From @uga.cc.uga.edu:LOJBAN@CUVMB.BITNET Fri Dec 4 13:21:42 1992 Received: from uga.cc.uga.edu by MINERVA.CIS.YALE.EDU via SMTP; Fri, 4 Dec 1992 08:35:32 -0500 Received: from UGA.CC.UGA.EDU by uga.cc.uga.edu (IBM VM SMTP V2R2) with BSMTP id 1130; Fri, 04 Dec 92 08:31:59 EST Received: from UGA.BITNET by UGA.CC.UGA.EDU (Mailer R2.08 PTF008) with BSMTP id 6267; Fri, 04 Dec 92 08:31:58 EST Date: Fri, 4 Dec 1992 13:21:42 +0000 Reply-To: CJ FINE <91909372@bradford.ac.uk> Sender: Lojban list From: CJ FINE <91909372@BRADFORD.AC.UK> Subject: Distribution problem X-To: Lojban list To: Erik Rauch Status: RO X-Status: Message-ID: First, some minor corrections on my previous posting. 1) As Iain spotted, I got seltanru and tertanru the wrong way round. 2) The point about selgadri, though correct, is irrelevant. The reason you cannot expand a connection inside a selgadri to connected sumti is nothing to do with jeks: it applies just as much with giheks: lo nixli gi'a nanla cu broda (1) is not expansibile to lo nixli .a lo nanla cu broda. (2) This is because the former is characterising one (possibly multiple) object, while the second is characterising two independent ones. In explicit terms, (1) is approximately su'oda poi nixli gi'a nanla zo'u da broda while (2) is approximately so'ada poi nixli ku'o so'ade poi nanla ku'o zo'u da .a de broda There happen to be some weak implications among pairs of statements of this sort, ((2) implies (1) but not the reverse, for example) but these depend both on the particular logical connectives and on the quantification. Thus selgadri are not particularly relevant to the issue. 3) My formal semantic account of kazytanru is slightly flawed, because it is clear to me that the semantic domain of a tanru is not strictly a restriction of that of the tertanru (got it right this time!) Thus I am clear that labno prenu is a reasonable tanru for 'werwolf' (also prenu labno), even though I deny that lo labno prenu cu prenu Thus, more accurately, I think that the semantic domain of, say, 'prenu' is surrounded by a kind of penumbra of person-like things (lo prenu simsa), and a tanru such as "labno prenu" identifies a restriction on this larger set (which restriction may lie entirely outside the original set, as in this case). I strongly doubt that it is possible to characterise this penumbra set either in general, or even for particular brivla: until you think of "labno prenu" and the like, it might not occur to you that such things lie close enough to "prenu" that a tanru fizo prenu could pick them out. I do not believe that any of these points affects my argument. Concerning Iain's point about the similar problems of distributing out of a connected selgadri and out of a seltanru: I suspect that they are a different order of problem nonetheless. I have not thought about it in detail, but I suspect that all the impliectaions among connected selgadri and connected sumti can be mapped considering only the connectives (logjonma'o?) and the quantification, whereas I am certain that in the case of tanru, the kazytanru (nature of the tanru modification) is significant. Short of dikyjvo (which I do not subscribe to) this will always be the case. I think there may well be a case for explicitly distributive jeks - as somebody (Iain?) said, a sort of abbreviation cmalu ckule gi'e melbi ckule (This is one interpretation of "cmalu je melbi ckule"). However, I'm dubious, because again, the semantics of the kazytanru in the two cases need not be the same). In answer to And(?) about werwolf - I would claim that "labno prenu", "prenu labno" and "labno joi prenu" (either way round) are all perfectly good expressions (NB everybody - the last is not a tanru! Word encompassing them all please?), the last being somewhat more precise than the others. Colin