Bruce Gilson skripted:
 
> > ...  I think my conviction will be borne out, sooner or
> >later, that it is a straight choice between t-derivation and POS
> >marking a-la-Ido.  This should not disturb us too much, as each is
> >indicative of a different level of schematism, and of a different
> >primary target audience.
>
> I really find this incomprehensible, and I would like James to explain. I see
> that there _is_ a conflict, but it has nothing to do with PoS marking at all.
> After all, even without the -r, there would still be all the difficulties
> in getting all the -ione forms correct (I assume this is what he means by
> "t-derivation"). I think the conflict is between recognizability and
> regularity, because natlangs have so many irregularities, and we would like
> to generate the largest number of recognizable forms with the smallest number
> of rules and exceptions to rules.
 
It's quite simple actually.  If you stipulate universal POS or word-group
marking, you will want a single vocalic ending for your verb ground forms.
But then, for t-derivation to work, you need four different vocalic
endings for ground forms.  Therefore, we have started from the assumption
that POS marking and t-derivation can coexist, and arrived at a contradiction.
(In maths we have a nice symbol for this, a sort of big cross with four
dots between the lines, but I won't try to reproduce it here!)
 
> I feel we've got a pretty good solution. James and Marcos probably disagree.
> I await the results of their work to see if they can do better.
 
Well, all you have to know about mine, is that I have t-derivation and
therefore the four vocalic endings.
 
James Chandler
[log in to unmask]