37

DISCLAIMER: As some readers have found the following question and prior title (How to best kill a manuscript as peer reviewer) outrageously unethical, I want to clarify that, by asking this question, I was simply and constructively trying to be provocative, not suggesting any means to hijack the peer review process.

I often peer review manuscripts for scholarly journals. In the most typical scenario, the editor asks me a quality appraisal and a priority appraisal.

For quality, the typical recommendations can be (mutually exclusive): accept as is, minor revision, major revision, and reject.

For priority, the typical alternatives can be (mutually exclusive): top priority, mid priority, low priority.

My experience and perspective is that most manuscripts do not deserve a rejection, as they most often have some merits, at least in the field of pragmatic cardiovascular research, where most works are only incrementally original.

If my will is to try to "kill the manuscript", i.e. maximize the likelihood it is eventually rejected because I find it unsuitable for that specific journal, but still finding it has some incremental value, I simply recommend minor or major revisions, but give low or very low priority.

Is this sound and appropriate? Am I wrong in this approach?

  • 51
    Note: Votes on questions should reflect whether a question is on-topic, well-posed, etc.; not agreement or disagreement with the idea reflected in the question. – ff524 Dec 8 '16 at 21:00
  • 2
    Comments are not for extended discussion; this conversation has been moved to chat. (Comments can't be moved to chat more than once, so future chatter can only be deleted instead.) – ff524 Dec 12 '16 at 10:21
  • @ff524 I seem to recall that there was a way to migrate a whole chat room into another. I forgot the specifics, but definitely recall doing it once or twice to merge multiple migrated-to-chat discussions. You may want to ask in TL. – a CVn Dec 13 '16 at 13:10
  • 1
    @FedericoPoloni I find this new title to be very unclear. Also, it does not seem to match the question body very well. – Raphael Dec 14 '16 at 10:30
  • 1
    The question body says "my will is to...maximize the likelihood it is eventually rejected". The title has NOTHING to do with that! I've changed the title by simply quoting the key line of the body, so now they agree. – David Ketcheson Dec 15 '16 at 13:39
9

I'm trying to understand the reasons for the harsh judgement of the OP reflected in many of the answers and comments. My reading of the question is that the OP feels uncomfortable recommending "reject" for sound, original, but not groundbreaking research, and is asking what the "best" course of action is when the submission is not sufficiently groundbreaking for the journal to which it has been submitted. (I believe others have interpreted the question as "What is the surest way to kill a manuscript?", which, to my mind, is not assuming good faith on the part of the OP.)

My brief answer is that one should recommend "reject", but should state in the report that the research is solid and original, and would be suitable for publication in a different journal. The reviewer might even make suggestions as to suitable publication venues.

The problem seems to be that the reviewer is being asked to assess quality factors including methodological soundness, originality, and clarity of the written report on scale of "accept as is", "accept with minor revisions", "resubmit with major revisions", and "reject", which do not form a scale of quality at all, but, rather, are recommendations as to course of action. I agree with (my interpretation of) the OP that "reject" is not an accurate quality assessment for research that is correct and original, although it might be the correct course of action for a particular journal. Thus it might seem that the only option left to a reviewer who does not want to send the wrong message regarding quality is to use the priority scale as a way to send the desired message regarding recommended course of action. As I state in the previous paragraph, however, one may escape from the straightjacket provided by the checkboxes on the reporting form by spelling out one's recommendation in the text of the report.

  • 7
    There is no way to assume good faith for the statement: "If my will is to try to "kill the manuscript", i.e. maximize the likelihood it is eventually rejected because I find it unsuitable for that specific journal...." The reviewer's job is not to "maximize the likelihood" of eventual anything. The reviewer's job is to recommend a course of action and give his/her reasons for that recommendation. Doing something else is inevitably in bad faith. – Wildcard Dec 11 '16 at 20:27
  • 3
    While I'm broadly in agreement with the other answers and with Wildcard's comment here, I think there is also room for the interpretation that `killing the manuscript' is just a loose, rather dysphemistic, phrase for what the OP proposes to do. Perhaps the OP is reluctant to simply recommend rejection, and seeks a more diplomatic verdict that achieves the same effect. As I say, I agree with the the consensus expressed in the other answers, but also agree with this answer that the implied judgement of the OP is perhaps harsh. – Shane O Rourke Dec 11 '16 at 20:46
  • 1
    @Wildcard: I think one can assume good faith. I don't see how a reviewer who chooses "reject" can be characterized as doing anything other than trying to "maximize the likelihood" that the manuscript will be rejected. This particular OP, however, appears to feel that, due to its overloading as a quality assessment, "reject" is not appropriate for many manuscripts, even when rejection would be his recommended course of action for that manuscript in that particular journal. – Will Orrick Dec 11 '16 at 20:47
  • 1
    @ShaneORourke: the OP has made a couple of edits, adding "because I find it unsuitable for that specific journal", "where most works are only incrementally original", and "but still finding it has some incremental value", all of which, to me, suggest trying to do right by the authors of the submitted manuscript, rather than the opposite. – Will Orrick Dec 11 '16 at 20:52
  • 2
    @WillOrrick, which just wastes everyone's time. In the field of dating there's even a phrase for this behavior: to lead someone on. "I don't want to turn her down for another date (even though I have no slightest interest in this person myself) because she would be a good match for someone...." It's not a nice thing, whether deliberately malicious or unintentionally cruel. However, I get where you are coming from. I just strongly disagree. – Wildcard Dec 11 '16 at 20:55
158

Is this sound and appropriate?

No. Your job as a reviewer is not to choose which manuscripts are accepted or rejected. Your job is to advise the editor as to the merits and the flaws of the manuscript, and then let the editor decide the fate of the manuscript.

Thus, you should give the manuscript the ratings that most accurately reflect your perception of it, not the ratings that you think will get it "killed".

I think that there is an underlying misunderstanding

I don't think I misunderstood the first version of your question, and my answer still applies. It is not up to the reviewer to maximize the likelihood of any particular outcome; just tell the editor what you think of the manuscript, and leave the rest to the editor.

This means that if you are asked by the editor to recommend an overall decision, you should recommend an overall decision. If asked by the editor to recommend a decision based on quality and specifically excluding the priority, novelty, and usefulness of contribution, then you should do exactly that. (If you aren't clear on what factors should be included your quality recommendation, ask the editor.) If you believe that a manuscript should be rejected from a given journal because its contribution is incremental, you can indicate as much in the part of the review where you're asked to assess the manuscript's originality and/or in a separate comment to the editor.

  • 1
    Comments are not for extended discussion; this conversation has been moved to chat. – eykanal Dec 12 '16 at 19:29
94

If your intention is to "kill" the manuscript, you should give a reject recommendation. Rejecting does not mean there are no merits to the manuscript. In fact, I would guess that most manuscripts are rejected because the journal is not a good fit, or the results are not strong enough for the journal, but might be appropriate for a lower tier or more specialised journal.

Of course you can and should discuss the merits of the manuscript in your report, as well as suggestions you think would improve the paper, even if your recommendation is to reject. It is then up to the editor to make the final decision.

In my opinion, you should only give a "revisions" recommendation if you will likely accept the manuscript after the requested revisions are made. Purposefully setting the bar for these revisions too high because your original intent is to "kill" the manuscript for that journal seems disingenuous to me.

  • 72
    It does not “seem” disingenuous, it outright is. Recommending “revisions” means: “I, as a reviewer, recommend the publication of this manuscript once these following revisions have been made”. If the revised manuscript then comes around and you fail to honour this promise, you’ve effectively lied to the authors (and the editor), and led them on a merry goose chase that, in many cases, has taken months and costs thousands of pounds in research money. Recommending revisions when you have no intention of accepting them amounts to theft of time and money, and is therefore deeply unethical. – Konrad Rudolph Dec 9 '16 at 12:51
  • @Konrad: Nice riff on "Nay, it is. I know not 'seems'." – Pete L. Clark Jan 2 '17 at 19:01
43

There is no "kill the manuscript" in a reviewer's job description.

If the reviewer thinks the manuscript is unoriginal or flawed (e.g. technically), that's what they should say in their review. They can recommend it to be rejected for the given journal.

However, the way the question is put sounds as if the reviewer may want to try to manipulate the editor, and if they do, they arrogate to themselves a role which they do not have.

Their job is to identify whether the paper is sound, original, (and possibly) fitting to the journal, according to their best knowledge and in good faith. A scientist thinking in terms of "killing manuscripts" effectively excludes themselves from the pool of available reviewers, hands down.

  • 5
    Could not agree more........Kill the manuscript........seriously, this is the most egregiously unethical thing I have read on SE so far. And what surprises me most, is it's not done through an anonymous posting, almost as if OP is proud of his actions >.< – NZKshatriya Dec 11 '16 at 2:13
  • @NZKshatriya Indeed, you nailed it. I almost thought this was a trolling attempt, but then came to the conclusion it was probably a seriously intended question. – Captain Emacs Dec 11 '16 at 2:41
  • I would have thought that "questions" such as these would be disallowed, especially here in Academia. – NZKshatriya Dec 11 '16 at 3:05
  • 5
    @NZKshatriya Questions like these should not be disallowed because the history of academia is full of unethical or borderline ethical practices and we should be able to give suitable answers pointing out the problems. – Massimo Ortolano Dec 11 '16 at 8:06
  • 2
    @MassimoOrtolano If the question was intended to be seriously asked, which we assume here, then answering it will give someone with complete lack of "natural intuition" concerning what is appropriate the opportunity to re-gauge their code of conduct. Some people may have grown up in less than ideal research cultures and need to learn what constitutes good/acceptable practice. – Captain Emacs Dec 11 '16 at 12:58
25

Your strategy may have the effect of killing the paper more broadly than you intend. Suppose the paper has both merits and flaws, but is not suitable for the journal for which you are reviewing.

Suppose you recommended "reject" with discussion of what is good about the paper and what could be improved, and maybe where it would be a better fit. If the editor accepts your recommendation the authors can immediately get on with making revisions taking into account your comments, and submitting it to another journal.

Now suppose you recommend revisions, with no intention of ultimately accepting the paper, and the editor accepts your recommendation. The authors are going to waste elapsed time, personal time, and energy on trying to get it into the current journal. It may take more than one round before they realize they are not going to be able to satisfy you and move on to a different venue. That delay and wasted effort could reduce the chances of the paper being published as authors move on to other research.

17

The question suggests that you have the whole procedure of reviewing wrong, so it's not sound nor appropriate.

Rejection or major or minor revisions answer to what revisions the manuscript might need or it has inherent flaws that cannot be amended with reasonable modifications. In any case, you have to justify your recommendation (and thankfully there are other reviewers also). So, if you recommend minor revision for a manuscript that should be rejected in your opinion, then you are a bad reviewer. The same if you accept a manuscript that should be revised or rejected.

Another recommendation that you are often asked is the priority. This asks for something completely different. Usually it refers to the novelty or the significance of the results.

Thus, it's possible for a manuscript to be acceptable but of low priority. This would normally lead it to another journal. Although it sounds contradictory, there might be a manuscript with very important or high impact subject or results, but with some experimental flaws that makes it worth for a rejection. Wouldn't you "kill" it? (there are some famous retractions when flaws or manipulations were later discovered)

It's true that our recommendations as reviewers are accepted often as they are from the editor... unless the reviewers disagree. In that case, the editor has to make their own opinion and then it's your validity as a reviewer that is on stake.

9

I think there is some linguistic confusion in some posts here. Deciding is not the same as recommending, and English language nicely gives different words for the two activities, viz “decide” and “recommend”. A reviewer makes a decision what to recommend to the editor; the editor makes a decision what to do with the manuscript.

As for my own opinion on the question itself, it seems a bit unfortunate to accept the responsibility of being a reviewer—which is a responsibility not just to the editor and to the journal but to the entire scholarly community—but then to try to game the system for whatever personal reason you might have. Regardless of your recommendation, if you think a paper has some merits, say that in your review, thereby making that part of the data which your review content constitutes. If you do not think a recommendation of “reject” will get a paper rejected, maybe you should discuss that with the editor.

4

Answers and comments here jumping at your throat have to do with the publishing culture in computer science and maths (what most users of this site do), which differs significantly from the one in biology or medicine. I think your question is not optimally phrased, but if curating for quality is your ultimate goal, I applaud that intent. Quality curation is a critical aspect of academic publishing and there generally is too little of it.

How to best kill a manuscript

Unfortunately, you can't.

Authors almost never resolve themselves to ditch a manuscript because that's very frustrating, and that would mean a shorter bibliography. You can be instrumental in its rejection from the journal you review for, but it's simply going to pop-up in the review queue of another less reputable journal.

How can I maximize the likelihood it is eventually rejected because I find it unsuitable for that specific journal.

By recommending rejection exactly for that reason. Write a review pointing out what you think should be improved in the paper, that's the service you're asked to do to the authors, and tell the editor you think the paper's contribution is too modest to warrant publication in that journal.

If you want to have more decision power in that process, you should become an editor yourself. But that is not something you typically decide on your own. If a community values your curating abilities, then you might get that job.

  • 3
    I'm in math and I think your answer is perfectly applicable there too. It's totally reasonable for a reviewer in any field to want low-quality work not to be published. And as you say, the one and only ethical way to approach that goal that is to write a review suggesting rejection and explaining your reasons, and hope the editor is convinced. The question seems to be suggesting "strategic" methods, whereby some other kind of report might actually lead to a higher probability of rejection, and I feel that is inappropriate. – Nate Eldredge Dec 12 '16 at 14:25
  • 1
    @NateEldredge I agree that one should not look for "strategic" ways of having papers rejected. I mentioned maths and CS because it seems these fields have more a "publish everything" kind of culture, at least that's what I gathered from my experience on this site. – Cape Code Dec 12 '16 at 15:00
  • I don't think I agree that math has a "publish everything" kind of culture, at least if by "publish" you mean "publish in a journal that the mainstream math research community regards as legitimate." (Maybe that's part of the issue? I would be willing to believe that there are more authors of math papers who are really not members of the community of research mathematicians than there are authors of medical papers who really are not members of the community of medical researchers.)... – Pete L. Clark Jan 2 '17 at 18:53
  • ...For instance I have a couple of papers that I submitted several times for publication, without success, and eventually I tired of the effort of trying to find a reputable home for them. When these papers are rejected, the referees and the editor did not identify any flaws and did not claim that there was no novelty, just that the results were not significant enough to be published. In fact I think that in some branches of math at least, "routine" or "incremental" work is really frowned upon, and in order to get published anywhere reputable, a paper should contain at least one "new idea." – Pete L. Clark Jan 2 '17 at 18:58

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.