From richard@rrbcurnow.freeuk.com Mon Jan 01 14:12:35 2001 Return-Path: X-Sender: richard@rrbcurnow.freeuk.com X-Apparently-To: lojban@egroups.com Received: (EGP: mail-6_3_1_3); 1 Jan 2001 22:12:35 -0000 Received: (qmail 86531 invoked from network); 1 Jan 2001 22:12:34 -0000 Received: from unknown (10.1.10.27) by l7.egroups.com with QMQP; 1 Jan 2001 22:12:34 -0000 Received: from unknown (HELO scrabble.freeuk.net) (212.126.144.6) by mta2 with SMTP; 1 Jan 2001 22:12:34 -0000 Received: from du-009-0075.freeuk.com ([212.126.152.75] helo=rrbcurnow.freeuk.com ident=root) by scrabble.freeuk.net with esmtp (Exim 3.12 #1) id 14DDBm-0001tl-00 for lojban@egroups.com; Mon, 01 Jan 2001 22:12:22 +0000 Received: from richard by rrbcurnow.freeuk.com with local (Exim 2.02 #2) id 14DCpP-00004b-00 for lojban@egroups.com; Mon, 1 Jan 2001 21:49:15 +0000 Date: Mon, 1 Jan 2001 21:49:15 +0000 To: lojban@egroups.com Subject: Re: [lojban] Commas and vowel pairs Message-ID: <20010101214915.A282@rrbcurnow.freeuk.com> Reply-To: Richard Curnow Mail-Followup-To: lojban@egroups.com References: <20001230223056.F292@rrbcurnow.freeuk.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i-nntp In-Reply-To: ; from slobin@ice.ru on Sun, Dec 31, 2000 at 09:49:02PM +0300 From: Richard Curnow On Sun, Dec 31, 2000 at 09:49:02PM +0300, Cyril Slobin wrote: > > As a result of that discussion, my lojban.vim marks all such diphtongs > as invalid anywhere expect cmene. Of cource I can change this behavour, > but I really want to know, what exact rule is. > > -- Following up on my earlier posting and Cyril's reply ... To recap, there is a statement in chapter 3 of the Reference Grammar "Commas are never required: no two Lojban words differ solely because of the presence or placement of a comma." This appears to contradict some statements in chapter 4, e.g. this example 7.11) bang,r,kore,a Korean (the language) 7.12) kuln,r,kore,a Korean (the culture) Note the commas in Examples 7.11 and 7.12, used because ``ea'' is not a valid diphthong in Lojban. Arguably, some form of the native name ``Chosen'' s In an attempt to resolve this conflict in a reasonable way, I have created the table below. For each case where two vowels are adjacent in a word, I have considered the cases of there being a comma between them and of there being no comma between them. The YES/NO validity is based on the best evidence available to me, however in some cases I was so unsure I have marked the entry with a question mark, notes follow below. I would be very keen to find out (soon!) about errors/disputes in this table, as I am about to start on including code to do vowel cluster checks in my revamp of jbofihe's morphology algorithms. Was the validity/invalidity of all these cases pre-planned into the language, and/or is there any "case law" dealing with the more obscure ones? If there are any cases that are considered ambiguous (i.e. subject to possible future ruling), I'd like to know where these are so I avoid an algorithm which makes those cases hard to change. General notes on the table : - the notation [aeo] means any one of the 3 letters a, e or o. So, for example, [eo][aeou] generates 8 pairings (2x4). - gismu are not included as they never contain vowel pairs. - lujvo can only be built from standard rafsi forms. In particular, vowel pairs only occur in CVV rafsi, and these are all of the first 4 forms. - cmavo (except VV) containing a vowel pair are of the form CVV; the vowel pair in all of these is one of the first four. - VV-form cmavo have a special waiver to allow the second block of ten (ui, ia etc) - fu'ivla can never contain the letter y - cmene can specifically include iy and uy. -------------+----------------+-------------------------------------- | | Valid in these types of word | | Vowel pair | Comma e.g. | cmavo VV- fu'ivla cmene | between | lujvo cmavo -------------+----------------+-------------------------------------- ai|au|ei|oi | NO ai | YES YES YES YES | YES a,i | NO?(1) NO?(1) YES YES | | [iu][aeiou] | NO ia | NO YES YES YES | YES i,a | NO NO?(1) YES YES | | a[aeo] \ | NO ae | NO NO NO?(2) YES?(3) [eo][aeou] / | YES a,e | NO NO YES YES | | [iu]y | NO iy | NO NO NO YES | YES i,y | NO NO NO YES | | [aeo]y | NO ay | NO NO NO NO?(4) | YES a,y | NO NO NO YES? | | y[aeiou] | NO ya | NO NO NO NO?(4) | YES y,a | NO NO NO YES? | | yy | NO yy | NO NO NO NO?(4) | YES y,y | NO NO NO YES? -------------+----------------+-------------------------------------- Notes ===== 1. Comma would never be needed as the syllables can't break between vowels. 2. fu'ivla are not allowed to have invalid diphthongs in them? 3. cmene can include any vowel pair (John Cowan's reply cited by Cyril Slobin) 4. .. but y doesn't count as a normal vowel in that sense (the book specifically allows iy and uy but doesn't mention y adjacent to anything else?) -- ---------------------------------------------------------------------- Richard P. Curnow rpc@myself.com Weston-super-Mare United Kingdom http://go.to/richard.curnow/