On Mon, 21 Jun 2004 17:56:26 -0700 (PDT),
enlil@... wrote:
>Miguel:
>>>Ur *mi and *ti is to be connected to *mu(n) and *tu(n).
>>
>> So what happened to the /u/? How can your so-called plural
>> forms actually be singulars?
>
>I think you're too ready to reject what I say before understanding
>what I've said and contemplating the evidence.
>
>The forms you cite for Uralic, *mi and *ti, are reflexes of *mu and
>*tu, there's no doubt.
Yes there is. I see no reason to assume that *mu and *tu
would have given Uralic *mi and *ti just like that.
Especially if we take into account the whole picture of
Proto-Nostratic 1/2(/3) sg. pronouns.
The Afro-Asiatic forms are *ni and *ka(m.)/*ki(f.).
Kartvelian has *mi/*ma (Svan mi, M/L ma(n), G. me(n)), *si
(Svan si, M/L si(n)), Uralic has *mi and *ti, Etruscan has
mi. The Afro-Asiatic dvelopment of *m- > *n- is only
comprehensible if the following vowel was *i (mi > ni),
whereas m- was preserved in the plural before -u (Hausa mu,
mun "we"). The fronting of *ki > *ti seen outside of
Afro-Asiatic and Kartvelian (assuming *si < *ci < *ki,
object marker 2sg. -g- < *-k-) is also best understood as
having resulted from /i/-vocalism. combined with influence
from the 2sg. verbal prefix *tV-.
PIE, Altaic and Chukchi *do* have *mu and *tu, just like
Afro-Asiatic has 2sg. m. *ka and 3sg. masc. *s^u. These are
variations that were already possible in the proto-language,
by superimposing the case markers *-u (active), *-a
(inactive) on the personal pronouns (as well as other
pronouns). It's unclear whether this was simple
substitution of -i (pl. -u) by -u or -a (pl. -i), as
suggested by PIE *tu and PAA *ka, or affixation of *-u, *-a
to the full form of the personal pronoun (as suggested by
Sem. s^u < *si-u, unless one wants to reconstruct the 3rd.
person suffix as *s^ instead of *s).
What I envisage is an original system which was simply:
1. *mi pl. *mu
2. *ki *ku
3. *si *su
(becoming *ni, *ki, *si in the South, *mi, *ti, *si in the
North).
Adding the case endings (< definite markers?) resulted in
the variant forms:
[active/nominative]
1. *mi / *m(i)-u pl. *mu-an / *mu-atu
2. *ki / *k(i)-u *ku-an / *ku-atu
3. *si / *s(i)-u *su-an / *su-atu
[inactive/accusative]
1. *mi / *m(i)-a pl. *mu-an / *mu-ati
2. *ki / *k(i)-a *ku-an / *ku-ati
3. *si / *s(i)-a *su-an / *su-ati,
where the first form is pausal, the second not.
All Nostratic forms can be derived from this, if we
understand that IE, Altaic and Chukchi generalized *mu-,
*tu- as the pronominal bases, while elsewhere it was mainly
*mi- and *ti-.
=======================
Miguel Carrasquer Vidal
mcv@...