The Royal Society has just published Sir Paul Nurse's annual address to the fellows. His speech this year focused on scientific advice to policymakers, focusing on two issues in particular - climate and genetic modification of crops.
His conclusions are rather good in places:
Scientific advice should be based on the totality of observation and experiment, be based on rational argument, and reflect the consensus views of expert scientists, views which have been rigorously peer reviewed by other independent experts. If there is no strong consensus or if knowledge is still tentative, then these uncertainties should be reflected in the advice.
The idea that scientific advice should be based on evidence and experiment is right up my street. However, the rest of the speech is not so good. Sir Paul rather shoots himself in the foot by harping on about the political views of those who disagree with him, referring darkly to the people who seem to have "political or ideological views that lead them to be unhappy with the actions that would be necessary if global warming is due to human activity". This kind of logical fallacy is a bit rum coming from someone who is lecturing us on rational argument. It's perhaps also a little strange coming from someone who was IIRC a former seller of Socialist Worker.
His views on the consensus are also suspect betraying a worrying ignorance:
The consensus view of the great majority of expert climate scientists is that the globe has increased in temperature by around 0.7 – 0.8oC during the last century, that this is largely due to increased greenhouse gas emissions as a consequence of human activity, and that a further rise of around 2 – 4oC can be expected during the next century.
Given that the attribution of warming "largely" to mankind is based on computer models that have proven to be almost devoid on any predictive ability, I think the suggestion that the extent of attribution can be represented as part of the consensus is therefore just a tad premature. In what way do computer simulations represent "evidence" or "experiment"? Does Sir Paul really think that there is "evidence" for attribution?
The same criticisms apply to predictions of warming. What part do computer models really have to play in informing public policy? As readers here know, the evidence - empirical evidence that is - suggests that climate sensitivity is low. Does Sir Paul agree that we should accept these empirical measurements over the model? And if so, why is he talking of 2-4degree warming? (And, by the by, does he have anything to say about the IPCC's rewriting of the scientific record in this area or is this another of those areas he would rather gloss over?)
Then, to round things off, he has a pop at the arguments of sceptics.
What appears to be happening is that the concerns of those worried about those types of action, have led them to attack the scientific analysis of the majority of climate scientists with scientific arguments that are rather weak and unconvincing, often involving the cherry picking of data.
Here, once again, we see Sir Paul apparently defending Phil Jones and his hiding of the decline. Given that hiding the decline was said by the Russell panel to be "misleading" it is hard to avoid the conclusion that Sir Paul is once again defending scientists who mislead policymakers. This position is doubly culpable given his statement that the scientific advice should reflect the uncertainties.
The Royal Society is not what it was, is it?