A journal accepted a paper but after receiving the decision letter, the corresponding author asked if one of the coauthors could be removed from the authorship list. This coauthor is now a prominent politician and felt that their new role would conflict with being an author on the paper.
The journal informed the corresponding author that they would need written confirmation from both the coauthor who wished to be removed from the authorship list and all remaining coauthors. They received this from every coauthor except one. The corresponding author reported that this coauthor lived in a country with potentially limited communication infrastructures and did not respond to several attempts at contact via email and telephone.
Questions for COPE Council
- Is it reasonable for the journal to go ahead with the removal of the name of the coauthor even though one of the coauthors has not given consent?
- Or should the journal still publish the article with the full authorship list on the assumption that they will never receive a reply from this coauthor?
Advice on this case is from a small number of COPE Council Members. Most cases on the COPE website are presented to the COPE Forum where advice is offered by a wider group of COPE Members and COPE Council Members. Advice on individual cases is not formal COPE guidance.
In general, changes to a manuscript’s list of authors should not occur so late in the publication process. The reason for this is that in general, by the time a paper has been accepted, the work of authorship has been done, and so the credit and responsibility of authorship should have already been determined with relative finality.
Changing authors at this late stage should raise an editor’s level of concern for ghost authorship (removing an individual who has met all the criteria for authorship). COPE, ICMJE, and many others have recommendations against ghost authorship (see the COPE flowchart here). Even if an author wishes to voluntarily take their name off a paper, there are good reasons why this should not happen. These include the hiding of relevant conflicts of interest or other relationships. Also, if the author who wishes to be removed has committed fraud or misconduct and this is discovered after publication, the remaining authors then would be responsible and it would be difficult or impossible to shift that responsibility to the ghost author.
If the reason the author wishing to be removed is related to a conflict of interest at a new job, then that could be added as a conflict of interest if at the time of the research the author was also pursuing activities that may have affected the research, and the authors and possibly peer reviewers may need to revisit the validity of the paper. If it is a new conflict of interest, there could be an editorial note saying exactly which part/s of the paper that author now disagrees with.
However, there are situations where an author might reasonably be removed from a paper; perhaps, for example, a disagreement with co-authors about the final message such that the individual no longer could stand behind the claims or arguments made in a manuscript. In this case, the editor might consider accompanying the published paper with a note from the editor or allow the author to be moved to the Acknowledgments if the contributions of that author are transparently stated. Ideally, the decision regarding who qualifies as an author and who takes responsibility for the article should lie with the authors themselves (and their institutions). Journals cannot be responsible for policing this.
The relevant COPE flowchart is Changes in authorship - removal of author before publication
According to the flowchart, all authors need to agree to the removal, and the author being removed needs to submit correspondence about their agreement and the reason. It is possible the coauthor that cannot be located may disagree, so the editor needs to wait until the last coauthor is contacted; the authors need to find a way to contact the last author, including contacting the last institution. The editor could ask for the institutions to be involved if the matter cannot be resolved. Making an exception to the rule that all coauthors must agree to a removal could set a bad precedent for the future. Until the authorship question is resolved, the editor should not publish the manuscript.
If the editor receives written confirmation from all of the coauthors agreeing to the removal of the name of the author, all contributions of that author would need to be removed with a full explanation, and the paper revised and reviewed again. The copyright transfer agreement may also need to be signed again.