[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
xorxes:
> la djorden cusku di'e
>
> > Yeah I don't think anyone thinks {ri} is vauge (except maybe xorxes)
>
> I don't think {ri} is intended to be vague. I do think
> there are hard to determine cases (embedded sumti
> for example). Also, I don't know if there is an authoritative
> list of which KOhA can be antecedents of {ri} (ce'u, ma, zo'e?,
> ....?) and which can't (most of them)
There isn't, afaik. But as part of the BF job I'll be trying to draw
one up. The ones you mention are straightforward. Ones like ma'a are
trickier. But ultimately it comes down to which KOhA do and don't
obligatorily preserve reference when repeated, so that is a general
problem that has to be resolved anyway, not just as part of sorting
out the rules for ri.
--And.