382 |
From Fortresses to Fragmentation |
materials and ideas between communities. Of the other three categories, the Ghodshali-Kedabeg horizon is so similar to Lchashen-Tsitelgori that it does not warrant being treated separately, while both the Samtavro and KaiakentChorochoi horizons are too inadequately articulated at present to provide an informed opinion at this stage.
Several factors have contributed to this bewildering situation, most of which we have dealt with before – the promotion of local sequences, the notion of ethnogenesis, self-referential perspectives adopted within modern geo-political borders, and an almost obsessive desire for the construction of elaborate relative chronologies without supporting radiocarbon readings. Complicating this situation is the loose usage of the term ‘culture’. In some instances, a mere handful of new material aspects – a different ornamentation on a clay pot, or a new metal item – have been enough to designate a new cultural community. In any case, I refer to some of the attributes of Samtavro, Ghodshali-Kedabeg, and Kaiakent-Chorochoi horizons in the following pages, which nonetheless will focus on the Lchashen-Tsitelgori and Talish traditions. The Koban and Colchian horizons will be dealt with in subsequent chapters.
THE LCHASHEN-TSITELGORI HORIZON (1500–800BC)
This cultural horizon covers an area from the Araxes valley in Armenia through the middle Kura valley around Tbilisi, to the southern slopes of the central Caucasus.There is a concentration in the Iori and Alazani regions of Kakheti, whereas the Surami Ridge and Mingechaur, in western Azerbaijan, define its western and eastern boundaries respectively.The emergence in the mid-second millennium of conspicuous fortresses defined by formidable stone walls is one of its most defining characteristics. As Adam Smith aptly points out, ‘the social inequalities visible in the kurgans of the early second millennium appear to have been formalised into a tightly integrated socio-political apparatus where the critical controls over resources – economic, social, sacred – were concentrated within the stone masonry walls of powerful new centres’.13 With these fortresses, which appear to have lower townships in some places such as Azerbaijan, came new types of cemeteries, often neatly and deliberately clustered.14 These novel paradigms of authority entrenched the notion of territoriality.
Fortresses
Hill forts are particularly characteristic of the rugged Armenian landscape, where they are well documented, and the nearby Dzhavakheti region. Many
13Smith et al. 2009: 30.
14Hammer 2014.
The Lchashen-Tsitelgori Horizon (1500–800 BC) |
383 |
fortresses have also been reported in the adjacent areas of southern Georgia, eastern Anatolia and north-western Iran, but none have been as thoroughly studied as those in Armenia.15 Built with roughly hewn stone blocks, some quite massive, these hilltop sites are located in strategic positions, typically perched on peaks that overlook plains and passes. As a further defensive measure, fortresses were often within sight of one another, which also enabled communication across a region. Geophysical prospection at Tsaghkahovit and Gegharot in Armenia, have also detected different circulation patterns around the fortresses, implying varying social networks.16 Exactly when these novel expressions of political power emerged remains unclear, but sometime around 1500 BC is a reasonable estimate.
The Armenian-American Project for the Archaeology and Geography of Ancient Transcaucasian Societies (Project ArAGATS) has conducted the most detailed and multi-scalar study of these fortresses.17 An intensive survey of the Tsaghkahovit Plain, on the northern fringe of Mount Aragats in central Armenia, revealed no less than ten Late Bronze Age (1500–1100 BC) fortresses and 199 cemeteries of similar age, accommodating some 5,970 tombs. This abrupt display of political centralisation, which appeared early in the Tsaghkahovit Plain, stands in stark contrast to the Middle Bronze Age, when mobile communities left behind shallow and transient settlements, and no more than fourteen burial grounds. One of the clearest indicators of this shift comes from the two main chambers of Barrow 1, situated below the fortress at Gegharot.18 One chamber (western) was transitional in character: a jumble of animal bones, no human remains, and ceramics that had clear Middle Bronze to Late Bronze features.The other compartment (central) revealed new sensitivities.The head and forelimbs of a horse were ceremoniously placed beside two skeletons, an adult male and an infant. This custom of placing animal extremities in tombs, plus the accompanying grave goods, clearly fixes the chamber to the Late Bronze Age.
The fortresses of Tsaghkahovit and Gegharot, some 8 km apart, are the two largest in the survey area. Tsaghkahovit, overlooking the plain 80 m below, encircles an outcrop over about 7.5 ha, and is crowned with a fl at citadel some 0.6 ha in area (Figure 8.2(4)). If we add the terraces, the residential semi-subterranean buildings at the foot of the citadel, and the surrounding terrain, the site reaches across 39.6 ha.19 Typically of the fortresses, the perimeter fortification wall is constructed with stones of varying size, some massive,
15 See Özfırat (2009a) for the latest of a series of reports on the easternmost provinces in Anatolia. Ceylan (2008) conducted a number of more wide-ranging surveys; Biscione 2009.
16Lindsay et al. 2014.
17Smith et al. 2004; 2009; Badalyan et al. 2008; 2014.
18Badalyan et al. 2008: 59–60.
19Smith 2015: 160.
Figure 8.2. Fortresses: (1–3) Abuli; (4) Tsaghkahovit; (5) Shaori (after Narimanishvili et al.2010b;2 courtesy G. Bedianashvili;3 courtesyArchaeological Survey in SamtskheJavakheti; 4 courtesy Project ArAGATS).
384
The Lchashen-Tsitelgori Horizon (1500–800 BC) |
385 |
placed on bedrock and reinforced with buttresses. The wall comprises two stone leaves filled with a rubble core. Both the exterior and interior faces of the stones were roughly smoothed for aesthetic purposes.Within the fortress there were no enclosure walls, but its builders carved a series of fl at areas – terraces that were faced with a single row of stones. Buildings are spread across the fortress hill and appear to cluster into three primary complexes.
Gegharot is smaller (3.43 ha) and quite eroded around the citadel,though the fortification wall is preserved to several courses in parts. Like Tsaghkahovit it is amongst the earliest settlements. Constructed just before 1500 BC, Gegharot was destroyed and re-built a century later, and provided with paved courtyards, storage areas, workrooms, and even a series of shrines. Gaps and features along the wall suggest a gateway and possible buttresses. Another striking feature of this Late Bronze Age landscape is the well-circumscribed clusters of cemeteries spread across the Tsaghkahovit Plain, mostly of the cromlech type: a stone cist tomb encircled by a single row of stones.20 The sharp boundaries between the cemeteries have been interpreted as a reflection of the territoriality of competing regional leaders extending into the mortuary landscape.21
Fortresses, often in an excellent state of preservation, also dot the south Georgian landscape – in the Tsalka and Dzhavakheti regions, and on the other side of the river in Meskheti. Leo Meliset-Bek’s investigations in the 1930s initiated an interest that was pursued by Kuftin and has only recently been re-kindled.22 Beshtasheni is unusual amongst the group. It has no fortification wall, but deep ditches provided it with a defensive apparatus. In 1990, the Tsalka expedition continued where Kuftin left off, and carried out exploratory excavations at Sabechdavi fortress (150 x 90 m) fitted with two entrances. The site comprises both a hill fort and an adjacent settlement.The stonework of the fortification wall is massive, with blocks up to 4 m in length.23 Thirty houses were documented within the walls, fringing the perimeter of the hill fort (Area A). Of the ten houses that were excavated, each was a single-roomed affair, covering an area of between 12 and 15 m2. A circular hearth, about 30– 35 cm in diameter, manufactured from coarse gritty clay, was the focus of each house. The cultural debris was not rich – coarse, obsidian-tempered pottery; obsidian debitage; and well-used basaltic grinding stones – but the ceramics can be comfortably accommodated into the Late Bronze Age (see below), which in the absence of radiocarbon dates provide a relative time slot.
Near the eastern entrance at Sabechdavi is an alignment of menhirs, the meaning of which eludes us, heading east towards Mount Eli-Baba and then
20 Cromlechs should be distinguished from kurgans. While their edges are both defined by a circle of stones, cromlechs have no earthen mound and do not exceed 10 m in diameter.
21Smith 2015: 164.
22Melikset-Bek 1938.
23Narimanishvili et al. 2010; Narimanisvili and Šanšašvili 2010.