From nicholas@uci.edu Tue Sep 18 22:31:52 2001 Return-Path: X-Sender: nicholas@uci.edu X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-7_3_2_2); 19 Sep 2001 05:31:51 -0000 Received: (qmail 38135 invoked from network); 19 Sep 2001 05:31:51 -0000 Received: from unknown (10.1.10.27) by l9.egroups.com with QMQP; 19 Sep 2001 05:31:51 -0000 Received: from unknown (HELO e4e.oac.uci.edu) (128.200.222.10) by mta2 with SMTP; 19 Sep 2001 05:31:51 -0000 Received: from localhost (nicholas@localhost) by e4e.oac.uci.edu (8.9.3/8.9.3) with ESMTP id WAA28839; Tue, 18 Sep 2001 22:31:51 -0700 (PDT) X-Authentication-Warning: e4e.oac.uci.edu: nicholas owned process doing -bs Date: Tue, 18 Sep 2001 22:31:51 -0700 (PDT) X-Sender: To: Cc: Nick NICHOLAS Subject: Re: [lojban] noxemol ce'u Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII From: Nick NICHOLAS I tuned out of this discussion, and now that I have caught sight of it, I wish I hadn't. I'm getting this annoying feeling that, just as historically bound-ka has been taken to be a property, and free-ka a quality, so here to bound-ni has been taken to be a property, and free-ni a quantity. So: {le pixra cu cenba le ni ce'u blanu} = {le pixra cu cenba le ka ce'u blanu sela'u makau} But conversely: {li 25 ni glare vi la melbn.} The discussion in the refgrammar certainly privileges the latter (your ni1); you don't measure properties directly in a colorimeter, but quantities. I mean, you don't get any more explicit than: "Semantically, a sumti with ``le ni'' is a number". So the official definition of ni is your ni1. And since a number is *not* intersubstitutable with a proposition, so ni1 and ni2 are irreconcilable (ni and ka should not properly be intersubstitutable in any context), we can do two things: 1) Say sentences where ni2 arises (as bound-ni) are wrong, and that you shouldn't say {le pixra cu cenba le ni ce'u blanu [kei]} at all, but {le pixra cu cenba leka leni ce'u blanu cu barda}; 2) (Messier, but I think far more desirable): do type-coercion: say that you're using ni to talk about a ka, and that you're doing it kind of elliptically, but without formally marking that ellipticality. So {le pixra cu cenba le ni ce'u blanu [kei]} would be called a convenient shorthand for saying {le pixra cu cenba leka leni ce'u blanu cu barda}. The type-coercion comes in in that clearly a number makes no sense as the x2 of cenba, as you rightly point out, so you behind-the-scenes turn it into something that does make sense. This really means {tu'a le ni...}, of course, and there's no precedent for type-coercion in Lojban: if you mean tu'a, you say tu'a. But I'd rather make an exception for {ni} than break the baseline --- which, sorrily, is arguably broken for {ni} just as it is for {ka}. We've had the Palace Coup for {ka}; let's not have one for {ni} as well. I assure you, noone really wants to say {le pixra cu cenba tu'a leni ce'u blanu}, and noone wants to give up {leni} evaluating to a number. Boy, am I glad I didn't include ni in the Lessons! I haven't been tuned in, so I may have misstepped here; but surely we don't want to start saying that a mother is the difference between Dubya and Chelsea. People aren't differences. Properties are differences. "The difference between Dubya and Chelsea is the fact that Dubya's mother is Babs, and Chelsea's mother is Hillary." Sounds to me like {la dubias. frica la tcelsis. leka makau mamta ce'u}. I see why you might want things to be differences or variations, so you can put {ni} in there with a free conscience; but that sounds mighty muddled to me. I mean, what is the difference extensionally, {la xilyris. fa'u la barbaras.}? Yuck. (Who does that mean I've just allianced to, And or xorxes? :-) Nick, trying to be all things to all people. -- == == == == == == == == == == == == == == == == == == == Nick Nicholas, Breathing I REJECT {gumri} nicholas@uci.edu (Lojban Wiki, Resurrected Gismu)