It’s probably a sad commentary on the way my mind works, but one of my more original insights into the writing process is this: writing well is like buying shoes.
You need some new hiking boots, so you go to Nugget Outfitter and browse the back shelves. You find a pair that’s not too expensive and just your style: purple with green laces. You turn to the ever-hovering salesman and ask for a pair in your size.
He brings you a pair in a box marked your size. You try them on but they’re too tight. You need a half-size larger. But they don’t have that size. So you try a whole size larger. It’s too big. You go back to the shelves. There’s no other style that really does it for you, but there’s one that kind of does, so. . . .
That’s writing: we’re always looking for the word that fits best. I draft a sentence, one word is close to what I’m trying to say, but not really. So I find a word that fits better. But then the new word will make me change other parts of the sentence in ways I don’t like and maybe even force changes to other sentences. So I look for another word that fits better, or I start changing the shape of the sentence, but then my meaning starts getting lost. So. . . .
The French call writing well the pursuit of “le mot juste” — the right word, the word that fits.
But a word can be right — and wrong — for many different reasons besides just its dictionary definition. One such reason — one that leads us into the heart of the nature of language — is hinted at in a recent email from my friend Bob Small.
Bob writes: “One of my scars from graduate school is to never write that data suggest something. Data INDICATE something, and humans SUGGEST what the data or results imply.”
As a scientist, a biologist, Bob wants his language to be precise and represent clearly and unambiguously the facts about the natural world that he’s trying to communicate. That’s what we all try to do in writing: find the words that have precise and unambiguous correspondences to what we’re trying to say.
So what do data do?
I am a big fan of etymology, the study of the origins of words, and that’s the first place I turn to answer a question like Bob’s. As it turns out, “indicate” comes from the Latin verb “dicere,” which means to talk, like our word “dictate.” And the word “suggest,” like the word “gesture,” comes from the Latin word “gesta,” which refers to non-verbal ways of communicating.
Etymologically speaking, “suggest” seems closer to what data do than what data don’t.
So what’s the right word in this instance?
Plato tells us a story in which Socrates is talking with his friend Cratylus about the origin of language. Socrates argues that all words originated as sounds of the things they represent — what we call onomatopoeia, words like “burp” or “crash” or “ka-ching.” Cratylus disagrees. Words, he says, have no correspondence to the things they represent other than what we agree to.
Cratylus is right. Language is a set of agreed-upon conventions. The word “four” means the whole number between 3 and 5 only because we all agree that that’s what the word “four” means. It could just as well mean “elephant” in some other language.
(Given my own personal relationship with mathematics, it might as well mean elephant, but that way madness lies.)
Dr. Small’s question about “indicate” and “suggest” leads us to two seemingly paradoxical truths about writing. The first is this: the meaning of a word is fluid, a created metaphor, a fictional correspondence to what the word represents, and that meaning can change.
The second fact is that words come with meanings already intact, meanings that we can’t change without confusing our readers. We have to write with a clear view of what our words mean for our readers. And the only way to know that, to know if the word fits, is to try it on and walk around a bit in your reader’s shoes.
• Jim Hale can be reached through his website at www.jimhalewriting.com