(A page from the Loglan web site.)

(From Lognet 94/1. Used with the permission of The Loglan Institute, Inc.)

(From the Logic-Worker = Logician)

by Randall Holmes

This will be an outline of what I am up to, rather than a formal column. For the last two or three years, I have been writing an equational theorem prover, first in the research language Standard ML and now in C++. The logical capabilities of the prover are more than ample for the needs of the logic implemented in Loglan, although the underlying logical paradigm is somewhat different. For example, I have written a "program" under the prover which automatically "simplifies" expressions in the propositional calculus and incidentally identifies tautologies. Propositional calculus is the logic of the propositional connectives no, a, e, o, u and their kin in Loglan. Quantifiers are a more complex problem. I will be attempting to design an interface which will allow one to talk to this prover in Loglan. Actually, I expect that the interface will at first only accept a subset of Loglan (but see below).

Some things which I could use from the Loglan computing community: a UNIX port of LIP (how many of the rest of you out there would like that?) [Dr. McIvor tells me that this is already available.--JCB], and some information about how LIP represents data internally and how it can be configured to talk to other programs. If the answers to these questions about LIP turn out to be the best possible, it might be possible to avert creation of a special Loglan subset and have the prover accept full Loglan from the outset. I hope that LIP has been designed with the thought that it might be needed as the "front end" of software which needs to "understand" Loglan!

On the question of what role a lodtua should play in the Loglan cummunity generally, I recently asked our sacdou what I should be up to and got the following reply [which I've neatened up for this unexpected reappearance!--JCB]:

Before I can answer that question, we need to know where are we in the logical language project. We've built a LIP-loadable grammar, one which, with remarkably few additions, has proved equal to everything we've asked it to do, or to be easily augmented to handle those few things it wasn't equal to. We're also building quite a large vocabulary with LOD (the "Loglan Online Dictionary"), one that is proving to be adequate for a wide range of literary purposes or, once again, to be easily expanded.

Finally, we have quite a few usages--by which I mean standard speech economies--although most of them, to be sure, are not especially "logical" in the sense of facilitating validity-conserving transformations. Still, all of our usages have, from the beginning of Loglandian time, been built with precise interpretability in mind. Just ask me, soi crano, and I'll tell you what they are, since I've been in on the construction of nearly all of them. In any case, strict interpretability has been a requirement of Loglan usages from the beginning.

For instance, we have the famous "observative", a routinely interpretable usage roughly equivalent to the sense of Fire! in English. Thus Lo simba! = Lions! may be interpreted as Mi katca su sanpa je lo simba = I observe one or more signs of lions. This is quite general. Thus Lo fagro! is interpretable as Mi katca su sanpa je lo fagro = I observe one or more signs of fire(s), and so is translatable as Fire!, while Lo gudbi monza! conveys the same cheerful observation that English Good morning! conveys. The observative is not a logical usage, to be sure; but it's a usage; and we will always know what to make of it.

Then there are the usages which together amount to our system of "implicit quantification". One element in this system is Ba preda = Something predas or is a preda. This can always be interpreted as Ba goi, ba preda = There is an x such that x predas. This is a logical usage in that it provides a tool that is directly useful in performing validity-conserving transformations.

Finally, for my third example, I will mention another element in the implicit quantification system but this time with the mechanism of conversion added: Da nu tugle teba = X is three-legged. This, too, is a logical usage in that it can be interpreted as Teba goi, ba tugle da = There are exactly 3 x such that x is a leg of X.

Obviously, we need 100's more of both kinds of usages. In particular, we need a logician to help us build logically defensible instances of the logical variety, and to teach our teachers how to express in Loglan the various tricky things we do in English often without even being aware that we are doing them, or of the logical structures we are implicitly utilizing. But we need to know how to do them explicitly in Loglan as well as how rigorously to interpret the Loglan usages we invent for such purposes when we want to transform them.

So what we need is a logician to keep a catalog of such inventions and their interpretations, his and those of others, and to publish, from time to time, editions of this catalog...at first a thin pamphlet, later a slim volume, much later a thick and finally definitive handbook which logli and teachers of logli can use for reference and instruction.

TLI will of course be glad to publish early versions of the "usage catalog" until it gets big enough and comprehensive enough for "us"--you and TLI--to take to an academic or commercial publisher.

Have I interested you in this formidable project? We surely need someone in this role, and I do not feel myself able to fill it any longer along with the several other things I am still trying to do for Loglan. Would you like to take it on?

Your column, by the way, could become the serial publication of pieces of that catalog. An early one, for example, could be addressed to the "wiles of English \plainand" and present a catalog of usages by which the various meanings and uses of and can be handled in Loglan. More ruminative essays like your last two could appear under their own titles in either LN or LL, depending on their length.

What do you say to this project/program? It could certainly help us achieve a higher and more uniform level of logical competence among our logli than is typical of human communities generally, thus giving, as you pointed out in your last paper, Loglan and the Sapir-Whorf hypothesis their best chance.

--Hue JCB

I agreed with JCB that Loglan needs a program such as he describes, and his "job description" does seem to cover my most useful role in it. I need to remind the rest of you, however, that, while I am a trained mathematical logician, I am not a Loglan speaker; I need a supply of proposed logical usages (or illogical usages) to analyze--supplying them will be your department, collectively--and I need time in which to analyze them (which may be a weak point of the whole idea) in order to make this work. In other words, I don't yet generate much Loglan myself. I do parse Loglan adequately and can evaluate proposed logical usages if you folks send them in to me.

My addresses are now on the back cover. Get in touch.

--Hue MRH

Copyright 1994 by The Loglan Institute, Inc. All rights reserved.

Send comments and corrections to:

djeimz AT megaseattle DOT com