(From Lognet 93/2. Used with the permission of The Loglan Institute, Inc.)
by James Jennings
(cejnoa = cenja norma = "change-average" = (3n) X, a movement or change-vector or movement in certain variables/parameters, is a trend perceivable/observable in domain/time-series/historical period Y by observer(s) Z. (1n) is a trend)
A Discussion Wherein the Author Projects Future Trends Based on Very Little Data and Then Attempts to Encourage Those Trends in Order to Prove His Thesis.
The story so far:
Over three decades ago, a linguist thought that an unambiguous language would be neat to study. He tried to build one and based it on formal logic because, well, how else do you do it? Fifteen years later, computers became a big thing and unambiguous grammars had a firm theoretical basis and, better yet, there were standard language designing tools. Our hero's language was re-engineered using the new tools.
Aside: The formal logic was not enough to make the language unambiguous. The machine grammar approach does not require formal logic. And yet, the formal logic is still there, giving the language certain special qualities.
Three decades ago, this same guy needed a vocabulary. The grammar was the most interesting thing, so he decided to keep the vocabulary simple: a very limited set of letter combinations, which forces one not to simply borrow words from natural languages. There was a distinction between "primitive" and "complex" words, but otherwise it was simplicity itself. It wasn't enough. Twenty five years later there was a much richer set of legal letter combinations and a powerful mechanism for borrowing from natural languages.
Three decades ago, the language needed a...a whatchamacallit...a...Well, I guess I'll call it an idiom. It needed a tradition for putting words together in order to express certain standard concepts. At the time, it was assumed that people would just create their idiom or standard usages as they went along. (People are pretty good at that.) Thirty years later, nothing has changed. People are still making it up as they go along, but for some reason there is not yet a large body of standard usages.
All natural languages have a grammar, a vocabulary, and what I have called an idiom. Obviously this new language is ready for an "idiom breakthrough."
Once we recognize the need to create an idiom the next question is, how do we do it? If you are of a theoretical bent (as I am) you will want a plan based on rigorous linguistic principles. We must choose a theory of "idiomatic usage" which will tell us what needs our idiom must meet and what limits there are to its generality.
Oops! No one seems to have heard of such a theory. (Sound of theorists cursing as they crawl back into their holes.) What do we do now?
Basically, we fall back to where we were before, inventing the new idiom as we need it. This is, of course, the way natural languages develop an idiom. The question is, can we speed up the process? That is, having recognized that idioms are born of necessity, we can go out of our way to "need" them more often?
Consider the pithy quote. When Julius Caesar said, 'Veni, vidi, vici' (I came, I saw, I conquered), he was using certain tricks of his native tongue to express a broad, sweeping idea in a compact, catchy way. He was making some standard Latin usages work for him. By doing the same thing in Loglan, we will exercise Loglan idiom and find out what needs to be worked on.
This is the work of Loglanists for the nineties!
How do we go about this? Start with any pithy quote in your native language that comes to mind and translate it into Loglan. What was the last pithy quote you've seen? How about
This is the work of Loglanists for the nineties!
First, look up the words. This = ti; work = turka (works at/on...with goal/purpose...); Loglanist: = logli; nineties: ... nineties?
Look! We found an idiom! The way English refers to a decade is to use the word in the name of the year that identifies the decade. "Nineteen-hundred-and-ninety-three" is one of the "ninety" years , and the ten of them together are "the nineties." Loglan has no plural and hence no mechanism for doing that. What should Loglan do instead?
Well, we could say it explicitly with something like, "the interval from 1990 to 1999" but that would defeat the purpose. We want something compact and catchy. Once you've invented a way to refer to a decade, you will want other people to use it from then on. Make it as easy to use and as logically transparent as you can.
How should we refer to a decade? Let's browse the dictionary a bit.
While we are browsing, we let our creative juices flow, and look for something promising. Perhaps we can count the decades from the start of the century.
No, wait. The first decade is the 'oughts (1900-1909), the second decade is the 'teens (1910-1919), etc. This is exactly the same as saying the 1900's are the twentieth century. We need
I think we've got it. The only thing missing is an attitudinal like oe (should).
Note that le neniri dekninkeo doesn't refer to which century the decade belongs to. Neither does the nineties. This is all right since the word le means that we effectively said the tenth decade that we have in mind. If the listener wants to know which tenth decade, da can always ask. [And we can answer Le neniri dekninkeo je le neveri hekninkeo.--JCB]
It sounds like we're done, but that's actually only the first step. The next step is very important: share your translation. Show it to a Loglan speaking friend or mail it to the Institute. (Electronic mail is very convenient here.) If you are told that your idiom is good, then you've learned something. If you are told that there is a better way, then you are about to learn something. How can you lose?
For example, after I had thought up le neniri dekninkeo, I told Bill Gober about it. He pointed out that the tenth decade might be different from "the nineties" because it began with 1991.
I had forgotten that there is no year zero (the year before 1 AD was 1 BC) and therefore:
If I were king, I'd put year zero back in, but since that isn't part of the Institute's Charter, soi crano, I'd have to say that le neniri deknie covers the years from 1991 through 2000.
Therefore, it's not an exact translation of the nineties (uu).
For many uses, however, it won't matter. People often use decades to identify general swings in fashion and culture which don't have clear cutoff dates. (eg: "The '60's ended in 1975.") For my purposes, le neniri deknie is good enough. If you want to be more precise, you can always be more long winded. One of Bill's suggestions was
The important thing is that idioms are built by communities. If you invent an idiom, share it. If you hear an idiom you like, use it. If you hear one you don't like, offer a different one. Or wait and see if it grows on you. Above all, be a part of the community in any way you can.
I'd like to finish with a case study. Sometime last year Kirk Sattley brought up
All that glitters is not gold.
and offered several Loglan versions. He started with the long-winded, logical claim:
Note that the English is not at all compact, but the Loglan is. It's one of the advantages of a logical language. I'm not sure that it's "catchy" though. The No, raba goi construction sounds very formal, while All that glitters is not gold sounds poetic.
Some of Kirk's variations:
Kirk unleashed his half dozen versions on the rest of us. The resulting discussion (between Kirk, Bob McIvor, and myself) yielded this astounding burst of creativity.
[This one needs a bit of work. Aurmo should be part of the predicate, not embedded in an argument, and I'd personally use punfo ('purely/homogeneously') instead of nercti here. The result would be: Lo brili ga no punfo aurmo = 'The mass of brilliant things is not purely gold.'--JCB]
[Djori doesn't really mean is a member of in the set-theoretic sense required here, although we've sometimes used it figuratively to mean that. We need a set-theoretic term and don't have one yet. Nursei won't do, for that would include all of a set's members. For the time being, I'd be tempted to use seidjo "set-member" for (2n) 'is a member/element of/is included in set...', and setpai "set-part" for (2n) 'is a set wholly included in/is a subset of set...,' whence nursetpai would be a superset.--JCB]
(meliusi from me liu si, a predicate made from a compound Little Word meaning 'having the characteristics of "at-most-one-of"')
[Meliusi doesn't quite do what the coiner intended. What the phrase me liu si actually says is "having some property of the word si". What the word-builder evidently wanted, however, was a predicate that would ascribe the properties of whatever the word si refers to, namely at-most-one-ness, in short, something that would mean "at-most-one-ish." We have an operator that will create that meaning, namely lie, the indirect designator. So this logically essential step was left out. Putting it in yields melieliusi, an interesting sort of compound, but rather frolicsomely one-of-a-kind, uu.--JCB]
This burst of creativity is larger than usual, and some of these examples might not be quite right, but it does show that there are many different ways to approach an idiom. Each of the above examples highlights a different feature of Loglan. And besides, writing them was a lot of fun.
So I entreat you to brush the dust off your copy of Bartlett's Familiar Quotations, and join the community.
Kanoi ba nu cnida purplikuo, e no nu dzamao, ki mu oa vetfa ba.
(purplikuo = purda plizo kusmo = "word-use-custom": (2n) is an idiom of language...)
Send comments and corrections to: