American Scientist

To the Editors:

I read “Authorship Diplomacy” (May–June) with great interest and commend Melissa Anderson and colleagues for providing real insight into a very vexed topic. I do think, however, that there is a solution to the problem of properly assigning authorship in the case of multi-author articles—a solution that is conceptually simple if politically complex. Journals can simply insist as a matter of editorial policy that each author reveal his or her contribution to the article in an accompanying appendix. What should be done when there are a hundred or more authors? They can be listed with their contributions only in the article’s Web version. If history is our guide, of course, journals will not easily adopt such a policy. Up to now, for example, many journals have resisted full transparency in regard to retractions, as documented by Reuters Health editor Ivan Oransky and Anaesthesiology News editor Adam Marcus in their blog, Retraction Watch.

Alan Gross
University of Minnesota

To the Editors:

In my laboratory at NASA Ames Research Center, I employed a simple procedure to determine which authors should be included and how credit should be allocated on scientific manuscripts. According to my system, the first author is the person who formulated the hypotheses, designed the study, supervised data collection and wrote the introduction and discussion, if not more. The first author must have been present at essentially all testing sessions. Other authors include all those without whom the study could not have been conducted, especially the technicians and graduate students. Order of authors is determined by time spent during the study, including data analysis and manuscript preparation. No ghostwriting or honorary authorship is allowed. Managers and administrators who did contribute, but not as scientists, can be included in the acknowledgments.

John Greenleaf
Saratoga, CA

Dr. Anderson responds:

Journals are increasingly adopting the sensible approach that Alan Gross recommends. Unfortunately, notations of author contributions can be manipulated almost as easily as the author lists themselves. Laboratory-specific rules for authorship, like those that John Greenleaf suggests, provide good guidance, but they sometimes conflict with international collaborators’ expectations, which may have more to do with hierarchy and politics than contributions. In any case, it is important to discuss authorship at the start of any collaboration and whenever collaborators begin work on a new paper.