From Lognet issue 90/1,

Sau La Sacdonsu (From the Start-Giver = Founder)

Our Arizona Trip: The cause of Loglan was considerably advanced during the two weeks centered on the first of March. That was the time my wife Evy and I spent in Arizona. We spent the first week in Flagstaff, which is where Northern Arizona University is, and the second in Phoenix, where the Seventh Meeting of “Contact” was being held. Contact is an annual coming together of anthropologists and science fictionists. At NAU I had a chance to work with a linguist friend of mine with whom I’m working on a paper for Language, and the NAU Anthropology Department—courtesy of Old Loglan-hand Professor Reed Riner—had also arranged with three other departments, Computer Science, Applied Linguistics, and Modern Languages, to co-sponsor me to give a three-hour Loglan workshop. This allowed me to present the Loglan idea to a partially prepared and reasonably heterogeneous academic audience.

The workshop was to be held two or three days after our arrival, giving Evy and me plenty of time to assess the interests of our various sponsors. We decided that, for the first hour, I would give a one-hour lecture on the history and character of Loglan, focussing on the many potential uses that the language now has for various kinds of scientists: anthropologists, experimental psychologists and language-teachers as well as the linguists, logicians and computer scientists who have been our limited public for too long.

The first hour was well-attended by faculty and graduate students from the four sponsoring departments as well as by students in Prof. Riner’s Human Futures class. After that I gave a two-hour demonstration of the various features of Loglan, allowing myself to be largely guided by the interests of the smaller, keener group that stayed with me for the whole three hours. For this “laboratory” phase of the workshop I had loaded the software I’d brought with me from G’ville—all four teaching programs, M1-3 and LIP—into a borrowed computer which my hosts had managed to hook up for me to a projector. This very flexible set-up allowed me to demonstrate nearly any feature of the language that my audience wanted to see or me to talk about. I found that LIP (the Loglan Interactive Parser) was especially useful in this setting. For example I ran the Chinese/Loglan/English triplet of sentences, Women chi ji, Mu titci lo henji, We eat chicken, through every possible word order in the three languages, using LIP to demonstrate the logical equivalence of all the numerous Loglan variations. So LIP turns out to be a useful laboratory tool in addition to its many other applications. 

My audience was also impressed with our teaching software, for I had time to put M1, M2 and M3 on the screen as well. One can never be sure of course, but my feeling is that we made a handful of stout academic friends for Loglan with that workshop. For example, one modern language teacher expressed a keen interest in helping me teach an intensive course in Loglan at NAU next Fall...should one take place. But more of that interesting possibility as it develops. 

A day or two later but still in Flagstaff, I gave a very different two-hour Loglan lecture-cum-de-monstration—again, computer-aided—to a group of gifted junior highschool students. One of my hosts in Flagstaff is a mathematician; and one of his daughters is in the "fast stream" at the local junior high. Through the child’s mother, whose idea this was, her teacher was easily persuaded to set aside two hours of the class’s time for me and Loglan. Evy and I were both amazed to see how swiftly these 12- and 13-year-olds learned! They gobbled up everything I put before them. I had the feeling that it wouldn’t have taken very many more of those two-hour sessions to have them speaking a very respectable variety of logically alert Loglan. 

Shortly after those two bouyant experiences, Evy, Reed and I drove down to Phoenix to prepare for the Contact meetings. Reed is one of the organizers of Contact and was soon absorbed in being just that. This gave Evy and me a chance to wander around among the other advance arrivals and get a feeling for which facet of Loglan we should try to put before this future-oriented group. Contact regulars are a mixed bag of anthropologists, engineers, hackers, and science fiction writers, artists and fans. As an old SF writer (circa 1950) I felt right at home among them. The Loglan workshop had been scheduled for two hours during the second day of the convention. So at the first day’s plenary session I asked those who were planning to sign up for it to indicate how they’d like it run. I then described the three or four tracks that had been suggested to me as having interesting possibilities. Well, 15 or 16 people did sign up, and a half dozen more came by to watch; and the votes of the former were virtually unanimous for the “subset Loglan” track. In short they decided they wanted to actually learn a small piece of this clever little language, a piece that I promised to carve out for them on the basis of a set of English sentences they’d turned in to me the night before. Talk about accelerated learning! I wasn’t sure that we could accomplish very much in so short a time, but it did seem attractively adventurous to even try.

In the event, those two hours did come surprisingly close to being enough to put them in command of a small subset of our language. We found a third, unscheduled hour that afternoon, during which we came even closer. I believe a fourth would have pulled it off. Clearly, the subset of Loglan I’d picked out for them was too large. In retrospect, I think there were just too many predicates. I’d had to tick off about 30 of them to handle the sentences they’d written up for me; and there were quite a variety of place-structures to be mastered. Still, everybody had a good time trying to pack them in. But there were just too many. In contrast, I don’t think the subset of grammatical rules I picked out for them was too large...although it was nearly the entire grammar of Loglan! (It’s hard to pick a subset out of Loglan grammar...it seems so all of a piece. You pull on any part of it and nearly all of it comes out!) Unlike the predicates, the little words I used to teach the grammar with seemed easy for them to learn. I got the impression that there was a good deal of one-shot grammar-learning going on. If this holds up, it would confirm a suspicion I have long held that there is no real difference between little-word learning and grammar learning in Loglan. The simplest little words are nothing more than sign-posts for a grammar rule. If the rule makes sense, so will the little word that calls it up.

Our heroes failed, of course. No one quite mastered Subset Loglan. But we all agreed that it was a magnificent try. It is even possible that making a large effort that failed was more rewarding for them than the easy success we might have experienced had we lowered our sights.

All in all the Arizona trip gave my wife and me an extraordinarily rich Loglandical experience, one that we will remember for a long time. We made many friends for Loglan. Some were old ones from the First Contact Meeting, which was held in California in 1983 and which Evy, I and Jenny attended; and also from our earlier trips to Flagstaff. But we also made some new ones.

Commercially, the expedition was also a success. The Institute sold nearly $1,300-worth of books, software and memberships in Arizona. As a result we have nearly 30 new L1-readers and five new members. We were given a table in the hall at the Contact Meeting, and the five who joined us did so after looking over our learning kit. I expect a good many of those other Arizona book-buyers will show up on our membership roles before long.

I want to thank NAU Anthropologist Reed Riner and his family for making this wonderful experience possible...indeed, for dreaming it all up in the first place!

New Members: Let me take a moment to welcome new members Meg Glasscock, Jim Smith, Alexis Wong, Marti DeMore, Karl Lueck and Larry Victor aboard our little ship. All these people came aboard in Phoenix except for one, Ms. Wong, who boarded us in Flagstaff. I thought at first that Ms. Wong was going to be our first native speaker of Chinese, but it turns out that she is mentally an English-speaker now, having for-gotten nearly all her childhood Chinese. Still, we came awfully close to getting our first Chinese student of New Loglan. We’re still looking!

Let me also take this occasion to welcome an earlier and larger group of new members: Garry Whyte, George Matthews, Dennis Smith, Mike Keller, Paul Crellin, David Gherson, Frank LaFontaine, Bob Boles, Bill Edwards, Cecil McGregor, Peter Ramsey, Harold Grovesteen and Mike Demoulin, all of whom have joined us since Lognet 89/1 went out. Welcome aboard, and we wish you all good loglanning!

It would be nice to hear from all of you in the pages of Lognet. A brief bio to introduce yourself would be all that is necessary. But it's your magazine now; make yourself at home.

Ads, Articles & Sales: Old Loglan-hands espe-cially will want to know how sales and publicity have been coming along. Well, four of our ads have appeared recently, as many of you have seen. One came out in Science in December, one was in the January Scientific American (but that came out in December too), and then there was a “free ad” in Algorithm that Wesley Parsons wrangled for us. (Thanks, Wes!) And one came out in Analog as well...or is about to.

A review has also just appeared: our own Rex May's "Tongues of Men and Computers" was published in this month's Liberty (March 1990, page 74ff). No other reviews or articles have actually come out yet, but three other Loglan-announcing pieces are either in the works or being considered by editors. In fact, four other magazine editors have called me recently to get information to do an article on us on their own. Looks like a new information wave is about to wash over us!

Financial Progress: Total revenues from both sales and dues have amounted to about $7,500 so far...counting from last July when we went public again. That’s not a lot of revenue as real businesses go, but it did put us in a position to pay back a substantial portion of our printing and other production bills. In fact, at present rates it won't be long until we’re actually in the black!

I almost said ‘again’. But that wouldn’t have been true. We have never actually had a positive net worth since work started on the 1975 dictionaries back in 1973. That's when we hired a staff of four young editors to execute the long-planned expansion of our word-list, and we had to go into debt to do it. That was a few months before The Institute was incorporated in 1974, and a small portion of those 1973-75 production debts are still with us. So it appears we have lived our whole corporate life so far in the red. Blessedly, that’s about to change.

How can you help put us on the other side of that fiscal colorshed? Pay your dues. Buy some software. Buy those new cassettes. (They’re great, by the way. Professionally recorded, this time. You'll find them very useful.) Talk about Loglan to your neighbors or workmates and get them to join you in supporting The Institute.

Membership Figures: Membership is, of course, the most important index of how we're doing. So you will want to know how that is coming along as well. The news is not quite so good here. At last count we’d acquired 35 new dues-paying members since last July. That's about a doubling of the figure with which we started, but a far cry from the goal of 200 members with which Lognet can turn into a bimonthly.

We've sold about 300 copies of the new Loglan 1 since going public, most of them quite recently. We sold a lot of other things as well, of course; but L1’s the “membership-prompting” purchase. Even if the “conversion ratio” of 33 members out of 300 book-buyers seems ominously low, we’re not particularly worried. There’s obviously going to be a lag between growth in the numerator of that fraction and relevant growth in its denominator. It evidently takes awhile, after coming into possession of the new L1—which is a pretty hefty book, after all—to make up one’s mind whether one is going to learn this new language or not...indeed, whether one wants to support The Institute or not. (Oh, how we could use that Loglan Zero (see p.14)!) We hope that eventually a sizable fraction of our bookbuyers will decide one or the other of these two questions affirmatively. In the meantime, we can expect membership to lag behind sales.

Welcome to Interface Specialists: One group of new members and customers has been especially interesting to me. These are the numerous bookbuyers who have told me that they are professionally involved in human/machine interfacing themselves, and wish to examine Loglan with a view to understanding how our interface function works. (The next paragraph is an open letter to this special group of newcomers.)

Many of you are apparently interested in what may be learned from the Loglan experience that will be applicable to your own work. Some of you are academics, teachers or researchers who work in universities. But more often you appear to be scientists working in private, commercial labs. From my correspondence with you, I gather that most of these laboratories are concerned with the development of “5th Generation” computer systems and products. Loglan may very well be your language. Let mewelcome you to its ranks. I will try to indicate in the next section how The Institute can serve you. Later, perhaps, you can decide how you can serve the language.

“Trade-Secret” Sharing: At the outset let me say that anyone interested in examining the source code, for example, of our machine grammar, or even of our whole grammar-writing system, is welcome to do so provided da is willing to acknowledge that what da is examining is proprietary. What this means is that our work may not be legally used for other people’s commercial advantage without The Institute’s permission. The Institute means to generate at least some of its revenues either by selling or licensing services or products that utilize or incorporate some or all of this grammar-writing software. So we'd have to know that this fact was understood, and its implications agreed to, before we’d be justified in “loaning” these systems out to others.

Let's consider an example. LIP, the Loglan Interactive Parser—a program that is currently being marketed by us as part of our teaching software (see Lo Nu Tifru)—incorporates the entire machine grammar of Loglan. In fact LIP is itself a part of the larger grammar-writing system which we call LYCES (“Loglan Yaccing and Corpus-Eating System”). LYCES is the tool we used to develop our unambiguous grammar in the first place; and it is also the tool with which we are currently maintaining it in ambiguity-free condition. We are scientists. Others are free to use our tools, or build other tools which incorporate any of our ideas, provided they will respect our own needs to survive institutionally in the process. We are, to be sure, a non-profit corporation. But we are unendowed, and currently unfunded except by private gifts, loans and dues from members... plus, of course, whatever we can earn commercially. Unfortunately, gifts and dues havenever been enough to support us; so we've had to borrow. So our continuing success as a research institute—indeed, as the caretaker, developer and teacher of Loglan—now depends on our becoming at least a modest success commercially; that is, as a creator and vendor of Loglan products and services.

It is from this commercial perspective that we are obliged to view the inner workings of LYCES. They are essentially what lawyers call “trade secrets”. What this means is that LYCES is an in-house tool of potential commercial value to us which we would be unwise to turn over gratis to competitors.

On the other hand if you are an academic, or part of a commercial research organization willing to cooperate with us and treat us fairly, we do not at all mind sharing our tools with you, or the knowledge of how they work. But in that case we should expect you to (1) pay us a reasonable fee for the privilege of using our trade secrets, or for examining them and using what you find useful in them, (2) agree to treat this information as proprietary, which means not disclosing it to others, except to such others and in such ways as we agree to—for example, partially and in a scholarly publication—and (3) agree to pay The Institute a small percentage (2% to 5%) of the retail price of any product or service which you or your company or institution ever creates which makes use of the proprietary information which you would in this way obtain from us.

If these provisions make sense to you, and you or your company or institution would like to explore these possibilities with us, please write me and we will ask our legal advisers to guide us through the thickets.

I should mention that The Institute has other “intellectual properties”—e.g., trademarks and copyrights—which may also in principle be shared. I will describe the Institute’s policies towards sharing these other two kinds of rights in the next SLS; but I am running out of space in this one.

Jury Duty: My fellow Academist, Dr. McIvor, spent several months in Florida this Winter, part of the time in a guest house on our land. Among other pleasant duties, his visit gave us an opportunity to explore the Eaton "jury work" announced in LN89/1:6. No other word-makers had volunteered to join us; so seizing the opportunity of working together on it, we took this happy duty on ourselves.

In the event, our experiment in "best word choosing" was a great success. We met for 20 consecutive days in late January and early February for a morning's work, and that was sufficient to make final word-choices for about 20% of the "Eaton Interface". (That's the block of "new complexes"—'83 and later—that currently comprises about half our new dictionary. At that rate, we have about 80 more days to go.

Next Winter we'll do it again. Perhaps others will care to join us. But such "judicial" work, we're learning, must be done interactively. There's no hope at all of doing it by mailed ballots. At least not well.

More Declensions? I have just enough space left to comment on Dr. McIvor’s “More Declensions for Loglan?" idea (LN89/1:6-8). I have read Wes Parsons' letter in this issue's Lo Lerci, and I believe he has given a strong argument against it. Wes's argument is going to have to be carefully answered; and I don't believe it has yet.

Unlike Wes, I was originally charmed by the "More Declensions" idea. At first glance, it seemed to me that the privilege of final-vowel variation ("declension"), for the purpose of making fine-grained semantic distinctions like the "ethnic" ones, could be given to a few more groups of primitive predicates without interfering with the affix system on which standard Loglan word-making depends. But of course I didn't know whether this could be done or not, and still don't. (I haven't made the requisite studies.) Still, essential non-interference with the affix system would have to be demonstrated before I could give my support to adding other ethnic-type declensions.

There's a more fundamental point to be made here. Whatever we do to the language from now on out is going to have to be in addition to the system we already have rather than a replacement of some major part of it. As I've promised many correspondents, the engineering phase of Loglan is over. The language is ready to be learned. I have given a number of people my personal word that Loglan will not change fundamentally in the future; that all future changes will be additive, or in some modest way corrective; that we will not change it again in ways that upset what you have already learned. That's my promise to all our members these days. Dr. McIvor's proposal, like all proposals the Keugru gets these days, will be viewed in its light. And that is, by the way, exactly the way he views it: as a possible augmentation of the existing word-making system.

Stephen Rice's proposals to add devices for handling Japanese-type register markers, as well as onomatopoeia and what he calls "pseudo-onomatopoeia", all of which were adopted by the Keugru in the interval since the last LN (see Sau La Keugru in this issue), are all excellent examples of additive, augmentative changes. They disturb nothing; they simply broaden our linguistic pallette.

Would a few more sets of declinable predicates augment Loglan? Or would they disturb the learning or the functioning of what we already have? Wes says 'No; it would disturb.' Rex says 'Yes, let's try 'em.' I'm not sure. What do you think?

Kerju tu. —JCB