From LOJBAN%CUVMB.BITNET@UBVM.CC.BUFFALO.EDU Thu Feb 8 09:08:16 1996 Received: from VMS.DC.LSOFT.COM (vms.dc.lsoft.com [205.186.43.2]) by locke.ccil.org (8.6.9/8.6.10) with ESMTP id JAA00854 for ; Thu, 8 Feb 1996 09:08:13 -0500 Message-Id: <199602081408.JAA00854@locke.ccil.org> Received: from PEACH.EASE.LSOFT.COM (205.186.43.4) by VMS.DC.LSOFT.COM (LSMTP for OpenVMS v1.0a) with SMTP id 18174C16 ; Thu, 8 Feb 1996 8:37:07 -0500 Date: Thu, 8 Feb 1996 07:48:09 +0000 Reply-To: ucleaar Sender: Lojban list From: ucleaar Subject: Re: brain fart metaphor X-To: lojban@cuvmb.cc.columbia.edu To: John Cowan Status: OR X-Mozilla-Status: 0011 Content-Length: 3639 Bojblab: > >Metaphor belongs to the extralinguistic domain of pragmatics. The > >grammar can (by definition) specify only what is intralinguistic. > >> My position, though, is that tanru are already metaphors, and don't, > >> therefore, require {pe'a} markers. > >Nothing requires {pea} markers. > Whaddaya mean "nothing"? > The place structure of a "broda brode" tanru is that of brode, with all > the places of broda implied using be/bei. Correct. But the nature of the relationship between these places becomes vague. Take the recently cited {rokci cinfo} - there is still an x2, but whatever fills this x2 needn't be in relationship "species of" to what fills x1; instead, they could be in relationship "in same universe as", or something equally uninformative and general. > If you wish any place structure OTHER than this, you need pe'a (you may > be able to use zi'o for some simple place deletions, of course, though I > might still mark it with pe'a). This has nothing to do with {pea}. If it is indeed possible to change the place structure (other than by zio) please tell me where in the refgrammar this is stated; I do not think it is possible. {pea} is a *****METALINGUISTIC***** indicator of figurativeness. > If you use a "pe'a te kafke" then you should have a "pe'a" either on the > kafke, or on a marker for the x3 place (more wordy, though). No. For reasons I have explained, it is not the business of any language design to say "you should have a {pea}". Maybe a manual of good usage could recommend it; but that's something that even I think should wait until a really active and proficient body of users comes into being. > >> Meanwhile, can anyone come up with a selbri for "metaphor"? I'd like > >> to move some of this discussion *back* to lojban, but it's tough when > >> I can't say "Tanru are metaphors" for lack of the darn word. > >Some ghastly attempts must exist in the archives, for Jorge & I both > >coined uncompelling lujvo for "metaphor". > tanru are a specific TYPE of metaphor - a "binary metaphor" having a > specific meaning. I'm not sure that tanru really are metaphors. Metaphors involve *resemblance* between signans/vehicle and signatum/tenor. Tanru don't necessarily. Furthermore, it is rather tricky to say what the vehicle of a tanru (i.e. what one might loosely and improperly call "the literal meaning") really is. > If you want a more generic word for metaphor, you need to define > the word first. Is it any figurative expression? Is it an expression > that suggests connotations? Is it a cultural or literary reference? > All of these can be definitions of metaphor in English, but which do > you want to express in Lojban? One realizes of course that the first step to lujvo making is to choose the appropriate ingredients. But sometimes that is very difficult. How about (i) figurative expression, (ii) sign based on resemblance {smisni}, (iii) figurative expression based on resemblance. What is "figurative expression"? Something not "asserted" but said only as the basis for some other idea to be inferred? "unasserted basis for inference"? What's the lujvo for "infer"? (i) should be something + smu. Then maybe (iii) could end up as something involving smi+sni+smu - how lyrical! > There will uindoubtedly be several words in Lojban for the several > meanings, and if you want a loose blanket word, you will be stuck with > a mediocre word for it - like maybe "meaning-suggest-ke-word-group" That's not "suggest" in the sense of {stidi}. How would you render that as lujvo? coo, mie and