Can you apply science to make writing better?

Thanks to Dr Phil Langton of Bristol University for bringing to my attention this interesting piece in Times Higher Education by Yellowlees Douglas on the way that 'understanding the reading brain can help academics and students improve [their writing skills].'

Douglas, an associate professor of management communication (no, really, they exist) at the University of Florida argues that we have a lot of data on the reading brain - how we take in information from the written word - and that we can make use of that to provide a series of rules for 'science-based writing' which could be taught in secondary schools to improve the quality of writing.

I don't doubt that we could do more to teach writing skills, but to my mind this is a craft, and benefits as much from practice and feedback as it does from a framework of rules, which the best writers break with ease anyway. However, there's something more dangerous here, which is the assumption that academic studies give us a picture of the real world.

Douglas points out that 'Readers best recall the last quarter of lists, sentences, paragraphs and documents', and I am sure that is true when it comes to students doing tests in some psych lab. However, in the real world I would suggest few people read in the same way they would when undergoing a lab-based test. I, for one, rarely read every word in a document unless it's either a contract or one of my books I'm proof reading. (If I'm honest, I don't think I read every word in Douglas's article.)

It's not for nothing that journalists put such a huge emphasis on the first paragraph, rather than the last as Douglas seems to. Because it's often the case when reading an article in a newspaper we don't get past the first paragraph - so it's essential in good writing in that kind of context to get a strong hook in the first paragraph, and a clear indication of what is to come. But the approach would be totally different for a literary novel. We read different types of material in totally different ways - but the most common thing, whatever the material, is that we tend to skip read. Many of us don't read word for word.

When, for instance, I get sent a press release, as I often do, I will usually read the title, some or all of the first paragraph and scan the rest. The whole thing takes about 20 seconds. That's it, unless it has really grabbed me. Delete already pressed. To be honest, even when I read a novel I tend to skim-read if it gets too descriptive and arty-farty.

So to suggest that you can build good writing on a rule like 'make sure your most important points are in the last quarter' (not explicit in the article, but implied) seems to be a product of a sterile university idea of what reading is like, rather than the real thing.

I'm not an arty type. I don't claim writing is purely an artform that has to come from the soul and can't benefit from technique and good writing skills. Like any craft, technique and skill are immensely important. But I'm highly sceptical that 'science' in the sense intended here can do a huge amount to improve the quality of the written word.