- JoNova - https://www.joannenova.com.au -

Cook’s 97% consensus is a case study of Agnotology – ignorance and misinformation

Agnotology is the study of how ignorance grows through repetition of misleading misinformation. You might never have heard of it, but it’s the perfect term for the climate science “debate”. Predictably its use began when those convinced of man-man global warming claimed fossil fuel groups were funding misinformation. But as per usual, unskeptical scientists opened a promising new front only to got burned by the evidence.

In the latest volley, from Legates et al 2013, John Cook’s “97% consensus” survey has become the case study in agnotology. Based on incorrect results, a flawed method, and a logical fallacy,  it kept key facts hidden while sloppily blending vague language into a form that is easily and actively misinterpreted. That it passed peer-review is another damning indictment of peer review.

Cook still refuses to provide about half the data, but the data that has been made public shows (after some digging) that a mere 41 papers out of 12,000 was called a 97% consensus. The trick is that Cook et al interchangeably use different definitions of consensus.

The Bait and Switch

The Bait: In the introduction Cook states that the reason for the paper is “to determine the level of scientific consensus that human activity is very likely causing most of the current GW [global warming]”.

The word “most” leaves no doubt that it refers to man-made forces causing “more than half”.

Cook categorizes endorsement of anthropogenic global warming into seven categories, only the first of which fits the aim stated in the introduction.

The switch:  Cook doesn’t provide results for Category one in the paper, even though that was the aim of the paper. To “simplify the analysis”, he blends together categories 1, 2 and 3 (which include two very different definitions of consensus). In the data there are only 64 papers of Type 1 — papers that state that humans are the primary cause. (Is that why he did not report the result?) Worse, Monckton read the abstracts and found that 23 of those are misallocated.

Did Cook really think he could get away with taking 41 papers out of 11,944 and claiming it was a “97% consensus”? Most of the thousands of papers included in his “97%” collection are merely me-too papers where the scientists have assumed the models are right and someone else has done the sums.

This statement in the abstract of Cook et al is so vague as to be useless.

Among abstracts expressing a position on AGW, 97.1% endorsed the consensus position that humans are causing global warming.

The statement refers to the broadest and weakest form of consensus, but is often mistakenly amplified to infer it causes “dangerous” global warming. By Cook’s own definition a mere 0.3% of papers actually show that “humans are the primary cause of recent global warming”.  Furthermore, being the primary cause does not necessarily mean the warming is also dangerous. The paper was not even designed to find out how many scientists endorsed dangerous warming. Yet in the media this is how the paper is being used. Tellingly, Cook has made no attempt to correct this misuse.

Note the strong terms of reference of the Legates paper — this is not about accidental misinformation, but intentional deception:

“…the focus will be on misinformation said to have arisen not through inadvertence, nor through any limitation in the state of knowledge, nor through any defect in teaching or learning, but through the self interested determination of some sufficiently influential faction to circulate misinformation calculated to sow doubt, to conceal a truth, or to promote falsehoods.”

The nub of the problem here is that this poorly done research is not contributing to human knowledge, but to the exact opposite. Cook uses a poor study of opinions to replace the empirical evidence he ought to have. And Bedford and Cook use agnotology as a method to shut down open debate about it.

Legates et al:

“Totalitarian regimes spread misinformation while demonizing their opposition. How is
it different here? Haud secus isti. If it is as Michael Oppenheimer argued earlier—though
the figure is wrong, the discussion is useful because it agrees with the consensus—then
misinformation is being used as information to support the consensus. In that instance,
agnotology takes on an added connotation—it includes the study of how misinformation is
spread as information by those espousing a contrived consensus to support one’s cause.”

Press release

0.3% climate consensus not 97.1%

MAJOR peer-reviewed paper by four senior researchers has exposed grave errors in an earlier paper in a new and unknown journal that had claimed a 97.1% scientific consensus that Man had caused at least half the 0.7 Cº global warming since 1950.

A tweet in President Obama’s name had assumed that the earlier, flawed paper, by John Cook and others, showed 97% endorsement of the notion that climate change is dangerous:

“Ninety-seven percent of scientists agree: #climate change is real, man-made and dangerous.” [Emphasis added]

The new paper by the leading climatologist Dr David Legates and his colleagues, published in the respected Science and Education journal, now in its 21st year of publication, reveals that Cook had not considered whether scientists and their published papers had said climate change was “dangerous”.

The consensus Cook considered was the standard definition: that Man had caused most post-1950 warming. Even on this weaker definition the true consensus among published scientific papers is now demonstrated to be not 97.1%, as Cook had claimed, but only 0.3%.

Only 41 out of the 11,944 published climate papers Cook examined explicitly stated that Man caused most of the warming since 1950. Cook himself had flagged just 64 papers as explicitly supporting that consensus, but 23 of the 64 had not in fact supported it.

This shock result comes scant weeks before the United Nations’ climate panel, the IPCC, issues its fifth five-yearly climate assessment, claiming “95% confidence” in the imagined – and, as the new paper shows, imaginary – consensus.

Climate Consensus and ‘Misinformation’: a Rejoinder to ‘Agnotology, Scientific Consensus, and the Teaching and Learning of Climate Change’ decisively rejects suggestions by Cook and others that those who say few scientists explicitly support the supposedly near-unanimous climate consensus are misinforming and misleading the public.

Dr Legates said: “It is astonishing that any journal could have published a paper claiming a 97% climate consensus when on the authors’ own analysis the true consensus was well below 1%.

“It is still more astonishing that the IPCC should claim 95% certainty about the climate consensus when so small a fraction of published papers explicitly endorse the consensus as the IPCC defines it.”

Dr Willie Soon, a distinguished solar physicist, quoted the late scientist-author Michael Crichton, who had said: “If it’s science, it isn’t consensus; if it’s consensus, it isn’t science.” He added: “There has been no global warming for almost 17 years. None of the ‘consensus’ computer models predicted that.”

Dr William Briggs, “Statistician to the Stars”, said: “In any survey such as Cook’s, it is essential to define the survey question very clearly. Yet Cook used three distinct definitions of climate consensus interchangeably. Also, he arbitrarily excluded about 8000 of the 12,000 papers in his sample on the unacceptable ground that they had expressed no opinion on the climate consensus. These artifices let him reach the unjustifiable conclusion that there was a 97.1% consensus when there was not.

“In fact, Cook’s paper provides the clearest available statistical evidence that there is scarcely any explicit support among scientists for the consensus that the IPCC, politicians, bureaucrats, academics and the media have so long and so falsely proclaimed. That was not the outcome Cook had hoped for, and it was not the outcome he had stated in his paper, but it was the outcome he had really found.”

Christopher Monckton of Brenchley, an expert reviewer for the IPCC’s imminent Fifth Assessment Report, who found the errors in Cook’s data, said: “It may be that more than 0.3% of climate scientists think Man caused at least half the warming since 1950. But only 0.3% of almost 12,000 published papers say so explicitly. Cook had not considered how many papers merely implied that. No doubt many scientists consider it possible, as we do, that Man caused some warming, but not most warming.

“It is unscientific to assume that most scientists believe what they have neither said nor written.”

Contact David Legates at Udel.edu for more information



Bedford, D. (2010). Agnotology as a teaching tool: Learning climate science by studying misinformation.
Journal of Geography, 109, 159–165.

Bedford, D., & Cook, J. (2013). Agnotology, scientific consensus, and the teaching and learning of climate
change: A response to Legates, Soon and Briggs. Science & Education, 22, 2019–2030. Abstract

Cook, J., Nuccitelli, D., Green, S. A., Richardson, M., Winkler, B., Painting, R., et al. (2013). Quantifying the consensus on anthropogenic global warming in the scientific literature. Environmental Research Letters, 8, 024024.

Legates, D. R., Soon, W., & Briggs, W. M. (2013). Learning and teaching climate science: The perils of
consensus knowledge using agnotology. Science & Education, 22, 2007–2017.

Legates, D. R., Soon, W.,  Briggs, W. M, & Monckton, C. (2013) Climate Consensus and ‘Misinformation’: A Rejoinder to Agnotology, Scientific Consensus, and the Teaching and Learning of Climate ChangeScience & Education  [Abstract]

Papers on: Agnotology (Climate)


8.9 out of 10 based on 98 ratings