The members of the Unicode Consortium had definitely not signed up to become the smiley faces people. In 2000, as emoji first started taking off in Japan, they politely declined to get involved, leaving DoCoMo and SoftBank and KDDI to hash out between each other the compatibility of picture messages (or, in some cases, the lack thereof). If sending small images encoded like text was just going to be a momentary fad in one country, that was below the pay grade of an international standards organization. But emoji hung on in Japan, and multinational companies started getting involved. Gmail needed its Japanese users to be able to send and receive emails with emoji in them. Apple wanted people in Japan to buy iPhones, but they wouldn’t buy a phone that didn’t support emoji. Ten years later, perhaps emoji weren’t just a fad anymore, so in 2010, into Unicode they went.
But which emoji? By this point, SoftBank’s initial set of ninety emoji had expanded as other Japanese carriers had come up with their own additions, so the initial set of emoji that were added to Unicode contained 608 symbols that were common in Japan. Now encoded, emoji arrived on Apple devices in 2011 and Android in 2013. The international support and cross-device compatibility solved a problem for Japanese texters, but it also helped emoji become popular outside of Japan. And become popular they did. Just five years after emoji entered the international stage, in early 2015, the most popular emoji, tears of joy , surpassed the usage level of the most popular emoticon, :).
As more and more people from around the world began using emoji, however, it became more and more apparent that even 608 symbols weren’t enough. People started questioning—If there’s a unicorn and a dragon, why no dinosaur? If there’s a man in a turban, why not a woman in a hijab? If there’s sushi and hamburgers, why no taco or dumpling? All of these emoji have been added, many in response to proposals from ordinary people who figured out the process from the Unicode website or with the help of grassroots emoji proposals organization Emojination. But the set of emoji remains a work in progress: the Unicode Consortium is still taking requests, still rolling out a hundred or so new emoji each year.
Even with these expansions, the official Unicode process is slow and deliberative by design. At its core, Unicode is still a unique, uniform, universal encoding system. The goal is to create symbols that work on every device in the world long into the future. No more blank boxes. This means that once Unicode adds a symbol, they never remove it—doing so would defeat the purpose of a unified standard. That’s also why Unicode doesn’t accept emoji proposals for celebrities and pop culture references. They’d be fun for a while, but our great-grandchildren won’t really need their keyboards cluttered with the faces of one-hit wonders from the early twenty-first century. To get around this, there are individual apps that use an emoji aesthetic with larger, more of-the-moment pictures, which are sent as normal image files rather than encoded like letters, known as customizable emoji keyboards and sticker apps. Plus of course, there are still gifs, and even the regular kinds of images that you find or make for yourself.
As the hype settles down, as we move into a stage where emoji use is becoming ordinary and unremarkable rather than the subject of daily news articles, we need to reckon with the enduring legacy that the first emoticons, and then even more strongly emoji, have left us with. In a few short years, in the span of an internet generation, we’ve radically changed our expectations for what we should be able to do with informal writing. We’re no longer content to leave full communication only to channels that allow faces and voices. We demand that our writing also be capable of fully expressing what we want to say and, most crucially, how we’re saying it. Any solution to this problem would have had to solve some similar problems to what emoji have, but why did emoji in particular catch on so quickly? What qualities do they have that any future would-be rival must meet or surpass?
Why Emoji Won
At a purely technical level, emoji come with some significant advantages, which we can see by comparing emoji to punctuation-based emoticons on the one hand, and animated gifs on the other. Emoticons are supremely easy to type, since they’re made of punctuation already on your keyboard, but there’s only so many recognizable figures you can make out of punctuation characters. Emoticons are good for your basic couple of smiley faces, but they become less practical as they get more elaborate—users of kaomoji inevitably find themselves either installing a kaomoji text expansion app or repeatedly googling “shruggie” and copy-pasting from the top search result. Gifs, on the other hand, are infinitely complex, with real faces and animations and full lines of dialogue written on as captions. But they have the inverse problem: there’s so many of them that it’s hard to get the one you want, and they’re so large and distracting that they don’t integrate seamlessly with a line of text (they take up their own line even when you’re in an app that has a built-in gif search). Gifs are fun to use occasionally, but they’re impractical to incorporate in every other sentence. Emoji strike a happy medium between the two: your most-used emoji show up in their own, easy-to-access section of your emoji picker, but there’s also more there if you want to go exploring. They intermingle easily with everything else you’re typing, rather than demanding their own new line, and they’re easy to copy-paste and send from one app or device to another, at least as long as Unicode has the ones you want.
But it’s certainly true that emoji, emoticons, and gifs all exist in the same ecosystem, and even certain words seem to have a related function. Instagram engineers looked at the most popular emoji used on their app and created a list of words people also use in a similar context. They found that people use the face with tears of joy emoji in the kinds of sentences where they might otherwise have used lolol, lmao, lololol, lolz, lmfao, lol, ahahah, ahaha, lmaooo, or lolll; they use the heart emoji in the same contexts as words like xoxoxox, xoxo, xoxoxoxoxo, xxoo, oxox, babycakes, muahhhh, mwahh, babe, and loveyou; and they use the loudly crying emoji like they use ugh, ughhhhh, wahhhh, agh, omgg, omfg, and whyyy. When one format isn’t available, the others can work as substitutes: a study by linguists Jacob Eisenstein and Umashanthi Pavalanathan showed that people who use more emoji rely less on other expressive resources, like plain text emoticons :), repeated letters (yayyy), acronyms (lol), and other creative respellings (wanna).
There’s a deeper question about the appeal of digital embodiment, though, regardless of whether it surfaces as emoji, emoticons, gifs, or another form. The facial expressions are by far the most popular, and yet there’s an important way in which they’re not like our ordinary kinds of facial expressions. When we’re interacting with other people, we find the most trustworthy kind of facial expression to be the kind that’s given off involuntarily: the burst of laughter or sob in the throat that’s difficult to fake. And yet you can’t involuntarily give off an emoji. They’re all given out deliberately—you choose exactly which one to send, and you know that everyone else does, too. Emoji and all of their relatives are fake by definition. If we try to say that they map directly onto our emotional facial expressions, then we have a weird mismatch. How is it that we’re so keen on such disingenuous symbols? What’s to enjoy about a world where everyone is wearing a mask?
A paper by linguists Eli Dresner and Susan Herring has a compelling answer. Rather than think about emoticons as emotional, they argue, we should think about them as deliberate cues to the intention of what we’re saying. Sometimes that intention does align with an emotion: if you say “I got the job :)” you’re indicating that you’re happy about it. But sometimes you put on a facial expression aspirationally, the way you might put on a polite social smile during a customer service interaction, even if you’re having a terrible day, just to make things proceed smoother. A smiley face might be used in a context like “I’m looking for some suggestions :)”—you might be anxious rather than happy about requesting feedback, but you’re using the smiley to make the request more polite. Moreover, people sometimes use smiley faces in contexts that aren’t happy at
all. Dresner and Herring quote a person saying “I feel sick and tired all the time :)”—the speaker isn’t happy or even smiling about feeling sick and tired, but might include the smiley to indicate that they don’t want their words to be read as a complaint. The same statement with :(, on the other hand, could be intended as a request for sympathy.
The basic smile emoticon :) or emoji is a versatile tool for this kind of contextualization. It can soften other kinds of harsh statements: making a demand into a softer request, or a seeming insult into softer teasing. As psychologist Monica Ann Riordan points out, saying an insult plus a smiley doesn’t mean smiling while insulting someone, or being happy about how terrible someone is: the smiley changes the intention behind the whole insult into a joke. A smiley can even indicate outright rejection, in a polite sort of way. Journalist Mary H.K. Choi did a series of interviews with a diverse cross-section of American teenagers about how they use technology and emoji for a 2016 article in Wired. One teen explained that he would exchange various heart emoji while flirting, but the worst emoji for a girl to send back was the smiley face—“Yeah, that’s the ‘Thank you, but I’m not interested.’”
Dresner and Herring point out that spoken language already has a well-established distinction, dating back to a British philosopher of language named J. L. Austin in the 1950s, between the actual words you’re saying and the effect that you mean to have on the world in saying them. (Whether you actually succeed, says Austin, is a slightly different philosophical category.) If I say, “There’s a car coming,” I may be intending a warning (Step back!), an insult (road rage), a promise (I have booked it for ten o’clock tomorrow morning), or a complaint (I thought we were alone on this desert island!). Or if I say, “That’s a nice shirt,” I could be complimenting it, hinting that I want to borrow it, or even criticizing it by calling attention to it at all (But why are you wearing it in this sauna?).
We have a lot of tools at our disposal for conveying our intended effect: we can add explicit, clarifying words like “Watch out!” or “I promise . . . ,” we can add strategic pauses and vocal inflections, we can rely on our shared knowledge of context, and we can gesture. It turns out that gesture linguist Adam Kendon has also invoked Austin’s idea of clarifying the intentions behind an utterance as a way of explaining what emblems do in communication. Think about saying “Good job!” along with a nameable, emblematic gesture: with a thumbs up, it’s a congratulations; with a wink, it’s a sly prod; with a facepalm, it’s a sarcastic acknowledgment of failure; with the middle finger, it’s an insult.
We’ve circled back to another reason why it makes sense to think of emoticons and emoji as gestural rather than emotional: thinking this way resolves the apparent contradiction between emotional facial expressions and the emoticons that supposedly represent them. Sure, it’s constructed, but a thumbs up is constructed, too, and both can still be genuine. If we say instead that people are consciously using them to guide their readers to the correct interpretation of their words, then emoticons become a positive, helpful, social behavior, a way of saying, “I want to clarify my true intentions for you.” It’s not the more negative behavior of putting on a mask. It’s true that a smiley face doesn’t always mean that the speaker is happy (an uncontrollable, genuine smile), but it does align with a deliberate, social smile, or the exclamation mark that proves you’re not a Stone-Hearted Ice Witch. All three can indicate that I’m asking politely, I don’t want to impose, I’m actually joking, I’m letting you down gently, or a passive-aggressive, “Oh no, of course I’m not mad.” It’s not so much that every emoji has a direct analogue in gesture; it’s that we can use them both to accomplish similar communicative goals.
Bodies don’t just communicate gesture: they also exist in space and time, and emoji can help us get across similar meanings in virtual space. Sometimes, you don’t actually have anything informative to say to the other person, and all you’re looking to communicate is subtext: “I see this,” “I’m listening,” or “I am still here and I still want to be talking with you.” In physical space, we often convey this through the body: you know when other people are near you and you can tell whether they’re paying attention to you or whether you’re looking at the same thing. Even if neither of you is saying anything, you can make eye contact, touch, or even just look over and see that the other person is still there. (Unless someone is being very, very sneaky.) In virtual space, sneaky happens by default. You’re only felt to be present when you’re saying something (save for a few limited exceptions, like videochat and avatars in Second Life or social games).
A simple way to let someone know you’ve seen their post is by liking it. This can be used for acknowledging big life events, like a wedding or a baby. Liking can be the precursor to something more: if you like a few of someone’s posts and they like yours back, you might take that as a sign they’re open to further conversation. It can also be a way of trailing off: by liking the final post in a thread, you indicate that you’ve seen the other person’s message and decided there was nothing left to say. Liking can also backfire: the “deep like” refers to a possibly accidental like on someone’s post from a long time ago, which implies that you were creepily looking back through their profile.
Emoji and gifs offer a way to indicate more active listening responses: not just “I’ve seen this,” but “I hear you and understand you.” In speech, we often indicate understanding by repeating the important part or mirroring each other’s gestures. If I say, “Sorry I’m late, I had a flat tire,” and you reply, “A flat tire!” you’re not being superfluous, you’re indicating understanding. Similarly, therapists and active listening coaches often recommend making people feel heard by restating their emotions to them. Thus, I could say, “Ugh, I got a flat tire on the way here,” and you might say, “Ooh, that’s frustrating.” Emoji can accomplish both kinds of reaction: if you say, “I want to go to the beach this weekend,” I can acknowledge the topic you’ve introduced by replying with fish and shell and crab emoji . Or if you say, “I miss you ,” I can share your sadness by echoing the same emoji or go one step further by finding a sad gif. Human-computer interaction researchers Ryan Kelly and Leon Watts interviewed a cross-section of young adults, primarily from the UK, about how they used emoji. One of their participants clearly illustrated the use of emoji to acknowledge a topic and close a conversation: “Yesterday we were talking about pancake day, so I just sent some pancakes [an emoji] and that kind of just, finished the conversation. It kind of just, yeah I think it says you have nothing else to say.”
Beyond single responses, sending messages back and forth can be a way of digitally hanging out: even when your messages have barely any textual meaning, they convey an important subtext: “I want to be talking with you.” The sending itself is the message, whether it’s emoji or stickers or selfies or gifs. This practice is especially common among teenagers, who often want to hang out with friends for hours on end in ways that seem trivial to adults around them. As a participant in the Kelly and Watts study put it: “You just start playing around with the emojis . . . like send a picture of a moon with a face on it, and then they would send me back like a cow, and I would send them back a turtle, and it doesn’t mean anything, but it’s just sort of funny. . . . Or like a little game, where you have to like guess what they’re trying to say with all the pictures.” One way that social tools catch on is when they facilitate conversation for the sake of conversation, allowing social interaction with less pressure, such as by encouraging us to send selfies or photos of our surroundings. We can’t be witty conversationalists all of the time, and embodied communication tools like emoji mean that we don’t have to be.
Sometimes even the fun of sending emoji or selfies back and forth pales. Our bodies—and the worlds they inhabit—are themselves colorful, animated, and interesting to look at. Words on a page, less so. After all, in the physical world, we don’t often sit around in undecorated, windowless rooms and
simply talk at each other. We do things together—we prepare and eat food together, we watch a show and talk about it later, we go for a walk or a car ride, we trade compliments, we point out the cute antics the dog or cat is getting up to, and so on. In digital conversations, we also bring in external objects as excuses to start up a conversation and ways to keep it flowing: a gif of a tiny turtle eating a strawberry, a sticker that makes a pop culture reference, a video that reminds us of someone’s interests, a link that supports an argument we’re trying to make, a camera filter that appears to give us cute animal ears. Studies find that looking at cute cat videos improves mood and that people have similar reactions to cute puppy photos as they do to cute baby photos, so gifs become a kind of emotional currency, a way of sending someone a tiny zap of positive feeling. A more involved way of digitally hanging out is by playing online games with your friends, whether that’s immersive-style games like Fortnite, League of Legends, and World of Warcraft, or casual games like Pokémon Go and Words with Friends.
* * *
—
Embodiment and projecting a virtual body may sound dangerously space-age—holograms!—but in many ways, embodiment is very old. Older than writing, as old as stories, perhaps as old as language itself. What does a storyteller do other than use their voice and body to project characters and feelings into the minds of their listeners? What is language other than a tool for transmitting new mental representations of the world into the minds of other people? Many theories about why language was evolutionarily useful involve things like collaboration and gossip—being able to plan together to hunt a mammoth, remember where the good berries are, or who can be trusted.
Like the failed proposals for sarcasm punctuation that we looked at in the previous chapter, generations of people have tried to reform English spelling, but eeven wen speling reeform iz perfektlee lejibl, sumhow it nevr katshiz on. The most we get is fragmentation, for example when some parts of the English-speaking world switched to -or or -ize while others stuck with -our and -ise, but multiple competing systems is not really an improvement. Same with other attempts at language reform: the international auxiliary language Esperanto is counted a success among constructed languages because perhaps two million people have learned it to varying degrees, while other, arguably better-designed, languages have languished in still greater obscurity. More than two million people use emoji every single hour.
Because Internet Page 19