Notes: Despite the root's somewhat expressive character, correspondences are regular, and the PNC reconstruction seems reliable. Traces of *-r-conjugation can be found in PA (stem II *q:ur-), and perhaps also in Darg. (medial -r- which is otherwise unclear) - although the decisive evidence of certain Lezghian languages is missing. See also comments to PEC *Harq_wV 'to dig'.
Notes: Trubetzkoy 1930,87 compares WC *q́I:a 'to be' with another PEC stem (*ʔikV q.v.), which is impossible for phonetic reasons. The comparison with *=Hŏq̇_Ē seems quite plausible semantically and phonetically.
A similar root exists in Kartvelian (Georg. q̇a-w- 'to be, to have'); its comparison with WC see in Charaya 1912, 37, Lomtatidze 1956, 225, 228; the nature of its connection with the NC root is yet to be cleared up.
Notes: Reconstructed for the PEC level. Although the root is attested only in three languages, correspondences are quite regular and the root seems to be well distinguished from several similar roots (see *HVq̇_Vr 'to drink', *HVlG_wVl 'to swallow', *=HVq̇_wVn id.).
Notes: Trubetzkoy (1930, 88-89) compares the WC root with other EC material (see *=ānʒŭ); the present comparison seems to us preferable, because the WC root contains not a hissing, but a hushing affricate (the correspondence PAK *ź : Ub., PAK *z points only to PWC *ž).
Notes: Reconstructed for the PEC level. Note the correspondence of PA *-j- to the PN development *ćw > *sṭ. The Nakh languages also have a root *=ēbc- / *=ōbc- 'to knit, plait' (Chech. =ūc-, Bacb. =epc-, term. =opc-) which may reflect the same verbal stem (with the plural infix -b-, which superseded *-j- and prevented the development *-jć- > *-sṭ-). Cf. also HU: Hurr. χuž- 'to bind, to detain', Urart. χuš- 'to annul (services), to stem (water)' (*'to untie'), see Diakonoff-Starostin 1986, 40.
Notes: The WC form has a usual labial prefix (a former class marker). Medial *-j- must be reconstructed to account for the development *-ś- > *-sṭ- in PN. Vocalic correspondences are not quite clear (due to the -j- and to the influence of the initial laryngeal); otherwise correspondences are regular.
Notes: The original meaning of the root should be probably reconstructed as "to go, walk (away from the speaker)". The etymology seems reliable phonetically and semantically.
Notes: Reconstructed for the PEC level. One feels tempted to compare also Lezg. al-ḳi- 'to stick, glue, adhere', k-iḳi- id.; but if this is the same verbal root as eḳi- 'to stick, be stuck', it should be kept separated from *=HVlḳ(w)Vn, because the Akht. correspondence (ak:ɨ-) points only to PL *-ƛ̣:-.
Notes: A common NC verb; due to its semantics subject to expressive changes (especially in the Av.-And. area) - but still the reconstruction seems both phonetically and semantically plausible.
Notes: An Aand-Lezg. isogloss. In Lezgian languages the root tends to contaminate with the reflexes of *HVq̇_Vr 'to drink' and *=Hoq̇_V 'to sink' (q.v.).
Notes: The verbal stem as such is reflected only in Av.-And.-Tsezian and WC languages. In most EC languages much more widely spread is the nominal derivate *nhewƛ̣_ŭ 'sleep' (a rather unique case of *n-prefix; perhaps it is actually *mhewƛ̣_ŭ with early dissimilation?), with either long or short first vowel (long *ē is reflected in PN, short *e - in other languages). Note a unique reflex *-wƛ̣- > *-ṗ- in PN, and the irregular development of *ƛ̣ in a tense word > *ƛ̣ in PL. The latter phenomenon is explained obviously by the influence of another NC root, *ɦemƛ̣V 'dream' q.v.
Notes: A secure common NC root, serving in both verbal and adjectival functions. It is usually accompanied by class prefixes; however, comparing PL *hIac̣ɨ- and the Av. Ablaut stem ʕorc̣- (with infixed -r-) we may reconstruct the initial laryngeal *ɦ-, superseded by class prefixes in most languages.
Notes: Reconstructed for the PEC level. Correspondences are regular (except for the nasalisation in PTs, which must be secondary - as well as the -n-conjugation in some Avar dialects).
Notes: Reconstructed for the PEC level. The verbal root is preserved only in PTs and Laki.; PN and PD reflect only the derivate *č_V-uV 'wound' ( = Lak. š:awu).