Would papers – or research projects, even – benefit from copying the arts in having an official producer, director, technical director, and starring cast?

A lot of academic research is done in an inherently messy and ad hoc way. Students and postdocs propose or are given projects, which are led by their group leaders. The leadership can be close-range or at arms’-length, and the work on the project may or may not be assisted by a shifting cast of technicians, collaborators, and facilities. At some point, an end will be declared – either because it feels like a story is finished, or people have run out of time, or some pre-set objective has been met…and then the process of writing up begins.

Writing-up also marks the point at which the jostling for credit starts. Early on in a scientist’s career, it’s important to be as buoyant as possible and try to rise to the top of the author list, while once a certain level of seniority has been reached the goal is to settle solidly in the last position. Only in academia is sinking to the depths considered a mark of professional esteem.

And so the squabbling begins. Who’s first author of the paper? Will there be co-first authors? Who’s corresponding author? Will there be co-corresponding authors? Who’s last author? Will there be co-last authors? And where in that potentially long chain of names between first and last does everybody else go, fighting to either sink or swim according to the demands of their ego? Benchwork is about using your hands but regrettably, authorship is often about using your elbows.

It’s a situation fraught with the potential for bad blood and misgivings, and fallings-out over authorship are common. People may feel that their contributions have been undervalued, that their status in the project has been undermined, or that they are being somehow sidelined. 

And it stems from the fact that the contributors to scientific papers have only one credit mechanism: their name. 

There’s nothing that differentiates the names on an author list. It’s assumed that the first author has done the bulk of the hands-on work, and it’s assumed that the last author has handled the bulk of the oversight. If the corresponding author is not the last author, then it’s a way of acknowledging a bit more credit for leadership. And between those poles, all manner of degrees of contribution are possible – second authors might be viable co-first candidates, or have chipped in a few hours’ work towards the end, depending on the paper. 

Recently, there have been attempts to provide a more fine-grained picture of author contributions. PLoS, perhaps the most visible proponent, offers the chance for authors to be credited for conceptualisation, data curation, formal analysis, funding acquisition, investigation, methodology, resources, software, supervision, validation, visualisation, writing original draft, and writing review/editing, though these sections are buried at the very end of the manuscript.

The problem with these categories is that they’re somehow hard to link to the product itself (for example, when someone is credited for “resources”, how do you know what they’ve done?), and they’re generally post hoc classifications that don’t reflect how the project was conducted in real time on the ground. Funding acquisition, supervision, writing are intuitively understandable; the others…less so. They’re also inevitably simplifications and sometimes exaggerations – saying that all authors were involved in reviewing the manuscript often just means that all of them were given the opportunity to provide feedback, not necessarily that all of them actually did. 

It’s deeply unsatisfactory, especially so in contemporary science where accruing credit for research makes and breaks entire careers, and where people have become technically specialised to such an extent that their contributions are often both essential but incapable of standing alone.

Might a solution be found in the arts?

The arts are arguably well ahead of the sciences in their acknowledgement of contributions and expertise by those not in starring roles. It might be the lead actors and the directors whose names are up in lights, but film and theatre professionals are acutely conscious of the critical role – and expert knowledge – provided by other members of the team. Anyone who’s read the programme notes for a play or sat through the credits for a movie (much more likely these days, thanks to Marvel standardising the practise of teaser trailers at the end) will know that an enormous number of people are involved in bringing these projects to fruition. 

At the Oscars, approximately one-third of the awards are in technical categories. On the other side of the Atlantic, the recent recognition of casting directors by the BAFTAs is just the latest nod to the essential work done by people who never step on the stage or appear in front of the camera.

The arts also provide a more intuitively understandable framing of people’s roles in a project that transcends their exact contributions. Who put up the money and promoted the story? The producer. Who brought that story to life? The director. (Who realised that vision on the set? Usually the First Assistant Director…) Who gets most of the lines or attention? The lead actor. And so on.

So why not do this in science? Instead of a single author line for papers, why not have three? The starring cast, the technical crew, and the production team, with the production further subdivided into writing/directing/producing? The appealing thing with such an arrangement is that it would be intuitively understandable who (generally) did what, regardless of the nitty-gritty of individual contributions. It would make explicit what is currently encoded, sometimes cryptically, in the author lists.

Importantly, it would immediately change author lists from being a single chain into a set of parallel objectives. It would compress the confusing set of 13 PLoS criteria into a clearer and more intuitively understandable set of ~5 categories – categories that bear a closer resemblance to how the work is actually done and what the relationships between the authors are – and then place those on the front page instead of tucking them away at the end. It would also allow people whose main expertise is technical to be properly recognised for that instead of always occupying some nebulous space in the middle of the author lists. 

Such an approach would serve to highlight the diverse strengths of the team involved in making a paper. By offering ~5 categories for recognition, each with ordered name lists, it would also provide a more realistic window onto contributions (e.g. is the “star” also the first name in the writing credits or is their supervisor taking pole position there?) and a better reflection of the organic way that research projects develop. 

The downsides? Yes, you might just turn the current squabble over order of authorship into a categorised scrap instead of a general free-for-all. In fact, that would almost certainly happen. But that doesn’t make it worse than the status quo, where authors are clinging to the top and bottom of the author list while others disappear into the vacuum of the middle. 

This way would produce more “first authors”, and that’s got to be a more inclusive, more realistic, and more generous system than what we have right now. Films, plays, and albums would never dream of acknowledging all involved personnel in a single list with no further accreditation, and nor – especially in these days of ever greater specialisation – should the sciences. It’s time to look to the Oscars.