From @YaleVM.YCC.YALE.EDU:LOJBAN@CUVMB.BITNET Mon Nov 15 21:59:45 1993 Received: from ELI.CS.YALE.EDU by NEBULA.SYSTEMSZ.CS.YALE.EDU via SMTP; Tue, 16 Nov 1993 03:02:17 -0500 Received: from YALEVM.YCC.YALE.EDU by eli.CS.YALE.EDU via SMTP; Tue, 16 Nov 1993 03:02:13 -0500 Message-Id: <199311160802.AA20103@eli.CS.YALE.EDU> Received: from CUVMB.CC.COLUMBIA.EDU by YaleVM.YCC.Yale.Edu (IBM VM SMTP V2R2) with BSMTP id 4885; Tue, 16 Nov 93 03:01:54 EST Received: from CUVMB.COLUMBIA.EDU by CUVMB.CC.COLUMBIA.EDU (Mailer R2.07) with BSMTP id 9577; Tue, 16 Nov 93 03:01:12 EDT Date: Tue, 16 Nov 1993 02:59:45 -0500 Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: another response on old posting - lujvo-making To: lojban@cuvmb.cc.columbia.edu Status: RO X-Status: In a posting last month (the same as I respond to in another posting tonight) John Cowan responds to an anonymous person: >> Rather than lean gismu or fat lujvo, I argue for weaking place >> structure. That each successive place of the word is weaker in >> its tie to the concept. >> For example: 'going' is central to 'klama'. A goer >> is the closest likely relation to going, thus X1 is for the goer. >> A destination is next closest/most-likely relation to going, thus >> X2 is the destination. At some point things related to going are >> sufficiently unlikely to be spoken of and/or sufficiently remotely >> connected to the action that we rely on the modifier words to >> attach places to connect them, like the reason for going, the time >> of going, the source of energy for going, the cost of going, etc. > >I think this analysis is very useful, although not absolutely true. It also makes an incorrect presumption about how places were selected and ordered in the place structures. In some cases, of course, we were merely patterning after English or some natural language, in deciding places and orders. In some cases, places were chosen so as to move likely abstractions to the end of the place structure, where there would be less need for elidables. More recently, we have tended to play games with moving agents to x1, and then deleting agents where it is plausible, since the agent can be added in more clearly than it can be deleted if there is an agent-less predicate that is related and meaningful. To adopt this analysis, we would have to analyze the place structures yet again to be consistent with the principle, and it ain't gonna happen. It would also play havoc with Nick's lujvo place structures system, since it is quite plausible that an unimportant place in a gismu could become important in a lujvo based on the gismu, and there is (I presume, not having studied Nick's work carefully) no principles for reordering such places based on some ad hoc 'importance' criteria. (I think it needs to be allowed, of course, when there is a good reason, one of the reasons I am skeptical of algorithmic lujvo place structures. lojbab lojbab