Sir

A recent leading article detailed some of the difficulties involved in arbitrating authorship of scientific papers and concluded that attempts to define rules for authorship are “doomed to fail” (Nature 387, 831 831; 1997). Although this is probably true, some of the thornier issues surrounding authorship could be mitigated if journals simply required the contribution of each author to be briefly stated. The rules the authors had applied in determining authorship would then at least be explicit.

For example, a paper (similar to that described in the leading article) on the discovery of a previously unknown hominid using a novel fossil detector might include the following statement: CM: fossil discovery, morphometry, principal author; PW: stratigraphy, assessed paper; MC: isotopic dating; LG: inventor of CRD (carbon replacement detector); PC: funding, intellectual contributions, co-authored paper.

Such a statement could be placed in the acknowledgements and would serve to allocate both credit and responsibility for the work being published. Such information, if generally available, would be widely useful in making hiring and tenure decisions, evaluating grant applications, judging published work submitted for doctoral theses, and, when necessary, determining responsibility for fraud.

Despite its utility, some investigators are likely to view this proposal as unnecessary, inconvenient or even demeaning. Journal editors will want to tread lightly so as not to alienate their contributors. Nevertheless, by taking the longer view (as you encourage authors to do), journals can serve the enterprise of science by helping authors to allocate credit fairly for their work and to accept responsibility for it.

Nature's pre-eminent position among journals of science gives it a unique opportunity in this respect. Few prospective contributors, I suspect, would forgo the opportunity of publishing in Nature simply because they were required to state their respective contributions.