Re[2]: [tied] Mother of all IE languages

At 18:14:32 on Wednesday, 3 December 2003, S.Kalyanaraman
wrote:

> Swadesh list is not the population representing languages.

This is incoherent. In any case I know what Swadesh-like
lists are. I also know their underlying rationale.

> Thundering conclusions sought to be drawn from a sample
> (which itself is a biased list) cannot be accepted as a
> good design for statistical tests.

I'll worry about it when you find someone actually making
calculations as if the Swadesh-like lists were random
samples. Possibly someone has done so, but I haven't seen
it.

> I do not accept everything in linguistics as sacrosanct.

I doubt that anyone here does. In any case this comment,
like what followed it, is a non sequitur.

Brian