From a.rosta@xxxxx.xxxx Tue Dec 7 07:05:07 1999 X-Digest-Num: 304 Message-ID: <44114.304.1658.959273825@eGroups.com> Date: Tue, 7 Dec 1999 15:05:07 -0000 From: "And Rosta" Date: Mon, 29 Nov 1999 16:07:32 -0000 > > > From: "Jorge Llambias" > > > > > > > What I have for dinner depends on what there is in the fridge. > > > > > > > > le nu mi citka roda poi mi citka ke'a cu jalge > > > > le nu rode poi ke'a nenri le lekmi'i cu nenri le lekmi'i > > > > "My eating that which I eat is a result of > > > > that which is in the fridge being in the fridge". > > > > > >I don't think this gets it. Yours (but not my original) would be > > >true if the fridge contents' being fridge contents had, say, > > >miraculously healed me of an inability to eat. > > > > I think you're right. What I would need is: "My eating that > > which I eat, and not something else, is a result of > > that which is in the fridge, and not something else, being > > in the fridge". > > > > Would that do it? > > I reckon so. Astrophysics's gain is linguistics' loss. > > [I can't explain the genitives in that sentence. They just sound > righter than the alternatives.] > > > > > It seems to work for other indirect questions as well: > > > > > > > > la djan djuno le du'u makau klama > > > > John knows who came. > > > > > > > > ro da poi ke'a klama zo'u la djan djuno le du'u da klama > > > > For each x that came, John knows that x came. > > > > > >I think you need to add > > > > > > ... and for each x that did not come, John knows that x > > > did not come > > > > Right, and I think I also need to add: "... and if nobody came, > > then John knows that nobody came." > > I don't see why. Ah, hang on. You're saying that "Everybody is > such that John knows they didn't come" doesn't entail "John > knows everybody didn't come", but that "John knows who came" > requires there to be such an entailment. Hmm. Not necessarily: > > A: You know who came. > B: Do I? I don't know of anyone's having come. > A: That's right; nobody came. > > In other words, B knows who came, but doesn't know that they > know who came(!) > > This is getting into greater subtleties than I'd originally > intended. I wonder whether it is "know" that is complicating > things here, rather than interrogativity per se. > > > And I need to add something like that in the prenex version > > of what I have for dinner, too, to cover the cases where > > I had nothing for dinner or where there is nothing in the > > fridge. > > > > Indirect questions are complicated beasts. > > Indeed. Oddly, I'm not aware of a profusion of studies of their > semantics in the linguistics literature. > > > >> > la djan djuno le du'u makau klama > > >> > John knows who came. > > >> > > > >> > ro da poi ke'a klama zo'u la djan djuno le du'u da klama > > >> > For each x that came, John knows that x came. > > >> > > >>I think you need to add > > >> > > >> ... and for each x that did not come, John knows that x > > >> did not come > > > > After some more thought, I think a better rendering would > > be: "For each x that came, and for no other x, John knows > > that x came." This is because if Paul didn't come, but > > John doesn't even know of Paul's existence, then saying > > "John knows that Paul didn't come" sounds wrong. Better > > to say that John doesn't know anything about Paul. > > > > That way it also fits better with the dinner one. > > So we ssem to be saying that interrogatives mean: > > "for set s, such that for every x x is in s iff [whatever], > for all?/some? y, such that y is in s, > for all?/some? z, such that z is not in s, > it is/being the case that y is in s and z is not in s" > > This would seem to cover: > > Where you live influences *what your insurance premiums are*. > Your living in that neighbourhood influences *what your > insurance premiums are*. > John knows *who came*. > > -- setting aside the issue of to what extent the knower knows that he > knows what he knows... > > For *whether he came*, s is the set of truth values of "he came". > I think. > > If you agree with this, my next question is how to strip the > repetition out of the formula. > > I think my former rendition of "know who came" as "for every x, know > whether x came" (with a further step to translate "whether" into > logical form) was simpler than what we are proposing here, but I > never got it to generalize to nonepistemic examples like the > insurance premium ones above. > > --And.