What Are Emoji and Where Did They Come From
The word "emoji" sounds English, and nearly everyone assumes it's a relative of "emotion". The coincidence has fooled a whole generation: it's actually pure Japanese, where 絵 (e) means "picture" and 文字 (moji) means "character". A picture‑character. And the etymology works as a pointer to the truth about their origin: emoji were invented in Japan, by one specific person in one specific year. They never grew gradually out of emoticons, though people assume exactly that.
Since we've started debunking, let's fix the main distinction right away, otherwise the whole history that follows reads wrong.
Emoticons are combinations of ordinary keyboard characters: :), ;-), <3. They existed long before emoji, since the 1980s. Emoji are built in a fundamentally different way: each one is a separate graphic character with its own code. :) takes two keystrokes, 😊 is one Unicode character with its own number U+1F60A. Everything ahead is the history of these codes.
One person and 176 pictures
Since emoji were invented, they have an author. It was Shigetaka Kurita: in 1999 a Japanese designer working on the i‑mode mobile internet service for carrier NTT DoCoMo. His task was utterly concrete. The screens of the first cell phones were tiny and monochrome, and conveying emotion or context in a short message with text alone was nearly impossible: characters were scarce and the tone got lost.
Kurita found the answer in pictures. He drew 176 icons at 12×12 pixels: a sun, an umbrella, a heart, a smiling face, zodiac signs, transport. His inspiration came from what was at hand in Japanese visual culture: simplified pictograms from manga, road signs, weather symbols from newspapers. The key find was technical. Each icon took the space of exactly one character, so it could be inserted right into text, like a letter. A sign in the line, with nothing to send as an attachment.
That sealed the format's fate. But so far — only in Japan.
Why Japanese pictures nearly stayed Japanese
For the next years emoji were a local phenomenon, and there was a reason they might have stayed that way. Each Japanese carrier (DoCoMo, au, SoftBank) had its own incompatible set of pictures. A message with emoji sent from one network to another turned into a mush of question marks: in place of a heart the recipient saw "we don't know what this symbol is". As long as everyone drew their own alphabet, no common language emerged.
For the pictures to work for everyone, the carriers would have had to agree on a single code, one any phone in the world would read the same. And there's only one global standard for text characters.
How emoji became part of the global standard
That standard is Unicode, the system that assigns a unique number to every text sign on the planet, from Latin and Cyrillic to hieroglyphs and math symbols. Adding emoji to it meant placing them alongside letters. The turning point took three years:
- 2007. Google
began adding support for Japanese emoji in Gmail
for the Japanese market and saw how popular it was.
- 2009–2010. Google
and Apple
jointly proposed to the Unicode Consortium that emoji be included in the standard on par with alphabets.
- 2010. Unicode 6.0 officially added the first 722 emoji. From that moment the symbols stopped being "a Japanese carrier feature" and became part of the internet's text infrastructure, like Cyrillic.
Then simple mechanics kicked in: since the symbols were now standard, they could be built into any device's keyboard. In 2011 Apple opened the emoji keyboard to all iOS users (previously a hidden option known mostly in Japan), and Google
added support to Android a year later. The symbols got a keyboard slot for hundreds of millions of people, and a growth began that hasn't stopped since. In 2015 the "face with tears of joy" 😂 became Oxford Dictionary's word of the year; skin tones were added the same year, then professional and gender diversity, and since then a new batch is approved every year.
Which naturally raises the question: who decides which emoji gets to exist?
Who decides what an emoji will be
New emoji aren't drawn by one person: they're proposed through a formal process at the Unicode Consortium, and anyone can submit — an ordinary user, a company, a designer. But wanting isn't enough. A proposal must prove the symbol is genuinely needed, via search‑query statistics, the frequency of similar words, or a clear gap in the set. That's how the wok, the "under construction" sign and dozens of national dishes entered the standard: their authors managed to justify the gap.
And here hides a subtlety that explains one of the most common gripes about emoji. After approval, Unicode fixes the code and short description of the symbol, but not its appearance. How exactly to draw the approved "smiling face" each platform decides itself. So the same code looks different on iPhone, Android and Windows, and sometimes the meaning shifts with it. That's a big topic of its own, covered in «Why emoji display differently».
Where the empty boxes come from
Separating the code from the picture also explains a second everyday puzzle: why an empty rectangle sometimes appears in place of an emoji. That rectangle has a name — tofu, for its resemblance to a piece of tofu.
The mechanics are direct. The device received the symbol's code, found it in the text and tried to draw it, but the installed font held no picture for that code. This happens with new emoji approved in last year's standard update: your phone already knows the symbol while your correspondent's phone, running two‑year‑old firmware, doesn't yet. The symbol itself isn't lost; it sits honestly in the message and will render as soon as the font updates.
Hence a practical takeaway for anyone putting emoji into newsletters and interfaces: fresh symbols give uneven results across a mixed audience, while a set from five years ago renders almost everywhere. How to account for that in copy is covered in emoji in marketing. You can check a specific symbol on Emojipedia: every emoji there lists the year it entered the standard and shows every vendor's artwork. A symbol older than five years renders almost everywhere, while a sign from the latest update reaches only recent devices.
How several codes assemble into one symbol
There's also the reverse case, where the picture is one and the codes behind it are several. Flags, families and professions with skin tones are technically assembled from base symbols glued together by the invisible ZWJ connector (zero‑width joiner).
It works like this: "woman" plus ZWJ plus "rocket" produces a woman astronaut. If the device understands the sequence, it draws one solid sign. If it doesn't, the glue comes apart and your correspondent sees two separate emoji in a row — a woman and a rocket. That's exactly why composite symbols break more often than simple ones, and flags behave especially capriciously: each is built from a pair of regional indicators, and a device without the right picture shows two letters instead of a flag.
Knowing this saves debugging time: if one sign turned into two, the issue is ZWJ support on the recipient's side, and rewriting the text won't help.
Can you make your own emoji
Since a consortium approves the symbols, a fair question follows: can you add your own sign, a company logo for instance. Into the standard — no. Unicode deliberately refuses to register brands and commercial symbols, because the standard describes humanity's writing systems, and a logo has no place there. A proposal carrying a brand mark gets rejected at the first stage.
A workaround exists, though, and it's more convenient anyway. Platforms allow their own sets on top of the standard: custom emoji in Slack and Telegram, sticker packs, reactions in corporate chats. Technically these are ordinary images bound to a text code like :logo:, and they work only inside their own platform. In exchange you can draw one in an hour and update it whenever you like, whereas the Unicode proposal route takes years and ends with a symbol that isn't yours.
The practical takeaway for brands: a company sign lives in stickers and custom emoji, while the standard set is used as it is, with no attempts to extend it.
How many emoji there are now
Since the set grows yearly, the count long ago passed the thousands. Unicode 15 holds more than 3700 emoji, composite symbols included. Each year adds another 30–100, from household objects to cultural symbols that were missing.
One practical takeaway follows from the whole history: emoji work as a symbol system with fixed meaning. This helps you pick them more precisely for marketing (a symbol goes through years of selection and carries a defined, approved meaning), tell emoji apart from emoticons in design writing, and understand why the same sign looks different to different recipients.
Where to download emoji PNGs
If after all the theory you need the symbol as a file, the Trace Logo's emoji catalog offers Apple and Google
versions for each. On an emoji page you can copy the symbol in one click, download the PNG of the vendor you want, and see the Unicode code and category. Search works in both Russian and English, so you'll find an emoji by name or just by meaning.
What to remember
Emoji were invented by one Japanese designer in 1999 for the tiny screens of the first cell phones, and from the very start they were signs in a line of text, equal in status to letters. They didn't stay local only because in 2010 they entered Unicode and got a single code any device understands; then Apple and Google
put them on the keyboard, and the growth never stopped. From that same split between code and picture grow all their everyday oddities: a different look on different platforms, tofu boxes in place of new symbols, and composite signs falling apart.
Ready PNGs from every vendor sit in the emoji catalog.