are continuous with the reasons we have for conferring benefits on others if we can.’4 It would appear that, if we have moral reasons to prevent people from being in disabling conditions, we might also have moral reasons to improve their conditions, whether they are disabling or not. Further questions are whether these moral reasons are (in at least some cases) moral obligations and whether it is wrong not to enhance individuals whom we could thus benefit. We shall assess some common objections to enhancement and conclude that there are at least three possible ways of conceiving the ‘morality of enhancement’ that are compatible with our account of disability. A. Against Social Conceptions of Disability John Harris has suggested that conditions are disabling if they are physical or mental conditions that constitute a harm to the individual which a rational person would wish to be without.5 According to this account, disabling conditions are not necessarily harmful relative to normal species functioning or species typical functioning. In what sense, then, are conditions such as deafness and Down’s syndrome disabling? What all disabling conditions have in common is that, to some extent, they cause harm to the people who are in them (by exposing them to risks, impairing them in what they do, limiting their opportunities or preventing them from having experiences that are worthwhile). On this view, disabling conditions constitute a disadvantage with respect to relevant alternatives, not 4 J Harris, ‘One Principle and Three Fallacies of Disability Studies’ (2001) 27 Journal of Medical Ethics 383, 386. 5 J Harris, Wonderwoman and Superman: Ethics and Human Biotechnology (Oxford, OUP, 1992) 84–97 and J Harris, ‘Is Gene Therapy a Form of Eugenics?’ (1993) 7 Bioethics 178, 180. Ch_03.qxd 11/21/2005 3:46 PM Page 33 Disability, Enhancement and the Harm–Benefit Continuum 33 necessarily with respect to the conditions of the typical human. The reason why the notion of ‘normal species functioning’ is unhelpful in defining disabilities is that it would make disability too narrow. Changing environmental factors, or new discoveries about the onset of serious diseases, for instance, might make it the case that typical conditions of our species become disabling. Consider the case of white-skinned people in an environment where the ozone layer is seriously depleted. They would probably be much more vulnerable to skin-cancer than brown- or black-skinned people and this would constitute a disadvantage. Being white-skinned is no more or less ‘normal’ than being blackskinned relative to our species functioning. Indeed, if we had evidence that women live longer because they are less vulnerable to heart disease than men,6 we might consider being a man rather than a woman a disabling condition! What reasons are there to adopt this conception of disability? First, it is sufficiently broad to cover all the harmful conditions that we might intuitively regard as disabling, whether the harm be primarily caused by medical conditions of the person, genetic or environmental factors or social context. It has clear advantages compared to a merely social conception of disability, as it is not committed to the rather implausible claim that all disabling features of the condition would disappear if society were inclusive and free from discrimination or prejudice. The definition of disability formulated by the Union of the Physically Impaired Against Segregation is an example of how disabilities can be conceived of as effects of social organisation: Disability: disadvantage or restriction of activity caused by a contemporary social organization which takes no or little account of people who have physical impairment and thus excludes them from the mainstream of social activities.7 Whereas it is certainly true that certain attitudes in society towards people who are perceived as different cannot but make things worse for disabled people, in many cases, perhaps most cases, their harmful condition—for instance, deafness of Down’s syndrome—would persist once society had been reformed. The only case in which a disabling condition would cease to be harmful if society were reformed is the case of a disability entirely caused by the adverse social context. For instance, history teaches us that being born female in 19th century Europe would have been a disabling condition when compared to being born male. Exceptions apart, women were prevented from exercising any form of autonomous decision-making, often did not have the opportunity to have a 6 The Guardian, 11 January 2005, reports that: ‘A group of scientists may finally have solved the mystery of why wives outlive their husbands after discovering that women’s hearts age better than men’s.’ 7 Reported by M Oliver, ‘Defining Impairment and Disability: Issues at Stake’ in C Barnes and G Mercer, (eds), Exploring the Divide: Illness and Disability (Leeds, The Disability Press, 1996) 29–54. Ch_03.qxd 11/21/2005 3:46 PM Page 34 34 Lisa Bortolotti and John Harris proper education and so on. Similarly, being born today in a developing country rather than in a developed country is, for most, a disabling condition. These harmful conditions would cease to be regarded as disabling if we could operate on the social, political and economic context. The ‘harmed condition’ conception of disability can recognise that disabling conditions might have a variety of causes without being committed to the view that by operating appropriately on social factors all disabilities would disappear. B. Moral Reasons against Creating Individuals with Disabling Conditions What are the consequences of this account of disability for the morality of reproductive choices? According to the definition of disability that we have defended in this chapter, a condition is disabling if it constitutes a harm to the individual which a rational person would wish to be without. Moreover, we argued that disabling conditions constitute a disadvantage with respect to relevant alternatives. Given this account, it is easy to see how we have moral reasons to prevent or eradicate disabling conditions, as part of our commitment to the basic moral principle of avoiding unnecessary harm. This means that, when we have the choice, we should bring into existence people without (known) disabling conditions rather than people with such conditions. But this of course says nothing about whether these moral reasons give rise to a moral obligation or about the way in which we should carry out such a moral obligation. There are many different ways in which creating a person with a disability can be avoided, including postponement of conception, behaviour modification, gene therapy, selection between pre-implantation embryos and abortion.8 One might recognise the moral obligation to prevent or eradicate disabilities and still object on moral or other grounds to the methods by which the obligation can be carried out. Moreover, the strength of the obligation might vary relative to the context of the disabling condition, and to the degree of harm that the disabling condition is likely to cause to future persons. 1. Serious Disability In the case of a person with very serious disability, there would be moral reasons not to bring her into existence. Imagine a child diagnosed in early pregnancy with 8 For a more comprehensive list and discussion, see J Harris, ‘Should We Attempt to Eradicate Disability?’ (1995) 4 Public Understanding of Science 233–42. Ch_03.qxd 11/21/2005 3:46 PM Page 35 Disability, Enhancement and the Harm–Benefit Continuum 35 holoprosencephaly,9 a structural brain disorder in which the two halves of the brain have failed to separate. In the full blown condition which we are imagining here to be present, there is severe brain damage and a high risk of intractable seizures. The condition is incompatible with long term survival. In such a case the child would have a hopeless prognosis and a brief life distressing for the child and the family with no hope of palliation, prevention or cure. Here there are surely strong moral reasons not to bring such a child into existence. 2. Slight Disability In the case of a person with a very minor disability, which is not thought of as truly disabling condition, there might none the less be moral reasons to eradicate the disability by selecting between pre-implantation embryos for example but perhaps not by abortion. Suppose that a woman has six pre-implantation embryos in vitro awaiting implantation. Pre-implantation genetic diagnosis (PGD) has revealed that three have asthma and three seem healthy. Which should she implant?10 Does she have any moral reasons to avoid implanting those who will develop asthma? Notice two features of this case. The woman is under no moral, nor any legal obligation to implant any of the embryos. The decision to implant some or none is entirely within her unfettered discretion. She doesn’t have to offer legal, moral or any other justifications to anyone if she decides to implant none of the embryos. Can she say, ‘It is a matter of moral indifference whether or not my resulting child has asthma and therefore I have no reason to select the healthier embryos’? This seems implausible. True, her choice is relatively free. None of the embryos has a right or an entitlement to be chosen rather than the others, since none is a person, none has begun the sort of biographical life that would give it interests, none has become a moral agent or a legal person.11 Nevertheless, she has a reason to do what she can to ensure that the individual she chooses is as good an individual as she can make it. She has a reason therefore to choose the embryo that is not already harmed in any particular way, however slightly, and that will have the best possible chance of a long and healthy life and the best possible chance of contributing positively to the world it will inhabit. If on the other hand she chooses to implant an individual destined to suffer an illness, however mild, she will have created that illness and any harm that it will do. 9 We are indebted to our colleague Dian Donnai for information about this condition. Further facts may be found at: http://www.ninds.nih.gov/disorders/holoprosencephaly/holoprosencephaly.htm. 10 This example is borrowed from J Harris, ‘Reproductive Liberty, Disease and Disability’ (2004) Reproductive Medicine Online. 11 See the judgment of the Grand Chamber of the European Court of Human Rights in Vo v France, Appl no 53924/00, ECHR 2004, judgment of 8 July 2004. Ch_03.qxd 11/21/2005 3:46 PM Page 36 36 Lisa Bortolotti and John Harris Notice that this woman has the same reason to select against an embryo with asthma as her sister who is told that if she conceives immediately she will have a child with asthma but that if she postpones pregnancy and takes a course of treatment she will have a healthy child.12 The sister and her partner in this case have a reason to maximise the opportunities and minimise the disadvantages of any child they will have and therefore to modify their behaviour so as to try and conceive the healthiest embryo possible. But if they only learn in the course of an already established pregnancy that the child the woman is carrying will develop asthma, the parents might consider that avoidance of asthma does not justify an abortion, since the problem is relatively slight and life with asthma is well worth living. 3. Merely Social Disability Social factors can be disabling but this does not imply that the person making reproductive choices has an obligation, or even a reason, to prefer creating an individual without socially disabling conditions such as those determined by gender or skin colour. In these circumstances the choice that needs to be made, ie whether to attempt to change society or to avoid creating individuals with features that will be socially disabling in the expected social or political context, is not value-free or politically neutral. The strategy of choice for minimising expected disability may be difficult to arrive at for individuals. Reforming society of course is never an easy option or a fast process, and there are known cases of mothers who chose to, say, have white-skinned children to prevent them from being harmed by prejudice and discrimination. Notice that these considerations make social factors significantly different from environmental factors. Recall the example of white-skinned people in an environment where the ozone layer is so depleted that their chances to get skin cancer are significantly increased with respect to darker-skinned people. In that case, it seems that we would be happy to acknowledge the need to ‘adapt’ to the changed environmental conditions to have a better and longer life. However, in the case of (merely) socially disabling conditions it is not at all obvious that we have the same need to adapt to the current organisation of our social environment and there is no morally neutral way to deal with the problem so as to minimise or eliminate 12 This situation is often discussed with a view to the so-called ‘non-identity problem’ which trades , on the possible moral significance of the fact that any child with asthma which the sister conceives now would not have existed at all if the sister had waited and only had a child later. Some would argue that this prevents us from viewing the child with asthma as having been harmed by its mother’s choice not to wait, and perhaps also from viewing her choice as morally wrong. For an exposition and discussion of the ‘non-identity problem’, see D Parfit, Reasons and Persons (Oxford, Clarendon Press, 1984) part 4 especially 358ff, and specifically in the context of disability J Harris, Wonderwoman and Superman: Ethics and Human Biotechnology (Oxford, OUP, 1992) ch 3. Ch_03.qxd 11/21/2005 3:46 PM Page 37 Disability, Enhancement and the Harm–Benefit Continuum 37 the disadvantage of the individuals who are likely to be harmed by prejudice or discrimination. Many, including the present authors, might feel a moral obligation to affirm the entitlement to equal concern and respect of society and their fellow citizens both for themselves and for their children; but the rationality as well as the morality of this can have limits. Many European Jews, for example, made a conscious decision not to reproduce during the Nazi period because they knew that whether or not their children chose to identify themselves as Jews, the mere fact of ‘having Jewish parents’ was enough to render their children liable to terrible cruelty and premature death. Another consideration in making these decisions is whether the factors that render a certain condition disabling can be altered. In the case of social organisation, alteration is always possible, though perhaps not realistic in a short span of time. In the scenario with the seriously depleted ozone layer the assumption is that we cannot do anything about the environmental change. But we can imagine scenarios in which the environmental factors in relation to which a condition becomes disabling can be altered and therefore the same choice that we have in the case of merely socially disabling conditions can present itself. C. Disability and Enhancement Suppose we accept that there are moral reasons to prevent or eradicate disability when possible. Does this commit us to recognising that we also have moral reasons to enhance? This question arises with particular force for those who believe that there is a continuum between harms and benefits. On this view the reasons we have to avoid harming others are continuous with the reasons we have for conferring benefits on others if we can. All actions are re-describable as omissions and vice versa. The decision to save a life is the decision not to allow someone to die. The decision to kill is a decision not to refrain from killing. This is not just a semantic quibble but is rooted in both theory of action and theory of responsibility. Where we can protect people from harms, to elect not to do so is to become responsible for the harms we might have prevented. Suppose a child can be protected from risk of polio in an environment where polio is rife. The enhancement (a polio vaccination) is available free of charge but the parents refuse. The child contracts polio. The chain of causation as well as the chain of moral responsibility is as clear as it could be.13 If we accept the view that there is a harm–benefit continuum, and we have a moral reason to avoid causing unnecessary harm to others, we also have a moral 13 See J Harris, Violence and Responsibility (Kegan Paul London, Routledge, 1980) and J Feinberg, Harm to Others (Oxford, OUP, 1984), ch 3. Ch_03.qxd 11/21/2005 3:46 PM Page 38 38 Lisa Bortolotti and John Harris reason to confer benefits on others, and this moral reason can become a positive obligation where the costs to ourselves are reasonable given the level of benefit. This is supported by the intuitive analogy between disability and enhancement. If disabling conditions constitute a disadvantage with respect to some relevant alternatives, enhanced conditions constitute an advantage. For instance, enhancing strategies such as taking a ‘longevity pill’ are likely to confer a benefit on those who appreciate life and want to live as long as possible. When we save a life, by whatever means, we simply postpone death. Life-saving is just death postponement. This is a truth from which it follows that life-extending therapies are, and must always be, life-saving therapies and must share whatever priority life saving has in our morality and in our social values. So long as the life is of acceptable quality (acceptable to the person whose life it is) we have a powerful, many would claim an overriding, moral imperative to save life, because to fail to do so when we can would make us responsible for the resulting death.14 Moreover, one can easily imagine scenarios in which not enhancing a person’s condition amounts to creating a disability. In an environment in which most people have had their long-term memory enhanced by 20%, people whose memory has not been enhanced are at a disadvantage in some contexts. Or, if a safe and effective vaccine against HIV/AIDS were to be developed, those not protected would be at a severe disadvantage. It is true that there are situations where it is more difficult to re-describe the choice to provide a person with an enhancement as a choice to prevent or avoid risk of harm. If there existed a free-of-charge vitamin supplement which a pregnant woman could take to prevent future ageing-related hair loss of the male child she is carrying, and if she nevertheless chose not to take it, she does perhaps cause, and become responsible for that person’s hair loss in late adulthood. But some would still baulk at describing ageing-related hair loss as a harm or a debilitating disadvantage, even if we imagine a society where hardly anyone still loses their hair at an advanced age. We nevertheless believe that our argument concerning the harm–benefit continuum also applies to these situations. First, it is clear that a reason such as ‘wanting the best life possible for one’s child’ applies equally to the avoidance of harm, to enhancements that prevent risks of harms and to ‘pure’ enhancements, which are merely likely to make the life of the child happier or better in some small but significant way. Secondly, in any consideration of ‘harms’—even if one rejects, as we do, a purely circumstantial or ‘social’ conception of harm and disability in which harms are simply ‘conditions considered disadvantageous by society’—the difference between a harm and a mere neutral condition is often one of degree only. The line between harms and non-harms fluctuates and is malleable over 14 See J Harris, Violence and Responsibility (London, Routledge and Kegan Paul, 1980). Ch_03.qxd 11/21/2005 3:46 PM Page 39 Disability, Enhancement and the Harm–Benefit Continuum 39 time. We therefore think that any differentiation between, on the one hand, harmpreventing enhancements, and on the other hand, further enhancements, is unwarranted and that the harm–benefit continuum applies to any change that can be brought about in a person that a rational person would consider a change for the better. D. Who is Afraid of Enhancement? In the bioethical literature, the press, and even in recent cinematography, enhancements are viewed with great suspicion.15 In this section we are going to briefly review some of the most common arguments that are often employed to show that enhancing is unethical. By ‘enhancement’ we refer to any procedure aimed at conferring benefits on people by improving their conditions, from immunisation against serious diseases to increases in intelligence or memory. We shall come back to the distinction between different types of enhancement in the next section. Here it will suffice to note that many of the following considerations have been relied on to argue against the prevention of disabling conditions. 1. Safety This family of arguments includes two main empirical worries. First, there are concerns about the safety of the procedures involved. Secondly, there are worries about the limited amount of knowledge even experts have about the consequences of, say, genetic engineering in certain domains. Although there might be very good reasons to decide against enhancement out of concern about safety, this is not an argument that could ever by itself show that enhancing is unethical. Should the science improve and with it the safety of the procedures, there would no objection left to proceeding with enhancement. Notice that in some cases you can have a sophisticated ‘safety’ objection to enhancement. Norman Daniels gives an interesting example.16 Suppose that we learn that an enhancement of short-term memory would benefit many of our cognitive processes and that we have the opportunity to enhance short-term memory by operating on embryos. This is how Daniels argues that we should not do it: 15 Let us offer just one example. In the movie ‘Gattaca’ (1997), a world where most people are genetically enhanced is presented as cold and soulless and those who are not enhanced are discriminated against by being prevented from playing certain valued roles in society. 16 N Daniels, ‘Can Anyone Really be Talking About Ethically Modifying Human Nature?’ in J Savulescu and N Bostrom, (eds), The Enhancement of Human Beings (Oxford, OUP, in press). Ch_03.qxd 11/21/2005 3:46 PM Page 40 40 Lisa Bortolotti and John Harris What else might we have to know to proceed ethically with the modification? We would have to know a lot more. We would have to know that the increased short-term memory involved here actually plays a role in enhancing the more complex cognitive task rather than, in the case of this pattern, interfering with it. Perhaps it creates too much noise that interferes with the complex task, or does so in some range of environments not tracked in the original finding of the association. ... Without some clear sense of these complex issues, we could not have any confidence that improving the component capability has the intended or desired effect on the more complex one. And all of this information goes well beyond the standard worry that the intervention itself carries with it risks that non-interference lacks.17 Daniels’ concern is that enhancements might not really improve the quality of life. Here the objection is context-specific. Daniels does not just reiterate the idea that there are always risks involved in changing something that is working well enough. He is saying that, given the nature of genetic modifications and the complex way in which we would need to assess their consequences, the fact that the capacity or trait to be enhanced is a necessary condition for better performance does not mean that by enhancing it we would produce an overall better offspring. Although his concerns are definitely to be taken into account, we do not believe that they justify the extremely conservative conclusion that enhancement should never be tried out. And this leads us to a brief examination of the precautionary principle. 2. The Precautionary Principle There is widespread acceptance of the (nonetheless incoherent) precautionary principle.18 The idea is the status quo has done well for us and is likely to continue to do well for us. Therefore in the absence of reliable data, any proposed changes are to be considered of uncertain consequence and presumed disastrous or, more weakly, dangers that are to be considered more probable and of greater magnitude than the expected benefits. The precautionary principle therefore applies and requires that we leave things well alone. A number of questionable assumptions are involved here. If enhancements might alter the course of evolution, or even effect permanent change, then one assumption that the precautionary principle requires is that our present point in evolution is unambiguously good and not susceptible of improvement. Secondly, it is assumed that the course of evolution, if left alone, will continue to improve things for humankind or at Ibid. In this section we draw on J Harris and S Holm, ‘Extended lifespan and the paradox of precaution’ (2002) 27 The Journal of Medicine and Philosophy 355–69 in which the incoherence of the principle is detailed. 18 17 Ch_03.qxd 11/21/2005 3:46 PM Page 41 Disability, Enhancement and the Harm–Benefit Continuum 41 least not make them worse. The incompatibility of these two assumptions is obvious. The status quo is a result of evolutionary change. Unless we can compare the future progress of evolution uncontaminated by manipulation of the human genome with its progress influenced by any proposed genetic manipulations we cannot know which would be best and hence where precaution lies.19 Put more prosaically, it is unclear why a precautionary approach should apply only to proposed changes rather than to the status quo.20 In the absence of reliable predictive knowledge as to how dangerous leaving things alone may prove we have no rational basis for a precautionary approach which prioritises the status quo. This point re-enforces our earlier remarks about the continuum between removal of disadvantage and implementation of advantage. 3. Priority of the Natural There is a strong and tenacious view that the natural is good and the unnatural bad, that we should, in short, give priority to the natural over the artificial. This seems untenable, as the following extract from a recent paper by Julian Savulescu and John Harris shows:21 Witness the obsession with natural foods over genetically modified and other artificial, Frankenstein foods. The belief that the natural should have priority over the artificial, though common, is mistaken. A child born by assisted reproduction is of the same kind with the same moral status as a child born by natural reproduction. In so far as naturally occurring foods are safer or healthier, there is a reason to prefer them. But in many cases, artificial preparations are healthier and safer. Indeed, they are preferred. The drug digoxin is used for heart conditions. It is always given in a highly purified pharmaceutical form as digoxin even though it occurs naturally as digitalis in the foxglove plant. Natural reproduction may be more fun than artificial reproduction. That is a reason to prefer it. Natural processes may be less costly or do less damage to the environment. These are reasons to prefer them. But there is no reason to prefer a natural process per se to an artificial process. Human beings are, after all, a part of the natural world. The mere fact that something is natural is not a reason to prefer it. The natural per se is morally neutral. Sometimes natural events are good, like a brilliant sunset or an abundant harvest. But often the natural does great harm—disease, pestilence, 19 While some maintain that human evolution of the Darwinian sort is at an end because most humans now survive long enough to reproduce, this view overlooks the role of parasites in evolution. It also of course ignores our deliberate interventions in the evolutionary process. 20 See FM Cornford, Microcosmographica Academica (London, Bowes and Bowes, 1908) ch 7: ‘Every public action which is not customary, either is wrong, or, if it is right, is a dangerous precedent. It follows that nothing should ever be done for the first time.’ 21 J Savulescu and J Harris, ‘The Creation Lottery: Final Lessons from Natural Reproduction: Why Those Who Accept Natural Reproduction Should Accept Cloning and Other Frankenstein Reproductive Technologies’ (2004) 13 Cambridge Quarterly of Healthcare Ethics 90–96. Ch_03.qxd 11/21/2005 3:46 PM Page 42 42 Lisa Bortolotti and John Harris floods, hurricanes, fire, landslides and the like can cause massive loss of human life. One might characterize the practice of medicine as the comprehensive attempt to frustrate the course of nature, because people naturally fall ill, are invaded by natural organisms like viruses and bacteria, and naturally die at a young age, often as babies. If we always prioritized the natural we would have to abjure the practice of medicine and the discoveries of medical science including vaccines and antibiotics. Our current fascination and worship of the natural should be tempered. We should remember how Thomas Hobbes so famously described life in a state of nature, ‘... and which is worst of all, continual fear, and danger of violent death; and the life of man, solitary, poor, nasty, brutish and short.’22 4. Self-evolution and the Post-human Future There are two objections that are often run together. One is the ‘playing God’ objection, the idea that by enhancing in certain ways we are guilty of hubris. Humans are not supposed to create better humans, because that would be arrogance on their part. They should just accept what God or Nature has given them without attempting to better it. This is not a very interesting objection and has been discussed at length, so we propose to move on to the second point. What are the consequences of playing God? By intervening on the genes, we might change human nature and self-evolve. What is wrong about that? One idea is that the human species as it is should be preserved. This stems from the belief that there is something intrinsically good about being human. As it happens, we also think that there is something very valuable about human persons, but we believe that what is really valuable in them is the capacity of persons to be aware of themselves, make decisions for themselves in a rational and autonomous way and have complex feelings and emotions. The fact that all the persons we know are human is just an accident. If we did find those characteristics of persons in non-humans, we would (or should) still appreciate them and cherish them. Our value and moral status does not depend on the species we belong to, but on the fact that we are persons and, as such, have interests of a certain kind. Some worry that, say, genetic prejudice might ensue and human rights might be denied to those who are no longer human as a consequence of genetic modification or self-evolution.23 But being human is neither necessary nor sufficient for having rights, as Juengst himself notices. Dead humans have no rights, whereas arguably intelligent animals do. If we recognise that what justifies granting rights 22 T Hobbes, M Oakeshot, (ed), Leviathan (Oxford, Blackwell, 1960) 82, cited in Harris and Savulescu ibid. 23 E Juengst, ‘“Alter-ing” Human Nature? Misplaced Essentialism in Science Policy’ in J Savulescu and N Bostrom, (eds), The Enhancement of Human Beings (Oxford, OUP, in press). Ch_03.qxd 11/21/2005 3:46 PM Page 43 Disability, Enhancement and the Harm–Benefit Continuum 43 to individuals is not the species they belong to but the interests they might have, then the question of whether to grant rights to post-humans is easily solved. The ‘human’ in the phrase ‘human rights’ is just supposed to highlight that differences of race, gender and wealth are not relevant to the question whether someone should be granted those rights. If we take seriously the concern that some philosophers have for another kind of prejudice or discrimination, speciecism, then the ‘human’ has to go and ‘human rights’ will just be the ‘rights of persons’. These objections, like the arguments in the next sub-section, refer to the political and social consequences of widespread enhancements and can be partially if not completely addressed if we operate in a framework in which we recognise the validity of the principle of genetic equity.24 This principle is simply an elaboration of principles of equality that have become well recognised and articulated in recent years. Such a principle highlights the requirement that each of us be treated as an equal, shown the same concern, respect and protection as is accorded to any other person.25 As John Harris and John Sulston have recently argued:26 All democratic societies must treat their citizens as equals, according to each the same concern, respect and protection as is accorded to any, both in the protection of their laws and customs but also in the allocation of public resources or indeed in the ways in which private resources are deployed to secure public goods. For example, if a private commercial organization, say a pharmaceutical company, develops a vital treatment or vaccine it is the obligation of societies to ensure that access to this drug or treatment is available in a way that does not unfairly discriminate against some sections of society, for example is not available only to those of a particular race or religion or to members of the governing party. The scope of this principle might of course be contested, for example as to whether or not it covers discrimination on the basis of ability to pay, or the degree of benefit to be derived from the treatment.27 Both in view of the above arguments but also because we believe equal respect for all persons is an integral and ineradicable part of democratic theory and 24 See J Harris and J Sulston, ‘Genetic Equity’ (2004) 5 Nature Reviews Genetics 796–800; and C Farrell, ‘The Genetic Difference Principle’ (2004) 4 American Journal of Bioethics 21–28. 25 See S Giordano, ‘Do Elderly People Have Human Rights? Respect for Equality and the Treatment of the Older Person. Declarations of Human Rights and Age-Based Rationing’ (2005) 14 Cambridge Quarterly of Health Care Ethics 83–93; J Harris, ‘Does Justice Require that we be Ageist?’ (1994) 8 Bioethics 74–84; J Harris, ‘Could We Hold People Responsible for their Own Adverse Health?’ (1996) 1 Journal of Contemporary Health Law and Policy 100–6; J Harris, ‘What is the Good of Health Care?’ (1996) 10 Bioethics 269–92; and generally R Dworkin, Taking Rights Seriously (London, Duckworth, 1977). For early philosophical discussion of the principle of equality, see Plato, AD Lindsay, (tr), The Republic (Dent/Everyman, 1993) Book V, 451–57 [Stephanus]; JJ Rousseau, Du Contract social, ou Principes du droit politique (Amsterdam, Chez Rey MM, 1972); JS Mill, Utilitarianism (London, Mary Warnock, 1960). Collins Fontana 26 J Harris and J Sulston, ‘Genetic Equity’ (2004) 5 Nature Reviews Genetics 796–800. 27 Original footnote omitted. Ch_03.qxd 11/21/2005 3:46 PM Page 44 44 Lisa Bortolotti and John Harris hence an essential feature of all liberal democracies, we believe that a principle of genetic equity should be explicitly added to our understanding of the requirements of both equity more generally and of the requirements of democratic polity. Societies have a fundamental obligation to ensure the protection of the life, liberty and health of each citizen impartially, and to provide access to beneficial health care, and to the fruits of research, on the basis of individual need so that each has an equal chance of flourishing to the extent that their individual genome and personal health status permits. This equal chance of flourishing should be protected by the state, regardless of such arbitrary features as race, gender, genome, degree of disability, wealth and power, religious belief, skin colour etc. Genes (or genetic constitution) are clearly some of the most important examples of those personal characteristics which most clearly make a difference to both how people are able to function and often also as to how they are perceived, but they are also features that as yet have not been adequately incorporated into our understanding of the requirements of both equity and democracy. Harris and Sulston therefore propose a ‘principle of genetic equity’, which they formulate as follows: Humans are born equal; they are entitled to freedom from discrimination and equality of opportunity to flourish; genetic information may not be used to limit that equality. It follows that neither genetic constitution nor genetic information should be the basis of discrimination or stigmatization of an individual, family or group. No one’s genes, or genetic information about them, can or should derogate from their equal standing and dignity in the human community and their equal entitlement to the concern respect and protection of others or of society.28 5. Unfair Advantage and Diminished Agency It is a common thought that some enhancing strategies such as genetic engineering are going to be very expensive and only the better-off in society will be able to afford them. As a consequence, the current divisions in society will become even less bridgeable. Notice that this is not an ethical objection to enhancing as such, but a concern about the distribution of resources. Actually, the worry about the ways in which enhancements will be distributed implies that enhancements are perceived as a good thing. Some believe that the practice of enhancing and genetically engineering capacities will lead to a revision of our conception of agency.29 Agents typically enjoy a certain Ibid, p 798. M Sandel, ‘What’s Wrong with Enhancement’ (2002), Council of Bioethics, available online at http://www.bioethics.gov.background.sandelpaper.html. 29 28 Ch_03.qxd 11/21/2005 3:46 PM Page 45 Disability, Enhancement and the Harm–Benefit Continuum 45 amount of freedom of action and are subject to judgments of praise for their achievements and of blame for their failures. But if the physical or intellectual achievement of the agent is only marginally due to effort and discipline and mainly produced by, say, a powerful drug, the achievement might no longer be a good reason to admire the agent. The argument is supposed to show that a pervasive use of enhancement might lead to a diminished sense of agency and responsibility. To assess the force of this argument one needs to be able to account for what the consequences of the practice of enhancement would really be for our conception of agency. Partly, this is an empirical question. We know what our current psychological reactions to illicit drug-taking by athletes are; we feel it is cheating. But the scenario in which everybody is given an opportunity to enhance their condition is significantly different and so our reactions might reflect that. It is not at all obvious that we would lose the sense of ownership of our own actions if the capacities that made it possible for us to achieve something desirable with our actions had been enhanced. One possible consequence of pervasive enhancement could be a ‘raising the bar’ effect that would subtract little from the merits of the personal achievements of the individual. That said, it seems as if the diminished agency objection is on to something. Suppose you are a runner and want to increase your speed by 20%. Also suppose that there are two methods by which you can achieve this target. You can take a pill that has an immediate enhancing effect on your speed or you can train for two months, three hours a day. (Notice that these are both enhancing strategies.) Now, you might have a morally relevant reason to prefer the hard way to the easy way. You might value self-discipline and think that you will grow as a person if you achieve this target by making a conscious effort to perfect your body during the next two months. You might believe that the sense of satisfaction you would get at the end of the training for having achieved the target is worth the time and the effort that are required. But all these valuable considerations do not amount to judging that it would be unethical for you to choose the easy option. E. Types of Enhancement It is not easy to provide a clear classification of types of enhancements. There are different ways in which the enhancement can be achieved and different aims that the enhancing strategies can have. 1. Methods One thought is that natural enhancing strategies are better than non-natural or artificial ones. But, first, the dichotomy between natural and artificial is not very Ch_03.qxd 11/21/2005 3:46 PM Page 46 46 Lisa Bortolotti and John Harris clear. Second, there don’t seem to be any good arguments for the view that the natural has moral value. Recall the example of the runner who wants to increase her speed by 20% and has a choice: she can either train for two months, three hours a day, or take an enhancing pill. What is the most natural of these two strategies? ‘Natural’ can refer to (1) the endowments we are born with and we have not acquired, (2) the features we regard as normal as opposed to those which result from illness, (3) the features which have remained the same as opposed to those that we have altered and (4) something that belongs to the world of nature and has not been processed or manufactured. One strategy does not strike us as obviously more natural than the other. It is true that regular training might not involve any of the artificial processing that might go into preparing the pill, but we do not seem to ethically object to this kind of artificial processing when we take medicines or eat food that we haven’t grown ourselves. Of course we might have preferences for avoiding excessive processing and opt for so-called ‘natural’ remedies or home-grown food when possible, but these preferences do not necessarily track moral reasons. We do no think that it is morally wrong to buy vegetables from a supermarket. And neither strategy is natural from the other points of view. Both strategies are aimed at the alteration of a feature (running speed) which is not innate but acquired. And running three hours a day is not something people normally do. 2. Aims There can be enhancing strategies whose main aim is disease prevention and increasing life expectancy and enhancing strategies that are aimed at improving cognitive abilities or physical appearance. While many feel an intuitive difference between these different goals for enhancement it is difficult to draw a clear line between them from an ethical perspective. Other things being equal, if it is not wrong to wish for something, it is not wrong, if we have the technology, to play fairy Godmother to ourselves and grant our wishes. So if it is not wrong to wish that our child were healthier, more intelligent or more beautiful it is difficult to see how it might become wrong to grant our own wish if we could, or of course for our fairy Godmother to grant it on our behalf. The difference of course between the various aims or goals of enhancement might rather lie in the risks that it would be worth running or justifiable to run in pursuit of them. While for example considerable risks might be justifiable in order to cure a terrible disease or protect ourselves from almost certain death in a pandemic, it would be difficult to justify exposing our children to risks simply to change the colour of their eyes or make them better tennis players. Ch_03.qxd 11/21/2005 3:46 PM Page 47 Disability, Enhancement and the Harm–Benefit Continuum 47 F. Is it Always Wrong not to Enhance? In the previous two sections, we have considered some standard objections to enhancements and challenged the view that, from a moral point of view, there is a sharp distinction between those enhancing strategies that have as their main objective a healthier and longer life and those enhancing strategies that are aimed at improving the cognitive capacities of a person. Here we face the question whether we have a moral obligation to eradicate disabilities and to enhance capacities where we safely can in human reproduction. Given our discussion so far, there are at least three positions that could be defended given the conception of disabling and enhanced conditions that we have presented, although not all of them seem compatible with the harm–benefit continuum and the thesis that we are responsible for both what we do and what we omit to do. 1. The Clear Moral Duty to Confer Benefits Option On this view, no matter how slight the disability or insignificant the enhancement, parents have powerful reasons, which are always moral reasons, to minimise harm for, or confer benefits to, the person they are bringing into existence (subject of course to a safe method of achieving this and to the unambiguously beneficial nature of the proposed outcomes). We can note that at the moment selecting between pre-implantation embryos in the context of medically assisted reproduction is reasonably safe. Dietary manipulation (folic acid for example) and avoiding smoking or alcohol are also safe. Genetic manipulation is as yet untried and unsafe although that may change in time. These moral reasons stem from the basic principles of beneficence and non-maleficence. 2. The Threshold Option On this view, parents have powerful moral reasons to prevent a disabling condition or enhance only when by not doing so they cause considerable harm to their children. This is a threshold option. In the context of reproductive choices, the parents’ actions cease to be morally neutral and become subject to moral approval or condemnation when their actions have significant effects on the person they are bringing into existence in terms of harmed conditions that can be prevented or other conditions that can be enhanced. If the disadvantage caused by not preventing a disabling condition or not enhancing another condition is below a certain threshold, then there are no moral reasons to act. However, if the disadvantage is significant, then there are. Ch_03.qxd 11/21/2005 3:46 PM Page 48 48 Lisa Bortolotti and John Harris 3. The Sliding Scale Option On this view, all actions are subject to moral scrutiny, not only those which have significant effects in terms of benefits and harms for future people. The idea is that parents always have powerful moral reasons to enhance and prevent disability, but there is an important difference between the clear-moral-duty option and the sliding scale. In the former, to confer benefits or to prevent harm is right and to withhold benefits or cause harm is wrong, no matter how slight the disability or valuable the enhancement. Here, whether the reason to act is a moral reason and the degree of moral appraisal or condemnation for our actions co-vary with the degree of benefit conferred or harm prevented. Therefore, the reasons to prevent the disabling conditions caused by holoprosencephaly have a much greater moral impact than the reasons to enhance intelligence by say 15%. We are not talking here about the degree of strength of the motivating reasons, which can of course vary equally in the three options described, but about their being moral reasons in the first place. Which of these positions is preferable? An advantage of the clear-moral-duty option is that it is a coherent and simple option that seems to be compatible with our conception of disability as a harmed condition and with the harm–benefit continuum. The problem with this option is that from a moral point of view some distinctions that many find intuitively strong (distinctions between preventing serious disability and enhancement for the purpose of minor benefits) are not transparent. Not to prevent disability where possible is clearly wrong, but so is withholding an enhancement, and both are moral issues. Of course where there are risks to the preventive or enhancing strategy, then the smaller the benefit the less it will rightly seem to be worth running risks. But apart from safety concerns, can the distinction between the two cases, which is intuitively recognised by many, be rationally justified, given our commitment to the harm–benefit continuum? We think that the distinction between preventing serious harm and preventing minor harm matters, but we believe that it is sufficient to mark such a distinction by considering the relative weight of the moral reasons we have. The moral reasons to prevent serious harm will be always stronger, all things being equal, than the reasons to confer minor benefits, although both sets of reasons will be morally relevant. There are two serious problems with the threshold position. One is epistemological. It is never easy to measure how harmful or beneficial a condition is, as it cannot always be done out of context or inter-subjectively. This kind of calculation becomes even harder when applied to future people, whose interests and inclinations we ignore. The loss of a finger might be more significant for someone who may develop the ambition to become a great pianist than for someone who will have no interest in playing musical instruments, although it is disabling for both. The other problem is whether this option is really compatible with the Ch_03.qxd 11/21/2005 3:46 PM Page 49 Disability, Enhancement and the Harm–Benefit Continuum 49 harm–benefit continuum. If we recognise that a condition is harmful for someone, it would seem to follow that there are moral reasons to prevent that condition independently of the extent to which it is harmful. Can actions aimed at preventing harm ever be morally neutral? The sliding scale option shares some of the epistemological problems that affect the threshold option but is overall more compatible with the harm–benefit continuum. What seems unattractive about this option is the conclusion that some reasons to act can be only partially moral reasons. Although it is perfectly reasonable to claim that we might have a variety of reasons to act in a certain way, it is more puzzling to believe that each of our reasons to act is only partially a moral reason. G. Conclusion In this paper we defend the view that there are moral reasons to prevent disabling conditions and to adopt enhancing strategies in reproduction, on the basis of our conviction that there is a clear moral duty to confer benefits and avoid harms. This duty is of course moderated by the usual risk/benefit calculation that we all have to make about every decision we face. First, we have defended a ‘harmed condition’ account of disability. Although there is always a social element to the disabling conditions people suffer from, some conditions seem to remain disabling even when the social context changes for the better. Since disabilities are harmed conditions, the moral reasons we have to prevent harm are also reasons to avoid creating disabling conditions. Our commitment to the harm–benefit continuum also suggests that we have moral reasons to confer benefits on our children when this can safely be done. This brings us to the debate on enhancements. We have assessed some of the common arguments against enhancements and found that either they were unjustified or that they did not offer any moral reason to reject enhancements. Finally, we have reviewed three possible approaches to the morality of enhancements and suggested that the most coherent option is to maintain that there are always moral reasons to enhance, although of course the strength of those moral reasons depends on the extent to which the condition to be enhanced would be an advantage to the future person.