[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
| In a message dated 11/6/2002 5:47:29 PM Central Standard Time, a.rosta@hidden.email writes: << I was arguing that {le'i} can be {le'i no}. If {le} is defined in terms >> Let's see. I agree that {le'i} should be able to be {le'i no} -- the empty set I pick out (typically what set I am thinking of it as a subset to). This entails a change from something pretty clear in CLL and your wiggle does not seem a convincing one at the moment. I can't think of a good reason for defining {le} in terms of {le'i} nor for doing it the other way (and the empty set problem is one of the reasons against doing either). I assume that when you say {ro le ro broda} = ro co'e je broda} you are assuming that {broda} is veridical (is that what "presuppositions apart" means? this doesn't look like a presuppositional thingy ). But if you do use this definition, how do you distinguish this from {ro lo ro co'e je broda}? Ah, you don't, which seems odd, somehow but does make a kind of sense -- {le broda} is just about a subset of {lo broda}. << I don't know what meaning {ro} has as a cardinality indicator. >> Probably not much more than {su'o} -- but that sets it off from {tu'o} at least. << Okay. so'e as a cardinal is not meaningless but is nonsensical. >> Well, not nonsensical in any technical sense, just not very helpful. (I suppose it could be taken as requiring at least three objects, but that is iffy.) << I reckon that ro was chosen for the default in the belief that it is tantamount to having no default. >> That may be right, but my recollection is that it was picked in the belief (over my objection, of course) that {ro} included 0. |