[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
| In a message dated 11/5/2002 5:32:44 PM Central Standard Time, a.rosta@hidden.email writes: << >Well, it is clearly not required that every member -- or any -- be a >> Well, {le du} does get rid of non-veridicality directly and {noi} carries it on nicely. << > Nor is it required > that there be a defining property -- you just pick 'em out somehow. My point is that what you say is true for {le (su'o)} but not for {le ro}. For {le ro} there must be a defining property. >> I don't see how this comes about -- {le} is defined as "the selected ones which I am describing as." I suppose the selection can be made on the basis of a property, but I can't think of any case where it has to be (well, picking out the prime numbers would have to be, but that has nothing to do with the choice between {su'o} and {ro} as internal quantifiers). << > And, of course, in Lojban as > opposed to Andban and Llamban perhaps, {ro} does not allow 0 -- this > is logic after all (16.8(399)). 16.8 pertains to {ro} as a quantifier, not as a cardinal number. As a cardinal number in the so-called "inner quantifier" position, {ro} is little different from {tu'o}. >> I'll take you word for the relation between {ro} and {tu'o} since you claim to understand it (though can't explain it). For the rest, I find it unlikely that {ro} changes its meaning when no other world in the class does. << As for Andban and LLamban, I've only said that I would have to unlearn less of {ro} as a quantifier did not entail {su'o}. Furthermore, I interpret 16.8 as making the existence follow from {da}. For instance, my interpretation is that >> Lojban is not responsible for your deplorable education. Where did you ever learn (let alone were ever taught, I hope) that a universal quantifier does not have existential import. I suspect that you somehow managed to scramble together the way that modern logic translates English "each/any/every/all" with the universal quantifier itself. The nonimporting translation results from the conditional after the quantifier, not from the quantifier itself, which imports as always. << For instance, my interpretation is that no da poi broda cu brode entails da broda >> How very odd! (coming from you, I mean). I would be perfectly happy to have that, but suspect that the opposite is more natural (though, I do think that {no broda cu brode} entails {da broda}. << It's not a snap if the set is a specific one with its own identity and is a subset of {lo'i broda}. >> Easiest thing in the world: the null set is always a subset and has its own identity. What more do you want? I suspect that what you want is {x: x broda gi'e brode}, for some {brode} and then want to avoid mentioning the {brode} -- why? If you use it, why be shy (or sly) about it. |