From LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU Sat Mar 6 22:51:58 2010 Received: from YALEVM.YCC.YALE.EDU by MINERVA.CIS.YALE.EDU via SMTP; Wed, 26 May 1993 00:17:04 -0400 Received: from CUVMB.CC.COLUMBIA.EDU by YaleVM.YCC.Yale.Edu (IBM VM SMTP V2R2) with BSMTP id 6492; Wed, 26 May 93 00:16:16 EDT Received: from CUVMB.COLUMBIA.EDU by CUVMB.CC.COLUMBIA.EDU (Mailer R2.07) with BSMTP id 5444; Wed, 26 May 93 00:17:20 EDT Date: Wed, 26 May 1993 00:15:47 EDT Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: TECH: more on xo'e X-To: lojban@cuvmb.cc.columbia.edu To: Erik Rauch Status: O X-Status: X-From-Space-Date: Tue May 25 20:15:47 1993 X-From-Space-Address: @YaleVM.YCC.YALE.EDU:LOJBAN@CUVMB.BITNET Message-ID: Cowan said, regarding the "xo'e" proposal to allow semantic place deletion: >I still cannot come up with any satisfactory scheme for marking place >deletion directly in the selbri, where it belongs. My general idea is >that there should be a new sort of SE-equivalent, similar to JAI, which >has the effect of deleting a specified place (swapping it with limbo?), >with a format something like "xi'o #" where "#" indicates which place to >delete. But all ways of specifying "#" (SE cmavo, FA cmavo, actual >numbers) seem unbearably ugly. Can anyone see how to do this? If the proposal is going to operate at the selbri level, rather than the sumti level, it must result in a renumbering of the places. You don't want to have to try to figure out what le xi'o-1 selbri is. (i.e. you delete the first place and then refer to it in the description.) In addition, it would seem counter to logic to deny the existance/relevance of a place and then leave it with a place number. Because FA operation do not result in renumbering, and SE operations DO result in renumbering, this suggests that a SE-related process is the way to go (at least at the selbri level). On the other hand, this has the problem (as And complained about for other reasons) that SE has no element for referring to the x1 place explicitly. One way that has not been cosidered, but which would have small effect on the grammar, would be to use "nai". I have mixed feelings about something like "senai" because the type of effect it has is drastically different from the effect of normal conversion, but modulo the missing x1 referent, this seems to meet the need. Another possibility would be to allow FA tags as inflections (they are otherwise similar to modals in grammar, and this would be parallel to using modals as inflections. Since "fa selbri" is undefined right now, it wouldn't even need an additional word like xi'o, though we might want one. If it was decided that there was merit in And's analysis claiming a need for a xo'e and a xe'o, one to delete the place as irrelevant, the other to assert the predicate with no value in that place, we have the capability (I think) to allow both FA and FA+NAI as inflection. Whether these plausible ways are good ones, I cannot say. BTW, the reason I claimed that this operation is metalinguistic (which John disagreed with) is PRECISELY because we are operating on a selbri which DOES include the place we want to delete. All methods proposed of deleting a place are inherently going to have to mention that place, which to me is on its face a metalinguistic reference. I'll accept that SOI may not do the job (though this may be a weakness of our definition of SOI, which has seen little real use), but SOI also is metalinguistic while making a secondary assertion (the 'vice versa' one). lojbab