45

I'm reviewing a paper after minor revisions, and can partially see another reviewer's comments in the response letter.

The other reviewer asks the authors to do X, but the authors argue that it's not worthwhile doing X. I've done things like X myself, and have found it not worthwhile (for the precise reasons the authors state). Thus, I strongly side with the authors on this matter, and I'm thinking about whether I should inform the editor. I'm not sure if I'm meant to do this...

Question: As a reviewer, is it inappropriate to make unsolicited comments on another reviewer's comments?

A related question is In peer-review, is it common for a reviewer to be shown and asked to comment on other reviewers' reports? However, I have not been asked to comment on the other reviewer's comments.

Rebecca J. Stones
  • 9,941
  • 6
  • 44
  • 49

2 Answers2

61

It is appropriate. You're acting in good faith after all, the editor is likely to be interested in what you say, and you're not harming anyone.

This is however the kind of thing to put in the confidential comments to editor box. The authors don't need to know of friction between reviewers.

Allure
  • 127,528
  • 50
  • 325
  • 493
  • 3
    Nothing wrong with a debate or disagreement as long as it's polite. If everybody agreed on everything nothing would improve. I don't consider this "friction between reviewers" unless you're disagreeing via ad hominem. – Lightness Races in Orbit Oct 24 '18 at 09:28
  • 8
    Nothing wrong with a debate with the other reviewer, but do it through the editor - it's not something the author should know about. – Allure Oct 24 '18 at 09:37
  • 12
    Why not? The whole enterprise is for the author's benefit. I see no reason for them not to be party to all the feedback and opinions therein. – Lightness Races in Orbit Oct 24 '18 at 09:41
  • 3
    It can cause the author to lose faith in the editor if the paper is rejected, or at least make the editor's job harder. See https://www.forbes.com/sites/dailymuse/2017/06/16/4-thoughts-that-go-through-your-boss-head-when-you-disagree-with-her/#12c79882e2ef point #3 - the same principle applies. Disagreeing with each other before a customer (author in this case) is not good. – Allure Oct 24 '18 at 09:47
  • 13
    I think it's dangerous to treat the scientific process like a business interaction with your boss, and social conservatism ("don't offend me!") has no place in a rigorous academic method, but we'll have to agree to disagree. – Lightness Races in Orbit Oct 24 '18 at 09:48
  • 3
    The avoidance of "don't offend me" is one of the reasons why it's anonymous, yet with the editor in-between to protect the author from actually offensive/insulting/unprofessional comments from reviewers. – gerrit Oct 24 '18 at 11:53
  • 2
    @LightnessRacesinOrbit Wouldn't you say that from the perspective of the journal, the purpose is usually for their own benefit; not that of the author? It could be beneficial for the author to see this disagreement; but for the journal, it might look unprofessional. If the journal is doing this for their own benefit, they would want to avoid that. If journals weren't a business, that would most likely be the best approach. – JMac Oct 25 '18 at 12:32
  • @JMac: That seems to me to be a valid perspective and observation. – Lightness Races in Orbit Oct 25 '18 at 12:56
  • 1
    @Allure You’re right, it might make the author lose faith in the editor — and this would be entirely correct in this context. – Konrad Rudolph Oct 25 '18 at 14:10
  • Seems like reviewers should have a chance to discuss after submitting reviews and update or provide additional post-discussion comments, but the journal does not provide a way to do this. Communicating with another reviewer via your review to the author is obviously bad. – usul Oct 25 '18 at 18:03
  • @KonradRudolph not necessarily, you can't see the entire context. For example if the other reviewer said "this paper has [critical error] and the authors should also [do this]", of which the former is not fixed while the latter is the item OP is referring to, then what? What if the other reviewer submitted confidential information that leads to a rejection? By agreeing with the author against the other reviewer in their presence you put pressure on the editor to accept the article. It's just not a good idea. – Allure Oct 25 '18 at 20:35
22

From my point of view, it is a task of the editor to make sure that you as a referee only get those parts of the communication which are meant for you. Since the response of the authors to another referee's comments is now available to you (intended or not), it is quite impossible to not factor it into your own consideration.

Therefore, I would point out to the editor that, strictly speaking, this thread of the comments is not originally yours -- but now that you have got insight to those points, you can just add a remark saying that in your professional opinion, you deem the authors' response convincing.

This makes clear to the editor that there might have been a mistake in providing that part of the communication to you, and the editor can decide for themselves, whether your stated opinion should be taken into account for the ultimate editor's decision, or not.

nabla
  • 1,239
  • 1
  • 17
  • 22
  • 16
    I agree with this course of action, but in my field it's normal for authors to write a single response letter to all referees, and nobody worries about referees seeing one another's comments. – David Ketcheson Oct 24 '18 at 07:34
  • 3
    @DavidKetcheson indeed. In all my response letters so far, I always included all reviewer comments in one document, without worrying about whether they see it or not. Never had a complaint by the reviewers or editors. I believe everyone sees the benefit in this, after all the purpose of this is to make the paper better. Why not? – Gimelist Oct 24 '18 at 08:29