To give an idea of some of the alternative suggestions about the prehistory
of the PIE personal pronouns...
Szemerényi's treatment can be summarized as follows (Introduction, pp.
216-220):
- the form *eg(h)om ~ *eg(h)o: is secondary, *(e)m- is primary (verbal
ending is *-m(i), not **-eg)
- the plural forms were:
1. *mes ~ *mos -> n.s-mes
2. *wes ~ *wos -> us-mes ~ us-wes
Under influence of *ns-, *mos became *nos.
1pl. *wei- is secondary (*we- "2" (dual!) + *-i (plural)
2pl. *yus is pronominal *y- + 2pl. *wes. ".. it seems clear that *wes is
itself simplified from *twes, the regular plural of *tu (cf. *mes from
*em-es); this is confirmed by the verbal ending -tes which represents a
simplification in internal position of *twes.
Martinet's theory is given in the following table ("Des steppes aux
océans", p. 202):
m.. "moi" tw.. "toi"
| |
\|/ \|/
m..s "moi et lui" (t)w..s "toi et toi" \
"moi et eux" "toi et lui" } sl. vy "vous"
"toi et eux" /
|
\|/ "toi et moi" --> got. weis "nous"
lit. mes "nous"
On the loss of t-: "On pourrait penser que *we-s (> skrt. vas) représente
<twe> deuxième personne du singulier, avec chute du t- initial explicable
dans un enclitique exposé à des contextes divers."
Martinet offers no explanation ("restent en dehors") for 1pl. *ns-/*nos or
2pl. *ju-. 1sg. *eg^- apparently also falls outside of the system.
Rasmussen's theory ("The constituent elements of the Indo-European Personal
Pronouns"), and Jens will correct me where I'm wrong, can be summarized as
follows:
The basic structure is person (1. m, 2. t, 3. s) + "stem" /w/ + number (sg.
0, du. -G-, pl. -D-) + case (nom. 0, acc -m(é)). In tabular form:
sg. du. pl.
1 nom (*m-w) *m-w(é)-G *m-w(é)-D
acc *m(e)-w-m(é) *m-w(e)-G-m(é) *m-w(e)-D-m(é)
2 nom *t(ú)-w *t-w(ú)-G *t-w(ú)-D
acc *t(e)-w-m(é) *t-w(e)-G-m(é) *t-w(e)-D-m(é)
3 acc *s(e)-w-m(é) *s-w(e)-G-m(é) *s-w(e)-D-m(é)
In ten steps, these forms develop into the attested forms:
sg. du. pl.
1 nom -- *wé: *wéy
acc *mé(:) *n.h3wé *n.smé
(*noh3) (*nos)
2 nom *tú(:) *yu(:) *yú(:)s
acc *twé(:) *uh3ó *usmé
(*woh3) (*wos)
3 acc *swé(:) *sphé *smé
(*sphoh3) (*sphos)
The 1sg. nom was replaced by *eg^.
I will only spell out exactly the developments in the 1 & 2pl. nom. forms:
1pl. *mwéD 1> *wéD 2> *wéy
(1: mw, wm > w; 2: D > y / (a/e/o)_#)
2pl. *twúD 1> *DwúD 2> *DúD 3> *yúz 4> *yús
(1: tw- > Dw- [before -D?]; 2: wu > u; 3: D- > y-, D > z / (i/u)_#; 4: -z >
-s)
=======================
Miguel Carrasquer Vidal
mcv@...