Re: Q (Caucasian Elf)

> Date:         Sun, 25 Feb 2001 20:21:25 -0600
> From: Danny Wier <dawier@...>
> It would help to look at genetic data from various people of Western Europe,
> especially the Basques and the Celtic peoples (including Galicians).  There
> are some major distinctive features, though I don't remember what.
A good place to get ideas --- but genetic data, like archeological
data, are at best a weak hint to how languages may have moved. We have
simply lost too many of the pertinent details.

How many percent of the native English-speaking population today share
the genetic makeup of the North Sea coast of Germany and the
Netherlands, with an admixture of mid-Italian? That is the time-scale
of the knowledge we have --- one language attested here, a similar one
there, a few thousand years later.

Yes, but, we know how English came to spread like that, you might
argue --- but who is to say that there was no similar cultural and
linguistic expansionism in pre-IE Western Europe? And the influx of IE
itself will have mixed things up for good.

To take the Basques: Yes, they may be a homogenoues population, never
mixed, that always lived around the Pyrenees and always spoke Basque.

They might also have started out elsewhere and been forced into the
area, by Celtic-speakers for instance.

They might be a genetically distinct population that always lived
around the Pyrenees, and adopted Basque when a small, technologically
superior group was forced into the area.

Or they might be a genetically distinct population that was forced
into the area and adopted Basque, which happened to be the trading
language there.

> > I also think that the glottalic theory of PIE makes sense.  The problem
> > of going from glottalized to voiced can be solved by an intermediate
> > stage:
> >
> > "Glottalic" PIE          "Intermediate" PIE      "Traditional" PIE
> >
> >  voiceless (aspirated) -> aspirated            -> voiceless
> >  glottalized           -> voiceless            -> voiced
> >  voiced                -> voiced               -> voiced aspirated
>
> Gamqrelidze and Ivanov's theory!  (Which I agree with by the way; Armenian
> and the Germanic languages are the best testaments to that.  I reflected
> that in my Calistan conlang.)  There may be a gap in the chart; traditional
> /p/ or "reformed" /p`/ is poorly attested and may not be an integral of IE
> phonology.
Isn't it traditional *b that is very rare or missing? That's what
corresponds to *p' in the glottalic version.

I like the idea that the 'glottalized' series was  actually ejectives
--- since, typologically, /p'/ is rarer than /t'/ and /k'/.

> Ural (now grouped with the small Yukaghir family of Siberia), Altaic (which
> probably does include Korean and Japanese) and Dravidian (and its likely
> ancestor, Elamite) seem to make up a fairly tight group of families, an East
> Nostratic branch perhaps.  (West would be IE and Kartvelian and *maybe* AA.)
> Sumerian's place in this is uncertain, and any relationship to Basque or
> anything else (except possibly Inuit-Aleut)
Elamite - Dravidian is not 'likely'. Some scholars think they are
directly related, some think the similarities are due to massive
borrowing. (Is Urdu a semitic language?)

The point is that the only reason why late Elamite cuneiform can be
read is that it's chock full of loan words from all over --- so the
Dravidian material might be borrowed too. The earlier pictographs are
not deciphered yet.

Similar comments apply for Korean and Japanese 'probably' being
Altaic. It depends on who you read.

You're welcome to an opinion on the likelihood of these relationships
being confirmed, but you shouldn't state it as fact. The most positive
thing you will ever be able to say is "there's a general consensus
that it's likely" --- and that's not true yet.

> In Germanic and especially in Scandinavian, there are two forms of Umlaut:
> i-Umlaut (fronting, produces Swedish a-dots, o-dots and y) and u-Umlaut
> (rounding, produces Swedish a-ring).  Swedish, Norwegian and Danish carry
> over this nine-vowel system (or ten if you count the schwa e/a), but not
> Icelandic (which has length distinction).
Swedish, Norwegian and Danish so have length distinctions. In Swedish
and Norwegian you might argue that it's conditioned by the following
consonant, but in Danish it's firmly phonemic.

U-umlaut didn't produce /