From LOJBAN%CUVMB.BITNET@uga.cc.uga.edu Sat Mar 6 23:00:03 2010 Received: from uga.cc.uga.edu by MINERVA.CIS.YALE.EDU via SMTP; Fri, 4 Dec 1992 04:07:45 -0500 Received: from UGA.CC.UGA.EDU by uga.cc.uga.edu (IBM VM SMTP V2R2) with BSMTP id 0424; Fri, 04 Dec 92 04:04:17 EST Received: from UGA.BITNET by UGA.CC.UGA.EDU (Mailer R2.08 PTF008) with BSMTP id 9781; Fri, 04 Dec 92 04:04:17 EST Date: Fri, 4 Dec 1992 08:55:10 GMT Reply-To: I.Alexander.bra0125@oasis.icl.co.uk Sender: Lojban list From: I.Alexander.bra0125@OASIS.ICL.CO.UK Subject: TECH RE: The Distribution Problem: An Ambiguity? X-To: lojban@cuvmb.cc.columbia.edu To: Erik Rauch Status: RO X-Status: X-From-Space-Date: Fri Dec 4 08:55:10 1992 X-From-Space-Address: @uga.cc.uga.edu:LOJBAN@CUVMB.BITNET Message-ID: Thinking about Colin's recent messages has prompted some more observations. > ... it is perverse to allow > cmalu je nixli ckule > to mean [small school for girls], because it is very hard > to find a meaning to attach to the constituent "cmalu je nixli" > that allow[s] this. Yes, I agree. We can take it to extremes by using an analogue of the selgadri example. nanla ja nixli ckule is the same as nanla bo ckule je nixli bo ckule for what I believe are essentially the same reasons as the selgadri situation. I've already shown the derivation in set theoretical terms. It's been more difficult finding an intuitive explanation which gets to the nub of the issue, but try di'e. I think the distinction comes from the point of view you take of the situation being described, su'anai whether you describe a set in terms of its subsets, or conversely describe subsets in terms of their parent set. It's not too surprising that this should result in different (in some sense, dual) expressions. ro verba cu nanla ba'eja nixli This describes {[lu'a] lo'i verba} in terms of the defining predicates (characteristic functions) of two constituent parts, whereas ro nanla cu verba .i ba'eje ro nixli cu verba describes {lo'i nanla} and {lo'i nixli} in terms of the common superset {lo'i verba}. I haven't worked out the {ckule} example in detail, but I suspect it happens because we are describing {le ckule} in terms of {lo'i ve ckule}, considered as the union of two subsets. And I imagine most other such situations could yield to a similar analysis. So, even if you were to arbitrarily specify that tanru modification may distribute over logical connectives, as one of the possible interpretations of tanru ambiguity, you would then have to live with nanla ja nixli ckule expanding to nanla bo ckule ja nixli bo ckule (by distributivity), or to nanla bo ckule je nixli bo ckule (by straightforward interpretation), which I certainly don't find attractive. That still leaves the question whether we want some mechanism to _abbreviate_ broda bo brodu je brode bo brodu as we would in NL. I certainly think it would be useful. doi djan. kau,n. your {xai} proposals are fine if you can make them work, but my instincts tell me that you're going to have difficulty defining the semantics rigorously (although I haven't had time to do a detailed analysis), which is why I suggested the me la fy.py. solution. (I think Colin's exposition probably contains the main point of the problem, even if you were to invent a modified-by-the-{xai} {je} which made the details different.) (ta'o doi kolin. I couldn't help feeling while reading your response to protin@com.usl that you'd got your seltanru and tertanru the wrong way round. I thought that the seltanru came first and modified the tertanru which was second, whereas you seem to have them vice versa.) mi'e .i,n. .alegZANdr.