Re: Afroasiatic, Eurasiatic, etc. (was Re: Japanese from Tungus)

Hallo!

On Mon, 31 Jan 2005 00:00:21 -0600,
"Thomas R. Wier" <trwier@...> wrote:

> From:    <joerg_rhiemeier@...>
> > > Actually, John Colarusso (certainly no Nostraticist) thinks that
> > > Indo-European and Northwest Caucasian, though not Northeast
> > > Caucasian, might be related to Indo-European, and he has come up
> > > with a number of cognates and sound change, including the word
> > > for "horse".
> >
> > I have heard of that.  However, to me the NWC languages look very
> > dissimilar to IE.  There are perhaps some phonological similarities,
> > but that reeks of areal or substratum influence.
>
> But the modern languages need not look very similar to be related.
I am aware of the difference between similarity and cognacy.
I am also very aware of the fact that neither multilateral comparison
nor glottochronology allow to establish linguistic relationships
for exactly that reason.  Yet there are hosts of scholars who try
to milk those bulls because the comparative method cow does not
give them enough milk.

> Afterall, Armenian and Arapaho were not recognized until very late
> as Indo-European and Algonquian, respectively, primarily because
> they both underwent a large number of phonological changes at one
> point.  What matters, as far as conservatives are concerned, is
> primarily lautgesetze, although as I have stated that's not the
> only criterion.
And one needs some really wild-out sound laws to connect *m with *s
in first-person pronouns, for instance.  Not to mention the other
vast differences between IE and NWC, such as the NWC languages
being mostly prefixing and head-marking.  A Finnish grammar (and,
to a somewhat lesser degree, a Turkish grammar) evokes a sort of
familiarity in a reader familiar with IE languages which an Abkhaz
or Kabardian grammar does not.

This does not mean that IE and NWC aren't related, but I think
the IE - Uralo-Siberian - Altaic connection holds more promise.

Greetings,

Jörg.