Two hats too many

We view ourselves as scientists as very intelligent, extremely hard-working, and highly creative, though perhaps more competitive than we care to admit. But no one seems to recognize the high incidence of a serious medical problem among investigators — 'multiple personality disorder'. Of course I say this with tongue planted firmly in cheek, but it's not too far off the mark when we consider our dual roles as authors and as peer reviewers. Most of us review the literature (hopefully beyond that accessible on-line) and cite earlier work, to give credit where it is due, to point out unsettled questions, and to support our conclusions. This is both expected and demanded of us by the norms of the field. One would thus imagine that as referees of submitted papers, we would bring to our evaluations the same careful analysis, thoughtful consideration, and scholarship that goes into our own research and manuscript preparation. Yet we are all too well aware that this is not always the case. Even to neutral third parties, many reviews seem to be hasty, intemperate, biased, and/or narrow-minded. Why is this? How do most of us, with a real love of what we do and commitment to getting to the 'truth' of the biology, wind up as Mr (or Ms) Hyde alternatives to our usual Dr Jekyll personas? In some (hopefully rare) cases, the evaluator is truly ignorant of the subject matter, or a review is intentionally damning to provide the referee with a competitive advantage, or the intent is to suppress work disagreeing with what the reviewer previously published. More often, time constraints, subconscious prejudice, and the difficulty of seeing past existing dogma are offered up as excuses for poor reviewing. But none of these seems irremediable. The major defects in most reviews fall into three general categories: gratuitous dismissal of the potential value of the work, statements of 'fact' left unsupported by specific citation, and demands for new experiments that are either impractical, patently impossible, or unlikely to change the proffered interpretation. The first is the easiest to eliminate — it simply requires the reviewer to step back from the initial emotional response to the paper ('What a crock!') and ask 'What if?'. What if the authors added some new data or revised their interpretation? Would the product be a significant contribution that enhances our understanding of a subject? We may be rightfully disappointed that …