From a.rosta@lycos.co.uk Sat Sep 07 15:32:05 2002 Return-Path: X-Sender: a.rosta@lycos.co.uk X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-8_1_0_1); 7 Sep 2002 22:32:04 -0000 Received: (qmail 32024 invoked from network); 7 Sep 2002 22:32:04 -0000 Received: from unknown (66.218.66.217) by m5.grp.scd.yahoo.com with QMQP; 7 Sep 2002 22:32:04 -0000 Received: from unknown (HELO mailbox-4.st1.spray.net) (212.78.202.104) by mta2.grp.scd.yahoo.com with SMTP; 7 Sep 2002 22:32:04 -0000 Received: from oemcomputer (host213-121-68-132.surfport24.v21.co.uk [213.121.68.132]) by mailbox-4.st1.spray.net (Postfix) with SMTP id 2F2511C3A2 for ; Sun, 8 Sep 2002 00:32:02 +0200 (DST) To: Subject: RE: [lojban] termsets Date: Sat, 7 Sep 2002 23:33:36 +0100 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) In-Reply-To: X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Importance: Normal From: "And Rosta" X-Yahoo-Group-Post: member; u=122260811 X-Yahoo-Profile: andjamin xorxes: > la and cusku di'e > > >This sort of coordination is very normal > >and unmarked in English. It's a shame that GA is not > >already equivalent to "nu'i GA" (so that all coordination > >is termset coordination), but the result is just averagely > >lojbanically clunky, and not downright unusable. > > I agree that the concept behind termsets makes sense, > but I don't think that its Lojban implementation is just > averagely clunky. At least I find it very difficult > to make it work with the rest of the sentence structure. > > The reason plain GA won't suffice seems to be that GA...GI... > doesn't have a terminator, so {ge ko'a gi ko'e ko'i} would > have {ko'e ko'i} as a termset. I don't think that would be > a bad thing though. You could always recover the present > reading with {ge ko'a gi ko'e vau ko'i}. But I guess that > will have to wait until the deadline ends (there is no danger > of termsets becoming popular in the meantime, so I expect it > will be easy to reform them away). In my own usage I just ignore the official grammar of GA and follow the principle that in GA X GI Y, the syntactic type of Y is determined by X. > But anyway, one trick to avoid termsets is this: > > ko'a dunda ko'e ko'i gi'e co'e ko'o ko'u > ko'a gives ko'e to ko'i and (does) ko'o to ko'u > > I suppose {go'i} won't work there, and I don't know > whether there is something more precise than {co'e}, > but if there isn't there very well could be. I don't like having to use a trick, though. Conceptually, coordination of single sumti ought to be seen as coordination of singleton termsets, since all sumti coordination is essentially an abbreviatory mechanism. > Compare with the equivalent "afterthought" termset form: > > ko'a dunda ko'e ce'e ko'i pe'e je ko'o ce'e ko'u > > which is longer and also requires some forethought for the > first {ce'e}. > > The forethought form with {co'e} is just as long as the > forethought termset form with {nu'i}, if the {nu'u}s can > be elided, but the co'e form is more flexible, so you can > say things like: > > ge ko'a prami ko'e gi ko'i ko'o co'e > > instead of the fixed order required by nu'i: > > nu'i ge ko'a ko'e gi ko'i ko'u prami > > which can also be replicated with co'e as: > > ge ko'a ko'e co'e gi ko'i ko'o prami > > So, my conclusion is that termsets can always be substituted > advantageously by another form. You'd have to find something better than "co'e". Maybe an experimental cmavo in GI that inserts and implicit GOhA. Your examples would then be: ge ko'a prami ko'e gi'ai ko'i ko'o ko'a ge dunda ko'e ko'i gi'ai ko'o ko'u --And.