PROCREATIVE BENEFICENCE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

In defence of Procreative BeneficenceJulian Savulescu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Why potential parents should select the best child of possiblechildren, and the necessity of a dialogue about the context of areproductive decision.

The principle of ProcreativeBeneficence is the principle of select-ing the best child of the possible

children one could have. This principle iselaborated on and defended against arange of objections. In particular, focus islaid on four objections that MichaelParker raises: that it is underdetermining,that it is insensitive to the complex natureof the good, that it is self-defeating andthat it is overly individualistic. ProcreativeBeneficence is a useful principle inreproductive decision-making. It is neces-sary to be more active in making selectiondecisions about what kind of child tohave.

Parker1 raises four objections to theprinciple of Procreative Beneficence (seepage 279). I will address these in turn.

(1) Procreative Beneficence isunderdetermining

Parker claims that ProcreativeBeneficence is underdetermining. By‘‘underdetermining’’, he means that theprinciple will not give clear and determi-nate answers as to which lives are betteror best. Parker argues that ‘‘rankingpossible lives as ‘‘better’’ or ‘‘worse’’ is‘‘highly problematic’’.

Ranking lives is a very complex matter.Let us distinguish between:

N the value of a whole life and

N the value of an individual feature of alife (eg, being short, having red hair,having a gene for baldness, and so on).

We should also distinguish betweenvaluation ex ante (prediction of the valueof a whole life or feature) and ex post(retrospective evaluation of a whole lifeor feature). In Procreative Beneficence, Ilikened genetic testing to playing thewheel of fortune.2 Just because we havea weak chance of winning, does not meanwe should not play the game. The onlyreason not to play a game that has a prizeis if the costs of playing are too high. Iaccepted the assumption that genetictests might only be weakly predictive oftraits that are only correlated with ahigher chance of having some valuableproperty, such as being less likely to have

a disease. Another valuable property ishaving some ability. When deciding touse a genetic test for a gene, we are onlymaking a decision about whether it isbetter to have that feature or not. We aremaking an ex ante trait evaluation. Weare not evaluating a whole person. Theseevaluations may come apart. A personwith an ability may still have a bad life, asthere are many other factors that make aperson’s life go well. The most talentedpainter might have his hands cut off by arogue farm machine.

Indeed, a trait that is generally goodmay itself contribute to a person’s lifegoing badly. Buchanan et al3 provides theexample of Cynthia, who uses her powersof empathy and her ability to understandother people’s feelings to con them.3

Imagine that Cynthia, as a result ofhaving the power and exercising it in thisway, ends up in jail, miserable andabused. We could plausibly say that shewould have been better off without thisability. This is an ex post evaluation of thetrait. Just because a trait might be bad forsomeone and be abused does not implythat we would not want our children tohave that trait or we should leave it tochance. Despite the possibility ofCynthias, it is rational to want ourchildren to be empathetic. Empathy is,on balance, likely to be a good thing for aperson to have. It is arguably a precondi-tion of being a moral person.

Moreover, we can grant, for argument’ssake, Parker’s claim that it is difficult toevaluate a whole life as better or worsethan another life. Smith gets cancer anddies a slow painful death at the age of78 years. Jones dies of a sudden heartattack at 72. Whose life was better?Indeed, it is hard to say, even if we knewmore details, or all the details.Nonetheless, we have a reason to preferan embryo that does not have a genepredisposing it to heart disease or cancerover ones that have these genes, otherthings being equal. And we have reasonsto prefer embryos with abilities ratherthan disabilities. The reason is given bythe badness of heart disease, cancer ordisability. That is the reason for couples

preferring not to have children withdiseases or disabilities. Even if we cannotknow the value of a whole life, we canknow that conditions are good or bad,and this provides a reason to prefer tobring children into existence withoutthose conditions. Indeed, all we can doin the world as it is to try to make our andour children’s lives go well, because weare not gods and we cannot control thefuture. Far from playing God, attemptingto control our genetic fate is ‘‘playinghuman’’—trying to improve the odds ofdoing well in an uncertain world ofdifficulty, threat and misfortune.Throwing up our hands and giving in toa sticky fate is hardly an admirablehuman trait, although some contempor-ary bioethicists seem paradoxically toextol it as virtue. I want to be the kindof human who lives longer and better, notshorter and badly.

Embryologists make these basic kindsof evaluation when they inspect in vitrofertilisation (IVF) embryos and selectthose whose gross morphology suggeststhat they are most likely to survive. Someembryos, they believe, are better thanothers. And when they are only going totransfer one from a batch of 10 or 12, theywant the best—the one with the featurespredictive of the best chances of survival.

Survival and health are not the onlystates that make it more likely that wewill achieve a better life. Buchanan et al3

have introduced the concept of general-purpose means—that is, traits that arevaluable no matter what kind of life aperson leads. Here are some putative all-purpose goods:

N Intelligence

N Memory

N Self-discipline

N Impulse control

N Foresight

N Patience

N Sense of humour

N Sunny temperament

N Empathy, imagination, sympathy, fair-ness, honesty, and so on

N Capacity to live peaceably and sociallywith others

Parker’s claim that ProcreativeBeneficence is underdetermining is notan objection to Procreative Beneficencebut an objection to placing a value on life.I have, so far, granted his assumptionthat we cannot value lives. There is,however, a large body of literature andapproaches to valuing lives.4–6 To claimthat we cannot evaluate life is to imply wecannot set priorities in health, research,social services and the distribution oflimited resources. It is to imply that we

284 REPRODUCTIVE ETHICS

www.jmedethics.com

group.bmj.com on August 11, 2013 - Published by jme.bmj.comDownloaded from

cannot say who has a better or worse life.But that seems radically wrong. We makethese kinds of evaluations all the time, forourselves, our children and others.Society is organised on the basis of suchevaluations. We have hospitals becausedisease and injury are bad, not becausethey are good or because we are indiffer-ent to them. Some people are held up asthe paradigm of a good life, whereasothers are said to have had a bad life. Theconcepts of virtues, strength of characterand character flaws all represent char-acteristics, which our normative languagerepresent as being good or bad—and thevirtuous person is a desirable person tobe.

It is important to remember that thealternative to selection is to leave thedistribution of traits to chance. Is ourignorance of what makes a good lifereally so great (or our trust in somesupernatural entity so strong) that wewant to leave the distribution of suchtraits to chance? Surely we can do betterthan chance? Evolution has selected forthe ability to survive long enough toreproduce. It is entirely indifferent toour well-being. Our environment hasradically changed in the past 10 000 yearsand even in the past 100 years.Reproduction (and, indeed, our biology)is not adapted to producing children bestsuited to living a good and fulfilling life inthese circumstances. But leading a goodlife is of great concern to most of us. If wecan do better than evolution, we shouldnot leave reproduction to chance andevolution.

As genetic tests multiply and moreinformation becomes available about dif-ferent embryos, it may be difficult tomake a decision about which constella-tion of genetic states is best. Is it better tobe more likely to be good at maths andabstract reasoning, self-absorbed withpoor interpersonal skills or have a sunnytemperament, optimism, good humourand good people skills? But we do notneed to provide a precise cardinal rankingof all possible lives in order for theprinciple of Procreative Beneficence tobe sufficiently determinative. Partialrankings may be possible. It may not bepossible to say whether A is better than B,but it may be possible to say that A or Bare better than C. This is enough torationally reject C. There will be constel-lations of traits that will be inferior toothers, and that is enough for ProcreativeBeneficence to be of value.

Parker raises the real case of Rachel,who is a carrier of spondyloepiphysealdysplasia tarda (SEDT). She is under-going IVF for infertility and wonderswhether she should use preimplantationgenetic diagnosis (PGD) to select embryos

that do not have this condition. Thecondition causes males to be short inadult life. They have a short trunk andbarrel shaped chest. Affected men tend toget some back and joint pain, and someosteoarthritis and restricted joint move-ment. In some, but not all cases, early hipreplacement (eg, in the 30s) and painmanagement is required.

The answer to whether this womanshould go on to have PGD is based in part,on the risks of PGD. With .1000 babiesborn after PGD, the risks appear minimal,although systematic follow-up isrequired. Let us assume that PGD is safe.She would then have a reason to test herembryos. This condition is a disability—resulting in stigma, pain and limited jointmovement. Hip replacements are notperfect and cannot fully restore comple-tely normal function. She should havePGD in this circumstance, at least in thesense that she has the most reason to usethe test. She would be irrational in failingto have this test. How is ProcreativeBeneficence underdetermining in thiscase?

Sometimes it is objected that condi-tions like SEDT are not disabilities. In asubsequent paper,7 I will argue for what Icall a biopsychosocial construction ofdisability. A disability is any state of aperson which:

(1) will reduce the goodness (value) ofa life (disability in the intrinsic sense), incircumstances, C; and/or

(2) reduces the chances of a personrealising a possible good life (disability inthe instrumental sense), in circumstancesin which the child will live, which we cancall ‘‘C’’.

On this account of disability, SEDT is adisability because it makes it moredifficult to achieve a good life.Importantly, circumstances C, constitutea complete relevant description of theworld and other aspects of a person. Theyhave biological, psychological, social andnatural external constituents. Given thelikely way in which society will beorganised (and the way the natural worldand people are likely to be), SEDT is likelyto be a disability— that is, an impedimentto the good life.

Parker claims that the concept of thebest life is dependent on a cluster of otherconcepts—a good life, flourishing, well-being. However, these concepts areroughly the same. I never denied thatthe best life requires a conception of well-being. He goes on to claim that ‘‘Thereference to the Use of such concepts …highlights too the fact that their meaningand the nature of their interrelatednessare sustained and transformed withinsocial practices.’’ I am not entirely surewhat this sentence means but if it means

that how well our lives go depends onsocial practices, that is clearly true. Whatconstitutes a disability depends on thecontext, on the way the world is. This hasnatural and social determinants. In thebiopsychosocial conception of disabilitythat I will offer,7 a state of the person(biology or psychology) could constitute adisability depending on the environment,including social practices, and alsodepending on that person’s other biolo-gical and psychological features. But themere fact that what constitutes a dis-ability is partly socially constructed, doesnot imply that we cannot evaluate somestates as disabilities relative to environ-ment, or select children on the basis ofthe way our society is likely to be.

Parker might alternatively mean thathow we define what constitutes a goodlife varies over time. He might be appeal-ing to a kind of cultural relativism aboutthe good. Such relativism as an accountof the good is highly suspect. Infibulationand child slavery are not good merelybecause some cultures approve of them.

Parker goes on to claim that ‘‘it is animplication of this that the interpretationin relation to particular cases of the dutyto have the best possible child is insepar-able from intersubjectively and sociallysustained discourses about human flour-ishing’’. It may be true that the applica-tion of Procreative Beneficence requires asocial discourse about human flourishing.It could also be true that what constitutesan ability or disability is partly sociallyconstructed. However, it does not followfrom these facts that ProcreativeBeneficence is underdetermining. I donot see how these arguments providereason to doubt that the ‘‘constellation of[these] concepts could be sufficientlyrobust to underpin a judgement in anyparticular case that what had beenchosen was the ‘‘best possible’ life’’. Theonly point that is of relevance in Parker’sargument is that disability and well-beingare context dependent. But the fact thatgrowing up being blind might be anadvantage in a world that suddenlydarkens does not show that blindness isnot a disability in our world and the wayour world is likely to be. SEDT is adisability in the world as it is likely to be,given the natural and social constraints,and this provides a reason to preferembryos that do not have the condition. Ican see no ‘‘good reason to doubt both thepractical utility and theoretical coherence’’of Procreative Beneficence based on thesearguments of Parker’s.

Parker then adds another alleged rea-son to doubt whether ProcreativeBeneficence is of any use: ‘‘it is extremelydifficult in advance, and perhaps alsoeven in retrospect, to say with any

REPRODUCTIVE ETHICS 285

www.jmedethics.com

group.bmj.com on August 11, 2013 - Published by jme.bmj.comDownloaded from

certainty what it is, or was, that makes(or made) a life go well. Is it true, forexample, that a life free of troubledinterpersonal relationships, free of suffer-ing, of loneliness or of misunderstandingis a better life, or even, taken as a whole, ahappier life, than one in which experienceof these to at least some degree hasplayed a part? Is it true to say that thegood life is the life free of any illness,disease or misfortune?’’

Parker here raises the problem ofprediction in the face of uncertainty. Ihave addressed this issue in my paper.7 AsI have argued here, when evaluating astate of biology, we are making an ex anteevaluation of whether that state consti-tutes a disability. Of course, we cannotknow what will be the best life, but weshould and do try to realise a good life forourselves and others. The person whoavoids any hard task or suffering, who isnot prepared to commit to pain ofrelationships and so misses out on thegoods of life, is just mistaken about whatmakes for a good life, and good may comeof adversity. But a rational person wouldnot choose to get cancer, unless there wasa good reason to have it. A parent whointentionally inflicted deafness on his orher child, or failed to treat it, would beabusing the child. Either illness, diseaseand misfortune make for a better life orthey do not—I do not hear any saneperson seriously suggesting we shoulddeliberately visit illness, disease or mis-fortune on people to help them live abetter life. Indeed, it would be a crime toharm them so. It is a mistake to thinkthat because life is unpredictable, andgood can come out of bad, we shouldchoose the bad, or be indifferent to it andallow it to occur. There is no difference,morally, between causing a harm anddeliberately and avoidably allowing it tooccur. Even if there is a difference, weshould not allow harm to occur when wecan easily and foreseeably avoid it.

Parker ‘‘gestures’’ towards a differentformulation of Procreative Beneficence:‘‘it is reasonable to expect that the childthey are thinking of conceiving is going tobe born under conditions conducive tothe possibility of a good life.’’ This,indeed, might be a different formulation.Rather than adopting a maximisingaccount, such as the one I have offered,we could construct a threshold account.What parents should aim at is having achild whose life is expected to be goodenough. The correlate of this is thatdisabilities should only be avoided if theyare severe enough. This threshold viewmight be what Parker means later whenhe states that ‘‘what is being rejected hereis only the pursuit of the best possible,not the pursuit of the good.’’

This threshold account has some sur-face plausibility. But its plausibility issmuggled in from intuitions about thecosts of attempting to maximise further,from the concept of opportunity costs.There is a good enough reason to acceptwhen the costs of gaining more informa-tion or seeking other options are prohibi-tive. Thus, satisficing consequentialism(the view that we should choose an actthe consequences of which are goodenough) is a version of maximisingconsequentialism. But as I argued pre-viously Procreative Beneficence providesone good reason for action.1 It is not theonly reason from which we might act. Theinterests of parents or reproducers alsoconstitute reasons. Couples should notundergo IVF and its risks if the harms aresignificant and the additional benefitssmall. An older couple with diminishingfertility might have more reason to accepta fetus with a cleft lip than a younghighly fertile couple who have no objec-tions to termination of pregnancy. Itmight be irrational for a couple to discardan otherwise healthy embryo that had amild predisposition to asthma andundergo a further cycle of IVF in anattempt to have a more healthy child.

However, when something better iscostlessly available, or available at a costthat is less than the benefit, there is noreason to settle for good enough. ImagineI win a prize to stay at the French Riviera.I am offered a very good hotel. The touristagency rings me back saying that I canexchange that hotel offer for anotherhotel that is excellent and better thanthe first. What reason can I have to refusethe offer? It might be that the first hotel iscloser to the water or has better food orlarger rooms. But this is just to questionwhether the second hotel really is betterthan the first. If it is equally close to thewater, has equally good food and equallylarge rooms and also has a balcony, or abidet, there is a reason to choose it.

In the same way, there is no reason fora person like Rachel who is alreadyundergoing IVF not to select the best ofthe available embryos she has (assumingthe test is safe). There is a reason infavour of the (apparently) normalembryo—that is, it does not have thegenetic disorder and the concomitantdisabilities. She might have a reason notto undergo further IVF if the risks aresufficient or the costs sufficiently high,even if say all the embryos in this batchare male and are affected by SEDT. But ifshe is already having IVF and PGD is safe,she should have PGD and select the best.

Now it may be that SEDT is sufficientlyminor that it does not warrant under-going IVF and PGD in cases of normalfertility for the purposes of selecting a

disease-free embryo. That is a moredifficult question, which depends on therisks of IVF and the couple’s attitudes tothose risks and a child with disability,and their other circumstances.

One upshot is that couples having PGDfor any reason should glean as muchinformation as possible from that test andutilise it in selection decisions. If a personis having PGD to exclude a geneticdisorder such as cystic fibrosis, theyshould also undergo chromosomal analy-sis. If a person is being screened for amajor genetic disease, they should also bescreened for all genetic abnormalities andobtain whatever genetic information thatcan be retrieved (subject to the con-straints of further harm or cost) has aprobabilistic relationship with the goodlife. So, if you can easily find out theexpected height of your child, or range ofintelligence, or some aspect of musical orathletic potential when looking forgenetic disease, you have a reason topeek into the genome and make adecision about which of your embryoshas, overall, the best genetic endowment,of those that are being considered fortransfer.

(2) Light, dark and the mingledyarn: the concept of the ‘‘best possi-ble child’’ is ‘‘paradoxical’’

In discussing Procreative Beneficence,it is important to separate two verydifferent questions:

(1) what constitutes a good life (or thebest life)?

(2) should we select an individual thatis expected to have better prospects of abetter life?

I have not committed myself to anyparticular substantive conception of thegood life. That is a complex question asold as philosophy itself. I believe the bestlife is a life of objectively worthwhileactivity that provides pleasure and isdesired.

Parker argues:

‘‘In All’s well that ends well,Shakespeare has a minor characterspeak the following lines, ‘‘The web ofour life is of mingled yarn, good andill together; our virtues would beproud if our faults whipp’d them not,and our crimes would despair if theywere not cherish’d by our virtues.’’In this, Shakespeare is not simplyreminding us that human lives are bytheir very nature characterised by bothgood and ill, and that we must learn tolive with these aspects of ourselves andof those around us. He makes thestronger and ultimately more interest-ing claim that both strengths andweaknesses of character and of our

286 REPRODUCTIVE ETHICS

www.jmedethics.com

group.bmj.com on August 11, 2013 - Published by jme.bmj.comDownloaded from

lives more broadly, are essential andinterdependent elements of the goodlife. Both aspects of our lives areinterwoven and, indeed, it is thisinterweaving and our struggles with itthat make us what we are andconstitutes in its interplay of light anddark, much that is of value andsignificance in human existence.

This echoes claims made by thePresident’s Council in Beyond therapy onthe value of suffering.

Traumatic memories, shame, andguilt, are, it is true, psychic pains. Inextreme doses, they can be crippling.Yet, short of the extreme, they canalso be helpful and fitting. They areappropriate responses to horror, dis-graceful conduct, injustice, and sin,and, as such, help teach us to avoidthem or fight against them in thefuture.8(p. 298)… there appears to be a connectionbetween the possibility of feeling deepunhappiness and the prospects forachieving genuine happiness. If onecannot grieve, one has not truly loved.To be capable of aspiration, one mustknow and feel lack.8(p. 299)

Let us assume that the best life requireswhat Parker describes, after Shakespeare,as light and dark. Fine. Then we shouldselect the right balance. Some have a lotof light and no dark; others are all dark.The issue is whether we should acceptwhat nature delivers up or make a choice.If we believe that it is better for people tohave some weaknesses, we had betterchoose embryos that have some weak-nesses. Parker simply denies this withoutargument, ‘‘This is not to say that the bestpossible life would be one in which acertain number of character flaws were‘‘thrown into the mix’’—for example,through preimplantation genetic diagno-sis.’’ But why not? Parker asserts thiswithout argument.

I believe it is perverse to suggest thatwe should allow biology, psychology orsocial situations to cause ill to people. It islike saying that we cannot have a happyor a good relationship without argu-ments. Arguments may occur, or theymay be necessary to resolve disagree-ments, but an otherwise perfectly happycouple should not start to argue, just tohave a mingled yarn.

Life will be a mingled yarn because ofthe unpredictable and uncontrollablenature of nature and life in general.People will necessarily face adversityand difficulty. And the pursuit of self-interest will always cause harm to others.

If, magically, we could remove the possi-bility of every natural disaster, everyhuman conflict, every human diseaseand make people instantly and perfectlyhappy, the question of how much ill theyshould experience might arise. But withthe world as it is and is likely to be evenin the face of greater technological con-trol, there will still remain plenty of meatof adversity, bad luck and human suffer-ing to get our teeth into.

Parker conflates the two questions out-lined above. His point is not about whetherwe should select, but about what consti-tutes a good life. Another possible inter-pretation of the light, dark and mingledyarn argument is that it is constitutive of agood life to accept what nature delivers.This is like Sandel’s9 claim that enhance-ment is wrong because we must be ‘‘opento the unbidden’’. We must remain open tothe mystery of life, which inevitably con-tains good and ill, and not seek to controlevery aspect of life. Again, plenty ofmystery will remain even if we seek toimprove our biological, psychological andsocial circumstances to make our lives gowell. One can choose to go to a good playrather than a poor one, and still experiencethe mystery of events as they unfold.

Moreover, insofar as this objection hasany force (which is, I think, minimal inthe predominantly uncontrollable worldof hurricanes, tsunamis, volcanic erup-tions, and human choice and evil), it hasno force when it comes to selection ratherthan enhancement decisions. Whenselecting among embryos, one is simplyselecting from what nature has created,rather than allowing nature or chance todecide which embryo is implanted. Sincenature rarely, if ever, creates a perfectembryo (all of us have 3–5 recessivemutations and countless genetic flaws),this version of the mingled yarn objectiondoes not apply to selection. Embryos,even the best of the bunch, will be amix of good and ill.

If one believes in accepting all sorts ofuncertainties, like the Dice Man,10 oneoption would be not to choose—not tohave the embryos diagnosed, or to roll adice even if the diagnosis is alreadyknown and leave it to chance. Such anapproach was disastrous for the DiceMan. We should not be Dice Men,maximally open to the unbidden.

(3) Self-defeatingParker claims it is self-defeating to

attempt to choose the best possible child.This is a familiar objection to consequen-tialism, which has been much discussedin the literature. It is related to Sandel’s9

objection that enhancement can result inparents being overbearing in trying torealise their child’s enhanced talents andultimately constrain that child’s life.9

I fail to see how the self-defeatingobjection can apply to our biological andpsychological potential and abilities. Howcan the capacity to remember thingsbetter, run faster or cope with psycholo-gical distress better imply that one is lesslikely to achieve the good life? It may beself-defeating in some circumstances to aimdirectly at achieving the good, but it issurely sensible to aim directly at achiev-ing the potential to be able to realise thegood. No one suggests it is self-defeatingto try to provide a good education to ourchildren because it is self-defeating todirectly aim for the good. If it is not self-defeating to alter the educational envir-onment to maximise our children’spotential and opportunities, why is itself-defeating to more directly intervenein their psychology or biology?

It is plausible that, sometimes, it is self-defeating to directly aim for the goods inlife, such as friendship.11 Sometimes, hardwork, difficult relationships and delayedgratification are necessary to achieve thegreatest goods. But how friendly andamiable we are, how hard we can work,whether we can delay gratification andeven our sense of fairness have somebiological basis.12 13 In so far as these arenecessary for the good life, we can selectembryos that have greater potential to(indirectly) pursue the good life.

(4) Overly individualistic, socialeffects and the public interest

Parker claims that ProcreativeBeneficence is overly individualistic. Thisagain is an objection to how we shouldconceive of a good life, and the factorswhich make for a good life, and not aboutselecting a child who is expected to havethe best chance of the best life. I have notdenied that our good is context depen-dent and that social and political factorsinfluence how well our lives go. This isnot an objection to ProcreativeBeneficence—it is an objection to aversion I did not offer.

There is a real sense, not alluded to byParker, in which Procreative Beneficenceis too individualistic. In some cases,10

choosing the child who has the best lifewill harm others. Say, for example, aperson is very manipulative, charismaticand dominating. Some great leaders mayhave been like this. It is plausible thattheir lives could go very well while thosearound them suffer. Bioconservativesstandardly oppose selection for highintelligence because they claim that thisis a largely positional good (I disagree)and that other less intelligent people areharmed. de Melo-Martin raises a versionof this objection in relation to physicalstrength, which she believes is a posi-tional good, and argues that procreativebeneficence would be self-defeating.14 It

REPRODUCTIVE ETHICS 287

www.jmedethics.com

group.bmj.com on August 11, 2013 - Published by jme.bmj.comDownloaded from

would be self-defeating, and also beharmful due to waste of resources thatare necessary to bring about zero benefits.She goes on to argue that, especially ifonly available on an ability to pay basis,selecting the best children would increaseinequality and injustice.14(pp 81–2)

It is plausible that their lives could govery well while those around them suffer.There may be legitimate public interestgrounds for interfering in reproductionand procreative liberty when that interestis also in accord with ProcreativeBeneficence. For example, I argued thatin certain rare and extreme cases, couplescould be coerced to have children who arefree of a strong genetic predisposition toviolent crimes.15 The public interest is arare although legitimate ground for inter-ference in procreative liberty.

The issue of positional goods is adifficult one that I cannot address fully.I will make several brief points.

First, in the case of selection (but notsome kinds of enhancement), we are onlychoosing between possible lives thatcould have existed. Natural inequalityexists—some people are born naturallysmarter than others. Allowing selectionwould, in one sense, only level up. Itwould reduce inequality, especially ifcheap and affordable.

Second, some kinds of inequality aretolerable. Just because we cannot makeeveryone live longer or be healthier doesnot imply that we should not make somepeople live longer and healthier lives. Ifthis applies to length of life and health, italso applies to what ultimately matters:well-being. Length of life and health areonly valuable insofar as they contribute toour well-being.

Third, when we withhold choice from acouple, we are responsible for the out-come, even if nature delivers it. When weknowingly and avoidably decide to pre-vent people from avoiding the naturallottery, we are responsible for its results.So the case of withholding selectionresulting in a child with worse prospects(for the public interest) is morally equiva-lent to the case of forcing parents to havea child with worse prospects (for thepublic interest). We would all recoil froma proposition that, because there are toofew people to do unpleasant and unpres-tigious jobs, natural selection be used toensure that some parents have childrenwith lower capacities but sunny disposi-tions who would be happy in these jobs.The decision to have a child with lessthan the best prospects in the publicinterest (even if indirectly through the

intentional use of the natural lotteryrather than deliberate selection) must bea last resort, if it is a resort at all.

Fouth, the objection that the use oftechnology to select better children willincrease inequality because it will only beavailable to the rich is a distraction. It is notan objection to Procreative Beneficence. Ifthe benefits are important, they should befreely available. And if they are not, this isreally a question of how far individuals canuse personal wealth to advance their ownand their children’s opportunities andwelfare. The same objection applies to thepurchase of many biological, psychologicaland social advantages. It is the sameobjection that applies to the purchase ofbetter healthcare, education, technologyand jobs for our children.

How we treat people is logically andpractically independent of what set ofbiological, psychological and social capa-cities and opportunities they are born with.Because somebody is born with a lower IQ,lower impulse control or is less attractive(whether or not these result from naturalor genetic selection), does not dictate howthese people are treated. That is our choiceand a matter for the social policies andlaws that we introduce. Natural inequalityexists and we require social institutions toensure that everyone has a fair go, a goodenough chance of a good life. The sameapplies to a world of selection.

FINAL REMARKSParker closes by suggesting that ‘‘wherehealth professionals have concerns aboutthe quality of the life being created, suchas— for example, in Rachel’s case above,it will be incumbent upon them to helppotential parents to think carefully aboutthe life they are about to create. Thehealth professionals involved will haveobligations to encourage people to reflecton their choices, to give reasons, and todebate with them the moral dimensionsof their choices’’. I suggested this pre-cisely 10 years ago in relation to thedoctor–patient relationship.16 In thereproductive sphere, we require a similardialogue about the context of a reproduc-tive decision and about the options andwhat makes for the best life in thatcontext. There might be circumstancesin which there is most reason to have achild with what appears to be a disability.But these will be rare. Although it may bemysterious what makes the best life, it isnot so puzzling or intersubjective whatconstitutes an ability, a talent, a potentialor an opportunity. Just as we seek toselect healthy children rather than those

with predispositions to disease, we shouldalso select children with abilities ratherthan disabilities. There are reasons toselect the best child, even if in the light ofsuch reasons and arguments presented tothem, parents should be free to select orhave less than the best child.17

J Med Ethics 2007;33:284–288.doi: 10.1136/jme.2006.018184

Correspondence to: Professor J Savulescu,Oxford Uehiro Centre for Practical Ethics,Littlegate House, St Ebbes, Oxford OX1 1PT, UK;julian.savulescu@philosophy.ox.ac.uk

Received 3 July 2006Revised 3 July 2006Accepted 4 July 2006

Competing interests: None.

REFERENCES1 Parker M. The best possible child. J Med Ethics

2007;33:279–83.2 Savulescu J. Procreative Beneficence: why we

should select the best children. Bioethics2001;15:413–26.

3 Buchanan A, Brock DW, Daniels N, et al. Fromchance to choice. Cambridge: CambridgeUniversity Press, 2000:167–70.

4 Harris J. The value of life. London: Routledge andKegan Paul, 1985.

5 Broome J. Weighing lives. Oxford: OxfordUniversity Press, 2004.

6 Singer P. Practical ethics. Cambridge: CambridgeUniversity Press, 1979.

7 Savulescu J. The biopsychosocial construction ofdisability: the moral obligation to have disability-free children. In preparation.

8 Kass LR, Blackban EH, Dressen KS, et al. Beyondtherapy: biotechnology and the pursuit of happinessWashington, DC, 2003:298, 299. http://www.bioethics.gov/reports/beyondtherapy/beyond_therapy_final_webcorrected.pdf (accessed5 Mar 2007).

9 Sandel MJ. The case against perfection. At Mon.2004: 293,51–62).

10 Rhinehart L. The Dice Man. London: HarperCollins,1999.

11 Cocking D, Oakley J. Indirect consequentialism,friendship, and the problem of alienation. Ethics,1995;106,86–111.http://links.jstor.org/sici?sici = 0014-1704%28199510%29106%3A1%3C86%3AICFATP%3E2.0.CO%3B2-L(accessed 5 March 2007).

12 Liu ZJ, Richmond BJ, Murray AE, et al. DNAtargeting of rhinal cortex D2 receptor proteinreversibly blocks learning of cues that predictreward. Proc Natl Acad Sci USA2004;101:12336–41.

13 Mischel W, Shoda Y, Peake PK. The nature ofadolescent competencies predicted by preschooldelay of gratification. J Pers Soc Psychol1988;4:687–96.

14 de Melo-Martin I. On our obligation to select thebest children: a reply to Savulescu. Bioethics2004;1:72–83.

15 Savulescu J. Why genetic testing for genes forcriminality is morally required. Princeton J Bioeth2001;4:79–97.

16 Savulescu J. Rational non-interventionalpaternalism: why doctors ought to makejudgements of what is best for their patients. J MedEthics 1995;21:327–31.

17 Savulescu J. Deaf lesbians, ‘‘designer disability,’’and the future of medicine. BMJ 2002;325:771–3.

288 REPRODUCTIVE ETHICS

www.jmedethics.com

group.bmj.com on August 11, 2013 - Published by jme.bmj.comDownloaded from

doi: 10.1136/jme.2006.018184 2007 33: 284-288J Med Ethics

 Julian Savulescu In defence of Procreative Beneficence

http://jme.bmj.com/content/33/5/284.full.htmlUpdated information and services can be found at:

These include:

References

http://jme.bmj.com/content/33/5/284.full.html#related-urlsArticle cited in:  

http://jme.bmj.com/content/33/5/284.full.html#ref-list-1This article cites 8 articles, 4 of which can be accessed free at:

serviceEmail alerting

box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in the

CollectionsTopic

(28 articles)Prison medicine   � Articles on similar topics can be found in the following collections

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:

group.bmj.com on August 11, 2013 - Published by jme.bmj.comDownloaded from