The development of the Classical Armenian aorist in modern dialects Текст научной статьи по специальности «Языкознание и литературоведение»

DOI 10.30842/alp2306573714107

H. Martirosyan

Leiden University

THE DEVELOPMENT OF THE CLASSICAL ARMENIAN AORIST IN MODERN DIALECTS

This paper aims to present a sketch on the development of the Classical Armenian aorist in the dialects with particular attention on Hamsen, Aramo, and other peripheral dialects, as well as some related issues, such as the aorist augment e- and the vowel -a- of the first person plural endings in the aorist and imperfect paradigms1.

1. ClassicalArmenianaorist

The active endings are: -i, -er, zero, -ak', -e/ik', -in; 3 sg. augment e- when the stem is monosyllabic.

Table 1. The aorist paradigms of Classical Armenian

lam berem sirem t'otum

'to weep, cry' 'to bring' 'to love' 'to let, leave'

1 sg. lac 'i beri sirec 'i t 'oii

2 sg. lac 'er berer sirec 'er t 'oier

3 sg. e-lac' e-ber sireac' e-t 'oi

lpl. lac 'ak' berak' sirec 'ak' t 'oiak'

2 pi. lac'e/ik' berek'/ik' sirec 'e/ik' t'oie/ik'

3 pi. lac 'in berin sirec 'in t 'oiin

2. Spread of e-

In a number of inscriptions one observes a spread of e-\ 3 pi. e-t'oiin (Vayoc' Jor, the village of T'erp', 1264 CE) instead of t'oiin (CArm. t'olum 'to let, leave', cf. 3 sg. e-t'of); 3 pi. e-barj-in in Ani instead of barjin (CArm. barnam < *barj-nam 'to lift, raise', cf. 3 sg. e-barj), etc. This is observable even in c '-forms the corresponding Classical paradigms of which never had an e-. Here are examples from

11 am greatly indebted to Kate Bellamy for checking my English. I am also indebted to Remy Viredaz, Nikolai Kazansky, and Petr Kocharov for making a few corrections.

the Lori and Tavus regions, the 13th century: e-gnec'i 'I bought' (Sanahin and Enok'avan); e/e-grec'in 'they wrote' (Ojun and Makaravank'). One also finds examples that combine the Middle Armenian forms with Classical ones, such as e-tui (vs. CArm. etu and MArm. tu-i), the 1 sg.aor. of tam 'to give' (Ani, Makaravank', etc.; the 11th century onwards); cf. 1 pi. e-tua(n)k' vs. CArm. and MArm. tuak'.

S. Avagyan [1986: 134-142] discusses the material and concludes that the vowel e- of these forms is due to insufficient grammar skills of the authors of these inscriptions. However, there is dialectal evidence pointing to the fact that at least in some cases we may be dealing with really existent forms. In § 6, the examples of etui and etuak' will be discussed in greater detail.

3. First person plural imperfect

The Classical Armenian 1 pi. imperfect ending is characterized by the presence of a vowel -a-; thus, for example, in the e-conjugation we have the following set of endings: -ei, -eir, -er, -eak', -eik', -ein. In Modern Armenian and most of dialects, this -a- has been analogically eliminated, but some dialects have preserved it intact. In the Tables 2-3, I present the Classical imperfect paradigm of the verb utem 'to eat' and its corresponding tense forms (imperfect or past subjunctive) in the dialects of Akn [Acaryanl911:223] (cf. ibid., p. 227 on Sebastia), Hamsen [Acaryan 1947: 136, 139], and SEA. I also include two paradigms from the region of Dersim, grem 'to write' in Xarberd/Erznka, and k'asem 'to pull, drag' in C'mskacag [Balramyan I960: 22, 30]; the latter preserves the -a- intact whereas the former is innovative. Interestingly, the forms that are characterized by the archaic -a- show innovations too; in Hamsen we see analogical insertion of -i/y-, whereas the others have the innovative nasal.

Table 2. The imperfect paradigm of the verb 'to eat' in Classical Armenian, the dialects of Akn, Hamsen, and Standard Eastern Armenian

CArm. Akn Hamsen SEA

1 sg. utei g-udei g-udeys k-utei

2 sg. uteir g-udeir g-udeyd(s) k-uteir

3 sg. uter g-uder g-uder k-uter

lpl. uteak' g-udeank' g-udayk 'a k-uteink'

2 pi. uteik' g-udeik' g-udeyk's k-uteik'

3 pi. utein g-udein g-udeyns k-utein

Table 3. The imperfect paradigm of the verbs 'to pull' in the dialect of C'mskacag and 'to write' in the dialect of Xarberd/Erznka

C'mskacag Xarberd/Erznka

1 sg. gg k'asei gs g'srei

2 sg. gs k'aseir gs g'sreir

3 sg. gs k'aser gs g'srer

lpl. gs k'aseank' gs g'areink'y

2 pi. gs k'aseik' gs g'sreik'

3 pi. gs k'asein gs g'srein

4. First person plural aorist and the e- augment

As we have seen in Section 1, the same 1 pi. -a- is seen in the aorist of Classical Armenian. Here again, some dialects preserve it intact, cf. e.g. the paradigms of xatam 'to play' and sirem 'to love' in Hamsen [Acaryan 1947: 128, 131].

Table 4. The aorist paradigm of the verbs xalam 'to play' and sirem 'to love' in Classical Armenian

xalam sirem

'to play' 'to love'

1 sg. xalac 'i sirec'i

2 sg. xalac 'ir sirec'ir

3 sg. xalac' sirec'

lpl. xalac 'ak' sirec'ak'

2 pi. xalac 'ik' sirec'ik'

3 pi. xalac 'in sirec'in

In old monosyllabic aorist forms of this dialect we find yet another archaism, the augment e-2. Here is the example of banam, aor. stem bac'- 'to open' [Acaryan 1947: 135].

Table 5. The aorist paradigm of the verbs 'to open' in Classical Armenian and the Hamsen dialect

CArm. act.

1 sg. bac'i

2 sg. bac 'er

3 sg. e-bac'

CArm. med.

bac'ay bac 'ar bac'aw

Hamsen

pac 'i pac 'ir

pac 'aw, e-pac'

2 For a discussion of this Hamsen archaism, see [Vaux 2007: 265-266]. It is also present in a neighboring dialect of Xotorjur [Kostandyan 1985: 60].

CArm. act. CArm. med. Hamsen

1 pi. bac'ak' bac'ak' pac'ak'

2 pi. bac'ek',-ik' bac'ayk' pac'ik'

3 pi. bac 'in bac 'an pac 'in

5. Hamsen asus, aor. stem ast- 'to say'

CArm. asem 'to say, tell, speak' displays aorist forms based on ast- in Hamsen. According to Acaryan [1947: 134-135], the -t- after -s- is an epenthesis of a phonetic nature, cf. almas 'diamond' > almast, t'asel 'to puff' > t'astus, etc. This explanation is not entirely satisfactory because: 1) it is not clear why the -t- is only found in the aorist; 2) we expect to see the reflexes of the aoristic c'-. I therefore propose the following solution.

In Classical Armenian, the aorist stem of this verb is asac'-. This subtype displays both non-syncopated and syncopated aorist forms in Hamsen, namely xalac'i and xalc'i (see [Acaryan 1947: 130131]). It seems therefore likely that Hamsen asti etc. derive from the syncopated forms *asc'i etc.; the development *asc'i > asti is probably due to dissimilation, which may have been triggered or reinforced by the aforementioned -(s)t- epenthesis. For the inclusion of the 3 sg. aor. med. into the main paradigm, compare the aforementioned pac'aw alongside e-pac' (cf. MArm. lsg. gtay vs. CArm. lsg. gti 'I found', see Karst 1901: 327).

In the Table 6, I have combined the Classical and Hamsen paradigms with that of Samaxi/K'yark'yanj, an easternmost dialect (on whichsee [Balramyan 1964: 166]):

Table 6. The aorist paradigm of the verbs 'to say' in Classical Armenian and the dialects ofHamsen and Samaxi/K'yark'yanj

CArm. Syncope Hamsen Samaxi

1 sg. asac 'i *asc 'i as-t-i asc 'i, assi

2 sg. asac 'er *asc 'er as-t-ir asc 'ir, assir

3 sg. asac' *as(a)c '-aw (med.) as-t-av asec', asec

lpl. asac'ak' *asc 'ak' as-t-ak' asc 'ink', assink'

2 pi. asac'ek\ -ik' *asc 'ek', -ik' as-t-ik' asc 'ik', assik'

3 pi. asac 'in *asc 'in as-t-in asc'in, assin

6. The aorist forms of tam 'to give' in Aramo and other dialects

In Classical Armenian, the irregular verb tam 'to give' is characterized by the e-augment throughout the paradigm except for 1 pi. and the absence of the vowel i/e/e in 1 sg., 2 sg., 2 pi., and 3 pi. In Middle Armenian (see Karstl901:333), dialects and modern Armenian one finds analogical developments:

1) the root *tu is recovered in the 3 sg. form: et ^ etu;

2) the augment e- is eliminated from the paradigm but in some dialects it is preserved in the 3 sg. form;

3) the regular endings are added to this new basis: etu ^ tu-i, etur ^ tu-i-r, etc.

The vowel -a- in 1 pi. forms and the augment e- are both present in the Aramo dialect, where, for example, sdvunk' reflects *e-tuank' and stands for the CArm. tuak', 1 pi. aorist of the verb tam 'to give'. In what follows I shall discuss this remarkable case in more detail since it simultaneously represents an illustrative combination of archaisms and innovations.

The Classical Armenian aorist forms etu, etur, and et, have been replaced in Aramo by sdva, sdvey, and ida respectively. Laribyan [1958: 47] (cf. ibid., pp. 19, 21) derives 3 sg. aorist ida from *etu (for the vocalic reflections, cf. ere/ikun 'evening' > irgaun, katu 'cat' > gada, heru 'last year' > hira, heru 'far' > hsrra, etc.) but gives no explanations for the other forms. The Classical 3 sg. form et was thus replaced by *etu, and all the other personal forms, as we shall see, have been reshaped by attaching the regular set of aorist endings to this base *etu.

The 1 sg. aor. form sdva may reflect *etu-i, in which the ending -i is taken from the Classical aorist ending -i. The vocalic developments are regular: 1) e- > 3-, cf. ezr 'edge' > azzir(a), erek' 'three' > ark', erkir 'earth' > srgayr; 2) word-final -i > -a, cf. agi 'tail' > akka, aygi 'garden' > akka, oski 'gold' > sska, etc. (see [Laribyan 1958: 19, 20]). The existence of such an analogical form is confirmed by some inscriptional evidence from the 11th century onwards, etvi (Ani, 1058 CE) and etui (Makaravank')3.

3 See [Orbeli et. al. (eds.) I960: 13, Nr 51;Avagyan 1986: 136-137].

The 2 sg. aor. form advey in all probability reflects an analogical *etu-er, with the CArm. 2 sg. aor. ending -er (cf. giser 'night' > kiseyr, ver 'up, above' > veyr, etc.). As to the alternation -ey vs. -£(y)r, compare the 1 sg. pres. perf. forms: affirmative adver im vs. negative c'am idvey. See below on the contrast between the archaic -er in Aramo and Svedia on the one hand and the innovative -ir (seen in Middle and Modern Armenian) in Zeyt'un on the other hand.

The 1 pi. aor. form advunk' reflects *e-tuank' according to regular sound laws (cf. aprank' > abrunk',p'orjank' >p'orc'unk'), and 2 pi. and 3 pi. forms derive from *etu-ik' and *etu-in respectively (cf. cic > jayj, hing > hayng), analogically acquiring thus the corresponding Classical endings -ik' and -in. The existence of 3 pi. aor. *etuank' is confirmed by, for example, an inscription from the Sxmurat monastery (Tavus, the village of Calkavan) in 1248 CE, which has the form etvank'; cf. three earlier attestations (12th and early 13th centuries) from the same regions with no nasal — etuak', etuak', and etvak' [Avagyan 1986: 136].

In the tables below, I have combined these paradigms with those of Middle Armenian [Karstl901:333 (= 2002:315-316); Ant'osyan 1975: 213; Hovsep'yan 1997: 68-69], Modern Western and Eastern Armenian [Ant'osyan 1975: 214], and the following dialects:

1) Western grouping: Svedia in Syria [Acaryan2003:494, 498], Zeyt'un in Cilicia [Acaryan2003: 243], Hamsen in the Northwestern intergroup [Acaryan 1947: 137], Mus in the South-Central group [Baldasaryan-T'ap'alc'yan 1958: 169] and neighboring Sasun [Petoyan 1954: 59], Polis/Constantinople [Acaryan 1941: 149], Moks [Muradyan 1982: 173]; cf. Satax, see [Muradyan 1962: 148], 1 sg. tavi), Van [Acaryan 1952: 173];

2) Eastern grouping: Marala [Acarean 1926:262], Agulis [Acarean 1935: 260], Samaxi [Balramyan 1964: 129, 162, 167], Krzen [Balramyan 1961: 151], Hadrut' [Polosyanl965:212], Ararat/Lori [Asatryan 1968: 141]. Brief outlines of individual dialects and dialectal maps can be found in [Martirosyan 2013].

Table 7. The paradigm of the verb 'to give' in Classical Armenian, the dialect of Amaro, inscriptions, and the Modern Armenian

CArm. Aramo Proto-Aramo Inscriptions (11th cent. +)

1 sg. etu sdva *etu-i etvi, etui

2 sg. etur advey *etu-e(r)

3 sg. et ida *etu

CArm. Aramo Proto-Aramo Inscriptions (11th

lpl. tuak' advunk ' *etua(n)k' e/ëtuak', ëtvank'

2 pi. etuk' advak' *etu-ik '

3 pi. etun advayn4 *etu-in

MArm. SWA SEA

1 sg. tu-i tu-i tv(ec')i

2 sg. tu-ir tu-ir tv(ec ')ir

3 sg. etu-r, eret, tu-aw tu-aw tvec '

lpl. tuak' tu-ink ' tv(ec ')ink'

2 pi. tuik ' tu-ik' tv(ec')ik'

3 pi. tuin tu-in tv(ec')in

cent. +)

Table 8. The paradigm of the verb 'to give' in the Western grouping of the dialects

Svedia Pr.-Sved. Zeyt'un Pr.- Zeyt'. Hamsen

1 sg. dva *tu-i dsve *tu-i dvi

2 sg. dvir *tu-er duvuy *tu-ir dvir

3 sg. i-dor *e-tu-r dovov *tu-aw dvav, ered

lpl. dvunk' *tu-ank ' dovonk ' *tu-ank ' dvak'

2 pi. dvak' *tu-ik ' dsvek' *tu-ik ' dvik '

3 pi. dven *tu-in dsven *tu-in dvin

Mus Sasun Polis Moks Van

1 sg. tav(ec ')i dvac '3 duvi tava tv(ic ')i

2 sg. tav(ec ')ir dvac 'sr duvir tavir tv(ic ')ir

3 sg. tavec ' dvsc' duvav i-tu tvec ', i-tu(r)

lpl. tav(ec')ink ' dvac'sk' (!) duvink ' tavink'y tv(ic')inkv

2 pi. tav(ec ')ik ' dvac 'ak ' duvik ' tavik'y tv(ic')ik'y

3 pi. tav(ec ')in dvac 'an duvin tavin tv(ic ')in

1 sg.

2 sg.

3 sg.

Table 9. The paradigm of the verb 'to give' in the Eastern grouping of the dialects

Marala

tuv/rum

tuv/rir

tuv/ric'

Agulis

tvem tves

Samaxi

tur/vi tur/vir

u-tur, tur/vav

4 Printed as anvayn in [Laribyan 1958: 47]; probably a misprint.

Marala Agulis Samaxi

lpl. tuv/runk' tvek' turink'

2 pi. tuv/ruk' tvek' turik'

3 pi. tuv/run tven turin

Krzen Hadrut ' Lori

1 sg. ts/uvi tuve tsvi

2 sg. ts/uvir tuver tsvir

3 sg. z-ret, tsvuc' tu/svav tsvuc'

lpl. ts/uvink' tuvek'y tsvink'

2 pi. ts/uvik' tuvek'y tsvik'

3 pi. ts/uvin tuven tsvin

Notes:

1) For Zeyt'un deve < *tui, cf. neveg < nuik 'arum lily' [Acaryan 2003: 60, 331].

2) All the phonological developments of the vowel i in 1 sg. {-i

> -a), 2 pi. (-ik' > -äk') and 3 pi. (-in > -en) are regular in Svedia. As for 2 sg. dvir, it goes back to *tu-er rather than *tu-ir, cf. giser 'night'

> g'isir, sisern 'chick-pea' > sisir on the one hand, and karmir 'red' > ¿ärmer, moxir 'ash' > mäxxer on the other. For the material, see [Acaryan 2003: 370-374, 377-383].

3) Agulis has no 3 sg. aor. in general [Acarean 1935:267]. According to Acaryan [1935: 145-146, 243], Agulis 1 pi. aor. -Vk' comes from-Vnk' through the loss of the nasal. This is not necessarily the case.

7. Conclusion

Various analogical developments have taken place in dialects; some peripheral dialects preserve archaic features. One finds more than one line of developments from Classical Armenian to modern dialects (not always through Middle Armenian). Inscriptions from North-East of historical Armenia in the 11th century onwards deviate from Middle Armenian but correspond to a dialectal development in Aramo (Syria), the farthest and most isolated dialect in the South-West corner.

Abbreviations

CArm. — Classical Armenian; MArm. — Middle Armenian; SEA — Standard Eastern Armenian; SWA — Standard Western Armenian.

Bibliography

Acaryan 1911 — Hr. H. Acaryan. Hay barbaragitut'iwn: uruagic ew dasawo-rut'iwn hay barbarneri (barbaragitakan k'artesov) // Эминский этнографический сборник. Т. 8. М.: Лазаревский Институт Восточных языков, 1911.

Acaryan 1913 — Hr. Н. Acaryan. Науегёп gawarakan bararan// Эминский этнографический сборник. Т. 9. Тифлис: Лазаревский Институт Восточныхязыков, 1913.

Acaryan 1926— Hr.H.Acaryan. K'nnut'iwn Maralayi barbari. Yerevan: EPH hratarakc'ut'yun, 1926.

Acaryan 1935— Hr. H. Acaryan. K'nnut'iwn Agulisi barbari. Yerevan: Pethrat, 1935.

Acaryan 1941 — Hr. H. Acaryan. K'nnut'yun Polsahay barbari // Gitakan asxatut'yunner 19, 1941. P. 19-250.

Acaryan 1947— Hr.H.Acaryan. K'nnut'yun Hamseni barbari. Yerevan: HSSR GA hratarakc'ut'yun, 1947.

Acaryan 1952— Hr.H.Acaryan. K'nnut'yun Vani barbari. Yerevan: EPH hratarakc'ut'yun, 1952.

Acaryan 2003 — Hr. H. Acaryan. K'nnut'yun Kilikiayi barbari. Yerevan: EPH hratarakc'ut'yun, 2003.

Ant'osyan 1975 — S.M. Ant'osyan. Xonarhman hamakargi zargac'man snt'ac'k's// L.S.Hovsep'yan (ed.), Hayoc' lezvi patmakan k'erakanut'yun. Vol. 2. Yerevan: HSSH GA hratarakc'ut'yun. P. 159-377.

Asatryan 1968 — M. E. Asatryan. Loru xosvack's. Yerevan: EPH hratarakc'ut'yun, 1968.

Avagyanl986 — S.A.Avagyan. Vimagrakan prptumner. Yerevan: EPH hratarakc'ut'yun, 1986.

Baldasaryan-T'ap'alc'yan 1958 — S. H. Baldasaryan-T'ap'alc'yan. Mso bar-Ьагэ. Yerevan: HSSR GA hratarakc'ut'yun, 1958.

Balramyan 1960 — R. H. Balramyan. Dersimi barbarayin k'artezs. Yerevan: HSSR GA hratarakc'ut'yun, 1960.

Balramyan 1961 — R.H. Balramyan. Krzeni barbars. Yerevan: HSSR GA hratarakc'ut'yun, 1961.

Balramyan 1964 — R. H. Balramyan. Samaxii barbars. Yerevan: HSSH GA hratarakc'ut'yun, 1964.

Hovsep'yan 1997—L. S. Hovsep'yan. ZG dari hayeren jeragreri hisataka-ranneri lezun. Yerevan: Van Aryan, 1997.

Karst 1901 — J. Karst. Historische Grammatik des Kilikisch-Armenischen. Strassburg: Verlag vonKarl J. Trübner, 1901.

Karst2002 — J.Karst. Kilikyan hayereni patmakan k'erakanut'yun (Modern Armenian translation from Karst 1901 by D. Alivnyak). Yerevan: EPH hratarakc'ut'yun, 2002.

Kostandyan 1985 — D. M. Kostandyan. Xotrjuri barbarayin yurahatkut'yun-neri lezvaasxarhagrakan bnut'agirs II H. D. Muradyan (ed.). Hayereni barbaragitakan atlas: usumnasirut'yunner ew nyut'er. Vol. 2. Yerevan: HSSH GAhratarakc'ut'yun, 1985. P. 46-68.

Laribyan 1958 — A. S. Laribyan. Hayereni norahayt barbarneri mi nor xumb. Erevan: HSSR GAhratarakc'ut'yun, 1958.

Martirosyan 2013— Г.К.Мартиросян. Армянские диалекты: характеристика отдельных диалектов II Ю. Б. Коряков, А. А. Кибрик, (ред.). Языки мира. Реликтовые индоевропейские языки Передней и Центральной Азии. М.: Academia, 2013. С. 334-385. [Н. Martirosyan. Armianskie dialekty: kharakteristika otdel'nykh dialektov II Iu. B. Koriakov, A. A. Kibrik. (ed.). Iazyki mira. Reliktovye indoev-ropeiskie iazyki Perednei i Tsentral'noi Azii [Languages of the World: Relict Indo-European Languages of Western and Central Asia], M: Academia, 2013: P. 334-385].

Muradyanl962 — M.H.Muradyan. Sataxi barbars. Yerevan: EPH hratarakc'ut'yun, 1962.

Muradyan 1982 — M. H. Muradyan. Urvagic Moksi barbari IIH. D. Muradyan (ed.). Hayereni barbaragitakan atlas: usumnasirut'yunner ew nyut'er. Vol. 1. Yerevan: HSSHGAhratarakc'ut'yun, 1982. P. 108-183.

Orbeli et. al. (eds.) i960-... — H. Orbeli, S. G. Barxudaryan et al. (eds.). Divan hay vimagrut'yan. Yerevan: HSSR GAhratarakc'ut'yun, I960-...

Petoyan 1954 — V. A. Petoyan. Sasuni barbars. Yerevan: HSSH GA hratarakc'ut'yun, 1954.

Polosyan 1965 — A.M. Polosyan. Hadrut'i barbars. Yerevan: HSSH GA hratarakc'ut'yun, 1965.

Vaux 2007 — B. Vaux. Homshetsma: the language of the Armenians of Hamshen II H. H. Simonian (ed.). The Hemshin: history, society and identity in the Highlands of Northeast Turkey. London — New York: Routledge, 2007. P. 257-278.