Could studying Shakespeare improve the way scientists write papers? Credit: © AP Photo

The job of a Nature editor is both enviable and happy. We sit at our desks, receiving, without the fuss of having to get up and go there, news of discoveries from around the world. But it is so painful when potentially important research is conveyed in the strangulated convolutions typical of the scientific paper.

We unscramble the well-meant prose of authors as much as we can, but it would be so much better were authors to try harder themselves. "If you want to make a silk purse out of a sow's ear," as someone once said to me in another context, "start with a silk sow."

When space in journals is at a premium, and when success is measured in terms of publication, it is in the interest of researchers to write clearly and plainly, and with feeling.

This is not literature so much as the delineation of territory: like dogs who mark fire hydrants, the message is not intended for everyone, only competitors.

Here's my advice. In true Nature style I shall give the punch line up front: when learning how to communicate effectively, study the works of people who really knew how to do it.

A benefit of a liberal education, especially in the United States, is that graduate students can study subjects that are conceptually far removed from science. Therefore I urge graduates to take a minor in English literature and concentrate on the works of authors from the early nineteenth-century, the time after the archaisms of Shakespeare, but before the excesses of Victoriana, when written English enjoyed a period of unforced clarity. My choice would be the shorter poems of John Keats and, especially, the novels of Jane Austen.

Night and day

When I was at school and hoping to study science at university, I realized that the timetable would not allow me to study chemistry and English at the same time. I opted for chemistry, deferring English to a later date.

That date arrived in my mid-30s, after I'd been at Nature for some years. I signed up with my local adult-education programme and, every Monday night for two years, studied for my A-level (high-school diploma) in English literature.

I read the classics from Shakespeare and Thomas Hardy to Arthur Miller and Virginia Woolf, but most of all I learned literary appreciation. I learned the craft of measuring the degree to which an author has succeeded in conveying a message.

I did this purely for pleasure, but to my surprise, I found that it had an immediate and beneficial effect on my work. I could articulate editorial decisions more succinctly, by treating manuscripts as literature.

Pride and prejudice

Scientists? convoluted prose produces a general frustration akin to that felt by the boxer who, while still gloved, tries to peel a banana.

Why are scientific papers sometimes so poorly written? The traditional answer is that poor scientific writing comes from a misplaced (I'd go so far as to say dishonest) objectivity, in which cause people are encouraged to remove themselves from the discourse, as if the experiments magically performed themselves.

Another possible cause is the desire of some scientists in certain specialities to write as densely as possible, so that the audience is restricted to a clique of peers. This is not literature so much as the delineation of territory: for the same reason that dogs mark fire hydrants in ways inaccessible to human apprehension, the message is not intended for everyone, only competitors.

But for those authors who are honestly striving to do a good job, much of the problem could be rooted far earlier, with what I perceive to be the low standard of English teaching in schools. It seems that formal teaching of grammar has been abandoned, possibly in the cause of social and linguistic relativism.

The situation is worsened by the decline of education in Latin, presumably because it is regarded as élitist. I learned Latin and am forever grateful for having done so, because through Latin I learned what few rudiments of English grammar I now possess. The fact that I was never formally taught English grammar I regard as a disability.

Death of the salesman

I support my case with the following evidence. First, people used to write much better in the 'old days'. Just pick up any issue of Nature from before the 1960s, and you'll find that, thanks to a grasp of grammar and vocabulary denied to almost everyone nowadays, the authors are able to articulate the subtlest issues with deft economy. Their prose is far from the roadkill we get these days (submitted by authors who are trying their best, but find themselves barely able to express the ideas boiling inchoately in their brains).

Their convoluted prose contains subordinate clauses stacked one after the other, indiscriminate neologisms and nouns prostituted as verbs. Reading it produces a general frustration akin to that felt by the boxer who, while still gloved, tries to peel a banana.

Second, the authors of the clearest manuscripts we receive are often those whose first language is not English, and who have been taught English the old-fashioned way. Their papers are, of course, the most enjoyable. And they are the most likely to win an editor's favour, emerging as they do like bright buttons from a larger pile of lexical sludge written in the customarily dreadful manner.

In the last analysis, when authors need to maximize every opportunity to get their message heard, literacy will be seen, increasingly, as something that could make or break a paper, and with it, the careers of authors.