[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
| A so des'! The recent discussion while I was asleep has been very illuminating. And what it shows is that at least part of the problem is that we are looking for at least three different things but calling them all {lo'e}. Nick is going for "the typical." I am going for the larger type of which "the typical" is a special case (along with "the average", "the normal" and Lord knows what else). Xorxes is after, it turns out, a non-raised version of sumti in {tu'a}. And is after -- well, I don't know what, though the squinty metaphors suggest something like "the typical" or "the generic." The point about xorxes came home when he commented that we got into trouble using {lo'e} in first position when it was usually elsewhere. That just did not jibe with the {lo'e} as I knew it. And then something else came along in his comments -- I can't now find just what -- that made me realize that he was talking about {tu'a lo ...}. From that point of view, putting {lo'e} in second place (which I assumed he had done just because {sisku} was so handy for what he was about) made sense (as -- sorta -- did starting with {sisku}). And so did the distinction between internal and external quantifiers, which is exactly right for that meaning. Of course, I think that there is an easier way to do this -- defining in terms of {tu'a} in particular, though that does not extend to predicates that don't take abstractions. On the other hand, I am not at all sure why we would want those terms elsewhere anyhow: I don't see hitting such a man, for example (making do allowance for the mess the metaphor makes "any man, it does not matter which) as something I would or could do -- though I can imagine wanting to do it and so on -- back in the original context again. And, even if it does work, I don't see it as being much like {lo'e}. It really is a singular term, though with a completely unspecified reference, whereas {lo'e} is not (per CLL and the cmavo list and Nick and me) not a singular term at all, but a convenient short form for a fairly complex claim -- an incomplete symbol that has meaning only in the unpacking of the whole sentence of which it has a part, typically as first argument. (I also am curious to see what the {kairbroda} generalization will be of predicates that normally take abstract arguments -- I suspect that we will find that our old familiar predicates are the {kair} forms really, and "the real" predicates are the ones that take xorxes' sumti.) Part of the collapse -- historically -- was the odd case of {nelci} which appears to take either abstract or (sorta -- to allow for generic cases) concrete second arguments. While I think this is just a mistake on somebody's part somewhere along the line -- inadequate analysis springs to mind, along with English unmarked subject raising. I suppose that xorxes formula works well here, except that it does seem to be about bits of chocolate, not generic chocolate. But then, that is probalby what {nelci} is about, too (real generics tend to go in subject position, I think, again). |