[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
pc: > arosta@hidden.email writes: > << > The solution I described in my original message is a general solution > (ke'a/ce'u goi ko'a zo'u): viz., ke'a/ce'u belong to the localmost > candidate bridi they occur in. > > I agree that backcounting is nightmarish if it's not made fundamental > to the language. > >> > It would be nightmarish to make it fundamental to the language > (sadistic, too). We all agree that forethoughting is a general > solution except for the forethoughting part and that the usual > afterthought devices won't work (unless you mean those > three-levels-up --and two-places--over types that are even less > practical than straight back-counting). No disagreement from me about this, except that I do feel that some sort of strategy that involves backcounting is probably the most effective way to get unambiguous anaphora. Obviously, faffing about with xi-subscripting is a bit traumatic, but were we doing LoCCan3 I'd have certain suggestions involving backcounting, by way of being the least bad solution. --And.