Re: [tied] Re: Convergence in the formatin of IE subgroups

On 2006-05-10 18:00, Patrick Ryan wrote:

> That "some authority had decided that Nostratic" (also
> Proto-Language) "is too old" was exactly what was asserted on this list.

Can you name names? "Nostratic is too old" is not an a priori assertion
or an authoritative prohibition but one of a number of possible
explanations (a posteriori) of the fact that the Nostratic hypothesis,
like many other long-range proposals, hasn't fared very well so far.

> The idea that "the methods (that) have been tried and haven't
> yielded much" is due to the lack of proficiency of the
> reconstructionists not to the model.

Oh, really? How do you know the model is sound if no consistent
reconstruction exists? By clairvoyance? That the methods haven't yielded
much is just a plain fact.

> And when you write that perhaps "we are indeed dealing with too deep
> chronologies", you are re-asserting — through a backdoor — the "too
> old" argument assuming the destructive effects of language change
> invalidating reconstruction attempts. Perhaps you do not even
> realize that.

I'm not going to prevent anyone from trying to demonstrate the validity
of long-range groupings. I can only wish such adventurous spirits
success. But somehow the task proves to be enormously difficult -- much
more difficult, at any rate, than the reconstruction of the ancestral
language of a family with the probable time depth of five or six
millennia. Information (and with it, evidence of relatedness) _is_
gradually lost over time as a side-effect of language change. This fact
_must_ make reconstruction problematic sooner or later.

Piort