Kibrik et al. 1977b: 277, 354; Chumakina et al. 2007; Dirr 1908: 47. Quoted as mirči ~ murči in [Mikailov 1967: 95, 193], but as marči in [Mikailov 1967: 156 (65)]. As noted in [Kibrik et al. 1977a 3: 277], mˈarči normally refers to substantives of classes 1 and 2, scil. human beings (the same in [Chumakina et al. 2007]).
Browsing through texts in [Kibrik et al. 1977b; Mikailov 1967; Dirr 1908] suggests that the adjective mˈarči is the most common expression for 'all (omnis)', which can be used as both attributive and non-attributive, referring to human beings. Examples are numerous, e.g., "All youths like this girl" and so on [Kibrik et al. 1977b: 9, 11, 66, 67, 90], "All men have come", "you do say to all the people that..." [Mikailov 1967: 95, 156, 158]. Examples for non-attributive mˈarči 'all (omnis)' referring to human beings are also well attested, e.g., "I have killed them all" [Kibrik et al. 1977a 2: 173], "Everybody was happy, when she died" and so on [Kibrik et al. 1977b: 9, 10, 11, 21, 27, 46, 104, etc.].
A reasonable solution should be to treat expressions for 'omnis (human beings)' and 'omnis (non-human or inanimate)' as synonyms, but the latter basic term cannot be established from available sources.
A possible candidate for 'omnis (non-human or inanimate)' could be the adverb/adjective kʼilliy-class-u [Mikailov 1967: 95, 186; Dirr 1908: 48] with polysemy: 'totus / omnis'. The following examples for 'omnis' are found: "All the bull calves went home" [Mikailov 1967: 95], "All the horses" [Dirr 1908: 48], and also referring to humans: "All the women" [Dirr 1908: 48]. In more modern sources, however, this word is quoted as kʼellˈey-class-u with the exclusive meaning 'entirely, wholly', i.e. 'totus' [Chumakina et al. 2007; Kibrik et al. 1977b: 264].
More rare seems to be the adjective bˈatː-ˈešːu-tːu-class 'totus / omnis' [Chumakina et al. 2007] (examples: "I have filled a whole notebook with my writing", "All the sheep have come back"), [Kibrik et al. 1977b: 200] from the adverb bˈatː-ˈešːu 'completely' ← the verb bˈatː-ˈešːa- 'to come true, be fulfilled (of wish); to pass, come (of time)'.
Cf. also various words for 'all (totus)':
The adverb ˈiːkʼen [Kibrik et al. 1977b: 251, 354; Mikailov 1967: 95, 182], which normally expresses non-attributive 'all (totus)' referring to inanimate and abstract objects. E.g., "She demonstrated her hair and all (the rest, i.e. her body)", "Then this girl said everything" and so on [Kibrik et al. 1977b: 32, 37, 41, 93, 94, 97, etc.]. But there is one example for attributive ˈiːkʼen 'all (omnis)' referring to inanimate objects: "He (the physician) cured all fractures" [Kibrik et al. 1977b: 116], and one example for attributive ˈiːkʼen 'all (totus)' referring to an inanimate object: "I have latched onto all of the property" [Mikailov 1967: 95].
The adjectives gʷˈey-class-u and gʷˈey-class-u-hˈoːnu, which are explicitly glossed as 'all (totus), whole' in [Kibrik et al. 1977b: 234, 352] and [Chumakina et al. 2007] (hˈoːnu means 'some, any' [Kibrik et al. 1977b: 240]) with examples like "He has left all his property to the son", "The whole girl has been covered with gold", "I was washing the dishes yesterday, but could not wash the whole lot" [Kibrik et al. 1977b: 41, 234; Kibrik et al. 1977a 2: 148]. In [Mikailov 1967: 95, 176] this adjective is quoted as gʷiy-class-u with additional examples for the meaning 'all (totus)'.
More marginal seems to be the adjective hˈannu-class-hˈoːnu 'all (totus), whole' [Chumakina et al. 2007; Kibrik et al. 1977b: 236, 354], consisting of hˈannu 'which one' + hˈoːnu means 'some, any'.
The last term is the adjective ˈobšːi 'all (totus/omnis)', glossed as 'all, whole, every' in [Chumakina et al. 2007] and supplemented by two examples: "All the people [sg.] went in different directions", "Everybody went to the fields". This word has not been found in other sources.
Kryts (proper):vari-1
Saadiev 1994: 422. Polysemy: 'totus / omnis'. Apparently a Turkic loanword: Proto-Turkic predicative *baːr 'there is, there exists', which shifted to 'all (omnis or totus)' in many Turkic languages, although Azerbaijani var is probably attested only with the meaning 'there is'.
Alyk Kryts:girt-1
Authier 2009: 107. Polysemy: 'totus / omnis'. The same term in other dialects: Dzhek, Xaput gɨrt [Saadiev 1994: 422]. Borrowed from Azerbaijani dialectal gird 'wholly, completely' or directly from Iranian, cf. Talysh gɨrd 'all (totus)'.
Budukh:pitin {питин}-1
Meylanova 1984: 119, 193; Alekseev 1994: 270; Talibov 2007: 130. Polysemy: 'totus / omnis'. Borrowed from Azerbaijani bütün 'all (totus/omnis)'.
Dzhamalov & Semedov 2006: 230. The meaning is 'all (omnis)'; it is the regular plural form from si-ne {сине} 'all (totus)' [Dzhamalov & Semedov 2006: 230].
Distinct from the paronymous säye-nä {саьенаь} 'all (totus)' [Dzhamalov & Semedov 2006: 224, 327] with the examples: "I have got it all" [Dzhamalov & Semedov 2006: 229b], "Everything depends upon you" [Dzhamalov & Semedov 2006: 23a], "He has sold everything" [Dzhamalov & Semedov 2006: 23b].
An additional term is bütün {буьтуьн} 'all (totus/omnis)' [Dzhamalov & Semedov 2006: 43], borrowed from Azerbaijani bütün 'all (totus/omnis)'.
Luchek Rutul:
Not attested.
Common Rutul notes:
Mukhad si-ʔin, siyä-nä and Ixrek si-ne, säye-nä apparently represent historical derivatives from the Rutul numeral sa '1'.
Dirr 1907: 28, 114, 170; Suleymanov 1993: 134. The form in ž- is from [Dirr 1907] (in some cases Dirr transcribes Gequn ǯ as ž). Meaning 'all (omnis)'.
Distinct from inherited kʼil-di 'all (totus)' [Dirr 1907: 28, 130, 169], an adverb derived from kʼil 'head' q.v.
Fite Aghul:ǯalːa-1
Suleymanov 1993: 134. Two synonyms are quoted in [Suleymanov 1993]: ǯalːa and wari, both of them probably with polysemy: 'totus / omnis'.
The word wari was ultimately borrowed from Turkic; see notes on Kryts (proper).
The second term (ǯilːa ~ ǯalːa) also seems to be a loanword. The Koshan and proper Aghul forms are quoted as ǯˈilːa ~ ǯˈalːa in [Suleymanov 2003: 80]; the initial stress points to a recent borrowing (see [Magometov 1970: 20]), although the source is not entirely clear (Arabic?).
Northern Tabasaran:warˈi ~ warːˈi-1
Uslar 1979: 145, 622, 991; Dirr 1905: 45, 159, 225. Polysemy: 'all (omnis) / all (totus)'. This is actually the word from the Khanag subdialect; the proper Dyubek term for 'all' is unknown.
The same in the Khyuryuk subdialect: warˈi {вари} 'all (omnis / totus)' [Genko 2005: 32].
Southern Tabasaran:wˈari {вари}-1
Genko 2005: 32. Polysemy: 'all (omnis) / all (totus)'. This is actually the word from the Khiv subdialect; the proper Kondik term for 'all' is unknown.
The same in Literary Tabasaran: wˈari {вари} 'all (omnis / totus)' [Khanmagomedov & Shalbuzov 2001: 93; Zhirkov 1948: 105]. There also exists the rare literary word marcːi-yi {марццийи} ‘all’ [Khanmagomedov 1957: 43] (not found in other sources; the exact meaning is unknown) - literally ‘cleanly’, derived from the adjective marcːˈi ‘clean’ [Khanmagomedov & Shalbuzov 2001: 224] with the common adverbial suffix -yi (for which see [Magometov 1965: 327]).
Common Tabasaran notes:
The loanword warˈi / wˈari is the Common Tabasaran term for 'all (omnis / totus)' [Magometov 1965: 189]. The form was ultimately borrowed from Turkic, see notes on Kryts (proper).
Gyune Lezgi:warˈi-1
Uslar 1896: 79, 366, 606. Polysemy: 'all (omnis) / all (totus)'. A close, but less frequent synonym is mixˈiz 'all (omnis/totus)' [Uslar 1896: 80, 507, 606].
The same in Literary Lezgi: wirˈi {вири} 'all (omnis/totus)' [Talibov & Gadzhiev 1966: 78; Gadzhiev 1950: 80; Alekseev & Sheykhov 1997: 46; Haspelmath 1993: 200, 511, 515] with the close, but less frequent synonym mixˈiz {михьиз} 'all (omnis/totus)' [Talibov & Gadzhiev 1966: 241; Gadzhiev 1950: 80].
The word wari ~ wiri was ultimately borrowed from Turkic; see notes on Kryts (proper). The second term mixˈi-z literally means 'cleanly', derived from the inherited adjective mixˈi 'clean' [Talibov & Gadzhiev 1966: 240] with the common adverbial suffix -z (for which see [Gaydarov et al. 2009: 220 f., 222 f.]).
Proto-Lezgian:
Not reconstructible.
Distribution: In the majority of languages, the meaning 'all' (omnis or with polysemy omnis / totus) is represented by Turkic or Iranian loanwords. Inherited forms are:
1) Tsakhur gɨrgɨ-n, derived from the old root for 'round' q.v.
2) Rutul si-ʔin, si-ne-bɨr, apparently derived from the old root for 'one' q.v.
3) Lezgi mixˈi-z, literally 'cleanly' from the adjective 'clean'.
4) Tabasaran marcːi-yi, literally 'cleanly' from the adjective 'clean'.
5) Caucasian Albanian ceχ and Archi mˈarči, both are etymologically unclear.
Tabasaran marcːi-yi and Archi mˈarči are indeed phonetically similar, but the affricate correspondence is irregular.
The same in the Khanag subdialect: ruqːˤ 'ashes' [Uslar 1979: 900, 994; Dirr 1905: 203, 229]. The same in the Khyuryuk subdialect: ruqːˤ {рюкъ} 'ashes' [Genko 2005: 135].
The same in the Khiv subdialect: ruqːˤ {рюкъ} 'ashes' [Genko 2005: 135]. The same in Literary Tabasaran: ruqːˤ {рюкъ} 'ashes' [Khanmagomedov & Shalbuzov 2001: 263].
Chumakina et al. 2007; Kibrik & Kodzasov 1990: 97; Kibrik et al. 1977b: 291, 363; Mikailov 1967: 196. Not found in [Comrie & Khalilov 2010; Dirr 1908]. Distinct from qal 'human skin (q.v.); fruit peel' [Chumakina et al. 2007; Kibrik et al. 1977b: 295].
The only term for 'bark' in [Dirr 1908] is, however, qal 'peel; bark; shell' [Dirr 1908: 162, 211]. It is very probable that this reflects an archaic usage, before qal 'skin / bark' was superseded by pˈaqʼut in the meaning 'bark' and its semantic usage was narrowed to 'skin' q.v. It must be noted that qal is also quoted in [Comrie & Khalilov 2010: 412] as the only translation for 'bark', but this may be an error.
It should be noted that Dirr quotes two different terms for 'bark': Mukhad qːabɨχ [Dirr 1912: 152, 192] (borrowed from Azerbaijani gabɨg 'bark') and Mukhad & Shinaz čʼɨχɨn [Dirr 1912: 181, 182, 192]. The latter can in fact mean simply 'birch bark', cf. the Luchek data below.
Kibrik & Kodzasov 1990: 97. It must be noted that in [Shaumyan 1941: 187] and [Suleymanov 2003: 114] Burshag gal is glossed only as 'shell', whereas in [Suleymanov 2003: 114] the meaning 'bark' is ascribed to the Burshag word ʁark.
In the Khudig subdialect 'bark' sounds as ʁark [Magometov 1970: 41].
Suleymanov 2003: 114; Shaumyan 1941: 187. Polysemy: 'bark / shell / peel'. The same in the Kurag subdialect: qːark 'bark' [Magometov 1970: 41].
Common Aghul notes:
It seems that qːark can assuredly be reconstructed as the Proto-Aghul term for 'bark / shell'. In some modern dialects this polysemy tends to be eliminated, with the meaning 'bark' transferred to other words - gužal ~ žigal in Keren or gal in Koshan (Burshag).
The same in the Khanag subdialect: ɣal with polysemy: 'bark / peel / skin (i.e. hide?)' [Uslar 1979: 653, 996; Dirr 1905: 162, 231]. The same in the Khyuryuk subdialect: ɣal {ггал} with polysemy: 'bark / peel' [Genko 2005: 39].
The same in the Khiv subdialect: ɣal {ггал} with polysemy: 'bark / peel' [Genko 2005: 39] (the plus sign is apparently omitted by accident, although the white space between the bracket and the siglum "Х." is present). The same in Literary Tabasaran: ɣal {гал} with polysemy: 'bark / rolled out dough' [Khanmagomedov & Shalbuzov 2001: 96].
The same in Literary Lezgi: čkːal {чкал} with polysemy: 'bark / peel' [Talibov & Gadzhiev 1966: 373; Gadzhiev 1950: 308].
Somewhat differently in the Akhty dialect: Khlyut kan 'bark' [Kibrik & Kodzasov 1990: 97]. According to [Kibrik & Kodzasov 1990], the second Khlyut word for 'bark' is čkːal. The difference between the two terms is unknown, but Khlyut kan is clearly secondary, because its literary counterpart kan means 'dandruff; epidermis' [Talibov & Gadzhiev 1966: 149] and the external comparison proves that the semantics of 'skin' is primary [NCED: 699].
Proto-Lezgian:*ƛːal1
NCED: 789. Distribution: An unstable word, superseded by a loanword from Azerbaijaniin Kryts, Budukh and Tsakhur and probably by a loanword from Avar in archaic Archi. Four inherited stems enter into competition: (1) *ƛːal (Udi 'bark / peel / shell', Tabasaran 'bark / peel' and Koshan Aghul 'bark'), (2) *čːukːa-la /*kːučːa-la (Rutul 'bark / peel', Lezgi 'bark / peel' and Keren Aghul 'bark'), (3) *parqʼːulː (modern Archi), (4) *qːärkʷa (Proto-Aghul 'bark / shell').
The last of these, *qːärkʷa, is formally a Proto-Aghul innovation (Lezgian comparison suggests the development 'a k. of hide' > 'bark' [NCED: 455 f.]).
Two stems with the best distribution - ƛːal and *čːukːa-la /*kːučːa-la - are formally equivalent candidates, and any historical scenario would imply certain zig-zag or parallel semantic shifts.
External North Caucasian comparison suggests that the original meaning of *čːukːa-la /*kːučːa-la was 'noodles' or 'a k. of food rising to the surface after boiling' [NCED: 439]. This semantics is retained in Budukh ('noodles'), but underwent such shifts as 'noodles' > 'sour cream' in Tsakhur, 'noodles' > 'bast, bass' in Kryts and 'noodles' > 'bark, peel' in Rutul, Lezgi, Keren Aghul. The latter shift is not a genetical feature of Rutul, Lezgi and Keren Aghul, but is either an independent development or, rather, an areal isogloss between adjacent territories (see [Koryakov 2006: map 13]).
As for *ƛːal, external North Caucasian comparison points to the original meaning 'color' with the shift 'color' > 'surface' [NCED: 789]; however, it is not particularly risky to assume the shift 'color' > 'bark, peel' for Proto-Lezgian. The meaning 'bark, peel' is retained in Udi and Tabasaran, and emerges as a secondary development in Koshan Aghul (in Proto-Aghul this root apparently meant 'shell').
The fourth inherited term for 'bark', *parqʼːulː, which is attested in the basic meaning in modern Archi, can be a recent introduction, if Dirr's data are correct. If so, this Proto-Lezgian stem demonstrates the shift 'burdock, plantain' > 'bark' in Archi [NCED: 865]. It is important that Lezgian *parqʼːulː finds Avar cognates with the meaning 'bark' [NCED: 865], but it is not clear whether *mVqʼːVr- can be reconstructed as the Proto-Avar-Andic term for 'bark' (perhaps we are dealing with independent developments in Archi and Avar). Note that in the Lezgian languages, *parqʼːulː demonstrates various consonants irregularities of dissimilative/assimilative nature [NCED: 865], particularly the Archi form is pˈaqʼut with lax qʼ (not **pˈaqʼːutpace [NCED]).
Gukasyan 1974: 202; Kibrik & Kodzasov 1990: 36; Comrie & Khalilov 2010: 222; Mobili 2010: 261. Glossed as 'belly, paunch, stomach' in [Gukasyan 1974: 202]. Quoted as a translation for 'stomach' in [Comrie & Khalilov 2010]. Glossed as 'belly' in [Kibrik & Kodzasov 1990], distinct from mäːda 'stomach' (< Azerbaijani mädä 'stomach') in the same source.
Gukasyan 1974: 93; Mobili 2010: 72; Fähnrich 1999: 11; Schiefner 1863: 102; Schulze 2001: 262; Starchevskiy 1891: 496. Glossed as 'belly, paunch, stomach, womb'. In [Gukasyan 1974] (followed by [Mobili 2010]) quoted with an error: buqːun {букъун} with a superfluous {ъ}. The variant bukun has also been checked by Yu. Lander (p.c.) in the modern subdialect of the Zinobiani (Oktomberi) village.
Schulze [Schulze 2001: 262] suspects a borrowing from Arabic butʼuːn (the plural form of Arabic batʼn- 'belly, stomach, womb'); the hypothesis is unconvincing both phonetically and morphologically.
Common Udi notes:
Both terms (tapan and bukun) look suspicious, although no potential sources of borrowing have been revealed up to now. If both forms are indigenous, it is rather tapan that should be postulated as the original Udi word for 'belly' in the light of external comparanda.
Caucasian Albanian: Unattested. Cf. kaq 'womb' in the collocation ne kaq 'mother's womb' [Gippert et al. 2008: IV-32].
Archi:lˈagi3
Chumakina et al. 2007; Kibrik & Kodzasov 1990: 36; Kibrik et al. 1977b: 271, 359.
According to [Chumakina et al. 2007; Kibrik & Kodzasov 1990: 36; Kibrik et al. 1977b: 271, 336, 359] there are two closely synonymous terms for 'belly, stomach': lˈagi and χˤurχˈu. Browsing through texts in [Kibrik et al. 1977b], however, suggests that lˈagi is applied to humans, whereas χˤurχˈu normally denotes stomach of an animal. Cf. the following examples for lˈagi: "I have a stomach ache" (lit. "my stomach aches") [Kibrik et al. 1977b: 27, 28, 271, 32, 33, 38, 39], "She is pregnant" (lit. "a child in her belly"), "stepbrother" (lit. "brother with a different belly") [Kibrik et al. 1977b: 271].
Examples for χˤurχˈu are: "Give me the ram's stomach" etc. [Kibrik et al. 1977b: 25, 26, 28], "I have given him (i.e., to the child of the king of beasts) the name Full Belly" [Kibrik et al. 1977b: 336], "A dried cow stomach" [Kibrik et al. 1977b: 336].
Two terms, lˈagi and χˤurχˈu, are discriminated in [Dirr 1908]: lˈagi is translated as 'paunch, belly (Russian: брюхо, живот)' [Dirr 1908: 164, 207], χˤurχˈu - as 'stomach (Russian: желудок)' [Dirr 1908: 190, 207].
Conversely, in [Mikailov 1967] two terms are opposed as lˈagi 'abdominal cavity (Russian: брюшная полость)' [Mikailov 1967: 190] vs. χˤurχˈu 'stomach, belly (Russian: желудок, живот)' [Mikailov 1967: 201a] (the Archi word is omitted due to typographic error).
We fill the slot with the etymologically obscure word lˈagi. This resembles Lak laqʼˤi 'belly, stomach', which is likewise unetymologizable, but the Archi term can hardly be a Lak loanword due to phonetic difference (for Lak loanwords in Archi see [Kibrik et al. 1977a 1: 53 f.]). It is proposed in [NCED: 755] that Archi lˈagi was borrowed from Avar lagˈa 'body part, organ; body-build, frame, figure; stature', but this solution seems dubious due to semantic difference (cf. the late transparent Avar loanword in Archi lagˈa ‘body part’ [Chumakina et al. 2007; Kibrik et al. 1977b: 271]).
Distinct from kummˈullin noƛʼ 'stomach, gaster (Russian желудок)', literally 'food's house' [Kibrik & Kodzasov 1990: 36] - a loan translation from Avar (in [Comrie & Khalilov 2010: 222] this is incorrectly quoted as the only Archi term for 'stomach').
Authier 2009: 23, 36, 49, 55, 105, 199, 268, 288, 321, 347. Paradigm: fan [abs.] / fun- [obl.]. According to examples, a generic term with wide application.
Distinct from the more marginal term fɨn {фын}, glossed as 'belly' in [Meylanova 1984: 143, 212] with the only example being: "to have a stomach ache".
Distinct from qːursaʁ 'stomach' [Kibrik & Kodzasov 1990: 36], borrowed from Azerbaijani gursag 'abomasum; stomach, belly'.
According to [Comrie & Khalilov 2010: 222], the meaning 'stomach' can also be expressed with the loanword maʔda {маъда} (< Azerbaijani mädä 'stomach' or rather directly from Arabic maʕd-at- 'stomach').
Dirr 1912: 174, 190; Ibragimov 1978: 39, 114. Polysemy: 'belly / stomach', applied to humans and animals. It should be noted that in [Makhmudova 2001: 43], uxun is specified as 'belly of animal' as opposed to iči 'belly' (i.e. 'belly of human'?) [Makhmudova 2001: 12, 41]; the latter is borrowed from Azerbaijani ič (possessed ič-i) 'entrails'. In [Ibragimov 1978: 39], however, Mukhad iči is glossed as 'pluck'.
Distinct from armaz 'belly, paunch', quoted in [Ibragimov 1978: 169, 222] without additional specifications.
Ixrek Rutul:ämräz {аьмраьз}-1
Dzhamalov & Semedov 2006: 21, 339. In [Ibragimov 1978: 198, 222], quoted as amraz. This seems to be the basic word for 'belly' in Ixrek, cf. the examples: "It became tight in the belly" [Dzhamalov & Semedov 2006: 88b], "Bellyaches have begun" [Dzhamalov & Semedov 2006: 166a], "To crawl on the belly" [Dzhamalov & Semedov 2006: 279b]. Ixrek ämräz ~ amraz, Mukhad armaz, Muxrek amraz 'belly' [Ibragimov 1978: 169, 186] looks like a recent loanword, although the source is not clear (cf. Arabic ʔamarr- 'intestines').
A second (apparently less frequent) term is the inherited ufun 'belly; stomach' [Dzhamalov & Semedov 2006: 257].
A third candidate is laqʼˤa 'paunch, belly' [Dzhamalov & Semedov 2006: 178], borrowed from Lak laqʼˤi 'belly, stomach'.
Luchek Rutul:iči-1
Kibrik & Kodzasov 1990: 36. Glossed as 'belly (Russian: живот)'. Borrowed from Azerbaijani ič (possessed ič-i) 'entrails'.
Distinct from šaxː 'stomach' [Kibrik & Kodzasov 1990: 36]. It must be noted that in [Suleymanov 2003: 115], 'stomach' is quoted as Koshan ʁʷag and šahar, but without exact specification of the subdialect.
The same in the Khanag subdialect: fun 'belly; offal' [Uslar 1979: 942, 989; Dirr 1905: 213]; cf. also the expression lekrin fun 'calf of the leg', literally 'belly of the leg' [Dirr 1905: 213]. The same in the Khyuryuk subdialect: fun 'belly; offal' [Genko 2005: 161].
Distinct from Khanag maʔad-an 'stomach' [Dirr 1905: 195, 228] (ultimately borrowed from Arabic maʕd-at- 'stomach').
The same in the Khiv subdialect: fun 'belly; offal' [Genko 2005: 161]. The same in Literary Tabasaran: fun 'belly; stomach' [Khanmagomedov & Shalbuzov 2001: 310].
The same in Literary Lezgi: rufˈun {руфун} 'belly' [Talibov & Gadzhiev 1966: 281; Gadzhiev 1950: 200; Haspelmath 1993: 505, 516]. Distinct from literary χuk [abs.] / χʷkːʷ-ˈadi- [obl.] {хук} with polysemy: 'stomach / gourmandizer' [Talibov & Gadzhiev 1966: 349; Haspelmath 1993: 512, 527].
The same in the Akhty dialect: Khlyut rɨfˈɨn with polysemy: 'belly / stomach' [Kibrik & Kodzasov 1990: 36].
Proto-Lezgian:*uo=ɬʷɨn ~ *ro=ɬʷɨn6
NCED: 771. Distribution: The best candidate is *uo=ɬʷɨn ~ *ro=ɬʷɨn, attested with the generic meaning 'belly' in all Nuclear Lezgian lects except for Budukh, but apparently totally lost in the outliers (Udi, Archi). External North Caucasian comparison also seems to confirm =ɬʷɨn as the Proto-Lezgian root for 'belly' [NCED: 771].
Another candidate is Nidzh Udi tapan ~ Budukh tǝpǝn 'belly', but these isolated forms seem very suspicious and rather look like a wandering loanword (although the source of borrowing is unknown). Vartashen Udi bukun and Archi lˈagi 'belly' are likewise isolated; they could only formally pretend to the status of the Proto-Lezgian term for 'belly'. It should be noted that the authors of [NCED] follow Gukasyan's typo that transcribes the Vartashen Udi form as buqːun - actually, the proposed Udi-Archi comparison [NCED: 297] should be rejected.
In some lects *uo=ɬʷɨn ~ *ro=ɬʷɨn was superseded with loanwords (Luchek Rutul < Azerbaijani; Ixrek Rutul < ?).
Gukasyan 1974: 131; Fähnrich 1999: 18; Dirr 1903: 2, 10; Schiefner 1863: 83; Schulze 2001: 288; Starchevskiy 1891: 492. Polysemy: 'big / old'. A second term for 'big' is quoted in [Fähnrich 1999: 18; Schiefner 1863: 105] and [Starchevskiy 1891: 492]: yeka ~ yeko 'groß, большой' (borrowed from Azerbaijani yekä ~ äkä 'big, large').
Common Udi notes:
Udi kala is an Iranian loanword: cf. Persian kalaː-n 'large, great, big; elder', Judeo-Tat kälä 'big, large', etc. Pace [Schulze 2001: 288], kala can hardly be borrowed directly from from Judeo-Tat kälä, since Judeo-Tat linguistic influence on Udi is very modest (if it exists at all) and the idea of a borrowing of such a basic term cannot be accepted.
Caucasian Albanian: boˤ-nʸi [Gippert et al. 2008: IV-11, 44], glossed as 'great, big, tall'. Apparently an important retention.
Dirr 1912: 166, 185; Ibragimov 1978: 69. Cf. Dirr's examples: "big house", "big (= important) person", "elder brother", "My son / the tree becomes big (= grows up) rapidly".
A second candidate is äyk-äd 'big' [Ibragimov 1978: 39, 125] with the more modern variant ekː-ed [Comrie & Khalilov 2010: 583; Makhmudova 2001: 95], borrowed from Azerbaijani yekä ~ äkä 'big, large'.
According to [Makhmudova 2001: 95], in the modern language the inherited form qʼux-dɨ 'big' is applied to animated objects, whereas the borrowed ekːe-d - to inanimate ones (apparently Makhmudova means human beings vs. animals/inanimate objects); cf. an example for qʼux-dɨ ~ qʼux-du: "big father" [Ibragimov 1978: 69]. We prefer to follow Dirr's norm.
Ixrek Rutul:qˤač-dɨ {хъаIчды}5
Dzhamalov & Semedov 2006: 278. Glossed as 'big, large'. According to examples that could be found, this seems to be the basic word for 'big', cf "large windows" [Dzhamalov & Semedov 2006: 68a], "big stones" [Dzhamalov & Semedov 2006: 95b], "The children became big (= grew up)" [Dzhamalov & Semedov 2006: 278a].
A second candidate is eke-dɨ 'large' [Dzhamalov & Semedov 2006: 309], cf. such examples as: "large (court)yard" [Dzhamalov & Semedov 2006: 80a], "large family" [Dzhamalov & Semedov 2006: 272a], "You became big (= grew up) rapidly", "to do a big job", "to hold a high post" [Dzhamalov & Semedov 2006: 309b]. Borrowed from Azerbaijani yekä ~ äkä 'big, large'.
A third candidate is qʼɨx-dɨ {кьыхьды} 'big' [Dzhamalov & Semedov 2006: 167, 323; Comrie & Khalilov 2010: 583] (in [Comrie & Khalilov 2010] erroneously quoted as kʼɨx-dɨ {кIыхьды}), but examples point rather to a more abstract sense, cf. "capital letters" [Dzhamalov & Semedov 2006: 79a], "He is a great dandy" [Dzhamalov & Semedov 2006: 90b].
In the Borch-Khnov dialect, only the Azerbaijani loanword äkːä-dɨ 'big' [Ibragimov 1978: 231] (quoted without specifications) is attested.
Both roots (Mukhad-Luchek qʼʷaxʸ- ~ qʼux- and Ixrek qˤač-) seem to be inherited, although their etymology is unknown. Final -dɨ / -d is the attributive suffix.
Kibrik & Kodzasov 1990: 237; Suleymanov 2003: 18, 190. In [Suleymanov 2003], erroneously quoted as ʁaba-ni-r {гъабанир} and ʁa-ni-r {гъанир}.
A different Burshag term for 'big' is quoted by Shaumyan: aHa-r 'big' [Shaumyan 1941: 176], and the same word in the Arsug subdialect: aHa-d 'big' [Shaumyan 1941: 176]. According to [Suleymanov 2003: 85], however, Koshan aHa-d {ахIад} has the more specific meaning 'big, huge, enormous'. On the contrary, the Arsug or Khudig word for 'big' is quoted as ʁaba-ni-d in [Suleymanov 2003: 18, 190].
It should be noted, however, that, according to [Magometov 1970: 86], the Richa term for 'big' is Ha- (ħa?). The same in the Usug subdialect: aχˤe-f 'big' [Shaumyan 1941: 176].
Dirr 1907: 99, 169; Shaumyan 1941: 176. Cf. the examples: "big river" [Dirr 1907: 13], "big house", "big woman" [Dirr 1907: 17, 19]. Not quoted in [Kibrik & Kodzasov 1990].
A second term is qːaba-n- (> qːaba-m-f), which is quoted in [Kibrik & Kodzasov 1990: 237] as a neutral term for 'big', but in [Dirr 1907: 128] is glossed as 'big, huge'. In the discovered examples qːaba-n- is applied to an evil monster [Dirr 1907: 83] or to a human penis, which cannot fit the fox burrow [Dirr 1907: 76].
A third term for 'big' is azmam-f [Kibrik & Kodzasov 1990: 237], borrowed from Azerbaijani azman 'huge, enormous', dialectal 'large' (ultimately from Arabic ʕazˤiːm 'big').
The same adjective in other subdialects: Kurag aHa-f 'big' [Magometov 1970: 170], Tsirkhe äχä-f 'big' [Shaumyan 1941: 176], Duldug aHa-f 'big' [Shaumyan 1941: 176].
Distinct from Tpig azman-f, glossed as 'big, large' in [Suleymanov 2003: 18] (an Azerbaijani loanword, see notes on Gequn Aghul).
Common Aghul notes:
Data from Koshan and Keren dialects are rather discrepant, but it is clear that two Proto-Lezgian roots enter into competition in Aghul: *ʔaχˤɨ- and *pːVhˤV- / *hˤVpːV-. Both scenarios are possible for Proto-Aghul: *ʔaχˤɨ- could be the neutral term for 'big' (retained in non-Koshan dialects, but shifted to 'huge' in Koshan), whereas *hˤVpːV- '?' acquired the generic meaning 'big' in Koshan. On the other hand, *hˤVpːV- could be the Proto-Aghul term for 'big' (retained in Koshan, but lost in non-Koshan dialects), whereas *ʔaχˤɨ- meant 'huge' in Proto-Aghul, retained in Koshan, but shifted to generic 'big' in the non-Koshan dialect. External Lezgian comparison speaks in favor of the second solution.
Keren baba- is of unknown origin. It is proposed in [NCED: 316] to treat baba- as a reduplicated cognate of Koshan ʢaba-.
On the sporadic reduction of initial a- in the Aghul dialectal descendants of *ʔaχˤɨ- see [Suleymanov 1993: 42 f.] with other examples of this process.
Final -d, -t, -f, -r are adjectival suffixes (fossilized class exponents) [Magometov 1970: 92; Shaumyan 1941: 45].
Kibrik & Kodzasov 1990: 237. The tenseness of the uvular is obviously secondary.
The same in the Khanag subdialect: aχˤˈi with polysemy: 'big / elder' [Uslar 1979: 600, 989; Dirr 1905: 154, 224]. The same in the Khyuryuk subdialect: aχˤˈi {аьхи} 'big' [Genko 2005: 24].
The same in the Khiv subdialect: aχˤˈu {аьхю} with polysemy: 'big / elder' [Genko 2005: 24]. The same in Literary Tabasaran: aχˤˈu {аьхю} 'big; elder' [Khanmagomedov & Shalbuzov 2001: 71].
Literary Lezgi: čʼeχˈi {чIехи} with polysemy: 'big / elder' [Talibov & Gadzhiev 1966: 378; Gadzhiev 1950: 59; Haspelmath 1993: 485, 516]. A close synonym is literary yekˈe {еке} 'big, large' [Talibov & Gadzhiev 1966: 117; Gadzhiev 1950: 59; Haspelmath 1993: 493], but this term is less frequent. Lezgi yeke was borrowed from Azerbaijani yekä ~ äkä 'big, large'.
Only the Azerbaijani loanword in the Akhty dialect: Khlyut yekˈe 'big' [Kibrik & Kodzasov 1990: 237].
Cf. the inherited term for 'big' in the Yarki dialect (Kyuri group): Nyutyug čʼeqi [Meylanova 1964: 96]. In the Yargun dialect (Quba group), čʼeχi means 'great; elder', whereas the generic word for 'big' is the borrowing yike ~ yeke [Babaliyeva 2007: 38, 48, 105, 106].
The historical phonetics of Lezgi dialects requires additional investigation, but the fluctuation q ~ χ (Gyune/Yarki čʼiqˈi/čʼeqi ~ literary/Yargun čʼeχˈi) seems irregular. Proceeding from general premises, one can suppose that the affricate q is primary here rather than the lenited χ, although external comparison speaks in favour of χ.
Proto-Lezgian:*pːVhˤV- ~ *hˤVpːV-6
NCED: 316. Distribution: From the distributional point of view, the best candidate is Proto-Lezgian *pːVhˤV- ~ *hˤVpːV-, which is retained in the generic meaning 'big' in Caucasian Albanian, South Lezgian (Kryts, Budukh) and Proto-Aghul, but underwent the shift 'big' > 'good' in Archi [NCED: 316].
The second candidate is *ʔaχˤɨ- [NCED: 511], attested as generic 'big' in Tsakhur, Tabasaran, non-Koshan Aghul. However, actually, *ʔaχˤɨ- might have been the Proto-Lezgian root for 'many' q.v. The assumed shift 'many' > 'big' cannot be an inherited feature of Tsakhur, Tabasaran and non-Koshan Aghul, but rather represents parallel independent innovations in Tsakhur and Tabasaran-Aghul (the Tabasaran-Aghul isogloss is apparently of an areal origin; the Proto-Aghul meaning of this root was probably something like 'huge').
The third candidate is *čʼaχV (~ -ä-) [NCED: 386], which means 'big (in general)' in Lezgi and 'massive' in Tabasaran. Despite some interesting external North Caucasian comparanda, this can hardly be considered a good candidate for Proto-Lezgian 'big'.
Various etymologically isolated roots for 'big' are found in Archi and the Rutul dialects; these do not look like loans, but lack Lezgian and North Caucasian cognates.
Inherited forms were superseded with loanwords in some lects (modern Udi < Iranian, Akhty Lezgi < Azerbaijani).
Chumakina et al. 2007; Kibrik & Kodzasov 1990: 86; Kibrik et al. 1977b: 286, 377; Kibrik et al. 1977a 1: 123; Comrie & Khalilov 2010: 141. Polysemy: 'bird (in general) / gusset (in a sleeve)'. Paradigm: nocʼ [abs.] / nacʼ-ˈa [erg.].
In [Mikailov 1967: 194] nocʼ is glossed as 'small bird (in general); sparrow', although translated as 'bird (in general)' in the texts: "A bird is sitting on the branch" [Mikailov 1967: 60], "He, like a bird, has thrown himself upon the sea steed" [Mikailov 1967: 160, 162]. Similarly in [Dirr 1908: 170, 219] nocʼ is glossed as "sparrow; small bird (in general)". It must be noted that in [Dirr 1908: 165, 219] lˈilǝχːˤu is quoted as 'bird (in general)', although the actual meaning of lˈilǝχːˤu is 'a mythological bird' [Chumakina et al. 2007; Kibrik et al. 1977b: 274; Mikailov 1967: 190] (borrowed from Lak liluχːˤi 'bird (in general)').
It is probable that in Proto-Archi the term nocʼ denoted a 'small/middle bird (in general)' as opposed to specific names of large predatory birds (a typologically possible lexicological situation).
Note that the Archi word is unjustifiedly labeled as "probably borrowed" in [Chumakina 2009].
In [Comrie & Khalilov 2010: 141] the word čaʁ {чIагъ} is also quoted as a synonym for 'bird (in general)', apparently < Lezgian *čːaqʼʷ(a) 'bird, small bird', although the expected Kryts form should rather be **čaqʼ {чIакъ} (Comrie & Khalilov's error?).
Alyk Kryts:qːuš-1
Authier 2009: 52, 76, 87, 181, 223, 259, etc. Borrowed from Azerbaijani guš 'bird'.
In [Comrie & Khalilov 2010: 141], 'bird' is glossed as ǯiv-ǯiv {джив-джив} - apparently a corrupted spelling of the onomatopoeic term ǯib-ǯib 'chicken' [Meylanova 1984: 57].
Kibrik et al. 1999: 888, 898; Comrie & Khalilov 2010: 141. In [Comrie & Khalilov 2010], quoted as šitʼ {шитI}, which is not an error, cf. the same form from [Ibragimov & Nurmamedov 2010] below.
Cf. examples in [Kibrik et al. 1999]: "If only a human had wings, he would fly like a bird" [Kibrik et al. 1999: 263]; "The bird ate the grain, but it was eaten itself by the hawk" [Kibrik et al. 1999: 677]; "The birds have flown away" [Kibrik et al. 1999: 218]; "Ali saw a rock, a bird was flying over it" [Kibrik et al. 1999: 391], "The bird is sitting on the tree" [Kibrik et al. 1999: 469]; "Ali fired at the bird that was flying over him" [Kibrik et al. 1999: 649]; "When the bird caught the snake, it ate it up" [Kibrik et al. 1999: 678]. Two first instances demonstrate that šitʼʸ generally denotes 'small/mid-size bird (in general)', but can be extended to the generic meaning 'bird'.
In [Ibragimov & Nurmamedov 2010: 221], only qːuš {къуш} 'bird' (borrowed from Azerbaijani guš 'bird') is quoted as a separate entry. Examples include: "There are a lot of birds in our woods" [Ibragimov & Nurmamedov 2010: 221], "A bird flies with the help of its wings" [Ibragimov & Nurmamedov 2010: 213], "The falcon is a predatory bird" [Ibragimov & Nurmamedov 2010: 217]. However, the word šitʼ / šitʼʸ- {шитI, шитIяр} is attested in examples like "He was snaring birds" [Ibragimov & Nurmamedov 2010: 51], "Morwennol is a pretty bird" [Ibragimov & Nurmamedov 2010: 199], "(Scarecrows) frightened birds on the grain field" [Ibragimov & Nurmamedov 2010: 354], "Babble of birds" [Ibragimov & Nurmamedov 2010: 403].
Kibrik & Kodzasov 1990: 86; Dirr 1913: 198, 236. A generic term, borrowed from Azerbaijani guš 'bird'.
Distinct from the inherited specific term šitʼʸi 'small bird (in general)' [Kibrik & Kodzasov 1990: 86] (in [Dirr 1913: 217, 236] quoted as šitʼ).
Gelmets Tsakhur:qːuš-1
Kibrik & Kodzasov 1990: 86. Borrowed from Azerbaijani guš 'bird'. In [Comrie & Khalilov 2010: 141], 'bird' is glossed with the enigmatic form šikʼʸey {шикIей}.
Common Tsakhur notes:
It is probable that in Proto-Tsakhur the term šitʼʸ(i) denoted 'small/middle bird (in general)' as opposed to specific names of large predatory birds (a typologically possible lexicological situation). Recently the Azerbaijani word guš has been borrowed for generic 'bird', although it should be noted that the old Tsakhur form šitʼʸ(i) seems etymologically obscure, so it could be an old loanword from an unknown source.
Mukhad Rutul:naχčir {нахчир}-1
Dirr 1912: 162, 198; Ibragimov 1978: 53. Oblique stem: naχčir-di-. A borrowed term, because the oblique stem marker -di- is characteristic of loanwords [Alekseev 1994a: 221] (the source is Iranian, see notes on Tabasaran).
Distinct from the inherited šuruk 'sparrow; small bird (in general)' [Dirr 1912: 183], 'sparrow' [Ibragimov 1978: 116], although in [Ibragimov 1978: 282; Makhmudova 2001: 12, 20, 86, 159] and [Comrie & Khalilov 2010: 141] Mukhad šuruk is glossed as the generic 'bird'.
Ixrek Rutul:šey {шей}-1
Dzhamalov & Semedov 2006: 302; Ibragimov 1978: 204. Polysemy: 'beast (in general) / bird (in general)'. In [Dzhamalov & Semedov 2006: 383], 'bird' is additionally specified as läwši-dɨ šey {лаьвшиды шей}, literally 'flying beast'. Cf. the example: "bird's wing" [Dzhamalov & Semedov 2006: 152], "The birds have flown from the tree" [Ibragimov 1978: 204]. Oblique stem: šey-di-. A borrowed term, because the oblique stem marker -di- is characteristic of loanwords [Alekseev 1994a: 221] (the source is Persian, see notes on Tabasaran).
A second word for 'bird' is qːuš {къуш} [Dzhamalov & Semedov 2006: 160], but it is probably less frequent - no examples found; borrowed from Azerbaijani guš 'bird'.
Distinct from the inherited šuruk 'sparrow' [Dzhamalov & Semedov 2006: 305] (which is quoted in [Comrie & Khalilov 2010: 141] as a generic term for 'bird').
Luchek Rutul:šey-1
Kibrik & Kodzasov 1990: 86. A Persian loanword, see notes on Tabasaran.
Distinct from the inherited širuk 'young (of animal), nestling; small bird (in general)' [Kibrik & Kodzasov 1990: 86, 219].
Common Rutul notes:
Borch-Khnov dialect: naχčir [Ibragimov 1978: 282] (a Persian loanword, see notes on Tabasaran).
As in some other Lezgian cases, it is likely that in Proto-Rutul the term širuk denoted 'small/middle bird (in general)' as opposed to specific names of large predatory birds. Recently several foreign words have been borrowed for the generic meaning 'bird'.
Kibrik & Kodzasov 1990: 86. It must be noted that in [Shaumyan 1941: 157], Burshag 'bird' is quoted as (borrowed) naχšir, whereas ǯaqʼʷ is glossed as 'sparrow' in [Suleymanov 2003: 80].
Keren Aghul:
Not attested. Cf. in the Usug subdialect: naχčir 'bird' [Shaumyan 1941: 157].
Distinct from Richa žaqʼʷ 'small bird (in general)' [Kibrik & Kodzasov 1990: 86], although Magometov translates it simply as 'bird' in examples [Magometov 1970: 136].
Gequn Aghul:naχšir-1
Dirr 1907: 136, 181; Shaumyan 1941: 156.
Distinct from inherited ǯaqʼʷ 'small bird (in general)' [Kibrik & Kodzasov 1990: 86; Dirr 1907: 114, 181].
Fite Aghul:
The generic term is unknown. Cf. ǯaqʼʷ 'small bird (in general)' [Kibrik & Kodzasov 1990: 86].
Aghul (proper):qːuš-1
Suleymanov 2003: 118. A recent borrowing from Azerbaijani guš 'bird', which has superseded the Tpig word naχšir 'bird' [Shaumyan 1941: 156]. The latter is of Iranian origin. Distinct from inherited ǯaqʼʷ 'sparrow' [Suleymanov 2003: 80].
In other subdialects: Tsirkhe naχšir, Duldug naχčir 'bird' [Shaumyan 1941: 156].
Common Aghul notes:
As in some other cases typical of Lezgian languages, it is likely that in Proto-Aghul the term ǯaqʼʷ denoted 'small/middle bird (in general)' as opposed to specific names of large predatory birds. It acquired the generic meaning 'bird' in Koshan (if Kibrik & Kodzasov's data is correct), while other dialects introduced the Iranian loanword naχčir ~ naχšir for 'bird (in general)' (see notes on Tabasaran).
Northern Tabasaran:
The generic Dyubek term is unknown.
In other subdialects the loanword šeyʔ is used for 'bird': Khanag šeyʔ 'a thing; bird' [Uslar 1979: 981, 1003; Dirr 1905: 220, 240]; according to [Dirr 1905: 218], there is also a more detailed expression for 'bird' in Khanag: čʼeyi šeyʔ, literally 'living/alive šeyʔ'), Khyuryuk šeyʔ {шейъ} 'a thing, creature, œuvre; bird' [Genko 2005: 189]. Another loanword in the Kumi subdialect: naχšˈir {нахшир} 'bird' [Genko 2005: 126].
Distinct from inherited Dyubek ǯˈaqʼ-a 'small bird (in general)' [Kibrik & Kodzasov 1990: 86], Khanag, Khyuryuk, Kumi ǯaqʼ {жжакь} 'small bird (in general)' [Dirr 1905: 169, 240; Genko 2005: 65].
Southern Tabasaran:
The generic Kondik term is unknown.
In other subdialects only loanwords are attested: Khiv ničχˈir {ничхир}, Tinit naχšˈir {нахшир} 'bird' [Genko 2005: 126, 127]; a second Khiv term is šeyʔ {шейъ} with polysemy: 'a thing, creature, œuvre / bird' [Genko 2005: 189].
Distinct from inherited Kondik, Khiv ǯaqʼʷ {жжакьв} 'small bird (in general)' [Kibrik & Kodzasov 1990: 86; Genko 2005: 65].
A different situation in Literary Tabasaran: inherited ǯaqʼʷ {жакьв} 'bird (in general); sparrow' [Khanmagomedov & Shalbuzov 2001: 157], which is opposed to borrowed ničχˈir {ничхир} 'wild bird; wild beast' [Khanmagomedov & Shalbuzov 2001: 244]. Cf. šeyʔ {шейъ} 'thing, object' [Khanmagomedov & Shalbuzov 2001: 349].
Common Tabasaran notes:
As in some other cases typical of Lezgian languages, it is likely that in Proto-Tabasaran the term ǯaqʼʷ denoted a 'small/middle bird (in general)' as opposed to specific names of large predatory birds. It acquired the generic meaning 'bird' in Literary Tabasaran (if Khanmagomedov & Shalbuzov's gloss is correct), while other dialects introduced the Iranian loanwords for 'bird (in general)'.
The forms naχšir~ ničχir originate from Iranian, cf. Persian naxčiːr 'hunting, the chase; game, wild animal', Tajiki naxčir 'wild animal', etc. The Literary Tabasaran meaning 'wild bird; wild beast' and the Gyune Lezgi meaning ‘game’ (see notes on Lezgi ‘bird’) directly correspond to the Iranian semantics. This term penetrated into some other Lezgian languages with the modified meaning 'bird (in general)'.
The word šeyʔ with polysemy: 'a thing, creature, œuvre; bird' was borrowed from Persian šayʔ 'a thing, something, object' (ultimately from Arabic šey 'a thing, something'). The semantic development 'a thing' → 'animal' can represent an internal Lezgian development. Cf. the polysemy in Ixrek Rutul: šey 'beast (in general) / bird (in general)' and the attested full collocation for 'bird': Northern Tabasaran (Khanag) 'living/alive šeyʔ', Ixrek Rutul (q.v.) 'flying šey'.
Gyune Lezgi:qːuš-1
Uslar 1896: 495, 628. Borrowed from Azerbaijani guš 'bird'. Distinct from the inherited Gyune word nükʼ 'sparrow' [Uslar 1896: 516]. Distinct from the Iranism ničχˈir 'game', luwˈan ničχˈir 'game bird' (literally 'ničχir with wing') [Uslar 1896: 516]. For the Iranian origin of ničχˈir see notes on Tabasaran.
The same Azerbaijani loanword in Literary Lezgi: qːuš {къуш} 'bird' [Talibov & Gadzhiev 1966: 195; Gadzhiev 1950: 671; Haspelmath 1993: 502, 516]. The second literary expression for 'bird (in general)' is luwan ničχir {луван ничхир} [Talibov & Gadzhiev 1966: 254; Gadzhiev 1950: 671], literally 'ničχir with wing', whose original meaning was the narrower 'game bird', see Uslar's data above. Modern literary ničχˈir {ничхир} means 'game bird, wild bird' [Talibov & Gadzhiev 1966: 254; Haspelmath 1993: 500, 516], originating from the wider meaning 'game', see Uslar's data above.
Distinct from literary inherited nükʼ {нуькI} 'small bird (in general), sparrow' [Talibov & Gadzhiev 1966: 255; Haspelmath 1993: 500, 516].
The same in the Akhty dialect: Khlyut qːuš 'bird (in general)' [Kibrik & Kodzasov 1990: 86]; distinct from the inherited Khlyut word nucʼ 'small bird (in general)' [Kibrik & Kodzasov 1990: 86].
Proto-Lezgian:*nɨсʼʷ1
NCED: 525. Distribution: Most of the attested Lezgian languages demonstrate the lexical opposition between a term for 'small/middle bird (in general)' and various terms for specific kinds of large (predatory) birds. There are no reasons not to project such a system onto the Proto-Lezgian level. Thus, we fill the Proto-Lezgian slot with the generic term for 'small/middle bird'.
Out of several inherited roots for 'small/middle bird', *nɨcʼʷ(a) possesses the best distribution: Archi 'small/middle bird (in general)', Lezgi 'small bird (in general)', also Khnyukh (subdialect of Mukhad Rutul) nicʼ 'a k. of bird' [Ibragimov 1978: 135]. This root also has a good North Caucasian etymology ('small bird' or 'bird').
Cf. other roots, attested with the meaning 'small bird' in Lezgian: *čːaqʼʷ(a) (Aghul, Tabasaran), for which external North Caucasian comparison suggests the shift 'a k. of small bird' > 'small bird' [NCED: 1105]; *š(ʷ)Vrtʼ / *čʼVˤrtʼ (Tsakhur), apparently with the shift 'a k. of small bird' > 'small bird' [NCED: 343]; *čɨraƛʼːʷ (Rutul), for which the Lezgian comparandum suggests the shift 'nestling, young of birds' > 'small bird' [LEDb: #161]; *čVpːV(?) (Udi), an isolated form [LEDb: #267].
The majority of Lezgian languages have recently borrowed their generic terms for 'bird' from Azerbaijani or Persian.
Replacements: {'a k. of small bird' > 'small bird'} (Aghul, Tabasaran, Tsakhur), {'nestling, young of birds' > 'small bird'} (Rutul).
Reconstruction shape: Correspondences are regular.
Semantics and structure: Primary substantive root. The oblique stem is *nVсʼʷa-.
Gukasyan 1974: 133; Comrie & Khalilov 2010: 233; Mobili 2010: 162. In [Comrie & Khalilov 2010] erroneously quoted as kaš-p-sun {кашпсун}. In [LEDb] this root is quoted as kːašˤ-, but without references. Polysemy: 'to dig (a hole etc.) / to bite / to sting'.
Gukasyan 1974: 133. Polysemy: 'to dig (a hole etc.) / to bite / to sting'. In [Schulze 2001: 289] attested only with the meaning 'to dig' (kašˤ-p-esun). Apparently the same verb is quoted in [Schiefner 1863: 82] as kːačː-p-esun {kać̣-} 'kauen' and in [Dirr 1903: 95 (l. 14)] as kač- {қач-} 'to dig'.
Common Udi notes:
Common Udi *kašˤ-p-esun 'to dig (a hole etc.) / to bite / to sting'; in [Gukasyan 1974: 133; Mobili 2010: 162] the reduplicated Nidzh-Vartashen variant kašˤ-kašˤ-p-esun {кашIкашIпесун} is also cited as a synonym. The morpheme -p- is a light verb: 'to say, to do smth. with the mouth; to do smth. (in general)' [Schulze 2005: 565 ff. (3.4.2.2 #15 ff.); Harris 2002: 204 ff.].
Not to be confused with kːacː-pː-esun {кIац́пIесун} 'to cut, split, cut down' [Gukasyan 1974: 141] = kːacː-p-esun {kac̣-} 'zerschneiden, zerstören, vernichten' [Schiefner 1863: 82], 'to kill, destroy, slay' [Schulze 2001: 291].
Distinct from Nidzh-Vartashen kːaram-p-esun 'to gnaw, gnaw round; to nag' [Gukasyan 1974: 140; Mobili 2010: 172] (translated as simply 'to bite' in [Starchevskiy 1891: 486]).
Distinct from the complex verb hˈaˤnčʼ-bo- 'to bite off; to nibble grass, depasture', formed with the suppletive light verb -bo- 'to say' [Chumakina et al. 2007; Kibrik et al. 1977b: 238].
Initial acʼ= is a prefix with general semantics [Ibragimov 1990: 124; Kibrik & Kodzasov 1988: 41]. Alternatively the Common Tsakhur verb can be analyzed as a=cʼakʼʷan- with the prefix a= (for which see [Ibragimov 1990: 123]), thus [LEDb].
Dzhamalov & Semedov 2006: 236, 351. Cf. an example: "He will not bite the finger put in his mouth" (proverb referring to a honest person) [Dzhamalov & Semedov 2006: 251b]. In [Comrie & Khalilov 2010: 233], the auxiliary verb is quoted with an error.
According to [Kibrik & Kodzasov 1988], one other synonym is the verb yitʼ- 'to bite', paradigm: r=itʼ-a-r- [imperf. 4] / y=itʼ-ɨ-r [perf. 4] / y=itʼ [imv. 4] (initial r= is the imperfective exponent; y= is a prefix with general semantics or the fossilized class 1/4 exponent). In the Borch-Khnov dialect, the corresponding verb w=itʼ- means 'to bite', applied to a snake [Ibragimov 1978: 276, 300]. This root is isolated within Lezgian, cf. [NCED: 227].
Common Rutul notes:
The analytic construction sɨs 'incisor / canine tooth' + (h=)aʔ- (/ h=äqʼ-) 'to do' represents the meaning 'bite' in all dialects. See notes on 'tooth'.
Borch-Khnov dialect: gɨčʼ haʔ- {гычI гьыъын} 'to bite' [Ibragimov 1978: 284], literally gɨčʼ + 'to do'. The Rutul noun gɨčʼ is unattested outside this expression, but regularly reflects Proto-Lezgian *gamčʼ 'canine tooth; molar tooth' [NCED: 430].
Kibrik & Kodzasov 1988: 159; Suleymanov 2003: 121; Magometov 1970: 229 (sentence 4). In [Shaumyan 1941: 136], the Burshag verb ayki- 'to bite' is quoted, which seems an inaccuracy, see notes on common Aghul.
A complex verb, consisting of the noun qʼacʼ, which means 'a piece' in Koshan [Shaumyan 1941: 186; Magometov 1970: 235 sentences 3] (cf. Magometov's example "The miller gave the fox a piece of bread"; in [Suleymanov 1993: 83], however, qʼacʼ is mentioned as the common Aghul substantive 'bite') plus the verb ʔaykʼi- (Koshan) / ʔikʼa- (Non-Koshan) 'to put in, move into (trans.)' [Suleymanov 2003: 210].
The same in Literary Tabasaran: qʼacʼ apʼ- {кьацI апIуб} 'to bite' [Khanmagomedov & Shalbuzov 2001: 209], literally qʼacʼ 'a bite; a piece' + apʼ- 'to do'.
A similar construction in the Khiv subdialect, but with a different noun: ʁancʼ apʼ- {гъанцI апIуб} 'to bite' [Genko 2005: 41], literally ʁancʼ 'a bite (e.g., of snake)' + the auxiliary verb ap- 'to do'. Distinct from Khiv class=is- {бисуб} 'to seize; hold; to bite (said of dog)' [Genko 2005: 29].
Common Tabasaran notes:
The collocation qʼacʼ 'a bite; a piece' + an auxiliary verb can be reconstructed as the Proto-Tabasaran term for 'to bite'. The Khiv construction with ʁancʼ 'a bite' is an innovation from the distributive point of view.
The same in Literary Lezgi: kʼˈas- {кIасун} 'to bite' [Talibov & Gadzhiev 1966: 212; Gadzhiev 1950: 326; Haspelmath 1993: 496, 516; Gyulmagomedov 2004, 1: 455]. Distinct from literary qʼa- {кьун} 'to hold; to seize; to bite (said of dog)' [Talibov & Gadzhiev 1966: 207].
Proto-Lezgian:*ʔeqʼːɨ-2
NCED: 559. Distribution: First of all, we need to exclude analytic patterns of the shape 'tooth' / 'a piece' / 'a bite' + an auxiliary verb, which are attested in Rutul, Aghul, Tabasaran. These look like recent introductions of areal origin.
After that, several verbal roots enter into competition: (1) Udi kašˤ-, isolated root; (2) *ʔeqʼːɨ- (Archi); (3) *kʼosʷɨ- (South Lezgian [Kryts, Budukh] and Lezgi); (4) the Tsakhur verb, whose morphological analysis is not entirely clear. From the distributional point of view, all of them are equivalent candidates. We choose *ʔeqʼːɨ- (Archi, lost in the rest of Lezgian languages), since it actually stems from a good candidate for the status of the Proto-North Caucasian verb for 'to bite' [NCED: 559].
If so, the Udi polysemy kašˤ- 'to bite / to dig' suggests an earlier shift 'to dig' > 'to bite', or else both synchronic meanings originate from *'to break' (i.e. 'to break' > 'to dig'). Pace [LEDb: #11], the Udi root has plain k-, not tense kː-, therefore, comparison with Proto-Lezgian *kʼosʷɨ- should be rejected.
The root *kʼosʷɨ- (South Lezgian, Lezgi) is an inner Samur introduction for 'to bite', although its semantic origin is unknown, since it lacks any cognates outside South Lezgian and Lezgi. Formally *kʼosʷɨ- can be reconstructed as the Proto-Nuclear Lezgian verb for 'to bite'. Cf. also the etymologically obscure Kryts form kʼɨp- 'to bite', which serves as the perfective stem in the suppletive paradigm.
Replacements: {'to dig' > 'to bite' (?)} (Udi), {'to do a tooth' > 'to bite'} (Rutul), {'to put a piece in' > 'to bite'} (Aghul), {'to put a bite' > 'to bite'} (Northern Tabasaran), {'to do a bite' > 'to bite'} (Southern Tabasaran).
Gukasyan 1974: 172; Fähnrich 1999: 23; Dirr 1903: 2; Schiefner 1863: 104; Starchevskiy 1891: 495. In [Fähnrich 1999] and [Starchevskiy 1891] quoted as maˤyn.
Common Udi notes:
The suffixal -n is not clear; in [Schulze 2001: 298] and [Schulze 2005: 131 (3.2.2.3 #5)] this is treated as a genitive exponent, that is, the underlying Proto-Udi stem should be substantival with the meaning 'blackness' (*'of blackness' > 'black'). It is not obvious, however, that all Udi nouns with the suffix -n- must be analyzed as old qualifying genitive forms.
Chumakina et al. 2007; Kibrik & Kodzasov 1990: 234; Kibrik et al. 1977b: 203, 388; Comrie & Khalilov 2010: 673; Mikailov 1967: 175; Dirr 1908: 134, 226. In [Comrie & Khalilov 2010: 673] there is also a corrupted variant beːχe-du-class. Regular participle from the stative verb beχːˤˈe 'to be black' [Chumakina et al. 2007; Kibrik et al. 1977b: 202].
The same in the Khanag subdialect: kʼerˈi 'black' [Uslar 1979: 799, 1010; Dirr 1905: 190, 246]. The same in the Khyuryuk subdialect: kʼerˈi {кIери} 'black' [Genko 2005: 112].
The same in other subdialects:Khiv kʼarˈu {кIару}, Tinit kʼarˈi {кIари} 'black' [Genko 2005: 111]. The same in Literary Tabasaran: kʼarˈu {кIару} 'black' [Khanmagomedov & Shalbuzov 2001: 213].
The same in Literary Lezgi: čʼulˈaw {чIулав} 'black' [Talibov & Gadzhiev 1966: 379; Gadzhiev 1950: 929; Haspelmath 1993: 485, 516].
The same in the Akhty dialect: Khlyut čʼlaw 'black' [Kibrik & Kodzasov 1990: 234].
Proto-Lezgian:*laχːˤV- ~ *loχːˤV-2
NCED: 748. Distribution: Four roots enter into competition here. Out of these, *laχːˤV- ~ *loχːˤV- seems to be the best candidate from the distributive point of view. It has been retained as the basic root for 'black' in Archi, on the one hand, and in South Lezgian (Kryts, Budukh) and Rutul, on the other, but got lost in other languages. Its only external comparandum is the basic Lak term for 'black' [NCED: 748].
The second candidate is *kʼarɨ-, attested with the meaning 'black' in Tsakhur, Aghul and Tabasaran. This was lost as a separate word in the rest of Lezgian languages, but survived in the compound *kʼarV-šːim 'charcoal', literally 'X pebbles' in Aghul, Rutul, Kryts, Budukh [NCED: 719]. External North Caucasian comparison could point, however, to the original meaning 'charcoal' for this root (cf. Proto-Nakh 'charcoal'), thus the Lezgian compound *kʼarV-šːim might actually mean 'charcoal pebbles' rather than the more trivial 'black pebbles'. If so, one must assume the shift 'charcoal' > 'black' that occurred independently in Tsakhur and Aghul-Tabasaran.
Two residual roots, attested with the generic meaning, should be excluded due to their distribution. The root *mičʼ[ä]- means 'black' in Udi, but 'dark' in other Lezgian languages including Archi, thus the Proto-Lezgian semantic reconstruction 'dark' is very probable; external North Caucasian comparison, however, is not unambiguous, because the Khinalug cognate of this Lezgian root means 'black' (further to Nakh 'yellow, orange', Avar 'dark grey, yellow', Lak 'blind') [NCED: 819].
The fourth root is *čʼulV (~ -o-) which denotes 'black' in Lezgi. This got lost in other Lezgian languages except for Aghul, where it survived in the expression for 'raspberry', literally 'dark/black berry' [NCED: 556]. Actually Lezgian *čʼulV possesses good North Caucasian (strictly speaking East Caucasian) comparanda with the meaning 'black' (Proto-Nakh, Proto-Avar-Andic), but it is hard to suppose that *čʼulV survived with its original meaning 'black' only in Lezgi; we should assume the meaning 'a k. of dark color' for Proto-Lezgian *čʼulV and the late development 'a k. of dark color' > 'black' in modern Lezgi.
Replacements: {'dark' > 'black'} (Udi), {'a k. of dark color' > 'black'} (Lezgi), {'charcoal' > 'black'} (Tsakhur, Aghul, Tabasaran).
Reconstruction shape: Correspondences seem regular, right up to the class prefix fusion in Archi.
Semantics and structure: Primary stative verbal root 'to be black'.
Following [NCED: 497, 879], we treat Rutul e-bir ~ e-bɨr as a compound of two old roots, where the second root acquired the final -r under the influence of the common plural exponent -b-Vr.
The same in the Khanag subdialect: i-fːˈi ~ ye-fːi 'blood' [Uslar 1979: 738, 996; Dirr 1905: 179, 232]. The same in the Khyuryuk subdialect: i-fːˈi {иффи} 'blood' [Genko 2005: 77].
The same in the Khiv subdialect: i-fˈi {ифи} 'blood' [Genko 2005: 77]. The same in Literary Tabasaran: i-fˈi {ифи} [Khanmagomedov & Shalbuzov 2001: 180].
Common Tabasaran notes:
Following [NCED: 497, 1064], we treat Tabasaran i-fːi ~ i-wi as a compound of two old roots.
The same in Literary Lezgi: i-wˈi {иви} 'blood' [Talibov & Gadzhiev 1966: 132; Gadzhiev 1950: 319; Haspelmath 1993: 493, 516].
The same in the Akhty dialect: Khlyut e-wˈi 'blood' [Kibrik & Kodzasov 1990: 48].
Following [NCED: 497, 1064], we treat Lezgi e-wi as a compound of two old roots.
Proto-Lezgian:*ʔäʔ3
NCED: 496. Distribution: A difficult case, because formal internal reconstruction contradicts external data. There are two main roots for 'blood' in Lezgian languages: *pːiy (~ pʼ-) [NCED: 879] and *ʔäʔ [NCED: 496].
The plain root *pːiy is attested in the meaning 'blood' in Caucasian Albanian-Udi and Archi. In West Lezgian (Tsakhur, Rutul), the compound *ʔäʔ-*pːiy 'blood' is observed. Finally, in East Lezgian we see the compound *ʔäʔ-*ɬːʷiy 'blood' (Tabasaran, Lezgi) and the plain *ʔäʔ 'blood' (Aghul). The root *ɬːʷiy is unattested outside this compound, but its external North Caucasian comparanda clearly point to the meaning 'vein' [NCED: 1064]. Formal distribution suggests that the Proto-Lezgian root for 'blood' should have been *pːiy, retained in both of the outliers (Udi and Archi), whereas in Proto-Nuclear Lezgian this was superseded with *ʔäʔ, which was normally used as an element of compounds. The problem of this solution is that the status of *pːiy in the Proto-West Lezgian compound *ʔäʔ-*pːiy appears to be unclear.
Both discussed roots - *pːiy and *ʔäʔ - possess external East Caucasian cognates with the meaning 'blood' [NCED: 496, 879], but it is actually *ʔäʔ which reflects the main candidate for the status of at least the Proto-East Caucasian term for 'blood', whereas the original meaning of *pːiy seems to have been 'blood vessel'. If *ʔäʔ is to be posited as the Proto-Lezgian root for 'blood', we must assume that the root *pːiy independently shifted from 'blood vessel' > 'blood' in both of the outliers (Udi, Archi). On the other hand, both of the attested compounds for 'blood', *ʔäʔ-*pːiy (West Lezgian) and *ʔäʔ-*ɬːʷiy (East Lezgian), acquire the identical structure 'blood + vein'.
In South Lezgian (Kryts, Budukh), the substantive 'blood' is transparently derived from the adjective *ʔirɨ- (~ ʔˤ-) 'red' q.v. [NCED: 519].
Kibrik & Kodzasov 1990: 39; Comrie & Khalilov 2010: 181. Distinct from a more specific term mäsir 'tubular bone (in general)' [Kibrik & Kodzasov 1990: 39].
In [Comrie & Khalilov 2010: 181], the word čanaq {чанахъ} is incorrectly quoted as a second term for 'bone', although in reality čanaq means 'hip bone' [Meylanova 1984: 151].
Tsakhur-Kum Tsakhur: barkʼʷ [Kibrik & Kodzasov 1990: 39]. Distinct from the more specific term cʼom 'tubular bone (in general)' [Kibrik & Kodzasov 1990: 39].
The same in the Khanag subdialect: yirkː 'bone' [Uslar 1979: 753, 996; Dirr 1905: 180, 231]. The same in the Khyuryuk subdialect: yirkː {йиркк} 'bone' [Genko 2005: 80].
The same in the Khiv subdialect: kʼurˈab {кIураб} 'bone' [Genko 2005: 113]. In Literary Tabasaran both terms are present: yirkː {йиркк} 'bone' [Khanmagomedov & Shalbuzov 2001: 185] and kʼurˈab {кIураб} 'bone' [Khanmagomedov & Shalbuzov 2001: 214].
Common Tabasaran notes:
External comparison suggests that Northern yirkː 'bone' is an archaism.
The same in Literary Lezgi: kʼarˈab {кIараб} 'bone' [Talibov & Gadzhiev 1966: 212; Gadzhiev 1950: 312; Haspelmath 1993: 496, 516].
The same in the Akhty dialect: Khlyut kʼrab 'bone' [Kibrik & Kodzasov 1990: 39].
Proto-Lezgian:*yirƛʼː1
NCED: 528. Distribution: Two complementarily distributed roots for 'bone' enter into competition in this criss-crossing situation. The first one is *yirƛʼː [NCED: 528], attested in Udi and Caucasian Albanian on the one hand, and in some Nuclear Lezgian lects (namely Aghul, Northern Tabasaran) on the other. The second one is *ƛʼorapː [NCED: 779], common in Nuclear Lezgian: South Lezgian (Kryts, Budukh), Rutul, Southern Tabasaran, Lezgi.
Since the Samur territory generally demonstrates a high number of post-split, contact-induced lexical isoglosses between Nuclear Lezgian lects, the distribution seems to speak in favor of *yirƛʼː as the Proto-Lezgian term for 'bone'. External North Caucasian comparison clearly supports such a solution. On the contrary, *ƛʼorapː in the meaning 'bone' looks like an inner Nuclear Lezgian introduction, which further spread across the Samur territory as an areal isogloss (cf. especially the opposition *yirƛʼː / *ƛʼorapː between two Tabasaran dialectal clusters). External North Caucasian comparison suggests that the Proto-Lezgian meaning of *ƛʼorapː was 'hand bone' [NCED: 779] (actually *ƛʼora-pː with the fossilized plural suffix, used for body parts).
In Archi, *yirƛʼː 'bone' was superseded with *läk, whose original meaning was probably 'leg bone' [NCED: 755]: cf. the meaning 'knee' in Caucasian Albanian (see notes on 'knee') and 'foot, leg' in Tabasaran-Aghul (see notes on 'foot'). In Tsakhur, th meaning 'bone' is expressed by *pːalkʼʷ, originating from the meaning 'a k. of bone' [NCED: 310] (cf. its reflexes in other Lezgian languages: 'cheek-bone', 'spine', 'rib').
In [Comrie & Khalilov 2010: 217] and [Kibrik & Kodzasov 1990: 22] the word döš is also quoted for 'breast, chest' (the only term in [Kibrik & Kodzasov 1990], as a synonym of šˤaqː in [Comrie & Khalilov 2010]); it is borrowed from Azerbaijani döš 'breast, chest'.
Distinct from cːicːikː {цIицIикI} 'female breast, nipple' [Gukasyan 1974: 234; Kibrik & Kodzasov 1990: 23; Mobili 2010: 253]. In [Comrie & Khalilov 2010: 217], 'female breast' is translated as äq {аьхъ}, which is both an incorrect translation (in fact, a Vartashen form aqˤ for 'human breast') and incorrect spelling for {аъхъ}.
Gukasyan 1974: 57; Mobili 2010: 114; Fähnrich 1999: 7. Polysemy: 'human breast, chest / slope (of mountain)'. In [Schulze 2001: 271; Schiefner 1863: 97] and [Dirr 1903: 2] only döš 'breast' is quoted (< Azerbaijani döš 'breast, chest').
Distinct from cːicːikː 'female breast' [Fähnrich 1999: 12].
Common Udi notes:
Both Nidzh šˤaqː and Vartashen aqˤ seem secondary with the common semantic shift 'mountain slope' > 'breast'. Vartashen aqˤ possesses Lezgian comparanda, whereas Nidzh šˤaqː is probably isolated.
Caucasian Albanian: Unattested.
Archi:χˈatum3
Chumakina et al. 2007; Kibrik & Kodzasov 1990: 22; Kibrik et al. 1977b: 332, 356; Comrie & Khalilov 2010: 217; Mikailov 1967: 200; Dirr 1908: 188, 206. In [Comrie & Khalilov 2010: 217], there is also a word ikʼ {икI}, quoted as a synonym for χˈatum - apparently a corrupted spelling for ikʼʷ {икIв} 'heart' q.v.
Distinct from mam 'female breast, nipple, baby's dummy' [Kibrik & Kodzasov 1990: 22; Kibrik et al. 1977b: 277] (incorrectly glossed in [Chumakina et al. 2007] as 'breast, nipple').
Kibrik & Kodzasov 1990: 22. Not found in [Meylanova 1984].
According to [Meylanova 1984: 55, 210], the word for 'breast (in general, both male and female)' is duš {душ} (polysemy: 'human breast, chest / slope (of mountain), plateau'). But in [Kibrik & Kodzasov 1990: 23] duš is quoted only for 'female breast'. Borrowed from Azerbaijani döš 'breast, chest'.
Distinct from the inherited mɨχɨ 'female breast' [Kibrik & Kodzasov 1990: 23].
Common Tsakhur notes:
In all the dialects the inherited term for 'breast (in general)' mɨχɨ ~ muχu was narrowed to the meaning 'female breast', having been superseded by the loanword koksi ~ koksɨ, borrowed from Azerbaijani köks 'breast, chest'.
Kibrik & Kodzasov 1990: 22; Suleymanov 2003: 136; Shaumyan 1941: 152. It must be noted that in [Magometov 1970: 23] the Burshag word for 'breast' is transcribed as muχːur - an obvious error. The same term in the Arsug subdialect: muχur 'breast' [Shaumyan 1941: 152].
Distinct from Burshag bizi 'female breast, nipple' [Kibrik & Kodzasov 1990: 23].
Suleymanov 2003: 136; Shaumyan 1941: 152; Magometov 1970: 195 strophe 6. According to Magometov's example, applied to both men and women. No specific Tpig terms for 'female breast' have been found in [Suleymanov 2003].
The same in Tsirkhe and Duldug subdialects: muχur 'breast' [Shaumyan 1941: 152].
Common Aghul notes:
The common Aghul term bizi 'female breast, nipple' (see also the additional forms in [Shaumyan 1941: 154]) is of unknown origin; cf. [NCED: 305] with hypothetical West Caucasian comparanda.
Kibrik & Kodzasov 1990: 22. Distinct from the nursery word nenˈey 'female breast' [Kibrik & Kodzasov 1990: 22].
The same in the Khanag subdialect: muχˈur 'breast' [Uslar 1979: 864, 992; Dirr 1905: 197, 227], nanˈay with polysemy: 'female breast, nipple / udder / pupil of the eye' [Uslar 1979: 868, 992; Dirr 1905: 197, 227]. The same in other subdialects:Khyuryuk, Kumi muχˈur {мухур} 'breast' [Genko 2005: 123], Khyuryuk nanˈay {нанай} with polysemy: 'female breast, nipple / pupil of the eye' [Genko 2005: 125].
Kibrik & Kodzasov 1990: 22. Distinct from the nursery word nanˈa 'female breast' [Kibrik & Kodzasov 1990: 22].
The same in the Khiv subdialect: muχˈur {мухур} 'breast', nana {нана} 'female breast, nipple' [Genko 2005: 123, 125]. The same in Literary Tabasaran: muχˈur {мухур} 'breast', nanˈa {нана} 'female breast, nipple' [Khanmagomedov & Shalbuzov 2001: 235, 241] (distinct, however, from ninˈi {нини} 'pupil of the eye' [Khanmagomedov & Shalbuzov 2001: 244]).
Uslar 1896: 578, 609. Distinct from Gyune mam 'female breast, teat; udder' [Uslar 1896: 501, 609].
The same in Literary Lezgi: χur {хур} 'breast' [Talibov & Gadzhiev 1966: 349; Gadzhiev 1950: 154; Haspelmath 1993: 512, 516]. Distinct from literary mam 'female breast, teat' [Talibov & Gadzhiev 1966: 230; Gadzhiev 1950: 154].
The same in the Akhty dialect: Khlyut χɨr 'breast' [Kibrik & Kodzasov 1990: 22]. Distinct from Khlyut mam 'female breast' [Kibrik & Kodzasov 1990: 23].
Proto-Lezgian:*moχor4
NCED: 829. Distribution: This stem is retained in the meaning 'breast' in all Nuclear Lezgian lects, except for Tsakhur, where the basic term for 'breast' is an Azerbaijani loanword, whereas *moχor narrowed to 'female breast'. In Archi *moχor shifted to 'brisket' ('breast' > 'brisket' is natural, but not vice versa), having been superseded with the etymologically obscure form χˈatum. No traces of the root *moχor in Udi. External North Caucasian comparison confirms *moχor as the Proto-Lezgian term for 'breast (in general)'.
In two Udi dialects, 'breast' is expressed by forms with synchronic polysemy: 'breast / mountain slope', apparently with the development 'mountain slope' > 'breast'. At least for the Vartashen Udi form *ʔawχː(a), both internal and external comparison suggest the original meaning 'slope' [NCED: 244] (the Nidzh Udi term is etymologically isolated). It should be noted, however, that theoretically both directions of the semantic shift between 'breast' and 'slope' are possible.
Comrie & Khalilov 2010: 66. Quoted as bokː- 'to burn (tr.)' in [Gukasyan 1974: 283] and as bokː-os-b-esun 'to set fire to' in [Gukasyan 1974: 88] and [Mobili 2010: 68].
Common Udi *bokː-b-esun, a transitive/causative from Nidzh-Vartashen bokː-(e)sun 'to burn (intr.)' [Gukasyan 1974: 88; Comrie & Khalilov 2010: 66; Mobili 2010: 68; Fähnrich 1999: 10; Schulze 2001: 259], formed with the light verb -b- 'to do' [Schulze 2005: 569 ff. (3.4.2.2 #22 ff.)].
The root bokː- is synchronically unsegmentable; as in many other cases with Udi roots in b-, Lezgian cognates suggest that b- is a petrified prefix (a former class exponent, see [Harris 2002: 72 ff., 215 ff.] w. lit. and discussion), thus b=okː-.
Caucasian Albanian: The transitive verb is unattested. Cf. bokʼ-okʼ-esun 'to burn (intr.)' [Gippert et al. 2008: IV-11]; morphologically can be analyzed as bokʼ-o-kʼ-esun with the light verb -kʼ- as, e.g., in bas-kʼ- 'to lie, sleep' q.v. (the second -o- is unclear in this case) or as partially reduplicated bokʼ-okʼ- (cf. [Gippert et al. 2008: II-78]).
Kibrik & Kodzasov 1988: 95; Dirr 1913: 146. It should be noted that in [Kibrik & Kodzasov 1988: 94, 95], a labile verb gʸ=oxːan / gʸ=oxːɨn ' burn (intrans.); to burn (trans.)' is postulated, which seems an inaccuracy; see notes on Common Tsakhur.
The stem represents a regular causative formation from gʸ=oxːʷan- 'to burn (intrans.)', formed with the verb (h=)aʔ- 'to do' [Kibrik et al. 1999: 58 f.; Kibrik & Kodzasov 1988: 129].
Note that xʷ is still retained in some forms in Mishlesh and Literary Tsakhur, but it has been totally superseded by x in Mikik under the influence of numerous forms with the regular delabialization oxʷ > ox (the Gelmets data are unknown).
Initial gʸ= is a prefix with general semantics [Ibragimov 1990: 124; Kibrik & Kodzasov 1988: 41].
Ibragimov 1978: 121; Comrie & Khalilov 2010: 66. Causative from ux- 'to burn (intrans.)' [Dirr 1912: 176; Ibragimov 1978: 120], modified with the additional prefix h=. It should be noted that in [Makhmudova 2001: 252] this word is quoted as h=uxʷ- {гьудхьвас} (with the labialized -xʷ- - an archaism) and treated as a labile verb 'to burn (intrans.); to burn (trans.)'.
Distinct from the causative l=ikʼʷ-a aʔ- 'to set fire to' [Dirr 1912: 157; Ibragimov 1978: 120] from l=ikʼʷ- 'to catch fire' [Ibragimov 1978: 120; Makhmudova 2001: 159, 253].
Distinct from l=ikʼʷ-ä haʔ- 'to set fire to' (with the example: "They lit a fire") [Dzhamalov & Semedov 2006: 180] from l=ikʼʷ- 'to catch fire' [Dzhamalov & Semedov 2006: 180].
Distinct from k=irš-e haʔ- [Dzhamalov & Semedov 2006: 144], which is translated by Dzhamalov & Semedov as 'to burn (trans.)', but the only example "He has lit a splinter" points to the meaning 'to set fire to smth.'. Causative from k=irš-, which is translated in [Dzhamalov & Semedov 2006: 144] as 'to burn (intrans.)' (examples: "The firewood burns", "The lamp is lit"), but such a glossing also seems an inaccuracy.
In [Ibragimov 1978: 224], both Ixrek forms, h=ux- and k=irš-, are quoted as synonyms for 'to catch fire'; they are semantic counterparts of Mukhad l=ikʼʷ- 'to catch fire'.
Kibrik & Kodzasov 1988: 95, 96. Polysemy: 'to burn (trans.) / to set fire to'. Causative from l=ikʼʷ- 'to burn (intrans.); to catch fire' [Kibrik & Kodzasov 1988: 94, 95].
Common Rutul notes:
In all the dialects the equivalents of the meaning 'to burn (trans.)' represent regular causative formations from different verbs for 'to burn (intrans.)', formed with the verb (h=)aʔ- (/ h=äqʼ-) 'to do'. Luchek l=ikʼʷ- in the generic meaning 'to burn (intrans.)' is an innovation; external Lezgian comparison proves that the primary meaning of this root was 'to catch fire' as in Mukhad and Ixrek.
Initial h=, l= and k= are prefixes with general semantics [Ibragimov 1978: 95; Alekseev 1994a: 227; Makhmudova 2001: 165].
Kibrik & Kodzasov 1988: 95; Suleymanov 2003: 159. Labile verb 'to burn (intrans.) / to burn (trans.)'. Derived from the verb uršː-a- 'to boil (intrans., trans.)' [Magometov 1970: 18, 23]; it should be noted, however, that in [Kibrik & Kodzasov 1988: 137] and once in [Magometov 1970: 57] the verb 'to boil' is transcribed as urš-a- with lax, not tense š. Etymologically corresponds to Keren (Richa) ruxʸ-a- 'to boil', Proper Aghul (Tsirkhe) urxː-e- 'to boil' (with the tense xː!), etc. [Kibrik & Kodzasov 1988: 137; Magometov 1970: 23, 57; NCED: 1030].
Distinct from k=eyx-i- 'to set fire to' [Kibrik & Kodzasov 1988: 96].
Suleymanov 2003: 159; Shaumyan 1941: 145. It should be noted that in [Suleymanov 2003: 160], the archaic Tpig form ugʷ-a- is also quoted as a variant of more common ug-a-.
Common Aghul notes:
Internal reconstruction as well as external comparison suggest that Koshan uršː-ana- 'to burn' is an innovation.
The labialization of gʷ in ugʷ-a- is almost eliminated in dialects due to the recent areal process of dissimilative delabialization uCʷ > uC [Magometov 1970: 26].
The same in the Khiv subdialect: ug-ˈ {ургуб} 'to burn (intrans.); to burn (trans.)' [Genko 2005: 154]. The same in Literary Tabasaran: ug-ˈ {убгуб} 'to burn (intrans.); to burn (trans.)' [Khanmagomedov & Shalbuzov 2001: 290].
The same in Literary Lezgi: kːu- [imperf.] / kːa- [perf.] / kːu-g [imv.]{кун, ккун} 'to burn (trans., intrans.)' [Talibov & Gadzhiev 1966: 166; Gadzhiev 1950: 199; Gaydarov et al. 2009: 181; Haspelmath 1993: 495, 516; Gyulmagomedov 2004, 1: 368].
Proto-Lezgian:*ʔokːʷɨ-1
NCED: 860. Distribution: Retained with the basic meaning in all the languages, except for some Nuclear Lezgian lects.
In West Lezgian (Tsakhur, Rutul), this root was superseded with the synchronic causative from *ʔeɬ(ː)ʷVr-, whose original meaning is 'to get heated' [NCED: 1036]. Subsequently in Luchek Rutul, it was superseded with the synchronic causative from *ʔikʼʷa-, whose original meaning is 'to catch fire' [NCED: 632]. Koshan Aghul, 'to burn (intrans./trans.)' is a suffixal derivation from the verb *ʔVrɬːan- 'to boil (intrans./trans.)' [NCED: 1030].
Gukasyan 1974: 176; Fähnrich 1999: 24; Schiefner 1863: 104; Starchevskiy 1891: 504. According to [Fähnrich 1999], polysemy: 'fingernail / claw' (in [Fähnrich 1999: 24] a phonetical variant muq is also quoted, apparently, in error). Not to be confused with the word miχ, glossed in [Schiefner 1863: 104] as 'Nagel', which apparently means 'nail, peg', as follows from the Persian gloss mex 'nail, peg', quoted by Schiefner as a source of borrowing.
Common Udi *muχ; it is claimed in [Schulze 2001: 299a] that Udi muχ was borrowed from Persian mex 'nail', which cannot be true, because Persian mex is a terminus technicus with the meaning 'nail, peg', not 'fingernail' (the Persian term was adapted as Vartashen miχ 'nail, peg').
Chumakina et al. 2007; Kibrik & Kodzasov 1990: 33; Kibrik et al. 1977b: 270, 370; Comrie & Khalilov 2010: 209; Mikailov 1967: 190. Paradigm: ƛʼontʼˈol [abs.] / ƛʼantʼl-ˈa [erg.]. In [Dirr 1908: 201, 215] a dissimilated form ƛʼontʼor is quoted. Polysemy: 'fingernail / toenail'.
As proposed in [NCED: 210, 1002], this is a compound: ƛʼon-tʼˈol, the second part of which (-tʼˈol) can be identified with Lezgian *tʼɨl (~ -o-) 'finger'. In turn, the first element ƛʼon- is analyzed in [NCED: 210] as ƛʼo-n- from the Archi verb ƛʼʷˈa- 'to slaughter', i.e. 'nail' as '[the part of the] finger which is being cut'. The latter solution is unlikely on two points. First, both the Archi verb ƛʼʷˈa- and its Proto-Lezgian ancestor *ʔirƛʼːʷär- mean 'to slaughter (an animal)', rather than the generic 'to cut, cut off'. Next, the semantic derivation 'to cut off' > 'fingernail' seems typologically problematic.
Alternatively analyzed in [Kibrik et al. 1977a 1: 301] as ƛʼontʼ-ˈol with a root ƛʼontʼ and the relatively common nominal suffix -ol.
Kibrik & Kodzasov 1990: 33; Comrie & Khalilov 2010: 209. Polysemy: 'fingernail / claw / hoof'. The final -ek is a diminutive suffix.
Alyk Kryts:dirnaχ-1
G. Authier, pers. com. Borrowed from Azerbaijani dɨrnag 'fingernail, claw, hoof'. In [Authier 2009: 25] Alyk 'fingernail' is quoted as meχ as opposed to the Persian loanword miχ 'nail, peg', but this is a confusion (in the discovered textual example the real meaning of meχ is 'nail, peg' [Authier 2009: 231], not 'fingernail').
Budukh:dɨrnaʁ {дырнагъ}-1
Kibrik & Kodzasov 1990: 33. In [Comrie & Khalilov 2010: 209] erroneously quoted as dɨrmaʁ {дырмагъ}. Polysemy: 'fingernail / hoof'. Borrowed from Azerbaijani dɨrnag 'fingernail, claw, hoof'. There are no terms for 'fingernail' or 'claw' in [Meylanova 1984].
Kibrik & Kodzasov 1990: 33; Dirr 1907: 127, 178; Shaumyan 1941: 193. Ergative form: kerk-u. As proposed in [NCED: 690], labialized kʷ in the absolutive form is secondary, due to the influence of the ergative kerk-u.
The same in the Khanag subdialect: šːaw with polysemy: 'nail / stone plate' [Uslar 1979: 984, 999; Dirr 1905: 220, 235]. The same in the Khyuryuk subdialect: šːaw {шшав} 'nail' [Genko 2005: 192].
The same in other subdialects: Khiv šib {шиб}, Turag šab {шаб} with polysemy: 'nail / stone plate' [Genko 2005: 188, 190]. The same in Literary Tabasaran: šib {шиб} with polysemy: 'nail / stone plate' [Khanmagomedov & Shalbuzov 2001: 349].
Common Tabasaran notes:
Note the retention of tense fricative šː in the Northern subdialects. The Kondik oblique stem šibu- (not **ši̥bu-) also points to an old tense consonant, although such paradigms are expected to get levelled after the monosyllabic absolutive form.
The same in Literary Lezgi: kek [abs.] / kʸkː-e- [obl.] {кек} with polysemy: 'nail / hoof / furniture leg' [Talibov & Gadzhiev 1966: 154; Gadzhiev 1950: 426; Haspelmath 1993: 494, 523].
The same in the Akhty dialect: Khlyut kek [abs.] / kʸk-e- [obl.] with polysemy: 'nail / hoof' [Kibrik & Kodzasov 1990: 33].
Proto-Lezgian:*mːäɬː1
NCED: 814 (as *ɬːämː). Distribution: This root is retained in the generic meaning '(finger)nail' in Udi, on the one hand, and in some Nuclear Lezgian lects, on the other: West Lezgian (Tsakhur, Rutul) and Tabasaran. External North Caucasian comparison confirms such a semantic reconstruction for *mːäɬː [NCED: 814].
Some local substitutions are observed in individual languages. In Archi the obscure form ʼontʼˈol is attested. In Kryts, the old root is superseded with *mičʼ 'hoof' [NCED: 819]. In Aghul and Lezgi, 'nail' is denoted by *kerk, whose original meaning is not clear and general antiquity is dubious [NCED: 689] (as proved by the Tabasaran data, *kerk is not the Proto-East Lezgian root for 'nail', but a late areal innovation).
Superseded with an Azerbaijani loanword in some Nuclear Lezgian lects.
Replacements: {'hoof' > 'nail'} (Kryts).
Reconstruction shape: Correspondences seem regular up to the metathesis *mːäɬː > *ɬːämː in Proto-Nuclear Lezgian.
Semantics and structure: Primary substantive root. The oblique stem is *mːiɬːV-.
Common Udi *aso(y), with a laryngeal prothesis in Vartashen (note, however, that normally in such cases the laryngeal prothesis is characteristic of the Nidzh dialect, not the Vartashen one). The suffixal -y of the Nidzh form is explained in [Schulze 2001: 282] as a result of analogy with the Nidzh word-formative suffix -oy, but actually the -y suffix is observed in some Nuclear Lezgic forms (namely Rutul), thus we rather deal with the archaic formation, not synchronic derivation (in such a case Vartashen haso can be interpreted as the occasiaonl loss of the final glide).
Caucasian Albanian: alʸeg 'cloud; mist, fog' [Gippert et al. 2008: IV-5]. Etymologically unrelated to the Udi term.
Kibrik & Kodzasov 1990: 205. Polysemy: 'sheep-flock / cloud'. In [Kibrik & Kodzasov 1990], two terms for 'cloud' are quoted: kʼäbäl and bɨlɨt; difference is unclear. It must be noted that kʼäbäl is a metaphoric expression with the main meaning 'sheep-flock' [Kibrik & Kodzasov 1990: 63], whereas bɨlɨt is borrowed from Azerbaijani bulut 'cloud'.
Distinct from ǯif 'fog' [Kibrik & Kodzasov 1990: 205].
In [Comrie & Khalilov 2010: 60] three terms - kʼabal (sic!), bɨlɨt and ǯif - are quoted as synonyms for 'cloud', whereas 'fog' is translated as zov, which in fact means 'sky' [Kibrik & Kodzasov 1990: 197].
Authier 2009: 49. For the Alyk dialect two words are translated as 'cloud' by Authier, but without specifications: metaphoric kʼapʼal with polysemy: '(small) flock / cloud' [Authier 2009: 49] and ǯif with polysemy: 'cloud / fog' [Authier 2009: 39, 280, 386]. The latter is a retention.
Budukh:bulut {булут}-1
Meylanova 1984: 31, 226; Kibrik & Kodzasov 1990: 205; Comrie & Khalilov 2010: 60. According to [Meylanova 1984], with polysemy: 'cloud / fog'. Borrowed from Azerbaijani bulut 'cloud'.
Distinct from inherited ǯuf {джуф}, glossed as 'black cloud; fog' in [Meylanova 1984: 59] and as 'fog' in [Kibrik & Kodzasov 1990: 205].
Ibragimov 1990: 38; Ibragimov & Nurmamedov 2010: 125. Missing from [Kibrik et al. 1999]. Polysemy: 'cloud / rheumatism'.
Another attested word is bulut {булут} 'cloud', borrowed from Azerbaijani bulut 'cloud' [Ibragimov & Nurmamedov 2010: 94]. In [Comrie & Khalilov 2010: 60], erroneously quoted as buput {бупут}.
Distinct from čamra 'fog' [Kibrik et al. 1999: 871] (in [Ibragimov & Nurmamedov 2010: 393], quoted as čaˤmra).
Despite the wide distribution of Rutul *kːɨbɨl, it is not entirely certain that this was the basic or at least the only term for 'cloud'. Another appropriate candidate is reflected as Khnyukh (subdialect of Mukhad) asɨy 'cloud' [Ibragimov 1978: 136], Ixrek äsʷäy 'cloud' (see above), Shinaz asay 'cloud' [Dirr 1912: 120, 196] (final -Vy may be a frequent nominal suffix with general semantics [Ibragimov 1978: 65]). Perhaps *kːɨbɨl possessed the generic meaning 'fog / rheumatism', whereas *asʷVy meant properly 'cloud'.
Kibrik & Kodzasov 1990: 205. For the Fite dialect two words are quoted as synonyms for 'cloud' in [Kibrik & Kodzasov 1990: 205]: qːiri and amsar; the latter is also attested in [Tarlanov 1994: 240], but without dialectal provenance.
Distinct from dif 'fog' [Kibrik & Kodzasov 1990: 205].
Aghul (proper):qːiri-1
Suleymanov 2003: 117; Shaumyan 1941: 188. The same in the Tsirkhe subdialect: qːiri 'cloud' [Shaumyan 1941: 188].
Distinct from inherited Tpig dif 'fog' [Suleymanov 2003: 77].
Common Aghul notes:
The widespread Aghul term qːiri was borrowed from some neighboring languages of the Dargi group (cf. Chirag qːiri 'cloud') or, rather, both words represent a common loanword of unknown origin. This means that Fite amsar (historically ams-ar with the fossilized plural suffix) is the only candidate for the Proto-Aghul term for 'cloud'.
Differently in the Khanag subdialect: difː with polysemy: 'cloud / rain cloud / foam' [Uslar 1979: 669, 999], opposed to ams 'fog' [Uslar 1979: 584, 1008]. Both Khanag words difː and ams are glossed as 'fog' in [Dirr 1905: 156, 164, 245].
Two words for 'cloud' are also quoted for the Khyuryuk subdialect: dif {диф} with polysemy: 'cloud / rain cloud / foam' [Genko 2005: 61] and ams {амс} 'cloud' [Genko 2005: 18] (semantic nuances are unknown).
Two words for 'cloud' are quoted for the Khiv subdialect: ǯif {жжиф} with polysemy: 'cloud / rain cloud' [Genko 2005: 68] and ams {амс} 'cloud' [Genko 2005: 18] (semantic nuances are unknown; the latter form is not explicitly marked as Khiv due to erroneous omission of the plus sign).
The corresponding Literary Tabasaran terms are better documented: dif {диф} with polysemy: 'cloud / rain cloud' [Khanmagomedov & Shalbuzov 2001: 149]; distinct from literary ams 'rain cloud' [Khanmagomedov & Shalbuzov 2001: 57]. According to [Khanmagomedov & Shalbuzov 2001: 149, 305a], in Literary Tabasaran 'fog' is expressed as ǯilˈin dif, literally 'earth's cloud'.
Common Tabasaran notes:
The distribution formally suggests that difː ~ ǯif was the Proto-Tabasaran term for 'cloud', whereas ams meant 'fog'.
The same in Literary Lezgi: cːif [abs.] / cʸf-ˈedi- [obl.] {циф} 'cloud' [Talibov & Gadzhiev 1966: 358; Gadzhiev 1950: 434; Haspelmath 1993: 483, 517]. According to [Talibov & Gadzhiev 1966: 358], in Literary Lezgi 'fog' is expressed as čːilin cːif, literally 'earth's cloud'. Distinct from kpːul ‘rheumatism’ [Talibov & Gadzhiev 1966: 163].
The same in the Akhty dialect: Khlyut cːif with polysemy: 'cloud / fog' [Kibrik & Kodzasov 1990: 205] (regular paradigm).
Proto-Lezgian:*ʔamsː1
NCED: 243. Distribution: A rather complicated criss-crossed situation with several roots entering into competition.
First, we must rule out the root *kʼapʼal, which shifted from its original meaning 'group, heap, sheep-flock' > 'cloud' in Kryts [NCED: 448].
Second, we may exclude the root *kːomːol, which means 'cloud, fog, rheumatism' in West Lezgian (Tsakhur, Rutul) and simply 'rheumatism' in Lezgi. This root has a fairly modest distribution in the meaning 'cloud' and can hardly reflect the Proto-Lezgian basic term. Its Proto-Lezgian or at least Proto-Nuclear Lezgian meaning can be either 'rheumatism' (if the shift 'rheumatism' > 'cloud, fog' is possible) or rather 'a k. of precipitation or wet weather' > 'cloud'. In this case, the shift 'a k. of precipitation or wet weather' > 'rheumatism' is a late Tsakhur-Rutul-Lezgi isogloss of areal origin, or the polysemy 'a k. of precipitation or wet weather / rheumatism' existed already on the Proto-Nuclear Lezgian level. Additionally, the retention of the root *ʔamsː with the meaning 'cloud' in some Rutul dialects may indicate that the semantic development *kːomːol 'a k. of precipitation or wet weather' > 'cloud' is a late Tsakhur-Rutul process after the split of Proto-West Lezgian. External North Caucasian comparison also proves that the original meaning of Lezgian *kːomːol was 'a k. of precipitation or wet weather' [NCED: 737].
The real choice consists of two Proto-Lezgian roots: *ʔamsː and *tːiɬːʷ, which are to be distributed among two specific meanings: 'cloud' and 'fog'.
On formal grounds, the stem *ʔamsː [NCED: 243] has a better chance to represent the Proto-Lezgian term for 'cloud'. It is retained as 'cloud' in Udi, on the one hand, and in Rutul and Aghul, on the other. The root *tːiɬːʷ [NCED: 400], whose Proto-Lezgian meaning in this case must have been 'fog', is retained as 'fog' in South Lezgian (Kryts, Budukh) and in Aghul. In Lezgi, *tːiɬːʷ expanded to 'cloud', having acquired polysemy 'fog / cloud' (whereas the old root *ʔamsː got lost). There are two difficulties with such a scenario. First, *tːiɬːʷ developed into 'cloud' ('fog' > cloud') in Archi. Next, both roots swapped their meanings in Proto-Tabasaran, where *ʔamsː probably meant 'fog' and *tːiɬːʷ meant 'cloud'; it should be noted, however, that the Proto-Tabasaran semantic reconstruction is not very certain.
The second scenario is that *ʔamsː meant 'fog' in Proto-Lezgian, whereas *tːiɬːʷ meant 'cloud'. This implies that *tːiɬːʷ has been retained as 'cloud' in Archi and probably Proto-Tabasaran, but independently underwent the shift 'cloud' > 'fog' in South Lezgian (Kryts, Budukh) and Aghul. In its turn, *ʔamsː 'fog' only retained its meaning in Proto-Tabasaran, but independently shifted from 'fog' > 'cloud' in Udi, Rutul and Aghul. It should be noted that both roots swapped their meanings in Aghul.
The first scenario is much more economical; thus, we follow the formal distribution and reconstruct Proto-Lezgian *ʔamsː 'cloud' and Proto-Lezgian *tːiɬːʷ 'fog'. The problem is that external North Caucasian comparison suggests that it should be *tːiɬːʷ that denoted 'cloud' in Proto-Lezgian. In any case, we must note that, due to natural reasons, both meanings, 'cloud' and 'fog', are frequently interchangeable in mountainous regions.
An additional term for 'cloud' is the etymologically obscure Caucasian Albanian form alʸeg. It is proposed in [Gippert et al. 2008: IV-5] that alʸeg originates from Lezgian *tːiɬːʷ, but the assumed sound shifts (not discussed by Gippert and Schulze) seem strange and irregular.
'Cloud' is expressed by Azerbaijani or Dargi loanwords in Budukh and Aghul.
Replacements: {'fog' > 'cloud'}, {'cloud' > 'fog'}, {'group, heap, sheep-flock' > 'cloud'} (Kryts), {'a k. of precipitation or wet weather; rheumatism' > 'cloud'} (Tsakhur, Rutul).
Reconstruction shape: Correspondences seem regular, except for the laryngeal prothesis in Vartashen Udi.
Semantics and structure: Primary substantive root. The oblique stem is not reconstructible. Udi and Rutul reflect the suffixed variant *ʔamsːV-y.
Fähnrich 1999: 12; Bezhanov & Bezhanov 1902 (Mt. 10.42 "a cup of cold (čaχ) water to drink"). Polysemy: 'cold (adj.) / cold (n.), frost / ice'.
Common Udi notes:
Common Udi *čaχ 'cold (n.); cold (adj.); ice'. In [Gukasyan 1974: 236; Mobili 2010: 80; Schiefner 1863: 87] and [Schulze 2001: 266], however, Nidzh-Vartashen čaχ is inaccurately glossed only as a substantive 'cold, frost; ice' (thus polysemy 'cold, frost / ice / cold (adj.)'); there is no equivalent for English 'cold (adj.)' in these dictionaries. Cf. also the common Udi substantive mi 'cold, frost' [Gukasyan 1974: 174; Schiefner 1863: 104; Fähnrich 1999: 23; Schulze 2001: 299].
Caucasian Albanian: mii [Gippert et al. 2008: IV-30] ("a cup of cold (mii) water"). Apparently [miʔi].
Chumakina et al. 2007; Kibrik & Kodzasov 1990: 245; Kibrik et al. 1977b: 335, 387; Mikailov 1967: 200; Dirr 1908: 189, 225. In [Comrie & Khalilov 2010: 686] a corrupted form χˤe-du-class is quoted. Regular participle from the stative verb χˤe 'to be cold' [Chumakina et al. 2007; Kibrik et al. 1977b: 335]; widely applicable.
Distinct from ˈoˤrču-tːu-class, which is quoted in [Dirr 1908: 172, 225] with the translation 'cold (adj.)' ("the cold water"). A more correct translation should be rather 'having gotten cold', a regular participle from the verb ˈoˤrča- 'to get cold' [Chumakina et al. 2007], Kibrik et al. 1977b: 291].
Kibrik & Kodzasov 1990: 245; Comrie & Khalilov 2010: 686. Participle from the verb s=aʁa- 'to get cold' [Kibrik & Kodzasov 1988: 97]. Initial s= is the preverb s-/c- with general semantics [Saadiev 1994: 424].
Meylanova 1984: 122, 249; Kibrik & Kodzasov 1990: 245; Comrie & Khalilov 2010: 686. Polysemy: 'cold / frozen, congealed'. Participle from the verb s=aʁa- 'to become cold; to feel cold' [Kibrik & Kodzasov 1988: 97]. Initial s= is a prefix with general semantics [Alekseev 1994: 271 f.].
Distinct from ʕatχa-lu {гIатхалу} 'cold (of weather)', a participle from ʕatχa 'to become cold (of weather), to catch cold (of human)' [Meylanova 1984: 46].
Dirr 1912: 161; Makhmudova 2001: 93, 181, 184; Comrie & Khalilov 2010: 686. Applied both to objects and weather. The assimilated variant with -u- comes from [Dirr 1912].
Kibrik & Kodzasov 1990: 245; Shaumyan 1941: 160. The same in the Khudig subdialect: mikʼ-le-d 'cold' [Shaumyan 1941: 160].
Derived from the substantive, attested as Burshag mekʼ (oblique mikʼ-la-) 'cold, frost' [Shaumyan 1941: 151].
It should be noted, however, that in [Suleymanov 2003: 144], 'cold' is quoted as urʁa-d {ургъад} and urʢa-d ~ urʕa-d {ургIад} (apparently urʢa-d), which represent forms of the Arsug or Khudig subdialect, as is evident from the suffixal -d.
Kibrik & Kodzasov 1990: 245. The same in the Usug subdialect: ruʁu-f 'cold' [Shaumyan 1941: 160]. Cf. also the Usug substantive mekʼ (oblique mekʼ-ala-) 'cold, frost' [Shaumyan 1941: 151].
Suleymanov 2003: 144; Shaumyan 1941: 160. The same root in the Tsirkhe subdialect: urʁu- 'cold' [Magometov 1970: 215 sentence 11 "cold water"].
In the Tpig subdialect, cf. the substantive mekʼ (oblique mekʼ-ila-) 'cold, frost' [Suleymanov 2003: 133; Shaumyan 1941: 151].
Common Aghul notes:
The distribution suggests that the Proto-Aghul adjective 'cold' was derived from the substantive 'cold, frost', attested as Gequn ruʁu 'cold, frost' (see above) and ruʁ 'cold, frost', quoted in [Magometov 1970: 46] without dialect specification (probably proper Aghul). Cf. also the paronymous verb, which is attested in non-Koshan dialects as Keren (Richa) ruʁ-a- 'to feel cold' [Kibrik & Kodzasov 1988: 166], proper Aghul (Tpig) ruʁ-a- 'to become cold' [Suleymanov 2003: 144]. For the Koshan (Burshag) dialect two synonymous verbs with the meaning 'to feel cold; to become cold' are quoted in [Kibrik & Kodzasov 1988: 165] and [Suleymanov 2003: 144]: ʡurʢ-a- and urʕ-a-. The former Burshag stem is perhaps to be analyzed as ʡ=urʢ-a-, although the prefix ʡ= seems very rare or unique (cf. Proper Aghul (Kurag) ʡ=aHa-r xa- 'to know' q.v.); phonetically =urʢ-a- normally corresponds to non-Koshan ruʁ-a-, see [NCED: 133]. The latter Koshan stem urʕ-a- is unclear; formally, it represents the result of secondary pharyngealization.
Only in the Burshag subdialect of Koshan (but not in other Koshan subdialects) this term was superseded with a new adjective, derived from another substantive with the meaning 'cold, frost' - mekʼ (oblique mekʼ-ila-), see the data above.
Final -d, -t, -f, -r are the adjectival suffixes (fossilized class exponents) [Magometov 1970: 92; Shaumyan 1941: 45].
Kibrik & Kodzasov 1990: 245. Distinct from Dyubek merčʼu-lˈi 'cool, chilly' [Kibrik & Kodzasov 1990: 245] (historically me-r-čʼu- with the fossilized class exponent -r-).
In the Khanag subdialect: mičʼi-lˈi 'cold' [Uslar 1979: 858, 1009; Dirr 1905: 196, 246], applied to both objects (e.g., water) and weather; distinct from aqˤˈi 'cold (said of weather)' [Dirr 1905: 152, 246].
Two words are quoted for the Khyuryuk subdialect: aqˤˈi {аьхъи} 'cold' [Genko 2005: 25], mičʼi-lˈi {мичIили} 'cold' [Genko 2005: 121] - application and semantic nuances of both terms are unknown.
Two words with the meaning 'cold' are attested in the Khiv subdialect: ʁ=ˈarʁu {гъаргъу} [Genko 2005: 41] (historically ʁ=ˈa-r-ʁu with the fossilized class exponent -r-) and mičʼ-li {мичIли} 'cold' [Genko 2005: 121] - both terms are applied to objects, but the semantic or pragmatic difference is unclear ('cold' and 'cool'?). Distinct from Khiv aqˤˈu {аьхъю} 'cold' [Genko 2005: 25], which is applied to weather, as may be seen from Genko's examples (cf. also an additional example in [Genko 2005: 147 sub tʼaqʼraqʼ]).
Three Literary Tabasaran terms are known: ʁ=ˈaʁu {гъабгъу} 'cold' [Khanmagomedov & Shalbuzov 2001: 329 sub qˤal] (the example: "cold water"); mičʼ-lˈi {мичIли} 'chilly' [Khanmagomedov & Shalbuzov 2001: 231]; aqˤˈu {аьхъю} 'cold (said of weather)' [Khanmagomedov & Shalbuzov 2001: 72].
Common Tabasaran notes:
A tangled situation with three competing roots: (1) *ʡiqˤä- (Northern aqˤˈi ~ aqˤ-lˈi, Southern aqˤˈu); (2) *ʔirqːe(r)- (Southern ʁ=ˈaʁu); (3) *meʼä- (Northern merčʼu-lˈi ~ mičʼi-lˈi, Southern mičʼ-li).
It seems that Northern me-r-čʼu-lˈi ~ mičʼi-lˈi, Southern mičʼ-li can be reconstructed with the specific meaning 'cool, chilly', as retained in both Northern and Southern dialects, although in the Khanag subdialect this acquired the generic meaning 'cold'. It is confirmed by comparative Aghul data that this adjective was derived from the oblique stem of the substantive 'cold, frost' (this substantive is retained as Tabasaran mikʼ ‘wind’ q.v.).
As for Northern aqˤˈi ~ aqˤ-lˈi, Southern aqˤˈu, its normal meaning is 'cold (of weather)', although in the Dyubek subdialect aqˤ-lˈi acquired the meaning 'cold (in general)' (apparently the Dyubek final -li is due to influence on the part of merčʼu-lˈi 'cool, chilly'). This term was derived from the verbal root that is retained in the prefixed form as Dyubek da=ʔaqˤ- 'to become cold' [Kibrik & Kodzasov 1988: 97], Khanag d=aqˤ- 'to become cool, chilly' [Dirr 1905: 163].
Southern ʁ=ˈa-class-ʁu 'cold' (not attested in Northern Tabasaran) is the synchronic perfect participle from the Common Tabasaran verb 'to become cold'. Cf. Northern: Khyuryuk aqː-ˈ {абкъув} 'to feel cold' [Genko 2005: 12], Kumi prefixed qː=aqː-ˈ {къаркъув} 'to feel cold' or 'to get cold' [Genko 2005: 100]; Southern: Khiv aʁ-ˈ {аргъуб} 'to feel cold; to get cold' [Genko 2005: 19], Khiv uʁ-ˈ {ургъуб} 'to freeze (trans., intrans.)' [Genko 2005: 155], Literary Tabasaran aʁ-ˈ {абгъуб} 'to get cold; to feel cold' [Khanmagomedov & Shalbuzov 2001: 44].
It is possible that Southern ʁ=ˈaʁu reflects the Proto-Tabasaran term for 'cold (in general)'.
Uslar 1896: 485, 638. Glossed as 'cold, cool'; applied to both objects and weather. Perfective participle from the verb räqːi- [imperf.] / qːa- [perf.] 'to get cold' [Uslar 1896: 495]; cf. also the parallel participle qːa-y 'cold wind, cool wind' [Uslar 1896: 485] (inaccurately glossed as 'cool, coolness' by Uslar; see note on 'wind'). Distinct from Gyune mäqʼˈi 'cold (of weather)' [Uslar 1896: 504].
The same in Literary Lezgi: qːa-yˈi {къайи} 'cold, cool (applied to objects)' [Talibov & Gadzhiev 1966: 176; Gadzhiev 1950: 916; Haspelmath 1993: 501, 517]; perfective participle from the verb reqːˈi- [imperf.] / qːa- [perf.] {къун} 'to get cold' [Talibov & Gadzhiev 1966: 193]. Distinct from literary meqʼˈi {мекьи} 'cold (of weather)' [Talibov & Gadzhiev 1966: 234], derived from the substantive meqʼ {мекь} 'cold (n.)' [Talibov & Gadzhiev 1966: 234].
Differently in the Akhty dialect: Khlyut meqʼˈi 'cold' [Kibrik & Kodzasov 1990: 245].
It is uncertain how the Proto-Lezgi word for 'cold' should be reconstructed. According to the external data, it is likely that Akhty meqʼˈi is an archaic form, whereas the Gyune participle qːa-yˈi represent a late introduction of areal origin, which superseded meqʼˈi in the meaning ‘cold (of objects)’.
Proto-Lezgian:*meƛʼä-4
NCED: 808. Distribution: It seems that in all (or almost all) attested cases the adjective 'cold' can be interpreted as a synchronic derivate from either the substantive 'cold, frost' or the verb 'to be cold' (participle pattern).
A rigoristic approach would surmise leaving the Proto-Lezgian slot empty. Nevertheless, we prefer to reconstruct Proto-Lezgian *meƛʼä- 'cold (adj.), derived from Proto-Lezgian *meƛʼ 'cold, frost' [NCED: 808]. This seems to be the most archaic Lezgian expression for 'cold (adj.)', and, indeed, the derivation 'cold, frost' > 'cold (adj.)' can theoretically be reconstructed for the Proto-Lezgian level.
The substantive *meƛʼ 'cold, frost' is a Common Lezgian stem, since it is attested in Udi and in Nuclear Lezgian: Aghul, Lezgi (also in Tabasaran, where it shifted to the meaning 'wind (in general)'). The adjective stem *meƛʼä- 'cold' also displays relevant distribution. It is attested in Caucasian Albanian, on the one hand, and in West Lezgian (Tsakhur, Rutul) and Proto-Lezgi, on the other. In some East Lezgian lects - Koshan Aghul (Burshag), Southern Tabasaran (Khiv) - the adjectives for 'cold' contain the same root, but these forms represent synchronic derivatives from the substantive *meƛʼ 'cold, frost'.
In Udi, 'cold (adj.)' is expressed with *čawχː- [NCED: 346], whose original meaning was substantival, cf. the synchronic polysemy in Udi: 'cold (adj.) / cold (n.), frost / ice'.
The verbal root *ʡiqˤä- 'to get cold' [NCED: 568] forms synchronic participles with the meaning 'cold (adj.)' in Archi and Northern Tabasaran (Dyubek).
The verbal root *ʔirqːe(r)- 'to get cold' [NCED: 649] forms synchronic participles with the meaning 'cold (adj.)' in some Nuclear Lezgian lects: South Lezgian (Kryts, Budukh), Proto-Aghul, Proto-Tabasaran, Gyune Lezgi.
The derivation 'to get cold' > 'cold' is likely to be a relatively late areal isogloss.
Gukasyan 1974: 122; Fähnrich 1999: 15; Dirr 1903: 10, 40, 51, 56, 69, 85, 89, 94, 95; Schiefner 1863: 77; Starchevskiy 1891: 488. Glossed as 'to come, to arrive'; but incorrectly as 'to go, walk' in [Schulze 2001: 275], cf. contexts like Mt. 8.9, where e(y)- 'to come' is opposed to ta(y)- 'to go': "I tell this one, 'Go (take),' and he goes (tanesa); and to another, 'Come (eke),' and he comes (enesa)" [Bezhanov & Bezhanov 1902].
In [Fähnrich 1999: 30] the variant eʁ-esun is also quoted - an important archaism, see notes on 'to go'.
Common Udi notes:
As described in [Maisak 2008a: 107 ff., 154 ff.], a suppletive paradigm: e(y)- (present-infinitive) / ar- (past) / eʁ- (future) / ek- (imperative). For the paradigmatic distribution of the Nidzh variants e- and ey- see [Maisak 2008a: 107].
Originally *e=iʁ (> e=y) / ar- / *e=iʁ- / e=k- with the preverb *(h)e- 'hither', see notes on 'to go'.
Caucasian Albanian: A suppletive verb heʁ- (present-infinitive) / ar- (past) / hekal- (imperative) [Gippert et al. 2008: II-44, 45, 51, IV-26]. With the exception of the imperative root, directly corresponds to the Udi paradigm. For further analysis see notes on 'to go'.
Chumakina et al. 2007; Kibrik & Kodzasov 1988: 75; Kibrik et al. 1977b: 188, 376; Comrie & Khalilov 2010: 495; Mikailov 1967: 171; Dirr 1908: 131, 219. Glossed as 'to come, to arrive'.
As described in [Kibrik et al. 1977a 2: 72; Kibrik et al. 1977a 3: 242; Chumakina et al. 2007], a suppletive verb: =ˈaƛi- [inf., imperf.] / =qˤˈa [perf.] / zˈa-ba [imv.] (-ba in the imperative stem is the light verb 'to say') / =hˈeˤ- [potential]. We treat =ˈaƛi- and =qˤˈa as synonyms. The latter is etymologically the same root as =ˈoqˤe- 'to go' q.v.
Kibrik & Kodzasov 1988: 75. Suppletive paradigm: ʕ=ušχä- [imperf.] / ʕ=uxu- [perf.] / ʕ=ušä- [imv.]. We treat ʕ=uxu- [perf.] and ʕ=ušχä- [imperf.] as synonyms. Both stems contain the preverb ʕ- 'in'; the imperfective stem is apparently a compound of two Proto-Lezgian verbal roots: ʕ=uš-χä-, thus [NCED: 657, 666]. Perfective ʕ=uxu- contains the same root as 'to go' q.v.
Kibrik et al. 1999: 68, 74, 77, 884, 897; Ibragimov & Nurmamedov 2010: 375. According to [Kibrik et al. 1999], a suppletive verb: qˤ=aʔ-a [imperf.] / q=aʔ-ɨ [perf.] / q=alʸ-es [fut.] / q=or-a [imv.]. Cf. synchronic forms: imperf. class 1/2/3/4 qˤaː, perf. 1/2 qarɨ, 3 qabɨ, 4 qadɨ, fut. 1/4 qalʸes, 2 qayeːlʸes, 3 qawalʸes, imv. 1/4 qora, 2 qeːra, 3 qiwora. Polysemy: 'to come / to bring (animated obj.)' in perf. & fut., although in imperf. & imv. the roots for 'to come' and 'to bring' are different.
There is also another suppletive verb with the more generic meaning 'to come, arrive / to bring (animated & inanimate obj.)', formed with the same roots aʔ- [perf.] / al- [imperf., fut., imv.] and the zero prefix (or with the prefix ʔ- - an automatic prothesis for vocalic onset): allʸes [Kibrik et al. 1999: 63, 869].
Ibragimov 1990: 183, 195, 213. Only the form of fut. is known: q-i-class-ʔ-ez ~ q-i-class-y-ez. The paronymous verb aʔ- (formed with the zero prefix) is a close synonym: fut. aryez, abyez, prohib. maraye, mabaye 'to come' [Ibragimov 1990: 196]. In [Comrie & Khalilov 2010: 495], only the second verb is quoted: fut. a-class-ʔ-ez (a-r-ʔ-ez {аръез}).
Common Tsakhur notes:
Initial q= is a prefix with general semantics [Ibragimov 1990: 124; Kibrik & Kodzasov 1988: 41].
According to the data in [Makhmudova 2001] and [Ibragimov 1978], the suppletive paradigm is as follows: class=iʔi ~ class=r=uʔ-u-r- [imperf.] / y=iqʼ-ɨ-r [perf.] / class=ɨqʼ-a [imv.]. In the imperfective form, with polysemy: 'to go / to come'.
According to the data in [Dzhamalov & Semedov 2006], the suppletive paradigm is as follows: class=iʔi [imperf.] / y=iqʼ-ɨ-r [perf.] / class=iqʼ-ä [imv.]. In the imperfective form, with polysemy: 'to go / to come'.
Suppletive paradigm: class=iʔi ~ class=r=uʔ-u-r- [imperf.] / y=iqʼ-ɨ-r [perf.] / class=ɨqʼ-ɨ [imv.]. In the imperfective form, with polysemy: 'to go / to come'.
Common Rutul notes:
An irregular verb (with two synonymous stems for the imperfective), but the paradigms generally coincide in all three dialects. Two imperfective stems with polysemy: 'to go / to come' are genetically related. See notes on 'to go'.
Initial y= is a prefix with general semantics [Ibragimov 1978: 95; Makhmudova 2001: 165].
Shaumyan 1941: 139. The following paradigmatic forms can be extracted from [Shaumyan 1941: 139] and [Suleymanov 2003: 18]: we-y- [imperf.] / ad-ina- [perf.] / ad-i- [inf.] / šab [imv.].
Common Aghul notes:
A very irregular verb, although the suppletive paradigms generally coincide in all dialects. In the imperfective and prohibitive forms, with polysemy: 'to go / to come'; further see notes on 'to go'. All dialectal perfective forms (arg-i-, ad-i-, ar-i-) are etymologically related, originating from one proto-root [NCED: 422].
The same in the Khiv subdialect: ʁˤ- [imperf., inf.] / af- [perf.] / ʁač [imv.] {гъюб} 'to come' [Genko 2005: 45]. The same in Literary Tabasaran: ʁˤ- [imperf., inf.] / af- [perf.] {гъюб} 'to come' [Khanmagomedov & Shalbuzov 2001: 117].
Common Tabasaran notes:
An irregular verb with three roots, although the suppletive paradigms coincide in both dialects. The imperfective stem Qˤ- contains the same root as 'to go' q.v.
The authors of [NCED], confused by the ambiguous Cyrillic orthography, transcribe the perfective stem as ata- with a subsequent incorrect connection to the Udi verb 'to go' q.v. [NCED: 423].
Proto-Lezgian:*ʔiqːʷˤä-1
NCED: 572. Semantics and structure: Primary verbal root, attested with several ablaut grades. Reconstructed as the imperfective stem with polysemy 'to go / to come'. Further see notes on 'to go'.
NCED: 1016. Reconstruction shape: The exact phonetic shape of a root with such a structure is not reconstructible. According to the table of correspondences in [NCED: 150], one could expect *ʔʷ > Udi p instead of observed 0, but this can hardly be an obstacle to the whole etymology.
Semantics and structure: Primary verbal root, used as the perfective stem for 'to come'. Further see notes on 'to go'.
Build tree