CC BY
56
УДК 94(479.22)
DOI: 10.24411/2713-2021-2020-00013
Vakhtang Licheli, Roberto Dan, Priscilla Vitolo, Andrea Cesaretti, Tamar Chogovadze, Tornike Chilingarashvili
CONTINUITY OR CHANGE? AN ANALYSIS OF FORTIFIED CHURCH COMPLEXES IN SOUTHERN GEORGIA*
This article describes and discusses a series of archaeological sites. All the sites contain an ancient enclosure built using clearly pre-medieval masonries, inside of which a number of churches have been built. The dates of these fortified structures, usually improperly defined as built using "cyclopean technique", cannot be established without archaeological excavations. Recent attempts have been made to establish functional continuity between the ancient enclosures and the churches inside them. The goal of this text is to present and discuss these interesting sites and analyse the available data relating to the function of these structures from a diachronic perspective.
Key words: Southern Georgia, Fortified churches, Bronze Age, Iron Age, Middle Ages.
About the authors: Vakhtang Licheli1, Professor (history and archaeology), Ivane Javakhishvili Tbilisi State University; Roberto Dan2, PhD (history and archaeology), ISMEO — The International Association of Mediterranean and Oriental Studies; Priscilla Vitolo3, PhD candidate (archaeology), ISMEO — The International Association of Mediterranean and Oriental Studies; Andrea Cesaretti4, PhD candidate (archaeology), ISMEO — The International Association of Mediterranean and Oriental Studies; Tamar Chogovadze5, PhD candidate (archaeology), Ivane Javakhishvili Tbilisi State University; Tornike Chilingarashvili6, PhD candidate (archaeology), Ivane Javakhishvili Tbilisi State University.
Contact information: 1,5,60179, Georgia, Tbilisi, 1 Ilia Chavchavadze Avenue, Ivane Javakhishvili Tbilisi State University; e-mail: Vakhlich@hotmail.com, vakhtang.licheli@tsu.ge; 5tatuchogovadze@gmail.com; 6tornike2210@mail.ru; 2,3,400186, Italy, Rome, Corso Vittorio Emanuele II, 244, ISMEO — The International Association of Mediterranean and Oriental Studies; e-mail: 2roberto_dan@hotmail.it; 3vitolo.pr@gmail.com; 4a.cesaretti12@gmail.com.
Вахтанг Личели, Роберто Дан, Присцилла Витоло, Андреа Кесаретти, Тамар Чоговадзе, Торнике Чилингарашвили
ПРЕЕМСТВЕННОСТЬ ИЛИ ИЗМЕНЕНИЕ?
АНАЛИЗ УКРЕПЛЕННЫХ ЦЕРКОВНЫХ КОМПЛЕКСОВ ЮЖНОЙ ГРУЗИИ
В этой статье описывается и обсуждается ряд археологических памятников. У всех них есть древняя ограда, построенная с использованием явно досредневековой каменной кладки, внутри которой было построено несколько церквей. Датировка этих укрепленных сооружений, которые обычно неправильно определяют как построенные с использованием «циклопической техники», не могут быть установлены без археологических раскопок. Недавно были предприняты попытки установить функциональную преемственность между древними оградами и церквями внутри них. Цель этой статьи — представить и обсудить эти интересные памятники и проанализировать имеющиеся данные, касающиеся функции этих структур, с диахронической точки зрения.
Ключевые слова: Южная Грузия, церкви-крепости, бронзовый век, железный век, средневековье.
Об авторах: Вахтанг Личели1, профессор (история и археология), Тбилисский Государственный Университет им. И.А. Джавахишвили; Роберто Дан2, PhD (история и археология), Международная
Статья поступила в номер 11 декабря 2020 г. Принята к печати 26 декабря 2020 г.
© Vakhtang Licheli, Roberto Dan, Priscilla Vitolo, Andrea Cesaretti, Tamar Chogovadze, Tornike Chilingarashvili, 2020.
ассоциация средиземноморских и восточных исследований; Присцила Витоло3, PhD candidate (археология), Международная ассоциация средиземноморских и восточных исследований; Андреа Кезаретти4, PhD candidate (археология), Международная ассоциация средиземноморских и восточных исследований; Тамар Чоговадзе5, PhD candidate (археология), Тбилисский Государственный Университет им. И.А. Джавахишвили; Торнике Чилингарашвили6, PhD candidate (археология), Тбилисский Государственный Университет им. И.А. Джавахишвили.
Контактная информация: 1,5,60179, Грузия, г. Тбилиси, пр. И. Чавчавадзе 1, Тбилисский Государственный Университет им. И.А. Джавахишвили; e-mail: Vakhlich@hotmail.com, vakhtang.licheli@tsu.ge; 5tatuchogovadze@gmail.com; 6tornike2210@mail.ru; 2,3,400186, Италия, г. Рим, Корсо Витторио Эмануэле II, 244, Международная ассоциация средиземноморских и восточных исследований; e-mail: 2roberto_dan@hotmail.it; 3vitolo.pr@gmail.com; 4a.cesaretti12@gmail.com.
This article examines a particular group of monastic complexes spread throughout southern Georgia, especially in Samtskhe-Javakheti and Kvemo Kartli regions (fig. 1). These monastic complexes are characterized by the reuse of pre-medieval fortified structures inside which churches and annexed structures were built in different periods. Recent proposed interpretations have suggested that the religious aspect of these medieval complexes was in continuity with the structures of the previous era within which the churches were established, trying to demonstrate the cultural continuity of these complexes. The aim of this article is to analyse precisely the remains present in these sites and verify the sustainability of this interpretative hypothesis. The inspiration for this work comes from the examination and study of some of the sites considered in the text as part of a Georgian-Italian research project named the Samtskhe-Javakheti Project (SJP), which has been active since 2017. The archaeological investigations are conducted in accordance with a Memorandum of Understanding agreed between the Archaeological Association of Georgia, which involves the collaboration of students from Tbilisi State University, and ISMEO — The International Association of Mediterranean and Oriental Studies1. The project is under the patronage of the Italian Ministry of Foreign Affairs and International Cooperation (MAECI; ARC-001675). From 2018, the mission has also been funded by MAECI (ARC-001737), as part of the newly established Archaeological Mission to South Caucasus — ISMEO (AMSC — ISMEO), which unites ISMEO's activities in Georgia and Armenia into a single major project. The first expedition of the Samtskhe-Javakheti Project took place in 18-26 May 2017, the second between 27 May and 12 June 2018 and the third between 2 June and 17 June 2019. The project's purpose is to study the archaeological remains of the region, especially the areas located near the southern border of Georgia (the municipality of Aspindza, Akhalkalaki and Ninotsminda), a little investigated area with a stunning archaeological potential. The main objectives are the following:
1) Production of an archaeological map of the Samtskhe-Javakheti Region (municipalities of Aspindza, Akhalkalaki and Ninotsminda) showing the distribution of archaeological sites;
2) Analysis of the establishment and development of the first proto-state communities in the area (Late Bronze Age — Early Iron Age);
3) Study of settlement patterns and the characteristics of local communities, focused on their relationship with the kingdom of Urartu, in order to understand the limits and expansion of Urartian cultural influence in this area (Middle Iron Age);
4) Study of the architectural, cultural and socio-economical features under Achaemenid influence (Late Iron Age, Post-Urartu/Achaemenid period).
1 A project of the Italian Ministry of Education, Universities and Research (MIUR) "Progetto MIUR. Studi e ricerche sulle culture dell'Asia e dell'Africa: tradizione e continuity, rivitalizzazione e divulgazione".
Archaeological sites with fortified churches
The following selected sites that feature the coexistence of fortification structures and medieval churches will be analysed from an archaeological perspective. Four of these (Ikhtila, Meghreki, Kilda and Aspara-1) were visited and documented directly as part of the Georgian-Italian expedition between 2017 and 2019. Four other sites are located in the same region — the Tsunda church and the sites of Balanta, Orlovka and Karzameti, but these have not yet been visited as part of the mission2. In any case, information on them is available in literature. Two sites (Meore Salamaleiki and Lipi) are located in the western part of the nearby Kvemo-Kartli region, in territorial continuity with the Javakheti plateau.
Ikhtila (SJP023)3
The site of Ikhtila is located in the south-western outskirts of the village of the same name, 70 m east of a small stream that crosses the village, and 16 kilometres north of Akhalkalaki. The site consists of a medieval church built in the middle of the remnants of a pre-medieval fortification (fig. 2—3). The church, which measures 12.12 x 9.72 m (Karapetyan 2011: 182) is dedicated to St Stephen. The church was built in the middle of a large sub-circular enclosure measuring 40 x 33 m, with a perimeter of 117 m and an area of 0.11 ha. The drystone walls are 1.86 m thick (measurement taken in the southern sector). In the eastern sector of the wall there is a door with an architrave composed of a single elongated block. The gate is 1.42 m wide, 1.89 m deep on the north side and 2.10 m deep on the south side. The threshold of the door is made of stone slabs and seems to be the original one. The southern sector of the wall is that best preserved, especially in the lower part, while the whole perimeter has been visibly remodelled (probably maintaining its general form) in medieval times, evidenced by the use of smaller, roughly worked stones in the upper part. The certainty that it is an old fortification is due not only to its architectural perspective — with the presence of colossal wall blocks in the lower part of the southern wall, as well as the general character of the building, but especially thanks to the discovery of pottery on the surface that can be dated to the Iron Age. This pottery was found on a small slope in the external south-western sector, where it is possible to see the trace of another wall, probably contemporary. The inner part of the fortified enclosure is completely filled with soil and houses not only the church but also part of the medieval cemetery, a sign of the site's constant rearrangement. Future archaeological investigation are needed, given the potential of the archaeological deposit.
Meghreki (SJP024)
The site of Meghreki, known locally also as Meghrek or Neghrek, is located 2.3 km east of the village of Ikhtila, and about 2 km west of Patara Samsari village. It is situated in the gorge formed by the course of the tributary Meghrek, which originates from Mount Samsari and is also known by the Armenian name of Vank (monastery).
The site is features the ruins of a settlement and a church, built inside the remnants of an ancient rounded fortification (fig. 4—5). The walled enclosure is oval in shape with dimension of
2 The visit to these sites was scheduled for the 2020 mission but was postponed due to the global pandemic.
3 In this text the names of the sites are followed by the site codes used by the expedition to classify them. The names used are always those already present in the specialist literature in the case of already known sites. The sites discovered as part of our activities have been named on the basis of the nearest settlement or using the names employed by local communities to identify them.
70 x 58 m, but unfortunately this is not its original size, since part of the walls were removed by the construction of the modern road which has destroyed an area about 60 m long. The current length of the wall is 140 m, and the area of the site is 0.30 ha; the original length of the wall must have been about 215 m and the total area 0.36 ha. The perimeter wall is without mortar; its two external faces are made of large rough stones. At the centre of the southern sector is the entrance door to the building, covered by a large stone lintel. On the crest of the walls traces of mortar are visible, indicating its reuse, probably in medieval times. The walls are preserved for a maximum height of 6—7 courses, i.e. about 3 m; they are 1.80 m wide in the southern sector and 1.65 m wide in the western sector. Although the conformation of the wall is perfectly legible, the whole south-east sector has been truncated by the construction of the new road that runs alongside the site. The perfectly preserved door is finely worked, with well-defined corner jambs, 1.04/1.27 m wide and 2 m high. The stone threshold consists of a single finely carved slab and is 44 cm wide. The jambs are well smoothed, formed on each side by two large slabs; internally, especially in the lower west part, the blocks have vertical recesses. The architrave is made up of three large stones. On the inside, the second stone has two semi-circular cavities on the long side, in front of the door: the eastern one of these is 17 cm in diameter, and the western one 19 cm; both are 30 cm deep. We can imagine a slab door with two poles inserted from the top of the wall and lowered from above to block it. The doorway forms a small entrance hall, made by five blocks before the architrave, 1.80 m wide and 1.75 m deep. The depth of the entranceway composed of the corridor and the door is 2.30 m. The external architrave has a small step to better accommodate the jamb (on the west side). Outside two single rows of stones create a corridor towards the door, but it is not possible to date them. The remains of a medieval village and those of a 12th — 13 th century CE church are located within the oval fortified enclosure and inside there are several smaller curved walls that butt up against the perimeter wall, clearly later in date than the latter. The church has a nave measuring 10.10 x 11.5 m constructed of finely-finished stone blocks (Karapetyan 2011: 187). The remains of structures can also be seen outside the external wall, especially on the north side. A single-row structure with an elongated rectangular shape aligned south-west/north-east overlooks the western area. There are also circular shaped heaps of small stones which might be due to the removal of structures to facilitate modern agricultural work. The comparison of the decoration of two — eastern and southern — windows shows a possible chronological gap between their construction — c. 10th and 14th centuries CE (Berdzenishvili et al. 2000: 66—67). But there are no signs of rebuilding; on the basis of its plan, stone masonry and window decoration we may deduce that the church was built at the end of the 10th century CE (Berdzenishvili et al. 2000: 67).
Kilda (SJP050)
This site is located in a flat area on the margin of the gorge overlooking the River Mtkvari/Kura Valley. It is located 4 km north-west of the village of Kumurdo, about 12 kilometres west of Akhalkalaki. The site is characterized by an extensive settlement area with a fortified structure inside which a medieval church was built (fig. 6). The village covers an area of 620 x 430 m for a total area of about 19 ha. At the north-eastern margin of the site there are the remains of a circular fortified enclosure on which a church was built in the 10th century CE (Elizbarashvili 1986: 77—78; Berdzenishvili et al. 2000: 96; Karapetyan 2011: 337). This situation is reminiscent of the sites of Meghreki (SJP024), Ikhtila (SJP023) and Aspara (SJP056). The probable area covered by this ancient structure is 0.12 ha, with an estimated perimeter of about 130 m and dimensions of 40 x 35 m. The church, which has two naves, measures 11 x 8 m and was evidently built with material reused from previous dismantled structures. The best preserved part of fortification is
located to the east, in correspondence to the entrance of the church; it is 8.10 m long and interrupted by a passage 1.30 m wide and 1.40 m deep, presumably placed in connection with the ancient entrance to the structure (fig. 7). In various parts of the site, among the ruins of the village, there are fragments of drystone walls made with large stones preserved to heights of up to 3 m. A series of impressive walled structures are found at the western edges of the site at the gorge. These are curved walls built against the natural rock walls overlooking the gorge, resembling situations seen during our activities in other sites such as Tsikhistavi-1 (SJP059) and Dakhvanda-2 (SJP048). The largest of these structures, which is oval in shape, has a length of 26 m and a width of about 2 m, built in a double-faced drystone wall. An ancient road started from the western edge of the site and descended towards the valley of the River Mtkvari/Kura. The current layout probably dates to medieval times, but the road could be older.
Aspara-1 (SJP056)
The site is located on the western shore of Lake Paravani, in a flat area located in the middle of the modern village of Aspara, 180 m from the shore. It is located 25 km north-east of Ninotsminda. The site is composed of church built on top of an older fortification (fig. 8—9). The fortified enclosure has a circular shape with dimensions of 20 x 17.5 m, a fortified perimeter about 60 m long and a surface area of 0.3 ha. The wall has a width of 2.8 m and is preserved for 5 courses of stones with a maximum height of 2 m. It was built with big rough stones in dry, double-faced technique and an internal fill of medium and small sized stones. The northern part of the wall is completely covered by soil and is not visible. A very well preserved gate, although partly buried by debris, is visible in the south-western part of the wall, with two architraves still in situ. The outermost stone of the architrave measures 1.10 m deep, 0.65 m high and 2.70 m long. The entire door is 2.80 m deep and 2.30 m high, with a 1.20 m wide passage. The ruined church dates to the 11th century CE (Berdzenishvili et al. 2000: 17). The nave is rectangular in shape and measures 11.85 x 5.84 m (Karapetyan 2011: 82).
Other complexes
A further six similar archaeological contexts are presented. They have not yet been visited as part of the project, but information is available in the literature. The site of Balanta is located at the north-western limit of the homonymous village, on a low hill, in Samtskhe-Javakheti region. It is characterised by a fortified enclosure of elliptical shape (fig. 10: A). This enclosure is 190 m long and covers an area of 0.27 ha. On the west side, a further wall is attached to the main enclosure, creating a sub-rectangular area covering 0.07 ha. The length of the second wall is about 77 m. Inside the enclosure there is a 19th century church restored in 1914, called the Church of the Nativity, surrounded by a cemetery.
Another church surrounded by a small fortified enclosure is located about 4.5 km north-west of Meore Salamaleiki village in Kvemo-Kartli region (fig. 10: B). The fortified enclosure has a wall that is 1.5—2 m wide and covers an area of 0.07 ha. The length of the wall is about 105 m. The structures identified there have been considered a shrine and classified among the Religious/Cult Buildings (Narimanishvili 2019: 449).
A further intriguing complex, again located in Kvemo Kartli region, is known as Lipi, and is located 290 m south-west of Tejisi village (fig. 10: C). There is a circular enclosure, built with double faced, unmortared walls of medium and large stones, 1.5—2 m wide. It has two entrances, one on the north side and one on the south, both characterised by huge monolithic stone architraves.
In the middle of it, another late medieval church called St Kostantine (nave measuring about 9 x 6 m) has been built. The enclosure has an area of 0.06 ha and the total length of the wall is about 90 m.
An additional similar complex is located about 5 km south of the Orlovka village4 in Samtskhe-Javakheti (fig. 10: D). Here, in the centre of a ruined Middle Ages village there are the remains of a church built in the middle of a fortified enclosure that is sub-triangular in shape. The enclosure, characterised by the presence of medium and large stones assembled into double-faced drystone walls, is badly damaged. It covers an area of 0.06 ha and the length of the wall is about 105 m.
Yet another site with similar characteristic is Karzameti, very near the border between Georgia and Turkey, again in Samtskhe-Javakheti (fig. 10: E). Here, on the northern margin of a ruined medieval village a Middle Ages church (measuring about 18 x 10 m) was built in the second half of the 13th century CE (Berdzenishvili et al. 2000: 125) in the centre of an elliptical fortified enclosure. This is considered to be one of the biggest churches with just a nave in all of Georgia. This drystone wall, double-faced and composed of medium and large stones, has a width of 1.5—2 m. The wall is about 135 m long and the enclosure has an area of c. 0.13 ha.
A final site, albeit with slightly different characteristics, that should be added to this group is the Tsunda church built in 12th — 13th century CE (Berdzenishvili et al. 2000: 109—111). Although this building does belong to the exact category discussed in this article, this church is important because it shows structural continuity over the millennia, being an example of how ancient abandoned structures in ruins were reused for the construction of churches or monastic complexes. This process is manifested in various ways, in most cases through the reuse of ancient fortified structures for the creation of monastic complexes (or defensive systems for religious buildings), while at Tsunda a different phenomenon has occurred. The church known as Tsunda is located about 170 m north-east of the village of Tmogvi, on the east side of the River Kura/Mtkvari at the bottom of the valley. The church of Tsunda (13.6 x 6.8 m), dedicated to St John the Baptist and built in the 12th — 13 th century CE, was clearly constructed on the remains of a fortified structure (Kurumidze 2000) built for military purposes, set on a natural rock outcrop, strategically overlooking the surrounding areas (fig. 10: F). Without more specific studies and investigations it is difficult to establish the chronology of this fortified structure, but we can currently hypothesize a generic dating to the pre-medieval era. It is important to underline that the phenomenon of the reuse of older structures is also attested in other parts of Georgia in areas not covered by this contribution. It will suffice to mention the very interesting church of Kaishaurni5 in Mtskheta-Mtianeti Region. Here, on the top of a horseshoe-shaped tower typical of the region, a small 17th — 18th century CE chapel has been built (fig. 10: G). It is morphologically similar to the one called Nijigori-1 (SJP011), for example, and to others identified in Georgia and the Ardahan region of Turkey6.
A last interesting example directly documented during our activities, although not connected with the construction of a religious building but of a military one, is the site of Aragva (SJP022)7, on the western bank of the River Baraletistskali/Arakvistskali, 8.5 km north of Akhalkalaki. Here a Middle Ages tower was built directly on the top of the remnants of a pre-medieval building, characterised by drystone walls made of large stones (fig. 11).
4 Called Gorelovka by Narimanishvili (Narimanishvili 2019: 408), but the closest village to the site is Orlovka.
6 Coordinates: 42°26'5.55"N 44°30'59.16"E; elevation: 1810 m a.s.l.
6 See for example the towers of Karakale and Ziyaretdere (Pataci 2016: fig. 20: 145—147).
7 Coordinates: 41°29'25.9"N 043°29'36.1"E; elevation: 1713 m a.s.l.
Site name Coordinates Altitude (m a.s.l.) Enclosure area
Aspara-1 41°27'09.07''N 43°46'53.04''E 2100 0.03 ha
Balanta 41°37'06.82"N 43°32'24.78"E 1897 0.27 ha
Ikhtila 41°33'10.08"N 43°32'20.04"E 1708 0.11 ha
Kaishaurni 42°26'05.55"N 44°30'59.16"E 1810
Karzameti 41°20'26.74"N 43°07'15.54"E 1954 0.13 ha
Kilda 41°26'21.22"N 43°20'24.30"E 1680 0.12 ha
Lipi 41°41'31.23"N 44°05'18.48"E 1780 0.06 ha
Meghreki 41°33'12.30"N 43°34'19.03"E 1850 0.36 ha
Meore Salamaleiki 41°28'54.46"N 43°59'58.84"E 1684 0.07 ha
Orlovka 41°10'27.02"N 43°38'30.69"E 2130 0.06 ha
Tsunda 41°24'27.36"N 43°20'04.23"E 1254
Analysis of the fortified enclosures
On the basis of recent studies (Licheli et al. forthcoming b) it has been proposed that during a still unspecified period in the Bronze Age or Iron Age, a particular type of defensive construction with quite distinctive characteristics spread over the territory of southern Georgia. These are fortified structures, usually circular or sub-circular in plan, characterized by walls often improperly defined as "cyclopean"8. These fortified structures have common characteristics: a single entrance, the absence of towers or buttresses on the external faces of the main walls, a construction technique involving the use of double-faced walls with external faces in large rough blocks and a filling of loose material. Almost all these structures are located in paradoxically unfavourable and poorly defensible geographical positions. In fact, they are usually found in flat areas on the edge of water sources (streams or lakes). Usually, with rare exceptions, they have rather small dimensions (between the 0.36 ha of Meghreki and 0.03 ha of Aspara) and very rarely contain the remains of structures inside them. This type of fortified enclosure with rather particular characters has been interpreted as walled areas that did not have strategic functions from a military point of view, such as the control of roads, or the defence of a particular access route or of goods that were contained inside. They probably served as shelters for the populations who lived in immediately adjacent areas at times when armed groups (who would have frequently crossed these territories) passed by9. In several of these structures, Christian churches dating back to different eras have been identified. The architectural relationship between these more recent churches and the older structures vary
8 An element of inconsistency which has probably made it more difficult to understand the chronology and the function of many fortified sites in the region is the generalized use of the term "cyclopean" in association with all the structures with drystone walls made of large stones, but which corresponds to no particular architectural or organizational features of the buildings. With the overall definition of "cyclopean fortress" there is a substantial minimization of the numerous chrono-typological problems affecting this type of context. Other scholars (see, for example, Biscione 2009: 223) have already made attempts to improve on these terms; here and in future work we will try to follow this trend in favour of more technical and less generalized definitions.
9 An interesting example of this kind of fortified enclosure/shelter is the small structure called Didi Khanchali (SJP020).
according to the size of the fortified structures; for example at Tsunda and Kaishaurni the fortified structures were used as foundations for the construction of the churches themselves. In other cases, such as at Lipi and Aspara, the church is contained within the fortified enclosure but there was evidently no space for the addition of other structures, whereas in the case of Meghreki, the size of the fortified structures allowed the church to be accompanied by a fair number of buildings, which are still visible on the ground.
From fortified structures to cult complexes: continuity or change?
For our purposes, it is important to discuss a recent proposal for the interpretation of most of the sites covered by this contribution. In a recently published a catalogue of fortified sites in southern Georgia, they are apparently united by the use of the term "Cyclopean". Analysing Urartian written sources, among these "Cyclopean complexes" three groups of "cyclopean buildings" have been identified. These are "Settlements", "Fortifications without settlements" and a last category of "Religious complexes", subdivided into "Sanctuaries" and "Cult centres". Although it is — rightly — admitted that "No data is preserved about the cult complexes of tribes in South Caucasus in Urartian written sources" — and, we may also add, from an archaeological perspective — it is affirmed that religious complexes "definitely existed on ancient territory of South Caucasus". To support this statement the interpretation of the improperly defined "Shaori Megalithic Complex" as a cult centre is proposed (Narimanishvili 2019: 78—79, 96, 369). This is not the place to discuss the Shaori site in detail. It is important just to bear in mind that a recent new study has made it possible to redefine its functional characteristics, that appear in all respects similar to those of the Abuli complex. For geographical and climatic reasons these high-altitude sites must be considered to have been temporary shelters, certainly not "royal cities" or "cultic centres"10. In any case, on the basis of these assumptions most of the sites discussed in this text have been placed in the "Sanctuaries" category. These have been described as "circular cyclopean construction on roads and passages. Most of them are destroyed or rebuilt in the Christian period. The diameter of that kind of constructions' walls is mainly 30—40 meters, with an entrance occasionally. It is worth mentioning, that walls of the "Sanctuaries" are without buttress" (Narimanishvili 2019: 79). Starting from the methodological approach mentioned above, the Ikhtila fortified enclosure is defined as an "important religious centre" or "sanctuary", as well as Meghreki. In the "Religious/Cult buildings" category are also inserted Karzameti, Kilda, Aspara, Orlovka, Meore Salamaleiki and Lipi (Narimanishvili 2019: 84, 90, 215, 299, 303, 337, 396, 408, 449, 474). Other sites, such as Kartsebi and Saghamo are considered religious centres, probably due to the presence of elliptical fortified enclosures11. Aspara, defined as a "prehistoric sanctuary", is considered to be a cultic complex of the late Bronze/Iron Age (Narimanishvili 2019: 83, 94). Unfortunately, given the available data, it is impossible to establish the chronology of these fortified enclosures, which could have been built at an unspecified time during the Bronze Age (including before the Late Bronze Age) or the Iron Age. Regarding their function, it must be underlined that the idea that these small enclosures were cultic centres because they later hosted a number of medieval churches is methodologically incorrect, and in any case a thesis difficult to support in the absence of additional evidence.
10 On this, see Licheli et al. forthcoming b.
11 The absence of churches seemed problematic after the typological proposals (Narimanishvili 2019: 334).
Conclusions
As we have seen, the phenomenon of the reuse of ancient fortified enclosures, generically attributable to a period spanning the Bronze Age and Iron Age, is widespread. These structures, often and unusually located in not particularly favourable or strategically useful places, but always close to streams or bodies of water, were reused in the Middle Ages, especially for the construction of structures with religious purposes, churches or small chapels. Recently some scholars have classified most of the places where these characteristic fortified churches are found as "religious/cult buildings", specifically as a subtype defined as "shrine". There is an obvious attempt to recognize in the choice of these fortified places a functional continuity that must have lasted until the Middle Ages. The choice to build churches within these fortifications is considered to have occurred because these same improperly defined "cyclopean structures" must have had a religious/cultic function. This type of interpretation is linked to a broader interpretative problem that involves the famous fortified site of Shaori, which over time was first associated with the Trialeti kurgan and later defined more generally as a "cult centre"12. Objectively, the data in favour of this interpretative approach are tenuous and the fortress of Shaori itself, together with that of Abuli, has been considered in the wider context of these shelter sites13. If we exclude the presence of medieval churches built on or within these fortified structures (none of them investigated by means of regular excavations), there is no archaeological or other evicence that could suggest a function related to the religious sphere for these structures. It is highly possible that the construction of these medieval churches within these fortifications has a merely functional basis connected to the generalized reuse of structures that could provide shelter without any cost associated with their construction.
References
Berdzenishvili, D. Elizbarashvili, I. Vacheishvili, N. Chachkhunashvili, T. 2000. Javakheti. Istoriul-
khurotmodzghvruli gzamkvlevi. Tbilisi: Nekeri (in Georgian). Elizbarashvili, I. 1986. Two-aisled church from the village Kilda. Dzeglis megobari 1, 28—33 (in Georgian). Karapetyan, S 2011. Javakh. Scientific Research Series of Research on Armenian Architecture. Vol. IX. Yerevan:
Research on Armenian Architecture (RAA) Foundation. Karumidze, T. 2000. Vol. VI. Tsunda Church, Essays of Georgian Museum of Art. Tbilisi: Georgian Museum of Art (in Georgian).
Licheli, V., Dan, R., Vitolo, P., Chilingarashvili, T., Chogovadze, T., Saccone, T. forthcoming a. Searching for the Northern Borders of Urartu: Preliminary Results of the First Two Seasons of the Samtskhe-Javakheti Project (2017—2018), Georgia. In: Kuntner, W., Heinsch, S., Dan, R., Gavagnin, K. (eds.). The Spread of Urartu: Proceedings of the Workshop, held in Innsbruck, 18 of July, 2018 at the 64th. Leuven: Peeters (Ancient Near Eastern Studies Supplement) (in print). Licheli, V., Dan, R., Vitolo, P., Chilingarashvili, T., Cesaretti, A., Chogovadze, T., Saccone, T. forthcoming b. Cyclopean fortresses, royal cities or mountain shelters? The Abuli and Shaori complexes in southern Georgia in the light of recent archaeological investigations. Ancient Civilization from Scythia to Siberia. Narimanishvili, D. 2019. Cyclopean Fortresses in Georgia. Tbilisi: "D&D" Ltd. (in print). Pataci, S. 2016. Ardahan Kale ve Kuleleri. Ardahan: Duru Ofset.
Shanshashvili, N., Narimanishvili, G. 2014. Environment and Dwelling in the Early and Middle Bronze Ages South Caucasus. In Narimanishvili, G. (ed.). Problems of Early Metal Age Archaeology of Caucasus and Anatolia. Tbilisi: Mtsignobari LTD, 245—264.
12 According to Goderdzi Narimanishvili, the fortress of Shaori shows a certain similarity in construction technique and materials to the Zurtaketi and Trialeti burial mounds (Middle Bronze Age), a circumstance which led the scholar to propose a possible construction date in the first half of the 2nd millennium BCE for the Shaori complex. In addition, he suggested that the road that leads to Shaori Area A might be interpreted as a ritual or processional street similar to those seen in the previously mentioned kurgans in Trialeti. In addition, a possible Early Bronze Age date has been suggested (Shanshashvili — Narimanishvili 2014: 247).
13 Licheli et al. forthcoming b. In this regard, we must note that the Abuli complex has been improperly defined as a "royal city" (Narimanishvili 2019: 78, 374).
PL Ikhtila
2. Meghreki
3. Kilda
4. Aspara
5. Balanta
6. Meore Salamaleiki1
7. Lipi
8. Orlovka
9. Karzameti
10. Tsunda
11. Aragva
-S \
a
(.»ieBSK
. ./5 JmfQjmj
s^i
«kSS j&s
-Aw
LAKE "- ■
■ ¿t
8 * .*'
r- * «»Pi
tabaskuri .
S
1 O HI ■ ■ ^ i '
. ...
tsalka reservoire
. ¿A T^^WvfjjÄ » K^T ' . - V' '
lake
paravani
V
Googje Earth
fflage.fsridsat / Coperniralr
&
■y
* v lake /
khozapini/kartsakhi
akhalkalaki
%. . V« ";. . -, ' " . ft
/ • . ." "
,y-» * immhw iwv
• &&
- ■ -
..üfai ■ meto
W
li-tr «TO
'-«ja
- . ..w.
. ■ * * • f > -A
fit-
Wr
v
■ *
v,
li-«'.' v .
TRH
\ IT**?»»
r
J
On
JQ
r £
ft
o
tr o
ÜQ
O
<
&
N
ft £
cr
<
£ CT* ft
"a
3.
ft 1
0 cr
1
p I
o*
ft p
O ft
s
Fig. 1. Distribution map of the sites mentioned in the text (Satellite picture after Google Earth).
£
> 9
Fig. 2. Aerial view of the fortified enclosure of Ikhtila (SJP023) and the church inside it (Samtskhe-Javakheti Project Archive).
Fig. 3. Aerial view of the fortified enclosure of Ikhtila (SJP023) with the church in it (Samtskhe-Javakheti Project Archive).
Fig. 4. Aerial view of the fortified enclosure of Meghreki (SJP024) with the church and other structures (Samtskhe-Javakheti Project Archive).
Fig. 5. Aerial view of the fortified enclosure of Meghreki (SJP024) with the church and other structures (Samtskhe-Javakheti Project Archive).
Fig. 6. Aerial view of the site of Kilda (SJP050) with the settlement and the church (Samtskhe-Javakheti Project Archive).
Fig. 7. The fortified enclosure and the church (Samtskhe-Javakheti Project).
Fig. 8. Aerial view of the fortified enclosure of Aspara (SJP056) and the church inside it (Samtskhe-Javakheti Project Archive).
Fig. 9. Aerial view of the fortified enclosure of Aspara (SJP056) and the church inside it (Samtskhe-Javakheti Project Archive).
hr
* J
tu
is
e ai
o o
O
r e afte
a <
t c
i
oo o
oo" vo m
e h
■s
il
e
As"
O <N
> h
im
ari
Z
am m
a d n u
<u /—) « 2
a
M
>
<N
O <N
i
m §
-o q •>
o to
en
I I
N
2 a z i — M I
to q
Fig. 11. Medieval tower of Aragva (SJP022) built on an older fortified building (Samtskhe-Javakheti Project Archive).
