Re: Armenian, Sumerian, Burushaski, and Turkic languages

On May 31, 5:24 pm, "Peter T. Daniels" <gramma...@xxxxxxxxxxx> wrote:
On May 31, 10:04 am, Darkstar <darkstar...@xxxxxxxx> wrote:





On May 31, 2:43 pm, "Peter T. Daniels" <gramma...@xxxxxxxxxxx> wrote:

On May 31, 5:52 am, Darkstar <darkstar...@xxxxxxxx> wrote:

The only problem being that the method you described has never been
consequently

consistently ["consequently" is a German false friend]

Noted.

applied to many well-known language groups and families.
What you're demanding from me, has never been done in much more common
and obvious cases. On this basis, I just suspected you of being
unfair.

Just because _you_ have never bothered to visit a linguistics library
does not mean that such a library does not exist.

Which well-established language families do you claim have never been
studied by the orthodox comparative method?

Armenian is just one good example. It has too many mysterious
"borrowings" from some "unknown Indo-Iranian substratum". I suspect it
has not been studied sufficiently.

(a) Armenian is not a language family, and (b) it is no less well
studied than any other Indo-European language.

Your "suspicions" are of no interest. You have never bothered to find
out the actual situation.

As to Germanic, how about Dalecarlian? Is that a dialect or a separate
language? Or even a special eastern subbranch of Scandinavian
languages? Who ever established regular correspondences for it?

I have never heard of "Dalecarlian," linguists don't do "language" vs.
"dialect," and why don't you look at the bibliography of wherever you
heard of this idiom to discover who has worked on it?

Some other IE examples probably without perfectly regular
correspondences: Dardic languages, Anatolian subgroup.

Your ignorance grows more and more oppressive. I take it you've never
heard of Braj Kachru (native speaker of Kashmiri and linguist who has
worked on his native language family, as well as Indian English and
world Englishes generally) or H. Craig Melchert, whose *Anatolian
Historical Phonology* (1994) is a classic.

Did you yourself read any of these works?

The Sino-Tibetan family is just another typical mess when judged by
your strict standards.

And who died and made you an expert in S-T? Have you studied
Matisoff's comparative phonology of S-T?

No one can ever produce a fully consistent and complete comparative
phonology of hundreds of S-T languages. I don't have to be an expert
in the field to make such an obvious conclusion.

Not to mention, Niger-Congo and many others.

Then don't.

Niger-Congo is one of Greenberg's fantasies -- the most likely of his
inventions to be valid, as it seems. Khoisan apparently isn't, and
Nilo-Saharan keeps changing: Lionel Bender, who knows more about N-S
than anyone else, keeps revising his models.

But that's not what we're talking about. You (and Nathan and the rest
of the modern Western linguistics) just seem to refuse to notice that
statistical conjectures are part of scientific life. You can't exclude
_all of them_ from your research. There are always some unpredictable
elements and phenomena in any complex system that you cannot account
for other than using some kind of stochastic approach. This discussion
of deterministic vs. indeterminisitic principles has been known in
philosophy probably since Epicurus. It has been later reiterated in
quantum mechanics as the famous Bohr-Einstein debate in the 20-s.
Einstein was trying to prove that quantum mechanics is wrong because
it can't fully predict the behavior of particals, while Bohr and the
others were saying that their behavior is in part fundamentally
unpredictable, so they don't actually have to. Etc, etc, etc.

Humans are not machines, and languages are not physical entities
governed by the laws of physics.

That's what I'm trying to say. (Their behavior is not completely
deterministic, that is).

There can simply be no such work for Cushitic or Omotic (as
particularly good example), because the procedure for finding regular
correspondences for hundreds of languages would be much too complex.

Bull***. There can be no such work because the fieldwork has never
been done. If we had as much data for Cushitic languages (where did
"hundreds of languages" come from, incidentally?

["Hundreds" was an exaggeration. Personally, I've only studied
numerals for 25 Central Cushitic, 7 South Cushitic and about 28 Omotic
languages.]

Which says very little about how many Cushitic and Omotic languages
there are ...

You don't even know
how many Cushitic and Omotic languages there are?) as we do for Indo-
European languages, the work would be just as detailed and just as
convincing.

If they haven't been described, on what basis do you include them into
any grouping?

Do you really know _nothing_ about the history of language research,
especially in Africa in the 19th century?
Have a look at the work of such as Koelle and Lepsius.- Hide quoted text -

Can't you really answer a direct question for yourself without sending
me on a wild goose chase?
Especially, such a simple question.


.

Quantcast