From Lognet 00/1. Used with the permission of The Loglan Institute, Inc.

The Future of Loglan

by Alex Leith

There are two main logical languages in the world. There is Loglan itself, which was the first of its kind. Beginning in the nineteen fifties, Dr. James Cooke Brown created it as a tool with which to test the Sapir-Whorf hypothesis, (that the language we speak constrains hat we are able to think).

That test has never actually been carried out: the original language JCB created turned out to be too small for the purpose. Over the years the language has grown, until we now have a fully speakable, yet grammatically unambiguous and computer parsible language with a rich and growing vocabulary. Loglanists nowadays are more interested in the language tself than in testing the Sapir-Whorf hypothesis.

The other logical language is Lojban. During the Eighties a schism took place in The Loglan Institute, and a group of Loglanists, led by Bob LeChevalier went off and created a separate language on the Loglan model. The Loglan vocabulary is protected by copyright wned by The Loglan Institute (TLI), whereas LeChevalier’s Logical Language Group (LLG) wanted their language to be in the public domain. So they devised an entirely different vocabulary, while using the grammar and much of the morphology of Loglan. I had not yet discovered Loglan at that time, so I cannot speak at first hand of the schism, but I know hat it generated hard feelings which sadly have persisted over the years.

Loglan and Lojban have developed along slightly different paths during the past fifteen years, and there are probably advantages on each side. Nonetheless the two languages are siblings rather than cousins.

One may regret the split that took LLG away from TLI: one may bemoan the dissipation of effort in the development of two languages, one may be sad at the division of the speaker base. These criticisms are true,but on the other hand one may take the existence of two closely similar languages as a confirmation of the fundamental validity of the logical language concept.

Whether one takes a negative or a positive attitude, the act remains that we have these two closely similar languages. What are we going to do about it? This question has been exercising my mind since I took ver from JCB as Chairman and CEO of The Loglan Institute.

I will say first of all that there can never be any question of yielding control of TLI to LLG (or indeed to ny other body). The Trustees of TLI will see that that never happens.

However I believe it is now time for the two groups to begin to live as neighbours. Rather than huddling apart like feuding clans in the Highlands of Scotland, we should admire and celebrate our different ways of dealing with the same set of problems. We should see how we may help each other, as farmers share heavy machinery at harvest time. After all, we are both part of a minority, the small population of logical language enthusiasts. Our differences should be nourishing our enthusiasm, not stifling it. I have recently begun to see a way in which this might happen.

The Logical Language Group last year baselined the Lojban language. This means that for five years the language will not change, thus offering an pportunity for people to become proficient in it without fear of being forced to unlearn something as changes are adopted. I suppose there will be people adding predicates as they find a need for them, but the grammar, embodied mainly in the little words, will remain stable. Loglan has undergone minor changes in its grammar during the past couple of years—the mia subjunctive form, and the set/multiple refinements in the tu and mu family are examples. Nothing too revolutionary, more a matter of fine-tuning, but the Keugru (Care-Taking Group) are still ready to consider any changes that might improve the language. Vocabulary development continues to be very important: JCB sometimes said he would like to have a different word to express every shade of meaning we can think of.

Although not baselined, Loglan has reached a point at which no major changes are envisaged. Yet we may well find ways in which the language could usefully be extended: we may yet add little words, but it is very unlikely that any will be removed or radically redefined. Those who learn Loglan now—and the teaching materials we now have should make this a pleasant task—can be confident that the ground is not going to shift beneath their feet: at most they may find that a new flower has taken root.

The relationship we may envisage between TLI and LLG is like that between the factory floor on the one hand, and the research and development department n the other. The explicit baselining of Lojban gives it an advantage in applying the language to practical uses out there in the world—the LLG have for instance been considering the evident advantages of a culturally neutral and computer parsible language as a medium for international patent descriptions. An inter-language for machine translation, despite the swift growth of computer power and advances in direct translation from one natural language to another, is still a field worth pursuing. These are projects which require a language that remains stable in all details.

This role for Lojban releases Loglan to continue as the linguistic test-bed, where we can try out new solutions and develop new software. We already have our admirable LOD (Loglan Online Dictionary), the LIP parser, and the MacTeach learning aids. JCB was in process of adding to the latter a module for teaching he little words, a valuable addition should someone wish o complete it. There are important projects already in hand: the Resolver and its extension into audio input and output is one. Unambiguous semantics, nd a concept set for artificial intelligence is another area hat we should start to explore. If the Sapir-Whorf ypothesis is correct, then any language tacitly ncorporates a world-view. Let Loglan render explicit the world-view that it incorporates: or rather, let it be able to incorporate, in a particular linguistic interaction, any one of a set of possible world-views, providing a clear context of which one is currently active.

The first project for our linguistic test bed is to develop a subset of Loglan which will be easier to learn and to use. Although Loglan is much simpler than any natural language, how many of us are able to actually use it with fluency and confidence? Many have found it surprisingly difficult to use. This may in some measure account for the small number of Loglanists there are in the world. Human minds, it seems, don’t take kindly to the rigorous punctuation that ensures parsibility, or to the intricacies of quantification. It’s a language in which one can say anything one wants, but actually saying it may be tricky.

Loglan’s criteria of syntactic unambiguity, and of perfect resolvability, ensure that a computer can resolve he individual words composing a speech stream, and can then parse the production. If we were temporarily to shelve these criteria, and if, as we do with natural languages, we were to rely on the contextual nowledge and common sense of our hearer, then we could use a much simplified form of Loglan. This might well facilitate the acquisition of vocabulary, y allowing learners to practise within an easier framework. It could stimulate members to use the anguage more in our mailing lists, and later perhaps in online chat, such as IRC. It might even be considered again as an international auxiliary language.

When I last visited JCB in San Diego, we brainstormed—among other things—the Easy Loglan idea, and he was full of enthusiasm for this new approach. It is not intended to supplant Power Loglan—as JCB dubbed the Loglan we struggle with—but to provide a gentler approach to it. I present it here in the hope that it might encourage and stimulate logli to add, perhaps bit by bit, the sophisticated structures of Power Loglan until they found themselves as much at home with that as they had been with Easy Loglan.

This issue of Lognet carries an initial description of Easy Loglan. After discussion and any necessary tweaking, we might baseline Easy Loglan and start work on properly organized teaching materials. We could be ready to do that about the end of this year. At the same time let us explore the nuances of Power Loglan and all its possible applications. These languages are so young; we have hardly begun to realize their potential power. Let’s forge ahead and delight in surprising ourselves.

Copyright © 2000 by The Loglan Institute, Inc. All rights reserved.

Send comments and corrections to: djeimz AT megaseattle DOT com