[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
I don't know why my messages to Jboske are taking 90 hours to get back to me, but for an updated version of the proposal in this message, see Experimental Anaphorics on the wiki. --And. > -----Original Message----- > From: And Rosta [mailto:a.rosta@hidden.email] > Sent: 30 October 2002 01:43 > To: jboske@yahoogroups.com > Subject: RE: [jboske] zo'au, vau > > > xorxes: > > la and cusku di'e > > > > >Is there anything better for it to be than a KOhA? KOhA is rather > > >inelegant > > > > The only alternative seems to be a new selmaho of its own > > So KOhA it must be, then > > > >Basically I want to say that in an ideal world, if X is within > > >the scope of Y, then X is in a lower bridi than Y > > > > This is not how it usually goes in Lojban though, because > > everything in the prenex is typically in the same bridi > > I like to think of each item in the prenex as implicitly starting > a new bridi. It does no harm, and keeps me happy > > > >I'm not sure if I'm being clear. Here's an example (Polish): > > > > > > Ex not Ay and Fxy Gxy > > > > > >can be rewritten with the hierarchical structure made explicit > > >thus: > > > > > > [ Ex [ not [ Ay [ and [F [x] [y]] [G [x] [y]] ]]]] > > > > > >If there was a way to encode this structure other than using linear > > >order, then the order could change without altering the formula, > > >e.g > > > > > > [ [ not [ [ [F [x] [y]] [G [x] [y]] and ] Ay ]] Ex] > > > > > >The closer the eventual solution is to this ideal, the fewer logical > > >problems could ensue > > > > I think what you're saying is that you don't want infixed > > structures, right? > > No, what I meant to say is, in effect, that if we had a way of > explicitly marking brackets (or starts of phrases), then the > head of the phrase would not have to be initial. In phrases headed > by a brivla, it indeed does not need to be initial. But if > a quantifier is head of the phrase it has scope over, in Lojban > that can only be indicated by linear order -- by placing the > quantifier before elements it has scope over. > > Feel free to ignore these remarks. All I'm saying is that in > principle, scope could be represented by some syntactic device > other than linear precedence, and that this would have been > desirable. > > > The problem with allowing the quantifier at either end is > > that then you need to mark the end of the scope as well > > Otherwise, if you allow some to be at the beginning and some > > at the end you can't determine their relative scopes > > > > I think the two workable options are: > > > > (1) the prenex is continued after zo'au. Anything put after > > zo'au works as if it were put in the prenex, in the same order, > > and after whatever is overtly there already > > > > (2) Anything after zo'au has scope over everything preceding, > > including the prenex > > > > I can see advantages for both. But any alternating system > > allowing more flexibility than that would require lots of > > cumbersome brackets > > I propose a compromise: > > Each zo'au is a continuation of a matching zo'u, with zo'u-- > zo'au correspondences nesting. If there is no overt zo'u, > an implicit one is inserted > > A B C zo'u > = (zo'u) ... zo'au A B C > = A zo'u ... zo'au B C > = (zo'u) B zo'u ... zo'au C zo'au A > = (zo'u) (zo'u) (zo'u) ... zo'au C zo'au B zo'au A > = A zo'u zo'u ... zo'au C zo'au B > = A zo'u C zo'u ... zo'au zo'au B > > A B C D zo'u > = A zo'u C zo'u ... zo'au D zo'au B > > This gives the needed flexibility and preciseness, while being > no more cumbersome than is required. > > If you approve of this scheme, I'll add it to the wiki > > It also has the entertaining property of simultaneously being > (a) genuinely useful is actual discourse (indeed useful only > for actual discourse), (b) baseline-conformant, (c) likely to > piss off reactionaries > > --And > > To unsubscribe from this group, send an email to: > jboske-unsubscribe@yahoogroups.com > > > > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ > > > >