1. Introduction
It is a point of essentially universal agreement in philosophical discussions of the ethical significance of the threat of severe climate change that we, that is, well off people in industrialized countries, have an obligation to engage in substantial mitigation by reducing our emissions of greenhouse gases in order to limit the damaging effects of climate change that future generations will have to endure. There is much less agreement, however, about how these obligations are best explained, or, in other words, about the morally relevant considerations that ground the obligations. Much of the disagreement about what considerations ground our mitigation obligations arises due to the difficulties presented by the well-known non-identity problem (NIP).Footnote 1 The NIP shows that we cannot explain our mitigation obligations by claiming that failing to mitigate will lead to particular future people being worse off than they would otherwise have been. This is because which people will come to exist in the future depends on whether we adopt a policy of mitigating or continue to allow emission levels to rise at expected rates. Of course, climate change is already generating significant negative impacts,Footnote 2 and it is possible that our actions now will affect at least some future people who will exist regardless of whether or not we engage in mitigation efforts. But these possible effects on particular people do not exhaust the reasons that ground our obligation to engage in mitigation efforts, and the reasons that they might provide could not explain the full range and extent of our mitigation obligations. This is because our actions now will also significantly affect the conditions of life many generations in the future, and the negative effects that will result from our continuing to emit at high levels would seem to give us strong reasons to reduce our emissions, even if those effects would only come about far enough into the future that no one who would exist if we fail to mitigate would also exist if we engaged in substantial mitigation.Footnote 3
One way of attempting to ground our mitigation obligations in light of the NIP is to appeal to impersonal factors. These are considerations that allegedly make an outcome, policy, or action better in some morally relevant respect, even though it isn’t better for any particular person. Derek Parfit offers an impersonal principle that he thinks provides at least part of the grounds of our obligations in non-identity cases, which he calls the Same Number Quality Claim, or Q. According to Q, ‘If in either of two possible outcomes the same number of people would ever live, it would be worse if those who live are worse off, or have a lower quality of life, than those who would have lived’ (Parfit Reference Parfit1984, 360). If we accept this principle, or something like it, then we are committed to thinking that there are moral reasons, all else being equal, to ensure that better off people come into existence rather than the same number of different, worse off people, when those are the only relevant alternatives. If there are such reasons, then, plausibly, they ground obligations in cases in which the cost to us of ensuring that better off people come into existence rather than worse off people is relatively low, and also in cases in which the difference in the expected quality of life for distinct groups of future people is large, even if the cost to us of ensuring that the better off people come to exist is fairly significant (as may be the case with respect to climate change mitigation).Footnote 4
An impersonal principle such as Q, then, can explain what is wrong with failing to mitigate, and, if true, would ground a robust and plausible set of obligations to engage in mitigation efforts.Footnote 5 We would be required to accept substantial (though not unlimited) costs in order to reduce emissions and thereby ensure that future people have a significantly higher quality of life than they would have if we continued to emit at high levels, even though none of the people who would exist if we accept those costs would also exist if we refused. And the explanation of the fact that we have these obligations would be that it is better, that is, impersonally better, that the better off people come to exist rather than the worse off people.
There has, however, been quite a bit of resistance to the view that our mitigation obligations are grounded at least primarily in impersonal factors of this sort. A number of philosophers have developed alternative accounts of the grounds of these obligations. Some aim to avoid appealing to impersonal factors altogether (Hanser Reference Hanser1990; Harman Reference Harman2004, Reference Harman, Roberts and Wasserman2009; Gardner Reference Gardner2015), while others argue that although impersonal factors can provide reasons to mitigate, our obligations are grounded primarily in, for example, the need to avoid violating the human rights of future people, or the need to avoid harming future people.Footnote 6 There are a number of motivations that have led some to seek an alternative to grounding our mitigation obligations primarily in impersonal factors. Perhaps the most important, however, lies in the thought that if an action is wrong, there should be at least one person who is entitled to complain about the action’s being performed on her own behalf. In other words, many are inclined to think that an action can only be wrong if there is at least one person who could claim to be wronged by its performance.Footnote 7 It does not seem, however, that the fact that one is worse off than different people could have been, had previous generations pursued different policies, could provide one with a basis for a complaint on her own behalf about what the previous generations have done, or with grounds to claim that she was wronged by those earlier generations. Because of this, many have rejected the view that the fact that one policy, for example mitigation, would lead to the existence of better off people than another policy, for example continued high emissions, is in itself a sufficient basis for an obligation to choose mitigation. They deny, then, that this fact can explain why failing to mitigate would be wrong, and insist that any adequate explanation of the obligation to mitigate must refer to complaints that individuals would be entitled to make on their own behalf against our failure to mitigate.
My aim in this paper is to argue that several prominent attempts to explain (at least the bulk of) our mitigation obligations without appealing to impersonal factors fail, since they either cannot account for a plausibly robust obligation to mitigate, or else have implausible implications in other cases in which we might act in ways that will lead to certain people coming into existence. I will argue, then, that despite the appeal of the motivations for rejecting the appeal to impersonal factors, such factors must play a prominent role in any plausible account of the grounds of our mitigation obligations.
2. Human rights and mitigation obligations
2.1. Can present actions violate future people’s rights?
Simon Caney claims that if we fail to engage in sufficient mitigation efforts, the effects of climate change will be such as to violate a number of the human rights of many future people. For example, some of the effects of unmitigated climate change are very likely to lead to many people being unable to meet their subsistence needs (Caney Reference Caney2006, 261; Reference Caney2008, 538; Reference Caney, Beitz and Goodin2009, 232; Reference Caney and Humphreys2010a, 80–82; Reference Caney, Pierik and Werner2010b, 27; Reference Caney, Hicks and Williamson2012, 99), unable to avoid serious threats to their health (Caney Reference Caney2006, 261–262; Reference Caney2008, 538; Reference Caney, Beitz and Goodin2009, 531; Reference Caney and Humphreys2010a, 78–80; Reference Caney, Pierik and Werner2010b, 27; Reference Caney, Hicks and Williamson2012, 99), and even unable to avoid dying as a result of climate change-related causes (Caney Reference Caney, Beitz and Goodin2009, 230–231; Reference Caney and Humphreys2010a, 76–78; Reference Caney, Pierik and Werner2010b, 27; Reference Caney, Hicks and Williamson2012, 99). The fact that failure to mitigate would lead to such human rights violations is, on his view, the primary consideration that explains our mitigation obligations.Footnote 8 We are obligated to mitigate, then, because if we fail to do so, our policy of continuing to emit at high levels will bring about a state of affairs in which many future people’s human rights will be unsatisfied. If we do in fact fail to mitigate, and many future people’s human rights therefore go unsatisfied, then those people will be able to complain about what we did on their own behalf; they will have been wronged by our failure to do what we are obligated to do.
Some will object to Caney’s view that our actions now cannot violate the rights of future people, since future people do not now have any rights, and an action cannot violate rights that do not exist at the time that it is performed.Footnote 9 Others will claim that future people who would not exist if we did mitigate will not, if they do in fact come to exist in virtue of our failure to mitigate, have any basis for complaint on their own behalf against our actions, at least so long as their lives are, on the whole, worth living. The thought here is that even if one at some point in her life is stricken with a climate change-related disease, or is killed by a climate change-related superstorm, if she wouldn’t have existed in the absence of the policy of high emissions that led to climate change being as severe as it is, and her life was on the whole not worse than no life at all, then she has no grounds for claiming that she was wronged by our failure to mitigate, since there is nothing that we could have done that would have made things any better for her.Footnote 10
The first objection is simply mistaken. If I plant a bomb today that explodes in one hundred years and kills someone who will not be born until fifty years from now (and would have been born whether or not I planted the bomb), it seems clear that my action will violate that person’s right to life (Elliot Reference Elliot1989, 162). Of course neither the person, nor her right, exists now, at the time that I act, but what I do now can come to constitute a rights violation in virtue of what happens at a later time.Footnote 11 The second objection is more plausible, but it can also be resisted. It is at least not obvious that someone who has a life that is on the whole worth living cannot have a legitimate complaint against the performance of an action that causes her to be in a condition in which her rights are violated, or causes her to be harmed, even if the performance of that action was also a necessary condition of her coming into existence.Footnote 12 Indeed, all of those who reject the view that impersonal factors have a role in explaining the wrongness of our failure to mitigate claim that individuals can have legitimate complaints against actions that were necessary conditions of their coming into existence, even if they have lives that are on the whole worth living. My defense of the need to appeal to impersonal factors will not rely on the rejection of that claim; I accept, for example, that the fact that an action of mine would violate the rights of a future person can ground a complaint on that person’s behalf against my performance of that action, and provide me with a reason not to perform it, even if my performing it is a necessary condition of the person’s coming into existence, and even if she would have a life that is on the whole worth living. Instead, I will consider the implications of accepting that individuals who will live in conditions of severe climate change if we do not mitigate can make versions of the particular complaints that Caney and others have suggested ground our mitigation obligations. In some cases, I will grant that there is nothing that makes accepting the complaints as legitimate clearly problematic, but argue that the considerations upon which they are based can only ground rather limited mitigation obligations, and therefore must be supplemented by additional considerations in order to fully explain our mitigation obligations. The relevant additional considerations must be either impersonal, or else provide the basis for further complaints on behalf of those who would exist if we fail to mitigate. But accepting that any of the considerations that might provide the basis for further complaints on behalf of future people in fact do so would, I argue, commit us to implausible views about the permissibility of acting in ways that will result in certain people coming into existence in other cases, particularly in cases in which the only alternative is that no one is brought into existence. To the extent that it might appear that allowing these complaints is compatible with avoiding the implausible implications, this is only because the complaints fail to function as they are meant to, that is, as complaints on behalf of the individuals allegedly entitled to make them; in these cases, I will argue, they in fact function as disguised articulations of the significance of impersonal factors.
2.2. Can human rights ground our mitigation obligations?
Caney’s human rights-based account of the grounds of our mitigation obligations appeals to what I will call a low-threshold account of the relevant rights.Footnote 13 The reasons that we have to mitigate, on his view, lie in the fact that failing to mitigate would lead to unmet rights to, for example, life, subsistence, and freedom from serious threats to health (Caney Reference Caney2006, 259–264; Reference Caney2008, 538–539; Reference Caney, Beitz and Goodin2009, 230–232; Reference Caney and Humphreys2010a, 76–82; Reference Caney, Hicks and Williamson2012, 98–99). On this view, if we fail to mitigate, then those who would suffer climate change-related death, serious illness, or lack of access to subsistence goods would have a legitimate complaint against our actions, even if they had lives that were on the whole worth living. This seems at least plausible, and I will not dispute it here.
If these are the only rights that we allow to ground complaints against actions that both lead to violations of those rights and are necessary conditions of the existence of the victims of the relevant violations, however, then we will not be able to explain, by appeal to human rights-based considerations, why we are obligated to mitigate more than just enough to ensure that all future people can avoid climate change-related death, serious illness, or lack of access to subsistence goods.Footnote 14 But surely it is implausible that we are only obligated to ensure that future people are protected from these human rights violations, if at modest further sacrifice to ourselves we can ensure that those who will live in the future will have much better lives than lives just above this threshold.Footnote 15
Imagine that we could, at modest cost to ourselves, mitigate enough to ensure that our actions will not exacerbate climate change to an extent that will cause further deaths, serious illnesses, or lack of access to subsistence goods. But this level of mitigation would lead to billions of future people being unable to achieve a standard of living much above subsistence and avoidance of serious illness. We could, alternatively, adopt more ambitious mitigation measures that, though fairly costly for us, would ensure that the different future people who would exist if we choose this option are all able to achieve a standard of living that is comparable to our own. It seems clear that in this case we would be obligated to adopt the more ambitious mitigation policy. But we cannot ground this obligation in low-threshold human rights, so if we accept a low-threshold account of human rights, it looks as though only impersonal considerations could explain it.
Those who wish to avoid appealing to impersonal considerations in order to explain why we are obligated to mitigate to a greater extent than is necessary to ensure that the human rights that Caney refers to are not violated would seem to have to endorse a much more expansive, high-threshold account of the content of human rights.Footnote 16 Perhaps, for example, we should hold that there is a human right to a standard of living that is comparable to that of, say, a typical middle class citizen of a moderately wealthy industrialized country, in addition to the minimal rights to which Caney appeals, so that we are obligated to mitigate enough that future generations will have access to that reasonably high standard of living, so long as the costs to us of doing so are not too extreme. On this view, anyone born without access to that standard of living as a result of our failure to mitigate would have a legitimate complaint on their own behalf against our actions, despite the fact that they would not have existed had we acted otherwise.
It is, however, not plausible that future people whose standard of living is just below that of a typical middle class citizen of a moderately wealthy industrialized country today would have a legitimate complaint, on their own behalf, against actions that were necessary conditions of their coming into existence. After all, if we think that these people would have such a complaint, then we are committed to thinking that anyone who is born without access to that standard of living has a legitimate complaint on their own behalf against any actions that caused them to lack access to the specified standard of living, even if those actions were necessary conditions of their coming into existence. This, however, is deeply implausible.
Imagine that my only options are to bring into existence Joe, who would have a life that is well worth living, but whose standard of living would be just below that of a typical middle class citizen of a moderately wealthy industrialized country, or to not bring anyone into existence at all. It seems clearly false that it would be wrong for me to bring Joe into existence, and it is implausible that he would have a legitimate complaint on his own behalf against my doing so. A high-threshold account of the content of the human rights that can ground our mitigation obligations, then, would commit us to unacceptable conclusions about the permissibility of causing people to exist in many cases in which the only alternative is causing no one to exist.
It might be suggested, however, that the content of the rights that can ground legitimate complaints on behalf of individuals against actions that were necessary conditions of their coming to exist can vary depending on the options that are available to those who might perform such actions. Perhaps, for example, the violation of a right to have access to a certain fairly high standard of living can only ground a legitimate complaint on behalf of a person who lacks access to this standard of living due to actions that were also necessary conditions of her coming into existence if those who performed the relevant actions could have acted differently and thereby caused different people to come into existence who would have had access to the relevant standard of living.
This view, however, doesn’t really avoid appealing to impersonal considerations; it simply masks that appeal by presenting it in the language of rights violations. What a violation of a person’s right to have access to a certain standard of living amounts to, on this view, is a failure to act in a way that would have caused a different person to exist who would have had access to that standard of living. If, in the case involving Joe described above, I could also have acted in a way that would have led to Jane coming into existence and having access to a standard of living above the threshold, then my causing Joe to come into existence would constitute a violation of his rights, whereas if the option to do what would cause Jane to exist was not available, it would not violate his rights (since he wouldn’t have the relevant right). But if this is our view, then Joe’s grounds for complaint in the case in which he is caused to exist rather than Jane must somehow refer to the fact that I could have caused Jane to exist instead. That I could have caused Jane to exist is an essential feature of the complaint that Joe can lodge against my action, since it is only because I could have caused Jane to exist that doing what caused Joe to exist counts as violation of Joe’s rights. But since this is the case, Joe’s complaint cannot really be a complaint on his own behalf. Instead, it is a complaint that at least implicitly refers to the impersonal consideration that the alleged rights violators could have caused a different, better off person to exist instead of Joe.
It might be suggested that in considering only low, high, and variable threshold accounts of the content of the relevant human rights, I have neglected the most plausible view, which would give to these rights some appropriately ‘moderate’ content. I acknowledge that there are other possible ways of spelling out the content of the relevant rights that I have not and cannot consider here, but I want to conclude this section by noting what would have to be true for any account to meet the challenge that I have set, and suggesting that there is reason to doubt that any view can in fact meet it.
In order to ground the bulk of our mitigation obligations, an account of the content of the relevant rights would have to meet the following two conditions:
-
(1) We would have at most relatively weak moral reasons to sacrifice in non-identity cases to ensure that future people’s lives are improved beyond the threshold(s) set by the rights, even if we could improve their lives a great deal at modest cost to ourselves.
-
(2) Anyone who suffers a deficit with respect to the relevant right(s) would have a legitimate complaint on her own behalf against actions that played a role in causing that deficit, even if those actions were necessary conditions of her coming to exist, and even if she has a life that is, on the whole, worth living. This complaint would be strong enough that doing what would cause a person who would suffer a rights deficit to exist would be wrong, even if the alternative is causing no one to exist.
I suspect that there is no way of specifying the content of the relevant rights on which it would be plausible to think that both of these conditions are met. If we make the threshold high enough that (1) is plausible, (2) looks clearly unacceptable. And if we make the threshold low enough that (2) seems plausible, (1) looks clearly unacceptable.
If this is right, then appeals to rights will not be able to ground the bulk of our mitigation obligations, and those who are inclined to resist appeals to impersonal factors in the explanation of our obligations in non-identity cases will need to find some other grounds for a plausible set of mitigation obligations, or else reconsider their resistance to impersonal explanations.
3. Harm and mitigation obligations
An alternative approach to grounding our mitigation obligations without appealing to impersonal considerations holds that if we do not mitigate, many future people will have harm-based complaints on their own behalf against our actions, and that these complaints can explain why our failing to mitigate would be wrong. Some harm-based accounts appeal to a threshold of well-being, claiming that an action that causes a person to be below that level harms her in a morally relevant sense, even if that action was a necessary condition of her coming into existence (Meyer Reference Meyer, Meyer, Paulson and Pogge2003). This view is equivalent to a version of a rights-based view, and is therefore subject to the objections of the previous section.Footnote 17 But other harm-based views rely on accounts of harm that are neither counterfactual accounts (which imply that future people cannot be harmed by actions that are necessary conditions of their coming into existence, at least so long as their lives are on the whole worth living) nor threshold accounts. Elizabeth Harman, for example, claims that we should accept an account of harm according to which one is harmed when she is caused to be in conditions of certain types. She says, for example, that ‘causing pain, early death, bodily damage, and deformation is harming’ (Reference Harman2004, 92), even in cases in which the actions that cause these harms are necessary conditions of the victims of the harms coming into existence. On this account of harm, one can be harmed by an action even if she is caused to be better off overall than she was before, or better off than she would have been had the action not been performed. Harman further claims that the fact that an action would harm someone in one of the relevant ways provides a reason not to perform the action, even if the action is a necessary condition of the person who would be harmed by it coming into existence. On her view, then, if we fail to mitigate, those who suffer harms such as early death or serious illness as a result of climate change would have complaints on their own behalf against our actions, and the fact that they would have these complaints explains why we are obligated to mitigate. She claims that this account avoids appealing to impersonal considerations in determining what we are obligated to do in non-identity cases (Harman Reference Harman2004, 90, 93).
The reasons against harming are, on Harman’s account, defeasible. They can be outweighed by other considerations. This is why it is permissible for a doctor to perform necessary surgery on me, even without my consent (supposing that I am incapable of either consenting or refusing). The surgery does harm me, since it causes pain and bodily damage, but the reasons that this provides against performing the surgery are outweighed by the fact that performing it would prevent me from suffering even worse pain and perhaps death in the future (Harman Reference Harman2004, 91).Footnote 18 Despite the fact that reasons against harming are defeasible on her view, Harman thinks that they are still quite strong, and typically take priority over other morally relevant considerations, such as increasing welfare (Reference Harman, Roberts and Wasserman2009, 139). Her explanation of why we are obligated to mitigate, then, is that failure to do so will lead to future people being harmed in certain ways. We have reasons to avoid acting in ways that will harm people, even if those people would not be made worse off overall by our acting in those ways, and even if our so acting is a necessary condition of their coming into existence and having lives that are on the whole worth living. Since if we mitigate sufficiently we will either avoid harming future people, or else minimize the amount of harm that we cause them, the harm-based reasons that count against failing to mitigate do not also count against mitigating, or at least count less against doing so. This is why we are obligated to mitigate. It is, on this view, the harmful effects that our actions would have on particular people if we fail to mitigate that explain why we are obligated to mitigate. This is why Harman thinks that her view succeeds in explaining our mitigation obligations without appealing to impersonal considerations.
I do not think, however, that Harman in fact succeeds in avoiding appealing to impersonal considerations. On her view, the reasons against harming a person in virtue of acting in a way that is also a necessary condition of her coming into existence can be overridden by other considerations, so that it is at least sometimes permissible to act in a way that will cause harm to a person who would not come to exist if one acted otherwise. For example, she suggests that it at least might be permissible for a woman, Patty, of whom it is true that any child she conceives will be deaf, to have a child, despite the fact that deafness is a harm (and so causing a deaf person to exist harms her; Harman Reference Harman2004, 102). In this case, the fact that Patty would be harming her child, though it provides a reason against conceiving, can be outweighed by the fact that the child would have a life that is on the whole worth living, the fact that she would be worse off without any children than she would be if she does have a child,Footnote 19 or some combination of these reasons. So although any deaf child that she does have would have some complaint against Patty’s decision to conceive, that complaint would not necessarily be strong enough to make it impermissible for her to conceive. At the same time, Harman insists that a woman, Tammy, who can either have a deaf child, or take a modestly costly precaution in order to ensure that she has a healthy child, would act wrongly by refraining from taking the precaution and thereby having a deaf child. Tammy’s deaf child would be harmed in exactly the same way as Patty’s deaf child, and so her complaint against Tammy’s action would be identical to Patty’s child’s compliant against Patty’s action. So what is it that, on Harman’s view, explains why Tammy’s having a deaf child would be wrong, while Patty’s doing so may be permissible? Harman claims that it is the harm that Tammy’s deaf child would suffer, and the complaint that this child could make on her own behalf, that explains why it would be wrong for her to have that child (Reference Harman2004, 101–102); but this cannot be the whole story, since Patty’s child would suffer the very same harm, and would therefore have an identical complaint. Harman adds that part of the explanation of the wrongness of Tammy’s having a deaf child is that, unlike in Patty’s case, ‘the benefits to the child cannot render the harm permissible because she has an alternative of providing parallel benefits without parallel harm’ (Reference Harman2004, 102). But she denies that this undermines her claim that it is the harm that Tammy’s deaf child would suffer that explains why it would be wrong for Tammy to have that child, rather than any impersonal considerations. Her argument for this is, roughly, that for a consideration to be the explanation, in the relevant sense, of the wrongness of an action, it need not be the case that nothing else needs to be added in order to fully account for why the action is wrong (Reference Harman2004, 102). It is true, she acknowledges, that we must add the fact that Tammy has an alternative that would involve providing the same benefits to a different person, without harming anyone, in order to fully account for the wrongness of Tammy’s having the deaf child. But this does not mean, she claims, that this apparently impersonal consideration, namely that Tammy has an alternative that would be better in a morally relevant sense (i.e. not harming anyone), though not better for any particular person, is an essential part of the explanation of the wrongness of Tammy’s having the deaf child.
This strikes me as an implausible account of what is necessary to have an explanation of the wrongness of Tammy’s having the deaf child. But, moreover, and more importantly, it also seems to clearly fail to account for the wrongness of Tammy’s having the deaf child without appealing to impersonal considerations. Even if Harman is right that citing the harm is enough to provide an explanation, in some relevant sense, of the wrongness of Tammy’s having the deaf child, it is clear that a full account of the conditions that make Tammy’s doing this wrong, while Patty’s doing it is at least potentially permissible, must include reference to the fact that Tammy, but not Patty, has an impersonally better alternative available to her.Footnote 20
When we consider our obligation to mitigate, we can see that Harman will also have to appeal to impersonal considerations to get plausible results. If we fail to mitigate, this will lead to many people being harmed, though, we can assume, they will nonetheless have lives that are on the whole worth living. And our failure to mitigate is a necessary condition of their coming into existence in the first place. If, instead of causing different, unharmed people to come into existence, mitigating would somehow cause no future people to come into existence at all, then it seems plausible that failing to mitigate would be permissible despite the fact that it would involve causing harm to future people. Their complaint against our harming them would not, in this case, be strong enough to ground the conclusion that our failing to mitigate would be wrong. When we have the alternative of mitigating and thereby causing unharmed people to exist rather than harmed people, on the other hand, it seems clearly wrong to refuse to mitigate. Any adequate account of the conditions that make refusing to mitigate wrong in this case must refer to the fact that we have an alternative available to us that is morally better in some respect, but not better for any particular people.
Harman’s account of harm, in combination with her view that harm has a special moral significance that makes causing it especially difficult to justify in comparison with, for example, refraining from providing additional benefits that one could have provided, also has some implications in non-identity cases that are difficult to accept. For example, consider a modified version of Tammy’s case in which it is, for whatever reason, predictable that the deaf child that she could have would have a somewhat higher quality of life than the hearing child that she could have instead, though both would have quite high qualities of life. Since on her view deafness is a harm, and the fact that an action would cause harm is a strong reason against performing that action, it would require considerations of great moral weight to justify having the deaf child rather than the hearing one. Certainly the fact that the deaf child would have a somewhat higher quality of life could not, on Harman’s view, count as a significant enough reason to justify having the deaf child, since doing so would impose a significant harm on that child (despite also providing the benefits of the high quality of life). But this is a rather implausible result. This might lead us to question Harman’s account of harm, but, perhaps more plausibly, it suggests that the fact that an action would cause harm to a person has significantly less weight in cases in which the action is a necessary condition of the person’s coming into existence than it has in ordinary cases, at least so long as the person would have a life that is on the whole worth living.Footnote 21 If this is right, then it is even more difficult to see how the complaints that those harmed by actions that were necessary conditions of their coming into existence could lodge against those actions could be significant enough to explain the wrongness of, for example, our failing to mitigate.
4. Reconsidering low-threshold rights
In light of the failure of appeals to high-threshold rights and harm-based considerations to allow us to explain what we take to be our mitigation obligations without committing ourselves to implausible views about, for example, the permissibility of bringing people with lives well worth living into existence when the alternative is bringing no one into existence, those who wish to reject appeals to impersonal considerations may suggest that we reconsider the objection made earlier to something like a low-threshold account of human rights as the basis of our mitigation obligations. Perhaps, they might suggest, we ought to accept that those obligations are much less extensive than we tend to suppose. They may argue, that is, that since we cannot accept that those who have lives that are well worth living have legitimate complaints, on their own behalf, against actions that were necessary conditions of their coming to exist, we should accept that those who perform actions that foreseeably cause people to exist who are worse off, even very much worse off, than different people whom they could have caused to exist instead would have been, do nothing wrong. Boonin (Reference Boonin2008, Reference Boonin2014) defends this view by arguing that we should reject the intuition that it is wrong to refuse to take even minimally costly precautions in order to ensure that a healthy child comes into existence rather than a different, worse off child with a genetic condition that is significant, but does not make the lives of those who have it so bad that they are not worth living (Reference Boonin2008, 147–154). On his view, then, not only is it permissible for Patty (of whom it is the case, recall, that any child she has will be deaf) to have a deaf child, but it is also permissible for Tammy to deliberately choose to have a deaf child rather than a different, healthy child, despite the fact that the healthy child would have a higher quality of life.
Boonin claims that we should reject the intuition that it would be wrong for Tammy to choose to have the deaf rather than the healthy child, since there are reasons to think that it is unreliable. Specifically, he argues that when we reflect on such cases, we will tend to fail to keep the fact that this choice does not make any particular person worse off than she would otherwise have been properly in view (Reference Boonin2008, 147–149). In other words, we will tend to assimilate cases like Tammy’s to more typical cases in which one’s actions do not determine which people will come to exist, and therefore will be pulled, implicitly, by the thought that Tammy’s choice to have the deaf child would make that child worse off than she otherwise would have been, to conclude that this choice would be wrong. Since our tendency to have this intuition is, according to Boonin, explained by our failure to properly attend to a relevant fact about the case, we have at least some reason to be skeptical of the intuition.
Boonin then argues that when we consider cases that are similar in morally relevant respects, but are not such that we will be liable to being affected by the implicit and mistaken thought that the person who is caused to exist has been made worse off than she otherwise would have been, our intuitions will provide support for the view that it is not wrong to cause a worse off person to exist rather than a different, better off person (Reference Boonin2008, 149–154). In the first case that Boonin offers in support of this claim, Fred can save either Billy or Timmy from drowning, and cannot save both of them. Billy is blind, while Timmy is not (and presumably as a result, Billy’s quality of life is lower than Timmy’s), and it would be slightly more inconvenient for Fred to save Timmy than it would be for him to save Billy (Reference Boonin2008, 150). Boonin suggests that, intuitively, it is permissible for Fred to save Billy, despite the fact that by doing so, he causes a blind person to exist rather than a sighted person (Reference Boonin2008, 152).
Even if we accept this intuition,Footnote 22 however, Boonin’s argument succeeds only if he is correct that there are no morally relevant differences between Fred’s choice to rescue Billy and Tammy’s choice, in a slight variant of Harman’s case, to have a blind child rather than a healthy child. After noting all of the ways in which the cases are analogous (e.g. both Fred and Tammy have a choice between making it the case that a blind person exists and making it the case that a different, sighted person exists), Boonin notes that ‘[i]t is true, of course, that [Tammy’s] choice involves choosing which of two lives to create while Fred’s involves choosing which of two lives to extend,’ but claims that ‘it is difficult to see how this difference in itself could matter morally’ (Reference Boonin2008, 150).
There are, however, at least two plausible moral bases for thinking that the fact that Billy already exists makes it the case that Fred is permitted to save him rather than Timmy, which would not also support the view that Tammy is permitted to choose to have a blind child rather than a sighted child. First, we might endorse an egalitarian or prioritarian principle according to which the fact that Billy has, up to the point at which both his and Timmy’s lives are at stake, had a worse life overall than Timmy, provides a significant reason to prefer giving the benefit of additional years of life to Billy. Such a principle would not, however, provide support for choosing to have a blind child rather than a different, sighted child, since merely possible people have no past or current quality of life that could provide the basis for a verdict under the principle.Footnote 23 Second, we might endorse a principle of equal respect for persons, or, similarly, a principle according to which all persons (or at least those with lives worth living) have equal moral status. Many plausible versions of such principles will imply that we ought not make decisions about, for example, whom to save from death, on the basis of facts about who would enjoy a higher quality of life going forward if saved.Footnote 24 These principles would not, however, provide us with any reason to think that it is permissible to choose to have a blind child rather than a different, sighted child, despite the fact that the sighted child would have a better life, since merely possible people do not already have an equal moral status that must be respected.
There are, then, plausible grounds for rejecting Boonin’s claim that Fred’s circumstances are, in all morally relevant respects, analogous to Tammy’s. But even if Boonin were correct about this, there would remain significant reasons to reject his view. Consider, for example, what accepting that view would commit us to saying about our mitigation obligations in the following case:
Delayed Climate Change: We have to decide between a policy of mitigation and a policy of continuing to emit at high levels. We would have a somewhat lower quality of life under a policy of mitigation than we would under a policy of continued high emissions, but our lives under mitigation would still be vastly better than lives that are just worth living. Our choice will have no effects on the quality of people’s lives for the next 200 years (though it will affect which particular people come to exist). If we choose mitigation, then 200 years from now, and for many centuries thereafter, the people who would live will have an extremely high quality of life. If we choose to continue emitting at high levels, on the other hand, then 200 years from now, and for many centuries thereafter, the different people who would live will have lives that are just slightly above the threshold at which we could plausibly think that their low-threshold human rights would be violated by actions that are necessary conditions of their coming into existence.
We may agree with Boonin that the intuition that it would be wrong for Tammy to choose to have a blind child rather than a different, sighted child, is not by itself a decisive reason to reject his view. But if we accept that view, then we are committed to accepting that it would be permissible to choose to continue to emit at high levels in Delayed Climate Change as well.Footnote 25 We are committed to accepting, that is, that it is permissible to do what would cause billions of people to come to exist with lives that are just above a low threshold, rather than doing what would cause billions of different people to come to exist who would have an extremely high quality of life. Indeed, since choosing mitigation would be somewhat costly for us, it appears as though Boonin is committed to thinking that we would be obligated to choose to continue to emit at high levels. After all, a policy of mitigation would be worse for particular people, namely us, while a policy of continuing to emit at high levels would be worse for no one.
This implication, it seems to me, provides us with a strong reason to reject Boonin’s view. But if we reject appeals to impersonal factors, or the somewhat weaker view that our mitigation obligations can be at least primarily grounded in such factors, then we have little alternative to accepting Boonin’s view. The failure of the high-threshold human rights and harm-based accounts of what most of us take to be our mitigation obligations, then, leaves those inclined to reject primarily impersonal explanations of those obligations with the choice between, on the one hand, giving up their opposition to primarily impersonal explanations, and on the other, accepting that our obligations are much more limited than we tend to think they are. It is difficult to believe, however, that many would find it more troubling to allow impersonal considerations a prominent role in explaining our mitigation obligations, than to allow that we have no such obligations beyond ensuring that future people’s lives will meet a rather modest threshold.
Acknowledgments
I am grateful to audiences at the 2015 Association for Social and Political Philosophy Conference, the 2015 Society for Applied Philosophy Conference, and the 2016 Inland Northwest Philosophy Conference. Amy Sepinwall provided helpful written comments along with valuable discussion. I also benefitted from discussions with Bert Baumgartner, Megan Blomfeld, Will Braynen, Vince Buccola, Mark Budolfson, Joe Campbell, Peter Conti-Brown, Nico Cornell, Blake Francis, Molly Gardner, Steve Gardiner, Michael Goldsby, Mark Goodwin, Gwen Gordon, Lisa Herzog, Rob Hughes, Hyunseop Kim, Alex Levitov, Andrew Light, Sarah Light, Brad McHose, Kieran Oberman, Eric Orts, Julie Rose, Debra Satz, Emma Saunders-Hastings, Liam Shields, Patrick Taylor Smith, and Tim Weidel.
