A Linguist Explains Emoji and What Language Death Actually Looks Like -The Toast

Skip to the article, or search this site

Home: The Toast

If you pay attention to the emoji news, you’ve probably heard the statement that “emoji will cause the death of English”. You’ve probably sensibly rejected this as doom-mongering hyperbole already. Which it is, so good job you.

But this statement isn’t just hyperbole, it’s got a truly spectacular number of incorrect assumptions packed into its seven words. In fact, it’s got so many levels of wrongness that it’s actually a fantastic opportunity to explore what language is, what language isn’t, and how emoji fit into this whole thing. Let’s put our linguist hats on and get ready to analyze.

The first level of wrongness is assuming that emoji are the kind of thing that could ever be used as a true replacement for English, that emoji are a language the way English is a language. The thing that gets me about this “language” hyperbole is that either the people saying it have never actually used emoji for more than five minutes, because if so they’d discover that emoji are woefully inadequate as a replacement for All Words Ever, or they’re somehow managing to use emoji and just be completely unreflexive about their own usage.

This comes, as far as I can tell, from the fact that we seem to mean two possible things by the term “language”. One is any means of communicating, like when we talk about body language, or the language of dance or art or flowers or bees. And of course, we can certainly communicate with emoji, just like we can communicate with improvised gestures in a foreign country. But ad-hoc gestures and dance and art have co-existed for a long time with languages proper.

And that’s the other thing that we mean when we say “language” — a particular, conventionalized system for representing abstract meaning, like English or Japanese or ASL. Here’s a fun twist on a classic book meme to prove my point: pick up the closest book to you, open it to a random page, point your finger at a random sentence, and try to “translate” it into emoji. Show this to someone with no context, and see if they can figure out what the hell you’re trying to say. Compare this to the experience of reading a book in translation. Sure, you’ll probably miss out on a few subtle nuances from the original, but you won’t be hopelessly lost.

Improvised tourist gestures are not like sign languages, by the way — signs are fully capable of communicating abstract, conventionalized meanings that are only understandable if you know the language and which don’t make any sense if you say them, for example, in ASL to someone who only knows BSL. And when you get people together who’ve all been using ad-hoc gestures, within a generation they invent a proper sign language instead.

There’s good reason to think that emoji are more like gesture than language. When you crunch the numbers (and I have), the face, hand, and heart emoji are by far the most popular — not the emoji that represent noun-like items. Furthermore, the vast majority of emoji are used beside words, not all by themselves in extended emoji-only stories. People aren’t using emojis as a substitute for language, they’re using it as an addition to it, just like you wouldn’t want to talk in person with your hands tied behind your back and a paper bag over your head.

Why, then, isn’t anyone proposing that interpretive dance or ad-hoc tourist gestures are going to drive out English? Time for another glorious misconception!

Emoji, you see, can be written all in a line with words. Using a keyboard even! Doesn’t that mean it’s basically like hieroglyphs?

Well, no. A language is not its writing system. English would still be English even if it had never been written down, or if it were instead written with the International Phonetic Alphabet, or Cyrillic, or Katakana, or an entirely novel writing system, however well or ill-adapted. When you say it aloud, it would sound the same.

In fact, a fair number of languages have historically changed their writing systems, like Turkish (Arabic alphabet to Latin alphabet), Korean (Chinese-based characters to Hangul), Turkmen (Arabic to Cyrillic to Latin), or, actually, Old English (runes to Latin). You have to retrain people in how to read, yes, but there’s no break in the spoken language when people stop understanding each other.

And in fact, this is the closest emoji come to being a threat to The English Language As We Know It, in that it is technically possible, if incredibly unlikely, that we could encode English sounds in emoji. We’d have to establish some conventions, say, that the tears of joy emoji stands for the first sound in “joy” and the hearteyes emoji for the first sound in “heart” and so on, but technically speaking, we could do it.

Of course, we’d face the massive task of convincing billions of English-speakers to learn and use it in place of the alphabet they already learned, but let’s say we somehow accomplish that. (Despite the fact that we can’t pass basic spelling reform or even a sarcastic punctuation mark.) But even then, we’d just have created a cipher, like those codes that convert A = 1, B = 2, and so on. We’d still be writing English: this would have no effect on our speech, since we’re not going to start carrying around giant flipbooks of emoji in order to spell out our words, any more than we currently carry around notepads so that we can write to each other in the Latin alphabet.

Or, we could write with emoji in a slightly different way, taking advantage of the fact that emoji have both sounds and meanings associated with them. For example, we could write “catalogue” using the cat emoji and the log emoji for sounds, together with the book emoji as a “written thing” classifier. We might want to smush them all together into a square so that we know they’re all part of the same word (or else how would we know that “book” wasn’t part of the next word, which uses the “book” emoji as a rhymes-with and the chair+house emoji to indicate part of a house, thus representing “nook”). If you add in a couple thousand years of the symbols getting more and more stylized and conventionalized, we’ve basically reverse-engineered Egyptian hieroglyphs or Chinese hanzi — both of which contain a combination of sound indicators and meaning indicators. Neither of them is as simple as “little pictures”— or else you wouldn’t need a trilingual carved rock or several years in school to learn them, respectively.

But the thing is, even if phonetic emoji or phonetic+semantic emoji are technically possible, they’re starting to seem awfully complicated. Sort of like a non-streamlined version of the complicated and arbitrary writing systems that we already have in most languages (with the exception of a few that were designed from scratch). That’s not what emoji were supposed to be! Weren’t emoji supposed to fix all our problems?

And that brings us to the next level of wrongness: first, the assumption that the mechanism by which one language drives out another is based on inherent properties of the language, its logic, its simplicity, its beauty, its capacity for complex expression — and second, that emoji are better at some of these things. Both of these are false, but let’s start with the second one.

We’ve established that emoji can’t say everything that English can say, but what if, says the hyperbolic futurist, that’s because they’re BETTER? After all, gosh, aren’t actual languages difficult to learn? You can spend months and years learning Swahili or Portuguese but with emoji, you understand them instantly! (As long as someone briefs you on what the eggplant means before you ask for a bunch of them from the supermarket.) Mightn’t that be because emoji are a “universal language”??

Yeah, so, that’s the problem. Emoji are a universal language the same way that pointing at stuff and grunting is a universal language. Useful, under a certain set of circumstances! But what makes language really powerful is its ability to talk about stuff beyond the here and now, beyond the easily visualizable. In other words, abstraction.

And you can’t have something that’s both abstract and universal at the same time. It’s a contradiction. If it’s universally, instantly understandable, it’s got to be really simple. If it’s abstract enough to talk about anything interesting, it gets that way because of a bunch of arbitrary associations of form and meaning that you just have to learn by rote. (It’s not a coincidence that learning about a new topic often involves picking up a bunch of new vocabulary.)

For example, look at the tremendous difficulty that scientists have had in communicating the fairly simple concept DANGER THERE IS NUCLEAR WASTE HERE STAY AWAY in a way that will continue to make sense to humans for the next 10,000 years. Circle with a slash? Nope, could be a sideways hamburger. Skull and crossbones? Nope, could refer to the Day of the Dead and/or pirates. Closer to home, there’s a considerable amount of work put into designing universal iconography for international purposes like traffic signs and airports and Olympic events, but even that relies on a mix of shared cultural awareness (e.g. that wavy lines represent water) and just plain arbitrary learning (e.g. that a red octagon means “stop”).

But even if emoji were a language, and even if that language was actually an improvement on English — I can definitely see the pros to a language with a phonetic spelling system and no irregular verbs, for example — that still wouldn’t be enough to cause English to die. In fact, people have designed arguably more logical or efficient languages according to various criteria, such as Lojban which has no ambiguity or Toki Pona which only has 120 base words, and I strongly endorse them as a reason to spend an afternoon reading Wikipedia. But none of them have mounted a serious threat to a natural language. (Esperanto is the conlang that’s caught on most, and it’s not especially logical.)

Thing is, languages don’t live or die on their grammatical merits. I too, enjoy learning about the history of English and the unique quirks of Englishes ’round the world and all the stuff we’ve borrowed from other languages. But there’s nothing about our sounds or our words or our spelling system or our grammar that makes English particularly fit to be a global language. English is a global language because English speakers have been global conquerors. It’s not about the quality of English nouns and verbs, it’s about the quality of English guns and money.

There are, of course, gun and money emoji. But there hasn’t been physical violence inflicted on people who refuse to use them. We can’t say the same for English nouns and verbs.

This gets us finally to the most troubling part — the idea that emoji might cause the death of English is a severe mischaracterization of what it actually looks like when a language dies.

Even if you don’t like a few details of how Young Women These Days are speaking it, a changing language isn’t a dying one, it’s a living one. And we can’t deny that millions of children go to school in English, play at home in English, and will one day be able to get jobs in English, not to mention the further billion or so adults who speak English as a second language.

No, English is in ruddy good health. In fact, it’s not only not under threat, it’s the aggressor. The only problem with English is the way it’s pushing out thousands of smaller languages. You see, it’s not that languages can’t die — quite the contrary. The precise numbers vary, but it’s commonly accepted that of the 7000 or so languages currently being spoken, at least half of them will no longer be active by the end of this century.

There are many reasons that the children of a community may stop speaking the language of their grandparents, whether because they can’t go to school in their mother tongue, because their parents decide that speaking a majority language will help them get a job, because of inaccurate advice that speaking multiple languages to a child will just “confuse” them, or because of governments that literally ban their language from being spoken.

That’s not to say there isn’t hope — language activists have been working on many projects to revitalize endangered languages, and even to reawaken “sleeping” languages from written records into active, daily life in their communities (see the documentary We Still Live Here for one example). It’s challenging work, and chronically under-funded, but it’s happening. And that makes it all the more frustrating when the kind of “language death” that makes the news is hyperbolic techpanic or simplistic language savior narratives.  

So what will make a language survive the 21st century? It’s being spoken by children and young people. It’s being considered “cool” by teenagers, important by young parents, and valuable by the societies they’re bringing their children into. And often, it’s the language being used in texts and tweets and Facebook posts and every other part of modern life. And therefore yes, it’s emoji.

It’s not that emoji are killing the English language — they couldn’t if they tried. But it may be that a language that people are not putting emoji next to — that’s a language that’s in trouble.


It’s been wonderful being your Resident Linguist at The Toast!

If you’d like to keep having regular doses of linguistics in your life, I have a daily linguistics blog called All Things Linguistic which can be followed via rss, tumblr, twitter, facebook, or google+ as well as a monthly newsletter (email/wordpress) which sums up my top few blog posts, any articles I write or get quoted in, and whatever else I’m doing. You can also follow my personal twitter for mostly-linguistic thoughts at unpredictable intervals.
I’m also writing a book about internet language (!!!), which will be published by Riverhead (Penguin). Here are all my blog posts about it and you can subscribe to the book’s very occasional email list if you want to make sure you don’t miss it when it comes out.

Add a comment

Skip to the top of the page, search this site, or read the article again