[YG Conlang Archives] > [engelang group] > messages [Date Index] [Thread Index] >
On Mon, Oct 15, 2012 at 11:04 PM, John E. Clifford <kali9putra@hidden.email> wrote: > On Oct 15, 2012, at 7:53 PM, Jorge Llambías <jjllambias@hidden.email> wrote: > > All we need is two forms that differ ony in > ni/nu. It doesn't really matter where the ni/nu appear within the > form, because "ni'u" will mark that place, whatever it is. The > substitution is: > > le (X nu Y li X ni Y) msteki > = la X ni'u(ka) Y msta > > where X and Y can be anything as long as "X ni Y" is a grammatical > formula. > > Well, X ni Y is (filled out phonologically) a legitimate grammatical form; > the problem is that its syntactic components are not X, ni, and Y. That's right, and they don't need to be syntactic components. > I really > think you have to go to basics here and creep forward as far as you can. La > je frma se je xsle pnsake ni'u si je xsli je pnsaki drxaki ... . The point of "ni'u" is not to repeat stuff, la je frmra se xsle je pnsake nu drxake lo je frmro se xsle je pnsoke ni drxoke mstako A/farmers(A) that for some donkey they own it whether or not they beat it, O/farmers(O) that for some donkey they own it and they do beat it, Most A's are O's. If you don't agree that the above captures (at least one reasonable reading of) the donkey sentence, then there's no point going forward. I do think it captures the meaning. If for you it doesn't, you can stop reading here since what follows won't be of interest to you. And's proposal is that given the similarity of the two arguments of "mstako" above, which only differ by the exchange of ni/nu, we can define ni'u to work in conjunction with a unary predicate "msta" such that the above can be condensed as: la je frmra se xsle je pnsake ni'u(ka) drxake msta co ma'a xrxe