Sarah Conly accomplishes the impressive feat of bringing population ethics down to earth – an earth threatened with environmental disaster in the coming century – while drawing on the more formal, abstract population ethics of recent analytic philosophy. She does so in clear, accessible writing that displays proficiency in several complex areas of practical ethics. A reader unacquainted with these areas will receive a good introduction; a philosophically informed reader will appreciate Conly's comfortable grasp of many of their nuances.
Conly argues that a global reduction in fertility to one child per couple is necessary and effective in mitigating the devastating effects of climate change. Although this claim is debatable, she offers enough evidence to make it plausible, and I will accept it in this review. I will focus instead on her claim that a one-child policy, enforced by criminal fines, would be morally justified. Conly maintains that such a policy would not be trumped by a right to have a larger family or to control one's own body, and that the likely consequences of such a policy would not be unjust or otherwise morally unacceptable.
Conly first rejects a countervailing right based on important interests (Chapter 2), then a right based on will or autonomy (Chapter 3). She contends that individuals have a right-grounding interest in having one child, but no more, and that the restriction of bodily control entailed by her one-child policy is justified by the harm of increased population. I found the first claim more persuasive: the interest in additional children, while significant, lacks the great weight of the interest in having and raising a biological child. I would, however, have liked Conly to address the stronger claim, made by some philosophers, that the interest in having even one biological child is not great enough to support a right if adoption is a feasible option. Her second rights claim – that the policy's restriction on bodily control is justified under a familiar harm principle – is less convincing, for reasons I consider below.
In Chapter 4, Conly argues that to be effective, a one-child policy must be enforced by criminal fines that carry moral disapproval as well as economic costs. Her distinction between taxes and fines, and her claims about the greater force of the latter, are both plausible, although her reliance on the criminal law, creates, as she recognizes, a heavier burden of justification. Chapter 5 defends Conly's policy against arguments that future people lack the rights that justify its imposition, or would have reason to waive those rights. Conly's argument that future people have rights is clear and cogent; below, I discuss whether she could have made a stronger case that the kind of rights they have would be violated by the failure to adopt a one-child policy. Chapter 6 addresses concerns that a one-child policy would damage the economy, exacerbate sex selection, or harm children by denying them siblings. Conly disposes of the last concern with plausible empirical evidence but offers only tentative, qualified reassurance concerning the first, which will not satisfy anxious sceptics. Her argument about the second is stronger, although I will argue below that other forms of selection warrant concern. Chapter 7 ends the book with a discussion of implementation.
There are several places in which Conly's generally cogent defence of a one-child policy could have been more thorough, albeit perhaps not in such a concise book. In offering a harm-based justification for restricting procreative and bodily liberty, she devotes surprisingly little attention to the difficulties of attributing responsibility for ‘incremental harms’ to individual procreators. Nor does she discuss the acute challenges in implementing her proposed policy in the low and moderate income countries that are, or are likely to become, major sources of the population and consumption increases she fears. And while much of what she says about the relevance of the non-identity problem is right and insightful, she could have made an even stronger case that it has little bearing on the duties of present consumers and procreators. Last, she understates some of the hardships likely to result from sanctions to enforce the policy, and from the policy's success. I have little doubt that Conly could have addressed these concerns more fully in a longer book, given the care and skill with which she has addressed such a wide range of issues in a short one.
1. INDIVIDUAL RESPONSIBILITY FOR OVERPOPULATION
Conly adduces the harm done by overpopulation to justify restrictions on the right to control one's body by having additional children. But as Travis Rieder points out in an earlier review (Reference Rieder2016), there is no, or negligible, expected harm from any single birth. This is a problem of ‘moral mathematics’ –Derek Parfit's term – where individual actions, themselves harmless or trivially harmful, contribute without coordination to massive cumulative harms. Conly recognizes the problem, but her brief discussion (96–98) does not give it the attention it deserves. To justify the assignment of responsibility to individuals, she offers a legal analogy in which Grandma receives a dose of poison from each of her 10 grandchildren, where just nine doses would be fatal. The grandchildren agree that if any of them fail to act, they will forfeit their share of the estate. Conley notes that the law would regard all 10 as having committed murder. ‘In the case of overpopulation, the harm isn't intentional, but the causal responsibility is the same . . .’ (97). But it isn't. It is the agreement that makes all 10 responsible even if none commits an individual act necessary or sufficient for Grandma's death. In contrast, people who have additional children do not coordinate their procreative acts. Moreover, their individual acts make a negligible contribution to the harms of overpopulation, unlike the poisoners, each of whom appreciably increases the risk of harm in the face of abandonment or failure by any one of the others.
There is an ongoing debate about whether uncoordinated collective harms can be analysed in a way that permits the attribution of responsibility to contributing individuals. John Broome (Reference Broome2012: 75) argues that the carbon emissions of each citizen in developed countries reduces total human life by six months. Elizabeth Cripps (Reference Cripps2016: 125) responds that this reduction counts as a harm only if we ignore the separateness of persons, since any given victim suffers only an infinitesimal loss of life from any given agent. Christian Baatz (Reference Baatz2014) suggests that because of the difficulty in assigning responsibility, lowering individual emissions is, in the absence of a binding law or policy, best seen as an imperfect duty.
I would argue that a perfect duty of self-restraint, in emissions or procreation, must arise from a just law, rather than justifying that law. An agent who violates the law does not become responsible for any part of the collective harm, but rather for a legally sanctionable wrong – not merely the violation of a just law, but a grave injustice to her fellow citizens. The one-child law itself is justified as the most fair and efficient way to avoid a great collective harm; the sanctions it imposes are justified as punishment for a refusal to assume the fair burdens it assigns. Compare a military draft imposed in response to an unprovoked attack by a neighbouring country; assume that a conscript army would have good odds of defeating the threat while a volunteer army would not. Most of us would see these circumstances as justifying a draft, though it would restrict the bodily control of draftees at least as much as a one-child policy would restrict the bodily control of fertile individuals. Sanctions for an individual act of draft dodging would not be justified because of its harm – we can assume that the army doesn't need every draftee in order to prevail, and that a single lapse in participation won't affect the draft's success the army's morale. Rather, sanctions would be justifiable punishment for free-riding that places one's own life and comfort above those of one's fellow citizens. Under a justified one-child policy, like a justified draft, free-riders are punished not for a negligible, causally tenuous contribution to the collective harm, but for the evasion of a fair duty.
2. THE GLOBAL EQUITY OF A ONE-CHILD POLICY
Conly defends the reasonableness of one-child policies in developed countries; she says much less about the difficulties of those policies in developing ones. The non-coercive measures she proposes – contraceptive dissemination and education – might be effective in developing countries if initiated or adopted by strong central governments. But the governments in many developing countries may resist such measures, some because they see greater population as a source of power, some because they fear widespread religious or cultural resistance, some because the policies conflict with their own religious or moral beliefs. Even in countries that would accept non-coercive measures, more coercive ones – taxes and fines – may not be politically or administratively feasible. Some of the countries facing the greatest barriers to adoption or enforcement may be among the more rapidly developing ones. Understandably, their leaders may be even less willing or able to take risks to reduce climate impact by population control than by environmental regulation. They may reject a demanding self-restraint that developed countries never attempted before they achieved political and economic ascendancy. Moreover, it is harder to argue that developing countries would be justified in adopting one-child laws that imposed sanctions and gave rise to individual duties. The burdens of compliance would be especially great in societies with strong social pressure to have large families, and for parents who relied on their children's labour for subsistence. Because some of that burden would be attributable to the long-term effects of colonization and trade policy, developed countries might be morally required to assume some of the cost of making a one-child policy fair and effective in many developing countries.
3. ARE ANYONE'S RIGHTS VIOLATED BY OVERPOPULATION?
It has been argued that no harm is done to the future ‘victims’ of a shortsighted climate policy, since they owe their lives to that policy, and would – if they had lives worth living – waive any rights against the unavoidably adverse effects of that policy. Conly denies that this version of the so-called ‘non-identity’ problem gets the present generation off the hook (161–169). First, she argues, there would inevitably be some overlap between the future populations of a one-child policy and a less restrictive one. Those people, though unidentifiable, would have been harmed by overpopulation, since they are worse off than they would have been under a one-child policy. Second, she writes, there is a significant risk that more of those people will have lives that are not worth living than under a one-child policy, as a result of overpopulation. Both arguments are well taken. Conly then tries to fortify her case with Dan Brock's (Reference Brock1995) Non-Person-Affecting Principle (NPA), which holds, in part, that it is wrong to create a person expected to face significant suffering or limitation rather than one who would not – a wrong not to either person, but to the world at large. I believe that NPA does not serve Conly's purposes, and that she could make a stronger argument, based on wrongs to future individuals, that captures what is plausible in NPA.
As formulated by Brock, NPA applies only to ‘same-number choices’ – choices affecting the identity but not the number of people who will ever live (Reference Brock1995: 273). Obviously, a one-child policy would not satisfy that condition. While Conly could substitute a same-or-lesser number condition, for choices resulting in an equal or lesser number of people ever living, there is a less tractable problem in applying NPA to a one-child policy. For Brock, it applies only when agents can avoid significant harm or limitation ‘without imposing substantial burdens or costs or loss of benefits on themselves or others . . .’ (Reference Brock1995: 273). Although parents may not have a right to a second child, refraining from having one would surely count as a ‘substantial . . . loss of benefits’ for many of them.
A duty to protect the environment for future generations can, I suggest, be more firmly grounded in a duty to the people who will make up those generations. We arguably have a duty as progenitors to attempt to secure our descendants – whoever they will be – a minimally decent environment. This duty would apply even if there were no overlap between future populations under one-child and less-restrictive policies. This duty is just a generational extension of the duty we have to our own future children, to make adequate provision for them to lead good lives. The latter duty applies even if our actions to fulfil it affect the identity of our future children, as they do if we delay procreation in order to provide children with a safe and comfortable home. An extended duty captures the responsibility of present individuals more intuitively and less problematically than NPA, by identifying their descendants as the rights-bearers.
4. THE ADVERSE CONSEQUENCES OF A ONE-CHILD POLICY
Conly does an admirable job of addressing some of the ‘unwanted consequences’ of a one-child policy. However, in some cases, she overlooks or understates significant concerns.
She argues persuasively that there's little danger of a substantial rise in sex-selective abortions in most developed countries under a one-child policy, although there may well be in some developing countries. The greater danger in developed countries, however, is that prospective parents facing a one-child cap will soon have a far greater capacity to exercise the finickiness or perfectionism that such a policy may exacerbate. The willingness to have a child with prenatally diagnosable disabilities, slight at present, may become negligible under a one-child policy. And a desire for positive attributes in future children, which cannot currently be satisfied and may not be very strong, may become more feasible with the advent of new genomic testing and more powerful with a one-child limit. Ironically, this risk may be increased by a morally attractive feature of Conly's policy: she would not grant an exception merely because the first child had a serious disability (the former Chinese policy). She suggests that an exception would be justified only if the disability were severe enough to make the relationship one which ‘no normal person could find . . . rewarding’; severe enough that the child was almost ‘as good as dead’ (58). The vast majority of genetically diagnosable disabilities would not satisfy these stringent conditions. Prospective parents unwilling to have, as their only child, one with any of those disabilities would likely abort and try again. Moreover, with non-invasive testing enabling them to start over after 7–10 weeks of pregnancy, prospective parents might be tempted to move from disability-aversion to perfectionism when facing the prospect of only one child.
Conly also ignores possible harms to the additional child and its siblings likely to result even from non-criminal sanctions like taxes – harms emphasized by Elizabeth Cripps (Reference Cripps2015). Arguably, such harms are no different in kind from the strained resources and crowded domiciles always faced by large families of limited means. But that cannot be said of fines, which are intended to convey disapproval. As Clare Heyward (Reference Heyward2012) suggests, the parents’ stigma is likely to be shared by the additional child as well as its siblings. They will of course be blameless, but that will hardly protect them from stigma. Bastards, after all, were not responsible for being born out of wedlock.
Finally, Conly displays excessive optimism about the fate of minority cultures under a one-child policy. Many minority cultures in America already face the risk of extinction, not from population control but assimilation. Population control would heighten the risk for some cultures, by threatening the core value of raising large families. More broadly, some cultures can compensate for shrinking numbers due to assimilation or population control by recruiting members with different ancestry. But for some, the recruitment of outsiders is slow and difficult; for others it is unacceptable. Conly could well maintain that cultures that fail to adapt, whether to an open society or a one-child policy, are simply doomed to the bloodless extinction many other cultures have faced. But she should not downplay the threat.
In conclusion, Conly is unlikely to persuade sceptics that present individuals have a duty to bear only one child in order to avoid environmental disaster in the next century. But for those who accept the need for, and efficacy of, a one-child policy in avoiding disaster, she makes a strong case for the morality of such a policy – if it can be implemented humanely and fairly. In emphasizing the cumulative impact of personal reproductive decisions on the fate of the earth, Conly has also made a valuable contribution to population ethics, illustrating how decades of theoretical debate can be brought to bear on pressing issues of global justice.
Disclaimer: The views expressed are the author's own. They do not represent the position or policy of the National Institutes of Health, U.S. Public Health Service, or the Department of Health and Human Services.
