Article Text

Download PDFPDF

Sparrows, hedgehogs and castrati: reflections on gender and enhancement
  1. John Harris
  1. Correspondence to Professor John Harris, Institute for Science, Ethics and Innovation and Wellcome Strategic Programme in the Human Body: its Scope, Limits and Future, School of Law, University of Manchester, Oxford Road, Manchester M13 9PL, UK; john.harris{at}


In a number of papers, including the one published in this journal, Robert Sparrow has mounted attacks on consequentialism using principally what he takes to be an important fact, which he believes constitutes a reductio ad absurdum of consequentialism in its many forms and of this author's approach to enhancement and disability in particular (see page 276). This fact is the current longer life expectancy of women when compared with men. Here the author argues that Sparrow's arguments and entire approach utterly fail. In doing so the author hopes to shed further light on the role of normalcy, normal species functioning and species-typical functioning in debates about enhancement and disability.

  • Enhancement
  • disability
  • normalcy
  • harm
  • harmed condition

Statistics from

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.

Robert Sparrow is definitely a hedgehog. No bad thing that, he is in good company (among many others he shares this intellectual trait with Ronald Dworkin and, more modestly, myself). While foxes know many things, hedgehogs notoriously know one big thing, and the merit of a hedgehogs' intellectual strategy is entirely dependent on the nature and importance of the big thing that he knows.

Unfortunately the big thing that Sparrow knows is that women currently have significantly longer life expectancy than men and he is trying to use this modest piece of knowledge to undermine some major approaches to human enhancement and the understanding of disability and impairment including my own. It would be a good trick if he can pull it off, but the auguries are not good.

The heart of Sparrow's problems with the part of my approach to disability and enhancement that he criticises are stated thus:

‘I will argue that Harris fails in his attempt to provide an account of a “harmed condition” that can stand independently of intuitions about what is “normal”. The fact that Homo sapiens is a sexually dimorphic species means that attempts to evaluate whether a given individual is in a harmed condition will sometimes require making reference to their sex—and therefore to the normal capacities of that sex. This, in turn, has implications for the plausibility of Harris's account of the reasons we have to enhance human beings.’1

Before coming to a discussion of disability and enhancement let us just dispose of Sparrow's complete misunderstanding of how the concept of normalcy functions in these debates. My use of the concept of a ‘harmed condition’ highlights the fact that the mainspring for moral concern and thus intervention both in therapy and enhancement is to remove or mitigate harms or to improve human life, not to restore the status quo ante or restore or otherwise endorse normality. If harms are normal so much the worse for them and those who experience them. Those who reject the ‘harmed condition’ approach to disability or indeed to enhancement tend to say that the purpose of therapy is to restore normal functioning or species-typical functioning and that that is all that is morally required. To this I respond, and have responded, that normal functioning is no comfort to those for whom normal functioning is a disaster or is, at any rate, harmful in circumstances in which normal functioning can be replaced with abnormal but better functioning.

Now of course a normal state can be good evidence that the individual or class of individuals in that state might also be in a harmed condition. A normal condition for men of my age is thus to need reading glasses. Here my age can function as a sign that makes us look for symptoms normal for men of my age. I have never denied or even implied ‘that attempts to evaluate whether a given individual is in a harmed condition will sometimes require making reference to their age’, or in Sparrow's example, their sex, in short to what is normal or expected. However, the problem is diagnosed independently of what is normal, the problem is short-sightedness or shorter life expectancy or whatever might be improved upon, not age or sex, not least because there are always exceptions to normal functioning, and the problem is identified as one that needs amelioration not by reference to normal functioning but by reference to what might be possible to make things better. The key issue is not that males have shorter life expectancy than females but they have shorter life expectancy than they need have, by reference, in short, not to normal functioning but to possible functioning. Contrary to Sparrow this emphatically does not ‘in turn, ha(ve) implications for the plausibility of Harris's account of the reasons we have to enhance human beings’.

Tom Kirkwood (more of whom later) famously pointed out that ‘we do not die of old age but of the diseases of old age’, and those diseases are normal and species typical for old people, but that does not remove our moral reasons for trying to combat them and taking the sting out of old age. Nor do we die of being a man rather than a woman. If we men die on average some years earlier than women, it is almost certainly because of the effects of testosterone on our physiology and the fact that these effects are gender typical is irrelevant. Of course what is normal may be a clue to a problem. Hypertension is more common in black Americans than non-black Americans, it is normal for black Americans to have more risk of hypertension; the Japanese have a higher incidence of stomach cancer than other populations, also normal for them; these facts are helpful clues in searching for a diagnosis or devising preventive strategies, but their normality in a population plays neither a normative nor a causative role in the incidence of disease. The harm of these conditions is in no way connected to, nor diagnosed by, the normality of their incidence in a particular population, and our moral reasons to treat these conditions are unconnected to their normalcy.

It bears repeating that humans naturally fall ill, naturally die prematurely; that pain, disease, misery and premature death are part of the human condition, are perfectly (sic) normal and natural and that these facts, their normality or abnormality, play no role whatsoever in our identifying such things as a curse and a disaster, yes and as positively harmful! We understand that our appreciation of how harmful such things are and of how important it is to ameliorate them if we can stand independently of intuitions about what is ‘normal’. It must do so or we could never have a motive for improving on ‘things as they are’ and trying to do something about such dreadful normal things. It is not an accident that Chapter 7 of ‘Enhancing evolution’,2 begins with some lines of Wallace Stevens. I will repeat them for the record:

‘They said, “you have a blue guitar,

You do not play things as they are.”

The man replied, “things as they are

Are changed upon a blue guitar.”’3

But let us turn to an even more irresistible subject.

The big thing—sex

Let us think about sex. Sparrow's complaint is that I (and others like Julian Savulescu4 who have followed similar lines of argument) cannot consistently recommend attempts to avoid disabilities and other ‘harmed conditions’ in future children without consistently selecting against male children and indeed without attempting to eradicate males altogether from our reproductive choices. He puts the point thus:

‘In fact, there is a significant difference between the sexes that means that male children suffer something that Harris explicitly mentions as a harm. In an early discussion of harmed conditions, Harris states:

“If the condition in question was one which set premature limits on their lifespan—made their life shorter than it would be with treatment, or was one which rendered her specially vulnerable to infection, more vulnerable than others, we would surely recognise that she had been harmed and perhaps to some extent disabled.

“Males do have significantly shorter life expectancies than females, dying on average roughly 3 to 7 years earlier, in part because males are indeed ‘specially vulnerable to infection’ compared to women.”

‘(v) Thus, to be born male is, according to Harris's own account, to be born “harmed and perhaps to some extent disabled”. Harris might want to insist that there are other benefits of being male…’.1

I had noticed the significant difference between the sexes to which Sparrow alludes and he is of course right, as I have always claimed, that, other things being equal, this gives us reasons to create children with better life chances. However, other things are so obviously not equal in the case of Sparrow's big idea, that it hardly bears pointing out. It may have escaped Sparrow's attention, but it is a commonplace that the cure can be worse than the disease, and when it is only masochists insist on the cure.

I am committed to humans flourishing, to making the world a better place. This is and has always been at the centre of my philosophy and indeed of my politics. I do not believe that the gradual eradication of men from the planet, by the erosion that would take place if reproductive choices systematically excluded them, would have a positive effect on human welfare, happiness, social life or any other aspect of existence for either of the two main sexes of people, or for societies. The creation of an all-female world would not be conducive to human flourishing. Sparrow must also believe this as it is at the heart of his attempted reductio ad absurdum of my position. Indeed, almost everyone believes it. This might not be enough, nonetheless, to make it right. As I argued in my book ‘The value of life’,5 there might be circumstances in which other values that we hold might make it difficult to avoid the creation or at least the acceptance of the arrival of an all-female world. I am not therefore averse to (or even shy of) thinking and discussing the unthinkable.

However, the all-female world is attractive neither morally, socially, politically nor indeed as anything that could be called an enhancement, let alone a viable reproductive goal. For the foreseeable future the ways in which the overwhelming majority of women might have their babies without men are limited, and worse they are problematical, dangerous and expensive. The only ‘reliable’ way so far is via human cloning with its attendant problems in the current state of science, technology and indeed prejudice.6

Other promising methods: manufacturing male gametes or embryos of either gender from mature cells via re-programming, or creating embryos via the artificial stimulation of parthenogenetic development, for example, are still problematical and many pork pies short of the sort of picnic that might celebrate a birth. Even frozen sperm from existing men will eventually run out. We are not in the sort of world that could contemplate Sparrow's glib criticism of me for not recommending the reproduction of females only, as anything other than a bad joke. This is perhaps one good reason for me not to have recommended it as a way of avoiding the birth of creatures who will have a lesser life-expectancy than existing human women. For the foreseeable future women will not be able to do without men if they want to have babies, and our species cannot systematically forego one of its main genders without the threat of extinction. Even if and when we have the technology to manufacture sperm or embryos safely enough for reproduction, the technology will be expensive, difficult and problematical for indefinite periods after it is first developed and therefore available to relatively few of that world's increasing supply of women.

Nor is a policy calculated to result in the survival of only one sex attractive biologically nor is it likely to be conducive to human survival for long. As Tom Kirkwood7 has pointed out, the sort of random combination of genes that results from the mixture of those from two different progenitors can keep us ahead of attacking parasites or produce random changes or mutations that may have beneficial effects:

‘Gene shuffling by way of sex not only helps keep up with the parasites, it is also a good way to speed up the trial and error of genetic novelty that you might need to get out of other potentially catastrophic fixes, such as a change in climate. If your world is changing fast enough, this advantage of sex might outweigh its twofold disadvantage.’7

For all these reasons, no sane person would now recommend the exclusive production of females as a reproductive method of choice. It is hard therefore to take seriously Sparrow's criticism of me for not wholeheartedly embracing this naive doomsday scenario.

Fortunately, in the case of reproductive choices, we have no good moral reasons at this time to think and do the unthinkable. This is because of the non-identity problem,8 which in terms of Sparrow's difficulties constitutes the non-identity solution. His big idea for a cure is definitely worse than the disease, I had not failed to notice this and therefore his criticisms ‘pass me by as the idle wind that I respect not’.9

The non-identity solution to Sparrow's little difficulty

Parfit's elaboration of the non-identity issues as they affect the present discussion may be summarised thus: It is usually the case that the children that result from reproductive choices are not the same children that would have existed had those choices not been made. So long as the children who result from those choices have, or can reasonably be expected to have, lives that are worthwhile overall those children have benefitted from the choices that produced them. Such children have no reason for rational regret that they were born. If we are thinking of the interests of the children who result from such choices existence is obviously in their interests. As I have argued at length elsewhere, such children may be harmed by existence in the sense that they are born with disadvantages relative to others, but as they cannot exist without such disadvantages and their existence is positively worthwhile overall, they have benefitted by being brought to birth. No one has done them wrong by creating them or permitting their creation.10

If, as I believe, the all-female world is in no-one's interests because it would be a worse world than the one we have (despite the suggested but problematical gains in health and life expectancy over the alternative world), then all people who will ever exist benefit from the fact that we have exercised reproductive choices without eliminating men and no-one has been wronged by such choices. Any harm involved is necessary because it is to the overall benefit of all who will ever exist and it creates a world better than the specific alternative of an all-female world.

Contrary to Sparrow's suggestion, nothing I have ever argued commits me to recommend the creation of a worse world than the one we have. I therefore do not need to resort to the escape route that Sparrow thoughtfully offers me when he says: ‘Harris might want to insist that there are other benefits of being male…’. It is not necessarily that there are other benefits of being male, (I don't need to think about those) it is that there are, I believe, overwhelming benefits for the foreseeable future of maintaining both species in rough equilibrium (if not in equality of numbers).

So far Sparrow does not have a leg to stand on. As I would not wish to leave him in such a disabled state if there is any remedy, we can now turn to some of his other claims this time, appropriately enough a claim concerning my arguments about the ethics of trying to remove or eliminate disability from present and future people.


I have argued, as Sparrow correctly suggests, that disabilities and indeed impairments are harmed states that are defined not relative to normal functioning or species functioning but relative to what might have been or what might be. In other words, states of being are ‘harmed’ relative to possible alternatives. I believe it is obvious that we have moral reasons for declining to create or confer even trivial harms, and we have the same reasons to confer and not withhold even small benefits. These moral reasons create obligations to prevent such harms or confer such benefits if we reasonably can. We need a word or two here about what this means. In the cases that preoccupy Sparrow we are talking about reproductive choices. In this context we have an obligation to make these decisions in a way that causes as much benefit and as little harm to future people as possible. So far so good and hopefully so obvious. However, because of the non-identity issues already discussed when we choose to create one child, rather than another but different possible child, our choices do not make a difference to a particular individual, but involve rather a choice between two possible individuals. If the individual chosen will have a life that is worth living, that can be expected to be worthwhile overall, then we will have benefitted that individual. The fact that we might have chosen to create a different individual whom we benefitted even more means that we have made a worse choice than we might otherwise have made, but this is not a choice that has been harmful overall to anyone. On the contrary, we have harmed no-one overall and benefitted someone who is pleased to have been brought into existence. Clearly here, although we have chosen less well than we might, we will still have done something good: we have created a child that is or will be pleased to exist (I ignore issues of population policy).

When I say we have an obligation to do something here, I mean nothing more than that we have good moral reasons to do it, not that we must do it come what may, nor that we are bound to do it, nor yet that it would necessarily be appropriate to compel us to do it or punish us if we do not. The strength of the moral obligation is directly proportional to the importance of doing whatever it is and to the seriousness of the consequences of not doing it. This by the way is always true of all obligations of whatever nature. Although this list of exceptions may seem to weaken this sort of obligation until it seems scarcely an obligation at all, this is not the case.

The obligation to do what we have moral reasons to do is robust. Such an obligation means that we are entitled to do those things to which we are obliged and must be free to do them, that we deserve praise if we do them and blame if we do not, that others, including the state and the courts have good reasons to support us in their accomplishment, that impediments in the form of laws, regulations, penalties, criticism and obstructive behaviour are unethical, that failure to do them may need to be explained and justified to ourselves and others.

It is a commonplace that parents have obligations to do the best for their children, and, I have argued, to have the best children that they can. It is also obvious to all of us who accept this simple truth, that all parents also neglect some of the things that they know would be good for their kids. Sparrow makes very heavy weather of this. He says:

‘It now appears that our obligations to avoid bringing children into the world in a “harmed condition” are incredibly—indeed implausibly—demanding: we will be morally obligated to provide each and every available enhancement to our children. Given that Harris has defined enhancements as things which make human lives go better and as it is rational to prefer that one's life goes better, it will be rational to wish to avoid being deprived of any enhancement. Parents who fail to maximise the well-being of their offspring by providing them with the full suite of available enhancements will have harmed their children.’1

Well they will have harmed their children to the extent that deprivation of those goods harms them. Sparrow of course is right about this, but not in any way that has a force to reduce what I have said (and what almost everyone believes) to absurdity. Rather, the absurdity is in the mock shock Sparrow evinces. We all know that every day there are things that we should do that we do not do, things we believe we ought to do that we fail to do, and even things that we have strong moral reasons to do that we neglect. That's life! and our failure is, to the extent of the importance of doing those things, often shameful, always regrettable. Failure does not show that we were wrong to believe we should do, ought to have done, have strong moral reasons not to have neglected, these things.

Morality is essentially utopian. We need high standards and of course we sometimes, sometimes often, fall short, but that does not show that the standards were inappropriate. If we set our standards lower the shortfall would be greater.

The emergency room test

Sparrow again:

‘The challenge that sexual dimorphism poses for Harris's “harmed state” account quickly becomes apparent when we introduce sexed bodies into the “emergency room test”. Imagine that:

‘A 14-year-old girl and a 14-year-old boy walk into a hospital emergency department, each complaining about their general health after experiencing some headaches and nausea. During the consultation it is discovered that neither has experienced menarche and on examination it is discovered that neither has a womb. Fortuitously, the technology to safely transplant wombs has recently been developed and (in one of those coincidences so beloved by medical ethicists) suitable donors are available in the next ward. It seems clear to me that doctors would be negligent if they did not make a womb transplant available to the 14-year-old girl in this scenario; she is indeed in a “harmed condition”. (ii) However, it does not seem to be the case that the doctors would be negligent if they did not offer a womb transplant to the young boy. (iii) Consequently, the boy is not in a harmed condition. Yet these intuitions rely crucially on the idea that it is normal for a woman to have a womb and not for a man.[25] That is, they rely upon an account of the capacities of a normal male and normal female human body. When it comes to deciding whether someone is an a harmed condition or not, we must first know their sex, and then compare their capacities with the normal capacities of their sex.’1

To take the last point first, Sparrow is wrong. When it comes to deciding whether or not someone is in a harmed condition, all we need to know is whether it would be possible to make their condition better, if we can, then their condition is harmed relative to that possible improvement.

Let us start further back. It is gratifying that Sparrow finds the ‘emergency room test’ a useful tool, and that I am subjected to a test that I myself first introduced in 1993.11

However, the example is inapposite. First, there is little reason to believe that a womb transplant will do the trick, any trick! A lot more would be needed to effect a real sex change, and again the evidence seems to be that successful sex realignment surgery requires that the subject wants to be, and feels comfortable with becoming (in this case), a woman and uncomfortable as a man, not simply that he wants a longer life and the capacity to bear children. Nor is such a transplant an unproblematical good for the man or for women or for humankind or our successors. First, there are the reasons already rehearsed not to embark on, or be part of, a road that leads inexorably to a single gender. There is also much more to a sex change than a womb transplant and, in the period of transition to an all-female world, this is unlikely to be beneficial to those poor men subjected to it or who subject themselves to it.

Finally, Sparrow has backed the wrong horse. The difference in life expectancy between men and women is almost certainly not located in the presence or absence of a womb, but in another altogether different part of the anatomy—the balls! Kirkwood again:

‘…neutered male animals often live longer than their intact counterparts. Indeed, the evidence supports the notion that male castration might be the ticket to a longer life’.12

Sparrow does not need to construct such an implausible case in which spare wombs just happen to be available next door—he can just offer all existing males castration and castrate all male babies at birth, and I hope I will not be anywhere near him when he makes this suggestion!

I believe I have given ample reason above why I have never advocated preferring women to men, and the foolishness of such a suggestion is a lesson I have not been late in the learning, so the reasons why I have never considered it and why it was and remains quite literally the last thing on my mind will be obvious to all who have studied these difficult matters.


Sparrow closes by attempting the philosophers nightmare—a charge of inconsistency:

‘In order to be consistent with his previously published claims, Harris will somehow need to explain why we have grounds for preference between any number of other small differences in capacity but no grounds for preference between the normal capacities of men and women—without making reference to the idea of the normal. This is, I think it is fair to say, a formidable task…

‘Until Harris provides such an explanation, his “harmed state” account cannot underpin a plausible ethics of human enhancement. When it comes to those aspects of human biology that differ between the sexes, the “emergency room test” works—if it works at all—by making reference to a notion of the normal capacities of each sex. Without the notion of the normal, we cannot avoid the conclusion that it is rational to prefer to be one sex rather than the other…

‘Thus Harris's account will imply that either male children or female children are born in a condition that it is rationally preferable to avoid and are thereby disabled. Moreover, parents will be obligated to use sex selection technologies to avoid harming their children by allowing them to be born the inferior sex; the force of this obligation will be precisely as strong as the obligation Harris believes parents have to enhance their children or to refrain from bringing them into existence with other unavoidable disabilities. The conclusion that 50% of the population are in a harmed condition—and that 50% of those parents who have conceived children naturally since the invention of sex selection technologies have harmed their children by allowing them to be born in this state—is too high a philosophical price to pay for Harris's enthusiasm for enhancement.’1

Sparrow is wrong when he says: ‘Harris's account will imply that either male children or female children are born in a condition that it is rationally preferable to avoid and are thereby disabled.’ (sic)

He is wrong because he is perversely narrow in his interpretation of what makes the lives of both genders worthwhile. My account might imply this, but only without the larger picture, which I have attempted to colour in above and elsewhere. This bigger picture shows the essential roles of men and women at least in the mainstream of reproduction unless genetic variety and novelty can be produced on a global scale by other means. This of course may eventually be possible, but probably not in a time scale or indeed on a population scale that will help Sparrow.

This context also shows why the notion of the normal plays no active, nor any theoretical, role at all. All we need is the big picture and a lively sense of what is or is not changeable, given our goals, which for me at least include human flourishing or the flourishing of our non-human descendents.

Of course you can divide any population of anything up in any way you choose and it would be remarkable if significant differences did not appear between the resulting groupings. It is no surprise to anyone that men and women differ in ways that for some purposes and in some circumstances confer advantages on one or the other group. I believe, and believe I have shown both here and elsewhere, that such intrinsic differences as exist are without moral or political significance, and I do not believe I am alone in finding it perplexing to decide which gender has the better life. For my own part I think the record shows that no one is more tenacious of life than me, nor more willing to strive for and accept any increases in life expectancy that might be available, and yet I would not change places with anyone. This of course may eventually be possible, but probably not on a time line, or indeed on a population scale, that will help Sparrow.

Sparrow has got one thing right. For the moment, and for the foreseeable future, we are going to need men and women and sexual reproduction or its equivalent so that we can say to those who might suggest a sex change (as Lola Montez may have put it) ‘absolutely not—Courage and shuffle the cards!’


The author would like to acknowledge the stimulus and support of the iSEI Wellcome Strategic Programme in ‘The human body: its scope, limits and future’, in the preparation of this paper, and to thank Sarah Chan for helpful comments.



  • Linked articles 039982.

  • Competing interests None.

  • Provenance and peer review Not commissioned; not externally peer reviewed.

Linked Articles

Other content recommended for you