*kop-í-lo-

It is commonly assumed that kappilazza-⁽ⁱ⁾ and its cognates derive an underlying nominal base *kappila(/i)- ʻwrath, enmityʼ, whose Hittite cognate is attested as an ablative kappilaz in a Middle Hittite text. This proposal is already found in Čop 1956a:19 and later picked up by most scholars (cf. Kronasser 1966a:477, HHw:72, Tischler 2008a:79, HW 2nd ed. K:109, Eichner 1979a:61, HED K:64, Tischler 1982c:34, Starke 1990a:102 n. 261, Yakubovich 2010d:409, CLL:101, DCL:107). From this base *kappila(/i)-, an unattested Luwian denominal verb in -ā(i)-^(di) was later derived, i.e. *kappila(i)-^(di). This formation constitutes the verbal base of the derivative kappilazza-⁽ⁱ⁾ and is also attested in Hittite as kappilā(i)-^(mi) (Oettinger 1979a:386, Melchert 1987a:199, Sasseville 2021a:440).Although Oettinger 1979a:386 suggests a Luwian origin for the Hittite verb, the word can also be genuinely Hittite and therefore be independently inherited from Proto-Anatolian.

However, the etymological interpretation of the root underlying the present word and its other cognates is difficult. A first suggestion made by Čop 1956a:19 has to be rejected on both semantic and phonological grounds and is outdated by now.Čop reconstructs a nominal formation *kapp-il-, *kapp-el- ʻcolère, haineʼ, which would continue a PIE root *ku̯ōp- (from *ku̯ēp- ʻto seethe, cook; to seethe (emotionally), to be in rageʼ). He compares the semantic development of such a formation to the alleged parallel case of Skt. kúpyati ʻil bouge, il bouillone; il s’irriteʼ. According to LIV²:359, Old Indic kupyati ʻs/he is angry, to bristle (with rage)ʼ < PIE *keu̯p- ʻto trembleʼ (“innerlich beben”) has to be separated from both PIE *ku̯eh₁p- ʻto seetheʼ and *?ku̯ep- ʻto breathe, aspirateʼ (“hauchen”), which would falsify Čop’s argument. See also under Lemma CLuw. kup-. His connection of Hitt. kappilā(i)-^(mi)  to PIE *ku̯ep- is thus rightly rejected by Brosch 2011a:62.

Puhvel (HED K:64-65) connects a putative base *kappil- to Greek κάμπτω ʻto bendʼ and Latin campus ʻfieldʼ. Regarding the alleged Anatolian cognate, he argues for a semantic shift paralleled by the development of Latin campus > OHG kampf ʻjoust, fightʼ. This proposal is firmly refuted by Kloekhorst (EDHIL:439), who classifies an Anatolian equivalent of the specific shift seen in campus as improbable. A different approach is taken by Eichner 1979a:61 (paralleled by Oettinger 1979a:241 n. 6), who reconstructs a nominal form *kappe/ila- deriving from the same Indo-European root as Old Indic śap- ʻto scold, to curseʼ. This proposal is rejected by EDHIL:439 on phonological grounds, since an underlying PIE root *k̑ep- should have yielded Luwian *zapp-. Even though an etymological connection to the PIE root *k̑ep- ʻto curseʼ (LIV²:327) seems tempting on semantic grounds, the only solution for a reflex kapp- would presuppose Hittite as mediator.Implausible Ünal 1996a:48 n. 106 proposing Hitt. kappi- ʻsmallʼ as the base of CLuw. kappilazza-⁽ⁱ⁾, kappilalla/i-, Hitt. kappilā(i)-^(mi) and also CLuw. kappiša.

If we accept that this root is genuine both in Luwian and Hittite, a Proto-Anatolian pre-form should be *kop- (or *kap-). Hittite kappilaz (abl.) as well as the base of all attested Luwian derivatives, going back to an original adjective in *-í-lo-. In order to arrive at the reconstructed base *kappila(/i)-, we can assume a starting point *kop-í-lo-.

[E.R.]

*This lemma was written together with the trainee Melvin Pötzsch.