Thursday, December 24, 2009

Research Ethics

There has been a reluctance on the part of many climatologists and other observers of Climategate to condemn the exposed behaviour. True, some said that the mails show that the researchers unfortunately do not appear as ‘nice people’ but that they have done top science.


This is not the point. The point is the attempt to manipulate the peer review process and the drafting of the IPCC reports. Some refuse to discuss the behaviour on the grounds that the emails have been obtained illegally (Stephen Schneider at Copenhagen press conference). Others say that scientists are humans and therefore their behaviour cannot be beyond the pale. The exposed practices are said to be ‘normal’ in many fields.

Both positions are untenable. The first is untenable as the CRU scientists did not contest the content of the mails. They have become a fact, no matter if you like it or not.

The second is untenable because the IPCC process and its policy relevance are unique. There are many anecdotes of how peer review has been corrupted in other fields of research, across the academic spectrum. We do not have systematic studies because of the nature of the process (anonymity and confidentiality). Thus it is normal that some people will claim that their field or journals are innocent in comparison whereas others will say that their field is similar to what has been shown at CRU. The difference is that IPCC relevant peer review has policy implications at the highest level. It is supposed to inform governments around the world with the best scientific knowledge available. The process used for this aim should be as robust as possible and withstand any amount of scrutiny. Sooner or later it will be put under the microscope.

More than half a century ago the sociologist of science Robert Merton said that science is a social institution which is guided by four norms: universalism (i.e. truth claims are evaluated in terms of universal or impersonal criteria, and not on the basis of race, class, ideology etc.); communism (free exchange of ideas and publication of research findings); disinterestedness (scientists do not have a stake in the outcome of research); and organized scepticism (research is checked by rigorous, structured scrutiny of peers). Do the practices highlighted by the emails exemplify any of the four norms?

Of course, it has been stated many times that Merton’s ideals were far removed from reality. But that reality should be removed is a much stronger claim and one that is made rarely. It seems to be made by some ill-advised people defending the climate cartell. Would science as a special social institution survive if we established new norms, effectively condoning and advocating such behaviour? Would we attract new generations of young researchers into this brave new world of ‘science’?

Michael Mann seems to realise the perils. He stated in the Washington Post: “I cannot condone some things that colleagues of mine wrote or requested in the e-mails recently stolen from a climate research unit at a British university. . . Some statements in the stolen e-mails reflect poor judgment -- for example, a colleague referring to deleting e-mails that might be subject to a Freedom of Information Act request -- but there is no evidence that this happened.”

Sure, this is lukewarm. But at least he acknowledges some normative standards of research ethics in this statement. It is time others wake up to the challenge.

1 comment:

Stan said...

Some have used the argument "this is just the way science is done" in several different venues. When AGW skeptics point out that no ever replicates the studies of others, we hear this same argument (because funding isn't available for replication). When skeptics point out how incompetent it was that no scientist ever bothered to check the siting of monitoring stations, we hear that it was unreasonable to expect any scientist to inquire into the quality of the data he used to make declarations of global doom. "That's just the way that science is done."

But those using the argument fail to take into account the moral foundation which must accompany a demand to impose massive infringements upon the lives, liberty, and property of the world's citizens. If 'science as usual' doesn't involve replication, it is inadequate for the task of supporting the proposed massive infringements. If 'science as usual' doesn't require that data quality be assured, it is inadequate for the task. If 'science as usual' doesn't reach a level of honesty and integrity, it can't be relied upon for massive changes to society.

The myopia of this type response by many scientists doesn't speak well for their moral maturity. Because their fellow citizens really don't give a damn whether the inadequacy of their science is "normal" or not. If they advocate changing the world, their evidence better be solid -- audited and replicated studies, verified and validated models, quality assured data, and all in a totally transparent environment built with integrity.

No more hide-the-ball. No more crappy amateur code and bizarre novel statistical garbage. No more unaudited and unreplicated studies. No more secret adjustments and peer review slanting. Business as usual is just BS.