From @uga.cc.uga.edu:lojban@cuvmb.bitnet Sat Jun 24 00:38:55 1995 Received: from punt2.demon.co.uk by stryx.demon.co.uk with SMTP id AA3582 ; Sat, 24 Jun 95 00:38:51 BST Received: from punt2.demon.co.uk via puntmail for ia@stryx.demon.co.uk; Fri, 23 Jun 95 19:38:21 GMT Received: from uga.cc.uga.edu by punt2.demon.co.uk id aa22532; 23 Jun 95 20:38 +0100 Received: from UGA.CC.UGA.EDU by uga.cc.uga.edu (IBM VM SMTP V2R2) with BSMTP id 7369; Fri, 23 Jun 95 15:36:04 EDT Received: from UGA.CC.UGA.EDU (NJE origin LISTSERV@UGA) by UGA.CC.UGA.EDU (LMail V1.2a/1.8a) with BSMTP id 6574; Fri, 23 Jun 1995 15:36:03 -0400 Date: Fri, 23 Jun 1995 15:39:32 EDT Reply-To: jorge@phyast.pitt.edu Sender: Lojban list From: jorge@phyast.pitt.edu Subject: Re: pc answers To: lojban@cuvmb.cc.columbia.edu Message-ID: <9506232038.aa22532@punt2.demon.co.uk> Status: R And: > > > > ro lo ci lo nanmu cu pencu ro lo ci lo gerku > > > > Each of three of all men touch each of three of all dogs. > > I was thinking that the outer > > quantifiers are nested in order of appearance and the inner ones > > are independent. > I take it that by "inner quantifier" you > don't mean "innermost", since everyone is agreed that the innermost > quantifier of {ci broda} is {ro} - {ci lo ro broda}. Yes, by "inner" I meant everything except the outermost ones. > If you do explain your version, could you also confirm that it > covers {le ci nanmu cu pencu ci gerku} (which has the 3-dog/9-dog > alternatives)? That one should be 9-dog as well. The outer quants are {ro} for the nanmu and {ci} for the dogs. For each man, three dogs. My idea is that any inside quantifier is in some sense part of the description, and independent of what's going on elsewhere, so, just like {lo gerku} is "at least one of all those that are dogs", then {lo ci lo gerku} is "at least one of some three that are dogs" and "all those that are dogs" and "some three that are dogs" are determined before the quantification of all the arguments. In other words, the outer quantifier in {lo gerku} selects from {lo'i gerku}, and the one in {lo ci lo gerku} would select from {lo'i ci lo gerku}, and all such sets are defined initially and not within the scope of other quantifiers. But in the case of {ci lo gerku}, (or {ci lo mu lo gerku}, etc) The selection of the cimei happens within the normal scope, because it is an outer quantifier. > > > Ev, v a cimei, Ew, w a cimei, Ax, x in v, Ay, y in w: x pencu y > > > instead of what we seem to be agreeing should be the default: > > > Ev, v a cimei, Ax, x in v: Ew, w a cimei, Ay, y in w: x pencu y > > Just as a reminder, these violate the goatleg rule. They would be > > the two possibilities for {su'oci remna cu pencu su'oci gerku}. > Could you remind me why (or whether) {ci nanmu cu pencu ci gerku} > doesn't violate goatleg? It depends how you expand it. If your expansion above was meant for {ci nanmu cu pencu ci gerku}, then it does, because your expansion permits that a {vomei} also satisfy it, but the goatleg rule doesn't allow that. The corresponding goatleg respecting expansions of {ci nanmu cu pencu ci gerku} are, for the 3-dog case: E1v, v a cimei, E1w, w a cimei, Ax, x in v, Ay, y in w: x pencu y and for the 9-dog that we prefer: E1v, v a cimei, Ax, x in v: E1w, w a cimei, Ay, y in w: x pencu y where "E1" stands for "there is one and only one". > But at this stage I would simply like to establish and get general > agreement that Lojban ought to have a comprehensive method of doing > afterthought scope. For me the upshot of The Any Debate was that there > is such a need. Well, I don't know. A comprehensive method probably can be worked out. But if it will require a lot of new cmavo, and a lot of complication in use, then I doubt it would be worth it. There is no point in having afterthought available if it will still require a lot of thought to use it right. I think Lojban ought to have it only if the comprehensive method is simple enough to use. I have to think more about it. Jorge