Comment: An expressive root, thus the correspondences are violated (excessive labialisation in Lezg., irregular correspondences between affricates). The Arch. word has been modified under Avar influence (cf. Av. ħanč̣-ize), but it must not be considered as a loanword because of its Lezg. and Tab. counterparts.
Comment: This is a rare, but archaic pronominal stem, nearly completely superseded in most languages by PL *u_o(-n) q.v. The stem *ʁu(-n) is used as a direct stem in the Nidzh dialect of Udi, in several Rutul dialects (Shinaz, Khnov, Ikhrek) and in Tsakhur (interchanging with the normal wu < *u_o(-n)). Most important, however, is the regular use of this stem as genitive in Tsakhur: jɨʁ-na (1-2 class), jɨʁɨ-n (3 class) < PL *-oʁ(ʷ)-. Other languages have not preserved this usage, but it is very close to the situation in Khinalugh, and allows to reconstruct the stem *-oʁ(ʷ)- as the original oblique stem of this pronoun. See Aлексеев 1985, 72.
Comment: Cf. also Lezg. Kub. ʁer, Akht. ʁir 'spittle'. The Ag. form is expressively reduplicated and somewhat irregular (ʁurʁur would be rather expected).
Comment: The stem is used separately only in Lezg. and Rut. (where it means 'that (already mentioned)'); in other languages it is added to other pronominal stems to emphasize their own meaning (in Arch. it is preserved only in h-inc 'now' = Ud. hajsa, Tab. hamus).
Comment: Arch. -r- before a uvular points only to PL *-l- (-r- would have disappeared). The root is attested only in Arch., but has good external parallels.
Comment: Obl. base hac̣i- ( < PL *ha(m)c̣e- ~ *ha(m)c̣ä- ~ *ha(m)c̣ɨ-); cf. also Tab. Düb. hac̣a. Isolated in Tab., but having probable external parallels.
Comment: Tab. waq̇Ii < *ʔamq̇Ii; Archi has an expressive prefix ṭ- (sometimes used in adjectival and verbal roots).
Besides *Hamq̇ʷV, Lezgian languages have a series of phonetically close forms with *-q̇:(ʷ)- and *-q:-: Tab. Düb. c-uq:i, Ag. q:uq:u-f (Burk. q:uq:u-f) 'hard', Lezg. weq:i 'rigid, rough', quq:ʷa-z 'to become rigid, to harden' (Khl. qʷq:ʷaj 'hard' = lit. quq:ʷaj 'hardened'), Rut. Mukh. (Ibragimov) ʁiʁ-dɨ 'hard, tough', Luch. q-u=ʁʷa- 'to harden, dry out' ( = Lezg. quq:ʷa-), ʁIɨʁI-dɨ 'hard'. It is rather hard to reconstruct a single protoform for all these forms: we probably deal both with expressive modifications of PL *Hamq̇ʷV and with assimilative developments of the reduplicated form *q̇Vmq:V (from the PEC variant *q̇Hwamq̇_(w)Ă).
