Return-Path: From: cbmvax!uunet!ee.mu.OZ.AU!nsn Message-Id: <9107300438.AA11734@munagin.ee.mu.OZ.AU> To: lojban-list@snark.thyrsus.com Cc: nsn@ee.mu.OZ.AU Subject: Where have all the theorists gone? Organisation: Department of Electrical Engineering, University of Melbourne Smiley-Convention: %^) Date: Tue, 30 Jul 91 14:38:40 +1000 Status: RO X-From-Space-Date: Tue Jul 30 03:08:33 1991 X-From-Space-Address: cbmvax!uunet!ee.mu.OZ.AU!nsn Oops, let's try that again. I'd hate to think that the deluge of lojbanistani here recently has stifled discussion on the more language-theoretical side; on the other hand, it is to be asserted that experience validates theory, in AL as in anywhere else (apart from real linguistics, of course %^). I found this in particular in dealing with my examples for BAI selma'o: the theory for semantic interpretation I had constructed collapsed readily, to be nebulously replaced by something less neat. Something similar is happening with my experience of lujvo construction, and my attempts to apply diklujvo to place structure (a topic which, one might well note, Jim Carter had not expanded on in his diklujvo paper). What this means is that the time for formalising such aspects of the language really isn't yet upon us, although some quite useful guidelines and trends can and should be noted in any teaching of the language. To get some kerfuffle back into the list, I propose a nice and vague topic to the more theory-minded: Metonymy. How should lojban handle it? Need we flag it explicitly? Will a LAhE word do the trick? In flagging it, are we flying against the face (or whatever that idiom is) of the NL trends of categorisation we have been brought up with? Will this make Lojban highly deviant sigmatically, and need we allow such deviance to take root? Indeed, given the degree of creolisation likely, can we expect to enforce such a policing of metonymy? (To say nothing of "true" metaphor). Someone out there please convince me there are more than five people active on this list. Yrs, Nick.