From lojbab@lojban.org Thu Mar 2 03:05:01 2000 X-Digest-Num: 380 Message-ID: <44114.380.2124.959273826@eGroups.com> Date: Thu, 02 Mar 2000 06:05:01 -0500 From: "Bob LeChevalier (lojbab)" Subject: Re: Use and abuse of sets At 01:16 PM 03/01/2000 -0800, Jorge Llambias wrote: > >Your body is a mass of organs > >This is starting to get fuzzy. Are you talking about the >lojban article "mass" or some other concept? I am talking about the Lojban *concept* mass which the gadri invokes. > >but the members of la xorxes are not le rango, > >of course, because {la xorxes} is not {le xadni be la xorxes}. Is your body not named that? %^) la xorxes is not a uniquely referential tag. > >What are the members of lei djacu? > >They are {le djacu}. > > >Are they individual molecules? > >No, they are quantities of water. Cowan put this better than I did - the quantities of water that comprise a mass of water are not a unique division of the mass. So if you say "le djacu" referring to those members, it is ambiguous which members these are. > >But if > >I pour 3 portions of water into a tub, I might well describe that as "lei > >ci djacu", yet it would be impossible to delimit what each of le djacu were > >that comprised the mass. > >Who denies that? They were identifiable before you poured >them, and then they were no longer identifiable. What is >the problem? Well now that you have this mass, how many members does this mass have? > >No, what makes a mass is the loss of individual identity of any > >identifiable components the moment that you start talking of their mass > >properties. > >Obviously we disagree. > > >Or shall we say that any possible manifestation of the mass > >properties is a putative member. > >I wouldn't say that. I don't know how to argue this with you. It is definitional. > >I would contend that there is no "current state of Loglan". > >I would contend you're being a bit unfair with TLI Loglan. >I can't think of any reason why... :) >L3 is a pretty good description of the language, at least of >some parts, and it was published with JCB's approval. I'll admit to not having read L3 closely; I was as unimpressed as others seem to be impressed. > > >One way to start to know is to look at what speakers of > > >the logical language Lojban are using and not using. > > > >Nick isn't saying much these days, and we have no others. > >That is a little insulting to the people who use Lojban >in jbosnu@onelist.com (the list for discussion in Lojban) >as well as to those who write in Lojban from time to time >in this list. I'm not meaning to be insulting. When talking about knowing what the language is or will be, I have my linguist hat on, and look at what fluent speakers do. The only way we will get fluent speakers of Lojban is for people to use Lojban. I believe that when last we discussed it, you did not feel yourself to be fluent, and yet I think most would consider you among the best of the active posters in the language. A lot of people who post in Lojban know only a subset of the language and therefore cannot possibly have the flexibility to use any or all parts of the language equally should the occasion arise when one of the more obscure parts is useful. > >(i.e. the point > >is that there is way too little usage to determine usefulness based on what > >is actually used, especially when talking about things that are intended > >for use in niche contexts). > >I agree we cannot determine anything yet. I only suggested >looking for trends, Whereas I think looking for trends too early is self-limiting; like a pre-election straw poll, doing poorly in a straw poll makes it more likely that you have no chance when the real vote comes because you have already been written out. This is why I am decidedly uninterested in what cmavo you feel are extraneous at this point. It is too soon to tell, and I don't wish to elevate a prediction to reality. > > >And the proposed Lojban way of reading mathematical expressions > > >is hoplessly inadequate for mathematicians and hoplessly > > >complicated for everyday use. > > > >How do we know? > >That it won't be enough for mathematicians can be known >by looking at any mathematical journal and trying to >read the more complicated formulas in Lojban. I helped design MEX and I don't pretend to be fluent enough to do this without grammar and wordlist in hand. But are you saying that you were unable to read such a formula, that the formula reading was ambiguous, that there were insufficient words for the operations? > That it is >complicated for everyday use I know from trying to learn >it and comparing it with the ordinary Lojban grammar. Whereas not having studied it of late, I recall it as being somewhat less complicated than the full grammar of sumti while being quite parallel in structure. > > >My point is that the parts of the grammar that are not used > > >in everyday conversation don't really belong in the language. > > > >None of Lojban is used in everyday conversation. > >Yes it is. For example in jbosnu. You were also present >when it was used during Logfest. I've been subscribed to jbosnu since it was announced. It is a great idea, don't get me wrong. But I have not gotten a digest everyday and I don't think it is a subscription bug. Most of the writings are shorter than the "conversations" you and Chris and Goran were having a couple years back and thus necessarily use a lot less of the language than longer texts would involve. There are after all some things that will not show up unless the text is of sufficient complexity. In a formal writing register these will be necessary. > > > What language > > >has a specialized grammar for technical fields? > > > >Mathematics. Computer languages that make use of natlang keywords (e.g. > >COBOL and the various restricted domain human-computer interfaces)) > >I meant human language, like Lojban. Mathematics is a human language. > I don't mind if someone >decides to use Lojban keywords for a computer language. I just >don't want to incorporate that computer language as part of >the Lojban grammar. I am thinking more of the versions of English that are being implemented into AI interfaces for natural language specialized domains. The moment you get into a specialized domain, there is a tendency to adopt non-standard styles of writing. For example in writing a play in Lojban, the method of identifying the speakers, of discussing stage directions and sets - these will involve significant use of metalinguistic structures or some other means to separate the play from the directions for the play. Yet these structures occur rarely in other parts of the language. Some of the fanciness of text context resetting with ki and multiple ni'o and no'i was invented with an eye to translating the stories within stories within stories of Burton's Arabian Nights. The grammar of such writing will not be the same as that of an IRC chat. Until something like that is written or translated, the potential won't be realized. > Air > >traffic control. > >They use a different grammar? Highly constrained and stylized so that all the non-fluent English speakers never run into something they cannot understand. Closer to a computer language than a conversational language, but potentially ambiguous and because it is spoken between humans it is considered bona fide language, albeit a jargon or cant. > >Probably a lot of others because there is little study of > >the specialization of language within a restricted discipline other than a > >focus on the jargon. > >Maybe you're right. I still think it is odd that to fully >know the grammar of Lojban you have to learn how to use a >huge number of selmaho that practically never come up in >real usage. It may be that no one person will know all of the language. The same is likely of English. 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!)