H. S. Teoh scripsit:

> LOL... I presume [tu] here is [t_hu]? That's quite an insult. It means
> "primitive", "uncultured", "aboriginal" (in the stereotypical negative
> sense).

The Monguor (< "Mongol") of eastern Qinghai province are still called
the Tu(ren) officially, despite the obvious pejorative nature of this.
Here's a posting I made elsewhere about an oddity of the
Monguor language:

Monguor is one of the offshoots of Mongolian that arose because the
Mongols conquered China and ruled it for centuries. Most of China's
border garrisons were composed of Mongol-speakers, some of whom took
root. For example, there are a few Mongol-speakers near the Vietnamese
border, others in northeast Manchuria (later relocated), others in the
provinces of Gansu and Qinghai, and some even in northeastern
Afghanistan. Their languages have of course diverged over the centuries
from the standard Mongolian of the Mongolian Republic and (Chinese)
Inner Mongolia.

The speakers of Monguor in Qinghai have been living with (Han) Chinese
and especially Tibetans for a long time. Unlike any other Altaic
language, their language has developed consonant clusters at the
beginning of syllables. Well, nothing strange about that: some short
vowels fell, and the result was a bunch of clusters: it's happened in
lots of languages, including English (more at the end of syllables,
though, as in /lIvd/ < /lIv@d/ 'lived').

But just which consonant clusters arose and which didn't? The Middle
Mongolian verb stem [hudaru] 'destroy' became [xtaru] in earlier
Monguor, and [stari] in the current language. But [hula:n] 'red'
did not become *[xla:n]. (In standard Mongolian, it has become
[ula:n], as in Ulaanbaatar 'Red Victory', the capital of Mongolia.)
What made [xt] privileged and [xl] not?

Well, most Monguor have for a very long time been bilingual in
Tibetan, or the local variety thereof, in which [xt] is a valid
cluster but [xl] is not. So when the Monguor borrowed the Tibetan
word [xtorma] 'sacrificial offering', they did not adapt it to their
Mongolian habits, but took it in directly as [xtorma]. Having
learned to pronounce [xt], they adapted some of their own words
to use it as well. In fact, vowels fell if and only if the resulting
cluster occurred in Tibetan as well.

In addition, the typical Altaic vowel harmony is completely gone
in Monguor.

Primary source: Rona-Tas^, A. "Remarks on the phonology of the
Monguor language", _Acta Orientalia Academiae Scientarum Hungaricae_
10.3 (1960), pp. 263-67.

Secondary source: Robert S. Ramsey, _The Languages of China_.

--
John Cowan <[log in to unmask]>     http://www.reutershealth.com
I amar prestar aen, han mathon ne nen,    http://www.ccil.org/~cowan
han mathon ne chae, a han noston ne 'wilith.  --Galadriel, _LOTR:FOTR_