From cbmvax!uunet!cuvma.bitnet!LOJBAN Sun Jan 12 08:07:51 1992 Return-Path: Date: Sun Jan 12 08:07:51 1992 Message-Id: <9201120632.AA23795@relay1.UU.NET> Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: gadri, and misc responses to Colin X-To: lojban@cuvmb.cc.columbia.edu To: John Cowan Status: RO Colin writes: >The further I get into playing around with Lojban, the less I think I >understand the different gadri. When I made the above comment, I was >somewhat infected with the ideas of the *loglan* articles 'le' and 'lo' >(which roughly correspond to 'le' and 'lei', but sometimes to 'le' and >'lo'). I am having doubts about the rule of thumb that 'le' is 'the' >and 'lo' is 'a'. Both are specific (the particular one or ones I'm >talking about) whether that specificity is external or just in the >speaker's mind - the difference is whether the selbri is being used as a >convenient description, or actually (incidentally) asserted to be true >of the thing thereby described. TLI "le" is the same as Lojban "le" TLI "lo" is the same as Lojban "loi" Lojban "lo" is the same as TLI "lea" but with a different default quantifier. TLI's quantifier is "all" making its "lea" = "all of the members of the set who actually satisfy the description selbri"; Lojban's default quantifier, su'o, makes "lo" mean "at least one/some of those who actually satisfy ...". Thus Lojban's is closer to a true indefinite, while TLI's similar word is only useful for the rare universal claim. "lei" and "lai" came about to fill what was obviously becoming a pattern, especially when we also added words to talk about the sets themselves (le'i/lo'i/la'i) as distinct from the members of the sets (le/lo/la). lojbab Our choices of quantifier make lo/le closer to teh English indefinite/definite. However, the Lojban 'indefinite is truly an indefinite, whereas English has several rhetorical uses for the indefinite that bespeak a particular one, a typical one, etc. that have other expressions in Lojban. The Lojban indefinite, not restricted by poi, is a very weak statement about the category described - that some member thereof fits the predicate being claimed by the bridi. poi is tied neither to lo nor le, but it does make a claim about whatever it is being talked about - a restrcitive one. In the case of da poi blanu, you are claiming that there exists something that is blue, and that thing is the subsject of the main bridi claim. vo'a was intended at creation to be restrictable to the present, previous, or some other bridi with "pedi'u" and the like. Indeed, the original default was the previosu bridi rather than the current one. But the current one is more useful in actual usage, it turned out. le go'i is a near synonym for vo'apedi'u but the nature of le does have an impact. lo go'i might be a better synonym. The place structure of facki and djuno, etc. was changed per the discussion about sumti raising in JL15. x1 knows/discovers (fact) x2 about x3 and x4 is an epistemology place in djuno. I suspect cilre may also share this pattern. zdile is indeed, as Colin suggests, entertained/occupied/diverted version of amusing rather than xajmi which is merely funny/humorous. This corresponds to the attitudinal .u'i amusement with 'opposite' .u'inai weariness, as distinct from zo'o - the humor attitudinal. lojbab