The Scientist is the source of our Casablanca flashback, with its report that an open access journal published by Bentham was willing to publish a ‘nonsense’ paper that supposedly passed peer review.  A Ph.D. student in science communications and a staffer at The New England Journal of Medicine have been testing journals peer review practices by submitting papers generated by computer program.  They document this particular incident on their blog.  In short, the journal agreed to publish the article, if the authors paid the fee, and asserted it had passed peer review.

At a minimum the publisher Bentham is guilty of allowing journals to assert peer review when none had taken place.  The scamming conclusion is reasonable, given the reports.  I’m not in agreement that open access journals are necessarily more suspect of putting out supposedly peer-reviewed articles that weren’t so reviewed.  Yes, they do charge more fees than traditional journals (who could be scamming authors for photo and chart fees, amongst other things), but an open access journal is not more likely to skimp on peer review than any other journal.

What bothers me is that it has to take generating obviously lousy articles to ferret out derelict peer review.  Given the volume of scientific publishing, there’s an enormous amount of implicit trust in the processes behind these articles that people will continue to exploit.  I wish I had even the germ of a possible solution here.