From lojbab@lojban.org Fri Feb 18 22:07:00 2000 X-Digest-Num: 368 Message-ID: <44114.368.2021.959273826@eGroups.com> Date: Sat, 19 Feb 2000 01:07:00 -0500 From: "Bob LeChevalier (lojbab)" Subject: Re: Dr. James Cooke Brown At 08:03 PM 02/18/2000 -0600, Rex F. May wrote: >Just out of curiosity, was any consideration ever >given, in either project, to replacing i and u with >y and w when they are to act as semivowels? It was considered as part of the morphology redesign the first weekend that we started on Lojban, and rejected as being insufficiently vital a change when we were trying for minimal change from TLI Loglan and no questioning of JCB's design decisions. We considered a lot of things, but the bottom line is that we felt that the goal was to change as little as possible and yet still have a "from first principles" implementation of Loglan that was on that algorithmic not subject to copyright. We changed the set of permissible initials and medials to make things linguistically symmetric (and hence more algorithmic) - GMR had replaced the original symmetry by an irregular collection of permissibles, based on "taste tests" that we knew were biased by English speakers and reportedly by controversial statistical analysis (RAM was the person who had raised the controversy and is more qualified than I am to discuss it). We added the apostrophe, and the corresponding close-comma for the syllable-dividing reasons I stated in another post. Otherwise we avoided any change to the morphology. We were not making a new or improved Loglan at that point, we were making a Loglan that reflected the existing design and the best linguistic knowledge we had as to how to implement that design in 1987. We required systematization, but not improvement. Several months later, we found an ambiguity bug in JCB's complex making algorithm, which he and I had written down in its then current form when I visited him in May 1986. I have read nothing to indicate that TLI has ever corrected that bug in their version - presumably RAM knows. (I'll provide details separately if anyone cares). Fixing that bug was the most significant change we made to the morphology after the start. We rejected and reversed the late changes JCB had made to the borrowings morphology as inconsistent with what had gone before - these are the changes made for the "international vocabulary of science" which include things like doubled "r" that are not found elsewhere in the language which violate audiovisual isomorphism (clearly vocalic r is spelled differently in borrowings than in other word types). The proposal to use y and w surfaced on occasion later, but always we found it to be one notch short of being a necessary change, and one which would make re-coalescing the two languages more difficult later (and such reunification was a major goal until JCB rejected our last attempt to negotiate after the court case was decided). More importantly for such a proposal was that, other than for reunification, language stability was considered to be the utmost priority. JCB never understood how many potential Loglan learners had turned away because something changed just as they had started learning, but that was the most common statement I received from old timers when I was trying to rebuild the community even before the split occurred. I gained support from many old-timers after the split by committing us to ending the change cycles as early as possible. As such, "improvement" was never really a factor in design decisions. The goal was to fix only things that were "broken", and to do so in ways that we believe a non-antagonistic JCB would find acceptable and consistent with the original design. "Broken" was always defined in terms of things JCB had himself written. After around 1990, almost all changes to Lojban were expansions to increase functionality in new areas explored by usage, expansions that did not force any significant amount of relearning to the prior language. lojbab ---- lojbab lojbab@lojban.org Bob LeChevalier, President, The Logical Language Group, Inc. 2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273 Artificial language Loglan/Lojban: http://www.lojban.org (newly updated!)