>From: Christophe Grandsire <[log in to unmask]>
>And probably the 30-
>odd language families we know exist in the world (there are maybe more or
>I'm taking an average number) come from 30 ancestral languages which, like
>11 right now, were spoken by a majority of the world population.

All thru reading and answering this series of posts I've been thinking about
a chart I once had in PDF format of what has been worked out about human
ancestry and language families. It took the language families back a certain
way, then continued the tree-structure of earlier common ancestors on the
basis of genetic and demographic considerations. I'm sure this sort of
research is what spurs on Greenberg and the rest of the "long-range"
historical linguists. Unfortunately I no longer have that chart, and I can't
remember what it was called. :( I'd like to find it again, it is very
pertinent to this discussion.

>The thing is that we know for sure that all IE languages come from a single
>ancestor from one reason: that's the basic requirement of comparative
>reconstruction. You cannot do reconstruction between languages which don't
>a single common ancestor. The very fact that we manage to reconstruct a PIE
>a proof of its existence as a single language. It was not a uniform entity,
>full of dialects and the form we reconstruct was already on the verge of
>splitting, but it was a single language. If it wasn't, the comparative
>wouldn't have led us to it. It's actually a surprising result, and people
>astonished that so many languages had a single common ancestor (with so few
>loans from other languages), but it cannot be otherwise, since it's if it
>we just couldn't have reconstructed it.

Correct and very succinctly put :)


Broadband? Dial-up? Get reliable MSN Internet Access.