Date: Sun, 31 Jul 1994 23:56:07 -0400 From: Logical Language Group Message-Id: <199408010356.AA28629@access1.digex.net> To: ucleaar@ucl.ac.uk Subject: Re: current cmene project Cc: lojbab@access.digex.net, lojban@cuvmb.cc.columbia.edu Status: RO X-From-Space-Date: Mon Aug 1 00:35:45 1994 X-From-Space-Address: lojbab UC> I think there's a fundamental difference between cmene and fuhivla, UC> whether they're nonce or conventional. The difference between nonce and UC> conventional concerns the interlocutors' knowledge of the word, while the UC> difference between cmene and fuhivla concerns the way the semantics of UC> these words work: fuhivla have definitions, while cmene only have UC> definitions in that 'la X' means 'lo thing-named-"X"'. Any conceivable UC> cmene already exists in Lojban, since Lojban fully specifies the meaning UC> of the cmene, whereas a nonce fuhivla is a genuine innovation in the UC> language, because of the novel meaning. 1. The definition of nonce/conventional also concerns the extent to which a word is established in the Lojban community, which, rather than any dictionary must eventually come to be the standard for what "is" Lojban (at least if what we are doing has any linguistic merit. We aren't even fixing the gismu semantics in concrete, thoughh they are a lot more solid than fu'ivla definmtions at this point. Thus conventional can only refer to that which has been adopted by the community, rather than by individuals. Nonce words might be made up and even spread to a part of the community before being rejected/superseded by other's ideas - they are still "nonce" in that they have not been accepted at-large, but they are no longer merely tied to a single interlocutor. 2. I want as many decisions about semantics to be made by Lojban speakers who are actually USING the language. All of us, even Nick and I as the best speakers/writers in the language, are still primarily thinking in some other language other than Lojban. We are not yet qwualified to make decisions as "Lojban speakers", and the language is still being prescribed from outside. 3. I think you have a fundamental misconception of the intended semantics of fu'ivla as distinct from cmevla. First of all, the kinds of cmevla we have been talking about ARE fu'ivla, just not fu'ivla brivla. My writings about fu'ivla have alwatys talked about type 1,2,3,4 etc. fu'ivla. Type 1 is other-language quoted text Type 2 is cmevla Type 3 is the kind of fu'ivla we have been talking about with classification tags Type 4 is the kind of fu'ivla that don't have tags, of which there are none yet defined in the language (the cultural gismu are the closest to these) but they will have the choice of the largest area of fu'ivla space. They just cannot reliably be made up ad hoc, and we will NOT make one up and verify its validity unless there is enough usage to demand it (though I may come up with a few examples for the dictionary merely as EXAMPLES, probably from among the non-gismu culture names). The idea is that most fu'ivla will, as you have more-or-less imploied, be referents to very specific concrete things. I expect most fu'ivla to serve the roles of "nouns" in Lojban, rather than as main selbri. The types of words that will rate Type IV status will be those that are usefull in the full range of gramamtical positions of a brivla, AND are frequent enough to make the Zifean shortening justifiable. These will also, probably be those that have come to have the broadest (or loosest) semantics of the fu'ivla, since frequency o use itslef will tend to cause some divergence in meaning as people's concepts of word meaning differ, and to some extent the differences between what in English are different parts of speech tends to cause some amount of meaning modification (for example, English table becomes in Lojban "jubme", which also has a sense of "table" as a verb - we have used it around here in such a capacity, such as when someone eats in the living room with "lo stizu cu jubme gi'o se cpana le sanmi". If we do adopt your argument that a cmevla is just "lo cmene be da bei de then it is even more clear that the semantics of cmene are not far removed from any other place of any other brivla. And when we have "me" to turn any cmene (or any other sumti) into its corresponding brivla, it takes more than hy hypothetical argument to justuify using the more difficult to mange fu'ivla space so lightly. lojbab