53

I received an invitation to review for the International Journal of Adaptive Control and Signal Processing. I started to analyze the manuscript by reading the title, abstract, introduction, results and discussion and conclusions before assessing the proposed methodology.

I detected that the authors copied approximately 7 pages from another article without a considerable change so that it looks like his own ideas (plagiarism). should I stop analyzing the paper (assessing the methodology), or continue? because from my experience I think that the editor will reject the paper even though there is a novel contribution.

Rodrigo de Azevedo
  • 1
  • 1
  • 6
  • 20
user137684
  • 969
  • 3
  • 8
  • 15
  • 19
    Can you give a little more detail about what type of content seems to be copied? Though, at 7 pages it seems unlikely this is borderline case. Personally, I would probably directly email the editor to bring it to their attention, noting which pages appear to be copied and the copied work. They can decide how it should proceed. This could save you, the other reviewers, and the editor a good amount of time. – user58322 Jul 01 '17 at 15:05
  • 5
    Are those 7 pages from a paper of the same author, or from someone else, and where did they appear before? Sometimes, authors of conference proceedings are invited to submit extended versions of their papers to journals; in such cases something like this could be expected and normal. – tobias_k Jul 01 '17 at 20:04
  • 4
    No, they are from a different paper with different authors. – user137684 Jul 01 '17 at 20:42
  • Just to be clear, they didn't copy from a paper one of them coauthored? – DonQuiKong Jul 01 '17 at 22:27
  • 3
    Extended versions usually include a reference to the original "unextended" version, at least in my field. – Fábio Dias Jul 02 '17 at 01:34
  • 12
    Just to make sure: those 7 pages are not "standard explanations, irrelevant to their findings"? Because in some fields, certain sections are intrinsically very similar to each other or are even allowed to be copied, e.g. the description of large physics experiments: if hundreds of paper describe in a single page how the experiment works, the wording will always be very similar or is even allowed to be copied from a certain source. This is not a problem, as the description just "has to be" in the paper but does not "contribute to the work". – Mayou36 Jul 02 '17 at 10:28
  • Is it not a double-blind review, so you know for sure they are not "self-plagiarizing"? – Headcrab Jul 03 '17 at 06:43
  • 4
    @Headcrab Even self-plagiarism at this scale is utterly unacceptable. – Walter Jul 03 '17 at 12:40
  • @Headcrab You would also know this if you were among the authors of the plagiarised paper. – cfr Jul 03 '17 at 23:59
  • @Mayou36 Even in cases like this, wouldn't an attribution still be required? – deckeresq Jul 26 '17 at 01:56
  • 1
    @deckeresq, no. You have to cite a source of course that provides the information of what you write about (and this is always cited, you have to be able to check the facts, of course) but not a source that covers the exact wording, as the wording does not add any relevant information to the topic. Actually, the wording I know of that can be copied (explicitly) is forbidden to cite, as it is more of an internal note. It is (slightly exaggerated) as if every chemistry paper contains an explanation of what an Erlenmeyer flask is. Which scientist should care whether the wording is "plagiat"? – Mayou36 Jul 26 '17 at 08:37
  • @Mayou36 That's extremely interesting! Thank you for the information. – deckeresq Jul 26 '17 at 13:58
  • @deckeresq but keep in mind, that this is more of a rare occasion and as common as 200+ authors and similar things (which is still common in certain fields), most of all it is very field and topic dependent. Make sure to check in your field. And usually, AFAIK, this does not cover 7 pages but is way shorter (half a page to a page). But it surely depends. – Mayou36 Jul 26 '17 at 14:19
  • @Mayou36 Definitely a good disclaimer. I don't think I've ever seen something like this in my field (CS), but perhaps it's just even rarer. – deckeresq Jul 26 '17 at 15:22

7 Answers7

105

Don't bother with more analysis, there is no point, since the paper should be withdrawn or rejected because of the plagiarism (if it isn't decline further review for this journal). Write a brief report along the lines

The manuscript contains 7 pages which are almost verbatim copied from < plagiarized paper >. In view of this I refrain from any further analysis as I expect you to reject the manuscript.

You should also check, whether the paper has already been published as a preprint (e.g. on arXiv) and if so inform the editor and suggest him to require the authors to withdraw from there too (I know of a case where this happened) but you may also contact arXiv directly (to protect your anynomity don't contact the authors or make your knowledge public).

Walter
  • 5,453
  • 1
  • 21
  • 30
  • 6
    I actually think the last part of the report (and expect you to reject the manuscript) is not necessary, as there should be fixed procedures for the journal to react to this anyway. It's not your job to investigate/punish/blame, just state the facts and let it develop. – skymningen Jul 03 '17 at 09:04
  • 10
    My first reading of the suggested text was "I require you to reject the manuscript", but rereading, I think you meant "I presume you will be rejecting the manuscript, so there doesn't seem any point wasting time". I agree though that the phrase doesn't really add anything. – Martin Bonner supports Monica Jul 03 '17 at 11:05
  • 6
    @skymningen it's just to explain why no further analysis is done. – Walter Jul 03 '17 at 12:38
  • 2
    I don't think "This is a plagiarized work." needs any further explanations why no further analysis is done. In grading student work, this actually is the explanation I sometimes give for not correcting the rest of the work. – skymningen Jul 03 '17 at 13:17
  • "In view of this I refrain from any further analysis as I expect you to reject the manuscript." Saves you time now. But if altered manuscript had those segments re-written, you later might need to only spend much time reviewing the re-written part. Editor might prefer that you only need to review the small update later. I'm highly speculating here, but perhaps saving time now might later equate to a more demanding schedule / time pressure. I know that stopping a job that was barely complete has sometimes paid off well for me (saving me time), and other times not-quite-so-much. – TOOGAM Jul 04 '17 at 09:43
  • @TOOGAM IMHO, the author should not be allowed to resubmit a revised version with this journal. – Walter Jul 04 '17 at 09:46
  • @Walter: Not disagreeing with that judgement call. You're probably pretty safe... as long as you get to make that judgement call (or the person who does have that authority actually ends up making the judgement call you're expecting/desiring) – TOOGAM Jul 04 '17 at 09:56
  • 1
    @TOOGAM Even if the authors are allowed to resubmit, the reviewer is under no obligation to accept the review the next time (and might even choose to stop reviewing for that journal completely if they do allow it). – Tobias Kildetoft Jul 04 '17 at 18:10
  • 1
    Using expect you to does carry some connotation of coercion. If that phrase needs to go in at all, use "expect that you will" or "anticipate that". It's the that which shifts the expectation to what is likely to happen rather than what you believe the editor should do. And using anticipate avoids any unintended nuance from expect. – Andrew Leach Jul 05 '17 at 08:34
  • @AndrewLeach I'm not a diplomat, but a reviewer and have only so much time for such reviews. The editor must allow for some slack in this regard. Actually, "expect you to" is more honest than your suggestion, since I would turn down any further review request from that journal should they not reject the manuscript. – Walter Jul 05 '17 at 08:38
  • Ah well, yes, if you do actually expect the editor to reject, then of course there is no problem with allowing that sense! – Andrew Leach Jul 05 '17 at 08:40
  • One of the basic functions of the reviewer is to make a recommendation, so I think the better phrasing would be, "In view of this I cannot recommend publication." – David Jul 05 '17 at 18:42
30

As with any other serious issue, you don't need to waste your time with the paper that is obviously not publishable. Please note that 7 pages is a extreme case of plagiarism, and there is absolutely no way that such paper could be accepted in any serious journal.

Reviewer's time is precious and I am sure editors know that.

As with papers that are unreadable (this is another case of serious issue, at least in my book), for example because of poor English that does not allow reviewers to actually get enough information from the paper to judge its merits, there is problem, what to review? Even if by any chance the authors get the opportunity to make a revision, which would remove the plagiarism, that would still left 7 pages which had to be written anew.

(And in my personal opinion, the author responsible for plagiarism should never get that chance, they should be tarred and feathered and barred from publishing for a year or two. But as said, this is only my personal opinion).

xmp125a
  • 3,915
  • 1
  • 16
  • 18
10

I had this exact situation happen to me before with a medical paper. I immediately sent a response to the editor (hidden from the authors), highlighting my concern about likely plagiarism and giving the reference of the uncited source work.

When I do reviews, I usually start by searching for "likely" phrases and even paragraphs in the usual repositories. You'd be surprised how many mediocre quality submissions contain even large sections that appear to be plagiarized wholesale from prior work.

Deepak
  • 1,315
  • 8
  • 14
10

Check:

  • Is the submission an extended version of a prior, shorter conference/workshop paper?
  • Is extending such a prior publication with additional material allowed by the guidelines of the journal? (E.g., some journals allow longer, full versions of prior short summaries. For example, one legit goal of journal papers in mathematics is to provide long proofs which have no space in a summary, while the statements of the theorems are present in both journal and conference versions. Related work, intro and conclusion may also be present in both in more-or-less the same form.)
  • If the authors are really different (not new pseudonyms of the old authors), is extending a summary written by someone else acceptable according to the guidelines? (Note: extensions of summaries of others are rare and should be always scrutinized carefully for plagiarism.)
  • Does the submitted material (paper and accompanying comments) observe the guidelines by, e.g., properly advertising the relationship to the prior submission?

If the answer to all the questions is "yes", continue reviewing. Otherwise, justify in which way the submission violates the guidelines of the journal and stop reviewing.

Also notice that papers in certain fields, such as mathematics, might require a prior build-up of notation and stating earlier or standard results in that notation before starting with actual contributions. There is little excuse to omit this standard stuff from journal papers where space is typically not an issue. Also, for estimating novelty, 7 pages is a lot in two-column 9pt style with Times fonts, but little in double-spacing narrow--single-column 12pt style with Computer Modern fonts (according to the guidelines of some journals): the absolute amount of copied material (say, measured in the number of words) may be smaller than it seems.

Summarizing, before blindly rejecting, take a careful look at whether your rejection is properly justified.

Leon Meier
  • 3,993
  • 1
  • 13
  • 37
  • Thanks for editing. I find your answer much clearer now. (It is also interesting because this obviously is very discipline-specific and it is always easy to think that things work just the same elsewhere, even when one knows they don't!) – cfr Jul 04 '17 at 16:17
  • @LeonMeier I can see no circumstances in which copying seven pages of of somebody else's work is acceptable. It doesn't matter what page layout anyone is using. Seven pages is a lot of text. In particular, I work in mathematics/theoretical CS: copying pages of background material would be unacceptable, and I can't think of any paper I've ever read that's needed more than two or three pages of such material anyway. – David Richerby Jul 04 '17 at 17:06
  • 3
    The possibility of extending one's own short conference paper into a longer journal paper is a good point but I would expect the asker would have mentioned this fact if it was the case. In particular, one would usually expect the two versions of the paper to have the same title and author list, and the long paper must include a statement about its relationship to the short version. (Typically something like "An extended abstract of this paper appeared in the Proceedings of the 12th International Conference on Interesting Stuff, 2017".) – David Richerby Jul 04 '17 at 17:09
  • 1
    @cfr Listing standard lemmas from a narrow mathematical area in your notation would do the job. Or listings of known data (figures, graphs) that your paper requires for deriving its own results. – Leon Meier Jul 05 '17 at 00:44
  • 3
    @LeonMeier I have never seen a paper that lists seven pages of lemmas from other papers. Please give some examples. – David Richerby Jul 05 '17 at 00:45
  • 3
    @DavidRicherby And even then, one would not just copy and paste those over. – Tobias Kildetoft Jul 05 '17 at 07:04
  • 1
    @LeonMeier This implies that the OP is concerned with self-plagiarism, which is a totally different beast. But, even 7 pages of self-plagiarism is a lot. – xmp125a Jul 05 '17 at 15:15
  • 1
    @xmp125a I am NOT saying that the OP should close their eyes on plagiarism or self-plagiarism. By no means. They should take (self-)plagiarism very seriously. I'm saying that sometimes there are legitimate ways of copying material which the OP could be not aware of and which do not count as (self-)plagiarism. These ways depend, more or less, on the policy of the journal. – Leon Meier Jul 05 '17 at 16:06
  • @LeonMeier I assume that the difference is mainly related to self-plagiarism, e.g. what amount of new material is required (there was the time in CS that 30% was the standard, if journal paper was based on a conference contribution). However, plagiarism rules (not the self-plagiarism rules) are, I assume, universally about the same. – xmp125a Jul 06 '17 at 10:41
  • 1
    @xmp125a Of course, in its basis. But the actual wording varies. – Leon Meier Jul 06 '17 at 15:56
3

No. stop reviewing it and return it to the editor with your findings, and wait for his advice. The editor needs to figure out if the authors have the permission to reuse the material (if it only is explanatory).

Sascha
  • 3,700
  • 10
  • 23
2

It is essential to report this to the editor. Your judgment as a referee is clearly already made, so subit a report as soon as possible wherein you declare the submission, in your opinion, is not appropriate for the journal.

  1. If this is allowed by the journal they will not contact you again, and you are ahead since you will not have to referee for a journal which an editorial policy you disapprove (although keep in mind the excellent points raised by @LeonMeir),
  2. If it is not allowed the editor will be grateful and you save some work.
1

I'll differ the tiniest bit from the existing answers. It can be helpful to look at the paper enough to help the editor triage.

Is there anything worthwhile in the paper? If not, you should point that out. If yes, as you noted in your recent experience, then provide this information, so the editor will be fully aware of the extent of the conundrum. You are the editor's fact finder -- let him/her benefit from your expertise.

But you don't need to look at the paper with as fine a view as if you were going to proceed with a normal review.

So, taste it enough to identify the poison before you spit it out.

aparente001
  • 38,999
  • 8
  • 65
  • 153
  • I think finding 7 pages of plagiarized content is more than sufficient to help the editor triage. No need to taste the poison. – JeffE May 31 '18 at 15:45
  • @JeffE - In some fields expectations of academic rigor are lower than others. As an editor, I've sometimes had to remove minor incidents of plagiarism caused by sloppy work habits. – aparente001 Jun 01 '18 at 04:08