The Nonidentity Problem

First published Tue Jul 21, 2009; substantive revision Fri Jul 19, 2024

The nonidentity problem raises questions regarding both the moralobligations agents have in respect of possible future people –people, that is, who do not yet but may exist at some future time– and how those obligations (to the extent we have them) aremost credibly explained. It today remains among the most challengingproblems in all of population ethics.

It is plausible to think that we are often obligated to make futurepeople – at least some of them – better off rather thanleaving them as they might otherwise be or making them worse off. Buton reflection it seems that any change we contemplate in respect ofany conduct we plan to engage inprior to a givenperson’s being conceived – even a change that would on itsface seem to represent a clearimprovement for that futureperson – will often fail to makethat person better offand often instead serve only to bring another person, abetteroff person but still anonidentical person, intoexistence in place of the one. Thus the phenomenon Gregory Kavkacalled the “precariousness” of existence (Kavka, 1982,93): we all just barely missed never coming into existence at all. Hadour parents, or any of our more remote forebears, done anything otherthan what they in fact did – had they, for example, with the aimof making things better forus, delayed their conception of achild for a few months or years until they were better prepared totake care of a child – the odds overwhelmingly seem to favorour never having existed at all. That is, after all, the wayof gamete production: neither sperm nor egg cells survive for verylong, and it is reasonable to suppose that a different egg and spermcombination would have produced a distinct person. Even delays not ofmonths or years but of split seconds would seem very probably toswitch out the sperm cell that is set to do the inseminating for oneof its many, many cohorts and thus the identity of the personconceived. Analogous points hold for changes in the manner ofconception and, even more obviously, for changes in the agent’s choiceof whom to conceive a child with. Any of your forebears’well-meaning changes in conduct may well have resulted in the cominginto existence of a better off child. But foryou the resultof any such change very probably would have beenyour neverhaving existed at all.

A similar phenomenon arises on a global scale. Suppose, for example,that we choose to take significant action today to mitigate theeffects of climate change in the future – and that we do so withthe aim of making future people better off. Will our choice make anyparticular future person or collection of future peoplebetter off? Probably not, according to the logic of the nonidentityproblem. After all, any significant climate action undertaken todaywill surely affect the timing and manner of human conception on a verywide scale over a very long period of time. Thus, according to thelogic of the nonidentity problem, our environmentally responsiblechoice very probably will not make any particular person or collectionof future people better off than that collection would have been underthe environmentally reckless choice. It very probably will serve onlyto bring an entirely distinct – better off but person-by-personnonidentical – collection of future people intoexistence instead.

It is when those graphic facts of life are combined with a highlyintuitive constraint on just when a choice is morally wrong and when agiven possible outcome, or possible world orfuture, ismorally worse than another – that is, theperson affectingintuition – that what is called thenonidentityproblem arises.

The person affecting intuition itself was described by Derek Parfit asthe idea that “what is bad must be bad forsomeone” (Parfit 1987, 363). Jan Narveson wittilyarticulated just how that intuition functions to constrain anotherwise compelling utilitarian sentiment that, taken too far, mayseem highly implausible: moral agents are (of course) in favor of“making people happy” but (surely) “neutral”about “making happy people” (Narveson, 1976, 73).

How, exactly, does that simple intuition give rise to a problem? Theperson affecting intuition, when combined with the fact that thecoming into existence of any particular person is highly precarious,seems to insist that, when it comes to the future-directed,identity-affecting choices that we make, we can “do nowrong.” Even imagined improvements in our original choice thatwe might contemplate making with the aim of making things better forsome particular future person or another very often will have theeffect not of making things better for that person but rather ofleaving that person out of existence altogether. On the assumptionthat that person’s existence under the original choice is worth havingand that no one else is affected, the person affecting intuition isgenerally understood to compel the conclusion that our originalchoice, not being “bad for” that person at all (if anything, itbenefits that person; in bestowing on that person the gift oflife, itincreases what we can call that person’swellbeing), isn’t “bad”: it isn’t, that is, morally wrong anddoesn’t make the possible future it helps to bring about morally worsethan any other.

Surely, however, it is not the case that we can “do nowrong” with respect to the future-directed, identity-affectingchoices that we make. More plausibly, such choices are on manyoccasions morally wrong and the futures that include such choicesmorally worse: as when, for example, we fail to prepare ourselvesproperly to care for our own future children or fail to serve asenvironmentally responsible moral agents on behalf of futuregenerations who, just like us, will be in need of a climate conduciveto good health.

By the 1980s, the nonidentity problem had become widely recognized.Once recognized, it then seemed immediately obvious to many moralphilosophers – including many consequentialists – how theinconsistency was to be avoided. It seemed obvious that a morecredible moral theory would simply reject the person affectingintuition and instead adopt a structure that isimpersonalrather thanperson affecting in nature. The paradigm exampleof such a theory is classic maximizing total utilitarianism (the“total theory”). On that theory, agents are to focus, not on makingparticular people better off, but rather on bringing it aboutthat, regardless of justwho does or will exist, thecollection of all those people has as much wellbeing on anaggregate basis as anyalternate collection ofpeople who could have existed instead. That our choice isn’t “bad for”anyparticular future person becomes, under the total theory,morally irrelevant: a choice may be deemed morally wrong, and thefuture that choice gives rise to deemed morally worse, even if thatchoice makes things worse for no one who does or will exist in thatfuture at all.

The move to impersonalism – to, for example, the total theory– doesn’t, however, end the discussion. For we then come face toface with still another challenging population problem, that is, therepugnant conclusion. Having dropped the person affectingintuition, the total theory would seem to obligate us (in Narveson’swords) to make people happyand to make happy people. But inbalancing those two moral desiderata, we find that our obligation onoccasion will be to make additional happy people even though our doingso comes at a grave cost to still other people, people who do or willexist and who would have been better off under the choice that createsfewer people rather than under the choice to create more. Thusconsider the collection “A” consisting of a large number of futurepeople whose lives will bewell worth living, and compare itagainst a collection “Z” consisting of all the A-people together witha very large number of additional future people, with each A-personand each additional person now having a life onlybarelyworth living. In a case where the future that contains the Z-peoplemaximizes aggregate wellbeing, the total theory deems that futuremorally better than the future that contains the A-people alone, withthe morally better future being the one that (given that there existsno still better future) we are morally obligated to aim to bringabout. But can that conclusion really be correct? Parfit himself foundit “repugnant” (Parfit, 1987, 380–390; see alsorepugnant conclusion). And it would seem to be, on its face, the sort of “moral”edict we would resist, indeed, defy – just as we would defy, inthe more local, closer-to-home case the edict that we are in manycases obligated to produce ever more offspring even when our doing sowould make things much worse for the (one or two or more) existing orfuture children we might ourselves otherwise have preferred toproduce.

Notably, the repugnant conclusion has itself been challenged, on thegrounds that either the conclusion itself is not truly repugnant orthat almost all credible population theories give rise to some form ofthe repugnant conclusion or another (Zuber et al., 2021, Tannsjo2002). Still, Parfit’s early assessment of the conclusion as“repugnant” continues to be the dominant view in moralphilosophy writ large, and the repugnant conclusion, together with thenonidentity problem, continues to define the basic contours of thecontemporary controversy surrounding the person affectingintuition.

It is thus generally thought that to solve the nonidentity problem isto produce a theory that accounts for the obligations we intuitivelybelieve we have in respect of possible future people without enmeshingourselves in any of the other serious problems that togetherconstitute the area of population ethics. It’s – amongother things – to identify a “Theory X” (Parfit1987, 378) that manages to avoid the nonidentity problemandthe repugnant conclusion. The path forward hasn’t, however, been easyor obvious. For that reason and others, the nonidentity problem opensthe door to significant and still unsettled questions regarding thenature and structure of moral law.

1. Construction the Problem; Strategies for Solving the Problem

As noted earlier, the person affecting intuition is a centralcomponent in the construction of the nonidentity problem. Indeed, itwas originally considered thetarget of the nonidentityproblem.

Many different formulations of the person affecting intuition appearin the literature (see Roberts 2024, 4–5 n. 8). For purposes of thisentry, formulations that are widely agreed to fail on grounds otherthan the nonidentity problem are set aside. John Broome, for example,considers a formulation he calls theneutrality intuition,according to which bringing an additional person into existence withina certain “neutral” range of wellbeing levels leaves the one future(other things equal) exactly as good as the other. As Broome argues,however, the neutrality intuition is clearly false (Broome 2004,143–149).

An initial, more cautious formulation, and one that tracks Parfit’soriginal formulation (“what is bad must be bad for someone,” Parfit1987, 363 (emphasis deleted); cf. Parfit 2017), is as follows: achoice made at a given outcome, or possible world or future, ismorally wrong, and makes the one future worse than a second, only ifthat choice makes things worse for, or harms, or is otherwise “badfor,” at least one person who does or will exist in the one future.

This initial formulation is cautious in three respects. First, itavoids assigning any special moral status to people who do or willexist in theactual world over people who do or will exist inalternatemerely possible worlds. Consistent with the initialformulation, we may freely assign moral status to merely possiblepeople who inhabit futures that themselves never in point of fact– that neveractually – obtain. (For argumentsagainstmoral actualism, see Hare 2007; Roberts 2024,45–49; see also part 4 below.) Second, it provides only anecessary, and not a sufficient, condition on when a choice is wrongor makes things morally worse; relatedly, the condition it provides isa condition only on moral wrongness and moral worseness, not on moralbetterness. (See Roberts 2024, 4–5n.8 and 60–61.) Third, it leavesopen the question of just which future it is that makes the person whoisworse off in the one futurebetter off. Often,the reference is assumed to be to the second future (we would say onefuture is worse than a second only if the one future is worse for aperson who does or will exist in that future than the second is forthat same person). In fact, however, to avoid the risk of starting offwith a formulation that cannot be made to work, the question of howthat blank is to be filled in must be left open.

Though cautious, this initial formulation of the intuition, togetherwith the phenomenon of the precariousness of existence and oncecombined with a handful of seemingly uncontroversial ancillary claims,is enough to trigger the nonidentity problem.

Some of those ancillary claims are moral in nature; others are moreconceptual, metaphysical or even epistemological in nature. When itcomes to identifying a credible way to avoid the inconsistency thenonidentity problem gives rise to, however, close scrutiny not just ofthe central claims but also of each seemingly uncontroversialancillary claim is in order.

The discussion that follows begins with a proposed solution thatboldly questions one of those ancillary (and already referenced)claims: that the future-directed, identity-affecting choices underscrutiny in the various nonidentity cases really are, in the main,morally wrong and really do make the futures that arise out of thosechoices morally worse than still other futures. As the nonidentitycases themselves suggest (for a list, see part 2 below), that claimcomes with strong intuitive support. Nonetheless, a few theorists haveaimed to solve the problem by rejecting that claim. And indeed anarrow class of nonidentity cases involving highly personal,procreative choices may well seem to throw that claim into question.Was it, for example, wrong for your parents to have a child with eachother and, eventually, to produce you rather than to have a child withan alternate partner who (let’s suppose) would have brought tothe table the sort of superior genetic contribution that would haveincreased wellbeing? We might think that it wasn’t. Fordiscussion of this “bite the bullet” proposal, see part 3.1 below.

More commonly, theorists, including many consequentialists, haveviewed the nonidentity problem as a counterexample against the personaffecting intuition and have thus aimed to solve the problem by simplyrejecting that intuition in favor of a purely impersonal, aggregativeapproach. Keeping the repugnant conclusion in mind, however, thosesame theorists have proposed novel versions of the total theory,versions designed to avoid the repugnant conclusion without the helpof the person affecting intuition. Still other theorists, who also canbe understood as working within a basic consequentialist framework,attempt to thread the needle between the nonidentity problem and therepugnant conclusion by adopting a pluralistic approach, one thatrecognizes both impersonal values (thus addressing the nonidentityproblem) and person affecting values (thus going some way inaddressing the repugnant conclusion). See part 3.2 below.

Still other theorists, seeking to retain the person affectingintuition without giving up the claim that the choices under scrutinyin the nonidentity cases really are wrong and really do make thingsworse, aim to solve the nonidentity problem by disconnecting thenotions of when it is that a choice makes thingsworse for aperson and when it is that a choice isbad for a person.Conceding that, given the precariousness of existence, the choiceunder scrutiny in any given nonidentity case does not make thingsworse for the future person it ushers into existence (assuming, ofcourse, that that existence is itself worth having), they argue thatthat same choice nonetheless harms or wrongs or is otherwise bad forthat person. See part 3.3 below,

A handful of theorists, also seeking to retain the person affectingintuition and the claim that the choices under scrutiny really arewrong and really do make things worse, have questioned some of theancillary conceptual or metaphysical claims that ground seeminglycompelling claims regarding the precariousness of existence. Thus inconstructing the problem certain very definite claims regarding whatit is for a personp to be the same person as a personq, when it is that a nonidentical personq existsinstead of a personp and, ultimately, when it is that “aperson” is made worse off are at work. The question has, however,emerged whether those claims – in effect, those assumptions– do not inappropriately block various natural descriptions ofthe cases that nicely avoid the nonidentity problem. See part 3.4below.

Still other theorists have questioned whether facts regarding theprecariousness of existence in fact support the position that a givenperson’s coming into existence is more probable under the choice thatwe agree is wrong than that same person’s existence is under analternate choice that we agree is permissible. That more epistemicproposal, which applies just to some but not all types of nonidentitycases, argues that the “no harm done” result rests on a certainfallacy in reasoning regarding the relevant probabilities. By avoidingthe fallacy, we can explain how the wrong choice after all harms– makes things worse for – the future person whose futureplight has triggered our concern and thus explain, in person affectingterms, why that choice is wrong and the future that it gives rise tois worse than others. See part 3.5 below.

Taking another tack entirely, many theorists are happy to reject boththe consequentialist framework itselfand the personaffecting intuition. Such theorists may aim to account for thewrongness of many future-directed, identity-affecting choices byreference to facts about the agent’s character, motives orintentions rather than by reference to the consequences the choiceunder scrutiny generates for anyone at all. See part 3.6 below.

Just as the repugnant conclusion (arguably) puts guardrails onproposed solutions to the nonidentity problem, so do certain otherproblems in population ethics, including what is called “theasymmetry.” This entry concludes with a note on that problem. See part4 below.

2. Nonidentity Cases

As noted earlier, the identity of any person plausibly depends on justwho the genetic parents of that person happen to be. As Parfitobserves, the woman who wonders who she would have been had her“parents married other people” “ignores theanswer,” which is “no one” (Parfit 1987, 351). Butthe timing and manner of conception are also critical. Had the timingof any one conception or the manner of that conception been changed,the result very probably would have been the conception of a distinctchild altogether. (Parfit 1987, 351–55; 2011 vol. 2,217–231). Moreover, the exact timing and manner of conception isitself highly susceptible to variations in whatever complex chain ofacts and events it is that has come before. Much of what has been donein human history, had it been done differently, would surely haveundone the conceptions of vast numbers of people. “[H]ow many ofus could truly claim, ‘Even if railways and motor cars had neverbeen invented, I would still have been born’?” (Parfit1987, 361).

Consider, then, a choicec1 that is made at a possible worldor futuref1 and that eventually results in one or morepeople having positive but low wellbeing levels inf1. Andsuppose that, at the time of choice, those very same people have notyet been conceived. Suppose, too, that the reason wellbeing levelswill be low is thatc1 performed atf1 will causethose very same people to suffer in various ways (say, from disease,deprivation or limitation). And suppose that that suffering cannot bejustified (on, say, the ground that it is necessary to save many otherpeople from suffering still more). When agents, at the time of choice,had an alternate choicec2 at an alternate possible futuref2 that would have avoided that suffering altogether, we maywell be convinced thatc1 atf1is wrong. The problemis that in many such cases – and it takes just one to generate acounterexample against the person affecting intuition – choosingc2 in place ofc1 very probably would also havechanged the timing and manner of the conceptions that take place atf1. Thusc2 would not have made things better forthe people who will eventually exist atf1 but rather wouldhave caused those particular people – each and every one of them– very probably to have been left out of existence atf2 altogether. To put the point another way, the sufferingc1 imposes on the people who will eventually exist inf1 seems very probablyunavoidable if thoseparticular people are ever to exist at all. (They all“owe” their “very existence” to agents’choosingc1 atf1 rather thanc2 atf2.) That, in turn, makes it hard to see howc1 hasmade things worse for, or harmed, or otherwise been bad for, anyonewho does or will exist inf1 in any “morally relevantway” (Parfit 1987, 361–63, 374).

2.1 Depletion

In Parfit’s depletion case, we are to suppose that agents as acommunity have chosen to deplete rather than conserve certain naturalresources. The consequences of that choice for the people who existnow or will come into existence over the next two centuries will be“slightly” better than under a conservation alternative(Parfit 1987, 362; see also Parfit 2011 (vol. 2), 218). After thatperiod, however, the quality of life would be much lower. Yet at thesame time the depletion choice would seem very probably to make thingsworse for, or harm, or otherwise be bad for, no existing or futureperson at all. After all, while distant future people by hypothesiswill suffer as a result of the depletion choice, it is also true thatfor each such person a conservation choice very probably would havechanged the timing and manner of the relevant conception. That change,in turn, would have changed the identities of the people whoeventually exist. Assuming that the existences those distant futurepeople will have are worth having despite depletion, we seem forced toconclude that the depletion choice does not make things worse for, orharm, and is not otherwise bad for, any among those distant futurepeople at all. The person affecting intuition then forces a furtherconclusion: that the depletion choice is not wrong and does not makethe future in which the depletion choice is made worse than anyalternate future. And thus the problem: for surely the “ greatlowering of the quality of life must provide some moral reason not tochoose Depletion” (Parfit 1987, 361–364). Surely agents ought tohave chosen conservation instead.

The risky policy case, also from Parfit, has a similar structure(Parfit 1987, 371–72). So does the climate change case (Broome1992; Parfit 2017, 122–123).

2.2 The slave child

The same type of nonidentity problem arises but at a more local levelin Kavka’s slave child case. In exchange for $50,000, a coupleenters into a binding contract according to which the couple willconceive and bear a child who will be transferred at birth to awealthy man as a slave (Kavka 1982, 100). The child is conceived andborn pursuant to the contract – and, as a slave, suffers butstill has a life worth living. We think what the couple has done is“outrageous” (Kavka 1982, 101). And at first glance we mayalso think that that is so because of what the couple’s choicedoes to the slave child – specifically, that it makes thingsworse for, or harms, or is otherwise bad for, that child. But thelogic of the nonidentity problem suggests otherwise. Had the coupledeclined to enter the contract but still taken steps to produce achild, the original child very probably would never have existed atall. For that alternate sequence of choices very probably would haveproduced a change in the timing and manner of conception. Theresulting child, as a nonslave, may well have been better off. But itvery probably would have been a child “nonidentical” to the original(Kavka 1982, 100 n. 15). On those facts, we seem forced to concludethat the couple’s choice makes things worse for, or harms, or isotherwise bad for, no existing or future person at all. According tothe person affecting intuition, what the couple has done thereforeisn’t wrong and doesn’t make things worse. But that result seemsclearly false.

Kavka’s “pleasure pill” case parallels the slavechild case (Kavka 1982, 98).

2.3 The fourteen year old girl

Another case from Parfit on its face avoids any reliance onprobabilities and thus can be considered of a different type. His“fourteen year old girl” case focuses on the fact that the egg cell ayoung girl will produce soon after she becomes capable of conceiving achild will not be the same egg cell as the one she will produce adecade or more later. In Parfit’s case, “[t]his girlchooses to have a child” (Parfit 1987, 358). Yet we think thatit would have been better for the girl to wait “for severalyears” to have a child and that, barring exceptionalcircumstances, the girl’s choice is wrong (Parfit 1987, 358).But it also seems that that particular child, even if burdened byhaving been born to a girl too young to be a good mother, practicallycould not (versus probablywould not) have existedat all had the girl waited “for several years” to have achild. Given that the child’s life is worth living, it’shard to see how the girl’s choice can plausibly be said to havemade things worse for, or harmed, or otherwise have been “badfor,” the child or (on some versions of the case) any otherexisting or future person at all. On those facts, the person affectingintuition insists that the young girl’s choice was not wrong and doesnot make things morally worse.

2.4 Wrongful life

The same type of nonidentity problem appears in defenses against thecause of action in the law of negligence calledwrongfullife. An action for wrongful life, by definition made on behalfof the child rather than the parent, makes one or both of thefollowing claims: (1) medical professionals have failed properly toadvise couples of their risk of producing a child with a seriousgenetic or chromosomal disorder or of the availability of technologiesthat would have enabled them to reduce that risk; and (2) medicalprofessionals have failed to implement those technologies effectively.The disorders at issue include Down syndrome and hereditary deafness,and the available technologies include preconception genetic testingand maternal blood screening. The proper use of the availabletechnologies could not have cured the underlying impairment. But theirproper use could have enabled the couple to produce a nonidentical,presumptively better off, child instead. At the same time, it seemsclear that in most cases the impaired child’s life will be (orat least can be made to be) worth living. Can the child claim to havebeen harmed – in the comparative, worse off sense of that termthat the law requires – by the very medical negligence to whichthe child owes its very existence? It seems not. The law of negligenceitself being deeply steeped in the person affecting intuition, amajority of courts have felt themselves forced to deny the child’sclaim altogether.

Parfit’s “two medical programs” case –specifically, the program that would encourage women who have acertain condition that “cannot be treated, but always disappearswithin two months,” to delay pregnancy – is structurallysimilar to the wrongful life case (Parfit 1987, 367–368; Parfit2011 vol. 2, 221–222; Parfit 2017, 121).

2.5 Historic injustices; reparations

The precariousness of existence raises the question whether historicalinjustices – for example, U.S. slavery and the Holocaust –can properly be said to make things worse for, or harm, or otherwisebe bad for, people who were not conceived until after the eventsconstituting the injustice have themselves concluded (Sher 2005;Herstein 2008; Cohen 2009; Smilansky 2013). Here tracking the sameprobabilistic logic that appears in the cases of depletion and theslave child, the argument is that the choices under scrutiny veryprobably do not make things worse for, or harm, and are not otherwisebad for, any of the later-conceived progeny of the historicalinjustice. Though those future people (stripped of family and familyresources) may be burdened in various ways as a result of the choicesthat create the injustice, it is also true that alternate, seeminglypermissible choices very probably would have changed the timing andmanner of the relevant conceptions and thus have changed theidentities of the later-conceived people. We then seem forced toconclude that the choices under scrutiny do not make things worse for,or harm, and are not otherwise bad for, any of those future people atall. Here, the person affecting intuition does not block the findingthat a wrong has been done or that the future that includes thehistoric injustice is not worse than still others – the factthat there existed manycontemporaneous victims of thehistoric injustice can’t be undone. But if reparations (likecompensation under the law of negligence) requires that the personclaiming reparations have been harmed or made worse off – if itrequiresvictims rather than apparentbeneficiaries– then we seem forced to conclude that reparations tolater-conceived progeny are not owed. While that result may seemplausible (“[H]ow can any person have a claim to compensationfor a wrong that was a condition of her existence?” (Cohen 2009,81)), many theorists resist it. (See alsointergenerational justice.)

3. Proposed solutions to the nonidentity problem

3.1 Seemingly wrong choice is not wrong; “biting the bullet”

For some philosophers, the person affecting intuition itself is beyondquestion. Combining that intuition with the phenomenon of theprecariousness of existence and a certain ancillary, conceptual claim,they reason to a conclusion that still other philosophers find deeplytroubling: that many, perhaps all, of the future-directed,identity-affecting choices under scrutiny in the various nonidentitycases are not, after all, morally wrong and do not make things morallyworse. They “bite the bullet.”

David Heyd accepts that conclusion even in the case in which theexistence that is brought about isless than worth having.Notably, Heyd’s position is driven mainly by his conceptual,rather than his moral, intuitions. Heyd thus rejects what can becallednonexistence comparability and instead takes theposition that comparisons between the wellbeing level a given personhas in a future in which that person exists and the wellbeing levelthat same person has in a future in which that person never exists arenot cogent. However dire the case, in other words, the choice to bringthe child into existencecannot make things worse for or harm(in the comparative, worse off sense of that term that Heyd accepts)the child and thuscannot be wrong (Heyd 1992, 30–33;Heyd, 2009). But for Heyd the conclusion that the choice is not wrongdoesn’t imply that the choice is morally permissible. (Here,Heyd rejects still another conceptual claim that others might findhighly plausible: the claim that choices that are not morally wrongare morally permissible.) Rather, according to Heyd, the choice fallsinto the category of thegenethical and not thestraightforwardly ethical: it’s neither morally permissible normorally wrong.

Philosophers who accept the person affecting intuition but also acceptnonexistence comparability remain free to limit their “bite thebullet” conclusion to the case where the existence underscrutiny is at least worth having. Thus, David Boonin remains at leastneutral on nonexistence comparability (Boonin 2008, 130 and 135;Boonin 2014; see also Bayne 2010; Risberg 2023). In that respect,Boonin’s position is more limited than Heyd’s: for Boonin,application of the person affecting intuition is restricted to thosecases in which the life at issue isn’t “worse than no lifeat all” (Boonin 2014, 2, 14, 17).

In another respect as well, Boonin’s approach is more limitedthan Heyd’s. At least some of Heyd’s language suggests theposition that the “no harm done” result is in order, notjust in the case where the flawed existence is (very probably)unavoidable if the person is ever to exist at all and where,in that strong sense, the person’s coming into existence“depends on” agents making the choice under scrutiny, butalso in the case where agents simply retain “control” overwhether that person comes into existence or not (Heyd 1992, 105 and,generally, 99–106; Heyd 2009, 15–17). According to thatprinciple, even if the couple has the alternative of bringing the samechild into a better existence – even if the existence is notunavoidably flawed – if the couple also retains theoption ofnot bringing the child into existence at all, thechoice under scrutiny cannot make things worse for the child. The ideahere would be that, until we actually produce them, “we mightdecide not to make them the subject of any kind of moral statuswhatsoever” – a situation that leaves them, even if theyin point of fact will (if not unavoidably) exist, with “no moralstatus of any kind, not even a weak one,” relative to the choiceunder scrutiny (Heyd 1992, 99).

But the conclusions Boonin reaches in another respect seem to gobeyond Heyd’s. Boonin typically describes the choices that giverise to worth-having existences as simply “not morallywrong” (Boonin 2014, 5, 27). But he, unlike Heyd, doesn’texplicitly resist the conceptual idea that choices that are not wrongare therefore permissible (Boonin 2008, 146–149; Boonin2014).

Not surprisingly, the “bite the bullet” strategy hasencountered substantial resistance. See, e.g., Parfit 2017,126–129. If forced to choose between the person affectingintuition and the intuition that what the agents have done in thedepletion case or the slave child case is wrong, we may well cling tothe latter. Among other things, it seems highly implausible that ourfuture-directed conduct should get a moral pass whenever it affectsthe timing and manner of conception.

Of course, as Heyd notes, in many cases a choice that creates a lowwellbeing level for a person whose existencedepends on thatchoice will also create low wellbeing levels for people whoseexistences areindependent of that choice. If an agent buriesglass in the wood prior to conceiving a child, then even if thatchoice affects the timing and manner of conception and thus (accordingto the logic of the nonidentity analysis) cannot make things worse forthe agent’sown child, that choice might still makethings worse for aneighbor’s child. On that ground,the choice can be declared wrong (Heyd 1992, 193–203). Thechoice of wrongful life, as well, can make things worse for peopleother than the impaired child – create burdens, in other words,for those other people that the impaired child itself may then berequired to share (Roberts 2009a).

Boonin offers another suggestion for making the “bite thebullet” strategy more plausible. According to Boonin, ourintuition that the choices under scrutiny – in, e.g., thedepletion case and the slave child case – are wrong is itselfrooted in the fact that we have a hard time keeping “presentbefore our mind’s eye” what makes the nonidentity cases“atypical” and “idiosyncratic” – namely,that in the nonidentity cases low wellbeing levels do not correlate inthe usual way or the way we have come to expect to a person’shaving been made worse off (Boonin 2008, 146–149; Boonin 2014).We then confuse the atypical case with the ordinary case in which lowwellbeing levelsdo signify a person’s having been madeworse off. We then, mistakenly, see wrongdoing not just in theordinary case but in the atypical case as well. Once we appreciatethat confusion, we should find ourselves more comfortable abandoningthe view that a wrong has been done in the atypical case.

Boonin’s deflationist suggestion seems to run counter, however,to our own lived experience, given that we – post-Parfit –feel strongly that wealways have the relevant distinctionclearly in mind but continue to consider the depletion choice wrong.

3.2 Purely impersonal views; pluralistic views

Philosophers who accept the various nonidentity cases ascounterexamples against the person affecting intuition have the optionof retaining the basic consequentialist framework but shifting from aperson affecting to an approach that is either purelyimpersonal or impersonal at least in part. Having thusrejected the person affecting intuition, their challenge then becomesto explain how a choice can be wrong, or make the future in which thatchoice is made worse, without making things worse for any person whodoes or will exist in that future –without generatingstill other population problems that seem at least as challenging asthe nonidentity problem itself.

3.2.1 Purely impersonal views: The total theory; the average theory; critical level consequentialism

Traditional forms of utilitarianism, including both the classicmaximizing form introduced above as the “total theory” and what we cancall the “average theory,” are purely impersonal innature: they flatly reject the person affecting intuition and, inaddition, have the resources easily to explain how a choice made at afuture can be wrong while making things worse for no one who does orwill exist in that future. If, by waiting a few years to have a child,the 14-year-girl could have produced a child who is better off butnonidentical to the child the girl in fact produced, both the totaltheory and the average theory, other things equal, imply that what thegirl has done is wrong and makes the future worse (Singer 2011,107–119). The key for both views is the position that afuture’s value, and thus the permissibility of the choice thatbrings about that future, is to be determined on anaggregatebasis: utilities correlating to the individual wellbeing levels foreach person who does or will exist in a given future are added up todetermine the value of that future, with, in the case of the averageview, that sum itself being divided by the number of people who do orwill exist in that future. On both approaches, it is immaterialwhether we create additional (total or average) wellbeing (a) bycreating additional wellbeing for a particular existing or futureperson or (b) by bringing a nonidentical but better off person intoexistence. (Seeconsequentialism.)

The nonidentity problem, according to Parfit, seems clearly todemonstrate that the moral principle we seek – the “TheoryX” – “will not take a person-affecting form”(Parfit 1987, 378). A theory that is impersonal – at least inpart – may seem the obvious alternative. The difficulty is thatboth the total theory and the average theory give rise to their veryown serious population problems. As noted earlier, the total theoryquickly generates the repugnant conclusion (Parfit 1987, 381–90;see alsorepugnant conclusion). The average theory does a good job with the nonidentity problemand the repugnant conclusion. But the average theoryimplausibly prohibits bringing even the very happy child intoexistence if it so happens that the people who are already inexistence happen to be even happier (Parfit 1987, 420; Feldman 1995,192–93). It also generates highly implausible results whenapplied to the case Parfit called “Hell Three” (Parfit1987, 422; see also Temkin 2012, 319–320). Surely, according toParfit, a situation that makes life hellish for everyone who does orwill exist is not improved by the addition of still more people whoselives aren’t quite so awful but are still far less than worthliving. The total theory and the average theory both, in addition,face perennial objections based on considerations of justice, fairnessand equality.

An attempt to modify the total theory with the aim of avoiding atleast the most extreme forms of the repugnant conclusion comes in theform ofcritical level utilitarianism (Broome 2014). Criticallevel utilitarianism is impersonal in nature – it rejects theperson affecting intuition and, as an aggregative, impersonal view,has the resources to provide a plausible account of the nonidentityproblem. Then, to avoid certain versions of the repugnant conclusion,it further insists that the life worth living, if it happens to fallbelow a certain critical level, willreduce, rather than addto, the value of the relevant future.

Critical level utilitarianism struggles, however, in the face of whatArrhenius called the “sadistic conclusion.” Thus incertain cases the theory will instruct that the future that containslives that are uniformly worth living and yet fall below the criticallevel isworse than the future that contains lives that areuniformly worse than lives worth living (Arrhenius 2000). That resultseems highly implausible.

3.2.2 Purely impersonal pluralism; radical pluralism; substitutional consequentialism

In response to objections against the total theory and the averagetheory, theorists more recently have outlined approaches that are, inone way or another,pluralistic in nature.

Thus Larry Temkin proposes that what makes a future better or a choicepermissible is determined by reference to theall thingsconsidered good, where the all things considered good is a matterboth of how much aggregate wellbeing a future contains and the extentto which that future realizes still other ideals or values, including,for example, fairness, desert, equality, human flourishing and theprioritization of the needs of the least well-off (Temkin 1993,221–27). In contrast, a second approach retains additivity forthe purpose of determining a future’s overall value but takesthe position that what is to be added up is indicated not byindividual wellbeing levels but rather by those levels adjusted insome way to reflect values beyond how much wellbeing a person has in agiven future. See Broome 2015; 1991; Feldman 1997.

Such forms of pluralism are consistent with the position that valueitself remains purely impersonal in nature. As such, they seem to havethe resources necessary to provide a plausible response to thenonidentity problem. Moreover, by recognizing values over and abovethe value of the maximization of aggregate wellbeing, they may alsohave the resources necessary to avoid certain versions of therepugnant conclusion – to explain, for example, by reference tothe ideal of human flourishing (Temkin) or the concept of desert(Feldman) or the concept of the personal good (Broome), how it can bemorally better to produce a somewhat smaller number of extremely welloff people rather than a very large number of people whose lives areonly barely worth living.

Still other theorists have proposed more radical forms of pluralism:theories that reflect both person affecting and impersonal values.Thus Jeff McMahan considers the question whether a couple ought toproduce a third child who would be very happy if he or she were toexist but whose coming into existence would create significant burdensfor that child’s two already-existing siblings. Then, we mightsay that the goal of maximizing wellbeing in theaggregateprovides the couple with a reason to have the third child while theperson affecting goal of avoiding burdens on behalf of thetwo siblings provides the couple with a reasonnot to havethe third child. What the couple ought to do is then determined by howthese opposing reasons balance against one another. See McMahan 2006;2009; see also Temkin 1993, 321–47; Temkin 2012, 363–456 (combining an“internal aspects” approach and an “essentially comparative”approach); Buchanan et al. 2000, 226 and 249 (adopting both principle“M” and principle “N”).

Parfit himself proposed serious consideration of what he called the“wide” “dual” person affecting principle: “One of two outcomeswould be in one way better if this outcome would together benefitpeople more, and in another way better if this outcome would benefiteach person more.” Parfit 2017, 154. According to Parfit,“this Wide Principle would be only one of our beliefs.”Parfit 2017, 157. It is not, on its own, the “Theory X”Parfit urged us to seek decades ago. Parfit 1987, 378.

It is not clear, however, that any of the forms of pluralism (radicalor not) outlined above, as they now stand, make any significantheadway in solving our population problems. The fact is that we arenot in a position to determine whether those theories have theadvantages that may seem to accompany them or not since the theoriesthemselves have not yet been sufficiently detailed to show how the(sometimes opposing) values that they recognize are to be balancedagainst each other. Without those implications at hand, we cannot testthe theory; and without testing the theory, we cannot determinewhether the theory generates plausible accounts of our problem casesor not.

Perhaps in part in search of principles that are more transparent andthus more easily tested, still other theorists have suggestedprinciples that require agents to create additional wellbeing,including by way of bringing a better off person into existence inplace of a nonidentical less well off person, while limiting thatobligation to the case where agents cansubstitute in thebetter off person for the less well off person on a one-to-one basis.Holtug 2009; 2010, 160–162; see also Parfit 1987, 369–71(discussion of what he calls “Principle Q”); Peters 2004,27–39 (concept of injury by failure to substitute).

3.3 Choice wrongs, or harms, or is bad for, person without making person worse off

Still other theorists, seeking to retain the person affectingintuition without giving up the claim that the choices under scrutinyin the nonidentity cases really are wrong and really do make thingsworse and while accepting the phenomenon of the precariousness ofexistence, aim to solve the nonidentity problem by taking the positionthat the choice that is “bad for” a person is not necessarily one thatmakes things worse for that person or that harms that person, in thetraditional, comparative sense of that term.

The positions noted in this part 3.3 are reminiscent of an idea thatKavka explores. Thus Kavka considers the position that, even if thechoices under scrutiny do not make things worse for or harm thatchild, they may nonetheless constitute instances in which the agenthas acted (by creating a less “intrinsically desirable,”or a “restricted,” life) “wrongly toward” thechild (Kavka, 1982, 97 and 104–105) and thus whose choice itselfis wrong.

3.3.1 Wronging without harming; rights-based accounts; exploitation account

Thus, one suggestion has been that what makes the choice underscrutiny wrong is that it violates the apparent victim’sright against being brought into a flawed existence (Woodward1986; Elliot 1989; Elliot 1997; Smolkin 1999; Velleman 2008; Cohen2009). Consider, for example, the slave child case. The nonidentityanalysis has it that the couple’s choice to enter into the slavechild contract does not make things worse for the child since thatparticular child very probably would never have existed at all in theabsence of that choice. Yet we may agree that everyone has a right notto be born, or made, a slave. The couple’s entering into theslave child contract and then proceeding to have a child under thatcontract thus violates the child’s right. On that explicitlyperson affecting basis, we can then say that the couple’s choiceis wrong.

Similarly, the ticket agent who refuses to sell an airline ticket toSmith on the basis of Smith’s race violates Smith’s rightagainst racial discrimination even in the case where the planesubsequently “crashes, killing all aboard” (Woodward 1986,810–11, citing Adams). “What makes racial discriminationwrong is that it is unfair and that it stigmatizes … and achoice may have that character – and be wrong for that reason– regardless of how it affects [a person’s] otherinterests” (Woodward 1986, 811).

Woodward goes so far as to say that the choice that brings a personinto a flawed existence and thereby violates that person’s right“harms” that person under what he calls a“nonconsequentialist approach” to harm (Woodward 1986,818). Once it is established that the right has been violated,however, any effort to establish harm as well would seem to bemotivated not by moral but by legal requirements (wrongful life) orquasi-legal requirements (reparations). For those purposes, however, anon-comparative account of harm becomes especially controversial.

The rights-based account does not require that we compile a definedlist of rights. We might take the more general view that each personhas a right to be created with “due consideration for his or herhumanity”; we can understand that life itself is a“predicament” for which one’s children need to bewell-equipped (Velleman 2008, 266 and 276). A “child has a rightto be born into good enough circumstances, and being born to [e.g.] afourteen-year-old mother isn’t good enough” (Velleman2008, 275). That a child was “glad to be born”doesn’t mean that that child has waived “hisbirthright” (Velleman 2008, 277). More concretely, one mightpropose that it is wrong to bring a child into existence when many ofhis or her rights as outlined in, e.g., the United Nations Convention,would be violated (Archard 2004, 403–20). Still another approachargues that future people come into existence with certain“claim rights,” or simply “claims” (Meyer2016, 49) that can be defined in part in either sufficientarian (Meyerand Roser 2009) or egalitarian (Meyer 2016, 52–58) terms. Agentswho fail to respect those claims fail to discharge a“duty” they owe “to the person who has theclaim” (Meyer 2016, 49).

Rights-based and claims-based solutions to the nonidentity problemface certain challenges. The first, from Parfit, is that it isquestionable whether the choice an individual would not“regret” and may even applaud can violate a right orconstitute a failure to discharge a duty. If it can, it (arguably)does so only in some technical way but not in a way supports the claimthat the choice is morally wrong. When respect for the right or claimcan be seen in advance not to promote the interests of therights-holder in any way, the better view might be that the right hasbeen implicitly “waived” or the violation implicitlyconsented to (Parfit 1987, 364–65 and 373–74) on behalf ofthe future person. By analogy, the right or claim against bodilyinjury may be considered waived, or the violation consented to, whenthe already existing person arrives at the hospital unconscious and inimmediate need of open heart surgery. We don’t think thatwhatever technical rights violation there may have been in such a casesupports the inference that what the surgeon has done is morally– or indeed legally – wrong.

The second objection asks whether a rights-based or claims-basedapproach proves too much. Consider the case where a couple is enslavedand has no means of escaping that status and where any child that thecouple will produce is sure to be born a slave. If producing a childin Kavka’s original slave child case described in part 2 above iswrong in virtue of the fact that it violates the child’s rightor disregards a child’s claim, it seems that producing a child in therevised slave child case must be wrong as well. But that negativeevaluation in the revised case seems implausible. While, in somesense, a right may have been violated or a claim disregarded, itremains unclear that the couple’s choice is morally wrong.

An alternate account that has features in common with the rights-basedapproach but that, by focusing more closely on the question of whetherfuture people have been “unfair[ly]” treated as compared againstexisting people and thus “wrongfully exploited”(emphasis added), goes some distance in avoiding both objections.Liberto 2014, 76–81. See also Kavka 1981, 106–109. Theidea is that, in cases where the exploitation is established as itselfwrongful, the first objection, based on waiver or consent, fails. Thesame idea provides a defense against the second objection. When theenslaved couple produce a child who is sure to be born a slave, thereseems to be no basis on which to claim that thecouple haswrongfully exploited the child. (Still other agents, in addition tothe broadercommunity, of course, would remain morallyliable.) A question for the exploitation view is whether it can beextended to address instances of the nonidentity problem in which itseems clear that a wrong has been done but future people aren’ttreated unfairly as compared against existing people. Thus forms ofdepletion, e.g, may still be considered wrong and to make things worseeven if existing people gain nothing and even if – due totechnological developments – future people are madesubstantially better off than existing people.

A third objection seems invulnerable to the reply based onexploitation. Narrowly focusing on the rights-based approach, thatobjection simply notes that, if thechild has the right notto be brought into an existence of a certain sort, it is plausiblethat the child’sparents have various rights as well.Thus, the couple who opts to produce a child with Huntington’sdisease or hereditary deafness in place of a relatively unimpairedchild may simply be exercising their rights of procreative liberty.They may, that is, be using their gametes and their labor, as a matterof right, in a way that suits them. We quickly see that thechild’s rights and the couple’s rights cannot both berespected – a fact that raises the concern that accounts of thenonidentity cases based on rights may themselves be rooted in a theorythat is inconsistent (Persson 2009).

3.3.2 Harming without making person worse off; non-comparative harm; list account of harm; threshold account of harm

Many articulations of the nonidentity problem take for granted that acorrect account of harm understands harm ascomparative innature. On such an account, a choiceharms a person only ifthat choice makes that personworse off than he or she wouldhave been under still other choices; a personp is not harmedat a given futuref1 if there exists no alternate f2f2such thatp has more wellbeing atf2 thanp hasatf1; a choice thatmaximizes wellbeing for a personat a futuref1 cannotharm that person atf1.At least: it cannot do so in any “morally relevant sense”(Parfit 1987, 374).

Some philosophers have suggested that we also have – or canconstruct – a second morally relevant concept of harm, one thatisnon-comparative in nature. Though narrower in some waysthan the comparative concept, in other ways the non-comparativeconcept is broader. Specifically, it can count bringing a person intoan existence that is in some way burdened asharming thatperson even in the case where that person’s wellbeing has beenmaximized. Having established that the choice under scrutiny harms theperson it brings into existence and burdens in some way, we can thenprovide a person affecting account of why that choice is wrong and howit makes things worse.

Shiffrin uses examples to explain her concept of non-comparative harmand give credence to the claim that non-comparative harm has moralrelevance. Thus she argues that, if you are hit on the head by a goldbar dropped from the sky as a gift to you, you have beenharmed even if you have been more than compensated for thatharm in virtue of the fact that you are now own a gold bar (Shiffrin1999, 120–135).

In the wrongful life case, a child is born with a serious genetic orchromosomal impairment, an impairment that is unavoidable if the childis to exist at all. On a comparative account of harm, bringing thatchild into existence does not harm that child (provided the existenceis not less than worth having). Elizabeth Harman points out, however,that it’s indisputable that bringing that child into existencedoes cause the child to experience “pain, mental orphysical discomfort, disease, deformity, disability, or death”;more generally, it produces a state that is bad for the child to be in– a “bad state.” (Harman 2009, 139). OnHarman’s view, that a choice imposes any of the listedconditions on the child is sufficient to establish that that choiceharms the child whether or not the child has been made worseoff (Harman 2004, 92–93 and 107; Harman 2009, 139). It would behard to argue, moreover, that harm in that form has no moralrelevance. Thus, even those who think that wrongful life doesn’tinvolve harm – harm, that is, in the comparative, worse offsense of that term – agree that any pain and sufferingexperienced by the child must be taken into account in order todetermine that the child’s existence is not less than worthhaving. (In the rare case in which the child’s existenceis less than worth having – in which, that is, thechild would have been better off never existing at all – wewould say that the child sustains both comparativeandnon-comparative harm.)

A related proposal that avoids the difficulties of tetheringnon-comparative harm to any particular list of burdensome conditionsmakes use of the concept of athreshold-dependent concept ofharm (Rivera-Lopez 2009, 342; see also Meyer, 2004). On this view,bringing a child into existence when that child’s wellbeing levelfalls below “some normal threshold of quality of life”counts asharming that child even if that existence is itselfworth having (Rivera-Lopez 2009, 337).

Objections to proposals that rely on non-comparative concepts of harmto solve the nonidentity problem focus on whether that concept itselfcan be clearly worked out. Thus, it may be hard to see how one whosewellbeing has beenmaximized can at the same time have beenharmed in any sense of that term that remains clearly withinour grasp. Moreover, as Parfit suggests, even if we think it correcton occasion to say that a person who has not been made worse off hasbeen “harmed,” we may still be unclear whether that personhas been harmed in a “morally relevant sense” (Parfit1987, 374). Finally, there is the question of limits: what burdensomeconditions are to appear on the non-comparative harm list? What is thethreshold level below which existence counts as harm? What is“normal”? And are we really willing to eliminate or atleast to mix up the distinction we think we now can so cleanly drawbetween, e.g., the case in which opening up one’s chest toexpose the heart serves no purpose at all and the case in whichopening up one’s chest to expose the heart is an essential partof an open-heart procedure that is itself both necessary andsufficient to restore life and health? For critical discussion of thenon-comparative account of harm, see Gardner 2015.

A strategy that avoids disputes surrounding the concept of harm wouldbe to argue that, even if the choice does notharm the child,the fact that it causes the child to suffer may mean that itnonethelesswrongs the child. Bonnie Steinbock thusrecognizes how hard it is to establish that the child brought into theunavoidably flawed existence has been genuinely harmed (Steinbock2009, 157–158). But she argues that we still have room to insistthat such a child has been treated unfairly or has been wronged incases in which the child’s existence fails to meet or exceed acertain “decent minimum,” which is itself achieved“only if life holds a reasonable promise of containing thethings that make human lives good,” such as “an ability toexperience pleasure, to learn, to have relationships withothers” (Steinbock 2009, 163–165). Another approach wouldbe to draw the line at average wellbeing (for discussion, see Rachels1998; Tooley 1998). Still another approach would be to focus onwhether the child will face unusual or severe hardships (Benatar 2000;Kamm 2002).

An underlying assumption of each of the approaches sketched in thispart 3.3.2 is that the questionable choice is properly considered tocause the serious hardship that initially gives rise to ourconcern. Arguably, in some nonidentity cases the choice does stand inthe appropriate causal relation to the relevant hardship, and in othernonidentity cases it does not (Hanser 2009). While Hanser puts thepoint in terms of harm, it could be extended to the notion of wrongingwithout harming that Steinbock suggests. See also Gardner 2015,2017.

3.4 Nonidentity problem rooted in conceptual or metaphysical mistakes regarding personal identity

3.4.1 Reexamining assumptions regarding the nature of personal identity and what it is to make “a person” worse off

The several proposals fall under this heading seek to retain theperson affecting intuition and the claim that the choices underscrutiny really are wrong and really do make things worse. But theyquestion some of the ancillary conceptual or metaphysical claimsgrounding the claim that the coming into existence of any given personis highly precarious. Proposal (a) below accepts that the futurepersonp is nonidentical to the personq conceivedfrom distinct gametes and then exploits the fact thatp andq may fall under the same description; proposal (b) arguesp andq may be identical despite a distinction ingametes (despite, that is, a distinction in the timing and manner ofconception); proposal (c) argues, on metaphysical grounds, that thenonidentity problem rests on a certain equivocation in the use of theterm “person”; proposal (d) exploits certain predictablemistakes many people will make in thinking about the identity betweenp andq; and proposal (e) explores a possibleequivocation on what it is to make a person “worse off.”

(a) Though Parfit briefly discusses and sets aside the descriptiveproposal (Parfit 1987, 359–60), interest in that proposal hasbeen renewed in recent years (Hare 2007, 512–23, advocating aconcept of “de dicto” harm; Reiman 2007, 78–90,describing a “veil of ignorance” with respect to theidentity of the person harmed; Wolf 2009; Räsänen 2023). Theproposal makes use of the fact that the same definite description canpick out distinct children in distinct scenarios. Consider the slavechild case. Letp be the child who in fact exists and suffersas a result of the couple’s entering into the slave childcontract. And letq be any one of the children who might haveexisted had the couple not entered into the slave child contract yetstill produced a child. The nonidentity problem argues (among otherthings) thatp andq are, at least very probably,numerically distinct, that is, nonidentical. At the same time, on oneconstruction,p andq, even if nonidentical, fallunder a common description: they are equally, for example, “thechild produced by the couple.” Moreover, since one choice makesp worse off than the other choice makesq, the onechoice, on a certain construction, makes “the child produced bythe couple” worse off than the other choice makes “thechild produced by the couple.” And on that basis we can then saythat the one choice makes things (“de dicto”) worse for“the child produced by the couple” than the other choicedoes.

(b) A second proposal challenges the metaphysical claims aboutcross-world identity that are inherent in the nonidentity problem. Onthis proposal, the childp produced as a slave when thecouple chooses to enter into the slave child contract and the childq produced as a non-slave when the couple refrains fromentering into the contract but still produces a child should notnecessarily be considered distinct.p andq may, inother words, bethe very same child despite the fact that(due to variations in the timing and manner of their conceptions) theyhave ended up with distinct genomes.

(c) Still another proposal rejects theories of personal identity thatreflect a rigid form of essentialism and instead posits that thereexist “many entities in the vicinity of” the child whoeventually comes into existence in the various nonidentity cases andis the subject of our person affecting concern. If we accept both theperson affecting idea that a choice is wrong only if it is bad for aparticular person – a person who does or will exist – andthe idea that a choice is bad for a person only if that person“would have existed” had an alternate choice been made, weare, without realizing it, focusing on distinct entities, with oneclaim holding of one entity and the other claim holding of another.Dasgupta 2018, 541–542; 550–554. For a proposal thatsimilarly rejects a strict form of the origins (genetic) account ofpersonal identity in favor of a Lewisian“counterpart-theoretic” approach for the purpose ofsolving the nonidentity problem and specifically aims to retain thecogency of certain harm claims in that context, see Wrigley 2012. Anaccount that also appeals to counterpart theory with the aim ofresolving not just the nonidentity problem but the full range ofpopulation problems can be found in Meacham 2012. For a more generaldiscussion, see Cooper 2015.

(d) A form of rule consequentialism, rather than struggling to avoidmistaken assumptions, simply exploits two naïve mistakes thatmany people may well make when they first begin to think about theirobligations in respect of future people. The first is the mistake ofthinking that the population produced when, for example, agents chooseconservation will be identical to the population produced when agentsinstead choose depletion. The second is the person affecting intuitionitself – the idea that a choice made at a future that makesthings worse for no person who does or will exist in that futurecannot be wrong. Tim Mulgan suggests that the “ideal code”– the code whose internalization by most of a given society willmaximize wellbeing on an aggregate basis – will incorporate andin effect deem correct both those mistakes. They are both, after all,easily and efficiently “taught” since (Mulgan thinks) weare prone to make them in any case and they in effect cancel eachother out. Because the ideal code is violated by the depletion choiceand the 14-year-old girl’s choice to have a child, both aredeclared wrong (Mulgan 2006, 155–56 and 204; Mulgan 2009).

(e) Cases involving causal preemption seem to show that the simplecounterfactual (“but for”) account of what it is to makethings worse for a person – a choice is worse for a person onlyif it makes that person “worse off than [that person] wouldotherwise be” – fails (Bontly 2016, 1236). For thepremises of the nonidentity problem themselves, then, to remainplausible, we must understand what it is for a choice to be“worse for” a person in another way. Bontly suggests thefollowing principle: a choice is worse for someone if and only if it“affects that person for the worse,” where it isunderstood that a choice that has an “effect” on a personthat is itself adverse in some way is sufficient to show that thechoice “affects that person for the worse” (Bontly 2016,1237–38). We then suspect an equivocation when we realize thatwe can’t both hold that understanding of what it is for a choiceto be “worse for” a person constant and at the same timeensure that still other premises of the nonidentity argument remainplausible. Specifically: while the claim that a choice is not worsefor a personif (sufficient condition) that person has a lifeworth living and “would never exist had [that] choice not beenmade” seems plausible, the claim that that choice does not“affect [that person] for the worse” given those sameconditions “seems rather more dubious” (Bontly 2016,1238–1241).

3.4.2 Is there an explanatory gap?

An ongoing controversy raised by the proposals described in part 3.4.1above is whether they successfully explain the wrongness of the wrongchoice in terms of the adverse consequences that choice hasforthe person whose plight has drawn our concern to begin with.

Parfit’s own concerns about proposal (a) – to take oneexample – are based on the apparent explanatory gap between theclaim that the couple’s choice makes things worse for “thechild produced by the couple” and the result we seek to explain– that the choice to enter into the contract is wrong. Thequestion is whether there is anything we can grasp in what the couplehas doneto the child in fact produced that explainswhy the couple’s choice is wrong. No “familiarmoral principle” takes us from the shorthand claim to theassessment we are aiming to explain (Parfit 1987, 359; Wasserman 2008,529–35; for discussion, see Weinberg 2008.)

More generally, do the proposals in (a) – (e) either (i) trace thewrongness of the wrong choice to something other than what is bad forthat particular person or (ii) ask us, without adequate justification,to abandon the handful of clear intuitions that we do seem to haveregarding personal identity? (i) Do they retain the person affectingintuition in a sufficiently robust form? Or are theypersonaffecting in name only and in fact root their explanation ofwrongness not in what is badfor the person but rather inwhat is badfor the world? (ii) Do they ask us toconcede thatp andq are one and the sameperson in cases where we really can’t bring ourselves tobelieve thatp andq are one and the sameperson? Could my father’s first-born child have beenmehad he never met my mother at all and instead produced a verydifferent child with a very different genome born at a very differenttime? We should agree we can figure out a theory that generates such aclaim of identity. What is unclear is the justification that compelsus to accept that theory.

If weboth confine our scrutiny to what has been done to theparticular childand insist on retaining some of our clearestintuitions regarding personal identity, a second concern arises. AsDavid Wasserman notes, if the child’s life is itself, thoughflawed, worth living and if that flaw is unavoidable if thatparticular child is ever to exist at all, it may seem that anydifficulty the child then faces is a “perfectly acceptable priceto pay for a life he [or she] could not have without it”(Wasserman 2006, 145). We again are left with the kind of explanatorygap that Parfit noted in connection with the descriptive view.Confining our scrutiny just to what has been done to the child andproperly taking into account the good that has been done for the childas well as the bad, we cannot discern why what has been done is, byany stretch, bad for the child and hence cannot discern any personaffecting ground for the claim that what has been done is wrong.

3.5 Nonidentity problem rooted in a mistake in reasoning about relevant probabilities

The nonidentity cases that seem most seriously to challenge the personaffecting intuition rely on claims about probabilities. Thus in theslave child case, the depletion case, cases involving historicalinjustices and many others, it is claimed only that “very probably”particular people would not have existed had agents chosendifferently. Closer scrutiny of the probabilities at stake in thosecases suggests a solution to the problem that allows us to retain theperson affecting intuition while recognizing the phenomenon of theprecariousness of existence and accepting the conceptual andmetaphysical assumptions other proposals seek to challenge. What thatcloser scrutiny shows is that the conclusion that the future personwho is the subject of our concern has not been made worse off, orharmed (in the traditional, comparative sense of that term), by thechoice under scrutiny is not one that we can validly reach. Once werecognize the fallacy in the argument, we can identify a basis for anaccount that recognizes that the choice under scrutiny, after all,makes things worse for, and harms, that person (Roberts 2024, 196–204;see also Roberts 1998; 2003; 2006; 2009; Roberts and Wasserman2017).

The probability-based proposal is based on the following formulationof the person affecting intuition: a choice made at a given futuref1 is morally wrong, and makesf1 worse than asecond futuref2, only if there exists an alternatepractically accessible futuref3 such thatf1 isworse for at least one person who does or will exist inf1thanf3, wheref3 may, but need not, be identical tof2. To this basic formulation is then added a furthernecessary condition, aprobability-based condition, on when achoice is wrong (Roberts 2024, chap. 6).

That formulation of the intuition, like the initial formulations setforth in the preamble and part 1, is cautious. It provides only anecessary condition for when one future is worse than another and whena choice at that one future is wrong; and the necessary condition itprovides is a condition only on when one future is worse (not better)than another. Also like those initial formulations, it is consistentwith the position that all people, including the merely possible, havethe same moral status. In contrast to those initial formulations,however, the formulation is explicit in nailing down a critical point:in determining whether a person has been made worse off in a givenfuture, the formulation requires an inquiry not just into how well offthe future person “would have been,” but also into howwell off that person practically (not logically) “could havebeen,” had an alternate choice been made. (In addition to givingsupport to the probability-based proposal, specifically to the firstpoint noted below, such an expansive test also helps to avoid Broome’sargument against the neutrality intuition (see part 1 above) as wellas issues that arise from simple, counterfactual “but for”accounts of when a person is made worse off or has been harmed. (The“but for” account of harm has been – it seems correctly –widely rejected, including in the 1930s by theFirst Restatementof Torts. See Bontly 2016; Carlson 2018; Roberts 2024, 4–5n.8, 19, 57–58, 196–97 n.20, 223; see also Roberts 2007; 2009;2011. For a defense of versions of the counterfactual approach, seeKlocksiem 2012; Bavli 2018.)

The probability-based proposal then makes two points. First, where itis stipulated as part of the case that the agents’ original, wrongchoice has, at a particular future, triggered a chain of events thatresulted in a particular child’s – say, Harry’s– coming into an existence at that future that is burdened by aparticular condition, on close inspection we can see that agents inthat same case also had an alternate choice at an alternate futurethat would have triggered an alternate chain of events that would haveresulted in thatsame child Harry’s coming intoexistence at that alternate future but now unburdened by thatcondition. Agents, in other words, prior to choice hadsomeway, if not a way they themselves were in a position to identify, tomake things better for Harry. (Thus nothing, for example, in the lawsof physics bars the agents in the slave child case fromdeclining to enter into the slave child contract and stilladhering to the same timing and manner for conception that they infact adhered to. Had they done so, Harry would have existed –but as a non-slave.)

That point leaves unanswered a critical question: even if Harrypracticallycould have existed under an alternate,permissible choice, howlikely is it that he would haveexisted under such a choice?

It is in fashioning an answer to that critical question that thesecond point becomes relevant. Many theorists accept that the rolethat probabilities play in evaluating choice can be best explained byreference to a standard form ofexpected value theory. (Itshould be recognized that standard expected value theory has becomeincreasingly controversial. For purposes here, however, whetherreference is made to standard expected value theory or instead to animproved successor view is immaterial.) So that agents can evaluatetheir available alternativesprior to choice, the expectedvalue calculation must itself be completed as of the moment just priorto choice based on information available to the agents at that moment.We agree that under any alternate, permissible choice (e.g., thecouple’s choice not to enter the slave child contract and yet stillproduce a child)Harry’s chances of existence would have beenvery, very small. What has been missed is that, under the agents’original, wrong choice,Harry’s chances of existence,calculated as of the moment just prior to choice and limited just toinformation available to the agents at that moment, were also very,very small. (How could agents have guessed, prior to choice,precisely when and how would conceive a child? To the extentthey had any rough idea at all when and how they might accomplish thatfeat, they could have allowed themselves to be guided by that roughidea under an alternate, permissible choice as well.)

The probabilities under the original, wrong choice and under variousalternate, permissible choices are, in other words, a“wash.” But the wellbeing that would have been created forHarry under any alternate, permissible choice made within the contextof any future that includes Harry and that (against all odds) happensto unfold is clearly greater than the wellbeing created for Harryunder the original, wrong choice within the context of the future that(against all odds) happened to unfold. The upshot under standardexpected value theory is that the expected value for Harry undervarious alternate, permissible choices is greater than the expectedvalue for Harry under the original, wrong choice. Thus the door isopen to an account that both accepts the person affecting intuitionbut can at the same time explain just why the couple’s choice to enterinto the slave child contract and produce a child under the terms ofthat contract makes things morally worse and is morally wrong: thewrong choice made things worse for a child, Harry, who does or willexist at the future in which he exists under that wrong choice. SeeRoberts 2022a, 2022b; and 2024, 196–204. For criticism of the proposaloutlined here, see Greene 2016; Smilansky 2017; Harney 2019.

One might object that in fact there isno probability Harrywould exist in the absence of the couple’s choosing to enterinto the contract and produce a child since, “but for”their entering into the contract, the probability that the couplewould together have produced any child at all is zero. That may be anaccurate prediction of what the couplewould have done, butit doesn’t bear on whether Harry has been made worse off orharmed by what theyin fact have done. (As noted above, thesimple, counterfactual, “but for,” account of when aperson has been made worse off, or harmed, has been widelyrejected.)

As to those nonidentity cases that seem not to rely on assessments ofthe relevant probabilities – including cases in which theexistence is worth having butunavoidably flawed due, e.g.,to a genetic or chromosomal disorder – consistent with theprobability-based proposal we may take the position that those casesare less worrisome in the sense that it may seem less clear to us thatthe choices under scrutiny are wrong or make things morally worse(Roberts and Wasserman, 2017).

3.6 Choice is wrong in virtue of agent’s reasons, attitudes and intentions

3.6.1 Choice violates principle of mutual respect

A contractualist approach to the nonidentity problem will focus on thelegitimate expectations future people have in respect of the agentswhose choices cause those future people both to exist and suffer. Suchexpectations can be violated – and future people wronged –even in cases where those people have not been made worse off, orharmed, by what the agent has done. What is important is not theoutcome for the person but rather the “culpable failure”on the part of the agent, including the failure to respect a futureperson’s “value as capable of rationalself-governance” by way of failing to take“risk-managing” measures (including, e.g, pre-conceptiongenetic testing) on behalf of that future person (Kumar 2003,104–114; Kumar 2018). An issue that immediately arises iswhether a measure that makes it the case that the person who willsuffer can never exist at all is genuinely risk-managinginrespect of that person. Drawing from Scanlon 1998, Kumardescribes an alternative position that focuses, not on particularpeople, but rather ontypes of persons, orstandpoints, and, specifically, “the reasons thatpersons in certain circumstances … typically have for caringabout or wanting certain things.” Principles reflected in suchreasons – principles “no one can reasonably reject”– include preferences against the risk of living a life that isseriously compromised. Failing to take such reasons into account(depending on other facts) is counted as awrong against(though not aharm to) any particular person who eventuallyexists and, as a result of the choice under scrutiny, suffers. Kumar2018. Left unclear is how to reconcile thestandpoint againstthepersonal risk. A future standpoint may include the“generic” preference not to suffer the side effects of aparticular infertility treatment. But if that treatment is necessaryfor the person ever to exist at all, it is arguable whether that thetreatment wrongs that person or not.

Finneron-Burns offers an alternative account of how Scanlon’scontractualism can be applied to solve the nonidentity problem. Incontrast to Kumar’s account, Finneron-Burns’ account avoids phrasingthe debate in terms of types or standpoints that need not be tied toany particular individual. It instead asserts that we can think aboutfuture people – people who will in fact one day exist – inthe same terms we think about existing people while also taking theposition thatmerely “possible people are not includedin the scope of those to whom we owe justification”(Finneron-Burns 2016, 1163; 2024). The upshot is that Finneron-Burns’account would seem to face the same hurdles as will any other viewthat claims that people who do or will in fact exist have a specialmoral status that the merely possible lack. See part 1 above; part 4below.

3.6.2 Agent fails to have appropriate concern for future person’s plight

A second approach that shifts the analysis of wrongdoing away fromwhat has been done to the future person and toward facts about theagent focuses on the agent’s reasons for making one choicerather than another. On this approach, attitudes themselves can be“morally defective” (Kahane 2008, 203; see also Noggle2019 (wrongdoing based on defective attitude toward morality; Bramble2021 (wrongdoing based on morally dubious character trait)). Whetherthe agent’s choice wrongs the future person will depend onwhether the agent is motivated by an appropriate level of concern forthe needs and interests of (among others) the future person. Is theagent appropriately sensitive both to the degree of suffering thatperson can be expected to endure and to the various aspects of thatperson’s life that can be expected to render that life (onbalance) worth living (Wasserman 2006, 146)? The parent may have aprincipled objection against pre-implantation genetic diagnosis butalso have an appropriate level of concern for the child’splight. In that case, the parent’s producing the impaired childin place of the better-off but nonidentical child does not wrong theimpaired child. On the other hand, if a parent never considers thechallenges that the child might face as a result of being bornimpaired, then what the parent has done is wrong. Thus, on this view,the choice’s permissibility is a function not of the expectedgood for the child in fact being counterbalanced by the expected badbut rather of the parent’s careful determination of how the onebalances against the other (Wasserman 2006, 146–151). Animplication of this view is that there need be nothing wrong inchoosing to have a less happy rather than a happier child.

Choices by policymakers (in, e.g., the context of the depletionexample) can be similarly evaluated. The parent’s and thepolicymaker’s roles may differ in respect of the future peopletheir choices will cause to exist but agents in both roles canplausibly be held to a role-appropriate standard of concern (Wasserman2009).

4. Related challenge to the person affecting intuition: the asymmetry

Part of the reason the person affecting intuition has been soattractive is the support that it provides to one half of a pair ofhighly intuitive claims that McMahan callsthe asymmetry(McMahan 1981; 2009). According to the asymmetry, it is wrong, andmakes a future morally worse, to bring amiserable child– a child whose life isless than worth living –into existence but it is perfectly permissible, and does not make afuture worse, to leave thehappy child out of existence. Thatlatter of those two claims is simply an implication of the personaffecting intuition.

Narveson was an early advocate of both the person affecting intuitionand the asymmetry. The asymmetry, however, has recently put forwardnot as an adjunct to the person affecting intuition but as stillanother way to undermine the intuition (Singer 2011, 88–89);MacAskill 2022, 172; see also McMahan 1981, 2009). The objection isjust this: the asymmetry is itself internally inconsistent. Since itis generally agreed that the choice to bring the miserable child intoexistence is, other things equal, surely wrong and makes the futuremorally worse, any internal inconsistency within the asymmetry must belaid at the feet of the person affecting intuition itself.

Consistency concerns regarding the asymmetry arise when attempts aremade to explain how it can be that the existence of the miserablechild makes thingsworse but the existence of the happy childdoesnot make thingsbetter. Other objections arisewhen attempts are made to avoid inconsistency by situating theasymmetry within the context ofmoral actualism or indeed anytheory that assigns a special moral status to actual people (or topeople who will exist if agents make the choice under scrutiny; or topeople who will exist independently of whether agents make thatchoice) that it does not also assign to merely possible people (or tonever existing people; or to people whose existence depends on whetherthat choice is made). See McMahan 1981; 2009; Persson 2009; Singer2011, 114; Parfit 2011 vol. 2, 224–225; see also cases involvingcycling, e.g., Parfit’s Tom, Dick and Harry case; Parfit 2017,140–146; see also Hare 2007, 503–507.

Claims of an internal inconsistency, however, have been challenged. Wecan recognize the importance of avoiding the claim that anyperson’smoral status is a function of thatperson’sexistential status, concede that leaving thehappy child out of existence makes that child worse off than thatchild might have been and still insist that a child’s being madeworse off by way of being left out of existence altogether lacks anymoral significance even though the child’s being made worse offby way of being brought into a miserable existence has full moralsignificance. The distinction that is at work in making that claimwould just be this: the happy child’s loss of wellbeing obtains withina future in which that child never exists at all, while the miserablechild’s loss of wellbeing obtains within a future in which that childdoes or will exist. See Roberts 2010; 2011a; 2011b; 2024,53–65.

5.Conclusion

Perhaps the most practically important contribution thenonidentity problem has made to our moral theorizing (whether asphilosophers, economists, lawyers or otherwise) is that it focuses ourclose attention on the precise nature of the obligations we have inrespect of possible future people, including future generations. Ifthe person affecting intuition means that – given theprecariousness of existence – our obligations are satisfied withimplausible ease, then the nonidentity problem provides us with acompelling objectionagainst that intuition. If rejectingthat intuition means that we are obligated, not just tomakepeople happy, but also tomake happy people and to do soeven at a grave cost to still other people, then we seem to have acompelling argument infavor of that intuition. The reasonthe nonidentity problem is of such intense and continuing interest isthat neither of those options seems even remotely plausible. Thesolution perhaps will involve a closer look at the intuition itselfalongside the various ancillary claims that we must accept in ordervalidly to argue for either one of those disturbing results.

Bibliography

Adams, R.M., 1972. “Must God Create the Best?”ThePhilosophical Review, 81(3): 317–332.
Archard, D., 2004. “Wrongful Life,”Philosophy, 79(3): 403–420.
Archard, D. and D. Benatar, 2010.Procreation and Parenthood:The Ethics of Bearing and Rearing Children, Oxford: OxfordUniversity Press.
Arrhenius, G., 2000. “An Impossibility Theorem for WelfaristAxiology,”Economics and Philosophy, 16: 247–266.
–––, 2009. “Can the Person-AffectingRestriction Solve the Problems in Population Ethics?” in M.Roberts and D. Wasserman (eds.) 2009, pp. 289–314.
Bavli, H., 2018. “Counterfactual Causation,” 51Arizona State Law Journal 879.
Bayles, M. (ed.), 1976.Ethics and Population, Cambridge,MA: Schenkman.
Bayne, T. 2010. “In Defense of Genethical Parity,” inArchard and Benatar (eds.) 2010, pp. 31–56.
Benatar, D., 2006.Better Never to Have Been: The Harm ofComing into Existence, Oxford: Clarendon Press.
–––, 2000. “The Wrong of WrongfulLife,”American Philosophical Quarterly, 37:175–83.
Bontly, T., 2016. “Causes, Contrasts, and the Non-identityProblem,” Philosophical Studies, 173: 1233–51.
Boonin, D., 2008. “How to Solve the Non-IdentityProblem,”Public Affairs Quarterly, 22(2):129–159.
–––, 2014.The Non-Identity Problem and theEthics of Future People, Oxford: Oxford University Press.
Bramble, B., 2021. “The Defective Character Solution to theNon-Identity Problem,”Journal of Philosophy, 118(9):504–520.
Broome, J., 1992.Counting the Cost of Global Warming,Cambridge: The White Horse Press.
–––, 2004.Weighing Lives, Oxford:Oxford University Press.
–––, 2015, “General and Personal Good:Harsanyi’s Contribution to the Theory of Value,” in I.Hirose and J. Olson (eds.),The Oxford Handbook of ValueTheory, Oxford: Oxford University Press, pp. 249–266.
Buchanan, A., D. Brock, N. Daniels and D. Wikler, 2000.FromChance to Choice: Genetics and Justice, Cambridge: CambridgeUniversity Press.
Carlson, E., 2018. “More Problems for the CounterfactualComparative Account of Harm and Benefit,”Ethical Theory andMoral Practice, 22: 795–807.
Carter, A., 2001. “Can We Harm Future People?”Environmental Values, 10: 429–54.
Cohen, A. I., 2009. “Compensation for Historic Injustices:Completing the Boxill and Sher Argument,”Philosophy &Public Affairs, 37(1): 81–102.
Cooper, R., 2015. “How Might I Have Been?”Metaphilosophy, 46(4–5): 495–513.
Dasgupta, S., 2018. “Essentialism and the NonidentityProblem,”Philosophy and Phenomenological Research,XCVI(3): 540–570.
de-Shalit, A., 1995.Why Posterity Matters, London:Routledge.
Elliot, R., 1989. “The Rights of Future People,”Journal of Applied Philosophy, 6(2): 159–169.
–––, 1997. “Contingency, Community andIntergenerational Justice,” in N. Fotion and J. Heller (eds.),Contingent Future Persons, Dordrecht: Kluwer, pp.157–170.
Feinberg, J. 1988. “Wrongful Life and the CounterfactualElement in Harming,”Social Philosophy & Policy,4(1): 144–178.
Feldman, F., 1995. “Justice, Desert, and the RepugnantConclusion,”Utilitas, 7(2): 567–85.
Finneron-Burns, E. 2016., “Contractualism and theNon-identity Problem,”Ethical Theory and MoralPractice, 19: 1151–1163.
–––, E. 2024.,What We Owe To Future People:A Contractualist Account of Intergenerational Ethics, OxfordUniversity Press.
Gardner, M., 2015. “A Harm-Based Solution to theNon-Identity Problem,”Ergo, 2(17),available online.
–––, 2017. “On the Strength of the ReasonAgainst Harming,”Journal of Moral Philosophy, 14:73–87.
Glover, J., 2006.Choosing Children: The Ethical Dilemmas ofGenetic Intervention, Oxford: Oxford University Press.
Gosseries, A., 2001. “What Do We Owe the NextGeneration(s)?”Loyola of Los Angeles Law Review,35(1): 293–354.
–––, and L. Meyer, 2009.IntergenerationalJustice, Oxford: Oxford University Press.
Greene, M., 2015. “Roberts on Depletion: How Much Better CanWe Do for Future People?”Utilitas, 28(1):108–118.
Hanser, M., 1990. “Harming Future People,”Philosophy & Public Affairs, 19: 47–70.
–––, 2009. “Harming andProcreating,” in M. Roberts and D. Wasserman (eds.) 2009, pp.179–199.
Hare, C., 2007. “Voices from Another World: Must We Respectthe Interests of People Who Do Not, and Will Never, Exist?”Ethics, 117: 498–523.
Harman, E., 2004. “Can We Harm and Benefit inCreating?”Philosophical Perspectives, 18:89–113.
–––, 2009. “Harming as CausingHarm,” in M. Roberts and D. Wasserman (eds.) 2009, pp.137–154.
Harney, J. 2019. “Comparative Personal Views and theNon-Identity Problem”,Intergenerational JusticeReview, 5(2): 52–53.
Herissone-Kelly, P., 2009. “Two Varieties of‘Better-For’ Judgments,” in M. Roberts and D.Wasserman (eds.) 2009, pp. 249–263.
Herstein, O.J., 2008. “Historic Injustice and theNon-Identity Problem: The Limitations of the Subsequent-Wrong Solutionand Towards a New Solution,”Law and Philosophy, 27:505–531.
Heyd, D., 1992.Genethics: Moral Issues in the Creation ofPeople, Berkeley: University of California Press.
–––, 2009. “The Intractability of theNonidentity Problem,” in M. Roberts and D. Wasserman (eds.)2009, pp. 3–25.
Holtug, N., 2009. “Who Cares About Identity,” in M.Roberts and D. Wasserman (eds.) 2009, pp. 71–92.
–––, 2010,Persons, Interests andJustice, Oxford: Oxford University Press.
Kahane, G., 2008. “Non-identity, Self-defeat, and Attitudesto Future Children,”Philosophical Studies, 145:193–214.
Kamm, F.M., 2002. “Genes, Justice, and Obligations to FuturePeople,”Social Philosophy and Policy, 19:350–88.
Kavka, G., 1981. “The Paradox of Future Individuals,”Philosophy & Public Affairs, 11: 93–112.
Klocksiem, J., 2012. “A Defense of the CounterfactualComparative Account of Harm,”American PhilosophicalQuarterly, 49(4): 285–300.
Kumar, R., 2003. “Who Can Be Wronged?”Philosophyand Public Affairs, 31(2): 99–118.
–––, 2018. “Risking FutureGenerations,”Ethical Theory and Moral Practice, 21(2):245–257.
Levy, N., 2002. “The Apology Paradox and the Non-identityProblem,”The Philosophical Quarterly, 52(208):358–68.
Liao, M. and O’Neil (eds.), 2017.Current Controversiesin Bioethics, London: Routledge.
Liberto, H., 2013. “The Exploitation Solution to theNon-Identity Problem,”Philosophical Studies, 167:73–88.
Lillehammer, H., 2009. “Reproduction, Partiality, and theNon-Identity Problem,” in M. Roberts and D. Wasserman (eds.),pp. 231–248.
McMahan, J., 1981. “Problems of Population Choice,”Ethics, 92(1): 96–127.
–––, 2006. “Paradoxes of Abortion andPrenatal Injury,”Ethics, 116: 625–55.
–––, 2009. “Asymmetries in the Morality ofCausing People to Exist,” in M. Roberts and D. Wasserman (eds.)2009, pp. 49–68.
Meacham, C., 2012. “Person-Affecting View and SaturatingCounterpart Relations,”Philosophical Studies, 158:257–287.
Meyer, K., 2018. “The Claims of Future Persons,”Erkenntnis, 83: 43–59.
Meyer, L., 2004. “Historical Injustice and the Right ofReturn,”Theoretical Inquiries in Law, 5(2):305–316.
Meyer, L and D. Roser, 2009. “Enough for the Future,”in A. Gosseries and L. Meyer (eds.), pp. 219–248.
Mulgan, T., 2006.Future People: A Moderate ConsequentialistAccount of Our Obligations to Future Generations, Oxford: OxfordUniversity Press.
Narveson, J., 1976. “Moral Problems of Population,” inBayles (ed.), pp. 59–80.
Noggle, R. 2019. “Impossible Obligations and theNon-Identity Problem,”Philosophical Studies, 176:2371–2390.
Parfit, D., 1976. “On Doing the Best for OurChildren,” in Bayles (ed.), pp. 100–115.
–––, 1982. “Future Generations: FurtherProblems,”Philosophy and Public Affairs, 11:113–72.
–––, 1986. “Comments,”Ethics, 96: 854–62.
–––, 1987 (orig. 1984).Reasons andPersons, Oxford: Clarendon Press.
–––, 2011.On What Matters (volumes 1and 2), Oxford: Clarendon Press.
–––, 2017. “Future People, theNon-Identity Problem, and Person-Affecting Principles,”Philosophy and Public Affairs, 45(2): 118–157.
Parsons, J., 2003. “Why the Handicapped Child Case IsHard,”Philosophical Studies, 112: 147–162.
Persson, I., 2009. “Rights and the Asymmetry BetweenCreating Good and Bad Lives,” in M. Roberts and D. Wasserman(eds.) 2009, pp. 29–47.
Peters, P., 2004.How Safe Is Safe Enough? Obligations to theChildren of Reproductive Technology, New York: Oxford UniversityPress.
–––, 2009. “Implications of theNonidentity Problem for State Regulation of ReproductiveLiberty,” in M. Roberts and D. Wasserman (eds.) 2009, pp.317–331.
Rachels, S., 1998. “Is It Good to Make Happy People,”Bioethics, 12(2): 93–110.
Räsänen, J., 2023. “Defending the De DictoApproach to the Non-Identity Problem,”Monash BioethicsReview, 41: 124–135.
Reiman, J., 2007. “Being Fair to Future People: TheNon-Identity Problem in the Original Position,”Philosophy& Public Affairs, 35(1): 71–92.
Risberg, O. 2023. “The Morality of Creating Lives Not WorthLiving: On Boonin’s Solution to the Nonidentity Problem,”Utilitas, 35: 88–97.
Rivera-Lopez, E., 2009. “Individual ProcreativeResponsibility and the Non-Identity Problem,”PacificPhilosophical Quarterly, 90: 99–118.
Roberts, M., 1998.Child Versus Childmaker: Future Persons andPresent Duties in Ethics and the Law, Lanham, MD: Rowman &Littlefield.
–––, 2002. “A New Way of Doing the BestThat We Can: Person-Based Consequentialism and the EqualityProblem,”Ethics, 112(2): 315–50.
–––, 2007. “The Nonidentity Fallacy: Harm,Probability and Another Look at Parfit’s DepletionExample,”Utilitas, 9: 267–311.
–––, 2008. “What is the Wrong of WrongfulDisability? From Chance to Choice to Harm to Persons,”Lawand Philosophy, 28(1): 1–57.
–––, 2009. “The Nonidentity Problem andthe Two Envelope Problem: When Is One Act Better for a Person thanAnother?” in M. Roberts and D. Wasserman (eds.) 2009, pp.201–228.
–––, 2010.Abortion and the MoralSignificance of Merely Possible People: Finding Middle Ground in HardCases, Dordrecht: Springer.
–––, 2011a. “The Asymmetry: ASolution,”Theoria, 77: 333–367.
–––, 2011b. “An Asymmetry in the Ethics ofProcreation,”Philosophy Compass, 6(11): 765–776.
–––, 2022a. “Nonidentity, Better Chanceand the Value of Existence: A Defense of Person BasedConsequentialism,” inThe Oxford Handbook of PopulationEthics, G. Arrhenius, K. Bykvist, T. Campbell and E.Finneran-Burns (eds.), Oxford University Press, pp.226–250.
–––, 2022b. “The Value and Probabilitiesof Existence,” inEthics and Existence, J. McMahan, T.Campbell, J. Goodrich and K. Ramakrishnan (eds.), Philadelphia: TempleUniversity Press, pp. 38–60.
Roberts, M., 2024.The Existence Puzzles: An Introduction to Population Ethics, New York: Oxford University Press.
Roberts, M. and D. Wasserman (eds.), 2009.Harming FuturePersons: Ethics, Genetics and the Nonidentity Problem, Dordrecht:Springer.
––– and D. Wasserman, 2017. “Dividing andConquering the Nonidentity Problem,” in M. Liao and C.O‘Neil (eds.), pp. 81–98.
Robertson, J. A., 1994.Children of Choice: Freedom and theNew Reproductive Technologies, Princeton: Princeton UniversityPress.
–––, 2004. “Procreative Liberty and Harmto Offspring in Assisted Reproduction,”American Journal ofLaw and Medicine, 30: 7–40.
Savulescu, J. 2001. “Procreative Beneficence: Why We ShouldSelect the Best Children,”Bioethics, 15(5/6):–413.
Scanlon, T., 2000.What We Owe to Each Other, revisededition, Cambridge, MA: Harvard University Press.
Schwartz, T., 1978. “Obligations to Posterity,” inObligations to Future Generations, R. Sikora and B. Barry(eds.), Philadelphia: Temple University Press, pp. 3–13.
Sher, G., 2005. “Transgenerational Compensation,”Philosophy & Public Affairs, 33: 181–200.
Shiffrin, S., 1999. “Wrongful Life, ProcreativeResponsibility, and the Significance of Harm,”LegalTheory, 5: 117–48.
Sikora, R. and B. Barry (eds.), 1978.Obligations to FutureGenerations, Philadelphia: Temple University Press.
Singer, P., 2011.Practical Ethics, 3^rd ed.Cambridge University Press.
Smilansky, S., 2013. “Morally, Should We Prefer Never toHave Existed?”Australasian Journal of Philosophy,91(4): 655–666.
–––, 2017. “The Nonidentity Problem:United and Unconquered,” inCurrent Controversies inBioethics, Matthew Liao and Collin O’Neil (eds.), London:Routledge, pp. 99–111.
Smolkin, D., 1999. “Toward a Rights-Based Solution to theNon-Identity Problem,”Journal of Social Philosophy,30: 195–96.
Steinbock, B., 2009. “Wrongful Life and ProcreativeDecisions,” in M. Roberts and D. Wasserman (eds.) 2009, pp.155–178.
–––, and R. McClamrock, 1994. “When IsBirth Unfair to the Child?”Hastings Center Report, 24:15–21.
Tännsjö, T., 2002. “Why We Ought to Accept theRepugnant Conclusion,”Utilitas, 14(3):339–359.
Temkin, L., 2012.Rethinking the Good: Moral Ideals and theNature of Practical Reasoning, Oxford: Oxford UniversityPress.
Tooley, M., 1998. “Value, Obligation and the AsymmetryQuestion,”Bioethics, 12(2): 111–124.
Velleman, J. David, 2008. “Persons in Prospect,”Philosophy & Public Affairs, 36: 221–288.
Wasserman, D., 2005. “The Nonidentity Problem, Disability,and the Role Morality of Prospective Parents,”Ethics,116: 132–52.
–––, 2008. “Hare on De Dicto Betternessand Prospective Parents,”Ethics, 118(3):529–33.
–––, 2009. “Harms to Future People andProcreative Intentions,” in M. Roberts and D. Wasserman (eds.)2009, pp. 265–285.
––– and M. Roberts (eds.), 2009.HarmingFuture Persons: Ethics, Genetics and the Nonidentity Problem,Dordrecht: Springer.
––– and M. Roberts, 2017. “Dividing andConquering the Nonidentity Problem,” inCurrentControversies in Bioethics, M. Liao and C. O’Neil (eds.),pp. 81–98.
Weinberg, R., 2008. “Identifying and Dissolving theNon-Identity Problem,”Philosophical Studies, 137(1):3–18.
Wolf, C., 2009, “Do Future Persons Presently HaveAlternative Possible Identities?” in M. Roberts and D. Wasserman(eds.) 2009, pp. 93–114.
Woodward, J., 1986. “The Non-Identity Problem,”Ethics, 96: 804–31.
Wrigley, A. 2012, “Harm to Future Persons: Non-IdentityProblems and Counterpart Solutions,”Ethical Theory andMoral Practice, 15: 175–190.
Zuber, S. et al. 2021 “What Should We Agree on About theRepugnant Conclusion?,”Utilitas, 33(4):379–383.

Academic Tools

How to cite this entry.
Preview the PDF version of this entry at theFriends of the SEP Society.
Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO).
Enhanced bibliography for this entryatPhilPapers, with links to its database.

Other Internet Resources

[Please contact the author with suggestions.]

Acknowledgments

I am grateful to Jacob Iacullo and Kevin Janas for their researchassistance in connection with the 2024 revision of this entry.

Open access to the SEP is made possible by a world-wide funding initiative.
The Encyclopedia Now Needs Your Support
Please Read How You Can Help Keep the Encyclopedia Free

Browse

About

Support SEP

Mirror Sites

View this site from another server:

USA (Main Site)Philosophy, Stanford University

Info about mirror sites

Library of Congress Catalog Data: ISSN 1095-5054

Movatterモバイル変換