Is Peer Review Better Anonymous or Signed?

A line in a lengthy post considering PubPeer (where we shall not go at the moment–PubPeer seems not to have gotten much of a foothold in the corner of earth science GG plays in) got GG a bit grumpy:

Those who critique science need to be protected, otherwise they will never be able to speak freely about a paper[‘]s problems. This is why scientific peer review is conducted anonymously, and why any website attempting post publication review also needs to guarantee the anonymity of its user base.

Actually, in earth science peer review anonymity is optional (Science and Nature are journals that make it very hard not to be anonymous).  More to the point, GG has always signed reviews. It would seem then that GG is a bit of an idiot.

So why sign reviews?  Here’s one reason: so people know when you were the one who suggested rejection of their paper (and yes, GG has signed such reviews).  Authors have the habit of trying to guess who wrote reviews; GG’s experience as an associate editor suggests that such guesses are typically wide of the mark, even for experienced senior scientists. By consistently not being anonymous, GG doesn’t take the blame for stuff he didn’t write. And you should take the blame for stuff you wrote that was bad.

Second, anonymity breeds contempt. Flame wars on websites make this obvious, but it extends into professional work as well. A review should be written as though you were on the opposite side of a table from the author, explaining what is unclear, what is demonstrably wrong, etc.  You are criticizing the work, not the worker, and there should be a common goal of making the work the best it can be. With only the very rarest of exceptions, this in the end improves the work the author is seeking to publish. Also, it means you have to be careful not to be flippant: if you sound off about something and the author shows you were fantasizing about how you think things should be, you look like an idiot to both the editor and the author. That extra sense of possible mortification can mean that you will construct your review more carefully.

Third, this blunts ad hominum attacks by the author in responding.  Occasionally GG has seen authors respond with vitriol to a review, claiming profound ignorance of the anonymous reviewer, utterly unaware that they are demonstrating a greater ignorance than the reviewer, who is often an equally accomplished scientist with strong grounding in the questions being discussed.  Knowing who was writing would force the author to be more contrite and careful in rebutting arguments.

Fourth, this provides the author an opportunity to note a potential conflict that the AE might have not known about.  GG has yet to see such an issue, but it isn’t impossible.

Fifth, the author will often acknowledge a reviewer by name if known.  This makes reviewers somewhat more responsible participants in the publication process (and also provide readers with some additional information as to how thoroughly a paper might have been reviewed).

The usual reason presented for protection by anonymity is to prevent retaliation.  For tenured professors, this is ridiculous; the only position where realistically somebody can cause you serious damage (provided they aren’t homicidal maniacs) is if they are a government grants officer (e.g., NSF program manager). Even if somebody really hates you and writes evil reviews on all your proposals, you know what? The panels or program officers evaluating those reviews will learn very quickly to discount them. Admittedly there is a greater risk for junior faculty and students, but there is also a greater upside that usually gets overlooked.  A careful and thorough review reflects a careful and thorough scientist; if your review is well done and persuasive, the author is apt to note who you are in a good way.  Now there are awards for reviewing that are ways for some individuals to gain recognition while preserving their anonymity on the particular review they wrote, but these are more impersonal and less apt to really make a strong impression.

Anonymity might bring out the full extent of the disdain of a reviewer for a piece of work, but it isn’t clear to GG that this is actually the best means of obtaining the desired results.  Maybe another day we shall revisit just what reviewing should and shouldn’t be.