From lojbab@lojban.org Fri Nov 29 21:09:46 2002 Return-Path: X-Sender: lojbab@lojban.org X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-8_2_3_0); 30 Nov 2002 05:09:46 -0000 Received: (qmail 21461 invoked from network); 30 Nov 2002 05:09:46 -0000 Received: from unknown (66.218.66.216) by m7.grp.scd.yahoo.com with QMQP; 30 Nov 2002 05:09:46 -0000 Received: from unknown (HELO lakemtao04.cox.net) (68.1.17.241) by mta1.grp.scd.yahoo.com with SMTP; 30 Nov 2002 05:09:46 -0000 Received: from lojban.lojban.org ([68.100.206.153]) by lakemtao04.cox.net (InterMail vM.5.01.04.05 201-253-122-122-105-20011231) with ESMTP id <20021130050946.OCKU1248.lakemtao04.cox.net@lojban.lojban.org> for ; Sat, 30 Nov 2002 00:09:46 -0500 Message-Id: <5.1.0.14.0.20021129223811.03a31ec0@pop.east.cox.net> X-Sender: rlechevalier@pop.east.cox.net X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Sat, 30 Nov 2002 00:01:54 -0500 To: lojban@yahoogroups.com Subject: Re: [lojban] Fu'ivla diphthongs was: Official Statement- LLG Board approves new baseline policy In-Reply-To: <0211292053270Q.02982@neofelis> References: <5.1.0.14.0.20021129201628.031287f0@pop.east.cox.net> <5.1.0.14.0.20021129125713.00abb680@pop.east.cox.net> <5.1.0.14.0.20021129201628.031287f0@pop.east.cox.net> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1"; format=flowed Content-Transfer-Encoding: quoted-printable From: Robert LeChevalier X-Yahoo-Group-Post: member; u=1120595 X-Yahoo-Profile: lojbab At 08:53 PM 11/29/02 -0500, Pierre Abbat wrote: >On Friday 29 November 2002 20:18, Robert LeChevalier wrote: > > At 01:41 PM 11/29/02 -0500, Pierre Abbat wrote: > > >what about {cipnrxakuila}? > > > > I dunno? What about it? If someone has said it is invalid, what was t= heir > > reason? > >No one said it's invalid that I know of, but it contains a diphthong that >does not appear in lujvo. > >According to chapter 3: > >The first four diphthongs above (``ai'', ``ei'', ``oi'', and ``au'', the o= nes >with off-glides) are freely used in most types of Lojban words; the ten >following ones are used only as stand-alone words and in Lojbanized names = and >borrowings; and the last two (``iy'' and ``uy'') are used only in Lojbaniz= ed >names. > >Thus a fu'ivla can contain {ua} or {io} or {ui}. Yes. But that does not mean that they can be used in unlimited ways. First I will quote on the alternate orthography as an argument: >On Friday 29 November 2002 13:01, Robert LeChevalier wrote: > > At 08:20 AM 11/29/02 -0500, Pierre Abbat wrote: > > >I don't use the TLI alternate orthography, so when I write {srutio}, I > > > don't mean {sruti'o}. > > > > But if srutio is a valid word, then it has to be usable by those who DO= you > > the alternate orthography. The book actually dealt with this: >=B7 =93i'a=94 through =93i'u=94 and =93u'a=94 through =93u'u=94 are = changed to =93ia=94=20 >through =93iu=94 and =93ua=94 through =93uu=94 in lujvo and cmavo other th= an=20 >attitudinals, but become =93i,a=94 through =93i,u=94 and =93u,a=94 through= =93u,u=94 in=20 >names, fu'ivla, and attitudinal cmavo. sruti'o is a lujvo in the alternate orthography become srutio. Thus we=20 cannot allow srutio in fu'ivla space. ckanku'a would fail slinkui so it is not a valid fu'ivla, but ckankru'a is= =20 a valid fu'ivla. In the alternate orthography, it would have to be written= =20 with commas ckankru,a, so ckankrua would be a distinct word. srutio might be a valid fu'ivla, if not for the alternate=20 orthography. However I personally would reject it as looking too much like= =20 a lujvo (or a typo for a lujvo). I still consider fu'ivla by intention to= =20 be second-class words in Lojban. I want them to be clearly seen to look=20 different, and that word form doesn't. Furthermore, the difficulty of=20 performing the slinkui test means that we shouldn't be trying to push the=20 limits on what fu'ivla are allowed on that basis. >From CLL >All fu'ivla: >::1) must contain a consonant cluster in the first five letters of the= =20 >word; if this consonant cluster is at the beginning, it must either be a=20 >permissible initial consonant pair, or a longer cluster such that each=20 >pair of adjacent consonants in the cluster is a permissible initial=20 >consonant pair: =93spraile=94 is acceptable, but not =93ktraile=94 or =93t= rkaile=94; >2) must end in one or more vowels; >:3) must not be gismu or lujvo, or any combination of cmavo, gismu,=20 >and lujvo; furthermore, a fu'ivla with a CV cmavo joined to the front of=20 >it must not have the form of a lujvo (the so-called =93slinku'i test=94, n= ot=20 >discussed further in this book); >:=93=94:4) cannot contain =93y=94, although they may contain syllabic=20 >pronunciations of Lojban consonants; >:5) like other brivla, are stressed on the penultimate syllable. >:Note that consonant triples or larger clusters that are not at the=20 >beginning of a fu'ivla can be quite flexible, as long as all consonant=20 >pairs are permissible. There is no need to restrict fu'ivla clusters to=20 >permissible initial pairs except at the beginning. (Note that this explicitly says that clusters larger than 3 are permitted=20 inside fu'ivla, to answer another post of yours. This does contradict=20 statements on pg 36 and 37 that says that clusters cannot occur in numbers= =20 larger than 3. It is clear that we did not carefully deal with fu'ivla=20 when Cowan wrote up the phonology rules. THIS is the sort of thing that=20 can be properly addressed by the byfy.) Rule 3 says that they cannot be lujvo or have the form of a lujvo. fu'ivla= =20 word space is defined as that brivla space which is left over when we=20 remove the forms reserved for other words individually or in any legitimate= =20 combination in the speech stream. >So {srutio} (a fu'ivla, >discarded in favor of {strutione} for "ostrich") is distinct from {sruti'o= } >(a lujvo: "penumbra", maybe?), and {ckankua} (a fu'ivla: "skunk") is disti= nct >from {ckanku'a} (a valsrslinku'i). > >As to the TLI alternate orthography, {i'o} in a fu'ivla is written as {i,o= }, >but in a lujvo {i'o} is written as {io}. Yes, and sruti'o is a lujvo, so how is it written? > > I don't pretend to be more a master of fu'ivla than others. So far as = I > > know, however, the rafsi fu'ivla like that one all work. > >By work, do you mean that you can make words like {cipnrxakuilykanla}? I meant "Type 3 fu'ivla", not "rafsi fu'ivla". When we wrote up the word resolution algorithm (which hadn't been formally= =20 proven, and that is the only reason it did not make it into the refgrammar,= =20 but it was intended that it end up in the baseline documentation), we tried= =20 allowing "iy" as a hyphen for fu'ivla to be made into rafsi, but it simply= =20 made the algorithm too complicated, so we now use zei: cipnrxakuila zei=20 kanla is a "lujvo" in principle (it is not a tanru in that it has a single= =20 specific meaning). The difference between zei and iy was insignificant in= =20 speech - they are both one syllable and no pause is required. In writing=20 you may need to write the space, but no one has tested to find out. The bottom line in cmene and fu'ivla is that we chose NOT to try to=20 maximize the use of the available space (which might have allowed ala'um=20 and srutio) because it is more important to minimize words being added in=20 error that could cause problems later. It is easier to make a blanket rule= =20 forbidding "la" than to come up with a SIMPLE set of rules that allows=20 someone to know when it is or is not allowed. It is easier to tell people= =20 to make Type 3 fu'ivla than to come up with valid Type 4s in the absence of= =20 a properly defined test to verify words (the word resolution algorthm was=20 not a valid test, since it would only detect a slinkui violation if the=20 word in question was preceded by a cmavo in the text. We need two things. A valid word-resolution algorithm that can be proven=20 (and which is verified to fit what was published in the refgrammar, since=20 we put it aside long before the refgrammar was published), and a word=20 tester that will classify all words allowing for ALL context situations=20 (thereby always detecting slinkui violations and alternate orthography=20 conflicts). If we have those, then it becomes useful to talk about how=20 closely to allow fu'ivla to encroach on the other word spaces. http://www.lojban.org/files/software/BRKWORDS.TXT which as you can see was INTENDED to be part of the baseline, needs to be=20 completed to fully discuss this matter. (the new policy does not define it= =20 to be part of the baseline, unless the byfy chooses to adopt something like= =20 this as a blanket resolution of the numerous glitches in the=20 phonology/morphology section like the question of 4 or more=20 consonants. But that requires that someone make it current and prove=20 it. Anyone who wants to undertake this is welcome to it. lojbab --=20 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