CC BY
17
P. Kocharov
The Etymology of Armenian gelardn 'Spear'*1
Резюме: Статья посвящена происхождению др.-арм. gelardn 'копье', которое до сих пор не имеет надежной этимологии. Предлагается возводить gelardn к и.-е. *welH-wr-ti- — деривату гетероклитической основы от корня *welH- 'бить', ср. хетт. walhzi 'бить', лат. vello 'ощипывать; разрушать' и т. д. Аналогичная морфологическая структура засвидетельствована в др.-арм. neard 'жила' из *sneh1-wr-ti- ^ *sneh1-wr-/-wen-, ср. др.-инд. snavan-, мл. авест. snavara, др.-гр. vsvpov, лат. nervus. В статье затрагиваются проблемы отражения в древнеармянском индоевропейской группы фонем *-Hwr- и гетероклитических основ на *-wr-/-wen-.
Ключевые слова: древнеармянский язык, древнеармянская этимология, индоевропейские гетороклитические основы, композиты.
Arm. gelardn 'spear' has not so far received a satisfactory etymological explanation. The word is amply attested in Classical Armenian; in the Bible, it renders a variety of Greek designations of spear such as Sopv [VT, Sam. I, 26: 7], loyxn [VT, Esr. II, 4: 16], and aeipo^aaxnq [VT, Num., 25: 7], see [NHB I: 535-536]. Nevertheless, this word is not mentioned in standard handbooks on Armenian etymology such as [Hubschmann 1897], [Solta 1960], [Olsen 1999], and [Martirosyan 2010]. However, it
* Статья написана в рамках проекта «Древние индоевропейские тексты», выполняемого по Программе ОИФН РАН «Текст во взаимодействии с социокультурной средой: уровни историколитературной и лингвистической интерпретации», направление V «Лингвистические аспекты исследования текста».
1 The etymology considered in the present article was inspired by the discussion of Arm. learn 'mountain' following a talk on the etymology of Gk. Xaaq held by Alexander Nikolaev at the Institute of Linguistic Studies (November, 2009). I would like to thank Prof. Leonard Herzenberg, Sergei Ivanov, Prof. Nikolai Kazansky, Hrach Martirosyan, Alexander Nikolaev, and Andrei Shatskov for many valuable comments and suggestions on the first drafts of the paper. All the responsibility for the actual content of the paper is mine.
is mentioned in [HAB I: 533]; as to the etymology of gelardn, H. Acaryan references to [NHB loc. sit.], where it is considered to be a compound of gelecik 'beautiful' or gelic 'wringing' and ardn 'lance', and to [Hiwnkearpeyentean 1894], where it is derived from a combination of Pers. gelan ‘big, thick’ and ardn. Although Acaryan does not express his opinion explicitly, he seems to support the etymological connection between synonymous gelardn and ardn, see also [HAB I: 310]. In the present article, I shall consider the above-mentioned etymological solutions and provide arguments for a new one.
The noun in question is attested as an n-stem in the early Classical Armenian texts and as a consonant stem gelard in some later texts, e.g. in Movses Kalankatuaci's Patmut'iwn Atuanic Asxarhi, II, 11: Ew areal gelard i jern... "And, having taken a spear in hand..." It is once attested in the form geleardn in Ephrem’s Meknowt’iwn Awetarani hamarbar (Venice, 1836; apud. [HAB I: 533]). The n-stem oscillates between the two inflectional types — the jukn-type (gen. sg. -an, nom. pl. -unk', gen. pl. -anc ) and the stin-type (gen. sg. -ean, gen. pl. -eanc ), see [NHB loc. sit]. The n-stem declension is beyond doubt relatively recent with this word: it must have developed according to the pattern of other secondary n-stems like elungn 'nail' (gen. sg. -an, nom. pl. -unk ), steln 'branch' (gen. sg. -an, nom. pl. -unk ), etc. This secondary -n should come analogically from neuter nom.-acc. *-m, e.g. tasn 'ten' < *dek-m 'id.' [Meillet 1936: 56], [Viredaz 2001-2002], even in the case of inherited heteroclitic stems like learn 'mountain', against [Olsen 1999: 122], where -rn is considered as the contamination of nom.-acc. sg. *-r and oblique *-n. The n-stem of gelardn may be due to the analogy with ardn.
Given the proposed etymologies of gelardn, I shall start with some remarks on Arm. ardn 'lance' (nom. pl. -unk', gen. pl. -eanc ). In the Bible, the word renders Greek apSig, Sopv, and Xoyxn and, as such, almost completely overlaps semantically with gelardn; in [VT, Job, 41: 18], ardunk and gelardunk are the reading variants that translate Gk. Xoyxoi, see [Cox 2006: 264].
As a second member of a compound, -ardn is only attested in erkardeay 'double-edged lance' in the 5th c. (?) text —
Patmut'iwn Atek'sandri Makedonacwoy (Venice, 1842; apud. [HAB I: 310]). No safe etymology has been proposed for Arm. ardn so far, see [HAB loc. sit.], [Alayan 1974: 31-34], and [Jahukyan 1987: 158] in the first place.
From my point of view, the comparison of gel- with a derivative from the verb gelum 'to wring', be it an agent noun gelic (as in NHB) or a participle, is the most reasonable of the proposals mentioned in HAB. Still, the semantic side of this etymology is not entirely convincing if we interpret gel-ardn as a compound '*the wrung/wringing lance' along with ardn 'lance'. In this connection, I would like to propose yet another etymological solution and consider gelardn as a continuant of a simplex, not a compound.
In my opinion, the inherited morphology of gelardn is comparable to that of leard 'liver' (gen. sg. lerdi) and neard 'sinew' (gen. sg. nerdi). While leard is most plausibly derived from PIE *(H)yekw-r-ti- (Ved. yakrt, gen. sg. yaknah) ^ PIE *(H)yekw-r-/-en- (see [Wodtko et al. 2008: 392-395] for an account of the proposed reconstructions), neard comes from *snehj-wr-ti- ^ PIE *snehj-wr-/-wen-, cf. Ved. snavan-, YAv. snavars, Gk. vevpov, Lat. nervus 'sinew' (see [Martirosyan 2010, s. v.] with references). In the same way, gelardn may be derived from *welH-wr-ti-, a derivative of heteroclitic *welH-wr-/-wen-. The root *welH- may be further compared to Hitt. walhzi 'to hit', Toch. A wallastar 'dies', Lat. vello 'I pull out', etc., see [LIV2: 679] and [Kloekhorst 2008: 945-946], where the PIE *welh3- 'to hit' is reconstructed.
Both neard and leard are *-ti- derivatives from the nominative case of inherited heteroclitic stems. It is difficult to decide whether *-ti- is original or an inherited *-t- stem that turned into the Armenian i-declination, see [Clackson 1994: 5556]. In the case of leard, the latter scenario is supported by Ved. yakrt. Furthermore, we may recall the secondary t-stems derived from n- and r/n-stems in Greek, see [Sihler 1995, § 293] and, from a larger Indo-European perspective, noun stems extended with a -t- in Hitt. ispan-t- 'night' (besides Ved. ksap-'id.'), whereas a -t- looks like a deverbal suffix of nomen instrumenti in Hitt. sas-t 'bed' (next to ses- 'sleep'). Still, there
are too few decisive morphological correspondences to establish t-stems as a type for Proto-Armenian.
Another solution is to reconstruct a pure *-ti- stem for the above-mentioned Armenian words. This accords with the pattern of substantivization of an underlying *-to- adjective discussed in [Nussbaum 2004]: PIE *keh3- 'sharpen' (Ved. sisati 'id.') ^ *koh3-t- 'the sharpening one' (Lat. cos 'whetstone') ^ *kh3-to- 'sharp' (Ved. sita- 'sharp', Lat. catus 'shrill') ^ *koh3-ti-/*keh3-ti- 'the sharp one' (Lat. cotes 'reefs'). The morphological and semantic aspects of the etymology of gelardn may be supported by such examples as Av. tiyra- 'sharp' ^ tiyri-'arrow' and Gk. aKpoq 'topmost' ^ aKpiq 'peak'. Another parallel is the insightful etymology of Gmc. *swerda- 'sword' from *s(h2/3)w-er-t-o- 'sharp' proposed by A. Nikolaev [Nikolaev 2009]. Without immediately accepting the hypothesis of the delocative origin of r/n-stems (*seh2/3u-'sharpness' ^ delocative derivative *s(h2/3)w-er-t- 'that which is in/of sharpness' ^ *s(h2/3)w-er-t-o- 'sharp'), I assume that both designations of weapons may ultimately come from similar Indo-European morphological structures. Unlike *per-wn-to- 'rocky' (Ved. parvata-, YAv. pauruuata- ‘id. ’) from *per-wr-/-un- 'rock', *s(h2/3)wer-to- is a derivative from the r-stem just like Arm. leard, neard, and gelard.
Thus, it seems plausible to reconstruct the following derivational chain: *welH- 'to strike, overturn' ^ neuter abstract noun *welH-wr-/-wen- 'strike' (heteroclitics often denoted inanimate entities capable of performing an action like 'fire', 'water', and 'sun' as well as abstract notions of action) ^ adjective *welH-wr-to- 'striking' ^ substantivized adjective *welH-wr-ti- *'the striking one' > 'spear'.
The development of PIE *-wr- is troublesome in Armenian. The most straightforward examples are aliwr/alewr 'flour' < *alewr < PIE *h2leh]-wr (Gk. alevpov, aleap '(wheat) flour')2 and albiwr 'fountain' < *arbewr < PIE *bhrehI-wr (Gk.
2 N. Kazansky [Kazansky 2001: 118] evokes Gk. a/pi and ^s/pi 'up to' along with Arm. merj 'near' from PIE *meghr- as a possible parallel to Gk. alsvpov and Myc. me-re-u-ro; if the two latter words belong
tppsdp), see [Martirosyan 2010, s. vv.]. The coherent evidence of neard < *near-ti-, leard < *lear-ti-, and gelard < *welar-ti- shall be correlated with the accent shift to a *-ti- suffix from the root. Here may also belong Arm. learn 'mountain' < *lear-no- if from PIE *leh2-wr-no- (Gk. laag, see [Nikolaev 2010] for this etymological solution). Still, the behavior of *-w- in such double outcome of *-H-wr- is enigmatic, see [Kortlandt 1993]. Arm. hur 'fire', gen. sg. hroy < *puro- < PIE *ph2-ur-/-un- (Gk. nvp, ON furr 'fire'; Arm. hnoc 'oven' from oblique n-stem) < PIE *peh2-wr-/-wen- (Hitt. pahhur; see [Wodtko et al. 2008: 540545]), shall be due to the root zero-grade and laryngeal metathesis that occurred with this word in some of the Indo-European languages. In view of the etymology of Arm. hur, it is tempting to postulate analogical reconstruction for Arm. sur 'sword, dagger' (gen. sg. sroy) < *suro- < PIE *kh3-ur-/-un-(against [Olsen 1999: 55] and [Nussbaum 2004: 1]). Arm. olorn 'drop' and olor 'twisting' are hard to explain as derivatives from PIE *kwlhI-wr [Olsen 1999: 139] on formal grounds, see [Martirosyan 2010, s. v.].
In view of the heteroclitic stem underlying gelardn, it is tempting to draw Gk. eilap 'protection; fence of canes' (e.g. [Od. 5: 257]) < *peXpap- to the comparison. Although, the conversive semantic shift from 'to hit' to 'to repel, protect' presumably bound to the middle voice connotations of the underlying verbal root *welH- may well appear to be a phantom. A semantic parallel, still, would be cognates Arm. slak' 'spear, dagger, arrow' and MPers. swl’ck 'grill' from *kuHlos [Mallory, Adams 2006: 245].
According to the proposed etymology, the Armenian word for 'spear' is a substantivized epithet — 'the striking one'. The archaic morphology of *welH-wr-t(i)- makes this designation of spear a suspect for the Indo-European lexicon.
Bibliography
together and are cognates to Lat. molo 'to grind', Arm. aliwr is not related to them. Still, the similarity of between Greek and Armenian, both formal and semantic, is hardly by chance [DELG I: 59].
Alayan, E. B. 1974. Baraknnakan ew stugabanakan hetazotut'yunner. Erevan.
Clackson, James. 1994. The Linguistic Relationship between Armenian and Greek. Oxford: Blackwell.
Cox, Claude E. Armenian Job: Reconstructed Greek Text, Critical Edition of the Armenian with English Translation. Leuven, Paris, Dubley, MA: Peeters.
DELG = Chantraine, Pierre. 1968-1980. Dictionnaire etymologique de la langue grecque. Heidelberg, Paris.
HAB = Acaryan, Hrac'ya Hakobi. 1971-1979. Hayeren armatakan bararan. 2nd ed., 4 vols. Yerevan (1st ed., 8 vols. Erevan, 19261936).
Hiwnk'earpeyentean, J. 1894. Stugabanakan bararan. Polis.
Hubschmann, Henrich. 1895-1897. Armenische Grammatik. I Teil: Armenische Etymologie. Leipzig: Breitkopf & Hartel.
Jahukyan, Gv B. 1987. Hayoc lezvi patmut'yun: naxagrayin
zamanakasrjan. Yerevan.
Kazansky, Nikolai N. 2001. PIE *megh-. Zeitschrift fur celtische Philologie 52, 118-120.
Kloekhorst, Alwin. 2008. Etymological Dictionary of the Hittite Inherited Lexicon. Leiden, Boston: Brill.
Kortlandt, Frederik. 1993. Intervocalic *-w- in Armenian. Annual of Armenian Linquistics 14, 9-13.
LIV2 = Rix, Helmut et al., eds. 2001. Lexicon der indogermanischen Verben. 2nd ed. Wiesbaden: Reichert Verlag.
Mallory, J.P., D.G. Adams. 2006. The Oxford Introduction to Proto-Indo-European and the Proto-Indo-European World. Oxford University Press.
Martirosyan, Hrach. 2010. Etymological Dictionary of the Armenian Inherited Lexicon. Leiden, Boston: Brill.
Meillet, Antoine. 1936. Esquisse d'une grammaire comparee de l'armenien classique. Vienne.
NHB = Awetik'ean G., X. Siwrmelean, and M. Awgerean. 1836-1837. Nor bargirk haykazean lezui. 2 vols. Venice: St Lazar (reprinted Yerevan: University Press 1979-1981).
Nikolaev, Alexander. 2010. Time to gather stones together: Greek laaq and its Indo-European background. Proceedings of the 21st Annual UCLA Indo-European Conference / edited by C. Melchert, B. Vine, and S. Jamison. Bremen: Ute Hempen Verlag.
—. 2009. The Germanic word for 'sword' and delocatival derivation in Proto-Indo-European. The Journal of Indo-European Studies 37, 3/4: 462-488.
Nussbaum, Alan. 2004. A -t- Party: Various IE Nominal Stems in *-(o/e)t-. Sixteenth Annual UCLA Indo-European Conference; November 5-6, 2004. Handout.
Olsen, B. A. The Noun in Biblical Armenian: Origin and Word-Formation: With Special Emphasis on the Indo-European Heritage.
Berlin, New York: Mouton de Gruyter (Trends in linguistics. Studies and monographs 119).
Sihler, Andrew. 1995. New Comparative Grammar of Greek and Latin.
Oxford University Press.
Solta, George R. Die Stellung des Armenischen im Kreise der indogermanischen Sprachen: eine Untersuchung der indogermanischen Bestandteile des armenischen Wortschatzes. Wien: Mechitharisten. (Studien zur armenischen Geschichte 9).
Viredaz, Remy. 2001-2002. Sur le traitement armenien des sonantes voyelles. Slovo 26-27 (Actes du 6e CILA, 5-9 juillet 1999, Inalco): 24-36.
Wodtko, Dagmar S., Britta Insler, and Carolin Schneider. 2008. Nomina im Indogermanischen Lexicon. Heidelberg: Winter.
