h₂uelk-

Grammatical information: f.

Meaning: `furrow' (Trag., Com.).

Other forms: Also αὖλαξ (Hes.), ὦλκα, - ας acc. sg., pl. (Hom.), Dor. ὦλαξ EM 625, 37and in ὁμ-ώλακες (A. R. 2, 396). Further εὑλάκᾱ `plough' with the Lacon. fut. inf. εὑλαξεῖν (Orac. ap. Th. 5, 16); and αὑλάχα ἡ ὕννις H. and *ὄλοκες (cod. ὀλοκεύς) αὔλακες H.

Origin: PG [a word of Pre-Greek origin]

Etymology: The relation between these forms was unclear. Solmsen Unt. 258ff. explained ὦλκα from *ἄϜολκα ( κατὰ ὦλκα Ν 707 for original *κατ' ἄϜολκα); it is strange that this form did not live on. Beside *ἀ-Ϝολκ- the zero grade would give *ἀ-Ϝλακ- in αὖλαξ. The root was supposed in Lith. velkù, OCS vlěkǫ, Av. varǝk- `draw'; one could assume * h₂uelk-. This is tempting, but must not be correct. If the Balto-Slavic words are isolated (there is further only Av. vǝrǝc-), the verb may be non-IE; also it is rather * uelkʷ-, which makes the connection with Greek impossible; further there is no trace of the verb in Greek, which has ἔλκω \< *selk-. εὑλάκα can no longer be explained from different prothesis, *ἐ-Ϝλακ-. But ἄλοξ cannot be explained in this way: metathesis of *αϜολκ- would give *αυλοκ-; an after the F had disappeared, metathesis was no longer possible (only contraction to *ωλκ-). - I see no reason to reject ὀλοκ-. ὦλαξ was perhaps taken from a compound, like ὁμώλακ-, which would give *ολακ-. - Pisani JF 53, 29 derived αὖλαξ from αὑλός and separated it from ἄλοξ etc., which is improbable. - The variants are strongly reminiscent of substr. words, as Beekes Dev. 40 held (withdrawn ib. 275-7). Variation of prothetic ε\/α\/ο\/αυ\/ευ is typical of substr. words, as is κ\/χ ( αὐλάχα). So more probably we have to assume a substr. word. The start with the Homeric form was wrong: it is the only form that has no vowel between λ and κ, and is therefore suspect. If we assume labialised phonemes, like lʷ, a reconstruction * alʷak- gives all forms: αὖλαξ (by anticipation of the labial feature; which gives ὦλαξ by contraction), ἄλοξ (influence on the second vowel ; ὀλοκ- on both vowels), interchange α\/ε gave εὐλακ-; see Beekes Pre-Gr., and cf. ἀρασχάδες etc. Homer might have had *κατ' ὠλακ(α), which became unclear during the tradition.

Page in Frisk: 1,77