Who Invented the Emoticon and Emoji? (And How They’re Forever Changing Written Language)

Historically, written language has always been rather limited in its ability to adequately represent the nuance of in-person communication, which comes with a whole slew of gesticulations, facial expressions, subtle shifts in tone and the like. As such, humans have been attempting to find ways around this problem seemingly as long as we’ve been writing stuff down.

Going back to the earliest alphabetic writings, their situation was unsurprisingly worst of all, with a complete lack of lower case letters, punctuation, and even spaces between words.  As you might imagine, for example in Ancient Greek, it was a rare feat for an individual to understand a given text they were reading the first time through, or even after several readings. By the 3rd century BC, punctuation in the West started to become a thing, though its spread was not without its detractors. For example, Roman orator Cicero (106-43 BC) scorned such punctuation, noting things like when and how long to pause for breath when reciting some bit of written text “ought to be determined not… by a stroke interposed by a copyist, but by the constraint of the rhythm.” In short, he felt adding punctuation was an inferior form of writing, and the person speaking that text should be skilled enough to figure it out without, based on the words alone.

Bizarrely, it would be another thousand years after rudimentary punctuation was introduced before spaces between words would become common, around the 7th and 8th centuries.

It would be another 800 years or so on top of that, thanks to the printing press, that humans thought it would be a good idea to once again improve on our ability to convey meaning in writing better, this time in more defined and consistent usage of punctuation and syntax. A major push towards this end came thanks to one Aldo Manutius the Younger (1547-1597) publishing a book on the subject in the early 1560s called Orthographiae ratio (A System of Orthography). In the book, Aldo built on the ancient Greeks’ minimal punctuation marks and designated that the comma would separate phrases and clauses, the colon would be used for lists, and the “full point” (read: period) would denote the end of the sentence. Aldo also explained the uses of other basic punctuation including the question mark, apostrophe, exclamation point and quotation marks.

On top of that, he explicitly noted that the purpose of these marks was more than merely as a rhetorical aid, but that such marks were necessary in order to express and preserve meaning.

Ben Jonson’s English Grammar (1640) further solidified this idea, illustrating how punctuation could help preserve an author’s original intention, rather than just giving a guide to how to read a text out-loud. Well received by most, by the time of the Restoration (1660), using punctuation for syntactical purposes was finally common, though much like with Cicero before, not without some detractors. A universal truth throughout the ages is that people hate change, no matter if clearly superior. And when talking about anything academic, generally think things are wrong or worse when not the way they were taught growing up, regardless as to whether it’s even possible in a given thing for it to be inherently right or wrong.

So what does any of this have to do with the emoticon and emoji? As alluded to, throughout this progression of written language, even through modern times, many of even the most distinguished writers have lamented that the basic written language tools available are insufficient to make the meaning of certain statements and words in the context of a written work inherently clear.

Enter emojis and emoticons, examples of which have been around for a shockingly long time, despite only recently widely used in communication.

In fact, it turns out the earliest known smiley on something goes all the way back about 4,000 years, found on a Hittite pot used for storing sherbet in. Apparently drinking said sherbet made the owner of the pot happy and he or she felt compelled to express that on the pot itself.

From there, we need to fast-forward to 1635 for the next prominent smiley meant to communicate something, penned by notary Ján Ladislaides of the town of Trencín in Slovakia. He drew this noteworthy little smiley next to his signature when he signed a document concerning the accounts of the Chamberlain of the Town Hall. Presumably this was to enhance his signature’s text to indicate he was not just signing off on the document’s contents, but happy with them.

The next known instance of a smiley in text occurred in 1648 in a poem by Robert Herrick, To Fortune, which has the lines:

Tumble me down, and I will sit

Upon my ruins, (smiling yet:)

If it was intentional, and not just a typo, this would make this the first known instance of a sideways smiley in text.

Moving on from there, on August 7, 1862, a New York Times writer also included the following when penning a transcript of a speech by Abraham Lincoln: “I believe there is no precedent for my appearing before you on this occasion, [applause] but it is also true that there is no precedent for your being here yourselves, (applause and laughter ;)…” This is yet another case of it not being clear if this was meant to indicate a smiley or merely a printing typo.

Whatever the case, becoming more explicit, on March 30, 1881, Puck magazine editors went so far as to create a set of typographical emoticons to help out in making certain writing more clear, with their set representing Joy, Melancholy, Indifference, and Astonishment.

Moving on to the early 20th century and in 1912, famed writer Ambrose Bierce proposed his own little smiley to indicate when one is joking, stating,

It is written thus  and represents, as nearly as may be, a smiling mouth. It is to be appended, with the full stop, to every jocular or ironical sentence; or, without the stop, to every jocular or ironical clause of a sentence otherwise serious- thus: “Mr. Edward Bok is the noblest work of God .”

Independently building upon this idea, in 1936, Alan Gregg of Harvard even attempted to create a whole set of emoticons, using (-) for the smile, (–) for laughter, (*) for a wink, and (#) for a frown.

Unfortunately in all the preceding cases and countless similar attempts to improve the tools of of writing to better capture what we humans can convey in person so easily with our body language and tone, Gregg’s ideas did not stick.

One reason for this was a lack of perceived need by the masses, as humans historically used writing very different than conversational speech. As illustrated by linguist John McWhorter of Columbia University,

When you write, because it’s a conscious process, because you can look backwards, you can do things with language that are much less likely if you’re just talking. For example, imagine a passage from Edward Gibbon’s “The Decline and Fall of the Roman Empire:” “The whole engagement lasted above twelve hours, till the graduate retreat of the Persians was changed into a disorderly flight, of which the shameful example was given by the principal leaders and the Surenas himself.” That’s beautiful, but let’s face it, nobody talks that way…

Casual speech is something quite different. Linguists have actually shown that when we’re speaking casually in an unmonitored way, we tend to speak in word packets of maybe seven to 10 words… That’s what speech is like. Speech is much looser. It’s much more telegraphic. It’s much less reflective — very different from writing. So we naturally tend to think, because we see language written so often, that that’s what language is, but actually what language is, is speech. They are two things.

Thus, up to this point while people were writing books, speeches, and letters and the like, these are all generally a much more thoughtful way of communicating, akin to highly structured monologuing at someone for some amount of time. Even in the case of many letters, this often expects a lengthy, thoughtful monologue both to someone, and in return. It’s not usually a conversation as we would do in person, particularly throughout much of history where the relative difficulty and time it took to send a letter to someone any distance away did not exactly encourage people to restrict themselves to a quick message.

Thus, while the telegraph would see advancements in the art of mimicking human conversation more accurately, as we’ll get to in the Bonus Fact in a bit, the stronger pressure to try to capture conversational human speech in text form wouldn’t really come until the digital age, when that’s exactly what we humans started trying to do. The problem, of course, is that we found the restrictions of traditional writing and the tools available insufficient for the task.

This all inspired the users of, for example, the PLATO IV system in 1972 to actually use a facility of that system to solve the problem, creating a whole slew of the first emojis and emoticons in the process.

And if you’re wondering about the distinction here, “emoji” derives from the Japanese for “picture” and “character”, so “picture character”. In contrast, “emoticon” derives from the English “emotion icon”. Thus, while you might think given the two words’ similarity and what they represent also being similar that one came from the other, this is actually purely coincidental.

In any event, going back to the PLATO IV system, with this system, users could press SHIFT-space and then a character to have that character plotted over the previous character without overwriting it. Particularly clever users used this fact to come up with all sorts of little images to represent various emotions and otherwise add context and meaning to a given bit of text, or sometimes to just have the thing stand alone to communicate something, like some sort of modern hieroglyphic. Eventually there were many hundreds of such symbols being used on this system.

However, despite how widespread the use of these emojis and emoticons were in this system, almost no one credits the users of this system as having “invented” either.

Similarly, one Kevin MacKenzie suggested in an April 12, 1979 digital message that “-)” should be used to represent when a statement was meant to be “tongue in cheek”. But MacKenzie, too, is generally not given credit for inventing the emoticon. In both of these cases, and many other such examples, this is simply because these instances were not the path to popularization.

So who is given the credit for inventing the emoticon and the emoji? As to the former, enter Computer Science professor Dr. Scott E Falman of Carnegie Mellon University.

Like others before him, his inspiration came from a chat message board, where people were struggling to mimic in-person human conversation in a digital, text space. Fahlman explains,

[The] Computer Science community at Carnegie Mellon was making heavy use of online bulletin boards or “bboards”. These were a precursor of today’s newsgroups, and they were an important social mechanism in the department – a place where faculty, staff, and students could discuss the weighty matters of the day on an equal footing. Many of the posts were serious… [or] of general interest, ranging from politics to abortion to campus parking to keyboard layout (in increasing order of passion). Even in those days, extended “flame wars” were common… Given the nature of the community, a good many of the posts were humorous (or attempted humor). The problem was that if someone made a sarcastic remark, a few readers would fail to get the joke, and each of them would post a lengthy diatribe in response… soon the original thread of the discussion was buried.

Most notable to the topic at hand, one such discussion in September of 1982 was started considering what would happen if the cables broke in the then perceived poorly maintained Wean Hall elevators. Among other musings was, for example, what would happen to the flame of a lit candle and a drop of mercury in the elevator as it plunged down, as well as what would happen if a pigeon was flying at the time inside said elevator. Things even progressed to what sound the pigeon would make if it previously inhaled helium as it fell- digital nerdery in all its exquisite glory.

While you might think this whole chat would be very apparently tongue-in-cheek, things changed when, on September 16, 1982, one Howard Gayle, referencing the candle and mercury discussion and what would happen after if they tested this in real life, wrote:

Because of a recent physics experiment, the leftmost elevator has been contaminated with mercury. There is also some slight fire damage. Decontamination should be complete by 08:00 Friday.

This ultimately got misinterpreted by some as being an actual experiment that happened, rather than just a continuation of the light hearted musings, as noted by one Neil Swartz the next day where he writes, “Apparently there has been some confusion about elevators and such. After talking to Rudy, I have discovered that there is no mercury spill in any of the Wean hall elevators. Many people seem to have taken the notice about the physics department seriously…”

Naturally, the conversation at this point shifted away from pigeons in plunging elevators to how to solve the problem of very clearly expressing that one is joking in written text. One suggestion made by Anthony Stentz was “How about using * for good jokes and % for bad jokes? We could even use *% for jokes that are so bad, they’re funny.”

To which Keith Wright replied,

No, no, no! Surely everyone will agree that “&” is the funniest character on the keyboard. It looks funny (like a jolly fat man in convulsions of laughter). It sounds funny (say it loud and fast three times). I just know if I could get my nose into the vacuum of the CRT it would even smell funny!

Leonard Hamey then appears to be the first in this discussion to think of trying to match symbols on the keyboard with a human face, stating, “I think that the joke character should be the sequence {#} because it looks like two lips with teeth showing between them. This is the expected result if someone actually laughs their head off. An obvious abbreviation of this sequence would be the hash character itself (which can also be read as the sharp character and suggests a quality which may be lacking in those too obtuse to appreciate the joke.)”

Building on this idea, but without the subtle irony, on September 19th at 11:44am Fahlman quickly wrote a comment that would finally, after literally thousands of years of certain humans lamenting this problem, spur us to come up with a near universally accepted solution. Unfortunately for Fahlman, one of the world’s leading minds in Artificial Intelligence, he did not know history would forever remember him for this comment, and has since lamented he didn’t even bother to proofread it, let alone attempt to sound remotely eloquent. His comment was:

I propose that the following character sequence for joke markers:

Read it sideways. Actually, it is probably more economical to mark things that are NOT jokes, given current trends. For this, use:

People on the Carnegie bboards quickly picked this sideways smiley and frowny face up and within 24 hours had also expanded on it, adding the |-: to indicate an alternate for a non-joke.

From there one of the early examples of the sideways smiley being used without explanation can be found in a discussion of aliens holding power tools… Written by one Guy Jacobson in a post titled “Holding a chainsaw?? :-)”, he asks:

Does anyone have a picture of R2D2 holding a seed auger in TeX format? Or how about a rendering of Yoda with a lathe for use with nroff? Any pointers to digitized images of short, cute aliens holding power tools would be greatly appreciated.

Spreading beyond universities, within a couple months, one James Morris in an message titled “Communication Breakthrough”, written to the staff at famed research lab Xerox Parc, stated,

Because you can’t see the person who is sending you electronic mail you are sometimes uncertain whether they are serious or joking. Recently, Scott Fahlman at CMU devised a scheme for annotating one’s messages to overcome this problem. If you turn your head sideways to look at the three characters :-), they look sort of like a smiling face. Thus, if someone sends you a message that says “Have you stopped beating your wife?:-)” you know they are joking. If they say “I need to talk to you :-(“, be prepared for trouble.

It’s at this point we feel it’s worth explicitly pointing out that- yes- the path to emoticons becoming popular did not come through so-called illiterate teenagers, but came from and were popularized by some of the world’s leading academic minds.

In any event, this notation likewise spread to the ARPANET, among other systems, with others along the way coming up with a number of other emoticons to further expand the range of ways to enhance written text.

Fast-forwarding to the 1990s and some were even going so far as to develop actual picture versions of emoticons, instead of just using symbols on the keyboard, more or less going back to what users of the PLATO IV had long before been doing, but using defined picture fonts this time. Examples of this can be seen in things like Microsoft’s Wingding font developed in 1990.

But wingdings and other similar things are not generally credited as launching the emoji- again, with these sorts of things, credit is generally given to the person who made the thing that ultimately resulted in the popularization, not who was actually first.

This could not be better illustrated by the fact that a set of 90 emojis were included with the release of the J-Phone on November 1, 1997. However, this set of emojis is not generally considered to be where the emoji originated either, primarily because the J-Phone at this point was outside of the price range most could afford and, thus, did not bring emojis to the masses.

For that, we turn to a worker at the telecommunications company NTT DoCoMo, Shigetaka Kurita. A year after the J-Phone emoji set, in 1998, Kurita was tasked with creating a set of pictures to sum up various emotions and ideas for NTT DoCoMo’s i-mode mobile internet system. Beyond aiding in conversational clarity, in this system, when sending messages, you only had a maximum of 250 characters to work with, making emojis even more handy for concisely expressing ideas.

And so it was that in a mere 5 weeks, Kurita came up with 176 12×12 pixel emojis. As for his inspirations, he stated, “Both emoji and kanji (Chinese characters used in the Japanese writing system) are ideograms, but I did not find inspiration for designing emoji in the kanji… In creating emoji, I found inspiration in pictograms, manga, and all sorts of other sources.”

When the emojis were released with the system in 1999, they proved to be extremely popular and were quickly copied by other companies around Japan and then beyond.

Since then, both emojis and emoticons have been providing an essential and ever expanding tool to enhance our written language communication, which today, at least when talking distance communication, has come to dominate the way we talk with one another. For example, texting alone accounts for approximately five times as many messages between people compared to phone calls, and when adding in email and social media posts, which also frequently use emoticons and emojis, the numbers skew even further.

Similar to Cicero lamenting early examples of punctuation, this marked shift in the way we communicate textually has been much to the chagrin of many, sometimes condemned, along with so called “chat speak”, as a sign of the degradation of written language.

For example, among other arguments we read against the use of emojis, was that the likes of the supposed pinnacle of masters of written language- Shakespeare- didn’t need emojis to convey subtle meaning… To which we feel compelled to point out that there are literally master’s level college classes offered at various universities to help people understand what the hell Shakespeare was talking about half the time… And even in his own era where more of it would have been inherently understood, Shakespeare was getting to rely on the actors on the stage making the more subtle meaning clearer by their presentation. And don’t even get us started on Shakespeare’s occasional lack of consistent spelling and punctuation (those two things not really being standardized at the time). On a related note, for those still thinking Shakespeare was representative of all things refined and upper class, we feel further compelled to point them to our article on How Far Back a Modern English Speaker Could Go and Still Understand People Speaking English, where we outline why it’s likely Shakespeare and his actors sounded, not like posh British English speakers today, but like Hollywood pirates… Not just in accent, but the bawdy nature of a lot of what they were saying too.

In the end, as Harvard linguist Steven Pinker sums up of emojis and emoticons, “as with many of our punctuation symbols, like a question mark or an exclamation point, [emojis] are there to convey some communicative force that would not be obvious just from the arrangement of words on the page…What the smiley in particular does, mainly conveying irony or levity, is often crucially important in getting a message across, because irony is often undetected… A gifted wordsmith can make it clear that something is intended in jest, however, if it is not hitting the reader over the head with unsubtle irony, if there is any attempt at wit, indirectness…then the readers … might find it going over their heads. In the cases where there’s that danger, the smiley face makes it clear — I’m telling a joke.”

When expanding to talk about the combination of emojis, emoticons, and general chatspeak, the aforementioned Columbia University linguist John McWhorter, rather than lamenting it, goes even further, expressing how revolutionary this is to written communication. He states,

What texting is, despite the fact that it involves the brute mechanics of something that we call writing, is fingered speech. That’s what texting is- now we can write the way we talk…. [It’s] easy to think that… it represents some sort of decline. We see this general bagginess of the structure, the lack of concern with rules and the way that we’re used to learning on the blackboard, and so we think that something has gone wrong… But the fact of the matter is that what is going on is a kind of emergent complexity…in this new kind of language, there is new structure coming up…

We’re seeing…a whole new way of writing that young people are developing, which they’re using alongside their ordinary writing skills, and that means that they’re able to do two things. Increasing evidence is that being bilingual is cognitively beneficial. That’s also true of being bidialectal. That’s certainly true of being bidialectal in terms of your writing. And so texting actually is evidence of a balancing act that young people are using today, not consciously, of course, but it’s an expansion of their linguistic repertoire. It’s very simple. If somebody from 1973 looked at what was on a dormitory message board in 1993, the slang would have changed a little bit since the era of “Love Story,” but they would understand what was on that message board. Take that person from 1993…and they read a very typical text written by a 20-year-old today. Often they would have no idea what half of it meant because a whole new language has developed among our young people doing something as mundane as what it looks like to us when they’re batting around on their little devices… [If] I could go into the future, if I could go into 2033… [I’d ask] please show me a sheaf of texts written by 16-year-old girls, because I would want to know where this language had developed since our times, and ideally I would then send them back to you and me now so we could examine this linguistic miracle happening right under our noses.

If you liked this article, you might also enjoy our new popular podcast, The BrainFood Show (iTunes, Spotify, Google Play Music, Feed), as well as:

Bonus Fact:

  • Speaking of chatspeak, it turns out it’s been around a lot longer than cell phones and computers. Exhibit A: telegraph operators and Morse Code. Given this form of communication did lend itself to quick messages, not the least of which because long messages were more expensive, such operators had a number of what we might describe as primitive emoticons in beep form, and firmly embracing the concept of what would in more modern times become “chat speak”. For example, rather than sending the full word “Yesterday”, users of telegraph operator Walter P. Phillips’ eventually popular “Phillip’s code”, created in 1879, would simply send the code for the letters “YA”. Or instead of “Tomorrow”, they’d simply send “Tw”. Interestingly enough, there was even a shorthand for something that sounds like it would come from a modern gaming chat, “Instantly killed”- “Ik”… As for the Morse code equivalent of emoticons, we have such things as the number 73 (and later switched to 88) which represented “love and kisses”. Also of note is some of Phillip’s code is still commonly used today, such as “POTUS” and “SCOTUS”, “President of the United States” and “Supreme Court of the United States” respectively.
Expand for References
Share the Knowledge! FacebooktwitterredditpinteresttumblrmailFacebooktwitterredditpinteresttumblrmail
Print Friendly, PDF & Email
Enjoy this article? Join over 50,000 Subscribers getting our FREE Daily Knowledge and Weekly Wrap newsletters:

Subscribe Me To:  | 


  • Greatly researched article, thank you for the additional literary insight!

  • “… with the release of the J-Phone on November 1, 1997.”

    What?? Please provide a source for this fantasy, as the J-Phone didn’t arrive until the year 2000.

    • Daven Hiskey

      You are thinking of the J-SH04. There was a predecessor nobody remembers, the J-SH03 which was made in 1997. But more to the point, 1997 was when Softbank (J-Phone) launched SkyWalker SMS that had the emojis as stated. You can see them here.

      • I needed to compliment the author, the writing style, phrasing, verbiage etc. I couldn’t just breeze by this article, almost felt physically impossible. And the research, so perfectly woven with your argument for emoticons. I, admittedly shockingly though, happily changed my “educated” opinion of emoticons, solely due to this piece. Bravo Mr. Hiskey, I hope you love what you do, because you do it brilliantly.

        Congratulations Sir,

        N. Robles

        • Daven Hiskey

          Thanks! That’s always really nice to hear. 🙂 I can tell you working online, it’s usually the opposite. 😉 So very much appreciated. 🙂

  • Some slight corrections. When I did the original post, I was posing a hypothetical physics problem: “What happens to mercury and a candle in free-fall?”. Later, one of the other responders made a joke about a mercury spill in an elevator in Wean Hall on campus. (There was nothing wrong with the elevators) I responded that it was a joke and we needed a way to mark jokes. I proposed the “*” and after several alternatives, Scott made his proposal.
    One of these years (maybe 2022?) I will get back to Pittsburgh for the Smiley Cookie day.