Credit: Illustration by Kai Grube

When it comes to publications, practically every researcher is aware of the potentially tricky politics surrounding the author list. A well-known case of gratuitous authorship involves Gamow's inclusion of Bethe's name in the author list of a cosmogenesis paper (Physical Review 73, 803–804; 1948) so that it would read Alpher, Bethe and Gamow — like the Greek alphabet. Unfortunately, most cases of authorship liberties are not so funny, or without serious consequence. And while crediting practices vary across communities, whether it be alphabetical or in order of 'importance', most people have a basic understanding of what is fair. All too often, though, this line is crossed.

A common form of deliberate authorship abuse occurs when the head of the group claims first authorship instead of the postdoc or student who actually did the work and, in some cases, wrote the paper. Other times, authors who haven't contributed very much are added for the promotion of their careers. Recently, another variant has come to light.

A professor had obtained a grant for a joint research project with a collaborator from a different country. For the professor, it meant travel money only, but his collaborator could claim research funding, owing to the differences between the two governments administering the international partnership. In fact, no collaborative work took place, for various reasons, so the professor was surprised when he accidentally came across a publication bearing his name, together with that of his erstwhile collaborator. Moreover, the subject of the paper was a material that the professor has consciously avoided during his entire career. Indeed, the collaborator has since acknowledged that no joint research took place, but that he assumed because they had obtained the grant together that both their names had to appear on all publications relating to the project.

Beyond this specific example, is it generally considered acceptable to include co-authors who haven't made a contribution? It's a bit of a no man's land within publishing, so some journals have tried to make author credits more explicit. The Proceedings of the National Academy of Sciences has an editorial policy that specifies “authorship should be limited to those who have contributed substantially to the work” and furthermore, “authors are strongly encouraged to indicate their specific contributions” as a footnote (http://www.pnas.orgmisc/iforc.shtml#Editorial%20Policies). Similarly, Nature and its sister journals also encourage such contributions in the Acknowledgements (see section 5.5 of the Nature Guide to Authors at http://www.nature.com/nature/submit/gta/index.html). It was the hope of Nature that this practice would spread and allow authors, editors and readers to appreciate who did what, and encourage fairness, but since this policy was introduced in 1999, not many authors have taken advantage of it. Whether this is due to a general lack of awareness or a more endemic problem is unclear. Should journals push harder for detailed author contributions? And if they did so, would it actually help?

So far, the system has largely relied on author integrity. In an ideal world, it would be the responsibility of all the co-authors to agree on an author list that reflects their degrees of involvement. And, more importantly, the co-authors would be able to vouch for the quality of the paper. But how many people would withdraw their names from an article that's likely to be published, even if they don't support (or understand) the results? Many researchers are desperate for publications, and so allow their judgment to be clouded by this need for 'CV points' because they think the next research grant or job depends on it. And how many students and postdocs would dare to stand up to their so-called superior and risk future repercussions when it comes to reference letters? Moreover, the inexperienced may think it's the norm.

Many blame the publish-or-perish culture, and as long as researchers are judged by the number of publications — one of the few quantifiable criteria for merit — the problem of giving (or not giving) contributors due credit will remain. Nature Materials is aware that this kind of misconduct happens, and when a specific case comes to light, swift action generally follows — in the form of a retraction in the worst case. Of course, it is impossible to catch all instances, but journals shouldn't have to. The scientific community must ultimately be responsible. Academic societies and educational institutions should take a leading role in establishing and propagating a code of fair practice, for without any well-defined rules, there can be no punishment.