Early Bronze Age migrants and ethnicity in the Middle Eastern mountain zone

Significance

This analysis shows the complex interaction of ethnic groups in antiquity, adapting to new locations and adopting and ultimately, assimilating into a majority culture. It occurs in a background of mountain valleys and highland plains, where ever-shifting populations carve out a living and an identity.

Abstract

The Kura-Araxes cultural tradition existed in the highlands of the South Caucasus from 3500 to 2450 BCE (before the Christian era). This tradition represented an adaptive regime and a symbolically encoded common identity spread over a broad area of patchy mountain environments. By 3000 BCE, groups bearing this identity had migrated southwest across a wide area from the Taurus Mountains down into the southern Levant, southeast along the Zagros Mountains, and north across the Caucasus Mountains. In these new places, they became effectively ethnic groups amid already heterogeneous societies. This paper addresses the place of migrants among local populations as ethnicities and the reasons for their disappearance in the diaspora after 2450 BCE.

Ethnicity is not important only as the prehistory of modern groups; it was an important structuring principle in many societies in the past. Understanding ethnicity, then, is a necessary precondition to adequate understanding of the past… (1).

The effort to understand the role of ethnic identity in forming distinct cultural traditions and defining group interactions requires analytical clarity. First, one must be clear that ethnicity only occurs(ed) in heterogeneous societies, which means societally complex ones. Second, an appropriate and specific definition is necessary. Barth (2) defines ethnicity according to four criteria. Ethnicity “1. is largely biologically self-perpetuating, 2. shares fundamental cultural values […], 3. makes up a field of communication and interaction, and 4. has a membership which identifies itself, and is identified by others, as constituting a category distinguishable from other categories of the same order” (2). A shorthand version of the last criterion is if I say I am and you say I am, then I am a member of a particular ethnicity. This idea of ethnic identity is not always immutable; it adapts to conditions. When I worked in the former center of Armenian occupation in modern eastern Turkey, the Muş Province, a local grocer whom I befriended declared himself a Muslim Kurd, although pictures of his grandparents holding Armenian crosses hung in his house. Other than the biological aspect of this definition, which archaeologists are only beginning to investigate, criteria by Barth (2) affect archaeologists’ analysis in that there are two distinct aspects of the definition. Gross (3) calls one the Essentialist definition, which emphasizes culture content: “Given the emphasis on culture-bearing aspect, the classification of persons and local groups as members of an ethnic group must depend on [consciously] exhibiting the particular traits of the culture” (2). Gross (3) contrasts this with the Boundary definition, which emphasizes, on the one hand, competition and conflict between ethnic groups and on the other hand, a possible division of labor, in which each has an occupational specialty on which the others depend. For example, the analysis by Barth (2) of the ethnic division of labor in Afghanistan cites the different ecological niches in which various ethnic groups lived.

A more traditional archaeological view of ethnicity often conflates ethnicity with culture (ethnos) and even race (4). The problem with the traditional definitions and the reason that I chose the modern anthropological one is that ethnicity is not, as I mentioned above, static. Although ethnic groups share(d) a common cultural heritage, which may have been derived from an original culture area, it can only be understood in the context where it currently exists or existed archaeologically. For example, Armenians constitute a strong, self-conscious ethnicity outside the nation-state of Armenia. However, to understand “Armenian-ness” in Los Angeles and Damascus as if it were identical is a mistake in my opinion. The cultural context, the nature of social boundaries, and the kinds of interactions are all different. As Roaf (4) asserts, ethnicity as such is hard to establish archaeologically, especially in prehistoric times. Nonetheless, to me, ignoring significant elements in an explanatory model because they are hard to assess is admitting defeat before one starts.

One should use the concept and its explanatory power when one can. Such a case, I believe, is the Kura-Araxes cultural tradition.

Kura-Araxes Case

The societies of the so-called Kura-Araxes cultural tradition that emerged in the highlands during the fourth millennium and continued into the early third millennium BCE (before the Christian era)—Shengavitian, Karaz, Pulur, Yanik, Early Transcaucasian, and Khirbet Kerak are some of its other names—present scholars of the Greater Middle East and Eurasia with a laboratory for studying the evolution of human cultures and the societies that they spawned in highland zones, a topic much studied in ethnography and less so in archaeology (5–8) (Fig. 1).

Fig. 1.

Distribution of Kura-Araxes cultural tradition.

Kura-Araxes as a Cultural Tradition.

How do we know that we are dealing with groups of ethnic migrants? The commonly cited answer is that the Kura-Araxes groups are marked by distinctive pottery styles (9–11) (Fig. 2). This pottery corpus consists of very distinctive handmade, black burnished pottery, often with incised or raised designs. In the latter, a thick layer is added, and then, all but the design is removed, like a faux appliqué. The cultural importance of this pottery style is that it dominates the area of the earliest appearance of the Kura-Araxes cultural tradition for a millennium or more. In areas out of the homeland of the South Caucasus [the current nation-states of Georgia, Armenia, Azerbijan (specifically Nachiçevan), and northeastern Turkey], the same pottery begins to appear, most prominently in the late fourth and early third millennia BCE. Its contrast with the local buff, wheel-made pottery makes the Kura-Araxes black ware seem out of place. Based on this fact, archaeologists have argued that this pottery represents migrants, who became, in effect, ethnic groups within the areas of the Taurus Mountains (12), the Zagros Mountains (13), the Amuq, Galilee (14), and north of the Caucasus Mountains (15). Whereas the idea that pottery style alone can define social groups—pots equal peoples (4, 11, 16)—is suspect and was used by many 19th and early 20th century archaeologists to reach rather unsophisticated and frankly, wrong conclusions about social process, in this case, it may be valid (11, 17). More than simple style is evident. Based on studies of the inclusions in the fabric of pottery bodies—pure clay cannot be used to make stable pottery without the addition of tempering materials—petrographers who study the makeup of pottery fabric determined that the residents of Bet Yeraḥ (Khirbet Kerak) in the Southern Levantine Early Bronze III (2650 BCE) were using the same techniques as those of Kura-Araxes sites hundreds of years earlier and as many kilometers away in Armenia (18, 19). The Kura-Araxes/Khirbet Kerak people and the local Early Bronze III people did not even use the same clay sources, except for some cooking pot wares.

Fig. 2.

Sample of Kura-Araxes pottery styles. Early Bronze Kura-Araxes: (A) Ravaz (north of Lake Urmia), (B1 and B2) Godin Tepe level IV, (C1 and C2) central western Zagros and Arslantepe (Malatya Plain), (D) Dimple and Line (Muş Plain), (E) Georgian, (F1–F3) Shengavit (Ararat foothills), (G) Karaz (Erzurum), and (H) Khirbet Kerak (Galilee littoral). Middle Bronze Bedeni/early Trialeti: (I1 and I2) Sos Höyük level IVA, Erzurum.

In addition, a repeated pattern of migration is evident. In the Muş Province west of Lake Van, new sites appeared in the mountainous areas with early Kura-Araxes pottery followed by the appearance of greater population (determined by number of sites and total hectares) in the valley bottom. In the latest phases, residents were using pottery that was a fusion of Kura-Araxes and local Late Chalcolithic techniques (20). Farther west in the Taurus and the Levant, another pattern appeared. Archaeologists found small percentages of Kura-Araxes pottery first at central (or town) sites followed by the appearance of small sites dominated entirely by people using Kura-Araxes–styled pottery (12). At the same time, Frangipane and Palumbi (21) argue that, in the Upper Euphrates area near Arslantepe (Fig. 1), the fourth millennium BCE black burnished ware is not the same as the Kura-Araxes ware, deriving instead from central Anatolia to the west. In the Zagros, Kura-Araxes wares, often with design techniques different from those farther west, appear in newly founded and central sites, whereas neighboring areas lack any evidence of Kura-Araxes presence (pottery) (13).

This ethnic identity is not, however, limited to pottery style. Archaeologists found remains of a common religious ritual that spans large areas of the South Caucasian homeland and the immigrant diaspora. This common religious practice is represented by a type of ritual emplacement with decorated fireplaces (either ceramic or a free-standing andiron) (22, 23). House construction and layouts are also distinct. These practices indicate common structures of social groups (family) and activities (13). Metals items also share common design patterns, especially in decorative objects. Spiral earrings and pins with two spiral ends occur across the range of the Kura-Araxes (11, 24).

Taken together, these different kinds of data suggest that there were Kura-Araxes ethnic groups within a large expanse of the upper Middle East, mostly in mountainous zones. To the west of the homeland, they were, in most cases, a minority within their new homelands and did not dominate or replace the local populations. In the Zagros to the east, they seem to be more dominant (see below). According to definitions of ethnicity by Barth (2) and Gross (3), they had the key elements of ethnic groups. They shared a symbolic identity, which was distinct and recognizable from their surrounding populations, and maintained a field of communication and interaction that extended all of the way back to their homeland. If the ritual practices are widespread—we have only a few excavated examples—their common religious ritual certainly represents a common set of values (23).

Furthermore, Anthony (25) relates these kinds of common ideological and ethnic identifiers through style (in this case, pottery, architecture, and metals) to language, one of the most prominent components of ethnographically and historically documented ethnic groups. Anthony (25) correlates distinctive regional languages and dialects to what he calls “Material-Culture Frontiers” (25). These frontiers are created by different sources of immigration and also, different ecological boundaries. The more self-sufficient that a group is in a particular environment, the more likely that they will tend toward a local language and a local, more homogeneous cultural style. This pattern may explain, in part, the many local variations in Kura-Araxes pottery designs and shapes, even in the homeland (26). However, in less certain environments or ones that require resources from a broad set of ecological niches, greater variability in language will exist. The latter describes, in part, the highland domain of the Kura-Araxes.

Kura-Araxes in Broader Ecological Context.

Issues of ecology and adaptation to local conditions within a broader geographical context are important in understanding ethnic identity and change. The Kura-Araxes homeland and the vast majority of its diaspora were mountainous environments. These environments go from the narrow valleys of the higher elevations in areas, like Shida Kartli (27), to the broader lower elevation plains, like Ararat (28, 29). All are in fairly marginal zones for agricultural, although rich in pastureland (30).

One of the largest issues remaining to be resolved is the nature of the broader effect of intercultural interaction on reasons for migration (see below) and the comparison of the lowland and highland cultural structures. Archaeologists are still investigating direct relationships between the South Caucasian homeland and its adjoining regions. The raw materials for metal production north of the Caucasus (the Maikop) certainly were mined in the South Caucasus (15). Whether the Kura-Araxes societies were the source of metal technology—this possibility has long been a supposition of archaeologists—is now being questioned (31). Certainly, the Taurus and Zagros, where Kura-Araxes migrants went, are the key resources areas for the developing state societies of Mesopotamia (32–34). The “supercities” of the Ukraine in the fourth millennium BCE, which are thought to be a result of exchange through the Mesopotamian Uruk expansion trading network (15, 35), had to pass through the South Caucasus. Archaeologists found fourth millennium BCE northern Mesopotamian pottery styles in the Kura-Araxes area of Georgia, and the precursors of the Kura-Araxes exhibit Mesopotamian-related styles (11, 31).

However, from what we know of the effects of the trading relationship with Mesopotamia on local societal development in other parts of the mountain zone (21, 36), such effects are not evident in the Kura-Araxes homeland. Societies and ethnic groups exhibiting the Kura-Araxes cultural tradition are a contrast to contemporary Mesopotamian societies. They were not urban societies, and as far as we can tell, they initially had subsistence economies. They had greater reliance on pastoral than agricultural products. Their productive technologies seem to have been tied to local resources, such as obsidian and flint; copper ores; semiprecious stones, like carnelian; and indigenous plants, such as wheat, barley, and grapes. Their social organization exhibited less centralization and less social differentiation than in Mesopotamia (23).

Causes of Migration.

To understand the role and the fate of ethnic groups in the diaspora, four questions need to be addressed. (i) Why did they emigrate in the first place? (ii) How does that reflect their place in the local societies? (iii) How did their new setting get reflected in their societies? (iv) In this case, why did they, in effect, disappear after 2500 B.C. as a distinct ethnic group?

The first question, after evidence of a migration has been established, is to ask why it took place in enough numbers that we find their remains in mounded sites. However, “while it is often difficult to identify specific causes of particular migrations, even with the help of documentary data, it is somewhat easier to identify general structural conditions that favor the occurrence of migrations. Moreover, particular structural conditions favor migrations of particular types” (37). In other words, we need to understand the natural, cultural, and sociopolitical environments in their homeland and outside in areas of migration at the time that we are studying, and then, if there were significant movements of human groups, goods, or information, we can begin to understand how that changed the adaptations of all of the societies involved (38, 39).

The natural environments and presumably, the local conditions of émigré starting points varied. I have come to see roughly seven environmental zones with populations exhibiting the Kura-Araxes cultural tradition: (i) the higher mountain valleys of the South Caucasus, such as Shida Kartli; (ii) the lower broad areas of the South Caucasus comprising the Ararat Plains into Erzurum and the area north of Lake Van; (iii) Iran east and south of Lake Urmia; (iv) the Taurus zone from the western side of Urmia to the eastern part of Elaziğ; (v) the area of Malatya near the Euphrates, including Arslantepe; (vi) Daghestan into the northern Caucasus; and (vii) the Amuq and Khirbet Kerak of the Levant. These zones tend to elicit somewhat different symbolic languages and somewhat different adaptations. For example, styles in pottery within the overall Kura-Araxes corpus varied among these zones (9). Economically, the more northern, higher elevation societies within the Kura-Araxes cultural tradition tended to have a higher percentage of domesticated cattle than sheep compared with societies in the lower elevation environmental zones. Like dialects in language, these variations were based on the degree of interaction of people in each of these zones (26) and their particular local customs and institutions. On a continuum, they are much more similar to the styles of other peoples of the mountains than those of lowland Mesopotamia.

So, why did they leave as emigrants from their homeland? Following the suggestion by Anthony (25), what we know of the social and economic organization of these societies is that they were small scale, were nonurban, and had subsistence economies in the fourth millennium BCE but may have intensified production through irrigation in some places during the third millennium BCE (40). Areshian (29) posits a population growth in the Ararat Plain at that time.

As I have argued (17, 26), the Kura-Araxes migrations did not take the form of a wave like what Ammerman and Cavalli-Sforza (41) proposed for the spread of agriculture and language into Europe from the Middle East during the Neolithic Period. I see it rather as a series of vectors of movements by small groups, perhaps clans, from place to place. Some have tried to match areas within the Kura-Araxes homeland with some variation in local pottery style within the environmental and style zones that I have listed above (9, 13, 42). As we gain more data, however, this proposed pattern does not seem to work all that well. Rather, a kind of crisscross pattern of groups moving and settling long enough to create mounded sites occurred (these people are not purely nomadic people, although they could be transhumant pastoralists). In northwestern Iran, for example, some small sites are dominated by pottery typical of that in the Shida Kartli highlands in Georgia, and others are dominated by pottery typical of that in the lower mountains of the Ararat Plain of Armenia (43).

So why did they move? Demographers speak of a push and a pull in all migrations (44–46). There must be some reason to be pushed out of an earlier homeland, but there must also be something that pulls the mobile population in a particular direction. Burney and Lang (10) and Areshian (29) have proposed that there was a sharp rise in population in the homeland in the early third millennium BCE. Primogeniture or other customs favoring one family member, kinship segment, or class over another, for example, may cause the less economically fortunate to emigrate. However, was it enough of a rise in population to exceed the carrying capacity of the homeland?

If it was not primarily a push, was it a pull? Moch (47) writes of modern migrations that “I argue that the primary determinants of migration patterns consist of fundamental structural elements of economic life: labor force demands in countryside and city, deployment of capital, population patterns (rates of birth, death, marriage), and landholding regimes. Shifts in those elements underlie changing migration itineraries” (47). At the same time, Moch (47) admits that configurations of political economy—this factor is what anthropologists would call social organization—underlie all of the different possible itineraries. As Anthony (46) points out, “[m]igration is a social strategy, not an automatic response to crowding” (46). In this case, environmental conditions were, in fact, propitious in the homeland, and therefore, a forced exit seems less necessary (30).

Therefore, it is possible that there was a push, not so much from demographic stress but from changes in social organization and access to resources, especially arable land and pasture. There are indications of intensification of agricultural production by the construction of irrigation systems to either feed more people or support the wishes of emerging leadership groups (29). At the same time, little evidence of the kind of authority necessary to control resources and the concomitant social stratification is in evidence (23). The beginning of mobile populations, marked by kurgans and the contemporaneous building of walls at Shengavit, Mokhra Blur (23, 28, 29), and Ravaz (48), is already evident in the early third millennium BCE. Ultimately, in the homeland, the Kura-Araxes adaptation would be displaced by a more mobile and militaristic one associated with the so-called Kurgan Cultures (15, 49). Is that a significant cause? Also, as the more mobile, militaristic organization of the homeland evolved after the end of the Kura-Araxes adaptation, the black burnished pottery continued (50), indicating some continuity of the cultural tradition.

The evidence, although still slim, suggests that emigration was catalyzed more by a pull than a push (17, 13, 51). If so, the likelihood is that some economic opportunities presented themselves in the diaspora. Three productive activities are possible: animal meat or byproducts (wool), metals and metallurgical skills, or viniculture.

To trace these possible pulls, I will take a more detailed look at a few of the migration sites. The first is Godin Tepe in the Kangavar Valley of the central western Zagros (13). Kura-Araxes migrants had entered the Zagros Mountains in the fourth millennium BCE. Archaeologists at Gijlar Tepe west of Lake Urmia uncovered 10 m of occupation with Kura-Araxes artifacts. Considerable time depth is also evident at Yanik Tepe on the eastern side of Lake Urmia (10) and Sangalan Tepe in Hamadan (the last destroyed by modern villagers in the 1970s). The story of Kura-Araxes people at Godin began in the late fourth millennium, when contact between the Mesopotamian lowland and the western Zagros along the High Road (later called the Silk Road) was established. Excavators recovered a few clearly Kura-Araxes potsherds in the oval-walled compound of late fourth millennium BCE Godin VI:1.

What would have pulled these Kura-Araxes people, who never went south of Kangavar or west into the Mahi Dasht, to Godin? The traditional answer has been metallurgy. Sagona (31) suggests that metallurgical work may not have been extensive enough to produce a regular flow of export products. However, there is a significant smelting installation in the third millennium BCE occupation of Godin IV:1a (13). There are, however, two alternatives. One is wine (52). Areshian et al. (53) have shown that the full process of winemaking was already known and practiced in Armenia in the fifth millennium BCE. The first clear evidence of wine at Godin Tepe occurred in VI:1 (54), just when the few Kura-Araxes potsherds appeared. Another less visible product is wool. Woolen cloth was supposed to be one of the major exports of Mesopotamia societies in the Uruk expansion (35). However, Anthony (25) points out that highland sheep had thick wool with long strands best for making yarn as opposed to the lowland sheep. After the end of VI:1 and a hiatus of about a century around 2900 BCE, significant numbers of Kura-Araxes migrants reappear at Godin Tepe. Now, rather than a small minority, they dominate Godin and the other medium-sized site in Kangavar, although the overall population of the valley remained the same as during VI:1. Sheep and goats in Godin IV tended to be killed at 4–6 y of age, old for their use as primarily meat sources but right for wool or milk production (13). The distribution of the Zagros variant of Kura-Araxes–styled pottery follows an old, inner-Zagros Mountain route that swings to the east south of the Caspian Sea. Exactly what was passed along that route we do not know, although lapis lazuli in the late fourth and early third millennia BCE followed a route to Susa (55).

For a period from 2900 to 2700 BCE, Kura-Araxes people seem to dominate the Kangavar Valley. No evidence exists of a mass departure of the local population, and some sites seem to use the older pottery. Others seem to adopt Kura-Araxes styles. However, by 2600 BCE, although the Kura-Araxes cultural tradition is still active in homeland South Caucasus, a transition occurred. Now, only 10% of the pottery was black ware, and it was undecorated. The architectural form and hypothetically, the family structure changed. As the Awan highland empire took over the Kangavar Valley (56), all vestiges of the Kura-Araxes cultural tradition disappeared. They had created a unique variant of the Kura-Araxes ethnicity in the Zagros ecological zone and then, could be conquered and assimilated by organized forces that were part of the ethnic mix all along.

To the west, the situation is quite different (12). The pattern is of small numbers of Kura-Araxes pottery and also, some ritual items, like ceramic decorated hearths at larger sites, like Korucutepe and Norşuntepe (57). This area between the Taurus massifs along the Murat River in Elaziğ Province consisted, however, of small clusters of largely decentralized communities (26). The key might be one of seeking new lands for pasture, or it might be related to Arslantepe. Arslantepe was a primate center during the fourth millennium BCE with a large temple/palace center (36). Its location at the nexus of routes from the northeast, east, west, and south made it a natural center for trade with Mesopotamia, just as Godin was but on a much smaller scale. The site was also near the largest copper mine south and west of the Caucasus, Zagros, and high Taurus at Ergani Maden. Frangipane and Palumbi (21) see the fourth millennium BCE black burnished wares at Arslantepe VIA as typical of central Anatolia, but after the collapse of the temple/palace organization, a combined population of people using a distinct variant of Kura-Araxes wares and local Plain Simple Wares occupied the site. Houses in the third millennium B.C. town had either black burnished or local buff wares (58), and some leader’s house was part of a larger feasting center, like the one at Godin. The immigrants, once a small percentage of the population in the fourth millennium BCE, had become a more significant, although still ethnically identifiable group, in the third millennium. The third millennium BCE also saw a marked increase in population around the site of Arslantepe (59–61), although many of these new sites seem to be short-term occupations on rocky outcrops. Perhaps these places are temporary sites for pastoralists, whom Frangipane and Siracusano (62) see as the Kura-Araxes migrants. An increase in the percentages of mobile sheep and goats at this time reinforces this theory (62). At the same time, a “royal” tomb over the remains of the abandoned temple/palace complex suggests closer cultural ties with the Kura-Araxes homeland and the Maikop cultures north of the Caucasus Mountains, especially in terms of metallurgy (63). We tend to want each ethnic or cultural group to be homogeneous, but because I believe that the zones of Kura-Araxes occupation are environmentally distinct, there is no reason to assume that they all shared the same economic adaptation in the diaspora. Some may have been primarily pastoralists; others may have been farmers and herders, who were more likely to settle for a longer time. Either group could have served as transmitters of goods and technologies.

A pull, again, for opportunities (perhaps metals and metallurgy, the byproducts of sheep and goat, or viniculture and winemaking) drew Kura-Araxes migrant clans over a few hundred years into the Elaziğ and Malatya areas. There is little indication that the interactions were at all violent. Ethnic communities over time integrated into these populations while keeping their identity. As in the Zagros, over time, these populations began to assimilate into the local populations. At Norşuntepe, the symbolic designs continued (57) but appeared as painted designs on buff wares rather than incised or raised designs on handmade black burnished wares. By the end or past the end of when the Kura-Araxes adaptation described the lives of people in the South Caucasus, the people at Norşuntepe built a Mesopotamian-style public building (“Palas”), and signs of the Kura-Araxes ethnic identity all but disappeared. Those same symbols of pottery style, despite the radical change in the lifeway of people in the South Caucasus, continued in the homeland.

Conclusion

This paper looked at the creation through migration of ancient populations related to the Kura-Araxes cultural tradition. The creation of distinct ethnic groups outside the homeland seems to have been based on a pull of population into areas where there were new opportunities to market skills in viniculture, metallurgy, and wool production. The migrants formed an identifiable ethnic group within local populations, as Barth (2) suggests, sharing values, identity, and communication networks while slowly integrating and ultimately, assimilating into local cultures. Only in the homeland, where a density of homogeneous cultural practices existed, did these same cultural traditions continue. This pattern was so, despite a major change in their lives from relatively peaceful, largely settled agropastoral and locally directed craft production practices to a mobile and military lifestyle. In many ways, this same pattern is evident historically, such as in the United States, where people of distinct and strong ethnic identities migrated for opportunity and to fill needed productive and labor roles. Over the less than two centuries since the high point of US immigration, a similar process of assimilation and slowly losing the elements of ethnic identity or incorporating them (like pizza, Chinese food, or participation in St. Patrick Day parades) into a new common, nonethnic social order continues to occur.