Date: Mon, 3 Nov 1997 14:24:01 -0500 (EST) Message-Id: <199711031924.OAA01564@locke.ccil.org> Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: le/lo X-To: lojban@cuvmb.cc.columbia.edu To: John Cowan X-Mozilla-Status: 0001 Content-Length: 3553 X-From-Space-Date: Mon Nov 3 14:25:03 1997 X-From-Space-Address: LOJBAN@CUVMB.CC.COLUMBIA.EDU >Essentially, there are two different, competing ways of interpreting "le" >and "lo": > >(1) Use a non-veridical descriptor to be veridical by default - I agree >that that's confusing, but it's also the standard way it seems to be used >by Lojbanists. I don't think that "le" is "veridical by default". "le" is "in-mind" by default. The speaker is relying on shared knowledge of the properties of the referent to label it. But that knowledge need not be accurate, so long as it is shared (i.e. the listener can identify the referent from the description). The goal is for the speaker to bring to the listsner's mind the referent that the speaker has in mind. IF the speaker succeeds in doing so, then there is no requirement that the description actually be correct. "lo" puts the added burden on the speaker of ensuring that the description actually applies to the referent accurately, and that is ALL that it imposes. However, the fact that the default quantifier on "lo" is "su'o" rather than "ro" means that it is a selection from the set of those described rather than all of them. That selection is nonspecific byt default, is has been noted (though I think there are ways to make it clear that it is intended to be specific) but "lo" indeed refers to all in the universe of discourse. I am inclined away from Bob C. on this in that the history of "lo" goes back to JCB's "lea" which means precisely "rolo", and was used for making those classic predicate logic statements like "All Xs are Y". These logical statements are seldom really true about the real world -or at least those statements that are indeed universally true don't really need to be said. I argued (with pc) to change the default quantifier to "su'o" because it is easy enough to explicitly say "rolo" and I wanted the default of the language to be something usable in real conversation. In English, using the word "all" in a claim is a mark of exaggeration. I do not want it to be so in Lojban, so "all" needs to be something that must be said explicitly in order to be claimed. I think similarly, if lo implies a selection from the universe of discourse, it is as unfair to assume a limited universe of discourse without marking it explicitly. If indeed we used limited universes of discourse in everday conversation, then the exaggeration of English "All Xs are Y" would not be an exaggeration - as long as at least one X is a Y, we can merely claim that our statement is true about a limited universe of ddiscourse that conveniently fits the true cases. No, I think that discourse limitation should be expressed with poi or pe. You can of course get a specific using "lo" with "lepalo broda. The fear I have of Bob C's version is that it will actually lead to increased English=like thinking. Using an old example and JCB's word "taksi" for a taxi, we will tend to translate "mi klama fu lo taksi" as "I came in a taxi" with virtually an exact mapping of "a" to "lo" in meaning. (Prior debate on this issue reached the conclusion that the correct descriptor is "loi". I won't attempt to reahsh this though.) lojbab ---- lojbab lojbab@access.digex.net Bob LeChevalier, President, The Logical Language Group, Inc. 2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273 Artificial language Loglan/Lojban: ftp.access.digex.net /pub/access/lojbab or see Lojban WWW Server: href="http://xiron.pc.helsinki.fi/lojban/" Order _The Complete Lojban Language_ - see our Web pages or ask me.