24

I did my PhD topic in mechanical engineering. A decade later, I agreed to review an article on the exact same topic, that has a great number of similarities with mine. OK, that can happen.

The authors do cite my PhD manuscript (not the related peer-reviewed articles, by the way) but only once throughout the paper and in a short misleading sentence. Let's say my PhD was about counting from 1 to 20, an analogy would be to say something like "X managed to count up to 2 [ref]". OK, reading too fast can happen too.

The main thing I am uncomfortable with is that the submission is supervised by someone who was in the jury of my PhD defense ten years ago and knows my work well and acknowledged it is a strong text. I am certain that this conduct is intentional. I see several reasons why they would try to minimize my work:

  • Their article brings nothing new (the novelty claims have all been already addressed during my thesis);
  • I proved that the scope of the article should be achieved computing some certain quantities, and they never did that (because it is much more difficult). They have several possibilities. Manage to compute the quantities: they apparently do not want to. Disprove my claim: very unlikely. Ignore my work from the literature review: dangerous. Inappropriately cite my work: what they chose.

I am convinced that they purposely omitted nearly all of my PhD thesis to have a higher chance of having their article published. They have been citing my articles inappropriately every couple of years, but the consequences were minor (in terms of science).

What would you do?

Edit One comment and one answer indicate that whether this behaviour is intentional or not does not matter at all. That is precisely what is not clear to me. When someone submits wrong data supporting a claim in an article, if they just plotted the wrong curve it is a mistake. If it they made up data, it is fraud. That makes a difference to me...

8
  • 1
    "I proved that the scope of the article should be done by X, and they never managed to do X." This is unclear. Dec 11 '20 at 0:46
  • 5
    Do not make assumptions about intention. Apart from that, you can kill the paper if they are not original, and that's the argument. Dec 11 '20 at 1:51
  • 4
    @justhalf No no I do not believe they made up data, this sentence if meant to show that intention/motivation can sometimes count! In short, the situation is: a professor is in my jury, acknowledge my work on "X", then years later writes an article stating that nobody has ever done "X".
    – anderstood
    Dec 11 '20 at 6:47
  • 1
    "They have been citing my articles inappropriately every couple of years" - for each of their papers that have already been published, you should call them out on PubPeer. Dec 12 '20 at 15:57
  • 1
    @DimitriVulis Thanks for PubPeer, I didn't know this relevant site.
    – anderstood
    Dec 13 '20 at 20:19
37

If you are certain the article is not novel, you should recommend to the editor that the manuscript be rejected. Give detailed references in your review.

It is impossible to know if the poor citations are deliberate or accidental. It also does not matter at all.

7
  • 3
    If your role was to investigate scientific misconduct (presumably in a paper already published) then the mistake vs. fraud issue would matter. Deliberate poor citation might be fraud, but I would not spend much time thinking about it unless someone was actually harmed by the fraud. Dec 11 '20 at 7:14
  • 8
    Upvoted for the answer, which is good. But probably almost every poor citation leads to a small amount of harm done (unfair credit taken) that could potentially turn into a greater amount of harm if the paper with the poor citation lifts off stronger than the first one. An interesting example is the Schmidhuber vs. Bengio & Hinton & LeCun controversy in deep learning. Dec 11 '20 at 8:32
  • 4
    @lighthousekeeper The harm is actually caused by using citations as a performance metric. Dec 11 '20 at 10:43
  • 2
    That's part of it. In the Schmidhuber case, deep learning caught on overwhelming societal impact and BHC seem to get more credit than they deserve (e.g., winning the Turing Award). Dec 11 '20 at 11:02
  • 1
    Upvoted, because as an outsider, one cannot conclude that the authors of the paper have conducted any type of clear misconduct. Being sloppy is not misconduct. Perhaps OP has a much better understanding of the field. Perhaps a supervisor first "did the right thing" and pointed a graduate student to OP's thesis but then did not check if the student understands the details of the work. I would even go so far to say that if the authors wanted to deceive, a more likely strategy is not citing OP's work and using unethical but presumably available practices to avoid OP as the reviewer. Dec 13 '20 at 13:27
6

I see OP mentions in comments and his clarifying edit that the core of the question is whether it makes a difference if this was a mistake or fraud.

I have to agree with other answers, and elaborate: it does not matter for a reviewer, from a practical point of view.

Missing a piece of related work happens all the time, and can be an honest mistake. A story of every paper is constructed to present the strong points of the proposed approach. Misleading citing, omitting citations, and similar dubious practices can appear exactly the same. If a reviewer spots the lack of novelty, context or a false claim, it means the review process works. The reviewer can then recommend revisions as required or a rejection all together if the work is not salvageable. If the editor then conditions the publication in including all the missing literature, and correcting all the false claims, the reviewer gets further re-assured in the quality of peer-review for that journal. If the editor rejects the reviewer's claims, the review might re-consider their opinion and confidence in the research published in that journal.

If all problems get fixed in the review process and the paper is deemed correct and a novel contribution, it doesn't matter what the original intentions of the authors were (besides, getting published). If a problematic paper gets denied publication, again, the original intentions of the authors don't matter.

You may equally replace "a piece of related work" with "a suspicious looking data point". The story is exactly the same.

And this is where the question ends from the "reviewer" perspective.


From a broader perspective as a researcher and a member of your research community, the scientific community and an active member of society, the mistake or fraud becomes a somewhat important question if the false claims get published.

If the work is somewhat false, and somewhat inconsequential, it will likely get ignored. If the work is substantially false and has important implications, it will ideally get debunked and hurt the reputation and credibility of the authors in the long run. This might happen directly, but this is somewhere you, as an ex-reviewer, but also somebody in a close research field, can contribute to: talk to the colleagues in your community about how journal X ignored the reviewers recommendations about bad science, false data, or plagiarised work. Publishing in dubious journals doesn't help credibility.

1
  • 1
    @lighthousekeeper Well I kind of try to imply that when I say "and this is where the question ends from the "reviewer" perspective". And I can see your point about submitting bad science, but honestly, I have reviewed a number of "bad science" papers in the last few years, and I promptly forgot the papers and the names after submitting my review (and even more so following a swift rejection by the editor)
    – penelope
    Dec 11 '20 at 11:26
4

You certainly can request in your revision that articles that should be cited, because they are highly relevant, be cited when they are not. You are also completely entitled to request that peer-reviewed articles be cited instead of PhD theses where such articles exist. They are the official original publications of the facts (at least that is usually the case) and very often better accessible, or at least it is more certain that they will remain accessible in decades to come. Also, they generate citations and h-index for their author, but that, even if relevant in life, is not relevant for the review.

If they claim that they did something novel when such a thing is in fact already published, you should certainly point it out in your review. If it makes the whole paper not novel enough, you can recommend rejection. Sometimes reviewers miss important differences from previous existing works, but in that case it is the task for the author to point out the differences clearly.

It is hard to prove that they omitted the article citations intentionally o that they misinterpreted their content intentionally so I would be careful with a claim that they did so. Claiming a priority for a scientific discovery, that is already published, is wrong or dishonest, but it is hard to prove the intention.

4

Recommend that the journal reject for plagiarism.

If you feel that an article that you're reviewing is plagiarizing your work, then the answer of what to do is simple: recommend that the journal reject for plagiarism, probably accompanied with a citation of the work you feel that they're plagiarizing. Plagiarism is a clear violation of academic ethics, and recommending a work be rejected for it seems like it would be straight-forward.

If a student in one of your classes plagiarized some of your research in one of their assignments, you'd reject their work and refer them to your university's academic integrity board, right? Why should you be more hesitant to recommend the rejection of plagiarism when it's done by your peers?

4
  • 1
    But there's no indication of plagiarism at all? The OP is just hurt that their thesis wasn't cited more often. Dec 11 '20 at 16:47
  • 1
    @NickMatteo It sounds like they're claiming that it's a copy of their work.
    – nick012000
    Dec 12 '20 at 6:34
  • 1
    I think plagiarism is to far streched. But lack of novelty is a reason to not publish it.
    – usr1234567
    Dec 12 '20 at 13:43
  • 1
    @NickMatteo My thesis is cited, so the number of citations is irrelevant. I am hurt of it being citing incorrectly on purpose. Regarding plagiarism, they surely reproduced what I have done without mentioning it, but I cannot claim that they would have done it differently if they were not aware of my work.
    – anderstood
    Dec 12 '20 at 17:02
2

Act as if you were not the author of your PhD thesis or the related article, just some random reviewer with excellent knowledge of your work:

  1. Point out that both or at least one of your works should be discussed in the literature review as they are relevant for their article.
  2. Make clear that the cited work contains more of relevance for the reader and must be explicitly mentioned.
  3. Further point out, that the article makes it not clear what is achieved beyond these works. An article has to claim something novel (either explicitly or implicitly).

Every single point is valid criticism. Non contains hurt feelings from you.

You can reply to the editor, that you are not sure whether you should recommend a major revision or a rejection, as you don't see the novelty of the work until point 3 is resolved.

2

There are two sides to this for me.

  • They are missing / omitting / misunderstanding parts from a reference, which makes their work less relevant.
  • This reference is your work.

Is your thesis publicly available? It does not matter, but if it is, the offending matter could have been found by anyone else. If not, you are on a thin ice about the reviewer's anonymity.

I suggest you list all the problems in your review. Typical review forms have two parts, "blind comments to the authors" (the actual review) and "comments to editor" (typically blank).

I would definitely write a comment to editor. It is your thesis they belittle. You happen to know the authors personally. Your reviewer identity might be compromised. The editor needs to know of all this.

It might happen, that you will be considered in a conflict of interests. But I doubt very much this paper would be accepted by the editor, if all the above issues are communicated to them.

1
  • My colleagues also believe I should write to the editor, which is now my plan. Yes the thesis is publicly available. About the anonymity: I intend to write the review as a random reviewer (cf usr1234567's answer), but as a side note, it does not really matter to me: I think the reviewers should have their name publicly attached to articles.
    – anderstood
    Dec 12 '20 at 16:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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