From LOJBAN@CUVMB.CC.COLUMBIA.EDU Sat Mar 6 22:53:44 2010 Return-Path: Delivered-To: veion@XIRON.PC.HELSINKI.FI Received: (qmail 29726 invoked from network); 23 Sep 1997 14:20:58 -0000 Received: from segate.sunet.se (192.36.125.6) by xiron.pc.helsinki.fi with SMTP; 23 Sep 1997 14:20:58 -0000 Received: from segate.sunet.se by SEGATE.SUNET.SE (LSMTP for OpenVMS v1.1a) with SMTP id <2.0A90C53D@SEGATE.SUNET.SE>; Tue, 23 Sep 1997 16:20:16 +0100 Date: Tue, 23 Sep 1997 15:00:37 GMT+0 Reply-To: And Rosta Sender: Lojban list From: And Rosta Organization: University of Central Lancashire Subject: Re: RV: na'e entails na? To: LOJBAN@cuvmb.cc.columbia.edu Content-Length: 2889 Lines: 79 Message-ID: Jorge Joaquin says: > la djan cusku di'e > > >Consider the following sentence: what is its truth value? > > > >1) li vo na'e sumji li re li re > > FALSE TRUE. But (1) is likely to pragmatically implicate a further proposition, "na sumji", which is FALSE. > >On the "na'e entails na" view, this means "4 is not the sum of 2 and 2" > >and is false. On the other (and/djan) view, it means "4 is a non-sum > >of 2 and 2" and is true, since 4 is a product (which is not a sum) > >of 2 and 2. > > No, this can't be right. If the non-sum is true, then not only has there to > be > another relationship, but the sum must be false. Otherwise na'e becomes > quite useless. For a given set of arguments, there are always any number > of relationships that hold among them, so that with your definition, for > any broda, {na'e broda} will be a tautology with any argument set. True, but you have to think of "na`e" as tanru-like. In a sentence like "da xekri gerku", "xekri" adds absolutely nothing to the truth-conditional meaning, because the relationship between "xekri" and "gerku" is unspecified. However, the sentence pragmatically implicates some further proposition where "xekri" does make a contribution. It might also help to think of the meaning when you go one step further than a tanru and create a lujvo. Even you would surely agree that "li vo narsumji li re li re" (I mean "na`e zei sumji") is TRUE. > Let's say {zmana'u} means "x1 is a positive number". Then > > li vo na'e zmana'u > 4 is non-positive > > According to you that is a true statement, since 4 is, for example, an > even number, so it is true that it is something other than positive, > besides being positive. I don't think that makes sense. {na'e broda} > does say that a relationship other than {broda} holds, but first it must > say that {broda} doesn't. > > With your definition, all of these are true: > > ro da cu na'e blabi > Everything is non-white. (Even the whitest of things.) > > ro remna cu na'e remna > Every human is a non-human. > (Since every human is, for example, a selmamta.) > > If John loves Mary, then it is true that > > la djan na'e prami la meris > John non-loves Mary > > because he is also looking at her. > > Could you give a sentence with your definition of na'e as > a selbri modifier that says something useful? You have to consider the role of pragmatics. In effect, na`e adds next to nothing to the logical information encoded in the sentence (just like tanru, as I pointed out above). But nonetheless na`e is communicatively useful. Further, I take it that it is possible to combine or conjoin na`e and na, if that is required. There is also a serious logical downside to na`e entailing na, which both you and I have pointed out - the scope interactions. --And