I've used "le du" as equivalent to english antecedentless personal pronouns and definite covert arguments. But "lo du" for "zo'e" seems extraordinary. It's fine where "zo'e" is understood as "something/someone", but seems perverse otherwise.
,, And
On 17 Oct 2011 03:24, "Martin Bays" <mbays@sdf.org> wrote:--
* Sunday, 2011-10-16 at 21:49 -0300 - Jorge Llambías <jjllambias@gmail.com>:
> Where the referent of "zo'e" is roses, yes.
> On Sun, Oct 16, 2011 at 7:23 PM, Martin Bays <mbays@sdf.org> wrote:
> > * Saturday, 2011-10-15 at...
>
> So far so good.
> > == {zo'e rozgu gi'e xunre}
> > == {zo'e xunre gi'e rozgu}
> > == {zo'e noi xunre cu ...
>
> But "zo'e noi xunre" could be changed to "lo xunre" only if the
> > == {lo xunre cu rozgu}
>
> referent of "zo'e" was red things. We already know that the referent
> of zo'e in all the previous sentences was roses, so we can't transform
> "zo'e noi xunre" into "lo xunre".
>
> Alternatively: "zo'e" is not like "ko'a", something that gets a
> referent and keeps it more or less indefinitely. The referent of
> "zo'e" gets assigned by context, and the sentence in which it appears
> is very much part of its context. "lo rozgu" == "zo'e noi rozgu" only
> in the context of defining "lo rozgu". With a different context,
> "zo'e" might get assigned a different referent and thus "zo'e noi
> rozgu" might no longer be replaceable by "lo rozgu".
>
> (All of which may amount to saying that "{lo} -> {zo'e noi}" should
> not be taken too literally, and "{zo'e noi} -> {lo}" is even less
> certain.)
OK.
I wonder then whether "{zo'e} == {lo du}" mightn't be a more accurate
summation of the relation between {zo'e} and {lo}.
(assuming that everything, pluralities and kinds included, dus itself)
Martin