Laryngeal hardening

Alb. <qengj> m, pl. <qengja> `lamb' is supposed to be borrowed from
the methetic from of *acing(u)lus <Lat agniculus (Skok). Bugge thinks
that it is from Rom *agnunc(u)lus, Meyer fellows Bugge. Camaj saw in
it as prefixcal derivative of q- and further comparison with <enjë>.
Çabej connects it qengj< *ken-k- to Gk kainos `new') etc.
Because the name of <lamb> appears in Gheg as <kingj> (cf. kepë,
besides Tosk <qepë> from probably Latin caepa `onion'), I guess that
in the case of Alb. <king/qingj> we have to deal with laryngeal
hardening (*H2 > /k/), deriving it from *H2egWh-nu- that yields PAlb
*kenj-u > kingj/qingj with analogy leveling of cluster *gWh-n to <nj>,
like in *g'ne:-sko > njoh `to know', *egni-/*ogni- `fire' > Alb.
<enj>, probably from *ogni-, attested in <e enjte> `Thursday' and
with /i/ vocalism due to i-Umlaut: Gk amnos: Lat. agnus: OCS jagne: OE
e:anian `yean'. (Pokorny agwh-no-s 9.)

Konushevci