I was personally offended by the revelations of e-mails among prominent climate scientists that were leaked in late 2009 and collectively became known as Climategate. Working in supporting government policy-makers in environmental and energy policy issues for the first half of my professional life left me with a finely-honed opinion about crossing the line from science to policy. Let’s just say I’m about as comfortable with scientists pushing their scientific authority to make policy value judgments as I would be with Deacons performing ordinations.
Although the widely-perceived failure of the Copenhagen climate negotiations suppressed the political urgency of the issue — even the follow-on talks this year in Cancun attracted little attention because they did little to actually impact any economic activity — I have continued to follow Climategate post-mortem analysis to see whether the flaws in the scientific process the e-mails revealed would be fixed. With the publication of the final British House of Commons report on those analyses themselves, it’s time to draw some specific conclusions about Climategate and see what they might suggest about more general issues regarding our responses to authority figures on vital public issues — both inside and outside the church.
The center of the scandal, for those not previously keeping score, was the Climate Research Unit (CRU) at Britain’s University of East Anglica (UEA), which explains the House of Commons interest in the scandal. Although it employed data collected by scientists throughout the world, the CRU was itself one of the top repositories of modern climate information and compilations of paleo-climate proxy data. (Paleo-climates have to be estimated by things such as annual ice cores or tree rings since they can not be measured directly after the fact.)
Moreover, the small group of climate experts is fully “intermarried” as authors, co-authors, and reviewers of each other’s major papers. Although referring to the relationship as incestuous might be too strong, none of the repositories of expertise can avoid having its work contaminated if the CRU messes up.
Even more importantly, the CRU scientists and their co-authors became heavily involved in UN-sponsored reports released through the Intergovernmental Panel on Climate Change (IPCC) that have provided the scientific backbone for climate negotiations for about two decades. The policy-makers depend on the integrity of such advice, because they can not otherwise balance climate demands against other public policy demands for economic resources.
The e-mails raised suspicions from outside observers that CRU had, indeed, messed up, and, worse, had systemically tried to hide the fact by refusing to provide data to scientifically qualified skeptics of their conclusions, even though sharing of information was required by British law.
Journalists quickly noted e-mails that shattered any illusion of detached scientific impartiality that is held up as the ideal. (Trust me! This is NOT unusual; scientists feel hurt when their work is criticized just like everyone else does.) The questions were whether the obvious antagonism toward skeptics — combined with pressure on scientific journal editors and discussion of “clever tricks” and “hide the decline” — was mere ego, or rose to the level of interference with the peer review process, or even misconduct under the law, and whether conclusions may thereby have been biased consciously or unconsciously.
As journalists explored these issues, they began to see an erosion of objectivity between the scientific conclusions, with their inherent uncertainty, and the “consensus” that simplified things for the policy-makers. For example, The Daily Telegraph produced a beautiful graphic showing what “hide the decline” was about.
Tree ring data doesn’t match other data recording modern temperatures for reasons that are not understood yet. However, the same data set is what the climate scientists were relying on to dismiss the reality of (perhaps equally) high temperatures around 1000 AD, when human CO2 emissions could not have been responsible. This issue was critical, was still being debated among the scientists themselves, but did not permit the IPCC to tell a definitive story that would motivate economic sacrifice in a Copenhagen treaty. The “hide the decline” obscured the issue by burying it in parts of the IPCC reports that the decision-makers, and the media, would not see. Figures like the heading graphic in this post, as the Telegraph graphic shows, artistically terminated the offending curve so its end would be unnoticed as it diverged from the other temperature data.
And the e-mails showed that pressure to obscure came from higher in the academic and governmental food chain than the scientists themselves.
Once that kind of thing was noticed, attention turned to the IPCC writing process more broadly, where it was quickly discovered that groups with potential conflicts of interest had been heavily involved. In fact, the top IPCC positions were unpaid and tended to attract people with strong feelings about the outcome because they were strongly concerned about the problem of warming beforehand. As noted in the New York Times, another official review panel recommended that the IPCC top positions be term-limited and paid positions in order to avoid any appearance of bias.
The leaked e-mails also included source code internal documentation. That documentation caught the attention of a lot of scientists outside of the usual skeptics. The documentation demonstrated that the climate scientists, however expert they were in their own fields, were struggling to deal with a lack of expertise in other scientific fields (like statistics and computer code configuration control) that had become extraordinarily more important as the policy stakes had grown.
From a CBS News report:
“As the leaked messages, and especially the HARRY_READ_ME.txt file, found their way around technical circles, two things happened: first, programmers unaffiliated with East Anglia started taking a close look at the quality of the CRU’s code, and second, they began to feel sympathetic for anyone who had to spend three years (including working weekends) trying to make sense of code that appeared to be undocumented and buggy, while representing the core of CRU’s climate model.”
The three-year effort failed. CRU had lost control and understanding of its own computer models. The models were not necessarily incorrect, but if they were wrong, the scientists would no longer be able to know it, or to know how to correctly incorporate new discoveries about climate into the models. By default, the models were “correct” if they reproduced the results that the scientists expected, but “incorrect” when they didn’t.
No Federal agency would currently issue a contract for development, or would accept as a deliverable a scientific software product with this level of programming quality control procedures. What works as adequate in an academic environment is not adequate to underpin a major restructuring of the world’s economy.
And so, numerous authoritative bodies agreed, as would be normal scientific procedure, that the integrity of the science must be reexamined. But then a curious — well, not so curious — thing happened. The same problems of political-scientific cross currents, the inability to find reviewers disinterested in the outcome, and the lack of expertise in all relevant sciences that plagued the original work popped up in the review process, compounded by the need to finish as much as possible before Copenhagen. The reviews became a review of the integrity of the scientists and never got around to examining the integrity of the science , as a “minority” conclusion proposed for the House of Commons report made clear. This conclusion was drawn exclusively from sub-conclusions drawn earlier in the Commons report. However, with pressure for binding treaty commitments receding, and Britain’s economy threatened, the political system was prepared to move on to other matters, and left the science unexamined
Do any of the limits of the “climate priesthood” apply to the religious priesthood?
- Are the authorities within the priesthood too ingrown to review each other’s work?
- Do they have access to all of the relevant expertise they require?
- Are they sufficiently transparent to inspire any confidence among the unconverted skeptics inside and outside the community?
- Are they more willing to “move on” than to reexamine and, if necessary, rebuild on earlier understandings?
- If any of the above are problems, are there things that can positively address their solutions?