T. A. McLeay writes:
> Henrik Theiling wrote:
> >> -ln-, -ls- clusters might be affected, however. Personally I'd limit
> >> it just to the one env., -l + plural s. It could happen, as we say.
> >
> > Ok, we'd get -i:s, -eis, -ais, -ois, -uis.  Some of these dipthongs
> > will only exist in this very place, I think, so there is not much
> > chance of collision.  But it would introduce more dipthongs, something
> > I wanted to avoid, too.
> Perhaps then:
>    -i:s, -eis > -i:s, -ais, -ous > -u:s, u:s
> which gives collision (which you might not like, but I certainly do ;)

Me, too. :-)

This is similar to:

-ous > -o:s
-eis > -e:s

which is also reasonable (cf. Japanese), but then, the whole shift is
just like assimilating -ls to plain -s.

> and a little bit of interest. But then, long vowels are non-phonemic in
> Terkunan, aren't they?

Yes.  Even more collisions!