28 November 2009

Mike Hulme on Climategate

These comments by Mike Hulme of UEA (location of CRU) were originally posted at DotEarth:

The key lesson to be learned is that not only must scientific knowledge about climate change be publicly owned — the I.P.C.C. does a fairly good job of this according to its own terms — but the very practices of scientific enquiry must also be publicly owned, in the sense of being open and trusted. From outside, and even to the neutral, the attitudes revealed in the emails do not look good. To those with bigger axes to grind it is just what they wanted to find.

This will blow its course soon in the conventional media without making too much difference to Copenhagen — after all, COP15 is about raw politics, not about the politics of science. But in the Internet worlds of deliberation and in the ‘mood’ of public debate about the trustworthiness of climate science, the reverberations of this episode will live on long beyond COP15. Climate scientists will have to work harder to earn the warranted trust of the public - and maybe that is no bad thing.

But this episode might signify something more in the unfolding story of climate change. This event might signal a crack that allows for processes of re-structuring scientific knowledge about climate change. It is possible that some areas of climate science has become sclerotic. It is possible that climate science has become too partisan, too centralized. The tribalism that some of the leaked emails display is something more usually associated with social organization within primitive cultures; it is not attractive when we find it at work inside science.

It is also possible that the institutional innovation that has been the I.P.C.C. has run its course. Yes, there will be an AR5 but for what purpose? The I.P.C.C. itself, through its structural tendency to politicize climate change science, has perhaps helped to foster a more authoritarian and exclusive form of knowledge production - just at a time when a globalizing and wired cosmopolitan culture is demanding of science something much more open and inclusive.


  1. I hadn't seen this post at DotEarth since Hulme's excellent comments were added- thanks for the link. What he's saying sounds exactly right, but I've got to wonder: will climate science will be able to resist staying political? These email show not just a field that has "circled the wagons", but one that is full of True Believers. As such, it seems to be inherently political and at risk of producing biasing results. So what would you do if, for example, you were Phil Jones' boss?

  2. I'd make him post his data and models on the internet, post his papers online in a format so that people could comment (even if a journal objected) and require him to answer the reasonable comments (reasonableness determined by a group of tenured faculty from related fields).
    For a start.

  3. Roger, is this story in the London Sunday Times correct or not?

  4. -3-bernie

    Yes, I think so, see:


    Though the CRU page on data availability linked there is not part of their "emergency server"

  5. I had seen this earlier note, but I assumed that there had been a new development. Shouldn't CRU do what they can to reconstitute the data? Wouldn't the National Weather Services have their own backups?

  6. Hulme seems to be part of group pushing climate science into something euphemistically called 'post-normal science', where scientists are advocates, and results are based on consensus among advocates, and not on a well defended thesis.
    I think that shluld be clarified before he is considered an authority on anything.

  7. I wonder which 'primitive cultures' Mr Hulmes refers to? Primitive cultures with unattractive social organizations? Are you British, Mr. Hulmes? Are your metaphors still a little bit shaped by colonialism? Just asking...

  8. Frontiers..

    Check out the wikipedia definition of "post-normal" science.
    I would argue that scientists "potentially being advocates and achieving their own consensus without a well defended thesis" appears to be the situation we have today.
    To quote the wikipedia definition,
    " They bring their "extended facts", that will include local knowledge and materials not originally intended for publication such as leaked official information. There is a political case for this extension of the franchise of science; but Funtowicz and Ravetz also argue that this extension is necessary for assuring the quality of the process and of the product."

    Since the quality of the process and the product currently seems to be the issue, it appears to be a great argument for climate science as post-normal science.

  9. Kwk asks ...

    "I wonder which 'primitive cultures' Mr Hulmes refers to? Primitive cultures with unattractive social organizations? Are you British, Mr. Hulmes? Are your metaphors still a little bit shaped by colonialism? Just asking..."

    Good point and I apologise for this expression - it was rushed off to Andy Revkin without time for reflection or even without knowing he was going to use it. What I should have said is that what we are seeing in climate science is a form of tribalism which anthropologists take great interest in studying in all cultures, including scientific ones. I am of course British.

    Mike H

  10. My experience of trying to publish in the climate literature (using time series modelling) is that there was very strong resistance to our non-mainstream methods from reviewers. They always wanted us to use GCM results in our analysis as well as actual temperature data and definitely didn't like our paper criticizing GCMs which never got published. In the end, I gave up pursuing this direction of research because it just seemed so hard to make headway and get cited by anyone. Our first paper on the topic was though published in Nature and got decent citations but it went downhill from there.

    On the other hand, it is very hard to get published in the mainstream economics literature if you don't use mainstream methods. All disciplines are probably very methodologically conservative.

  11. Now that Judy Curry and Mike Hulme introduce the term 'tribe' it is perhaps in order to point out that the sociology of science has long studied this phenomenon of gatekeepers and groupthink. A term used by HM Collins is CORE SET.

    Core sets of research are to be found at a few prestigious institutions and laboratories and are linked to other networks at prestigious institutions. Leadership personalities in these groups control the access to key resources of research such as labs, publication opportunities and finance (cf. Hagstrom 1965; Traweek 1988). They also decide in which direction the field will move, what the next ‘hot’ topics will be and where the boundary between science and non-science has to be drawn (Gieryn 1995; Jasanoff 1990).

    The acknowledgement as an expert in the field is done by these core sets of scientists. This is gate keeping activity or boundary work.

    I came across some interesting examples in the study of scientists studying stratospheric ozone in the mid 1990s. The following interview excerpt is kind of intriguigng, given the fallout from CRU. Here a prominent scientists muses about the merit of including skeptical voices in the literature:

    "There are also scientists, people who have done good work in the past, such as S. and E. But they have the wrong arguments. They do not publish in our journals. They do not get through with their arguments. Now one might say that this is the scientific Mafia which stops them but it really is too ridiculous."

  12. Dr. Hulme, as a prelude to all the sociological chin-tugging, why not post a clear statement of whether you believe global warming is occuring, and whether human activity is a significant driverm, based on curred evidence? That would speak to 90% of the 'denialist' blather I see on the blogosphere, which starts with the presumtion that AGW is a 'lie'.