Must the identity of an article referee remain confidential? How about the referee report?

Some journals strongly assert that reviewers have a duty not to out themselves. For example, the Nature group says

We ask reviewers not to identify themselves to authors without the editor’s knowledge. If they wish to reveal their identities while the manuscript is under consideration, this should be done via the editor, or if this is not practicable, we ask authors to inform the editor as soon as possible after the reviewer has revealed his or her identity to the author.

I suppose the question to ask is: does the fact that you were a reviewer for this article change anything?

I’d suggest that it shouldn’t make any difference – rather what actually matters are the issues, and that they are identified them as something that may (or may not) need to be addressed, it should not matter who identifies the issues themselves.

I’m sure it’s exceedingly unlikely that any refereed publication has had unanimous agreement from all it’s refeeres, and the fact that an article has been published does not in any way suggest that you (or any of the other referees) 100% completely agree with it – if that were the case, you would be perfectly justified in outing yourself.

It’s also unnecessary to out yourself because we have ways for you to express your concerns. Obviously depending on the publication, you could well write a letter to the publication airing the concerns (again, there is no need to out yourself as a referee when doing this), make public statements and posts, or if these concerns are serious enough to warrant the effort, write a full article (which of course, will be refereed by others who may or may not disagree with you) that airs the concerns in a subsequent issue of the same publication.

Bonus points if you collaborate with this third party whom agrees with you.

Tom Hurka: many journals will not pass on names of referees under any circumstances.

This thread will work better if we all focus on the question at hand: a reviewer outing herself to a third party *after* the paper has been published — not the perennial of whether or not reviewers should be anonymous to the author during the review process.

Irrespective of journal policies (which I suspect do not usually address this question) I would tend to come down, as a default, on the side of not saying “I reviewed this”–the argument being partly the same as that for reviewer anonymity at the review stage, namely to keep personal tensions down. Saying “I reviewed this and made the following criticisms [which were not heeded]” sends the message that the author is (in your view) unreasonable. Perhaps, though, this is the message you want to send — in which case I can’t see any reason why you shouldn’t be able to.

It seems to me the first comments address the question of a reviewer outing himself to the author, whereas Eszter is considering outing herself to a third party.

Personally, I wouldn’t do this, and so I think you made the right decision. My reasons are as follows. First, even if authors should make honest efforts to submit papers of the highest quality possible, it is understood that a paper might be faulty for many reasons. By submitting his paper, I don’t think the author agreed (even implicitly) to make this potential faults public. He rather agreed to have his paper evaluated, and if accepted, to have his work made public. So I don’t think it very polite to discuss the version you read as a referee with a third party and make public your review of it unless it is exactly identical to the published version.

Now, if the paper has been published and if your opinion of it was favorable, I don’t think it is a good idea to mention that you were the reviewer because that could create wrong incentives, as mentioned above (“Oh X has reviewed favorably my article, what a nice guy, I’ll be sure to recommend his articles if they stumble on my desks”). Just say openly that you think the paper is good.

If the paper has been published and your opinion of it was unfavorable (the case at hand, if I understood correctly) then this means that your contribution as a reviewer has been disregarded, and hence minimal. In that situation, I don’t see that your position is very much different from that of any other reader of the published article and thus I don’t see that the little time you could have saved by forwarding your review inside of copying it was really worth breaking the (usually implicit) contract of confidentiality that bound you, the author and the editors.

To sum up, I would say that anonymous evaluation is a valuable good for a healthy scholarly community and so I would err on the side of prudence and break it only if I had really compelling reasons to do so.

Saying “I reviewed this and made the following criticisms [which were not heeded]” sends the message that the author is (in your view) unreasonable.

In my field, I think this would (mostly) send the message that the editors of the journal are unreasonable. Other than that, I agree with what DCA and hb write.

None of the journals for which I have reviewed seem to have a policy that requires that I not disclose my status as a reviewer or the review after the fact. So as a matter of what one may be obligated to do, I’d say it’s up to you. (Although as a matter of practice, I would probably write the relevant journal editor first.) Our annual review process (back before I retired) did require that, if I claimed professional service credit for being a referee, I had to include copies of my reviews with my annual report.

This has happened a few times to me: reviewers have told me they reviewed my pieces, that is, and in one case they disclosed their identity (with the editor’s permission) before the process was over, so that I could contact them if I wanted further guidance on their suggestions. It didn’t seem to me to pose any problems, because these were cases where the reviews were positive. I’ve also revealed my identity, after the fact, to people whose papers I have reviewed favourably. Again, it seemed a nice thing to do. But cdm’s point, in the very first comment, does give me pause for thought. Does the practice leave people feeling beholden to each other? Would it be harder for me to be critical about a paper (where that was appropriate) if I knew that its author had recently been nice about a paper of mine? Anonymity means no-one is obliged to ‘return favours,’ of course, but would the temptation nevertheless be there? Quite possibly, if I’m honest with myself, and I think this provides a reason for not disclosing.

None of this applies to the original question, of course, which was about disclosing yourself to third parties. Here I think there’s a quite different reason for not caution. The reason is this: we understand that in publishing our work, we’re opening ourselves up for public criticism. But someone who submits their work to a journal can be taken to have opened themselves up to PRIVATE criticism and not necessarily to public criticism. They might be mortified to know that the weaknesses of this private version of their papers are then going to be aired before the public. If there are mistakes in the paper, that’s between the author and the editor and the reviewers, they might feel, and they should get a chance to correct them before publication. This is not how I tend to feel: for me, if it’s good enough to send to a journal, it’s also good enough to put up on the web, in general. Usually, anyway. But I also recognise that others feel may differently, and hence that for me to discuss pre-publication, non-publicly-available work with third parties is potentially a (small) breach of trust. At least, I think that angle needs considering.

From what is written in the post I don’t see a compelling reason to break confidentiality. Certainly many people divulge their identities as reviewers, especially when they have been favorable. The reason I was taught that this is unethical is that there is at least the possibility of an unspoken request for a quid pro quo. And this seems a real risk in the usually small circles of qualified and likely reviewers, particularly for high impact journals. I remember being appalled when a prominent scientist (and competitor) gave my supervisor a call to say he had just recommended our paper for publication in a top journal, including enough info for us to know which review was his/hers. Maybe the intent was simply to congratulate us for superlative work, but I certainly interpreted it as corrupt. Outing yourself as a negative reviewer (at least to the author) seems less likely to result in a conflict of interest, and probably the strongest criticism of confidential review is the suspicion that reviewers use the cloak of invisibility for nefarious ends: delaying or placing unreasonable burdens on the submitter. This problem is probably better met by the so-called transparent review process as practiced by EMBO Journal. However, reviewer confidentiality remains the rule there, because it is believed to protect the integrity of the process. Outing yourself to a third party but not the author seems to me to raise the suspicion that the intent is to curry favor or gain influence. In any case, in my experience you have agreed to confidentiality when you agree to review and I do not think that agreement expires upon publication or rejection. Is that wrong?

It didn’t seem to me to pose any problems, because these were cases where the reviews were positive. I’ve also revealed my identity, after the fact, to people whose papers I have reviewed favourably. Again, it seemed a nice thing to do. But cdm’s point, in the very first comment, does give me pause for thought. Does the practice leave people feeling beholden to each other?

It is not just a matter of making people feel beholden. If I positively review a paper, and later reveal myself, this has the potential to benefit me personally. Academia is a small world, and careers (especially for those of us still in the early phases) can be shaped by small acts of favoritism.

Having positively reviewed someone in a department I may eventually be applying to work in (or who one of my students will be applying to) will not get me (or my student) the job. But it may ensure that the application is carefully read, and may get me/my student onto the short list. Or may help getting a panel accepted at a conference, a grant application favorably read, etc.

Positively reviewing someone’s paper and then revealing oneself can also lead to a professional friendship. I personally know of two cases, both friends of mine, who became friends with much more senior people in their fields this way. (In one case it was the senior person who revealed themselves as the reviewer.)

The point is: revealing oneself as reviewer can be of great personal benefit. This in itself presents a temptation to write a positive review. So I say the practice should be explicitly banned.

In my field, computer science, publications at some high-profile conferences carry more weight than most journal articles. In practice what this means is that an annually rotating set of people for a given conference, the senior program committee, supervises the reviews of a much larger set of reviewers. At a guess, I’ve written more than a hundred paper reviews over the years, so there are a lot of people who have seen them. Most of my colleagues are in the same boat. Even with double blind reviewing it’s sometimes not hard for an author and a reviewer to guess each other’s identity, and there’s been some discussion about relaxing anonymity on both sides.

This is different from the situation Eszter describes, in which someone outside the review process is given information about who was involved. I bring it up because it’s part of the complex power dynamics that Eszter mentions and NR fleshes out. Newcomers to a given area don’t have the same connections and background knowledge that old timers do, which puts them at a disadvantage, and that’s not a good thing.

Once in a while I’ve probably said in passing to a colleague, while discussing research by another person, “I was a reviewer on that paper…” I won’t do that any more, given the discussion here.