From LOJBAN%CUVMB.BITNET@UBVM.CC.BUFFALO.EDU Sun Feb 11 03:41:17 1996 Received: from VMS.DC.LSOFT.COM (vms.dc.lsoft.com [205.186.43.2]) by locke.ccil.org (8.6.9/8.6.10) with ESMTP id DAA29354 for ; Sun, 11 Feb 1996 03:41:11 -0500 Message-Id: <199602110841.DAA29354@locke.ccil.org> Received: from PEACH.EASE.LSOFT.COM (205.186.43.4) by VMS.DC.LSOFT.COM (LSMTP for OpenVMS v1.0a) with SMTP id C1F26547 ; Sun, 11 Feb 1996 3:09:02 -0500 Date: Sun, 11 Feb 1996 03:07:31 -0500 Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: Re: X-To: sbelknap@UIC.EDU X-cc: lojban@cuvmb.cc.columbia.edu To: John Cowan Status: OR X-Mozilla-Status: 0011 Content-Length: 7899 >Don't know what a "sei" metalinguistic phrase is. Do you mean the rafsi for > or , the cmavo for self-directed? no that is se'i SEI takes a selbri, with possibly leading terms before it (trailing terms MUST be attached with be/bei - this greatly cuts down on the need for the elidable terminator SEhU coreresponding to SEI). This in effect allows a person to make any metalinguistic bridi attach to any of the grammatical structures to which free-modifiers can attach (a couple of dozen of these in cluding almost all structures witha right-terminator). Since you can express a fairly elaborate bridi using the subset of the grammar permitted, this allows all manner of things metalinguistic to be identified or discussed, including in this case, fuzzy truth values. All you need is a selbri/brivla that has the right places and you can cover ANY fuzzy logic convention you can think up, rather than locking us in to specific ones that may or may not be sufficient. The cost is a couple of syllables for a very precise construct, and no where near 15 syllables. >>Why must people insist on being so bloody metaphorical. what is wrong >>with "nalsatci" as the critical modifer of the defining tanru. > >Been there, done that, and rejected it. Unless I misinferred a design >principle of the language. I thought that the lojban designers were intent >on doubling the number of necessary gismu by using the "antonymal >language-design buddy system": > > :: for weight > :: for length in longest dimension > :: for length in the second longest dimension > :: for length in shortest dimension No. These words were added largely because JCB had them, and JCB in turn had them because he based his gismu list in part on word/concept frequencies in Helen Eaton's book. In a few cases such words were added simply to complete a paradigm that was largely complete already - thus if we had had 5 of those last 6 words in the 3 length pairs, we would have added the 6th rather than taken away the other two opposites. We also make conisderabel analysis for usage in tanru and lujvo making. Both words and their opposites are heavily used in natlangs for acompound building. That this does not apply to satci can be seen in English at minimum, by the fact that we say light/heavy and not light/unlight long/short and not long/unshort thick/thin andnot thick/unthick But we use exact/inexact - a scalar negation form just like Lojban nalsatci would be. I really don't see in any event why you "rejected it", unless you see a meaning to nalsatci that is other than what you intend. >Seems wasn't good enough. No doubt they did this so that they >could fill gismu space with culturally neutral debris and so they could >later complain about there being too many gismu. :-) Insulting even with the smiley. When me made the gismu, there was no particular philsophy about how many gismu there should be, and in any case we were making gismu to pressure JCB into reforming TLI and not to really create a separate language. That came about AFTER the gismu were done and JCB made clear that no meaningful negotiations were posssible. The too-many-gismu faction started to develop right at the end of the gismu-making period, and did not culminate until we baselined the list more than 3 years later, then had 2 dozen new proposed gismuy by the following year. Most of which were at least questionable in terms of real usefulness (being mostly terms that in many cases would be fu'ivla if they were not basic to some cultures enough to argue them as "primitive" - cultures that we do not yet have much audience in %^) >I searched high and >low for an antonymal buddy for but came up empty. As the gismu >creating is semifrozen, I searched for alternatives. There is NO requirement that an "antonymal buddy" be a gismu. lujvo are perfectly good words - really they are! If nalsatci or the scalar antonym of satci has the right meaning, then it is an appropriate basis for "fuzziness". Certainly FAR more appropriate than the figurative kerfa and mursi references. >Besides, don't you mean ? (Has a rather doubleplusungood >quality to it.) Why? What is the double plus ungood quality? Length? No. If the "ka" were necessary to the metaphor (and to me this might be the distinction between "fuzzy" and "fuzziness", then it would be kamselsatci, staying one syllable longer than the corresponding English word, which is probably at least average if not better than average for less-used concepts. But I was sayingh that nalsatci would be the critical part of the metaphor, and not necessarily the complete on. Fuzzy truth values could be as "long winded as "nalsatci ninje'u" (still only one syllable longer than the English, BTW). the point is that the concept can be clearly expressed, and I dare say that these tanru/lujvo would be FAR more understandable than lojrfuzi. We are NOT trying to borrow all of the words of science and technology from English, and there are many people who would revolt at any major effort to do so EVEN IF that is what is largely happening in natlangs. Lojban is more like Chinese and German in attitudes towards borrowing, we prefer internal compounds that may be loan translations, and the Germans will prefer a 10 syllable superword to a simple English borrowing. >>"ckilyjetnu"? nalsatcyninjetnu? You don;t need to >>invoke fur or twilight or chocolate confections to convey the concept. >>Sheesh! > >Or apparently talk about fuzzy logic at all, as you have determined that >fuzzy logic is not important! Slow down, Logical Bob :-) I have not determined any such thing. I have said merely that there is no evidence in natlangs thatfuzzy concepts are more important to normal expression than we have provided for in the existing language. We are not trying to change normal linguistic use in areas where we do not have to, hence fuzziness is at best an option in the language. If we can express that option with at least as much specificity simplicity as natlangs, we are doing fine. I have no doubt that our tools for fuzziness are far MORE powerful than the ones natlangs have, even if the aren't up to "convenient and brief" forms that might be needed if people wanted to put fuzziness into a large portion of their everyday speech. But until people are doing the latter and finding the language too unZipfeanly verbose to convey the needed brief ideas, there is no NECESSITY to come up with shorter forms - only aesthetetic reasons. But if people DID find they needed a gismu for fuzziness because of Zipfean problems with nalsatcyninje'u or the more terse but looser meaning of nalsatci, then there are grouinds to consider a new gismu. But the usage is not there in natlangs, nor yet in Lojban. Indeed the fact that we have gotten by in the Lojban net corpus of the last 3 years with so little need for new concept words suggests that there really will be little Zipfean pressure for more gismu in a longer term study like the post-baseline period. >The situation regarding fuzziness appears unsatisfactory. To whom for what purpose. I wil concede that seems to you to be unsatisfactory for your purposes. >Next: Why fuzziness matters. I do not contest that the idea is interesting to logicians and some scientists, but whether the precise version of fuzziness that some scientists use has major import LINGUISTICALLY is what is in question. I have yet to read what Mccawley says in his book about fuzzy logic, but if he says it is relatively unimportant under current theories, then you might need a fair amount of supporting literature to contradict him. lojbab