Date: Thu, 4 Dec 1997 14:20:45 -0500 (EST) Message-Id: <199712041920.OAA15158@locke.ccil.org> Reply-To: Thanatos Sender: Lojban list From: Thanatos Subject: Domains of Discourse and lo mlatu X-To: Lojban List To: John Cowan X-Mozilla-Status: 0001 Content-Length: 1153 X-From-Space-Date: Thu Dec 4 14:21:04 1997 X-From-Space-Address: LOJBAN@CUVMB.CC.COLUMBIA.EDU From: And Rosta >I do feel that our logic debates do eventually lead to wider >understanding of the issues. For example, in recent discussions >of LE/LO, several people besides me & Jorge contributed, and all >understood the difference correctly, except Bob (and maybe you). Not to belabor the point, but to put another perspective on the issue. Bob seemed to want to switch the domain of discourse from one set of cats to another. If you do that, {mi nelci lo mlatu} and {mi na nelci lo mlatu} are true if you've switched domains between the two. It seems reasonable that {lo mlatu} could be drawing from some subset of all the cats in existence if the conversation has already been limited, i.e. the domain of discourse set, to some subset of all cats. However, changing domains in the middle leads to confusion, such as switching between different subsets of all existent cats. Without mentioning a particular subset, it would seem that {lo} would draw from all the cats in existence, setting the domain to all cats, in which case both {nelci lo mlatu} and {na nelci lo mlatu} could not both be true. -- Erik W. Cornilsen