n today’s lookit-me culture of selfies, Twitter, Facebook and (ahem) endless blogging, the notion of anonymity is about as welcome as a case of hemorrhoids. But Paul Hanel thinks it may be key to correcting some fundamental problems in science.
Hanel, a psychologist at Cardiff University in the United Kingdom, posted a manuscript recently calling for anonymity in science articles. More than that, Hanel suggests stripping identifiers from virtually all academic output: doing away with name-based citations, CVs on researchers’ web sites, author names on book chapters, titles on academic journals, and more.
The immodest proposal — made available on arXiv, a preprint server, before peer review — is akin to destroying the academic village in order to rid it of pests. But while some of what Hanel recommends is impossible at best, and perhaps even counterproductive, his overarching point seems pretty solid. When it comes to protecting the scientific literature from bias, the safeguards that academics now use are sorely inadequate.
article continues after advertisement
Consider: Blinded peer review, which is intended to act as a bulwark against shoddy science, turns out to be largely ineffective, prone to manipulation, and not so blind after all. Journal editors appear to be more welcoming of the work of the big names in the field, while reviewers seem to reinforce this bias by casting less critical eyes on manuscripts from high-profile researchers.
Nature, arguably the world’s most prestigious scientific journal, recognized this earlier in the year when they announced that authors “will be able to request that their names and affiliations are withheld from reviewers of their papers — a form of peer review known as double blind. At present, the process is single blind: reviewers are anonymous, but they know the authors’ identities.” The move, said the journal, could get rid of “personal biases, such as those based on gender, seniority, reputation, and affiliation.”
Similarly, once published, the papers from scientists with name recognition receive more citations than those of less well-known scholars, even after accounting for the quality of the studies. Science also is rife with gender inequity, with female scholars suffering the “Matilda effect” in which their feminine names put their work at a relative disadvantage to that of their male colleagues — although what effect that has on peer review outcomes is unclear.
In short, Hanel is pointing out an uncomfortable fact: Science is not a pure meritocracy, or even close to one. Scientists are human, subject to the same biases and foibles as everyone else. And one of those failings quite clearly is a weakness for worth by association.
Hanel is not the only academic to play this tune. Earlier this month, Michael Eisen, a biology professor at the University of California, Berkeley, created an online CV stripped of all of the names of journals he’s published in. Members of the lab “do not believe that journal titles convey useful information about the quality or value of published works,” he wrote, and others agreed.
But not everyone thought Eisen’s idea was a good one. One scientist, Claus Wilke of the University of Texas at Austin, wrote that “hiding journal names from the publication list is directly at odds with the principles of openness and egalitarianism that people like Michael Eisen so strongly promote. Therefore, to put it bluntly, I think this practice stinks.”
The issue is certainly one that divides scientists. Even some who might agree with Eisen’s basic premise — that work should be judged purely on its merit, not on who published it — may not be comfortable with other kinds of anonymity. Philip Moriarty, of the University of Nottingham, thinks impact factor — a much-abused way to rank journals by how often their papers are cited — is a useless metric. At the same time, he is concerned with the growth of anonymity in forums dedicated to picking apart studies after they’re published.
Indeed, it’s easy to see how accountability and transparency likely would be among the first casualties in the nameless regime. Identifying and tracking conflicts of interest and instances of fraud would become far more difficult, while public scrutiny of waste and abuse would be virtually impossible. The best solution to all of these problems remains elusive.
In the meantime, perhaps researchers could repeat a 1987 experiment by William Hoover, a physicist at Lawrence Livermore National Lab. Hoover was a bit frustrated after having had a paper rejected by two journals. So he made up a name, Stronzo Bestiale, added it to his coauthor list, and resubmitted the paper with a different title. One of the journals that rejected the original paper accepted the new version.
Just one problem: Stronzo Bestiale isn’t just a fake name, it’s also (as a quick Google will tell you) a very rude phrase in Italian.
The paper still stands — proof that maybe names don’t matter after all.