The 2009 Smiley Award Winner: Kēlen

I am very pleased to present the 2009 Smiley Award to Kēlen, an engineered language with the soul of an artistic language, created by Sylvia Sotomayor. Congratulations to Sylvia!

Smiley Award 2009

http://www.terjemar.net/kelen.php

Elves and Dragons

Who can resist them, I ask you? After all, dragons are like dinosaurs, but they can fly and breathe fire (or sometimes snow or lightning)! It was certainly not in Sylvia Sotomayor's power to resist their allure, when, while in junior high, she encountered Tolkien for the first time, and concocted the wild and subversive notion that she could create her own language. Thus Kēlen was born, the language of the Elven Kēleñi.

Most first-time languages stop about here. Fleshing out a culture is complicated, and once you actually get into language, it becomes evident that it's, you know, difficult (just think about the term "laid-back". How do you create that?!), and as the challenge and the learning curve steepen, interest wanes, and pretty soon it becomes more interesting (and simpler) to draw pictures of elves riding dragons and write stories about finding ancient swords than to fiddle with noun cases, subordinate clauses and deixis.

I imagine there are probably thousands and thousands of juvenile languages like this scrawled on bits of paper scattered about here and there, forever lost to the wastebins of time. Of those thousands and thousands of languages, though, there's always one or two that refuse to die—whose creators are undaunted by the incredible task that grammar creation represents. Kēlen turned out to be one of those languages.

Where Everything Went Wrong

It's only with a slight wink that I say that formal linguistic training ruins people—especially those that find it enjoyable (that's the worst). I think Sylvia and I shared the epiphanic experience of kind of falling in love with the idea of linguistic analysis, likely taking the same courses from some of the same professors at UC Berkeley, albeit at different times. It's hard to explain to someone who doesn't really dig language (and I won't even try, since it seems unlikely that anyone who doesn't really dig language will be reading this).

Most American linguistics undergraduate programs proceed in roughly the same fashion. You start out as an L1 English speaker (probably) in an introductory linguistics course that fulfills some sort of a gen. ed. or breadth requirement, and the job of the professor is to blow your mind with natural language data. After the first week, you should pretty much be thinking, "Whoa! Language can do that?!" After a few weeks, the prof. introduces some hard stuff (e.g. Turkish morphology) to weed out the so-sos, and then after the intro, when the majors have been roped in, they unload the hard stuff on you as students start taking introductions to the various subfields of linguistics (phonetics, morphology, semantics, historical, etc.).

Anyway, somewhere near the beginning, usually, students are introduced to linguistic typology, and the concept of linguistic universals. Linguistic universals are just fantastic. They're a great big list of things that languages can and can't do. It's as if language is a sport, and these are the rules. How linguists and others interact with these rules is fascinating. After all, the "rules" themselves began their existence as observational rules (i.e. linguists like Joseph Greenberg looked at a bunch of languages and wrote down how it was they varied, and noted what it was that none of them seemed to do). Pretty soon, though, certain linguists started to treat these like mandatory rules, or laws, so that if they were presented with a piece of data that seemed to violate a particular universal, they'd try to explain how it was that either the data was wrong, or it had simply been misanalyzed.

For a conlanger attempting to create a more or less natural language, linguistic universals present a terrible challenge. Their very existence is problematic. If you don't pay any attention to the universals, then how can your language be natural? If you flout a universal, intentionally or unintentionally, that's enough for some to claim that the language is unnatural and alien. If on, the other hand, you make sure to bow to every linguistic universal, then that too makes the language artificial. What's one to do?

What Sylvia Did

On Sylvia's very well-designed Kēlen website, she quotes a familiar absolute universal: that all languages draw some sort of formal distinction between nouns and verbs:

The distinction between nouns and verbs is one of the few apparently universal parts-of-speech distinctions. While the universality of even that distinction has sometimes been questioned, it now seems that the alleged counter-examples have been based on incomplete data, and that there are no languages that cannot be said to show a noun-verb distinction when all relevant facts are taken into account. [Schachter: 1985]

To be sure we're on the same page here, there are plenty of languages (e.g. Hawaiian) where, for example, noho can mean "chair" (a noun), "to sit" (a verb), "sitting/seated" (an adjective), without any change in the form (all three are noho). The language, though, distinguishes quite clearly between the three in their distribution (e.g. on account of the strict VSO word order, the noho that means "sit" only occurs in the verbal position and gets verbal particles, and it never means "chair", and vice-versa). After dealing with even a small amount of data, one is forced to conclude that Hawaiian draws a clear dinstinction between nouns and verbs, even though a single phonetic form can be used in many different ways. What no language does (allegedly) is have, say, only nouns or only verbs; such a thing has been judged impossible, or, at the very least, alien.

This is where Kēlen starts to get controversial. For whatever reason (and I can think of at least four good ones), Sylvia took issue with this absolute universal. Why couldn't a language fail to make a distinction between nouns and verbs? Why couldn't it have one class that was either all nouns or all verbs? As a result, Kēlen became a verbless language: Nothing but nouns. (And, hey, think about it: that's the tougher route. I mean, a language with no nouns? Easy! "To boy girlingly huggingly" > "The boy hugs the girl". Done!)

So How Does It Work?

The first question one might ask is, how can a language exist without verbs? Sylvia writes the following:

Possibly the language would have a small number of words that do the functions of verbs without any of the semantic content. In other words, words that would tell how many arguments to expect and what the relationship is between these various arguments.

To sum up, I think a verbless language must satisfy the following:

Without verbs, a language must have something that informs the user how one noun relates to another.
In doing so, these elements should not relate any contentful information (e.g. in "The man hugs the fish" and in "The man cooks the fish", the man and the fish are in some relation where the man is acting upon the fish. That's all these verbless elements should encode—not the hugging or the cooking).
To make a good faith effort, this set of verbless elements should be small.
Finally, the class of verbless elements must be closed (i.e. in natural languages, we can create new verbs whenever we want, but we can't really create new determiners [e.g. a new "the" or "a"]. In this language, the verbless elements are like determiners, and about as contentful).

This is precisely what Sylvia set out to do.

The verbless elements Sylvia decided on are called relationals, and this is them (or if you prefer, these are they):

LA takes a single object: something that exists in a state or a location.
NI takes a single object: something that has undergone a change of state or location.
SE takes a single object: something that has a source and/or a goal.
PA takes two objects: a whole and its part.

I remember the first time I saw pa, I was like, "Seriously?! You could've done wild stuff with that fourth one!" But, of course, that's not the point. It's cool, though, just think about this. Take any verb, like "bake". If I bake a tasty and delicious cake, that cake has undergone a change of state (so this would be NI [which is written ñi when used, but in the abstract, it's printed in all caps]). Now here's the question: What do you need the verb for? Just what does baking involve? As I see it, it involves the following (or, at least, this is the impression I get from observation, as my baking skills are...suspect, at best):

Some set of ingredients is taken from a raw state to a cooked state.
Some agent initiates this act.
Some instrument effects the change (say, an oven).
This happen over a fixed amount of time, set by the heat of the oven, the ingredients, and the competence of the baker.

Now, think about it: What about this requires a verb? NI pretty well covers what's happening to the cake as it's getting caked, so to speak. Nominal phrases can then be used to take care of the rest: the baker is the agent, the oven is the instrument, the time period is fifteen minutes (I have a particular cake in mind), and the catalyst, I suppose, is heat. Each of those nominal items can be assigned a semantic role, and there you have the verb "bake". (It's kind of like peeking behind the curtain at a magic show, huh?)

Here are two simple examples I like because they illustrate something I usually never bother with in any of my conlangs (but which, of course, shouldn't be ignored). This will show us the difference between LA and NI:

1. Ōrra la jacēla jahūwa.
2. PAST LA bowl broken-thing
3. "The bowl was broken." (Lit. "The bowl was a broken thing.")

1. Ōrra ñi jacēla jahūwa.
2. PAST NI bowl broken-thing
3. "The bowl broke." (Lit. "The bowl became a broken thing.")

Okay, let's not get clever with the English translation of (1c) there: that's not a passive. It's a simple stative sentence saying that at some point in the past, the bowl was in the state of being already broken. What's interesting here is that the thing that changed is the relational. It captures the essence of what it means to be broken. (On an entirely separate note: What a disquieting notion that this language has a separate word for a broken thing... And it's not anything in particular—it could be a broken anything. The only coherent bit that ties them all together is broken—literally.) To be broken entails that something was once whole. From that point of view, then, a broken item has three states: not yet broken, in the process of breaking, and broken. And the difference between the three? Nothing but time.

Hey, just for fun, here's the fourth relational with the formula above:

1. Ōrra pa jacēla jahūwa.
2. PAST PA bowl broken-thing
3. "The bowl has a broken part."

That's how I think it translates, anyway. It's a lot of fun just to imagine up sentences in Kēlen. It's liberating, in a way—in a way that, perhaps, only conlangers can understand. Specifically, I hate verbs. I hate them! They're so complicated! I don't like creating them, I don't like using them, and I don't like looking at their ugly mugs. Nouns, on the other hand, are concrete and easy to get a hold of. Give me cases, O Conlangus (that's the god of conlanging, Conlangus)! Let me bask in their glorious certainty! I would love cases to be a vat of foam cubes that I could dive into and bounce around in (or with). I'd love to have a foam cube case fight, where we throw them at each other, perhaps dividing into teams beforehand and devising some sort of scoring system (there's over thirty of us in this vision, by the way). I imagine that if you nailed someone with the illative, it could really do some damage... But see, that's why they're foam!

But, yeah, just try to do that with verbs; I dare you. Verbs be all like: "Let me check my appointment book... It looks like at 2:39 p.m. and twenty-three seconds I'll consider deciding what I'm going to do today. That will have to be noted in my datebook and cross-referenced with my spreadsheet on the computer. Once that's done, I'll have to check and see how many people are in my house and what gender they are, and make a note of that on my shirt. And once that's finished, well, the day will be half over, and I might as well stay inside. But you go on ahead, if you feel it's important! Just wipe your feet before you leave. And don't bang the door shut!" Pff... Verbs.

How Kēlen Has Made Me Smile

You know what makes me smile about Kēlen? The fact that I've said all this, and this isn't even a tenth of what I could say. Kēlen is enormous! I suppose if one had to categorize it it would be an engelang, but unlike most engelangs I've seen, whose descriptions are usually about three hundred times the size of its vocabulary, Kēlen has the life and vibrancy of a full-fledged artlang. Take the relational se, for example. In fact, go here right now, scroll down a little more than halfway, and take a look at the conjugation (for lack of a better word) of se. That's not something a typical engelang does. What function does it serve, after all? It's opaque! (Well, aside from the 3rd.sg./3rd.plu conjugation sexme and the 3rd.sg. reflexive conjugation semeñ. You can't fool us with that tilde; we know what's up!)

And that's just the language! Sylvia's got a ton of information on Kēleni culture and society, a description of the Kēlen calendar, this crazy divination thing that I don't really get but really like to look at (it has a button you can press, and stuff happens when you press it!), not one, not two, but three scripts, a translation of an utterly intolerable prose passage, an online dictionary that immediately makes me want to stop using it and do something else because I'm so jealous of how well the freaking thing works and so frustrated with my feeble attempts to try to create something that's even half as good as it, and tons more!

Oh, and please don't let me gloss over the Kēlen Ceremonial Interlace Alphabet. It's awesome! This (right now, at the time of writing) is probably my eleventh time trying to figure out how it works, and I still don't get it. Just take a look at this:

The letter 'p' in Sylvia Sotomayor's Ceremonial Interlace Alphabet.

How is that p?! I just want to see my last name spelled out! Even the little animated example doesn't make it clear for me (even though the little animated example is genius).

And then, wouldn't you know it? Sylvia was so nice that she went and did it! She made my last name in the Ceremonial Interlace Script! Here it is, for all to enjoy and marvel at:

Peterson in the Ceremonial Interlace Script.

The point, I suppose, is this. Whenever you hear Kēlen mentioned, it's referred to as the language without verbs, as if that were it. And that is cool, and the system deserves the attention it gets. But that's far from the whole enchilada. Kēlen isn't an experiment the way many engelangs are: it's a language. It's a labor of love that has lasted a lifetime, and which lives on. And, in case you haven't clicked on any of the dozens of links I've provided all over this page, I think we should all be grateful for the Kēlen website which details the language so faithfully, and is designed so damnably well. It's one thing to create a fantastic language, but quite another to create and maintain a user-friendly site that allows others to be able to fully appreciate your creation.

It's been my pleasure to spend time with Sylvia at LCC1, LCC2 and LCC3, and I have to say, I'm so happy to be able to present the 2009 Smiley Award to her wonderful language Kēlen. Congratulations, Sylvia! I offer to you the following divination I randomized up from your site, part of which kind of looks like a "Y" for "Yay!"

A sample divination image from Sylvia's site.

Back to the Smiley Award Main Page