Wouldn’t you know? There’s rhyme and reason to this phonetics lark after all. Not a kind I’m a natural at, but eminently sensible. The key, as always, is in giving things names. Even and especially those things whose existence we’ve taken for granted while using them to name everything else along the way. I don’t think too highly of the particular choices at work in the field’s notoriously cryptic symbols, but just because I wasn’t on the committee doesn’t mean there’s no good to them. In fact I ought to get to the bottom of the complete set.
You see, I’ve actually been on the brink of phonetics all along.
Once upon a time, in the hazy disconnected funk of the last century, before even this story’s ancient genesis, I cobbled up a little speech synthesiser. Give it a listen, I dare! This was 1999, and what else were you going to use your computer for when kicked off the family modem? Civilisation, usually, but sometimes I did projects like this. That one little demo sounds like I was onto something. I remember showing it off, well pleased with my invention, until people asked to make it say something else. Bugger.
The particular wheel I’d reinvented was concatenative synthesis, a fact I only checked up on years later. I’d figured out a basic set of sounds, and wrote out and sampled myself saying a list of words which contained the lot. Then I snipped the recordings right down to their individual units, named them by my system, and saved the lot to a trio of folders from which my synthesiser script stacked them end to end when asked to play. The result was pretty choppy. That’s because the art of gathering sounds was more nuanced than I thought. I hadn’t taken stress into account at all, or intonation. But the part of my system which stayed in mind is still present in my work. The bit where I identified and named those sounds.
My little proto-phonetic system was so simple that I’m still proud. It works like this. Capital letters represent their “name”, so [A] rhymes with hay, and [E] rhymes with tree. (I’m borrowing the phonetic habit of wrapping square brackets around my old symbols, I didn’t consider such matters of style at the time.) Lowercase letters represent the “sound”, rather like phonics, so [k] rhymes with cat, not hay, and [p] rhymes with plop, not tree. I think I just called the capitals the “names” and the smalls the “sounds”, but I’d been taught the difference when I learned to read so it was hardly a giant leap. Indeed, I was initially so naïve as to ignore the difference between consonants and vowels entirely. So capital [B], [C], [D] and so on existed too, although the lot of them were of course just [bE], [sE], [dE] and so forth. I culled them soon enough, along with such phonetic absurdities as the letters C, K, and Q all representing the same thing. (Interesting to me, the one I kicked the others for was [k], the same choice as made in phonetics.) There was one more group of sounds besides these two. I’d also figured that “th” and “sh”, and indeed “ch” and “ng” (think the end of “ting”) weren’t just awkward spellings but true, separate sounds. So I included a third list, that I merely titled “specials”. I recorded samples for those, and wrote them with parentheses: [(th), (sh), (ch)] and so on.
So in my old system, “speech synthesis” is written: [spE(ch) sin(th)esis]. Almost readable. But how about this: Project Andala is written [projekt andala]. Very almost identical. In fact, so much Anatara is exactly that way. Because this is how I seem to approach sound naturally.
It may be my way, but it’s not phonetics. Yet it was a start.
Something I’ve learned since is that there’s more to vowels than I thought. I’d made just the two versions of each. [A] is in say, [a] is in sad; [E] as in knee, [e] as in head; [I] as in bye, [i] as in sit; [O] as in hope, [o] as in hop; and [U] as in huge, [u] as in hug. But [U] was an awkward exception. Does it sound like “you” or “ew!” How about “oo”? I added [(oo)] to the specials list and left [U] as it was, but I could tell there was something awry. Indeed, the multitude of vowels is something I’ve come to appreciate since my first stab at collecting them. As vowels are every bit as important as all those consonants we bung between them, one kind so very often after the other.
In fact, vowels are the main sounds that we make. If you play around with a waveform editor, as I did in the process of making my synth half a lifetime ago, you’ll see the graphic truth. Consonants are the busy little valleys, full of whispers, between the mighty vowel hills. Speech is their sequence. And what a complex sound it is.
Fortunate for us, our brains are well tuned for it. Like the maddening complexity of what our eyes do and do not see, sound is handled as an unconscious process for us by the superhuman genius with which we share our heads. What we can “hear”, rather than wiggling charts of noise, is what the book I’m reading describes just perfectly as “gestures of speech”. In fact, let me quote the opening of its first chapter:
We will begin by describing how speech sounds are made. Most of them are the result of movements of the tongue and the lips. We can think of these movements as gestures forming particular sounds. We can convey information by gestures of our hands that people can see, but in making speech that people can hear, humans have found a marvelously efficient way to impart information. The gestures of the tongue and lips are made audible so that they can be heard and recognized.
Now that’s just a superb way of putting it! Speech is a uniquely fast and precise way to articulate ourselves. But fundamentally it is just the same as waving our arms or making gestures with our faces, as exploited by sign languages the world over to great effect. The core of it is recognition. We understand what others are saying, right back when we begin, by expressing it with our own mouths. We hear when someone has their tongue to their teeth, or when they pop their lips. The gestures themselves are at once universal and yet quite meaningless. Language is how we’ve come to put them together, to make meaning from our shared fluency in sound.
The way I was going to make my matrix for Aznush, and so also Anatara, was as a grid of vowels and consonants. It still is, but now I’m learning that there are more of these sounds than my own ancient alphabet described. From the lexis I’ve already contrived, I can tell I need a few. Not least if Azutara is indeed to be the subject of its own writing system! I’m forever into minimalism — keep it simple, stupid! — until the moment I actually get to work. Languages, notoriously, are anything but logical, optimised, minimal sets. They are organic. So it’s only right that I should learn on my feet, as we all do while acquiring, living and making them for ourselves.