Hostname: page-component-68c7f8b79f-b57wx Total loading time: 0 Render date: 2026-01-10T17:13:42.705Z Has data issue: false hasContentIssue false

The bema in the Tangchaodun Christian Church site (Xinjiang) in the context of Syriac Christianity

Published online by Cambridge University Press:  08 January 2026

Yuge Xing*
Affiliation:
School of Sociology and Anthropology, Sun Yat-sen University, Guangzhou, China
Rights & Permissions [Opens in a new window]

Abstract

This article examines the Tangchaodun bema—a liturgical structure dating to the Gaochang Uyghur Kingdom in Xinjiang—as a regional manifestation of the architectural and theological tradition of the Church of the East, shaped over centuries of transmission and adaptation. Through comparative analysis of archaeological remains and liturgical texts from Syria, Mesopotamia, and Central Asia, the study argues that the Tangchaodun bema follows the ‘eastern-type bema’ model rooted in the East Syrian tradition. Its spatial configuration and ritual function reflect established Mesopotamian patterns, particularly in the mirroring of bema and sanctuary, while also incorporating localised features shaped by visual adaptation and intercultural contact. Bilingual inscriptions and iconographic traces further attest to this integration of tradition and regional context.

Rather than existing in isolation, the Tangchaodun bema forms part of a broader historical development in East Syriac ecclesiastical architecture. By positioning the site within this extended line of transmission, the article shows how sacred space operated as a medium of both theological continuity and cultural dialogue across Asia. In so doing, it offers new perspectives on the role of Christian architecture in the Tang to Yuan Dynasties and contributes to a more integrated understanding of the Church of the East in its easternmost reaches.

Information

Type
Article
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - ND
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives licence (http://creativecommons.org/licenses/by-nc-nd/4.0), which permits non-commercial re-use, distribution, and reproduction in any medium, provided that no alterations are made and the original article is properly cited. The written permission of Cambridge University Press must be obtained prior to any commercial use and/or adaptation of the article.
Copyright
© The Author(s), 2026. Published by Cambridge University Press on behalf of The Royal Asiatic Society.

Introduction

In 2021, Renmin University of China conducted an excavation at the site of an East Syriac church (Figure 1) located north of the ruins of Tangchaodun 唐朝墩 ancient city in Qitai County, Changji Hui Autonomous Prefecture, within the Xinjiang Uyghur Autonomous Region (Map 1). Located within the central area of the nave, in the north aisle of the church, one can see a rectangular structure constructed out of adobe (Figure 2). The raised platform can be dated to the period of the Gaochang Uyghur Kingdom, between the tenth and fourteenth centuries, at which point the Tangchaodun site was abandoned.Footnote 1

Source: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 2.

Figure 1. Plan of the Tangchaodun Church site.

Source: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 3.

Figure 2. Bema in the Tangchaodun Church site (southwest–northeast).

Map 1. Ecclesiastical sites in Xinjiang Uyghur Autonomous Region, Gaochang Uyghur period.

According to the architectural terms given by Īshōʿyahb I (Figure 3)Footnote 2, also with reference to J. M. Fiey’s restoration of the church building plan of Chaldean-rite Christianity (Figure 4),Footnote 3 it can be identified that the platform is a fundamental liturgical furnishing of East Syriac Christianity, namely a ‘bema’, serving as the place at which the Liturgy of the Word, which is the first part of the Sanctification, is conducted.

Source: Harrak, Law Code of Īshōʿyahb I, fig. 1.

Figure 3. Church structure according to Īshōʿyahb I.

Source: Fiey, Mossoul Chrétienne, pl. ii

Figure 4. Plan of Chaldean Church (Légende du plan de l’ Église Chaldéo-Nestorienne).

For the detailed structure of the Tangchaodun bema, the central part of the eastern wall features a staircase that gives access to the elevated platform, with a heptagonal column on each of the south-east and north-east sides. At the symmetry point, a niche is carved inside the central part of the western side of the bema, flanked by two symmetrical columns. Located at the base of the central niche is an octagonal pedestal, and the walls on either side bear discernible remnants of mural paintings and inscriptions. The Tangchaodun bema is characterised by the presence of four shallow niches (or frames) on both its south and north walls. A further crucial characteristic is the central part of this elevated platform, which contains an arched doorway connecting the north to the south side.Footnote 4

The emergence of the Tangchaodun bema is rooted in the liturgical system of Syriac Christianity. Meanwhile, as a product of the Church of the East within the transmission route from Syria—Mesopotamia (Persian Gulf)—Central Asia—Xinjiang, its architectural form, detailing, and iconography reflect the localised adaptation of cultural exchange. This article addresses the following key issues from this perspective:

  1. 1. Within the chronology of the Syrian bema tradition, what are the typological classification and structural characteristics of the Tangchaodun bema?

  2. 2. How does the Tangchaodun bema, along with its associated liturgical furnishings, reflect the East Syrian liturgical tradition and the symbolic function of sacred space?

  3. 3. In what ways does the Tangchaodun bema embody both Syriac Christian liturgical practices and localised adaptations—architecturally and iconographically—amid the eastward transmission of the Syrian bema?

Through a comparative analysis of bemata across Syria, Mesopotamia, and Central Asia, this article seeks to demonstrate the significance of the Tangchaodun bema and to enhance the understanding of how East Syrian Christian liturgical furnishing was transmitted and adapted across cultural frontiers.

Tangchaodun bema in the Syriac Christianity tradition

Transition and evolution of bema types from Syria to Mesopotamia

As a key architectural feature in Syriac Christian churches, the bema not only coordinates with other furniture but also adheres to fundamental liturgical principles. The main focus of this section is to establish the chronology of bema-tradition development, set within the context of Syriac Christianity, across the spatial range from Syria, Mesopotamia, to Central Asia and Xinjiang, with consideration of cross-regional architectural traditions. The appearance of the Tangchaodun bema is ‘inlaid’ in the sequence and, through this, examines the localisation of East Syriac Christianity.

Classification and characteristics of the ‘north-western Syria-type bema’

In the early twentieth century, H. C. Butler’s systematic survey and documentation of churches and monasteries in northern Syria introduced the terms ‘exèdre’ (exedra) and the Syrian-style ‘ambon’ to describe the bema.Footnote 5 Building upon early fieldwork, J. Lassus’s foundational researchFootnote 6 and G. Tchalenko’s influential monographFootnote 7 played a vital role in popularising the term ‘bema’. Further research by R. TaftFootnote 8 on the attributes of a bema and its associated liturgical practices advanced the scholarly understanding of its significance.

According to Taft’s four categories of church distribution, a bema is commonly found in churches of northern Syria, particularly in the Syrian Limestone Massif. The standard form of the bema in this area is straight-sided and horseshoe-shaped, with origins traceable since the fourth-century Antioch region. In addition, updated learning suggests that similar bema forms found within the territory of Syria II should also be considered as part of the system.Footnote 9

Tchalenko mapped the distinctive features of horseshoe-shaped bemata in Syria, analysing their architectural elements.Footnote 10 Following his logic, this part proposes to uniformly categorise these forms as the ‘north-western Syria-type bema’,Footnote 11 which could be subdivided into three types: Type A (Figure 5a), characterised by a podium closest to a semicircle; and Type B (Figure 5b) and Type C (Figure 5c), which typically feature a rectangular-like main structure in the east, with a ‘bema-throne’Footnote 12 set in the semicircular west part. From the fifth to sixth centuries, these structures evolved from wood to stone, becoming standardised within the Diocese of Antioch. The Type C north-western Syria-type bema—the most complicated due to its three-part design—provides remarkable insights into the specific functions of the bema within Syrian liturgy. Moreover, as this form spread eastwards into the Syro-Mesopotamia frontier region, significant formation changes occurred, ultimately influencing the architectural styles of East Syriac Christianity, including the Tangchaodun bema in Xinjiang, as discussed later in this article.

Source: Donceel-Voûte, Les pavements, p. 154, fig. 126.

Figure 5a. Type A north-western Syria-type bema in Jerade Church.

Source: Donceel-Voûte, Les pavements, p. 241, fig. 220.

Figure 5b. Type B north-western Syria-type bema in Qalb Loze Church.

Source: Donceel-Voûte, Les pavements, p. 278, fig. 257.

Figure 5c. Type C north-western Syria-type bema in Holy Cross Church, Resafa.

Liturgical system and ‘cosmological symbolism’ in sacred space

The design of the bema aims to facilitate the weekly re-enactment of the crucifixion and resurrection through its role in the Liturgy of the Word.Footnote 13 This tradition, derived from the so-called ‘Nestorian liturgy’,Footnote 14 is compatible with the architectural layout of Antioch and seeks to express theological concepts through symbolic spatial arrangements.Footnote 15

Archaeological evidence from Syria suggests that the fifth and sixth centuries marked the maturation period for Syrian church architecture. During this time, architectural focus shifted towards integrating the sanctuary with other functional areas, leading to a distinct style that set it apart from ‘Roman models’.Footnote 16 It is notable that the semicircular apse structureFootnote 17—a distinctive feature of Syrian church architecture—became standardised during this period. Concurrently, the ‘horseshoe-shaped bema’ (Type A and Type B) had also attained full development by at least the mid fifth century.Footnote 18

Drawing on Jewish prototype cosmological symbolism, the bema, as a symbol of sacred space, contributes to the overall liturgical space system alongside other areas. The sanctuary represents heaven, the qestroma corresponds to paradise, the nave symbolises the earthly church, and the bema embodies the earthly Jerusalem.Footnote 19

Based on both archaeological evidence and historical records, it can be inferred that the core design principle of ‘bema mirrors the apse, replicating its form’ was influenced by the early Antioch Church. The close correlation between the shape of bema and apse creates a visual ‘mirroring relationship’,Footnote 20 reinforcing its liturgical function and positioning church architecture not merely as a place for religious activity but as a tool for educating believers and strengthening Christian faith.

Origin, distribution, and characteristics of the ‘eastern-type bema’

In addition to the north-western Syria-type bema, archaeological excavations since the twentieth century have revealed four rectangular/square-shaped bemata in north–central Syria (see Table 1 and Figures 6a, 6b, 6c, and 6d). Given the cultural context of Syro-Mesopotamia (central Syria), this article uses the term ‘eastern-type bema’, emphasising its geographical characteristics rather than confessional affiliation.

Source: Tchalenko, Églises syriennes, p. 247, fig. 15.

Figure 6a. Plan of Faferteen Bema Church.

Source: Harper and Wilkinson, ‘Excavations at Dibsi Faraj’, fig. D.

Figure 6b. Plan of Citadel Church at Dibsi Faraj.

Source: Donceel-Voûte, Les pavements, p. 145, fig. 117.

Figure 6c. Plan of Halawe Bema Church.

Source: Majcherek, ‘Excavating the basilicas’, fig. 6.

Figure 6d. Plan of Great Northern Basilica IV in Palmyra.

Table 1. Rectangular/square-shaped bemata found in Syro-Mesopotamia

Placing the rectangular/square-shaped bema in its historical and geographical context, it is clear to find that, except for the Faferteen Church site,Footnote 28 Dibsi Faraj and Houeidjit Halaoua bemata are within the geographical and cultural sphere of the Middle Euphrates region (see Map 2).Footnote 29 Based on the rough north-eastern limes of the Middle Euphrates—Palmyra (central Syria), the westward and eastward development of the bema diverged. In regions distant from Antioch, such as Euphratesia and Osrhoene, the eastern-type bema retained the essential symbolic element of the north-western Syria-type bema—that is, the so-called ‘bema-throne’—but transformed the horseshoe shape into a rectangular form.

Source: Leone and Sarantis, ‘Middle Euphrates’, fig. 1.

Map. 2. Square/rectangular-shaped bemata in the Middle Euphrates region.

Since antiquity, the Syro-Mesopotamia region, in which the eastern-type bemata are distributed, has been a borderline between Rome and Persia.Footnote 30 The religious architecture in this region blends Sasanian artistic traditions with native Semitic styles, creating a second Christian language parallel to the West Syrian tradition.Footnote 31 Persian influence is evident in church buildings from Tur ‘AbdinFootnote 32 to the Iraq–Persian Gulf.Footnote 33 Although there is no direct archaeological evidence of the developmental sequence of the square/rectangular-shaped bema in Syro-Mesopotamia, as a key liturgical furnishing centre located in a church layout, the form of the eastern-type bema may also had been permeated by Persian style—particularly the ideal preference for a square outline in core areas, drawing from both secular buildings (such as royal palaces) and sacred structures (such as Zoroastrianism fire temples).Footnote 34

Evolution and spread of the eastern-type bema tradition

Building upon the distribution, structural features, and cultural factors of the eastern-type bema, tracing its evolution path on the Church of the East reveals a key liturgical and architectural intersection: the Type C north-western Syrian-type bema, exemplified by the Basilica of the Holy Cross (Basilica A) in Resafa (Figure 7). Resafa—also known as Sergiopolis, named after Saint SergiusFootnote 35—offers critical evidence. Its bema includes an underground chamber for housing relics and a central ciborium for their ritual display during festivals.Footnote 36 Similar relic-veneration features appear in other Type C bema sites, such as Seleucia Pieria and Qausiyeh (the martyrium of St Babylas).Footnote 37

Source: Bogisch, ‘Qalat Seman and Resafa/Sergiupolis’, fig. 2. Original source: Resafa, Church of the Holy Cross, plan of the pilgrimage complex (after Ulbert, Die Basilika des Heiligen Kreuzesin Resafa-Sergiupolis, pl. lxxx, 1).

Figure 7. Plan of Holy Cross Church in Resafa.

Uniquely, the Basilica of the Holy Cross also retains both the synthronon in the apse and bema (Cathedra), making it the only known site at which these two liturgical devices coexist. This dual presence provides compelling evidence for transformation of the Antiochene bema tradition during the sixth to seventh centuriesFootnote 38—preserved, notably in Resafa, geographically distant from the Antiochene core region.

By the sixth century, the Resafa ecclesiastical complex had become a major cross-cultural and inter-religious pilgrimage centre, drawing worshippers from Byzantine, Sasanian Persian, and Arab communities.Footnote 39 In this context, the Syrian bema tradition—diverse in its architectural expressions (namely the rectangular-shaped Type C north-western Syria-type bema or the rectangular eastern-type bema) incorporating Persian architectural features—likely spread eastwards along the pilgrimage route of palmyra–Resafa–Disib Faraj to Mesopotamia.Footnote 40

Mesopotamia as a convergence area: regional variations in the bema and the related liturgical system

From the Sasanian period onwards, Christian traditions in Mesopotamia, including the Church of the East and Miaphysite Christianity, flourished under a confluence of cultural and theological interaction (see Map 3a and Map 3b).Footnote 41 Currently, three known cases of a bema belonging to the East Syrian tradition have been confirmed (see Table 2 and Figures 8a, 8b, 8c, and 8d). Analysing the similarities and differences in bema churches between northern and southern Mesopotamia provides a significant perspective into how architectural form and its associated liturgies were localised, and how they ultimately extended eastwards to Central Asia and Xinjiang.

Source: Narmen, ‘Monastic church of Bāzyān’, plan 1.

Figure 8a. Plan of Bazyan bema church.

Source: Harrak, ‘Christian archaeological sites in Iraq’, p. 36, fig. 2.

Figure 8b. Plan of Qalʿat Šīla bema church.

Source: Rice, ‘Oxford excavations’, p. 280, fig. 1.

Figure 8c. Plan of Hira Church XI.

Source: Hunter, ‘Christian matrix of al-Hira’, pl. iii.

Figure 8d. Bet-šqaqone in Hira Church XI.

Map 3a. Syriac Christianity sites in northern Mesopotamia. Source: Amin Ali and Brelaud, ‘Churches’ building in Northern Iraq, fig. 1.

Source: Briquel-Chatonnet and Debié, Syriac World, p. 116, ‘Areas of Syriac influence in the South: Arabia, the Arab-Persian Gulf, Ethiopia (5th–8th centuries)’.

Map 3b. Syriac Christianity sites in southern Mesopotamia.

Table 2. East Syrian tradition bema sites in Mesopotamia

The shape and spatial organisation of bema churches across northern and southern Mesopotamia reveals regionally distinctive patterns rooted in hybrid indigenous contexts.

Bema churches in northern Mesopotamia exhibit a ‘distinct borderland identity’.Footnote 42 In addition to common characteristics found throughout Mesopotamia, regions closer to the Byzantine–Persian border incorporate more pronounced ‘Western elements’ in their church designs.Footnote 43 The bemata at Bazyan and Qalʿat Shīlā, for instance, conform to the Type A north-western Syria-type bema, shaped by the centre the of Diocese of Antioch–Edessa–Tur ‘Abdin–Mesopotamia ecclesiastical expansion network.Footnote 44 The Gola Bema site in Tur ‘AbdinFootnote 45 further supports the diffusion of this Syrian tradition. However, church layouts in this region frequently diverge from the Antiochene formula of ‘bema mirroring the apse’. Instead, they feature rectangular sanctuaries paired with semicircular or horseshoe-shaped bemata, echoing Persian architectural conventions.Footnote 46 As for the southern counterpart, Hira Church (Mound XI) in the Persian heartland,Footnote 47 with its eastern-type bema and near-square (cruciform-shaped) sanctuary, more clearly demonstrates the normative East Syrian theology.

Nevertheless, it is noteworthy that, beyond the shape of the bema, a more significant divergence occurred in how the bema-related liturgical office/service and ‘sacred context’ formed in northern and southern Mesopotamia.

According to East Syrian ecclesiastical law, the bema is restricted to the cathedral-based service and absent from monastic rituals.Footnote 48 This aligns with archaeological findings from northern Syria, which consistently reflect this usage. Yet, in northern Mesopotamia, ecclesiastical complexes increasingly blurred the boundaries between two sets of liturgies. For example, the Bazyan bema, situated within a monastic church, allowed lay participation in both parish and monastic services,Footnote 49 which is evidenced by architectural modifications for pilgrimage, such as the prominent apse that may have functioned as a Bēt sāhdē—that is, a martyrium housing relicsFootnote 50—suggesting that Bazyan’s saint-veneration function paralleled that of Resafa. Despite adherence to early traditions, the East Syrian cathedral office may have occasionally ‘retreated to monasteries’ as did its counterpart Syrian Orthodox in northern Mesopotamia to preserve its liturgical heritage amid periods of intense inter-religious conflict involving Muslim and Christian groups.Footnote 51

In contrast, the southern heartland of the Church of the East maintained a clearer distinction between parish and monastic liturgy systems. The Hira XI Bema Church exemplifies the former, while the latter is represented by monastic churches built in the seventh/eighth to ninth centuries (such as those at Kharg in Iran,Footnote 57 Al-Qusur in Kuwait,Footnote 58 and Ain sha’ia in IraqFootnote 59) that omitted bemata and qestroma due to simplified hierarchy and liturgical practices.Footnote 60 These churches, through their baptistery and reliquary, as well as spatial design of central and side aisles, incorporated a ‘pilgrimage flow design’ that allowed pilgrimage activities to occur without disrupting the monastic daily routine.Footnote 61

Despite local variation, one feature exhibits remarkable consistency across Mesopotamia: the bet-šqaqone—a processional passage linking the bema to the sanctuary, symbolising a bridge between the heavenly and earthly realms.Footnote 62 Its substantial origins lie in the ambo with the solea of Byzantine liturgical architecture in Greek-speaking Syria. In Mesopotamian churches, this symbolism was localised through the construction of ‘walled pathways’Footnote 63 that materialised the heavenly–earthly transition. Not limited to the Church of the East, even East Syrian Jacobites of the Maphrianate of Tikrit incorporated the bet-šqaqone into their churches, underscoring its Mesopotamian regional consistency. Although textual reference to the bet-šqaqone appears only in the eleventh-century Kitab al-Muršid by Yahya ibn Garir,Footnote 64 its architectural manifestations suggest a much earlier origin and indicate a distinct Mesopotamian liturgical system diverging from northern Syrian models. A representative example is Hira XI Church (Figure 8d), in which 12 arched wall niches along the bet-šqaqone may have once contained statues of the Apostles,Footnote 65 visually enacting the liturgical passage from Earth to Heaven. This symbolic arrangement provides a crucial context for interpreting similar iconographic programmes, such as those found in the Tangchaodun bema.

With the gradual strengthening of the influence of East Syriac Christianity in Central Asia since the sixth to seventh centuries,Footnote 66 these architectural and liturgical models, particularly the eastern-type bema enhanced in southern Mesopotamia, the concept of the bet-šqaqone, and pilgrimage–monastic integration, were transmitted eastwards from the Persian Gulf into Central Asia. These features played a critical role in shaping the principles of the East Syriac church building in regions such as Central Asia and Xinjiang, China.

The eastward transmission of the East Syrian bema tradition: from Mesopotamia to Central Asia (Xinjiang)

Controversial evidence (Urgut site) in Central Asia and the bema-related liturgical system

In post-Sasanian Central Asia, three undisputed East Syriac Christian sites have been identified: two at Ak-Beshim (Figure 9)Footnote 67 and one at Urgut (Figure 10).Footnote 68 This section focuses on the Urgut site and its purported bema structure, examining how Syrian tradition spread to the Christian buildings in Central Asia and potentially impacted the Xinjiang region.

Source: Savchenko, Roads and Kingdoms, p. 200, fig. 104.

Figure 9. Plan of Ak-Beshim ‘Building VIII’.

Source: Ashurov, ‘Sogdian Christianity’, p. 134, fig. 1.

Figure 10. Ground plan of Urgut Church.

Excavator A. Savchenko initially proposed that the Urgut ecclesiastical complex features two aisles divided by a central raised platform that was accessible through an aperture in the western wall, possibly linked to a now-lost mud brick or loess stairway (Figure 11).Footnote 69 B. Ashurov, building on this, argued that the ‘low narrow passage’ connecting the bema to the near-square cross-shaped sanctuary is a ‘substantial structure’ associated with the Mesopotamian bema—specifically the bet-šqaqone.Footnote 70

Source: Savchenko, Roads and Kingdoms, p. 37, fig. 18.

Figure 11. View over the monastic and parish churches, facing east.

However, in 2022, Savchenko revised his interpretation, suggesting that the Urgut site comprises three main parts from the north to the south: the refectory, monastic church, and parish church (Figure 12).Footnote 71 The so-called ‘central raised platform’ appears to function as a division between the latter two building sections. Savchenko also noted niches on the southern wall of the monastic church, which housed oil lamps—indicating that the platform is part of the solid wall rather than an independent bema structure.Footnote 72 More importantly, the small room to the east of the platform, connected by steps to the sanctuary of the monastic church, is probably an attached vestryFootnote 73 that does not form a mirrored structural pair with the previously mentioned platform. Based on this, combined with the layout of Ak-Beshim Building VIII,Footnote 74 the author also agrees that the raised platform is presumably NOT a ‘ritual structure’ such as the bema.

Source: Savchenko, Roads and Kingdoms, p. 47, fig. 24.

Figure 12. Idealised plan of the monastery building.

In the Urgut ecclesiastical complex, notable features include the narrow passage connecting the nave and the sanctuary, as well as the spatial organisation within the sanctuary (see Figure 13).Footnote 75 The better-preserved monastic church presents a spatial division between the ‘altar at the eastern (front) end’ and the ‘platform at the western (rear) end’. The altar area serves as the core space for the Eucharistic celebration, evidenced by the priest’s seat and symmetrical ledges raised above the floor for the placement of liturgical implements. The platform, with ‘brick pillars probably serving as two lecterns’, suggests a daily function for scripture reading.Footnote 76 This architectural pattern is representative of East Syrian churches in a monastic compound, in which ‘monastic liturgy’ takes place entirely within the sanctuary, as seen in the Kharg monastery church (Figure 14), where its western part functions as an alternative to the bema.Footnote 77 Simultaneously, the parish church to the south is accessible to laity and is connected to the monastic church via a narrow corridor ‘to bring a relic of a saint from the monastic church to the parish church on the yearly feast of Remembrance without leaving the building’.Footnote 78 A similar layout design is found at the southernmost ‘parish church’ (Building A) of the Ak-Beshim VIII site, where the southern aisle also provides an efficient path for pilgrims wishing to attend shrines and altars.Footnote 79

Source: Savchenko, Roads and Kingdoms, p. 53, fig. 31, and p. 52, fig. 29.

Figure 13. Spatial division and structures in a monastic church and a parish church (top: altar; bottom: corridor space connecting the nave and the sanctuary).

Source: Steve, L’île de Kharg, fig. 12.

Figure 14. Spatial division of the sanctuary in a Kharg monastic church.

Inferring from the above evidence, it could be presumed that, in ecclesiastical complexes of Central Asia that serve both parish and monastic functions, the use of a bema follows a pattern that is similar to that of East Syriac Christianity in southern Mesopotamia and the Persian Gulf. Inscriptions found near the Urgut site indicate that Christians from the heartland of Mesopotamia and Turkic-speaking migrants from the East settled here,Footnote 80 suggesting that the region not only adopted the Syriac–Mesopotamian ‘Martyrs’ Cult’,Footnote 81 but also developed into a regional pilgrimage centre. While Christian communities in Central Asia during the Mongol era included other churches from the Syriac milieu, such as the Syrian Orthodox church and Melkite church,Footnote 82 archaeological evidence manifests that the East Syrian bema tradition and its associated liturgical system remained dominant. Through the network-connecting Samarkand, Ak-Beshim, and even as far as Turfan in Xinjiang, the relative explicit influence of southern Mesopotamia also reached the Tangchaodun bema church during the Gaochang Uyghur period, which will be discussed in detail later.

Tangchaodun bema in the ‘bema development chronology’

In summary, the origin of the bema can be traced to the centre of the Antioch Diocese. As various Syrian Christian sects and their related liturgies evolved, the bema gradually spread eastwards and developed in various forms, as shown in Table 3.

Table 3. Chronology of bema development from Syria and Mesopotamia to Central Asia (Xinjiang)

The north-western Syria-type bema of the Antiochene tradition in northern Syria and the eastern-type bema of the East Syrian tradition in southern Mesopotamia are the undisputed representatives of these two liturgical systems. However, within the multicultural context of Syro-Mesopotamia, there is no strict ‘correspondence’ between the shape of the bema, its liturgical function, and the relative Christian denominations. On the other hand, Mesopotamia developed distinct architectural features related to the bema and a dual-purpose functional service system for both monks and laity. This regionality was also integrated into the East Syrian tradition, profoundly influencing the transmission of the bema tradition eastwards.

Considering the temporal and spatial evolution described above, the Tangchaodun bema can be placed within the ‘East Syrian tradition’ milieu of Central Asia (Xinjiang). Its form and church layout design distinguish it from the ‘monastic system’ seen at Urgut and Ak-Beshim. The Tangchaodun bema aligns with the East Syrian theological concept of ‘the bema mirroring the apse’ and can be compared to the bema (church) at Hira. Unlike the isolated monastery located on the outskirts of the capital of Gaochang Uyghur Kingdom in Turfan,Footnote 83 the primary function of Tangchaodun Church was as a parish church serving the laity. Additionally, the church site in Qocho ancient city was excavated and briefly documented by A. von Le Coq; the detailed internal architectural features were not preserved, but its function and attributes should be consistent with those of Tangchaodun Church.Footnote 84 (For the locations of the three sites, see Map 1.) This highlights that, at a time when the Syrian bema tradition in the Mesopotamian heartland of the Church of the East was on the brink of extinction, the distant East preserved this ancient tradition.

Sacred contexts, architectural structures, and liturgical functions of the Tangchaodun bema

Layout of Tangchaodun Church and the bema-centred liturgical process

The Expositio officiorum ecclesiae (Commentary of the Ecclesiastical Services) is the primary work that provides detailed information on the religious symbolism of church architecture and liturgical practices associated with the bema. The book was authored during the early to mid ninth century by a resident of Mesopotamia who was devoted to East Syriac Christianity. Its purpose was to address inquiries regarding the liturgical innovations implemented by East Syriac Catholicos Īshōʿyahb III.Footnote 86 The works of various East Syrian theologians, including Gabriel Qatraya (and also Abraham Bar Lipheh),Footnote 87 from the sixth century onwards also contain material related to this theme.

When observing the layout of Tangchaodun Church from the perspective of the cosmological symbolism of the Antiochene tradition,Footnote 88 the structure of the ‘Heaven (sanctuary)–Paradise (qestroma)–Earth (nave)’ is clearly revealed. Based on the ‘whole world’ conception, as shown in Figure 15, this article attempts to restore the ritual process related to ‘sanctuary–bema’ (see Table 4)Footnote 89 within the church space.

Sources: left: Fiey, Mossoul Chrétienne, pl. ii, fig. 4; right: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 2 (see Figure 1).

Figure 15. Illustration of ‘whole world’ in Tangchaodun Church.

Table 4. Bema-related liturgy in Tangchaodun Church and corresponding symbolism

The liturgy takes place on the bema, fostering tighter interactions between the clergy and the laity. As E. Loosley notes, bema-related liturgy has a very clear purpose: ‘for those who could not undertake a pilgrimage to Jerusalem the clergy enacted a weekly ceremony that symbolically drew a map of the world for the faithful.’Footnote 90 Referring to Canon XV from the Synod of Mar Isaac held at Seleucia-Ctesiphon in 410 CE,Footnote 91 ‘liturgy in the midst of people’ dates back to the beginning of the fifth century, just before the initial eastward spread of East Syriac Christianity; this idea is also preserved in church buildings from Central Asia to Xinjiang. In conjunction with the liturgical ceremony, the bema and its imagery served as powerful media for disseminating the faith and informing believers.

Liturgical furnishing on the platform of the Tangchaodun bema

The central one-third of the Tangchaodun bema platform exhibits a slight concave shape, with a preserved notch located in the middle of the eaves on both the south and north sides of the platform. The layout can thus be divided into three distinct parts: (a) east, (b) centre, and (c) west, each of which will be examined in detail. Considering the role of the bema in the ritual process as described in the written sources, this section analyses the liturgical furnishing structure of the Tangchaodun bema plan in relation to the aforementioned eastern-type bema, which shows a relatively complete structural preservation in comparison with the architectural details of the Resafa bema, which is significant in the transformation of the bema pattern.

Theoretically, the Tangchaodun bema is similar to the Hira bema, based on standard East Syrian tradition (see Figure 16).

Sources: left: Rice, ‘Oxford excavations’, fig. 1; right: drawn by the author.

Figure 16. Three-part division of the Hira and Tangchaodun bemata.

Part a: The easternmost part contains the ascending steps up to the bema platform and comprises two pulpits: one located in the south-east corner and the other in the north-east corner. The bema of the East Syrian tradition consists of two parts: one for recitation of the Old Testament and the other for the New Testament. The preaching is oriented towards the east, with all parties facing that direction during prayers and Eucharistic liturgy.Footnote 94

Part b: The central area is the heart of the bema and representative of the centre of earthly Jerusalem, where tradition records that the tomb of Adam was found. This area of the bema, which is used to hold the Gospel and the processional cross, symbolises Golgotha.Footnote 95 The Gospel and cross echo the sanctuary symbolising the Resurrection and Ascension of the Lord, reinforcing in another form the ‘mirroring relationship’ between the bema and the sanctuary in East Syrian theology. The archaeological remains of columns in bemata in Syria suggest the possibility of a corresponding ciborium above the bema.Footnote 96 Regarding the Tangchaodun bema, there is no evidence that it was accompanied by a fixed ciborium.

Part c: As the Expositio officiorum ecclesiae notes, the bishop’s throne is located at the western end of the bema, facing east towards the sanctuary. To the left of the bishop’s throne is the seat of the archdeacon.Footnote 97 The Hira and Bazyan bemata show the evident remnants of a cathedra. By inference, the Tangchaodun bema should be similar to those cases.

In addition to the aforementioned significant liturgical furnishings, it is also important to mention the seats for the concelebrating priests. This is a designated area, known as the space for 12 men (symbolising the Apostles) that is connected to the bishop’s throne.Footnote 98 It is associated with the upper room and the Last Supper. Both the East and West Syrian traditions agree concerning this space, on the basis of written sources and archaeological evidence.Footnote 99

Based on the typical north-western Syria-type bema and the case associated with the Church of the East, we can infer that, apart from the bishop and archdeacon positioned at the west end and readers situated in the two pulpits at the east end, the remaining clergy likely sat on the south and north sides of the bema. This is supported by evidence from the Hira bema as well. P. Donceel-Voûte examined the arrangement of bemata in churches in Farfenteen and Dibsi Faraj, which belong to the Antiochene tradition: In Farfenteen, the bema is laid out with a ‘west bema-throne/pulpit + bench seats on the north and south sides’.Footnote 100 In the similar case in Dibsi Faraj, the layout is enhanced with addition of a ciborium in the centre and a small vestibule at the east end (Figure 17).

Sources: left: Tchalenko, Églises syriennes, p. 279, fig. 52; right: Rice, ‘Oxford excavations’, p. 280, fig. 1.

Figure 17. Seat arrangements of the Faferteen and Hira bemata.

The article will now examine the Tangchaodun bema in more depth. Donceel-Voûte noted of the Dibsi Faraj bema that its seats were approximately one metre in width—greater than the width of other bema seats that include a backrest.Footnote 101 By comparison, the breadth of the eaves on the platform of the Tangchaodun bema is comparable to the width of the bench seats found in the south aisle of a church. These eaves measure approximately 0.3 to 0.4 metres in width and 0.3 to 0.5 metres in height. Nevertheless, there is a substantial difference in the height between the eaves and the bench seats, and there is no evidence of any permanent or supported wooden structures, such as the backrest of a chair. Despite the extensive damage and deterioration over the centuries, the adobe-rammed eaves of the Tangchaodun bema do not have brick borders that the synthronon in the Bazyan bema has. Presumably, the Tangchaodun bema platform only allows the placement of individual, portable wooden seats. With reference to the Syro-Mesopotamia bemata, the seats for the clergy, including the bishop’s throne, were primarily situated in the western part, as the eastern and central areas were already properly organised. Therefore, excluding the bordering section, the Tangchaodun bema measures approximately five to six square metres, providing space for approximately four or five individuals to sit. Based on the way in which the bema-related liturgy is carried out, priests, deacons, and vigilers have the opportunity to alternate their presence on the bema and engage in the ceremony. Through this approach, the available space can be efficiently utilised to convey the significance of the cosmological system to the faithful through the liturgy.

Considering the administrative level of Tangchaodun ancient city,Footnote 102 though we lack sufficient evidence to know for sure, it seems likely that Tangchaodun Church was a parish church, not a cathedral with a bishop, but it is reasonable to assume that the local bishop (perhaps in Qocho?) would have visited Tangchaodun from time to time. As Loosley notes, the West Syriac Ordo quo episcopus (The order for a bishop) highlights the great importance of the bema for the Liturgy of the Word and the instructional elements of the service, particularly when bishops from nearby areas attended.Footnote 103 According to the Expositio officiorum ecclesiae Footnote 104 and assuming that Tangchaodun Church was indeed a parish church and not a cathedral, the liturgy would normally have been presided over by a local priest(s).

Stairs and height of the Tangchaodun bema

The Tangchaodun bema contains a staircase that provides access to the platform, located in the centre of the east side. The staircase is around 0.7 metres wide and consists of five steps, which gives the Tangchaodun bema a height of two metres. This is in contrast to the bemata in Mesopotamia, in which there are typically two or three steps, comparable to the number of stairs leading up to the sanctuary,Footnote 105 which aims to create a symbolic theological system that corresponds to the concepts of the earthly Jerusalem and Paradise–Heaven by making the height of the bema nearly identical to that of the altar in the apse.

Beyond its practical function, the ‘staircase’ of the Tangchaodun bema may carry additional symbolic significance. Unlike typical East Syrian bema churches in Mesopotamia, Tangchaodun Church lacks any trace of an independent bet-šqaqone—that is, a ‘wall-flanked passage’. In East Syrian theology, the bet-šqaqone is an abstract concept often associated with Jacob’s ladder—an ‘elevated’ motif that may parallel the symbolic role of the staircase.Footnote 106 Thus, the staircase not only serves the liturgical function of the Tangchaodun bema but also subtly contains the invisible conception of a ‘bridge connecting Heaven and Earth’ and distances it from the distinct architectural identity of Mesopotamian churches, adapting it to the Xinjiang region.

Following the Expositio officiorum ecclesiae, Fiey highlighted the fact that, apart from the main staircase, the construction of the bema may have included two other sets of steps for those who chant the zummārā (a psalm or canticle between the epistle and the Gospel). This was done to prevent them from using the same stairway as those who read the epistle.Footnote 107 The Resafa bema is an exception to this rule: there are stairs located on the south and north sides near the western synthronon. It is possible that this is an ancient tradition that was maintained in Mesopotamia. The Tangchaodun bema does not have these extra steps, and clergy participating in bema-related liturgy would use the primary stairway located on the east side of the bema; no further steps were required, possibly due to the ecclesiastical rank of Tangchaodun Church.

There is also a difference in the height between Tangchaodun and Syro-Mesopotamian bemata. Tchalenko stated that the typical height of bemata in Limestone Massif in north-eastern Syria is approximately 0.5 metres.Footnote 108 The height of the bench, including its backrest, measures around 1.5 metres. The same can be observed about the bemata associated with the Church of the East. Based on the photographs in the excavation report of the Bazyan site, the height of the bema is no more than 0.4 to 0.5 metres—about knee-level for an average adult.Footnote 109 There is less evidence available regarding the bemata at Hira. In contrast, the Tangchaodun bema has a platform height that is much greater than those of the typical Syrian or Mesopotamian bemata. The Byzantine ambo’s influence is evident in the bema’s distinctive structure, which features a central arched doorway connecting north and south; see the discussion in the following part.

The probable association of the Tangchaodun bema with the ambo and cultural influence from the West

This article has analysed the regional development chronology and related liturgical pattern of the Tangchaodun bema, demonstrating its significant East Syrian traditional features. Most importantly, its core function still aligns with Chaldean-rite Christianity norms. However, aside from this, by examining the development and spread of the ambo and its associated religious concepts, and comparing the structural features that are similar to the ambo in the Tangchaodun bema, this article offers valuable insights into the study of East Syriac Christianity during the Gaochang Uyghur period.

In the sixth to seventh centuries, in the province of Syria Prima, where Greek-speaking tradition and Syrian tradition overlapped, the Syrian bema gradually evolved into the Byzantine ambo.Footnote 110 In the Syro-Mesopotamian region, the interaction between Greek-speaking and Syriac-speaking traditions became more frequent. For instance, the Sogitha on the Church at Edessa (sixth century) defines the bema as being more similar to the Greek ambo. It describes the bema (ambo) as based on the model of the Upper Room at Zion, supported by 11 columns, symbolising the ‘hidden eleven apostles’.Footnote 111 Although archaeological evidence may not fully support this ‘idealised model’,Footnote 112 the architectural pattern of a ‘circular platform supported by several columns’ had widespread influence in the Syro-Mesopotamian region.Footnote 113 In fact, even in the very heartland of Byzantine tradition, the ‘religious conceptual model’ of the ambo closely resembles this form. For example, in On the Divine Liturgy by St. Germanus of Constantinople (eighth century), the ambo is defined as a hill placed on a flat and level area.Footnote 114

However, while such ambos have the key structural feature of a hollow interior, the Tangchaodun bema’s feature of an arched doorway connecting the north side to the south side remains distinct. Therefore, this feature should be examined within the broader context of the ambo development system. J. Davies analysed ambo remains in the Byzantine empire’s core region (Asia Minor-Greece), outlining three developmental stages for the ambo:Footnote 115 generally speaking, the standard Byzantine ambo began in the fifth century as a ‘monolithic Ambo with a single flight of steps’, evolving into a ‘semicircular structure with two short steps leading up from the same side to a height-increasing platform’ in around the sixth century. Notably, the third stage, during or after the sixth century, was characterised by the ‘high platform’ and ‘hollow under the ambo-platform’, with two flights of stairs on both the east and west sides.Footnote 116 The high-platform ambo became the most common ambo form along the eastern Mediterranean coast after the sixth century and even continued through the middle–late Byzantine period.Footnote 117

How, then, should we interpret the unusual ‘Byzantine architectural feature’ in the Tangchaodun bema from the perspective of religious and cultural background and transmission pathways?

From the perspective of the Christian sacred liturgical setting, in the sixth to seventh centuries, while the East Syrian tradition did not explicitly adopt the Byzantine ambo form in Syro-Mesopotamia and beyond, the region had clearly been influenced by the ‘ambo–solea’ feature. Therefore, it can be inferred that the ambo and its religious symbolism—‘(The herald) ascending the mountain (Ambo) to announce the good news of Christ’s Resurrection’Footnote 118—would have been known to the followers of the Church of the East, even probably as part of the eastern-type bema’s transmission. At least they would not have explicitly rejected its associated religious meanings. From the seventh to eighth centuries onwards, after the ‘unified Islamization’ of the Syro-Mesopotamian to Syria-Anatolia region, elements of Syrian and Byzantine Christian architecture began to blend, and this influence spread eastwards once again to Xinjiang.

Other ‘foreign elements’ in the Tangchaodun bema and its ecclesiastical complex can also serve as evidence of this process. For example, the inscription on the upper part of the eastern side of the Tangchaodun bema is a mirror image (Figure 18)—that is, both sides are the same word, reading ‘rk’wn (Alaph, Resh, Kaph, Alaph, Ayin, Waw, Nun), transliterated as ärkä‘ün, which is the Uyghur word meaning ‘Christ’,Footnote 119 in which the ‘Alaph’ is written in the West Syrian Serto script. This inscription shows the blending of eastern Estrangela and western Serto features.Footnote 120 The ‘mirror inscription’ form may have developed as a decorative calligraphic style in Syriac and Arabic during the Abbasid period or later.Footnote 121 Additionally, glazed pottery (a pot or vase) found at Tangchaodun Church features Arabic script in repetitive and symmetrical forms (Figure 19),Footnote 122 reflecting the influence of the Western world that prevailed in the Gaochang Uyghur Kingdom.

Source: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 4.

Figure 18. East side of the Tangchaodun bema, with inscriptions and murals.

Source: Ren and Wei, ‘2022 nian Tangchaodun gucheng’, fig. 9.

Figure 19. Glazed pottery with inscription unearthed in Tangchaodun Church site.

From a broader perspective, the cultural context of Tangchaodun also includes the discovery of a Romanesque bath site on the eastern side of Tangchaodun Church,Footnote 123 which further serves as evidence of Roman–Byzantine architectural and cultural influences. C. Ge suggests that the Tangchaodun city, since its establishment during the Tang Dynasty, likely functioned as a post-station settlement for merchants and pilgrims, possibly part of a post-station network along the Silk Road.Footnote 124 Although no parallel bema/ambo church sites have been found in Central Asia and Xinjiang, the earlier discovery of the Urgut ecclesiastical complex proves the influence of Christian traditions from West Asia, suggesting that the Tangchaodun bema was probably part of this systematised network.

In conclusion, the basic attribute of the Tangchaodun bema is indisputably the eastern-type bema belonging to the Church of the East. However, the appearance of a multicultural fusion ambo-like structure in far-off Xinjiang, distant from the heartland of East Syriac Christianity, does require further archaeological discoveries to verify its regional development chronology. Nonetheless, East Syrian theology had already undergone a process of borrowing and theoretical fusion with the Greek-speaking tradition by the fifth to sixth centuries.Footnote 125 Thus, it is reasonable to infer that the incorporation of ambo-like architectural features into bema building during the Gaochang Uyghur Kingdom manifests a subtle reflection of the historical trend.

A preliminary interpretation of the iconography of the Tangchaodun bema

The excavation team at Tangchaodun have found traces of wall paintings on both the east and west sides of the bema. The paintings on the north part of the Tangchaodun bema’s east wall depict a figure on a red horse or donkey. Furthermore, in the centre of the west side of bema is a niche, about one metre wide and 0.3 metres deep, flanked by two symmetrical decorative columns on either side, all of which is built on top of a rectangular earthen platform (Figure 20). An octagonal pedestal can be seen at the bottom of the centre of the niche, approximately 0.45 metres in diameter. There are traces of white plaster and mural paintings on the wall surface, but these are covered by plaster and damaged so seriously that it is impossible to recognise the specific original content of the mural paintings.Footnote 126

Source: Ge, ‘Yi lu si cheng’, fig. 4.

Figure 20. The western wall of the Tangchaodun bema (looking eastwards).

Drawing on bema-related liturgy, its symbolic significance, and the religious traditions rooted in its Syro-Mesopotamian origins and expansion, as well as the development of the Church of the East in the Gaochang Uyghur Kingdom, this section analyses archaeological and textual evidence to demonstrate that the Tangchaodun bema comprises two interrelated iconography–narrative systems. These systems complement each other, deepening the bema’s religious significance and playing a vital role in shaping the sacred space of Tangchaodun Church.

Iconography system I: substantial depiction of Holy Week

The first iconography system of the Tangchaodun bema likely represents Holy Week. Clues to this system appear on the eastern and western walls of the bema.

Drawing on iconographic parallels from Qocho ChurchFootnote 127 and the Xipang Monastery site (Figure 21), which were built and developed simultaneously, the mural on the east wall—depicting Christ wearing a cross-adorned crown and holding a cross sceptre (Figure 22)—suggests a motif of the ‘Entry of Jesus into Jerusalem’. The inscription ärkä‘ün (see Figure 18) at the top of the eastern side of the bema corresponds to the black-ink inscription of the Master Yelikewen (Figure 23) on the south side of the north-eastern pillar, reinforcing Christ’s identity and linking the bema structure to his life. This scene also parallels the symbolic significance of the liturgical step, where the bishop and clergy exit the sanctuary and ascend the bema. Just as Christ’s earthly journey was accompanied by the Twelve Apostles, the niche statue carved in the bet-šqaqone of Hira Church embodies this concept.Footnote 128 Although the typical Mesopotamia-style bet-šqaqone is absent in Tangchaodun Church, it is reasonable to infer that, following standard East Syrian tradition, the east wall of the bema—directly facing the sanctuary—depicts the beginning of Holy Week, symbolising Christ’s earthly experiences.

Source: Liu et al., ‘2021 nian Xinjiang Tulufan Xipang Jingjiao siyuan’, fig. 4.

Figure 21. Head of Christ wearing a crown adorned with a cross.

Source: Drawn by the author.

Figure 22. Line drawing of Christ’s entry into Jerusalem, on the east side of the Tangchaodun bema.

Source: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 8.

Figure 23. Inscription of Master Yelikewen in Uyghur script.

The imagery on the Tangchaodun bema’s western wall follows a ‘sculpture-in-central-niche and paintings-on-both-sides’ arrangement. Based on the Syrian tradition and Eastern Christian art, this wall likely draws a Traditio Legis-like motif, depicting Peter and Paul as representatives of the Apostles proclaiming Christ’s law and the Gospel to the nations from Jerusalem.Footnote 129 This theme aligns with the symbolic meaning of the bema as earthly Jerusalem.

A common element in Traditio Legis scenes is a row of lambs/sheep beneath the Apostles flanking Christ, sometimes with Christ accompanied by two lambs. These elements likely symbolise believers from all nations.Footnote 130 Similar symbolic representations of the Apostles and believers appear in Syrian bema sites. For example, the bema mosaic at Uqayribat features a palm tree, representing the Tree of Life, alongside three sets of heraldic lambs (Figure 24).Footnote 131 Given that the fruit of life was promised by the Tree of Life,Footnote 132 this mosaic likely allegorises the apostolic adoration of Christ. Likewise, the two lambs depicted in the Rayan bema mosaic (Figure 25) probably hold comparable religious significance.Footnote 133

Source: Giovanna, ‘Geological materials in late antique archaeology’, fig. 11.

Figure 24. Bema mosaic at Uqayribat: palm tree and heraldic lambs.

Source: Donceel-Voûte, Les pavements, fig. 242.

Figure 25. Bema mosaic at Rayan: vase and heraldic lambs.

Another significant case of archaeological and iconographic evidence in the Syriac tradition is the Dedication (folio 14r) from the miniatures of the Rabbula Gospels (586 CE),Footnote 134 likely produced in the region between Antioch and Apamea; this manuscript provides vital evidence from which to infer that the prototype of the Traditio Legis originated in the Syriac region.Footnote 135 Beyond the standard connotations of the Traditio Legis in Syrian bemata iconography, what is more noteworthy is its localised adaptations.

The Dedication depicted an enthroned Christ with two monks in the monastery at Beth Zagba, offering their dedications to Him, under the sponsorship of two indigenous patron saints.Footnote 136 Early Christian imagery was rarely a mere illustration of biblical texts, but instead carried symbolic and allegorical significance. Thus, while the Traditio Legis serves as a compositional reference, the murals flanking the Christ sculpture in the Tangchaodun bema may not necessarily depict Peter and Paul; the alternative could feature saints venerated in the Church of the East or even monks who lived at Tangchaodun, aligning with the cult of saints, which is further explored in ‘Iconography system II’ of the Tangchaodun bema.

Beyond explicit iconographic evidence, the Hetoimasia motifFootnote 137 in early Christian art—particularly the representations of Golgotha and the altar on the bema—serves as an indispensable part in constructing Holy Week. The Hetoimasia—the Throne of Christ’s Sacrifice—symbolises the Holy Liturgies that sanctify believers and prepare them for the Last Judgement. Structurally, the bema itself can be seen as an ‘empty throne’, representing Christ’s presence through the Gospel and the cross, as confirmed by bema screens and mosaic inscriptions found in Syria.Footnote 138 At the practical level of a real building context, the ‘empty throne’ corresponds to the episcopal throne, which, in the absence of the bishop, serves as a perpetual reminder of his authority.Footnote 139 Within the architectural framework of the bema, all imagery ultimately reinforces the supreme authority of Christ and theoretically, by extension, his representative, the bishop.

Above all, considering the theological symbolism in Syriac Christianity and the doctrinal principles of the Church of the EastFootnote 140—rooted in the School of Nisibis and favouring the theologia prima of liturgy over academic theological reflectionFootnote 141—this article seeks to reconstruct the iconographic and liturgical furnishing system of the Tangchaodun bema and to explain how it visually elaborates Holy Week (Table 5).

Table 5. Substantial depiction of Holy Week in the Tangchaodun bema

Iconography system II: tradition of saints’ veneration

The second iconography system of the Tangchaodun bema consists of images of the south and north sides. This structure exhibits a symmetrical design, featuring a total of eight enclosed frames that are recessed inwardly by approximately two centimetresFootnote 143 (see Figure 26). Additionally, some of these frames still bear inscriptions in ink along their edges. The prevalence of Buddhist grotto art in the Gaochang Uyghur KingdomFootnote 144 suggests that such niches typically housed seated or standing figures. Given their shallow depth—unlike the main sculpture on the west wall—the images within were likely flat or rendered in low relief. In Buddhist art, figures with inscriptions in frames often represent donors.Footnote 145 By analogy, in the localised Christian context, they were presumably saints or martyrs subordinate to the Lord God.

Source: Ge, ‘Yi lu si cheng’, fig. 5.

Figure 26. The southern wall of the Tangchaodun bema (four niches) (looking northwards).

As noted earlier, the north-western Syria-type bema in northern Syria (Antioch) was closely associated with the tradition of saint/martyr worship from the fourth to fifth centuries. Archaeological evidence indicates that this tradition persisted and even intensified in the eastern-type bema of Syro-Mesopotamia from the sixth century onwards. The canon of the Church of the East explicitly requires bishops to preach on Sundays, feast days, and saint commemorations as well,Footnote 146 underscoring the significance of saint veneration in the East Syrian tradition.

As East Syriac Christianity spread from Sasanian Persia to Central Asia, this tradition remained strong, at least through the tenth to thirteenth centuries. For inferences about the identity and attributes of the characters, reference is made to Syriac Christian literature unearthed in Turfan from the ninth to tenth centuries and onwards: Christian figures revered by Christians in the Gaochang Uyghur Kingdom, including the saints George, Sergius, Bacchus, Cyriacus, and Julitta, were adored not only in Central Asia but also in Western Asia and Mesopotamia, and in Greek and Latin Christendom. In addition to these widely recognised figures, Central Asia has its own local saints who played a significant role in introducing Christianity to the region, namely Mar Barshabba, the Persian queen Mart Shir, and Zarvandokht.Footnote 147 The northern and southern sides of the Tangchaodun bema may have been adorned with depictions of some of these saints. It is plausible that the north and south sides of the Tangchaodun bema were adorned with depictions of some of these saints. In a religiously pluralistic Gaochang Uyghur Kingdom, in which Christianity coexisted with Buddhism and Manichaeism, images of saints and martyrs visually reinforced the Christian spirit of dedication during daily liturgical practices.

Regarding saint veneration within the liturgical context of the bema, Catholicos Patriarch Mar Awa III has analysed Syriac manuscript MIK III 45 from Turfan, containing 61 folios from the Ḥūḏrā of the Church of the East, from which he concluded that the week-long festive observance of the commemoration of the saints resembles the liturgy of the hours of the Lenten season. As Holy Week is an essential part of the Lenten season, this connection further underscores the religious significance of the bema.

In sum, the Tangchaodun bema embodies dual religious systems: (1) Holy Week (Great Lent), which prepares the faithful not only to commemorate but also to partake in the Passion and Resurrection of Christ; and (2) the Cult of Saints and Martyrs, represented by the Ceremony of the Saints. These two dimensions, mutually reinforcing and intertwined, condense multiple liturgical and theological functions into the architectural space of the bema.

Conclusion

The Tangchaodun bema stands as a compelling testament to the eastward transmission, preservation, and transformation of the East Syrian Christian liturgical tradition. Rooted in the Antiochene ecclesiastical framework and structurally aligned with the eastern-type bema, it reflects the core spatial and theological principles of the Church of the East. Its architectural form and symbolic organisation bear strong affinities with liturgical developments in Mesopotamia, especially the heartland of East Syriac Christianity, suggesting deep regional influences along the eastward movement of Syriac Christianity. The Tangchaodun bema, closely aligned with ecclesiastical complexes in Central Asia, represents a regional manifestation of the same theoretical model that structured the Church of the East’s liturgical architecture throughout its eastern expansion. While some design elements evoke distant affinities with Byzantine ecclesiastical architecture, these are better understood as formal echoes shaped by long-range cultural contact rather than as direct architectural borrowings.

Through its spatial organisation and symbolic features, the Tangchaodun bema exemplifies the cosmological logic embedded in East Syrian theology. In its structural configuration, it coheres with both the architectural hierarchy of church building and functional requirements of East Syrian liturgical practice. Meanwhile, traces of iconography and bilingual inscriptions point to a visual and textual synthesis shaped by intercultural currents, revealing how the Gaochang Uyghur context reinterpreted established Christian forms within a pluralistic religious landscape.

Far from being a passive recipient of Christian forms, the Gaochang Uyghur Kingdom, and sites such as Tangchaodun in particular, actively reinterpreted and localised the tradition of the bema. Its synthesis of the Syriac liturgical structure with regional material culture resulted in a sacred architectural language that was at once orthodox in theology and adaptive in form. In this sense, the Tangchaodun bema does not merely preserve a fading tradition from the Mesopotamian heartland but reanimates it at the cultural frontier, affirming the vitality of Christian expression in the medieval East.

Acknowledgements

I would like to express my immense gratitude to Dietmar Winkler and Li Tang for allowing me to present an initial version of this article at the 7th Salzburg International Conference: Syriac Christianity in China and Central Asia in 2023, Samarkand. The article also owes a great debt to Erica Hunter and Mark Dickens, as well as two anonymous reviewers, whose suggestions and comments helped clarify the structure and strengthen the argument. In addition, I sincerely thank Valerii Kolchenko for providing me with important archaeological materials on Christian remains in Central Asia, which have supported part of the key arguments in this article.

Conflicts of interest

None.

References

1 Ren Guan and Wei Jian, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan yizhi kaogu fajue zhuyao shouhuo’ 2021年新疆奇台唐朝墩景教寺院遗址考古发掘主要收获 [The main achievements of the 2021 archaeological excavation of the Eastern Syriac Christianity temple site in Tangchaodun City, Qitai, Xinjiang], Western Regions Studies 3 (2022), pp.  106–113.

2 A. Harrak (ed.), The Law Code of Īshōʿyahb I, Patriarch of the Church of the East (New Jersey, 2022), p. 7, fig. 1.

3 J. M. Fiey, Mossoul Chrétienne: Essai sur l’histoire, l’Archéology et l’etat Actuel des Monuments Chrétienne de la Ville de Mossoul (Beirut, 1959), pl. ii.

4 Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, pp. 107–108.

5 H. C. Butler, Early Churches in Syria: Fourth to Seventh Centuries (New York, 1929), p. 215.

6 J. Lassus and G. Tchalenko, ‘Ambons Syriens’, Cahiers Archéologiques 5 (1951), pp. 75–122.

7 G. Tchalenko, Églises syriennes à bêma (Paris, 1990).

8 R. F. Taft, ‘Some notes on the bema in the East and West Syrian traditions’, Orientalia Christiana Periodica 34 (1968), pp. 326–359.

9 J.-C. Balty, ‘Églises d’apamée et d’apamène’, Topoi, Orient-Occident 12.1 (2013), pp. 199–221.

10 Tchalenko, Églises syriennes, pp. 241–251.

11 The dating of bemata and bema churches is defined in E. Loosley, The Architecture and Liturgy of the Bema in Fourth- to Sixth-Century Syrian Churches, new edn (Leiden, 2012), ‘Appendix 3: The distribution pattern of bemata: neighbouring villages, the age of churches with bemata, the age of stone bemata’. Also, the names of all bemata mentioned in this article are based on the English spelling of the names used in Loosley’s work.

12 Generally, in the West Syrian tradition, this structure modelled after an episcopal throne emphasises both the anthropomorphism and the exalted status of the Gospel. See Taft, ‘Some notes’, p. 332.

13 E. Loosley, ‘The early Syriac liturgical drama and its architectural setting’, in Case Studies in Archaeology and World Religion, (ed.) T. Insoll (Oxford, 1999), pp. 20–21.

14 J. Lassus, ‘Liturgies nestoriennes médiévales et églises syriennes antiques’, Revue de l’histoire des religions 137.2 (1950), pp. 246–247.

15 The Antiochene-style organisation could be known in the third and fourth centuries through the Didascalia Apostolorum, the Apostolic Constitutions, and John Chrysostom. See W. Khoury and B. Riba, ‘Peut-on discerner des modèles reliés à des communités ecclésiales ou linguistiques en Syrie du Nord? (Ive-Xe siècle)’, in Les églises de la Syrie du Nord dans le contexte des luttes christologiques, (ed.) F. Briquel Chatonnet (Paris, 2013), pp. 115–117.

16 W. Khoury and B. Riba, ‘Les églises de Syrie (IVe-VIIe siècle): essai de synthèse’, in 10e Table-ronde de la Société d’études syriaques Nov 2012 (Paris, 2012), pp. 43–44.

17 ‘These sites have in common a Π-shaped sanctuary and chancel barrier, but in these churches, there is an inscribed apse that has a room to either side of it.’ See B. Mulholland, The Early Byzantine Christian Church: An Archaeological Re-Assessment of Forty-Seven Early Byzantine Basilical Church Excavations Primarily in Israel and Jordan, and Their Historical and Liturgical Context, vol. ix (Bern, 2014), p. 53. Of course, there were flank-wall sanctuary church buildings in the Syriac region, such as the fifth-century Qirkbize Church, Mezera ‘A El-‘Oulia church. See Tchalenko, Eglises syriennes, pp. 242–256, ‘Illustration of plans of church structure’. The reason could be that the early Christian buildings ‘converted from an original domestic use would possess a flat eastern wall […] perhaps the sanctuary resembles the cella of the pagan temple’; see Loosley, Architecture and Liturgy, pp. 44–45. At the same time, the form of sanctuary of the church building, which is relatively far from the centre of the town, may also reflect the local characteristics and fit in with its overall relatively simple architectural style, while adopting the ‘rectangular sanctuary’. See K. Wedad and P. Castellana, ‘Frühchristliche Städte im nordlichen Jebel Wastani (Syrien) Villes antiques chrétiennes du nord du Jebel Wastani (Syrie)’, Antike Welt 21.1 (1990), pp. 14–25, Abb. 21 Maurasras, Basilika, Grundriß. However, the above ‘architectural variation’ is not mainstream and did not affect the construction of the standard ‘Syriac traditional church architectural model’.

18 Tchalenko, Églises syriennes, p. 322, ‘Résume chronologique’.

19 S. Y. H. Jammo, ‘La structure de la messe chaldeenne du debut jusqu’a Panaphore’, Orientalia Christiana Analecta 207 (1979), pp. 191–193; P. Maniyattu, Heaven on Earth: The Theology of Liturgical Spacetime in the East Syrian Qurbana (Rome, 1995).

20 F. Briquel Chatonne, ‘Les églises dans les textes’, in Les églises en monde Syriaque (Études Syriaques 10), (ed.) F. Briquel Chatonnet (Paris, 2013), p. 30, fig. 2.

22 R. P. Harper and T. J. Wilkinson, ‘Excavations at Dibsi Faraj, northern Syria, 1972–1974: a preliminary note on the site and its monuments with an appendix’, Dumbarton Oaks Papers 29 (1975), pp. 319–338.

21 With the chronology for the pavements of the Citadel Church at Dibsi Faraj, this period should begin in the first quarter of the fifth century, and the whole process lasts half to three-quarters of a century. See P. Donceel-Voûte, Les pavements des églises Byzantines de Syrie et du Liban: décor, archéologie et liturgie, vol. ii (Louvain-la Neuve, 1988), pp. 74–76.

23 The general form of the original structure—a vast basilica with a flat apse, side aisles flanked by external porticoes, and an interior space organised into broad bays—is not a common model. As far as the form and dating of the church are concerned, given the scarcity of comparative examples in the region, it is risky to use the Faferteen bema as a decisive argument for dating the earliest phase of this building. Ibid, p. 74.

24 See G. Tchalenko, Églises de village de la Syrie du nord, Planches (Paris, 1979), vol. ii, p. 44; see also Tchalenko, Églises syriennes, pp. 41–45; Butler, Early Churches in Syria, p. 33, fig. 32.

25 Tchalenko depicts a cutaway view of a bema combined with the ground floor of the church building, and it appears that the church originally had a wooden bema before the later introduction of the stone structure; see Tchalenko, Églises syriennes, p. 297, ‘13. FAFERTIN I, milieu du 5e s’ in ‘III—Bemas à superstructure en bois élevée sur un socle de pierre posé sur le sol achevé de la nef’ and p. 301, ‘26. FAFERTIN II, milieu du 6e s’ in ‘V -Bêmas à superstrueture primitive en bois, remplacée par la suite par une superstructure en pierre’.

26 Donceel-Voûte, Les pavements, pp. 145–150.

27 G. Majcherek, ‘Excavating the basilicas’, Studia Palmyreńskie 12 (2013), pp. 254–255. Similarly to Resafa city, in addition to the Syriac tradition, the ecclesiastical complex here also shows the influence of Byzantine architectural and liturgical models. For example, in Basilica III, contemporary with Basilica IV, the remains of a Byzantine solea (‘a similar pathway leading to the ambo’) have been found; see G. Majcherek, ‘Palmyra: Season 2002’, PAM 14 (Reports 2002) (2003), pp. 279–290.

28 It seems that, in the fourth century, there were not yet strict and common architectural rules in the region of Faferteen Church. Each locality in the mountains apparently had its own tradition. Therefore, the community of Faferteen obeyed Antiochene organisation as well as being ‘capable of building the walls of its church independently’. Therefore, Faferteen bema may well be a special case of an early, unsophisticated architectural model within a small-scale Christian community and its direct connection to the Middle Euphrates area cannot be confirmed. See Tchalenko, Églises syriennes, p. 44.

29 A. Leone and A. Sarantis, ‘The Middle Euphrates and its transformation from the third to the seventh centuries: the case of Dibsi Faraj’, Journal of Late Antiquity 13.2 (2020), pp. 320–329.

30 A. Stein, ‘The Roman Limes in Syria’, Geographical Journal 87.1 (1936), p. 70.

31 E. Loosley, Architecture and Asceticism: Cultural Interaction between Syria and Georgia in Late Antiquity (Leiden, 2018), p. 110. The architecture and art of Syro-Mesopotamia reflect a fusion of Roman structural principles, Persian decorative styles, and local traditions, evident in sites such as the Sanctuary of Allat, churches in Nisibis, and Dara’s fortifications. Palmyra also exemplifies this overlap, drawing from both Western Classical and Eastern Persian traditions, showcasing the complex cultural interplay that shaped the region’s unique identity. See F. Millar, ‘Eastern Syria Phoenice: mountain, oasis and steppe: 9.4. “Palmyra”’, in The Roman Near East, 31 B.C.–A.D. 337, (ed.) F. Millar (Cambridge, 1993), pp. 319–336; B. Dignas and E. Winter, Rome and Persia in Late Antiquity: Neighbours and Rivals (Cambridge, 2007), pp.  18–49; D. Wielgosz-Rondolino and D. Tarara, ‘When architecture meets art: the case of the early architectural elements from the sanctuary of Allat’, in Life in Palmyra, Life for Palmyra: Conference in Memory of Khaled al-As’ad, Warsaw, April 21st–22nd, 2016, (eds.) M. Gawlikowski and D. Wielgosz-Rondolino (Leuven, 2022), pp. 529–540.

32 G. Bell, The Churches and Monasteries of Tur ‘Abdin (London, 1982), p. 56.

33 Y. Okada, ‘Early Christian architecture in the Iraqi southwestern desert’, al-Rafidan 12 (1991), pp. 71–83.

34 The ecclesiastical building forms of the Church of the East across the Persian empire are strongly associated with Parthian Temple/Sasanian basic Iwan architecture, in which the sanctuary was a square room similar to the throne room of the Sasanian royal palace. See O. Reuther, ‘Sasanian Christian churches’, in History in Survey of Persian Art Volume II, (eds.) A. Upham Pope and P. Ackerman (London, 1964), pp.  560–566. In addition, the prominent religious architectural form of ‘Chahārtāgh’ in Fire Temples during the Sasanian period also provides relevant comparative evidence. See A. Hozhabri, ‘The evolution of religious architecture in the Sasanian period’, SASANIKA, Archaeology 18 (2013), pp. 1–40.

35 Donceel-Voûte, Les pavements, pp. 273–279.

36 Loosley, Architecture and Liturgy, p. 30.

37 Loosley pointed out that bema churches were not ‘inextricably linked’ to martyria or reliquaries (see ibid, p. 32). But it is an undeniable fact that the cult of martyrs in Syria appears to have been very prevalent (since the fifth century). See E. Hunter, ‘An inscribed reliquary from the Middle Euphrates’, Oriens Christianus 75 (1991), pp. 157–158.

38 During the sixth to seventh centuries, the West Syrian bema in the province of Syria Prima gradually died out, spreading westwards and transitioning to the Byzantine ambo, which implied likewise a related shift in the liturgical pattern; however, in Sassanid Persia, ‘a purer form as a recognisable bema’ was retained. See Loosely, Architecture and Liturgy, p. 10; Khoury and Bertrand, ‘Les églises de Syrie (IVe-VIIe siècle)’, pp. 41–84.

39 E. K. Fowden, ‘The spread of the Sergius cult in Syria-Mesopotamia’, in ‘Sergius of Rusafa: Sacred Defense in Late Antique Syria-Mesopotamia’ (unpublished PhD dissertation, Princeton University, 1995), pp. 83–111.

40 M. Bogisch, ‘Qalat Seman and Resafa/Sergiupolis: two early Byzantine pilgrimage centres in northern Syria’, in Byzantino-Nordica 2004: Papers Presented at the International Symposium of Byzantine Studies Held on 7–11 May 2004 in Tartu, Estonia, (eds.) I. Volt and J. Päll (Tartu, 2005), pp. 52–72.

41 In the Persian empire, the Church of the East (centred in Seleucia-Ctesiphon) and West Syriac Christianity (based in Takrit and Mar Mattai) competed for influence, fostering theological debates and adapting to Sasanian and Islamic rule. This rivalry enriched the cultural and religious diversity of Christianity in Mesopotamia. See F. Briquel-Chatonnet and M. Debié, The Syriac World: In Search of a Forgotten Christianity (New Haven, 2023), pp. 140–168. For a comprehensive map of Syriac Christianity in the Roman and Sasanian periods, see https://gedsh.bethmardutho.org/Maps, ‘Map I’.

42 E. Keser-Kayaalp, ‘Boundaries of a frontier region: late antique northern Mesopotamia’, in Bordered Places, Bounded Times, Cross-Disciplinary Perspectives on Turkey, (eds.) E. L. Baysal and L. Karakatsanis (London, 2016), pp. 135–147.

43 M. Cassis, ‘Kokhe, cradle of the Church of the East: an archaeological and comparative study’, Journal of the Canadian Society for Syriac Studies 2.1 (2009), pp. 62–78.

44 Loosley, Architecture and Liturgy, p. 55.

45 See the semicircular bema in the parish church of Tur ‘Abdin Church in Gola, Mardin. The site is located at the intersection of three ancient transportation routes connecting Edessa, Amida, and Nisibis, no later than the seventh to eighth centuries. E. Keser-Kayaalp, Church Architecture of Late Antique Northern Mesopotamia (Oxford, 2021), p. 145, fig. 2.7.8.

46 An important feature of the southern Mesopotamian churches’ plan is the rectangularity of the sanctuary and its eastern annexes. See Okada, ‘Early Christian architecture’, pp. 71–83; S. J. Simpson, ‘Christians on Iraq’s desert frontier’, Al-Rāfidān 39 (2018), pp. 1–30.

47 For Hira’s significance in the history of the Church of the East, see P. Wood, ‘Al-Ḥīra and its histories’, Journal of American Oriental Society 136.4 (2016), pp. 785–799.

48 Taft, ‘Some notes’, pp. 344–356.

49 Harrak suggests a parallel situation for the Chenīsā monastery. The site at Takrit, presumably built in the sixth century and abandoned in the late twelfth century, probably belonged to the Syriac Orthodox Church. However, the form of ‘bema’ is doubtful and the possibility of a similar form to ‘ambo’ cannot be ruled out. See A. Harrak, ‘Christian archaeological sites in Iraq before and slightly after the Arab conquest’, Bulletin d’études orientales 68 (2023), pp. 52–55.

50 N. Amin Ali, S. Brelaud, V. Déroche, and J. Gaborit, ‘The fort and the church of Bāzyān, latest discoveries and new hypotheses’, Bulletin d’études orientales 68 (2023), pp. 121–141.

51 Loosley, Architecture and Liturgy, pp. 90–92.

52 M. A. A. Narmen, ‘The monastic church of Bāzyān in Iraqi Kurdistan’, Journal of the Canadian Society for Syriac Studies 8 (2008), pp. 74–84; N. Amin Ali and V. Deroche, ‘The site of Bazyan: historical and archaeological investigations’, in The Archaeology of the Kurdistan Region of Iraq and Adjacent Regions, (eds.) K. Kopanias and J. MacGinnis (Oxford, 2016), pp. 11–18.

53 The initial construction of the church (and bema) may be dated to the late sixth to seventh centuries, but, on the basis of carbon-14 dating data and analysing the artistic style of the stucco panel decoration, Lic suggests that the core of the building was in use in the eighth to ninth centuries, and that it is not possible to exclude the possibility that the bema was being gradually ‘built up to completion’ at this time. See A. Lic, ‘Chronology of stucco production in the Gulf and southern Mesopotamia in the early Islamic period’, Proceedings of the Seminar for Arabian Studies, Archaeopress 47 (2017), pp. 151–161.

54 N. Amin Ali, ‘Discovery of two new churches in the north of Iraq: a church with a bema and a martyrion’, Journal of the Canadian Society for Syriac Studies 19.1 (2019), pp. 52–58.

55 Adiabene, also historically known as Nod Ardashirakan or Hadyab in the Sasanian empire, was located in the province of Erbil, modern-day Iraq, where the Church of the East had established a dense and evenly distributed episcopal system by the early fifth century CE. See P. Wood, ‘Christians in Umayyad Iraq: decentralization and expansion (600–750)’, in Power, Patronage, and Memory in Early Islam: Perspectives on Umayyad Elites, (eds.) A. George and A. Marsham (Oxford, 2017), pp. 255–274.

56 D. T. Rice, ‘The Oxford excavations at Hira, 1931’, Antiquity 6.23 (1932), pp. 276–291.

57 M.-J. Steve, L’île de Kharg: une page de l’histoire du Golfe Persique et du monachisme oriental, Civilisations du Proche-Orient, Série I, Archéologie et environnement, vol. i (Neuchâtel, 2003).

58 V. Bernard and J. F. Salles, ‘Discovery of a Christian church at al-Qusur, Failaka (Kuwait)’, Proceedings of the Seminar for Arabian Studies 21 (1991), pp. 7–21.

59 H. Fujii, ‘Excavations at Ain Sha’ia ruins and Dukakin caves’, al-Rafidan 10 (1989), pp. 27–88.

60 M. Cassis, ‘The bema in the East Syriac church in light of new archaeological evidence’, Hugoye: Journal of Syriac Studies 5.2 (2002), pp. 195–211.

61 J. Bonnéric, ‘Des églises monastiques de pèlerinage? Organisation spatiale et analyse fonctionnelle des églises à murs-piliers du Bas-Irak et du golfe Arabo-Persique à l’époque umayyade et au début de la période Abbasside’, Bulletin d’études orientales (2023), pp. 95–119.

62 Loosley, Architecture and Liturgy, p. 9.

63 Besides the Syrian bema, the Byzantine ambo-like structure is also present in northern Mesopotamia, such as the circle platform with nine supporting columns in Tell Ḥasaka, Khabur River bank in eastern Syria (circa late sixth to seventh centuries). See W. Khoury and B. Riba, ‘Appendice: l’église du Tell de Hassaké’, in 10e Table-ronde de la Société d’études syriaques, pp. 85–106.

64 Taft, ‘Some notes’, pp. 353–355.

65 E. C. D. Hunter, ‘The Christian matrix of al-Hira’, in Les Controverses des Chrétiens dans l’Iran Sassanide, (ed.) C. Jullien (Paris, 2008), pp. 41–56.

66 A. Mingana, ‘The early spread of Christianity in Central Asia and the Far East: a new document’, Bulletin of the John Rylands Library 9.2 (1925), pp. 297–371.

67 A. Savchenko and M. Dickens, ‘Prester John’s realm: new light on Christianity between Merv and Turfan’, in The Christian Heritage of Iraq: Collected Papers from the Christianity of Iraq I–V Seminar Days, (ed.) E. C. D. Hunter (Piscataway, NJ, 2009), p. 125.

68 See A. Savchenko, ‘Excavations 2005: brief report’, https://www.exploration-eurasia.com/inhalt_english/frameset_projekt_2.html. The excavation team date it provisionally to around the eighth to thirteenth centuries based on the pottery recovered.

69 Ibid.

70 B. Ashurov, ‘“Sogdian Christianity”: evidence from architecture and material culture’, Journal of the Royal Asiatic Society 29.1 (2019), pp. 133–135.

71 A. Savchenko, Roads and Kingdoms: Two Encounters with the Nazarenes Beyond the River (Leiden, 2022), vol. xxxii, p. 47.

72 Ibid, p. 50, fig. 28.

73 Ibid, p. 51, ‘5.1.3.1.4 Vestry’.

74 Ak-Beshim Building VIII ‘partitioned or juxtaposed, double or triple church layout’ (see Figure 9—the most comparable configuration is characterised by the ‘connecting space’ between buildings B and C) resembles Urgut Church and the dates of construction (not earlier than the tenth century) almost coincide, so the two could be seen as cross-references. See W. Khoury, ‘Churches in Syriac space: architectural and liturgical context and development’, in The Syriac World, (ed.) D. King (New York, 2018), p. 545.

75 Savchenko, Roads and Kingdoms, pp. 48–57, figs 29, 30, 35.

76 Ibid, p. 51.

77 Steve, L’île de Kharg, p. 97.

78 Savchenko, Roads and Kingdoms, p. 56.

79 Ibid, pp. 199–201; see also W. Klein, ‘A newly excavated church of Syriac Christianity along the Silk Road in Kyrghyzstan’, Journal of Eastern Christian Studies 56.1 (2004), pp. 25–47.

80 M. Dickens, ‘Syriac inscriptions near Urgut, Uzbekistan’, Studia Iranica 46.2 (2017), pp. 205–260.

81 Since the sixth century, the ‘martyrs’ cult’ of Sasanian Persia has been gradually expanding from the metropolitan areas of Seleucia-Ctesiphon to the countryside and the eastern periphery. See R. Payne, ‘The emergence of martyrs’ shrines in late antique Iran: conflict, consensus, and communal institutions’, in An Age of Saints? Power, Conflict and Dissent in Early Medieval Christianity, (eds.) P. Sarris, M. Dal Santo, and P. Booth (Leiden, 2011), pp. 89–113; S. Brelaud, ‘Commémorer et vénérer les martyrs de Perse: le témoignage des passions syriaques’, Bulletin d’études orientales (2023), pp. 207–225.

82 K. Parry, ‘Byzantine-rite Christians (Melkites) in Central Asia in late antiquity and the Middle Ages’, Modern Greek Studies, Australia and New Zealand 16 (2012), pp. 91–108.

83 Liu Wensuo, Wang Zexiang, and Wang Long, ‘2021 nian Xinjiang Tulufan Xipang Jingjiao siyuan yizhi kaogu fajue de zhuyao shouhuo yu chubu renshi’ 2021年新疆吐鲁番西旁景教寺院遗址考古发掘的主要收获与初步认识, Western Regions Studies 1 (2022), pp. 74–80.

84 A. von Le Coq, Chotscho: Koeniglich Preussische Turfan-Expeditionen (Berlin, 1913), p. 7, ‘Wandbilder Aus Einem Christlichen Tempel, Chotscho’.

85 R. Taft, ‘On the use of the bema in the East-Syrian liturgy’, Eastern Churches Review 3 (1970), p. 38.

86 J. M. Fiey, ‘Išo‘yaw le Grand: Vie du catholicos nestorien Iso‘yaw III d’Adiabene’, Orientalia Christiana Periodica 36 (1970), p. 7.

87 Abraham bar Lipeh is likely a seventh-century relative of Gabriel Qaṭraya and authored a liturgical commentary that is largely an abridgement of Gabriel’s work. See: ‘Abraham bar Lipeh (7th cent.)’ in Gorgias Encyclopedic Dictionary of the Syriac Heritage: https://gedsh.bethmardutho.org/Abraham-bar-Lipeh. Also see: M. Kozah, ‘Abraham Qaṭraya bar Lipah’s commentary on the liturgical offices’, in An Anthology of Syriac Writers from Qatar in the Seventh Century, (eds.) M. Kozah, A. Abu-Husayn, S. Shaheen Al-Murikhi, and H. Al Thani (Piscataway, NJ, 2015), pp. 97–147.

88 For the liturgy steps and corresponding symbolism, see Lassus and Tchalenko, ‘Ambons Syriens’, p. 91; K. E. McVey, ‘The domed church as microcosm: literary roots of an architectural symbol’, Dumbarton Oaks Papers 37 (1983), p. 103; Loosley, Architecture and Liturgy, pp. 89–90.

89 See Jammo, ‘Structure de la messe chaldeenne du debut jusqu’a Panaphore’, pp. 191–193; Taft, ‘Some notes’, pp. 335–337; D. Hickley, ‘The ambo in early liturgical planning: a study with special reference to the Syrian Bema', Heythrop Journal 7.4 (1966), pp. 90–99.

90 Loosley, ‘Early Syriac liturgical drama’, p. 19.

91 ‘Tout diacre qui se trouve dans la ville, n’étant pas gravement malade, doit se trouver dans le temps de la prière, parmi le peuple, au rang des diacres’ (Every deacon who is in the city, not being seriously ill, must be found in the time of prayer among the people, in the rank of deacons); see J. B. Chabot (ed. and trans.), Synodicon orientale ou recueil des synodes nestoriens (Paris, 1902), p. 28, [267].

92 The two deacons are called Michael and Gabriel, after the archangels. The one who returns from sanctuary to direct people during the anaphora is called Michael; see Taft, ‘On the use of the bema’, p. 35.

93 Ibid, pp. 34–35.

94 According to the Expositio officiorum ecclesiae, Taft explained that the priest descends to place the Gospel on the golgotha after the reading, symbolising the crucifixion. This analysis suggests that the two platforms should be positioned above the bema platform or a similar elevated structure, such as a bookcase; see Taft, ‘Some notes’, p. 335.

95 Eastman pointed out that Gabriel Qatraya sees the Gospel and the cross as significant in the interactions between the space of the sanctuary and bema, symbolising ‘Christ’s rational soul’ and ‘the body of Christ that was crucified’, respectively; see D. Eastman, ‘How to read the Gospel(s): reading and ritual in late ancient Syria’, Journal of Theological Studies 70.1 (2019), pp. 271–316.

96 Unlike the East Syrian tradition, the West Syrian tradition does not have a golgotha altar on the bema. However, the Gospel is recited beneath the ciborium. Archaeological evidence indicates the presence of ciboria on moderately large bema structures. As an illustration, the tiniest bema at Qirq Bizeh measures approximately 3.7 metres × 3.5 metres and only preserves the throne at the end. The bema at Behyo, on the other hand, has dimensions of about 5.75 metres × 4.25 metres, with the ciborium positioned in the middle of the bema. The scale of the Behyo bema closely resembles that of the Tangchaodun bema, allowing for some degree of reference to the layout and proportions of the Behyo bema. See Lassus and Tchalenko, Les Exèdres du Gebel el ‘Ala, in ‘Ambons Syriens’, pp. 96–108.

97 Taft, ‘On the use of the bema’, p. 30.

98 Loosley, Architecture and Liturgy, p. 116.

99 In the East Syrian tradition, for example, we read in the Commentary on the Liturgy by Gabriel of Beth Qaṭraye: ‘The sitting down of the priests on the Bema at the time of the Readings is for a demonstration of what our Lord said to his Apostles’; see S. P. Brock, ‘Gabriel of Qatar’s commentary on the liturgy’, Hugoye: Journal of Syriac Studies 6.1 (2010), p. 208. In the West Syriac tradition, ‘[f]or the reading of the gospel in the centre of the bema, all the priests and deacons mount the bema with the priest who is to read the gospel and surround him like the disciples around the Lord’. The current West Syriac texts were largely derived from the Jacobites, described above in ‘Kitāb al-murshid’ by Yaḥyā ibn Jarīr (1058–1104 CE). See G. Khouri-Sarkis, ‘Le “Livre de Guide” de Yaḥyā ibn Jarīr’, Orient Syrien 12 (1967), pp. 303–354.

100 Donceel-Voûte, Les pavements, p. 514, fig. 460.

101 Ibid, p. 521.

102 Tangchaodun city was the seat of Pulei County 蒲类县, which was built from the early Tang Dynasty (640 CE) and was used during the Gaochang Uyghur Kingdom, the Western Liao Dynasty, and the Mongol-Yuan period, and was probably abandoned during the Chagatai Khanate in the fourteenth century. See G. Ren and T. Rong, ‘Archaeological achievements and preliminary study of Tangchaodun ancient city in Xinjiang Qitai (新疆奇台县唐朝墩古城遗址考古收获与初步认识)’, Western Regions Studies 1 (2019), pp. 142–145.

103 Although this text is West Syriac in origin, it seems to show influence from the Church of the East. See Loosley, Architecture and Liturgy, pp. 72–73.

104 Synod of Mar Isaac (410 CE): Canon XV: ‘Que l’eveque ait dans sa ville un archidiacre special […] de proclamation des diacres, il lira l’evangile’ (Let the bishop have a special archdeacon in his city […] of the deacons’ proclamation, he will read the gospel). ‘Quand l’eveque est absent, l’archidiacre designera le pretre qui doit offrir l’oblation; si l’eveque est present, il peut designer qui il veut.—De meme pours les diacres’ (When the bishop is absent, the archdeacon will designate the priest who is to offer the oblation; if the bishop is present, he may designate whomever he wishes. The same applies to deacons); see Chabot, Synodicon orientale, p. 28, [267–268]. Synod of Mar Isho’yahb I (585 CE): Canon II: ‘L’archidiacre designe celui qui doit consacrer; si le president de l’office n’est pas present, celui qui remplace l’archidiacre, ou celui qui officie au pupitre, designe, avec la permission de l’eveque, celui qui doit consacrer’ (The archdeacon designates who is to consecrate; if the president of the office is not present, the one who replaces the archdeacon, or the one who officiates at the lectern, designates, with the bishop’s permission, the one who is to consecrate); see ibid, p. 170, [430].

105 Harrak, ‘Christian archaeological sites in Iraq’, p. 66.

106 Loosley, Architecture and Liturgy, p. 89.

107 Fiey, Mossoul Chrétienne, p. 77.

108 C. Mango, ‘Églises syriennes à bêma (Inst. Français d’Archéologie du Proche-Orient, Bibl. archéologique et historique, t. CV.)’, Journal of Theological Studies 43.1 (1992), pp. 263–266.

109 Narmen, ‘Monastic church of Bāzyān’, p. 74, fig. 2.

110 Loosley points out that the Ambon tradition of the Greek-speaking areas of the Byzantine church (Asia Minor, Syria Prima II, Arabia) and the bemata tradition of the Syriac-speaking Church of the East (northern Syria, Edessa, Tur ‘Abdin, and Mesopotamia) began to diverge in the sixth century and even the former may have replaced the latter. For example, in the churches of Ba‘udeh and Bafetin within Prima I, traces of a bema can be discerned under the rounded outline of the ambo. See Loosley, Architecture and Liturgy, pp. 28–38.

111 The Sogitha on the Church at Edessa (the sixth century) provides a comparable account of the bema (similar to the Greek ambo). According to The Sogitha, the bema (ambo), inspired by the Upper Room at Zion, is upheld by 11 pillars symbolising the ‘11 apostles who were hidden’. See K. McVey, ‘The Sogitha on the church of Edessa in the context of other early Greek and Syriac hymns for the consecration of church buildings’, ARAM Society for Syro-Mesopotamian Studies 5 (1993), pp. 329–370.

112 For example, the ambo (early seventh century) in the basilical church of Aghios (Deir ‘Ain’ Abata site in Jordan), located in the north-east corner of the nave. The ambo is connected to the chancel by steps on the east side, with seven columns supporting the upper platform. See K. D. Politis, ‘The monastery of Aghios lot at Deir ‘Ain ‘Abata in Jordan’, Byzanz–das Römerreich im Mittelalter, Mainz: Verlag des Römisch-Germanischen Zentralmuseums (2010), pp. 1–24, fig. 14.

113 In northern Mesopotamia, a ‘bema with bet-šqaqone’ coexists with an ‘ambo with solea’. The correspondence between their architectural and liturgical patterns remains controversial, and there are scholarly opinions that refer to them collectively as ‘bemata’, which is considered to be merely a difference in architectural form. See N. Amin Ali and S. Brelaud, ‘Churches’ building in northern Iraq from the late Sasanian to the early Islamic period’, in From the Tigris to the Ebro Church and Monastery Building Under Early Islam, (eds.) M. de los Ángeles Utrero Agudo and S. Pierre (Madrid, 2024), pp. 127–149.

114 A. Kazhdan, ‘A note on the “Middle-Byzantine” ambo’, Byzantion 57.2 (1987), pp. 422–426.

115 J. G. Davies, The Origin and Development of Early Christian Church Architecture (New York, 1953), p. 95.

116 Ibid, pp. 95–96.

117 A well-known visual example of this fully developed form appears in the fresco ‘The Elevation of the Cross’ (Church of the Mother of God, Monastery Gračanica, circa 1321), where the ambo is depicted as a high raised platform with a hollow space beneath and two opposing flights of stairs. See J. Bogdanović, The Framing of Sacred Space: The Canopy and the Byzantine Church (Oxford, 2017), p. 75, fig. 2.18.

118 St Germanus of Constantinople, (trans.) P. Meyendorff, On the Divine Liturgy (New York, 1984), p. 63.

119 F. Briquel-Chatonnet, ‘Writing Syriac: manuscripts and inscriptions’, in The Syriac World, (ed.) D. King (London, 2019), pp. 257–258.

120 L. van Rompay, ‘Les inscriptions syriaquesdu couventdes syriens (Wadi al-Natrum, Égypte)’, in Les inscriptions syriaques, Études Syriaques 1, (eds.) F. Briquel-Chatonnet, M. Debié, and A. Desreumaux (Paris, 2004), pp. 66–67.

121 Ren Guan and Wei Jian, ‘2022 nian Tangchaodun gucheng yizhi kaogu gongzuo de zhuyao shouhuo’ 2022年唐朝墩古城遗址考古工作的主要收获 [The main achievements from the 2022 archaeological works in the Tangchaodun ancient city site of the Tang Dynasty], Western Regions Studies 2 (2023), pp. 132–138, fig. 9: F15–3:1.

122 See Ren Guan and Wei Jian, ‘Tangchaodun gucheng yuchang yizhi de faxian yu chubu yanjiu’ 唐朝墩古城浴场遗址的发现与初步研究 [The discovery of and preliminary study on bath ruins found at the Tangchaodun site], Western Regions Studies 2 (2020), pp. 58–68.

123 C. Ge, ‘Zuo lu si cheng: Xinjiang Qitai Tangchaodun gucheng kaogu de xin renshi’ 驿路寺城:新疆奇台唐朝墩古城考古的新认识 [Post-road and temple city: new insights into the archaeology of the Tangchaodun ancient city in Qitai, Xinjiang], 世界宗教研究/Studies in World Religions 1 (2024), pp. 83–89, 129–132.

124 Having examined the Semitic and Hellenic influences on East Syrian theology, Brock categorises the evolution of Christian theology in Syria into three distinct phases: prior to the fifth century, the theological expression of Syriac Christianity was not influenced by Greek elements. However, in the fifth and sixth centuries, Greek and Syriac elements were combined in a creative manner, resulting in the Hellenisation of the theological system of Syriac Christianity. Syriac Christian writings from the sixth and seventh centuries exhibited a notable inclination towards Hellenistic philosophy. See S. P. Brock, ‘From antagonism to assimilation: Syriac attitudes to Greek learning’, in East of Byzantium: Syria and Armenia in the Formative Period, (eds.) N. G. Garsoïan, T. F. Mathews, R. W. Thomson (Washington, DC, 1982), pp. 17–34.

125 Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, pp. 107–108.

126 ‘On the north wall of the eastern hall, the faint figure of a horseman could still be discerned (though it could not be captured in a photograph), carrying over his right shoulder a banner pole ending in a cross'; see von Le Coq, Chotscho, p. 7. The complete restored image can be seen in P. Y. Saeki 佐伯好郎, Shina kirisutokyō no kenkyū 支那基督教の研究, vol. i (Tokyo: Shunjū-sha shōhaku-kan, 1943), p. 475.

127 Hunter, ‘Christian matrix of al-Hira’, pl. iii.

128 R. Couzin, The Traditio Legis: Anatomy of an Image (Oxford, 2015), pp. 1–4.

129 R. Hvalvik, ‘Christ proclaiming his law to the Apostles: the Traditio Legis-motif in early Christian art and literature’, in The New Testament and Early Christian Literature in Greco-Roman Context (Leiden, 2010), pp. 434–435.

130 See B. Giovanna, ‘Geological materials in late antique archaeology: the lithic lectern throne of the Christian Syrian churches’, Heritage 4.3 (2021), pp. 1883–1898, fig. 11.

131 ‘Hymns on Paradise VI, 7–8 & Hymns on Virginity XVI,10’, in M. C. Aphrem, Symbols of the Cross in the Writings of the Early Syriac Fathers (Piscataway, NJ, 2004), pp. 45–46.

132 See Donceel-Voûte, Les pavements, p. 267, fig. 242.

134 O. M. Dalton, Byzantine Art and Archaeology (Oxford, 1911), p. 664.

135 D. H. Wright, ‘The date and arrangement of the illustrations in the Rabbula Gospels’, Dumbarton Oaks Papers 27 (1973), pp. 197–208.

136 For example, the ‘Hetoimasia’ unearthed from a temple in the East Mediterranean (likely from Syria, the fifth to sixth centuries) is probably a depiction of the bema. See Donceel-Voûte, Les pavements, p. 515, fig. 461. An empty throne with cushion and crux gemmata, flanked by Saints Peter and Paul, can also be seen in several mosaics from Ravenna and Rome. See Ş. Mera, ‘Hetimasia’s throne’, Anastasis Research in Medieval Culture and Art 1.1 (2014), pp. 134–160.

137 Example 1: a Greco-Syriac inscription found in the ruins of a bema in Zebed (the fifth century): Ada Raboula made this throne [thronos]; see Butler, Early Churches in Syria, p. 217, fig. 217. Example 2: a mosaic panel of the central nave of Taybbat al Imam Church (442–447 CE) near Hama: Theodose praying to God with (his) wife and sons paved with mosaic (the area) behind the throne; see A. R. Zaqzuq and M. Piccirillo, ‘The mosaic floor of the Church of the Holy Martyrs at Tayibat al-Imam-Hamah, in central Syria’, Liber annuus /Studium Biblicum Franciscanum 49 (1999), p. 452.

138 ‘The altar and the bishop’s chair or cathedra are presented as symbols of the promised salvation of the worthy Christian.’ A. J. Wharton, ‘Ritual and reconstructed meaning: the neonian baptistery in Ravenna’, Art Bulletin 69.3 (1987), p. 365.

139 Aphrem, Symbols of the Cross, p. 3.

140 B. A. Mattam, ‘Historical setting of the East Syriac theology’, in East Syriac Theology: An Introduction, (ed.) P. Maniyattu (Satna, 2007), pp. 27–28.

141 It can be inferred that the iconography on the south part of the east side of Tangchaodun bema may echo the theme of ‘Palm Sunday’ by depicting the ‘Ascension’ (e.g. folio 13v of the Rabbula Gospels) to form the beginning and end of Holy Week.

142 Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, p. 108.

143 E.g. Bezeklik Grottoes in Turfan, which was initially built in the late North and South Dynasties, and has been the centre of Buddhism in the Gaochang for seven centuries through the Tang to Yuan Dynasties. During the most prosperous period of the grotto complex, it was once the temple of the royal family of Gaochang Uyghur Kingdom, with most of the surviving caves dated to around the tenth to thirteenth centuries. See E. Zeren, ‘The material cultural heritage of Buddhist Turks’, International Journal of Emotions, Expressions and Dimensions (JEED) 1.2 (2021), pp. 64–76, figs 13–16; see also ‘Bezeklik Caves’, in Wiki Encyclopedia, https://en.wikipedia.org/wiki/Bezeklik_Caves.

144 For donor paintings in Bezeklik Grottoes, see Ma Zhenlin 马振林, ‘Baizikelike shiku Gaochang Huihu shiqi gongyang ren tuxiang yanjiu’ 柏孜克里克石窟高昌回鹘时期供养人图像研究 [A study on donor paintings in the Bezeklik Grottoes during the Gaochang Uyghur Period] (unpublished master’s dissertation, Xinjiang Arts University, 2020). In addition, similar donor paintings with inscriptions in frame have been found in Buddhist temple sites of the same period in Tangchaodun ancient city. See Ge, ‘Yi lu si cheng’, fig. 15.

145 The Synod of Catholicos Patriarch George I (676 CE): Canon I: L’évéque préchera les dimanches, les jours de fetes et les jours de commémoraison des saints (The bishop will preach on Sundays, feast days, and saints’ days); see Chabot, Synodicon orientale, p. 217, [482].

146 E. Hunter, ‘Commemorating the Saints at Turfan’, in Winds of Jingjiao: Studies on Syriac Christianity in China and Central Asia, (eds.) L. Tang and D. W. Winkler (Vienna, 2016), pp. 89–104.

147 Mar Awa III (Royel), ‘From Mosul to Turfan: the ḥūḏrā in the liturgy of the Assyrian Church of the East: a survey of its historical development and its liturgical anomalies at Turfan’, Ex Fonte—Journal of Ecumenical Studies in Liturgy 1 (2022), p. 49.

Figure 0

Figure 1. Plan of the Tangchaodun Church site.

Source: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 2.
Figure 1

Figure 2. Bema in the Tangchaodun Church site (southwest–northeast).

Source: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 3.
Figure 2

Map 1. Ecclesiastical sites in Xinjiang Uyghur Autonomous Region, Gaochang Uyghur period.

Figure 3

Figure 3. Church structure according to Īshōʿyahb I.

Source: Harrak, Law Code of Īshōʿyahb I, fig. 1.
Figure 4

Figure 4. Plan of Chaldean Church (Légende du plan de l’ Église Chaldéo-Nestorienne).

Source: Fiey, Mossoul Chrétienne, pl. ii
Figure 5

Figure 5a. Type A north-western Syria-type bema in Jerade Church.

Source: Donceel-Voûte, Les pavements, p. 154, fig. 126.
Figure 6

Figure 5b. Type B north-western Syria-type bema in Qalb Loze Church.

Source: Donceel-Voûte, Les pavements, p. 241, fig. 220.
Figure 7

Figure 5c. Type C north-western Syria-type bema in Holy Cross Church, Resafa.

Source: Donceel-Voûte, Les pavements, p. 278, fig. 257.
Figure 8

Figure 6a. Plan of Faferteen Bema Church.

Source: Tchalenko, Églises syriennes, p. 247, fig. 15.
Figure 9

Figure 6b. Plan of Citadel Church at Dibsi Faraj.

Source: Harper and Wilkinson, ‘Excavations at Dibsi Faraj’, fig. D.
Figure 10

Figure 6c. Plan of Halawe Bema Church.

Source: Donceel-Voûte, Les pavements, p. 145, fig. 117.
Figure 11

Figure 6d. Plan of Great Northern Basilica IV in Palmyra.

Source: Majcherek, ‘Excavating the basilicas’, fig. 6.
Figure 12

Table 1. Rectangular/square-shaped bemata found in Syro-Mesopotamia

Figure 13

Map. 2. Square/rectangular-shaped bemata in the Middle Euphrates region.

Source: Leone and Sarantis, ‘Middle Euphrates’, fig. 1.
Figure 14

Figure 7. Plan of Holy Cross Church in Resafa.

Source: Bogisch, ‘Qalat Seman and Resafa/Sergiupolis’, fig. 2. Original source: Resafa, Church of the Holy Cross, plan of the pilgrimage complex (after Ulbert, Die Basilika des Heiligen Kreuzesin Resafa-Sergiupolis, pl. lxxx, 1).
Figure 15

Figure 8a. Plan of Bazyan bema church.

Source: Narmen, ‘Monastic church of Bāzyān’, plan 1.
Figure 16

Figure 8b. Plan of Qalʿat Šīla bema church.

Source: Harrak, ‘Christian archaeological sites in Iraq’, p. 36, fig. 2.
Figure 17

Figure 8c. Plan of Hira Church XI.

Source: Rice, ‘Oxford excavations’, p. 280, fig. 1.
Figure 18

Figure 8d. Bet-šqaqone in Hira Church XI.

Source: Hunter, ‘Christian matrix of al-Hira’, pl. iii.
Figure 19

Map 3a. Syriac Christianity sites in northern Mesopotamia. Source: Amin Ali and Brelaud, ‘Churches’ building in Northern Iraq, fig. 1.

Figure 20

Map 3b. Syriac Christianity sites in southern Mesopotamia.

Source: Briquel-Chatonnet and Debié, Syriac World, p. 116, ‘Areas of Syriac influence in the South: Arabia, the Arab-Persian Gulf, Ethiopia (5th–8th centuries)’.
Figure 21

Table 2. East Syrian tradition bema sites in Mesopotamia

Figure 22

Figure 9. Plan of Ak-Beshim ‘Building VIII’.

Source: Savchenko, Roads and Kingdoms, p. 200, fig. 104.
Figure 23

Figure 10. Ground plan of Urgut Church.

Source: Ashurov, ‘Sogdian Christianity’, p. 134, fig. 1.
Figure 24

Figure 11. View over the monastic and parish churches, facing east.

Source: Savchenko, Roads and Kingdoms, p. 37, fig. 18.
Figure 25

Figure 12. Idealised plan of the monastery building.

Source: Savchenko, Roads and Kingdoms, p. 47, fig. 24.
Figure 26

Figure 13. Spatial division and structures in a monastic church and a parish church (top: altar; bottom: corridor space connecting the nave and the sanctuary).

Source: Savchenko, Roads and Kingdoms, p. 53, fig. 31, and p. 52, fig. 29.
Figure 27

Figure 14. Spatial division of the sanctuary in a Kharg monastic church.

Source: Steve, L’île de Kharg, fig. 12.
Figure 28

Table 3. Chronology of bema development from Syria and Mesopotamia to Central Asia (Xinjiang)

Figure 29

Figure 15. Illustration of ‘whole world’ in Tangchaodun Church.

Sources: left: Fiey, Mossoul Chrétienne, pl. ii, fig. 4; right: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 2 (see Figure 1).
Figure 30

Table 4. Bema-related liturgy in Tangchaodun Church and corresponding symbolism

Figure 31

Figure 16. Three-part division of the Hira and Tangchaodun bemata.

Sources: left: Rice, ‘Oxford excavations’, fig. 1; right: drawn by the author.
Figure 32

Figure 17. Seat arrangements of the Faferteen and Hira bemata.

Sources: left: Tchalenko, Églises syriennes, p. 279, fig. 52; right: Rice, ‘Oxford excavations’, p. 280, fig. 1.
Figure 33

Figure 18. East side of the Tangchaodun bema, with inscriptions and murals.

Source: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 4.
Figure 34

Figure 19. Glazed pottery with inscription unearthed in Tangchaodun Church site.

Source: Ren and Wei, ‘2022 nian Tangchaodun gucheng’, fig. 9.
Figure 35

Figure 20. The western wall of the Tangchaodun bema (looking eastwards).

Source: Ge, ‘Yi lu si cheng’, fig. 4.
Figure 36

Figure 21. Head of Christ wearing a crown adorned with a cross.

Source: Liu et al., ‘2021 nian Xinjiang Tulufan Xipang Jingjiao siyuan’, fig. 4.
Figure 37

Figure 22. Line drawing of Christ’s entry into Jerusalem, on the east side of the Tangchaodun bema.

Source: Drawn by the author.
Figure 38

Figure 23. Inscription of Master Yelikewen in Uyghur script.

Source: Ren and Wei, ‘2021 nian Xinjiang Qitai Tangchaodun Jingjiao siyuan’, fig. 8.
Figure 39

Figure 24. Bema mosaic at Uqayribat: palm tree and heraldic lambs.

Source: Giovanna, ‘Geological materials in late antique archaeology’, fig. 11.
Figure 40

Figure 25. Bema mosaic at Rayan: vase and heraldic lambs.

Source: Donceel-Voûte, Les pavements, fig. 242.
Figure 41

Table 5. Substantial depiction of Holy Week in the Tangchaodun bema

Figure 42

Figure 26. The southern wall of the Tangchaodun bema (four niches) (looking northwards).

Source: Ge, ‘Yi lu si cheng’, fig. 5.