--- In cybalist@..., "Piotr Gasiorowski" <gpiotr@...> wrote:
> ----- Original Message -----
> From: tgpedersen
> To: cybalist@...
> Sent: Wednesday, April 03, 2002 2:31 PM
> Subject: [tied] Re: Misra, Bryant and Indigenous-Nationalist
> > One argument (cogent or otherwise) would be the Austric(?)-
Sumerian-AfroAsiatic-PIE extended Manansala list
> > but we are not many believers in that.
> No, you aren't ;).
> >> ... That would require a number of really massive but
archaeologically unsubstantiated exoduses -- a high price to pay for
abandoning a model that requires a single movement in the opposite
> > A-hem. Didn't you just say that migrations rarely leave
archaeological traces? How high is the price then?
> I didn't say they rarely do. I said they often don't, which is not
quite the same assertion. Anyway, it's easier to accept a model that
assumes _one_ migration wave leaving disputable archaeological traces
(there are quite good candidate cultures one could associate with
common Indo-Iranian, and it's just the actual entry into India that
causes still unsolved archaeological problems) than a model that has
to propose several such waves (some kind of periodic branch-
generator) leaving India quite invisibly.
> > The contact has always been assumed to be from west to east.
> Has it? There's little doubt that early Indo-Aryan absorbed more
than one non-IE substrate, including languages related to modern
That's no movement. Never mind.
> > How do we know that the loaning (and of course there are clear
west-to-east examples) wasn't in the other direction? Eg. the snake-
> The snake : na:ga connection is fanciful. Since when do non-IE
loans show the mobile *s-?
At some time mobile *s- was functional, rather than a relic; Møller
equates it with Semitic causative *s-, and such a function for mobile
*s- in IE would make at least *some* sense (e.g. melt/smelt).
Whatever was borrowed before that time (if it's not some Ur-
vocabulary, à la the "real" Albanian words in Albanian) would have
participated in formations with *s-
OE snaca and ON snákr are probably 'crawlers', connected with OHG
snahhan 'crawl', while the Indo-Aryan word, if not a non-IE loan, may
represent *nogWo- 'naked creature' > 'snake; elephant'.
A confession: even Manasala kept the two naga's apart. I merged them
on the list since they began to encroach on each other's semantic
territory as I searched wider.
I don't know if it's just me; somehow "nekkid" doesn't make me
> > As to heterogeneity in the supposed center of origin - yes, I
believe in that too, but where is the heterogeneity of dialects
around the city of Rome? Perhaps after the last non-IA migration some
pre-proto-As^oka came along and spread one language in India?
> There's _a lot_ of dialectal variation in Italy, and some of it
still reflects dialectal splits within early Romance (rather than
purely inner-Italian divisions).
I don't doubt that, the question was whether you find greater
linguistic diversity near Rome than in any other Romance-speaking
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.
Georgiev was able find a layer of "foreign" words in Greek (those in
*ps-, *ks-) where no one else had looked (although I don't know how
mainstream his ideas are now). Did anyone look for IE, non IA stuff
The only possible exception known to me is the putative centum
substrate in Bangani, but that's in the Western Himalayas (north of
Dehra Du:n), on the northern fringe of the Indian linguistic area,
where a single minor "lost tribe" (like the Tocharians) may have
ended up at an unknown time in the past, coming from Central Asia.
The Nuristani languages (not to mention Burushaski) found refuge not
far from there.