Notes: Reconstructed for the PEC level. The original root structure was *Hirǯ_V, with sporadic metatheses typical for adjectival stems. Most languages do not use class prefixes with this root, so prefixation in some Tsezian languages, as well as Khin. b- is apparently secondary.
Notes: One of several interrogative stems reconstructed for PEC. It has a general interrogative meaning and is most often used in oblique cases.
The root *hī- is frequently encountered in combination with *-nV-, probably as an oblique stem construction *hīnV- 'whose, which, what'. We have mentioned above the secondary development *hīnV- > nV- in some Lezgian languages, but the same process probably had taken place in other languages too. Cf. PA *(h)i-n- (in *(h)in-da 'when' > And. inna-l, Cham. inna, Tind. hin-da-la, God. in-da-q:i,*(h)inV-l 'where' > And. inu-l, Tind. ini-la etc.) = PTs *hɨ̃tǝ / *nitǝ ( < *hini-tǝ ) 'when' (Inkh. ito, Tsez. neti, Gin. nete, Bezht. nito, Gunz. hɨ̃dǝ), PTs *na / *nija 'where' (Inkh., Tsez. na, Gin. ni, Bezht. nā, Gunz. nijõ). It is significant that an identical combination *-a-nǝ- is also attested in WC languages: PAT *-anǝ- 'when' (a verbal infix > Abkh. -an(ǝ)-, Abaz. -an(ǝ)-), Ub. aná-(n) 'at the time, when...". We can thus safely reconstruct the morpheme combination *hīnV- for the PNC level.
Notes: A common NC expressive verb. As with other verbs of "beating", several languages have derivates meaning "wound" (cf. Chech. darχ-, PD *χIa-na < *(Hīr)χIV-nV = PL *jirχIona).
Notes: Reconstructed for the PEC level. The stem contains an element *-ɫī (av. xibi-l, PC *žeb-lV, PD *š:a-li < *š:aw-li) which is present not in all languages and may be an old suffix. But if it is not, then this is one of several very rare cases of tetrasyllabic roots in PEC. Quite enigmatic is the Lak. form čul 'side' (why č-?).
Notes: Reconstructed for the PEC level. Since reflexes of the root are known only in four languages, there is some uncertainty about the initial laryngeal and the medial resonant (several alternative reconstructions are possible).
Notes: Reconstructed for the EC level. The PD (*dulk:ʷa < *duluk:a) and PL form point to the PEC structure *HRVCV̄; this should have given *HVRCV (in this case, *HulŁV) in Avaro-Ando-Cezian languages. However,the root was preserved only within a compound with *kwīlʡɨ'hand', and the first syllable was reduced both in Avaro-Andian and Cezian languages. In general, the comparison seems quite plausible.
Notes: A common NC term. The affricate *ƛʷ in PWC can be explained if we assume a fusion with a dental prefix (*ƛʷa < *T-ʎʷa), as in some other similar cases.
Notes: A widely spread common NC root (there also may have existed a variant *Hnŏƛ̣ŭ which can be reflected in some subgroups). The EC-WC comparison see in Abdokov 1983, 97.
Notes: Correspondences are regular, and the etymology seems quite reliable. The original meaning is certainly 'winter'; the meaning 'summer' in PN comes from the intermediate stage 'year' (cf. the meanings in Tsezian languages).
See Trubetzkoy 1922,241 (for EC). Abdokov (1983, 91) proposes a different comparison for EC: PWC *bž́V 'winter, autumn', which seems impossible for phonetic reasons.
Notes: The root is preserved only in a few EC languages, but appears to be archaic. Correspondences are regular. One phonetic problem, however, is that we should expect *ƛ: in PWC before an originally long vowel - this makes the comparison less reliable. The Khin. form can belong here if -id may be considered as a suffix, and näḳ- < *läḳ- (although reasons for *l- > n- are not quite clear).
Abdokov (1983,73) also compares the PAK form with EC; the Bezht. form ƛ̣a 'behind' which he quotes, may in fact belong here too (although the source of it is unknown to us); however, other forms that are adduced, either do not belong here (as Lezg. ḳul), or are misspelled (as Akhv. raƛ̣:aƛi - instead of raq̇:ʷaƛ:i).
Notes: One of the few verbs with the root structure *HRVCV. Some irregularities (reduplication, sporadic glottalisation in WC) are conditioned by its expressive nature. See Абдоков 1983, 166.
Notes: Reconstructed for the PEC level. There is some variation between the root structures *HRVCV̄ and *HVRCV̄ (which is not unusual in adjectival and verbal roots); otherwise correspondences are regular.
Notes: Reconstructed for the PEC level. Length of the final vowel is unclear: the Av. paradigm C and the structure of the root in PA (*RVCV) point to a short vowel; however, Eastern Dagestan forms presuppose a form like *Hläƛ̣V̄. Metathesized *Hƛ̣älV might have also existed (cf. Avar ṭul - if it belongs here, see above), in which case we can also compare Urart. zel-dǝ 'liver' (see Diakonoff-Starostin 1986, 48-49).
Abdokov's (1982, 83) attempt to compare this root with WC *c̣ʷA 'liver' should be regarded as unsuccessful (for phonetic reasons, and because of a better etymology for the WC form, q.v.).
Notes: A widely spread common NC root. In PL and PAand. the root has a class prefix *w- (in PAand. it has replaced the original r-: we would expect in PA *rišo- as a regular reflex of the structure *HRVCV̆); we should stress that in PD and Lak. initial d- and t:- are not class prefixes, but regular protheses in this phonetic structure.
Notes: In PL and Khin. there occurred a secondary metathesis (we should expect a form like PL *Holƛ̣:) - probably to avoid a cluster of two laterals. Otherwise the correspondences are regular. PWC has a prefixed *b- (as in many other cases). See Abdokov 1983, 81 (where the root is mixed up with *rĕmƛ̣e 'cubit' q.v.)
