Dangers of scientific publicity machines
A hyped-up fossil find highlights the potential dangers of publicity machines, according to one of the Editorials in today's Nature (459, 484; 2009 - free to access online). The Editorial describes last week's publication of paper describing a 47-million-year-old fossil primate with a remarkable degree of preservation, which quickly led to enormous media and internet coverage, including claims that the fossil is a "missing link" in human evolution. The Editorial describes how, "in the paper the authors explicitly state that Darwinius masillae "could represent a stem group from which later anthropoid primates evolved, but we are not advocating this here, nor do we consider either Darwinius or adapoids to be anthropoids". The authors also refrain from claiming that the fossil changes our understanding of primate evolution."
But the circumstances surrounding the paper's publication were anything but normal. Before the paper had even been submitted, a television documentary and an accompanying book about the find had been commissioned; a week publication of the paper, the book is out and the documentary has been aired on TV in several countries. The Editorial continues:
"Both book and documentary include the the suggestive words 'The Link' in their titles. A press release associated with the New York press conference at which the fossil was first officially described claimed that the fossil represents revolutionary changes in understanding. The History Channel website calls the find a "human ancestor", and the BBC website describes it as "our earliest ancestor".
To be fair, the authors' claims at the press conference were appropriately measured. Nonetheless, the researchers were fully involved in the documentaries and the media campaign, which associate them with a drastic misrepresentation of their research.
Another damaging aspect of the events was the unavailability of the paper ahead of the press conference and initial media coverage. This prevented scientists other than those in the team from assessing the work and thereby ensuring that journalists could give a balanced account of the research.
There is no reason to think that PLoS ONE's editors and reviewers did less than their duty to the paper. Nonetheless, the clock was ticking at the time of submission. Nature has over the years received occasional offers of papers associated with television documentaries, and the offers usually come with broadcast dates attached. Where the refereeing process might have been compromised, we have always said no to these papers. When time is tight, there is a risk that the broadcast will go out even if any problems uncovered by peer review cause the paper to be delayed or rejected.
In principle, there is no reason why science should not be accompanied by highly proactive publicity machines. But in practice, such arrangements introduce conflicting incentives that can all too easily undermine the process of the assessment and communication of science."
Amid all the vast quantity of media and internet coverage of this research, I highly recommend this excellent post, 'Why Darwinius is not our ancestor', by Karen James at her Data Not Shown blog. I also recommend this Nature Network blog post by senior Nature editor Henry Gee.
Nature journals' embargo policy.
Nature journals' policy on confidentiality and pre-publicity.
Comments
"Another damaging aspect of the events was the unavailability of the paper ahead of the press conference and initial media coverage."
It's unfortunate timing that the same issue of Nature contains a piece on peer review which is about a paper that has also not been published.
Posted by: Bob O'H | May 28, 2009 03:44 PM
Maxine adds: the NatureJobs article to which you refer is about a reanalysis, by the two sets of authors concerned, of the data underlying their two earlier published papers about possible gender bias. Did you find out whether a preprint of the reanalysis is available upon request? I'd be very surprised if it isn't.
In fact, the reanalysis is said to show that the findings of gender bias during peer-review of grant applications by one of the earlier studies is incorrect. And in this NatureJobs piece, the authors are correcting their earlier paper. They write: "At least for grant applications, all of the co-authors from each of the research teams agree that the weight of evidence suggests that the applicant's gender has no effect on the outcome of peer review, and that these findings are robust and broadly generalizable." So it seems to me, that contrary to "hyping" and "publicity machines", these authors are promptly correcting their published record.
Posted by: Maxine | May 28, 2009 04:13 PM
Maxine, I should ask about a preprint anyway (there are some red flags about subgroup analysis), but the Ida Who Lived paper was also available when the publicity drive started.
It seems to me that the correction of the record is what the paper will be about (when it appears). Neither of the earlier papers were published in Nature, so I can't see why that is the place to announce their new findings.
I guess the authors' mistake was not to hype their findings. If only they'd done that properly, they could have had a whole TV series out of this! (mmm, that's given me an idea for a Friday blog post...)
Posted by: Bob O'H | May 28, 2009 07:31 PM
I have obtained a clarification from the NatureJobs editors. The "in press" paper described is not under embargo. Interested readers can obtain a copy directly from the author of the NatureJobs Prospect, or by sending an email to NatureJobs.
Posted by: Maxine | June 1, 2009 08:52 AM