Sketches toward a history of the logical language project.
(All possible caveats and disclaimers apply: at this point this is done from memory, without additional checks, on material that is either hearsay or based on interested personal experience. This is also be barest fragment of a skeleton on which to hang all the meat of history that we can dig up. It will eventually also flesh out the timeline to which it is attached. Comments, additions, corrections, documentations and whatever else are welcomed at or on the lojban-list.)

James Cooke Brown, Assistant Professor of Sociology at University of Florida, Gainesville, decides to test the Sapir-Whorf Hypothesis. Brown, a social psychologist with a degree from U/Minnesota, was well trained by second generation Logical Positivists (e.g. May Broadbeck) and so thought to use something like a logically perfect language, that is the First Order Predicate Logic, for the test.

The details of the progression here are hazy, but during this period most of the basic guidelines of the language were developed as well as some of the details: the shape of predicates, names, and little words, little word vocabulary and some predicate vocabulary.
Speech stream uniquely segmented into words, word stream uniquely parsed into sentences (from the language of logic).
Metaphysically neutral (for the SWH).
Grammatically simple, for easy learning (practical).
Vocabulary derived from major languages for easy learning (practical)
The technique for deriving predicates was also at least outlined and some predicates derived by hand.
Sometime in this period, Brown invented the game of Careers, which was marketed by Milton Bradley Co. and eventually provided Brown with a decent living, independent of the usual academic jobs.

June: “Loglan” in Scientific America
There was a flurry of responses to this article from a range of people: linguists, philosophers, logicians, artificial language buffs. The mail was largely unanswered, though a few letters, e.g. from Quine, were saved for later use. A later examination of those letters shows that many people who later played some role in Loglan were already inquiring about it (pc sent from RAND Corp. asking about the use of Loglan as an intermediate language in machine translation, for example).

Brown had an NSF Grant to cover the computer time to develop the primitive predicates according to his algorithm (with some handwork at the end). This resulted in the first edition of Loglan 4 and 5, the dictionaries between Loglan and English, work done with his wife Lujoye Fuller Brown.
There was apparently a group of people around Brown who worked on the language: learning it, using it, discussing it and revisions. It is not clear who all was involved, though some names have turned up in various contexts. For them, at some time, was written the first draft of
Loglan 1, the general description of the language. Also during this time or shortly thereafter a longer version of L1, the only finished versions of L2, the explanation of the language, and L3 a (very restricted) primer were written and “published” by University Microfilms.

A dark period. Brown’s (second?) marriage ended in divorce and he fled with his infant daughter (?) to Europe, first England and eventually Ibiza. Apparently various people managed to track him down in these places and discuss Loglan with him. He worked on it some, eventually producing the (third edition of) L1 and getting it published just prior to his return to the US.
In the US before his return (exact date unclear) some friends of Brown set up an organization, The Loglan Institute, to deal with Loglan related matters. This was not a non-profit organization and was, in fact, a DBA for Brown, though his friends were listed as officers of the organization. (I am unsure whether the organization was even incorporated at that time.)
Sometime in here JCB wrote The Troika Incident, which Lyman Sargeant called the best utopian novel of the period. In it Loglan is talked about under the name Panlan, but the samples, e.g. ai mi betgo, are recognizable as the Loglan of the period. The novel also contains a description of another of JCB’s projects, the Jobs Market (seen by some as a practical application of Careers), on which he worked for many years.
Loglan is mentioned by name in Heinlein’s The Moon is a Harsh Mistress but there are no samples, indeed no evidence that Heinlein knew anything about it other than the name.

Loglan 1 was announced with fanfare in an ad in the August 1975 Scientific American. A couple hundred orders for the book were submitted, but the addresses given for the orders was not quite correct and so there was some delay in shipping. The address for contacting Brown on the back of the book was also incorrect and so the considerable body of mail that resulted was delayed and largely went unanswered.
Brown eventually came to the US to take over, settling eventually in San Diego. There he began teaching the language to several interested people: Bob Jenner from York, ME, first of all, John Parks-Clifford at the end of 1976, and other soon thereafter, forming a group in San Diego.
In the fall of 1975, Brown visited Parks-Clifford in St. Louis (a letter had gotten through to him and he had replied, with a dialog ensuing). In the course of the visit or soon thereafter, pc was asked to edit a journal of Loglan and accepted.
The first issue of The Loglanist (la loglentan) bore the nominal date of 1976 (when the editing was mostly done) but actually appeared early the next year.
The first issue was largely pc and Brown though there were already some reader- supplied pieces, with responses from JCB and pc.

In The Loglanist there was a considerable amount of discussion of a sort (and indeed on topics) that are familiar from the Lojban List and wiki of the present. In some cases these convinced JCB to make changes in the language, mainly adding new words or, in one case, a new letter, h, and then reconstruct a few words as a result. There were also a number of translations of short passages. The changes were summed up in a special issue of The Loglanist in the fall of 1980.

During this period interested people met in informal get-togethers, Logfests, at various places in New England, usually in July (pc was in Maine for estivation, Jenner and several other participants lived in ME, MA or close enough). Some proposals, both about the language and about the Institute arose from these meetings.

Along with the discussion of the issues there developed of course people who held different opinions about how things should be done from those that were the official choice. For the most part this was confined to constantly raising the issues, though at least one regular discussant also set up his own alternate language (not directly in competition with Loglan, however) and another was (more or less jokingly) accused of setting up a competitor that did everything backwards of Loglan, so Nalgol.

Three larger issues had come in for discussion and had moved JCB into action during this period: unambiguous grammar, creating compound words, and the organization of The Institute. The grammar had always been said to be unambiguous, but that claimed had never been proven and the first few attempts to prove it showed it not to be true as matters stood. This was partly due to the fact that the grammar given was not well formalized beyond the pieces that were directly borrowed from the (demonstrably unambiguous) grammar of the language of logic. JCB, guided by Jeff Prothero and others, set about to write a formal grammar. The target was to get a parser by using YACC, since that would guarantee the language was LALR1 and so (from other considerations) unambiguous. This project continued for several years, with various successes being announced along the way but each failing to be a complete grammar.
The issue of compound predicates, built up from the primitive predicates of the language became pressing as more text was developed and new words were needed. Loglan 1 had left the method of forming such words up in the air and, as a result, several techniques had been used, mainly just ramming (what were taken to be significant or easily identifiable) parts of two (or more) primitives together in a way that seemed likely to be recognizable. So, in the example above, betgo for an English speaker in the context of use pretty clearly means, “go to bed” and is derived from betpu, “bed,” and gotso, “go.” Other cases – particularly those by speakers of languages other than English – were less clear to the usual audience. And some cases were just perverse: the Jenny method, for example. And people were also coining words from non-Loglan sources but also fitting as closely as possible into the CVCCV and CCVCV formats. JCB became convinced that this could not continue and so reworked the whole morphology of the predicates, giving many of them special reduced forms to be used in constructing compounds and insisting that all compounds be made using these forms where possible and full forms where not. This project took a while, being announced finally in 1983.
JCB wanted to make the Institute tax-exempt and most of the participants in the Loglanist wanted it to be a membership organization. In 1979, then, the Institute was reorganized as a membership organization and tax-exempt status as a research organization was applied for. The first election by the members of the board of the organization was held in 1980. Once the Institute became a membership organization, the board urged JCB to copyright (or whatever turned out to be the appropriate protection) the material: name, lists, examples and so on. He refused to do so at the time (though he did eventually do a part of it).

The Loglanist was slow to appear usually and so a new publication was started to carry intermediate news, including social events, personal information, and business of the Institute that needed membership input. It soon came to also include some discussion of issues.
In 1980 there was the first general election for officers of the Institute, JCB still as President but the rest of the board being drawn from the members. In 1982, for the second election, JCB stepped down as President but took the office, Chairman of the Board – ultimate control but freedom from day-to-day matters; pc was President.
The third election was 1984. For whatever reasons (apparent loss of control?) JCB decided to return to a Board of cronies, but was late in getting his slate out and did not inform the rest of the board that he was preparing one. The board cobbled together a slate drawn entirely from members and presented it simultaneously with JCB presenting his slate. JCB persuaded the membership slate to withdraw and declared that the slate had been presented too late under the bylaws. His slate was elected. He then attacked the people who had put together the other slate in a variety of ways (though never actually accusing them of betrayal). They quit the organization.

1984 –
JCB turned to completing the Great Morphological Revolution – already pretty much done – of getting all primitives equipped with affixes to build compound word and of fixing the form for such compounds as well as some rules about borrowings. He also sought legal protection for all of the elements of the language: the YACC grammar, the whole vocabulary and many of the other programs, which were used (not Logflash, a computerized version of his flashcard routine, which had been created by Nora Tansky and never given to the Institute).

Sometime around 1986, Robert LeChevalier came to aid JCB with computer related projects, tidying up the YACC grammar and computerizing the very haphazard membership and book-order files of the Institute. When LeChevalier moved to the Washington DC area, he took some of the membership data with him to continue trying to bring some order out of it. In Washington LeChevalier started a local Loglan group, which produced some discussion, and eventually a newsletter, which was sent to all the Loglanists in LeChevalier’s database (The Loglanist was still being published with a new editor and joined with the old Institute newsletter to offer a more regular – though less meaty – periodical). Though the members of the Washington group were for the most part also member of the Institute, JCB took the position that the group was unauthorized and was trespassing on Institute business (though the Institute had no other local groups, even in San Diego). He further held that mere membership in the Institute did not entitle a person to make use Loglan material – all under copyright – and so even citing Loglan words was forbidden as were definitely writing Loglan sentences or citing bits of the grammar.

In response to this as well as the accumulated suggested additions and “corrections” to the Institute’s language, the Washington group – with some remote adherents – decided to create a new version of Loglan from scratch. The outlines were thrashed out in a series of Logfests at LeChevalier’s house in Fairfax VA; the details – the derivation of new primitives from a revised list of the most widely used languages, the reorganization of the phonology and morphology, the revision of the grammar were done by LeChevalier and John Cowan with a variety of helpers in one area and another. The result was launched in 1987 as “Lojban, a realization of Loglan.” JCB fairly quickly filed a copyright infringement suit, which went on for several years, finally resulting in the withdrawal of copyright from “Loglan.” Meanwhile, although the newsletters from the groups eventually ceased, discussions continued with ever-increasing ease on the Internet. Finally, the creation process culminated in the publication of The Complete Lojban Language (CLL) by John Cowan in 1997. By this time, the group had been formalized as a tax-exempt educational and research organization, the Logical Language Group, and the Logfest had become the annual meeting of the Group, regularly electing LeChevalier (now “lojbab”) as President and the active people present as the Board.

Of course, publication of an official guide to the language did not stop discussion of what the language should be, starting with disputes about what the official text actually meant and then what it should have meant – or said. However, the guide did open the way for a number of large-scale translations: xorxes did Alice and The Little Prince and a number of other works, including game scripts were made. There was also a serious and growing amount of spontaneous communication in Lojban over IRC, generating another large corpus of usage. In 2002, the accumulated discussion including especially the practical problems that turned up in translation and chat work, led to the formation of BPFK (baupla fuzykamni – the committee responsible for a language plan) to amend CLL to remove unclarities or uncertainties, to fill gaps and, if need be, to alter bits to deal better with perceived problems. This committee is meant to be almost the last step in fixing the language, setting up a baseline (which CLL and word lists had already largely done) so that users would not have to worry about the language changing (as Loglan has done without stop since its inception) out from under them. This work has proceeded with increasing speed – though occasional snags – since, under the overall leadership of Robin Lee Powell, who was also elected President of the Group at the 2002 meeting and has provided a number of Internet services, fore mostly the wiki, for the Group over the years.