Search This Blog

Thursday, January 03, 2008

Caia - Design Goals & Stuff

So, what ever happened to Caia, anyway? For those of you who forgot all about it, or never read the couple of posts I mentioned it in, Caia is my pet language - one of them, anyway. I can't remember exactly why, but at one point I decided I wanted to make my own language - perhaps it was because I thought I could make a more logical language than natural languages. There were several design goals with Caia.

Caia was meant to be a real, usable language; that is, you should be able to convey any meaningful thought through it. Furthermore, it should be sufficiently efficient, by various definitions, to actually be used as a native language, rivaling those naturally occurring in the world. Among other measures, it should be able to convey meaning in a short enough amount of time to be practical for real-world communication, able to compete with natural languages.

It was intended to be both simple (as much as a usable language can be) and intuitive. Natural languages are notorious for the amount of content that is either illogical or needlessly complex - some things you simply cannot reason out if you don't know them, and must simply be memorized. Our brains allow us to effortlessly learn even such illogical and complex languages as children, but it can be painful to learn some languages as second languages, or to process languages with a computer. I had hoped to create a language that is much easier to learn, more subject to reasoning - as opposed to memorization - and perhaps can even be analyzed by a computer.

Similarly, it should be strongly structured, in both form and meaning. In English, there are plenty of examples of ambiguity - things such as "The man shot the soldier with the gun", but other languages can be even worse. For example, it's common to completely omit the subject of sentences in Japanese, to save the time used to specify the subject; e.g. you'd probably say "mise ni okonatta" ("went to the store") instead of "watashi wa mise ni okonatta" ("I went to the store"), unless the previous sentences made it very likely that, if the subject were not specified, you would be talking about somebody else (random fact: this type of thing is actually appearing in English, these days, especially in instant messaging and text messaging; another random fact: the common Japanese exclamation "kawaii!" is a complete sentence omitting the subject, meaning "[it] is cute"). And even the (typical) manner of specifying the subject in Japanese, when the subject is specified at all, is often ambiguous; the exact same sentence in Japanese ("[watashi wa] maika desu") could mean both "[I] am a squid" and "[I] would like the squid" (think of ordering at a restaurant). A third example would be Japanese adverbs; for example (take this one with a grain of salt, as I'm not certain my understanding of the grammar in this particular example is precisely correct), "sugoku futoi", which could both mean "is incredibly fat" and "is incredible and fat". I wanted to create a language more precise than English, and dramatically more precise than loose languages like Japanese.

Everything satisfying these rules was left for me to do as I pleased, in one way or another. In some cases, I chose what I thought would be optimal. For example, I thought it made the most sense to structure a language such that the more important words always come before the less important words; the result of this is the verb-subject-object word order and right-branching pattern seen in Caia. Similarly, I chose to make it a primarily analytic, isolating language, to make it easier for computers to parse, yet use agglutination for some things, to significantly reduce the time to convey meaning in some specific areas.

In other cases, there was no 'optimal' answer, and I merely did what I wanted. For example, the sound of Caia ("kai'ja" in IPA notation) is purely aesthetic - it was simply what I think would be nice to hear. For an example of a 'sentence' in Caia (it doesn't actually have any meaning, as I haven't defined the vocabulary of Caia, yet): "Vaga ran mezh kana sit" ("va'ga ɾan mɛʒ ka'na sit" in IPA notation). You might notice that it sounds a lot like Latin; this is thoroughly unsurprising, as I think Latin sounds both cool and moderately pretty. On the other hand, there are some additional sounds I think are pretty that don't occur in Latin, which result in it having a bit of a middle-eastern or Indian sound (or at least of stereotypes of them).

However, the freedom of creating such a language is less than you might suspect. Some constraints listed are very harsh, and significantly limit what I can do with it. The time efficiency constraint has been a particularly sticky point, especially given that English is my native language. It's surprisingly difficult to create a language as efficient with time as English, as English uses a number of tricks - things such as complex syllables and ablaut - to achieve quite impressive efficiency, at the cost of other things, such as complexity. I've spent a large portion of the total time thinking about Caia in thinking how to minimize the time it takes to say things, while still retaining the relative simplicity, elegance, and strong structure desired.

So, where am I going with this? Well, that'll have to wait for next post (assuming I don't lose interest in this train of thought before then), as this post is pretty long, already.


Anonymous said...

"Caia was meant to be a real, usable language; that is, you should be able to convey any meaningful thought through it. "

So you shouldn't have any reason to use it...

Justin Olbrantz (Quantam) said...

So, I was wondering something, and I think you're just the person I was considering asking. Is 'smartass' a proper subset of 'ass', or are the two just partially overlapping sets?

Anonymous said...

Continuing from my comment before about subjectless sentencess being the possible end of mankind (perhaps even moreso than nuclear war or global warming), you might want to take a look at Phoenician since it's the forebear of almost every modern IE language to the west of the Euphrates.