Wednesday, October 25, 2006

In defence of consensus

If I were to hope for psychological subtlety from soap operas, or historical accuracy from Dan Brown, I’d have only myself to blame for my pain. So I realize that I am scarcely doing myself any favours by allowing myself to be distressed by scientifically illiterate junk in the financial pages of the Daily Telegraph. I know that. Yet there is a small part of me, no doubt immature, that exclaims “But this is a national newspaper – how can it be printing sheer nonsense?”

To wit: Ruth Lea, director of the Centre for Policy Research, on the unreliability of consensus views. These are, apparently, “frequently very wrong indeed.” The target of this extraordinarily silly diatribe is the consensus on the human role in climate change. We are reminded by Lea that Galileo opposed the ‘consensus’ view. Let’s just note in passing that the invocation of Galileo is the surefire signature of the crank, and move on instead to the blindingly obvious point that Galileo’s ‘heresy’ represented the voice of scientific reason, and the consensus he opposed was a politico-religious defence of vested interests. Rather precisely, one might think, the opposite of the situation in the climate-change ‘consensus.’ (The truth about Galileo is actually a little more complicated – see Galileo in Rome by William Shea and Mariano Artigas – but this will do for now.)

In any case, the rejoinder is really very simple. Of course scientific consensus can be wrong – that’s the nature of science. But much more often it is ‘right’ (which is to say, it furnishes the best explanation for the observations with the tools to hand).

As further evidence of the untrustworthiness of consensus, however, Lea regales us with tales of how economists (for God’s sake) have in the past got things wrong en masse – apparently she thinks economics has a claim to the analytical and predictive capacity of natural science. Or perhaps she imagines that consensus-making is an arbitrary affair, a thing that just happens when lots of people get together to debate an issue, and not, as in science, a hard-won conclusion wrested from observation and understanding.

Ah, but you see, the science of global warming has been overturned by a paper “of the utmost scientific significance”, published by the venerable Royal Society. The paper’s author, a Danish scientist named Henrik Svensmark, “has been impeded and persecuted by scientific and government establishments” (they do that, you know) because his findings were “politically inconvenient”. What are these findings of the “utmost significance”? He has shown, according to Lea, that there has been a reduction in low-altitude cloudiness in the twentieth century owing to a reduction in the cosmic-ray flux into the atmosphere, because of a weakening of the shielding provided by the Sun’s magnetic field. Clouds have an overall cooling effect, and so this reduction in cloudiness probably lies behind the rise in global mean temperature.

Now, that sounds important, doesn’t it? Except that of course Svensmark has shown nothing of the sort. He has found that cosmic rays may induce the formation of sulphate droplets in a plastic box containing gases simulating the composition of the atmosphere. That’s an interesting result, demonstrating that cosmic rays might indeed affect cloud formation. It’s certainly worth publishing in the Proceedings of the Royal Society. The next step might be to look for ways of investigating whether the process works in the real atmosphere (and not just a rough lab simulacra of it). And then whether it does indeed lead to the creation of cloud condensation nuclei (which these sulphate droplets are not yet), and then to clouds. And then to establish whether there is in fact any record of increased cosmic-ray flux over the twentieth century. (We can answer that already: it’s been measured for the past 50 years, and there is no such trend.) And then whether there is evidence of changes in low-altitude cloudiness of the sort Svensmark’s idea predicts. And if so, whether it leads to the right predictions of temperature trends in climate models. And then to try to understand why the theory predicts a stronger daytime warming trend, whereas observations show that it’s stronger at night.

But that’s all nitpicking, surely, because in Lea’s view this new result “seriously challenges the current pseudo-consensus that global warming is largely caused by manmade carbon emissions.” Like most climate-change sceptics, Lea clearly feels this consensus is pulled out of a hat through vague and handwaving arguments, rather than being supported by painstaking comparisons of modelling and observation, such as the identification of a characteristic anthropogenic spatial fingerprint in the overall warming trend. It is truly pitiful.

“I am no climate scientist”, says Lea. (I take it we could leave out “climate” here.) So why is she commenting on climate science? I am no ballet dancer, which is why, should the opportunity bizarrely present itself for me to unveil my interpretation of Swan Lake before the nation, I will regretfully decline.

Simon Jenkins has recently argued in the Guardian that science should not be compulsory beyond primary-school level. I don’t think we need be too reactionary about his comments, though I disagree with much of them. But when a director of a ‘policy research centre’ shows such astonishing ignorance of scientific thinking, and perhaps worse still, no one on a national newspaper’s editorial or production team can see that this is so (would the equivalent historical ignorance be tolerated, say?), one has to wonder whether increasing scientific illiteracy still further is the right way to go. In fact, the scientific ignorance on display here is only the tip of the iceberg. The real fault is a complete lack of critical thinking. There are few things more dangerous in public life than people educated just far enough to be able to mask that lack with superficially confident and polished words.

But it’s perhaps most surprising of all to see someone in ‘policy research’ fail to understand how a government should use expert opinion. If there is a scientific consensus on this question, what does she want them to do? The opposite? Nothing? A responsible government acts according to the best advice available. If that advice turns out to be wrong (and science, unlike politics, must always admit to that possibility), the government nevertheless did the right thing. If this Policy Research Centre actually has any influence on policy-making, God help us.

No comments: