Return-Path: Message-Id: From: cowan (John Cowan) Subject: Re: This Chemical Element Stuff To: lojban-list Date: Thu, 2 May 91 12:14:04 EDT In-Reply-To: <9105020340.9263@mullian.ee.mu.OZ.AU>; from "mullian.ee.mu.OZ.AU!nsn" at May 02, 91 1:40 pm X-Mailer: ELM [version 2.2 PL13] Status: RO X-From-Space-Date: Thu May 2 12:15:07 1991 X-From-Space-Address: cowan la nitcion. cusku di'e: > The fact is that le'avla are a > massive pain, but we haven't realised it yet, because we've gismu'd the > ones we wanna talk about. The only argument left for gismuing these things > is frequency of use, and use in metaphors in source languages is a (very poor) > heuristic to find this out. If you want the argument of frequency for gismu- > ification, state it. And by the time you get to chem elems, you'll find your > heuristic is pretty poor. {nikle} indeed! In an earlier phase of the Loglan Project, the "Eaton project" was just such an attempt. "Eaton" is a large but rather outdated study (from the 1920's, I think) which purported to provide information on frequencies of concepts in various Euro languages, a kind of sorted multilingual thesaurus. An attempt was made to guarantee that everything in the top N concepts was representable by tanru-ing the then-current Loglan gismu set. Again, the gismu list is a rag-bag. There is no theory behind it considered as a whole. Theory only applies when we consider incremental changes to the list >now<: we add a gismu to complete a semantic set, or to provide a concept for which there is no good tanru and from which new good tanru can be built. > Hm. In my 36 hour experience of Grammar, I see Fillmore's case grammar dissed > a lot. Is it really that worthless, bearing in mind that we are still in > a prescriptive, rather than descriptive phase of the language? The last 10 years in the case grammar field have been spent picking Fillmore apart. The consensus around here is that there isn't enough that's stable enough to serve as a comprehensive guide, so we are stuck with our idiosyncratic place structures. > co'omi'e mi. You are you, huh? -- cowan@snark.thyrsus.com ...!uunet!cbmvax!snark!cowan e'osai ko sarji la lojban