Re: [tied] Re: Latin barba in disaccord with Grimm's Law?

On 2006-06-20 17:29, mkelkar2003 wrote:

> The debate is between Grimm's law and Glottalic theory. Depending on
> which way you go would dramatically alter the reconstructed
> vocabulary.

No, it changes nothing as regards the phonemic make-up of reconstructed
words. No phonological contrasts are affected. It's just the phonetic
realisation of the stops (or, more precisely, their phonation type) that
is different. No big deal. One could say that G&I's Indo-European is the
same old IE as before, but pronounced with a different accent.

> Let us read some more "nonsense"
>
> "This contention is illustrated in the evolution of the English
> word "cow" (in German, Kuh). In Sanskrit, the word for "ox" is ganh

The citation form is <gauh.>

> and in Greek it is bous. In the classical system, this word is *gwou
> in PIE,

*gWo:us (nom.sg.)

> where the new system described claims it to be *kwou1.

*k'Wo:us, with a glottalised (ejective) initial stop! It's surely an
important point, given that the whole theory is called "glottalic"!

> Obviously, the classical system describes a word very much similar
> to the Sanskrit, where Gamkrelidze puts it closer to Germanic.

_Only_ as regards the pronunciation of its stops, and even there our
brilliant revolutionaries have to cheat a little. After all, the initial
consonant in Germanic is /k/. It may be aspirated in this position (e.g.
in English), but it certainly isn't glottalised anywhere in Germanic, so
the similarity is less than perfect.

Of course in many other respects it's Sanskrit, not Germanic, which
remains more conservative. But then, in one way or another, PIE is
similar to all its daughters. You can find some interesting archaisms
(lexical, phonological or morphological) in practically _any_ IE language.

> Says
> Gamkrelidze, the new system "has brought the protolanguage closer to
> some of its daughter languages without resorting to such difficult
> phonological transformations as that from /g/ to /k/."

I see. Ejective /k'/ to voiced /g/ (in about ten branches!) is fine and
dandy, but simple devoicing (in two branches) is "difficult". If one
accepts the standard model, Grimm's Law is not difficult at all. After
the change of the *gH and *k series into fricatives (*G and *x,
respectively) the only remaining stops were the *g series. As they
didn't contrast with anything in terms of voicing, their [+voice]
feature was redundant. So what they did was the most natural thing to
do: they switched to the unmarked value (which is [-voice] for plain stops).

> Whether it is *kwou or *gwou is more than just a matter of
> intellectual curiosity, right? I have already quoted BArber (2001)
> to the effect that who changed what and how has an impact on
> deciding the homeland and hence the history of real people.
>
> http://everything2.com/index.pl?node_id=1674658

*kWou- and *gWou- point to different homelands? Just how do they do so?

> "In revising the consonant system of the Indo-European
> protolanguage, Gamkrelidze has also called into question the paths
> of transformation into the historical Indo-European languages.
> Grimm's assumption (known as "the classical system") was that
> Germanic, Armenian, and Hittite daughter languages underwent a
> systematic sound shift and Sanskrit remained faithful to the
> original consonants. Gamkrelidze contests--and, in fact, reverses--
> that idea. The diverging pathways of linguistic transformation,
> Gamkrelidze says, can now be traced back to a convergence in the
> Indo-European protolanguage and its homeland. "
>
>
> "And its homeland" "And its homeland" "And its homeland"
>
> So if you go for *kwou, Germanic can somehow be clubbed together
> with Armenian, Hittite and then it must bolster the G & I theory.
> May be that is why they keep repeating it ad nauseaum as you have
> indicated in the earlier post. By the way, auto plagiarism is like
> murder in academia.

While we're at it, Hittite is not a member of the same club, whatever
G&I say. It has no fricatives or aspirates where the majority of IE
languages have voiceless stops, and there's no trace of devoicing in the
*g series (G&I's *k'). Anatolian spelling systems are one hell of a
problem, but what seems to have happened in Anatolian was the falling
together of the *g and *gH series while *k remained distinct. More like
the Celtic, Albanian or Balto-Slavic pattern than Germanic or Armenian.

Piotr