Article Text


No sex selection please, we’re British
  1. John Harris
  1. Correspondence to:
 John Harris
 Institute of Medicine, Law and Bioethics, School of Law, Williamson Building, University of Manchester, Oxford Road, Manchester M13 9PL;

Statistics from

There is a popular and widely accepted version of the precautionary principle which may be expressed thus: “If you are in a hole—stop digging!”. Tom Baldwin, as Deputy Chair of the Human Fertilisation and Embryology Authority (HFEA), may be excused for rushing to the defence of the indefensible,1 the HFEA’s sex selection report,2 but not surely for recklessly abandoning so prudent a principle. Baldwin has many complaints about my misrepresenting the HFEA and about my supposed elitist contempt for public opinion; readers of this exchange will decide for themselves.


Baldwin begins with a piece of wishful thinking:

Harris objects that in this recommendation “an absurdly high standard of caution is employed”, since a theoretical risk is associated with almost all medical procedures. This objection is misplaced: as paragraph 142 of the report indicates, the phrase “theoretical risk” is to be understood here in the light of the earlier discussion of the risks arising from the fact that flow cytometry exposes sperm to laser energy, a procedure which is known to be liable to damage DNA.

Paragraph 142 does not make that clear. It does indeed refer back to a set of earlier paragraphs but these give, if anything, an upbeat assessment of the safety of flow cytometry. Paragraph 121 states: “However whilst potentially less intrusive, and with potentially lower risk to the health of patients, flow cytometry …” (my emphasis).2 But even if the overall burden of the report does indicate unresolved fears, the standard is still absurdly high. However, so far from endorsing the report’s judgement that flow cytometry has “potentially lower risk to the health of patients”, Baldwin now regards the risk of flow cytometry as “serious”1:

Since the application of flow cytometry to humans is a new procedure, the risk of human genetic defects caused this way is still only “theoretical”; nonetheless it is serious enough to warrant caution—hence the recommendation that at present flow cytometry should be available only when its use brings clear medical benefits.

If Baldwin’s present view is right then we have been seriously misled by the safety assessment in the HFEA report and one is compelled to ask whether it can be consistent with the HFEA’s statutory requirement to have regard for the welfare of the child to be born, to permit children to be exposed to risks as serious as this even to secure medical benefits? After all people do not have to have children, and if there is a “serious enough” risk of damage, even medical benefits would probably not justify subjecting children to such a risk. Remember that the children exposed to the risks are not the ones whose health is being protected by the medical benefits. Indeed no child is being protected. The medical benefits result in different healthier children being born. These children would have been healthy in any event (had they been chosen) so do not benefit and the children selected against certainly do not benefit. The benefit is a public health benefit, a eugenic benefit, not a benefit to the child to be born.3


Next Baldwin complains “it is worth emphasising here that Harris’s description of the HFEA inquiry as one which involved only ‘respondents to a consultation which necessarily samples a tiny fraction of the population’ is misleading.” He then proudly notes that the HFEA polled 2000 people and received “641 replies” to its consultation, “including 66 from organisations representing large numbers of people”. That makes 2641 responses out of a population of around 60 million. We are not told the “large numbers” that these replies represent, but nothing Baldwin says should give anyone confidence that I am wrong to suggest that these only represent “a tiny fraction of the population”. I am sure that scepticism about the legitimacy of proposals to deny choice to the entire community made on the basis of a poll of 2000 people and 66 collective responses from pressure groups, cannot reasonably be described as being “dismissive of public anxieties”.


Since Baldwin knows that there is no empirical evidence available which points to any harms from sex selection per se, nor even any plausible guess as to what that evidence might turn out to be, he turns to the mystical sermonising4 of Jürgen Habermas.

Baldwin tells us that1:

Habermas argues that an essential ingredient of our conception of ourselves is that we should be able to regard our embodied character (Leibsein) as a natural phenomenon, and not something which has been, in some respect, deliberately imposed upon us by others, even by our parents. Of course, we must also recognise that in many ways we have been formed by the genes we have inherited from our parents; these genetic predispositions are, however, our bodily inheritance and there is no way in which a human being can be created without some such genetic inheritance. But where a fundamental characteristic such as one’s sex has been deliberately selected for things are different ... There is an inescapable alien intrusion into its subjective sense of itself.

Baldwin offers no reason other than the authority of Habermas as to why “an essential ingredient of our conception of ourselves is that we should be able to regard our embodied character (Leibsein) as a natural phenomenon.” But again we should be cautious about accepting any of this at face value. Before Darwin, it might have, not implausibly, been asserted that an essential ingredient of our conception of ourselves was that we were created as human beings. Now we know we have evolved in a seamless transition from our common ancestor with chimpanzees, but most of us seem to have adapted well to this dramatic change in Leibsein. The evidence is that human beings are fairly robust and well able to adapt to new conceptions of themselves and their place in the universe. The observations of Galileo and Copernicus were equally, perhaps more, momentous for our conceptions of ourselves and our place in the universe and in the scale of things, but again we seem to have come through.

Although Baldwin accepts the non-identity of the children who result from different choices, he constantly talks as if this point has not really penetrated very far into his Leibsein. For example it “was her parents’ choice that their child should be a daughter and in that sense her femininity is indeed imposed on her …” seems to imply that their child might not have had her femininity imposed on her, that she might have been, and might have preferred to be, a boy. In case Baldwin thinks that this is an unfair characterisation of his words he should ask himself how she might have escaped this fate and avoided this imposition of femininity. We should perhaps recall and emphasise the ways in which sex selection is an excellent precursor of other forms of genetic selection and indeed genetic manipulation and what such choices can and cannot achieve.


Sex selection is an excellent exemplar for discussions of genetic determination because:

  1. Sex is genetically determined.

  2. Gender is harmless, being a boy or a girl is not bad for you.

  3. Selection does not involve shaping the individual in any way, nor can it conceivably make the individual worse off than either (a) she was or (b) she might have been.

The first point needs no further explanation. Let me say something about the claim that gender is harmless. By this I mean simply that it is not harmful to the individual to be a man or a woman. Men and women have existed since humans have and although there have been severe power imbalances between the two genders for most of human history the damage that this has caused is contingent, not a necessary part of maleness or femaleness. If gender is what I call a “morally neutral trait” then it cannot be morally wrong to be a man or a woman and so it cannot be morally wrong to create a man or a woman. The only remaining question is whether it can be morally wrong to create a man rather than a woman or vice versa.5

The third element is very important. There is no complaint the “victim” of sex selection can make because for her there was no alternative but never to have existed. “She” could not have been a boy. This is because the boy that might have been selected or created instead of her would not have been “her” only with a different sex. It would not have been a case of sex reallocation. It would simply have involved the creation of an entirely different person.

Although this is hardly ever noticed the same is true for any significant genetic manipulation that might be made to an embryo or indeed to the gametes prior to conception, if this ever becomes possible. So complaints that parents who would use sex selection are attempting to shape or mould their child are simply incoherent. They may of course be choosing what sorts of children there will be, but none of those children have any legitimate or even coherent complaint, for they could not have had an alternative life free of such externally imposed choices.


Finally we must turn to the HFEA’s rejection of what I have called the “democratic presumption”. And this is down to the HFEA and not the Human Fertilisation and Embryology Act 1990. Introducing his discussion of this Baldwin says1:

In the final part of his paper Harris criticises the way in which the HFEA, having given a “firm and consistent statement” of the liberal presumption that the state should not intervene in family life except to prevent serious harm, abandons this principle in the light of the public hostility to sex selection revealed by its consultation. This is the most important of his criticisms and in responding to it I should acknowledge that the discussion of this matter in the HFEA report is not as full as, in retrospect, was desirable. … In doing so even he acknowledges that “there is no widespread agreement as to the nature and scope of this right”; so, on the face of it, even he should accept that he cannot just invoke this right in order to legitimate sex selection … this area of reproductive liberty is not fundamental to liberal democracy, as Harris ludicrously maintains …

Although we both agree that the force of this presumption is really the key issue, Baldwin, I am sorry to say, completely misunderstands what the democratic presumption entails.

First, I did not “invoke” any right of reproductive choice in order to legitimate sex selection. I set out the arguments for such a right to show that claims in this area were not simply idle preferences; however, nothing in my argument turns on establishing or even accepting reproductive liberty as a fundamental right. The democratic presumption does not protect only established rights. Moreover it is difficult to understand how Baldwin can claim with a straight face that I “ludicrously” maintain that this area of reproductive liberty is fundamental to liberal democracy. I maintain no such thing. I invoke the democratic presumption exactly in the way the HFEA report does (at paragraph 132)2 as a presumption that we all accept and that must somehow be met, not evaded or, in the case of the HFEA reformulated so that it is emptied of content.

The point is not whether or not this particular dimension of reproductive liberty is fundamental to liberal democracy; that I concede is hardly likely. But that does not mean that the denial of reproductive liberty or this dimension of it is not fundamental to liberal democracy, nor that it is ludicrous to suggest it is. What is fundamental to liberal democracy is that the liberty of citizens should not be abridged unless good and sufficient cause can be shown as to why this is required. This is true whether the liberty asserted is the freedom to drink tea or the freedom to reproduce. When, for example, we say that “due process of law” is a principle of liberal democracies, that it protects all, we are not saying that the alleged commission of any particular offence goes to the heart of liberal democracy, nor that the wrongful or even the unsound conviction of any individual threatens democracy. We are saying that the institution of due process is essential and that deliberately setting it aside (as sometimes in terrorism cases) requires the weightiest justification.

As I argued in my original paper, even idle preferences command respect and their denial requires justification. But serious moral claims require the greatest respect and the weightiest reasons must justify their denial, even where those claims are just that, “claims” and not established moral rights. It is not the particular liberty that has to be important, what is important is that the burden of proof is on those who would curtail liberty. Since the HFEA report invokes the democratic presumption in precisely this context and in this way Baldwin’s comments must be taken as an attempted criticism of the HFEA report. If my invocation of this principle is ludicrous then so is that of the HFEA, if not then the HFEA has me to thank for defending it against one criticism from the Deputy Chair which is way off target.

If as Baldwin states, there are powerful arguments on both sides then the presumption must be in favour of liberty. However I did argue, and still maintain, that reproduction clearly is an important liberty by any standards and while maybe not itself be constitutive of liberal democracy it is by any reckoning of an importance which requires serious engagement if it is to be set aside. In this it may be different from drinking tea. I also argued and nothing that Baldwin has produced casts any doubt on this whatsoever, that there are no powerful arguments against the exercise of this liberty by sex selection. No evidence of harm was produced by the HFEA (only evidence that some people don’t like the idea of it) and Baldwin now concedes the absence of any evidence by calling in aid the mystical, and in so far as they are coherent, doubtful assertions of Habermas.


Indeed, when Baldwin claims of his invocation of Habermas: “In drawing on Habermas in this way my aim has been to show that the objection to the exercise of reproductive liberty in this area is not just a matter of disputable empirical harms. The issue goes much deeper …” he shows that he has given up on evidence and argument altogether. Baldwin is taking the “argument” beyond disputable harms because he knows not only that such harms have not been and cannot be established, he knows that there is no credible reason to think that any serious harms might eventuate. Harbermas does not simply take the argument “deeper”, he takes the debate to a depth that neither rationality nor evidence can reach. Safe in the impenetrable murk that pervades such depths Baldwin can assert what he likes, knowing that because the fears that there find expression are not supported by evidence or by argument or by reason they are truly unreasonable. As Jonathan Swift is believed to have said, “it is hopeless to reason a man out of something that he has not been reasoned into”.

Given an absence of evidence and a very low probability of harm, it is surely perplexing that the HFEA did not take the cautious but democratic step of saying—yes you can do this but initially it should be regulated which will involve contributing to research studies that will evaluate the impact of this new technology.6


My own proposal, which I first published in 1998,5 would be that a society like ours, of about 60 million people, should perhaps begin with cautious regulation. We surely could “afford” to licence, say one million, procedures for sex selection over a 10 year period with options to revise the policy if severe imbalance seemed likely or if unforeseen harms began to emerge. We could then see what patterns of selection and motivation emerged and whether any significant problems were caused to individuals or to society. Even if all choices went one way, the imbalance created would be relatively small before detection, and a halt could be called if this seemed justifiable. I doubt that the places allocated on such a programme would be taken up (it would of course be self-financing and would not be part of the public healthcare system). It must be remembered that those who opted for sex selection would (with current technology) have to be very circumspect about their procreation and use assisted reproduction or the various methods so well described in the HFEA’s report. This would not, I guess, be wildly attractive or indeed particularly reliable. For the foreseeable future the take-up will also by limited by the availability of clinics offering the service. In any event, the way forward for a tolerant society respectful of autonomy, and mindful of the democratic presumption, would surely be not to rush to prohibitive legislation, but rather to licence the activity with regular monitoring and follow up studies and see whether anything so terrible that it required prohibitive legislation emerged.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.