What constitutes proof? How certain do you need to be before you take a risky decision? How do you weigh up the evidence in a situation where both action and inaction carry unquantified and unquantifiable risks?
It depends. In a criminal trial the burden of proof is ‘beyond reasonable doubt’ – this is often interpreted using Blackstone’s formulation – ‘better ten guilty men escape than that one innocent suffer’ – implying a requirement of 90% certainty. In civil cases it is the ‘balance of probabilities’ – or 50%. Most scientific trials seek confidence at the 95% or 99% level, while in order to claim the discovery of a new sub-atomic particle requires a five-sigma level of certainty – 0.57 parts per million – or greater than 99.9999% confidence that the result did not occur by chance.
So what level of proof will be required for the deployment of a geoengineering technology? Nobody knows, but what we do know is that the basis on which decisions are taken in the real world are often far from the rational evaluation of the evidence that we might wish.
Would you be willing to be the first human subject for a new treatment to cure an otherwise fatal disease? The answer will depend on the circumstances – if you were fit and healthy, you would be highly unlikely to agree, while if you were seriously ill you might feel you had nothing to lose.
While we hope that geoengineering is not necessary – that efforts to curb emissions are successful and that the response of climate to greenhouse forcing is not worse than anticipated – it is prudent to be prepared against these possibilities. And that preparation involves several aspects:
Failure to undertake research due to an understandable squeamishness about the hubristic nature of the field would deprive future policymakers of evidence that they might require to make decisions. If we reach a situation where the effects of climate change are causing widespread disruption then the pressure to ‘do something’ will become intense. Better to know which techniques we really should not be considering than risk deploying an under-researched technique. The obverse risk is that lack of research could leave a technique that could be employed without creating countervailing side-effects unused.
The question is sometimes asked ‘how could we reach agreement on the deployment of geoengineering technology if we can’t agree to reduce emissions?’ The situations are different. It is the nature both of the human condition and our political processes to think in short time-frames. But when a crisis erupts, action can be swift as was seen during the bail-out of the financial system – whether or not that was the correct course of action is not the subject for this blog - but it would be helpful to create an anticipatory governance framework for the eventuality of geoengineering deployment being considered
But surely all this preparatory work – the research, the governance framework – points to a desire to deploy geoengineering? Am I an ‘axe-handle’ – not the perpetrator of violence, but an enabler of such violence? I would argue most strongly that I am not. Research can help us exclude those techniques that would be too risky to deploy and a governance framework thought through ahead of a crisis sounds to me like a better idea than no governance framework in place when the situation requires.
There are those who will put forward geoengineering as a ‘silver bullet’ or a ‘get out of jail free card’ and cite geoengineering as a reason not to mitigate emissions. This is wrong, both factually and morally. The evidence that geoengineering techniques could mitigate the effects of climate change is scant and to slow down reductions of emissions on such a basis is morally reprehensible – it would be as if a doctor hailed early-stage research into a drug to cure cancer as a reason not to quit smoking – premature and reckless.
Instead we need appropriate messaging – research into geoengineering is not an excuse to reduce the pressure to cut greenhouse gas emissions, and geoengineering research and development of governance frameworks does not presuppose a desire to deploy.
We can hope that we never need to employ geoengineering, but we may have to. It would be a mark of our failure as a species. Perhaps part of our squeamishness about this subject is because it confronts us with that reality. But should we arrive at such a sorry pass, where the risks associated with climate change are considered so severe that we would consider taking a step into the different unknown of geoengineering, it would be better to be as prepared as we possibly can be.