Rust Family Foundation: Archaeology Grants Program

Forging Communities: the Metal Production Complex at Mtsvane Gora, Georgia


Principal Investigator: Dr. Nathaniel Erb-Satullo
Lecturer in Archaeomaterials, School of Archaeology, University of Oxford

Importance of the site and project

Between 1500 and 500 BC, a series of fortified hilltop sites dominated the settlement
landscape of the South Caucasus. Investigations of these sites suggest that a range of residential, craft production, and ritual activities took place there. However, our understanding of the social
dynamics of these fortresses remains incomplete.

Fig.1: Location of Mtsvane Gora in southern Georgia .

The primary goals of 2017 fieldwork were to recover additional samples of metal production debris and to clarify further the context and dating of iron production at the site. Scientific analysis of iron production debris will identify the stages of production taking place at the site, and illuminate the technologies of metal production. Because iron production debris dating to before 500 BC is rare in the ancient Near East, especially outside the Levantine corridor, examination of iron production practices and their contexts helps to explain why iron use spread across the region.

Previous Research in the region
Between 2013 and 2016, the Archaeological Research in Kvemo Kartli (ARKK) Project carried out an archaeological survey of hilltop sites in a transitional zone between the Kura River Basin and the Lesser Caucasus highlands. This foothill zone is characterized by a long history of metallurgy, and Project ARKK seeks to
investigate the social and technological transformations at a time when iron was first introduced to the region. 

One of the sites surveyed, called Mtsvane Gora, contained traces of metal production and was selected for further investigation. While metal production debris is occasionally mentioned
in excavation reports of Late Bronze Age fortresses, Project ARKK is the first to undertake comprehensive analysis of iron production debris at a Late Bronze-Early Iron Age fortress. Preliminary ceramic analysis and radiocarbon dates demonstrate that the site dates to before 500 BC, meaning that the iron production remains are some of the earliest known in the region.

Fig.2: Aerial drone photo of Mtsvane Gora, showing its position within the landscape.

2017 Funded Research  Project (RFF-2017-28)


The 2017 excavations at Mtsvane Gora were designed to explore the technology and organization of iron production within this fortified complex. Initial excavations in 2015 had uncovered clear iron production debris and yielded radiocarbon dates from several secure contexts. However, a number of aspects remained unclear.

Fig.3: Site plan of Mtsvane Gora, showing wall and excavation trench locations.

The specific goals of the 2017 season were:
- To identify and recover more slags and other metal production debris at the site, in order to improve our understanding of the technologies of metal production during this period.
- To improve our understanding of the chronology of the site and the metal production within it through greater exposure of secure contexts.
- To understand the spatial organization and chronology of metal production at the site.
- To test whether the undercutting of rocky outcrops at the base of the hill is in fact
evidence of mining activity.


Excavations continued in the areas of the site investigated in 2015, and three additional 5 5 excavations squares were opened in directly adjacent areas.

The team consisted of seven archaeologists (including both Georgian and American students) and roughly the same
 number of local workers, though numbers occasionally fluctuated with the need for more or fewer workers. Fieldwork lasted for six weeks, with in-country time before and after spend organizing logistics, transporting and storing finds, etc.

Fig.4: Aerial drone view of Trenches 1,4,5, and 6.

Trench 1
As anticipated, continued excavation in Trench 1 uncovered a clear floor level, with a rich assemblage of flat-lying sherds, many of which joined to others nearby. The floor surface itself was made of packed clay, better preserved in some areas, but very clearly extending over much of the trench and running up to the large fortification wall in the southeastern corner (fig.5). A small wall, two possible post-holes, and a circular stone post-base suggests the presence of some kind of structure built within the large fortification wall, a conclusion s
upported by the slightly different character of the floor assemblage on either side of the small wall. 

Fig.5: Excavation in Trench 1, where team member Karam Mirzoev sweeps the beaten clay floor.

In the northwestern corner of the trench, the packed clay floor surface disappeared and the floor assemblage continued directly on bedrock. The rather modest construction of the internal wall, coupled with the steepness of the slope, meant that the walls of the structure were not well preserved.

Trench 4
Adjacent trenches to the west, southwest, and south provided an clearer picture of the nature of deposits within th
e large encircling wall, and clarified structure of encircling wall itself. In Trench 4, a series of installations, post-holes, and ashy deposits were found, many of which were situated on or built against bedrock. Several post holes were identified, which likely represent a second site of the structure investigated in Trench 1.

In the northwestern corner of Trench 4, an unusual installation with several complete or nearly complete vessels were identified. The installations consisted of several large flat stones facing an embankment, along with numerous pieces of clay hearth material. The ceramics included very large piece of a greyware storage vessel, very typical of Late Bronze-Early Iron Age, alongside a goblet and an unusu
al vessel interpreted as an incense burner (fig.6). The vessels were found within an extremely fine, gray, ashy deposit, suggestive of a hearth. Other installations with evidence of burning were recovered elsewhere in the trench.

Fig.6: Selection of complete or nearly complete pottery vessels recovered in excavation. Incense burner is on the left.

 While a number of additional pieces of slag were recovered in the 2017 excavations of Trench 1 and Trench 4, no clear pieces of slag were found on the clay floor surface or unequivocally within the fire-using installations. The floor assemblages did include a number of hammer and grinding stones, which may have had a metallurgical use. Indeed the association of smithing slag and hammer stones in some contexts supports the assumption that at least some of them were used in smithing. Nevertheless, the presence of hammerstones in floor contexts without slags hints that they may have had other uses as well.

Overall, the stratigraphic evidence and the production debris suggest that iron working took place within the large encircling fortification wall, probably a little way up hill from the excavated trenches. The fact that the overwhelming majority of diagnostic ceramics clearly date to the late 2nd and early 1st millennium BC strongly suggests that occupation at the site is restricted to the Late Bronze and Early Iron Age.

The lack of slag finds in situ on the floor surface suggests that the debris post-dates the initial foundation of the settlement. The precise spatial context of the metal production
workshops within this compound remains unclear, and given presence of a modern communications tower at the top of the hill and the decreased depth of deposition farther up the hill slope, it is unclear whether further excavations would resolve these questions. While we cannot be certain that the wall remained in good repair during the later periods when iron smithing took place, this assumption is not unreasonable. Situating a smithy at the top of an otherwise uninhabited hill makes little sense.

One of the key research goals was to place metal production debris within a clear spatial (and by extension social) context. Excavation in Trenches 1, 5 and 6 showed that the wall encircling the top of the hill is roughly 2 to 2.5 meters thick. The better preserved sections show that the wall was constructed by laying facing stones on each side and then filling the interior with stone rubble. Facing stones reached sizes of up to 70 cm in diameter, but most were more modest in size. Thus, while the wall is significant in size and probably required a significant investment in labor, it is conceivable that it was within the capabilities of a small community pooling labor resources.

Finally, a small test pit excavated directly in front of the rocky outcrops at the base of the hill showed unequivocally that it is not an ancient mine. No traces of ancient activity, mining or otherwise, were discovered in these excavations. Most probably, the hollow in the outcrop is the result of natural weathering processes, rather than human activity.


Fieldwork at Mtsvane Gora in Summer 2017 yielded a number of new samples of
production debris, the analysis of which will reveal further details about the metal production technologies used at the site. Analysis of slag deposits collected in 2017 is ongoing, but initial analysis shows that they too are related to iron production. Excavations demonstrated unequivocally that metal production debris dates to before c. 500 BC, making the site one of the oldest sites with evidence of iron production in the region. Finally, the excavations suggest that metal production took place within the walls of a fortified hilltop, suggesting that iron resources were deemed important enough to warrant protection behind a substantial wall.
Aside from metal production, the large quantities of animal bone, ceramics, and obsidian offer possibilities for additional research projects. Obsidian from the site will be analyzed to map patterns of movement between lowland and highland areas.

Recent Foundation grants: general        Archaeology Grants Program w/map  

Copyright    2018 Rust Family Foundation