Jens:
> My use of just about any device to get the message across that there
> were cases of lengthening of already-long vowels in the history of
> PIE has been fiercely criticised as theoretically impossible by a
> brain which now fully allows for it in the above statements.
Leave my brain alone. You got your own brain. Whatchoo messin' with
mine for, Mr Smartypants? >:P Besides, it tickles when you do that.
I don't "_now_ fully allow for it". I ALWAYS allowed for it. Obviously
so, if I knew about Estonian and the like! My, you like to twist
other people's words but that's no way to debate unless you wish to
avoid what the opponent is addressing.
In your Danish examples, you know as well as I do that you are playing
grammatical magick in order to induce [a::], [a:::], etc. This is
a different situation than in Estonian where there can be an honest
distinction between short, long and double-long in a single _unaltered_
word. If we were to accept your examples as showing true double-length,
then every language on the planet can conceivably have double-long
vowels but that still doesn't make double-long vowels true units,
that is, when properly accounting for the morphology of the language.
> If it is now acknowledged that [a], [a:], [a::] and [a:::] may coexist
> in the same language
But it doesn't except in the most superficial of analyses. It is in
reality a:-a, a:-a-a, etc. No language operates the way you want it
to in order to justify your preIE madness.
> then it can no longer be dismissed as a priori impossible that there
> was such a stage as *[pé::dz]
My whole point is that we don't need to do this. There is simpler,
like *pa:ts, and that theory does just fine to explain the paradigm
of *pod-.
= gLeN