Endorsement in the scientific literature
The first point to note is that the articles represented in my literature survey did not come from the scientific literature at random. Each appeared because it answered to the search terms “global warming,“ “global climate change,” or “climate change.” In the majority of cases, the authors themselves used one of those terms in their title or abstract. These authors have all chosen to write about one of those three topics. Would thousands of authors write about a theory they do not accept, yet never say so? What would be the point?
The Cook et al. paper used explicit endorsement as the criterion for counting a paper as being part of the consensus. That required them to rule out two-thirds of the published literature an AGW. But you cannot rule out the majority and then derive the consensus, because the consensus is what the majority believe. Moreover, as the four examples below show, to use endorsement as the criterion of consensus is to fail to understand the way publishing scientists treat their ruling paradigm.
This theory has been the ruling paradigm of geology for half a century. (Here I use paradigm in the sense of Thomas Kuhn: “Universally recognized scientific achievements that, for a time, provide model problems and solutions for a community of practitioners (Kuhn 1962).”) I used the Web of Science to review 500 recent articles on “plate tectonics.” Not one endorsed plate tectonics directly. Nor did any reject the theory. Some pointed out unsolved problems, but those did not cause the authors to question plate tectonics itself.
To provide a second example, I reviewed the abstracts of articles in the Journal of Evolutionary Biology from 2000 through 2014. Of the 303 articles, 261 had abstracts. Not surprisingly, none of the 261 rejected the modern evolutionary synthesis. But neither did any “endorse” it. The closest any came in my opinion is this statement: “A long line of biologists have followed [Darwin] in seeing, in the concept of ‘descent with modification’, a framework naturally able to incorporate both adaptation and constraint.” The authors of these articles are writing about evolution; again, why would they do so if they did not accept it? And since evolution in virtually unanimously accepted by biologists, why would a publishing biologist feel it necessary to express that acceptance directly?
As a third example, consider the origin of lunar craters by meteorite impact. As recently as 1964, nearly every scientist who had studied the moon believed that her craters were volcanic. Then in July of that year the first successful Ranger mission returned thousands of photographs showing that the moon exhibits craters ranging in size from the colossal to the microscopic. Except for a few senior holdouts, scientists quickly embraced the meteorite impact theory. A WoS search for “lunar craters” turns up 185 articles stretching back to 1920. I reviewed the abstracts of the most recent 100, which go back to 1997. As with plate tectonics and evolution, none of the 100 reject meteorite impact as the cause of lunar craters, but neither does any explicitly endorse impact. The closest any comes may be this sentence: “It is known that most of the craters on the surface of the Moon were created by the collision of minor bodies of the Solar System.” This is not an "endorsement" but a simple statement of a scientific fact.
Environmental Research Letters
I used the Web of Science to review articles from a single journal, Environmental Research Letters, for 2013 and 2014, searching under “global warming” and “climate change.” I found 283 unique articles. None rejected AGW and only one might be said to endorse the theory. Its title, “The role of reduced aerosol precursor emissions in driving near-term warming,” is self-explanatory. The authors concluded that, “In the near-term, as in the long-term, GHG [greenhouse gas] increases are the dominant driver of warming.” In my opinion, this is not an endorsement but simply a statement of a scientific finding of the kind one might find in any number of articles. The Cook et al. (2013) article was one of the 283. Even though we know these authors accept AGW, their own method places them in the no position category and rules them out of the consensus.
It is an evident fact of scientific publication that authors writing about their ruling paradigm almost never endorse it directly. The Cook et al. results prove this very point. As discussed here, only 0.5% of articles (64) in their database, according to their classification, fully endorsed AGW. But upon examination, few if any of the 64 are true endorsements; rather they are statements of a scientific conclusion or ones that the authors made to set the stage for their article. Thus we can conclude that well over 99.5% of the authors in the Cook et al. database do not endorse AGW, just as the authors of the other theories above do not endorse their ruling paradigm. But from a multitude of other evidence, we know that AGW is the strong consensus, just as we know it is for those other theories. The Cook et al. results falsify their own starting assumption and their methodology.
It is a truism of scientific publication that authors who question or reject a theory say so; those who accept it rarely say so directly. As one example of dissent from a consensus, consider this paper: Keith, M. (2001). Evidence for a plate tectonics debate. Earth-Science Reviews, 55(3), 235-336. In it, the distinguished geologist Mackenzie Keith presented a 101-page critique of plate tectonics, showing how in his opinion, much of the evidence could be interpreted differently. It is salutary that instead of declaring plate tectonics false, Keith made the case for caution and a more extended debate. But his arguments did not find many supporters.
If one considers Keith's paper a rejection, and if there were no direct endorsements in 2001, the Cook et al. method applied to that year would leave us dividing zero (endorsements) by 1 (endorsements plus rejections).
The scientific literature itself falsifies the use of endorsement as the criterion of consensus among scientists. This is not merely my opinion: it is an evidentiary fact. If to count as part of the consensus, scientists must directly endorse a theory, as Cook et al. required, then there is no consensus on evolution, global warming, meteorite impact, plate tectonics, or, likely, any widely accepted theory.