The suicide earlier this week of Yoshiki Sasai, one of the scientists who worked on the now-discredited STAP stem cell work, was a startling and sobering reminder to the research community and the public that misconduct can take a heavy human toll – even on people like Sasai, whom by all accounts only had the misfortune of working with a dishonest colleague.
The tension surrounding this case and others was well-captured by University of California, Berkeley biologist Michael Eisen, who himself lost his father, also a scientist, to suicide in 1987 in a case that he said had haunting parallels to Sasai’s:
Obviously, fraud is a terrible thing. Nothing provides as deep an existential threat to the scientific enterprise than making up data. But as bad as it is, there is something deeply ugly about the way the scientific community responds to misconduct. We need to deal swiftly with fraud when it is identified. But time after time I have watched the way not only the accused, but everyone around them, is treated with such sanctimonious disdain it is frankly not surprising that some of them respond in tragic ways.
Can we ask whether prominent Web sites like Retraction Watch contribute to this toxic “witch hunt” environment?
We have heard this implied — and sometimes very direct — criticism before. Sometimes it is from people who believe that what we do is dangerous to science because it will diminish trust and therefore funding — a “shoot the messenger” argument we respond to here.
But the suicide of a scientist understandably unearths raw feelings and emotions. Mercifully, of the hundreds of cases like this that we have covered over the past four years, only two have involved the intentional death of a researcher. Although even that number is too high, the vast majority of scientists who fabricate their data or otherwise commit professional misdeeds endure the publicity and shame their actions ultimately bring on without taking their own lives.
The question, however, is whether the risk that someone might not be able to withstand public scrutiny of his or her actions justifies quashing open discussion of such cases. We believe, emphatically, that the answer is no.
Our operating system at Retraction Watch runs on a fairly basic principle: Bright sunlight, as Louis Brandeis put it, is the best disinfectant, which means that fair and honest reporting of retractions and other aspects of scholarly publishing is the best defense against cheating and obfuscation. We simply report the facts as we know them in a style we believe to be engaging and respectful. Others agree.
One of the criticisms we’ve seen in recent days — and which echoes past critiques — is that our focus on retractions leaves out the larger context of science, particularly the pressures that today’s publish or perish incentives bring to bear on researchers. That context, however, informs our work every day, and has formed the basis of some of our big pieces, including this invited Comment in Nature in 2011, for example. Some believe that we don’t do enough to emphasize the small fraction of published papers that retractions represent. We do, however. Here’s just one example, an interview Ivan did on a Boston NPR affiliate this week that begins with him placing the numbers in context.
Still, we could always include more context, so we appreciate this constructive criticism. But we firmly believe that cataloging and probing the symptoms of some of these problems — in our case, that means retractions — is a good way to check the health of transparency in science.
What goes along with that is our belief that a vigorous and open debate is crucial to science. For that reason, we allow our thousands of commenters substantial latitude in their posted opinions. We draw the line at remarks that are libelous, that generate heat without providing any light, or are otherwise inappropriate.
Indeed, we receive as much criticism from those who accuse us of being too strict and deliberate with our screening of comments as we do from those who say some of the comments we let through are too intemperate. What has become clear this week, in public comments and in private conversations, is that when some critics say that Retraction Watch is contributing to a witch hunt-like atmosphere, they are more than likely thinking of our comments than our posts.
We can always do better in our comment moderation. But we hold firm to the notion that the more robust the conversation, the better the science. We are, as we’ve said a number of times, big fans of PubPeer, where those conversations have led to important corrections and retractions.
Those conversations are not easy for those whose work is being questioned. And we are neither ignorant nor indifferent to the potential personal ramifications for those whose publications are being discussed. But to repeat: The more robust the conversation, the better the science.