> Dene-Caucasian SinoTib AC Basque
> > *m-lir "ear" *nli? b/ni? belarri
>The TB word has a -a, never a -i, unless you know a language that
>has??..
[lui :]
Hmm, well I wouldn't protest to an ST form *nlya? either.
[moi :]
Well, -ja in TB gives -ya in tib. and in burmese. It cannot be the case.
Thought you'd like that one. I think it's better than pleading for an
unlikely "borrowing" from an AC form with an extra /-p/ and a different
vowel.
Well, analogy took place AFTER it was borrowed.
>>Huh? Where does the /-r?/ come from? I think you mean that b/lhuj?
>>from *sluj? so that we can explain the otherwise problematic lh- >>
sound,
>>just as AC /lha/ is from *sla? "rodent", yes?
>
>I told you already, sy- in MC comes regularly from b/lh-.
No problem. I'm not reconstructing AC with /sl-/, I mean _SinoTibetan_
*sl,
although it would seem that AC /slh-/ would still be proper since the
sibilant seems to carry through to Chinese unaffected.
[moi :]
slh- in AC would give s- in MC. Sorry for that one. However, You can
still argue
that there was an iambic prefix that disappeared without leaving
traces. Difficult
to prove internally in AC.
>Final -j reflects often -r or -l, as final -n does also. This is
>reflected
>in early Zhou riming where words in -j and -n rime with >each other,
and of
>course in word families and phonetic series.
>Note that AN has a very clear root luR "flowing", possibly related to
>chinese.
element *il- - perhaps "falling water". The AC finals could be
"rhyming"
because these phonemes share a common characteristic like
palatalisation,
resonance, etc and not because they are the same phoneme - that's
nonsensical. They are written differently because they are different
phonemes, duh. :P
[moi :]
You are mistaking articulatory and acoustic resemblance. -l, -j and -n
are articullatorily very similar, that is why and -l, which is harder
to pronounce at the end of a syllable because it implies a burst at the
end to be discriminated (or it changes to velar -l, that changes the
formants of the preceding vowels to be more perceptible) can change to
-j and -n. However, these three sound are widely different on a
spectrogram,
nobody would mistake one for another.
Rimes are based on perception rather than on articulation aren't they ?
Besides, if what you said were true, how can you explain that only some
words in -j rime with some words in -n and that there are often xiesheng
and word-families relationships between them. Finally, how come words
in -j and -n rhyme with each other only in the ldest parts of the Book
of
Odes (sywijX < b/lhur? rimes with -n in ode 183, which was composed
roughly
in the 8-7th century), and never in the younger parts ? It suggests the
change
-l > -j /-n (dialectal isogloss, and then dialect mixture) occurred in
the 6th century BC. I think almost any serious specialist in AC
phonology
agrees with that (well, there are still "irreductibles" in Taiwan, Japan
and China, but do they really count ?)
Speculation without any substantiation is never a good rebuttle. Your
point
is that there is no observable /-h/. So DC *? can very well become AC
/h/
and it's not an unusual sound change either.
[moi :]
I think I am less speculating than you are. To prove it, I would need
to
find a language that loaned -h and -s words with otherwise similar
sound
correspondance with AC. However, this is not yet ready although I think
my
hypothesis is testable.
I now wonder if one can find this same contraction in the proximate
Na-Dene
too which contains oddities like pre-glottalized final nasals which
seem
perfect for my evil linguistic purposes. It's really looking less and
less
like SinoT is close to AustroN, don't you think? Or are you still in
denial?
[moi :] Your arguments me laissent complètement froid. You say sometimes
interesting things but I think your attitude as to AC / TB is alarming.
You
seem to accept any reconstruction without knowing the phonological
systems
of all the languages you are talking about (eg : your "ST" ***lnya?).
The NWC-SinoTibetan connection also serves to explain the eerily close
similarity in numerical systems such as the common word for "nine" with
velar. This would lend clout to my view that Starostin is wildly off
the
[moi :] Well, is a velar initial enough to establish cognacy ?
>By the way... Starostin proposed a comparison of chinese bjun < b/byn
>"big horned goat with various TB words. However Sagart showed that
>the
>primary meaning was 'big horned', not 'goat'. Sorry for that one.
I don't see how this is necessarily unrelated within SinoTibetan. Terms
such
as these can end up having slightly different meanings anyway. Consider
IE
*loks-. Is it "salmon" or "salmon trout"? How does Sagart go about
proving
such an impossible arguement? How does he know that it truely meant
"big-horned"? My, there's alot of information encoded in them-there
syllables! :P
[moi :]
Sorry Glen, you are not a specialist of old chinese, so don't talk about
things you don't know anything about. You didn't even look up a
dictionary
of old chinese, did you ?
I think the problem with Starostin is that he doesnt know properly his
old
chinese (he doesn't speak mandarin anyway). The word bjun < b/byn is
very old,
it has many derived meanings, such as "big tortoise" (religious item)
or "big cauldron". "big horned" is one of them. It is written with the
same
xiesheng. "big horned" is a rather marginal gloss. This word is quite
uncommon
in texts.
Guillaume