From lojbab@lojban.org Wed Mar 1 05:58:02 2000 X-Digest-Num: 380 Message-ID: <44114.380.2103.959273826@eGroups.com> Date: Wed, 01 Mar 2000 08:58:02 -0500 From: "Bob LeChevalier (lojbab)" Subject: Re: Use and abuse of sets At 08:34 AM 02/29/2000 -0800, Jorge Llambias wrote: >Yes, that's what I read. The arguments were presented >by JCB himself, so probably biased towards his position, >but even so, I don't see the benefit of the set articles. >Every time they are used in Lojban they can either be >substituted by masses without loss of clarity, but only by bastardizing the mass concept. Example: your "chief of mass" - masses being amorphous and not having members, it doesn't seem to make sense that they can have a chief member. > or they >should be substituted by masses because they're not used correctly. That is a job for better teaching. I admit that we have a ways to go here. > When you start to learn the language you imagine >that since they're there you have to use them, and that's >how they get misused. I agree that this is not the reason to use a cmavo. > >I believe I looked at what he said and found that he had gone back > >to what we now convey as "ro lo" (their lea), but I may be recalling > >incorrectly. > >I think you are recalling incorrectly. Their sets now >correspond exactly to our masses, except perhaps for the >default quantifier. Their leu = our {lei}, their lea = >our {piro loi}. This is explained in lesson 11 of L3. I'll look again, but I am basing my argument on the original discussions that brought these into the language, and not so much on JCB's last word, which resulted in dissent. The original examples that justified including lea (which was etymologically a contraction of "le ra" BTW) were not mass examples. I also suspect that you and I disagree as to what "piro loi broda" means anyway. > >Prior to that they had no equivalent to "lo" > >They still don't, they use "ne" (Lojban {pa}). It is not clear to me that their indefinites are veridical. >But even Lojban {lo} is not strictly necessary, since >{lo broda} is always replaceable by {su'o broda}, and >inner quantifiers with {lo} are practically useless. I think it depends on the context. But the point of arguing inner quantifiers was to acknowledge that they had default values for each of the articles, justifying their parallel grammatical structure. That the value is usually uninteresting does not mean that it doesn't exist. > >(their "lo" is > >our "loi", except when it is our "lei") > >I disagree. We don't really have anything quite like >their "lo", which is not really very well defined anyway. It has been well-defined several times, with some degree of self-contradiction. I base my claim on the classic definition. All the stuff that has happened in recent years has been JCB trying to work out the impossible contradiction of that classic definition. The result is that the definition that was clear (and corresponded to our loi, except for one example) is less well-defined. >It has several different translations to Lojban. >They use it for our "observatives", so they would >say "Lo fagro!" for "Fire!", whereas we just say {fagri} >plus perhaps an attitudinal. This one fits loi. Our observatives might also have used loi, but I think it weakens the predicate structure of the grammar to focus on sumti rather than on relationships when expressing an observative. When Lojbanists say "klama", they are observing a going, and not a goer. When they say "fagri" they are observing an instantiation of the predication and not the x1, which is what a "lo" would imply. By contrast, a command would seem to inherently put the commanded one into the relationship in some way, and this justified "ko". >They use it for "I'm waiting for a taxi", or "I need a box", >and things like that, that have to do with reduced scope and >opaque contexts and such, for which we don't have a final >determination in Lojban, either {tu'a} or, the one I prefer, >{lo'e}. If I had to choose a one word translation for >Loglan "lo" it would be {lo'e}, but it doesn't really >have anything to do with averages. You seem to be observing the muddle that resulted from their trying to solve problems and figure out how to use the concepts in recent years without having anyone able to point out the problems. The closedness of the TLI community in the last 10 years silenced people who could have pointed out that they were muddling what had been clear if not much used concepts. > >We split out the triple descriptors based on differences in usage within > >Loglan, so there somewhere was discussion supporting each of the > >interpretations of "set". JCB at last decided to stop contradicting > >himself, and chose the most useful given that TLI did not have the triple > >descriptors. > >I like the final state of the Loglan articles better than >the Lojban one. Fortunately the Lojban system can evolve >painlessly and naturally to the Loglan one if it proves >to be better. It isn't and it won't. > >TLI Loglan has no cmavo for most of the needed MEX structures. > >Aren't they lucky! > >What do you mean by "needed"? They are not used in Lojban yet, >so it must be a theoretical need. It is a need specified in JCB's vision of the language. He actually had a MEX in 1962, but threw it out when it a) proved less than adequate and b) would not YACC. Of course it is a theoretical need. Much of Loglan is purely theoretical. We cannot know what speakers of a logical language will need. But the fact is that mathematicians do talk about mathematics and read off mathematical expressions, and the current English way of reading mathematical expressions is grammatically ambiguous. > The recent translation we >did about the hunger problem had lots of numbers and statistics >(I thought it was excelent as a test of Lojban's abilities for >handling numbers, which is rather tricky) and we did not use >a single MEX structure. Where is this supposed need coming from? JCB "Four score and seven years ago" was one phrase that required MEX consideration. > >We > >considered it part of fulfilling the JCB commitment to have a MEX solution, MEX = "mathematical expressions" The problem is the expression in language of that peculiar symbology that we call mathematical notation. Most of MEX may have little use outside of that domain, but it was a problem that JCB identified early on as being of interest, was of early interest to Loglanists (I think it was also discussed in the first issue of The Loglanist, but surely one of the first few; the first assistant editor, Mengarini, considered MEX the most interesting problem). Not all of the language is used to handle problems that occur in everyday English, or any natlang conversation. Loglan was always intended to go beyond that in potential. (How often to we use some of the obscure logical connectives, yet these seem quintessential Loglan?) lojbab ---- 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 (newly updated!)