From cbmvax!uunet!CUVMA.BITNET!LOJBAN Fri Feb 28 17:41:27 1992 Return-Path: Received: by snark.thyrsus.com (/\==/\ Smail3.1.21.1 #21.19) id ; Fri, 28 Feb 92 17:41 EST Received: by cbmvax.cbm.commodore.com (5.57/UUCP-Project/Commodore 2/8/91) id AA14884; Fri, 28 Feb 92 17:02:27 EST Received: from cunixf.cc.columbia.edu by relay1.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA19871; Fri, 28 Feb 92 16:50:34 -0500 Received: from cuvmb.cc.columbia.edu by cunixf.cc.columbia.edu (5.59/FCB) id AA02456; Fri, 28 Feb 92 16:49:40 EST Message-Id: <9202282149.AA02456@cunixf.cc.columbia.edu> Received: from CUVMB.COLUMBIA.EDU by CUVMB.COLUMBIA.EDU (IBM VM SMTP R1.2.1) with BSMTP id 7607; Fri, 28 Feb 92 16:48:09 EST Received: by CUVMB (Mailer R2.07) id 1932; Fri, 28 Feb 92 16:47:24 EST Date: Fri, 28 Feb 1992 12:56:42 PST Reply-To: cbmvax!uunet!pyramid.com!cuvmb.cc.columbia.edu!fschulz Sender: Lojban list From: cbmvax!uunet!PYRAMID.COM!cuvmb.cc.columbia.edu!fschulz Subject: word morphology X-To: lojban list To: John Cowan , Eric Raymond , Eric Tiedemann Status: RO I read the lojban morphology document. Very difficult. To aid in my understanding, I contructed a simpler morphology that I hope to use as a stepping stone to the lojban morphology. First cut: --- .V .VV .V'V CV CVV CV'V are cmavo Note that when strings like this are run together in connected speech or writing that unique word resolution is possible. Finally I really understant why the pause appears in the cmavo list at the front of some words. Note that stress is not needed for word resolution at this stage. Second cut: --- .V .VV .V'V CV CVV CV'V are cmavo --- CCV are gismu, string together for lujvo. Note that these CCV will always combine without creating new consonant clusters. The hyphens "n" "r" "y" are not needed. Lojban has 48 initial CC clusters giving 48 * 5 = 240 gismu. This is short of the 1400 needed. Note that the word class, cmavo or gismu can be determined by looking at the first two characters of the word, needing no lookahead buffering. Stress is not needed for word resolution. Third cut: --- .V .VV .V'V CV CVV CV'V are cmavo --- CCV are gismu, string together for lujvo. --- yC[CV]*. are le'avla --- yV[CV]*. are cmene le'avla start with a consonant to match the CCV gismu. cmene start with a vowel to distinguish from le'avla. Word resolution is still possible without using stress. The first two characters determine which of the four word classes a word belongs to. At this point I believe I have all the word classes that are in lojban and all the resolution properties that lojban has. I am not sure if lojban allows pauses in cmene or le'avla so this might be a difference. Does anyone see any lojban word resolution capabilities this simple morphology is missing? Have I missed covering some morphological class? Also lojban cmene may not have "la" and other stuff. I do not understand the reason for the restriction. Does this alternate morphology remove the restriction, with reducing the resolution capability? -- Frank Schulz ( fschulz@pyramid.com )