From LOJBAN%CUVMB.BITNET@UBVM.CC.BUFFALO.EDU Mon Jan 1 15:50:55 1996 Received: from vms.dc.lsoft.com (vms.dc.lsoft.com [205.186.43.2]) by locke.ccil.org (8.6.9/8.6.10) with ESMTP id PAA04425 for ; Mon, 1 Jan 1996 15:50:45 -0500 Message-Id: <199601012050.PAA04425@locke.ccil.org> Received: from PEACH.EASE.LSOFT.COM (205.186.43.4) by vms.dc.lsoft.com (LSMTP for OpenVMS v1.0a) with SMTP id 4A91E7E0 ; Mon, 1 Jan 1996 15:28:40 -0500 Date: Mon, 1 Jan 1996 12:11:15 -0800 Reply-To: "John E. Clifford" Sender: Lojban list From: "John E. Clifford" Subject: opaque X-To: lojban list To: John Cowan Status: OR X-Mozilla-Status: 0001 Content-Length: 2142 &: For shifts to other worlds, we need something in NAhE, as I've said before pc: NAhE has the wrong grammar for the present question, which is just about referential expressions in otherwise normal contexts. NAhE is useful for shifting the whole bridi into Nephalococcygia - - although I would have thought that the home of the usual modals, CAhA, would do it even better. What we need for the present problem is a sumti-to-sumti function that projects a new point of evaluation for the sumti inside, LAhE as ever was. &: I accept that in principle there could be a new addition to LAhE that means "the quantification on this sumti does not belong in the bridi the default rules would locate it in". This in effect is what Jorge proposed virtually at the start of the debate on "any" - I think he called it {xee}. pc: I guess we had to go through all this to get me to understand that that was the point, if it was -- I am not convinced, looking through the records of that year-and-some-long discussion. We need (in the current depressing state of the language) some way to get opaque contexts on the surface so that we can go unicorn hunting and the LAhE solution seems to be the one that causes least hassle (except for my relearning cost -- and everyone else's when they start to see what follows). But this is a very different (though related) problem from the "any" case, which was getting out of acknowledged opaque contexts with terms known to refer into the upper level world. Yes, they are both "evaluate in a different world from the obvious," but the direction to look for the appropriate world is opposite. And the vague xe'e (?) flag allows for uncertainty about what world to go to when there is or might be a number of layers. Does this mean more damned subscripting? gaxfesti (Most of the "any" problems seem to me to have arisen from the ongoing problem -- that will now get worse -- of failing to note opaque contexts when they occurred or were intended, using simple sumti when nu sumti were required in the old cases, now failing to mark opaque sumti when intended.) pc>|83