Disputes over who truly deserves the credit — or blame — for published work can result in bruised egos, damaged careers and court cases. Helen Pearson looks at ways to avoid fights over authorship.
A flurry of squabbles about high-profile biological research is prompting scientists to revisit a perennially touchy subject: how should credit for scientific findings be assigned?
In recent months, a panel on research integrity at the University of Pittsburgh, Pennsylvania, scolded cloning expert Gerald Schatten for his limited contribution to papers he co-authored with a South Korean team. Ian Wilmut of the Roslin Institute in Edinburgh has been criticized for taking most of the credit in 1997 for the cloning of Dolly the sheep. And a co-author has accused Alison Murdoch, of the University of Newcastle upon Tyne, UK, of hogging the credit for an advance in cloning human embryos (see ‘Cloning clashes’).
These quarrels all involve cloning, but they resonate with a wider audience. Scientists often work for years towards a breakthrough; the recognition for it can bring prestige, promotion and pay. Small wonder, then, that they care about how credit is assigned.
“This is about one of the most important questions you can ask, because credit more than anything drives science,” says Drummond Rennie, deputy editor of the Journal of the American Medical Association (JAMA). But he and others say that there are some simple courtesies and practices that could, if widely adopted, help ease any tension.
Most concern focuses on who gets listed as an author on a publication. Tallies of such publications are one of the main measures of scientific achievement. And it is particularly important for a key finding: “The difference in authorship at the time can end up being the difference between a Nobel Prize and not,” says Nicholas Steneck, an expert in research ethics at the University of Michigan in Ann Arbor.
By some measures, abuses of and disputes over credit are commonplace. In 2005, a team of researchers from Minnesota found that 10% of US scientists surveyed admitted inappropriately assigning authorship in a paper1. And in a study of young physicists, some 4% reported that they had seen cases in which relevant studies were not cited in a paper2. The practice infuriates many scientists, because they feel they are being cheated of acknowledgement, and because citation counts are another measure of a paper's — and an author's — worth.
The central problem is that there are no set rules for assigning credit. Biologists tend to place a supervisor or lab head last in an author list; organic chemists might put him or her first. Until recently, it was standard for the head of a German department or institution to take credit on a paper regardless of input. Graduate students and postdoctoral fellows frequently complain of being pushed down the author list, if they are included at all.
These differences in expectation create fertile ground for disputes. And many say that the number and ferocity of such clashes is rising with the number of collaborative projects, which often have multiple authors spanning disciplines and national borders. In genome sequencing and particle-physics collaborations, for example, a paper's author list can run into the hundreds. Some of these fights become public when there is disagreement about who should talk to, and be acknowledged in, the press. “Some people are very good at co-opting the media and blowing their own horn,” says Martin Blume, editor-in-chief of the American Physical Society.
Many institutions, societies and journals have guidelines on what deeds warrant a name on a paper. According to the International Committee of Medical Journal Editors (ICMJE), authors must have made a substantial intellectual contribution to a study's conception and design, or to the acquisition, analysis or interpretation of data. They must also have drafted or revised the article's intellectual content, and approved the final version.
On this basis, a technician who performed routine genetic sequencing, a colleague who supplied an antibody, or a departmental bigwig who supplied funding would be relegated to the acknowledgements. The ICMJE guidelines also say that authors should be able to take public responsibility for appropriate portions of the work — in other words, they must take the flak if the results come under fire, as Pittsburgh's Schatten found.
Whatever the rules, many scientists feel that the reality is different. “We all know labs where powerful figures appear on papers that are not within their field of expertise,” says Ray Dolan, who heads a neuroscience imaging department at University College London. A particular problem, he says, is researchers who invent a technique and then demand credit every time it is used, even after it becomes standard procedure.
One way to avoid ambiguity is for each author to spell out their contribution in the paper. Many medical journals, including JAMA, require this. The practice is voluntary at Nature; editors at Science are considering whether to introduce it. “It's moral, it's easy and it takes up virtually no space,” says Rennie, who has pushed for adoption of the practice.
Training scientists in the ethics of authorship could also help; many such courses already exist. In his ethics course, neuroscientist Michael Zigmond, of the University of Pittsburgh, asks graduate students and postdocs to consider a case study of ten people and reach a consensus on whose contribution merits authorship. The participants, he says, often disagree at first, but eventually agree on criteria.
Indeed, many scientists say that the simplest way to avoid fights is to discuss authorship when a collaboration begins. “Proactive discussion is really important,” says Kate Kirby of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts, who co-authored the survey of young physicists.
The stakes are perhaps highest for the most prestigious scientific prize, the Nobels. Goran Hansson, who chairs the Nobel Committee for Physiology and Medicine, says it would help his task if the research community could agree on criteria for co-authorship. “We would still have to be very meticulous,” he says, “but it would be a bit easier.”