On Wed, 02 Jul 2003 20:56:10 +0200, Miguel Carrasquer <mcv@...> wrote:

>the transitive verb inflections (definite conjugation) obviously
>originated as possessive constructions. There was no distinction
>originally between the possessive and the definite verbal suffixes.
>
>The indefinite conjugation, however, was originally distinct.
>
>In most modern Uralic languages, the indefinite conjugation has influenced
>the definite, and viceversa, but the original state of affairs must have
>been something like:
>
> indefinite definite/possessive
> (stative) sg.poss. du.poss. pl.poss.
>1. *-k *-m& *-g(&n')-m& *-d-m& > *-n&
>2. *-n *-d& *-g(&n')-d& *-d-d& > *-t&
>3. *-0 *-sa *-g(&n')-sa *-j-sa
>du.
>1. *-g-m&n' *-m&n' *-g(&n')-m&n' *-d-m&n' > *-n&n'
>2. *-g-d&n' *-d&n' *-g(&n')-d&n' *-d-d&n' > *-t&n'
>3. *-k, *-g&n' *-san' *-g(&n')-san' *-j-san'
>pl.
>1. *-d-m&k *-m&t *-g(&n')-m&t *-d-m&t > *-n&t
>2. *-d-t&k *-d&t *-g(&n')-d&t *-d-d&t > *-t&t
>3. *-t *-san *-g(&n')-san *-j-san

To expand on this:

Proto-Uralic apparently did not make a big distinction between nominal and
verbal roots. Personal suffixes could indifferently be added to nouns
(resulting in stative / possessive) or to verbal roots (resulting in
intransitive / transitive conjugations). The modern languages have
introduced distinctions between nominal and verbal forms, and have merged
and/or eliminated some of the old categories:

Finn. Saami Mari Mordv Udmurt Komi Hung Mansi Xanty Nenets Selq
stative x x x
possess. x x x x x x x x x x x
poss(du) x x x x
poss(pl) x x x x x x
vb (intr) x x x x x x x x x x x
vb trans x x x x x x
vb tr(du) x x x
vb tr(pl) x x x x


Finnish has only a possessive and a verbal comjugation. The possessive
forms are a mix of the old possessive w/ sg. possessum and w/ pl.
possessum:

1. -ni (pl.poss. *-n-(m)i)
2. -si (sg.poss. *-ti)
3. -nsä (pl.poss. *-n-sa)
1. -mme (pl.poss. *-n-mek)
2. -nne (pl.poss. *-n-tek)
3. -nsä (pl.poss. *-n-sek)

The verb has a mix of intransitive and transitive forms:

1. -n < *-m (tr. *-m)
2. -t (tr. *-t)
3. -0 (intr. *-0)
-sen, -zen (tr. *-sa(n))
1. -mme (intr. *-t-mek)
2. -tte (intr. *-t-tek)
3. -t (intr. *-t)
-zet (tr. *-sat)

In Mari, possessive and verbal endings are virtually identical, except in
some 3rd. person forms:

1. *-Vm (tr/poss sg.poss)
2. *-Vt (tr/poss sg.poss)
3. *-s^V, *-z^V (tr/poss sg.poss)
*-V (intr)
1. *-(V)na (tr/poss pl.poss.)
2. *-(V)ta, *-(V)Da (tr/poss sg.poss.)
3. *-(V)s^t (tr/poss sg.poss.)
*-Vt (intr)

Mordvin retains all categories except the dual:

poss.sg. poss.pl. stat. intrans. trans.sg.obj. pl.obj.
-m -n -n -n -0, -n -n'
-t -nt -t -t -k, -t' -t'
-zo -nzo -0 -0 -si -n'z'e
-mok -nok -tano -tano,-tama -n'ek,-s'k -n'ek,-s'k
-nk -nk -tado -tado,-d'ä -nk, -s't' -nk,-s't'
-st,-sk -st,-sk -t -(i)t' -z' -z'

There is also a whole grid of forms with 1/2 sg/pl object agreement, which
appear to be a development within Mordvin (there is considerable variation
between the main dialects). There is some syncretism in the forms given
above (2pl. -nk is a pl.poss./pl.obj. form transferred to the
sg.poss./sg.obj. paradigms).

Komi has a single possessive and a single verbal category:

poss. vb
1. -&(j) < *-Vm (sg) -0 < *-m (tr)
2. -Id < *-Vd (sg) -n (intr)
3. -Is < *-Vs (sg) -s, -0 (tr; intr)
1. -nIm < *-n-Vm (pl) -m, -mnIm (tr; tr+poss.pl.)
2. -nId < *-n-Vd (pl) -d, -dnId (tr; tr+poss.pl.)
3. -nIs < *-n-Vs (pl) -nIs (poss pl.)

Note the transfer of the plural possessum forms to the category of plural
possessor. Noteworthy is also verbal 2sg. -n, from the old
stative/intransitive.

Udmurt only has transitive/single possessum/object forms:

poss/vb.
1 -(V)
2 -(V)d
3 -(V)z
1 -m(I)
2 -dI, -tI
3 -zI, -sI

Hungarian has sg. and pl. possessum forms, and intransitive and transitive
conjugations:

poss.sg. poss.pl. intrans. trans. 1>2
1 -m -im -Vk/-m -m -lak
2 -d -id -sz/-l -(V)d
3 -0 -i -0 -0
1 -nk -ink -Vnk -(u)k
2 -tok -itok -tok -tok
3 -k -ik -k -k

Like Samoyed, Hungarian has generalized -i- (not -n-) in the plural
possessum forms. The difference between intransitive -0 and transitive
*-sa in the 3rd. person has been obscured by the loss of -s- here.
Noteworthy are the 1sg. intransitive -k (as in Selqup Samoyed), and 2sg.
-sz/-l (< *-n', -r'?). There is a special form for 1st person acting on
2nd person, -lak.

For the Ob-Ugric and Samoyed forms, I refer to the Uralic chapter of my
"Nostratic Tour".

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