Every few years, it seems, we need to be reminded that no matter how educated and experienced any one of us is, we are all fallible. My most recent examples involve the fabrication or manipulation of scientific data for the sake of publication in a top-tier journal and the glory that goes along with that achievement (the publication, not the fabrication).
Science and Nature are both in the throes of dealing with the repercussions of South Korean stem cell biologist Hwang Woo-Suk, while the Lancet is faced with the prospect of having to retract a paper published last October by a Norwegian scientist on the efficacy of a drug targeting oral cancer. So much for the power of peer-review.
In an effort to democratize science, and as electronic publishing has become simpler, there have been moves on several fronts to create new scientific journals that lie purely in the virtual domain. But this effort has largely met with derision from the major scientific publishing houses, who argue that such efforts only serve to dilute the standards they have created and cannot offer the quality control that the time-tested peer-review process provides. With many of the biggest names in science invested in this process, as the reviewer and the reviewed, it becomes difficult for fledgling publications to become accepted.
Recent events, however, dictate that the "quality control" protections offered by peer review are largely illusory and that there will always be people who manage to sneak through fraudulent data. If anything, the latest computer technologies make it easier than ever to fabricate data and nothing short of a television CSI lab is going to be able to detect the fraud. And with all due respect, because of the impact of publication in top-tier journals on annual reports and grant applications, the temptations for scientists to target these publications with explosive, if fraudulent, findings are probably higher.
I don't think we should punish the journal editors or manuscript reviewers for missing the fraud. Having been involved in journal publishing myself, I can vouch that these are highly skilled, highly dedicated people working to the best of their abilities. If we ask them to be more vigilant than they already are, we are asking the already-slow scientific publishing process to grind to an absolute halt.
Do I advocate the elimination of peer-review? No. I think it is important for journal editors to get secondary opinions on the validity of a scientific pursuit. I would just like to see publishers get off their high-horses about it and see it for what it truly is: a fallible mechanism.
When we come right down to it, real peer review happens after the paper is published and the global scientific community gets a shot at it. If that means that a few unscrupulous scientists force us to shine a light on our communication processes, so be it. In the long run, they'll have done all of us a favor.
NOTE: Before publication, this commentary was peer-reviewed by two editors, two publishers, and my wife, none of whom I expect to go out on a limb for the opinions expressed.