Return-Path: Message-Id: From: cowan (John Cowan) Subject: Re: Interpreting sumti (was: anaphor means what?) To: lojban-list Date: Wed, 1 May 91 11:41:43 EDT In-Reply-To: <9104301552.AA14152@euphemia.math.ucla.edu>; from "math.ucla.edu!jimc" at Apr 30, 91 8:52 am X-Mailer: ELM [version 2.2 PL13] Status: RO X-From-Space-Date: Wed May 1 11:42:20 1991 X-From-Space-Address: cowan djim. kartr. writes: > What I really want to hear about is the "problem of the empty set". When > is a sumti veridical -- which I take to mean has a hidden implicit > existential quantifier in it? And when is a void referent set simply > ignored, producing a vacuously true assertion? Take for an example: > > The dodo lives on Ascension Island > > to be translated with "lo". The X1 sumti (all things that really are > [alive] dodos) has a void referent set. What problems arise here? You have the distinction exactly backwards. (First, a minor side issue: Lojban "da" is Old Loglan "ba", not Old Loglan "da". Do you have this firmly in mind? Some of your remarks on "da poi" make me wonder if you are hearing this "da" as free rather than bound. Non-Old-Loglanists, ignore this.) A veridical description is precisely >not< one that has a hidden existential quantifier; "da poi" is the one with the implicit existential quantifier. Therefore, "da poi " can't ever refer to the null set, because "da" is implicitly existentially quantified unless some other quantifier appears on it. "lo ", on the other hand, is free to reference the null-set-in- extension, in other words nothing. lo cipnrdodo cu xabju le daplu se cmene zo .asencn. Those-which-are bird-"dodo"s dwell-on the island named "Ascension". is vacuously true, whereas da poi cipnrdodo cu xabju le daplu Something such-that [it] is-a-dodo dwells on the island. is false, because it contains an implicit (not hidden) existential quantifier on "da", and is in fact the same as: su'o da poi cipnrdodo zo'u da xabju le daplu There exists a dodo: it dwells on the island. "Veridicality", OTOH, has to do with whether the object described is required to actually meet the description. The sentence mi catlu lo cribe I look-at a bear. is true only if, in fact, the object I am looking at is a bear, whereas mi catlu le cribe I look-at the bear. can be true even if the object I am looking at turns out to be a raccoon. "le" descriptions are "non-veridical" because the truth of the bridi in which they are embedded is independent of the truth of the description. (It seems to me, based on your use of the phrase "in-mind selection" when explaining -gua!spi "xe", that the superficially analogous !ji /knl !berw is true only if the referent of X2 place is in fact a bear, although I am free to choose any bear I like without sacrificing truth.) -- cowan@snark.thyrsus.com ...!uunet!cbmvax!snark!cowan e'osai ko sarji la lojban