One day I had a sudden burst of artistic inspiration as I was eating some buttered toast. Hey, Jonathan, said the obtrusive muse, you know what you need? You need a language with a writing system that grows like a tree. “A script that grows like a tree?” I replied, crunching down, “What ever do you mean? How would that even work?” I dunno, you’re the linguist. Figure it out. I just gave you the idea. My muse was being a tease. “Oh goodness,” I sighed, for I was now properly distracted. “ I suppose that you could write it down from bottom to top...” I placed the toast down. “ with words written in columns, and sentences in rows.” I was now standing up, and searching for paper and a pen. Upon finding these, I scribbled down the following sketch.
Ok, I think to myself, that’s not too bad. We have here a basic geometry. A visual guideline for the shape of the letters. I imagine that this shape is a full word. I don’t know what it sounds like yet. But something like this is what I want the words to look like. Since I’m designing a script with an arboreal aesthetic I should break down the visual components of a tree branch.
Now, I know that “real” artists would probably rip me a new one for doing this but I decide to identify three parts to a tree branch: 1. The main body 2. The extending branches 3. The leaves and fruits. I represent the main body of the branch with a curved glyph. Just a simple curve. And then, for the extending branches, I can just add a little offshoot from the main curve. And finally for the leaves, I can add a little ornament off the extending branch.
I start at this point to feel a tingling sensation at the tips of my fingers, because I have established discrete visual elements that I can manipulate spatially to make different glyphs that look like they belong to the same script! After all, if what I’m doing is re-arranging the same three things, then there’s no danger of the script having strange out-of-place symbols in it. Finding the abstract basic shape for my glyphs permits me to be faithful to the first rule of designing an orthography
All elements must be spatially proportionate to all other elements
With the basic shape established, I began to sketch out some other arrangements of elements.
I have no idea so far what these glyphs sound like, but that doesn’t seem to matter much at this point, I’m just having fun tracing out arboreal glyphs. After sketching about a dozen or so of these, I think to myself that I really ought to decide on how they’re going to be pronounced. I mean, it does seem a tremendous waste to have all these symbols with no meaning behind them. And here is where I encounter my first problem.
My muse gave me no information on the sound of the language! I just heard “Tree Script,” and I was off. I start to regret going down this rabbit hole. Was I too hasty? Too ambitious? I’m gonna have to slow my roll. I’m gonna have to work out a sound inventory, and assign sounds to symbols. I’m going to need to decide what vowels I want, and what consonants I want, and how differences between similar sounds are encoded, and all that nonsense--and that promises to be slow, long, laborious work, which I don’t want to take the time to do. Isn’t there some shortcut I can use?
Look at the shapes, my muse whispers
So, I do look, and what do I notice?
For one thing, the individual letters all are made from variations on a basic shape. This visual systematicity suggests a corresponding phonological systematicity. That is to say, if I’m using a certain visual pattern to draw the sounds of the language then I should use a corresponding sound pattern to pronounce the language. This gives me information on the kind of writing system that I have devised. It appears that I have made a writing system with regular patterns in the representation of sounds. This would suggest that similar sounds would be represented by similar glyphs. The arboreal writing system is starting to look more and more like a syllabary.
I gotta quickly take a break here and explain what a syllabary is. There are three broad ways that languages use writing to represent words.
You’ve got your standard alphabets, where every individual sound is represented by a different symbol. You’ve got your syllabaries where every individual syllable is represented by a different symbol. And you’ve got your logographies where each individual word is represented by a different symbol.
The Arboreal script seems to be a syllabary because its visual patterning mimics the visual patterning of elements in a syllabary.
In a syllabary you have syllables patterned according to what the central vowel is in a syllable. To give another example of how a syllabary would work, let’s say you have 10 Consonants and 2 vowels in your language. An alphabet for that language would have 12 characters in it; a syllabary for that language would have 20 characters in it, if the language only allowed syllables to have an initial consonant and a vowel.
The choice to use a syllabary rather than an alphabet is related to the way that a language organizes its sounds. For example, English permits very long syllables with at least three consonants on either side of the vowel, e.g. sprints. An English syllabary would have hundreds of discrete elements. It would be horrible to learn. Languages like Korean or Cherokee, which do use syllabaries, have much shorter syllables, which makes using a syllabary as a writing system more feasible. I might also mention here that Cherokee, while it has shorter syllables than English, has much, much longer words. So using a syllabary in Cherokee is still more economical than using an alphabet.
But what does all this mean for my script? Why, it means that I know what the structure of the sounds is going to be! This language has a simple syllable structure, permitting only Consonant-Vowel combinations, leading it to adopt a syllabary as its writing system! Everything is starting to come together. I have activated retroactive creativity.
And now that I know the syllable structure of this language, I can generate a basic phonetic inventory.
Let’s say that we have the basic three vowels
[a, i, u]
and 13 consonants
[p,b, t,d, k,g, r,l, m, n, s, ʃ, h]
We can change these up later on, but these will do for now.
(the ʃ symbol stands for the sound represented by “sh” in English. All other symbols have approximately the same value as their English counterparts)
The only thing left to do now is actually devise the syllabary. We will start with three basic curve shapes to represent a syllable that’s just a single vowel.
And now all I have to do is add the necessary ornaments and sub-branches to flesh out the syllabary entirely. I’m not going to represent the entire syllabary here, because in the final draft I ended up adding some more vowels and consonants leading to an inventory of 86 discrete elements, which is too many to share on substack. I’ll settle for showing you just 21 elements. Here they are.
So, how would you actually read this? The top row shows what the central vowel of the syllable is going to be. The left column shows what the initial consonant of the syllable is going to be. (Recall that our Arboreal Language only allows Consonant-Vowel syllables and not, for example, Consonant-Vowel-Consonant syllables). The first top glyphs are read as “ahh,” “oo,” and “ee” respectively. The next row down would be read “pah,” “poo,” “pee.” And the next row would be read “bah,” boo,” “bee.” And you keep on reading them in that manner, systematically varying the vowel and the consonant depending what the central branch is and what the sub-branches and ornaments are.
If you look carefully, you’ll see that I encoded consonant voicing as an ornament to a sub-branch. The difference between “tee” and “dee” is a single thing. The vocal folds are vibrating when you day “dee” and they are still when you say “tee.” You can verify this for yourself by placing your hand on your throat and comparing the vibrations as you say “ta...ta...ta” to those when you day “da...da...da.” Nothing else about the sound changes, however. Your tongue and lips and lungs are doing the same thing, with just one difference. When the vocal folds vibrate during the production of a sound, linguists call that a voiced sound. Vowels are always voiced (expect in special circumstances). Consonants can be either voiced or voiceless. The distinction between voiced and voiceless consonants is important enough for the Arboreal Language that it’s encoded in its writing system, which will give me something to work on later.
But for now, I am simply happy to have been gifted the idea. To have grown a sound system and a full orthography from a single visual cue. I hope that laying out my process in this manner will inspire your own creations and fortify you with the knowledge that there really is no “right” way to make art. You can start from any point, and proceed along manifold paths. Indeed, for my next article, I’m going to consider a Writing system that I developed from an entirely different perspective--from the point of view of a linguist trying to accommodate a language with an already developed sound system, full of contradictions and complexities.
Beautiful and fascinating. Tell your muse I said, « good work! Keep it up. »
Can’t wait to see a full sentence.