14

For transparency, I would like to share my accepted papers' anonymous peer reviews, if my coauthors unanimously agree. Some journals partner with websites such as openreview.net (which share reviews even for rejected papers), others may have that included in their process, but in most cases, nothing is said about sharing reviews by the authors (though it is sometimes explicitly said that reviews cannot be shared by the reviewers).

Are there any ethical issues with that?

I could perhaps see intellectual property issues (are reviews the property of the journal? of the reviewers?), though in our field, journals at least allow to have the author's version of the paper downloadable from the authors' website (so, why not the accompanying reviews?). More importantly, nothing is explicitly said about that on the websites of editors and journals.

5
  • 10
    "Transparency" is no value of its own. What do you want to achieve?
    – Karl
    Feb 18 at 13:37
  • 2
    Several things. First, showing strengths and weaknesses of the paper as evaluated by experts and how we ultimately responded. Second, showing that in our community, reviews are of extremely good quality (typically, 3 to 6 reviews of consistently high quality and can be a page each), and this can help young reviewers. Third to show if the paper was "barely" accepted or a strong accept. Fourth because except copyright issues as pointed out here, I do not see any reason to hide them. Entire communities are moving towards websites such as openreview.net for these reasons.
    – nbonneel
    Feb 18 at 14:30
  • 2
    Showing that reviews are of high quality adds credibility to the paper (e.g., predatory or very low quality journals generally have barely any reviewing), and can clarify the accept decision process (we've all seen low quality papers that were surprisingly accepted, and wondered if reviewers missed the problems or if the editor did not take reviews into account in the final decision -- e.g., the joke article "SARS-CoV-2 was Unexpectedly Deadlier than Push-scooters: Could Hydroxychloroquine be the Unique Solution?" and its open peer review report: sdiarticle4.com/review-history/60013).
    – nbonneel
    Feb 18 at 14:59
  • 9
    I get your point, but relevant readers know which journals are crappy. And you also get lousy reviews for very good papers, even in good journals. Will you show the bad (superficial) ones, too? And anyway I never want to rely on experts and authorities, much less on unknown experts, for my judgement. I don't have to bother what the reviewers said, because I can read myself.
    – Karl
    Feb 18 at 15:54
  • 1
    Even good peer-reviewed publication venues regularly produce a significant lot of crappy reviews, and publishing crappy reviews is worthless. Publishing only good reviews would be biased, so why bother publishing any of them at all? Moreover, papers are often improved based on the reviews, and so most reviews may be completely irrelevant to the published paper.
    – user21820
    Feb 19 at 19:58

4 Answers 4

21

Unless it is an explicit part of the reviewing process, or there is some rare overriding concern, you musn't publish the referee reports you've received.

This is actually a rare situation where copyright law and academic norms coincide. Basically, you cannot publish stuff other people have written unless there is something that says you may.

Initially, copyright of the report lies with the referee, as they have written it. By submitting it to the editor, there is an implied license to use the report internally and to share it with the authors. Barring other arrangements, there is no basis for assuming that the journal has additional rights to the report. As the author, you don't need any particular rights to do stuff with the report, so there is no implied license here. [Individual parts of the report may come with a very generous license, eg if the referee suggests formulations.]

7
  • 4
    Another exception worth mentioning would be: "you have the explicit consent from the referees". One can always ask, and in this case of an anonymous review, the editor can act as the communication channel. Feb 18 at 11:39
  • 2
    Mustn't publish doesn't imply that you can't share reviews with a limited set of people. I don't think there are ethical issues with this. Feb 18 at 20:17
  • 1
    Ethically, perhaps not, but legally, it would still be a copyright violation unless you can argue otherwise based on fair use or some similar exception defined in copyright law. Anyway, a separate point: the answer seems confusing where it says "you don't need any particular rights to do stuff with the report". I'm probably misunderstanding what that's supposed to mean, but at face value it seems wrong because you most certainly do need some rights if you want to do stuff - e.g. sharing on a website - with the referee report.
    – David Z
    Feb 19 at 0:04
  • 1
    @DavidZ: in order to accomplish what the report was conveyed to you for ("do stuff"), you don't need any rights for sharing etc., so no such license is implied for authors (unlike editors). Feb 19 at 5:50
  • 1
    @DavidZ I think it's intended to be read as "do the stuff authors do with reports". Feb 19 at 17:15
7

It is NOT a good idea to do this:

  1. you risk identifying referees through their writing or formatting styles. There might be specific turns of phrases or expressions specific to a referee (especially if the referee did not write in their native language) that could be tracable to papers authored by this referee.
  2. What do you do if one review is positive but another is not so positive? Do you cherry pick only the positive reviews?
6

I think the trouble is then that people would rely on the reviews instead of reading the article. These are only in place to allow for formal acceptance or formal rejection, the reviewers can miss very obvious and important mistakes.

An expert reader will not care much about these reviews as they can read the article and come to their own judgment.

1
  • 1
    Good answer (+1). One comment though: An expert reader will not care much about these reviews may not always be the case. When reading some posts (notably Hacker News) I usually start with the comments because of the high quality of the exchange. I read the actual article in maybe 20% of the cases - when the comments make it worth it. This of course relies on the comments being high quality and is especially true for areas with a wide range of topics.
    – WoJ
    Feb 19 at 8:53
3

Aside from copyright issues, there is the issue of basic politeness and kindness.

If you're sharing a piece of writing someone wrote privately, you should get agreement from that person. Not from your coauthors. If you cannot get agreement from that person, you should not share it.

If a journal partners with openreview.net or a similar website, then the reviewers know ahead of time that their reviews will be visible publicly. That's not the case if you publicize the reviews of your paper unilaterally. The reviewers were writing for an audience of you and your coauthors.

I know I for one would feel very uncomfortable if my reviews of papers were made public. I think very differently about the tone of my writing when I have a much larger potential audience, and I think the result is much better when I only have to ask myself, "How will the authors of the paper (and the editor) understand this?" and not, "How will a random person on the internet understand this?"

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .