Date: Sun, 7 Nov 1999 08:43:41 GMT

From: Nick Trefethen lnt@comlab.ox.ac.uk

To: brosa@Mailer.Uni-Marburg.de

Subject: anonymous refereeing

Ulrich,

It's a pleasure to hear from you. Incidentally, I was just at a EUROFTAC meeting in Delft (Nieuwstadt et al.) on transition, and your name came up a number of times. The non-modal point of view, just ten years after your paper, is now more or less mainstream.

I enjoyed your essay on refereeing, and I'd be happy to read an English translation and suggest perhaps a few improvements. I can't say I entirely agree with you, however. I agree with you about the problems with anonymity, but I think you underestimate its virtues. I've been an editor of various journals for about twenty years, and in my field at least, a great number of the papers that come in are dim, badly written, little thought out, ignorant of the literature. Anonymous refereeing enables you to get rid of such papers quickly. (Yes, undoubtedly a mistake is made occasionally.) With signed referee reports, the process would get harder for two reasons:

(1) you'd have to spend hours putting together a legalilstic case to argue why the paper is dim (e.g. by tracking down predecessors in the literature that the author was too dim to find for himself)

and more interestingly,

(2) The people who write dim papers are precisely those people who will not understand what is wrong with them!

Item (2) is a killer. You can spend hours writing a careful argument showing why a paper is empty and incoherent and unoriginal, but believe me, the sort of person who will understand your argument is not the sort who writes empty, incoherent, unoriginal papers. You end up arguing forever and everyone gets upset.

Nick