Archiver > GENEALOGY-DNA > 2006-12 > 1166416968

From: "Lawrence Mayka" <>
Subject: Re: [DNA] Indo-European (was, Aryan)
Date: Sun, 17 Dec 2006 22:42:48 -0600
In-Reply-To: <002d01c72259$ea3d7b40$6400a8c0@Ken1>

> [mailto:] On Behalf Of Ken Nordtvedt
> If that is the order of the linquistic splits, is there not something
> unusual? Celtic is said to have split from Italo-Germanic
> before the
> latter two split according to Atkinson and Gray? Yet
> geographically the
> Celts of Central Europe end up BETWEEN the Italics and
> Germanics.

Various combinations have been proposed, including one that ties
Balto-Slavic with Germanic until a rather late date.

But one approach is to apply the well-known evidential principle of
'surprise'. In other words, two language families in close proximity are
*expected* to have some borrowings between each other, and hence at least a
small amount of such borrowing must be discounted when reconstructing the
original breakup. And conversely, two language families that are closer
than their geographical distance would indicate, may have been united at an
earlier point.

My impression is that the authors may have indirectly taken this approach,
by discarding elements that clearly appear to be borrowings rather than
shared heritage.

> And those
> three dates for the splits are quite close together; throw in
> the confidence
> intervals and what is really being said?

Yes, the bottom line here is the statement quoted by Ellen:
The consensus tree also shows evidence of a period of rapid divergence
giving rise to the Italic, Celtic, Balto-Slavic and perhaps Indo-Iranian
families that is intriguingly close to the time suggested for a possible
Kurgan expansion.

'Rapid divergence' is an understatement; a linguistic 'explosion' might be
an apt description. (Whether or not this relates at all to a Kurgan
expansion, or to some other historical event, is a separate issue.)

This thread: