Luke’s Languages – Sound and Symbol




Huh, that’s funny. Why does my spell-check say that’s wrong? “Hello” is a word, “everyone” is a word… oh, wait. That’s right. Unlike when we speak, when we write we use capitalization, spaces, punctuation, and other funny squiggles.

This week’s topic is the differences between written and spoken language, and what they mean for linguistics.

Something linguists have recently been noticing—or admitting—is that although we have a huge corpus of writing to draw data from, the data might be very different from how the language is actually spoken.

Why? Am I saying that when we’re writing a language, we’re not actually writing it? How does that make sense?

Well, the simplest difference is what you just saw above: one is communication through the sense of sound, the other through sight. They both have advantages and disadvantages. Written language has some useful punctuation, only some of which has been translated back into face-to-face language—like “air quotes”. On the other hand, I was just saying to a friend the other day that we need some way to represent stress and intonation in writing. It doesn’t work so well to, say, capitalize the stressed parts of EVery sentence.

Speaking of which, sarcasm might just be a good thing to be to mark, too.

And did you know that in the 16th century, someone created the backwards question mark to signal rhetorical questions

But the whole punctuation thing is actually a little misleading. This past summer I transcribed about 10 hours of recordings of conversation word for word (don’t try this at home, kids), and I realized that the writing system we have just isn’t designed to transcribe how people actually speak. They start new sentences in the middle of old ones, sometimes landing combos of four or five in a row. They fill gaps with “you know” and “I mean”. They mark some things as jokes by laughing. They speak over each other. Even at a structural level, they design their whole message differently. They repeat things they wouldn’t normally repeat; they build redundancy into their sentences and their paragraphs.

And they definitely use a different vocabulary; they’re more familiar with their tongues than with their fingers. The way people write, you might have a million lines of text and miss half of the casual phrases that make up our daily conversation.

Writing systems, among other things, are cognitive tools. They give us another look at how our language works, another means of communication, that involves a whole different skillset and mastery from spoken language.

When I mentioned this to a certain editor I know, from whom I learned everything I know about language, she pointed out that even though there was no good-looking way to represent all the half-finished sentences, the trailing off, the asides, and all, it doesn’t sound bad when a person speaks. Even though the best written transcription of their dialogue looks like a mess, these people are described as very articulate speakers.


Perhaps the most startling quality of written language was suggested by, among others, Kieran Egan. In The Educated Mind, he says that “natural mind development” mirrors the developmental stages of historical thought. In particular, he sees historical thought going through stages of increasing abstraction that line up with writing systems of greater abstraction. He cited an example of a non-literary Slavic tribe whose members, when asked, “Everywhere up north has snow; Siberia is up north; does Siberia have snow?”, responded, “I don’t know; I’ve never been there!” He suggested that literacy correlates with logical reasoning.

Now, keep in mind that his work is more theoretical than empirical. A counterexample would be the Chinese writing system, in which each symbol essentially makes no reference to individual sounds, only to concepts. In Egan’s model, their culture shouldn’t have so easily  and quickly had the philosophical, literary, and technological advances it in fact did.

But I do like one idea in Egan’s theory: that writing systems, among other things, are cognitive tools. They give us another look at how our language works, another means of communication, that involves a whole different skillset and mastery from spoken language.

Take, for example, a typo someone found in last week’s episode: “Imagination if there was…” That article, like all our articles, was edited several times. And it’s not that we’re shoddy editors, as we hope the quality of our paper in general makes evident.

No, the mistake illustrates a point: how differently the brain treats what it reads from what it hears. Can you imagine anyone making that mistake in speech? It stands out sharply to the ear, but to the eye, it and “Imagine” have a more similar shape. Written language works so differently from spoken language—I mean, seriously, how else could something so obvious have passed inspection by three different pairs of eyes?