Re: STAN vs DCA : round 2

>If you find a copy of Karlgren's grammata serica recensa (1967, The
>museum of Far eastern antiquities), I will tell you the references to
>related phonetic series (this book gives old reconstructions that
>nobody on earth believes, but puts together words with the same
>phonetic, which makes this book an unvaluable reference.

I would have expected you would call it "in-valuable" but since you place no
value on them whatsoever, then I won't either :P

Alright, I'll yield! I will review my views on the form of SinoTibetan based
on what you're saying and swallow my pride... but only very cautiously... I
remain thouroughly unconvinced of a close Austronesian relationship although
I don't deny a remote one to DeneCaucasian as a whole with ST included.

>There are NO real reconstructions for Sino-tibetan. The only I am >aware of
>are Starostin & Pejros 1995, South Coblin 1986 and Benedict >1972.

Hmm, well it's not Starostin that I saw because it's void of laryngeals. I
have to check two things: the author of the reconstruction I'm seeing and
the authors for the SinoTibetan text in Encyclopaedia Brittanica.

So, you agree to the possibility of ST and not denial. Alright. Got it. Now
on to the show...

>>The forms that I recognize are the forms I see from many sources
>>together with much direct and self-evident justification: *nga "I"
>>and *nei "you". No
>
>OK, look at Chunqiu chinese pronominal system :
>
>1sg a/nga 1pl b/ngaj? MC : ngu ngaX
>2sg b/na? 2pl b/naj? MC : nyoX nyeX
>
>Now look at Shang oracle bone chinese :
>
>1sg a/la 1pl b/ngaj? MC : yo ngaX
>2sg b/na? 2pl b/naj? MC : nyoX nyeX

Alright. A pronoun "la" is attested in the first person singular? This could
be related to DC *tLu "we". Certainly this isn't outside the bounds of
reasoning if the pronouns had lost their sense of "plurality" being
interchangeable for either number. A similar thing seems to have happened
between Nostratic and DC whereupon *ni > Nostratic *nu "I, me" [ergative]
(with regularisation of pronouns with *-u) and *tLu > *u "I, me"
[absolutive]. The plural forms were then derived out of the now suppletive
singular pronoun stems. No big whoop.

At any rate, there is still a strong connection. The forms in the plural
still seem to derive from the singular (as per the 2nd person at least) and
the plural forms seem derivative based on their relative complexity to the
singular. Thus there is still room for DC *ni (*tLu)/*ngu in SinoT studies.
You haven't found a second person looking like *Lu lying around have you?

>kulou < MC khu luw < a/kha a/lu (in fact, in my opinion, rather a/k->lu,
>with a prefix.

Skeptical but I'll let it pass due to my lack of knowledge to follow
through.

>>Dual, Trial? Only numerals up to six?? What on earth are you babbling
>>about. This is an assertion of a negative that is unprovable.
>
>provable : australian or papuan languages often lack numerals above
>three. Besides, many languages such as Austro-asiatic count in base >5.

Yes, they are called "Australian" and "Papuan" languages because they are
different from the "Altaic" languages whose Turkic branch, you claimed,
actually lack "six" and "seven", which remains unprovable and worrisome as
an assertion.

>There is a trace of this in chinese :
>"two" nyijH < b/nits
>"seven" tshit < b/s-hnit

Erh, this isn't a trace of much. Granted the forms have influenced each
other because of the obvious semantic relationship but notice the
similarities between *sNit/*bryat (or s-hnit/brat, if you will). Now compare
Basque zortzi/bederatzi. Anything familiar? We should in fact expect "seven"
in ST to be *sryat but the numeral was certainly influenced by "two" in ST
itself.

>AC : baet < a/brat
>tibetan loaned it strangely, it added a yod. Anyway, it might be due >to
>the coloration of the vowel by -r- by the time of loaning. Compare
>bjenH < bron-s siamese plian.

Alright, we'll drop the *-y-. No prob.

>>*minrac, whose reasons for the semantic shift have already been
>>explained). Plus, medial *m regularly becomes Vasconic *w and then
>>often disappears
>>altogether in Basque as in *sulmu -> *hirwu -> hiru or *limu ->
>>*lawu -> lau
>
>funny, it "looks like" AN lima, paiwan rima "5". But it is the wrong
>number.

Note: Nostratic *lil, *lilmu "four"
Sumerian lim, limmu "four"
Uralic *nelj�
Dravidian *na:l
Altaic *n�r(b�n) (Turkish d�rt, Mong d�rben, OJap y�-tu)
-> NEC *libu "three" !! (<-my reconstruction: Starostin NC *lHe)

We would have expected **nele- (or **nene-) in IE if it weren't for that
damn *kWetw�res word. Maybe there is something to the AN *lima form
ultimately.

> > DeneCaucasian *m-hutL "eye":
> >
> > SinoTibetan (OChin my�k, Tibetan myik)
>mjuwk < AC b/mriwk (This word is nasty, Sagart proposed b/mr-liwk to
>explain word family connections but I am not convinced)
>the tibetan word is mig, not "myik".

Whoops, sorry. I knew I would goof on one of these words. Hmm, as far as the
theories that I was in contact with go, the SinoTibetan *r was supposed to
be uvular, not trilled. Is this Ancient Chinese -r- in /mriwk/ uvular? If
so, it could be a voiced attestation of DC *-h- in ST and demonstrates a
regular correspondance of *-tL to AC -k
(*rutL "six").

- gLeN
______________________________________________________