Origins, Homelands and Migrations: Situating the Kura-Araxes Early Transcaucasian ‘Culture’ within the History of Bronze Age Eurasia
Posted by Sjur Cappelen Papazian on October 13, 2013
The Copper Age of the Caucasus or, more precisely, the immediately pre-Maikop and pre-Kura-Araxes horizons of, respectively, the northern and southern Caucasus appears remarkably impoverished relative to Chalcolithic developments farther south in northern Mesopotamia or even more to the spectacular Cucuteni-Tripol’ye complexes and related cultures in southeastern Europe during the 6th through the first half of the 4th millennium BC.
Even more striking is the underdevelopment of the northwestern Caucasus prior to the emergence of the famous Maikop culture, which most specialists now date as beginning at least towards the middle of the 4th millennium, if not earlier, to the end of the first quarter of the 4th millennium.
Such underdevelopment in Chalcolithic times, of course, contrasts sharply with what occurred during the Early Bronze Age when the Caucasus became one of the main suppliers of arsenical copper/bronzes to the peoples of the steppes, particularly to the Pit (Yamnaya) and Catacomb Grave cultural communities, and to Near Eastern cultivators farther south.
The northern Caucasus from Maikop times through the Middle Bronze period may have functioned as the critical intermediary for receiving metals, many of which may have originated in the southern Caucasus, and for producing and transshipping arsenic copper/bronze artefacts to the steppes.
Clearly a major shift in interregional relations occurred initially sometime around the second quarter to middle of the 4th millennium BCE that brought the Caucasus onto the main stage of developments encompassing both the steppes to the north and the mixed agricultural/herding and settled agricultural regions of the Ancient Near East to the south.
The emergence and development of the Maikop culture-historical community must ultimately be related to the subsequent advent of the Kura-Araxes culture-historical community. While these broadly defined Early Bronze archaeological cultures represent distinct formations in terms of most features of their material remains (architecture, subsistence economy, burial practices, etc.), they nevertheless formed integral parts of the same overarching system, a field of shared technologies that partially define what the Russian archaeologist Chernykh has termed the “Circumpontic Metallurgical Province”.
The Maikop parallels with northern Mesopotamia or, more broadly, with the Ancient Near East, and the seemingly consistent and growing number of calibrated radiocarbon determinations not only more securely date the ‘Maikop phenomenon’, but also suggest some connections—albeit hard to specify—with larger historical processes, such as the northern Mesopotamian incursion into the Caucasus and then the later so-called ‘Uruk expansion’ north along the Upper Euphrates into eastern Anatolia.
Pre-Kura-Araxes/Late Chalcolithic materials uncovered from the settlement of Boyuk Kesik and the kurgan necropolis of Soyuq Bulaq in northwestern Azerbaijan, and materials from a pre-Kura-Araxes kurgan, Kavtiskhevi, in central Georgia, can be related to remains from the metal-working Late Chalcolithic site of Leilatepe on the Karabakh steppe near Agdam and from the earliest level at the multi-period site of Berikldeebi in Kvemo Kartli. They reveal the presence of early 4th millennium raised burial mounds or kurgans in the southern Caucasus.
Similarly, likens chaff-faced wares collected at Hanago in the Sürmeli Plain and Astepe and Colpan in the eastern Lake Van district in northeastern Turkey are similar with those found at the sites mentioned above and relates these to similar wares (Amuq E/F) found south of the Taurus Mountains in northern Mesopotamia.
These resemblances is interpreted as representing an intrusion of north Mesopotamian immigrants, if not colonists, into the southern Caucasus prior to the well-known ‘Uruk expansion’ north along the Upper Euphrates River. Their arrival is seen as roughly contemporaneous with the seemingly sudden emergence of the Maikop culture of the northwestern Caucasus with its wealth of metal vessels, tools, ornaments and weapons.
The latter, however, is not interpreted as a direct consequence of this northern Mesopotamian incursion into the highlands, but is viewed as also somehow related to the collapse of the earlier Southeast European hearth of metallurgical activity or the so-called Carpatho-Balkan Metallurgical Province.
The new high dating of the Maikop culture essentially signifies that there is no chronological hiatus separating the collapse of the Chalcolithic Balkan centre of metallurgical production and the appearance of Maikop and the sudden explosion of Caucasian metallurgical production and use of arsenical copper/bronzes. More than forty calibrated radiocarbon dates on Maikop and related materials now support this high chronology; and the revised dating for the Maikop culture means that the earliest kurgans occur in the northwestern and southern Caucasus and precede by several centuries those of the Pit-Grave (Yamnaya) cultures of the western Eurasian steppes.
The calibrated radiocarbon dates suggest that the Maikop ‘culture’ seems to have had a formative influence on steppe kurgan burial rituals and what now appears to be the later development of the Pit-Grave (Yamnaya) culture on the Eurasian steppes.
Roughly 150 Maikop burial complexes have been excavated, while there are only ca. 30–40 known Maikop settlements (or even fewer, cf. Korenevskii 2001), only a handful of which have been substantially excavated. The mortuary assemblage to settlement ratio for Maikop remains is heavily weighted towards the former, and this situation is almost the opposite of what is known for the Early Bronze Kura-Araxes culture-historical community of Transcaucasia to the south, which probably begins slightly later towards the middle or third quarter of the 4th millennium. Hundreds of Kura-Araxes settlements have been found, scores of which have been excavated, while very few Kura-Araxes cemeteries have been located and investigated.
It is primarily this difference in the nature of the archaeological evidence that explains the apparent greater wealth of the Maikop metals relative to that of the Kura-Araxes culture. Both areas were working – and probably producing – metals on a large scale, though there are more metal artefacts from the Maikop culture just because more rich ‘royal’ kurgans have been uncovered.
Kurgan burials are not characteristic of northern Mesopotamia, but at least eight Chalcolithic and presumably pre-Maikop kurgans have been excavated in the central northern Caucasus and in the Kuban area. Early kurgans with Maikop or Maikop-related materials also appear on the Middle and Lower Don on sites of the so-called Konstantinovka culture, some materials of which, such as characteristic asymmetric flint arrowheads, show clear parallels with Maikop remains.
Two early pre-Pit-Grave kurgan burials with the actual remains of wooden wheels have been found respectively in the Lower Don (Koldyri, burial 7, kurgan 14) and Kuban (Starokorsun, burial 18, kurgan 2) areas. Their appearance in these latter areas is due to “the migration or re-settlement of groups from the agricultural population” farther west.
The latter burial, which also contained the remains of a wagon with wooden wheels (ca. 60 cm in diameter), has been attributed to the “early Novosvobodnaya” phase of the Maikop culture, and the partial remains of a similar wheeled cart were found in a kurgan at Tsagan-nur in Kalmykia to the northeast that also apparently contained Maikop-related materials.
This latter collection of heavily worn and new metal tools, including pickaxes and adzes, may document the long-distance exchange of finished arsenical copper/bronzes with high nickel content, but it also underscores the similarity of highly functional metal tools and weapons shared by the distinct but interconnected Maikop and Kura-Araxes culture-historical communities.
The ‘homeland’ (itself a very problematic concept) of the Kura-Araxes culture-historical community is difficult to pinpoint precisely, a fact that may suggest that there is no single well-demarcated area of origin, but multiple interacting areas including northeastern Anatolia as far as the Erzurum area, the catchment area drained by the Upper Middle Kura and Araxes Rivers in Transcaucasia and the Caspian corridor and adjacent mountainous regions of northeastern Azerbaijan and southeastern Daghestan. While broadly (and somewhat imprecisely) defined, these regions constitute on present evidence the original core area out of which the Kura-Araxes ‘culture-historical community’ emerged.
Kura-Araxes materials found in other areas are primarily intrusive in the local sequences. Indeed, many, but not all, sites in the Malatya area along the Upper Euphrates drainage of eastern Anatolia (e.g., Norsun-tepe, Arslantepe) and western Iran (e.g., Yanik Tepe, Godin Tepe) exhibit – albeit with some overlap – a relatively sharp break in material remains, including new forms of architecture and domestic dwellings, and such changes support the interpretation of a subsequent spread or dispersal from this broadly defined core area in the north to the southwest and southeast.
The archaeological record seems to document a movement of peoples north to south across a very extensive part of the Ancient Near East from the end of the 4th to the first half of the 3rd millennium BCE. Although migrations are notoriously difficult to document on archaeological evidence, these materials constitute one of the best examples of prehistoric movements of peoples available for the Early Bronze Age.
The term ‘Kura-Araxes’ were coined to describe the very recognizable black- and red-burnished, hand-made ceramics now attributed to the ‘Kura-Araxes’ or, in the Western literature, ‘Early Transcaucasian’ (ETC). This because at that time all the sites with these materials, particularly the very diagnostic ceramics, which were known at that time, were found in the greater catchment areas of the Kura and Araxes basins.
We now know that these materials, first discovered at places like in the Gyandzha region of Azerbaijan and at the site of Shengavit now in Yerevan and often found in the lowest levels of many later Bronze Age sites, is distributed far beyond Transcaucasia itself, spreading at some point southeast along the eastern slopes of the Zagros at least as far as west central Iran (e.g., at the Godin IV settlement) and into eastern Gilan province of northern Iran and onto the central Iranian Plateau and also southwest across northeastern Anatolia at least as far as the Amuq Plain and into northern Israel during the Early Bronze III period.
Sometime around the middle of the 4th millennium BCE or slightly subsequent to the initial appearance of the Maikop culture of the NW Caucasus, settlements containing proto-Kura-Araxes or early Kura-Araxes materials first appear across a broad area that stretches from the Caspian littoral of the northeastern Caucasus in the north to the Erzurum region of the Anatolian Plateau in the west. For simplicity’s sake these roughly simultaneous developments across this broad area is considered as representing the beginnings of the Early Bronze Age or the initial stages of development of the Kura-Araxes/Early Transcaucasian culture.
The internal periodization of the Kura-Araxes ‘phenomenon’ within Transcaucasia has been worked out by several scholars, and are generally divided into four sub-periods (EBI–IV), extending from ca. 3500 – 2300 BCE, even if it still must be emphasized that much guesswork is involved and that the internal sequence requires additional refinement and corroboration.
The Kura-Araxes culture seems to have emerged in different places—northeastern Anatolia, the broad area of the southern Caucasus drained by the Upper and Middle reaches of the Kura and Araxes Rivers and the adjacent region of northwestern Iran, and the Caspian coastal corridor and contiguous mountainous regions of northeastern Azerbaijan and southeastern Daghestan—exhibiting different regional features at approximately the same time, towards the middle of the 4th millennium BCE.
Kura-Araxes sites are found throughout all areas of Transcaucasia, except for the sub-tropical Colchidean basin of western Georgia, and are located in markedly different environments at different altitudes, including burial sites even in the valleys of the high Central Caucasus Range, such as at Giorgitsminda and at Mutso, on the border between Georgia and Chechnya in Pirikita Khevsureti (Gogochuri and Ghlonti 2003); Kura-Araxes settlements high in the Great Caucasus Range have also been documented at Shatili, Vakissopeli and Sviana-rostiaanebi, and this last settlement is also supposed to contain later Bedeni and Trialeti-like materials (Gogochuri, personal communication). Not surprisingly, settlement sites found high in the Great and Lesser Caucasus Ranges or on highland plateaus typically contain stone architecture and have relatively thin cultural deposits (sometimes barely exceeding 1 m).
Sites farther south on the fertile Ararat Plain of southern Armenia and Nakhicevan or in the eastern piedmont between the Guru and Kandalan Rivers in southeastern Azerbaijan (at Garakapektepe) or even farther south in northwestern Iran (e.g., Geoy Tepe, Yanik Tepe and Tappeh Gijlar) or in eastern Anatolia (e.g., Pulur [10 m] and Karaz [9 m]) are often multi-period tells formed by the decomposition of mudbrick architecture with very thick cultural deposits, at times exceeding 10 m (at Tappeh Gijlar, for example, which is located west of Lake Urmia in northwestern Iran, the unmixed Kura-Araxes levels [period B] are nearly 11 m thick).
It is very difficult to correlate precisely such differently formed settlements. Thus, it is suggested that the initial dispersal of the Kura-Araxes culture is to be found in “the flatlands of the southern Caucasus” (i.e., on the Ararat Plain and farther east in the interfluve between the Guru and Kandalan Rivers of southeastern Azerbaijan) and that it was a movement from these plains into the highlands, associated ultimately with a more productive agricultural economy and consequent population increase.
Regions immediately adjacent to the southern Caucasus (i.e., to the south and west of the Middle Araxes River or northeastern Anatolia and farther to the north and east into southeastern Daghestan) should also be seen as part of the formative area for this culture.
Possibly, but the reverse process could also be argued on the basis of the archaeological evidence (earliest sites possibly located in Shida Kartli or even in higher areas) and is more consistent with the historical pattern of mountain valleys becoming overcrowded and sending their surplus population down onto the plains. For example, the Ossetians are known to have moved down from both sides of the Greater Caucasus and into the broader valleys of central Georgia during relatively recent historical times. Current evidence does not allow us to resolve this problem.
The postulated movement over time of surplus populations from the restricted mountain valleys onto the plains is consistent with the original local formations of this culture – quite literally, northeastern Anatolia and in the high Caucasus mountains, and such movements in search of more arable land may constitute one of the mechanisms driving the peoples out of Transcaucasia and southeast into Iran and farther southwest into the Upper Euphrates basin and beyond.
To add further ambiguity to the situation, while some areas exhibit a break in material culture remains, others, such as Sos Höyük near Erzurum, show continuity from its earliest so-called Late Chalcolithic level into later Early and Middle Bronze times. While never densely occupied with Early Bronze remains, the Erzurum region in northeastern Anatolia may lie within the original formative region of the Kura-Araxes ‘culture-historical community’, and, consequently as such, did not experience the later dispersal or intrusion of Kura-Araxes peoples into other areas of northeastern Anatolia, such as into the Malatya region (e.g., Norşun-tepe, Arslantepe, etc.).
The characteristic red-and-black burnished wares, one of the hallmark features of Kura-Araxes material remains, may actually have originated at some sites beyond the catchment basins of the Kura and Araxes in today’s northeasternmost Anatolia and subsequently spread east into Transcaucasia as conventionally defined.
There seems to have been fairly rapid intra- and inter-cultural communication among these different contiguous regions, leading relatively quickly to the emergence of a recognizable Kura-Araxes koine or broadly defined ‘cultural-historical community’.
Even within this broadly defined area, some thicker multi-period tells that contain earlier pre-Kura-Araxes Chalcolithic levels show a gap or period of abandonment between the latest Chalcolithic and earliest Kura-Araxes occupations (e.g., at Kyul’tepe I in Nakhicevan), and other multi-period tells with thick Kura-Araxes deposits (e.g., Dzhraovit and Metsamor on the Ararat Plain or Garakepektepe in Azerbaijan) are inadequately published and work on them has only preliminarily, if at all, plumbed the earliest Kura-Araxes levels. That is, we know little about the beginnings of these latter settlements.
Problems of interpretation are further exacerbated by the distinct regional variants of this ‘culture-historical community’. This pronounced regional diversity may, of course, also be explained in part chronologically and suggests that this ‘culture-historical community’ was quite heterogeneous, never representing a single unity or polity.
Some Kura-Araxes sites are located near steep ravines or in fairly inaccessible settings (e.g., Garni), and some (e.g., Shengavit, Mokhra-Blur) appear to have been fortified or located on naturally protected promontories or terraces (e.g., at Kvatskhelebi in Shida Kartli), though it must be emphasized that the dating of such fortifications to the Kura-Araxes occupation has not been established in all the claimed cases.
Many sites, including those most carefully excavated, such as Karnut in northwestern Armenia, were not fortified but represent simple open villages with separate or clustered one-room houses with central hearths, often set at the southern foot or along the lower slope of a local large hill (e.g., the sites of Satkhe and Amagleba in southern Georgia).
Certainly most Kura-Araxes settlements and their accompanying materials in the southern Caucasus exhibit far less emphasis on militarism and defence, reflective of politically insecure and unstable times, than is characteristic for the later Transcaucasian Late Early and Middle Bronze and, particularly, Late Bronze/Early Iron periods (from the second half of the 3rd through the beginnings of the 1st millennium BCE).
Our understanding of the Kura-Araxes ‘phenomenon’ is incomplete, and surprises, like still await us. It is also possible that much larger Kura-Araxes settlements lie buried beneath more massive Late Bronze and Early Iron deposits (e.g., possibly at Metsamor). Based on the currently available published evidence, however, most Kura-Araxes settlements in Transcaucasia are small (rarely exceeding 5 ha in size) and show very little evidence of internal social differentiation. The dwellings in the largest sites, such as Arich (12 ha) on the southern edge of the Shirak Plain in northwestern Armenia or Amiranis-Gora (ca. 4 ha) near Akhaltsikhe in southern Georgia, a site which shows evidence of deliberate terracing, are quite dispersed, not densely packed together.
At most, some Kura-Araxes settlements can be considered fortified towns, but not cities, and such towns do not constitute evidence for a sharply differentiated three-tiered settlement hierarchy. Thus, for example, the Early Bronze occupation at the site of Arich, which is located on a naturally fortified promontory drained by a stream flowing down from the northwestern slope of Mt. Aragats, is surrounded by Late Bronze/Early Iron dwellings and burials and even later (Classical?) fortifications that cover the ca. 12 ha area of the site. It is very difficult to estimate the extent and density of its Kura-Araxes occupation.
Possibly, the largest and most impressively fortified Kura-Araxes town (excepting the major Khirbet Kerak settlements of northern Israel [e.g., Bet Yeraḥ]) is the ca. 15 ha site of Ravaz, which is located southwest of Maku in northwesternmost Iran; its well-built fortification wall with rounded towers separates a densely populated acropolis or core area on a raised promontory from a more sparsely settled lower town. The comparable or even larger Kura- Araxes settlements may lie buried beneath later deposits on multi-period tells, such as Dvin, that are found on the fertile Ararat Plain of southern Armenia, though again, even in such cases, it is unlikely that these buried settlements, which are located presumably in the Kura-Araxes ‘heartland’, could be characterized as real urban centres or cities.
Although the mortuary evidence is fragmentary and unexpected discoveries, like the rich burial at Arslantepe, may occur and alter our understanding, the currently available record does not suggest that the Kura-Araxes societies in Transcaucasia at least were torn apart by internal social divisions. In this sense, the Kura-Araxes materials contrast strongly with those of the Maikop culture to the north or with what appears in the southern Caucasus during the immediately succeeding late Early Bronze period or the time of the monumental ‘chiefly’/‘royal’ kurgans. Individual flat-grave burials have been excavated both within settlements and in cemeteries outside the settlements, as well as small kurgans or barrows associated with or in immediate proximity to Kura-Araxes settlements (e.g., at Satkhe in Djavakheti, cf. Kohl, Carson, Edens and Pearce 1993).
None of these southern Caucasian Kura-Araxes burials has yielded evidence for an accumulation of wealth comparable with that seen in the burial at Arslantepe or in those of the Maikop culture-historical community of the northwestern Caucasus.
The available evidence does unequivocally show that all areas of Transcaucasia (excepting the distinct region of western Georgia bordering the Black Sea) were occupied during the initial Early Bronze period in the second half of the 4th millennium. Kura-Araxes settlements, now numbering in the hundreds (Kushnareva 1997: 44), are found throughout the region, even at very high altitudes, suggesting possibly seasonal occupations and some form of transhumance, and their association with terraced agriculture in some mountainous areas seems well-established. These ‘peoples of the hills’ knew how to adapt to different altitudinal zones, settling in high mountain valleys, on broad volcanic uplands or on lower-lying fertile plains.
Given their occupation of these different altitudinal zones, it is not surprising that the materials used in the construction of their houses varies from stone and wattle-and-daub with wooden post structures in the intermontane valleys and higher plateaus to circular and sub-rectangular mudbrick structures sometimes with stone foundations in the lower plains. We know that they herded sheep and goats and, to a lesser extent, cattle and it is hypothesized that some flocks may have been driven to higher pastures during the summer by transhumant pastoralists as occurs today on the passes into and on the plateaus of Djavakheti from the Adzhari and Imereti regions.
The incredible profusion of small Kura-Araxes settlements throughout Transcaucasia and northeastern Anatolia may reflect both population increases over time and the periodic settlement of new areas suggestive of a form of extensive shifting cultivation, an interpretation consistent with the apparent sudden abandonment of several Kura-Araxes settlements. Kura-Araxes houses, such as those uncovered at Karnut on the Shirak Plain of northwestern Armenia, often contain large complete artefacts, such as storage jars and the characteristic, distinctly modeled andirons or figured portable hearth supports. It appears almost as if the people had planned to return to the settlements that they had mysteriously and suddenly left.
Whether it was the search for more arable land to support their burgeoning populations and/or their displacement with the arrival of new groups from the north with four-wheeled, ox-driven wagons, the Kura-Araxes peoples moved over some extended period beginning towards the end of the 4th millennium far to the southwest across the Anatolian Plateau to the Amuq Plain and beyond to northern Israel and to the southeast into northern Iran and along the Zagros Mountains and into eastern Gilan Province south of the Caspian Sea and onto the Iranian Plateau as least as far as Qazvin.
This spread of ‘Early Transcaucasian’ settlements has long fascinated archaeologists, many of them speculating on the ethnic/linguistic identity of these migrants and interpreting them as ancestral to Hurrians or Hittites or other later historically attested peoples. The question of the spread or migration of Kura-Araxes related peoples remains inadequately investigated.
Calibrated radiocarbon dates are beginning to yield a consistent picture for the timing of this dispersal. The relevant VIB period (though Kura-Araxes ceramics first appear earlier in Period VII) at the extensively excavated site of Arslantepe near Malatya dates ca. 2900–2700 BCE. This date essentially coincides with the appearance of Kura-Araxes materials from neighbouring sites such as Norsuntepe.
Related Khirbet-Kerak materials from northern Israel have been dated roughly from 2700–2450 BC (Miroschedji 2000: 258), suggesting an initial dispersal into the Upper Euphrates basin at the very beginning of the 3rd millennium (and after the collapse of the Uruk expansion), followed by a subsequent movement to the southwest in the second quarter of the 3rd millennium.
The overall pattern seems reasonably clear: an initial spread across eastern Anatolia to the Upper Euphrates basin at the very end of the 4th and beginning of the 3rd millennium followed by a relatively rapid diffusion (during the course of a century or so?) farther southwest, ultimately to the eastern Mediterranean coast.
Sites in the Urmia basin of northwestern Iran with relevant materials (e.g., Geoy Tepe and Yanik Tepe) seem to have been occupied already in the last centuries of the 4th millennium. ‘Early Transcaucasian’ materials appear to be intrusive in this region as well; i.e., they represent a break with earlier Chalcolithic remains on these sites, but this movement appears to predate the spread into the Upper Euphrates area.
One can only speculate that the lack of an Uruk presence in northwestern Iran may have facilitated this earlier movement from the southern Caucasus to the southeast. Their spread farther south into central-western Iran occurred later, though precisely how much later is still unclear.
These ‘peoples of the hills’ seem to have consciously avoided certain regions, including large settled areas on the northern Mesopotamian Plain. Less than a handful of Kura-Araxes sherds, for example, have been found at Tell Brak (J. Oates, personal communication).
Movements across the Anatolian Plateau and into northern Mesopotamia and regions farther west were undoubtedly very complex and involved more than just these dispersals from Transcaucasia. Other groups may have crossed the Caucasus from the northwest and then intermingled with both the local peoples and the Transcaucasians with whom they came into contact. A chain reaction was set in motion with incoming groups successively displacing one another.
There also remained relatively empty places that the southern Caucasians could easily settle. They possibly destroyed or overran some settlements, while they avoided or left others alone, presumably because the polities that occupied them were more powerful.
While our knowledge of the distribution of the sites containing Kura-Araxes materials is obviously dependent upon the nature and extent of the surveys conducted throughout these different regions, which manifestly are not commensurate with one another, it also seems clear that not all contiguous zones were equally affected by these dispersals.
The spread was not continuous and there are clear gaps in the distribution of sites containing these materials, such as the dense concentration of Early Transcaucasian sites in the Malatya region of eastern Anatolia or the gap in known sites with Early Transcaucasian/Khirbet Kerak ceramics in Syria and Lebanon between the Amuq Plain and northern Israel, a break possibly to be explained by coastal rather than overland contacts and movements of peoples.
Despite the uneven coverage, these gaps to some extent must reflect the historical reality that the newcomers from the north only occupied certain selected regions. It is also obvious that for the most part these dispersals do not represent armed military invasions and that the movements involved considerable assimilation with pre-existing local traditions, exacerbating the archaeologists’ task of recognizing them. Populations expanded and intermingled with one another. In these processes, social structures obviously must have changed.
It is an archaeological truism today to note that pottery styles do not equate with peoples, and the temptation to do so must be resisted. Nevertheless, the very frequency of distinctive, seemingly intrusive ceramics and other items of material culture, such as the highly specific figured andirons, suggest that this phenomenon, however short-lived, must have been reasonably substantial.
It is unclear what was driving these dispersals. Possibly, the peoples involved were in search of new sources of metal in Jordan or, more convincingly, in Cyprus (cf. the recently excavated, Kura-Araxes-like hearth stands and evidence for migrants from southwestern Anatolia at the Early Bronze Age site of Marki Alonia.
The settlers from Transcaucasia were skilled metallurgists, but why leave a metalliferous region like the Caucasus for unknown sources? Moreover, Khirbet Kerak materials are not found in the metal-bearing Wadi Feinan area south of the Dead Sea. Perhaps they were simply in search of more and better arable land with natural population increases, replicating on a much larger scale the movements from the highlands to the plains that may have characterized the initial spread of Kura-Araxes settlements within Transcaucasia?
Another factor may also have been at work: people were not only moving south out of the Caucasus, but also may have been moving into Transcaucasia from the north—at least at some point during the first half or towards the middle of the 3rd millennium (see the new calibrated dates for the ‘early kurgan cultures’ of Transcaucasia). It is hard to distinguish cause from effect here: did peoples move into the rich Alazani and Kura Valleys because others had moved out or were the Kura-Araxes peoples moving south due to the incursions of more mobile peoples from farther north?
Traditionally, Early Bronze materials from the Caspian coastal plain have been interpreted as a relatively late and northernmost manifestation or regional ‘variant’ of the Kura-Araxes culture-historical community. This view needs to be totally reconceptualized.
Recent excavations at the Kura-Araxes related site of Velikent in southeastern Daghestan north of Derbent have documented that it was initially occupied ca. 3600–3500 BCE or just subsequent to the initial appearance of the early Maikop culture in the northwestern Caucasus.
The Daghestan ‘variant’ of the Kura-Araxes cultural tradition contains specific distinctive features related, on the one hand, to local Chalcolithic developments in mountainous Daghestan immediately to the west and, on the other, to Maikop remains to the northwest and to materials found even farther north on the western Eurasian steppes, particularly in terms of metals and polished stone weapons, such as shaft-hole axe/hammers (battle-axes) and perforated mace heads.
The architecture on Early Bronze Caspian coastal plain sites varies from circular mudbrick freestanding architecture (e.g., at Serker-tepe in northeastern Azerbaijan) to deeply dug oval and circular pit-houses and even sunken, multi-roomed public structures.
There is practically no earlier evidence for Chalcolithic sites on the Caspian littoral plain, as there is in the mountains, suggesting that the coastal plain was first settled during the middle of the 4th millennium BCE (Fig. 5) or slightly later than early Maikop sites to the northwest and roughly contemporaneous with the initial appearance of Kura-Araxes sites in Transcaucasia to the south.
The different components or regional variants of the Kura-Araxes culture-historical community seem to emerge simultaneously at a time that roughly coincides with the so-called Uruk expansion up the Euphrates and onto the eastern Anatolian Plateau.
The Early Bronze Age site of Velikent was occupied from the mid-4th to the early 2nd millennium BCE (or ca. 3600–1900 BCE as based on a series of calibrated radiocarbon determinations). Its cultural remains, which consist of separate burial and settlement areas set on the top of five natural clay terraces formed by a transgression of the Caspian Sea, extend discontinuously over more than 30 ha. Some of the features connect the steppe world to the north with the sown agricultural world to the south.
Excavations at the type-site of Velikent have been the most extensive and have yielded the most materials, particularly from its collective catacomb burials where hundreds of metal and polished stone objects and complete ceramic vessels have been recovered. Large circular dwellings with internal features such as hearths and benches and made of dried mudbricks, some of which were occasionally fired, characterized the earliest building horizon.
Subsequently, the architectural tradition changed, and deeply dug pit-houses became the norm. An even later multi-roomed building, which had been extensively burned, was excavated above a series of these deep circular pit-houses, though the rooms of this building, which were reinforced with wooden posts, flat river boulders and even columns of stones set on top of one another reinforcing the corners, were dug down into the natural clay terrace and not built-up as in the first building horizon. This multi-roomed burned building was not a domestic structure but served some public function, possibly associated with ceramic production and storage.
Thus, there was a very significant shift in building traditions not long after Velikent had been initially settled. The earliest horizon has numerous parallels with Kura-Araxes materials from sites to the south, while the later levels, which are deeply dug down from the surface, may reflect more northern influences as well as represent a unique local adaptation to the dense clay terraces into which they were dug and into which they also dug their collective catacomb-shaped burials. The forms of the tombs with their attached entrance pits closely resemble or even consciously emulate the deeply sunken circular pit-house dwellings.
Most significantly, the initial settlers at the site arrived with metal working skills, since arsenical copper/bronzes and ceramic molds for casting objects appear in the earliest levels. They also initially produced very fine, highly fired ceramics with impressed designs that may have been finished on a slow wheel.
These ‘high-quality’ wares, which constitute ca. 10% of the total ceramic assemblage in the early levels, are also found on other sites to the south on the coastal plain of northeastern Azerbaijan and to the west in Chechnya, though their quality of manufacture suggests possibly a connection with northern Mesopotamia (perhaps related to the earlier incursion of settlers from the south?).
These ‘high-quality’ wares were not found in the multi-roomed building or in the latest excavated pit houses on the northern settlement mound at Velikent; i.e., they disappear at some point during the later 3rd millennium occupational sequence at the site.
Several collective catacomb-shaped tombs have been excavated at Velikent that have yielded a wealth of material remains, such as hundreds of complete ceramic vessels, arsenical copper/bronze (and tin-bronze) ornaments, and tools and weapons, and the polished and perforated stone mace heads and hammer axes (or battle-axes).
Radiocarbon dates and typological parallels with other materials suggest that the tombs thus far uncovered are somewhat later in date, possibly beginning in the second quarter of the 3rd millennium BCE. A few highly burnished, occasionally incised vessels and fragments have been recovered from these collective tombs that closely resemble so-called Bedeni vessels found in the large early pre-Trialeti kurgans in the Kakheti and Kvemo Kartli regions of eastern Georgia.
Wheeled wagons, driven by oxen, first are found in Transcaucasia in these Bedeni kurgans, though, as previously mentioned, a few Maikop-related kurgans with wheeled wagons have been found north of the Caucasus and probably date somewhat earlier to late 4th millennium times.
Other parallels can be drawn between materials from Velikent-related sites on the Caspian coastal plain and those from the early monumental Transcaucasian kurgans to the south and to those from kurgans found in the steppes immediately north of the Great Caucasus Range. Thus, a stone battle-axe of a type common in the northern Caucasus and southern Russia was found in one of the Martkopi kurgans, as were gold spiral pendants presumably worn about the temples or in the hair, pear-shaped ground marble mace-heads, and perforated animal-toothed pendants. Very similar materials are found in the collective catacomb tombs at Velikent, and the stone battle-axes are also very typical of the Bronze Age Novotitorovskaya culture (3300–2700 BC) remains in the Kuban region of the northwestern Caucasus.
Novotitorovka culture was immediately to the north of and largely overlapping portions of the Maykop culture facing the Sea of Azov, running from the Kerch Strait eastwards, almost to the Caspian, roughly coterminous with the modern Krasnodar Krai region of Russia. It is distinguished by its burials, particularly by the presence of wagons in them and its own distinct pottery, as well as a richer collection of metal objects than those found in adjacent cultures, as is to be expected considering its relationship to the Maykop culture. It is grouped with the larger Indo-European Yamna culture complex, and in common with it, the economy was semi-nomadic pastoralism mixed with some agriculture.
Most significant, of course, is the parallel appearance of oxen-driven wooden wagons in Novotitorovskaya kurgans and, more generally, in kurgans on the western Eurasian steppes from Novosvobodnaya, Pit-Grave (Yamna culture) of the Southern Bug/Dniester/Ural region (the Pontic steppe), Early Catacomb occupying essentially what is present-day Ukraine, Kemi-Oba at the northwest face of the Sea of Azov, the lower Bug and Dnieper Rivers and the Crimea, and other related culture sites north of the Great Caucasus Range and in the large monumental kurgans of the Late Early Bedeni and Middle Bronze Trialeti-related cultures in the southern Caucasus.
The remains of more than 250 wagons have been excavated in kurgans from the Kuban area of the northwestern Caucasus and across the southeastern European steppes, 115 of them to be attributed to the Novotitorovskaya culture of the former region.
The parallel appearance of similarly constructed wagons on both sides of the Caucasus cannot be coincidental but must be historically related, possibly representing the continuous movement of cattle herders north to south along the Caspian corridor or coastal plain to circumvent the Great Caucasus Range. The earliest remains in Eastern Europe of a wheeled cart were found in the “Storozhova mohyla” kurgan (Dnipropetrovsk, Ukraine) associated with the Yamna culture.
One cannot also fail to observe the nearly simultaneous appearance of such oxen-driven wagons in the ‘royal’ tombs of southern Mesopotamia (e.g., at Ur), tombs that contain striking parallels in precious jewelry and bronze weapons with remains from the southern Caucasus.
The societies responsible for the construction of the large late Early and Middle Bronze kurgans in Transcaucasia were not egalitarian but must have been ruled by a paramount leader or chief who was capable of waging war and amassing labour on a significant scale to raise these monumental mortuary mounds.
The number of known settlements decreased dramatically from the earlier time of the Kura-Araxes culture-historical community, and the later Middle Bronze settlements that have been excavated, such as at Uzerlik-Tepe, were heavily fortified, safely encircled behind massive stone walls, again reflecting unsettled, perpetually bellicose conditions.
It is thought, though not yet conclusively demonstrated, that the earliest fortresses with cyclopean stone architecture, which typically are located in steep or relatively inaccessible locations, such as the citadel of Schaori on top of a steep peak overlooking the western shore of Lake Paravani in Djavakheti, may first date to the Middle Bronze Age. Later, during the Late Bronze/Early Iron Age, such citadels become a characteristic, if not dominant, feature of the Transcaucasian landscape.
The emergence and subsequent dispersal of the Kura-Araxes/ETC ‘culture-historical community’ seems to be situated in a broader chronological and geographical context that reveals historical connections and developments over a broad area stretching from the southern Russian steppes to the southern Levant.
It seems that their sudden emergence and more gradual disappearance in the northeastern and southern Caucasus in terms of larger developments is associated with movements of peoples south to north (e.g., the Uruk expansion) and north to south (e.g., cattle herders with oxen-driven wheeled wagons, traversing the Caspian coastal plain and burying their leaders in monumental kurgans with rich mortuary remains). Current evidence allows us to sketch such processes only cursorily, not paint them in precise detail. The archaeological record is cumulative, and new discoveries will both complicate and clarify our understanding of this diverse, archaeologically defined ‘phenomenon’.
New evidence from Iran seems to support the further dispersal of Kura-Araxes folks east across the central Iranian Plateau and into the thickly wooded province of Gilan south of the Caspian Sea. Kura-Araxes/ETC sherds have been collected from the sites of Diarjan, Doranabad and Qoli Davis.
Some of these collected sherds, particularly those incised and even occasionally inlaid, seem to resemble most closely those from the Godin IV settlement, and perhaps ultimately they will help define an eastern or distinctively Iranian regional variant of the Kura-Araxes culture-historical community.
These materials are not far removed spatially from those recovered from another area with its own distinct grey ware ceramic tradition, incorporating Damghan (Tepe Hissar), the Gorgan Plain (Shah Tepe, Tureng Tepe and countless other sites), and the Sumbar Valley of southwestern Turkmenistan (the Parkhai Cemeteries).
Materials, that were nestled in the Kopet Dagh Mountains of the Upper Sumbar Valley, the Sumbar River being a right bank tributary of the Atrek that runs essentially parallel to the Gorgan River, both of which flow into the southeastern corner of the Caspian, particularly ceramics, from the Late Chalcolithic/Early Bronze Parkhai II Cemetery of southwestern Turkmenistan clearly relate both to materials excavated on sites, such as Ak-depe and Kara-depe, in the Kopet Dagh piedmont farther to the northeast, and to sites on the Gorgan Plain, such as Shah Tepe, and on the north-central Iranian Plateau, such as Tepe Hissar, which are located southwest of the Sumbar Valley.
The most distinctive and striking ceramics from the Parkhai II Cemetery consist of highly burnished, sharply carinated biconical grey ware bowls that can be paralleled on these sites located farther northeast and southwest and clearly relate to the development of the northeastern Iranian grey ware tradition.
More distant parallels can also be detected with materials from Transcaucasia and the western Caspian littoral plain: the collective tombs from the Parkhai II Cemetery resemble the catacomb tombs from Velikent in that they represent collective tombs that were periodically opened from a separate entry chamber to the side of the principal vault and were reused with earlier burials pushed to the side of the chamber for the most recently interred individual.
Numerous double spiral-headed toggle pins were found at Parkhai II and these have a broad spatial and possibly temporal distribution, which also includes Transcaucasia (though they are not found at Velikent). The burnished carinated bowls remotely resemble the metallic-like Bedeni vessels that are rarely encountered at Velikent but that are characteristic of the Bedeni early monumental kurgans from the southern Caucasus.
It seems that there were a mid-3rd millennium movement from the Caucasus across northern Iran into southwestern Turkmenistan. The parallels seem to be too specific and numerous to be totally fortuitous. The Kura-Araxes ‘culture-historical community’ or, more ambiguously, ‘phenomenon’, clearly emerged and developed in a much larger inter-connected world.
Cradle of Civilization 
Leave a comment