[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
| In a message dated 11/7/2002 12:23:35 PM Central Standard Time, jjllambias@hidden.email writes: << >Now subtract the three blocks: thy're >> No, it is not {abu joi by joi cy bevri le pipno}, since at no time are all three in it together, though each of them is in it sometimes It is not G but {ge D gi E ije ge E gi Fije ge D ge F} . I suspect this is just a different way of understanding the additive properties of masses. In the first (carry three blocks) though at no time were all three involved, what each pair achieved was only part of the whole property in question. In the second case (carried) still all three are not involved at any one time yet each pair achieves the whole of the property in question. My point is that this is still {piro lei ci nanmu cu bevri} but not {abu joi by joi cy bevri}. That is, G is sufficient but not necessary for {piro lei ci nanmu...} << I think {lei ci nanmu} is not more than {abu joi by joi cy}. A claim about {[piro] lei ci nanmu} cannot be broken down logically in terms of submasses like {abu joi by}. For some particular predicates we might infer something about the submasses, but not due to some general logical rule. >> You have the problem backwards here -- and also wrongly answered. The first point is that several things other than G entail {lei ....} and, consequently, {lei ...} does not entail G flat out. In a particular case it may, but it need not. And similarly for the other cases of partial masses. In any case, the main point is that masses are not individuals but round about ways of talking about individuals. As such they cannot directly instantiate universal or generalize to particulars. The fact that a mass of individuals moved a piano does not mean that there is something, in the {da} sense, that moves a piano. And similarly for other variables and other such indirect references to individuals (though not sets, oddly) and similar variable devices. << None of this seems to relate to whether or not it makes sense to use {ko'a goi lei broda} within the scope of something else. >> The point is that this {ko'a} is not marked and can thus be used illicitly to generalize or instantiate, like the above case. << >). If we move on to {pira'eci lei ci nanmu cu bevri le pipno} > >can > >only mean {A ija B ija C}. > >I think that would correspond to {su'o pira'eci lei nanmu ...}. > >> >And ordinarily {pira'eci lei nanmu} is {ro pira'eci lei nanmu}, that is, A >ije B ije C? This seems very odd. No, no! {su'o pira'eci} means "at least one third". I've no idea what {ro pira'eci} might mean, that one is odd indeed. A ija B ija C won't be {su'o pira'eci} either, actually. >> I suppose I would have said {pisu'ora'eci} if I had wanted to say that. But your point is that I need XOR and that is correct. << a device (which I can't find >anywhere) for clustering sentences and then specifying properties on the >whole set: "exactly one of" being the desideratum here. I don't think we have such a device. We have {pa lu'a} to select exactly one member of a sumti, but nothing like that for sentences, as far as I know. >> I think we bunced it about a couple of times and maybe even got it to lexed status, but I can't find it -- I can't even think what to search on. The need is rare but real -- as any attempt to express what I want succinctly shows. |