On Wed, 02 Jul 2003 02:25:08 +0000, Rob <magwich78@...> wrote:

>--- In Nostratica@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>
>> Where does that leave *swe/*sewe, which look like they're structured
>> exactly like *twe/*tewe?
>
>I must admit that I am unsure as to the answer to this question.
>Perhaps there was another pronominal stem *se(i)? However, I was
>told on Cybalist that the root of Latin su- "self" was not a 3rd
>person pronominal stem.
>
>What is your origin for *so and *swe/sewe?

I think *so and *to are derived from the same demonstrative root. *so is
from the old active/ergative *tu-, while *to continues the old
inactive/absolutive *ta-. As in all non-personal pronouns, these roots
were extended with the anaphoric pronoun *iz, giving:

nom.anim. *tu-íz > *sWóz > *s(W)ó
acc.anim. *ta-ím > *tóm
neuter *ta-íd > *tód
oblique *ta-á- > *tó-

The third person pronoun *su (analogous to 1. *mu, 2. *tu), made an
accusative/genitive *sú-ma > *séwe (*swé with shift of accent). The
meaning got changed to a reflexive pronoun, and the non-oblique forms were
lost. The plural and dual forms mostly merged with those of the 2nd.
person (e.g. Hittite enclitic -smas "you, them").

>
>> Latin nos is derived from the plural oblique *n.s-mé, which is from
>> *mes(W) "we". The *n- is not original (cf. vos < *wos < *ws-mé < *us(W)
>> "you"). There is no 1 pl. *n- in (pre-)PIE.
>
>Hmm. But doesn't PIE *-n.- give Latin *-en- (> -in-)?

The dual and plural enclitics go back to PIE. The accusative forms were:

1pl. *ns-mé
1du. *nh3-wé

2pl. *us-mé
2du. *uh3-wé

The enclitics were derived from this by chopping off the accusative ending
*-mé, *-wé and inserting *o (compensatory lengthening?) in the remaining
form:

1pl. *nos, *no:s
1du. *noh3
2pl. *wos, *wo:s
2du. *woh3

>> The same alternation -t-/-j- in the plural possessum forms is also
>found in
>> Eskimo-Aleut, e.g. Yupik:
>
>What's the reason behind such an alternation?

My theory is that the plural endings were *-atu, oblique *-ati. This gives
IE *-esW, *-ey, PUral-Eskimo *-d, *-d^. In the forms with plural
possessed, the possessed is in the oblique, while the possessor is in the
nominative/ergative.

"my house" kota-m& -> kotam&
"my houses" kotad^-m& -> kotan&

"your house" kota-d& -> kotad&
"your houses" kotad^-d& -> kotat&

In the third person, the construction was apparently slightly different:

"house of-him/of mother" kota sa-n / kota emä-n
"houses of-him/of mother" kotad^ sa-n / kotad^ emä-n

In the Auslaut *-d^ > -j, and only then is -sa/-sä agglutinated, so we get:

kota-sa(n)
kotaj-sa(n)

>> Elsewhere the *-n- of the 1st persons was analogically extended:
>>
>> 1. -n-n&/-n-m& -n-n&k/-n-m&k (Finnish: -ni -mme(k)
>> 2. -n-t& -n-t&k *-nti -nne(k)
>> 3. -n-sa -n-sak -nsä -nsä(k))
>
>I disagree with the Finnish reconstructions. Your *-nti for 2sg
>possessor would've given *-nsi via assibilation. The Finnish 2sg
>possessive suffix is instead *-si, with no nasal element.

That's because in Finnish, the possessives with singular, dual and plural
possessed have merged. There were once 2 different paradigms:

sg.poss. pl.poss.
-mi -me(k) -ni -nme(k)
-ti -te(k) -nti -nte(k)
-sä -sä(k) -nsä -nsä(k)

The have merged as Finnish:

-ni -mme
-si -nne
-nsä -nsä

>> You mean plural nouns?
>
>Perhaps.

You said:

-When it
-was deemed necessary to distinguish grammatical number in pronouns,
-typically a plural formant was added to the preexisting personal
-suffixes (this was often the same as the plural formant for the
-plural pronouns)

which makes no sense. I assumed you meant to write: "... which was often
the same as the plural formant for the plural nouns"

This is not the way it usually works. There are plenty of languages that
do not distinguish singular from plural nouns, but still distinguish number
in the pronouns. I can't think of any language right now where the
converse is the case.

In fact, the Uralic nominal plural (nom. *-t, oblique *-j) is in origin a
pronominal plural. The Nostratic nominal plural was probably *-abu, *-abi,
as we see in Kartvelian pl. -eb(i), PIE obl. *-bhi-, G.pl. *-om, and
perhaps Egyptian -aw. A trace of it is perhaps still present in Mari,
which has the nominal plural -Blä, -Blak, which apparently contains the
Uralic collective suffix -la (reinforced -la-k) appended to *-B.
Apart from Mari, the plural in *-B was lost (probably for phonetic
reasons), and was replaced by the pronominal plural *-t / *-j (which was
perhaps already attached to definite nouns in the plural).


=======================
Miguel Carrasquer Vidal
mcv@...