Why scientific writers should study writing scientifically: the value of research into written communication

Abstract

In the November 2016 EASE Newsletter, Julian Venables proposed "Five golden rules for writing a good scientific paper. "1 Rule #5 is "Be clear -sentences should be less than 25 words long. " He continues: "The 20 + 10 rule says that for each 10 words above 20 a sentence becomes twice as hard to read. So for a 70 word sentence, which is not uncommon for francophone scientists, there is an 'excess of 50 words' ie 5 extra 10s, or a 2x2x2x2x2 = a 32-fold increase in difficulty!" [Spelling and punctuation are as published.] Unfortunately, Rule #5 won't necessarily improve writing.2-6 It's actually a myth, and one that harms us professionally when we perpetuate it.7 I don't mean to be critical of Dr Venables. Most of my fellow medical writers and clients who are nonnative English speakers are also unaware of the research base and advanced skills that are a part of our profession. I learned them in the 1970s, when I was trained to be a technical writer by technical writers. My sense is that people entering the profession of medical writing in mid career do not get the same orientation or training. This lack of awareness is curious, if not disturbing, for another reason however. If physicians study medicine, psychologists study psychology, and historians study history, why don't writers study writing? Why the rule won't necessarily improve writingThe Rule is puzzling. What does it mean if a text is "twice as hard to read?" What is a "32-fold increase in difficulty?" I suspect that "reading difficulty" is the concept of "reading grade level, " which is the outcome of some readability formulas. The usual understanding of reading grade level is that a text written at, say, an 8th-grade level can be understood by a reader with at least an 8th-grade education. Not so. It means that half the 8th graders tested for their comprehension of a standardised text can answer correctly half of the questions about it.8 So, if we test 10 adolescents and only 5 answer 5 of 10 questions correctly-the circumstance that defines reading grade level-the "comprehension success rate" is only 25%. In fact, readability formulas have long been discredited as a way to guide revision or to evaluate writing.2,9-12 Consider the fact that a text and the same text written backwards, making it meaningless, have the same reading grade level.Back to Rule #5. Shortening sentences, by itself, does not necessarily improve comprehension, 4,5 and may even reduce it.6 Complexity, not length, is what reduces comprehension, and longer sentences have more opportunities to be complex. 13 Sentence length is a factor in many readability formulas, however, which penalise longer sentences as surrogates for increased complexity. The advice to use shorter sentences is not necessarily bad, it's just simplistic; based on correlation, not causation; and not supported by the research. Why the rule harms scientific writersConventional notions of writing are founded on what we learned in college composition classes and practiced in writing term papers in college; that is, on ...