I. M. Diakonoff

The Pre-history of the Armenian People

(Continued from Previous Page [115])

Chapter 3.
The Formation of the Armenian People

2.4. The Problem of the Mushki

Now we can turn to the analysis of the data on the Mushki and the Urumeans. We shall begin with the problem of the Mushki.

In favor of the fact that the earlier (Eastern) Mushki were speakers of Proto-Armenian is the time of their appearance and the place of their settling, which from the earliest times has been considered the homeland of the Armenian nation (61). It is also important that their designation coincides in Assyrian with that of the Phrygians. Note that there are are two different groups called "Mushki" in the Assyrian sources: one group of Mushki captured Alzi and Purulumzi (near the confluence of the Arsanias and the Euphrates) around 1165 B.C. They are in evidence as an agricultural population in this region right up to the beginning of the 9th century B.C. It is impossible to establish whether this group of Mushki lived only in this region or also in others, to the west, which they may have occupied on their way through Asia Minor, since the Assyrian annals in which they are mentioned naturally speak each time only about those territories which the Assyrian army contacted at the moment. But actually there is ground to suppose that this group of Mushki spread also to the western bank of the Upper Euphrates.

The other group of Mushki are mentioned in connection with the campaigns of the Assyrian king Sargon II (722-705 B.C.) and the Urartian king Rusa II (first half of the 7th century B.C.) as dwelling to the west of the Cilician Taurus. They are unquestionably to be identified with the Phrygians (62).

It might seem strange that the Eastern Mushki, whom we identify with the Proto-Armenians, are no longer mentioned in the Assyrian and Urartian sources after Assurnasirapal II. But this is explained by the fact that Alzi, the new home of the Mushki, as we have seen, was subjected in 856 B.C. by the Assyrian king Shalmaneser V (in his inscription it is called by the apparently archaic term Isua, i.e., Isuwa) and was included in the "province of Nairi, Alzi, and Suhmu," which became part of Assyria. Between 799 and c. 780 B.C. it was conquered and annexed by Minua, king of Urartu. The inscriptions of both these kings, as was often the case in texts of this sort, do not mention tribal or ethnic designations, but only topographical and political ones. From this time on, to the end of the existence of Assyria and Urartu, no more campaigns were conducted into the territory of Alzi. Therefore there was no longer any reason to mention the Eastern Mushki in the inscriptions (63) Only two of the [116] "countries" in which the Mushki may have partly settled, remained independent. These were Shubria to the south, and Melid-Kammanu to the west of Alzi (across the Euphrates). The first was subdued in 673 by Assyria for a short time but probably regained its independence at the end of the 7th century B.C. The second was in part re-settled by people replaced by the Assyrians from beyond the Euphrates, but remained subject to Assyria for only thirty years.

However the problem of the identification of the ethnic term Mushki remains complicated. Goetze compared the Mushki (on the basis of external similarity of the word's structure) with the Kaska (Kashka), but an ethnic connection between them is most improbable (64). In contradistinction to the Kaska, the Mushki are not known to the Hittite sources (65). Several scholars connect the Mushki with the tribe of Moschi attested by the Greek sources in Pontus, and with the Georgian tribe of Meskhi. Note that while with the Assyrians, Urartians, and ancient Hebrews Mushki was undoubtedly a name for Phrygians, the Greeks neatly differentiated the Phrygians from the Moschi.

The data of the classical authors are contradictory. Hecataeus (66) speaks of the Moschi as a "Colchian," i.e., probably Georgian-speaking people who lived in the neighborhood of the Matieni, or in other words, the Hurrians. Herodotus (III, 95; VII, 78) lists them among the peoples of the XIX satrapy of the Achaemenian Empire, i.e., Pontus, uniting them with the Tibareni, who lived near Cotyora (modern Ordu, cf. Xenophon, Anabasis, V, 1 ff.). It follows that they lived to the west of the Colchi (67) and, in any event, did not live to the east of the Upper Euphrates valley. However half a millennium later Strabo places the Moschi in two different places, but in both cases far from Pontus. The first location is somewhere in modern Abkhazia on the eastern shore of the Black Sea (XI, 2, 12 ff.; Stephan of Byzantium also places them there, quoting the 6th-century-B.C. writer Hellanicus--possibly mistakenly, as Melikishvili suggests, with a reference to Kiessling). The second is in the mountains where Colchis, Iberia, and Armenia meet (XI, 2, 18). The later Moschi are obviously the Meskhi /Mesxi/ of the later writers, both Byzantine and Georgian, and it is possible that we have here a simple confusion of two similar terms or an attempt to identify a later term with one known from the ancient authors. Melikishvili considers the term Meskhi, as used for a Georgian-speaking tribe in this region, to be a later, and perhaps foreign term. The Urartian sources do not know either the Moschi or the Meskhi in this area (68). Note that the term Mosok is used in Avarian (Daghestan) as the name of the Georgians as a whole (69). All of this makes us suspect that the term Moschi was not an ethnonym at all but more likely some sort of nickname, which could have been [117] applied to various tribes. This is also the case with the Chalybes: the Greeks evidently used the name Chalybes for all of the inhabitants of the Pontus who traded in iron ore (see Pauly, s.v. Chalybes); in some cases we might suspect that the Moschi (in Pontus) and/or the Chalybes were foreign designations for the Chaldians (Halitu, Xaghtik') who actually did live between the Tibareni in the west, the Mosynoeci in the north, and the Matieni in the southeast (70).

It has been suggested that the Mushki of the Ancient Oriental texts which we identify as Phrygians were actually a Georgian-speaking tribe and that Tabal, a region in the Cilician Taurus, corresponds to the Pontian Tibareni (71). The latter allegedly migrated here from Tabal (or vice versa, from here to Tabal). At the same time the Tibareni, by reason of a rather doubtful similarity in the sound of the term, were thought to correspond to the Iberians, i.e., the eastern Georgians. Melikishvili expresses himself more cautiously as to Tabal (72), only admitting the possibility of a presence of Georgian-speaking tribes in the Cilician Taurus (Tabal) at the beginning of the 1st millennium B.C.73 But according to him the Mushki should be regarded as Georgian. He supposes that the Assyrian sources mention at least two different invasions of these tribes into the region of the Armenian Taurus; the first is evidenced by Tiglath-pileser I for the first time around 1165 B.C., and the second is connected with the reference to the Mushki in the same region in the 9th century B.C.(74) Moreover Melikishvili compares the name of the king of Phrygia, Gordias, (and the name of the Phrygian capital, Gordion) with the name of the mountain tribe in the eastern part of the Armenian Taurus and the mountaineers of Gordyene or Corduene (modern Kurdistan), viz., the Carduchi, which name he also connects with the self-appellation of the Kartvelians. All of this, however, is most strained and unconvincing. Linguistically we can hardly accept a comparison of remotely similarly sounding names without an explanation of the regularities of the changes (k > g, t > d, etc.) (75). Moreover, according to Melikhisvili's own remark, the suffix -uchi in the name of the Carduchi more likely indicates their Hurro-Urartian affiliation. The habitat of the Carduchi was the valley of the river Kentrites-Bohtan and the mountains around the upper reaches of the Upper Zab (i.e., the regions where Assyrian sources attest Hurrian mountaineer tribes, who, like the Carduchi of the 5th-4th centuries B.C., kept the surrounding valleys in fear) (76). Corduene had no contact either with the place of habitation of unquestionably Georgian tribes or with the place of settlement of the Eastern Mushki (even if we place the latter, with Melikishvili, in the northern Mesopotamian mountains of Kashiari-- Tur-'Abdin, which from our point of view is [118] incorrect) (77). And, finally, it is completely impossible to derive the name of a tribe from the name of a person who lived a thousand miles to the west and of a city named for this person. (Or, vice versa, why should the Phrygian king and the Phrygian capital be named after the Carduchi?) Thus nothing remains of all the argumentation, except for a similarity between the name of the Mushki and the name of the Moschi and the Georgian tribe of the Meskhi.

Melikishvili writes (78): "The name of the Tabalians may have been used by the Assyrians (like the name of the Mushki) in a comprehensive sense, designating tribes of varying origin; however, among them (at least as one composing part) we must probably suppose the presence of Kartvelian (particularly of western Georgian) tribes as well." Here a correction is needed: the Assyrian sources do not know of a "Tabalian" tribe or nation, but speak only of "Tabal" (79), as the name of one of the regions with Luwian population (although other ethnic groups are also possible here). As far as the term Mushki is concerned, a comprehensive meaning is actually possible. But it is in any event improbable that the Assyrians, Urartians, and Hebrews should designate Thraco-Phrygian tribes and state formations by the term Mushki after a Georgian tribe (the Moschi), as Melikishvili, Cavaignac, Khazaradze, and others suggest (80). The alternative explanation, which I have suggested, is that certain Georgian tribes were called Moschi because at one time they lived in territory subject to Phrygia (Mushki), or had a Phrygian-type culture (81).

It is important to determine the original phonetic form of the term in question. There was no x ("loch")-sound in it: it does not appear in the Assyrian transliteration (Mushki, Mushki, Musku), or the Urartian (Mushki-) (22) nor in the Luwian Hieroglyphic Muska-(a /sh/ phoneme did not exist in Luwian), nor in Hebrew, where all the forms which have come down to us can be traced to *moshk-, nor in Ancient Greek (Gk. Moschoi has the aspirated /kh/ sound which corresponds to Armenian /k'/ and not the Armenian /x/ (as in "loch") (83). Therefore, the original form of the term must have been /*mushk'-/ or /*musk'-/ (thus perhaps preferably to /*moshk'-/ or /*mosk'-/). But as pointed out above, Greek kh could also be used to transcribe the foreign phoneme /x/ which was absent from the Greek phonological system. Hence Strabo's Moschi and Moschian Mountains correspond to Georgian and Old Armenian Mesx-, but whether Georgian Mesx- (Meskh-) with, actually, only two phonemes, /m/ and /s/, coinciding with those in the term /*mosk'-, *musk'-/, can be identical with the latter, is very doubtful, since both Georgian and Armenian distinguish /k'/ aspirated from /x/ (84). It is improbable that at a time when the invasion of the Thraco-Phrygians was spreading throughout Asia Minor, [119] some of their tribes should have penetrated into Pontus and even into Transcaucasia and subsequently become speakers of Georgian. Perhaps certain Georgian tribes, which were for a period subject to Phrygia, received the nickname "Phrygians" (Mushki) because of the amount of Phrygian cultural borrowings. Or, finally, perhaps this nickname was widely used for various tribes and peoples by reason of some common historico-cultural features or associations.

One thing is clear, however: in the Ancient Oriental sources the term Mushki was used to designate Phrygia and the Phrygians, whose Indo-European linguistic affiliation is unquestionable (85) Therefore other Thraco-Phrygian tribes may also have been so designated, including the Proto-Armenian ones. Note the important suggestion of A. Goetze (86) that the term Mushki originally referred to the Thraco-Phrygian tribe of the Mysians in northwestern Asia Minor and Troad, and to the province of Moesia in the Balkans. (Gk. Mysoi, read /Musoi/; the stem is *mus-, /sh/ did not exist in Greek.) The Mysians may have been the first Thraco-Phrygian tribe whom the inhabitants of Asia Minor learned to know. Later their name spread to all the tribes related to them or close to them in culture (87).

Should we assume that the Eastern Mushki in Armenia Minor, mentioned by Tiglath-pileser I, Tukulti-Ninurta II, and Assurnasirapal (as well as the Muska- of King Iariris of Carchemish) were not Proto-Armenians, but Proto-Georgian tribes which temporarily found themselves here (88), then the question again arises, in what manner, when, and from where did the Indo-European Armenians come, who subsequently were the stable population of the country in question? We have already indicated that in the 8th-7th centuries B.C. a cattle-breeding, mobile population is not noted in the adjacent areas. Therefore the most probable date for the appearance of the Proto-Armenians in the Armenian Highlands is the 12th century B.C., the century when there actually is historical evidence of great migrations (including, in the first place, the Thraco-Phrygian migration, of which the Proto-Armenians must have been a part). And when we have direct evidence about the appearance, in precisely this place and at precisely this time, of a new tribe bearing the designation which was unquestionably applied to certain Thraco-Phrygian tribes, it can logically be inferred that this tribe, i.e., the Eastern Mushki, should be identified as the Proto-Armenians (who continued to live here also in the future), and not as Georgians, who never lived here afterward, and had they appeared here, this could anyway have happened but accidentally and temporarily. In addition the inscription of Assurnasirapal bears witness to the fact that the Eastern Mushki changed here to a settled way of life.


2.5. The Problem of the Urumeans. The Theory of S. T. Eremyan

The same inscription of Tiglath-pileser I which speaks of the first appearance of the Mushki in the valley of the Upper Euphrates also mentions two more tribes which arrived here at the same time: the Apeshlaian Kaska and the Urumeans. Therefore we must analyse the available data about the Urumeans as well (inasmuch as the Apeshlaians were already discussed above in Chapter One). In connection with this it is necessary to dwell on the theory of Eremyan, which in essence is an attempt to reconcile the "Haiasa" and "Mushki" theories of the Proto-Armenian ethnogenesis, and also the theory of J. Markwart, who saw the ancestors of the Armenians in the "Arimi" (or "Arima") of Homer (89).

In Eremyan's opinion, the Urumeans are to be identified with the Arimi, whom he locates, for the Hittite period, in the territory of Haiasa or thereabouts. Carried along by the general ethnic migrations of the 12th century B.C., the Arimi-Urumeans, along with their neighbors the Kaska-Apeshlaians, descended into the valleys of the Upper Euphrates and the Arsanias. Here, in the region of the modern city of Mush and in the mountains of Sasun, they formed the "country" which is called Urumu, Urme, or Arme (90) in the sources, and which merged with the Hurrian "country" of Shubria. The mountainous region of Sasun could not be subjugated either by the Assyrians or the Urartians, and the nucleus of the future Armenian people and Armenian state system was formed here. It is precisely the Arimi-Urumeans who were the speakers of Proto-Armenian; they merged with the Mushki, or Phrygians, who were related to them by language and who introduced the Thraco-Phry-gian element into Old Armenian.

The theory of Eremyan needs several modifications. First of all, the Arimi are not known to the Hittite sources either in the region of Haiasa or anywhere else (91) The Arimi, whose "right" to claim the role of ancestors to the Armenian people is based only on a certain similarity of names, are once mentioned in the Iliad in a most vague context (in an extended simile) (92). From it there is no way to decide where the author of the poem thought them living; it is also unclear whether he means a tribe of Arimi or a city of Arima (93). All that we can grasp is that he is talking about a volcanic locality. It in no way follows that the Arimi lived in Asia Minor at all, and the ancient commentators placed them in different countries by pure guesswork (94). Moreover, as we have already seen, it is not proven that any Indo-European-speaking tribes lived in the region of Haiasa in the 2d millennium B.C., much less tribes speaking languages of the Proto-Armenian type. And finally, if the [121] Urumeans and Mushki were tribes of such different geographical background, their languages could not be perceived as similar. As far as the Thraco-Phrygian (95) element in Old Armenian is concerned, it is not another stratum of borrowings (viz., from the Phrygians) but is the basic stock of vocabulary.

Just as in the case of the Mushki, we must try to establish the precise ancient phonetic form of the name of the Urumeans. If we connect this term with the Arimi, we have to explain the difference in the vocalization in accordance with the rules of Indo-European apophony, unless we can prove that there was a regular historical development of the phoneme /a/ into /u/ or /o/. Note that the Armenians themselves never, as far as is known, called themselves "Oromi," nor even "Armenians." The medieval term Armn was introduced by foreign conquerors.

It is difficult to say just who the Urumeans were. The source says of them only that they acted jointly with the Apeshlaian Kaska and that they originated from the "country of the Hittites," i.e., that they came from the west, from beyond the Euphrates. They could even have been another Kaska tribe, or a tribe related to the Kaska, or the name could be another designation for the Mushki (this, however, is improbable, since both the Mushki and the Urumeans are mentioned in one and the same inscription although under different years). Finally, they could have been another Thraco-Phrygian tribe of the same origin as the Mushki. The Thraco-Phrygians still lived in a tribal society and must have consisted of a number of small tribes, each with its own name. The tribes as a whole could hardly have had one common self-designation (autonym). But if they were another Mushki tribe, we must explain why the Urumeans entered into an alliance, not with the rest of the Mushki, but with the Apeshlaians, who spoke a foreign language. The Urumeans no doubt finally became part of the Armenian nation, but there is no ground for attributing a more important role to them than to the Mushki.

2.6. The Territory of the Settling of the Speakers of Proto-Armenian

Thus we come to the conclusion that the speakers of Proto-Armenian, known to the ancients under the name of Mushki (the eastern ones), and possibly also of Urumeans, came to the valleys of the Upper Euphrates and the Lower Arsanias in the second quarter of the 12th century B.C., after having destroyed the Hittite Empire.

We may attempt to establish the territory where they settled. The first Assyrian information seems to indicate that the place where the Mushki and the Urumeans settled was the country of Alzi (later [122] Aghdznik', Arzanene), which included at that time the lower part of the Arsanias (Muratsu) valley and the regions to the south of this river. But, as we have seen, there is some reason to believe that the Mushki also settled in the valley of the Upper Euphrates, on both sides of the river. The area of settlement of the Urumeans must probably be identified as the country of Urumu or Urmie; the term Arme seems to denote another region. The land Urmie is first named in the Urartian text of King Minua (UKN 41 c). In the fragment b of the same inscription the region Qulmeri asune is also mentioned, but it has been convincingly argued that the two fragments do not join, so that Qulmeri need not be placed in Urmie. The Assyrian sources place Qul(lim)meri in Shubria (96). The latter is the same term as the Akkadian Subaru, Shubre "Hurrian"; however the ending -ia shows that in this particular form the term is a borrowing in Assyrian, probably from an Urartian dialect. Hence although we still have no documentary proof, it is probable that Shubria is not the same as Urmie, but was known also to the Urartians as *Shubria and/or Qulmeri. Urmie (or Urmio) is also mentioned in the inscription of the Urartian king Argishti I (UKN 155 A 22), but its location cannot be decided from the context. The "Inner Urumu" of Assurnasirapal (Annals, II, 2-14; Grayson [1972-76],2, 8, 551) was situated near Shubria but was different from it (97). According to Esarhaddon's Letter to the God Assur, Shubria had a Hurrian dynasty as late as 673 B.C., and some of the personal names of Shubrians in the letters found in the Assyrian state archives seem also to indicate that the population was Hurrian (98). However Esarhaddon led away the people of Shubria, and it is quite probable that beginning from that time the land was settled by Proto-Armenians. There is evidence that deportees from a region west of the Upper Euphrates were settled there. As to Arme and its city Neheria (99), it is mentioned by Sarduri II (UKN 156 DI DII 11-12). The term probably means "Aramaic country," but may have been used specifically of an Aramaic enclave amid the territories of local speakers in the region in question. The city Neh(e)ria is known since the early 2d millennium B.C. and was situated somewhere between the sources of the Tigris and the valley of the Euphrates. The identification with Nep'erkert-Mayafarkin is probably wrong. The early settlement of Proto-Armenians in Neheria is possible and even probable.

There is no doubt that after coming from Inner Anatolia to the Upper Euphrates valley, the Mushki must have first settled on the right shore of the Euphrates. Along the Euphrates they apparently contacted Carchemish in the 8th century B.C. (100). If Tiglath-pileser I makes no reference to this, it is probably because in that [123] particular context he was not interested in the area beyond the Euphrates. The capital of the XIII ("Armenian") satrapy under the Achaemenids in the 6th-4th centuries B.C., (see above, p. 88), was Melid (Eski Malatya), and at the beginning of the 5th century Herodotus (I, 72, 180) assigns to Armenia, not only the sources of the Euphrates, but also the territory up to the watershed where the river Halys has its beginning. The presence of Thraco-Phrygians in this region is supported, as we have already mentioned, by the "Old Phrygian" archeological culture reaching Eski Malatya (101). But we have also pointed out that all the dynasties on the right shore of the Upper Euphrates, including those in Melid, were Luwian right up to the end of the 8th century B.C. (102)

Thus as the territory where the speakers of Pro to-Armenian originally settled we must consider the area from the Northern Taurus to the spurs of the Armenian Taurus near the sources of the Tigris River (the Sasun Mountains), including the valley of the Upper Euphrates on both sides of the river, i.e., what in the Hittite period were the "countries" of Pahhuwa, Zuhma, Tegarama, Isuwa, Maldia, and Alzi. The area in the center of this territory was called Supa (UKN 39, 3, 10; 128 A2, 22). This is the later Sophene, a term which, as a result of political events, came to be applied in a much broader sense than under the Hittites (103). The aboriginal inhabitants whom the speakers of Proto-Armenian found were mainly Luwian and partially Hurrian on the right shore and mainly Hurrians on the left shore, although, as we have seen, the Luwian element had also penetrated there (see above, pp. 57, 74).

3. The Formation of the Armenian Nation

Up to this point we have been dealing with the problem of finding out who the speakers of Proto-Armenian were. The formation of the Armenian nation itself constitutes a separate problem.

3.1. The Number of Proto-Armenians in Comparison with the Local Population

There is absolutely no doubt that the Proto-Armenians of the Thraco-Phrygian linguistic group were not the only and not even the main component in the formation of the Armenian nation. The number of Mushki and Urumeans could not have been great. The annals of Tiglath-pileser I speaks about the penetration of 20,000 Mushki warriors into Kadmuhi (it is possible that some of these warriors actually might have been local inhabitants of Alzi; and we must not ignore the fact that the figure is almost certainly exaggerated) (104). He also mentions about 4,000 Kaska and Urumean [124] warriors. If we consider that in those times perhaps one out of every four men and women was a warrior, we must assume that the total number of the invading tribes might have been 100,000 or, say, 200,000, including those who did not participate in the campaign against Kadmuhi. But the aboriginal population was more numerous. According to the data of Sarduri II's annals, within the Urartian Empire he abolished the duty of military service for 350,000 men (105). This means that the population of Urartu certainly exceeded one million and may have reached two or three million. The rich valleys of the Upper Euphrates and the Arsanias constituted not less than 25 percent of the imperial territory, so we may assume that its inhabitants comprised at least one quarter of its total population, i.e., from nearly half a million to three-quarters of a million. Thus the number of the local inhabitants exceeded the number of invading tribes about three times or more (*3).

There are no data which would make us suppose that the newcomers supplanted or annihilated the local population, whom the Assyrian sources, as we have seen, indicated to have been active supporters of the Mushki against their common enemy. After the invasion these areas are still called "Subarean," i.e.. Hurrian, and Hurrian and Luwian dynasties (106) continued to rule there.

3.2 The Period of Bilingualism

The very considerable layer of substratal vocabulary in Old Armenian clearly shows that the local inhabitants had not been ousted. For purposes of comparison, note that there are only a few words of the Celtic (Briton) substratum preserved in Anglo-Saxon and in modern English. Or, taking another example, note that even in Georgian the amount of substratal vocabulary is very small (107). Moreover it has been shown that the phonetics of Armenian reproduce, on the whole, the Urartian, and not the Indo-European phonetic system (108). These phenomena speak beyond any question of a long period of bilingualism, when the local population, in the process of changing over to Old Armenian, continued to use the older language as well, speaking Old Armenian according to the rules of pronunciation of its former native language (109) and introducing a great number of words from the latter into Old Armenian.

As we have seen, by the time of the appearance of the Mushki and the Urumeans the local population of the Upper Euphrates valley was speaking Luwian and Hurrian, i.e., it was already very much mixed. The Urartian dominion which had continued for about two hundred years must have also introduced an admixture of an Urartian-speaking population. It is precisely this mixed linguistic [125] nature of the local population, which used no less than four languages in everyday life (Proto-Armenian, Luwian, Hurrian, and Urartian), which explains the fact that in the end, under the conditions of a new political and economical unity, it at first began to use, along with its native language, a second language for mutual understanding, and then changed over to a single language (110).

It might have been expected that this language would be Urartian. However the ancient empires never thrust their language on their subjects (111), being satisfied with gathering tribute and organizing obligatory labor. They were not interested in a cultural subjugation of the population. Relocating inhabitants, as they did, from one end of the empire to the other, they were even interested in a Babel of tongues, which did not permit the subjects to plot among themselves. But the people, feeling the need for mutual understanding, developed a language common to all (at first a lingua franca, i.e., a commonly understandable language for individual instances of communication with neighbors who spoke a foreign language, and then a koine, i.e., a common spoken language, while preserving the local languages and dialects only in household use). As a common language the mass of the population accepted the one that was easier to learn, either by virtue of its being widely spread, or by virtue of its simplicity. For the Hittite Empire the common spoken language apparently was not Hittite-Nesite, but Luwian, which therefore survived the collapse of the empire. In the Assyrian Empire the common language which emerged was not the Assyrian dialect of Akkadian, but the Aramaic of that comparatively recently arrived population, which was still partly nomadic. This was exactly why that language was so widely spread. For the inhabitants of the western areas of the Urartian Empire, Urartian was the language of the officials, which the people did not know; that language prevailed among the elite, to whom the people were strangers. But shortly after the arrival of the Mushki and the Urumeans even those who spoke Hurrian and Luwian probably became only a small local ruling stratum, which was annihilated during the Assyrian and Urartian conquests, just as the Akkadian aristocracy of Assyria was destroyed during the Median conquest.

The spread of Proto-Armenian was greatly assisted by the policy of the Urartians themselves, who used to resettle conquered inhabitants. Thus we know that when the Urartian king Argishti I built the fortress of Erbune at the site of modern Erevan in 782 (or 776) B.C., he settled there people brought from Supa (Cop'k', Sophene) and Hate (Melitia-Melid) ( 112), i.e., precisely from the Upper Euphrates valley, with its mixed Proto-Armenian-Luwian-Hurrian [126] population, which at that time undoubtedly was already using Proto-Armenian as a second and perhaps even as its main language (113). There were many such instances in the history of Urartu, so that ethnic mixing occurred on a very large scale, just as in the case of Assyria.

An important factor in the strengthening of the significance of Proto-Armenian (to the detriment of Hurrian and Luwian, which by about the 8th century may have been spoken only by the local ruling class) could have been the existence of such centers to which the exploited population of Urartu and Assyria could flee: one of them was Shubria. But it was Melid that seems to have been the nucleus around which the Armenian nation formed, perhaps because the social conditions brought by the pastoralists were more liberal. Here the people easily changed over to a common language (just as in Assyria and later in Babylonia the common people began to speak Aramaic earlier than the aristocracy and the citizens of the privileged cities).

3.3. Name

Since the population of the Upper Euphrates valley at the beginning of the 1st millennium B.C. was mixed in its ethnic composition, it probably did not have a common self-appellation which could have been generally used as its name by its neighbors. Therefore the neighbors called the whole group by the name of the one Proto-Armenian-speaking subgroup closest to themselves. For the Georgians such an area was Suhmu, in the north of the valley; for the Aramaeans (*4) it was Arme, which bordered on the northernmost of the regions with an Aramaic population (Amed, Carchemish, etc.). Therefore for the Georgians an inhabitant of the Proto-Armenian-speaking countries was a Somexi (114), while for the Aramaeans he was an *Armnaia (115). The ancient Persians borrowed this term, as well as other geographical terms of Western Asia (116), from the Aramaeans, for it was they who supplied the Achaemenian chancellery with scribes and officials (117). And the Greeks borrowed this name from the Persians, although originally they had used another terms, viz., Melittenians.

As to the Urartians, for them all the inhabitants to the west of their empire were "Hittites" (Hatine), and all the area west of the Euphrates was called Hate. The term "Luwians" is unknown in the 1st millennium B.C. For the inhabitants of western Asia of this time all the Luwians were "Hittites," and they also probably called themselves by that name (118). Apparently the entire mixed population of the right bank of the Upper Euphrates valley were bracketed as "Hittites" (119).

[127] If, as we suppose, the Proto-Armenians lived not only on the left bank, but also on the right, it would be completely natural for Urartians to call their language "Hittite," and the population "Hittites." That was probably the name of the language during the long period of bilingualism. Subsequently, when the Urartians themselves changed over to Old Armenian and merged with the Armenian nation (in which they probably constituted the majority), the name "Hittites" became their own self-designation. In Proto-Armenian this name could have sounded like *Hatiyos; according to the rules of Old Armenian phonetics, this had to develop into hayo- (120).

3.4. The Components of the Old Armenian Nation

Thus from our point of view the Old Armenian nation was formed in the Upper Euphrates valley from three components--the Hurrians, the Luwians, and the Proto-Armenians (the Mushki, and possibly the Urumeans). Moreover the Hurrians and later the Urartians constituted the main mass of the nation and determined the basic line of biological succession, while the language of the nation, because of a certain historical situation, was taken over from the Proto-Armenians. Less significant, apparently, was the contribution of the Luwians (121). This process began in the 12th century B.C. and in the western part of the Highlands was completed by the 6th century B.C. (122)

When in the 6th century B.C. the Old Persian and Greek sources begin to mention the "Armenians" and "Armenia," the first of these terms is applied either to the entire population of the Highlands or to the newly formed Old Armenian nation in the western part of the Armenian Highlands (thus Herodotus) (123).

As to the term "Armenia," it is used in two different senses. Sometimes the source (as, for instance, the Bisitun inscription) does not distinguish between the XIII and the XVIII satrapies; then by the Persians and the Greeks the whole Highlands is called "Armenia," while the Babylonians (124) and probably the Hebrews (125) use the old term Urartu (Urashtu, Ararat). But when the two satrapies are distinguished, then the term "Armenia" applies to the XIII, or western, satrapy (although the Babylonians use the term Melid (126), probably after its capital). The XVIII satrapy is called Urashtu by the Babylonians; Herodotus (III, 94; VII, 79) does not give it a name but states that it was inhabited by the Alarodians (or Urartians), the Matieni (or Hurrians), and the Saspiri (probably western Georgians). Apparently the main part of the population of the XVIII, or eastern, satrapy still spoke Urartian, although the extension of the name "Armenia" to the whole Highlands may be [128] evidence that the Proto-Armenian language had also spread to its eastern half already by the 6th-5th centuries B.C.

The Armenian kingdom probably already existed in the period of Media's hegemony (as Xenophon informs us in Cyropaedia, and which may also be inferred from Ezekiel and from the Old Armenian legends transmitted by Moses Xorenac'i) (127). If so, this fact could have been favorable for the spread of the Proto-Armenian language to the entire territory of the Highlands (128). After the 5th century B.C. the Alarodians are no longer mentioned in history. In all probability the final merger of the Urartians with the Old Armenian nation was accomplished by the period of the Armenian state of the Eruandides/Orontids (4th-2d centuries B.C.) and of Greater Armenia of the Artashesides/Artaxiads (2d century B.C.) (129). With respect to numbers and the importance of their cultural contribution, the Urartians may have been the strongest component of the Armenian nation.

3.5. The Hurrians, Urartians, and Luwians Beyond the Confines of the Old Armenian Ethnic Area. The Georgians

We must not forget the enormous cultural impact the Urartian state and culture also had beyond the confines of the Urartian linguistic territory proper. In this sense the cultural heritage of Urartu is the property of all the peoples of Transcaucasia and not of the Armenian nation alone. Moreover during the period of Urartian domination a certain number of Urartians must have settled throughout the entire territory of the empire (130) and thus merged with the local inhabitants, and not just with the Armenian-speaking population.

The main mass of the Urartian-speaking people, however, lived inside the territory of the formation of the Armenian people and merged with it. As far as the Hurrians, the "Etio," and the Luwians are concerned, a considerable part of them lived outside of this territory and thus did not merge with the Armenians. In the Middle Ages, instead of the mountaineer Hurrians (?) in the eastern part of the Armenian Taurus and Kurdistan Mountains (the Carduchi), as well as instead of the Quti, we find the Kurds, who speak a language which is obviously a descendant of Median, but which developed in a specific direction. In it we should perhaps see the influence of a Hurrian substratum (131). A considerable portion of the "Etio" must have merged with the Albani in the east of their area and with the Georgians in the west (132), while within the confines of the Ararat valley and the adjacent territories they must have merged with the Armenian nation. Likewise the northwestern group of Hurrians may have merged with the [129] Georgians [133], who, as we have surmised, spread out widely during the course of the 12th (?) to 8th centuries B.C., taking in the aboriginal population of many areas of Transcaucasia and Pontus. From this time on the Georgians appear as one of the leading ethnic groups of the territory under study. Geographically, culturally, and politically, they occupied a very important area, sharing a common cultural substratum with the Armenian nation (134).

As for the Luwians, a great portion of them lived far to the west of the main region of the formation of the Armenian nation. Their descendants were, in all probability, the inhabitants of Cilicia, Cataonia, Pisidia, Lycia, and Caria of the Hellenistic period; they were subsequently Hellenized. Considerably later some eastern groups of the long since Hellenized Luwians were also subjected to Armenization.

3.6. Conclusion

From all that has been set forth it is evident that the history of the Armenian people is a direct continuation of the history not only of the Proto-Armenians, but also (and to no lesser degree) of the Hurrians, the Urartians, and the Luwians. The main mass of the Armenian nation consists of their descendants; there was a historical moment when a person might speak Old Armenian, his father, grandfather, or great-grandfather was more likely to have been bilingual, while his forefather was a pure Hurrian or Urartian. It is difficult to determine the moment of change. There are no written sources in Old Armenian before the 5th century A.D., and the proper names preserved by the older sources do not tell us much. From the Armenian historical tradition we know, for example, that the king Artashes I in the 2d century B.C. was an Armenian, but he bore an Iranian name and used the Aramaic language and the Aramaic writing system for official purposes. In view of this, and inasmuch as the Armenian historical tradition does not go back any further than the 2d quarter of the 1st millennium B.C., how can we be sure that all the dynasties of the Highlands in the 9th-7th centuries B.C., which bore Hurrian or Luwian names and used Luwian Hieroglyphic writing or cuneiform, were still really Hurrians and Luwians (135)? But from the historical point of view this is not too important: their subjects, in any event, were the direct ancestors of the Armenian-speaking inhabitants of these lands in the following centuries. Therefore in studying the ancient socio-economical or cultural history of the Armenian people, we cannot begin it from scratch and look for primitive tribal relations in the 6th and 5th centuries B.C., as has frequently been done. There is no question but that ancient Armenian history can correctly be understood only [130] as the continuation of the more ancient history of the Hurrians and the Urartians, as well as that of the Luwians (136).

It is now recognized that the rudiments of the Armenian state go back not only to the epoch of the collapse of Urartu and Assyria, but even further. Piotrovsky believes that its nucleus was the kingdom of Shubria, which he identifies with Arme. He supposes that a Scythian-Armenian league emerged here in the 7th and 6th centuries B.C., but that Shubria itself already had been a state earlier. Perhaps it is more likely that the nucleus of the Armenian nation is to be sought in the kingdom of Melid, capital of the XIII satrapy of Armenia in the 5th century B.C., and perhaps also the capital of the Armenian kingdom of the legendary Tigran I in the 6th century B.C. Melid was also the "Land of Hatti" of the 12th-8th centuries B.C. The Mushkian kingdom of Alzi in the 12th to 9th centuries B.C., later included in Urartu, can also be viewed as one of the nuclei of the Armenian state, but to a certain extent we can view as such nuclei any Hurrian, Urartian, or Luwian kingdom in the Armenian Highlands. These states were also created by ethnic groups which were not foreign to the Armenians, by people whose descendants merged with the Armenian nation, although at the time they themselves still spoke other languages.

These links open in separate windows:


Return to First Page of Article
Hittite, Hurrian, Urartian Page
Classical Antiquity Menu
--   This is a mirror of one of Robert Bedrosian's web pages   --