[YG Conlang Archives] > [engelang group] > messages [Date Index] [Thread Index] >
On Sun, Sep 9, 2012 at 2:13 PM, John E Clifford <kali9putra@hidden.email> wrote: > > (though we have yet to see what Xorban will do with similar situations of > collapsing parallel sentences, for example). For the case of multiple predicates sharing the same argument, no special treatment is needed: la nnla je le zrce klmake le ldre trvcnake Boy-A: store-E: goes(A,E) & milk-E: buys(A,E) For the same predicate applying to several arguments, one strategy is to collect the arguments: la nnla le nxle li gnmikake lo zrco klmiko Boy-A: girl-E: they(I,A,E)-I: store-O: go(I,O) We already have a special union operator in place to make this shorter: la ga nnla nxla le zrce klmake Boy+girl-A: store-E: go(A,E) > As for speakability, > Xorban is on the road to a serious memory problem, keeping all the variables > straight over long stretches of conversation, so it will eventually have to > consider some forms of redundancy to overcome this, whether reupping or some > other device. You shouldn't have to remember variables over long stretches of conversation, since they will be constantly reused. The rule for the binding of unbound variables would normally apply to a variable that appeared in the previous sentence, or even the current sentence, not one from several sentences ago since it's unlikely that the variable would not have been used again. ma'a xrxe