From @uga.cc.uga.edu:LOJBAN@CUVMB.BITNET Mon Nov 30 12:06:40 1992 Received: from uga.cc.uga.edu by MINERVA.CIS.YALE.EDU via SMTP; Mon, 30 Nov 1992 17:09:47 -0500 Received: from UGA.CC.UGA.EDU by uga.cc.uga.edu (IBM VM SMTP V2R2) with BSMTP id 4183; Mon, 30 Nov 92 17:06:24 EST Received: from UGA.BITNET by UGA.CC.UGA.EDU (Mailer R2.08 PTF008) with BSMTP id 0204; Mon, 30 Nov 92 17:06:23 EST Date: Mon, 30 Nov 1992 17:06:40 -0500 Reply-To: "Mark E. Shoulson" Sender: Lojban list From: "Mark E. Shoulson" Subject: The Distribution Problem: An Ambiguity? X-To: lojban@cuvmb.cc.columbia.edu To: Erik Rauch In-Reply-To: John Cowan's message of Mon, 30 Nov 1992 12:44:24 -0500 Status: RO X-Status: Message-ID: I don't know about historical grounds, but from my perspective of Lojban it seems I must disagree with you, John. I agree that the current state of affairs, leaving it ambiguous, is less than acceptable: tanruic ambiguity does not include grouping ambiguity within the tanru. However, it seems to me that the default grouping for logical and non-logical connectives alike should be the non-distributive grouping. Distributive seems convoluted and non-obvious syntactically. After all, complex tanru, connectives or no, are interpreted as their components modify each other as tanru elements, in the order of grouping. You don't usually see the nature of one element affect how modifies the next larger one. Just as {melbi cmalu nixli ckule} is {(((melbi cmalu) nixli) ckule)}, that is "((pretty type-of small [thing]) type-of girl) type-of school", with each element modifying the next one out in "unmarked tanruic" form (i.e. somehow having to do with), so too {cmalu je nixli ckule} must be {(cmalu je nixli) ckule}, or "(small and girl) type-of school." The nature of the first tanru element, that it is a logical connective, ought to have no bearing on the fact that it simply modifies the second one in umarked method. Thus, we have a school that is somehow associated with things that are small-things and girlish. Similarly, (here I may be weakening my point), {cmalu je nixli ja ckule} would be {(cmalu je nixli) ja ckule}, that is, "(small and girl) or school", so it might describe something which is either (a) both small and girlish or (b) a school, (and possibly both). Basically, the way tanru elements modify each other should be dependent *only* on the various cmavo we have for just that purpose (JA, JOI, etc.) and *not* on some magical characteristics of the elements themselves. What would happen if we somehow crammed in a connective by means other than JA or GUhA? Like mela'elu....li'u? Heaven only knows what's inside of that! How do we know when (or how!) to apply distribution? If you want to throw a cmavo in to flag distributivity, that's a different kettle of worms, and I wouldn't mind that. I think the A je B xai C method makes the most sense: again, we have a cmavo between tanru elements to flag the way the modification happens. And as for your example (1), for me, {melbi je cmalu nixli} makes eminent sense as "(beautiful and small [thing]) type-of girl", which would likely expand using parallel tanru method to {melbi je cmalu je nixli}, a beautiful and small girl: a thing that is beautiful and is small and is a girl (or, to use a sumti, {lo nixli poi ke'a melbi je cmalu}). There are, of course other interpretations. Maybe a girl who collects small-and-beautiful things? ~mark