Printer Friendly

Highland fortress-polities and their settlement systems in the southern Caucasus.

[ILLUSTRATION OMITTED]

Introduction

In the Late Bronze Age (LBA, c. 1500-1150 BC) and Iron Age (LA, c. 1150 BC-300 AD), fortresses dotted hilltops and rock outcrops in the highlands of eastern Turkey, north-western Iran and the southern Caucasus (modern Armenia, Azerbaijan and Georgia; Figure 1). After a long period of mobility and pastoral economy that left few archaeologically visible settlements during the preceding Middle Bronze Age (MBA, c. 2400-1500 BC; Figure 2), LBA and LA fortresses marked the reappearance of a more settled mode of life, the transition to an agricultural economy and the emergence of the first complex polities in the region (Smith 2005: 264-68). Fortresses were frequently placed at the edges of arable plains, suggesting a concern with the surveillance of agricultural land. Spatial analyses of fortress distribution and hierarchy have connected these sites to changing political configurations of the LBA, Early Iron Age (EIA, c. 1150-800 BC) and Middle Iron Age (MIA, Urartian period, c. 800-600 BC) (Kleiss 1976; Zimansky 1985; Biscione et al. 2002; Badaljan & Avetisyan 2007; Smith et al. 2009).

[FIGURE 1 OMITTED]

Survey and excavation of fortresses has led to the definition of a 'Caucasian model of development' characterised by non-urban complex societies led by military aristocracies that were housed within their walls (Masson 1997: 127-32; Smith & Thompson 2004). This model describes Caucasian communities in opposition to those of the 'core' area of civilisation in the ancient Near East, where dense, urban occupation occurred on artificial settlement mounds (Biscione 2002: 360-65). However, archaeologists have constructed this model with only the most impressionistic ideas about the potential regional population. Little is known about the location, size and character of any domestic settlements associated with these fortresses or the identity and subsistence strategies of their inhabitants (but see Khatchadourian 2008; Lindsay et al. 2010; Lindsay 2011 for recent data on fortress settlements in Armenia). Many analyses explicitly ignore the presence and size of towns (Biscione 2009, 2012; Biscione & Dan 2011), ranking fortresses and inferring territorial boundaries solely on the basis of fortress perimeters. As most excavated examples do not contain domestic architecture, the fortresses themselves likely did not accommodate the populations that constructed and maintained them. Furthermore, their locations on hilltops would have been impractical for agriculturalists farming the valleys below.

[FIGURE 2 OMITTED]

Most previous research in the southern Caucasus (with important exceptions) has adopted a site-centric approach (Smith 2005; Lindsay & Smith 2006). This emphasis on fortresses rather than settlements and landscapes may have led to misleading conclusions about the size, spatial organisation and economic base of MBA, LB A and IA communities in the region. Landscape and settlement-focused archaeology has the potential to reshape the 'Caucasian model of development' by identifying stratified sites and territorial markers in the vicinity of major fortresses and investigating the spatial organisation of these settlements. Elsewhere in the Near East, landscape archaeology has identified small and unstratified sites; in the southern Caucasus even large, stratified settlements are not highly visible and require intensive survey methodologies for detection (Lindsay et al. 2010: 16-17).

Landscape-oriented research in the Serur Plain of Naxcivan, Azerbaijan, has revealed a fortress-settlement that is earlier, larger and more deeply stratified than the chronological and developmental frameworks discussed above would predict. Oglanqala, the dominant fortress on the plain, was part of a settlement complex consisting of two fortresses and a domestic settlement of the MBA and LA, all of which may have been surrounded by a wall enclosing at least 490ha. The size of the potentially enclosed area is particularly significant: in the southern Caucasus, pre-medieval settlements rarely encompass more than 10ha, including their fortification walls (Figure 3). The position of the complex at the entrance to a river pass may have facilitated control of highland pastoral and lowland agricultural resources. These survey findings demonstrate that a large, potentially urban settlement emerged in Naxcivan during the 'nomadic' MBA, and indicate that when the IA Urartian state (ninth-seventh centuries BC) expanded eastward it did so into an area with pre-existing complex political and settlement traditions.

[FIGURE 3 OMITTED]

Methods

Recent agricultural land-use, ancient settlement characteristics and topography create difficulties for landscape archaeology in the southern Caucasus. Settlements associated with fortresses are most likely to lie in the valleys or plains below them, but these areas also have been transformed topographically and hydrologically in recent decades by Soviet land amelioration programmes (Smith & Greene 2009). Because ancient settlement did not typically result in artificial mounds as it did elsewhere in the Near East, archaeological sites in river valleys are often not highly visible in satellite imagery or on the ground. Important sites and features related to fortress-settlement systems may be located outside the river plains, but systematic survey of these areas is frequently made difficult by extreme topography, and identification of stone architecture may be complicated by linear natural rock outcrops.

These difficulties make targeted, systematic, pedestrian survey essential for identifying and investigating fortress-settlement systems. In July to August 2012, an archaeological survey was initiated to explore off-site features and less visible sites along the north-eastern edge of the [section]3rur Plain, surrounding the fortress of Oglanqala (Figure 4). The survey team employed a stratified methodology. Possible sites and features in all ecological zones (the river plain, steppe and highland zones) were first identified using DigitalGlobe QuickBird imagery and then visited. GPS receivers with mobile GIS were used to map features and record surface pottery concentrations. The spatial limits of significant ceramic scatters were documented by walking regularly spaced transects (15m spacing) and recording the location of individual artefacts. Within the cultivated plain, the survey team investigated areas with the most visible ground surfaces, such as canal banks and fallow fields. For each site, transect or collection area, all ceramics were collected for later comparison to excavated sequences from Oglanqala and other sites in the surrounding region.

Previous archaeological survey in Naxcivan had focused on locating the core areas of settlement mounds and fortress sites within and on the edge of alluvial plains (e.g. Bahsaliyev 1997; Belli & Sevin 1999; Parker et al. 2011; Ristvet et al. 2011); the present project represents the first systematic study of landscape-scale features and sites that were not highly visible. The overall goal was to examine the spatial relationship between Oglanqala and the surrounding communities who provided it with products and labour.

Oglanqala and its settlement system

Fortresses

Previous excavations at Oglanqala laid the foundation for intensive survey work on the Serur Plain. Oglanqala was a large, walled administrative centre initially constructed in the eighth century BC, although there is also evidence of earlier settlement. Many LA fortresses in eastern Turkey, Armenia and north-western Iran have been firmly connected through texts and material culture with Urartu, a kingdom centred in eastern Turkey during the ninth to seventh centuries BC, and one of the world's earliest highland empires (e.g. Zimansky 1985; Biscione et al. 2002; Koroglu & Konyar 2011; Kroll et al. 2012). However, it remains unknown if and when the Serur Plain was incorporated into this polity. Material culture differences between Urartian sites and Oglanqala open the possibility that the fortress could have been part of a local independent polity on the edge of the Urartian realm (Ristvet et al. 2012: 355-57). At 12ha, the Oglanqala citadel is the largest in a group of inter-visible IA fortresses on the Serur Plain and along the Arpacay Valley that may have guarded the territory of the independent polity or the frontiers of Urartu (Ristvet et al. 2012: 324-26). As at most MIA fortress sites in the region, there is no evidence for domestic structures dating to this period within the citadel (Ristvet et al. 2012: 330-35).

Earlier work in the region had also briefly investigated a much smaller 2.1 ha fort located approximately 2km north of Oglanqala (Ristvet et al 2011). A re-survey of this site, Qizqala 1, revealed a dense MBA ceramic scatter with some LA material and collapsed stone fortification walls of an unknown date. As Qizqala 1 has not yet been excavated, its designation as a MBA fort partially reused in the IA is at the moment based on ceramics scatter and its location on a steep, artificially terraced rock outcrop overlooking the Arpacay (Arpa River) with a direct sightline to Oglanqala.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

Long wall segments

The discovery of three long stone wall segments in surrounding uncultivated steppe and highland zones has transformed the understanding of territorial boundaries around Oglanqala (Figures 5 & 6). The alignment of these wall segments and their shared masonry style suggest that they were part of a fortification line, or perhaps an enclosure wall.

Wall segment A extends 850m through the uncultivated steppe from Qizqala 1 and terminates on a rocky precipice overlooking the plain. Concentrations of large, roughly worked stones were visible inside and along the side of a modern canal at the foot of this precipice, in an area that is otherwise free of them, suggesting that the wall may have continued south into the plain and towards Oglanqala. Wall segment B extends at least 530m into the steppe east of Oglanqala. A CORONA satellite image from 1966 suggests that wall segment B at that time extended another 440m south into the plain (Figure 7). Today the wall is only preserved outside the plain, where its foundation was protected from destruction by agricultural and settlement activity. Wall segment C extends over 370m along a rocky ridge north-east of Oglanqala. All three wall segments are c. 2m thick. Although much of the wall was levelled to the ground, portions of segments B and C are preserved to a height of c. 1.25m.

[FIGURE 6 OMITTED]

The date of the wall's construction is not yet known. The few sherd scatters near the wall segments were largely undiagnostic. All three wall segments are made of the local black limestone without the use of mortar. Two faces, constructed of flat stones up to 1.25m in length/height, enclose a rubble core of smaller, rounder stones 100--300mm in diameter (see Figure 6). The excavated MIA (800-600 BC) walls of Oglanqala (Ristvet et al. 2012: 329-31) also used this construction technique. Wall segment C has a foundation partially constructed upon bedrock, also similar to several walls of the Oglanqala citadel.

The position of the wall segments suggests that they were part of a single wall. It is possible that this wall was merely a fortification line across the entrance to the river gorge. However, its extension along the western and eastern edges of the plain and the indications that the wall extended into the plain make it more likely that the wall encircled Oglanqala and Qizqala 1, creating a fortified area of at minimum 490ha that enclosed the edge of the [section]3rur Plain, the Arpa9ay Gorge and surrounding highlands.

[FIGURE 7 OMITTED]

Although the wall is not very thick, its defensive purpose is suggested by the fact that it encloses the high peak (1090m asl, c. 230m above the plain) on the opposite side of the Arpagay from Qizqala 1. This peak is extremely steep and rocky, and the survey team found no surface indications of habitation on its summit or slopes. As it is significantly higher than Qizqala 1 (1003m asl) and Oglanqala (1000.5m asl, c. 130m above the plain), this peak would have been a liability to both fortresses if left outside the walled perimeter.

Lower town and domestic spaces

The long wall enclosed a domestic settlement located between Oglanqala and Qizqala 1. Ceramic scatters documented along the banks of a large irrigation canal south of Qizqala 1 (750m north-east of the base of Oglanqala) show that there was an extensive ancient settlement in an area that has been destroyed by modern activity. A trench dug for the construction of the canal and an adjacent area terraced for a dirt road show at least 10m of archaeological stratigraphy, including large stone wall foundations, smaller mudbrick walls, ceramic ovens, and multiple living floors (Figure 8). Ceramics dating to the MBA, IA and the medieval period were eroding out of the canal profile. Survey along the nearby, uncultivated edges of the valley also yielded dense scatters of ceramics of the same periods (Figure 5). The spatial distribution of the surface ceramic scatters outside the plain and the stratigraphy visible in the canal cut correspond to the position of wall segment A, extending from Qizqala 1. In the uncultivated steppe, the survey team documented a clear drop-off in ceramic density beyond the wall, suggesting that wall segment A delimited the settlement in this area.

[FIGURE 8 OMITTED]

Survey transects in the fields immediately north of Oglanqala mapped surface ceramic concentrations indicating the potential spatial extent of the settlement first identified in the canal cut (Figure 9). As there were very few fallow fields in July-August 2012, it is difficult to draw conclusions at this stage. However, the densest concentrations of sherds (including both MBA and IA ceramics) were found approximately 200m north of the base of Oglanqala along a recently placed irrigation pipe. This location agrees with a 1930s survey of the area around Oglanqala, which indicated the existence of an ancient settlement at the foot of the fortress, before the introduction of mechanised agriculture (Alekperov 1937).

In future seasons the survey team will continue to investigate the spatial extent of the settlement. It is already clear on the basis of preliminary investigations, however, that the entire area within the large wall did not have a uniform density of occupation. The long wall encloses several landforms whose slopes are too steep and rocky for settlement. Also, sample areas within the possible enclosure that were chosen for survey transects had highly variable surface ceramic densities. While the transects in fields between Oglanqala and Qizqala 1 revealed clusters of sherds, transects in fields and uncultivated areas east of the Arpacay recovered very few artefacts. It remains to be determined whether these areas are empty of ceramics due to taphonomic processes or because they were ancient agricultural fields or pasture areas. A non-uniform settlement density would be consistent with finds from Ayanis, Turkey, the only intensively surveyed Urartian domestic settlement (Stone 2012).

[FIGURE 9 OMITTED]

Even if only the area to the west of the Arpa9ay between the two fortresses was settled, it covers approximately 80ha. That is about the same size as the Urartian settlement at Ayanis (Stone & Zimansky 2003: 215) and more than twice the size of the less systematically surveyed Urartian lower town at Karmir Blur (c. 30ha) (Oganesian 1955: 9-35). Two large Urartian fortresses near Oglanqala, in Iran, have been surveyed, and the sizes of their associated lower towns have been estimated at only 3.5ha (Verahram) (Kleiss 1974: 87) and 8ha (Bastam) (Zimansky 1985: 35).

The Oglanqala-Qizqala 1 settlement complex: interpretation and comparisons

The fact that Oglanqala and Qizqala 1 are spatially associated with a long wall and are connected by a settlement indicates that they were two elements of a single fortification-settlement system for at least part of their histories. If the long wall does indeed enclose the fortresses and intervening settlement, the minimum enclosed area (490ha) and minimum settlement size (80ha) are enormous for a site in Naxcivan, where excavated and surveyed settlements generally cover less than 10ha. The largest known Urartian fortified settlement in the southern Caucasus, at Armavir, Armenia (Martirosyan 1974), encloses approximately 75ha. Intensive survey in the Tsaghkahovit Plain of western Armenia found that fortified citadels of the LBA and IA were no larger than 1.6ha, and the unfortified settlements below them did not exceed 40ha (Smith et al. 2009: 281-329). The one exception to this was Berdidosh, a fortress that had both LBA/EIA and Urartian/Achaemenid ceramics on its surface and had fortification walls encircling a hill summit of 210ha (Smith et al. 2009: 287). The discrepancy in size between the Oglanqala-Qizqala 1 settlement complex and other known sites in the southern Caucasus may be partly due to the fact that most areas have not been intensively surveyed and that artefact scatters, walls and other features have not been sought beyond the walls of other fortresses.

Large settlement areas incorporating fortresses have been documented by survey in Armenia at Horom (c. 400ha including cemeteries; Badaljan et al. 1992), Uits (c. 200ha including cemeteries; Zardaryan 2012) and Artashat (c. 100ha of fortified hills and an additional c. 500ha of settlement; Zardarian & Akopian 1995: 173-80). However, neither Horom nor Uits were surrounded by enclosure or fortification walls, and the various components of their settlement complexes date from the Early Bronze Age to the medieval period. The area occupied in any one period may have been smaller, and their large expanse may be partially due to horizontal stratigraphy. The extensive fortifications and city at Artashat belong to its LIA occupation, and the extent of the site in earlier periods remains unknown (Zardarian & Akopian 1995: 179). Oglanqala may be fundamentally different from Horom and Uits and more similar to Artashat in that its spatial extent shows the size of the fortified site and settlement during a single period.

Together, Horom, Uits and Oglanqala demonstrate the value of landscape archaeology and survey for understanding the scale and spatial organisation of the settlements associated with fortresses in the southern Caucasus. When Horom was originally studied in the early twentieth century, only the two fortified citadels, 4ha in total area, and an extensive but unmapped area of graves were known. Through pedestrian survey, collection of surface materials and topographic mapping, archaeologists in the 1990s demonstrated that the fortified citadels and unfortified settlement area covered 40-55ha. Further, they found that these remains and various clusters of graves in the surrounding area collectively occupied more than 400ha (Badaljan etal. 1992: 31-35), but there is no enclosing wall as at Oglanqala-Qizqala 1.

The possible dual fortress or 'bi-nuclear plan of the Oglanqala-Qizqala 1 settlement complex has few parallels in the region. In form, the Oglanqala complex could most closely resemble the Urartian fortress Armavir (Argistihinili), located around 120km away on the Ararat Plain in Armenia. Argistihinili consists of two separately fortified citadel areas on east and west hills with a settlement between them. Most significantly, there is fragmentary evidence of a wall enclosing c. 75ha surrounding the complex (Ghafadarian [Kafadarian] 1978). Other Urartian sites have two citadels but do not have evidence of an enclosure wall: Anzaf in Turkey (Burney & Lawson 1960) and Qal'eh Ismail Aga in Iran (Pecorella 1980: pl. 4) both consist of upper and lower fortresses separated by less than 500m. Horom similarly has two fortresses located 320m apart (Badaljan et al. 1992: 31-35, 41-45, 1993: 1-6, 21-22). Both citadels had ELA material, but only one has evidence of ELA occupation; any ELA occupation levels on the northern citadel were removed during Urartian-era construction. Other EIA sites in Armenia have multiple dispersed citadels that may not have been inhabited simultaneously (Smith & Kafadarian 1996: 29-31).

[FIGURE 10 OMITTED]

The position of Oglanqala and Qizqala 1 at a mountain pass and the presence of the long fortification/enclosure wall sealing off the only entrance to this pass probably indicate an effort to control a major route between the high-altitude Dareleyez Mountains and the low-lying Serur Plain (Figures 1, 10 & 11). North of Oglanqala, cliffs and steep slopes flank the Arpacay. The long wall would have thus been an effective way of ensuring that the fortresses would see all traffic entering the pass. The position of MIA fortresses at the edges of river plains in Iran, Turkey and Armenia has been interpreted as indicative of control of agricultural production (Zimansky 1990: 8-9). The placement of Oglanqala and Qizqala 1 and their associated enclosure wall could certainly be related to agriculture or to trade, but equally could be related to the seasonal use of highland and lowland resources. Mobile pastoralism has been an important land-use strategy in the southern Caucasus for millennia, and both texts and ethnographic analogy suggest that it was also important in Urartian times (Yakar 2011; Burney 2012). Oglanqala and Qizqala 1 may have been involved in the surveillance or taxation of merchants as well as mobile pastoralists who annually migrated between highland and lowland pastures. Further survey and excavation will evaluate this hypothesis by investigating the economic strategies of people living within settlements on the Serur Plain.

[FIGURE 11 OMITTED]

The position of the fortification/enclosure wall at the mouth of the Arpacay Gorge is particularly interesting in light of the fact that a concentration of LA. forts linked by a 2.8km fortification line has been found in the Mtnadzor-Martuni Valley of Armenia (Joj Kogh 1; Biscione et al. 2002: 178-79). The Joj Kogh 1 and Oglanqala-Qizqala 1 fortification walls could be indicative of efforts to seal the northern and southern ends of the c. 80km route between them via the Selim Pass, although more research is necessary to test this hypothesis (Figure 11).

Potential MBA urban origins in the southern Caucasus

The most significant questions arising from the identification of the Oglanqala-Qizqala 1 settlement complex may be those related to its MBA beginnings. Qizqala 1 and the settlement between the fortresses were founded in this period, centuries before the first (IA) fortress at Oglanqala. Greater chronological precision is impossible at the moment; future excavations in the stratified areas of the settlement will be essential to refine the ceramic chronology of this long period. Although the long wall most likely dates to the IA, excavation is necessary to determine whether segments of the long wall could date to the MBA occupation. Preliminary analyses of the distribution of pottery in the settlement between the fortresses suggest that the MBA occupation was spatially extensive, during a period where few settlements are known and most information comes from the excavation of kurgan burials. Two of the few recorded MBA settlements are also found nearby on the edges of the Araxes Valley: Kultepe II in Naxcivan (Ristvet et al. 2011) and Metsamor in Armenia (Khanzadian et al. 1973).

Archaeological research in Armenia and Georgia has led scholars to propose that, after a period of widespread nomadism in the MBA, complex political organisation in the region first appeared in the LBA (Smith 2005, 2012). The possibility that a fortress and large, fortified settlement already existed at Qizqala 1 in the MBA could indicate that politically complex, sedentary settlement began earlier in Naxcivan. This idea fits well with data from the c. 3.2ha fortified MBA site of Uzerlik Tepe in western Azerbaijan (Kushnareva 1997: 129-44). The landscape of stone fortresses known from the LBA/EIA and later in the southern Caucasus might have been first introduced in the MBA (Kohl 2007: 120).

Conclusions

Preliminary landscape work around Oglanqala demonstrates the need for more systematic survey in the southern Caucasus. Fortresses in this region are not isolated but instead may all be part of much larger complexes that included residential areas and empty enclosed space. The recent survey around Oglanqala indicates that the societies encountered by Urartu in the Naxcivan area were large, complex communities focused on controlling trade and pastoral transhumance routes. The MBA origin of the fortress settlement and Qizqala 1 suggests that complex, spatially extensive, sedentary settlement may have emerged in Naxcivan earlier than previously expected.

Forthcoming field campaigns will investigate the chronology and the path of the long wall as well as the spatial organisation of the fortress-settlement. More research is needed to determine how they fit with demographic and functional variables used to define urbanism cross-culturally (Cowgill 2004: 543). The Oglanqala-Qizqala 1 complex does have several characteristics commonly associated with urban sites: it contains a spatially extensive, clearly delimited and fortified settlement of significant chronological duration. On the other hand, the morphology probably indicates that the population density could not have approached that of extremely densely occupied BA and IA cities of neighbouring Mesopotamia. Recent research in many areas of the world has investigated ancient sites that are spatially large but in other ways (such as their low-density, dispersed form) challenge standard notions of an urban-rural dichotomy and suggest alternative definitions of urbanism (e.g. McIntosh 2005; Kohl 2007: 39-46; Smith 2010; Al Quntar et al. 2011; Fletcher 2012).

The identification and excavation of settlement complexes associated with fortresses such as Oglanqala-Qizqala 1 will provide a springboard for future research into the economic base and spatial organisation of BA and IA communities in the southern Caucasus. Such research carries broad significance for understanding the timing and characteristics of the earliest urban centres and polities in the southern Caucasus region, the relationship between local and regional (Urartian) forces in these developments, and the dynamics between sedentary agricultural and mobile pastoral groups.

Acknowledgements

The American Research Institute in the South Caucasus, the Davis Center for Russian and Eurasian Studies at Harvard University, the American School of Prehistoric Research, and the University of Pennsylvania Museum generously supported fieldwork in summer 2012. The field team consisted of the author, Lauren Ristvet, Voli Baxseliyev, Emin Mammadov, Lara Fabian, Stephanie Martin and Jennifer Swerida. The article benefited from suggestions made by Hilary Gopnik, Lauren Ristvet, Jason Ur and two anonymous reviewers. Original ASTER GDEM data is a product of the Ministry of Economy, Trade and Industry of Japan and the United States National Aeronautics and Space Administration (NASA). Original SRTM data is a product of NASA and was provided by the CGIAR.

References

AL QUNTAR, S., L. KHALIDI & J.A. UR. 2011. Proto-urbanism in the late 5th millennium BC: survey and excavations at Khirbat al-Fakhar (Hamoukar), northeast Syria. Paleorient 37 151-75.

ALEKPEROV, A. 1937. Krashenaia keramika Nakhichevanskogo kraia i Vanskoe tsarstvo. Sovetskaya Arkheologia IV: 249-63.

BADALJAN, R.S. & P. AVETISYAN. 2007. Bronze and Early Iron Age archaeological sites in Armenia, volume 1: Mt. Aragats and its surrounding region (British Archaeological Reports international series 1697). Oxford: Archeopress.

BADALJAN, R.S., C. EDENS, PL. KOHL & A.V. TONIKJAN. 1992. Archaeological investigations at Horom in the Shirak Plain of northwestern Armenia, 1990. Iran 30: 31-48.

BADALJAN, R.S., C. EDENS, R. GORNY, P.L. KOHL, D. STRONACH, A.V. TONIKJAN, S. HAMAYAKJAN, S. MANDRIKJAN & M. ZARDARJAN [ZARDARYAN]. 1993. Preliminary report on the 1992 excavations at Horom, Armenia. Iran 31: 1-24.

BAHSALIYEV [BAXSELIYEV], V. 1997. Nahcivan arkeolojisi [The archaeology of Naxcivan]. Istanbul: Kanaat Basimevi Arkeoloji ve Sanat.

BELLI, O. & V. SEVIN. 1999. Nahcivan da Arkeolojik Arastirmalar [Archaeological survey in Naxcivan]. Istanbul: Kanaat Basimevi Arkeoloji ve Sanat.

BISCIONE, R. 2002. The Iron Age settlement pattern: pre-Urartian and Urartian periods, in R. Biscione, S. Hmayakyan & N. Parmegiani (ed.) The north-eastern frontier: Urartians and non-Urartians in the Sevan Lake basin. Volume 1: the southern shores: 351-70. Rome: CNR Istituto di Studi Sulle Civilta dell'Egeo e del Vincino Oriente.

--2009. The distribution of pre- and protohistoric hillforts in Iran. Studi Micenei ed Egeo-Anatolici 51: 123-43.

--2012. Urartian fortifications in Iran: an attempt at a hierarchical classification, in S. Kroll, C. Gruber, U. Hellwag, M. Roaf &C P. Zimansky (ed.) Biainili-Urartu: the proceedings of the symposium held in Munich 12-14 October 2007: 77-88. Leuven: Peeters.

BISCIONE, R. & R. DAN. 2011. Dimensional and geographical distribution of the Urartian fortifications in the Republic of Armenia. Aramazd: Armenian Journal of Near Eastern Studies VI: 104-20.

BISCIONE, R., S. HMAYAKYAN & N. PARMEGIANI (ed.). 2002. The north-eastern frontier: Urartians and non-Urartians in the Sevan Lake basin. Volume 1: the southern shores. Rome: CNR Istituto di Studi Sulle Civilta dell'Egeo e del Vincino Oriente.

BURNEY, C. 2012. The economy of Urartu: probabilities and problems, in S. Kroll, C. Gruber, U. Hellwag, M. Roaf & P. Zimansky (ed.) Biainili-Urartu: the proceedings of the symposium held in Munich 12-14 October 2007: 53-60. Leuven: Peeters. BURNEY, C.A. &C G.R.J. LAWSON. 1960. Measured plans of Urartian fortresses. Anatolian Studies 10: 177-96. http://dx.doi.org/10.2307/3642434

COWGILL, G. 2004. Origins and development of urbanism: archaeological perspectives. Annual Review of Anthropology 33: 325-49. http://dx.doi.Org/10.l 146/ annurev.anthro.32.061002.093248

FLETCHER, R. 2012. Low-density, agrarian-based urbanism: scale, power and ecology, in M.E. Smith (ed.) The comparative archaeology of complex societies: 285-320. New York: Cambridge University Press. http://dx.doi.org/10.1017/ CBO9781139022712.013

GHAFADARIAN [KAFADARIAN], K. 1978. Architecture de la ville d'Arghistihinili, in G. Ieni & L.B. Zekiyan (ed.) Atti delprimo Simposio Internazionale di Arte Armena, Bergamo, 28-30 Giugno 1975: 197-212. Venezia: San Lazzaro.

KHANZADIAN, E.V., K.A. MKRTCHIAN & E.S. PARSAMIAN. 1973. Metsamor. Yerevan: Akademii Nauk Armjanskoj SSR.

KHATCHADOURIAN, L. 2008. The 2005-2006 excavations of the Iron III settlement at Tsaghkahovit. Aramazd: Armenian Journal of Near Eastern Studies 3: 91-100.

KLEISS, W. 1974. Planaufnahmen urartaischen Burgen und Neufunde urartaischer Anlagen in Iranisch-Azarbaidjan im Jahre 1973. Archaologische Mitteilungen aus Iran 7: 79-106.

--1976. Urartaische Platze im Iran (Stand der Forschung Herbst 1975). Archaologische Mitteilungen aus Iran 9: 19-43.

KOHL, P.L. 2007. The making of Bronze Age Eurasia. Cambridge: Cambridge University Press.

KOROGLU, K. & E. KONYAR (ed.). 2011. Urartu-Doguda Degijm [Urartu--Transformation in the East]. Istanbul: YapiKredi.

KROLL, S., C. GRUBER, U. HELLWAG, M. ROAF & P. ZIMANSKY (ed.). 2012. Biainili-Urartu: the proceedings of the symposium held in Munich 12-14 October 2007. Leuven: Peeters.

KuSHNAREVA, K.K. 1997. The southern Caucasus in prehistory. Philadelphia: University of Pennsylvania Press.

LINDSAY, I. 2011. Holding down the fort: dynamics of Late Bronze Age community formation and political landscape production in the southern Caucasus, in P.G. Johansen & A.M. Bauer (ed.) The archaeology of politics: the materiality of political practice and action in the past. 151-77. Newcastle-upon-Tyne: Cambridge Scholars.

LINDSAY, I. & A.T. SMITH. 2006. A history of archaeological practices in Armenia and the South Caucasus. Journal of Field Archaeology 31:165-84. http://dx.doi.org/10.1179/009346906791072016

LINDSAY, I., A.T. SMITH & R. BADALYAN. 2010. Magnetic survey in the investigation of sociopolitical change at a Late Bronze Age fortress settlement in northwestern Armenia. Archaeological Prospection 17: 15-27. http://dx.doi.org/ 10.1002/arp.369

MARTIROSYAN, A.A. 1974. Argistichinili. Yerevan: Akademii Nauk Armjanskoj SSR.

MASSON, V.M. 1997. 'Kavkazskiy put' k tsivilizatsii: voprosy sotsiokul'turnoy interpretatsii. Drevnie Obshchestva Kavkaza v Epokhu Paleometalla (Rannie Kompleksnnye Obshchestva I Voprosy KuTturnoy Transformatsii): 124-33. Saint Petersburg.

MCINTOSH, R.J. 2005. Ancient Middle Niger: urbanism and the self organizing landscape. Cambridge: Cambridge University Press.

OGANESIAN, K.L. 1955. Karmir Blur IV: Arkhitektura Teishebaini. Yerevan: Akademii Nauk Armjanskoj SSR.

PARKER, B., L. RISTVET, V. BAXSEOLIYEV, S. ASUROV & A. HEADMAN. 2011. In the shadow of Ararat: survey in the Araxes River region, Naxcivan, Azerbaijan. Anatolica 37: 187-205. http://dx.doi.org/10.2143/ANA.37.0.2073788

PECORELLA, P.E. 1980. L'Urartu sulle rive del lago di Urmia: una relazione preliminare su tre campagne nell' Azerbagian Iranico. Studi Micenei ed Egeo-Anatolici 22: 337-47.

RISTVET, L., V. BAXSELIYEV [BAXSSLIYEV] & S. ANUROV. 2011. Settlement and society in Naxcivan: 2006 excavations and survey of the Naxcivan Archaeological Project. Iranica Antiqua 46: 1-53. http://dx.doi.Org/10.2143/IA.46.0.2084412

RISTVET, L., H. GOPNIK, V. BAKHSHALIYEV [BAXSeLIYEV], H. LAU, S. ASHUROV [ASUROV] & R. BRYANT. 2012. On the edge of empire: 2008 and 2009 excavations at Oglanqala, Azerbaijan. American Journal of Archaeology 116: 321-62. http://dx.doi.org/10.3764/aja. 116.2.0321

SMITH, A.T. 2005. Prometheus unbound: southern Caucasia in prehistory. Journal of World Prehistory 19: 229-79. http://dx.doi.org/ 10.1007/s 10963-006-9005-9

--2012. The Caucasus and the Near East, in D.T. Potts (ed.) A companion to the archaeology of the ancient Near East, volume IT. 668-86. Chichester: Wiley-Blackwell.

SMITH, M.E. 2010. The archaeological study of neighborhoods and districts in ancient cities. Journal of Anthropological Archaeology 29: 137-54. http://dx.doi.Org/10.10l6/j.jaa.2010.01.001

SMITH, A.T. & A. GREENE. 2009. Remote sensing data and analysis, in A.T. Smith, R. Badalyan & R Avetisyan (ed.) The archaeology and geography of ancient Transcaucasian societies, volume 1: the foundations of research and regional survey in the Tsaghkahovit Plain, Armenia. Chicago (IL): Oriental Institute of the University of Chicago.

SMITH, A.T. & K. KAFADARIAN. 1996. New plans of Early Iron Age and Urartian fortresses in Armenia: a preliminary report on the Ancient Landscapes Project. Iran 34: 23-37.

SMITH, A.T. & T.T. THOMPSON. 2004. Urartu and the southern Caucasian political tradition, in A.G. Sagona (ed.) A view from the highlands: archaeological studies in honour of Charley Burney. 557-80. Herent: Peeters.

SMITH, A.T., R. BADALYAN [BADALJAN] & P. AVETISYAN. 2009. The archaeology and geography of ancient Transcaucasian societies, volume 1: the foundations of research and regional survey in the Tsaghkahovit Plain, Armenia. Chicago (IL): Oriental Institute of the University of Chicago.

STONE, E. 2012. Social differentiation within Urartian settlements, in S. Kroll, C. Gruber, U. Hellwag, M. Roaf & P. Zimansky (ed.) Biainili-Urartu: the proceedings of the symposium held in Munich 12-14 October 2007: 89-99. Leuven: Peeters.

STONE, E. & P. ZIMANSKY. 2003. The Urartian transformation in the outer town of Ayanis, in A.T. Smith & K.S. Rubinson (ed.) Archaeology in the borderlands: investigations in Caucasia and beyond: 213-28. Los Angeles (CA): Cotsen Institute of Archaeology.

YAKAR, J. 2011. The ethnoarchaeology of the socio-economic structure of east Anatolia in the Urartian period, in K. Koroglu & E. Konyar (ed.) Urartu-Doguda Degijm [Urartu--Transformation in the East]: 126-49. Istanbul: YapiKredi.

ZARDARIAN [ZARDARYAN], M.H. & H.P. AKOPIAN. 1995. Archaeological excavations of ancient monuments in Armenia 1985-1990. Ancient Civilizations from Scythia to Siberia 1: 169-95. http://dx.doi.org/10.1163/157005794X00057

ZARDARYAN, M.H. 2012. Archaeological investigations in Syunik Region Armenia, and the problems of preservation of historical and cultural heritage. Regional Co-operation for Cultural Heritage Development E-Journal 1. Available at: http://www.nxhd.icomos.org.ge/?l=E& m=4-4& JID=1& AID=11 (accessed 20 February 2014).

ZIMANSKY, P. 1985. Ecology and empire: the structure of the Urartian state. Chicago (IL): Oriental Institute of the University of Chicago.

--1990. Urartian geography and Sargons eighth campaign. Journal of Near Eastern Studies 49: 1-21.

Received: 13 June 2013; Accepted: 9 September 2013; Revised: 17 October 2013

Emily Hammer *

* The Oriental Institute, University of Chicago, 1155 East 58th Street, Chicago, 1L 60637, USA (Email: emily.hammer@gmail.com)
COPYRIGHT 2014 Cambridge University Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Research
Author:Hammer, Emily
Publication:Antiquity
Article Type:Report
Geographic Code:9AZER
Date:Sep 1, 2014
Words:5664
Previous Article:Neolithic foundations in the Karama valley, West Sulawesi, Indonesia.
Next Article:The socioeconomic status of Iron Age metalworkers: animal economy in the 'Slaves' Hill', Timna, Israel.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters