Chaucer 49; Dickens 20; JK Rowling 12* – mean sentence lengths (MSL) in words. And it’s no coincidence that the more recent the author, the shorter the MSL. Admittedly, the works of JK Rowling are not strictly comparable with those of Chaucer and Dickens. Still, both Rowling and Dickens appeal(ed) to a broad public, and literary analysts agree on the general trend in MSL–a trend that is accelerated by computers and the Internet. My own analysis has a simple message: full-length science papers need substantially shorter sentences if they are to be thoroughly read by most of their potential readership. We’re all in the same quandary – BioEssays too – so I’ll make no pretence: it’s not easy for any of us to write really short sentences, editors included.
But we can no longer regard the scientific literature as a communication form with completely separate rules, or exceptions, from the rest. The gaps – “communication niches” – between lay and professional communication have rapidly closed. The Internet places diverse genres of written works side-by-side. Ever more researchers use Google et al. to find relevant literature; and if a reader finds one particular paper too taxing to read, an alternative source of the information – in more digestible form – is increasingly frequently just a click away. Ever more, bloggers and other science writers write for audiences that include researchers. And a growing number of scientists – some of them almost professional bloggers themselves – write brief communications for their own community. I believe that scientific articles are, increasingly, in competition with such writings. Whether scientists or not, few of us will disagree that the short-winded sentences of science writers are usually more pleasing to read than the average peer-reviewed scientific article.
So what are we up against in concrete terms? Here is my preliminary analysis** of the peer reviewed literature: MSLs of between 25 and 30 likely make up the majority. A paper with an MSL of 17 did crop up, but such works are rare. And the “competition”? Some prominent lay science writers regularly come in at around 15 words, I found. Even some scientists writing blogs for members of their own community manage a respectable 20; sometimes fewer.
Clearly this “competition” is not active rivalry. However, at the point of reading online, articles vie for attention. Good Internet writers do a great service to science. Some even contextualize and add critique. However, such digests of articles are inevitably interpretations by more generalist writers: they are not intended to be as thorough as the source article. For lay readers, that might not matter: for scientists it should.
Commentators on modern culture such as Andrew Ferrier, Seth Godin or Frank Schirrmacher regularly remark that computer/Internet use reduces users’ attention spans. More interesting – as they also observe – is the neural mechanism: a defined craving for short-term gratification. And in the case of articles it seems to me that the gratification is uniquely about understanding. A short sentence is easier to understand than a long one. Even if a short sentence gives only a fraction of the final information, it is enough to keep us going: we have learnt something new.
MSL aside, articles frequently contain a few sentences with over 60 words: one in my analysis “captured” the experimental motivation and design in 69 words. Crucial information should be written in short chunks. A few massive sentences can seriously diminish reader understanding, and hence gratification! Here are some practical suggestions: Don’t start a sentence with a qualifier such as “although”, “because” or “since”. Whatever we were taught at school, starting a sentence with “And” or “But” is a permissible way of making a break. Splitting sentences with a colon “:” or similar is also a good idea.
We can’t do much as individuals about the march of Internet- and computer-driven behavior. If we want the fullness of science in necessarily long papers to be appreciated, it must increasingly be written in short sentences.
*Numbers are rounded to the nearest integer. Sources, respectively: Sherman LA. 1803. Analytics of Literature: A Manual for the Objective Study of English Prose and Poetry. Ginn & Company, Boston; Gunning R. 1964. How To Take The Fog Out Of Writing, Dartnell Corp.; analysis of “Harry Potter and the Sorcerer’s Stone”
**Full-length cell/molecular biology articles, three each from BioEssays, Aging Cell, Nature, Science. 30 sentences from each were analyzed: 10 consecutive sentences from three different sections of body text were counted using word-count function in Word. Citations and bracketed references to figures, tables etc. were subtracted. Colons and semi-colons were taken as sentence breaks.
