It surprises a lot of people, I think, that there’s no official way in which science is Done. instead, there’s a collection of customs and practices, including things like referring to previous work in the same area, making your data available for others to verify your results, and the whole process of peer review. The basic idea of peer review is pretty simple: when you submit a paper to a peer-reviewed journal, the paper is shown to anonymous, confidential, and relatively disinterested reviewers who read the paper, comment on it, and recommend for or against publication. Usually, an affirmative recommendation is “publish with recommended changes” because reviewers, like all editors, figure they could have done it better if they’d just had the time.
People may recall one of my first Climategate pieces in which I argued that the people at University of East Anglia had violated that social contract in several ways.
I think it’s fair to say this continues in climate science. Yesterday Ryan O’Donnell published a lengthy post at Climate Audit and other places, in which he broke with the confidentiality usually expected to reveal that Eric Steig had been one of the reviewers of a paper (I’ll update with a link to the actual paper in a bit). His justification — which I think is completely appropriate, by the way — is that Steig’s reviews and following behavior is itself over the line, and a deep violation of those scientific norms. What O’Donnell explains is that:
- Steig, first of all, was asked to review a paper that directly criticized some of Steig’s work. (Disinterested?)
- In that review, Steig asked for major changes to the methodology. As O’Donnell puts it, there were eight times as many pages of corrections as the original paper.
- Steig, having apparently voted “publish with changes” on the paper, then attacked the paper in the Real Climate blog, which is a major advocacy blog for the “anthropogenic global warming” hypothesis.
- And in fact attacked it for using the very method he insisted on in his comments. (Among other things.)
Normally, Steig would have remained anonymous, except that O’Donnell recognized some of the critical points, and confirmed with Steig that he had indeed been the reviewer.
Frankly, this should lead to resignations by the journal editors, and looks to me like scientific misconduct.
(Updated: linked my “social contract” post, corrected spelling.)