Peer review vs. editorial review

A few years ago, Bruce Charlton was fired as the editor of Medical Hypotheses for his principled refusal to withdraw a paper. The paper in question was by Peter Duesberg, the well-known HIV/AIDS skeptic and a molecular virologist at UC Berkeley. It started a firestorm of, well, not exactly criticism, but of calls for the head of whoever let the paper see the light of day, that head belonging to Bruce Charlton. The scientific publisher Elsevier listened respectfully to the baying wolves, and when Dr. Charlton would not bow to Elsevier’s demands, notably both withdrawing the paper and changing the format of Medical Hypotheses from one of editorial review to one of peer review, they sacked him.

Dr. Georg Steinhauser et many al.s have written a paper that touches on this controversy, Peer review versus editorial review and their role in innovative science. The abstract:

Peer review is a widely accepted instrument for raising the quality of science. Peer review limits the enormous unstructured influx of information and the sheer amount of dubious data, which in its absence would plunge science into chaos. In particular, peer review offers the benefit of eliminating papers that suffer from poor craftsmanship or methodological shortcomings, especially in the experimental sciences. However, we believe that peer review is not always appropriate for the evaluation of controversial hypothetical science. We argue that the process of peer review can be prone to bias towards ideas that affirm the prior convictions of reviewers and against innovation and radical new ideas. Innovative hypotheses are thus highly vulnerable to being “filtered out” or made to accord with conventional wisdom by the peer review process. Consequently, having introduced peer review, the Elsevier journal Medical Hypotheses may be unable to continue its tradition as a radical journal allowing discussion of improbable or unconventional ideas. Hence we conclude by asking the publisher to consider re-introducing the system of editorial review to Medical Hypotheses.

This paper ironically had to go through peer review, and also ironically is being published by an Elsevier journal, Theoretical Medicine and Bioethics, having been rejected by several other journals and then subject to extensive editorial revision (something of course not uncommon to scientific papers).

Having written a fair amount on this topic at the time of l’affaire Medical Hypotheses, I was asked to place my name among the names of the authors, which I happily did, although I’m not responsible for any of its content. The other signers number something close to 200 I believe, and most of the other names I don’t recognize. However, a few of them I do, among them Seth Roberts and Michael A. Woodley.

image_pdf

Leave a Comment:

4 comments
JP says October 26, 2012

Peer review can be prone to bias towards ideas that affirm the prior convictions of reviewers and against innovation and radical new ideas.

OK, but editorial review can be prone to bias towards ideas that affirm the prior convictions of the editor and against innovation and radical new ideas.

Why is it better to have one dumb or biased guy than three?

Reply
Chuck Currie says October 26, 2012

It seems, lately, peer review has been flying a lot of dirty laundry, and I suspect there is a lot more to come, and therefore, quite suspect in and of it’s self. Peer review has become friend review to the skeptic.

Will Professor Google save us all?

Cheers

Reply
Mangan says October 26, 2012

JP: I don’t think the paper or anyone really advocates doing away with peer review altogether. It’s just the opposite, in fact, and MH was, I believe, just about the only editorial-reviewed journal in existence. That couldn’t be tolerated.

Reply
eah says October 27, 2012

Recently on Sailer’s site:
Social psychologists are conformist and discriminatory

And so why wouldn’t the output of any peer review by ‘social psychologists’ be affected by that? Why would it be expected that any science, even the hard sciences, would be largely immune to a similar expression of bias?

Reply
Add Your Reply