Month: October 2018

I talked about this a while back, but now it’s for real. Today, I introduce to you a new conlang: 🖼🗣. Or, to put the name in something pronounceable, Pictalk. Yes, the glyphs making up the name are emoji. Yes, so are all the characters used in the entire language.

Strictly speaking, Pictalk isn’t a full-fledged conlang. It’s written-only, first of all. There is no true spoken form. Instead, it should be considered something closer to a conscript, an artificial writing system, modeled after hieroglyphic and ideographic scripts. But that’s enough to encode ideas, thoughts, sayings, and anything that might need to be written in this modern, digital age.

Glyph inventory

The hardest part about making Pictalk is the very restricted set of available glyphs. True, there are over 1200 emoji characters available, and they cover a wide variety of concepts, from animals to emotions to transportation and more. But I don’t have control over which symbols the Unicode Consortium adds to the list. While that list will grow (they add more each year, it seems), there’s little rhyme or reason to which new characters come in.

But that’s okay. We can do this. English only needs 26 letters, right?

Even with the wide array we have, it’s safe to discard quite a few right off the bat. First, I’ll drop the “cat face” group, such as 😸, because they really only repeat the normal human smileys. Next, toss out the handful of CJK ideographs in circles or squares, like 🈹—I’m an English speaker, and even Unicode gives up on giving them reasonable names. The skin tone modifiers (🏻 and friends) don’t make sense in the context of language; Pictalk thus won’t give them meanings, but will allow them to modify other symbols as a kind of synonym.

Likewise (and here’s where we start getting into the grammar bits), gendered forms like 👩‍🏫 or 👨‍⚕️ are synonymous with their “base” forms. With many languages, particularly in the West, where there is no neuter form, masculine is considered the default. Pictalk, however, is gender-neutral. That’s not out of some misguided idea of social justice or diversity, but simple expedience. Unicode has neuter forms for most of what we might call agentive glyphs. Where it doesn’t, we can use either, and that’s fine.

Last, flags. These take up a good chunk of the emoji list (about 15%, all told), and they’re mostly country flags. Well, for Pictalk, those flags represent their countries, and that’s that. Unlike most other characters, they don’t really participate in the construction processes we’ll see later on.

Non-emoji characters

Before we get to that, let’s go over the rest of Unicode. Obviously, since the whole point of Pictalk is to create a hieroglyphic script using the emoji characters, they’re the focus. But we’ve got a few other options available. One I won’t use is Latin letters. Or, for that matter, any other alphabetic script. In earlier versions of the language, I did utilize them for derivation and some small grammatical particles, but I’ve since removed the need for them. Only proper names use alphabetic characters; these are written as they would be in either the speaker’s or the audience’s preferred language.

Numbers, on the other hand, are perfectly usable. They’re already a little bit ideographic, after all, so it wouldn’t destroy the purity of Pictalk to include them. So 0-9 work exactly as they would in English: as the numerals zero through nine. And you can build on that as you do in English. (Pictalk is base-10, by the way.)

Punctuation works the same, as it’s very difficult to design a conlang that doesn’t need it. So sentences can end with a period, question mark, or exclamation point. Quote marks work for, well, quotes. Commas aren’t as necessary, but you can still use them to mark off clauses. Colons, besides having their normal English function, are used as attention-getters, in a sense, following the intended recipient of a statement or question. And we’ll see the other “special” characters as they come up.

Building words

Quite a few emoji work as words by themselves. Think of 🐕, 😄, or ✈, for instance. In Pictalk, that’s the most basic sort of word, and most symbols can function alone. Some are considered nouns, others adjectives or verbs, but there’s always a way to convert them.

Other symbols are “bound”, in that they can only occur fixed to others. An example here would be the (optional) plural marker ➿. By itself, it has no meaning. Suffixed to a root, whether a single symbol or a string of them, it gains meaning: 🐕➿ “dogs”.

More complex are the compound symbols that make up the bulk of the lexicon. In general, nominal compounds are head-final, as in 🐕🏠 “doghouse”, while verbal compounds are often head-initial, as with 📖🏫 “study”, from 📖 “read”. I’ve tried to refrain from being cute with meanings, striving instead for transparency, but some compounds remain idiosyncratic in meaning.

Last, a form of word-building that English doesn’t often employ comes into its own in Pictalk. Reduplication is productive for many basic words. For nouns, it can create a kind of collective sense: 🏠🏠 “neighborhood”. Verbs instead use reduplication as an intensifier: 💭💭 “to contemplate” (or possibly “to overthink”).

Moving on

All in all, I think this just might work. We can make words using only emoji characters. Next up, we’ll see how far we can go in making a language.

Today, most of the world uses alphabetic scripts, or something fairly close to them. With the major exception of Chinese (and the writing systems derived from it, such as those in Japan and Korea), alphabets, consonantal scripts, and the like reign supreme. They’re easier to learn, obviously, and far more suited to computers, so it’s only natural. Simple scripts, in the vast majority of cases, work just fine, so that’s what we use.

But it wasn’t always this way.

If you look back at the history of writing, you see that alphabets were not the original form of script. Indeed, assuming current theories are correct, writing developed first as pictorial representations of people, animals, etc. Abstractions came in later, as did the practice of using glyphs to represent spoken language, rather than as something closer to an aide mémoire.

The oldest evidence of writing we have all points in the same direction. Egyptian hieroglyphs, Sumerian cuneiform, and ancient Chinese symbols share the common feature of being, at least in some part, logographic scripts. The same may be true of other, mostly undeciphered writing, such as the Proto-Elamite script of that of the Indus Valley—given their age, it doesn’t seem out of the realm of possibility. While China kept its style of writing through the millennia, occasionally simplifying but never throwing away, the rest have mostly died out, replaced by Latin, Greek, Cyrillic, Arabic, the various scripts of India and Southeast Asia, and so on.

Enter madness

But wait. Anyone with a cellphone (which is to say, well, anybody) has at their disposal a vast and growing collection of bona fide ideograms: emoji. Can we use those as the basis for a modern-day hieroglyphic script?

I know what you’re thinking. “Michael, you’ve gone completely crazy!” you probably shouted at your computer screen.

You’d be right, but hear me out. I am being totally serious. Think about it. As of 2018, there are over 1000 emoji symbols in the Unicode standard, and they’re adding more with every update. Granted, most of the new ones are gender-specific versions of older ones, but you still see a genuine emoji every now and then. (“Lobster” was in the newest batch, I think.)

Most emoji fall into one of two categories. One is clearly nominal in nature: animals, vehicles, people, and so on. The other is the emotional set: grinning faces, smilies, and the like. Those can be considered adjectives, if you look at it the right way. Verbs, now, those are harder, but not impossible.

So here’s what I propose. Take the emoji, minus a few that aren’t really all that useful to English speakers (think the “cat faces”, or the numerous symbols containing Japanese writing), and construct a script. Or, if you will, a written-only conlang. Technically speaking, it would be something more akin to a pidgin. It would have no vocabulary of its own, and the grammar would necessarily be very stripped-down.

The limitations are severe, but operating under limiting conditions is the time-honored path of the hacker (in the original sense of the word). Here, we have no control over the inventory of symbols, no convenient way of even typing them, much less pronouncing them. And there’s no real payoff, either. If I did this, it would be for fun, not for glory.

Yet none of that ever stopped me before, so why should it now?

If you’re interested, stick around. I’ll post something more about this mad scheme in the coming weeks.