etymology of karN

[Dominique.Thillaud]

S. Palaniappan answered me:

><<        My point of view:
>         Semantically, it seems we have a general meaning "prominent,
> protuberant": Russian "celo" is also "front" and Greek words give related
> meanings: "kolophOn" (top, achievement); "keleontes" (vertical weaving
> loom's uprights). But the meaning "to pierce" is in other words: "kolaptO"
> (to gash, to peck, to stake out); "kolios", "keleus" (woodpecker).
> Actually, a metaphor "to pierce"/"to rise up" is known by the French
> "percer" which can be used for "to become famous" for a man, or "to come
> up" for a plant.
>         Phonetically there is a difficulty in Greek with the initial stop,
> the forms excluding a labiovalar ("po", "te" expected); an evolution *qo >
> Greek ko is just known in the case of a dissimilation by "u": boukolos <
> *gvou-qolos (cattleman, see gocaraH) against hippopolos (horseman).
> Possibly, we can imagine a *q(near L) > Gr. k as in *wLqos (vRka) > Gr.
> lukos (the L in *qL becoming normally "ol" in Greek), but there are other
> explanations of the difficult word "lukos". Moreover, we don't have never
> *qe > Gr. ke. Perhaps the easier way would be to suppose a root *kel-,
> fitting well for the meaning with Latin "celeber" (famous), and to explain
> karN (*zarN expected) by an labialization induced from "l" (the same thing
> occuring in Lituanian)???
>         Mythologically, that's easy to understand the hero's name karNa as
> "famous, out of the common": eldest son of kuntI, despite his obscure birth
> he is supposed becoming even better than arjuna. The reference to the
> earrings could be a secondary motivation of the name (his Greek parallel
> Glaukos bears just a golden plate).
>  >>
>
>Am I right in understanding that the IE etymology is not very satisfactory?
>Before I post the following in Indology, I want to get your view.

        You're perhaps right.
        But what is "very satisfactory"? Personnally, I'm "normally"
satisfied by this IE analysis where the only problem is the supposed
assimilation *kol-no- > *qol-no- which would need a proof. Usually, such a
proof needs two parts:
        - phonetic: it must be acceptable regarding general rules. That's
here the case: the "l" sound is known to have a labial feature (see in
ancient French the evolution chivals > chivaus which give the modern
irregular plural cheval/chevaux) and the "o" being also labial is able to
inforce the process of anticipation.
        - other examples of such a treatment: finding a specific groupment
of phonems attested in the specific languages is not always easy, but here
we can consider the old high German "klagOn" (to moan, complain, giving
German "klagen") which is coming from a *galgh-. If we accept (not a
difficulty) a semantic comparison with Skr. "garhati", "gRhuH" we have our
example of labialization (*gva > ga but *ga > ja). Unfortunately, the two
well known roots *kel- (to hide) and *gleubh- (to carve, sculpt, bark,
hollow) are not attested in Sanskrit and we are limited to one example.
        Do YOU are "very satisfied"? "normally"? not?
        I, normally :)
        What is the opinion of other Eurindian linguists on the list?

        Alas, I'm strictly unable to comment your own hypothesis on the
word, completely out of my competence.
        Best regards,
Dominique

Dominique THILLAUD
Universite' de Nice Sophia-Antipolis, France