Kevin Boone

Technical writing: what do we lose, when we write more readably?

It’s easy, especially in science and technology, to write text that nobody wants to read. Scientific writing is often over-long, over-reliant on the passive voice, and full of vague abstractions. We often avoid referring to specific individuals, even to ourselves, the writers. Generality and objectivity are worthy goals in science, but readers prefer the specific and the personal.

You don’t have to look very far to find guidance on how to write readable English text. There was an explosion of publication on this subject between the end of the 19th and the mid-20th centuries. It’s still worth reading early examples like Mark Twain’s Fenimore Cooper’s Literary Offenses (1895), and the Fowler brothers’ The King’s English (1908). This eventually became Fowler’s Dictionary of Modern English Usage, which is still in print. Of course, there are modern books on writing style too, including some for scientists, like Anne Greene’s Writing Science in Plain English.

Novice writers have to struggle with issues of grammar and spelling, but computers increasingly help here. This article is more about problems of style, which goes beyond simple rules that can be learned. Still, it’s easy enough to distill general principles of good English style from the various publications on the subject. Here are a few which have been suggested for at least a hundred years.

My question, in this article, is whether applying these principles to scientific and technical writing loses subtleties that a reader would have understood. The answer will be “maybe”, but it might still be worth doing.

I’m taking as my example a slightly modified version of a sentence that I read in an academic article about healthcare in the developing world. Sadly, I could have closed my eyes and stuck a pin randomly into any scholarly journal, and found something just as apposite. My example doesn’t demonstrate terrible English writing: frankly, it’s the kind of stuff I’m used to reading. Still, I think I can apply various stylistic principles to it, and end up with something that might have satisfied Mark Twain.

Sentence a is the original; here it is in all its stuffy glory.

a It could no longer be denied that the poor health of the Country’s population seemed to be attributable to insufficient cultivation of agricultural produce.

So – where to start? I’m not very keen on “seemed to be attributable”. All too often, “to seem” is a weak, indefinite verb, which we ought to treat with a bit of suspicion. So let’s start by removing the “seemed”.

b It could no longer be denied that the poor health of the Country’s population could be attributed to insufficient cultivation of agricultural produce.

“Could be attributed” is only marginally better than “seemed to be attributable”, but it’s shorter, and a bit less vague. Still, we’ll have to come back to it later.

What about this “insufficient cultivation of agricultural produce”? I imagine that the problem here is simply a lack of food, not all kinds of agricultural output. So:

c It could no longer be denied that the poor health of the Country’s population could be attributed to insufficient production of food

What about “could no longer be denied”? This is a stock phrase, and doesn’t really provide a lot of information. Perhaps “was found” instead?

d It was found that the poor health of the Country’s population could be attributed to insufficient production of food

“It was found” is still a passive voice expression, and we can easily make it active if we wish:

e We found that the poor health of the Country’s population could be attributed to insufficient production of food

Of course, we can only write “we found” if “we” were the ones who did the finding. Depending on the circumstances, we might say “The public health service found…”, or “Red Cross medics found…”, or whatever. It’s certainly worth making the sentence start with the active voice if it matters, even a little, who actually did the finding. We can only justify the clumsiness of the passive voice if we’re adamant that we don’t want to identify the subject of the verb.

Let’s come back to “could be attributed to”. Perhaps we could simplify this to “resulted from” or “followed”:

f We found that the poor health of the Country’s population resulted from insufficient production of food

I’m not entirely happy with this, because it still leaves the main actor in the sentence vague. But let’s press on.

“Insufficient production of food” gives us three nouns in a row. It might be better to turn the noun “production” into the verb “produce”. We’re saying the same thing, but in a more immediate way. In addition, “the Country’s population” can probably be shortened to “people”.

g We found that people were unhealthy because the Country didn’t produce enough to feed everyone.

This change gives us the bonus of removing the vagueness of “resulted from”.

What next? “Unhealthy” is an abstract state. It does actually matter what health problems people suffered from: mostly scurvy and rickets, in this article, because of insufficient fresh fruit and dairy produce. Using these more definite terms gives us:

h We found that people were suffering from scurvy and rickets, among other conditions, because the Country didn’t produce enough to feed everyone.

“The Country didn’t produce enough” is a little impersonal – it’s not the “Country” that produces food, but people in it; but we can make the sentence even more personal by modifying it to focus on what people really experienced: they didn’t get enough food.

i We found that people were suffering from scurvy and rickets, among other conditions, because they didn’t get enough to eat.

I’m a little worried about “scurvy and rickets, among other conditions”. It really isn’t any better than “scurvy and rickets, etc”, and “etc” is a real no-no. It doesn’t help the reader to know that people suffered from “other conditions”, if we decline decline to tell them what these other conditions were. Either the reader knows what kinds of medical problems are associated with malnutrition, or not. If not, saying “and other conditions” doesn’t add anything. So we can, perhaps, leave this phrase out.

j We found that people were suffering from scurvy and rickets because they didn’t get enough to eat.

“were”, in this case, isn’t just a linking verb – it’s not merely saying that two things are connected. Still, we might consider shortening this phrase, just to make the sentence a little snappier.

k We found that people suffered from scurvy and rickets because they didn’t get enough to eat.

These previous changes are all fairly minor; we can hugely increase the impact of the sentence by making it refer to specific individuals. If the article had earlier mentioned some particular group of people, even as only an illustrative example, we can refer to them here.

l We found that Tomasina and her co-workers suffered from scurvy and rickets because they didn’t get enough to eat.

We could go on, if we had the time and enthusiasm, but I think sentence l probably satisfies Mark Twain’s criteria for sound writing. The original sentence a almost certainly doesn’t.

But…

While sentence l has the advantage over sentence a that somebody might actually read it, each of the changes between sentences a and l modifies the meaning a little.

I struck out “it could no longer be denied” because of its verbosity, but it doesn’t mean precisely the same as “we found”. “It could no longer be denied” quietly implies that, at some point, somebody was denying it. Perhaps the writers accepted the conclusion – that lack of food caused health problems – only with reluctance, when the weight of evidence became insurmountable? “We found…” is completely neutral in this respect – maybe we were surprised by the result, maybe we weren’t. Does this matter? In some circumstances it might.

I modified “the Country’s population” to “people”, again for conciseness, but also with a subtle change of meaning. When we refer to an unhealthy “population” it implies that many people are sick – perhaps everybody is sick. “People” could legitimately refer to a single region, or even a single village. We could, perhaps, say “many people suffered…” but, unless we’re prepared to quantify “many”, the word doesn’t add much context. Perhaps, instead, we could say “people all over the Country…”, but that’s no shorter or more engaging than “the Country’s population”. Brevity, here, comes with costs.

And what about moving from “poor health” to “scurvy and rickets”? I’m following the Fowler brothers’ advice to prefer the concrete to the abstract, and added some useful, accurate information. But have I gone too far? Scurvy and rickets are (usually) disorders of malnutrition; at least, they are organic conditions amenable to medical treatment. But what about the mental health problems that might afflict the unfortunate populace? By drawing attention to particular, organic medical problems, I imply either that scurvy and rickets are the only relevant problems, or at least any other problems were similar. And that, in fact, was untrue in this case.

I replaced “agricultural produce” with “food”. But “produce” implies a deliberate, organized attempt to farm food for the population. If I’m short of “food”, that might be unrelated to agricultural activity; perhaps my foraging activities were hindered for some reason? In my country, we forage only for fun, not out of necessity; the country in sentence a was one one where many people depended on wild food to supplement their diets.

So much for the problems, but are there any solutions? That is, could I rewrite sentence a so that it is both short and impactful, and carried exactly the same information? I tried very hard, but I couldn’t find a way. Every change that made the sentence easier to read, made it mean something just slightly different.

In my view, it’s fine to write in a way that preserves a subtle meaning, if

For example, I might write “cultivation of agricultural produce” in preference to “food” if I know, and I know that the reader knows, that there are kinds of “food” that are different from “agricultural produce”. That’s different from writing “agricultural produce” because I’m too lazy to think about whether there’s a simpler way to express the same thing. Or, worse, because “food” isn’t highbrow enough.

Even if I’m using this phrase deliberately, and I want to make the point that I’m only considering commercial-scale agricultural activities, I still have to wonder whether the reader will pick up on the distinction, just from the wording I’ve used. Perhaps it would be better to make the point explicitly, if it’s important to the message?

In summary, I think we can apply common guidelines on style to technical writing, and end up with more readable text. However, I’m not convinced that we can always do this without losing subtle shades of meaning.

It takes a lot of experience for a writer to know whether revising a document to use a more readable style will lose or change information. More importantly, perhaps, it takes a lot of thought and time.