Do parents have a special duty to mitigate climate change?

On shared duties

Shared duties require a number of persons to organize in pursuit of some end which none can secure alone. This might involve different levels of collective action, from the weakest (a combination of individual actions with each aware of the combined effects) to becoming a formally structured group capable of intentional action over and above the actions of individual members.

The shared duties defended here are differently derived from those familiar from the cosmopolitan global justice literature (Goodin, 1985: 134–141; Shue, 1980: 35–64). ² General duties requiring the global affluent to organize to protect the vulnerable or secure basic rights are acquired in the first instance as shared: because some morally salient end could be achieved by cooperation among a number of persons at relatively low cost to each. Here, the shared duty is derived from special duties held by each duty bearer to other specific individuals. Cooperation is ultimately owed by each parent to her own child, rather than by all parents to all children.

Such duties are quickly defended. Suppose (as later sections will argue) a number of individuals (parents) have special obligations to a number of other individuals (children), fulfilling which requires collective action. This might be an action taken together only with others in this potential group or also with motivated third parties. If we take these duties seriously at the individual level, we must for consistency take seriously the claims they make on duty bearers to act together.

Imagine that several fathers and children are stranded for years somewhere without schools. None could adequately educate his child alone but, given complementary areas of expertise, they could educate all the children between them. Without the prospect of cooperation (or of cajoling or bribing the others to teach his child), any father could invoke the ‘ought implies can’ proviso to reject any duty to educate his child. The option of collective action undermines this response, at least until he has tried to secure cooperation. The shared duty follows from the duty each has to his own child.

On parental duties

Parents have a duty to do certain things for their children. This much is uncontroversial. However, I take it as a starting assumption that these are special duties in a strong sense: moral duties owed by one person to another by virtue of some particular event, transaction or relationship. ³ They are not ‘special’ only in the pragmatic sense that would arise had general duties, owed by all to all, been socially allocated to parents for convenience or efficiency. ⁴

In terms of how these particular special duties are acquired, most of this article is neutral between two influential accounts. One is a causal view: parents, having caused someone to exist in a state of vulnerability to certain harms, acquire a negatively grounded duty to protect them from those harms as far as necessary and possible (Archard, 2010; Blustein, 1997). The other is an intentionalist or voluntarist account, on which parents acquire these special duties by committing, perhaps implicitly, to rear and care for a child (Blake, 2010; O’Neill, 1979). ⁵ Given this neutrality, three of the four arguments to follow apply to all parents, rather than only procreative parents. (The exception is the direct duties to further descendants argument, which assumes a causal account.)

A duty to protect one’s children from climate change makes certain assumptions about the content of parental duties. It takes for granted that parents should not only if necessary provide basic needs (nutrition, shelter, healthcare) in childhood and prepare their children to live as fully functioning, independent adults in their community (education, social acclimatization, emotional grounding) but also provide for the continuation of the background conditions against which those children can meet their basic interests as adults. ⁶ By ‘basic interests’, I mean uncontroversially central human interests such as opportunities for physical health, emotional development, affliation, and exercise of practical reason. This goes beyond what is explicit in much of the debate, although it is implicit in some discussions of how fairly to allocate the costs of children (Olsaretti, 2013; Tomlin, 2014). My claim, however, is that if parents have the first two sets of duties in any meaningful sense, they must have this third one. ⁷

Begin by assuming a duty to prepare one’s child for adulthood, emotionally and educationally. If the child is starving and the parent could but does not prevent this, continuing only to give hugs and bedtime stories, it would seem a mockery of caring for them. Equally, if the child will face starvation or serious disease in the future and the parent could but does not prevent this, the same can be said for her other efforts to prepare that child for adulthood. If the parent is to have any special duties reaching beyond her child’s active dependence, these must include duties to do what is both possible and necessary to enable that child to continue to live a decent life.

I am not making the general claim that whenever something (x) is missing without which the fulfilment of a duty would be futile or impossible, the duty bearer acquires a further obligation to provide x. This would be absurd: they could equally well cease to have the duty altogether. Suppose Joe has promised to play tennis with Sam but Sam’s bicycle breaks and he cannot get to the court. We do not automatically assume that Joe should drive him there.

Rather, I am drawing on something specific to the origin and nature of parent–child duties. This applies on either an intentionalist or a causal view of that duty’s acquisition. On the former, the parent has committed to do certain things for her child towards the general end of making the child’s life go well. If circumstances change, the spirit of the commitment would require her to change what she does. Suppose Joe’s promise was part of a broader commitment to help Sam turn his life around. Then he might be expected to assist Sam by getting him to the court, if necessary. ⁸

Alternatively, suppose that ‘whosoever is causally responsible by their actions for the existence of a child is responsible for ensuring that the child does not suffer the harms that it undoubtedly will in the absence of proper parental care and protection’ (Archard, 2010: 113–114). What this requires depends on what those harms are and when they will occur. ⁹ If a dangerous childhood disease is a significant risk in their society, parents have a special duty to do what they can to prevent their child from contracting it (vaccination, other precautionary measures). Equally, given the likelihood of a dangerous disease at age 40 unless some otherwise benign drug is administered at the age of six, parents should protect their children by administering that drug, if they can.

It is important to clarify the claim made in this section. I have argued that parents owe it to their children to protect their opportunity to secure basic interests, by virtue of the special relationship between them. This is so regardless of whether those same interests are also protected by duties of justice held either by all the children’s fellow citizens or by all affluent fellow humans. If they are, I see no difficulty with parents having simultaneously a general and a special duty to do the same thing. If not – if neither general cosmopolitan duties to the vulnerable nor extensive duties of social justice could be defended – then it is an advantage of this article that most of its arguments would hold despite this. (The exception is the just world argument.)

Indeed, I can remain neutral about who has primary responsibility for securing the background circumstances for flourishing the next generation of adults. Irrespective of whether this is parents, the state or all of us, the same practical point can be made. If the interests are not being secured at some other level, parents, by virtue of this special relationship, must take action to secure them if they can. ¹⁰ If they are, action by parents may not be necessary. ¹¹ Recall the disease example, above. If there is a free vaccination programme provided through schools, a parent need to allow her child to take part in this. If only private options are available, she must pay for these (if she can) or work with other parents to change the system.

One last preliminary point: a likely concern regarding this expanded account of parental duties is that it overburdens parents. This can be eased: recall the ‘as far as possible’ clause, above. In determining whether a duty is appropriate to all things considered, it matters what the cost will be to the duty bearer and what other duties she has. ¹² In the case of shared duties, whatever cost to the individual would render it morally permissible not to fulfil the duty must be measured in terms of both the achievement of cooperation and the cost of doing one’s part in whatever collective scheme is established. In other words, there must be some collective scheme which distributes the burdens so as to avoid unreasonable cost to any of its participants, and no parent can be expected to take on unreasonable cost in attempting to establish such a scheme. I will return to this caveat in the climate change case but note, anticipating this, that my arguments apply primarily to affluent parents.

The children’s interests argument

Climate change threatens fundamental interests, including the lifetime of today’s children. Health, life and community are all at risk. Diseases are predicted to rise, dangerous heat waves and other weather events to increase and whole coastal communities to be threatened even in affluent countries (IPCC, 2014c: 13; Karl et al., 2009: 47–48; Stern, 2006: viii; Zsamboky et al., 2011). Ongoing, significant, individually unavoidable risk to basic human interests renders those central interests insecure and threatens human flourishing (Cripps, 2013: 40–43; Wolff and De-Shalit, 2007: 64–84). Such risk requires one to endure the associated psychological and other costs even if one’s health, life or community is not actually snatched away. Which, of course, it might be.

From this, and from the previous section, a special parental duty follows to protect one’s children from climate change, at least to the extent of ensuring that their opportunity to secure central human interests is not undermined. Governmental and intergovernmental action has hitherto been inadequate, ¹³ but parents (especially affluent parents) could pursue mitigation or adaptation between them. They could campaign for governmental and intergovernmental action or they could commit to a global, emissions-cutting ‘citizens’ movement’ (Jamieson, 2011: 36) and invest in adaptation technology. Thus, on the apparently obvious children’s interests argument, they have a shared special duty to do this. For mitigation, this requires global-level cooperation; for adaptation, it might be achieved at the local or state level.

However, this step is too quick: it ignores the possibility that many parents could protect their own children more directly. Let us begin with mitigation. To defend a global-level shared parental duty to mitigate climate change, mitigation would have to be necessary to secure each child such fundamental interests as continued life, health and community. This is questionable. ¹⁴ In fact, adaptation could keep serious health impacts in Europe and North America within or below ‘medium risk’ levels until well into the second half of the century (IPCC, 2014c: 21–22). Moreover, it would be several decades before mitigation efforts now would take effect (Stern, 2006: i; IPCC, 2014c: 9–11). ¹⁵ The claim is not that mitigation could not or would not improve prospects for today’s children in later life. (Given that many will still be alive in 2080–2100, it could.) It is that their prospects for a flourishing adulthood could be reasonably secured without mitigation.

Turn, then, to the possibility that the children’s interests argument could yield a shared special duty to cooperate for adaptation. Without adaptation, severe impacts are expected for the globally vulnerable, and even many developed states will see their citizens face significant risks, over a more immediate timescale (IPCC, 2014c: 21–23). This suggests a shared parental duty to exert collective pressure on state or local governments, or if necessary to invest together in adaptation technology.

Even here, however, sweeping conclusions must be avoided. Some such potential groups would fail even a moderate cost condition: parents in the most vulnerable communities or states could not secure adaptation even between them, at least without serious sacrifice. At the other extreme, some developed countries may already have enough state adaptation in train to protect current children’s future health, life and community. ¹⁶ Some very affluent parents may be able to protect their children on an individual basis, for example by funding them to avoid the worst hit areas. I am not suggesting that this would be generally sufficient: only that it might be for some parents and could be easier and less costly than promoting collective-level adaptation.

Thus, this first – and most obvious – argument yields only the limited conclusion that parents owe it to their children to protect them from climate change and that this sometimes implies a shared duty to organize for state- or local-level adaptation. I turn, then, away from the impact of climate change on such uncontroversially ‘basic’ elements of our children’s lives as their own physical health and from the indirect impact of its devastating implications for others.

The indirect further descendants argument

For most of us, the welfare of certain others can determine how well our lives go. We cannot flourish without affiliation. Moreover, our emotional health is bound up in those to whom we are closest. This is especially so with our children: someone whose child becomes or is likely to become seriously ill has had their life made worse – has, in a sense, themselves been harmed – in a way for which it is not possible to compensate. However, it also applies to other loved ones, including grandchildren and perhaps even great-grandchildren.

This requires us to revisit what is required for parents to protect their children’s opportunity to secure the central elements of a decent human life: they cannot do so without also protecting at least some interests of those children’s immediate descendants. Thus, building on the discussion of shared and parental duties, parents have a duty, where necessary and possible, to secure these more distant interests.

In the climate change case, a special shared parental duty is defended. Even over this century, risks to central interests dramatically increase (IPCC, 2014c: 21–22). It also becomes less plausible that individuals could protect their own grandchildren or great-grandchildren without at least community or state-level adaptation. Given the prospect of some crisis situation, current parents cannot even rely on the continuation of a social structure which permits individual insulation at the expense of others. Thus, relatively affluent parents would share a duty to secure adaptation at local or state level.

The need for mitigation also becomes evident. By 2080–2100, when the children of current children would not be beyond middle age, a 4 per cent temperature rise would impose serious health risks which adaptation, though necessary, could only curb to a limited extent. This is so even in more developed areas (IPCC, 2014c: 21–22). ¹⁷ From 2100, there is a ‘high to very high risk of severe, widespread and irreversible impacts’ without dramatic global emissions cuts over the next few decades (IPCC, 2014a: 18). The parental duty to mitigate climate change is shared at a global level but, again, would be likely in practice to require action only by more affluent parents.

This is one version of the indirect further descendants argument. The other follows in the next section. It is important to emphasise that the duty is still to one’s own children. This sets this argument apart from ‘chain of concern’ approaches to intergenerational justice which use this ‘stake’ in overlapping generations to ground a general duty to protect the fundamental interests of the next generation, owed to parents or potential parents (Howarth, 1992; Passmore, 1974: 82–91; Rawls, 1971: 254–255). ¹⁸ Moreover, whatever duties parents owe directly to further descendants, or, however, motivated they are to protect them, this argument would hold if they had none, or were not so motivated.

One obvious objection is that parents do not know that their children will have children and grandchildren. Another is that it will be their decision to do so and parents need not protect them from the consequences of bonds they have chosen to forge. However, not being able to have children would, for many, be a central interest deprivation. ¹⁹ Parenting is so valuable an activity for those who undertake it that it might plausibly be asserted that there is a central human interest in having a secure opportunity to be a parent (Brighouse and Swift, 2006: 91–101; Gheaus, 2016; Overall, 2012: 19–33, 180–184; Robeyns, unpublished; Robertson, 1994: 24). ²⁰ Still less controversial is the claim that there is a central human interest in being able to develop and live by one’s own plan of life, at least to some basic degree (Nussbaum, 2000: 78–79). In practice, for very many persons, the opportunity to parent is integral to this (Cripps, 2016a: 7). ²¹ Parents cannot assume that their children will be the exception.

On either account, given that a central element of a human life is insecure unless it can be maintained without exposing another to serious risk (Wolff and De-Shalit, 2007: 70–72), it would seem a violation of parental duties not to spare one’s child the choice, if possible, between leaving her children and grandchildren liable to great suffering, and not having them at all.

As in my earlier discussion of parental duties, this is a special duty held by parents because of the parent–child relationship. On an intentionalist model, it would be inconsistent with a commitment to enable one’s child to live a flourishing life not to prevent a situation in which she can only secure one central human interest by jeopardizing another. On a causal account, this is an area of significant future vulnerability for those whom parents have brought into the world, from which they are owed protection. This holds regardless of whether parents and others also have general duties to render current children’s central interests mutually secure. ²² I certainly do not deny such duties: whether as duties of social or of global justice. Nor need I assume even that primary responsibility lies with parents. Given that the necessary action is not being taken elsewhere, parents would owe it to their own children to step in now, wherever the primary responsibility lay.

The just world argument (and the indirect duties to further descendants argument revisited)

The next argument – and another version of that just given – relies on a certain view of our position as individual human beings, which goes beyond the perspective of our own interests. These interests give us central reasons to act, but we also have central reasons of other kinds. As has been implicit throughout this article, one set derives from our close relationships with others. Call these interpersonal reasons. As moral agents, we also have central reason to avoid causing, and to prevent, the serious suffering of our fellow human beings. Call these impersonally moral reasons. ²³

Pervasive conflict between or within these sets of central reasons can present individuals with choices that many will experience as marring or ‘tragic’. Consider the position of a woman whose brother’s genetic condition will incapacitate him as he gets older. This gives her a central reason to invest ever more time and money in preparing to care for him and/or in promoting medical progress. Such efforts will pull against her own interests, other interpersonal ties and other actions that she has central impersonally moral reasons to perform.

This is not a claim about what can be demanded of each individual as an impersonally moral or interpersonal duty: what she can be criticized for not doing. Serious conflict with her interests or (other) interpersonal ties counts against the all-things-considered attribution of any such duty. (I will come back to this, in the climate change case, in conclusion.) It is about the pull that the severe suffering of others and the needs of those closest to us have on us even after we have expended significant effort trying to redress or secure them. There will always be more we could do and, from the interpersonal or impersonally moral perspective, have central reason to do. Note, moreover, crucially for the current discussion, that such ongoing conflict between central reasons to act tends to arise in situations of collective failure to tackle wide-scale suffering which, if adequately addressed collectively, could be alleviated while leaving individuals with ample scope to pursue their own lives.

Building on my earlier account of parental duties, I suggest that parents owe it to their children to spare them such pervasive conflict. This will require action by those parents if this is reasonably possible (individually and collectively) and not being done elsewhere. As in the previous section, the point can be made on either a causal account, on which parental duties derive from having caused someone to exist in a condition of vulnerability, or an intentionalist one on which parents have implicitly or explicitly committed to caring for their child. Because ongoing incompatibility of central reasons can make a person’s life hard for them to live, in a very real sense, it is an area of significant vulnerability. It is also in the spirit of any commitment to enable that child to live a flourishing life that they be spared such conflict.

Against this background, another version of the indirect further descendants argument can be offered. Collective failure to mitigate climate change will leave today’s children facing constant tension between their own interests, other interpersonal ties and the fundamental prospects of their children and grandchildren. At some point, continuing to attempt to secure their offspring’s future against climate change would conflict with other things they had central reasons to do for those children: even parental duties they have central reason to fulfil. Thus, parallel conclusions can be drawn to those of the previous section: parents have a shared duty to secure adaptation at local or (probably) state level and mitigation at the global level.

The just world argument can also be made. Radical global injustice, including climate injustice, leaves individuals unable to reconcile the impersonally moral perspective, the personal perspective, and many aspects of the interpersonal perspective. (By radical global injustice, I mean widespread failure to provide human beings with even the minimum necessary for a decent life.) This is because there will always be more each could do – and has central impersonally moral reason to do – to protect the vulnerable, in this or in future generations.

True, the arguments above do, in one way, reduce such conflict: they give parents central reasons to do something for their children (mitigate climate change) which they also have impersonally moral central reason to do. But individuals have other morally impersonal central ends, including pursuing global-level adaptation. Devoting all one’s effort to these would be incompatible not only with pursuing mitigation but also with satisfying one’s own interests or pursuing other central interpersonal ends, including prioritising local and state-level adaptation to protect one’s own children.

Current parents already face a version of this predicament. If the current opportunity to avoid dangerous climate change is missed, and more extensive adaptation becomes necessary, the tension will escalate. Individuals will face choices many will find searingly difficult. Their parents, as parents, must spare them this if they can. In the climate change context, this yields the furthest reaching conclusions yet on the shared parental duty. Addressing radical climate injustice requires collective action not only for state or local-level adaptation and global-level mitigation but also for global-level adaptation.

Two clarificatory points must be stressed. These arguments do not conflate individual flourishing with being ‘good’, either in virtue ethics terms or in terms of avoiding serious harm to others. The point is simply this. As well as being with our own interests and relationships, we are moral agents. As such, some of us find it very difficult to live with ongoing conflict between the central reasons to act given by each of these perspectives. Arguably, everyone should so find it. Parents, then, should operate on the assumption that their child might be so placed.

Moreover, as with duties to render children’s future interests secure, I do not deny that there may also be general duties to spare other pervasive conflict between central reasons for action. Indeed, I have defended such duties elsewhere (Cripps, 2013: 169–196). ²⁴ However, points from my earlier discussion apply again here. Parents (also) have such obligations as special duties and so are doubly required to act on them, if such action is necessary.

The direct duties to further descendants argument

The last two sections argued that parents owe it to their own children to cooperate to mitigate and enable adaptation to climate change, because of the impact on those children of serious climate harms to others. Now let us consider another possibility that parents owe this directly to their own more distant descendants.

The motivating idea is that parents have a special moral duty to their grandchildren, great-grandchildren and further descendants to do what is reasonably possible towards securing them the background conditions for basic flourishing. To defend this, I must abandon my former neutrality and assume that parents have special duties to their children because they have caused those children to exist in a state of vulnerability to certain harms. I must also assume that being deprived of the opportunity to have and rear one’s own biological child would, for very many, be a central interest deprivation. This is more controversial than the view of human flourishing needed for the indirect duties to further generations argument. As such, the argument here is not only rather more speculative than those above but also applies only to procreative parents rather than to parents in general.

The argument, briefly, is as follows. In reproducing, parents bring not only their child into the world but the prospect of a whole line of descendants, all also vulnerable in central ways. If parents know this and know, too, that they (individually or collectively) could protect the fundamental interests of their grandchildren, great-grandchildren, and so on, the same initial responsibility that grounds their special duties to their children may ground a special duty to do this.

This premise accepted, the direct duties to further descendants argument parallels the children’s interests argument but with further reaching conclusions. Without collective action on climate change – adaptation and mitigation – future generations will face serious suffering. Central interests will be rendered insecure in the affluent as well as the developing world (IPCC, 2014c: 12–25). Parents cannot, in isolation, protect their descendants from this. However, they could secure progress at the necessary level by organizing for collective action, including if necessary institutional reform. (As with the indirect duties to further descendants argument, this may be at local or state level for adaptation but would be global for mitigation.)

Two obvious objections, and one query, should be anticipated. Firstly, in parallel to an objection to the indirect duties to further descendants argument, if I have a child and she has a child, it is her action, not mine, that places this new person in a situation of vulnerability. Thus, the objector insists that any special duties lie with her, not me.

Clearly, I do not deny that future parents will also have causal responsibility for bringing children into the world, and so acquire special duties to them. This would be counterproductive, since it would imply that current parents, for many of whom not procreating would also have been a great sacrifice, lack such duties. Nor do I deny that parents have a more primary responsibility to secure their children’s interests. In other words, insofar as they can, it is parents, rather than (great-) grandparents, who should be doing this. ²⁵

Also clearly, individuals cannot be said to have causal responsibility for (and so negatively grounded moral duties relating to) everything following from events they cause. In general, others’ choices break the chain. However, there are exceptions. If I (foreseeably, intentionally) cause something to happen which will lead to some other event unless another person makes a very great sacrifice, it seems reasonable to consider me partially responsible for that further event. ²⁶ For example, compare the following two scenarios. In (A), Anna pushes a boulder down a hill but it will not hit anyone unless Jane, who happens to be nearby, knocks it off course. In that case, it will hit and seriously injure Sara. In (B), the boulder will crush Jane’s leg unless she diverts it, but by diverting it she will roll it onto Sara. Whatever we feel about (A), it seems clear that Anna has at least some responsibility for any harm to Sara in (B).

I argued in defence of the indirect duties to further descendants argument that for many people being deprived of the opportunity to become parents would amount to deprivation of a central interest. If we assume, going beyond what was suggested there, that it would be such a loss – or sacrifice – for them not to become procreative parents, some moral responsibility for the existence of a child, and so some causally acquired duties to her, belong to those who brought her parents (and theirs) into the world.

The next objection is that this too dramatically increases the demands that can be made on current parents, as potential (or actual) future grandparents, great-grandparents and so on. This can be alleviated. Parents not only (as stressed above) also have a special duty to care for their children and secure their basic needs but also have a more primary responsibility for fulfilling that duty and are generally better placed to do so than grandparents or great-grandparents. ²⁷ For anyone more distant than a great-grandparent, duties requiring immediate contact are ruled out on ‘ought implies can’ grounds.

The climate change case is exceptional in that current parents not only can adequately protect their further descendants’ basic interests but, in terms of timely mitigation, may be the only generation who can do so. As already noted, without mitigation, serious health and other risks will spread across even developed countries by 2080–2100 and grow still worse and more pervasive thereafter (IPCC, 2014a: 18; IPCC, 2014c: 21–22). ²⁸ This renders our grandchildren, great-grandchildren and so on peculiarly liable to certain harms ‘in the absence of proper [(great-)grand]parental care and protection’ (adapted from Archard (2010: 114)).

The further query is this: across how many generations do duties to descendants extend? ²⁹ Further, evidently, than those upheld by the indirect further descendants argument. There, parents were required only to protect those further descendants with whom their children could build a relationship. But must parents now protect all possible descendants, no matter how far into the future, from the risks of climate change? I think not, although even a tentative defence of this depends on a particular interpretation of the causal derivation of parental duties.

On this, special parental obligations apply only if bearing a child is the reasonably foreseeable result of reasonably avoidable action. Compare a couple having a planned child with parents who conceive despite scrupulously careful contraceptive use and for whom an abortion is unavailable, or a woman who chooses to leave her house, is raped and is denied an abortion. ³⁰ Thus, the probability that having children will mean having descendants in y-generations time affects whether the parents have special duties to those descendants.

Suppose all families were single-child, with a known chance greater than 50 but less than 100 per cent that one’s child would go on to have a child, an equal chance that child would and so on. Looking forward across generations, the chance steadily reduces of any given current parent continuing to have descendants. ³¹ Eventually, it becomes smaller than that of the careful couple conceiving, and finally smaller even than that the third woman faces of being raped and conceiving. And this ignores the possibility – small at any given point but aggregating over time – of some one-off external event wiping out one’s descendants, one’s society or even humanity altogether. At some point along this scale, the chance would be too small to justify assigning special duties to the many-greats grandparent.

Stepping back from this artificially simplified case, my point is this. Depending on factors such as the number of one’s children, the number of children they can be expected to have (factoring in choice and chances of fertility problems) and so on, and the possibility of external events rendering whole families, communities or populations extinct, there must be some limit to the number of generations unto which parents have special duties to their descendants. However, the argument would apply for at least several generations.

Concluding remarks

This article has defended a special, shared parental duty to organize to act on climate change, through both adaptation and mitigation. Without such action, parents cannot secure their own children’s indirect, or relational, interest in not seeing their immediate descendants condemned to great suffering nor leave them able to reconcile their own interests and relationships with the perspective of the moral agent. Nor can they fulfil special duties owed, at least arguably, to their more distant descendants.

Two possible courses of action have been indicated. Parents might form a ‘global citizens’ movement’ (Jamieson, 2011: 36), each committing to cut her own emissions. ³² They might collectively commit to investing in adaptation technology. Alternatively, parents might use their collective political influence to bring about effective state- or global-level political change, for example by pressuring governments to fulfil and increase their commitments under the Paris Agreement. ³³ In neither case need cooperation be by all parents with all other parents, even at the relevant level (local, state, or global for adaptation; global for mitigation). Nor need it be only with other parents. Recall that this shared duty arises because parents have individual duties which they can only fulfil between them. Suppose this could be achieved most effectively (and at reasonable cost) by throwing the collective weight of those parents motivated to be part of such action behind existing efforts to combat climate change. This would be a legitimate way to fulfil the duty. ³⁴

One outstanding objection remains: that of over-demandingness. ³⁵ In this context, the question is: are the costs to parents unreasonable, either of securing collective parental action on climate change, or of that action itself? ³⁶ I think they need not be, whether we consider only costs directly to duty bearers or also opportunity costs in terms of other special duties.

I have left open what counts as ‘unreasonable’, although the threshold is probably higher for parental duties than for general positive duties. But it seems likely both that a reasonable cost condition would exclude very impoverished parents and that some collective endeavour could put the burden primarily on the remaining parents (or even some significant proportion of them) without being excessively burdensome or jeopardizing other special duties. Estimates have been made of 1–2 per cent of GDP for mitigation and moving towards climate change resilient infrastructure (Stern, 2006: xiii, xxi). Given existing networks, cooperation itself should not be unfeasibly costly. Most affluent individual parents could do much more than at present to promote such cooperation without undertaking excessive burdens or undermining interpersonal ties. ³⁷

In terms of moral opportunity costs in a broader sense, a full discussion is needed of whether shared parental climate duties are (collectively or individually) compatible with other duties of justice and whether they can legitimately be prioritized if not. As such a discussion lies beyond the scope of this article, no all-things-considered conclusions can be drawn as to how parents should act. Recall, though, that in defending a shared special duty to mitigate climate change, this article serves in one way to ease the tension between parental and general duties. It requires affluent parents to do for their own children something also demanded of them on almost all accounts of climate justice. ³⁸

I must end with a note of caution. My optimism regarding demandingness concerns only the present. Suppose the global window of opportunity for keeping climate change below 2 or even 3°C were to be missed. With escalating costs, even the duties defended in this article could then become mutually incompatible. Many affluent parents, even acting collectively, could have to choose between local- or state-level adaptation to protect their children’s lives, health and livelihoods from already significant warming, and mitigation to prevent the already terrible prospects for future generations, including their own descendants, from getting worse. In other words, there would be no way they could act that would not leave their own children with significantly marred opportunities.