Torsten: >They are indeed very strong reasons, but for what? How does
the law
>of palatals argue against PIE presence in India? Suppose it took
>place there, after the non-II branches left?
Something like that could have happened but to prove it would involve
a major reworking of the law of palatals and assimilation of the
vowels. So far, no one has thought about this much.
> I suggested re-introducing Cuny's old proposal
>that the Ablaut vowel, traditionally rendered e/o/nothing should be
>represented as รค/a/nothing, from which (but after palatalization)
>there wouldn't be far to Indo-Iranian's a/a/nothing.
Where can I find more information about this?
Piotr:
>Since there's no linguistic evidence of a primary split into Indo-
Aryan and non-Indo-Aryan (whereafter the latter would have moved to
a "secondary homeland" to split into more branches), the only
solution that respects the structure of the IE family tree (barring
the possibility of pre-Indo-Aryan first migrating out of India and
then coming back home to the subcontinent, which would satisfy
neither the indigenists nor the migrationists) is to have several
successive IE migration waves leave India and populate Europe and
parts of Asia over the millennia.<
I asked an Indologist friend who is doing some thinking about this
(although he is far from convinced) and he thinks the only likely
scenario is that if PIE was in India, it was not Sanskrit but a
dialect that had many non-IA features like centum forms, etc. This
dialect migrated to Central Asia (BMAC) and became what is generally
considered PIE as proposed by Nichols and Lal. Some remnants of this
or its related dialects *might* be seen in centum languages like
Bangani and Tocharian. A different IIR dialect would have emigrated
in order to form Iranian. Sanskrit continued its separate evolution.
Thus you require only one migration (plus the smaller Iranian one.)
This PIE would then have interacted with Uralic/Finno-Ugric since it
is defensible that the loans are only one-way into those languages
families. After it left India, you can take your pick about where it
might have settled to play the role of PIE.
>Areal and substrate studies don't favour an Indian homeland.
Substrate studies could be just as easily explained as adstrate.
Regarding areal studies: It is obvious that IA is unique and that
many words for plants and animals are lost at the Hindu Kush. But
these could just as easily have been lost LEAVING India when they
resided in Central Asia and they no longer needed the words for South
Asian flora and fauna that they didn't see anymore; or at least given
different meanings.
> One node deeper, there are uniquely Indo-Iranian words that have
not diffused to the other branches....Then, there are several
historical layers of borrowings -- including those from pre-Proto-
Indo-Iraniana and Proto-Indo-Iranian proper -- in Finno-Ugric.<
Both explained above.
>What's also suspect is the genetic homogeneity of the IE languages
in India. All of them belong to the same subbranch of IE, and there
is no evidence of former differentiation -- an unlikely situation for
a homeland area.<
There are several historical examples of later language movements
destroying earlier languages. The later Scythian/Iranian and then
Turkic advances are suggested as reasons why there is no direct
evidence for PIE in Central Asia. Also, Nichols points out that
homogeneity is actually a sign of a homeland and diversity is found
on the edges.
It is very clear that there were many Sanksritic and other IA
dialects in India that have been lost. Witzel has written several
papers on this. Also it seems that the modern IA languages aren't
directly connected to recorded Sanskrit -- there was another dialect
that gave rise to them. So we already have evidence that at some time
in the past the Indian languages became homogenized for some reason
perhaps due to Brahminical influence during the freezing of the
language in the Vedic texts.
Tocharian and Bangani are some non-homogeneous examples. More might
surface if they were sought out. The predominance of the AMT thinking
has put blinders on much of this research.
>Even if we assume that IA spread at the expense of other IE
languages in India, it's funny that the latter should have left no
substratal, toponymic or any other traces, while numerous non-IE
languages have. <
These non-IE words could actually be part of the original Indian PIE.
Something is said to be PIE if it is found in several IE languages.
If words were not part of the Indian PIE that left India, or if they
were lost during the sojourn in Central Asia, then they would not
appear to be PIE because no other language would show them.
>I sincerely hope that unless the Bangani substrate should prove to
be an illusion (which I don't think it is: Claus Peter Zoller's
defence of his original findings is very convincing) we shall hear
more about it; <
Hock has also come out supporting Zoller. It seems that Van Driem was
too hasty in denouncing him.
>(c) The history of movement of I-I speakers down through Iran into
India,which can be traced in toponyms.<
If these are valid (and some of them are questionable) they could
just as easily be evidence of going the other way.
Same with much of the archaeological evidence.
>Isn't it so that Sanskrit shares northern animals with the west,
southern ones with the east?<
Where could I read more about this?
>What about the words for 'winter' (*g^Hjem-) and 'snow' (*sneigWH-),
which seem to be the most widely and most securely
attested "weather/climate" terms in IE?<
India has lots of snow and winter. The Himalayas and the Hindu Kush
are both on the borders of the proposed Indian Urheimat.
>The borrowings in Finno-Ugric come partly from Proto-Indo-Iranian,
which means that they are more archaic than either Proto-Indo-Aryan
or Proto-Iranian forms.<
The proposed Indian PIE going to Central Asia would handle this.
>There may be still older (Proto-Satem [?] or dialectal PIE) loans in
FU, sometimes also in Samoyedic, though these are not uncontested<
I contest them! :-)
But where can I read more about it?