On Writing

Really good scientific writing is a rare thing. Very few master the art. I wish I were among their number. Among the stacks of papers, overdue for review and that await my attention, I know I will encounter few that possess that clarity, concision and felicity of expression which mark good scientific writing. Reading good writing is a pleasure in itself, but more, it is, of course, difficult to write bad science well. Good scientific writing leaves shoddy ideas or incoherent reasoning exposed and a scientist with the taste and skill to write well will generally have the sense to consign these to the bin.

It helps, naturally enough, to have good English. Scientists whose mother tongue is not English have a massive hill to climb. The excellent English that so many overseas scientists display is something for which I have awed admiration. I am convinced however that much of the advantage that UK and US science enjoys in league and citation tables is the by-product of the dominance of English as the language of science. Many talented Asian scientists are held back by their language skills to the detriment of science. A small number of non-native English speakers succeed in writing well by adopting a direct and simple style that is well suited to conveying scientific ideas. A significant number of native English speakers adopt a convoluted and baroque style that is much more painful to read than writing that contains grammatical errors.

I would really like to understand how to teach good scientific writing. Obviously practice at writing helps and so does critical feedback. I am less convinced that detailed corrections are of much assistance, but perhaps I am lazy. It is easy to correct the obvious minor grammatical errors and choices of words but this only makes poor writing slightly less grating. The real problem commonly lies in sentence and paragraph structure. Correct these and you soon find yourself rewriting the entire text, hardly the point. I tend to pick illustrative examples and leave the corrections to my students. I think they sometimes find this frustrating, but they are usually too polite to say so.

For the most part I suspect that problems in scientific writing are problems of reading. Poor writers have simply not read enough. Not read enough good writing generally and not read enough well written scientific papers. Of course they have often not read their own writing either! Simply asking a student to read back, aloud, tutorial style, their own drafts often has a striking and salutary effect. Almost all the good scientific communicators I know are people with a broad literary education.

The presentation of text on paper is also important. A poorly typeset paper with ugly fonts, ill-spaced lines, awkward indents and headings that are wrongly sized usually signals a writer who does not care for their craft. Strangely, by encouraging a student to present their work well, you can improve their writing.

I would be most interested to hear from readers who have experience in teaching writing and to get pointers to writing that can serve as models for my students. Feel free to criticise my writing, just do not be surprised if I delete your comment.