Glasnost factor in peer reviews

IT SEEMS impossible not to endorse the various flaws in peer review of applications for research grants and academic promotions.

We are admonished to go for peak-quality performance, yet power and judgements really seem to reside in a charmed circle of uneven expertise and prejudice whose members understandably are not minded to consider radical changes to procedures that are serving them well. This is especially so when it comes to innovation, originality or unfashionable research. Much potentially important progress can thus be missed and promising newcomers discouraged.

The purpose of this letter is to stress two points: 1. What has been said about grants and promotions applies equally to publications, especially articles in "refereed" journals which, at least in theory, help to determine the outcomes of applications for grants and promotions.

Wearing one's various hats of author, editor and referee, one may occasionally come across quite weird malpractices, for example attempts by one referee (usually only one) rudely to rubbish work with which they disagree or which they do not understand, or to delay publication by repeated piecemeal and sometimes self-contradictory fault-finding.

2. Of the various remedies suggested, and sometimes already tried, perhaps the most powerful would be to ask referees to sign their comments, with the understanding that they are likely to be passed on to the applicants or authors.

Although not problem-free, this measure would not only provide valuable feedback, but refereeing would also benefit. Lest secrecy be thought essential to protect referees and ensure that they tell the truth, with a specialised manuscript it is often relatively easy to figure out the referees' identities from internal evidence but, more important, it is quite possible to be constructively critical without being offensive. This skill could be taught to beginners by elder statesmen, and would probably do away with frivolous condemnations and petty or ignorant nit-picking.

Perhaps the best criterion for referees to bear in mind while formulating their comments would be "could they defend them in a court of law?" Hanna Steinberg

Professor emeritus University College London Visiting professor Middlesex University