Sir

In your Editorial 'Working double-blind' (Nature 451, 605–606; doi:10.1038/451605b 2008), you suggest that authors may be vulnerable to bias if referees guess their identities — for example, bias about their previous work, their gender, their nationality or their being new to a field. A high reputation could inhibit impartial assessment, although this bias is more likely to disarm than antagonize a reviewer.

The days when the gender or the nationality of a scientist could always be ascertained by her/his name are long gone; for example, try guessing my gender or nationality. I have met people from a non-European background who thought that the French name Jean-Marie was a female name.

An author (or an entire team) embarking on a new area of research is easily spotted: the bibliography will contain no or few references to previous works by the author(s) of the paper under review. Scientists who frequently refer to their own work are readily identified; one does not have to be a Newton to be recognized by one's paw prints (Nature 333, 592; 1988). Double-blinding will not curb excessive self-citation because those who succumb to vanity know that they will be recognized anyway.

Regarding the open sharing of information, readers of a paper could benefit not just from the published paper, but also from the referees' reports if these were pithy and incisive.

Once upon a time, some journals appended to each paper the comments and the names of the reviewers. How about adopting this as an alternative to double-blind review, and rewarding scholars for all their contributions, original as well as critical?