Comment: Cf. also Ag. Burk. χalaʁ 'roof' ( = Darg. Chir. qalʁ). The oblique base can be reconstructed (with Ablaut) as *qola- (cf. Tab. χula-, Ag. χula-, Rut. χálɨ- and the derivate χɨla-dɨ 'roof', Tsakh. Gelm. χawa-). Belongs to the 4th class in all class-distinguishing languages (except Kryz., where it belongs to the 3d class). In Arch. cf. also the derivate χali 'family' (clearly derived from 'house' and thus indicating that the meaning 'nest' in Arch. is secondary).
Comment: Cf. Lezg. erg. χk-uni. 3d class in Rut. Medial -m- reconstructed on basis of external evidence (both in Lezg. and in Rut. it is lost before velars).
Comment: PL paradigm: *χara, obl. *χorä- (cf. Ag. χura-, whence also the direct base χur; in other languages there had been a vowel levelling of the obl. stem: Lezg. χaru-, Tab. χara-). Cf. also Lezg. Khl. χar, Tab. Düb. χar, Ag. Fit. χur, Bursh,, Burk. χar (χaru-). Kryz. has a reduplicated form (as in some other names of agricultures).
Comment: Cf. also Lezg. Akht. mihman-χaš 'spider', Rut. Shin. χašamkal id., Tab. Düb. χašχaš id. The original hissing sibilant is preserved in Tsakh.; it shifted to *š (*šʷ) in other languages due to a merger with another PL root, *χʷärčVj 'worm' q.v. - both roots are very actively contaminating in modern Lezghian languages.
Meaning:1 sleeve 2 branch 3 wing 4 one of a pair 5 thing
Lezghian:χel 1,2
Tabasaran:χil 1
Agul:χil (Tp.) 3
Kryz:χäl-χäǯ 1,χel 4
Budukh:χel 5
Archi:χala-c̣i 1
Comment: Cf. also Lezg. Khl. χIäl 'branch; sleeve'; Tab. Ulz. χil 'feather' ( < 'wing'); Ag. (Rich.) χilib 'sleeve'. Words for 'wing' in other Tab. and some Ag. dialects are also formed from this root: Tab. χl-inc:, Düb. χil-imc:i 'wing' etc. (formally cf. Arch. χalac̣i). Although there is no Tsakh. form (with -a), most probable for PL is a paradigm dir. *χäla, obl. (with Ablaut) *χilä- (cf. Ag. Tp. χili-, Lezg. χile-). The semantic development in Bud. ('thing') is explained by the closely related Kryz. χel 'one piece of a pair, couple'. In Arch. this root must have influenced the development of PL *χ:ɨl 'hand, arm' q.v. (it yielded Arch. χol instead of the expected *χ:ol). The two roots are close phonetically and semantically; therefore in some cases it is hard to attribute some reflexes precisely: such derivates as Rut. χɨli-dɨ 'mitten', χuluk 'sleeve' ( = Tsakh. χuk id.) could be derivedfrom both PL forms.
Comment: Cf. also Tab. Kand. qif (probably with a new q-, dissimilated from the final fricative), Düb. χif:i, Khür. χif:, Lezg. Khl. χʷäIχI. In Lezg. the final -χ is a result of assimilation < -x; pharyngealization in Ag. is secondary (after a uvular in a position before a front vowel; cf. in other dialects: Bursh., Burk. χew, Fit. χiw).
A variant *χeƛ̣:ʷ is reflected in Rut. χik (obl. base with levelling χikɨrɨ-, but cf. Mükhr. χik,χiggirɨ-); Tsakh. χek (obl. base χek:a-), Ud. uq:I (3d class in Rut., Tsakh.). Just as PD *χiɣʷ, this variant reflects a contamination or compound of PEC *χäwx_wV and *ʔwǟrƛ̣_V (q.v.).
Comment: Cf. also Tab. Kand. χir, Düb. χIar, Ag. Burk. χer 'meadow'. Pharyngealization in Tab. and Ag. dialects is automatic (before a front vowel); otherwise we would expect Ag. X̀-. Obl. base *χera- (cf. Tab. χIara-, Ag. Burk. χera-).
Comment: Cf. also Tab. Kand. χIarǯ 'linden'. This form (with palatalisation *d > ǯ), as well as Tab. lit. obl. stem χIardi-, Ag. Bursh. obl. stem χirdi-, Tsakh. obl. stem χide-, point to the PL oblique stem *χert:e-. Pharyngealisation in Tab. is secondary (before a front vowel). 3d class in Kryz. and Tsakh. The two stems: *χer (reflected in Kryz. χir) and *χert: (reflected in other languages) go back to different PEC variants.
Comment: Cf. also Tsakh. Gelm. wɨχIɨna, (Tsakh.) wuχIana. 4th class in Rut. and Tsakh. The Tsakh. word contains a prefix wu- , probably < *ru- (or *ʔur(V)-).
Comment: The Rut. word is a compound (for the second part, -χIul, cf. perhaps PL *χIul 'fat' q.v.?). The Tsakh. form has a prefixed *ʔVr- (whose meaning is unclear, as in a number of other PL etyma).
Comment: Cf. Lezg. erg. χp:e; cf. also Ag. Fit. χIub (pl. χIup:-ar), Burk. X̀a, pl. X̀aw-ar. The PL paradigm must be reconstructed with Ablaut *χIap:, *χIop:a- (cf. Lezg. χp:e-, Ag. X̀up:a-, Rut. χIɨbɨ-); the oblique base vocalism influenced the direct base in Ag. and Rut. 3d class in Rut.
Comment: Cf. also Tab. Düb. χIaw, Lezg. Khl. χIäw, Ag. Bursh., Burk. X̀aw, Tp. ħaw. Obl. base *χIowa- (cf. Lezg. χiwe-, Rut. χIɨwɨ- - whence also the direct base χIɨw, Tsakh. χIowa-). 3d class in Rut. and Tsakh.
