29 August 2009

Scientific Arguments as Tribal Politics

Michael Tobis, a climate scientist who can always be relied on to say what he really thinks, provides another remarkable glimpse inside modern climate science as he explains how he evaluated the arguments in Klotzbach et al., our recent paper that he continues to dismiss as "nonsense":
No scientist really knows everything he or she claims to know from direct experience. Most of what we know as individuals comes from two factors: 1) a network of trust and 2) the test of coherence. . .

Consider how I got mixed up with the Klotzbach paper. Prominent naysayer Joe d'Aleo's blog alleged that most warming on land was due to a bias in the land surface temperature record, referring to Klotzbach's paper. Klotzbach's claim was not consistent with my coherence network. Therefore, I resolved to figure out what the unclear claim really amounted to. The deeper I got into it, the stranger it got. Eventually, one of the authors was compelled to admit that the word "bias" did not actually indicate an error in temperature. Further reading revealed many other flaws in the work, although the key one, which appears to have been reported (somewhat at second hand) by James Annan, is something I still don't entirely understand.

This is where the first principle cuts in. Should I further investigate the key claim, still contested by the authors? Well, I know James to be an extraordinarily careful and precise thinker, and it's already demonstrated that his opposition is not. Since boundary layer meteorology is not my forte, and since the rest of the paper is flawed in many ways, I feel satisfied that it's best to put my attentions elsewhere.

The main point for present purposes is that I immediately questioned the result claimed by d'Aleo on the basis of its incoherence with everything else I know. And my questioning turned out to be justified. The publication, though it passed peer review, probably should not have done so. It looks like science from a distance, but up close it looks like nonsense.
Remarkably, Michael admits to having no expertise in the subject of our paper and not to understand the arguments made in it or about it, but because the paper challenges the beliefs that he and his tribe ("coherence network") firmly hold, he concludes that the scientific arguments in our paper must be wrong. He in fact knew this to be the case before even reading our paper. Michael speaks of an "opposition" -- and presumably these are the people in other tribes who hold views that are different than his own. I guess I am in that "opposition" because someone Michael dislikes happened to blog on our paper (d'Aleo). Michael reminds us that, after all, some other climate scientist in Michael's tribe (James Annan) who also admitted to having no expertise in the area of our paper said it must be wrong.

Will Michael's or James' critiques of our work appear in the peer reviewed literature? Of course not (because their critiques are off target and simply wrong). But their critiques do serve an important sociological function by reinforcing the tribal network and also give some comfort to those who evaluate arguments solely by their degree of conformance with views already held. Had Michael been blogging around the time of Copernicus, he would have explained to his readers that the world is in fact at the center of the universe, and that Copernicus guy must be wrong, because Michael and all of his Ptolemian friends said the world was at the center of the universe, so those saying differently must be wrong because they do not jibe with his "coherence network."

The troubling thing about this is not that people evaluate arguments based on trust -- we all do and it is a necessary part of life. What is troubling is that Michael suggests that he has found our paper "is flawed in many ways" but the basis for this is a felt need for tribal affirmation and not having his firmly held beliefs challenged. His scientific judgment on our paper is not grounded in the logic, data or analysis found in our paper which passed peer review and is published in a leading journal in the field. To the extent that Michael's views of scientific arguments are shared among his climate science peers, that community loses its ability to evaluate arguments based on their merits rather than by their putatuve tribal characteristics. I don't think that all or even most climate scientists think or act this way, but my experience is that enough do to create an unhealthy degree of politicization within the community. Michael is to be applauded for his candor, but what it reveals is not a pretty sight.

So here is an offer to Michael Tobis: Write up a serious critique of our paper's logic, data and analysis and I will publish it prominently here. Prove that our paper is in fact "nonsense" based on science and that your evaluation of it is not simply tribal politics played out on a blog.