On 05-01-14 16:19, Joao wrote:
> There's > also a trait of PIE in Portuguese sapo "toad" (<*sapp-,
> akin to Slavic zaba ?)
Impossible. The Slavic word is actually *z^e^ba with the common
retraction of *e^ to /a/ after a "shibilant". The source of the *e^ must
be pre-Slavic *e:, otherwise it wouldn't have palatalised the initial
consonant. The *z^ may reflect earlier *g, *gH-, *gW or *gWH, so without
taking any extra-Slavic evidence into account the most likely
reconstruction is something like *g(W)(H)e:bH- (cf. Pokorny's *gWe:bH-).
If you want me to go on speculating about the underlying root, PIE
*g(W)eh1(i)- (or *g(W)eih1-?)/*g(W)ih1- 'sing, call' (= Pokorny's
"long-diphthong" *ge:i, cf. Skt. ga:yati, gi:tá-, Av. ga:þa:, Lith.
giedóti) looks like a promising guess. The *-bH- may be the same
zoonymic suffixoid that we find in Gk. elapHos and Skt. s'arabHá-,
perhaps a compositionally reduced form of *bHuh2-o-/-ah2 'being,
critter' with *-bHu[h2]-o- > *-bHw-o- > *-bH-o-.
Piotr