Academic Authorship - Growing Number of Authors Per Paper

Growing Number of Authors Per Paper

From the late 17th century to the 1920s, sole authorship was the norm, and the one-paper-one-author model worked well for distributing credit. Today, shared authorship is common in most academic disciplines, with the exception of the humanities, where sole authorship is still the predominant model. In particular types of research, including particle physics, genome sequencing and clinical trials, a paper's author list can run into the hundreds. In large, multi-center clinical trials authorship is often used as a reward for recruiting patients. A paper published in the New England Journal of Medicine in 1993 reported on a clinical trial conducted in 1,081 hospitals in 15 different countries, involving a total of 41,021 patients. There were 972 authors listed in an appendix and authorship was assigned to a group. In the summer of 2008, an article in high-energy physics was published describing the Large Hadron Collider, a 27 mile long particle accelerator that crosses the Swiss-French border; the article boasts 2,926 authors from 169 research institutions.

Long author lists strain guidelines that insist that each author's role be described and that each author is responsible for the validity of the whole work. One Big Science facility, the Collider Detector at Fermilab (CDF), in 1998 adopted a highly unorthodox policy for assigning authorship. CDF maintains a standard author list. All scientists and engineers working at CDF are added to the standard author list after one year of full-time work; names stay on the list until one year after the worker leaves CDF. Every publication coming out of CDF uses the entire standard author list, in alphabetical order. Such a system treats authorship more as credit for scientific service at the facility in general rather that as an identification of specific contributions.

Large authors lists have attracted some criticism. One commentator wrote, "In more than 25 years working as a scientific editor ... I have not been aware of any valid argument for more than three authors per paper, although I recognize that this may not be true for every field." The rise of shared authorship has been attributed to Big Science -- scientific experiments that require collaboration and specialization of many individuals.

Alternatively, the increase in multi-authorship might be a consequence of the way scientists are evaluated. Traditionally, scientists were judged by the number of papers they published, and later by the impact of those papers. The former is an estimate of quantity and the latter of quality. Both methods were adequate when single authorship was the norm, but vastly inflate individual contribution when papers are multi-authored. When each author claims each paper and each citation as his/her own, papers and citations are magically multiplied by the number of authors. Furthermore, there is no cost to giving authorship to invidivuals who made only minor contribution and there is an incentive to do so. Hence, the system rewards heavily multi-authored papers. This problem is openly acknowledged, and it could easily be "corrected" by dividing each paper and its citations by the number of authors.

Finally, the rise in shared authorship may also reflect increased acknowledgment of the contributions of lower level workers, including graduate students and technicians, as well as honorary authorship.