On Thu, 05 Dec 2002 00:15:09 +0000, "Glen Gordon"
>>We have different Ablaut grades *g^e, *g^o, *g^ (*g^he, *g^ho,
>>*g^h). In word final, *-g^ (> *-k^) is attested in [...]
>That's beside the point. The *-g- in *ego: is the emphatic
>_particle_ (not suffix) which by its nature must have had at
>least one syllable as it is ATTESTED! There is no **-g. The
>final vowel of the much-used pronoun has simply eroded away in
Vowels do not erode away just like that. For instance, I'm not aware
of any vowel eroding in Hittite, and we still have uk, ammuk, zik, tuk
in that language. In these personal pronouns, the particle was simply
univerbated at an earlier stage, and became subject to zero grade.
That the particle remained productive and we have attested froms with
*-ge (e.g. Greek eme-ge) is not a problem. The demonstrative *to-
remained productive and was even affixed anew to nouns and verbs (Arm.
article -d, Slavic 3sg. -tU?), but surely that doesn't mean that the
earlier 3sg. verbal ending *-t, or, as you claim, the n.sg. ending *-d
cannot have the same origin?
Latvian and Armenian <es> show that *g^ stood in word-final position.
If the vowel had simply "eroded away", we would have *e:z and *ec.
That's enough to prove that *eg^ is as real as *eg^hom (I-I) and
*eg^o: (Lat & Grk).
Miguel Carrasquer Vidal