Monday, March 1, 2010
Geoenigneering and risk: more
Let us assume, though limited experimentation, we can be confident that we can effect an overall cooling of the plant by sulfur injection – we just don’t know what the risks of such planetary insertion are. Insertion might buy us 50 years to develop other technologies to deal with the underlying problem. We also don’t know the risk of climate warming which we could avoid by sulfur insertion – at least for those 50 years. (Assume here what most climatologists take to be the case: these are our likely choices, even if we ceased all carbon output today, because of warming already built into the climate system.) So here is a choice in which we are ignorant on both alternatives when it comes to some worst case scenarios. Why isn’t the rational choice to bracket these and make the decision based on other considerations? Am I evincing a prejudice against advertent intervention over inadvertent intervention? I think there is a different consideration that makes my argument non-prejudicial. Assume global warming carries some risk known only to God as it were. Geoengineering is an imperfect offset for that warming at best. For example, it does not counter ocean acidification. As such it is a mistake to treat the choice as an exclusive one between climate change and geoengineering. We can’t rule out that geoengineering might make things worse (at least locally) than they would be with climate change alone. Still, we make choices in everyday life in the face of worst case scenarios that loom in the background. Crossing the street, I may be run down. The plane I ride on may crash. The food I consume may be fatally contaminated. But these are cases in which we can be crudely guided by expected utility. (“Crudely” because as behavioral psychology has show, we discount the future more than we should, we overvalue the familiar, and so on.) Courting death may make sense (for me at least) when I chance giving up longevity for a life of gluttony given my preferences. But what about when I can’t assign probabilities to such outcomes? You are dying, nothing more can be done for you as things stand. The doctor comes in with news – there is a new idea. We have no sense whether it will help or hurt. Should you try it? Why not, you have nothing to lose. But of course, that is only because all is lost. But that is not the risk posed by climate change. Things are (possibly) very bad. But we don’t face extinction (in the short term). The parallel here is that you are very sick with an unknown prognosis and no treatment alternatives. The doctor comes in again with the same offer. If you are a gambler by preference perhaps you should double up on risk. Or perhaps you should toss a coin. But the problem of geoengineering is one level more complex than this. For all the unknowns concerning risks, there is some reason to believe that those risks, if they exist are not going to be equally spread throughout the planet. For example, one of the unknowns about sulfur insertion is its effect on participation patterns as opposed to temperature. Hence those populations that are vulnerable to precipitation stress are more at risk than other if there is indeed risk. A lottery in which (we hope) most will benefit and some will be hurt is one thing when everyone has an equal chance. But without that an equal chance, incentives for those less likely to benefit to agree to such a lottery decline.