Re: Mapping the Origins and Expansion of the Indo-European Language

--- In cybalist@yahoogroups.com, "Richard Wordingham" <richard.wordingham@...> wrote:

>
> > Can you please explain what do you mean by "fit"?
> > What would be a few salient features of Germanic which does not make it "fit"?
>
> Vocabulary. It is generally reckoned to have a very high proportion of non-IE, or at least, hard to recognise vocabulary.
>

Yes, and the study of this lexicon gives us precious insights about the prehistoric linguistic landscape of Europe. Unfortunately, many IE-ists dismiss this evidence because it doesn't fit their own theory.

> Vocabulary actually pulls it two ways - connecting to Italic and

Celtic, and connecting to Balto-Slavonic. I think that's just evidence of an intermediate position in a dialect continuum of disintegrating IE.

>

Lexical isoglosses are usually the consequence of contacts between neighbouring languages, which in this case also include Uralic.

As I said earlier, the classical tree model is grossly inadequate for the IE family, which appears to be the result of multiple (and possibly complex) language replacement and contact processes.

> Grimm's shift makes it stand out, though that's possibly dubious as a reason to exclude it from other groups.
>

In the traditional model, Armenian has a similar consonantal "shift", although it doesn't bear Grimm's name. IMHO the Germanic and Armenian stops reflect a different system than the one found in other IE languages and which is better described by the glottalic theory and similar models based on macro-comparative evidence.
 

> The sound shifts from PIE don't show much commonality with any other group. *o > *a is one of the few, but that's a fairly common change.
>

As in the case of Grimm's Law, this "shift" appears to be so in the traditional model, but IMHO Germanic and other IE languages reflect an older 4 vowel system /i, e, A, u/ like the one found in Etruscan. According to the Spanish IE-ist Villar, the introduction of a new central vowel (presumably from *H2e), caused the original *A to be backed to /o/ in some IE languages, giving a 5 vowel system /i, e, a, o, u/, while in others both vowels merged into /a/.

My own view is similar to Villar's except that the "intruding" vowel is a non-ablauting /a/ different from the ablauting vowel, which possibly was originally /@/.