1. Introduction
What is the practical relevance of the Non-Identity Problem (NIP) for our climate change-related duties?Footnote 1 Does it make any difference to what we should do about one of the most pressing issues of our time?Footnote 2 If so, how extensive is that difference? Climate change and the NIP are often discussed together. When the NIP is discussed in more general terms, climate change is by now one of the most often mentioned applications.Footnote 3 Furthermore, works with a focus on climate ethics will often feature sections or chapters on the NIP.Footnote 4 And perhaps most importantly, there are a number of papers whose primary purpose is to investigate the connection between climate change and the NIP.Footnote 5 Given this wealth of material, it is surprising that there is very little explicit engagement with the questions raised above. Climate change is often used as a mere backdrop to discuss the NIP in more general terms, without going into the specifics of our current real-world situation. These specifics, however, do determine the practical relevance of the NIP for climate matters. This is why previous papers do little to tell us what practical difference the NIP makes in relation to one of the most pressing real-world problems of our time. This article aims to provide an answer to this question.
My central claim will be that the NIP makes a relatively minor difference to what we should do about climate change. This may seem like an uncontroversial finding, but I will argue that it remains true even if we forgo the two most obvious ways to justify it: switch to a view not threatened by the NIP in the first place or modify one of the views threatened by the NIP. In other words, I argue that even if we stay with a view that is clearly subject to the NIP and concede that the NIP can make a difference to what this view tells us to do, if we ‘bite the bullet’, so to speak, the case of climate change is such that what we should actually do about climate change changes only in relatively minor ways. The key to reaching this conclusion is pointing out the consequences of our climate change-related decisions for people whose existence does not depend on these very decisions.Footnote 6
I begin by briefly outlining the NIP itself, typical strategies for dealing with it, and its place in climate ethics (section 2). My own argument then proceeds in two stages: first, I will give reasons for why the NIP makes at most a minor difference to our climate duties (section 3) before showing why it would be wrong to say that the NIP makes no difference whatsoever to our climate duties (section 4).
2. The Non-Identity Problem and its place in climate ethics
The NIP is arguably the most important puzzle in intergenerational ethics.Footnote 7 To briefly summarize it: some of our acts will cause future people to have lives that are in some sense ‘flawed’.Footnote 8 Perhaps we will cause them to have a bad childhood, or suffer serious ailments, or, to stay with our example, they will live in a world marked by dangerous anthropogenic climate change. However, upon closer inspection it turns out that the same acts that cause these flaws are also necessary preconditions for the existence of these particular people. Had we chosen, for example, a different climate policy, the world would have been changed in such a way that different people meet and procreate or the same people meet but procreate at different moments. Either way, different people would be born. The only chance for the ‘climate change-people’ to be born is the one that includes what mars their lives. As far as our actions are concerned, we could not have provided them with better lives. Still, it seems that at least some of these actions are morally impermissible. Can we still explain why? This is the NIP.
Several strategies to respond to the NIP can be found in the literature. One of them is what Roberts (Reference Roberts and Zalta2020) calls the ‘bite the bullet’ strategy: stay with a view subject to the NIP and accept that the NIP sometimes makes a difference to what this view demands, allows, and forbids us to do. Boonin (Reference Boonin2014) and Heyd (Reference Heyd1992), (Reference Heyd, Roberts and Wasserman2009) are the two most prominent theorists who have argued for this response to the NIP. This article investigates the practical implications of choosing this strategy for the case of climate change. It therefore presupposes the ‘bite the bullet’ strategy rather than arguing for it being our best, or least bad, option. Still, I want to comment on two of the alternative strategies to motivate why choosing to ‘bite the bullet’ is not just an interesting theoretical challenge, but a relevant one given the problems of the alternatives.
One alternative strategy is to opt for one of the so-called ‘impersonal views’. These views reject the idea that what is bad must be worse for someone and can thus remain unbothered by the discovery that some supposedly immoral acts are worse for no one. Classical utilitarianism is the prime example for such a view.Footnote 9 One of the problems with this strategy is that impersonal views tend to arrive at troubling conclusions when applied to choices that determine how many people will come into existence.Footnote 10 The ‘repugnant conclusion’ is arguably the most famous of these.Footnote 11
A second alternative strategy is to stick with an approach subject to the NIP but try to modify it in such a way as to produce plausible implications in NIP-cases. There have been many attempts to modify what is known as ‘narrow person-affecting views’ in such a way as to make them NIP-proof.Footnote 12 Such narrow person-affecting views include, for example, a rights-based view and a harm-based view. For the rest of this article I will, in order to streamline the discussion, focus on the harm-based view. It seems to me that structurally similar considerations to the ones I will develop in relation to the harm-based view can be brought forward for other narrow person-affecting views. There can be many harm-based views but their common core is the idea that very weighty pro tanto reasons speak against harming other people (exceptions like self-defense aside, which need not worry us for the case at hand) and that these reasons frequently, but arguably not always, make harming other people morally impermissible.Footnote 13 Such a view is threatened by the NIP because it seems that in order to harm someone we have to cause that person to be worse-off than they would otherwise be.Footnote 14 As long as the lives of future people are still worth living, this will never be the case for people whose identity depends on us not acting according to our supposed climate duties.
Attempts to modify the harm view have often centered on formulating a ‘non-comparative’ conception of harming.Footnote 15 Elizabeth Harman (Reference Harman, Roberts and Wasserman2009: 139), for example, argues that we harm someone if we cause that person to be in a bad state. Because the conception of harming is non-comparative, we can say that people are harmed by our actions even if there was no way to make them better-off. There are several problems with such non-comparative conceptions of harming. David Boonin (Reference Boonin2014: 73) shows how they might be a bad fit with our linguistic intuitions regarding whether or not somebody is harmed in cases in which we cause somebody to be in a bad state but also confer greater benefits upon that person.
But to my mind the biggest challenge to this strategy is a different one: we can assume that the ‘non-identity people’ we create will, after understanding the precariousness of their existence and as long as their lives are still worth living, welcome the choices we have made.Footnote 16 And as long as this is the case, it seems doubtful whether our actions can be wrong because of the effects on these people, never mind whether we can categorize these actions as instances of us harming these people or not. This is a fundamental problem for all narrow person-affecting views, not just for a harm-based one.Footnote 17 I do not want to claim that it is insurmountable, but it is substantial enough to provide further motivation for the project of this article: if there are major doubts about whether the modification strategy can succeed, it is all the more relevant to investigate the practical implications of ‘biting the bullet’.Footnote 18 This is what I will do for the case of climate change.
Before I do so I want to briefly provide some context regarding the way the NIP is dealt with in the climate ethics literature and outline what separates my project from previous ones. Page (Reference Page, Kim and Gene Frankland1999), for example, follows the strategy of modifying a narrow person-affecting view. The question whether the NIP makes a practical difference to our climate duties if we ‘bite the bullet’ is therefore never tackled. McMahan (Reference McMahan2021), Nedevska (Reference Nedevska2019), and Woollard (Reference Woollard2012) explicitly exclude the effects of our climate action on fixed-identity people from their focus and they can therefore be understood as talking about a counterfactual version of climate change or about just one aspect of real-world climate change (its effects on a sub-class of future people) instead of the whole phenomenon.Footnote 19
Broome (Reference Broome2012) is the most interesting case. He acknowledges the effects of our climate change-related decisions on fixed-identity people but nevertheless goes on to claim that the NIP makes a difference to our climate duties.Footnote 20 In particular he argues that the public morality in relation to climate change (roughly speaking the collective duties that are the subject of this article) is marked more by duties of goodness rather than by duties of justice at least in part because of the NIP.Footnote 21 What remains unclear, however, is why the effects of our actions on fixed-identity people do not make it the case that justice-based justification strategies, like the harm-view that Broome favors, are sufficient to justify the full range of our collective climate duties. Broome acknowledges that some duties of justice are explained by the effects of our actions on fixed-identity people but it remains unexplained why these effects cannot make it the case that the NIP makes no difference at all to what we should do about climate change.Footnote 22 The question regarding the practical relevance of the NIP for our real-world climate duties therefore remains unanswered.
3. The Minor Difference Claim
I will now present two claims regarding the practical relevance of the NIP for our climate change-related duties. I endorse the first, the Minor Difference Claim, which I will discuss in this section, and reject the second, the No Difference Claim, which will be the subject of the next section.
The Minor Difference Claim says that the NIP makes at most a minor difference to our climate duties. How should we understand this claim? Two clarificatory notes are in order. First, ‘our climate duties’ is meant in a practical way as in ‘the things we should actually do about climate change’ rather than in a theoretical way. It is immediately conceded that the NIP makes a great difference to the theoretical picture of our climate duties. For example, if it weren't for the NIP, we would surely have the duty not to harm distant future generations by means of climate change. The Minor Difference Claim simply states that this potentially vast difference to our ‘theoretical climate duties’ makes at most a minor difference to our ‘practical climate duties’, that is, the things we should actually do about climate change. If I continue to talk about ‘our climate duties’, this is the sense I have in mind.Footnote 23
Second, ‘our climate duties’ is a very succinct term for what is potentially a large number of duties. To investigate whether the NIP makes a practical difference to every single one of our climate duties, and if so, how extensive that difference is, is impossible. What is needed is a more general description that captures the big picture of what our supposed climate duties are. Luckily, the climate ethics literature has indeed produced something close to a consensus view regarding the general extent and content of these duties.Footnote 24 It can be expressed as the following call for ‘robust climate action’:
Robust Climate Action Claim
We have a moral duty to pursue robust climate action. This must include extensive mitigation, adaptation, and compensation efforts. Perhaps most importantly, our greenhouse gas (GHG) emissions must quickly be reduced and reach net-zero during the next decades.
The claim is painted in broad strokes, only the mitigation-related duties are spelled out in more detail. This is necessary because a more concrete version runs the risk of being too specific to capture the consensus at the heart of the debate. I do contend that the claim is specific enough to be far from vacuous. Since climate ethics is a theoretically diverse field, I have formulated it in such a way as to be compatible with different lines of justification. My exclusive focus on a harm-based view and its ability to justify the Robust Climate Action Claim should not make us think that other theoretical approaches cannot be used to justify it.
I take the following to be a sufficient condition for the NIP to make at most a minor difference to our climate duties: if we can justify the Robust Climate Action Claim while ‘biting the bullet’, that is, while arguing based on a view subject to the NIP that we do not modify in order to accommodate the NIP, the NIP makes at most a minor difference to our climate duties. I will now argue how one particular unmodified justification subject to the NIP, namely the harm-based view sketched above, can still justify the Robust Climate Action Claim even if we ignore the harm we supposedly do to future people whose existence depends on us not undertaking robust climate action.
The positive case for why the harm-based view can justify the Robust Climate Action Claim even under non-identity conditions is straightforward: our real-world situation is such that failing to undertake robust climate action will massively harm a great number of people whose identities are in the relevant sense ‘fixed’, either because they are already born or because their identities do not depend on the decision for or against robust climate action. Because this is the case, a harm-based view will tell us to undertake robust climate action irrespective of the harm we supposedly do to people whose identity is not in the relevant sense fixed. In what follows I will address a number of reasons to doubt this claim and argue that none of them are convincing.
3.1. Objection 1
First, one could argue that dangerous climate change is only to be expected at some point in the future when harmful effects on fixed-identity people are no longer to be expected. I think this argument has by now become something of a non-starter. As anyone who has followed the news in recent years is no doubt aware, dangerous – and, to cut the euphemism, deadly – climate change is already here. It can get much worse than it currently is, but if there ever was such a thing as non-dangerous climate change, we are past it. I only mention this objection because the phrase ‘preventing dangerous climate change’, which dates back to the UNFCCC Convention of 1992 and which indicates that dangerous climate change is a merely possible future state of affairs, is still used in climate ethics.Footnote 25
3.2 Objection 2
Second, one could argue that dangerous climate change may be happening now, but even robust climate action would only start to affect the progress and the impacts of dangerous climate change with some delay. We can only change the further future, when non-identity will be the norm, the argument goes. In response to this objection, it is first worth pointing out how long it will take until we have reached conditions of complete non-identity even if we focus only on the people whose identity is fixed in the most straightforward of senses: those already born. Two out of three children born in 2020 will live to see the twenty-second century. The average girl born in Singapore today is expected to die in 2117. There are people alive now who might still be alive in the 2140s.Footnote 26 Even those effects of our present actions that are delayed by decades are thus highly likely to have effects on fixed-identity people. Furthermore, the objection is clearly unsuccessful for adaptation and compensation efforts. Dykes, air-conditioning systems, and cash payments have an immediate effect. One could think that duties to help with adaptation in particular cannot be justified by a harm-based view in the first place since these are duties of beneficence, but this would be wrong. The duty not to harm plausibly includes duties to aid with minimizing the negative consequences of the harmful processes we have played a major part in causing.
However, the Robust Climate Action Claim also outlines our mitigation duties and here the objection is not as hopeless. The delayed effect of emissions reductions has been emphasized in the climate ethics literature. Darrel Moellendorf, for example, states that ‘[d]ue to inertia in the climate system and long residence of time of CO2 in the atmosphere, mitigation policies will most benefit the very young who are now alive and people who will exist in the future’.Footnote 27 And Dominic Roser and Christian Seidel write that ‘the effects of many greenhouse gases are time-delayed’ and that our mitigation measures ‘do not necessarily affect those who are alive today, but above all affect future generations’.Footnote 28 But this is only half-true. Due to climate inertia and the long atmospheric lifetimes of some GHGs, emissions continue to induce climate change decades, centuries, perhaps even millennia later.Footnote 29 But GHGs are not time bombs that go off only long after they are emitted. They immediately contribute to a worsening of climate change once they are emitted into the atmosphere.Footnote 30 Hence it would be wrong to conclude that 2021s emissions are solely responsible for the harms that climate change brings about in, say, 2025. But how much we go on to emit plays an important part in determining how much warming we must expect in the next years and decades.Footnote 31 Our emissions can thus be considered harmful soon after they are emitted.
3.3 Objection 3
The third objection is the one I consider to be the most powerful: it concedes that dangerous climate change is happening and that our decisions regarding whether or not to undertake robust climate action, including mitigation, will make a difference for whether or not we harm fixed-identity people. But it goes on to claim that the extent of these harms is not sufficient to justify the extent of climate action roughly sketched in the Robust Climate Action Claim. If we abstract from non-identity cases, the harm we do is simply too insignificant to justify such an extensive response. It would perhaps be laudable to undertake it, but to demand it, like the Robust Climate Action Claim does, would be overly demanding.
One of the reasons why this objection is hard to refute is that it plays on two imprecisions in the argument. The Robust Climate Action Claim is not fully precise. It cannot be if it is to cover the consensus in climate ethics. And the harm-based view is not fully precise, either. How strongly does the harmful character of our actions speak against them? This imprecision, too, cannot easily be eliminated. An extreme position that says that it is all-things-considered always impermissible to harm others is implausible and to pick one more moderate position would be arbitrary. These imprecisions mean that my response to the third objection cannot be as definitive as the last two responses (hopefully) were. Still, I think there are good reasons why a harm-based perspective can justify the extent of our climate duties sketched in the Robust Climate Action Claim.
There are two reasons in particular that I want to highlight: first, it is widely thought that duties not to harm people are overly demanding only under extreme conditions, if ever. Second, if there are cases in which not harming people is overly demanding, there are good reasons to think that our situation regarding robust climate action is not one of them.
First, there is a reason why demandingness objections are usually raised against duties of beneficence. Helping others might sometimes be overly demanding, but we usually think that demandingness objections carry far less power (if any) when applied to instances of omitting to harm. David Sobel (Reference Sobel2007: 4), for example, argues that it is morally impermissible to harm someone by taking their kidney against their will even if that is the only way to save one's life. He states that ‘those who champion the Demandingness Objection overwhelmingly believe that morality could ask so much’.Footnote 32 The climate harm we cause is different from such paradigmatic cases because it is collectively caused, but I don't see why the basic idea that demandingness objections carry less weight if applied to cases of omitting harm should not apply here as well.Footnote 33
Second, even if there are cases in which not harming people is overly demanding, there are good reasons to think that our situation regarding robust climate action is not one of them. The first reason is that failing to undertake robust climate action would not only cause minor harms to a few fixed-identity people, but would cause major harms, including lethal ones, to a great many fixed-identity people. Whether or not omitting to harm is overly demanding surely depends not just on the demands placed upon the harming parties but also on the severity and the extent of the harm itself. And the harm we would cause by failing to act in accordance with the Robust Climate Action Claim would be immense. John Nolt (Reference Nolt, Murphy, Gardoni and McKim2018: 46) has recently compiled a few of the more prominent attempts to estimate the annual fatalities of climate change and they are within the same order of magnitude:
• Global Humanitarian Forum (2009): 300,000 additional deaths every year, estimate concerns the present and is from 2009
• Development Assistant Research Associates (2012): 400,000 additional deaths every year, estimate concerns the present and is from 2012, 700,000 per year by 2030
• WHO (2018): 250,000 additional deaths every year in the 2030–50 period
• Springman et al. (Reference Springman2016): 500,000 additional deaths every year from 2050 onward from reduced food availability alone
As Nolt notes, these numbers and the differences between them should be treated with caution and there is reason to believe that they underestimate the fatalities rather than overestimate them.Footnote 34 Given these numbers we can expect climate change to threaten the lives of dozens of millions of people, perhaps even those of 100 million people or more, during the twenty-first century alone. Furthermore, given the all-encompassing character of climate change, we can expect that for every person who dies because of climate change, there are many who are otherwise harmed, often severely so.
It is important to keep in mind that not all of these potential harms are caused by us, not all of them can still be prevented, and not all of them will befall people whose identities are in the relevant sense fixed. However, the real-world situation is such that more than half of the GHG emissions since industrialization began were emitted after 1990.Footnote 35 They can thus (very roughly) be attributed to the current generation. The harm of climate change is not caused by us alone, but we have played a very major part in causing it. Regarding our means of prevention: it is true that it is unpreventable that there will be climate harms years and even decades from now. But this is completely compatible with the view that our means of mitigation, adaptation, and compensation can make a great difference to the lives of very many people. Luckily, I see no reason to doubt this claim. When it comes to climate action, it is both ‘too late’ and ‘definitely not too late’.
And, finally, 2.7 billion people have been born since the year 2000,Footnote 36 many of them in poor societies that will be disproportionately affected by climate change.Footnote 37 There will be billions of fixed-identity people around for decades to come who might be harmed by our failure to pursue robust climate action even if we ignore all future people who would have been born irrespective of whether or not we undertake robust climate action. To sum up, these complications only carry so far. They do not decisively alter the verdict that the fixed-identity, preventable climate harm we would cause would be immense. Its extent is truly catastrophic. If we really believe that not harming others should play an important role in our moral thinking, I do not see how omitting to cause this kind of harm could be overly demanding.
This gets even clearer if we recognize that the Robust Climate Action Claim is not that demanding in the first place. Acting in accordance with it need not result in climate policies that are so demanding as to cost people their lives. It does not, for example, demand a sudden end to all GHG emissions from fossil fuels, which would wreak havoc, but only to reach net-zero emissions at some point during the next decades. If anything can make not harming many millions of people overly demanding, it would surely be the fact that the survival of very many people depends upon it. Perhaps such ‘suicidal climate action’ would be overly demanding. But the Robust Climate Action Claim simply does not demand that much.
Taken together, these reasons refute the third objection. We can therefore conclude that none of the objections succeed. Since I see no further objections that stand a better chance of succeeding, I conclude that the harm-based view can in fact justify the Robust Climate Action Claim. This in turn is a sufficient condition for the NIP making at most a minor difference to our climate duties. The Minor Difference Claim is true.Footnote 38
4. The No Difference Claim
The Minor Difference Claim leaves it open whether the NIP makes a minor difference or no difference to our climate duties if we ‘bite the bullet’. In this section I will argue that the former is the case by rejecting the No Difference Claim. The No Difference Claim consists of the proposition that the NIP never makes any difference to our climate duties. This claim is false. Consider our duties to adapt to climate change: if, for example, we must decide how high a new dyke should be, the NIP can conceivably make a difference. Say a certain height is needed in response to storms that will become a real possibility only in the second half of the twenty-second century. And let us, to simplify matters further, say that for some reason today is either the only or by far the best opportunity to build the kind of dyke that will safeguard people even 150 years from now and that our decision will have strong identity-changing effects. If the harms we supposedly do to non-identity people give us reasons to act, we should now opt for the higher dyke.
A harm-based view that focuses only on harms to fixed-identity people will be unable to explain why we should do so. Opting for the lower dyke seems perfectly compatible with acting in accordance with a duty not to harm. Generally speaking, whenever a supposed climate duty can be justified only by pointing to the supposed harm (or, to name another line of reasoning susceptible to the NIP, the supposed rights violations) we do to people whose existence depends on us acting that way, the NIP will make a difference if we stick with the ‘bite the bullet’ strategy. Furthermore, there can be cases where the supposed harm we do to non-identity people is a necessary part of a justification that also includes harms to fixed identity-people.Footnote 39 In these cases, too, it would be true that our duties would be different if it weren't for the supposed harm we do to non-identity people.
Given that the topic of this article is real-world climate change and not climate change-related thought experiments, it is an important question how likely such cases will be in real life. It seems to me that cases like these, while perhaps not common, can't be discarded as unrealistic thought experiments either. The question whether or not to keep nuclear energy as a transitory technology could be an example. Long-term harm to people who would never be born if we phase out nuclear energy earlier could conceivably make a difference to our moral evaluation of this policy choice. Long-term side effects of certain carbon dioxide removal strategies could be another example. The potential risks of storing captured carbon underground that may not materialize for many years come to mind here.Footnote 40
It bears repeating that ‘our (supposed) climate duties’ describes a wide and diverse array of (potential) duties. This article does not aim to investigate the relevance of the NIP for all of them but restricts itself to making two big picture claims: that the NIP makes no more than a minor difference to our overall climate duties insofar as it does not threaten the Robust Climate Action Claim (the Minor Difference Claim) and that we have sufficient reason to reject the claim that the NIP never makes any difference to our climate duties (the No Difference Claim).
5. Conclusion
Deadly climate change affects both the present, the near future, and the further future. Because it affects the further future, the Non-Identity Problem has a role to play in climate ethics. But because climate change affects the present and the near future, too, and in such a catastrophic way, the practical relevance of the Non-Identity Problem for climate ethics is in fact only a minor one. What I take to be the consensual core view of climate ethics, the Robust Climate Action Claim, can be argued for while still relying on the moral perspectives susceptible to the problem. The Non-Identity Problem may be one of the most interesting and theoretically challenging parts of climate ethics, but when it comes to finding out what we actually should do about climate change, it is not of great relevance. One could be forgiven for thinking that this is good news, but in fact the opposite is true. The practical relevance of the Non-Identity Problem for the ethics of climate change is but a minor issue because catastrophic climate change is no longer something that may or may not happen to some unidentifiable people at some point in the further future. It could get much worse still, but catastrophic climate change is already here and will get worse soon.Footnote 41
Competing interests declaration
The author declares that there is no conflict of interest.
