On Thu, 29 Mar 2001 20:42:17 , "Glen Gordon"
<
glengordon01@...> wrote:
>Miguel:
>>Unrelated forms (that also includes, unless I'm mistaken, the
>>"woopsies" under /w-/).
>
>Obviously /wupse/ isn't easy to connect to the rest but it would seem that
>/woli/, /woro/, /we/, /wedi/ are relatable. Why is there an objection to
>relating /ga:de/, /waga:de/, /ga:de/ and /ka:di/? Afterall, it all belongs
>to the same grouping and these forms merely rhyme with the f- and w- forms.
>It seems like you're trying desperately to push a *p.wat- reconstruction
>that isn't there.
*p.wat.-, actually. It would be fruitless to speculate on the exact
etymology of Logone and Kotoko "four", given that I know nothing about
these languages except for their numbers 1..10. Suffice it to point
out that the other numerals are also, on the surface, quite different
from the other ones in the same "grouping" (but the subgrouping of
Chadic is a moving target anyway). Paul Newman reconstructs
Proto-Chadic *fwad.@, and that's OK with me, given his status as one
of the world's leading specialists on Chadic (I did some library
research for him when he was teaching at Leiden University, so this is
no hear-say).
>>>So you're saying that the word for
>>>seven in AA was with a */p./... like *sap.x-?!
>>
>>Yes (*sap.Ga-?).
>
>An ejective with a laryngeal right after.
So? Happens in Arabic all the time (i.e. emphatic followed by
laryngeal).
>How odd. Have you ever tried pronouncing your reconstructions?
They're not mine. There may of course have been an intervening vowel
at some stage. The status of what I ambiguously notated as *G is
unclear. It may have been a velar or uvular fricative, or a uvular
stop. It may not have been voiced. Proto-Afro-Asiatic still needs a
lot of work.
Just saw an interesting suggestion by Václav Blaz^ek. He points out
that Eastern Chadic has a form *sab.u "three" (besides *kanu(di)
elsewhere), and that compositions with "three" to make the numeral
"seven" are not uncommon (and occur within Chadic itself), either as
10 - 3, or as 4 + 3. Blaz^ek goes on to link Chadic implosive /b./
with the Semitic cluster /b3/ as in <sab3at->, but it's obviously
preferrable to equate it with the ejective *p. (> Sem. /b/) in medial
position. The `ayn might simply be a remnant of the other half of the
composition (i.e. "10" or "4" and/or anatomical references to the
hand).
=======================
Miguel Carrasquer Vidal
mcv@...