Tuesday, 4 January 2011

Is sex selection on non-medical grounds justifiable?

Advances in medical technology now allow parents to choose the sex of their baby without having to undergo excessively invasive or dangerous procedures. Yet most societies have recoiled from this possibility, and have severely limited parents’ freedom to choose. In the UK, for example, HFEA (Human Fertility and Embryology Authority) regulations only permit sex selection for medical reasons: to avoid serious sex-linked conditions, such as haemophilia.

Yet any close scrutiny of the common justifications for preventing parents from choosing the sex of their child shows them to be misinformed, unreasoned or weakly rationalised gut reactions. Any review of either the ethical literature or the numerous public consultations on the subject is bound to uncover the same set of objections to non-medical sex selection. This essay addresses each of the most common concerns in turn, and seeks to demonstrate that they are misplaced or mistaken.

This is a strategy that needs defence in itself, since it assumes that the burden of proof is on those that would restrict non-medical sex selection. Yet this presumption does not seem unwarranted. Forbidding selection is clearly a restriction of freedom, since it prevents someone (parents) from doing something that they wish to do (choose the sex of their baby). There is not a single moral or political theory which would deny that all else being equal, more freedom or preference satisfaction is better than less. Nor does any philosophy I am aware of permit the restriction of freedom or the frustration of preference without any justification whatsoever. To put it another way, there is no ethical system which requires us to get permission before acting – if things are not expressly forbidden, we can assume they are permitted. The consideration is just a weak one – as soon as there is a countervailing reason against an act, we may be justified in restricting it. However, it hardly seems unreasonable to require some such convincing reason against sex selection as a necessary precondition of restricting it. This article/talk attempts to show that no such compelling reason exists.

The ‘Feminist’ argument: sex selection reduces the status of women

One common concern regarding non-medical sex selection is what David Heyd calls the ‘feminist’ argument (though without referencing any actual feminists who he believes to hold it). The basic idea is that allowing parents to choose the sex of their children will allow them to favour boys over girls, and will further diminish the status of women in society.

Heyd’s first response to this line of argument is to observe that non-medical sex selection does not create or promote sexism, it merely reflects the underlying prejudices of the society. He urges feminists to reconsider their targets, and to focus on the underlying root causes of gender inequality, rather than being distracted by symptoms. For example, Heyd implies that attacking the institution of dowries would make girls less expensive, and would make parents less likely to choose boys.

This response is not fully persuasive. Even if sex selection is just a manifestation of gender inequality, it can still be harmful independently of its root causes. Sex selection could perpetuate and normalise sexist prejudices, by reassuring people that it is acceptable to hold such views and to act upon them.

Heyd also refers to the empirical work of Marcia Guttentag, who suggests that historically, the proportion of women in a society is inversely related to their status. If women are relatively rare in a society, they will be valued more, given more power and freedom. This further undermines the empirical claims of the ‘feminists’.

Heyd’s arguments are interesting and subtle. However, the strongest grounds for resisting the ‘feminist’ criticism is the assumption that boys would, in fact, be preferred to girls. In reality, this premise does not seem to hold in Western countries. Gleicher and Barad’s analysis of choices made by parents in an American clinic that offered sex selection found that those wanting boys and girls balanced each other out. Numerous surveys have established that, if anything, there is a slight preference for girls in Western countries.

The Problem of Sex Ratio Imbalance

These findings undermine perhaps the most common objection to non-medical sex selection: the fear that it will lead to a major imbalance in the sex ratio. Yet this fear is clearly unjustified. As we have seen, those that do opt to choose the sex of their children tend to be evenly split between those who want boys and those who want girls. Even if there were an imbalance in this population, the overall effect would be attenuated by the fact that most people probably wouldn’t choose the sex of their babies in the first place. Hall et al’s review of survey data demonstrates most people do not want to decide.

Of course, these findings are limited to Western countries. In many parts of the world, there would be a strong tendency for parents to choose, and in particular to choose to have boys. In countries where sex imbalance and discrimination are genuine possibilities, it would be irresponsible to facilitate them. But when a society is deciding on policies and restrictions which are binding only on it, the effects that policy would have on other societies not subject to their laws is surely irrelevant.

The only reason I can imagine for the UK or other Western societies to consider the effects of permitting sex selection on other societies is that it might leave them open to the charge of hypocrisy in dealing with those other societies. Thus if the UK permitted sex selection but tried to criticise (say) China for doing the same, it would appear inconsistent. Yet it is clear that the inconsistency is only apparent, an acknowledgement of and response to the fact that the UK and China are different social contexts. It is not that different principles are being applied to different societies, rather the application of the same principle requires different responses in different contexts.

In any case, it is important to remember that even if every parent chose the sex of their child, this would still only have a small effect on overall sex balance because sex selection is currently open to so few parents. Sex selection is limited to those who undergo assistive reproductive treatments such as IVF. These methods have much lower success rates than natural conception, so it seems implausible that many parents would jeopardise their chances of conceiving at all just to ensure they could choose the sex of their child.

Inequality of access

The fact that so few will in practice be able to choose the sex of their children may, though, be used as an argument against permitting sex selection. It simply illustrates that access to this technology will almost certainly be extremely unequal. This concern can come from two sources: the worry that the rich will be able to afford services that the poor cannot, or that those who conceive naturally will be unfairly disadvantaged.

The notion that the rich should not be able to buy a better standard of healthcare is problematic because it brings to mind the ‘levelling down’ objection against extreme versions of egalitarian thought. If nobody is made better-off by equalisation, it is hard to see why we should pursue absolute equality. If we cannot make the blind see, does this mean we should all put our eyes out to ensure we can all see equally badly? In the same way, just because some people cannot take advantage of the option to choose the sex of their child, does this mean nobody should have that option at all?

Yet even if we accept that unequal access to sex selection is wrong, it does not follow that we should ban it. Rather, it makes more sense to extend access to the technology to all. Subsidisation of the service or its provision on the National Health Service would give the poor equal opportunity to choose the sex of their children. Perhaps this focuses the issue: there are certain essential medical treatments that we believe all should have access to, rich or poor, such as emergency care, and so we ensure public provision of these. There are others which are not so crucial which we do not mind the rich having greater access to, such as cosmetic surgery, and so these are generally provided privately. The question then becomes whether sex selection is more like emergency care or cosmetic surgery, essential or not.

If we even up access to sex selection between rich and poor, this still fails to address the fact that only the infertile will be able to choose the sex of their baby (Of course, it is not only the literally infertile who will choose their child’s sex – fertile parents who carry sex linked diseases may do the same. However, the point still holds because in forgoing natural conception they are reducing their chances of conception to that of the infertile). This is a peculiar argument, since it depends on us accepting that the infertile are unfairly advantaged. Stated in these terms it seems plainly absurd: if anything the ability to choose the sex of their baby is a sort of compensation for parents’ inability to conceive naturally, and an inadequate one at that. It is helpful to consider the issue in terms of Michael Walzer’s concept of complex equality. According to Walzer, to determine whether two people are equal, it is not sufficient to look at just one dimension, or ‘sphere’ (such as income), but we must consider all indices. Thus while infertile couples are disadvantaged in the sphere of fertility, they are advantaged in the sphere of reproductive choice.

Misuse of medical technology

The distinction between essential and non-essential medical services leads us to another common objection to non-medical sex selection. Many believe that medicine should only be about curing disease. On this view sex selection is therefore a misuse of doctors’ time and resources. Yet the example of cosmetic surgery illustrates the fact that doctors can and do devote their time to other things beyond curing disease. Not many people would go as far as to call for the banning of cosmetic surgery on these grounds, however, as this position would appear to commit them to.

In any case, it is hardly self-evident that doctors have no duties beyond fighting disease. Many insist that enhancement of patients’ wellbeing is an equally important facet of medicine, for example, reducing pain.

‘Slippery slope’ to designer children

Another common fear regarding non-medical sex selection is that it is the ‘thin end of the wedge’, and that accepting it will inevitably lead to the development of ‘designer’ babies. The thought is that if people are permitted to choose one characteristic of their children – sex – then it will be impossible to prevent them from choosing other attributes, such as eye colour, hair colour, intelligence. Even accepting that this would be undesirable (not everyone agrees - see Nicholas Agar’s case for liberal eugenics), it is not clear why exactly the descent down the slippery slope is inevitable. There is no logical certainty that sex selection entails any other form of selection.

Indeed, as Heyd points out, there is a notable discontinuity between sex selection and the choosing of other attributes. While sex selection can be done through sorting of embryos or sperm cells, full-on designer babies would require genetic manipulation. This difference is key: clinical sex selection involves no interference with the human genome at all, and so is crucially different from the production of designer babies.

It is also worth remembering that most people are willing to accept sex selection on medical grounds without fearing the inevitability of designer babies. Yet this is also a case of choosing the characteristics of one’s child – so how is it any different from sex selection on non-medical grounds?

Argument from religion/nature

It is worth pressing the objection to designer babies a little harder to see why exactly these are perceived to be so undesirable. One strong strand of the argument is a religious one, which sees such tampering as questioning the will of God, and undermining His authority. On this line of thinking, to fail to leave the matter of the sex of our baby to divine providence is to fail to respect God’s wisdom and benevolence. Secularised versions of this argument are often founded on the thought that sex selection is in some way unnatural, or represent humanity in some way ‘overstepping’ its remit. For example, Michael Sandel sees the desire to master the ‘mystery of birth’ as ‘hubris’.

None of these can in themselves provide persuasive accounts of what is supposed to be wrong with non-medical sex selection. The fact that something is natural does not in itself make it right or wrong. This much has been established in ethics since G.E. Moore pointed out the ‘naturalistic fallacy’. There are all sorts of established human practices and behaviours that could hardly be classified as ‘natural’, and yet which hardly seem ethically problematic. From wearing clothes to driving cars to artificial reproduction, to reject all that is unnatural would involve rejecting most of modern human life.

The explicitly religious argument is problematic too. Even if we concede the existence of God, it is unclear why we should spend much time trying to second guess His desires and intentions. To the argument that the conception of a boy or girl is in some way the revealed will of God, we can respond that if God did not want us to choose the sex of our children, He would not have provided us with the technology and expertise to do so. God helps those who help themselves, as the old joke reminds us:

There was an old man sitting on his porch watching the rain fall. Pretty soon the water was coming over the porch and into the house.

The old man was still sitting there when a rescue boat came and the people on board said, "You can't stay here you have to come with us."The old man replied, "No, God will save me." So the boat left.

A little while later the water was up to the second floor, and another rescue boat came, and again told the old man he had to come with them.The old man again replied, "God will save me." So the boat left him again.

An hour later the water was up to the roof and a third rescue boat approached the old man, and tried to get him to come with them.Again the old man refused to leave stating that, "God will grant a miracle & save him." So the boat left him again.

Soon after, the man drowns and goes to heaven, and when he sees God he asks him, "Why didn't you save me? I thought you would grant me a miracle and you have let me down."God replied, "You idiot, I don't know what you're complaining about. I sent three boats after you!!"

The ‘Parental Love’ Objection

Perhaps we are being unjust in writing off people’s unease as superstition or luddism. Some philosophers maintain that the unease many feel with non-medical sex selection is rooted in concern for the welfare of the resultant child. This certainly appears to be the position of the HFEA, which seems to accept the argument that children who have their sex chosen by their parents will be somehow disadvantaged. Peter Herissone-Kelly tries to rigorously formalise this intuition, among the most common gut reactions to the matter of sex selection, into what he calls the parental love argument. This also seems the most charitable interpretation of Sandel’s argument above.

On Herissone-Kelly’s account, what is wrong with non-medical sex selection is that it fails to display ‘proper parental love’: “the sort of love that we think parents ought, simply in virtue of being parents, to have for their children”. Presumably, the implication is that if a parent cannot provide such love, they should not be given children. (Note possible difference between ‘giving’ and ‘letting have’)

Proper parental love must satisfy the ‘any incumbent’ model. For A to love B with proper parental love, A’s love must have only one (necessary and sufficient) condition: that B is A’s child. Another way of putting this is that A would love anyone who happened to be their child. Since parents who choose the sex of their children seem to be making their love contingent on them being the ‘right’ sex, they violate the any incumbent model, and so fail to display proper parental love.

This does not follow. These parents may seem to be making their love contingent on their children’s sex, but this is not necessarily the case, nor is it even that likely. The claim is that all parents who would prefer a boy to a girl would therefore not love their child if they were a girl (or vice-versa). This might hold in a minority of cases, where parents would rather have no child than a child of the ‘wrong’ sex. Yet there are almost certainly many parents who would love and be happy with a child of either sex, but who simply have a preference one way or the other.

It is important to separate two distinct concerns here. On one hand, we may be worried that parents with a sex preference may end up with a child of the wrong ‘sex’, and therefore fail to be good parents to it. On the other, the problem is that even if the parents get the sex they wanted, this reflects a deficiency in their attitude – a lack of proper parental love – which is bound to adversely affect their child. This draws out a central plank of the parental love objection – that our attitude to hypothetical children in different possible worlds is an indicator of our strength of feeling for our real children.

Observe that the ‘any incumbent’ model places very strong restrictions on what can be described as proper parental love. If you can imagine any circumstance under which your love for your child would falter, then you do not love them ‘properly’ as a parent. This is a test that many, if not most, real-life parents would fail. The conditions of their love may be few and weak, but only the most saintly could still love a child who lied, stole, raped, killed without repentance. Parents do disown their children, and often because of the despicable characteristics of the child. Herissone-Kelly accepts this argument, and insists that parents who would not love a ‘diabolical’ child cannot hold proper parental love for their real, non-diabolical children.

However, Herissone-Kelly is very careful to point out that he is not doubting that parents who insist upon this ‘non-monstrosity’ condition do love their children. His point is merely that they do not display ‘proper parental’ love. But what is the difference between ‘proper parental’ and other kinds of love? And why should we care about the absence of ‘proper parental’ love, if Herissone-Kelly accepts that some form of love is present? Herissone-Kelly’s answer is that parental love ought to be reliable, and that making your love contingent on a person’s characteristics makes it unreliable. This links his argument to the virtue ethics rejection of sex selection.

The virtue of ‘acceptance’

Herissone-Kelly explicitly builds on Rosalind McDougall’s virtue ethics approach to the question of non-medical sex selection. Virtue ethics is the philosophy that moral rightness consists in possessing and displaying ‘virtues’ rather than in any given action. Virtues are usually defined as dispositions or mindsets, which entail not only acting in a certain way but also certain “emotions and emotional reactions, choices, values, desires, perceptions, attitudes, interests, expectations and sensibilities”. Thus an honest person – one who displays the virtue of honesty – does not merely tell the truth, but does so because of the values, beliefs and motivations that we would associate with honesty.

McDougall seeks to adapt the virtue ethics approach to the question of sex selection, and argues that non-medical sex selection is wrong because it fails to display parental virtue. She posits that the criterion of right parental action is that an action is right if and only if it is what a virtuous parent would do. Traditionally (at least within the Aristotelian tradition), virtues have been regarded as those traits favourable to human flourishing (the Greek concept of eudaimonia). In McDougall’s account, what makes for a parental virtue is that a disposition is conducive to the child’s flourishing, based on immutable facts about human reproduction and rearing.

According to McDougall, one such immutable fact is unpredictability. Children are always capable of placing unexpected strains or demands on their parents, and so it is in the interests of a child’s flourishing to have a parent capable of adapting to this. Even if we could know a child’s entire genetic code, McDougall alleges, environmental factors are constantly capable of surprising us. For example, the child could be blinded in an accident. This fact implies that acceptance is a parental virtue. Parents must be capable of taking their children as they are, even when they fail to meet their expectations because the sheer unpredictability of children means that they will not always meet their parents’ expectations. This seems to be Sandel’s point when he argues that choosing the sex their child “deprives the parent of the humility and enlarged human sympathies that openness to the unbidden can cultivate”.
The virtue ethics account, like the parental love objection, often seems to read too much into parents’ desire to choose the sex of their baby. This one choice need not imply total obsessive control freakery, a desire to dominate every part of their child’s life. It is just one decision, one feature of the child. There is nothing inconsistent about having a preference for the sex of your child and being generally accepting of its other features: appearance, talents etc. Moreover, it does not follow from the fact that a parent has a preference for a certain sex that they would not accept and love a baby of the ‘wrong’ sex. Thus a person can have a sex preference and possess the virtue of acceptance at the same time.

McDougall accepts that it is not clear that having a preference is in itself wrong, but holds that acting on this desire would be wrong. She also insists that even if the parent’s future actions displayed the parental virtues, including acceptance, this would be irrelevant to this moral question, because it is only the act of sex selection itself that is at issue. Yet this seems to break dramatically with the spirit of virtue ethics, which evolved as a rejection of moral philosophy’s obsession with evaluating individual acts. Rosalind Hursthouse bemoans “the extreme recklessness of attributing a virtue on the basis of a single action”. One lapse does not make a person vicious, it is their general character that must be considered. Therefore, a virtue ethics account of sex selection has to account for the motivations, attitudes and characteristics of parents before pronouncing on their moral status.

However, there is a deeper point to be made against both the parental love and virtue ethics objections to sex selection: that they expect too much of parents. Even if we accept that a parent who fails to show proper parental love or the virtue of acceptance towards their children is in some way imperfective, it is not enough to show that they are negligent or incompetent. Herissone-Kelly has to concede that there are numerous well-functioning happy families where parents love their children, but simply fail to meet the (incredibly high) standards he demands. Similarly, many parents fail to be fully accepting, blame children for faults that are not their own and push them to be something they are not. Yet this one failing does not mean that they are unfit parents, or even that their children will be particularly harmed. Most parents are imperfect, rarely is love truly unconditional.


  1. An interesting piece, though, being rather pushed for time, I have at the moment only skimmed it. I promise to give it a more thorough read later on. Natural human vanity meant that I gave most attention to the bits that deal with my own attempts to construct a justification for the parental-love objection. It seems to me that some of your reporting of my view gets it a little wrong, but I'm confident that that's a result of my not having been as clear as I might have been in my papers on this topic. When I have more time, I'll post again with responses to what you have to say. That might involve my first having to reread my Cambridge Quarterly papers, though ... it's a few years since I wrote them, and, shamefully, I don't have as detailed a memory of them as I should have!

    Peter Herissone-Kelly

  2. In your article you have tried to discuss is this justified to use gender selection natural introduction and is it worth it or not. I think this method is used by several women and the results are effective as well cost-free.