Writing “smart” versus smart writing

An impression of intelligence is readily achievable, even in the absence of significant information value, through the expedient of adhering to the expected usages of a genre associated with intellectual output.

Let me put that another way: You can sound smart without saying much by just following the rules of the “intelligent writing” game.

We all know this, of course. You can use ten-dollar words instead of two-bit ones, and the mental effort associated with their retrieval and decoding will stand in for the mental effort associated with working out information-rich content. More than that, though, words are known by the company they keep; words seen in “smart” content will cue your mind that what you’re reading is smart. It’s just like going to a restaurant with expensive décor and smartly dressed waiters: they could serve you frozen dinners and cheap wine and you’d still assume, at least at first, that the food and bev were of high quality.

But wait. There’s more.

It’s not just what words you use. It’s how you use them. You can—and probably will—lean towards a different grammatical style when writing in formal academic prose than in informal communications, even if you’re saying exactly the same thing. In fact, whatever kind of thing you’re writing or saying, you will tend to adhere to certain rules, certain syntactic and morphological preferences, and you will associate things you read and hear with specific contexts and purposes just on the basis of their style of usage.

Here, try these:

Lay fish flat on plate.
Put the fish flat on the plate.
The fish are to be placed flat on the plate.

Those aren’t the cakes Jack baked.
Those are not the cakes that Jack baked.
The cakes in question were not baked by Jack.

Now, your high school English teacher probably taught you a few “rules” that you try to abide by. Most of those aren’t real rules; they’re just etiquette for certain genres. Some of them are quite the opposite of the expected style in some other genres. Passive voice, for instance, is to be avoided (though not strictly) in lively writing, but in many academic journals you may not use first- or second-person active voice: you must write not “We selected the specimens” but “Specimens were selected.” (This is thanks to an ideology of objectivity: the idea is that this is science and so the subjective has no place in it—though of course anyone who actually understands how it works should know that the specific position and pragmatics of the observer are unavoidable and should be acknowledged, not masked.)

Beyond that, the mental schemata of different approaches to information and communication vary considerably. We conceive of knowledge as being like a library of things, both solid and abstract, and of states attributed to things. We conceive of stories as being sequences of actions. We think of science as requiring explicitness (except for that bit where we pretend that we as the observers and actors don’t exist); we think of social interaction as being easy and casual and relying on implied details to create a communal bond through shared understanding.

And so we get different genres. A genre is a style of text (written or spoken) suited to a specific context. It implies a certain kind of author or speaker, a certain kind of audience, a certain kind of context, a certain purpose for existing; it has an expected structure (letters often begin with “Dear —:”; ads often end with a call to action; stories normally have a certain kind of narrative arc); it tends to use a certain set of words and grammatical constructs (also known as a register—a term that can be used overlappingly with genre depending on who’s writing).

Various people have studied genre in various ways. One of my favourites is an American linguist named Douglas Biber, who has analyzed numerous texts of different genres, identifying frequency of certain features of usage. For his book Dimensions of Register Variation, he ran multifactorial analyses on these and identified several dimensions that account for some of the differences between genres. Each dimension is associated with various tendencies in usage.

So, for instance, of the 11 dimensions Biber identified, number 1 accounts for 26.8% of shared variance between genres; number 2, 8.1%; and on down to numbers 10 and 11, which account for 1.9% of the variance each. Dimension 1 has a positive loading of 0.96 for private verbs (e.g., assume, believe, doubt, know), 0.91 for that-deletion (I think you will in place of I think that you will), 0.90 for contractions, 0.86 for present-tense verbs, 0.86 for second-person pronouns, and on down through several more features, and it has a negative loading of –0.80 for nouns, –0.58 for longer words, –0.54 for prepositions, and so on through a few more. The genre most strongly positive for dimension 1 is face-to-face conversations; the written genre most strongly positive is personal letters. The genre most strongly negative is official documents, followed by academic prose and press reportage. On the basis of all this, Biber has named this dimension “Involved versus Informational Production.”

Dimension 2 is “Narrative versus Non-narrative Discourse” and loads most strongly positively for fiction (of course) and most negatively for broadcasts, followed by hobby texts, official documents, and academic prose. Dimension 5, also of interest for us, is “Non-abstract versus Abstract Style”; it loads negatively for technical and engineering prose, academic prose, and official documents, and most positively for fiction and face-to-face conversations. It is associated entirely with negative features: –0.48 for conjuncts (e.g., consequently, furthermore, however) –0.43 for agentless passives, –0.43 for past participial adverbial clauses (Viewed objectively, this tells us…), –0.41 for by-passives. It accounts for 2.9% of shared variance between genres.

What this all says is that if you want to sound like you’re writing academic prose, you can load up on nouns and prepositional phrases, use longer words, avoid contractions and other hallmarks of informal usage, use passives more, and avoid talking about yourself and your reader.

The problem is that this is also much more likely to bore the readers. It’s abstract and uninvolved. Any writing coach or storyteller—and certainly any advertising copywriter—will tell you that if you want to get people to pay attention and remember things, you need to draw them in. Give a narrative, give a personal connection; breathe hot wet life into these dusty cool pages. People buy a lot more novels than textbooks when left to their own devices.

Our dilemma is that if you write scientific fact in too engaging a way, you risk people not taking you seriously. The engaging results in disengagement. A spoonful of sugar may help the medicine go down, but a bitter taste is required for them even to believe it is medicine and take it in the first place. If you’re doing a double-blind placebo-controlled trial of a certain drug, you will find that the placebo likely has a stronger placebo effect if it’s believably medicinal. And so you can write vapid prose in a dry scholarly style and achieve a textual placebo effect, while if you write information-filled prose in too lively a style they just… won’t take it. Seriously.

This is not getting better, either. Historical studies of different genres have shown that while some genres have increased in informality—plays, newspapers, and political speeches are all much more involved and non-abstract than they were a century or two ago—scholarly writing has actually tended to get drier, more abstract, less involved. There’s a great insecurity, it seems. I found that certain kinds of particularly vivid words (e.g., glow, glare, spray, splash, twirl, dump—all having sound clusters that associate them with a physical expressiveness) correlated fairly well in use with Biber’s dimensions 1 and 5, but most notably were strongly avoided by texts that were intended to sound formal and intelligent. The use of such words could make a text seem a little too… well, childish. Welcome to the wonderful world of linguistic insecurity: self-presenting intellectuals trying to shout down their impostor syndrome. And readers judging accordingly too, some more harshly than others.

So you have to gauge your audience. Use features that will cue the engagement just enough, while still calling up features that signal that the text should be taken seriously. This will vary according to the audience and content, and it requires developing your sense of what will work, and where, and how.

The point is this: It’s possible to sound intelligent without saying much of value by following the rules of an “intellectual” genre, and it’s possible to sound unintelligent while conveying valuable information; it’s also possible to have the best of both worlds. For example, a writer who writes in a non-abstract, narrative way, but who is sparing in use of first- and second-person pronouns as well as contractions and other informalities—and who perhaps tosses in the occasional sesquipedalianism—may thread the needle quite well.