Notes: A rather widely represented common NC root. The etymology on the whole seems quite satisfactory, but there are some irregularities: in PD we would rather expect *z, not *c:. We may think that the root consonant here was expressively geminated (PD *ʔizz- > *ʔicc:-), and indeed we see traces of this gemination in Ak. izz-es,izz-ala. It is even possible that the gemination occurred as early as in Proto-Darg.-Lezg. - cf. a quite analogous gemination in Rut. =ädda- (normally -d- would be expected).
Cf. also Hurr. azz-u/ošχǝ 'tabooed, impure, sinful, sick (or the like)', see Diakonoff-Starostin 1986, 40.
Notes: Although the root is not very widely spread in EC, it seems to have been the basic PNC word for 'mother' (opposed to *ʔŏbV̄(jV) 'father'), superseded in most EC languages by other expressive "children's words" (*babV, *jājV, *dajV) or its reduplicated variant *nǟnV (q.v.).
Notes: The PN form (as well as PAK *c̣ǝ-na) contains a suffixed *-nV (originally participial); the weak initial syllable with a laryngeal is reduced (which probably reflects the Ablaut grade *ʔĭc̣wĂ). The WC form contains a labial (class) prefix, with a usual in this case delabialisation of the affricate. The comparison of PAK *c̣ǝ-na with Lezg. was proposed by Rogava (1956, 81); see also Абдоков 1983, 150-151.
Notes: Reconstructed for the PEC level. One of several common EC trisyllabic roots, which is clearly seen in PTs and PD reflexes; it may be connected with PEC *=Harq_wV 'to dig', although the derivation model is not quite clear.
Notes: Reconstructed for the PEC level. Correspondences are regular, except for the excessive -n- in PD (all other languages reveal absolutely no trace of any medial resonant); it may have penetrated from an oblique base like *ʔaq-nV- (or, more probably be a result of contamination with another root: PEC *HēnqwV 'meadow, plot' q.v.).