The face that won Word of the Year

Every November, Oxford Dictionaries announces a Word of the Year — the term that, in its editors’ judgement, best caught the mood of the preceding twelve months. Vape had taken it in 2014. In 2015 the winner was 😂 “Face with Tears of Joy.” Not a word at all, but an emoji. For the first time in the prize’s history, the Word of the Year had no letters in it.

It was a slightly cheeky choice, and Oxford knew it. But the data behind it was real: working with the keyboard maker SwiftKey, Oxford reported that this single face accounted for roughly a fifth of all emoji used in the UK that year, and about a sixth in the US. A picture had become so common, so useful, so sayable, that a dictionary felt it had to notice. The question worth asking is how a little face ever got good enough to beat actual words — and the answer runs back through three dates most of us have never thought about.

Read it sideways

The story usually starts on an electronic message board at Carnegie Mellon University, on 19 September 1982. Posts in these early online threads kept going wrong in a particular way: someone would make a joke, a reader would take it literally, and an argument would break out over nothing. A computer scientist named Scott Fahlman suggested a fix. “I propose,” he wrote, “that the following character sequence for joke markers: :-)” — and then the crucial instruction — “Read it sideways.” For sad or serious remarks, :-(.

Tilt your head and a colon, a hyphen and a bracket resolve into a grinning face. Fahlman did not claim to have invented the idea alone — he calls himself the inventor “or at least one of the inventors” — and other people had floated other joke markers before him. But :-) is the one that stuck, and it founded a whole tribe: the emoticon, a face assembled out of ordinary punctuation. Its genius was that it needed nothing new. You could already type it.

A joke that arrives without a face is a joke that half the room will take seriously. The emoticon was a tiny repair kit for a brand-new problem.

A picture character

The leap from a sideways smiley to the full, coloured zoo on your phone today happened on the other side of the world. In the late 1990s, mobile phones in Japan were learning to send text, and text alone kept causing the same misreadings Fahlman had worried about — only now between millions of strangers. At the carrier NTT DoCoMo, a young employee named Shigetaka Kurita and his team drew a solution: a set of 176 tiny pictograms, each a mere twelve-by-twelve grid of pixels, to carry the tone a bare sentence couldn’t. They released it in 1999 for DoCoMo’s i-mode mobile-internet service. Kurita has said he drew on the visual shorthand all around him — the symbols in manga, the icons on weather forecasts, the dingbats in a typeface.

That set became famous enough that in 2016 the Museum of Modern Art in New York acquired the original 176 for its collection. It is worth being careful here, though: “the inventor of emoji” is a tidy label the history keeps resisting. Kurita himself has stressed he was one of several hands, and later research has pointed to an even earlier pictogram set from the rival carrier SoftBank. Invention is rarely a single person in a single room; it is usually a good idea that several people reach for at once.

There’s a smaller legend that explains why DoCoMo cared so much — and it’s a lovely one, if told with the caution any well-worn story deserves. On the company’s earlier pagers, wildly popular with Japanese teenagers, you could send a little heart symbol. When a later, more business-minded model quietly dropped it, the young users were not grateful for the extra professionalism. They were furious. The lesson, as the story is usually told, landed hard: a single tiny picture could carry something a page of text could not, and people would not give it up without a fight.

The word that was never about emotion

Here is the twist almost nobody sees coming. The word “emoji” has nothing to do with “emotion.” The resemblance is a complete coincidence.

“Emoji” is a Japanese word — 絵文字 — built from e (絵, “picture”) and moji (文字, “character” or “letter”). It means, simply, picture character, and lexicographers note it was in Japanese use long before smartphones — the Oxford English Dictionary traces the term back to at least 1928. The fact that it sounds like “emotion” and like Fahlman’s “emoticon” is pure chance — a happy accident that, etymologists suggest, probably smoothed the word’s path into English, because it already felt like it belonged. (We’ve seen this trick before: a word that looks like it came from a language it never did.)

How a face left Japan

For years, Kurita’s pictures were trapped. Each carrier had its own set, and a face sent from one network could arrive on another as a meaningless box. What freed emoji was the quietest and most powerful force in modern writing: standardisation. In 2010, the Unicode Consortium — the body that assigns every character on the planet a single agreed code, so that an a or a あ or a ا means the same bytes everywhere — folded emoji into Unicode 6.0. From that moment a face typed in Tokyo would arrive, as the same face, in Toronto.

That is the unglamorous machinery behind the 2015 prize. An emoji could win Word of the Year only once it had become something everyone, on every phone, could read — a shared alphabet of feeling, agreed on by committee and adopted by billions. In barely three decades, a sideways joke marker had grown into the youngest writing system on Earth.

Why the dictionary points the wrong way

And here is where emoji stop being a tech-history curiosity and start sounding a lot like the rest of this journal. Because the official meanings of emoji are, increasingly, wrong.

Unicode still lists 😭 as “loudly crying face.” Yet scroll any group chat and you will find it stamped under the funniest message of the day, or the most overwhelming piece of good news — it has drifted to mean something closer to “I cannot cope, this is too much,” in delight as often as in grief. The skull, 💀, is not about death; it means “I’m dead” — as in, dead from laughing. The literal description on the character chart would lead you to read the room exactly backwards, the same way knowing that Italian cavolo means “cabbage” tells you nothing about the temper of the person shouting it.

Nobody ever handed you a glossary of emoji. You learned that a skull means laughter the way you learn everything that matters in a language — by watching it happen, over and over, until the meaning is simply there.

Think about how you actually came to know these. There was no lesson, no flashcard reading 💀 = hilarious. You met the little faces in context — in a friend’s reply, under a video, at the end of a text that made you laugh — and you absorbed not just a definition but a register: who uses it, when it’s warm and when it’s sarcastic, how it changes an entire sentence sitting at the end of it. That is real vocabulary knowledge, and you built it the only way such knowledge is ever built: from repeated encounters in the wild.

Where vocabla fits

Emoji are the proof, running live on a billion phones, of the thing vocabla is built around: a symbol’s meaning lives in its use, not in its label. A flashcard that reads 😭 = crying would be technically correct and would leave you misreading your friends. What you needed instead was the sentence it sat in, the mood it carried, the hundredth time you saw it land.

That is exactly why we show you new words inside real sentences rather than as bare translation pairs — so each one arrives with its temperature and its situation attached, and met again and again until it stops being a shape you decode and becomes one you simply know. You already did it once, effortlessly, for a whole alphabet of little faces nobody ever taught you. A French, German, Italian, Spanish or Russian word is learned the very same way. 😊