A recent post on Scholarly Kitchen raised some interesting points about the ethics surrounding citation, and specifically self-citation. Previously, COPE has discussed related issues surrounding self-citation by journals and editors and citation of preprints. During this forum, we’d like to broaden the discussion to include some of the questions related to self-citation by authors in scholarly publication.

Questions for discussion:

Where is the line between what’s appropriate in terms of self-citation, and what’s not?

If an author did not cite any of their previous works, this could be considered to be an inappropriate deception, indicating that the present work is novel, and unrelated to past works. But on the other hand, too many citations to previous works by the same authors are also inappropriate, considered to be a potential attempt to manipulate an author’s own h-index. So where’s the line?

What is a peer reviewer’s responsibility in terms of calling out self-citation?

While journals often expect that reviewers will read a paper inside and out, front to back, how much time does a peer reviewer actually spend looking at the reference list? And even if they do take a look through the reference list, is it biased to call out an author for citing too many of their own works, especially if in their report, the reviewer asks for the author to add references to the reviewer’s work?

Should journals have specific policies about self-citation, and the peer review of reference lists?

Is there a certain number, or percentage, of references in a given work that can or should be a limit in referring back to the authors’ own publications? Is there any guidance journals or publishers can give to authors about what is considered “appropriate” vs. “excessive”? Are there, or should there be rules about changing reference lists during revision, in particular in adding more self-citations if not specifically requested by the editors/reviewers?

The Forum is open to COPE members who need to register by Friday 10 November. Comments on this discussion are open to all below:

This will be discussed at the next COPE Forum on Monday 13 November 2017. Please do leave any comments below, whether or not you are planning on joining the meeting

Comments are reviewed and, on approval, added below.

Please login as a member or register as a non-member to post comments


  • Posted by Daniel Kulp, 10/11/2017 6.58pm

This is a very difficult and delicate issue. In general, I would expect some self-citation since most work is built off previous studies by the same group. When this gets to be "too much" is difficult to assess. For the APS journals, we directly ask the referees "Are there appropriate and adequate references to related and previous work?" We definite want to insure that the work is properly placed within the context of previously published work.

  • Posted by Zoe Mullan, 13/11/2017 9.37am

For a primary research article, all types of citation should be used to justify the need to do the current research and to put the findings into the context of existing evidence. That could legitimately involve self-citation, and, in some very niche fields, it could involve quite a lot of self-citation. So for original research, I don't think it's appropriate to put a percentage on it. For review papers and opinion pieces, one could be a lot more critical.

In looking for appropriate peer reviewers, an editor can probably get a feel for whether or not the authors are dominant in the field or whether they are omitting others' work. But reviewers are probably better placed to assess this, and so I think an instruction phrased along the lines of Dan's is a good one.

  • Posted by Charon Pierson, 13/11/2017 3.46pm

I agree with both Dan and Zoe on this. I think this issue may be more straightforward in the sciences where as mentioned, research is an iterative process and the full picture can only be viewed with an understanding of previous work by the author and other colleagues doing similar work. In the humanities, and social sciences, there is likely some difference, although, a scholar could easily develop a highly focused research trajectory requiring a lot of self citation. Even in that situation, there is a need to cite contradictory work to flesh out the arguments. There is a need for careful review of the content and the references in the peer review process, but there is also a need for careful consideration by the editor to determine if the citations are appropriate. I do not believe any kind of "rule" is possible limiting the percentage of self-cites. One other factor to consider is the education of early career researchers and doctoral students on the purpose of citations. I don't remember learning anything like that in my education - it was very much an individual situation depending on the professors. Above all, students need to understand the importance of accuracy of citations and references.

  • Posted by Geraldine Pearson, 13/11/2017 3.54pm

I think my additional comment would involve education of reviewers around self citation and the balance they look for as they review papers. References need to be examined just like the rest of the submission.