There is a group in our small field that almost exclusively cites themselves. Their latest paper has 35% self-citations, which is 80% of the citations within the field. They only reference their own work in talks, presenting it (e.g. to industry) as if they are the only group doing research in this field. When they refer to the work of other groups they do it with a handwave, e.g. in a little corner of their slide with misspelled names.

Their name is attached to a well-known university and they had influential and excellent papers in the field, so most people cite them in their papers, but it increasingly feels as if we are helping a career-boosting strategy here. I have witnessed that they were already criticised in review rounds for ignoring other people's work and not giving credit, but they continued the self-citing in new publications, either seeing nothing wrong with it or following a strategy. They often seem to get away with it. About two years ago they had a publication where the main result was already published by another group in our field years ago, and they did not cite them.

Is there a counter strategy against this behaviour?

  • 47
    In the special case you mentioned in the end, that the result was already known and not cited, a message to the editor might help. Of course assume in the email that they simply didn't know of the previous work, don't go claiming that they deliberately ignored it. – Dirk Liebhold Sep 6 at 10:33
  • 2
    What is a typical self-citation rate in your (sub) field? – Azor Ahai Sep 6 at 16:42
  • 17
    Publish a paper in a statistics journal analyzing the effect of their self citing behavior and draw correlations between inflated scores, failure to cite others, and increased funding and job opportunities for the grad students. Then, either you will find something interesting and shame them, or find out that their self citing is not actually a big problem. – aidan.plenert.macdonald Sep 6 at 22:57
  • You can make the point graphically about the clique by plotting the network graph of publications, citations, people, groups and cliques – smci Sep 7 at 22:59
  • It really sucks but sadly enough if they are at some fancy place there is usually very little you can do about it (from bitter experience). – user_of_math Sep 9 at 14:55
up vote 64 down vote accepted

First, I think it's important to have a more precise diagnosis of the problem.

Self-citations by themselves are not necessarily a problem---even 30% self-citations might be quite reasonable depending on circumstances (e.g., work in a very narrow but deep niche). Likewise, re-inventing a result is not necessarily a problem if the prior result is in another community and your community was genuinely unaware of it.

Failing to properly credit the work of others when made aware of it, on the other hand, is definitely a problem, and can be addressed by the community of peers:

  • For journal articles, the peer reviewers can simply insist on citations. It's very easy to add a citation in revision and pretty much never a reason not to cite closely related work. Furthermore, reviewers get to see it again and don't have to say "accept" until they are satisfied. Thus, unless the editor thinks the failure to cite is OK, it's easy to force the citations to be added as a condition of publication (or they can go to a crap journal instead, which is worse for them).

  • For conference papers, it is more difficult to insist on citation, since there is often no second round of review. Here, I would recommend invoking the ongoing pattern of failure to cite in a review as a reason for giving a low score, thus signaling that this does not appear to be an accidental oversight.

  • For slides, of course, there is no pre-presentation review, but there are post-presentation questions. "I notice that you failed to mention Smith's prior work on this problem, would you care to comment on its relationship to your work?", "How does your results compare to Xu et al's results on the problem?" Embarrass them in front of the peers, and they will be motivated to improve---and more importantly, the peers will be motivated to make the same requests.

None of these strategies are things that you can really do just by yourself. But if it's a legitimate issue (as opposed to, for example, a potential distorted view of the significance of your own work), then other peers will hopefully have noticed as well and can be doing the same.

  • 26
    If it really is a small field, being a pain as a referee might actually be quite effective. Sure, you probably won't get to review all their papers, but if you get them to cite paper X once, maybe others will say "you cited X in your last paper, but not this one. Why?" during the next submission, and so on. And editors will likely catch the drift, if they haven't already. – Anyon Sep 6 at 11:26
  • 4
    I agree with this answer and @Anyon. Just wanted to add that self-citations are sometimes a good way to help "new readers" catch up or follow the paper, particularly if you want the paper to be used by others to do more research. However, it appears this group is abusing this mechanism for self-promotion instead of scientific purpose. – SecretAgentMan Sep 6 at 13:56
  • 22
    From experience, the questions during a presentation tactic is much less effective when the presenter has no sense of shame – Phil Miller Sep 6 at 15:48
  • @PhilMiller I often feel does realisation only comes with age, and truth indeed, from SE stats 40-50 olds are a minority here. – Rui F Ribeiro Sep 9 at 11:37

What you encounter is unfortunately common, even 20 years ago. I don't know if there is an English equivalent, but it is called in German Zitierkartell (citation cartel). I encountered it the first time when I researched phase-unwrapping algorithms for speckle interferometry in my Diplom, there were people working on network algorithms, region-growing algorithm, branch-cut algorithm and least-squares algorithm. Every single one was citing only their own research and of friendly groups.

What can you do against it? Essentially nothing from a higher instance. The only thing which prohibits it is a thing called "scientific ethos" which has now the conservation status of "Vulnerable" or even more pessimistic, "Endangered". It is not forbidden to ignore other people's work. Worse, it does work. By citing own and friendly work and getting cited in return, the citation index is inflated. It's the same method as the Salamitaktik, salami-slicing, instead of putting content in one paper, spread it thin over five to get more citations (in English: "least publishable units").

So the only way to punish the perpetrators is doing the same in return, so in time several citation cartels begin to grow, each fighting for funding. As citation cartels are unethical and unethical behavior seldom comes alone, you can look up their papers and search intently for data fudging and scientific misconduct. It goes without saying that you should have an impeccable record because you can expect retaliation in return.

  • 14
    Now I want to know all the German words for academic misconduct. – Azor Ahai Sep 6 at 21:37
  • 7
    Citation cartel is in some use in English too, but citation ring is probably more common. I personally think cartel is a better term. – Anyon Sep 6 at 22:05
  • 1
    Thanks for the insights. We have all the technical means now to make the excessive self-citing part of citation cartels more visible, and it is also necessary due to now-accepted measures such as the h-index. Maybe GoogleScholar will consider making self-citations more visible to reveal the ones who exceed reasonable amounts. – Marius Sep 7 at 11:45
  • So the only way to punish the perpetrators is doing the same in return — Or you could, you know, do the ethical thing instead. – JeffE Sep 9 at 12:14
  • @JeffE It is a typical prisoner dilemma: If everyone cooperates you are much better off than if everyone prefers rivalry. But if one person breaks the cooperation, he gets a temporary advantage. One of the best and easiest strategies who allows cooperation, but punishes violators is Tit for tat. If the ethical stance does not influence jerks, it is of no use. And no, they don't understand why you are doing this voluntarily. – Thorsten S. Sep 9 at 16:45

Science is not a competition.

The "counter strategy" is to ignore the "problem" and do excellent science yourself. They are making a fool of themselves, and most other scientists will realise that as well. Behave like a responsible scientist, cite them if you need to, and only cite their relevant and interesting papers (as you would normally do anyway).

  • 123
    Science might not be, but funding is. – Peter Taylor Sep 6 at 10:35
  • 28
    " you can only hope for funding decisions to be reasonable" Hah!!! I've had a funding request rejected, with a single person giving the following two reasons: (1) the proposed work is impossible to achieve and (2) the proposed work has already been done and published. Illogical, captain? – alephzero Sep 6 at 13:38
  • 21
    Science is indeed not a competition, but that doesn't mean that unethical behaviour in science doesn't exist. If OP's impression is correct, and this group's behaviour is so bad that it amounts to a misrepresentation that is negatively and significantly impacting non-specialists' perceptions about the relative impact of the different contributions to the field, then it is certainly problematic and potentially amounts to research misconduct. That isn't "just deal with it" material. (Assuming, of course, that OP's impression is correct. It may well not be, but the if-then still holds.) – E.P. Sep 6 at 14:32
  • 12
    "There is no way you can change the behaviour of others" Not true at all. That's the whole point of this question (and jakebeal's answer). – user76284 Sep 6 at 15:43
  • 4
    Everything in life is a competition. Science is race. If not, why do we care so much about who did something first? Moreover, science is a popularity contest. If not, why would we care about citation counts and impact factors? Just like any other formal contest, if someone brakes the rules then there should be penalties, disqualification, suspensions etc. To make it more fair and to create disincentives for those who want to cheat. – Asdf Sep 7 at 0:37

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.