me lu ju'i lobypli li'u 7 moi

From Lojban
Revision as of 00:09, 25 May 2014 by Mukti (talk | contribs) (Conversion up to "A Very Short Introduction")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

For a full list of issues, see zo'ei la'e "lu ju'i lobypli li'u".
Previous issue: me lu ju'i lobypli li'u 6 moi.
Next issue: me lu ju'i lobypli li'u 8 moi.

Ju'i Lobypli Number 7 - November-December 1988

Front matter

 Copyright, 1988, 1991, by the Logical Language Group, Inc.
 2904 Beau Lane, Fairfax VA 22031-1303 USA
 Phone (703) 385-0273

All rights reserved. Permission to copy granted subject to your verification that this is the latest version of this document, that your distribution be for the promotion of Lojban, that there is no charge for the product, and that this copyright notice is included intact in the copy.

Ju'i Lobypli Number 7 - November-December 1988

 Published by: The Logical Language Group, Inc.
 Bob LeChevalier, President and Editor
 2904 Beau Lane, Fairfax VA 22031
 (703) 385-0273

Lojban Grammar Completed

Lojban Ready for Classes! Are You?

This newsletter is a publication by and for people interested in the development of a 'logical language', based on concepts described in the June 1960 Scientific American article Loglan, by Dr. James Cooke Brown (JCB), and in later publications written by Dr. Brown and by others. The Loglan Institute, Inc. of Gainesville, Florida has obtained a registered trademark on the name 'Loglan' for its line of grammars and dictionaries. This acknowledgement is not to be considered acceptance of the validity of this trademark.

The use of the name 'Loglan' in these pages by the editor or by correspondents is generally not in reference to the trademark, or the products covered by it. Rather it is descriptive of either the movement of supporters of which this newsletter is the spearhead, or to the language development effort proposed and initiated by Dr. Brown, which has now culminated in Lojban. 'Lojban' generally refers to the current version of the language developed by The Logical Language Group, Inc. The editorial policy of this newsletter is to print references to 'Loglan' and 'Lojban' exactly as written by the authors to eliminate possible confusion of their intent.

The Logical Language Group, Inc., to be known in these pages as la lojbangirz (/la lozh,BAN,geerz/), is a non-profit organization incorporated in the Commonwealth of Virginia, and has its principle office at the masthead address. Copies of the Articles of Incorporation and Bylaws are available at cost. la lojbangirz is not connected with The Loglan Institute, Inc., and no publication of this organization is, or purports to be, approved of by Dr. Brown or The Loglan Institute, Inc. Dr. Brown has expressly disapproved of this publication effort.

Ju'i Lobypli is distributed for free. Donations are needed in order to cover our publication costs and to support our development work. We have implemented a voluntary balance system - you may maintain a balance from which we will deduct our production costs for materials sent to you. Details may be found below. la lojbangirz is seeking tax-exempt status from the IRS as a 501(c)(3) organization. This may allow your donations, in excess of costs for materials you receive, to be tax-deductible. Upon obtaining IRS approval, we will notify all previous donors as to how much, if any, of their donations may be considered tax-deductible.

Our bylaws require us to inform prospective donors that we are limited to spending no more than 25% of our income on administrative expenses not directly related to the purposes in our charter. We will account at least annually in these pages to show you that we have met this limit. You are encouraged to make donations contingent upon our meeting this requirement.

This newsletter is being sent to about 400 people in about a dozen countries. This newsletter completes a year of regular quarterly publishing.

First, a Commercial

We need your help (and money)! The last issue, our largest, cost us over $1500.00 to print and distribute, while it generated the smallest response yet in terms of donations and contributions towards voluntary balances. This more than used up the surplus we had from JL5, which had the highest response rate yet. Since I'm still not working, Nora and I cannot afford to continue paying the bills for JL. Thus, if you want to keep seeing this newsletter, we need a lot more of you to contribute to its upkeep. While about 3/4 of you have so far indicated that you want to receive JL, only about 90 of you have positive balances. Moreover, a half dozen people: Nora and I, Jeff Taylor, Mike Yoder, Jack Waugh, and Jeff Prothero, have contributed over half the $8000.00 we've spent this year. We hope that you will remedy this as we bring the year to a close.

I believe we've earned your support. We have delivered regular newsletters, and committed to continuing their regularity; these newsletters, by your accounts, are more readable and interesting than The Loglanist was. More than that, we have pretty well met the goals for this year, and did so reasonably close to our scheduled dates. We have re- announced the language to the public and we have completed the basic development of the language (more on this below). I believe we've also been more responsive to your letters and orders than was The Loglan Institute, Inc. And finally, we are not asking nearly the amount that organization has asked ($50 per year) for your right to be kept informed of the language we want you to learn and use.

The four issues this year totalled less than $15 for those who have paid - less than the cost of most magazine subscriptions. Since we do not sell advertising or our mailing list, you get only what you pay for - news about the logical language effort. I'd like to think that you have read far more of our material than most people read of the magazines they subscribe to. Your support costs less than a typical dinner at an average restaurant, and the result is generally more rewarding and lasting (unless you need to diet like I do). And unlike most fund-raising projects, we actually give you something tangible for your contributions.

Think about it. And respond with whatever you can spare for our support - and at least enough to cover what we've been sending you.


If you ask to be kept on our mailing list, you will receive at least one more issue, even if you do not have a positive balance. If you ask to be dropped, we thank you for giving us a chance (We don't expect many of you to drop - in over 200 responses so far, only 6 have asked to be dropped). But we need to hear from you, especially if we haven't heard from you before.

(end of commercial)

We had intended to make this issue shorter, due to our finances and some feedback that the last newsletter was too long. However, we had so much good material that we couldn't decide what to cut. So we will let you decide what you want to read, and will use your questionnaire responses on the subject to determine the future policy. This issue has a significant amount of Lojban text, in addition to your letters of general interest, and to responses to those letters. There will be some 'teaching' going on in those responses, and newcomers will find answers to a lot of their most common questions there. The letters are on various topics, unlike those of last issue, and I recommend at least skimming each letter and response to see whether it contains material that you are interested in.

We anticipate that the next issue, and all succeeding issues, will come out approximately one month earlier. I hate having to compete with the Xmas mailing season - and I hope all of you get this before the new year. We also want the May 1989 issue to be received early in May, so that people are reminded of LogFest 89 (June 17-18, 1989) early enough to make plans to attend.

We regret to inform the community that Joel Shprentz's computer, formerly supporting the Capital Loglan Bulletin Board, has met an untimely but permanent death. Joel is looking for a new PC computer, but CLBB will be down until further notice.

Bulletin Board services to the community are available via a Loglan conference on the AMRAD (Amateur Radio) BBS, also local to Virginia/DC

 (703) 734-1387
 300/1200/2400 Baud
 Lojban Data and Shareware File Upload and Download; PC Pursuitable. Courtesy of Lawrence Kesteloot.

Send messages to the Editor via UseNet/ARPAnet/UUCP care of Joel Shprentz at his address:



The use of Lojban vocabulary in the following pages is intended to aid readers in learning the gismu, which were enclosed with JL6. Certain words, defined in our glossary, are used to free us from all the misleading connotations of their closest English equivalents. Thus, the glossary probably should be the first thing a newcomer reads. If you do not have the gismu list, please write and order same, or order LogFlash 1/MAC LojFlash 1 for PCs and MacIntosh computers, respectively. These programs teach you the gismu, and give you an on-line list to work from, as well. For those who have the Synopsis of Lojban Phonology and Morphology, skimming it for italicized words (which are defined where they are marked) will give you a greater understanding of the terminology.

Your Mailing Label

The top line of your mailing label gives the status of your voluntary balance. A typical balance line looks like:

C 50.00 -D 10.00 -S 21.25 =B 18.75

This says that we have $50.00 in Contributions from you, including matching funds and volunteer/award credits. You have Donated $10.00 of this as a gift not to be repaid. We have Sent you $21.25 in materials, resulting in a net Balance for you of $18.75. For foreign recipients, this balance does not include the postage for this newsletter, which we won't know until we take the envelope to the post office for shipping.

la lojbangirz Incorporated

The Logical Language Group, Inc. was officially incorporated by the Commonwealth of Virginia on 3 November 1988. The initial Board of Directors and members are identical: Nora and myself, Tommy Whitlock, John Parks-Clifford (pc), and Jeff Taylor. No one else has asked to become a member yet - the qualifications sought are a substantial commitment to the success of the project, and corresponding contributions in time and/or money to that end. We will consider all applicants.

There is no discrimination against new people either. Tommy was not significantly involved in the project until he helped with the development of the gismu list. Athelstan first heard about us at Evecon last January, and first became active in June at LogFest 88. He has been asked to be a member, but has put off accepting until his life gets better organized. This has not stopped his outstanding participation and significant contributions to development effort. We welcome the rest of you to participate in whatever ways you can.

As stated previously, Nora and I are no longer matching contributions to balances or donations. This is partly due to our finances, and partly because, with incorporation as a 501(c)(3) non-profit organization, we expect that your donations will be tax deductible from the date of incorporation. (We have not gotten approval of this status yet, and it could take several months to obtain it. It appears, though, that your donations prior to approval but after our incorporation date are covered - though possibly not your voluntary balance contributions, since the IRS could decide that you are receiving goods for those contributions. Please consult your tax advisor before deciding whether to deduct any amounts you've contributed. When we get IRS approval, we will notify all donors of any eligible amounts that you've donated, but do not expect this prior to next April.

The Board has already had its first meeting. A proposed set of Bylaws has been drafted and is being circulated; it should be in place by the end of the year. The business of getting established as a business is high priority, but we have the higher priority of preparing materials for you, and especially for the upcoming classes starting in January.

Our annual meeting will occur at LogFest 89 next June 17 at 2PM. You need not be a member to participate, and are welcome to attend. Several decisions about our future activities are likely to be made. Only members have a vote, but in such a public forum, you can be sure that those present will be carefully listened to.

While have had a few new people overseas since we started, contacting Professor Semenov could be a major breakthrough for Lojban. The people of the Soviet Union the are known for their interest in artificial and international languages. This is perhaps due to the large number of nationalities and languages spoken within the Soviet Union, and perhaps to a greater interest and awareness of language than most Americans have. With the recent developments under Premier Gorbachev, Lojban can serve as one of the basis for further intellectual ties between the US and the USSR. We will be communicating further with Professor Semenov, attempting to get his opinions and assistance on developing a community within his country.

With the completion of the major development efforts needed to teach Lojban, we are finally ready to significantly expand our community. A new revision of the brochure is being prepared, and should be available by January. We intend to translate this brochure into other languages,including at least German, French, Spanish, and Russian (for whom we have identified linguists). We seek linguists to translate the brochure into other languages, most especially into Japanese, Chinese, Arabic, Hindi and Esperanto. If you are reasonably fluent in any non-English language, or know someone who would be willing to assist our translation efforts, please contact me.

You also can get or make copies of the brochure for your own recruiting efforts. Posting a couple dozen copies in the Computer, Anthropology, Sciences, and Linguistics departments at your local colleges and universities may be all that is needed to build a local group of Lojbanists and enable a class to get started. (We are willing to sponsor student groups; some universities require an organized student group before allowing such postings. Let us know what you need from us.) Or post a few at work, especially if you work in the computer field. Companies are quite willing to support employee activities of this sort - Brooke Albert has gotten significant assistance from UNISYS in setting up the Boston class.

New Recruits - An International Language At Last?

We had a fall-off in new responses in September and October, but November nearly made up for it. Mike Gunderloy has continued to review JL in his quarterly review publication Fact Sheet 5. Each issue brings us a few new responses. Brooke Albert has recruited 9 new people from his workplace, who hopefully will form the nucleus of the Boston class to start early next year. Jamie Bechtel put up 20 copies of the brochure at Univer- sity of Nebraska, Omaha. Nearly all were gone within a month, although we haven't yet heard from those who picked them up. Athelstan takes recruiting honors this time. He has brought in a dozen new Lojbanists, many from the Philcon 88 SF convention.

Athelstan may have made an even greater contribution by going to Philcon for us. Staying at the hotel, but not for the convention, was Professor Alexey Semenov, Deputy Minister of Education for the Soviet Academy of Sciences. Professor Semenov was reportedly quite interested in Lojban, took a full set of material, and gave us his ad- dress and phone number in Moscow (I wish we could afford to use it).

We will shortly be starting on a short video about Lojban, aided by Brad Lowry. When complete, it will be approximately comparable to the brochure as an introduction, but will have some actual conversation, and will be suitable for spreading the word about Lojban via public television and science fiction conventions.

By the way, several others of you have been mentioning Lojban at SF conventions. Keep it up - we occasionally hear from new people who mention hearing of us at conventions that I've never heard of. For our benefit, it wouldn't hurt if you dropped us a postcard telling us that you recruited people at a convention, so we know who did the good work. Also be sure to send us any names and ad- dresses you gather. One recruit who recently moved to the DC area heard about Lojban at a Minneapolis convention and gave his name and address to the person who told him about Lojban, but neither we nor he ever heard from the recruiter (who is still unknown).

After the first of the year, we will be contacting various people who might be able to help us spread word about Lojban in computer, scientific, science fiction, and linguistic circles. Articles in magazines are much more effective than ads, so we will be working to get such articles written.

I will also be trying to sign on to Compuserve and other major networks using the free time that came with my modem purchase. I will upload copies of the new brochure and try to get some discussion going on those networks. I may stick with one of them to supplement our bulletin board support, but haven't decided whether we can afford it. If any of you are involved in one or more of these networks and want to take this task from me, I would appreciate it. I suspect I won't be that efficient, since I am an inexperienced user of these boards. We also have a local volunteer who runs a FIDOnet bulletin board. I don't understand the workings of this net very well, but we can probably support FIDO to some extent if someone else knows better how to best use it to support our needs. I still am looking for a good means to hook into the Usenet/Arpanet/Internet system, where a good deal of discussion on linguistics and languages apparently takes place. We occasionally are mentioned in traffic on that system, but I would like to be able to directly answer what apparently are a lot of questions.

Remember: Lojban is your language. Help us make it successful.

Lojban Grammar and cmavo List Completed

Immediately after JL6 went out, I turned my attention to the grammar and cmavo lists, with excellent results. It took only a week or so to incorporate the rest of the machine grammar to a level matching JCB's last available grammar. Then, using ideas proposed at LogFest and in further work with pc and Jeff Taylor, among others, I was able to devise solutions or responses to all of the known criticisms of JCB's grammar. We eliminated JCB's 'machine lexemes', although we had to add 'lexer lexemes' of a related sort; these latter, however, are generated by rules that are verified unambiguous by YACC. Lexer lexemes are inserted to mark cmavo that are normally compounded into single words, such as nuke, compound tenses, and multi- digit numbers. Jeff Taylor, and possibly others, are working on parsers which will process and add in these lexemes; Jeff's current approach is to use two YACC parsers, one to do the compounding, and one to process the main grammar. He plans to take some vacation time shortly to work on this, and hopes to have a parser done before our next issue.

We also were finally able to abandon using LYCES, the development tool written by Scott Layson, which JCB offered to all Institute members who wanted to work with the grammar a few years ago. JCB of course disapproved of our use for purposes he did not intend. Upon incorporating all of the compounding rules, we were overflowing Scott's conversion of YACC. I luckily located a work-alike alternative - PCYACC, put out by Abraxas Software in Portland. These people gave me outstanding support - Lojban is unlike most computer languages in what it requires of YACC, and they had to tailor a special version for us.

The final grammar is significantly longer than JCB's versions of the grammar, since it includes all of the compounding rules. Incorporating those rules proved valuable - I found several ambiguities hidden in the hitherto unwritten compounding rules, ambiguities which violate the Lojban standard of a single interpretation for any string of sounds.

Another reason the grammar is longer is that I have gone through all issues of TL, looking at proposals that were made during the 1976-84 period. Many of these seemed to have been adopted, but were forgotten by JCB; others had been deferred until they could be looked at with the machine grammar, and still others were probably intended to be covered within the compounding rules. In general, if a proposal was made and found significant support, we have either incorporated it in the grammar, devised an alternative, or identified specific compatibility reasons for excluding it. Thus, the efforts of all of you who worked to learn the last version of the language have finally borne fruit.

The last major accomplishment of the new grammar is that it finally incorporates a presumably workable MEX - a grammar of mathematical expressions. The MEX grammar simplifies to something resembling the old simple usages (and also English mathematical expressions), but is capable of going far beyond what any other language has been able to do with mathematics, and can do so unambiguously. For those who received The Loglanist, TL Issue 5/1 contains a specification of requirements for MEX written by Bill Mengarini.

As far as I can determine, our new MEX grammar fully meets the criteria Bill set forth, and does a few more things besides. Our main limitation will be in coming up with a written description of the MEX grammar and its capabilities. Most of its capabilities are very specialized in application and will not be taught in any depth in the first textbook. It will remain for mathematicians and computer people to take what I've done, test it, and determine the usages that will make it useful for mathematics. Since the MEX grammar operates in parallel to the rest of the grammar, those who learn the basic language will find it reasonably easy to learn MEX.

I also finished a revision to the cmavo list, incorporating dozens of comments, as well as some of the nuances implied by the new grammar. However, I found that my descriptions were in some cases inaccurate, and that I don't know the new grammar well enough yet to catch all of the inaccuracies. I therefore started to write a technical grammar synopsis that delineated the design of the grammar and all the details of usage implied by the design. After doing some 40 pages of description, and making several minor adjustments to the grammar as questions arose, I sent this partial grammar draft and cmavo list out for review to about 20 of you who had expressed the strongest interest in the grammar, or who I knew had expertise with machine grammars. I haven't heard from all reviewers yet, but I have a clear idea of what needs to be done.

Unfortunately, this review leaves both the cmavo list update and the grammar description incomplete. What has turned out to be needed is a complete rewrite of what I've already done, a rewrite that will have to wait until after the classes get started. The material I wrote is com- prehensible to someone who takes time to study it and who has some prior general knowledge of the language.

It is not very useful as a teaching document for the community (and wasn't intended to be). My reviewers agree, however, that the rewrite needs more overview and general explanation of what's going on in the grammar, and better definitions of terms. This will take quite a while, and cannot be allowed to interfere with the textbook and class materials preparation that will be our main emphasis in the next few months. The good side of this is that, of the reviewers that have responded, all but one (Jim Carter) believes that the grammar and cmavo list are basically ready to be taught; Jim would like us to adopt more new ideas from his derivative language (see below). pc has some small quibbles on usage, which so far sound convincing to me (he's assembling a copy for me that I haven't yet received.

The result of all this is that I can safely say that LOJBAN IS COMPLETED AS A LANGUAGE, even if the writing about it still has a ways to go. It is ready to be learned. There is no reason to believe that any changes that might surface in the teaching or grammar description process will be significant to someone who has started learning the language. The types of rules questions that come up are comparable to English quibbles such as split infinitives, who/whom, etc. that do not generally interfere with communication. Most are rules that humans speakers will often ignore - rarely do language speakers, even fluent ones, speak without minor grammatical error. By the time you speak or write Lojban fluently enough to care, any such issues will be resolved.

Since we have a cmavo list, albeit incompletely described, and a partial grammar description that contains significant useful information, as well as the completed machine grammar, we do have some material we can distribute to those that want to tackle it. If you remember that this material is draft only, you will find it useful and possibly interesting. It is probably comparable in usefulness and stability to JCB's Notebooks 1 and 2 at their time of publication; there is a lot of information there, with much not explained; the material is generally accurate and enough close to final to be useful, but individual items are not cast in concrete and may change in the next few months. The text amounts to about 100 pages, and thus is expensive for us to reproduce. We thus cannot afford to send it out unsolicited, so see the information on ordering below to obtain it. We are happy to accept comments and questions from those who receive it and wish to review it.

Jeff Prothero has now incorporated the new grammar in a version of PLOP (Public-Domain Lojban Parser). It needs the work of a good 'C' programmer to complete a lexer, since it handles only a limited subset of the language without it. Jeff and I can supply the information needed to make these modifications. We will send PLOP to anyone who asks for it. When this, or another parser is completed, it will serve as a standard for proper Lojban grammar for anyone who uses it.

LogFlash/LojFlash Status

We sent out some 85 copies of LogFlash 1 (PC) with the baselined gismu list in September. I have had no significant feedback on how people are coming, except that most people seem to be using lesson sizes around 20 instead of 40. This has caused us to consider a revision that will be added before the next release, as described below.

Richard Kennaway offered his MacIntosh version of the teaching program (Mac LojFlash) to JCB as well as to us. He asked us to accept the same royalty terms that JCB wanted, thus not favoring us financially. In return for this, he gave us source and permission to modify and adapt the program if we wished. We had trouble getting the MacIntosh diskette merged with the new gismu list, but fi- nally sent copies to some 16 users in early November. Both Mac LojFlash and LogFlash PC take text files of a particular format for its word list. For anyone still uncertain of whether our language version will survive, either version of the program can easily be used with other language versions, or even with natural languages, provided that word list data is prepared. Thus your investment in these programs is quite secure.

We still have ongoing reports of portings to the Amiga, UN*X C, and Apple 2, but none are complete, though the Amiga version may be close. Our Atari ST conversion volunteer indicated that PC emulation boards for the Atari are able to run LogFlash, and are becoming sufficiently widespread that he doesn't think the market justifies the conversion time (only 2 others of you have indicated interest in an Atari version).

Art Wieners has a UN*X C flashcard program that operates differently from LogFlash. It is built around a voice synthesizer; the program causes the word to be spoken along with the screen display. Art likes it, and will be trying to make it available to other northern New Jersey Lojbanists who want to use his machine - possibly as part of a class there. Others who have a UN*X machine and are willing to invest in a voice synthesizer can probably get a copy from Art. Let me know if any of you are interested. Art's efforts have led us to try to get voice synthesis added into the Mac LojFlash program as an enhancement for the future. (It possibly can be added to other machines as well; we have working voice synthesis algorithms for Lojban, and merely need the programmers to add them into the specific machine versions - as well as to finish the original LogFlash porting.)

I have spent the last month as Nora's guinea pig for LogFlash 2 PC, which teaches the rafsi. They are somewhat easier to learn than the gismu, once you have gotten the latter up to a consistent 90% recall level. I was able to run 60 word lessons in the same 1 to 1 1/2 hour time I used for 40 word lessons the first time I used LogFlash 1. An improvement to the algorithm, has led to substantially increased recall results - I have reached the 90% recall level on rafsi after only two times through the words. The change is minor - we add an Under Control lesson that feeds you words that would otherwise not be seen until you've completed the entire word list. For people using small lesson sizes, going through the entire word list can take a couple of months, and this partially random selection ensures that you don't forget the words from the beginning by the time you get to the end of the list. The extra lesson takes less than 5 minutes per session - I exceeded 90% retention on these words even on the first time through.

As the only LF2 user, I can offer testimony as to its effectiveness. Moreover, it has an unexpected psychological effect as well. Completing the gismu list with LF1 gives a tremendous sense of achievement and accomplishment - a real feeling that you have learned something. I also obtained insights into the semantics of the words we chose that were not possible before I had learned them all. LF2 gives a totally different feeling - I can only describe it as a feeling of tremendous power. With what I have learned with LF1 and LF2 gives me a Lojban vocabulary that is larger than that of English, and I truly feel that vocabulary at my metaphorical fingertips. I feel creative and unconstrained. While I know only about 20% of the cmavo and not much more of the new grammar, nothing really keeps me from saying practically anything I want in Lojban.

I haven't written a poem in English since 6th grade. I was inspired to Lojban poetry after 2 weeks with LF2. That first Lojban poem, which uses only a couple of lujvo, proved to me what I had already believed to be the case - tanru and lujvo probably make Lojban poetry easier to write than English poetry. That first poem is awful; I wrote it at 3AM in two hours. I've gotten the inspiration for the next one already - and it will be longer and undoubtedly better. Bob LeChevalier writing Lojban poetry - a truly unexpected Whorfian effect.

LogFlash 2 PC will be available as of the beginning of next year at our standard $20 Shareware price, with updates for minimal cost. While we need the money from program sales, I would not recommend using the program until you are close to 90% with LogFlash 1. I have no reports yet that anyone except Nora and myself are ready for LF2.

We have designed LogFlash 3, which will teach the cmavo and much of the simple grammar, but I need to make up a data set. I expect to announce its availability next issue. Again, it is recommended that you substantially finish LF1 before going on to LF2 or LF3.

To make LF2 and LF3 a little more attractive, we will be bundling each with a second program that is useful for learning but does not tie into the LogFlash teaching algorithm. LF2 will eventually have a lujvo-making program that will build lujvo from either user-entered or random tanru. Good practice, and it may prove more useful than a LogFlash lujvo-making teacher. The initial release of LF2 will not have this program.

LF3 will be bundled with an update to Nora's random sentence generator. This program uses the complete grammar to generate sentence at four selectable levels of difficulty, and also uses the complete gismu and cmavo lists. The program is now available, but is only useful to those with a cmavo list. We will include it with orders for the latter when requested. The program is neat, and definitely mind-stretching as you attempt to interpret the sentences it generates. More importantly, it can give you a feel for the grammar, and helps you learn the place structures.

I have a tentative teaching order for the gismu in the textbook. This will eventually be added into LF1 and LF2 in place or in addition to the current random selection of words. The long-promised tape will also be prepared using this order. The minor changes needed to the programs will be made at the same time as the enhancement mentioned above, sometime in the next few months, after LF3 is completed.

The teaching programs are keeping ahead of the students, as we have promised. For those with the right computers, now is the time to get started. When the textbook is done, or when you start a Lojban class, you will find that memorizing words is the critical step that will slow you down. Learn them now, and you will be ahead of the game then.

News from the Institute

In November, JCB sent a 'Bulletin' to a rather scattered mailing list of people.

JCB reports that he is spending his time rewriting L1 to reflect his latest iteration of the language. He has at least one more chapter to rewrite, and is trying to figure out how to index it. He apparently plans a several month review cycle, followed by further revision. Bob McIvor (RAM) is giving him some assistance, producing some teaching programs, and helping rewrite Notebook 3, the proprietary language description JCB published last year. JCB's language version has apparently undergone some considerable changes and/or refinements since NB3 was published.

JCB states that he has enlisted the support of an academic linguist who strongly supports the logical language research effort, which is excellent news. There is talk of a paper. la lojbangirz is offering to review this paper before publication, given the opportunity, to ensure that as many technical challenges to the quality of such a paper are resolved before publication, thus presenting the best possible picture of the logical language community is presented to the academic world We extend our offer to share all of our research results, and to fair and timely review, and reaffirm our commitment to integrity in such cooperation.

Now that we are incorporated for this purpose, we will be contacting linguistic, scientific, and computer publications, the National Science Foundation, as well as other forums that may be interested in logical languages. We will offer our services as a source of considerable expertise and credentials in support of the peer review process in our field. By working within the system and in support of it, we hope to ensure fair consideration of all ideas, including both JCB's and our own.

JCB has asked any who disapprove of our policies to tell us so and to ask us to drop them from our list. Since we don't want to send to people who aren't interested, we encourage this, especially if you tell us why. We have received no such requests for deletion, including any from JCB or his Board, but will respect any that we do receive.

JCB quotes a recent letter from pc; pc and I have agreed to the following response and clarification: JCB misinterprets pc's letter as calling for two mutually competitive and secret research efforts culminating in simultaneous publications. This was not pc's intent, and is not the policy of la lojbangirz. All of our research activities are non-proprietary and public, and will continue to be. We are not directly competing with JCB; we certainly are not stressing academic publication, since we believe that that community will take us more seriously once we have proven the language through usage. We view any real advances JCB makes as advances to our efforts as well. Lojban, and la lojbangirz are the culmination of JCB's efforts, and not a contradiction of them. We would like nothing more than to be working in cooperation with him.

How Complicated is Lojban?

Several correspondents indicate that Lojban appears complicated. This apparent complexity is mostly an illusion, which I'll try to deal with briefly.

There are several possible explanations for the illusion, depending on what feature of the language is involved.

Phonology: Lojban has one of the smaller sets of phonemes among human languages, and all phonemes occur in most of the world's most widely used languages. (Lojban 'x' is not found in English, but is more common in other languages than several other English sounds.) The sound set was defined based on familiar linguistic principles, and is easily explained and taught. Permissible consonant clusters were selected for linguistic harmony and easy resolution by a listener, even in a noisy environment. Vowel sounds are pure, and are maximally different for their number.

The phonology is also audio-visually isomorphic: each sound is represented in print in only one way, and vice versa. Spelling is completely phonetic. With the exception of simple and invariant stress rules, the language ignores intonation. All punctuation is spoken. The phonology and morphology rules allow completely unambiguous resolution of correctly spoken speech of any speed, regardless of whether the words are run together. Lojban phonology is closer to English, and easier to learn for English-speakers, than nearly all other languages.

Print: Lojban uses a subset of the Roman alphabet, with only the apostrophe ' as a mandatory special symbol. Commas and capital letters are permitted when Lojbanizing a non-Lojban name to indicate deviations from standard Lojban pronunciation rules. Punctuation marks are optional, but all are represented by Lojban cmavo and are spoken, except the period, which represents a single 'sound' (a pause or glottal stop). The Lojban character set is of course compatible with standard computer keyboards and screens. The lack of capitalization makes typing in the language easier and faster than English. Only a few letters have pronunciations that are significantly different from their most common English/European pronunciation.

Morphology: Word length, the existence of consonant clusters, stress, and pause together serve to completely an unambiguously define the morphology. Content words, the brivla are easily distinguished from structure and support words, the cmavo. The 1316 gismu, serving as roots, and the few hundred cmavo, are a minimal vocabulary to support conversational fluency, compared to most languages. Most other words in the language are compounds, or lujvo, which are built from a set of resolvable affixes (rafsi), each uniquely representing one of the gismu or a small number of cmavo.

The rafsi were selected for maximum statistical word- building efficiency given the set of gismu, and are chosen from a limited set of letter-patterns found with the gismu. Where more than one rafsi exists for a word, the morphology allows the selection of any of them for use in constructing the lujvo, such that the resulting lujvo meets the morphology rules. rafsi are combined in a strict modifier- modificand relationship. The building of lujvo is sufficiently simple that someone who has memorized the rafsi can create lujvo for new concepts while speaking or writing, with minimal disruption. This minimizes the need for dictionaries, and makes the full power of the language available to a new speaker with minimal memorization compared to any natural language.

Borrowing is permitted, subject to adaptation to the Lojban phonology and morphology - a requirement of nearly all languages. There are fairly simple rules for Lojbanization, and for identifying borrowings. There are conventions that make borrowing even easier, and which make a 'nonce' borrowing more easily recognizable to the listener. These are based on categorizing, which is used in Chinese speech for a similar purpose.

Grammar: Lojban has a completely unambiguous grammar, verified by computer analysis. The grammar is described by a rule set of less than 500 rules in the standard format for such rules, and could be written in less than 400 with some loss of clarity. About half of these rules define permissible compounds of cmavo that have the same grammar as the simpler forms. By comparison, English grammar is estimated to have in excess of 50,000 such rules and is believed undefinable in its entirety. Also by comparison, commonly used computer programming languages have about 200 such rules (C, PROLOG, and PASCAL have been compared); ADA, the new military standard language has about 500 rules and is very close to Lojban's complete grammar in complexity.

In spite of the small number of rules, the count overstates the actual complexity of normal conversational usage, which is expected to use less than half of the rules permitted. (The other rules might be used in writing, especially in complex discourse where elaborations, inferences, and qualifications are commonly used.) Following this section, we present a set of about 120 simple sentences which demonstrates most of the conversational grammar rules. The structure implied by these rules is fairly transparent in translation.

Included in Lojban's grammar is a grammar for incomplete utterances, for continued speech, and for mathematical expressions, all of which are unambiguous. The mathematical grammar is consistent with the grammar of the rest of the language.

Lojban's grammar is a predicate grammar, which is not found in any of the natural languages. It is found in logic and in some computer programming languages and is simple in concept, although very different in its implications for language.

Given all of this data which suggests that Lojban is far simpler than any natural language, and indeed is comparable to computer languages in simplicity, why does the illusion persist.

One reason is that Lojban is different from English, though less so than most natural languages. Anything different seems complex until you know the rules, in which case it becomes simple.

Another reason is that there is a fair amount of initial memorization in our current teaching strategy, an exercise which is unappealing to most people. But far less has to be memorized of Lojban to achieve conversational fluency than any natural language. The average American 6- year old has a greater vocabulary than is needed to speak Lojban. The typical college graduate English-speaker has a vocabulary of 50,000 to over 300,000 words, suggesting that several thousand words per year are learned during our development.

A third reason is that most people don't realize just how complex a language is. The typical foreign language program seldom teaches more than a fraction of what is used by fluent language speakers. English, itself, due to its melting pot tradition and extensive world use, is among the most complex of these languages, but this complexity is transparent to a native English speaker.

The fourth reason is that Lojban text, as shown in translations such as The Lord's Prayer (in JL6 and revised in this issue) appear to be longer in Lojban than in the English. Almost any fairly literal translation will tend to be longer in ANY other language. For something as culturally sensitive as The Lord's Prayer, translating must be literal rather than figurative; paraphrase tends to be culturally offensive in such contexts. As is shown by the translations of the Lojban-original texts below, such texts are longer in English when all the hidden structures and optional omissions which are implied in Lojban are explicitly stated. When re-expressed as English idiom, the text becomes shorter but more ambiguous; Lojban as yet has no true idiom to shorten translations, and any idioms would have to memorized if they did exist.

This leads to the final reason why Lojban appears complex. Most things we have translated have been poetry and short aphorisms. These are among the most ambiguous expressions in English; they depend on ambiguity to be interesting; they rely on idiom to be comprehensible. Many are not even grammatical English. People have chosen them for us to translate primarily because of this complex and ambiguous nature (You don't want to make it too easy for us, do you?) If you write the Lord's Prayer in modern grammatical English sentences, so as to make it clear what every line means, it will probably be even longer than my Lojban translation.

The added length of Lojban, by the way, is relatively insignificant by natural language standards. Professionally translated German has about 1.3 times as many words in English as in the original. Several natural languages have words or sentences, even in idiomatic speech, that are much longer than English.

The other reason why Lojban is longer is of course due to the fact that it IS unambiguous in grammar. There is more information, so it takes more data to communicate it. Lojban's phonology and morphology are close to the minimal limits on redundancy needed to allow communication, so the length is actually quite short for the amount of data involved. You must remember that Lojban is a 'logical language'. Logical does NOT mean short, though it may be simple.

English and other languages leave out considerable amounts of logical structure needed to be unambiguous, information that you cannot omit in Lojban. My comments on Kenneth Clark's letter below include an example of an apparently simple English sentence with a significant amount of hidden logical structure. When you learn to look for the that hidden structure, as Kenneth has apparently been able to pick up just by trying to write a few Lojban sentences, you will quickly see why there are more words in typical Lojban sentences. Kenneth's self-taught progress, without any more material than any of you have received, shows that the language structure is actually quite simple to learn, and is relatively intuitive. He doesn't yet know the cmavo and their grammar, but his grasp of the language structure will allow him to add that information in easily.

One other comparison is relevant for the proponents of 'Anglan', a Lojban with English-based roots. These people would be wise to study the problems of BASIC English. This 'simplified English', dating from over 50 years ago, is sufficiently powerful that the New Testament has been translated into the language. It apparently is limited in dealing with technical literature, however. Most important, it is harder for English speakers to learn than for speakers of other languages. Why? English speakers have to memorize the same list of allowable words, comparable in size with that of Lojban, and then remember that NONE OF THE OTHERS ARE ALLOWED. Remembering what isn't permitted is much harder than remembering what is. Nora and I have a similar problem - we thoroughly learned JCB's gismu list, and our own intermediate lists using LogFlash - and we have had our greatest trouble forgetting the old words, which occasionally crop up to haunt us.