Return-Path: Received: from SEGATE.SUNET.SE by xiron.pc.helsinki.fi with smtp (Linux Smail3.1.28.1 #1) id m0t69fT-0000ZOC; Fri, 20 Oct 95 07:06 EET Message-Id: Received: from listmail.sunet.se by SEGATE.SUNET.SE (LSMTP for OpenVMS v1.0a) with SMTP id B84D4E5B ; Fri, 20 Oct 1995 6:06:54 +0100 Date: Fri, 20 Oct 1995 01:06:05 -0400 Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: lujvo-making X-To: lojban@cuvmb.cc.columbia.edu To: Veijo Vilva Content-Length: 3264 Lines: 69 Jorge: >And: >> > {nacpoiba'u}. >> Shd that be {nacpoirbahu}? > >That's what I thought some time ago, but it seems it shouldn't. There >should be no stress of the first syllables, so there is no chance of it >breaking appart: nacpoiBA'u CVV rafsi ONLY add the r (or n or l) when they are in initial position, and there is one exception in that a CVVCCV form 2-part lujvo (e.g. fu'ivla) does not need the hyphen. The rule is that if a hyphen is not needed, it is forbidden. >I'm not sure whether using the rafsi "por" instead of "poi" is anygood, >because {nacporBA'u} could be {na cpo-r-ba'u}, or couldn't it? The "r" >is not necessary there, so I'm not sure if {nacporba'u} would break or >not. Because lujvo-space takes precedence over fu'ivla-space, this cannot fall apart: "cporba'u". would be a fu'ivla. But we have to be strict about hyphenation rules specifically to prevent things like this. If you could arbitrarily add an 'r' hyphen after any CVV or CCV non-final rafsi, and a 'y' between any CVC and the following C initial rafsi, then you would expand the potential lujvo space greatly at the expense of fu'ivla space, but you also would make more situations where you would HAVE to put a hyphen for fear of the word breaking up. Under liberal rules that would allow "cporba'u" as a lujvo of cpo+ba'u, then you would need the 'y' hyphen for nacyporba'u. There are only two word forms that require tosmabru-type hyphen inserting. Both involve CVC rafsi at the beginning, AS WELL AS in all positions up to final term CVCCV or to the first hyphen. And of course the first consonant cluster must be a permissible initial. TOS-MABRU has CVC in initial position and CVCCV in final, and the permissible initial cluster, so it can become TO SMABRU which looks like two CCV rafsi. TOS-TOM-Y-MABRU has CVC's up to the first hyphen, and can therefore fall apart into TO STOMY-MABRU, where the Y is taken as the replacement for the final vowel of a CCVCV gismu. Note that these rules must be completely blind as to whether the rafsi or gismu actually have meaning in Lojban - the morphology is independent of, and completely preceding, the association of the lexicon to semantics. All other forms do not break up, because taking the initial CV (or CVV) off does not yield a valid lujvo. The 'r' after a CVV makes the following portion of the word start with an rC cluster, which is never a permissible initial, so it cannot fall off. In two term lujvo, CVV must have an r before another CVV, or it fall apart into two cmavo, but before CCV, it cannot fall apart since CCV is not a standalone word. The 'y' hyphen of course prevents a permissible initial from forming. TOSMABRU constructions are only an issue between CVCs because in CVCCVVXXX you would end up with CV+CCVVXXX which is never a valid lujvo form and CVCCCVXXX, where CV+CCCVXXX is not a valid lujvo form. ---- lojbab lojbab@access.digex.net Bob LeChevalier, President, The Logical Language Group, Inc. 2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273 For the artificial language Loglan/Lojban, see ftp.cs.yale.edu /pub/lojban or see Lojban WWW Server: href="http://xiron.pc.helsinki.fi/lojban/"