The place of Armenian in the Indo-European language family: the relationship with Greek and Indo-Iranian Текст научной статьи по специальности «Языкознание и литературоведение»

Urach Martirosyan Leiden University

The place of Armenian in the Indo-European language family: the relationship with Greek and Indo-Iranian*

The main purpose of this paper is to present lexical correspondences that unite Armenian with Greek and/or Indo-Iranian. They include shared innovations on the one hand, and isolated lexemes on the other. These two lexical corpora — lexical innovations on an inherited basis and isolated words — can be placed within the same temporal and spatial framework. After the Indo-European dispersal Proto-Armenian would have continued to come into contact with genetically related Indo-European dialects. Simultaneously, it would certainly also have been in contact with neighbouring non-Indo-European languages. A word can be of a substrate origin if it is characterized by: (1) limited geographical distribution; (2) unusual phonology and word formation; (3) characteristic semantics. The material presented here, albeit not exhaustive, allows to preliminarily conclude that Armenian, Greek, (Phrygian) and Indo-Iranian were dialectally close to each other. Within this hypothetical dialect group, Proto-Armenian was situated between Proto-Greek (to the west) and Proto-Indo-Iranian (to the east). The Indo-Iranians then moved eastwards, while the Proto-Armenians and Proto-Greeks remained in a common geographical region for a long period and developed numerous shared innovations. At a later stage, together or independently, they borrowed a large number of words from the Mediterranean / Pontic substrate language(s), mostly cultural and agricultural words, as well as animal and plant designations. On the other hand, Armenian shows a considerable number of lexical correspondences with European branches of the Indo-European language family, a large portion of which too should be explained in terms of substrate rather than Indo-European heritage.

Keywords: Armenian historical linguistics, Armenian etymology, Indo-European comparative linguistics, Indo-Iranian lexicology, Greek lexicology, Mediterranean substrate.

Introduction

1. Method

2. Phonological isoglosses

3. Morphological isoglosses

4. Lexical isoglosses between Armenian, Greek and Indo-Iranian

4.1 Material

4.2 Pair isoglosses

5. Lexical isoglosses between Armenian and Indo-Iranian

5.1 Armenian and Indo-Iranian

5.2 Armenian and Indo-Aryan

5.3 Discussion

5.3.1 Native or loan?

5.3.2 Armeno-Indo-Iranian poetic or mythical lexicon

5.3.3 Other issues Table set A (sections 4-5)

6. Lexical isoglosses between Armenian, Greek and European dialects

6.1 Armenian and Greek: innovations

6.2 Armenian and Greek: isolated words

* I am greatly indebted to Kate Bellamy for checking my English. I am also much obliged to James Clackson, Frederik Kortlandt, Alexander Lubotsky and the editorial staff of JLR for their thorough and helpful comments.

Journal of Language Relationship • Вопросы языкового родства • 10 (2013) • Pp. 85 — 137 • © Martirosyan H., 2013

6.3 Armenian, Greek and Albanian

6.4 Armenian, Greek and Latin

6.5 Armenian, Greek and Germanic and/or Celtic

6.6 Armenian, Greek and Balto-Slavic

6.7 Armenian and Greek in a broader European context 7. Armenian, Greek and the Mediterranean/European substrate Table set B (sections 6-7)

Preliminary conclusions

Introduction

The dialectal position of Armenian has attracted the attention of Armenologists ever since Heinrich Hubschmann (1875/1877) proved that Armenian does not belong to the Iranian group of Indo-European languages and should be treated as an independent branch of the Indo-European family. Mainly under the influence of the centum / satam division, Armenian was considered to be in close relationship with the Aryan and Balto-Slavic languages for a long period, until Pedersen (first in 1906: 442), Meillet and others noted that the number of Greek-Armenian agreements is greater than the number of agreements between Armenian and any other Indo-European language. The relations between Armenian and Greek are sometimes regarded within a larger Balkan context including Phrygian, Thracian and Albanian.

Some scholars argued that there are a large number of similarities between Greek and Armenian, which allow for the postulation of a common Graeco-Armenian language.1 It is now clear, especially after Clackson's (1994) thorough, albeit somewhat hypercritical treatment, that this case is not as strong as it is for Indo-Iranian and Balto-Slavic. The contact relations between Proto-Greek and Proto-Armenian may have been intense, but these similarities are considered insufficient to be viewed as evidence for discrete Proto-Graeco-Armenian.2

There are also connections between Armenian and Indo-Iranian on the one hand, and between Armenian, Greek and Indo-Iranian on the other. Armenian is usually placed between Indo-Iranian to the east and Greek to the west, and on the northern side it might neighbour Balto-Slavic (and/or Germanic and others). The dialectal unity of Armenian, Greek, (Phrygian,) and Indo-Iranian is often discussed.3

1 Note the term Helleno-Armenian in Hamp 1979: 4-5; 1983: 6; 1992: 58.

2 For such a moderate approach, see Meillet 1936: 9, 141-143. For more literature and different discussions of this and related issues, see Meillet 1922/1950; Pedersen 1924; Bonfante 1937, 1981; Adjarian 1937, AcarHLPatm 1, 1940: 23-99; Makaev 1967; Schmitt 1972-74: 34-40, 64-67; 1975: 27; Sirokov 1977, 1980; de Lamberterie 1978-79; 1992: 236-239; Jahukyan 1980; Wyatt 1982; Hamp 1983a; Beekes 2003: 152-153; Ringe / Warnow / Taylor 2002: 102106. See especially the monographs: Arutjunjan 1983; Clackson 1994. For the relationship of Armenian with Balkan (and Asia Minor) languages, see Jahukyan 1970; 1987: 296-306; Diakonoff 1984: 103-104, 110-112, 120-121, 18417, 188-190; Holst 2009: 49ff; Kortlandt 2010: 4-6, 31-32, 78; de Lamberterie 2012 and 2013; individually: Phrygian (Pedersen 1925: 44-49; Haas 1939; Bonfante 1946; Jahukyan 1968; Orel 1993; Clackson 2008: 124), Albanian (Pedersen 1900; Kortlandt 1980, 1986; Rusakov 1984), Thracian (Kortlandt 2003: 83-87; Beekes 2003: 153). For Urartian names of Balkan origin, see Petrosyan 2002: 179-182 and 2005 (with literature).

3 For various views and discussions, see Meillet 1896: 149-155; 1936: 142; Pedersen 1924: 224-225 = 1982: 307308; Specht 1935: 29-30, 102-103; 1939: 8, 12-14; AcarHLPatm 1, 1940: 85-86; Thieme 1954: 582-590; Porzig 1954: 162-164; Birwé 1956: 6; Solta 1960: 459ff; Schmitt 1967: 259-260; Makaev 1967: 453-455, 461; Xac'aturova 1973, 1979; Pisani 1979: 210; Euler 1979: 18-23; Jahukyan 1980: 4; Wyatt 1982: 27; Gamkrelidze/ Ivanov 1984, 1: 417-418 = 1995:

After the well-known large-scale investigations of Porzig (1954) and Solta (1960), in the period between 1970's and 1990's there have been made a number of attempts to study the relations between Indo-European branches by means of statistics: Tischler 1973, Davies / Ross 1977, Bird 1982 (updated 1993), Coleman 1992, etc. With respect to Armenian one has to mention especially the works of Jahukyan, 1980, 1983, and 1987: 86-222. For a critical account of these studies, see Clackson 1994: 7-8, 193-198. For a recent attempt to recover the first-order subgrouping of the Indo-European family using a new computational method, see Ringe / Warnow / Taylor 2002: 102-106.

As far as the morphological and lexical isoglosses are concerned, in these statistical investigations Greek and Indic mostly appear among the closest languages to Armenian. As an example, in the table below I present Jahukyan's data on the first five language branches displaying the highest number of common features taken from his lists of 27 phonetic, 35 morphological, and 1400 lexical isoglosses.

Phonological Morphological Lexical

1. Greek 14 Indic 20,5 Greek 878

2. Phrygian 12,5/13,5 Greek 18 Germanic 783,5

3. Thracian 11/13 Anatolian 17 Indic 661,5

4. Slavic 11 Tocharian 16,5 Italic 636

5. Iranian 10,5 Italic 16 Baltic 625,5

6. Baltic 10 Iranian 13,5 Slavic 579,5

7. Celtic 9,5/10,5 Slavic 13 Celtic 542,5

Rather than discuss here the different theories of the relationships between Armenian and other language branches and tackle every individual isogloss, which would require a copious monograph, I shall limit myself to a general outline of the most relevant issues regarding Greek and Indo-Iranian. After a short methodological outline (§ 1) and sections on phonological and morphological agreements (§§ 2-3), I shall turn to the main goal of this paper, the lexical material. Applying the methodology outlined in § 1, I shall select the most illustrative examples from the lists that have been used before and will add some new material that has not been discussed in this context before. Additionally I present a number of new etymologies which are marked as HM.

In cases where I give no references, the relevant etymological material can be found in HAB and Martirosyan 2010 s.v.

1. Method

The methodological and thematic background of this paper largely coincides with that of Clackson 1994, so I simply omit these discussions and refer the reader to this exemplary monograph.

A crucial methodological point of departure is that archaic features and independent developments are not significant for determining a close genetic relationship between two languages or dialects. Instead, one should rely on shared innovations from the outset. The draw-

365; de Lamberterie 1986; Manczak 1987; Pisowicz 1987; Schmidt 1987; Hamp 1992; Lehmann 1993: 19; Clackson 1994: 201-202; 2008: 124; Mallory/Adams 1997: 29; 2006: 78-79, 109-110, 455; Stempel 2000; Ritter 2006; Schmitt 2007: 22-23; Fortson 2010: 203, 383.

back with this method is that there is often (if not always) the possibility of independent innovations yielding similar results. Nevertheless, the cumulative evidence decreases the likelihood of chance in such cases.

On the other hand, if a competing feature is present in a certain group of dialects that is otherwise confirmed by a number of isoglosses, it should be taken into account even if we cannot formally decide whether we are deling with an archaism or innovation. Thus Armenian, Greek, Phrygian and Indo-Iranian make use of the e-augment (§ 3.1) whereas other dialects do without it, and it is impossible to decide whether the presence or absence of an augment is to be regarded as dialectal innovations made in late Proto-Indo-European. Nevertheless, this is a significant isogloss, because a similar dialectal distribution is found, as we shall see, for a number of morphological and lexical variables.

In the case of, e.g., the genitive ending *-osio-, however, some archaic traces are also found in other branches, for example Italic and Celtic (see § 3.2). This is reminiscent of e.g. IE *fener, gen. *hinr-6s 'man' that is basically represented by the dialect area under discussion (Armenian ayr, gen. am, Skt. nar-, Greek avf\p, gen. avSpoc, Phryg. avap, cf. also Alb. njeri 'human being, person') but has also left some traces in Italic (Osc. ner-) and Celtic (Mir. ner 'boar', MWelsh ner 'chief, hero').4 A similar case is *h2erh3-uer/n- 'arable land'. Such cases cannot be regarded as significant for the purpose of subgroupping or establishing areal contacts.

When an etymon is only found in two or three non-contiguous dialects, it may theoretically represent an archaic PIE lexeme that has been lost elsewhere5 and is thus not significant for our purpose. But when an etymon appears in a few dialects that can be regarded as contiguous at a certain stage, we should take it seriously even if the etymon has no PIE origin and cannot be thus treated as a shared innovation in the genetic sense. Two Indo-European dialects that were spoken in the same geographical area at a period shortly before and/or after the Indo-European dispersal could both develop shared innovations as a result of their interaction with neighbouring non-Indo-European languages.

After the Indo-European dispersal Proto-Armenian would have continued to come into contact with genetically related Indo-European dialects. Simultaneously, it would certainly also have been in contact with neighbouring non-Indo-European languages. A word can be of a substrate origin if it is characterized by: (1) limited geographical distribution; (2) unusual phonology and word formation; (3) characteristic semantics.

Theoretically, these two lexical corpora — lexical innovations on an inherited basis and substrate words — can thus be placed within the same temporal and spatial framework. As far as the relationship between Armenian and Greek is concerned, matters are particularly complicated for two principle reasons: (1) it is often very difficult to know whether we are dealing with an innovation or a substrate / cultural word, and (2) aside to lexical correspondences confined to only Armenian and Greek, there are also a large number of lexical agreements between Armenian, Greek and a few other European dialects. Many of these words belong to the semantic fields of the physical world, fauna, flora, agriculture and crafts. They may, therefore, shed some light on cultural and geographical characteristics of the environment where Proto-Armenian might have contacted Proto-Greek and other dialects in a late period around the time of the Indo-European dispersal. For these reasons, in section six, following the subsection on Armenian and Greek only, I also provide a few characteristic examples reflecting the broader areal context. I then present a brief discussion on

4 On this etymon see de Vaan 2008: 406-407; Matasovic 2009: 289; Martirosyan 2010: 61-62; Beekes 2010, 1: 103-104.

5 For the problem of determining a PIE word, see e.g. Mallory/Adams 2006: 107-110.

the substrate (section 7) and add a summarizing table divided into semantic fields. Wherever a lexical agreement is likely to be an innovation rather than an isolated etymon, I mark it by shading.

2. Phonological isoglosses

2.1. A major and very complicated issue is the centum / satdm division which puts Greek and Armenian on different sides of the line. Together with Indo-Iranian, Balto-Slavic and Albanian (the situation in Luwian is disputed), Armenian belongs to the satdm group of languages which show palatalisation of the palatovelars and absence of a labial element in their reflexes of the labiovelars.6

2.2. Another phonological feature that unifies Indo-Iranian, Balto-Slavic and partly Armenian is the ruki/iurk-rule, the special development of *s after r, k, i, and u.7 It is interesting that Armenian shows a twofold development of *-rs- as reflected in t'arsamim and t'aramim 'to wither'; the -rs- reflex is in line with the ruki-development, whereas -r- betrays an assimilation of *-rs- to *-rr-, also seen in Greek. This issue can be placed within the framework of the development *s > h in Armenian, Greek, Iranian, Phrygian and Lycian (and also Brythonic Celtic). Both developments may have resulted from a common change, although independent innovations are not unlikely either.8

2.3. The so-called "prothetic vowel", viz. Gr. a- (and o-) : Arm. a-, and Gr. e- : Arm. e- vs. zero in other languages, is now interpreted as a vocalized reflex of the PIE initial laryngeal followed by a consonant (see the table below). It is considered an important isogloss shared by Armenian and Greek, and possibly also Phrygian and Albanian.9 Clackson (1994: 36) notes that this may represent an areal feature since initial laryngeals might also have left vocalic reflexes in the Anatolian languages. However the latter statement seems to be uncertain.10

PIE Gloss Greek Armenian Other

*hi regwos- 'darkness' ëpefioç erek(-oy) Goth. riqis

*h\neun 'nine' èvvéa inn Skt. nava

*h2Ïe/o(u)pek- 'fox' àAwnrj^ aiuës Skt. lopasa-

*hister- 'star' àax^p asti Hitt. haster-

^néhmn 'name' ovo^a anun Skt. naman-

6 See Pedersen 1925: 7, 44-47; Allen 1978; Shields 1981; Gamkrelidze/ Ivanov 1984, 1: 417 = 1995: 365; Clackson 1994: 54-55. It is remarkable that Luwian preserves the original labialized reflexes of labiovelars (for relevant literature, see Szemerenyi 1996: 61i).

7 See Martirosyan 2010: 709-710; Beekes 2011: 30, 126-127, 137. I do not share the view (see Olsen 2011: 26-27 with lit.) on the final -r as a ruki-development in Armenian.

8 For references and a discussion of these two issues see Clackson 1994: 54, 21086; for *s > h, cf. Szemerenyi 1985; Schmidt 1988: 602. Note that the change s > h in Lycian and Brythonic Celtic is certainly not a common archaism with Greek, Armenian, and Iranian according to received opinion on the matter (cf. already Meillet 1896: 151 on Celtic).

9 For literature and a discussion, see Martirosyan 2010: 714-716; de Lamberterie 2013: 29-34. See also Schmidt 1988: 602. For Phrygian, see Ligorio / Lubotsky forthc. (section 4.3); for Albanian, see Demiraj 1994.

10 For a discussion, see Kloekhorst 2006 and his manuscript monograph on Hittite accentuation, notably the section "Words containing aC(-)".

2.4. Vocative accent. Armenian manuscripts and dialects provide rich evidence for vocative forms accented on the first syllable: hayrik 'o father', mayrik 'o mother', Karapet, etc.; dial: T'iflis axper 'o brother', vurt'i 'o son'; Lori ordi 'o son'; Hamsen hayr-i 'o father', mayri 'o mother'; Akn harsnuk 'o sister-in-law', marik 'o mummy'; Moks xrotper 'o uncle', t'agavur 'o king', Nor-Bayazet hars-e 'o sister-in-law', Hofomsim, Mayran, Margarit, tnakolner "you whose house may be destroyed!", even word combinations, such as iurban harse "you, dear sister-in-law (to whom I may be sacrificed)", Hfop'sama Xat'un 'o you, Lady Hrop'sim'.

The vocative with initial accentuation may be considered an Indo-European inheritance. In Vedic Sanskrit, the vocative, when accented, has the acute on the first syllable, e.g., voc. pitar vs. nom. pita. The same is found in Greek: aSeAtye vs. aSeAtydc 'brother'; Seonoza vs. Seondzric 'master (of the house), lord'; nazep vs. nazrp 'father', etc.; in modern Iranian languages: in Persian, the stress is on the initial syllable of the vocative noun or phrase. In Kurdish Awroman, when no vocative particle is present the stress is brought forward to the first syllable of a noun.11 This isogloss is highly hypothetical.

3. Morphological isoglosses

3.1. One of the most significant morphological isoglosses shared by Armenian, Greek, Phrygian and Indo-Iranian is the e-augment (cf. § 1), e.g. Arm. 3sg aorist e-ber 'brought' from PIE *e-bher-et: Skt. a-bhar-at, Gr. e-tyep-e; Arm. 3sg aorist e-git 'found' from PIE *e-uid-et: Skt. a-vid-at, Gr. elSe < e-piS-e; Arm. 3sg aorist e-d 'put' from PIE *e-dhehi-t: Skt. a-dha-t, Gr. dial. e-Orj, cf. suffixed forms, Gr. e-Orj-Ka, Phrygian e-daes.12

3.2. The genitive ending *-osio- (Skt. -asya, Gr. -oio, Arm. -oy,13 etc.) of the nominal o-stems has been taken over from the pronominal declension. It is basically restricted to Indo-Iranian, Greek and Armenian and has been interpreted as either a dialectal Indo-European innovation or a morphological isogloss.14 Given the appearance of this genitive singular ending in Italic (-osio in early Faliscan inscriptions and in one early Latin inscription, the Lapis Satricanus, c. 490 bc, and in the name Mettoeo Fufetioeo) and Celtic (-oiso in three or four Lepontic inscriptions from before 400 bc), it is now possible to argue that the spread of a genitive singular*-? took place relatively recently, not much earlier than the period of Italo-Celtic unity. It has been argued that the ending *-osio- was also present in Anatolian. As an archaism it cannot, therefore, be used as an isogloss. Nevertheless, it is somehow significant that, as in case of the e-augment, Armenian sides with Greek and Indo-Iranian in having *-osio- as a specific genitive marker of o-stems.15

11 See Martirosyan 2010: 748-749 and Martirosyan forthc.

12 For a discussion, see Meillet 1950: 97-101; Birwe 1956: 18-19; Meid 1975: 214-215; Schmidt 1980: 2-5; 1987: 39; 1988: 601-602; Gamkrelidze/ Ivanov 1984, 1: 388-390 = 1995: 340-341; de Lamberterie 1986: 48-49; 1992: 237; Abajyan 1991; Clackson 1994: 9-10; Meier-Brügger 2003: 182; Mallory/Adams 2006: 75; Kocharov 2008: 32-33; Fort-son 2010: 92, 101, 392. For Phrygian, see also Ligorio / Lubotsky forthc. (section 5.3)

13 Meillet 1900: 17.

14 See e.g. Gamkrelidze/Ivanov 1984, 1: 375-379 = 1995: 329-332; Kortlandt 1984: 99-100 = 2003: 47.

15 For some other possibly related forms and a general discussion, see Lehmann 1981; Gamkrelidze / Ivanov 1984, 1: 3771 = 1995: 3303; Schmidt 1987: 40-42; Beekes 1990-92; Morani 1992; Hamp 1992: 59; Clackson 1994: 8, 14; Szemerenyi 1996: 184, 1876; Clackson/Horrocks 2001: 16-17, 32, 69; Eska/Wallace 2001; Fortson 2010: 127. On Anatolian, see Szemerenyi 1996: 184; Kloekhorst 2008: 216; and especially Yakubovich 2008.

3.3. A commonly cited morphological feature found in Armenian, Greek and Indo-Iranian (and perhaps also Celtic) is the instrumental marker *-bhi(s). Furthermore, Greek and Armenian share the use of *-bH- as the instrumental singular marker, probably due to extension of the athematic instrumental plural marker *-bHs that is also shared by Indo-Iranian. After a lengthy discussion, however, Clackson (1994: 68-74, 87) concludes that the two languages are likely to have made independent developments and denies the significance of this isogloss. He does admit the importance of this feature, however, for the dialect group Armeno-Graeco-Indo-Iranian.16

3.4. *mehi prohibitive particle: Arm. mi, Skt. ma, Av. ma, Gr. Alb. mo.17 The Armenian prohibitive particle mi is probably reflected in Urartian me(i).18 The value of this isogloss is uncertain in view of Toch. AB ma 'not, no', which expresses both simple negation and prohibition (Adams 1999: 445-446).

3.4. *h2oiu-kwi(d): Arm. oc' 'not', Gr. ovx, ovkL 'not'. However, an inner-Armenian development is not excluded.19

3.5. *-nu-presents are attested in Armenian, Greek and Indo-Iranian in a number of verbs that lack them outside this area: *hir-nu-: Arm. arnum 'to gain, obtain, take' (Armenian, Greek, and probably Iranian, see § 4.1.9); *ues-nu-: Arm. z-genum 'to put on clothes' (Armenian and Greek, see § 6.1.16); *gwher-nu-: Arm. jernum 'to be/become warm, burn' (Armenian and Indic, see § 5.2.13).20

As an example of the -nu-extension on Armenian grounds, note Arm. lnum, 3sg.aor. e-lic' 'to fill, be filled' from QIE *pleh1-: Gr. nL^nArj^i, -a^.ai 'to fill, make full', uAewq, Ion. nAeoc, 'full', Lat. plere 'to fill', Skt. par' 'to fill', pres. *piprati, etc. (cf. Arm. li 'full, abundant, whole' and lir, i-stem 'plenitude' vs. Gr. nArprjC 'full; in full'). The aorist e-li-c' derives from *e-ple-ske, with *-ske/o- added to the old root aorist *ple-(s)-, cf. Ved. apras, Gr. enArjoe, etc.

3.6. The *ni- preverb in Armenian and Indo-Iranian.

*ni-si-sd-e/o-: Arm. nstim, 3sg.aor. nst-a-w, impv. nist 'to sit' < *nihist-e-; Skt. ni sidati, Av. nishiSaiti, MPers. nisastan 'to sit'. The form is based on the reduplicated present form *si-sd-from PIE *sed- 'to sit': Skt. sidati, Gr. IZ<, Lat. sido, etc. The verbal form *ni-si-sd-e/o- 'to sit' is a significant isogloss shared by Armenian and Indo-Iranian. Other languages only have the de-verbative noun *ni-sd-o-: Lat. nidus m. 'bird's nest, residence', OHG nest 'nest', cf. Arm. nist, o-stem 'seat, site, base; royal residence, capital', Skt. nida- m.n. 'nest, lair, bird's nest', etc.21

3.7. *-n-presents in Armenian (-anem) and Greek (-av<).

*li(n)kw-n- 'to leave': Arm. lk'anem, 3sg.aor. e-lik' 'to leave', Gr. AeLnw, Aiy.nav<x> 'to let, leave'; cf. Skt. rec-, pres. rinakti 'to leave, let, release', Iran. *raic 'to leave, let, abandon', Lat. linquo, liqui 'to leave, quit, forsake; to abandon', OIr. leicid 'leaves'. Arm. 3sg.aor. e-lik' is de-

16 For references and a general discussion of the *-bhi- ending, see Meillet 1896: 153; Pedersen 1924: 223 = 1982: 306; Gamkrelidze / Ivanov 1984, 1: 379-382 = 1995: 332-335; Kortlandt 1984: 101-102; 2010: 40, 44-45; Schmidt 1987: 40; Martirosyan 2010: 751; Beekes 2011: 30-31, 187-189.

17 Martirosyan 2010: 468-469. For Albanian mo, see Demiraj 1997: 275-276.

18 Jahukyan 1963: 124; Arutjunjan 2001: 454b; Yakubovich 2010.

19 For references and a critical discussion, see HAB 3: 561-562; Clackson 1994: 158; 2004-05: 155-156; Martirosyan 2010: 531. The most recent treatment of this correspondence is found in de Lamberterie 2013: 21.

20 See Clackson 1994: 83-84, 178-180 and Martirosyan 2010 s.vv., also Schmidt 1988: 601; Fortson 2010: 97, 214. For an extensive discussion on nw-verbs I refer to Kocharov 2008: 39-40, 126-155, 182-185.

21 See de Lamberterie 1986: 49-57 and Martirosyan 2010: 505-506 with lit.

rived from thematic aorist *e-likw-e-t, cf. Gr. eArne, and the imperative lik' reflects IE *likwe, cf. Gr. Aim. PIE nasal-infixed present *li-n-kw- was remodelled to *li(n)kw-n-: Gr. Aiynavw and Arm. lk'anem (cf. *bheg- 'to break', nasal present *bh-n-eg-: Arm. bekanem, 3sg.aor. e-bek 'to break', Skt. bhanj-, bhanakti 'to break, shatter', OIr. bongid, -boing 'breaks', etc.). I agree with the view22 that this is likely to be a shared innovation (pace Clackson 1994: 84-85).

This type of presents became productive in Armenian, cf. e.g. *prk-ske/o- (sk-present): Arm. harc'anem, 3sg.aor. e-harc' 'to ask, question, inquire', Ved. prcchami, MPers. pursidan 'to ask', Lat. posco 'to ask, demand', etc. Arm. 3sg.aor. e-harc' derives from thematic imperfect *e-prk-sk-et, cf. Skt. aprcchat. Note also Arm. imper. harc' vs. Skt. prccha.

3.8. The *-n-presents (see the previous paragraph) and a few other Graeco-Armenian iso-glosses are treated by Clackson (1994: 74-87) as ambiguous with respect to the question of whether they represent shared innovations or independent developments: the suffix *-ola- in Greek -6Ar\g (e.g. yaivdArig 'raving, frenzied') vs. the Armenian quasi-participles in -ol, the usage of the PIE verbal suffix *-sk- (Greek -ok- in Ionic iteratives and -c'- in the Armenian aorist) with restriction to past time, peculiar verbal reduplication seen e.g. Gr. 5ai5aAAw 'to embellish' and Arm. cicalim 'to laugh', etc. Naturally, one should welcome such a sound and cautious approach. However, the cumulative strength of these morphological (and a few phonological) features and a great number of such lexical agreements gives additional weight to the evidence.

4. Lexical isoglosses between Armenian, Greek and Indo-Iranian

4.1. Material.

4.1.1. *plhiu- 'Pleiades': Arm. alaw(s)unk' 'Pleiades', YAv. acc.pl. f paoiriiaeiniias < *paruiiaini-, NPers. parvin 'Pleiades', Gr. nAeiaSeg 'id.'; cf. Arm. y-olov 'many' < *polhius: Gr. noAvg 'many', Skt. puru- 'much, abundant', etc. (Martirosyan 2010: 12-13).

4.1.2. *polio-/*polieh2: PArm. *(p)oliya- > Arm. ali-k'i, obl. ale-a- 'wave'; ali-k'2 obl. ale-a- 'grey hair; old age'; Gr. noAidg, fem. noAiag 'whitish grey (of hair and of foaming seas)' (cf. especially noAiaL 'grey hair' which stands for Arm. alik' e.g. in Proverbs 20.29); MPers. pir 'old, aged' < *parya-, Kurd. pel 'wave, billow', etc. 23 In view of Mycenaean po-ri-wa, the Greek word has been reconstructed as *uoAlyo- and its close connection with Arm. ali-k' has been doubted (Clackson 1994: 163-164). Beekes (2010, 2: 1219), however, notes that the appurtenance of the Mycenaean word is quite uncertain and prefers to reconstruct *polio-.

The Armenian, Greek and Iranian (if *parya- is reliable) words are particularly close to each other in having both meanings ('wave' and 'grey hair, old') and reflecting *polio-. Perhaps we can also add Skt. palita- 'grey, grey of old age, aged', though this is uncertain.24 Other languages have *poluo- 'pale, grey', which seems to be unrelated: Lat. pallidus 'pale' < *palwo-, OHG falo 'faded' < PGerm. falwa-, OCS plavb 'white', etc.25

4.1.3. *h2(e)lhi-/*h2\-n(e)hi-: Arm. alam 'to grind' < *al-n-, Gr. aAew 'to grind', MInd. ata 'flour', Av. asa- 'ground' < *arta-, MPers. ard 'flour' < *arta-, Khot. arr- and Sogd. 'rn 'to grind' from Iran. *arna-. See also § 4.1.4.

22 Hamp 1975; Wyatt 1982: 29; Stempel 2000: 517. For an extensive discussion I refer to Kocharov 2008: 34-39, 73-101, 172-180.

23 Martirosyan 2010: 14-15.

24 See Mayrhofer EWAia 2, 1996: 103-104.

25 See de Vaan 2008: 440-441; Derksen 2008: 412.

4.1.4. *h2(e)lhi-tr-i-: Arm. alawri, ea-stem 'mill; female grinder (of corn)', Gr. àAexpiç 'woman who grinds corn'. If Pers. as, asya 'mill', Sogd. 'r5 'mill' and other Iranian forms reflect *a/ar6ra- 'mill', a similar *-tr-formation of *fe(e)lhi- 'to grind' (see § 4.1.3), then this is a lexical isogloss between Armenian, Greek and Indo-Iranian, as is the root *fe(e)lhi-. Note also Arm. alewr 'flour' and Gr. âAevpov 'flour' (§ 6.1.1).26

4.1.5. *hilôpe/ëk- / *h2le/oupek- 'fox': Arm. aluës, gen. aiues-u 'fox', Gr. àAùnri^, -exoç 'fox', Skt. lopasâ- probably 'fox', Proto-Iranian *raupasa- 'fox' (Parth. rwb's [robas], MPers. robah 'fox', Oss. ruvas/robas 'fox', Sogd. rwps-, Khwar. rwbs 'fox', Khot. rruvasa- 'jackal', etc.). Despite the vocalic problem, I agree with Mayrhofer EWAia 2, 1996: 482 in that Indo-Iran. *Raupaca- is "nicht zu trennen" from Arm. aiuës and Gr. àAœnrj^. Further, note Av. urupi- 'dog', raopi- 'fox, jackal', Celtic *lop-erno- (Welsh llewyrn 'fox', Bret. louarn 'fox', etc.), Lat. volpës f. 'fox', Lith. lapé, Latv. lapse 'fox', etc.27

4.1.6. *h2(e)ig- 'goat': Arm. ayc, i-stem, 'goat', ayc-i, obl. aycea- (probably from fem. in *-ih2-), Gr. aïE„ alyôç f. 'goat' (compositional alyi-), YAv. îzaëna- 'leathern', perhaps also Skt. eda- m. 'a kind of sheep' and Alb. dhi f. '(she-)goat'; note also the i-less form: Skt. ajâ- 'goat', YAv. aza- 'goat', Lith. ozys 'goat', etc.28

4.1.7. *h2nër, gen. *hinr-ôs 'man': Armenian ayr, gen. arn, Skt. nâr-, Av. nar-, Greek àvrjp, gen. àvSpôç, Phryg. avap, cf. also Alb. njeri 'human being, person'; note also traces in Italic (Osc. ner-) and Celtic (Mir. ner 'boar', MWelsh ner 'chief, hero'). For the areal distribution, compare, e.g., gen. *-osio- and *h2erhs-uer/n- 'arable land'.

4.1.8. PIE *hirs-en- 'male, male animal': Arm. arn 'wild ram' (acc.pl. z-arin-s), Gr. apoqv, -evoç, Att. âppqv, Ion., Lesb., Cret. ëpoqv, Lac. âpoqç adj. 'male', Av. arsan- m. 'man, male', OPers. arsan- 'male, hero, bull', cf. Skt. rsabhâ- m. 'bull'.

In view of the vocalic discrepancy in the Greek forms eporjv and apor\v, two different roots may be posited: *hirs-en- (with Arm. arn and Indo-Iran. *Hrsan-) and *hiufsen- (with Skt. vrsan- 'manly; male animal, bull, stallion, etc.', Lat. verres 'boar', Lith. versis 'bull, ox, ox calf', etc.), respectively.29 According to Pronk (2010), the second part of the Proto-Indo-European determinative compound *gw(e)h3u-ursën 'bull', lit. 'cow-male' (Toc. A kayurs 'bull', B kaurse 'bull', OIc. kursi, later kussi 'bull calf', Skt. gô-vrsa- and gô-vrsabha- 'bull', etc.), was reanalyzed in Greek, Indo-Iranian and Armenian as *-rsen and started to lead an independent life.30 Whether one accepts this attractive scenario or not, we are nonetheless dealing with a lexical isogloss between these three branches (pace Pronk 2010: 176i4). Note the abundance of such isoglosses in the domain of animal husbandry (see Table set A).

4.1.9. *hir-nu-: Arm. arnum 'to gain, obtain, win, take, grasp', Gr. àpvv^at, aor. àpô^.ev 'to win, gain', probably also Av. drdnauu- 'to grant, allot, provide' (see § 3.5 on nu-verbs).

4.1.10. *sru-ti/to-: Arm. aru, i-stem, o-stem, a-stem 'brook, tributary; channel, ditch, trench'; the threefold declension of the Armenian word points to different derivatives: *sru-ti- (cf. Skt. sru-ti- f. 'way, path', Gr. pvoiç f. 'flowing, flow', etc.), *sru-to- (cf. Gr. pvzôç 'flowing') or *sroutos- n. (cf. Skt. srôtas- n. 'stream, current', OPers. rautah- n., Pahl., NPers. rod 'stream'),

26 See Bailey 1979: 22a; Clackson 1994: 90-95; EtimSlovIranJaz 1, 2000: 200-204; Beekes 2010, 1: 65; Martiro-syan 2010: 13-14, 26-27, 31.

27 Clackson 1994: 95-96; Martirosyan 2010: 42; Beekes 2010, 1: 78-79.

28 Euler 1979: 167-168; Clackson 1994: 88-90; Martirosyan 2010: 58; Beekes 2010, 1: 40-41; cf. also Clackson 1994: 182, 2376/4.

29 For references and a discussion, see Martirosyan 2010: 112.

30 Note that, in the Atharva-Veda, Skt. rsabha- is usually a real male animal, whereas vrsabha- is generally used symbolically, often referring to, e.g., Indra or Agni (Lubotsky apud Pronk 2010: 172, 175-176).

*sr(o)u-ieki (cf. Lith. srauja, Latv. strauja 'stream', Russ. strujá 'stream', etc.). The forms derive from PIE *sre/ou- 'to stream, flow': Arm. a/oiog(an)em 'to water, irrigate', Gr. pé< 'to flow, stream', etc.

4.1.11-12. *kirgipió-: arcui 'eagle' and *tkiH-(i)no-/*tkiH-eno-: c'in 'kite' (§ 4.2).

4.1.13. *urkién, gen. *urkino-: Arm. gain, in/an-stem: gen. garin, instr. garam-b, nom.pl. garin-k', gen.dat.pl. garan-c' 'lamb', Skt. úran-, nom. úra, acc. úranam m. 'lamb', NPers. barra 'lamb' < PIr. *varn-aka-, Gr. áp¡v m., ^ap¡v 'lamb', noÁv-ppr¡v-eg 'possessing many lambs' < *-urki-n-.

4.1.14. *dhmbh-: Arm. damban, dambaran 'tomb, grave'; Gr. za<p¡ f. 'interment', iá<og m. 'funeral rites; grave, tomb', %á<pog f. 'ditch, trench', §ánx<x> 'to bury' from *dhmbh-io. Probably here also belongs YAv. daxma- 'grave' (dissimilated from *dafma- < *dhmbh-mo-). The appurtenance of Old Pruss. dambo 'ground' is uncertain. PArm. *damb(a)r- 'tomb' (< *dhambh-ro-/-reki-, cf. Gr. %á<pog) may have been borrowed into Abkhaz a-damra 'tomb, grave, dolmen'. Note also Arm. t'umb 'mound; fence, wall around a house' and Gr. zvy.¡og m. 'mound, burial mound, grave' (see § 6.1.18). In view of the aberrant vocalism comparable to burgn and durgn, as well as Arm. t'- instead of d-, here we may be dealing with a substrate intermediation.31

4.1.15. *kiegwh-i-: Arm. iz, i-stem 'viper', Gr. éxiQ, -£<g, gen. exiog 'viper; name of a monster', Skt. áki- m. 'snake, adder', YAv. azi- m. 'snake, dragon'. The assibilation *-gwh¡- > *-yzy-and the problem of the Armenian vocalism are due to a generalization of the genitive *efyo-from gen. *ki(e)gwh-i-ós (cf. Gr. gen. extog). If the Indo-Iranian forms belong rather to PIE *ki(e)ngwh-i- (Arm. awj, i-stem 'snake', Lat. anguis m.f. 'snake', Lith. angis f. 'snake'), then we are left with a correspondence between Armenian and Greek.

4.1.16. *gerH-: Arm. cer, o-stem 'old man; old', cer-anam 'to become old', Skt. jar 'to age, grow old', jarás- f. 'old age', YAv. zar- 'to age, grow old', yépag n. 'gift of honour' (originally 'old age'); *gerH-ont-: Arm. cer-un(-i) (ea-stem) 'old', Skt. járant- 'old', Oss. zxrond 'old', Gr. yépwv 'old man'. A different formation: OIc. karl 'old man', OHG karal 'old man', etc.32

4.1.17. *gwou-io- (or *gwk3eu-io-): Arm. kogi, gen. kogw-o-y, ins. kogw-o-v 'butter', Skt. gávya-, gavyá- 'consisting of cattle, coming from or belonging to a cow (as milk, curds, etc.)', YAv. gaoiia- 'coming from cattle, consisting of cattle', Gr. adj. -¡¡o(p)iog, e.g. évveá-¡otog 'worth nine beeves'. This isogloss33 is based on the PIE word for 'cow' (Arm. kov; cf. nom. arew vs. oblique areg- 'sun'). see author's addition on p. 137.

4.1.18. *k2erk3-uer/n-: Arm. harawun-k' (acc.pl. karawun-s) 'sowing, seeds; sowing-field; arable land', Gr. apovpa f. 'tilled or arable land; pl. corn-lands, fields', Skt. urvára- f. 'arable land, field yielding crop', Av. uruuara- f. pl. 'food plant, plant, ground covered with plants, flora'. As in cases of e.g. gen. *-osio- and *k2ner- 'man', Celtic and Italic are added: MIr. arbor, NPl arbanna, OIr. gen. arbe 'grain, corn', Lat. arvum 'ploughed land'. Armenian, Greek and Indo-Iranian are unified by the *-r/n- heteroclitic declination (seen also in Celtic) and the semantics. If the original meaning was 'grain, crop' (cf. Iranian and Celtic), we might treat the semantic shift as an innovation. However, the *-r/n-declension is rather archaic. The value of this iso-gloss is uncertain.

4.1.19. *mrto-: Arm. mard, o-stem 'man, human being' (renders Gr. av$p<nog or ¡pozóg in the Bible), Gr. ¡pozóg m., f. '(mortal) man; mortal'; Skt. mrtá- 'died, dead'(verbal adj.), Av. mardta- 'dead'; cf. privative *n-mrto-: Skt. amfta- 'immortal', YAv. amasa- 'immortal', Gr. ay.¡pozog 'immortal, divine'; with different vocalism: Skt. márta- m., Av. marata- m. 'the mortal

31 For a discussion, see Clackson 1994: 120-121; Martirosyan 2010: 232-233; Beekes 2010, 1: 534, 1517-1518.

32 Mallory/Adams 1997: 409-410; Olsen 1999: 611; Martirosyan 2010: 339; Beekes 2010, 1: 268-269.

33 See already Meillet 1896: 152.

one, man', Gr. popxÖQ^ avOpwnoQ, OvqzÖQ 'man' (Hesychius); the other cognates continue a form in *-tu-o-: Lat. mortuus, OCS mrbtvb 'dead', etc.34 Armenian, Greek and Indo-Iranian agree in *-to- as well as the semantic shift from 'dead' to 'mortal' and therefore represent a significant isogloss (cf. Meillet 1896: 151).

4.1.20. *kie/ou-: Arm. c'uem 'to go, set forth, march off, break camp', c'og-, suppletive aorist of ert'am 'to go, set off'; Skt. cyav- '(to start) to move, stir; to undertake'; OAv. sauuaite 'to move', YAv. fra-süiti- f. 'approach', OPers. siyav- 'to set forth, go, march'; Gr. oevo^ai 'to be in violent motion, hurry; to walk, rush (to)'; participle *kiu-to- 'moved': Arm. c'u, o-stem 'setting out, departure; campaign, expedition; journey', Skt. cyuta- 'moved (wankend, in Bewegung geraten)', YAv. mainiiu.süta- 'vom Geist angetrieben', Gr. euL-oovtoq 'rushing, gushing'. This isogloss is based on PIE *kei(hi)-, cf. Gr. k'lo, klvem 'to set in movement, drive away, shake', Lat. ciere 'to move, stir up', citus 'fast'.35

4.1.21. *gwher-os- 'warmth', PIE s-stem neuter: Arm. jer, o-stem 'warmth; warm and bright weather; warm', Skt. haras- n. prob. 'flame, glow', Gr. SepoQ n. 'summer; harvest'.36

4.1.22. *(p)ste/en(-o)-: Arm. stin, gen. stean 'breast of a woman'; Skt. stana- m. 'breast of a woman, mother's breast, nipple'; YAv. fstäna- m 'breast of a woman', MPers., NPers. pestän 'breast'; Gr. oxfviov- ot-Soq (Hesychius); probably also Toch. A pässäm, B päscane dual 'woman's breasts' < PIE thematic dual *pstenö. The other cognates have an initial *sp-, cf. Lith. spenys 'nipple', OIc. speni 'teat, nipple', etc.37

4.1.23 *k(e)rH- 'to tie, attach, bind': Arm. sarem 'to form, make; to equip, prepare; to stretch; to weave, etc.' (Middle Armenian and a number of non-contiguous dialects); sard, i-stem 'spider' (Bible+; dial.) from *kr(H)-ti-; Iranian *sar- 'to tie, attach, link': OAv. sär- 'to mix, unite with', Parth. sar 'community' (only in pd ... sr 'together with'),38 Pashto sard adv. 'together', etc.

*k(e)r(H)-ieh2 'band': Arm. sari-k', ea-stem 'chain, fetters, bands' (5th century onwards); Gr. KaipLa 'tape or cord used for ligatures', KeipLa f. 'girth of a bedstead; swathing-band, bandage', Kaipooewv (Homer) 'close-woven', Kaipö« 'tie the Kalpoi onto the loom'.39

The Iranian verb is usually derived from IE *kerh2- 'to mix, tie': Skt. a-sirta- 'mixed (with milk)'; Gr. Kepavvvpi, aor. Kepao(o)ai 'to mix, mix up (especially of wine with water); to temper (of the climate)'. However, this is uncertain, as is the appurtenance of Skt. srnkhalä-, srnkhala- 'chain, fetter'.40

4.1.24. *(s)peud- 'zeal, haste': Arm. p'oyt', o-stem (also i-stem) 'zeal, diligence; haste; zealous, diligent; hastily', p'ut'am 'to hasten, hurry, strive'; Gr. onov8-f f. 'haste, zeal', onevSw 'to hasten, hurry, strive'; MPers., NPers. pöy- 'to run', ManParth. pwd- 'to hasten'. The problem of Arm. -t' can be solved by positing *(s)peud-to- > *phoy(t)tho-/41 Beekes (2010, 2: 1381-1382) notes

34 Clackson 1994: 237^ Mayrhofer EWAia 2, 1996: 318, 327; de Vaan 2008: 389-390; Derksen 2008: 342; Marti-rosyan 2010: 452-453; Beekes 2010, 1: 242-243. It has been assumed that Lat. (im)mortälis might be based on earlier *morto- rather than mors 'death' (de Vaan 2008: 390; cf. Euler 1979: 125), but this is not compelling.

35 Schrijver 1991: 237-238; Mayrhofer EWAia 1, 1992: 552-553; Olsen 1999: 41; de Vaan 2008: 113-114; Martiro-syan 2010: 547-548; Beekes 2010, 1: 700, 707; 2: 1321-1322. On the pure velar *k- in *kei-, see Beekes 2011: 126.

36 On the isogloss, apart from the standard literature, see Euler 1979: 224; Wyatt 1982: 31-32; Schmidt 1987: 37; de Lamberterie 2013: 19-20; cf. Hamp 1992: 57-58. For the Armenian etymon, see Martirosyan 2010: 556-557.

37 See Euler 1979: 33-34; Stempel 1990: 52; Olsen 1999: 135-136; Martirosyan 2010: 584-585.

38 Durkin-Meisterernst 2004: 308b.

39 HAB 4: 183-184, 186, 187-188; Jahukyan 2010: 670-672 (mentioning only the Greek cognates).

40 For the forms and a discussion, see Schwartz 1986: 359-360; Mayrhofer EWAia 1, 1992: 178; Clackson 1994: 139-140; de Vaan 2003: 99-100; Cheung 2007: 337; Martirosyan 2010: 573-574; Beekes 2010, 1: 617, 664, 675.

41 Klingenschmitt 1982: 167; Clackson 1994: 155; Martirosyan 2010: 652 (with more references).

Lith. spausti 'to press, squeeze; to push, drive on; (intr.) to hurry' as the only certain cognate for the Greek and does not mention the Armenian and Iranian forms. If the Baltic form is indeed related, the isogloss becomes less significant, although the semantic identity of the Armenian, Greek and Iranian cognates is more impressive.

4.2. Pair isoglosses. If two lexemes are contextually related with each other and both represent the same dialect area, the significance of these isoglosses increases. In this section I present two such pairs.

4.2.1. *h2Tgipio-: Arm. arcui, ea-stem: gen.sg. arcu-o-y, gen.pl. arcue-a-c' 'eagle'; Skt. rjipya-'epithet of an eagle', m. 'eagle', YAv. drdzifiio.pardna- adj. 'having eagle-feathers', MPers. ''lwf 'eagle' (= phonetically /aluf/), aluh 'eagle', etc.; Gr. aiyvniog m. 'vulture', aiyino^ aezog vno MaKeSovwv, cf. also apyionovg• aezog. MaKeSoveg. The formal difficulties of Gr. aiyvniog (the expected form is *apyi-niog) may be due to folk-etymological association with a'iE, 'goat', ainvg 'high and steep, sheer' and yv^ 'vulture'.42

4.2.2. *tkiH-(i)no- or *tkiH-eno- 'bird of prey': Arm. c'in, o-stem 'kite', Skt. syena- m. 'bird of prey, falcon, eagle', Av. saena- 'a big bird of prey', Gr. iKzivog m. 'kite'.43

Discussion: *h2fgipid- and *tkiH-(i)no-/*tkiH-eno- (4.2.1+2). In RV 4.38.2, etc. the horse Dadhikra- is compared with rjipyam syenam. Vedic rjipya- is an epithet of syena- 'bird of prey, falcon, eagle'. It is remarkable that both *hirgipio- 'epithet of a bird of prey' and *tkiH-(i)no- or *tkiH-eno- 'bird of prey' belong to the Armeno-Graeco-Indo-Iranian dialect group. Within this group we can speak also of the Armeno-Aryan poetic language, notably arew 'sun', erg 'song', ji 'horse', perhaps also surb 'pure, holy' (see §§ 5.2-5.3).

Arm. arcui (gen. arcu-oy) is the principal word for 'eagle', and its derivation from *h2rgipio-in native terms is secure both formally and semantically. The contextual relation with *tkiH(i)no-/*tkiHeno- 'bird of prey' (from which Arm. c'in, o-stem 'kite' certainly derives as a native word) that belongs to the same isogloss area makes the native origin of arcui impeccable. I therefore see no reason for denying a direct derivation of arcui from late Indo-European *h2rgipio- and treating it as an Iranian or Urartian loanword. Urartian arsibi- that is found in a horse-name and has no Hurrian match should be regarded as an Armenian loanword.

Armenian arcui largely functions in poetic association with a swift horse and in figures such as 'eagle-winged' and 'sharp-flying as an eagle'. In the epic fragment on the abduction of the Alan princess Sat'inik by King Artases (Movses Xorenac'i 2.50), the horse of Artases is compared with arcui srat'ew 'sharp-winged eagle'. In Azdahak's dream (Movses Xorenac'i 1.26), the dragon-riding hero was dashing with eagle's wings: arcuoy imn ardarew slac'eal t'ewovk'. In a kafa-poem to the Alexander Romance we find srdnt'ac' arciw 'sharp-riding eagle'.

These figures probably go back to the Armeno-Graeco-Indo-Aryan poetic language, cf. Skt. asu-patva 'swift-flying' as epithet of syena- 'eagle' (cognate with Arm. c'in 'kite'), Gr. <x>kv-nezr\g 'swift-flying' (used of horses and hawks), <KV-nzepog 'swift-winged'; cf. also Av. drdzifiio.pardna- 'eagle-feathered (arrow)', Lat. acci-piter 'hawk', etc.44

4.2.3. *ke/omieh2: Arm. pl. sami-k', gen. samea-c' 'the pair of yoke sticks; rudder'; Skt. samya- 'pin of a yoke, peg, wedge', yuga-samya- n. 'yoke and yoke-pin'; Av. sima- f. 'yoke-pin' (Yast 10.125, perhaps for *sdma-), dual yuii(uu)o.sdmi- '(having) yoke and yoke-pin' (Videv-

42 See Pedersen 1924: 224 = 1982: 307; Schmitt 1970: 66-6717; de Lamberterie 1978: 251-262; Euler 1979: 88-89; Meier-Brügger 1995; Watkins 1995: 170-172; Balles 1997: 148-150; Beekes 2010, 1: 33, 126; Martirosyan 2010: 139-141.

43 Clackson 1994: 45-46, 143-144; Martirosyan 2010: 627.

44 Watkins 1995: 170-172, 252-253. For references and a thorough etymological discussion of Arm. arcui, see Martirosyan 2010: 139-141.

dad 14.10) for *yuuo.sami- from Indo-Iran. *iuga-cam-i;45 Western Iranian: Takistani same, As-tiyani sama 'yoke-peg'; Sughni, Bajui sim-Sorg (with Sorg 'wood, stick' < Iran. *daruka-), Khufi sim 'peg for fastening yoke to bullock's neck'.46 Outside of Indo-Iranian, note Gr. Kaya^, -aKog f. m. 'pole, shaft; pole to support the vine; shaft of a spear; tent pole', MHG hamel 'shaft, pole', etc.

4.2.4. *dehi- 'to bind': Arm. *ti- 'tie, bond' in *sami-a-ti > sameti-k' and sametay(-k'/n) 'the tie of sami, yoke band', Gr. Sew 'to bind', Skt. da-/dyati 'to bind', Av. da- 'to bind'.47 This etymon is restricted to Armenian, Greek, and Indo-Iranian, possibly also Hittite, tiie/a-zi 'to bind?' (cf. Skt. dyati), tiiamar / tiiaman- n. 'cord, string'48 and Alb. duaj 'sheaf'.49

To the best of my knowledge, the Armenian by-form *tay- 'bond' (sametay-k'; dial. *sametay-n, *samotay; other dial. compounds: *beran-tay, *bn-a-tay, *vz-tay, etc.) has not yet received an explanation. I propose to derive it from *dhi-ti-: Gr. Seoig 'binding, joint', and Skt. -diti- 'Gebundenheit, Fesselung' (in a-ditih 'boundlessnes').

Discussion: *ke/omieh2 and *dehi- (4.2.3+4). Armenian sami-k', gen. samea-c' 'the pair of yoke sticks; rudder' (Severian of Gabala, John Chrysostom, Grigor Narekac'i, etc.; preserved in a number of dialects) is mostly attested in a compound with *ti/tay 'tie, band' which is represented in several forms: sameti-k', ins.pl. sameteawk' (Sirach 28.23-24, 30.27);50 samete-k', acc. samete-s, ins. sameteiw-k' (Jeremiah 5.5, Severian of Gabala, John Chrysostom, etc.);iii samet, i-stem (ins. pl. samet-i-w-k' [var. lect. sameteawk', sameteiwk', etc.] in Job 39.10, see Cox 2006: 251); sametay-k' (Commentary on Jeremiah by Mxit'ar Gos, 12th cent.); samotik' (Grigor Narekac'i, Oskip'orik); sameten-k' (Grigor Tat'ewac'i), all meaning 'the tie of sami, yoke band'. The compound corresponds to Gr. Seoyog 'band, fetter' or iyag 'leathern strap or thong' in the Bible translation.

The component *tay in sametay, albeit attested in a Middle Armenian source only, seems to be reliable and old since it is confirmed by data from both western and eastern dialects. In a folk incantation against the Evil Eye from the Javaxk' region one finds samota < *sam(w)oy-tay (Lalayeanc' 1892: 13a). Identical to this are samoda and somat'a found in ritual songs of Palm Sunday in Basen and Javaxk' respectively (Grigoryan 1970: 323). In Xotorjur, a dialect that is both geographically and linguistically close to the Karin/Erzrum group, to which Basen and Javaxk' belong too, one finds samotek' (YusamXotorJ 1964: 506b), obviously from *samotay-k'. In these forms the first component comes from sam(w)oy, the genitive singular of sami. It is also found in samotik' (Grigor Narekac'i and Middle Armenian). As for the eastern dialects, we find tarabal, Hadrut', etc. sambetan and tazax sametan,52 which presuppose *sametay-n. The widespread form sameten may also be derived from *sametayn, with a common development ay > e.

45 For the Indo-Iranian forms, see Mayrhofer EWAia 2, 1996: 412-413, 613; Skj^rvo 1997: 119-121; de Vaan 2003: 470.

46 Morgenstierne 1962: 207; 1974: 31b, 73b.

47 Bugge 1893: 25; Hübschmann 1897: 488; HAB 4: 403-404; Mayrhofer EWAia 1, 63, 1992: 716-717; for the Indo-Iranian forms, see also Mayrhofer KEWA 1, 1956: 29, 547; 2, 1963: 69; EtimSlovIranJaz 2, 2003: 444-445; Cheung 2007: 47).

48 See Kimball 1998: 338; Melchert apud Kloekhorst 2008: 880-881; Beekes 2010, 1: 321-322.

49 Frisk GEW 1, 1960: 374-375; Mayrhofer KEWA 2, 1963: 69; Demiraj 1997: 128, 149-150.

50 In the Sebastac'i Bible, we find slt'ay 'chain' instead of sametik'.

51 The form samete (ins. sameteiw-k') vs. sameti is reminiscent of the puzzling auslaut of aste / aste (ins.pl. asteiwk') 'spear' from Iran. *arsti- 'spear', cf. OPers. and Av. arsti-, etc. (HAB 1: 221a; Olsen 1999: 865). If the -e proves to be original, one might think of a possibility that IE *dh1-ti- yielded *tey- > *te, with a development *-h1- > Arm. -e- between consonants. More probably, however, -e was taken from obl. aste- and samete-.

52 Davt'yan 1966: 467; HayLezBrbBar 5, 2008: 259b.

In dialects we also find compounds with beran 'mouth' (*beran-tay), bun 'trunk, shaft' (*bn-a-tay), or 'buttocks' (*or-ti-k' and *or-tay-n), viz 'neck' (*vz-tay), etc.

Since Bugge,53 Armenian sami-k' has been interpreted as an inherited word. Some scholars are inclined toward an Iranian origin of the Armenian word.54 However, there is no compelling reason for this. There are no Iranian forms that would be formally and semantically compatible with the Armenian word as a source of borrowing.55 Note that the second component of the compound sameti is not attested independently, and this is another (albeit not decisive) indication that sami is archaic.

It is especially important that both sami and *ti/tay belong to the same dialect area, namely Armeno-(Graeco-)Indo-Iranian. This situation is reminiscent of another case, PArm. *andi-'doorframe, threshold', that has been preserved only in the compound dr-and-i (ea-stem) and can be derived from *h2(e)nHt-ieh2-, which is also to be regarded in terms of an interchange between feminine suffixes, cf. YAv. qiSiia- f.pl. 'door-post' vs. Av. a$a- 'house', Skt. ata- f.pl. 'door-frame, door-posts' and Lat. antae f.pl. 'square pilasters, wall posts of a temple'.

Taking into account all that has been said above, I am inclined to treat Arm. sami-k' / samea- 'the pair of yoke sticks' as a native match of Skt. samya- 'pin of a yoke'. Theoretically, the Armenian form may be derived from *samiya- < *komieh2.

5. Lexical isoglosses between Armenian and Indo-Iranian

5.1. Armenian and Indo-Iranian.

5.1.1. *h2enHt-i(e)h2-: Arm. (dr-)and, i-stem, (dr-)and-i-, ea-stem 'door-frame, threshold, vestibule' (perhaps also 'house, estate', cf. dial. *andiwor 'family'), (h)and, i-stem and o-stem 'cornfield, arable field, pastureland'; YAv. qiSiia- f. pl. 'door-post', a&ahuua 'house' (loc.pl. of a$a- 'house', with extension of 'doorposts' to 'house', cf. Arm. *and-i- 'house', 'cornfield'); further: Skt. ata- f. pl. 'door-frame, door-posts', Lat. antae f. pl. 'square pilasters, wall posts of a temple', OIc. Qnd f 'front room, corridor'.56

5.1.2. *n-bhudhno- 'bottomless': Arm. andund-k', o-stem, Skt. a-budhna- 'bottomless', MPers. a-bun 'baseless, bottomless' (compare Skt. budhna- 'bottom, depth, the root of a tree', Gr. nv^^qv, -evoc 'bottom, depth, base', Lat. fundus 'bottom', OHG bodam, etc.).

The close relationship between the Armenian and Aryan words is also seen in the mythological context: Arm. Andndayin awj 'the Abyssal Serpent' (in an incantation against the snake and scorpion); a black serpent (sev oj) at Andndayin car 'the Abyssal Tree' (in an incantation from the Akn area); Andndayin t'agavor "Abyssal King" in a New-Year's ritual formula related to a spring in Kamarkap, a village in the same area of Akn.57 Compare the Rigvedic primordial Serpent of the Depth, Ahi- Budhnyd-, whose origin and abode is the dark bottom of the waters, as well as the Cosmic tree in the bottomless (a-budhna-) abyss.58

53 Bugge 1893: 24-25; Hubschmann 1897: 488; HAB 4: 167; Jahukyan 2010: 665b.

54 Benveniste 1964: 2; Olsen 1999: 906; Mayrhofer KEWA 3: 302 (not mentioned in EWAia 2, 1996: 613); hesitantly: Pokorny 1959: 556. The etymon is absent in Gamkrelidze/Ivanov 1984 and Mallory/Adams 1997. Beekes (2010, 1: 629) is sceptical about the connection.

55 Jahukyan 1987: 174-175, 551; Hovhannisyan 1990: 213-215; cf. also Schmitt 1983: 108; 1987: 446b. Jahukyan (2010: 665b) is quite positive about the native origin of the word.

56 Martirosyan 2010: 72-77.

57 Canikean 1895: 48-49, 167; K'ec'ean/ Parsamean 1952: 78; Lanalanyan 1969: 110Nr304; S. Harut'yunyan 2006: 91a, 152a.

58 Martirosyan prepar 2.

5.1.3. *bhekig-\ Arm. bak, a-stem 'courtyard; sheep pen; sun or moon halo' (this tentative etymology implies that the original meaning of the Armenian word was 'landed allotment, encircled estate'); Indo-Iranian *bhag-â-: Skt. bkagâ- m. 'part, portion, share, lot, destiny' (RV+), OAv. baga- 'part', Aram.-Iranian *baga- 'landed property, estate, fief', Sogd. ß'y 'garden', MPers. baw from bay 'garden, orchard', NPers. bay 'garden, orchard', Tadj. boy, Baluchi bag 'garden', etc.; cf. Skt. bkâga- m. 'prosperity, good fortune, property, personified distribution' (RV+), bkaj (pres. bkâjati, aor. âbkaksi, perf. babkaja, etc.) 'to share, partake, divide, distribute, apportion; to receive, enjoy' (RV+), bkakti- f. 'distribution, apportioning' (RV+), OAv. baxsta 'genießt, hat Anteil', YAv. baya- m. 'god, distribution', baxta- 'allotted; allotment', OPers. baga-m. 'god', bafi- m. 'tribute, tax', MPers. baf 'tribute, tax', bay [bg] 'god, lord', baxs 'allotment, grant', baxt 'luck, fate, fortune', Parth. baxs- 'to divide, distribute, bestow', baxtag 'divided'. Further: Gr. (ayeïv 'to eat, consume, swallow' (< *'to enjoy, share').59

The etymon may broadly be ascribed to the Armeno-Graeco-Indo-Iranian dialectal area. Arm. bak 'courtyard, pen, circle, halo' matches the Indo-Iranian noun *bhagâ- from *^ekigo-both formally and semantically. The semantic specification 'portion, share, allotment' > 'landed allotment' is also seen in Iranian languages (Aram.-Iranian *baga- 'landed property, estate, fief', Sogd. ß'y 'garden', MPers. bay 'garden', etc.) and may be due to independent developments. Compare the case of karaw 'south'.

However, the Armenian word has an a-stem instead of the expected o-stem. If the a-stem is old (note that we have no evidence for any declension class from the so-called Golden period), we can posit a feminine or collective *bhekig-(e)ki. Alternatively, we might assume an old Aryan borrowing: *bhagâ- 'portion, share, allotment' > PArm. *bhag-a- > *bak-a- 'landed allotment, encircled estate', with the consonant shift g > k (cf. the well-known case of partez 'garden', which is usually treated as a very old Iranian loan reflecting the devoicing shift d > t).

The basic meaning of the Armenian word thus is 'landed allotment, encircled estate', which easily developed to 'courtyard', 'sheep pen', 'circle', etc. For the semantic fluctuation between 'courtyard, pen' and 'garden, estate', note, e.g., Goth. garda 'Viehhürde', gards 'house, family; court', OHG garto 'garden', OEngl. geard 'enclosure', Engl. yard 'yard', Lat. kortus 'garden; pleasure-grounds', etc. For the semantic shift 'courtyard, pen' > 'halo', cf. Turk. ayal and kutan, both displaying the meanings 'overnight sheep pen' and 'moon halo'. An older example is Hitt. Éhïla- c. 'courtyard; halo (of the moon or the sun)'. We can see that this pattern is widely represented in Asia Minor and adjacent areas.60 (HM)

5.1.4. *k\egh-ik2-: Arm. ezn, gen.sg. ezin, nom.pl. ezin-k', ezan-c' 'bullock, ox'; Skt. akt- f. (vrkï-inflection) 'cow, female of an animal' (RV), Av. azï- (devï-inflection) 'milking (of cows and mares)'; the appurtenance of OIr. ag n. 'cow, cattle' (< *aghes-) is uncertain. Arm. ezn (gen. ezin) may be a frozen accusative in *-iki-m. The gender change is somehow reminiscent of the other important designation of bovids, Arm. kov 'cow' from the PIE generic name for 'bovid'.

5.1.5. *pro-kienk\-o- 'breath, air': Arm. eran 'gentle breeze; winnowing wind; a wind-spirit' (*pro-kân-o- > *e-ra(k)ân-o- with a regular prothesis before #r-); Skt. pranâ- m. 'breath, breathing out, air'; MidIran. *frana- 'air' (cf. Sogd. ßr'n, ßr''n, etc.), *pati-frana- 'ouverture d'aération' > Arm. patukan 'window'. The Indo-Iranian form is composed of PIE *pro- (cf. Skt. prâ 'before,

59 The Tocharian (A pak, B pake 'part, portion, share') and the Slavic (OCS bogatb 'rich', u-bogb, ne-bogb 'poor', Russ. bogatyj 'rich', OCS bog~b, Russ. bog 'god') cognates reflect loanwords from Iranian (see Adams 1999: 363-364 and Derksen 2008: 50).

60 If this etymology of Arm. bak is accepted, we have to treat Georg. bak'i 'hedged pen for cattle; yard; moon halo' and the related forms as armenisms. Georgian-Zan *baga- 'sheep-pen, goat-pen, crib', if related, can have been borrowed from Proto-Armenian *baga- prior to the devoicing consonant shift.

forward, forth, in front', Arm. era- 'first', etc.) and *h2enhi- 'to breathe': Skt. an' 'to breathe', 3sg.pres. aniti, cf. *h2enhi-mo-: Gr. avey.oc m. 'wind', Lat. animus m. 'mind, soul; the element of air (as the principle of life); spirit', anima f. 'breath; soul, life; disembodied spirit, soul, ghost; soul, spirit; air as the substance of wind, an air current, wind, breeze', etc. (HM)

5.1.6. *loukeno-: Arm. lusin, o-stem 'moon', Skt. rocana- n. 'light, luminous sphere, firmament', YAv. raocana- adj. 'shining, light' (see § 5.3.2 for the discussion).

5.1.7. *prHuo-: Arm. haraw, o-stem 'south; southern wind', Skt. purva- 'being before, going in front, first, former; eastern', OAv. pouruuiia- 'first, intial, former', YAv. pauruua-, paouruua-, pouruua- 'being in front, first, former, southern'; OCS pnwb 'first', Toch. B parwe 'earlier; first', Alb. pare 'first', etc.; with a different suffix: Lith. pirmas 'first', Lat. primus 'first'.61 This etymology has been proposed by Jahukyan62 and, with few exceptions,63 has largely remained outside the scope of Indo-European etymological studies. Armenian and Indo-Iranian are unified by *-uo- (this is also found in Slavic and Tocharian, for example) and the meaning 'going in front' with a further shift to a compass direction. The direction ('south') is identical in Armenian and Iranian, but it is difficult to ascertain whether this is due to chance or not.

5.1.8. *h3meigh-o-: Arm. meg, o-stem (also i- or a-stem) 'mist, fog, darkness', Skt. megha- m. 'cloud, gloomy weather', Av. maeya- m. 'cloud', Parth. meg 'cloud, mist'. The other cognates continue *h3migh-leh2: Gr. dy.ixArj 'mist, fog', Lith. migld, OCS mtgla 'fog'. Arm. meg may also be an Iranian loanword. However, this is not compelling. Note the o-stem of the word, as well as the dialectal *mg-l-im 'to cloud' comparable to Dutch dial. miggelen 'staubregnen', etc.64

5.1.9. *pe/ork-u- 'rib, side': Arm. yorsays adv. 'supinely, lying on the back' (John Chrysos-tom, Philo, etc.), yorsayseal 'id.' (Proverbs 6.9;65 yorsayseal ankeal in Canon Law, with ankanim 'to fall down'), yorsaysem 'to cause to lie down; to let fall, overthrow' (Paterica, Grigor Narekaci, etc.); without y-: orsays 'lying on the back' (Paterica), orsayseal 'supine' (Movses Xorenac'i 1.12);66 Skt. parsu- f. 'rib', parsva- n. 'the region of the ribs, side, flank', YAv. parasu- 'rib', Khot. palsua- 'rib; spoke', MPers. pahlug, NPers. pahlu 'side, rib', Oss. fars 'side', etc. The connection of this Indo-Iranian word with OCS pnsi 'chest, bosom', Russ. persi 'breast, bosom', Lith. pirsys 'chest of a horse', etc. is considered uncertain.67 This attractive etymology has been proposed by Jahukyan (1991: 42; 2010: 556a)68 but has remained outside the scope of standard Indo-European etymological studies.

The Armenian word is composed of the prefix y- 'at, in, on' and an otherwise unattested word for 'rib, side', *ors- < *pork-u-. The semantic pattern is widespread in Armenian: t'ekn 'shoulder, back' > t'ikn tal / t'iknel 'to recline', kot 'rib, side' and kotmn 'side' > an-kotnim and dn-kotmanim 'to recline, lie down', krt'-un-k' 'back' > krt'n-il 'to lean, recline', parak 'rib, side' > parakim 'to lie down'.69

61 For the forms, see Mayrhofer EWAia 2, 1996: 157; Demiraj 1997: 311; Adams 1999: 360; Derksen 2008: 430; de Vaan 2008: 488-489; Martirosyan 2010: 393-394.

62 Jahukyan 1973: 20-21; 1986-87: 30; 1987: 143, 186; 2010: 450-451.

63 Clackson 1994: 39; Olsen 1999: 26.

64 Martirosyan 2010: 457-458, 466, 715 (with a hypothetical explanation for the loss of an initial *h3- before *-m- in Armenian); Beekes 2010, 2: 1077.

65 In Proverbs 6.9 (Zöhrapean 1805, 3: 149): Minc'ew yerb yorsayseal kas ov vat, kam erb i k'noy zart'ic'es "How long wilt thou lie (ÖKvripe KaraKeiaai), o sluggard, and when wilt thou awake out of sleep" (Brenton 1851: 793).

66 In Movses Xorenac'i 1.12 (1913=1991: 38L17; transl. Thomson 2006: 86): Ew zdastn arewelic' gogc'es imn ibrew orsayseal "The eastern plain, you might say, was supine".

67 Mayrhofer KEWA 2, 1963: 229, 261; EWAia 2, 1996: 100-101; Derksen 2008: 429.

68 No acceptable etymology is recorded in HAB 3: 408-409.

69 See HAB s.vv. and Martirosyan 2010: 368-369, 376.

Jahukyan does not specify the nature of -ays. I propose to explain -ay as a (collective) suffix seen in a few formations such as ap'-ap'-ay-k' 'rocky and steep place', bat-ay-k' and batbat-ay-k' 'pretext, ostensible reason or accusation'.70 It probably has an adverbial origin, cf. Arm. i ver-a-y 'on, over', Gr. vnep 'over, plenty; beyond; above', vnepa, pl. -at f. 'upper ropes on the sails', Lat. s-uper 'above, on, over', supra 'above, over, on the upper side of' (see Martirosyan 2010: 592-593). We can posit an underlying *orsay(-k') 'the region of the ribs / Rippengegend'.

The "pure" form *(y)orsay has been preserved in the dialect of Nor-Jula: hoisa angnel 'to lie down or recline like a superior' from *yorsay ankan- (cf. yorsayseal ankeal in Canon Law), and *yors-ank- > hoisang tal 'id.'; the development y- > h- is exceptional in this dialect.^

The o-vocalism is found in a number of words in the same semantic field, such as kot 'rib, side', otn 'spine, back', or 'buttocks', p'or 'belly' and k'ov 'side'. An astonishing parallel for y-orsays-eal 'supinely, lying on the back' (from *orsay- 'rib, side') is y-otn-eal 'id.' (from otn 'spine, back').

The final -s points to a frozen accusative-locative plural *ors-ay-s. There are a number of similar frozen formations belonging to the same semantic sphere, such as *beran-s-i-vayr 'lying face downward' > Nor-Jula b'arazver (with beran 'mouth'), *p'or-s-i-vayr '(lying) belly downward' > tarabal p'arsavaer (with p'or 'belly').72

5.1.10. *ke/omieh2: Arm. pl. sami-k', gen. samea-c' 'the pair of yoke sticks; rudder'; Skt. samya- 'pin of a yoke, peg, wedge', Av. sima-, -sami- f. 'yoke-pin'; further: Gr. KayaE,, -aKog f. m. 'pole, shaft', MHG hamel 'shaft, pole', etc. (for a thorough discussion, see § 4.2.3-4).

5.1.11. *kubh-ro- or *(s)kubh-ro-: Arm. surb, o-stem 'pure, clean; holy', Skt. subhra- 'shining, glimmering, beautiful', cf. sobh-/subha- 'to be beautiful; to shine', subh- f. 'beauty, splendour, ornament', etc. Note also Khotanese suraa- 'clean, pure' (for a discussion, see § 5.3.1).

5.2. Armenian and Indo-Aryan.

5.2.1. *sm(H)-eh2- 'year': Arm. am, a-stem 'year, age', Skt. sama- 'year, season'; further: YAv. ham-, OIr. sam 'summer'; note also Arm. amain and OHG sumar 'summer'. The original meaning was 'summer'.73 Arm. am and Skt. sama- agree both in *-a- and semantics.

5.2.2. *h3nebh- 'nave, hub of wheel': Arm. aniw, o-stem (also a-stem) 'wheel; axle of a chariot; wheel as a torture instrument'; Skt. nabhi- f. 'nave, hub of wheel; centre; navel (of the body or the world); origin, relationship, family', nabhya- n. 'nave, hub of wheel'; cf. also YAv. nafa-m. 'navel, origin, blood relationship' (for the semantic relationship cf. Arm. port 'navel', 'tribe, generation'), OPr. nabis 'hub, navel', OHG naba 'hub', nabalo 'navel', Lat. umbilicus m. 'navel; centre, middle' < *h3nbh-, Gr. oycpaAog m. 'navel, umbilical cord'. This isogloss can be considered valid only if Skt. nabhi- indeed reflects *h3nebh-i- (vrddhi-derivation) rather than *h3nobh-i-.74

5.2.3 *s(e)Hdh-io- 'successful' and *s(e)Hdh-ie/o- 'to succeed, be successful': Arm. aj, o-stem 'right', aj-ol and y-aj-ol 'skilful, successful', (y-)aj-ol-em 'to have success'; Skt. sadhya- m. 'a class of divinities', sadhati 'to succeed, reach the goal', siddha-, sidhra- successful', sadhu-'straight, effective', sidhyati 'to succeed, be successful'.75

70 For more examples and a discussion, see Jahukyan 2010: 796.

71 HAB 3: 408-409; Acaiean 1940: 21, 76, 127, 378a.

72 Alternatively, one might treat the final -s as a relic of *si- from PIE *kei- 'to lie' (cf. Gr. xelpai 'to lie', etc.), which is visible in Armenian hiwsis(i) 'north' if indeed from *seukoi-ki(y)o- (see Martirosyan 2010: 412 with ref.). This is less probable, however.

73 The original paradigm has been interpreted as follows: *s(e)m-eh2-, genitive *smh2-os.

74 Ritter 1983; Martirosyan 2010: 89-90.

75 Martirosyan 2010: 99-100.

5.2.4. *h2reu-i-: Arm. arew, u-stem, old gen. areg 'sun; sunlight; life': Areg k'aiak' 'the city of the Sun' (Gr. 'HAiov noAic,, e.g. Genesis 41.45, 50), areg, gen. aregi 'the 8th month', areg 'eastern', areg-akn 'sun', etc.); Skt. ravi- m. 'sun, sun-god' (Upanisad+), ravi-putra- m. 'son of the Sun' (Kathaka-Brahmana); probably derived from a PIE verb that is reflected in Hitt. haru(ua)nae-zi 'to become bright, get light, dawn'. According to Demiraj (1994: 71), Alb. (ve) re 'klar, deutlich machen, sehen' also belongs here.

In view of the -i of Sanskrit ravi-, Arm. arew, u-stem 'sun' and gen. areg < *areg-i- may be interpreted as reflecting an old HD i-stem: nom. *hireu-oi > PArm. *arew-u(y), gen. *h2r(e)w-i-os (rather than *hireu-os, as is frequently assumed) > PArm. *areg-i-.

5.2.5. *Hkek-ih2-: Arm. ak'is, i-stem 'weasel'; Skt. kasika- f. 'Ichneumonweibchen' or 'weasel' (RV 1.126.6), and kasa- 'weasel'. Skt. kasika- f. is considered a derivation from *kasi- f. The connection with MPers., NPers. xaz 'marten' and Lith. seskas 'Iltis' is uncertain. The absence of palatalization of *-k- before a front vowel in Armenian is perhaps due to dissimila-tive influence of the palatal *-k-: *k-k > k'-s. Possible reconstruction: nom. *Hkek-s, oblique *Hkek-. Compare the analyses of aiues 'fox' (u-stem) and iz 'viper' (i-stem). We may be dealing with a common borrowing from an unknown source. (HM)

5.2.6. *ueh2g-nu-: Arm. gang 'sound', Skt. vagnu- m. 'sound, noise'; cf. Lat. vagire 'to cry, wail', etc. For the metathesis (*wagn- > *wang- > *gang-), cf. *bhudhno-: Lat. fundus 'bottom' and Arm. andund-k', o-stem 'abyss' vs. Skt. a-budhna- 'bottomless'. (HM)

5.2.7. *hierkw-o-: Arm. erg, o-stem 'song; poem; 'playing (music)', ergem 'to sing; to play a musical instrument; to praise'; Skt. thematic noun arka- m. 'ray, light, shine; song, magic song', root noun rc- f. 'song of praise, poem, stanza, verse', arcati 'to sing; to praise; to shine'; Hitt. arku-zi, arku- 'to chant, intone' (from *hierkw-/*hirkw-), Toch. A yark, B yarke 'worship, reverence', probably also OIr. erc 'sky'. Arm. erg, o-stem 'song' and Skt. arka- m. 'shine, song, magic song' represent a thematic noun and should be regarded as a shared innovation.

5.2.8. *singho-: Arm. inj, u-stem 'panther, leopard' (renders Gr. napSaAic 'panther, leopard' in the Bible, e.g. nman anju : o^oiov napSaAei in Revelations 13.2; i leranc' anjuc' : and dpewv napSaAew in Song of Songs 4.8); Skt. simha- 'lion'. The connection with Toch. A sisak, B secake 'lion' is uncertain.76

The assumption that the Armenian word has been borrowed from an unattested Iranian form is not compulsory. Nor is it plausible, since: (1) none of the Iranian languages have preserved a trace of this etymon; (2) the loss of the PIE initial *s- is regular in native Armenian words, whereas in Iranian loanwords Armenian preserves the h-; and (3) the semantic difference indicates that it is an old word. Note especially that Arm. inj and Skt. simha- are found in a comparable mythical context. In Armenian incantations the Evil Eye often appears as an inj 'panther', an ariwc 'lion', and a visap 'dragon' coming up out of the sea (compare inj etc. in the famous dream of Daniel 7,77 and in Azdahak's dream in Movses Xorenac'i 1.26); he roars like a cloud (amp/b) or a lion (ariwc/aruc) or a bull (c'ul).78 The animals inj 'panther' and ariwc 'lion' are also listed with gel 'wolf' and oj 'snake' in a Daralagyaz incantation (K'ajberuni 1902:

76 Hübschmann 1877: 25; 1897: 450; Meillet 1936: 142; HAB 2: 243; Specht 1939: 14; Xacaturova 1973: 196; 1979: 363; Gamkrelidze/Ivanov 1984, 2: 507 = 1995: 427; Clackson 1994: 13; Olsen 1999: 110. Müller (1870: 452-453) considers a loan from lost Iranian *hinza- beside Skt. simha- 'lion'. Bailey (1979: 484a, cf. 421; 1987: 461a; see also May-rhofer EWAia 2, 1996: 727; EtimSlovIranJaz 3, 2007: 332) connects the Sanskrit word with Iran. *haiz-, *hiz- 'to mount' assuming a basic meaning 'pouncer' and considers Arm. inj as a loan from an Iranian unattested *hinzu-. Thieme's (1954: 54-56) sceptical approach is rightly criticized by Makaev 1967: 453-454.

77 Cowe 1992: 195-199.

78 Abeghian 1899: 124; Alisan 1910: 412; Ödabasyan 1986: 141.

112aNr1), which starts with the name of Surb Daniel 'St. Daniel' (cf. the aforementioned vision of Daniel).79

For the association of Arm. inj (*singho-: Skt. simha-) with thunderous cloud note RV 5.83.3 where the thunder of Parjanya is compared to the thunder roar of a simha- 'lion'.80

In view of the absence of secure IE cognate forms, one may assume that the Aryan and Armenian (possibly also Tocharian) forms reflect a common borrowing81 from a North Pontic or Near-Eastern source. Possibly related forms in non-IE languages are: the old Central Asian word for 'lion', *sengha/singha, Tibetan senge, singe, Zhan Zhun sango, etc.; North Caucasian: PEC *cä:nqV 'lynx, panther', Chechen coq 'snow leopard', Avar cirqq, Akhvakh ciqqo 'lynx', Lak ciniq 'tiger, leopard', Akusha cirq 'panther', etc.; Akkad. sin/mk/gurru 'a hunted mountain feline, gepard'; Chadic: Kwang semk-, semgi 'lion', Chibak zing'e 'lion', etc.82

5.2.9. *gwou-d.hehi-eh2- 'a lizard', lit. 'cow milker/sucker': Arm. kov-a-di-ac' (also kovideay, kov-di-c') 'a kind of lizard' (renders Gr. xaAaßwTri^ 'spotted lizard, gecko' in Leviticus 11.30) reshaped from an older *kov-di-a-; Skt. godha- f. 'Iguana, a species of big lizard'. In later literature (Nonnus, Galen) and dialects the Armenian word has been replaced by kov(a)cuc 'a kind of lizard', composed of kov 'cow' and cuc 'sucking'. There are many semantic parallels in other languages: Xurasani Pers. boccos (preverb bi + cos- 'Sauger') 'eine Art Eidechse, die nach dem Volksglauben nachts in die Hürden schleicht und den Ziegen am Euter saugt', Ukr. moloko-sis 'lizard', etc. 83

5.2.10. *h3eui-peh2- 'shepherd': Arm. hoviw, a-stem 'shepherd', Skt. avi-pa-lä- 'shepherd', cf. also go-pa- m. 'herdsman', lit. 'cowherd'.

5.2.11. *ghei-o-: Arm. ji, o-stem 'horse', Skt. haya- m. 'horse'. Skt. häya- is usually derived from hay- vs. hinoti 'to impel, set in motion; to hurl; to help' (presumably derived from PIE *ghei- 'to drive; to throw'). Arm. ji 'horse' and Skt. häya- m. 'horse' represent a poetic word, belonging to the "language of gods", as opposed to the PIE word for 'horse', viz. *hiekuo- > Arm. es 'donkey'.84

Kurdish delazi 'horse' is only recorded by Chodzko in 1857 among the Kurds of the Rishvand tribe in Iran near Alamut, between Qazvin and Rudbar. It is composed of del 'female' (cf. delagur 'female wolf') and the otherwise unknown zi, which was earlier considered to be a loan from Arm. ji (HAB 3: 152b). Garnik Asatrian (1997)85 rejects this view saying that this Kurdish dialect had no contact with Armenian during the whole period of its history. He therefore treats this word as the only remnant of Iranian *zaya-, the theoretical cognate to Skt. häya- 'horse'. If this interpretation is accepted, we are dealing with an isogloss between Armenian and Indo-Iranian.

5.2.12. Arm. marmin, o-stem 'body; flesh'; Skt. marman- n. 'vulnerable point of the body', MInd. mamma- n. 'weiche Körperstelle'. There is no consensus about the origin of the Indo-Aryan word. If the Aryan word is related with Lith. melmenys 'die um die Nieren liegenden

79 For a discussion and parallels from other traditions, see Ivanov/Toporov 1974: 169-170; Watkins 1975: 20f; Gusejnov apud MifNarMir 2, 1982: 342; Mawet 1983: 182-183; Gamkrelidze/Ivanov 1984, 2: 500-503 = 1995: 420423; Cowe 1992: 399-401; Petrosyan 2002: 16. For biblical parallels and an extensive discussion, see Xalat'janc 1896: 172-200; see also Thomson 2006: 112230.

80 Cf. Gamkrelidze/Ivanov 1984, 2: 510i = 1995, 1: 43027.

81 Cf. Blazek 2005: 67-68, 91-92.

82 For the forms and a discussion, see Pinault 2002: 330-331; Witzel 2003: 14-15, 45, 47; Behr 2004-05; Blazek

2005.

83 Martirosyan 2010: 372-373; Olsen 2011: 25.

84 The vocalism is slightly problematic. Alternative: a substantivized *-to-participle *ghi-to- (de Lamberterie apud Olsen 1999: 40; see also Viredaz 2005-07: 7-9).

85 I am indebted to Garnik Asatrian and Vardan Voskanian for this information and a copy of the paper.

Fleischteile' and others, Arm. marmin cannot be treated as a native word because of the vocal-ism and the -r-.86 Compare Arm. salil 'raw flesh, body, corpse' vs. Skt. sarira- n. 'the body, bodily frame, solid parts of the body'.

5.2.13. *gwh(e)r-nu- 'to be warm, burn': Arm. jernum or jeranim (aor. ]er-a- from sigm. aor. *gwher-s-) 'to be/become warm, burn'; Skt. ghrnoti 'to glow, light' from *g*br-n(e)u- (see § 3.5 on nu-verbs).

5.2.14. *kHl-eh2: Arm. sal, i-stem (prob. also a-stem, as reflected in a borrowing into Georgian sala 'a flat roundish stone to play with') 'a large flat block of stone; anvil'; Skt. sila- 'stone, rock, crag'. The appurtenance of Goth. hallus 'rock' and others is uncertain.

5.2.15. *kHs-ti-: Arm. sast, i-stem 'rebuke, scolding, censure, castigation, punishment, chastisement, threat; indignation, wrath',87 sastem 'to rebuke, remark indignantly, reprimand angrily, threat', sastik 'intense, hard, violent, rigid'; Skt. sisti- 'punishment, command, instruction'; from the root *k(e)Hs-: Skt. sasti 'to punish, control, command, instruct', Av. sah-'to teach', sasti 'lehrt', saxvan- 'doctrine', sasna- 'doctrine, command', sastar- 'commander', ManMPers. and Parth. sastar 'commander, tyrant'88, probably also Alb. thom 'to say' and Toch. A kas 'to scold, reprimand'.89

Since Hubschmann,90 Armenian sast has been interpreted as an inherited word although some scholars are inclined toward an Iranian origin instead.91 However, there is no compelling reason for the latter, and I do not understand the extreme confidence of, e.g., Xacaturova (1979: 372), who claims that Arm. sast is indisputably ("бесспорно") an Iranian loanword. There are no Iranian forms that would be formally compatible with the Armenian word as a source of borrowing. Besides, the Armenian word is semantically closer to the Indic than the Iranian forms. I therefore follow Hubschmann, Acaryan and Jahukyan in directly deriving Arm. sast, i-stem 'rebuke, punishment, indignation' from *kHs-ti- and thus identifying it with sisti- 'punishment, command, instruction' and positing an Armeno-(Indo-)Aryan isogloss.

5.2.16. *ker-e/os- n. 'cream of milk': Arm. ser 'cream of milk, skin on milk or sour clotted milk' (Zgon-Afrahat, Middle Armenian; widespread in the dialects); Skt. saras n. 'cream, skin on milk'. 92 Mayrhofer (EWAia 2, 1996: 617-618) hesitantly derives the Sanskrit form from the root sar- (srnati) 'to smash, crush, break' from *kerh2-, cf. Gr. xepatZw 'to destroy', etc. (HM)

5.2.17. *skHel-: Arm. sxale/im 'to err, be mistaken; to stumble; to fail, miss'; Skt. skhalati 'to stumble, stammer, fail'. Here belong also Arm. sel 'slanting, crooked, oblique', sil 'squint-eyed', dial. 'mistake; disorder'; Lat. scelus, gen. sceleris n. 'misdeed, crime', Gr. океААдд 'crook-legged', окоАюд 'wicked, crooked' (on the other hand, note Gr. офаААоуас 'to fall, stumble, be mistaken'). The twofold development of *skH- as Arm. s- and sx- is puzzling. Most probably, sxale/im is a loan from the Indo-Aryan language of the Near East,93 while sei/sil has been inherited from Indo-European.

86 For a critical discussion of the etymology of marmin, see de Lamberterie 2013: 44-47.

87 In the Bible translation, Arm. sast corresponds to, e.g., Gr. enixi^rjoiQ 'castigation, censure, criticism' and ayavaKTr\OL^ 'vexation, wrath'.

88 See Durkin-Meisterernst 2004: 306a.

89 For the forms and a discussion, see Mayrhofer KEWA 3, 1976: 319-320, 330-331; EWAia 2, 1996: 632-633; Schrijver 1991: 101; Demiraj 1997: 399-400.

90 Hübschmann 1897: 488-489; HAB 4: 178; Jahukyan 1987: 130, 173, 551 (hesitantly); 2010: 669a.

91 Pokorny 1959: 533 (with a question-mark); Benveniste 1964: 2; Mayrhofer KEWA 3, 1976: 330 (not mentioned in EWAia); Schmitt 1983: 108; 1987: 446b; Olsen 1999: 906.

92 Martirosyan 2010: 574-575; welcomed in Olsen 2011: 26.

93 Another possible case of such a borrowing is Arm. burn (i-stem, cf. adv. brn-i-w 'violently' in Eusebius of Caesarea) 'strong, violent', 'violently', 'violence, strength; tyrant', if indeed from Skt. bhurni- 'zealous, wild'.

5.3. Discussion.

5.3.1. Native or loan? The examination of Armenian-Indic correspondences is complicated in several respects. Firstly, scholars often state that Arm. arew, erg and others were borrowed from Aryan in the middle of the second millennium BC. This view is untenable since at that period the development PIE *e > Aryan a had already taken place, as is seen in Mitanni panza 'five'. Besides, these poetic words are culturally and/or semantically associated with each other and are all Armeno-Indo-Aryan (or Armeno-Graeco-Aryan) correspondences, and some of them clearly preclude the loan theory: arcui 'eagle', ji 'horse', c'in 'kite', etc. For the association between 'bird, eagle', 'horse' and 'sun' in poetic language, cf. e.g. Skt. patanga- adj. 'flying', m. 'bird; flying horse; sun'.

Secondly, there is always a possibility that the Indic might have had an Iranian cognate even if it is not attested in the Iranian languages themselves. This point is often illustrated using the Armenian word nirh 'dormancy, slumber'. On the basis of its appearance, the word is seen as a loan from an Iranian *ni5ra-. In Iranian such a word is not attested, but we do know that it exists in the Indo-Iranian subgroup at large because of Vedic Skt. nidra- f. 'slumber, sleepiness'.

Thirdly, in individual instances it is often very difficult to identify a word as an inherited word or an Iranian borrowing. Armenian and Iranian are independent branches of Indo-European but sometimes parallel phonetic developments complicate a judgement on the status of a lexeme. A frequently cited example is Arm. naw 'boat, ship': is it an Iranian loan (cf. Oss. naw/nawx 'boat', Khot. no 'boat', Parth. nawaz 'skipper' > Arm. nawaz 'boatman') or an inherited word next to Skt. nau- 'boat', Gr. vavQ 'ship', Lat. navis, is 'ship', OIr. nau 'ship'?

It is usually the cumulative evidence that tips the balance. Arm. surb, o-stem 'pure, clean; holy' (Bible+) has been taken as cognate to Skt. subhra- 'shining, glimmering, beautiful'. On the other hand, the Armenian word may have been borrowed from Middle Iranian *subra-, itself a lost cognate of Skt. subhra-. A number of circumstances point to the native origin of the Armenian form, though, individually taken, none of them is decisive: (1) the o-stem of the Armenian; (2) the metathesis *-bhr- > Arm. -rb-; (3) the semantic difference; (4) the absence of direct evidence for this lexeme in the Iranian language group. The last two arguments have become insignificant in view of Khotanese suraa- 'clean, pure', which has been regarded as reflecting the theoretical Iranian form *subra-ka-.94 On the whole, it seems more likely that we are dealing with an Armeno-Indo-Iranian lexeme rather than an Iranian loanword in Armenian.

See also the discussions on ji 'horse' (§ 5.2.11) and sami (§§ 4.2.3-4).

5.3.2. Armeno-Indo-Iranian poetic or mythical lexicon. We have discussed poetic words inherited from the Armenian-Greek-Indo-Iranian dialect union (see arcui 'eagle' and c'in 'kite'). As for the Armeno-Aryan poetic words, we have already discussed ji 'horse' and arew 'sun'. We have also discussed two Armeno-Aryan words in the mythological context: andund 'abyss' and inj 'panther'. Here I shall elaborate on 'sun' and 'moon'.

Arew, gen. Areg- 'Sun God' is attested in Movses Xorenac'i 2.8 and in folkloric texts. Most explicit is the following folk prayer from Larabal: Astco c'ncuin tvac arignak, im eress k'o otand takd, du im xoxek's pahes "O du göttlich strahlende Sonne! Dein Fuss ruhe auf meinem Antlitz! Bewahre meine Kinder" (transl. Abeghian 1899: 43). Note also that this word appears as an oath formula or as an interjection of astonishment. Arm. arew/g- 'sun, Sun God' and Skt. ravi- m. 'sun, Sun God' (Upanisad+) derive from a proto-form *hireu-i- and may be regarded as an Armeno-Aryan poetically or sacredly marked designation of 'sun' replacing the PIE unmarked profane word for 'sun', *sehiul-.95 This is reminiscent of the case of Arm. ji vs. Skt. haya- which we have already discussed.

94 Emmerick apud Schmitt 1987: 446b; Emmerick / Skj®rv0 1997: 155; see also Lubotsky 2001: 5151.

95 Martirosyan 2010: 135-138.

It is remarkable that the Armenian word for 'moon', lusin, is also in a way related to the Armeno-Aryan unity. The word has an o-stem (abl. i lusn-o-y in Eznik Kolbac'i, ins. lusn-o-v in Jeremiah 8.2) and is usually derived from *loukeno-: Skt. rocana- n. 'light, luminous sphere, firmament', YAv. raocana- adj. 'shining, light'. Next to this, however, there is also reliable evidence for gen. lusn-i (abundant in the Bible) which may point to both i- and a-stems; this can be confirmed by ins. lusn-i-w (Movses Xorenac'i 2.77, etc.) and ins. (z-)lusn-a-w (Anania Sira-kac'i, 7th cent., Abrahamyan 1940: 58L20f) respectively.

This leads us to the derivation96 of lusin from *louksneh2-: Lat. luna (Praeneste losna) f. 'moon, month' and OCS luna f. 'moon', cf. Av. raoxsna- adj. 'shining', OPr. lauxnos nom.pl. 'luminary', as well as Arm. lusn 'white spot'. The internal -i- may be analogical (cf. katin 'acorn' vs. Gr. fiaAavoc 'acorn'). In view of the o-declension of lusin, however, it is tempting to assume a blend with *loukeno- 'light, luminous (sphere)'.

My working hypothesis can be formulated as follows: Armenian inherited PIE *louksneh2-f. 'moon', cf. Lat. luna (Praeneste losna) and OCS luna f. 'moon'. In a late period around the Indo-European dispersal, Proto-Armenian shared the thematic innovation *loukeno- 'light, luminous (sphere)' with Indo-Iranian (cf. Skt. rocana- and YAv. raocana-). Subsequently, PArm. *lusna- f. 'moon' blended with *lowsino- 'luminous' and resulted in lusin 'moon', displaying o-, a-, and i-stem forms.

5.3.3. Other issues. A lexical correspondence, albeit perfect both semantically and formally, cannot be considered as significant for the purpose of this paper unless we demonstrate that we are dealing with a shared innovation rather than an archaism. In some cases we are dealing with very interesting correspondences, the nature of which is quite hard to determine. Such ambiguous correspondences, even those which are more likely to be archaisms, should not be ignored if they display recurrent patterns. Future studies should gather all such correspondences and try to estimate their cumulative strength.

Armenian y-arnem (aor. stem y-ari, imper. ari) 'to rise, arise, get up, stand up, wake, resurrect' derives from PIE *h3r-i- 'to rise': Hit. arai-i / ari- 'to rise, arise, lift; to raise', CLuw. ari(ia)-'to raise' < *h3r-oi- / *h3r-i-; Lat. orior, -iri, ortus 'to appear above the horizon, rise; to rise from bed, get up; to begin, be born'; Skt. ar-, 'to set in motion, move; to arouse, excite', rnvati 'to rise, move', Av. ar-, redupl. pres. ira- 'to reach', ira- n. 'attack', YAv. aranao- 'to set in motion'; Gr. opvv^L or -vtx), med. opvv^aL 'to rise, rouse, stir (up), urge on, move'.

3sg.pres.act. iyarti, med. irte < *Hi-H(a)r-. Armenian *y-ar-i- and impv. *ari derive from *h3r-i- (cf. Hit. arai-i / ari- 'to rise', perhaps also Lat. orior, -iri 'to rise'). The initial y- in *y-ar- (vs. imperative *ar-) is puzzling. It is tempting to explain the problem by assuming a redupl. pres. *Hi-H(e)r- > PArm. *Hiyar- > *(i)yar-. This would match Skt. iyarti (next to ar-). One is tempted to treat this as an Armenian-Aryan isogloss; note especially *ni-si-sd-e/o-: Arm. *nihist-e-, Skt. nisidati and Av. nishiSaiti 'to sit' (see § 3.6). However, the reduplicated present seems to be an archaic feature in Indo-European and is not productive in Armenian. The reduplicated structure of PArm. pres. *(h)ipe- (with suppletive aor. arb-) 'to drink' from PIE *pi-ph3-e- > *pibeti (Skt. pibati, Lat. bibo, OIr. ibid) ceased to be sensed at a very early stage, and a new present was made by a nasal affix: *(h)ipnem(i) > ampem.

Another complicated but intriguing example is Armenian targal 'spoon' (attested in Movses Xorenac'i 2.47 and ubiquitous in the dialects) that seems to derive from *dru-, a zero-grade form of the PIE word for 'wood'. A perfect semantic match is Skt. darvi f. / darvi f. 'spoon', though this has a full grade in the root. But now we have a wonderful match that can solve even the problem of the suffix: Hitt. GlHaru-ali- n., which refers to an implement used for

96 Meillet 1936: 21. For a full discussion of lusin and related words, see Martirosyan 2010: 320-322.

grinding or crushing, probably something like 'pestle'. For *-al(i) in designations for implements or the like cf., e.g., Hitt. Gl%ulali- n. 'distaff'. I wonder, therefore, whether Arm. targal is an Anatolian loanword.

There is a better Armenian match for Skt. darvi f. and darvi f. 'spoon', namely torg 'wooden framework, loom'. Here again we find an interesting Anatolian cognate: HLuw. tarw-i(ia)-prob. 'wooden beam'. Further, note Arm. torn 'pestle' and Skt. drona- n. 'wooden vessel, trough, bucket'.97

Onomatopoeia and nursery words are usually considered insignificant for the problems of reconstruction. However, identical onomatopoeia and nursery words are not necessarily independent creations. Here again, cumulative strength can play a certain role in estimating a genetic relationship between two languages or dialects. Note correspondences such as Arm. alalak, obl. aiaiak-a- 'shouting' vs. Skt. alala and Gr. aAaAayrj 'shouting'; Arm. atta 'mother' vs. Skt. atta 'mother, older sister' (other cognates differ in their semantics).

Table set A (sections 4-5)

Lexical isoglosses: Armenian, Greek and Indo-Iranian.

Physical world, time, space.

Proto-form Gloss Sanskrit Iranian Armenian Greek

*plhu- Pleiades nAeiâSeç

*polio-/-ieh2 wave; grey (palitâ-) *parya- ali-k' noAiôç, -iâç

*sm(H)-eh2- year sâma- (ham-) am, -a-

*n-bhudkno- bottomless a-budhna- MP a-bun andund-, -o-

A *sru-ti- stream, etc. sruti- aru, i-stem pvaiç

B *sr(o)u-to- stream, etc. srotas- OP rautah- am, -o- pVTÔÇ

*h2reu-i- sun ravi- arew

*pro-hienht- air, breeze pranâ- *frana- eran

*loukeno- luminous rocanâ- raocana- lusin, -o-

*prHuo- east./south. purva- pauruua- haraw, -o-

*hsmeigh-o- cloud, mist meghâ- maeya- meg, -o-

*kHl-ehi stone, rock sila- sal, -i- (-a-)

*kubh-ro- shiny, pure subhrâ- Khot. suraa- surb, -o-

Human, age, kinship.

Proto-form Gloss Sanskrit Iranian Armenian Greek

man nâr- nar- ayr, gen. arn àvrip

A *gerH- old jarâs- zar- cer yépaç

B *gerHont- old jârant- Oss. zœrond cer-un(-i) yépav

mortal -mfta- -masa- mard, -o- Ppojôç

97 For a discussion of all these words, see Martirosyan 2010 s.vv.

Body, perceptions, mentality, belief.

Proto-form Gloss Sanskrit Iranian Armenian Greek

uehig nu sound vagnú- gang

*pr(e/o)Hkt- buttocks (prsthá-) (parsta-) erastan-k' npœKTÔç

*h erkw-o- song arká- erg, -o-

*m(e/o)rmen- body márman- marmin

*skHel- to err skhálati sil / sxal 0<ftáAAo^.at

*pe/ork-u- rib, side pársu- parasu- (y-)ors-

A *gwher-os- warmth háras- jer, -o- $époç

B *gwh(e)rnu- to be warm ghrnóti jernum

*(p)ste/ên- wom. breast stána- fstana- stin 0TfVioV

Movements, speech and other activities.

Proto-form Gloss Sanskrit Iranian Armenian Greek

*s(e)Hdh-io- succeed sídhyati aj(-)

*hir-nu- gain; allot aranauu- arnum àpvv^ai

A *kie/ou- to go, move cyav- 0£VOm

B *kiu-to- moved cyutá- -00VTOÇ

punish, etc. sisti- sast, i-st.

*(s)peud- zeal, haste Parth. pwd- p'oyt' 0nov8-f

Fauna.

Proto-form Gloss Sanskrit Iranian Armenian Greek

*hilV(u)pek- fox

*h2rgipió- eagle, etc.

*tkiH-(e)no- bird of prey syená- saêna- c'in, o-stem ÎKTÏVOÇ

*Hkek-ih2- weasel kastka- ak'is, i-stem

*hegwh-i- snake, adder áhi- azi- iz, i-stem ëXiç

*singho- lion, panther simhá- inj

*gwou-dheh- lizard godha- kov-a-di-ac'

*ghei-o- horse háya- (*zaya-) ji, -o-

Animal husbandry.

Proto-form Gloss Sanskrit Iranian Armenian Greek

*hi(e)ig- goat (eda-) (tzaena-) ayc aï^

*hrs-en- male anim. (rsabhá-) arsan- arn àP0nv

*urhiën lamb úran- *varn- garn papfv

*hiegh-ihi- cow, ox ahí- azt- ezn, g. ezin

*gwou-io- of cow gávya- gaoiia- kogi -3o(p)ioç

*h^eui-pehi- shepherd avi-pa-lá- hoviw, -a-

*ker-e/os- cream sáras ser

Agriculture.

Proto-form Gloss Sanskrit Iranian Armenian Greek

alewr àAevpov

(MInd. ata) *arna- aiam àAéw

*a/ardra- aiawri àAe%piç

*h2erh3uer/n- arable land àpovpa

*ke/omieh2 yoke-pin sami, ea-st. (xâpcxÇ)

House, housekeeping, crafts, implements, building.

Proto-form Gloss Sanskrit Iranian Armenian Greek

*h2énHt-ieh2 threshold (ata-) qiSiiä-

*h3nêbh- nave, hub aniw 'wheel'

*bheh2g- lot, estate (bhaga-) *baga- bak (^ayecv)

*dhmbh- tomb, grave *daf-ma- damban râiftoç

*k(e)r(H)- to tie, form *sar- sarem

*k(e)r(H)ieh2 band sari-k', -ea- Ke/aipia

to bind da-/dyati da- *ti- ôéw

bond -diti- *tay Séaiç

6. Lexical isoglosses between Armenian, Greek and European dialects

6.1. Armenian and Greek: innovations.

6.1.1. *hilehi-ur: Arm. alewr, aliwr, gen. aler (later also o-stem) 'flour', Gr. aAevpov, aAevpog 'flour'. See also aiam 'to grind' and alawri 'mill, female grinder' (§ 4.1.3-4).

6.1.2. *agu(s)iehi-: Arm. acu 'garden-bed', Gr. ayvia, pl. ayviai f. 'street, road'; probably a shared innovation based on PIE *hieg-: Arm. acem 'to bring, lead', Skt. ajati, Gr. ay<x> 'lead', etc.98 For the semantic relationship between 'garden-bed' and 'street', compare Arm. marg 'meadow' (dial. 'garden-bed'), which has been borrowed from Parth. mary 'wood, meadow'. Sanskrit has mrga- m. 'wild animal' (cf. Wakhi merg f. 'female ibex') and marga- '(wild) path, road'. The latter is comparable to the Armenian dialectal meaning 'garden-bed'.99

6.1.3. *h3kwkwon 'eye': Arm. akn gen. akan 'eye', Gr. dxxov dtyOaAyov 'eye' (Hesychius); derived from PIE *h3(o)kw- 'eye': Skt. aksi-, Gr. oooe, Arm. ac'-k', etc.

6.1.4. *anter / *an(n)er 'cave': Arm. ayr, i-stem 'cave', Gr. avzpov n. 'cave'.100 The development of *-nt- is problematic, however.101 Perhaps one can assume a substrate origin with a nasal vacillation, *an(n)er vs. *anter, somehow comparable to another substrate term, Arm. kamurf 'bridge' vs. Gr. yetyvpa 'bridge'. The by-form *an(n)er could easily develop into ayr (cf. *hiner > ayr 'man').

98 On Gr. ayvia, Arm. acu and various explanations of -u, see Clackson 1994: 117, 225m; Martirosyan 2010: 1718; Beekes 2010, 1: 17 (not mentioning the Armenian word).

99 For these Armenian and Indo-Iranian words, see HAB 3: 275-276; Mayrhofer KEWA 2, 1963: 626, 669; EWAia 2, 1996: 370-371; Dockalova / Blazek 2011: 323, 327.

100 De Lamberterie 1978: 243-245; 1992: 238; Olsen 1999: 92; Martirosyan 2010: 62-63 with lit.

101 Clackson 1994: 98; Beekes 2010, 1: 110.

6.1.5. *h\0s-r-(e)hi 'harvest, summer': Arm. *ar-a- 'harvest, harvest time', seen in ar-a-c' 'harvest time, harvest of grape/fruit', the sixth month (17th August to 15th September); Gr. дп-шра f., Lac. дп-ара 'end of the summer, beginning of autumn; harvest, fruit'; cf. also CS jesenb, Russ. osen' f. 'autumn', Goth. asans f. 'harvest, summer', OHG aran, Germ. Ernte 'harvest', etc. Arm. *ar-a- derives from PArm. *o(h)ar-a- < neuter plural or collective *hios-r-h2 'harvest, summer' (or *hios-r >> fem. *h\os-r-ehi). Note the remarkable contrast with the preceding month name, k'al-o-c' 'mowing time', deriving from k'alem 'to pluck, weed, mow, harvest' < *(s)kl-nelo-, which is a Graeco-Armenian agreement too, cf. Gr. окаААш 'to stir up, hoe' from *OKaA-v<x> (see § 6.1.30).

For the typology of such a contrast between the fifth (reaping/mowing) and the sixth (harvest — grape/fruit) months, compare e.g. the contrast between the fifth month (July-August) as "reaping/mowing time" ("урожайная пора") vs. the sixth month (August-September) as "beginning of the pressing of grape-juice" ("начало выжимания виноградного сока") in the Khwarezmian calendar.102 (HM)

6.1.6. *h2er- 'to fix, put together': Arm. arnem, lsg.aor ar-ar-i, 3sg.aor. ar-ar 'to make; to create': Gr. apapioKw, aor. rjpapov 'to fit, equip', etc. The agreement is unobjectionable both formally and semantically, but it may be an archaism.103

6.1.7. *Hehim-(d)r, gen. *Hhim-(e)n-: Arm. awr, gen. awur, instr. awur-b 'day; time, age'; Gr. щар, Arc. ayap, -атос n. 'day', щера, Dor. ацера 'id.'. Arm. aw(u)r may be explained as follows: *alamor > PArm. *amur > *awmur > *awur > Arm. awr, gen. awur.104

6.1.8. *h3bhel-: Arm. awel 'broom', later denominative awelem 'to sweep, broom'; Gr. дфеААы 'to sweep, broom', дфеАуа, o^eATpov 'broom'; Arm. *awel- 'increase' in aweli 'more', ar-awelum 'to increase', y-awelum 'to add to'; Gr. дфеААы 'to increase, enlarge, augment, advance' (cf. Myc. no-pe-re-a2 /nopheleha/ 'useless' < *n-h3bhel-es-h2: ^шфгАцс). There is no cognate to this root in other Indo-European languages.105

Jahukyan (1970: 2I39) admits the possibility that Arm. awel- 'to increase' has been borrowed from Urartian abili-d(u) 'to join, increase'. Arutjunjan (1983: 339ш) notes that in this case the comparison between Arm. y-awelum and Gr. дфеААы would be impossible. However, the etymological connection between these Armenian and Greek words is unobjectionable, and the apparent contradiction can easily be removed if we assume the opposite direction of borrowing, namely from Armenian into Urartian.106 Compare the cases of arcui 'eagle' and burgn 'tower' (§§ 4.2.1-2 and 6.1.10).

The remarkable agreement between Armenian and Greek in both meanings, 'sweep' and 'increase', makes this one of the most important isoglosses.

6.1.9. *bhh2-ti-: Arm. bay, i-stem 'speech, word, verb', Gr. фаосс, фатсс f. 'declaration, enunciation, rumour'; a zero-grade ti-derivative of PIE *bheh2- 'to speak': Arm. bam 'to speak, say' vs. фцус 'to say'.

6.1.10. *burgh- 'tower': Arm. burgn, gen. brgan 'tower; pyramid'; Gr. nvpyoc, m. (also фрркос) 'tower'. Notwithstanding the formal problems, which might suggest a substrate in-

102 See Martirosyan prepar. 1.

103 Clackson 1994: 101-102; Martirosyan 2010: 112; de Lamberterie 2013: 18.

104 The appurtenance of OIr. amm 'time, season' (from *Hh2m-n-?) and Arm. amanak 'time' is uncertain. For a discussion, see Clackson 1994: 96-97; Martirosyan 2010: 46, 156.

105 HAB 1: 356-358 with lit.; de Lamberterie 1992: 238; 1992a; Clackson 1994: 33-35,156-158; Olsen 1999: 211, 436; Beekes 2010, 2: 1133.

106 Jahukyan 1987: 433; 2010: 100-101. For the Urartian word and its connecton to Armenian awel-, see N. Arutjunjan 2001: 431a.

termediation, this cultural term seems to be based on *bhergh-,*bhrgh-u-, *bhrgh-(e/o)nt-: Arm. barjr, gen. barju, -berj 'high', barnam 'to lift, raise' < *barj-nam; Hitt. parku- 'high', Skt. brhdnt-'large, wide, abundant, lofty, high, strong, dense, loud', etc. Urart. burgana 'fortress' (if the meaning is reliable) may be an Armenian loanword. For another cultural term of a similar structure, cf. durgn, gen. drgan 'potter's wheel' vs. darnam 'to turn; to return' < *darj-nam (see § 6.5.2).

6.1.11. *dHhiro-: Arm. dalar, o-stem 'green, fresh'; dalar-i, dalarw-o-y, -o-f 'greenery, grass, herb'; Gr. SaAepog 'blooming, fresh'. The root is visible in Gr. $aAA< 'to bloom, flourish, grow', SaAog n. 'sprout', SaAAoQ m. 'green twig, esp. of the olive, sprout', MIr. duilne, duille 'leaf, foliage', OEngl. dile 'dill', Alb. dal 'to sprout', etc., as well as Arm. del, o-stem 'herb; medicine; poison'. Notwithstanding the problems concerning the reconstruction of the root (*dHh\- or *dhehil-) and the suffix (*-ero- or *-ro-),107 I see no solid reason for separating Arm. dalar (o-stem) from Gr. SaAepog.

It is also worth considering whether Arm. del 'herb' and Gr. $aAA6g m. 'green twig, sprout' derive from an old n-stem: nom. *dhel(H)-n-, gen. *dH-nos.

6.1.12. *dh(e)his- 'god': Arm. di-k', gen.pl. di-c', instr.pl. di-a-w-k' (compositional diwc'-) 'god', Gr. $eog 'god'. With a different meaning: Lat. feriae < OLat. fesiae 'festival days', festus 'festive', fanum < *fas-no-m 'hallowed place', Osc. fiisnu 'templum', etc.

6.1.13. *pr(e/o)Hkt-/*pr(o)kt-: Arm. erastan-k', a-stem 'buttocks', Gr. npwKxog m. 'anus'. Clackson (1994: 166-167) takes this as an Armenian-Greek-Aryan correspondence, cf. Skt. prsthd- n. 'back, mountain-ridge, top', prsti- f. 'rib', cf. YAv. parsta- m. 'back, spine, support in the back', parsti 'back'. However, the Indo-Iranian words appear to be derived from *pr-sthi-o-and are thus unrelated.108 Even if they are related (which would create an isogloss between Armenian, Greek and Indo-Iranian), Armenian and Greek are certainly closer to each other since they agree in both vocalism and semantics.

6.1.14. *pre(i)sgwu-: Arm. erec', u-stem (adj.) 'elder; presbyter'; Gr. npeopvg m. 'old man; the elder; ambassador; president'; perhaps also Lat. priscus 'ancient'.109

6.1.15. *duehi-ro-: Arm. erkar 'long' (in both temporal and spatial aspects), Gr. ^d^apSg: Sqpog, Dor. Sapog 'lasting long'; with a different vocalism: Lat. duro 'to endure, last out, survive', Skt. durd- 'far', etc.; *dueh2-n-: Arm. erkayn 'long', Gr. Srjv 'long, far' < *5pav.110

6.1.16. *ues-nu-: Arm. z-genum, 3sg.aor. zge-c'-a-w 'to put on clothes', Gr. evvv^.L 'to clothe; to put on, clothe oneself'; cf. Hitt. ues- 'to be dressed', Skt. vdste 'to be clothed, wear', etc. (see § 3.5 on nu-verbs).

6.1.17. *pter- 'feather, wing, blade': Arm. t'er (widespread in the dialects: Hamsen, Axalc'xa, tarabal, Ararat, Jula, Sebastia, etc.) 'leaf; leaf of dough or paper; petal', *t'el (dial.) 'id.'; t'er, abl. i t'ere 'side' (from earlier 'wing, feather');111 t'ert', i-stem 'leaf of a flower, plant or paper; plate' (Philo, Paterica, etc., and dialects); Gr. nzepov n. 'feather; bird's wing; wings of a bat and of insects; any winged creature, such as the Sphinx; anything resembling wings or feathers, such as oars and parasols; side-walls of Egyptian temples; drawbridge', mepvE, f. 'wing of a

107 For a discussion, see Clackson 1994: 118-120; Matasovic 2009: 88, 102-103; Martirosyan 2010: 231-232, 237238; Beekes 2010, 1: 530-531.

108 Mayrhofer EWAia 2, 1996: 165-166; Beekes 2010, 2: 1244. For a discussion of the Armenian word and literature, see Martirosyan 2010: 258.

109 Clackson 1994: 165; Martirosyan 2010: 262-263; Beekes 2010, 2: 1231-1232; de Lamberterie 2013: 15.

110 Meillet 1924: 1-4; de Lamberterie 1992: 257; Clackson 1994: 114-115; Martirosyan 2010: 266-267.

111 For the semantic development, cf. the meanings 'side-walls of Egyptian temples' and 'drawbridge' of Gr. nrepov 'wing'. Note also Arm. kurn 'back, arm' and 'side'; Engl. wing 'wing' and 'side, flank'.

bird; winged creature, bird; blade'; the other cognates represent *pet-r-: Skt. patra- n. 'wing, feather, leaf', OHG fedara 'feather', Hitt. pattar, obl. pattan- 'wing, feather', etc.; derives from *pet(H)- 'to fly': Skt. patati 'to fly, rush, fall', Gr. nex-o-^ai, m-e-oSai 'to fly', etc.

Arm. t'er and Gr. nzep- correspond to each other both semantically (pace Beekes 2010, 2: 1248) and formally. Notwithstanding the formal problems, here may also belong Gr. mLAov n. 'soft feathers, down; wing (properly of insects); the wing-like membrane on a kind of serpent',112 Arm. t'el 'leaf, leaf of dough' and 'wing of a bat' (the latter meaning is seen in mask-a-t'el 'bat', with mask 'skin' as the first member) and redupl. t'it'eln / t'it'ern 'butterfly'.113

6.1.18. *tumbo- 'mound': Arm. t'umb 'mound; fence, wall around a house', Gr. zvp.pog m. 'mound, burial mound, grave' (see § 4.1.14 on *dhmbh-: Arm. damban 'tomb, grave').

6.1.19. *gelhi-os, gen. *glh2-s-e/os:n4 Arm. calr, gen. cai-u 'laugh, laughter; joke, mockery'; Gr. yeA«Q, -wtoq m. 'laughter', yeAao^a 'laughing', yaAfvrj f. 'stillness of the sea', yeAaw 'to shine'. Note also Arm. catik 'flower' and the Hesychian gloss yeAelv Aa^neiv, avdelv 'shine, bloom'.115

6.1.20. *gwlhi-eno-: Arm. kalin, o-stem 'acorn', Gr. fiaAavog f. 'acorn'; with a different suffix: Lat. glans, glandis f. 'acorn', SCr. zelud 'acorn', Lith. gile, etc.116 It is tempting to identify Arm. dial. tarabal tkoien 'hazelnut' (< *tu-kutin < *tu-kalin) with Gr. paAavog 'chestnut' (cf. Lat. iuglans 'walnut') from *diuos-gwlhi-eno- 'divine acorn' (Martirosyan 2010: 348-349).

6.1.21. *gwnehiik- 'woman': PArm. *kan-ay- (seen in pl. kanay-k' vs. sg. kin), Gr. yvvai-x-, voc. yvvai, nom. yvvf f. 'wife, woman'. This is a remarkable agreement, though its nature is debated.117

6.1.22. *per-(i)on- 'piercing implement': Arm. heriwn, ins. hereamb 'awl', Gr. nep-dvn f. 'pin or tongue of a brooch or buckle' from IE *per- 'to pierce': Gr. neip< 'to perforate, pierce, pervade', etc.ii8 The suffixes are different in the two languages.ii9

6.1.23. *mar-mar-: Arm. dial. *mar-m(a)r-il 'to shimmer, flicker, glimmer, extinguish gradually (said of e.g. a candle)', Gr. y.apy.aipw 'to flash, sparkle, gleam' (said of any darting, quivering light), which is analyzed as a reduplicated intensive yod-present *mar-mar-ie-. For the other Armenian and Greek forms, see HAB 3: 248-249, 262, 263, 365 and Beekes 2010, 2: 906-907. (HM)

6.1.24. *mehitrui(e)hi 'stepmother': Arm. mawru, a-stem 'stepmother' (dial.: Hamsen moru 'stepmother', Mus muri 'step-', Satax muru mer 'stepmother', Mus / Bulanax xort'umuru < *xort'-u-moru); Gr. ^n%pvia 'stepmother'; further: OEngl. modrige (n-stem) 'mother's sister'. This is an innovation shared by Armenian and Greek (and, more distantly, Germanic). It is based on PIE *mehiter- f. 'mother'.120

6.1.25. *me-ghsr-i 'near', lit. 'in the hand': Arm. merj 'near', merjenam < *merji-anam 'to approach, touch'; Gr. y.expi 'as far as; up to, about, nearly; until; as long as, whilst'.

112 For various views and references, see Beekes 2010, 2: 1249.

113 For a thorough discussion of all these Armenian words, see Martirosyan 2010: 286-294, 450-451.

114 Alternative: an old u-stem with nom. *-ou(s).

115 De Lamberterie 1978: 269-276; Klingenschmitt 1982: 147-148; Clackson 1994: 126-132; Kortlandt 2003: 117119; Martirosyan 2010: 336-338, 340-341; Beekes 2003: 193-194; 2010, 1: 257-258, 264-265.

116 Clackson (1994: 135-136) is positive about this isogloss.

117 Clackson 1994: 136-137; Martirosyan 2010: 363-365; Beekes 2010, 1: 291-292.

118 Hübschmann 1897: 467; Pedersen 1924: 225 = 1982: 308; HAB 3: 86; Meillet 1936: 142.

119 De Lamberterie 1982: 66-67; Clackson 1994: 159; Olsen 1999: 492.

120 For a discussion and literature, see HAB 3: 246b; Szemerenyi 1977: 60; Beekes 1976: 55-58; Clackson 1994: 145-147; Martirosyan 2010: 453-454; Beekes 2010, 2: 949. For the element *-u-, cf. Gr. ^TpwQ m. 'male relative of the mother, maternal uncle, grandfather' from *meh2tr-ou- 'relative of the mother', perhaps also Arm. mi-a-mawr, gen.pl. -u-c' 'the only (offspring) of one mother'.

6.1.26. *med-es-(e)hi 'mind, counsel': Arm. mit, a-stem, mostly in pl. mit-k', gen. mt-ac' 'mind, intelligence', Gr. ^fSea 'counsels, plans, arts' (pl. of the unattested *^fSoQ, -eoQ, s-stem neuter), y.fSoy.ai 'to be minded, intend; to take care, keep watch', cf. ^.eSw 'to protect, rule over', ^.eSo^ai 'to provide for, be mindful of; to plan, contrive, devise', Lat. medeor 'to heal, cure', Umbrian mers 'law, justice' < *medos, etc. from PIE *med-. The Armenian and Greek forms agree in both vocalism and semantics.

6.1.27. *h1en-h3orgH- 'testicled, uncastrated, male (ram or buck)': Arm. y-orj, i-stem 'male sheep, ram'; Gr. ev-opxiQ 'provided with testicles, uncastrated', cf. ev-opx-oQ, ev-opx-rjQ meaning also 'buck'; note also Arm. xol-orj(n) 'orchis' vs. Gr. opxiQ m. 'testicles', 'the plant orchid (because of the shape of the root)'.121 (HM)

6.1.28. *h1e/oghjn(i)o- 'hedgehog': Arm. ozni 'hedgehog', Gr. exivoQ m. 'hedgehog, sea-urchin'; cf. OHG igil 'id.', Phryg. eZiQ 'hedgehog', Lith. ezys, Russ. ez 'id.', etc.; note also Oss. wyzyn/uzun 'hedgehog'.

6.1.29. *suekur-(e)h2: Arm. skesur, a-stem 'husband's mother', Gr. exvpa 'mother-in-law'. Other cognates continue *suekru-h2 f.: Skt. svasru-, NPers. xusru, Pashto xwase, Lat. socrus, OHG swigar, OCS svekry, etc. Arm. skesur, -a- and Gr. exvpa derive from QIE fem. in *-ur- which has been taken from the PIE form for 'father-in-law', *suekur-o-: Skt. svdsura-, YAv. xvasura-, Gr. exvpoQ, Lat. socer, OLat. socerus, OHG swehur, CS svekn (*swesur- was replaced by *swekr < *swekr- analogically after svekry 'mother-in-law'), Lith. sesuras, etc.

6.1.30. *(s)kl-ne/o-: Arm. k'alem 'to pluck, weed, mow, harvest', Gr. oxaAAw 'to stir up, hoe' prob. from *oxaA-v<; further: Lith. skeliu, skelti 'to split', etc.122 Note Arm. k'al-o-c' 'mowing time', the 5th month (18 July-16 August). See ar-a-c' 'harvest time, harvest of grape/fruit', the sixth month (§ 6.1.5).

6.2. Armenian and Greek: isolated words.

These words have no Indo-European etymologies and may therefore be treated as words of substrate origin, that is to say, common borrowings from an unknown language (for a discussion of the substrate, see section 7).

6.2.1. *anth(-r)- 'coal': Arm. ant'-el 'hot coal, ember' (tazar P'arpec'i, Hexaemeron, etc.), dial. *ant'(e)l-oc' 'metal rod for poking or stirring a fire, poker', dial. *ant'-(a)r- 'coal, ember' in *ant'-r-oc' and *ant'-ar-oc' 'poker' (note also ant'ayr 'spark' in Bargirk' hayoc' and NHB, probably from *ant'-ar-iV-); Gr. av&paE, m. 'charcoal'.123

6.2.2. *drepan-eh2: Arm. artewan-un-k', gen.pl. artewan-a-(n)c' 'eyelashes; eyebrow' (Bible+), Gr. Spenavq, Spenavov 'sickle' (from Spenw 'to pluck, cut off'). According to this fascinating etymology suggested by de Lamberterie (1983; 1992: 239; 2013: 22), the human eyebrow (and/or eyelash, see below) is taken as sickle-shaped.124

The basic meaning of artewanunk' is usually presented as 'eyelid' since it usually corresponds to Gr. fiAefyapov 'eyelid' in the Bible. NHB and HAB, however, describe the Armenian word as 'eyelashes', and so does de Lamberterie (1983: 21) in French, 'cils'. Indeed, in some

121 Martirosyan 2010: 329-331, 538-540.

122 Beekes 2010, 2: 1340-1341.

123 HAB 1: 194; Martirosyan 2010: 85; Jahukyan 2010: 57a; Beekes 2010, 1: 105 (here the dialectal forms in -rare not mentioned). For the dialectal forms, see NHB 2: 1060abc; Amatuni 1912: 4a; Acarean 1913: 48b, 98-99 HayLezBrbBar 1, 2001: 9a. The nasalless by-form at'ar-oc' may be due to folk-etymological association with at'ar 'dry dung used as fuel'. For the suffix -oc', see Olsen 1999: 533-537.

124 For a discussion, see Clackson 1994: 109-112; Olsen 1999: 296-297; Beekes 2010, 1: 353; cf. Jahukyan 2010: 96b. Aiayan (1974: 34) derives the word from *drep- 'to see'.

biblical passages 'eyelash' (or 'eyebrow') would make more sense than 'eyelid', e.g. Proverbs 6.25 (mi yap'stakic'is artewanambk' nora "do not be captivated with her eyelashes/eyebrows") or Jeremiah 9.18 (ew artewanunk' jer btxesc'en fur "and let your eyelashes drop water"). Note also that the derivatives of Gr. ¡3Ae<apov display a semantic vacillation between 'eyelid' and 'eyelash'. And finally, a few passages from original (non-translated) literary sources make the meaning 'eyelash' quite clear (e.g. maz artewanac' "hair of eyelashes").

In a remarkable passage from Movses Xorenac'i 2.42 we reed: "a multitude of vineyards resembled the beautiful crescent of thick lashes (zartewananc' xit ew getec'ik cir); on the northern side its curved form truly imitated the arching brows of charming maidens (getawor kusic' yonic')" (transl. Thomson 2006: 180). We can see that artewanunk' cannot mean 'eyelid' here since it is compared to vineyards. Nor does it mean 'eyebrow' since the latter is present here by its main designation, yonk'. There can be no doubt that Thomson's translation as 'eyelash' is correct.

6.2.3. *seph-s- or *seps- 'to boil, cook': Arm. ep'em 'to cook, boil'; Gr. 'to boil, seethe (of meat and the like); to smelt, refine (of metals)'.125

6.2.4. *t(a)rp-eh2: Arm. t'arp' / t'arb (abl. i t'arb-e) 'large wicker fishing-basket, creel', Gr. zapnri f., zapndg, zepndg m., zapndvr f. 'large wicker basket'; probably a common borrowing from a lost source.126

6.2.5. *ginyl(u)m- 'hinge': Arm. clxni, ea-stem (loc.sg. i ctxnw-of, gen.dat.pl. ctxn-e-ac'), cxni, cx/tan, dial. clxan 'door hinge'; Gr. yi(y)yAu^.og m. (dimin. yi(y)yAv^.iov n.) 'hinge, joint, pivot, gudgeon'. Mediterranean word (Martirosyan 2012). (HM)

6.2.6. *kalam- 'aspen; plane': Arm. kalamax(i) 'white poplar, aspen'; Gr. xaAay.iv-8ap• nAazavog r\8ovielg 'plane', obviously with *dar 'tree' (Hesychius); in neighbouring non-Indo-European languages: Salmast Turk. k'dldm-bdr 'aspen', T'avriz Turk. qdldmd 'poplar'; in Daghestan languages: Lak kalaxi, Rutul kalax 'aspen'. For the semantic relationship, cf. Arm. candar 'poplar' and 'plane'; op'i 'poplar, aspen' and tarabal *hop'i 'plane'. The ending -ax in Armenian may be a suffix, possibly seen also in mei-ex 'the handle of an axe' (if related with Gr. y.eAia 'manna ash, ashen spear') and taws-ax 'box-tree'. The correspondence Arm. k vs. Gr. k here and in a few of the lexemes that follow points to a later stage of Mediterranean substrate vocabulary.

6.2.7. *kast(an)- 'chestnut': Arm. kask-eni 'chestnut-tree' if from *kast-(u)k-eni (for the suffix, cf. hacar-uk and dial. hacar-k-i 'beech-tree'); Gr. Kaozavov n. 'chestnut', Kaozavea f. 'chestnut-tree'.

6.2.8. *karid- 'crayfish': Arm. karic, a-stem 'scorpion', dial. 'crayfish' < *karid-ia f.; Gr. Kapig, -i8og, -l8og (also Koupig, K<pig) f., probably a general term for small crustaceans, including shrimp and prawn. For the (old feminine) suffix *-iehi, note Arm. dial. *mormonf 'ant' < *mormon-ieh2 (cf. morm 'tarantula' and Gr. Mop^.<x>v 'bogey, bugbear', see § 6.4.8). Note also Arm. kor, gen. kor-i 'scorpion' (Dersim dial. gor-f), which is reminiscent of the Greek by-forms Koupig, K<<pig.

6.2.9. *gorio- 'drain': Arm. kori 'drain, channel', Gr. yopyvpiov n. 'subterranean channel'. (HM)

6.2.10. *gw(e)m/bhurieh2 'bridge': Arm. kamurj, a-stem 'bridge', Gr. ye<vpa f. (Boeot. fie<upa, Cret. 8e<upa, Lac. 8i<oupa) 'bridge'; in non-Indo-European languages: Hatt. hamuru(wa) 'beam', Abkhaz *qwd(m)bdld-ra 'beam over the hearth, cross-beam', etc. The Proto-Armenian theoretical by-form *kaburf- may have been reflected in Urart. qaburzani possibly meaning 'bridge' in a recently discovered inscription.

125 NHB 1: 705c; Hübschmann 1897: 446; HAB 2: 72-73; Arutjunjan 1983: 282-283; Clackson 1994: 172-173; Beekes 2010, 1: 492.

126 Clackson 1994: 183. For an etymological discussion, see Martirosyan 2010: 281-283. The Armenian form is absent from Beekes 2010, 2: 1453.

6.2.11. *mosgh-o/io- 'young bovine': Arm. mozi 'young bovine, calf',127 dial. mozi or diminutive mozik, mostly 'male or female calf', in some dialects: 'young ox', 'female foal, filly', 'young buffalo'; Gr. yooxoQ m.f. 'calf, young bull, any young animal', metaphorically 'boy' or 'girl, maid', m. 'young shoot or twig', yooxaQ, -aSoQ f. 'shoot, slip; heifer', yooxiaQ 'like a calf (used of any young animal); three-year-old ram'; diminutives: yooxiov 'young calf', yoox'tStov 'small shoot', yoox-apiov n. 'young calf'.128

Clackson (1994: 153-154) assumes that the word "is of later origin in Armenian, and it may even be a loan from Greek". This is not plausible, however. This Armenian word, albeit attested late, is reliably old since it is found in a great number of non-contiguous dialects, such as Hamsen, T'iflis, Ararat, Karin, Mus, Svedia and tarabal. Besides, Arm. -z- can hardly be explained from Gr. -ox-. In my opinion we are dealing with a Mediterranean word: *mosgh-o-'young bovine' (with Gr. yooxoQ m.f. 'calf, young bull, any young animal') > Parm. *moz(o) + -i as in other animal designations, such as ayci 'goat', mari 'female bird', mak'i 'ewe' (Martirosyan 2010 s.vv.), or directly *mosgh-iio-/-iia- (cf. Gr. yooxiov 'young calf', yooxiaQ 'like a calf; three-year-old ram', etc.) > Parm. *mozziyo/a-129 > Arm. mozi 'young bovine, calf'. For the semantic relationship between 'young shoot' and 'young animal', compare, e.g. ClArm. erinf 'heifer' > Ararat dial. erinf 'a three-year-old sprout of grapes'.130

6.2.12. *notiieh2 'wetness': Arm. nay, gen. nay-i 'humid, moist; wetness, moisture; (phonet.) liquid' (Dionysius Thrax, Book of Chries, Grigor Magistros, Esayi Nc'ec'i, etc.), nayac'uc'anem 'to wet, moisten, water, irrigate' (Philo, Book of Chries), nayakan 'humid, moist' (John Chrysostom); Gr. vozia, -ir\ f. 'wetness'.

The Armenian word is usually derived from PIE *sneh2- 'to swim': Skt. snati 'to bathe', Gr. ve<, vfx< 'to swim', Lat. nare, natare 'to float, swim', etc.131 The Armenian word may be derived from *n(e)h2-ti-, cf. Avest. u-snaiti- f. 'Abwaschung'. However, semantically more attractive is the comparison of Arm. nay with Gr. vozia, -irj f. 'wetness' (cf. Scheftelowitz 1904-05, 2: 24). According to Beekes (2010, 2: 1025), the latter should be separated from *(s)neh2- 'to swim' because of the vo-calism and may be Pre-Greek. Gr. vozia f. 'wetness' and Arm. nay, i-stem 'humid, wetness' may have been borrowed from a (substrate) proto-form like *notiieh2 > PArm. *notiya > *nayi(ya) > nay.

6.2.13. *(H)olur-: Arm. olorn, an-stem 'pea, bean', dial. hule(o)rnd (Goris, tarabal hulernd); Gr. oAvpai f. pl. 'spelt; rice-wheat' (cf. Akkad. halluru, hi/ulluru, etc.).

6.2.14. *osp- 'pulse, legumen': Arm. ospn, an-stem 'lentil', Gr. oonpiov n. 'pulse, legumen'.132

6.2.15. *pyortho- or *(t)por^o- 'sprout, young twig': Arm. ort', o-stem 'vine', Gr. mopOoQ m. 'sprout, shoot, young twig', nopOoQ• nzopOoQ, xAaSoQ, fiAaoxoQ (Hesychius).133

127 Attested in the 11th century commentary of Grigor Magistros on the Armenian translation of Dionysios Thrax (Adonc 1915: 240L7, 241L6):

128 The appurtenance of Skt. mahisä- 'great, mighty; buffalo', Lith mäzgas 'bud' and others is uncertain. For an etymological discussion, see Hübschmann 1883: 43; 1897: 475; Meillet 1898: 282; Patrubany 1902-03: 124; Scheftelowitz 1927: 226, 232; HAB 3: 338; Pisani 1950: 171; Jahukyan 1987: 139, 298-299, 302; Clackson 1994: 152-154; Olsen 1999: 489; Beekes 2010, 2: 970-971.

129 The pretonic *-o- has not yielded -a- because the syllable was closed due to the geminate -zz- (see Kortlandt 2003: 40; Beekes 2003: 157).

130 For more detail and other examples, see Clackson 1994: 230214; Martirosyan 2010: 264-265, 785-787.

131 HAB 3: 426-427; Pokorny 1959: 972; Frisk GEW 2, 1970: 310-311, 324-325; Schrijver 1991: 168-169; Mayr-hofer EWAia 2, 1996: 769-770; Beekes 2010, 2: 1012-1013.

132 NHB 2: 522a (s.v. ospneay); Olsen 1999: 141; Holst 2009: 126, 143, 188, 231. According to Katz (2000: 84-85), Gr. oonpiov derives from *uospr- 'having a shroud, covering'.

133 Petersson 1916: 271-273; Pokorny 1959: 823; Furnee 1972: 317, cf. 261; Kortlandt 1986: 40 = 2003: 70; Olsen 1999: 24; Beekes 2008: 52; 2010, 2: 1250; cf. Kloekhorst 2008: 645-646.

6.2.16. *khsan-t(e)r- 'wool-carder, comb': Arm. santr / sandr, ins.sg. santr-o-v (tazar P'arpec'i 3.61), abl.sg. i sandr-e (Ephrem) 'comb; weaver's comb', dial. sander-k' (Karin santr-e-k') 'weaver's large comb';134 Gr. £aivw 'to card, comb wool', £av%rg m. 'wool-carder'E,avxpiai (title of a play by Aeschylus).135 In view of the incompatibility of the Armenian initial s- with Greek *ks- in Indo-European terms, this comparison is considered to be uncertain.136 I assume that we are dealing with a substrate word: *khsan-t(e)r- > Arm. sandr, pl. sander-k'.

6.2.17. *si/ekhu- 'melon, gourd': Arm. sex (gen. sexoy in Hexaemeron) 'melon', Gr. oiKva, Ion. -vr f. 'bottle-gourd, round gourd; gourd used as a calabash', oeKoua 'id.' (Hesychius), oiKuog, oiKuOg m., oiKug f. 'cucumber', oiKuog nenov 'a kind of gourd or melon, not eaten till quite ripe'. The relationship with Russ. tykva 'pumpkin' and Lat. cucumis 'cucumber' is unclear.

6.2.18. *keno-/*ken(e)uo-: Arm. sin, o-stem 'empty', Gr. Att. Kevog and Ion. Keivog from *Kevpdg, Epic Keve(p)0, 'empty, idle'.

6.2.19. *ste/oibo- or *ste/ibeh2: Arm. step, o-stem, a-stem 'haste, alacrity; zeal, diligence; frequent, frequently; hastily, quickly, stipem 'to constrain, compel, force; to urge, hasten'; Gr. ozeifiw 'to tread (on something), densify by treading, trod, trample, trend', ozoifif f. 'stuffing, cushion, bulge', ozifiog m. 'trodden road, path, footstep, trail'. The appurtenance of OLith. staibus 'strong, brave' and other cognates is uncertain.137

6.2.20. *srungh- 'snout, nostrils': Arm. rungn, mostly pl. rng-un-k', instr. rng-am-b-k', *rung-k', a-stem 'nostrils'; Gr. pvyxog, pvyxeog n. 'snout (e.g. of a pig), muzzle, beak'.

6.2.21. *ps(e)ud-e/os-: Arm. sut, o-stem 'false; falsehood, lie', Gr. tyev8og n. 'lie', also ^v8og.

6.2.22. *skorp-i-, gen. *(s)krp-i-os: Arm. k'arb, i-stem 'basilisk, asp'; Gr. oKopniog m. 'scorpion; a sea-fish', oKopnig, -i8og f. 'a sea-fish'. These words have been claimed to be derived from IE *(s)ker(-p)- 'to cut': Arm. k'er-(t')-, k'er-b/p'- 'to scratch, chop, carve', Gr. Keipo 'to cut (off), shave, mow off, ravage', OHG sceran 'to cut', OEngl. sceorfan 'to scratch', etc. However, scholars are now more inclined towards a substrate origin.

6.3. Armenian, Greek and Albanian.

6.3.1. *h2n(e/or)io- 'dream': Arm. anurj-k', i-stem, o-stem 'dream, day-dream, prophetic vision, vision'; Gr. oveipog m. 'god of dreams, dream', Aeol. ovoipog m., Cret. avaipov oveipov, cf. ovap n. 'dream', especially 'fortune-telling dream, vision', avap• ovap (Hesychius); Alb. dderre (Geg.), enderre (Tosc.) 'dream' from *anderre < *Hnr-io/a-. Probably derived from PIE *h2enhi- 'to breathe' > '(vital) breath, energy' (de Lamberterie 2012a).

6.3.2. *bhe/or-(e)n- 'load': Arm. bern, gen. berin, ins. beram-b, vom.pl. berin-k' 'burden, load; bag, sack; freight, cargo';138 Gr. <epvfj f., Aeol. <epeva 'dowry', Dor. <epva f. 'god's share at the sacrifice', Alb. barre 'burden, load; freight, load; foetus' < *bhor-neh2; with a different meaning: Lith. bernas 'boy, (farmer's) servant', Latv. bprns 'child, baby'; with o-grade: Goth. and OIc. barn n. 'child' < "what was borne".139 This word is a verbal noun from PIE *bher- 'to bring, bear'.

The Greek meaning 'dowry' probably derives from 'load, bag' (both nuances are attested in Armenian). Concerning the meaning of Dor. <epva, 'god's share at the sacrifice',

134 See Acarean 1913: 954; HAB 4: 174-175; HayLezBrbBar 5, 2008: 266.

135 Liddell / Scott / Jones 1996: 1188a.

136 Hübschmann 1897: 488; HAB 4: 174-175; Jahukyan 2010: 668a.

137 HAB 4: 273-274; Olsen 1999: 196; Beekes 2010, 2: 1393-1394.

138 In the Bible, bern corresponds to ^opxtov 'burden' (Job 7.20), ßaaray^a 'load' (Jeremiah 17.21), ^.apainnog 'bag' (Genesis 44.11, 13), yo^ot; 'freight, cargo' (Revelations 18.11), etc.

139 Hübschmann 1897: 429; Pedersen 1905: 217 = 1982: 79; HAB 1: 440a, 441a; Pokorny 1959: 129; Chantraine 1968-80: 1180; Stempel 1990: 51; Clackson 1994: 111, 135, 183; Demiraj 1997: 92-93; Olsen 1999: 120-124, 140, 833834; Beekes 2010, 2: 1562. For Slavic *berm§ 'load, burden', see Derksen 2008: 37.

compare the semantic shift 'share' > 'dowry' (cf. Arm. bazin 'share' > bazin-k' 'dowry', Marti-rosyan 2010: 803).

We can postulate *bher-(e)n- 'load' ('that is borne'), a verbal noun from PIE *bher- 'to bring, bear' (cf. Arm. berem, Gr. fyepw, etc.) shared by Armenian, Greek, Albanian, Baltic and Germanic. The Armenian is particularly close to the Greek (e-grade and semantics) and is seman-tically identical with the Albanian. They point to *bhe/or-neh2 'load, freight', with a secondary transfer to the n-stems in Armenian (cf. eln 'deer cow', § 6.7.3) or *bhe/or-(e)n-.

6.3.3. *ghri(dh) 'barley': Arm. gari, ea-stem, o-stem 'barley'; Gr. xpi§-f f. 'barley-corns', usually pl. 'barley', from an original root noun *xpi& > Ep. xpl n.; Alb. drithe 'cereals, wheat, barley'. A different formation: Lat. hordeum 'barley', OHG gersta 'barley', and Hitt. karas n. 'wheat, emmer-wheat'.140

6.3.4. *sk(h)odoro- or *sk(h)orodo- 'garlic': Arm. xstor, i-stem and o-stem, sxtor 'garlic'; Gr. oxdp(o)Sov n. 'garlic', Alb. hurdhe, also hudher (Schriftsprache) f. 'garlic'.

6.4. Armenian, Greek and Latin.

6.4.1. *bhug/g- 'dewlap': Arm. erbuc, o-stem 'breast of animals', Gr. fyapvyE,, gen. -vyoQ, -vyyoQ 'throat; dewlap of a bull', Lat. frumen 'throat' < *frug-smen.

6.4.2. *el(e/a)iw- vel sim. 'olive, oil': Arm. ewl, gen. iwl-oy, dial. *el 'oil'; Gr. ¿Aaia, Att. ¿Ada, Ion. ¿Aaiq f. 'olive-tree; olive', eAaioQ m. 'wild olive', eAaiov n. 'olive-oil; anointing-oil; any oily substance'; Lat. oleum n. 'olive-oil; oil'.

6.4.3. *ptel- 'elm': Arm. t'eli 'elm'; Gr. meAe-a, Ion. -r\ 'elm, Ulmus glabra'; cf. also Lat. tilia 'linden'.

6.4.4. *thuoiko- or *tu(i)ko- 'fig': Arm. t'uz, o-stem 'fig', dial. (Aslanbek and Ozim) 'female genitals'; Gr. ovxov, Boeotian zvxov n. 'fig; pudenda muliebria, female genitals'; Lat. ficus, i and us, f. 'fig; fig-tree'.

6.4.5. *h1ul(e/o)h\r-o- n.pl. *-h2 'rope, thong, rein': Arm. lar, o-, i- and a-stems 'rope, rein, cable, cord, string; plumbline of stone-masons; snare; tendons of the neck; string of a musical instrument'; Gr. evAqpa, Dor. avAqpa, Hesychian apAqpa n.pl. 'reins'; Lat. lorum, -i, n. 'thong, rawhide whip, rein'. We can reconstruct a Proto-Armenian paradigm *uldr-o-, pl. *uldr-a.

6.4.6. *g\gt- 'milk': Arm. kat'n, gen. kat'in, instr. kat'am-b 'milk', Gr. yaAa, yaAaxxoQ n. 'milk', Lat. lac, lactis n. 'milk'. The *-l- has been preserved in the Armenian dialects of Agulis and Melri, where we find kaxc' pointing to *kalc' (the development a > Agulis o has been blocked in position before l). Kak'avaberd has kaxc' in the village of Varhavar vs. kdt'nd in the other three villages of the region. The conditions responsible for the loss or preservation of the *-l- are not clear.141 Nevertheless, I do not think that this comparison should be abandoned.142 We are probably dealing with a cultural word of Mediterranean origin.

6.4.7. *mor- 'blackberry': Arm. mor, gen. mor-i'blackberry (fruit of the bramble)', mor-(en)i 'bramble, blackberry (plant, shrub)', dial. mor-m-eni 'blackberry', mo(r)s(-i) 'tamarisk; blackberry, bramble'; Gr. yopov n. 'black mulberry; blackberry', yopea, -er\ f. 'mulberry-tree, Morus nigra'; Lat. morum, i, n. 'fruit of the black mulberry', morus, i, f. 'black mulberry-tree' (sometimes considered a Greek loanword).143

140 Demiraj 1997: 145-146; Orel 1998: 75; de Vaan 2008: 288-289; Kloekhorst 2008: 444-445; Martirosyan 2010: 199; Beekes 2010, 1: 779.

141 For references and a discussion, see Martirosyan 2010: 345-346.

142 Pace Olsen 2011: 24. The etymology is accepted in de Vaan 2008: 320 and Beekes 2010, 1: 256.

143 The Celtic forms (Welsh mer-wydden 'mulberry, blackberry', Olr. smer 'blackberry', etc.) point to a different proto-form, namely *smero- (Matasovic 2009: 347).

6.4.8. *mor-m- 'she-monster, spinning demon/goddess': Mid Arm. and dial. mor-m and mor(i) 'spider, tarantula, phalangium'; Gr. Mopy.<, -oog -ovg, Mopy.<v, -0vog f. 'she-monster, bogey' (also used by nurses to frighten children), generally 'bugbear', Lat. formido, inis f. 'fear, terror; a thing which frightens, bogey'. (HM)

The Greek and Latin words are related, either etymologically or secondarily, with the word for 'ant', cf. Lat. formica f. 'ant', Gr. ^.vp^rE, -IKog, Dor. -aKog m. 'ant; fabulous

animal in India' (by-forms: y.vpy.og, fivpy.a£, fidpy.a£, op^iKag), etc. This connection or conflation becomes quite transparent in view of the following forms and meanings: y.upy.fK-eiov n. a species of <aAayyiov, the latter being 'a kind of venomous spider, especially Lathrodectus or malmignatte', y.upy.fK-iov n. 'a species of spider'. Note also y.dpy.opog and y.vpy.og 'panic fear' (glossed by <dj3og in Hesychius), the former of which strikingly resembles Armenian dialect of Polis/Stambul *mormoroz, Crimea and Nor Naxijewan *mrmras 'Easter bogey'.

A similar kind of conflation is seen in some dialectal forms of mrfiwn, the Armenian word for 'ant': Lori mormonf and Samaxi mormorinf. Since Gr. Mop^.<v is feminine, one may identify it with Lori mormonf, which probably reflects fem. *mormon-iehi. Structurally, compare another insect-name of Mediterranean origin: karic, a-stem 'scorpion', dial. also 'crayfish' < *karid-ieh2, cf. Gr. Kapig, -i/l8og f. 'Crustacea' (§ 6.2.8). That 'ant' is associated with 'bogey, ghost' is not surprising. According to Armenian folk beliefs, the ant, sometimes called a 'devil', is an evil night animal. Like the snake, frog and other fauna, it causes the skin disease called mrfm-uk 'little ant', cf. Gr. ^up^rK-ia 'wart that spreads under the skin, also the irritation caused thereby, which was compared to the creeping of ants' (from ^vp^rE, -IKog 'ant') vs. ^upy.fK-iov n. 'a species of spider'.

It is remarkable that Armenian has both *mor- and *mor-m, whereas Greek and Latin only display forms going back to *mor-m-. The root *mor- is probably related with European forms reflecting *mora-: OIc. mara, OHG mara 'nightmare', Germ. Mahr 'nightmare', Engl. (night)mare; OIr. mor-(r)igan 'lamia', lit. 'Alpkonigin'; Bulg. mora 'nightmare', dial. 'evil spirit', SCr. mora 'nightmare, incubus', mora 'a kind of night butterfly', Czech dial. mora 'night butterfly; a mythical evil creature', Russ. dial. mora 'a female mythical being which in the night, under the moonlight, spins the yarn that has been left unfinished by a woman', Russ. kiki-mora 'a kind of brownie who spins in the night', etc.144 We may posit European substrate *mor-a- 'nightmare; a mythical creature; an insect that is associated with mythical beings (butterfly, spider. etc.)' vs. Mediterranean substrate (probably a broken reduplication145) *mor-m- 'she-monster, bogey; spider, tarantula'.

Armenian thus takes an intermediary position because it has both forms. We may tentatively assume a PArm. *mor(m) referring to Lady of the Beasts, an Athena/Artemis-like female mythical personage, a protector of weaving and spinning (cf. the semantics of Russian mora) and personified as a spider. Note also Armenian dialectal mamuk 'spider', literally 'grandmother'. That this goddess is continued by a mythological figure of lower rank is not surprising. A very clear case is that of the Armenian goddesses Astlik and Anahit, attested as being of the state pantheon in Classical Armenian sources, and as female spirits or nymphs in a few later sources. In some folk tales, Anahit is represented as a wise queen who makes wonderful rugs, or the daughter of a dragon. Compare also Russian Mokosi, a female divinity who spins wool and whose name comes to denote 'evil monster'.146

144 For the forms reflecting *mora, see Pokorny 1959: 736; EtimSlovSlavJaz 19, 1992: 211-214, cf. 17, 1990: 204207; HerkWort 1997: 434b; Derksen 2008: 324-325; Matasovic 2009: 278. For the forms reflecting *mor-m-, see Pokorny 1959: 749; de Vaan 2008: 234-235; Beekes 2010, 2: 967. Both are discussed in Nocentini 1994: 399-401. For an etymological discussion of the Armenian word, see Martirosyan 2010: 478-480, 787.

145 Cf. Gamkrelidze/Ivanov 1984, 1: 222 = 1995, 1: 191.

146 For more data on this and a thorough discussion, see Martirosyan prepar. 3.

6.4.9. *sphongos / *phsongos 'sponge, mushroom': Arm. sunk/gn '(tree-)mushroom', Gr. ondyyoQ, ofyoyyoQ m. 'sponge; any spongy substance, e.g. tonsils', Lat. fungus m. 'fungus, mushroom'.

6.5. Armenian, Greek and Germanic and/or Celtic.

6.5.1. *breh1-ur (obl. *bhrun-): Arm. albewr, albiwr, r-stem: gen. alber 'fountain, spring', Gr. fypeap, -azoQ 'an artificial well, spring'; cf. Goth. brunna, etc.

6.5.2. *dhorgh-/*dhrogh-: Arm. durgn, gen. drgan 'potter's wheel', Gr. zpoxoQ m. 'wheel; potter's wheel', OIr. droch 'wheel'. Notwithstanding the formal difficulties, this etymon may be derived from IE *dhfgh-'to turn': Arm. darj-, darnam 'to turn; to return' < *darj-nam, cf. Gr. zpex< 'to run, hurry', etc. For another cultural term of a similar structure, cf. burgn, gen. brgan 'tower; pyramis' vs. barnam 'to lift, raise' < *barj-nam (see § 6.1.10).

6.5.3. *trso-/*tors-eh2: Arm. t'ar 'perch, roost for birds' (MidArm.; ubiquitous in the dialects), 'bar for drying grapes' (Mid Arm.) < *trsos: Gr. zapooQ, Att. zappoQ m. 'frame of wicker-work, crate, flat basket for drying cheeses on', zpaoia, zapoif, zepoia (-e- after Tepoopai) f. 'hurdle for drying figs; dried figs; drying-place for corn, cheese or bricks'; Germ. *torsa: OHG darra f. 'apparatus for drying fruits, etc.', Swed. and Norw. tarre m. 'frame for drying malts, etc.'.147 This implement designation is usually derived from PIE *t(e)rs- (cf. Skt. tars- 'to be thirsty, crave', Gr. zepooyai 'to become dry', OHG derren 'to make dry', Hitt. tars- 'to make or become dry', Arm. t'aramim / t'arsamim 'to wither', etc.),148 although there are phonological difficulties in Greek (Beekes 2010, 2: 1453-1454). We may posit an innovation shared by Armenian, Greek, and Germanic.

6.5.4. *iork-o- 'deer, roe, game': Arm. ors, o-stem 'hunt, game'; Gr. SopxaQ, -aSoQ f. (Herodotus 7.69), ZopxaQ (Herodotus 4.192), SdpE,, SopxoQ, Z,op<i, lopxoQ, etc. 'a kind of deer, roe, antelope, gazelle'; Corn. yorch 'roe', MWelsh iwrch 'roe-deer (caprea mas)'. (HM)

6.5.5. *ki(u)on 'column, pillar': Arm. siwn, gen. sean 'column, pillar'; Gr. xiwv, -ovoq 'column, pillar; flogging post', Myc. ki-wo-qe 'and a pillar'. According to Clackson (1994: 141, 142143), *kiuon represents a borrowing into Greek and Armenian from a lost non-Indo-European source. Recently,149 these words have been derived from PIE *(s)kiHu- 'shin', which is attested in Balto-Slavic and Germanic languages, cf. Russ. cevka 'bobbin, hollow bone, shin-bone', OCS cevbnica 'flute', OEngl. scia 'shin, leg', MHG schie 'post', OHG skena, skina 'post', etc. Lubotsky (2002) connected these words to Skt. asthivd(nt)- 'shin, shank' with Av. ascuua- (attested in ASg. ascum) 'shank', reconstructing an Indo-Iranian *HastciHua-. He interprets this form as a compound of the word for 'bone', viz. Skt. dsthi- and Av. ast-, with a reflex of PIE *(s)kiHu- 'shin'.

The second member of the Indo-Iranian compound, viz. *ciHua-, is not attested anywhere independently. However, the existence of Iranian *civa- 'shank' can be proven by Arm. ciw 'shank, leg'. This is attested in Classical Armenian only in the compounds men-a-ciw 'having one hoof' (translating Gr. yovavvxoQ) and erk-a-ciw 'having two hoofs'. Further, civ is attested in Middle Armenian and in dialects, mostly meaning 'leg', 'shin, shank'.150

As for the semantic development from 'shin, shank' to 'pillar, post', Lubotsky mentions a number of parallels: Engl. shank 'shank' and 'shaft of a column'; Latv. stulps 'shank' and 'post, pole'; OEngl. scia 'shin, leg', scinu 'shin', MHG schie and OHG skena, skina 'post'.151 The last set

147 HAB 2: 154-155; Arutjunjan 1983: 284-285.

148 Martirosyan 2010: 281. On Hittite, see Kloekhorst 2008: 848.

149 Praust apud Lubotsky 2002: 323b.

150 Martirosyan 2005; 2010: 579-580, 803-804.

151 Note also Arm. dial. Moks srungy 'the stem ends of wheat remaining attached to the soil after mowing (stubble)' from srunk' 'shin, shank'; Oss. zxng / zxngx 'shin' and 'stalk' (Martirosyan 2010: 585-586, 804).

of words is etymologically related to the etymon being discussed here. Note that Germanic *skino 'post' is reminiscent of the n-stem in Armenian siwn and Greek Kiwv 'column, pillar'.152 It is remarkable that we have yet another lexical correspondence with the same meaning and dialect distribution, namely *st\neh2 'post, pillar': PArm. *stat-a-, Gr. ozfAr, OHG stollo (see § 6.5.7).

If this attractive explanation is accepted, then this is a shared innovation between Armenian, Greek, and, somewhat distantly, Germanic. Otherwise, it should be treated in the section on substrate.

6.5.6. *keudh- 'to hide': Arm. soyz-: suzanem, 3sg.aor. e-soyz 'to plunge, sink; to hide, cover' (probably from sigmatic aorist *keudh-s-);153 Gr. KevOw, aor. Kevoai 'to cover, hide; to be concealed, lie hidden', KeuOfveg 'subterranean deities'; OEngl. hydan 'to hide' from *hudjana-, Goth. huzd 'treasure' from *kudh-to-.154 Skt. kuhu- f. 'new moon' and others are hardly related.155

6.5.7. *st\neh2: PArm. *stalna- > *stal-a- 'post, pillar' in ara-stal, a-stem 'ceiling, roof'; Gr. ozfAr 'block or slab used as a memorial; monument, gravestone; post, pillar; boundary-post'; OHG stollo, MHG stolle 'support, post'. Probably derived from IE *stel-: Gr. ozeAAw 'to put in order, equip, prepare', OHG stellen 'to array, establish, arrange', OCS stblati 'to spread', etc.156 Note the semantic closeness to another agreement between Armenian, Greek and Germanic: Arm. siwn, Gr. xiwv, and OHG ske/ina (see § 6.5.5).

6.5.8. *dig/gh- 'goat': Arm. tik, a-stem 'wineskin, a vessel made of an animal's skin (for wine, oil, water, etc.)' < 'goat, skin of goat', dial. Agulis tagy/taygy pointing to an older *tig; Gr. Laconian SiZa■ ai£. AaKwveg (Hesychius), unless a corruption for *alZa; Germ. *tigo 'goat, he-goat': OHG ziga, Germ. Ziege, etc.). The vacillation *-g/gh- points to a non-Indo-European origin; a well-known root structure constraint does not permit two unaspirated voiced stops within a root. Compare Arm. kacan 'path' vs. SCr. gaziti 'to step, trample, wade', Arm. karkut 'hail' vs. OCS gradb and Lat. grando 'hail'.

6.5.9. *p(o)Hl- or *p(o)lH-: Arm. ul, o-stem (secondarily: u-stem) 'kid' (probably also amul 'childless', as well as al-oj 'female kid' from a zero-grade form, with an ending comparable to orof 'lamb'); Gr. ntiAog m. f. 'young horse, foal, filly', secondarily also of other young animals, metaph. 'young girl, youth'; Goth. fula, OHG folo 'id.'; Alb. pele 'mare' < *pol-n-.157

6.6. Armenian, Greek and Balto-Slavic.

6.6.1. *h2(e)gH- 'darkness, fog, mist': Arm. alj- 'darkness, fog, twilight' (aif-a-muif 'darkness', aif-ut'iwn-k' 'darkness', aif-aif 'fog'); Gr. axAvg, -vog f. 'mist; darkness', OPr. aglo, u-stem n. 'rain'. Arm. *aif- probably reflects a frozen locative *h2(e)ghl-i > PArm. *algi (regular metathesis).

6.6.2 *h2(e)rti 'now, near' (probably a locative formation from the root *h2er- 'to fit together, join', with the original meaning 'fittingly, suitable, at hand'): Arm. ard(i) 'now', dial. ard-ak

152 Beekes 2010, 1: 707.

153 For literature and other examples of sigmatic aorist in Armenian, see Martirosyan 2010: 757 and s.vv.

154 For the etymology and various explanations of Arm. -z, see Bugge 1893: 38-39; Pedersen 1906: 381-382, 425 = 1982: 159-160, 203; de Lamberterie 1978: 281; Mallory/Adams 1997: 268b; Olsen 1999: 782; Kortlandt 2003: index; Kocharov 2008: 100; Beekes 2010, 1: 682. The solution with the sigmatic aorist (*-dh-s- > -z-) is advocated by Pedersen, Kortlandt and Kocharov; Kocharov explicitly cites Gr. aor. Kevaai in this context. Some scholars are sceptical about the etymology: Meillet 1925: 5; HAB 4: 241-242; Clackson 1994: 116; Jahukyan 2010: 687a. On Germanic, see also Mayrhofer 1986: 11678; Lehmann 1986: 196b.

155 See Mayrhofer EWAia 1, 1992: 383; cf. Beekes 2010, 1: 682.

156 Martirosyan 2010: 107-108; Beekes 2010, 2: 1397-1398, 1404; for Slavic, see Derksen 2008: 473.

157 Lehmann 1986: 130b; Clackson 1994: 183; Demiraj 1997: 314; Martirosyan 2010: 15-16, 53, 631-632. Sceptical on the etymology: Olsen 1999: 185 and Beekes 2010, 2: 1266.

'immediately, on the spot'; Gr. apzi 'just now', apzioc 'suitable; ready'; Lith. arti 'near' (referring to proximity of space rather than time).158

6.6.3. *dghuH- 'fish': Arm. jukn, gen. jkan, pl. jkun-k' 'fish'; Gr. ix&vg, -voq m. 'fish'; OPr. suckans, Lith. zuvis, Latv. zuvs 'fish'.

6.7. Armenian and Greek in a broader European context.

6.7.1. *h1o(h1)i-ueh2: Arm. aygi, ea-stem 'vineyard; vine', Gr. o'lq f. 'service-tree', Lat. uva 'grapes', Russ. iva 'willow', Czech jiva 'willow', SCr. iva 'willow', Lith. ieva 'bird-cherry', OIr. eo m. 'yew', OHG iwa f. 'yew', OPr. iuwis 'yew'. The appurtenance of Hitt. GI$eian- n. 'a tree (evergreen)', perhaps 'yew' is uncertain (Kloekhorst 2008: 233-234).

6.7.2. *ureh2d- 'branch; root': MidArm. and dial. argat 'superfluous branches cut off from the vine and used for kindling', Gr. pdSiE, m. 'branch, twig', Lat. radix f. 'root; radish', ramus m. 'branch, twig' if from *wrad-mo-, MWelsh gwreid < *urh2d-io- 'roots', OIc. rot, Goth. waurts 'root', Alb. rrenj/e, -a (Tosk), rra(n)je (Gheg) 'root'. The appurtenance of Toch. B witsako (from *urdi-k-eh2-?) is uncertain.

6.7.3. *h1el-(h1)en- or *h1el-no-: Arm. eln, nom.pl. elin-k', gen.pl. elan-c' 'deer cow, hind'; Gr. ¿AAoq 'deer-calf, fawn' < *h1el-no-, eAafyoc, m. f. 'deer; deer cow, hind' < *h1el-n-bho-; cf. also *h1el-h1en-i- 'deer, hind': OCS jelenb 'deer', aHnii 'doe', SCr. lane 'doe', Russ. lan' 'fallow deer, doe', olen' 'deer, stag-beetle', dial. elen' 'deer, stag-beetle', Lith. elnis 'deer'; further: MIr. ailit f. 'doe, hind' < *hel-(H)n-t-iH- or *hel-en-t-iH-, etc.

If Arm. eln derives from *h1el-no- (with Gr. ¿AAoq 'deer-calf, fawn') rather than *h1el-h1en-(with Balto-Slavic), it parallels bern, pl. berin-k', gen. beran-c' 'burden, load' from *bhe/or-neh2: Gr. fyepvrj f. 'dowry' (see § 6.3.2).

6.7.4. *lu(n)k- 'lynx': Arm. *lusan-n (pl. lusanunk') 'lynx', dial. *lus(e)amn also meaning 'hyena' and 'marten'; Gr. AvyE„ gen. Avyxdc (-yydc) 'lynx'; Lith. lusis, dial. (Zem.) ltnsis, ltnsi, OPr. luysis, Russ. rysb; MIr. lug; OHG luhs 'id.'.

6.7.5. *glieh2 / *glineh2 'glutinous substance, clay': MidArm. and dial. kaljin 'mortar, clayey soil'; Gr. yAia f., yAoidc m., yAivq 'any glutinous substance, gum'; Russ. glej (dial.) 'clay, loam', glina 'clay', Ukr. hlej 'moist clay', Lith. gleine 'moist clay'; OEngl. clxg, Engl. clay and MLG klei 'clay' from Germ. *klaiia-, etc. (HM)

6.7.6. *kiker- 'chick pea': Arm. sisern, gen. sis(e)ran 'chick pea', Gr. xixeppoi 'bird's pease' (He-sychius: Maced.), Lat. cicer n. 'chick pea', OPr. keckers 'chick pea', Alb. thjer(r), thierr 'lentil, Er-vum lens'.

7. Armenian, Greek and the Mediterranean/European substrate

In recent years, the methodology of dealing with substrate words has been developed and applied by several scholars.159 It has been pointed out that an etymon is likely to be a loanword if it is characterized by some of the following features: (1) limited geographical distribution; (2) unusual phonology and word formation; and (3) specific semantics.160

The Armenian words that are frequently considered to be of Mediterranean origin are: gini 'wine', ewl/iwl 'oil', t'uz 'fig', spung 'sponge', sring 'pipe, fife', sunk/g(n) 'mushroom'.161 The

158 See Clackson 1994: 103-104; Martirosyan 2010: 135.

159 Kuiper 1995; Beekes 1996, 1998, 2000, 2003; Schrijver 1997; Lubotsky 2001a.

160 See Schrijver 1997: 293-297; Beekes 2000: 22-23; Lubotsky 2001a: 301-302.

161 In fact, gini 'wine' and spung 'sponge' should be excluded from the list since the Indo-European origin of the former is more probable, and the latter is likely to be a Greek loan.

actual number is much higher. In Martirosyan 2007 and 2010, I have applied the aforementioned methodology to a number of such words, mostly plant names, animal names and cultural words. In these cases, an etymon is attested in Armenian, Greek, Latin and/or another Indo-European language of south-east Europe (such as Albanian or Phrygian) or Anatolian, but the phonological or word-formative correspondences are irregular with respect to the Indo-European system, and they cannot be considered loanwords from one another.

Bearing in mind that Greek and Latin on the one hand and Armenian on the other are historically located on the opposite sides of the Black Sea, as well as that in some cases Mediterranean words have related forms in the Caucasus and Near East, I prefer not to confine myself strictly to the notion of so-called Balkan Indo-European. I conventionally use a term Mediterranean-Pontic Substrate. In some cases (e.g. Arm. pal 'rock' vs. Gr. neAAa 'rock', OIr. ail 'cliff' < *pal-i-, MIr. all < *p\so-, OIc. fell 'mountain, rock', OHG felisa 'rock, cliff' prob. from *palis-), an etymon is also present in other European branches, such as Celtic and Germanic, thus we are faced with the European Substrate in the terms of Beekes 2000. Whether the Mediterra-nean-Pontic and European substrata are identical or related is difficult to say with confidence.

There are words belonging to the same semantic categories (plant names, animal names, cultural words) that may be treated as innovations shared by Armenian and Greek etc. For instance, the morphological agreement between Arm. kalin, o-stem 'acorn' and Gr. fiaAavog f. 'acorn' (vs. Lat. glans, glandis f. 'acorn, beech-nut', Russ. zelud', SCr. zelud 'acorn', Lith. gile, dial. gyle 'acorn', Latv. zile 'acorn', etc.) may reflect a common innovation undergone jointly by Greek and Armenian (Clackson 1994: 135-136, 200/2372). Such words do not belong with the substrate since they are of Indo-European origin and do not reflect any phonological or morphological deviation. Nevertheless, these innovations are relevant to our topic in that they may be ascribed to the same Mediterranean-Pontic area and period. In other words, after the Indo-European dispersal, Proto-Armenian, Proto-Greek and some other contiguous language-branches (e.g. Phrygian and Thracian162) may have remained in contact somewhere in the Mediterranean and/or Pontic areas, probably in the 3rd and 2nd millennia bc and have shared both IE innovations and substrate words.

The consonantal correspondences between substrate words in Armenian and other languages are of two kinds:

1) archaic, matching the correspondences of the native Indo-European heritage:

*-ri- > Arm. -rf- and *g/gw > Arm. k, e.g. Arm. anurf 'dream' vs. Gr. Ove/oipog, Arm. kamurf 'bridge' vs. Gr. yetyvpa;

*k > Arm. s, e.g. Arn. sisern 'chick pea' vs. Lat. cicer 'chick pea', Arm. siwn 'column, pillar' vs. Gr. xiwv;

*g > Arm. c, e.g. Arm. erbuc 'breast of animals' vs. Gr. <\>apuyE, gen. -uyog, -uyyog 'throat, dewlap';

*p- > Arm. h- or zero, e.g. Arm. aiawni (*aiawun), ea-stem 'pigeon, dove' vs. Lat. palumbes 'wood-pigeon, ring-dove' (*p\h2-bh-on, gen. *-bh-n-os); Arm. hec', gen. hec'-i 'felloe', if from *pelk-s, cf. OHG felga, OEngl. felg(e) 'felloe', etc.; Arm. ort', o-stem 'vine' vs. n(x)6pOog 'sprout'.

2) relatively young:

*k > Arm. k, e.g. Arm. kaiamax(i) 'white poplar, aspen' vs. Hesychian KaAa^ivSap 'plane'; karic, a-stem 'scorpion', dial. 'crayfish' vs. Gr. Kapig, -iSog, -ISog 'crayfish';

*p- > Arm. p, e.g. Arm. pal 'rock' vs. OIr. ail 'cliff' < *pal-i-, MIr. all < *p\so-, Gr. neAAa 'rock';

162 Cf. Kortlandt 2003: VIII, 83-87.

*s > Arm. s (unless these words have been borrowed from lost satam-forms in *k), e.g. Arm. sayl, ¿-stem and o-stem 'wagon; Ursa Major and Minor, Arcturus' vs. Gr. oazivq f. 'chariot' and Hesychian oaziAAa nAeiat; to aazpov, the constellation being regarded as a car (considered to be of Phrygian or Thracian origin); Arm. sring 'pipe, fife' vs. Gr. ovpiyE,, -lyyoQ f. 'shepherd's pipe, panpipe', which is considered to be of Phrygian or Mediterranean origin.

This implies that we have to deal with at least two chronological layers, and that the Proto-Armenians must have remained in or close to the Mediterranean-Pontic areas for a long period of time.

In Table set B, the lexical correspondences of section 6 are grouped according to semantic fields. As in the Table set for sections 4 and 5, here also the correspondences that are likely to be innovations are marked by shading. The others probably have a substrate origin. Needless to say, all the lists in this paper are provisionary and are subject to corrections and additions.

If we collate the two sets of tables, we observe that both sets have a roughly equal number of lexical agreements in each semantic field, with a remarkable exception: in A, we find zero and five lexemes in the domains of flora and agriculture respectively, whereas B has 13 lexemes for each domain. Especially remarkable are sets of correspondences within a narrow semantic group, e.g. the three designations of plants of the legume family, all of Mediterranean origin: olorn 'pea, bean', ospn 'lentil', and sisern 'chick pea'. Interestingly, all three Armenian words display an additional -n and belong to the an-declension class.

Another remarkable difference is that, in the domain of technical activities, set A has lexemes with more general meanings, such as 'bond', 'grave' and 'threshold', whereas B displays a number of specific technical terms such as 'bridge', 'drying implement', 'hinge', 'pillar', 'potter's wheel' and 'rein'.

These sketchy conclusions probably indicate that, after the separation of the Indo-Iranians, Proto-Armenian remained close to Proto-Greek and some other dialects and, approaching Mediterranean or Pontic regions, developed a high number of lexical agreements, both innovations and borrowings from neighbouring non-Indo-European languages, especially in the domains of agriculture and technology.

Table set B (sections 6-7)

Lexical isoglosses: Armenian, Greek, etc.

Physical world, time, space.

Proto-form Gloss Armenian Greek Other 1 Other 2

*anth-r- coal *ant'-(a)r- àvdpaÇ

*Me)ghl~ mist, fog aij- àxAvç OPr. agio

*an(t)er cave ayr àvTpov

*h2(e)rti now, (near) ard(i) (Lith. artî)

*Héh2in-(ô)r day awr, g. awur f^ap

A *dueh2-ro- long erkar uô^àpOQ

B *dueh2-n- long erkayn us^àv

tUTUlnO- mound

*mar-mar- to shimmer *mar-m(a)r-

Proto-form Gloss Armenian Greek Other 1 Other 2

*me-ghsr-i near Véxpi

*notiieh2 wetness nay, ¿-stem voTÍa

*ken(e)uo- empty sin, -o- *Kevpôç

Human, age, kinship.

Proto-form Gloss Armenian Greek Other 1 Other 2

*pre(i)sgwu- elder

*gwnehiïk- wife, woman

*mehitruieh2 stepmother

*suekur-eh2 moth.-in-law

Body, perceptions, mentality, belief.

Proto-form Gloss Armenian Greek Other 1 Other 2

*hsk.wk.won eye akn ÖKKOV

*h2n(e/or)io- dream anur], -i-/-o- ôve/otpoç Alb*andërrë

*d.h{e)his- god di-k' §£0ç (Lat. feriae)

*pr(e/o)Hkt- buttocks erastan-k' npwXTÔÇ

*ues-nu- put on cloth. z-genum ëvvv^i

*g(e)lh2-s- laughter calr, g. calu yéAœç, yaA-

*mëd-es-h2 mind mit-k', mt-a- V^öea

*mor-m- she-monster mor-m Mop^á(v) Lat. formïdô

*srungh- snout rung(n) pvyx(e)oç

*ps(e)ud-os- false, lie sut, -o- ^evôoç

Movements, speech and other activities.

Proto-form Gloss Armenian Greek Other 1 Other 2

A *hsbhel- awel(-) o^éAAw

B *hsbhel- awel- o^éAAw

*bher-(e)n- load bern ^epvrf Alb. bârrë ('child')

*bhh2-ti- word,rumour bay, i-st. fyàotç, tftâTtç

*k(e)r(H)ieh2 band sari-k', -ea- Ke/aipia

*keudh(-s)- to hide soyz Kev9œ OEng. hydan

Fauna.

Proto-form Gloss Armenian Greek Other 1 Other 2

*hiel-(hi)n- deer Lith. élnis MIr. ailit

*pter- feather, wing

*lu(n)k- lynx lusa(m)n- AvyÇ Lith. lusis OHG luhs

A *karid(-ia) crayfish karic, -a- Kàpiç, -Uïôoç

B *ko/or-i- crayfv scorp. kor, -i scorp. Kovpiç Kœpiç

Proto-form Gloss Armenian Greek Other 1 Other 2

*dghuH- fish jukn ix$vç Lith. zuvïs

*h-ie/ogkïno- hedgehog ozni èxlvoç (OHG igil) (Oss. wyzyn)

■iork-o- deer, game ors, -o- ïopxoç Zopx- Corn. yorch

*sk(o)rp-i- asp, scorpion k'arb, i-stem axopnioç

Animal husbandry.

Proto-form Gloss Armenian Greek Other 1 Other 2

*hi(e)ig- goat ayc aï^ (Skt. eda-) (Av.ïzaëna-)

*bhrug/g- dewlap erbuc, -o- Lat. frumen

*äst- milk kat'n yáAaxi- Lat. lact-

*mosgh-o/io- young bov. mozi ^oax-

*h-ienhsorgk- male anim. y-orj ëv-opx-

*dig/gk- goat tik, -a- SiZa OHG ziga

■poHl-/pöl- young anim. ul, -o- nwAoç Goth. fula Alb. pélë

Proto-form Gloss Armenian Greek Other 1 Other 2

*h1oi(h1)ueh2 yew, vine aygi, -ea- oïn Lat. uva OHG iwa

*ureh2d- branch, root argat paSï£ Lat. radix Goth. waurts

A *dklh-lro- green, fresh dalar, -o- SaAepôç

B *dk¿l(H)-n- twig, herb del, -o- SaAAôç (MIr. duilne) (OEngl. dile)

■ptel- elm, Ulmus t'eli nxeAéa (Lat. tilia)

*fku(ö)iko- fig, (vulva) t'uz, -o- avxov, Tvxov Lat. ficus

*gwlhi-eno- acorn kaUn, -, (SCr. zelud)

■kalam- aspen, plane kaiam-ax(i) xaAa^iv-Sap

■kast(an)- chestnut kas(t-)k- xáaravov

*mor- blackberry mor(s) ^ópov (Lat morum)

Yor^o- sprout, twig ort',- o- n(r)ôpdoç

■hsorgH- orchis

*spongos mushroom sunk/gn an/^ôyyoç Lat. fungus

Agriculture.

Proto-form Gloss Armenian Greek Other 1 Other 2

alewr àAevpov

aiawri àAe%piç Iran*ar9ra-

■hios-r-(e)hi harvest ■ar-a- (on-)úpa (Goth asans) (CS jesem)

*drepan-ehi *sickle artewan, -a- Spenávq

*gkrt(dk) barley OHG gersta

*el(e/a)iw- olive, oil ewi, iwi-o- ëAaioç Lat. oleum

*sk(k)odoro- garlic xstor/sxtor axóp(o)Sov Alb. hùrdhë hùdhër

Proto-form Gloss Armenian Greek Other 1 Other 2

*gorio- drain kori yopyvpiov

*(H)olur- pea; spelt olorn, -ran öAvpai

*osp- legumen ospn 'lentil' öonpiov

*si/ekhu- melon,gourd sex aixva

*kiker- chick pea sisern, -ran (KÎxeppoi) Lat. cicer IPr. keckers Alb. thjer(r)

*{s)kl-ne/o- to hoe, weed (Lith. skélti)

House, housekeeping, crafts, implements, building.

Proto-form Gloss Armenian Greek Other 1 Other 2

*agu(s)iehr- path, road

*hier- fix, put tog.

*bhurgh- tower

hr ts * wheel II1B iiioi^^ IIB

*seph-s- to boil, cook ep'em ë^w

*t(a)rp-eh2 basket t'arp' / t'arb Tàpnn

*t(o)rs- drying impl. « OHG larra

*hiul(e/o)hiro- thong, rein lar, -o/i/a- evAripa Lat. lörum

*ginyl(u)m- hinge clxni, -ea- yi(y)yAv^.oç

*glieh2/glineh2 clay kaijin yAia, yAivq Russ. glina Engl. clay

*gwem/bhurieh2 bridge kamurj, -a- yé^vpa

*per-(i)on- awl, pin

r 1

*^san-t(e)r- comb sant/Ir Sâwp-

*(s)kt(u)ön pillar ssw OHG sk.na

*sttneh2 post, pillar

Preliminary conclusions

We may preliminarily conclude that Armenian, Greek, (Phrygian) and Indo-Iranian were dia-lectally close to each other or even formed a dialectal group at the time of the Indo-European dispersal. Within this hypothetical dialect group, Proto-Armenian was situated between Proto-Greek (to the west) and Proto-Indo-Iranian (to the east). On the northern side it might have neighboured, notably, Proto-Germanic and Proto-Balto-Slavic.

After the Indo-European dispersal, Armenian developed isoglosses with Indo-Iranian on the one hand and Greek on the other. The Indo-Iranians then moved eastwards, while the Proto-Armenians and Proto-Greeks remained in a common geographical region for a long period and developed numerous shared innovations. At a later stage, together or independently, they borrowed a large number of words from the Mediterranean / Pontic substrate lan-guage(s), mostly cultural and agricultural words, as well as animal and plant designations.

On the other hand, Armenian shows a considerable number of lexical correspondences with European branches of the Indo-European language family, a large portion of which too should be explained in terms of substrate rather than Indo-European heritage.

Literature

Abajyan, A. H.

1991 Anc'yal kataryali cagman harc'a. In: Banber Erewani hamalsarani 1991, Nr 1: 162-167.

Abeghian, Manuk

1899 Der armenische Volksglaube. Leipzig: Druck von W. Drugulin.

Acaryan, Hrac'ya Hakobi

1913 (Acarean), Hayerên gawarakan bararan. T'iflis (Ëminean azgagrakan zoiovacu; 9).

1937 (Adjarian), De l'affinité du grec et de l'arménien. In: Mélanges Émile Boisacq. Annuaire de l'Institut de Philologie et d'Histoire Orientales et Slaves. Bruxelles. 1: 3-4.

1940-51 (AcarHLPatm 1-2): Hayoc' lezvi patmut'yun, vol. 1, Yerevan: University Press, 1940; vol. 2, Yerevan: State Press of Armenia (HayPetHrat), 1951.

1971-79 (HAB), Hayerên armatakan bararan, in 4 vols (second edition). Yerevan: University Press. [The original edition: 1926-1935, in 7 volumes, Yerevan: University Press].

Adams, Douglas Q.

1999 A dictionary of Tocharian B. Amsterdam/Atlanta: Rodopi (Leiden studies in Indo-European; 10).

Adonc, N.

1915 Dionisij ©rakijskij i armjanskie tolkovateli (texts and study). Petrograd: Tipografija Imperatorskoj Akade-mii Nauk. (Bibliotheca Armeno-Georgica, 4).

Atayan, Ë. B.

1974 Barak'nnakan ew stugabanakan hetazotut'yunner. Yerevan: Academy Press.

Alisan, Lewond

1910 Hin hawatk' kam het'anosakan krönk' Hayoc'. Venice: S. Lazar.

Allen, W. S.

1978 The PIE velar series: neogrammarian and other solutions in the light of attested parallels. In: Transactions of the Philological Society 1978: 87-110.

Amatuni, Sahak Vardapet

1912 Hayoc' bar u ban. Vaiarsapat.

Arutjunjan, Covinar R.

1983 Armjano-greceskie leksiceskie izoglossy. In: G. B. Dzaukjan, L. A. Saradzeva, C. R. Arutjunjan. Ocerki po sravnitel'noj leksikologii armjanskogo jazyka. Yerevan: Academy Press: 221-354.

Arutjunjan, N. V.

2001 Korpus urartskix klinoobraznyx nadpisej. Yerevan: Academy Press 'Gitutjun'.

Asatryan, G. S.

1997 Remnant of an Indo-European poetic term for 'horse, stud' in New Iranian: Kurdish dêlazï. In: Iran and the Caucasus 1: 179-180.

Bailey, Harold W.

1979 Dictionary of Khotan Saka. Cambridge, London, New York, Melbourne: Cambridge University press.

Balles, Irene

1997 Reduktionserscheinungen in langen Wortformen als Ursprung morphologischer Doppelformen im Urindogermanischen: die Suffixformen *-io- und *-iio-. In: Die Sprache 39.2: 141-167.

Beekes, Robert S. P.

1976 Uncle and nephew. In: JIES 4: 43-63.

1990-92 The genitive in *-osio. In: Folia linguistica historica (Societas Linguistica Europaea) 11.1-2, 1990 [1992]: 21-25.

1996 Ancient European loanwords. In: HS 109.2: 215-236.

1998 The origin of Lat. aqua, and of *teuta 'people'. In: JIES 26, 3-4: 459-466.

2000 European substratum words in Greek. In: Michaela Ofitsch & Christian Zinko (eds.), 125 Jahre Indogermanistik in Graz. Leykam: 21-31.

2003 Historical phonology of Classical Armenian, apud Kortlandt 2003: 133-211.

2008 Palatalized consonants in Pre-Greek. In: Evidence and counter-evidence: essays in honour of Frederik Kortlandt (ed. by Alexander Lubotsky, Jos Schaeken, Jeroen Wiedenhof, with the assistance of Rik Derksen and Sjoerd Siebinga). Vol. 1: Balto-Slavic and Indo-European linguistics; vol. 2: General linguistics. Amsterdam, New York: Rodopi. (Studies in Slavic and general linguistics; 32-33). 1: 45-56.

2010 Etymological Dictionary of Greek (with the assistance of Lucien van Beek). 2 vols. Leiden: Brill.

2011 Comparative Indo-European linguistics: an introduction (2nd edition revised and corrected by Michiel da Vaan). Amsterdam, Philadelphia: John Benjamins.

Behr, Wolfgang

2004-05 Hinc sunt leones: two ancient Eurasian migratory terms in Chinese revisited (1). In: International journal of Central Asian studies. Seoul, Institute of Asian culture and development; Tashkent, International association of Central Asian studies. 2004: 1-24; 2005: 1-28.

Benveniste, Émile

1964 Éléments parthes en arménien. In: Revue les études arméniennes n.s. 1: 1-39.

Bird, Norman

1982 The distribution of Indo-European root morphemes: (a checklist for philologists). Wiesbaden: Otto Harrassowitz.

1993 The roots and non-roots of Indo-European: a lexicostatistical survey. Wiesbaden: Harrassowitz Verlag.

Birwé, Robert

1956 Griechisch-arische Sprachbeziehungen im Verbalsystem. Walldorf-Hessen: Verlag für Irientkunde Dr. H. Vorndran.

Blazek, Vaclav

2005 Hic erant leones: Indo-European 'lion' et alii. In: JIES 33.1-2: 63-101.

Bonfante, Giuliano

1937 Les isoglosses gréco-arméniennes. I. Faits phonétiques. In: L. Hjelmslev, C. Höeg, Ch. Moller, Ad.

Stender-Petersen (eds.), Mélanges linguistiques offerts à M. Holger Pedersen à l'occsion de son soixante-dixième anniversaire, 7 avril 1937. Aarhus: Universitetsforlaget; Kobenhavn: Levin & Munksgaard: 1533.

1946 Armenians and Phrygians. In: Armenian quarterly 1.1: 82-97.

1981 Hayereni dirk'a hndevropakan lezuneri mej. In: Patma-banasirakan handes 1981, Nr 2: 54-67.

Brenton, Lancelot Charles Lee

1851 The Septuagint with Apocrypha: Greek and English. London: Samuel Bagster & Sons, Ltd. (= Hendrickson Publishers, 1986/2009).

Bugge, Sophus

1893 Beiträge zur etymologischen Erläuterung der armenischen Sprache. In: ZVS 32: 1-87.

Canikean, Y. K.

1895 Hnut'iwnk' Aknay. T'iflis: Tparan M. D. Rötineanc'i. (T'iflisi hayoc' hratarakc'akan ankerut'iwn; 89).

Chantraine, Pierre

1968-80 Dictionnaire étymologique de la langue grecque: histoire des mots (4 volumes). Paris: Klincksieck.

Cheung, Johnny

2007 Etymological dictionary of the Iranian verb. Leiden, Boston: Brill. (Leiden Indo-European Etymological Dictionary series; 2).

Clackson, James

1994 The linguistic relationship between Armenian and Greek. Oxford, Cambridge: Blackwell (Publications of the Philological Society; 30).

2004-05 Review of Kortlandt 2003. In: Annual of Armenian linguistics 24-25: 153-158.

2008 Classical Armenian. In: Roger D. Woodard (ed.), The ancient languages of Asia Minor. Cambridge University Press: 124-143.

Clackson, James & Horrocks, Geoffrey

2011 The Blackwell history of the Latin language. Wiley-Blackwell.

Coleman, R. G. G.

1992 Le lexique grec et le lexique indo-européen: étude comparative. In: Françoise Létoublon (ed.), La langue et les textes en grec ancien: actes du colloque Pierre Chantraine (Grenoble, 5-8 septembre 1989), Amsterdam: Gieben: 169-179.

Cowe, S. Peter

1992 The Armenian version of Daniel. Atlanta, Georgia: Scholars Press (University of Pennsylvania Armenian texts and studies, 9).

Cox, Claude E.

2006 Armenian Job: reconstructed Greek text, critical edition of the Armenian with English translation. Leuven — Paris — Dudley, MA: Peeters. (Hebrew University Armenian studies, 8).

Davies, Paul & Ross, Alan S. C.

1977 'Close relationship' in the Indo-European languages. In: Transactions of the Philological Society 1975 [1977]: 82-100.

Davt'yan, K. S.

1966 Lernayin tarabaii barbarayin k'artezd. Yerevan: Academy Press.

Demiraj, Bardhyl

1994 Bemerkungen zur Entwicklung der anlautenden idg. Laryngale im Albanischen. In: In honorem Holger Pedersen: Kolloquium der Indogermanischen Gesellschaft vom 25. bis 28. März 1993 in Kopenhagen (hrsg. von J. E. Rasmussen, unter Mitwirkung von Benedicte Nielsen). Wiesbaden: Dr. Ludwig Reichert Verlag: 57-76.

1997 Albanische Etymologien: Untersuchungen zum albanischen Erbwortschatz. Amsterdam, Atlanta: Rodopi (Leiden studies in Indo-European; 7).

Derksen, Rick

2008 Etymological dictionary of the Slavic inherited lexicon. Leiden, Boston: Brill. (Leiden Indo-European Etymological Dictionary Series; 4).

Diakonoff, I. M.

1984 The pre-history of the Armenian people (transl. from D'jakonov 1968 by Lori Jennings; with revisions by the author). Delmar, New York: Caravan Books (Anatolian and Caucasian studies).

Dockalova, Lenka & Blazek, Vaclav

2011 On Indo-European roads. In: JIES 39.3-4: 299-341.

Durkin-Meisterernst, Desmond

2004 Dictionary of Manichaean Middle Persian and Parthian. Turnhout: Brepols. (Corpus Fontium Manichaeorum; 3.1).

Emmerick, R. E. & Skj^rvo, P. O.

1997 Studies in the vocabulary of Khotanese III. Wien: Verlag der Österreichischen Akademie der Wissenschaften. (Österreichische Akademie der Wissenschaften, philosophisch-historische Klasse; Sitzungsberichte 651; Veröffentlichungen der Kommission für Iranistik hrsg. von H. Eichner und R. Schmitt, Nr. 27).

Eska, Joseph F. & Wallace, Rex E.

2001 Remarks on the thematic genitive singular in ancient Italy and related matters. In: Incontri linguistici 24: 77-97.

EtimSlovIranJaz

2000- V. S. Rastorgueva & D. I. Edel'man, Etimologiceskij slovar' iranskix jazykov. Vol. 1, 2000; 2, 2003; 3, 2007 (in progress). Moscow: "Vostocnaja Literatura" Publishers.

EtimSlovSlavJaz

1974- O. N. Trubacev (ed.), Etimologiceskij slovar' slavjanskix jazykov. Vol. 1, 1974 — vol. 32, 2005 (in progress). Moscow: "Nauka".

Euler, Wolfram

1979 Indoiranisch-griechische Gemeinsamkeiten der Nominalbildung und deren indogermanische Grundlagen.

Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck (Innsbrucker Beiträge zur Sprachwissenschaft).

Fortson, Benjamin W.

2010 Indo-European language and culture: an introduction (2nd edition). Chichester, U.K.: Wiley-Blackwell. (Blackwell textbooks in linguistics; 19).

Frisk, Hjalmar

1960-72 (GEW 1-3), Griechisches etymologisches Wörterbuch. Vol. 1, 1960; vol. 2, 1970; vol. 3 (Nachträge, Wortregister, Corrigenda, Nachwort), 1972. Heidelberg: Carl Winter.

Furnee, Edzard Johan

1972 Die wichtigsten konsonantischen Erscheinungen des Vorgriechischen: mit einem Appendix über den Vokalismus. The Hague, Paris: Mouton.

Gamkrelidze, Th. V. & Ivanov, V. V.

1984 Indoevropejskij jazyk i indoevropejcy: rekonstrukciya i istoriko-tipologiceskij analiz prayazyka i protokul'tury. With a preface by R. Jakobson, 2 vols, Tbilisi: University Press.

1995 Indo-European and the Indo-Europeans: a reconstruction and historical analysis of a proto-language and a proto-culture. Part 1. The text. Transl. by J. Nichols, ed. by W. Winter. Berlin, New York : Mouton de Gruyter (Trends in linguistics. Studies and monographs, 80). [English translation of the Russian original 1984].

Haas, Otto

1939 Über die phrygischen Sprachreste und ihr Verhältnis zum Armenischen. In: Handes amsöreay 53.7-12: 225-235.

Hamp, Eric P.

1975 On the nasal presents of Armenian. In: ZVS 89: 100-109.

1979 Indo-European *gwen-Ha. In: ZVS 93.1: 1-7.

1983 Philologica varia. In: Revue des études arméniennes n.s. 17: 5-12.

1983a On the Helleno-Armenian shared lexicon. In: Annual of Armenian linguistics 4: 63-64.

1992 On subgrouping and Armenian. In: Annual of Armenian linguistics 13: 55-61.

Harut'yunyan, Sargis

2006 Hay hmayakan ew zoiovrdakan aiot'k'ner (Armenian incantations and folk prayers). Yerevan: University Press.

HayLezBrbBar 12001 A. Sargsyan et al. (eds.), Hayoc' lezvi barbarayin bararan. Vol. 1, 2001 [a-z]; vol. 2, 2002 [e-c]; vol. 3, 2004 [k-c]; vol. 4, 2007 [m-c'];vol. 5, 2008 [letters p-s]. Yerevan: Academy Press "Gitut'yun".

HerkWört

1997 Herkunftswörterbuch der deutschen Sprache (bearb. von Günther Drosdowski). Mannheim, Leipzig, Wien, Zürich: Dudenverlag. (Duden, 7).

Holst, Jan Henrik.

2009 Armenische Studien. Wiesbaden: Harrassowitz Verlag. Hovhannisyan, L. S.

1990 Hayereni iranakan p'oxarut'yunnera. Yerevan: Academy Press. Hübschmann, H.

1875/77 Ueber die stellung des armenischen im kreise der indogermanischen sprachen. In: ZVS 23 [N.F. 3.1]: 5-49.

1883 Armenische Studien. 1/1. Grundzüge der armenischen Etymologie. Leipzig: Breitkopf & Härtel.

1897 Armenische Grammatik. 1. Theil: Armenische Etymologie. Leipzig: Breitkopf & Härtel.

Ivanov, V. V. & Toporov, V. N.

1974 Issledovanija v oblasti slavjanskix drevnostej: leksiceskie i frazeologiceskie voprosy rekonstrukcii tekstov. Moscow: "Nauka".

Jahukyan, G. B.

1963 (Dzaukjan), Urartskij i indoevropejskie jazyki. Yerevan: Academy Press.

1968 Hayerena ew p'ryugerena. In: Patma-banasirakan handes 1968, Nr 3: 121-137.

1970 Hayerena ew hndevropakan hin lezunew. Yerevan: Academy Press.

1973 Stugabanut'yunner. In: Lraber hasarakakan gitut'yunneri 1973, Nr 6: 17-22.

1980 (Djahukian), On the position of Armenian in the Indo-European languages. In: John Greppin (ed.), Proceedings of the First international symposium on Armenian linguistics (11-14 July 1979). Delmar, New York: Caravan Books: 3-16.

1983 (Dzaukjan), Opyt semanticeskoj klassifikacii i areal'nogo raspredelenija indoevropejskoj leksiki armjanskogo jazyka. In: G. B. Dzaukjan, L. A. Saradzeva, C. R. Arutjunjan. Ocerki po sravnitel'noj leksikologii armjanskogo jazyka. Yerevan: Academy Press: 5-116.

1986-87 L'étymologie des mots haraw "sud", hiwsis(i) "nord" et le système d'orientation des anciens Arméniens. In Revue des études arméniennes n.s. 20: 29-33.

1987 Hayoc' lezvi patmut'yun: naxagrayin zamanakasrjan. Yerevan: Academy Press.

1991 Stugabanut'yunner. In: Patma-banasirakan handes 1991, Nr 2: 36-44.

2010 Hayeren stugabanakan bararan (ed. by Vahan Sargsyan). Yerevan: "Asoiik".

K'ajberuni

1902 Haykakan sovorut'iwnner. In: Azgagrakan handês, year 7, book 9: 83-116. Katz, Joshua T.

2000 Evening dress: the metaphorical background of Latin uesper and Greek ëanepoç. In Karlene Jones-Bley, Martin E. Huld, Angela Della Volpe (eds.), Proceedings of the eleventh Annual UCLA Indo-European conference. Los Angeles, June 4-5, 1999. Washington, DC: Institute for the Study of Man. (Journal of Indo-European Studies Monograph Series; 35): 69-93. K'ëc'ean, Arak'ël & Parsamean, Mkrtic'

1952 Akn ew Aknc'ik'. Initiated and compiled by Arak'ël K'ëc'ean; composed and edited by Mkrtic' Parsamean. Paris: Compatriotic Association of Eguen. Kimball, Sara

1998 On the origins of Hittite verbs of the type têhhi, dai. In: Jay Jasanoff, H. Craig Melchert & Lisi Oliver (eds.), Mir curad: studies in honor of Calvert Watkins. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck (Innsbrucker Beiträge zur Sprachwissenschaft; 92): 335-344. Klingenschmitt, Gert

1982 Das altarmenische Verbum. Wiesbaden: Dr. Ludwig Reichert Verlag. Kloekhorst, Alwin

2006 Initial laryngeals in Anatolian. In: HS 119: 77-108.

2008 Etymological dictionary of the Hittite inherited lexicon. Leiden, Boston: Brill. Kocharov, Petr

2008 Otrazenie indoevropejskix nazal'nyx prezensov v drevnearmjanskom jazyke (thesis). Saint Petersburg. Kortlandt, Frederik H. H.

1980 Albanian and Armenian. In: ZVS 94: 243-251.

1984 Proto-Armenian case endings. In: International symposium on Armenian linguistics (Yerevan, September

21-25, 1982): Reports. Yerevan: Academy Press: 97-106. 1986 Armenian and Albanian. In: Maurice Leroy & Francine Mawet (eds.), La place de l'arménien dans les

langues indo-européennes. Louvain: Peeters: 38-47. 2003 Armeniaca: comparative notes; with an appendix on the historical phonology of Classical Armenian by Robert S. P. Beekes. Ann Arbor: Caravan Books.

2007 Italo-Celtic origins and prehistoric development of the Irish language. Amsterdam, New York: Rodopi. (Leiden studies in Indo-European; 14).

2010 Studies in Germanic, Indo-European and Indo-Uralic. Amsterdam, New York: Rodopi. (Leiden studies in Indo-European; 17). Kuiper, F. B. J.

1995 Gothic bagms and Old Icelandic ylgr. In: NOWELE 25, March: 63-88. Lamberterie, Charles de

1978 Armeniaca I-VIII: études lexicales. In: BSL 73: 243-285.

1978-79 Armeniaca X: une isoglosse gréco-arménienne. In: Revue des études arméniennes n.s. 13: 31-40.

1982 La racine *(s)per-gh- en arménien. In: Revue des études arméniennes n.s. 16: 57-68.

1983 Une bagatelle étymologique (artewanunk'lôpenâvrf). In: Revue des études arméniennes n.s. 17: 21-22. 1986 Deux isoglosses entre l'arménien et l'indo-iranien. In: Maurice Leroy & Francine Mawet (eds.), La

place de l'arménien dans les langues indo-européennes. Louvain: Peeters: 48-61. 1992 Introduction à l'arménien classique. In: LALIES 10 (Aussois, 1988-1989): 234-289. 1992a Le problème de l'homonymie: les trois verbes ô^éAAw en grec ancien. In: Françoise Létoublon (ed.), La langue et les textes en grec ancien: actes du colloque Pierre Chantraine (Grenoble, 5-8 septembre 1989), Amsterdam: Gieben: 201-217.

2012 L'arménien et l'indo-européen balkanique. In: IX International Conference on Armenian linguistics (7-9 November 2012, Saint Petersburg). Abstracts: 109-111.

2012a Sur le nom du "rêve" en indo-européen balkanique. In: Etymology and the European lexicon: 14th Fachtagung of the Indogermanische Gesellschaft (17-22 September 2012, University of Copenhagen). Abstracts: 69.

2013 Grec, phrygien, arménien: des anciens aux modernes. In: Journal des Savants (janvier-juin 2013): 3-69. Lanalanyan, Aram

1969 Avandapatum. Yerevan: Academy Press.

Lehmann, Winfred P.

1981 The genitive singular ending in *-syo: how an Indo-Europeanist works. In: Yoël L. Arbeitman & Allan R. Bomhard (eds.), Bono homini donum: essays in hitorical linguistics, in memory of J. Alexander Kerns. Amsterdam: John Benjamins, vol. 1: 179-188.

1986 A Gothic Etymological Dictionary (based on the 3rd ed. of Feist's dictionary). Leiden: E. J. Brill. 1993 Theoretical bases of Indo-European linguistics. London, New York: Routledge.

Liddell, Henry George, & Scott, Robert

1996 A Greek-English Lexicon. A new edition revised and augmented throughout by Sir Henry Stuart Jones, with the assistance of Roderick McKenzie and with the co-operation of many scholars. Oxford: Clarendon Press [1940/53].

Ligorio, Orsat & Lubotsky, Alexander

forthc. The Phrygian language (forthcoming). Lubotsky, A. M.

2001 Reflexes of Proto-Indo-European *sk in Indo-Iranian. In: Incontri linguistici (Pisa, Roma) 24: 25-57. 2001a The Indo-Iranian substratum. In: Christian Carpelan, Asko Parpola & Petteri Koskikallio (eds.),

Early contacts between Uralic and Indo-European: linguistic and archaeological considerations. Papers presented at an international symposium held at the Tvärminne Research Station of the University of Helsinki (8-10 January, 1999). Helsinki: Suomalais-ugrilainen seura. (Mémoires de la Société Finno-ougrienne, 242): 301-317.

2002 The Indo-Iranian word for 'shank, shin'. In: Journal of the American Oriental Society 122.2: 318-324. Makaev, È. A.

1967 Armjano-indoevropejskie leksiceskie izoglossy i areal'naja lingvistika: In: Zeitschrift für Phonetik, Sprachwissenschaft und Kommunikationsforschung 20: 449-464. Mallory, J. P. & Adams, D. Q.

1997 (eds.), Encyclopedia of Indo-European culture. London, Chicago: Fitzroy Dearborn Publishers.

2006 The Oxford introduction to Proto-Indo-European and the Proto-Indo-European world. Oxford University Press. Manczak, Witold

1987 The relationship of Armenian to Greek and Persian. In: Annual of Armenian linguistics 8: 19-28. Martirosyan, Hrach K.

2005 On Armenian ciw 'shank'. In: Iran and the Caucasus 9.1: 81-84.

2007 Mediterranean-Pontic substratum words. In: Aramazd: Armenian journal of Near-Eastern studies, vol. 2 (Yerevan: Association for Near Eastern and Caucasian Studies): 88-123.

2010 Etymological dictionary of the Armenian inherited lexicon. Leiden, Boston: Brill. (Leiden Indo-European

Etymological Dictionary Series; 8). 2012 European and Mediterranean substrate words in Armenian. In: Etymology and the European lexicon: 14th Fachtagung of the Indogermanische Gesellschaft (17-22 September 2012, University of Copenhagen). Abstracts: 75-76.

forthc. Vocative strategies and accent in Armenian: synchrony and diachrony. To appear in Sprachtypologie

und Universalienforschung (ed. by Anaid Donabedian and Vladimir Plungian). prepar. 1 The Armenian month names. In preparation.

prepar. 2 The Indo-European and Iranian layers of Armenian mythology. In preparation.

prepar. 3 Lernayin akunk'i farm, Arsaluysi dic'uhin ew glxi u mort'u zohacesa visapakot'olneri imastabanakan henk'um. In preparation. Matasovic , Ranko

2009 Etymological dictionary of Proto-Celtic. Leiden, Boston: Brill. (Leiden Indo-European Etymological Dictionary Series; 9). Mawet, Francine

1983 The motif of the bird in Armenian epic literature and its relations with Iranian tradition. In: Thomas J.

Samuelian & Michael E. Stone (eds.), Medieval Armenian culture. Chico, Calif.: Scholars Press, 1984 (University of Pennsylvania Armenian texts and studies; 6): 179-193. Mayrhofer, Manfred

1986 Lautlehre: segmentale Phonologie des Indogermanischen. In: Indogermanische Grammatik 1.2. Heidelberg: Carl Winter, Universitätsverlag.

1956-80 (KEWA) Kurzgefaßtes etymologisches Wörterbuch des Altindischen (4 vols.). Heidelberg: Carl Winter Universitätsverlag.

1992-2001 (EWAia) Etymologisches Wörterbuch des Altindoarischen (3 vols). Heidelberg: Universitätsverlag C. Winter.

Meid, Wolfgang

1975 Probleme der räumlichen und zeitlichen Gliederung des Indogermanischen. In: Helmut Rix (ed.), Flexion und Wortbildung: Akten der V. Fachtagung der Indogermanischen Gesellschaft (Regensburg, 9-14 September 1973). Wiesbaden: Dr. Ludwig Reichert Verlag: 204-219.

Meier-Brügger, Michael

1995 Griechisch (aiyv)niôç, vedisch (rji)pya- und Verwandtes. In: HS 108: 50-55.

2003 Indo-European linguistics. In cooperation with Matthias Fritz and Manfred Mayrhofer; translated by Charles Gertmenian. Berlin, New York: Walter de Gruyter.

Meillet, A.

1898 Étymologies arméniennes. In: MSL 10: 274-282.

1900 Notes sur quelques faits de morphologie. In: MSL 11: 6-21.

1896 Varia. In: MSL 9: 136-159.

1924 Remarques étymologiques. In: Revue des études arméniennes 4, fasc. 1: 1-6.

1925 Remarques sur l'étymologie de quelques mots grecs. In: BSL 26: 1-22.

1936 Esquisse d'une grammaire comparée de l'arménien classique. Wien: Mechitharisten.

1950 Les dialectes indo-européens (nouveau tirage avec une introduction nouvelle et des additions). Paris: Librairie ancienne Édouard Champion; Librairie de la Société de Linguistique de Paris.

MifNarMir 1-2

1980-82 Mify narodov mira: ènciklopedija. Redactor-in-chief: S. A. Tokarev. Vol. 1, 1980; vol. 2, 1982. Moscow: "Sovetskaja ènciklopedija".

Monchi-Zadeh, Davoud

1990 Wörter aus Xurâsân und ihre Herkunft. Leiden: E. J. Brill. (Acta Iranica 29).

Morani, Moreno

1992 Pour une histoire du génitif arménien. In: Muséon 105.3-4: 303-319.

Morgenstierne, Georg

1962 Feminine nouns in -a in Western Iranian dialects. In: A locust's leg: studies in honour of S. H. Taqizadeh: 203-208.

1974 Etymological Vocabulary of the Shughni Group. Wiesbaden: Dr. Ludwig Reichert Verlag. (Beiträge zur Iranistik; 6).

2003 A new etymological vocabulary of Pashto. Compiled and edited by J. Elfenbein, D. N. MacKenzie and Nicholas Sims-Williams. Wiesbaden: Dr. Ludwig Reichert Verlag. (Beiträge zur Iranistik; 23).

Müller, Friedrich

1870 Armeniaca II. In Sitzungsberichte der philosophisch-historischen Classe der Kaiserlichen Akademie der Wissenschaften. 64.2. Wien: 447-456.

NHB 1-2

1836-37 G. Awetik'ean, X. Siwrmëlean, M. Awgerean, Nor bargirk' haykazean lezui, 2 vols., Venice: St Lazar (1836-1837); reprinted: Yerevan: University Press (1979-1981).

Nocentini, Alberto

1994 Le radici ie. *mor-/morm- 'incubo' e *bü-/bub- 'spauracchio' e il sistema fonologico indeuropeo. In: P Cipriano, P. Di Giovine, M. Mancini (eds.), Miscellanea di studi linguistici in onore di Walter Belardi. 1-2. Roma: Il Calamo. Vol. 1: 399-415.

Ödabasyan, A. A.

1986 Tiezerasteicman araspeli veraprukayin tarrern axtark'um. In: Patma-banasirakan handes 1986, Nr 3: 138-149.

Olsen, B. A.

1999 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).

2011 Review of Martirosyan 2010. In: Kratylos 56: 14-30.

Orel, Vladimir E.

1993 The position of Armenian. In: Annual of Armenian linguistics 14: 55-67.

1998 Albanian etymological dictionary. Leiden, Boston, Köln: Brill.

Patrubâny, Lukas v.

1902-03 Idg. *elozg1h im Armenischen. In: IF 13: 124-125.

Pedersen, Holger

1900 Albanesisch und Armenisch. In: ZVS 36: 340-341 (= 1982: 1-2).

1905 Zur armenischen Sprachgeschichte. In: ZVS 38: 194-240 (=1982: 56-102).

1906 Armenisch und die Nachbarsprachen. In: ZVS 39: 334-485 (= 1982: 112-263).

1924 Armenier. B. Sprache. In: Max Ebert (ed.), Reallexikon der Vorgeschichte. Vol. 1. Berlin, pp. 219-226 (= 1982: 302-309).

1925 Le groupement des dialectes indo-européens. Kobenhavn.

1982 Kleine Schriften zum Armenischen (hrsg. von Rüdiger Schmitt). Hildesheim, New York: Georg Olms Verlag.

Petersson, Herbert

1916 Beiträge zur armenischen Wortkunde. In: ZVS 47: 240-291.

Petrosyan, Armen

2002 The Indo-European and ancient Near Eastern sources of the Armenian Epic. JIES Monograph Nr 42. Washington DC: Institute for the Study of Man.

2005 Urartui vernaxavi ët'nikakan patkanelut'yan xndira. In: Patma-banasirakan handes 2005, Nr 3: 208-226.

Pinault, Georges-Jean

1989 Introduction au tokharien. In: Lalies: Actes des sessions de linguistique et de littérature 7 (Aussois, 27 août — 1er septembre 1985), Paris: 5-224.

2002 Tokh. B kucanne, A kucim et skr. tokharika. In: Indo-Iranian journal 45: 311-345.

Pisani, Vittore

1950 Studi sulla fonetica dell'armeno. I. Palatalizzazioni ed esiti di s + gutturale; esiti delle semivocali. In: Ricerche linguistiche (Roma) 1/2: 165-193.

1979 (Pizani), Hndevropakan lezuneri caguma. In: Patma-banasirakan handes 1979, Nr 1: 204-212.

Pisowicz, Andrzej

1987 Some remarks on Greek-Armenian-Persian lexical parallels a propos W. Manczak's "The relationship of Armenian to Greek and Persian". In: Annual of Armenian linguistics 8: 29-33.

Pokorny, Julius

1959 Indogermanisches etymologisches Wörterbuch, vol. 1. Bern, Stuttgart: Francke.

Porzig, Walter

1954 Die Gliederung des indogermanischen Sprachgebiets. Heidelberg: Winter. (Indogermanische Bibliothek. 3. Reihe: Untersuchungen).

1964 (Porcig, V.), Clenenie indoevropejskoj jazykovoj oblasti. (Russian translation of Porzig 1954; ed. by A. V. Desnickaja). Moscow: "Progress".

Pronk, Tijmen

2010 Sanskrit (v)fsabha-, Greek àpo-qv, ëpar^v: the spraying bull of Indo-European? In: HS 122: 170-181.

Ringe, Don, Tandy Warnow & Ann Taylor

2002 Indo-European and computational cladistics. In: Transactions of the Philological Society 100.1: 59-129.

Ritter, Ralf-Peter

1983 Eine verkannte Etymologie für armen. aniw 'Rad'. In: Münchener Studien zur Sprachwissenschaft 42: 191-196.

2006 Zur Frage der Reste indogermanischer Dichtersprache im Armenischen. In: Georges-Jean Pinault et Daniel Petit (eds.), La langue poétique indo-européenne: actes du Colloque de travail de la Société des Études Indo-Européennes (Indogermanische GesellschaftlSociety for Indo-European Studies), Paris, 22-24 octobre 2003. Leuven, Paris: Peeters: 413-417.

Rusakov, A. Ju.

1984 Albanskij i armjanskij: nekotorye problemy. In: International symposium on Armenian linguistics (Yerevan, September 21-25, 1982): Reports. Yerevan:

Scheftelowitz, J.

1904-05 Zur altarmenischen Lautgeschichte. In: Beiträge zur Kunde der indogermanischen Sprachen, Göttingen (ed. Ad. Bezzenberger), (1): 28, 1904: 282-313; (2): 29, 1905: 13-71.

1927 Idg. zgh in den Einzelsprachen. In: ZVS 54: 224-253.

Schmidt, Karl Horst

1980 Altarmenische Miszellen. In: Annual of Armenian linguistics 1: 1-5.

1987 The Indo-European background of the Classical Armenian noun declension. In: Annual of Armenian linguistics 8: 35-47.

1988 On Armenian phonology and Indo-European reconstruction. In: Mohammad Ali Jazayery and Werner Winter (eds.), Languages and cultures: studies in honor of Edgar C. Polomé. Berlin, New York, Amsterdam: Mouton de Gruyter. (Trends in linguistics. Studies and monographs, 36): 601-610.

Schmitt, Rüdiger

1967 Dichtung und Dichtersprache in indogermanischer Zeit. Wiesbaden: Otto Harrassowitz.

1970 Der 'Adler' im Alten Iran. In: Die Sprache 16.1: 63-77.

1972-74 Die Erforschung des Klassisch-Armenischen seit Meillet (1936). In: Kratylos 17: 1-68.

1975 Von Bopp bis Hübschmann: Das Armenische als indogermanische Sprache. In: ZVS 89.1: 3-30.

1983 Iranisches Lehngut im Armenischen. In: Revue des études arméniennes n.s. 17: 73-112.

1987 Armenia and Iran I. Armenia, Achaemenid province. IV. Iranian influences in Armenian. 1. General. In: Encyclopaedia Iranica 2, 1987: 417-418; 445-459.

2007 Grammatik des Klassisch-Armenischen mit sprachvergleichenden Erläuterungen (2., durchgesehene Auflage). Innsbruck: Institut für Sprachen und Literaturen der Universität Innsbruck. (Innsbrucker Beiträge zur Sprachwissenschaft).

Schrijver, Peter C. H.

1991 The reflexes of the Proto-Indo-European laryngeals in Latin. Amsterdam, Atlanta: Rodopi. (Leiden studies in Indo-European; 2).

1997 Animal, vegetable and mineral: some Western European substratum words. In: A. Lubotsky (ed.), Sound law and analogy: papers in honor of Robert S. P. Beekes on the occasion of his 60th birthday. Amsterdam-Atlanta, GA: Rodopi B.V.: 293-316.

Schwartz, Martin

1986 Coded sound patterns, acrostics, and anagrams in Zoroaster's oral poetry. In: B. Forssman, K. Hoffmann, J. Narten (eds.), Studia grammatica iranica: Festschrift für Helmut Humbach. ("Münchener Studien zur Sprachwissenschaft", Beiheft 13, N.F.). München: R. Kitzinger: 327-392.

Shields, Kenneth Jr.

1981 A new look at the centum/satem isogloss. In: ZVS 95.2: 203-213.

Sirokov, Oleg

1977 Armjano-greceskie ètnogeneticeskie kontakty po dannym sravnitel'no-istoriceskoj fonologii. In: Patma-banasirakan handes 1977, Nr 1: 85-100.

1980 Mesto armjanskogo jazyka sredi indoevropejskix i problema armjanskoj prarodiny: avtoxtonnost' armjan po dannym sravnitel'no-istoriceskogo jazykovedenija. In: Lraber hasarakakan gitut'yunneri 1980, Nr 5: 80-93.

Skj^rv0, P. O.

1997 Avestica II. Yokes and spades and remnants of the "tripartite ideology". In: Münchener Studien zur Sprachwissenschaft 57: 115-128.

Solta, G. R.

1960 Die Stellung des Armenischen im Kreise der indogermanischen Sprachen: eine Untersuchung der indogermanischen Bestandteile des armenischen Wortschatzes. Wien: Mechitharisten. (Studien zur armenischen Geschichte; 9).

Specht, Franz

1935 Zur Geschichte der Verbalklasse auf -e: ein Deutungsversuch der Verwandtschaftsverhältnisse des Indogermanischen. In: ZVS 62: 29-115.

1939 Sprachliches zur Urheimat der Indogermanen. In: ZVS 66.1/2: 1-74.

Stempel, Reinhard

1990 Die Entwicklung von auslautendem *m und das Problem sekundärer n-Stämme im Armenischen. In:

IndogForsch 55: 38-62.

2000 Armenisch und Indoiranisch. In: Bernhard Forssman & Robert Plath (eds.), Indoarisch, Iranisch und die Indogermanistik: Arbeitstagung der Indogermanischen Gesellschaft vom 2. bis 5. Oktober 1997 in Erlangen. Wiesbaden: Reichert Verlag: 515-522.

Szemerényi, Oswald

1977 Studies in the kinship terminology of the Indo-European languages: with special reference to Indian, Iranian, Greek, and Latin. In: Varia 1977 (Acta Iranica, Série 3. Textes et mémoires, 7): 1-240. Téhéran-Liège: Bibliothèque Pahlavi; Leiden: Brill.

1985 Armenian between Iran and Greece. In: U. Pieper and G. Stickel (eds.), Studia linguistica, diachronica et synchronica Werner Winter sexagenario anno MCMLXXXIII. Berlin, New York, Amsterdam: Mouton de Gruyter: 783-799.

1996 Introduction to Indo-European linguistics (translated from the 4th edition, 1990, with additional notes and references). Oxford: Clarendon Press.

Thieme, Paul

1952 Studien zur indogermanischen Wortkunde und Religionsgeschichte. In: Berichte über die Verhandlungen der Sächsischen Akademie der Wissenschaften zu Leipzig: Philologisch-historische Klasse, 98.5. Berlin.

1954 Die Heimat der indogermanischen Gemeinsprache. In: Abhandlungen der Akademie der Wissenschaften und der Literatur in Mainz: geistes- und sozialwissenschaftliche Klasse, 1953.11 [1954]: 35-86. Wiesbaden.

Tischler, Johann

1973 Glottochronologie und Lexikostatistik. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck. (Innsbrucker Beiträge zur Sprachwissenschaft; 11).

Thomson, R. W.

2006 Moses Khorenats'i, History of the Armenians: translation and commentary on the literary sources (revised edition of 1978). Ann Arbor: Caravan Books.

Vaan, Michiel de

2008 Etymological dictionary of Latin and the other Italic languages. Leiden, Boston: Brill. (Leiden Indo-European Etymological Dictionary Series; 7).

Viredaz, Rémy

2005-07 Notes d'étymologie arménienne I. In: Revue des études arméniennes n.s. 30: 1-14.

Watkins, Calvert

1975 La famille indo-européenne de grec ôqxl€: linguistique, poétique et mythologie. In: BSL 70.1: 11-26.

1995 How to kill a dragon: aspects of Indo-European poetics. New York, Oxford: Oxford University Press.

Witzel, Michael

2003 Linguistic evidence for cultural exchange in Prehistoric Western Central Asia. Philadelphia: University of Pennsylvania, Department of East Asian Languages and Civilizations. (Sino-Platonic papers; 129).

Wyatt, William F.

1982 Lexical correspondences between Armenian and Greek. In Annual of Armenian linguistics 3: 19-42.

Xac'aturova, E. G.

1973 Ob armjano-indijskix leksiceskix izoglossax. In: Patma-banasirakan handes (= IFZ) 1973, Nr 2: 191-199.

1979 Drevnejsie armjano-indo-iranskie jazykovye kontakty. In: Hayoc' lezvi hamematakan k'erakanut'yan harc'er. Yerevan: Academy Press: 325-375.

Xalat'eanc', Grigor

1896 (Xalat'janc), Armjanskij èpos v istorii Armenii Moiseja Xorenskago: opyt kritiki istocnikov (Part 1, Issledo-vanie. Part 2, Materialy). Moscow: Tipografija Varvary Gatcuk.

Yakubovich, Ilya

2008 The origin of Luwian possessive adjectives. In: Proceedings of the 19th Annual UCLA Indo-European conference (Los Angeles, November 2-3, 2006). Washington, DC: Institute for the Study of Man. (Journal of Indo-European Studies Monograph Series, 54): 193-217.

2010 Morphological negation in Urartian. In: Aramazd: Armenian journal of Near Eastern studies, vol. 5.1: 141-165.

YusamXotorj

1964 H. Yarut'iwn v. Hulunean & H. Matt'ëos v. Hacean, Yusamatean Xotorjuri. Vienna: Mechitharisten.

Zöhrapean, Yovhannës

1805 Astuacasunc' matean Hin ew Nor Ktakaranac' (4 vols.). Venice: S. Lazar.

Грач Мартиросян. Место армянского языка в индоевропейской семье и его связь с греческим и индоиранским.

Основной задачей статьи является каталогизация лексических этимологий, сближающих армянский язык с греческим и/или индоиранским. Предлагаемые списки включают в себя как общие инновации, так и изолированные ареальные термины. Обе группы этимологий (армяно-греческие и армяно-индоиранские) могут рассматриваться в рамках одной и той же пространственно-временной модели. В процессе расселения носителей индоевропейских языков протоармянский продолжал контактировать с как соседними индоевропейскими диалектами, так и с неиндоевропейскими языками. Слова субстратного происхождения выделяются ограниченной географической дистрибуцией, необычной фонологией или словообразованием, характерной семантикой. Материал, представленный в настоящей работе, не претендуя на исчерпывающую полноту, позволяет, тем не менее, сделать предварительный вывод о диалектной близости армянского, греческого, (фригийского), и индоиранских языков. Внутри данной гипотетической диалектной группы ареал протоармянского языка занимал промежуточное положение между ареалом протоиндоиранцев на востоке и ареалом протогре-ков на западе. Впоследствии протоиндоиранцы сдвинулись далее на восток, тогда как протоармяне и протогреки продолжали оставаться в смежных географических ареалах на протяжении длительного прериода, развив многочисленные общие инновации. В последующий период они заимствовали, совместно или поодиночке, множество лексем, принадлежащих к черноморско-средиземноморскому субстрату, по преимуществу культурные и сельскохозяйтвенные термины, а также обозначения флоры и фауны. С другой стороны, армянский язык содержит значительное число лексических соответствий индоевропейским языкам Европы, существенная часть которых также должна объясняться в терминах общего субстрата, а не индоевропейского наследия.

Ключевые слова: армянская историческая лингвистика, армянская этимология, индоевропейская сравнительная лингвистика, индоиранская лексикология, греческая лексикология, средиземноморский субстрат.

AUTHOR' S ADDITION TO P. 94:

4.1.17. *gwou-io- (or *gwh3eu-io-): Arm. kogi, gen. kogw-o-y, ins. kogw-o-v 'butter', Skt. gavya-, gavya- 'consisting of cattle, coming from or belonging to a cow (as milk, curds, etc.)', YAv. gaoiia- 'coming from cattle, consisting of cattle', Gr. adj. -3o(p)io^, e.g. evvea-^0i0Q 'worth nine beeves'. This isogloss33 is based on the PIE word for 'cow' (Arm. kov; cf. nom. arew vs. oblique areg- 'sun'). Armenian and Indic are closer to each other since they show a semantic development to 'a dairy product'. Now we also have a wonderful match in Toch. B, kewiye 'pertaining to cow; butter' < *kawiya < *kawaya < *gwow-iyo-.33a Although this makes the isogloss less significant, I nevertheless included it in order to emphasize its semantic closeness to another agreement between Armenian and Indic: Arm. ser 'cream' and Skt. saras n. 'cream' (see § 5.2.16).

33 See already Meillet 1896: 152.

33a Pinault 1989: 53. I am indebted to James Clackson for this information. For the Tocharian word, see also Adams 1999: 198.