For these and related reasons, in the last two decades geoengineering proposals have played a marginal role in discussions of climate change policy. But last year the scientist Paul Crutzen, a Nobel Laureate, reignited the debate by suggesting that we ought to embrace an ambitious geoengineering research program. Crutzen's favoured proposal was that we should explore the possibility of injecting sulphur particles into the stratosphere in order to suppress global warming by simulating volcanic eruptions. One reason that this suggestion has gained some traction is, I think, that unlike some previous advocates for geoengineering, Crutzen appears to take seriously both the gravity of our current environmental problems and people's initial reservations about geoengineering. His main argument goes as follows.

To begin with, Crutzen acknowledges that the main focus of climate change policy should be mitigation and adaptation. So, he is not any kind of sceptic. But he goes on to claim (plausibly) that, so far, progress on mitigation and adaptation has been slow and minimal, and that there is good reason to fear that this situation may continue. (Elsewhere I call this "the problem of political inertia"). Given this inertia, Crutzen proposes that we must take seriously the possibility that substantial mitigation and adaptation will not occur, and that this may mean that by the middle of the century the planet faces the imminent prospect of runaway climate change.

In such a scenario, Crutzen thinks, we will be forced to choose between allowing catastrophic impacts to occur, or engaging in geoengineering. Both of these, he admits, are bad options. But, he asserts, engaging in geoengineering is less bad than allowing catastrophic climate change. Therefore, if we end up facing the choice between the two, we should choose geoengineering. But here's the rub. Unfortunately, if we do not start doing serious research on geoengineering now, then we will not later be in a position to choose the geoengineering option should the unfortunate scenario just described actually arise. Therefore, Crutzen concludes, we should start doing serious research on geoengineering now.

In summary, Crutzen's argument is that geoengineering, though arguably an evil in itself, might turn out to be a lesser evil than the likely alternative; hence, he thinks, we should prepare just in case we are compelled to endorse that evil. This is a familiar and straightforward kind of moral argument. Moreover, arguments of this form have wide currency in the geoengineering debate, appealing to both enthusiasts for and detractors of the approach. Still, it is worth raising some concerns.

To begin with, arguments from moral emergency are perennially popular in both private and public life and for an obvious reason. Clearly, part of the point of claiming that one is in morally exceptional circumstances is in order to secure an exemption from the usual norms and constraints of morality. But this fact should give us pause. After all, there will always be those who would prefer that morality not apply to them or their projects, and all of us are vulnerable to such thoughts at some time or other. Morality sometimes seems inconvenient to us (like truth, as Al Gore reminds us) - and in such cases we'd often like to have an exemption.

In short, we should be wary of arguments from emergency. Clearly, they are open to manipulation. (This helps to explain why arguments from emergency - and declarations of states of emergency when normal political processes and rights are suspended - are so often employed by political despots.) So, such arguments should be subjected to special scrutiny.

What would "special scrutiny" amount to in this case? The first, and obvious, point to make is that we should not simply accept as a stipulation that some policy that is said to be an evil (like geoengineering) should be endorsed because it is a lesser evil than some other policy (such as catastrophic climate change). Instead, we should ask important questions, such as: how likely is the emergency situation (where one has actually to decide between these two options) to arise? Is it true that these are the only alternatives? Is the lesser evil really lesser, all-things-considered?

Now, as it happens, the answers to these questions seem very much in doubt in the geoengineering case. For example, for a decision-maker actually to be considering deployment of geoengineering in the situation Crutzen envisages, that decision-maker would seem to need to know that the planet was on the verge of very serious climate impacts, that geoengineering was very likely to - and the only thing likely to - prevent them, and that the side-effects of geoengineering would be minor in relation to the harm prevented. But this would be a pretty unusual scenario, and it is not clear how likely it is to arise.

Them, not us
More importantly, this is not the situation that Crutzen claims that we actually face. He says only that we must prepare for the possibility of an emergency, not that we are actually in one. This makes at least two crucial differences.

The first is that talk of the choices that "we" might face obscures the fact that if Crutzen's scenario does arise, the people facing the tough choices won't be us. Crutzen is thinking of the threat of runaway climate change emerging fifty years or more down the track, and so confronting future generations. This is important, since some people (myself included) believe that one cause of political inertia, and a big part of the moral problem more generally, is that climate change allows the current generation to unfairly pass on its costs to future people.

The second difference is that one of the usual effects of actually being in an emergency is to make considerations of how the situation arose much less salient. (For example, if someone sees a small child drowning in a pond, we do not usually think it appropriate for them to stop to ponder how it came to be there. The relevant question is what to do here and now.) But this is not the case if one is anticipating an emergency. Then it is perfectly appropriate to consider how the emergency might arise.

First, it may be that the best way to plan for an emergency is to prevent its arising. In that case, the fact that we would choose a (lesser) evil if it did arise is usually not of much relevance to what to do now. Moreover, there is a real worry that Crutzen's proposal is not neutral here. It is not silly to think that substantial investment in geoengineering will itself encourage political inertia on mitigation and adaptation, and also facilitate the actual deployment of geoengineering "solutions". In short, Crutzen treats the decision to do research and the decision to deploy as if they were causally isolated. But it is not clear what justifies this assumption - indeed, the history of technological innovation suggests otherwise.

Second, the fact that we can prepare allows us to consider putting other options on the table. Crutzen is implicitly suggesting that the best we can do now to help future people faced with catastrophic climate change is to research geoengineering. But presumably there are other possibilities. For example, we could invest in alternative energy in order to prepare future people for a massive emergency deployment of alternative energy sources (e.g. by establishing a "Strategic Solar Panel Reserve"), or we could establish a massive international climate assistance and refugee programme, or we could do both of these things, together with any number of other alternatives.

Third, there seems to be an important difference between preparing for an emergency and preparing for an emergency that is to be brought on by one's own moral failure. Crutzen's idea is that we ought to pursue mitigation and adaptation, but we probably won't; therefore, he says, we should research geoengineering. Now, many things might be said about this. But here let me make just one remark: if we put someone in a very bad situation through our own bad actions, usually we don't think it enough to respond merely by offering them an evil way out. Instead, it seems that we have substantial obligations not only to help them to find better alternatives but also, if none is to be found (or if there is a substantial risk of this), to find a way of compensating them for our failure. If this is right, then even if Crutzen's argument were correct in other respects, we should not conclude from it that we owe future generations only research on geoengineering; much more seems to be required (e.g. a very substantial compensation fund).

This brings us nicely back to the issue of political inertia. For it seems likely that the same forces that oppose substantial mitigation and adaptation measures would also oppose substantial compensation proposals (and perhaps a huge investment in geoengineering too, if that were being proposed); and this brings a larger issue into focus. Crutzen's claim is that research on geoengineering acts as a kind of insurance policy. But there are many such policies; and there is a real concern that the narrow one of "Geoengineering Research Only" gains prominence among them only because it is the one that seems most congenial to us, the present generation. In short, perhaps we'd be happy to spend a few million dollars researching technology our generation won't have to bear the risks of implementing, and even happier to think that in doing so we were making a morally serious choice in favour of protecting future generations. But thinking so does not make it the case. The worry is that Crutzen's proposal - however well-intentioned - is at least as likely to facilitate the problems of political inertia and intergenerational moral corruption as to solve them. Surely we can, and ought, to do better.