Ethics of publishing received peer reviews

What are the ethics of publishing (on the web, etc.) peer reviews received for a paper? Does your answer change if the paper was accepted or rejected?

I am specifically interested in the "one-shot" case typical of computing science conferences, without rebuttals, where there is no or limited dialogue between an author and the reviewers.

This is quite distinct from Can I publish the reviews I write? as here I am talking about reviews I have received, not those I have written - the reviews' authors retain their anonymity, and presumably the paper would be included alongside those reviews.

Vijay's response below includes a summary of much of the other responses and my comments on those.

Answer

Let me add some data points to this discussion. Some items below are about discussing reviews and editors in a public forum and not about publishing reviews.

• Jens Teubner makes available the reviews to his papers online. I do not know him but I have heard from someone who does that he said he has not received negative feedback about doing this. Maybe you can ask him for more information.


• Doron Zeilberger has published reviews in Opinion 87 and for his paper Automatic CounTiling.


• Doron Zeilberger's Opinion 61 is about rejections and accountability (it has a response from Luca Trevisan) and his Opinion 81 is about rejection and snobbery.


• Peter Clark's mail to Doron Zeilberger about some material Zeilberger published online. This is only tangential to your question, but I think it's good to keep in mind that when you go down the road of open publication, you should be ready for others publishing material about you or your reviews.


• The Writings of Leslie Lamport chronicles in very direct terms the stories behind his papers including some notes about editors and reviewers (for example papers 62, 122,129, ).



• In The Writings of Leslie Lamport Paper 132 he talks about having written an 'unkind review'.

I have been meaning to publish my reviews for multiple reasons and I am glad to see that other people have been thinking the same and some have done it. I do not think the result or answer should depend on whether the paper is accepted or rejected. If reviews are published, I believe that one should also publish the version of the article that was used to make the reviews. Otherwise, the reviews are like quotations taken out of context. If there are coauthors, one should obtain their permission first, or at least include a disclaimer that you are publishing your reviews on your behalf only. As an example of a disclaimer, see Lamport's page:

Some of the stories read like complaints of unfair treatment by editors or referees. Such cases are bound to arise in any activity based on human judgment. On the whole, I have had little trouble getting my papers published. In fact, I have profited from the natural tendency of editors and referees to be less critical of the work of established scientists. But I think it's worth mentioning the cases where the system didn't work as it should.

The notification of acceptance is usually signed by the editor of the venue, even in the case of anonymous review, so one should ideally obtain their permission if you will reveal who the editor is. The same applies for reviewer permission in signed reviews. I cannot tell whether people in the links I give above obtained editor permission first so there may be precedent for not doing so. Publishing reviews intended for private circulation still takes them out of context. I would add a disclaimer that the reviews and notification letter were written as private communication and if published without permission of named entities, I would note that too. The latter is in case your intention is to protest the status quo by subverting standard conventions.

I haven't answered your question because I don't have a clear answer. Publishing reviews is not conventional academic behaviour. Doing so can be construed as unprofessional depending on how you publish them and what additional commentary you add. Calling it "unethical" seems a rather heavy handed judgement to make. There are things an anonymous reviewer can do that are clearly unethical (steal research, suppress publication, circulate the manuscript) or questionable depending on context (force citations, comparisons, reject without reading, write ad hominem reviews etc.). There is very little an author can do to wrong a reviewer that is remotely comparable and publishing a review does not seem remotely on that scale to me.

There are multiple reasons to publish reviews including accountability for all parties involved (including authors) and as a form of protest. I believe this intent is important to consider because forms of protest do subvert what may be considered acceptable behaviour. A useful thought exercise might be to put yourself in the reviewer seat and ask if you would be fine with similar treatment. I would not have an issue if the reviews I have written were made public (even if I knew that stylometric techniques could be used to identify me). I do not think all reviewers feel that way.

Finally, let me point again to Jens Teubner's page, which comes across to me as a model of how to publish reviews while retaining professionalism and dignity.