Древности Семидворья I. Средневековый двухапсидный храм в урочище Еди-Евлер (Алушта, Крым): исследования и материалы [Drevnosti Semidvorja I. Srednevekovyj dvuxapsidnyj xram v uročišče Yedi-Evler (Alušta, Krym): issledovanija i materialy]

International audienceThe collective work presents the study and publicationof excavated materials of an archaeologically known twoapsechurch from the first half of the 9th to the first half ofthe 10th century at the top of the Tuzluk Hill in the Yedi Evlerarea, Crimean Peninsula, near the village of Semidvorie(Alushta, Crimea, Ukraine). This sanctuary was linked tothe large agricultural and pottery producing settlement thatexisted in this economically developed and populous regionin the second half of the 8th/9th - first half of the 10th century.The settlement was situated 150-250 meters to the east andsoutheast from the church. Just 350 meters southeast fromthe church was a medieval cemetery of the “Suuksu” typeof the 7th – 8th / 9th (?) centuries existed which was left by thepopulation usually identified as Crimean Goths tribes.In 2007, an area of around 96 m2 was investigated andchurch ruins and surrounding cultural layer were studied.The stratigraphical analysis managed to identify here 44archaeological layers or contexts, one medieval grave withdouble burials, and a Bronze Age cultual place. The studyof ruins shows that the sanctuary was rebuilt multiple times.The church consisted of two communicating compartmentsof different sizes. As for characteristic features, the southernmain apse is bigger in size than the northern one, andthere was an entrance in the main part of the church throughthe northern compartment as well as two other doorwaysfrom the west and from the south. The western portal of thenorthern compartment was completely open and no traces ofwall masonry here were attested. In contrast to the southerncompartment, the foundation of the northern part was cutin natural. The three-layer masonry wall was made of localpoor faceted rectangular stones of various sizes. For buildingmortar, mud solutions with clay loam as a binder elementwere mostly used. The inside of the southern churchwalls was plastered with lime mortar, which in some placesis preserved in situ, and painted with red linear and geometricpatterns including letters or even inscriptions that aretoday illegible. The roof likely had two slopes covered bylocally made tiles of different types.The overall dimensions of the church were: width – 5.60-5.70 m, length - 8.50 m. The thickness of the wall was about0.7 m. Structure remains are preserved to a height of 0.80 m.Both apses have shoulders connecting apsidal semicirclesand walls. The external diameter of the southern apse is 2.13m. The internal dimension of the southern main compartmentis 2.34×4.15 м. The external diameter of the northernapse is 1.20 m, while the internal is 0.63 m. The width in thewestern part of the northern compartment is 1.34 m, and inthe eastern part it is reduced to 1.26 m. The church was orientedto the northeast. The azimuth of its central axis is 47°,which roughly corresponds to the azimuth point of sunriseduring the summer solstice for Crimean latitude.SUMMARYIn the first chapter, written by V. Kirilko, the buildinghistory of the church and its architectural peculiarities arepresented. The double apse sanctuary belongs to the relativelyrare type of churches of the Middle Byzantine periodthat could be described as a two-apse church with unequalapses of different sizes. G. Dimitrokallis (1976), the authorof the most representative corpus of double apse byzantinesanctuaries, classified them as “pseudobiconques.” Thereare some examples of double apse churches in the Crimea(Sotera near Alushta, Sudak, Funa near Luchistoe settlement,Chembalo fortress in Balaklava). Yet, these sanctuariesmainly date back to the 14th century, with the one exceptionbeing the Sotera church that belonged to the periodof the 8th-10th century, and none of them provides an exactparallel to the church of Yedi Evler.During the short period of its history, the church wascompletely rebuilt at least once. The first building periodinvolved the creation of the main southern church with theapse and the three entrances from the west, south and north.It is highly likely that the church was intentionally conceivedby priests, ktitores or the Christian community as a doubleapse and two-part building. Immediately after the perfectionof the southern church, the additional northern compartmentwith open western portal and separate apse was added. Thispart of the church was connected to the main church via aspecial doorway in the wall dividing the compartment thatpreviously served as the northern entrance to the southernchurch. In fact, the second building period is distinguishedonly theoretically as a final step in the construction of thechurch. The chronology of the first two periods of the building’shistory, based mainly on the study of pottery and ceramicmaterials from the complex, dates back to the firsthalf of the 9th century, or more precisely the second-third tothe middle of the century.After a short period the church was completely destroyed,most likely due to inadequate construction worksor an earthquake. The third building period is determined as860-880s, when the sanctuary was rebuilt and reconstructed.After reconstruction, the northern compartment was buriedby earth and ruined stones and preserved according tocanon law practices for unused sacral Christian objects. Inthe third building period, the northern part was not active asa liturgical zone. The sanctuary became an ordinary ruralByzantine one-apse, one-nave church. A narthex was constructedin the eastern part of the sanctuary. The doorwaybetween the southern and northern parts was closed off bywall masonry. During the third building period, only twoentrances — the southern and western — were still active.The main entrance was the southern one, which was addedby a wooden apprentice. After the second deterioration ofthe church in the first half of the 10th century, no more renovations were carried out. The ruins were reused by the localpopulation for ordinary purposes no earlier than in the secondhalf of the 14th -15th century, as pottery fragments fromthe ruins show. Most probably, the narthex and apse wereused at this time as a temporary living structure in what isregarded in the chapter as the fourth building period. Theauthor proposes graphical reconstruction of the sanctuaryaccording to fourth building periods and shows architecturalparallels to this building among contemporary churches ofthe Northern Caucasus and Minor Asia.Chapter two, author I. Teslenko, deals with the stratigraphyof the site and description of archaeological layers.The analysis of excavated materials provided in the chapterallowed for the presentation of all steps of anthropogenicactivity on the Tuzluk Hill from the Bronze Age to moderntimes. The description of materials is organized by archaeologicallayers, with general characteristics of different findsincluded. Every layer inside and outside the church is attributedto a corresponding building period. A hypothesison the formation of each layer and its causes are also given.The most important layers are linked to two dilapidationsof the church, and some of them are attributed to regular liturgicallife and different rituals practiced in and around thesanctuary. Several layers may be left from construction andreconstruction works. A detailed description of the archaeologicalfinds and a cultural and liturgical interpretation ofstructures, layers and bones are given in the next chapters.In the third chapter, I. Teslenko provides an analysis ofceramic and pottery materials from the church. During theexcavation, 2,589 fragments of roof tiles and kalypters (55%of all ceramic materials), 637 fragments of kitchen and tablewares (13.5%) and 1,485 pieces of pithoi and amphora (31.5%) were recorded. Among them 9 intact rectangular rooftiles that were still preserved and 5 kalypters can be fragmentarilyreconstructed. Several tiles have a constructionsign or craftsmen marks as tridents and Greek letters «λ»,«ρ», «π» «В», «V». A theoretical estimation on the numberof tiles, including kalypters for covering the roof, has beendone. The amount is between 374 tiles / 376 kalypters and396 tiles / 397 kalypters in the second and third buildingperiod respectively. Accordingly, in the second period theweight of the roof was about 3893-3897 kg, for the thirdperiod – 4118-4122 kg.Nearly all excavated ceramic materials came from localproduction. The author lists the characteristics and providesa description of clay pottery and ceramic items, which showtwo craftsmen traditions. The first one emerged locally andis characteristic of primitive treatments, the use of a handpottery wheel and unsatisfactory baking. The second craftsmentradition reflects well-organized, high-technology commodityproduction oriented on the external wine trade. It ispresented specially by amphora. Today, there are more than40 known pottery workshops with high-technology kilns inthe southern part of the Crimean peninsula. Such a potterytradition was most likely brought here in the 8th-9th centuryfrom Minor Asia. The author discusses chronologies ofvarious types of local pottery, particularly amphora, and hemakes comparisons to groups of amphora known from differentregions of the Byzantine World. Local amphoras arepresented by so-called “Black Sea type” (second variant),which was produced until the mid-10th century, according tothe author. At the archaeological site, only two fragments ofimported pottery have been recorded: the bottom of a highneck brown clay jug with wide flat handles, no earlier thanthe mid-9th century, and a fragment of Glazed White Ware II,according to J.W. Hayes, from 10th century Constantinople.The kitchen pottery which were in use in Khazar kaganateis also absent. Ceramic finds in the church date back mainlyto the end of 8th-10th century; only several fragments of twored glazed sgraffito bowls and one fragment of a brown unglazedpot come from the 14th-15th century.The fourth chapter presented by I. Teslenko and A.Musin describes and studies the collection of glass lampfragments (342 items) that are partially not indentified.The bulk (91%) of the lamps comes from the third buildingperiod and is concentrated near the southern entrance tothe church, where the liturgy should start. Precisely withinthe same zone, micropieces of flint made by strike-a-lightfor making “liturgical fire” were recorded, and kitchen andbone remains from community meals were also attested.Glass lamps are presented by two main groups: polycandelonor beaker-shaped lamps with hollow stems, and singlelamps with handles on the rim. All lamps have close parallelsamong glass finds from other Middle Byzantine sanctuaries,for instance, Myra-Demre in Turkey, Thessaloniki inGreece, Chersoneses in Crimea, etc. The glass is mainly coloredlight green and blue. A slowly increased percentage ofpotassium oxide recorded by optical emission spectroscopymay point to glass production centers in the southeasternpart of Asia Minor or Levant.Chapter five, written by A. Musin, analyzes and classifiesmetal crosses found in the church. The excavation recordedat least 30 crosses and their fragments. Crosses wereused throughout the entire period of the church’s existence.Crosses are regarded as an ex-voto offering. Most of themwere concentrated in the altar zone of the sanctuary andnear the southern entrance to the church. Two crosses wereput in wall masonry that closed the doorway between thenorthern compartment and the main church during the thirdbuilding period, evidently with apotropaic magic purposes.Presumably, crosses were suspended on the church wall oron elements of the church’s interior, or inserted in them. Thecorpus of crosses is divided into five typological groups.The main group consists of iron crosses with an extendedlower branch made of two plates connected with a rivet thatderived from individual processional crosses and turned inex-voto. Some crosses with splayed arms were cut from thinsheet-metal, including copper alloy and probably silver,and decorated with punch ornamentation. Two crosses weremade of silver coins: Umayyad dirham (661 – 750 AD) andimitation of Arab-Sassanian half-drachma of the Sassanidking Kosrou II (590-629 AD).The two last groups of crosses can be compared to thecrosses of the type 1.2.2 according to J. Staecker found inEarly Rus’ and Scandinavia in the 10th – 11th century, especiallyknown to be in graves in Birka (Sweden), Gnezdovonear Smolensk, Timerevo near Yaroslavl (Russia), Kiev,Iskorosten (Ukraine) and other political and economic centersof the formation of early medieval states in Russia andSweden. Several scholars have insisted that the crosses havean Anglo-Saxon origin and appeared in Sweden around930-940s AD with the mission of bishop Uni from BritishIslands. However, after the Yedi Evler excavation, the Byzantineorigin of these crosses is quite clear. Crosses fromEastern and Northern Europe may have been created usinga Byzantine example or brought directly from this regionin several cases. During the cultural transformation of theChristianization period, crosses that initially belonged to liturgicalpublic culture were turned in barbarian society intoprivate devotion objects and used as an element in burialcustoms.Nearly all crosses found in the Yedi Evler church haveparallels in other regions of the Byzantine Empire and theneighboring region in the Black Sea coastland, Mediterranean,Asia Minor, Northern Caucasus and Balkans. Suchex-voto crosses illustrate a special feature of post iconoclasticculture in the beginning of the Middle Byzantine period,as well as large distribution of personal reliquary-crossesof the end of the 9th – 11th century. However, prior to becomingan ex-voto offering in church interior, both types ofcrosses were generally used in private Christian devotion.It is largely accepted that the 9th -11th century was a periodof increasing individualism, social atomism and growingemphasis on personal piety. With that in mind, individualcrosses were evidence of the new post-iconoclasm Orthodoxyas a manifestation of personal activity in church lifeand a sign of the victory of polis community tradition overimperial tyranny.The process of donating personal crosses to churchesshould be regarded as a special way of reconciling personaldevotion with the liturgical needs of the local communityencouraged by Church hierarchy. The present hypothesisis confirmed by information in the Byzantine MonasticTypikons, especially that of Empress Irene Doukaina Komnenefor the Convent of the Mother of God Kecharitomenein Constantinople founded between 1100 and 1118, whichprescribed that each Saturday laymen would offer crosses-stauria in the sanctuary for the commemoration of thedeceased, and that other crosses must be brought similarlyeach Sunday on behalf of the living who are recorded on thediptychs. Crosses from the Yedi Evler church and in othercases should be regarded as an archaeological illustration ofsuch a ritual.Other small finds from the church like nails, chain linksfor the suspension of lamps, fragment of bronze wire, leadplates from a wick holder, buttons of bronze, small greenglass beads, and an iron arrow-head characteristic of EasternEurope military culture in the 10th/11th - 13th century aredescribed and analyzed in chapter six by I. Teslenko. Twoamulet-pendants found in the church that are made of clamshell of Cerithium vulgatum and tooth of deer of Cervuselaphus, which could also be offered in the sanctuary asex-voto, are presented in chapter seven by G. Gavris and I.Teslenko.Chapters eight to twelve compiled by G. Gavris, V.Logvinenko, and S. Leonov deal with bones and faunisticremains including birds, mammals, fishes, marine mollusks,and land snails recorded during the excavations. As a result,information is exhausted on the repertoire of animal sacrifices,a normal practice in rural parish Byzantine churches,and the composition of church festive meals has been determined.Among 139 identified bones of mammals, 64% belongto Ovis aries and Capra aegagrus hircus, 16% to Sus scrofadomesticus, 6% to Lepus europaeus and 2 % to Bos Taurus.Birds are presented with 148 individuals of 19 species,including 78% of Gallus domesticus and Gallus domesticussm. and an insignificant quantity of bones of Otis tarda,Cygnus olor, Perdix perdix etc.It is quite interesting to note that fishes are nearly absentfrom the collection, and consequently, on the table of parishmen who lived along the sea coast, only 13 bones ofAcipenser gueldenstaedtii and Perciformes were recorded.Evidently, bones from the excavation present the remainsof a festive meal and not an everyday diet. However, shellfishesare recorded here in 1900 fragments of Mytilus galloprovincialis(95% of mollusk) and a small number ofPatella ulyssiponensis and Ostrea lamellose. Eriphia spinifronspresented in 4-5 individuals should also be noted. Terrestrialgastropods mollusks are mainly presented by Helixalbescens (72.4%), Monacha fruticola (24.2%) Chondrulatridens (3.2%), and only one shell of Brephulopsis cylindrical.Some remarks on the distribution of animal bonesin the excavated complex will be provided in the followingchapters.In chapter thirteen, I. Teslenko proposed and arguedthe chronology of the site based mainly on pottery analysis.Coins from the 7th – mid-8th century that were used forthe manufacturing of crosses give only large terminus postquem for the church building. Amphora with small horizontalmultiple grooves on the surface well-known in Crimeanot later than the beginning - first half of the 9th century arenot recorded among the excavation materials; so the beginningof the church complex must date back to the secondthird-middle of the 9th century. The find of the fragment of ahigh neck jug with wide flat handles in layers of the secondbuilding period, and their absence later on, puts the date ofthe rebuilding of the church at 860-880 AD. The presence oflocal “Black Sea type” amphora of the second variant andthe absence of forms similar to amphora of types I and IIbaccording to N. Günsenin allow to propose the first half –mid of the 10th century as the final stage of the church’s existenceand that of surrounding settlements. Another find isthe fragment of Glazed White Ware II, dated no earlier thanthe beginning of the 10th century. The history of the churchactually spans about 100 (± 20-25) years.Chapter fourteen by A. Musin discusses liturgical ritualspracticed in the sanctuary against the large background ofByzantine church culture and shows parallels from relatedterritories. To explain the meaning and origin of the two unequalapse church building in the Yedi Evler area, the authorprovides a thorough account of the phenomenon of doubleapse churches with unequal apses from Transcaucasia andthe Northern Caucasus through Asia Minor and the GreekIslands up until biapsidal churches were recorded in medievalItaly in the 9th-13th century. As a result, a conclusionhas been made that the Mediterranean World did not havea unique genesis of double apse churches. Late Antiquitychurches with two symmetrical naves and apses cannot beregarded as a direct prototype for the Yedi Evler church andrelated building. The architecture of Transcaucasia and theNorthern Caucasus sometimes gives similar features, forexample Mgvimevi, Georgia, the end of the 13th century,but all of them were built later than the monument underconsideration.The “pseudobiconques” churches with a reducednorthern apse are also known in medieval Italy and Corsicaof the 10th-12th century (see for example: San Venerio,La Spezia-Migliarina, Liguria; San Tommaso al Poggio,Rapallo, Liguria; Santa Maria della Chiappella, Rogliano,Haute-Corse; Santa Maria di Sibiola, Serdiana, Sardegna).However, they hardly could be a source of inspirationfor builders of the Yedi Evler church for cultural andchronological reasons. The Italian architecture of the “chiesebiabsidate” did, however, deeply influence the appearanceof two apse churches in Crimea and Muscovite Russia inthe end of the 14th-15th century. Nevertheless, early Italiantwo apse sanctuaries, especially with different apses and anadditional northern entrance, could initially reflect the sameprocess of the change of liturgical planning as in the YediEvler church.It should be acknowledged that “pseudobiconques”churches are not very characteristic for the Greek Island.Some indirect parallels can bee seen in the planning ofthe church of St Spyridon – Panagia Protothroni Halkia,Halki, Naxos Island; church of St Pantaleon, Kotraphi,Peloponnesus; church of St Athanasius, Phaturu, PatmosIsland; church of St Athanasius, Phaturu, Patmos Island. Inall cases, it is difficult to say whether the additional reducedcompartment was initially intended for this or that particularliturgical ritual. It is quite possible that both naves wereused for the Eucharist. However, in the Middle Byzantineperiod, the appearance of double churches of Sts John andGeorge, Sarakini, Samos, and the Monastery of St JohnChrysostomos at Koutsovendis, Cyprus can be attested.The double apse church was renewed in the 10th century inÜçayak, near Kirşehir, Central Anatolia, Turkey. The mostnotable fact is that the high density of two apse middlebyzantine churches, including the “pseudobiconques”sanctuary, is known to have existed in the ancient Pontprovince and near Trabzon, Turkey, for example in Koralla,Görele Burunu fortress or Gantopedin fortress (Matzouka,Zana Kale), Labra, Maçka Dere, near Köpruna Köy. Thisregion always had direct ties with the northern Black Seacoast and Crimea during Antiquity and Middle Ages.At the same time, the closest parallel to the Yedi Evlerchurch can be seen in the 10th-11th century double apsechurch in the Upper City of Middle Byzantine settlementin Boğazköy (Hattusa, Asia Minor), Turkey, excavated by P.Neve in th