[Arctic ice cap shrinks dramatically as global warming denialists prepare for new "surge."]
Naomi Oreskes is an historian of science whose PhD work was entirely concerned with understanding a major scientific controversy-the early 20th Century controversy over continental drift, and its rejection by American geologists. [Book: The Rejection of Continental Drift: Theory and Method in American Earth Science]. She went on to edit a second book [Plate Tectonics: An Insider's History of the Modern Theory of the Earth] about the eventual triumph of continental drift, which came about only in the 1960s.
Thus, as Oreskes told me when I interviewed her in 2005 in the aftermath of Katrina, she was intimately familiar with what such controversies look like, and knew from her own research that what is going on in the field of climate science is nothing remotely like a fundamental scientific controversy. In 2004, she set out to conduct an experiment to produce hard numbers to docuiment what she already knew as a trained observer. The results were published in a brief article in Science magazine, "Beyond the Ivory Tower: The Scientific Consensus on Climate Change," (Science 3 December 2004, Vol. 306. no. 5702, p. 1686), and were expanded on in a book chapter published in 2007, "The scientific consensus on climate change: How do we know we're not wrong?" [PDF] (in Climate Change: What It Means for Us, Our Children, and Our Grandchildren, edited by Joseph F. C. DiMento and Pamela Doughman, MIT Press, pp. 65-99.)
In her article, Oreskes first took note of the existing consensus, and its expression in a variety of forms through public statements by major scientific bodies with relevant expertise and responsibility. She said:
The scientific consensus is clearly expressed in the reports of the Intergovernmental Panel on Climate Change (IPCC). Created in 1988 by the World Meteorological Organization and the United Nations Environmental Programme, IPCC's purpose is to evaluate the state of climate science as a basis for informed policy action, primarily on the basis of peer-reviewed and published scientific literature (3). In its most recent assessment, IPCC states unequivocally that the consensus of scientific opinion is that Earth's climate is being affected by human activities: "Human activities ... are modifying the concentration of atmospheric constituents ... that absorb or scatter radiant energy. ... [M]ost of the observed warming over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations" [p. 21 in (4)].
Oreskes then went on to cite other organizations who had explicitly supported the IPCC-identified consensus, including the National Academy of Sciences (quoting from its report, Climate Change Science: An Analysis of Some Key Questions), the American Meteorological Society, the American Geophysical Union, and the American Association for the Advancement of Science.
From this starting point, it's clear to anyone who understands how science works that there is already a clearly visible consensus. Anthropogenic (human-caused) climate change is real, and is recognized by an impressive array of relevant scientific bodies. The only real question is, "How strong is this consensus? Are there any serious challenges to it?" This is the question that Oreskes set out to investigate. Science being what it is, there are virtually always naysayers about, if one looks hard enough. So what about them? How widespread are they, and do they have any merit?
Oreskes herself put like this:
The drafting of such reports and statements involves many opportunities for comment, criticism, and revision, and it is not likely that they would diverge greatly from the opinions of the societies' members. Nevertheless, they might downplay legitimate dissenting opinions. That hypothesis was tested by analyzing 928 abstracts, published in refereed scientific journals between 1993 and 2003, and listed in the ISI database with the keywords "climate change" (9).
The 928 papers were divided into six categories: explicit endorsement of the consensus position, evaluation of impacts, mitigation proposals, methods, paleoclimate analysis, and rejection of the consensus position. Of all the papers, 75% fell into the first three categories, either explicitly or implicitly accepting the consensus view; 25% dealt with methods or paleoclimate, taking no position on current anthropogenic climate change. Remarkably, none of the papers disagreed with the consensus position.
Admittedly, authors evaluating impacts, developing methods, or studying paleoclimatic change might believe that current climate change is natural. However, none of these papers argued that point.
Two points need to be stressed here: First, as already noted, Oreskes was not proving that there is a consensus on human-caused climate change. The existence of that consensus had already been proven by the statements of the relevant scientific bodies she cited. Rather, Oreskes was demonstrating the length, depth and breadth of that consensus. It did not emerge as a result of research during the 1990s, but was present from the very beginning of the period she studied, and it was sufficiently broad and deep that she found no dissenting articles in a survey of 928 abtracts drawn from a broadly-defined search using an inclusive database.
Oreskes herself explained the state of the science she was describing in the book chapter cited above:
[T]o say that global warming is real and happening now is not the same as agreeing about what will happen in the future. Much of the continuing debate in the scientific community involves the likely rate of future change. A good analogy is evolution. In the early twentieth century, paleontologist The Scientific Consensus on Climate Change George Gaylord Simpson introduced the concept of ''tempo and mode'' to describe questions about the manner of evolution-how fast and in what manner evolution proceeded. Biologists by the mid-twentieth century agreed about the reality of evolution, but there were extensive debates about its tempo and mode. So it is now with climate change. Virtually all professional climate scientists agree on the reality of humaninduced climate change, but debate continues on tempo and mode.
Second, Oreskes was not claiming that there were no published papers disputing the consensus. Historians of science are well aware that small pockets of dissenting views can often be found when there is no serious challenge to the consensus view within a field. This often happens, for example, when a few die-hard proponents of an earlier view simply refuse to ever accept a newly-established consensus. They may still be respected, and public occassionally, but no one is newly pursuaded by their work.
Thus, it would have been perfectly undertstandable to find a small sample of dissenting articles, and such could still exist, since Oreskes was examining a representative sample, not the entire universe of peer-reviewed papers. But the sample was sufficiently large that the absence of any dissenting papers was surely significant: Whatever objections there may be to the consensus, they do not play a significant role in deliberations within the field.
Indeed, in the book chapter, Oreskes wrote:
The total number of papers published over the last ten years having anything at all to do with climate change is probably over ten thousand, and no doubt some of the authors of the other over nine thousand papers have expressed skeptical or dissenting views. But the fact that the sample turned up no dissenting papers at all demonstrates that any remaining professional dissent is now exceedingly minor.
There was a subsequent attempt to challenge her work was mounted by Benny Peiser, who claimed that Oreskes had misclassified 34 papers that challenged the conensus. His claim was immediately criticized, and scaled back, but it took a full year for him to admit that he was wrong about at least 33 of the 34 papers-97%. [Wrap-up summary with links here. Significantly, denialists continue to pretend that Peiser somehow undermined Oreskes' credibility.]
Claims Of A Consensus Overturned
Now a much more ambitious-sounding challenge to Oreskes' work has been mounted, headlined by the claim that a majority of climate scientists no longer support the consensus. To anyone who understands the story so far, this claim is clearly absurd. If true, it would be very big news indeed. It would mean that the IPCC, the American Meteorological Society, the American Geophysical Union, and the American Association for the Advancement of Science had all been wrong, and had now reversed their positions.
Of course, this is nonsense. Nothing of the sort has occured.
If anything, evidence has reached a new crescendo of intensity since Oreskes first published, and even the Bush Administration has reluctantly backtracked modestly from its hardline opposition-though not from doing its best to suppress the science involved. [They are now trying to implement a system of high-level security clearances, including extensive FBI background checks on climate scientists at NASA's Jet Propulsion Laboratory and Goddard Space Flight Center, for example-LINK.] But ignoring the fundamental framework of agreement that Oreskes noted, the new paper claims that an analysis of peer-reviewed literature since 2004 shows much lower levels of support for the consensus view, and much higher levels of dissent.
Given that there has been no erosion of the existing consensus, the most likely explanation is that the new study is simply flawed. That it either does not follow the same protocols that Oreskes used, and thus cannot be compared, or that it codes papers erroniously, or both.
In fact, despite the fact that only true believers have seen the paper so far (and only two people, apparently, have written about it), we have evidence that the new study is flawed in both ways.
The New Studies Flaws, Part 1: Bad Coding
So far, we have only a small set of examples to look at that supposedly examplify rejection of the consensus. These appear in an online paper, "Consensus"? What "Consensus"? Among Climate Scientists, the Debate Is Not Over by The Viscount Monckton of Brenchley (July 2007). In this paper, we're told how superior the new work by Schulte is. For example, we're told, "Unlike Oreskes, who does not quote even one of the 928 papers upon which her analysis was based, Schulte cites some of the counter-consensual papers from his sample." Of course, such citations are not a normal part of scientific protocol, where investigator integrity is assumed. But in this case, we're fortunate, since it saves so much time.
The third such example given is particularly telling:
Leiserowitz (2005) reports -
"results from a national study (2003) that examined the risk perceptions and connotative meanings of global warming in the American mind and found that Americans perceived climate change as a moderate risk that will predominantly impact geographically and temporally distant people and places. This research also identified several distinct interpretive communities, including naysayers and alarmists, with widely divergent perceptions of climate change risks. Thus, 'dangerous' climate change is a concept contested not only among scientists and policymakers, but among the American public as well."
Clearly, this paper has nothing at all to say about climate science per se--and thus cannot represent a rejection of the consensus. It is a paper about attitudes regarding climate science. Any lingering doubt whatsoever is easily resolved by looking at the paper itself, available here in PDF. If Schulte thinks this paper disputes the concensus, he is clearly incapable of the basic analytical task that his work requires. What's more, The Viscount Monckton is similarly disqualified, since he passed this on as a purported example of Schulte getting the goods on Oreskes.
Blogger Tim Lambert, an Australian computer scientist who blogs about politicized scientific issues, goes even farther in an incisive post, "Classifying abstracts on global climate change", analyzing all seven papers:
...three of them really so reject the consensus.
Cao just says that there are uncertainties in our understanding of the carbon cycle. Leiserowitz just studied public opnion of the risks of climate change. Moser was not one of the 576 papers. Lai et al ends up implicitly endorsing the consensus by suggesting that reducing CO2 emissions will reduce global warming.
The three that do reject the consensus are Gerhard, which was published in the American Association of Petroleum Geologists Bulletin; Shaviv arguing for cosmic rays, which doesn't explain how they could make a difference over the past 50 years when the cosmic rayflux hasn't changed over that period; and Zhen-Shan and Xian, which is just a rubbish paper that should not have been published. (What is the next number in this sequence? 60. Their answer is 60.)
So, a 42% accuracy rate in coding papers. Not exactly ready for peer review. And these are presumably the best examples he has.
Not only that, but the first few commentators to Lambert's post raised questions about two of the three identified as rejecting the consensus. As a member of the American Association of Petroleum Geologists noted:
The Gerhard (2004) paper from the AAPG Bulletin is in a section of the journal called "Geohorizons" described as: 'Geohorizons papers focus on current technical methodologies and should be written for the general readership rather than for a limited audience of experts in a given field. (see Gerhard's article here).
So, it is essentially a review article. I would not describe it as a "research paper". If I were doing a compilation of climate research papers, this would not be on the list.
Another commentator remarked:
Zhen-Shan and Xian indeed looks shaky, but it's not 100% clear to me from the abstract that it's necessarily a challenge to the consensus. In a strict sense perhaps, but a bit like the apparently much-superior Tsonis paper, it argues that the anthropogenic signal is overlaid on natural climate cycles, and (diverging somewhat from Tsonis, who finds a smaller-amplitude natural signal) that the amplitude of the anthro signal hasn't yet gotten to the point that it overwhelms the natural cycle.
Thus, we really only have one of seven papers that unambiguously meets the criteria of explicitly rejecting the consensus view. That's a 6-out-of-7, 86% error rate. With that sort of stupendous error rate, nothing claimed for the paper should be believed. There is simply no reason to trust anything about it.
The New Studies Flaws, Part 2: Changed Protocols
In her original study, Oreskes sorted the papers as follows:
The 928 papers were divided into six categories: explicit endorsement of the consensus position, evaluation of impacts, mitigation proposals, methods, paleoclimate analysis, and rejection of the consensus position.
As for the results, Oreskes wrote:
Of all the papers, 75% fell into the first three categories, either explicitly or implicitly accepting the consensus view; 25% dealt with methods or paleoclimate, taking no position on current anthropogenic climate change. Remarkably, none of the papers disagreed with the consensus position.
In contrast, Monckton presents both the abstract of Dr. Schulte's paper, and what he claims is a table of results. The table by itself is inconclusive-it could simply be one way among several of analyzing the data-but if it is indicative of Schulte's primary coding protocol, it is clearly incompatible with Oreskes', which makes any direct comparison of their results utterly impossible:
|Abstracts on ISI Web of Science||Oreskes (2004) ||Schulte's review|
|Period under review:||1993 to 2003||2004 to 2007|
|Quantity of documents reviewed: || 928 documents||539 papers|
|Mean annual publication rate: || 84.3 documents.yr -1 || 254.6 || (+201%)|
|Explicit endorsement of the consensus: || Not stated||7%||(38 papers) |
|Explicit or implicit endorsement: || 75%||45% ||(244 papers) |
|Explicit rejection of the consensus|| 0%||1.3%||(7 papers) |
|Explicit or implicit rejection:||0%||6% ||(32 papers) |
|New data / observations on climate change:||Not stated||24% ||(127 papers) |
|New research on the consensus question:||Not stated||2%||(13 papers) |
|Quantitative evidence for the consensus:||Not stated||0%||(no papers) |
|Mention of "catastrophic" climate change:||Not stated||0%||(one paper) |
While it's certainly possible that this is a secondary table, supplmenting a primary one meant to show the two results in direct head-to-head comparison, it is still troubling in at least four ways.
First, it makes no attempt to be comprehensive. Neither of the columns adds up to 100%. Even a secondary table in a serious research paper ought to have its columns sum to 100%, with appropriate catch-all categories for those items not being specifically focused on.
Second, because the columns don't sum to 100%, it's not even possible to tell if the categories are meant to be mutually exclusive. Are papers categorized under "New data / observations on climate change" excluded from "Explicit or implicit endorsement:"? Included? Some included, some not? And, in any case, why?
Third, although it's impossible to say without seeing the underlying abstracts referred to, but it seems exceeedingly likely that most if not all of those categorized under "New data / observations on climate change" could reasonably be categorized as implicitly endorsing the consensus. The reason is simple: the consensus has been in place for quite some time now, and is very broadly shared. If someone makes observations that conflict with the consensus, that's news. Observations that confirm it are not. Therefore silence is generally indicative of implicit endorsement of the consensus.
What's more, what looks like silence to the untrained eye is very often not. Experts in the field are the best at picking up implicit messages, while trained historians such as Oreskes are capable of learning to read such messages nearly as well. However, a medical doctor with no such interpretative training is exceedingly unlikely to pick up such implicit messages. The chance of subtley misreading such papers is high-especially given the evidence of gross misreadings in the section above.
Fourth, why is there a category for "Mention of 'catastrophic' climate change"? This was certainly not part of Oreskes' original protocol, nor is it representative of how climate scientists ordinarily might be expected to write. It is, however, a direct reflection of the denialist's agenda-the setting up of a straw man.
The combination of these four observations creates the distinct impression that even if this is a secondary table, it is shows troubling signs of unreliability at best, as well as indications of being created for purposes that are much more likely to be propagandistic than to be objectively informative. Given that we haven't seen the whole paper, it is impossible to be certain, but this is not a good sign of what lies within. There is a very strong probability that the high level claim that fewer than half the papers now endorse the consensus is simply wrong due to methodological errors.
In any event, the further claim that lack of evidence is evidence of lack has no support whatsoever. Indeed, it is advanced as a logical sleight-of-hand. This brings us to our next topic-the nature of posts and sites promoting this story.
But this post is already more than long enough. That will have to wait for Part II.