[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
| In a message dated 10/17/2002 2:27:31 PM Central Daylight Time, jjllambias@hidden.email writes: << Compare: >> I worry a bit about the loose movement back and forth between different kinds of quantifier expressions, that is, about the "="s. I am not at all sure at what level these exchanges work -- and I don't think it is exactly settled. So, the claimed differences here presumes one way of doing this, which is not obviously the correct way -- and, of course, consequently, one way of reading the questioned sentence, which is not obviously the correct way either. On the whole I agree with xorxes about how it should be, but don't see any very strong reason (CLL being silent on this as most technical logical questions -- go figure) to think that that is not the right way to go. Thanks for pointing it out, but now what? << >Taking the {na} out >would give >{naku ro da poi cipni zo'u ga da vofli ginai da blabi} , which seems to say >the same thing though less clearly. Right. But compare it with {lo cipni na vofli}. >> Oops, in the fog, I did that backwards: it should have been something like {ro da poi cipni zo'u ga da naku vofli gi da naku blabi} : "every chicken that flies is non-white" stronger than what was wanted, but what is wanted was never said anyhow: {lo cipni na vofli} is just {no cipni cu vofli}, from which the above follows. (I was about to start an old row by pointing out that {lo cipni na vofli} at least implies what you want to say, apparently, ({lo cipni cu naku vofli}) but skip that.) << lo cipni na vofli gi'e blabi 1) {gi'e} has scope over {na} This is how the parser groups it, and it is how it should be if it is going to be equivalent to {lo cipni cu vofli nagi'e blabi} and to {lo cipni cu blabi gi'enai vofli} and to {lo cipni cu blabi gi'e na vofli}. >> But why should {na vofli gi'e blabi} be equivalent to { vofli nagi'e blabi} in the critical way? One is (as you would have it) a compund predicate with a negated component, the other is just a compound predicate -- with a different compounding operator. To be sure, in some contexts, {na-- e ...} interchange with {-- na e ...}, but this may not be one of those cases. << 2) {su'o} has scope over {gi'e} This is just order of appearance. CLL studiously avoids mixing connectives and quantifiers anywhere, and also sumti connectives with bridi-tail connectives, so there is no strict official rule about this, but order of appearance is the rule for each of these things, so it is the natural rule when combined also. >> Yes, this makes sense, but is not necessarily correct for all that. I take it that you are finding reasons to doubt it. But in this case, of course, the original rule was that {g'ie} has scope over sumti: A broda ije A brode => A broda gi'e brode. But this cannot be the rule that applies here, since the two {lo cipni} are not correferential. So, how did this sentence come to be? May it be that it can't and thus is a false case for {lo cipni na vofli ji'e blabi} where {na} takes the whole comlex as its scope and all is right with the world by the old rules? Or it is from some deep structure where the two sumti are correferential but the second cannot reliably come to the surface, i.e, the one with blabi is a kind of {no cipni} that, for various reasons, gets absorbed over to the {naku lo cipni} in the first. Or.... In short, we don't know what the rules are or whether they have anomolies until we know what the situation really is -- and, despite the claims that deep structure is near the surface in Lojban, what is going on is not always very clear. As here. << Now, (1), (2) and (3) cannot all be correct, because scope is transitive. One of them has to be broken. >> Even this is open to some question: the combination may create a situation where the separate rules do not combine as expected. So, the rest does not follow. |