By Robert LeChevalier, lojbab@lojban.org
Lojban is the current viable version of the artificial language
Loglan, supported by The Logical Language Group, Inc., a non-profit
organization. Lojban is a public-domain Loglan, with 300-1500
supporters depending on how you count, about 300 people claiming to
be learning the language, and over a hundred having completed a
course and/or otherwise demonstrating skill in the language.
The inventor of Loglan, Dr. James Cooke Brown (JCB) continued
work on a separate version of Loglan, supported by his organization,
The Loglan Institute (TLI). While he does not advertise it as such,
much of his language design is considered proprietary, and TLI
claims copyright, royalty and publication rights in his version of
the language, and requires trade secret agreements to get complete
language details. To keep his control over the language in the face
of 'competition' from Lojban, few supporters of 'Institute Loglan'
are given direct access to others, and there is little spontaneous
activity going on. There are by latest report 110 supporters, many
of whom are supporting both versions of the language. Most of these
stem from a largely unsuccessful advertising campaign in 1988 when
JCB published a revision of his book on the language. There are no
known speakers of TLI Loglan, and only a couple of people try to
write in the language (and all writings have to be vetted by JCB
before publication). JCB is now (2000) in the process of retiring as
leader of his organization, and there is no evidence that his
handpicked replacements command the authority and respect needed to
sustain the effort without him. While I am obviously biased, I
firmly believe that the TLI version of Loglan is dead and going
nowhere. In any event, the proprietary claims on the language give
it little attraction to most potential users of the language.
The two versions of the language are close, but drifting slowly
apart, with Lojban being a superset of JCB's version with a
completely substituted vocabulary list (derived algorithmically in
the same manner as the original). Both versions are substantially
different from the earlier published versions of the language, with
significant known flaws remaining in JCB's version (detailed review
available via postal mail or on the Planned Languages Server).
More details follow:
Loglan is based on predicate logic, and was started in 1955 for the
purpose of testing the Sapir-Whorf hypothesis. Other linguistic
research and education applications have been added since, along
with computer and AI applications associated with its unambiguous
grammar and morphology (flawed in the case of JCB's version.)
Loglan was first publicly announced in the June 1960 Scientific
American. JCB had not completed the language design, and did not for
several years to come. However, several science fiction authors took
note, most notably Robert Heinlein, who referred to the language in
his novel 'Moon is a Harsh Mistress', and Robert Rimmer, who made
several mentions of the language in various utopian novels, most
especially 'Love Me Tomorrow. JCB himself wrote a utopian science
fiction novel, 'The Troika Incident', in which Loglan figures
prominently.
In 1965 and 1968, limited editions of a book called 'Loglan 1' (L1)
were published, the latter in microfilm. After several years outside
the country, JCB republished L1 and a dictionary called Loglan 4/5
(L4/5) in 1974-5, and incorporated The Loglan Institute, Inc. (TLI).
Over the next few years a Loglan journal 'The Loglanist' (TL) was
started edited by Dr. John Parks-Clifford, (pc) Professor of Logic
at University of Missouri at St. Louis. About 2000 copies of the
books were sold, with about 250 purchasers remaining active for the
next year or two. Of these perhaps half a dozen temporarily learned
the language well enough to carry on conversation with direct
assistance from JCB. Meanwhile, JCB sought grants for research from
the National Science Foundation and was turned down 3 times, at
least partially for irregularities in the proposals as compared to
NSF standards.
The influx of new people led to many new ideas and the detection of
several major problems in the language design. Computer technology,
notably that of 'YACC' the UNIX compiler development tool, was
applied to Loglan, and the supposedly unambiguous language was found
flawed in this aspect. Meanwhile in response to criticism, a more
robust way of making compound words that comprise the bulk of the
Loglan vocabulary was proposed. The twin redevelopments of the
grammar and the morphology paralyzed the language for several years,
during which JCB and TLI relied on Loglan supporters to financially
keep the project afloat. Shortly after the redevelopment was
complete, in the 1982-4 era, there was a big power struggle between
JCB and the TLI Board of Directors over control of the language and
the organization. JCB won because there was seen to be no hope of
completing the language without him, but the bulk of the supporting
membership and the most active volunteers dropped out. The new
language design was incompletely documented and there was no
in-language activity; old Loglanists forgot much of what they had
learned. 20% of the root vocabulary and 100% of the compound
vocabulary was changed, and there were many changes to the grammar.
The journal TL suspended publication, as pc dropped out.
JCB virtually alone set about rewriting the language definition, but
progress was infinitesimal. In 1986, I (Bob LeChevalier) volunteered
to work on a dictionary update, and tried to stimulate new volunteer
activity. I was successful, and even got some of the old-timers
involved again, but JCB saw this activity as a threat. Nora Tansky
(now my wife) and I completed an MS-DOS based flash-card vocabulary
program. JCB insisted that we sign a copyright acknowledgement and
agree to pay royalties, claiming copyright on the individual words
of the language. Legal threats followed, and the mass of Loglan
volunteers decided to reinvent a public domain vocabulary and
grammar while seeking a negotiated solution. Negotiations failed,
and the legal threats culminated in a trademark registration battle,
which our organization won in 1992, with a ruling that 'Loglan' was
a generic name. Loglan/Lojban design was completed, and we have an
active if cash-poor organization with language students on 5
continents.
To sum up, the underlying dispute is over whether an artificial
language can hope to succeed if its inventor tries to control its
development too closely, whether intellectual property claims are
valid given that they are being raised after people had contributed
time and money for years without such claims, and whether people
will support a language when information is kept secret from them.
This is a historically recurring battle in the history of artificial
languages. To the extent any AL has succeeded, it has been through
openness and letting the language supporters control the language.
I will be happy to respond to questions on any of the above. Details
and documentation are enormous as I maintain as complete an archive
as possible.
Copyright, 1991, 2000 by the Logical Language Group, Inc. 2904 Beau Lane, Fairfax VA 22031-1303 USA Phone (703) 385-0273 lojban@lojban.org
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.