Re: Reflexivity and {ri}

[John Cowan <cowan@LOCKE.CCIL.ORG>]

la dilyn. cusku di'e

> > (1)     la djan. prami la djan. .i la suzn. go'i
> > (2)     la djan. prami ri .i la suzn. go'i
> > (3)     la djan. prami vo'a .i la suzn. go'i
> >
> > It seems to me that (1) means "John loves himself.  Susan loves John"

la xorxes. cusku di'e

> I think that's undisputable.

dy. cusku di'e

> > and (3) means "... John loves herself", while (2) is either (1) or
> > (3)--probably (3).

xy. cusku di'e

> I agree that (2) and (3) should mean the same, but I'm not sure that
> they mean that Susan loves herself. I have to think more about it.

They mean the same, and they mean that Susan loves John.  Anaphora binding
is done just once, in the first bridi, and the bound meanings are picked
up by the the "go'i" and unchanged  (eager evaluation, in CS terms).

If you want the other behaviour, use "go'i ra'o", which causes the anarphora
to be rebound in their new context, getting the result that Susan loves Susan.

> It is not even clear to me which sumti are supposed to be transparent
> to {ri}. Are lerfu pro-sumti, for example?

Lerfu words are transparent to "ri", although the reference grammar doesn't
say so.  (It will in the next release.)

--
John Cowan                                              cowan@ccil.org
                        e'osai ko sarji la lojban.