Rock-cut façades: conveyors of ‘false’ monumentality in Byzantine Cappadocia

Fatma Gül Öztürk Büke

doi:10.1017/byz.2022.8

Rock-cut façades: conveyors of ‘false’ monumentality in Byzantine Cappadocia

Published online by Cambridge University Press: 04 July 2022

Fatma Gül Öztürk Büke

Show author details

Fatma Gül Öztürk Büke*: Affiliation:
Çankaya University, Ankara gulozturk@cankaya.edu.tr

Article contents

Abstract
Visibility and impressiveness
Styles and types
Inspirations and origins
Concluding remarks: patronage and monumentality
Footnotes
References

Rights & Permissions

Abstract

The monumental rock-cut façades of the tenth to eleventh century-mansions – so-called courtyard complexes – in Cappadocia, central Turkey, are rare examples of secular Byzantine architecture. While these symmetrically designed façades adorned with superimposed arches differ from the simpler ones (both carved and built) in the region, they bear striking similarities to others from the broader Mediterranean basin. This article offers new insights into the discussion on the uniqueness of the rock-cut façades of courtyard complexes and reconsiders the raison d’être of this ‘false’ monumentality in the rural setting of Byzantine Cappadocia.

Keywords

Byzantine Cappadocia courtyard complexes monumental rock-cut façades horseshoe-shaped arch

Type: Article
Information: Byzantine and Modern Greek Studies , Volume 46 , Issue 2 , October 2022 , pp. 158 - 175

DOI: https://doi.org/10.1017/byz.2022.8 [Opens in a new window]
Copyright: Copyright © The Author(s), 2022. Published by Cambridge University Press on behalf of Centre for Byzantine, Ottoman and Modern Greek Studies, University of Birmingham

In Cappadocia, a region in central Turkey renowned for its idiosyncratic volcanic landscape and rock-cut architecture, the majority of surviving structures carved out of the soft tuff stone date back to the Byzantine period. Hundreds of the rock-cut churches found here feature well-established Byzantine ecclesiastical plan types adapted into this unique natural setting, but usually on a smaller scale.Footnote ¹ By contrast, several large rock-cut complexes, generally identified as tenth to eleventh-century mansions of the local aristocracy, are among the rare examples of Byzantine secular architecture: not much has survived elsewhere in the empire.Footnote ² Although they generally have single-storey interiors, these rock-cut mansions are decorated with engraved façades that often evoke a multi-storey architectural perception (Fig. 1).Footnote ³

Fig. 1: Açıksaray, Area 1 (background) and Area 2 (foreground) (photo: Aykut Fenerci)

There are more than forty such mansions spread out across the volcanic valleys of Cappadocia, either as isolated estates or as groups forming settlements such as at Çanlı Kilise, Selime-Yaprakhisar, and Açıksaray (Fig. 2). Since many of these are organized around naturally or artificially formed three-sided (U-shaped) open courtyards, they have come to be referred to in the literature as courtyard complexes. Echoing the over-generalized and largely unfounded identity of Cappadocia as a supposedly ‘monastic centre’, these complexes too were initially labelled as monasteries. It was only recently that they began to be reconsidered to be mansions belonging to the elite, which is now widely accepted.Footnote ⁴

Fig. 2: Map of Cappadocia, distribution of courtyard complexes (drawing: author)

The courtyard complexes usually feature a central core, obviously used for reception purposes and often occupied by two halls: a vestibule and the main hall, these perpendicular to each other and forming an inverted-T plan. Service spaces, such as kitchens, stables and occasionally a humble chapel secondary to the halls, were also carved into the rock around the courtyard (Fig. 3). Nevertheless, the most apparent common qualities of the courtyard complexes are their two to four-storied rock-cut façades decorated to imitate built architecture (Figs 4–5).Footnote ⁵ Ironically, while these façades carved onto the living rock have survived, their potential prototypes among built architecture have been almost entirely lost.Footnote ⁶ It is the absence of built structures of their kind and their uniqueness among other rock-cut structures that make the façades of courtyard complexes stand out. Comparison of extant façades demonstrates that the symmetrically organized monumental façades of courtyard complexes differ significantly from the often simple and haphazardly carved façades of the region's religious establishments, such as hermitages, free-standing churches, and probable monastic complexes.Footnote ⁷ The rest of the Cappadocian medieval settlements usually evolved organically without any identifiable layout, and they often lack façade decoration altogether.

Fig. 3: Açıksaray, Area 5, plan (survey/drawing: author and Aykut Fenerci)

Fig. 4: Açıksaray, Area 5, the main façade (photo: Aykut Fenerci)

Fig. 5: Açıksaray, Area 5, restitution of the main façade (drawing: author)

Besides their distinctiveness in Cappadocia and the Byzantine Empire, cross-cultural stylistic features, such as the extensive use of horseshoe-shaped arches, attested in the broader territory within and beyond the Mediterranean, make the façades of these courtyard complexes all the more exciting and worthy of closer study. This article offers new insights into the discussion on the uniqueness of the façades of the courtyard complexes and questions the raison d’être of the ‘false’ monumentality in the rural setting of Cappadocia, testing it through three pairs of concepts: visibility and impressiveness; styles and types; inspirations and origins.

Visibility and impressiveness

Rock-cut architecture is practised not by an additive construction process but by extracting voids from the existing rock mass. Therefore, the three-dimensional adaptation of the exterior following the internal organization – in other words, the expression of the inner space in the outer mass, as in conventionally built architecture – is not necessary and is rare in Cappadocia.Footnote ⁸ Most church entrances in the region exhibit necessary openings with a minimum of decoration. Nonetheless, the courtyard complexes with their elaborate rock-cut façades, are the closest in appearance to conventionally built architecture (Figs 4–5).Footnote ⁹ The sculpted main façade of a courtyard complex ensures visibility and indicates the rock-cut architecture behind. The façades of the courtyard complexes often look as if they belong to multi-storey buildings, although the internal organization was, in most cases, only at the ground level. Like a three-sided (U-shaped) courtyard artificially carved into the rock, a high sculpted façade transfers courtyard complexes from the realm of the natural environment to the realm of built – human-made – environment.Footnote ¹⁰

In a typical courtyard complex, a three-sided (U-shaped) open courtyard, the sculpted main façade, and spaces of reception formed by the horizontal vestibule lying behind the main façade and the main hall lying perpendicular to the vestibule, were all aligned along the same central axis (Fig. 3). The axis that indicates a processional approach occasionally ended with a niche carved into the main hall's furthermost end. The niche points to a hierarchical organization of the space, and it may have been where the patron received guests.Footnote ¹¹ The axis was underlined by the symmetrical decoration of the main façade, and the central door leading to the core of reception behind the façade. The axis was often further emphasized through a variation in the carved decoration on the ceiling of the vestibule and the main hall. Likewise, the main hall's lateral walls were occasionally decorated either with arcades or rows of blind niches that flanked and underlined the axis. Along this axis – the processional way – the consistency of decorative details, which differ from settlement to settlement, is also noteworthy. Only in few cases does the façade decoration continue along the courtyard's lateral walls, and even then only to a limited extent, since primary importance was given to the main façade in the centre and, accordingly, to the core used for the reception of guests.

The highly decorated façades of the Cappadocian courtyard complexes are not only instruments of the hierarchical and processional arrangement, enabling the rock-cut architecture to be perceived as ‘real’ and impressive. In a practical sense, they serve to make the complexes visible where they would otherwise merge with the surrounding landscape and disappear into it. In this sense, with their likeliness to be noticed from a distance, the façades of the courtyard complexes resemble the highly decorated portals centrally attached to the long blank walls of medieval Seljuk inns. From the thirteenth century onwards, these caravanserais were built at intervals along trade routes throughout Cappadocia and elsewhere in Anatolia. However, unlike the strictly introverted caravanserais, courtyard complexes are usually open on one or both sides, and so it can be concluded that safety was not a priority, as was the case with the caravanserais.Footnote ¹² Above all, they differ radically from other Cappadocian rock-cut structures, such as the underground cities hidden below ground or behind cliffs. Obviously, the tenth to eleventh-century inhabitants of courtyard complexes felt safe and possessed a certain power: they chose visibility, in order to attract and impress people beyond their immediate boundaries, over seclusion and concealment.Footnote ¹³

Styles and types

In general, while the number and state of surviving façades of Byzantine ecclesiastical built architecture allow us to discuss stylistic variations and relative chronology, this is not the case with secular architecture, whether built or rock-cut.Footnote ¹⁴

Most of the courtyard complexes were probably decorated with façades, but many of these have not survived, due to erosion and human intervention. Based on a comparative architectural study of 43 courtyard complexes conducted by the author, 26 of 43 bear evidence of a sculpted façade.Footnote ¹⁵ Noticeable care was taken to cut and level the irregular rock into a vertical flat surface, and in many cases, horizontal mouldings divide this surface into two to four registers to give the appearance of a multi-storey building. Vertical pilasters are also used in some façades further to divide the registers into odd numbers of bays, and while these pilasters do not always continue through all the registers, in all cases the layout is symmetrical, and the principal entrance to be found in the centre.

The central door is usually a rectangular opening set in a horseshoe-shaped recess, with occasionally a pair of small keyhole-shaped windows cut into the lunettes above the door (Fig. 6). While horseshoe-shaped rather than semi-circular arch is used for the principal entrance, occasionally both types of arches are used together in the same façade. In some cases, additional entrances or blind niches of the same shape and size can be found on either side of the central entrance, which further underlines the symmetrical layout. The arrangement of the façades indicates a desire for monumentality and a desire for distinctness among its kind.Footnote ¹⁶

Fig. 6: Açıksaray, Area 1, the main façade (photo: author)

In some cases, as exemplified at Açıksaray, Area 1, the individual parts are emphasized by carving the pseudo-structural elements deeply (Fig. 6). In other cases, the total composition that combines the various elements in shallower carvings stands out as a whole as at Açıksaray, Area 5 (Figs 4–5).Footnote ¹⁷ Discussing courtyard complexes in Yaprakhisar, Veronica Kalas differentiates between two types of façade layouts, one where blind niches were carved in separate bays (as at Yaprakhisar Area 11), the other with a continuous blind arcade (as at Yaprakhisar Area 14).Footnote ¹⁸ Likewise, Robert Ousterhout differentiates between two systems of façade articulation: ‘a system of superimposed arcades, with minimal horizontal relationships, as at Açıksaray Area 5’ and ‘a sort of grid of horizontals and verticals, as in most of the Yaprakisar facades’.Footnote ¹⁹

A further distinction is achieved by using diversity in the arches and arcades in terms of shape, size and number. Furthermore, stepped carving is applied alternatingly either to the arch itself or to its frame, or the mouldings and pilasters. Surprisingly, although the ceilings and walls of the main halls and vestibules are often decorated with crosses, these are seldom found on the façades and are not in the foreground when they are. Likewise, although occasionally seen in the interiors, figure carvings are not used on the façades, while the lack of finely carved tangled decorations may be attributed to the limitations of the soft tuff stone.Footnote ²⁰ Instead, some façades bear traces of painted decoration, such as zigzag motifs of red paint on a white background, or chequerboard motifs in red and white paint.Footnote ²¹

As for the courtyard's lateral walls, the shape and size of the lateral rock would often be unsuitable for carved decoration to the same extent as the main façade, nor was decoration the carvers’ aim. The processional central axis, adorned with the main façade and ending with the reception area at its heart, was emphasized to dominate the complex (Figs. 3–5). Even in the exceptional four-sided courtyard complex at Eski Gümüş, which is arranged around an enclosed courtyard, only the wall facing the entrance to the inner courtyard is decorated (Fig. 7).Footnote ²² On the other hand, some of the more distinctive details that dominated the main façade of a courtyard complex would often be repeated in the interior decoration of the vestibule, main hall and even of the attached church – where there was one – of the same complex, or of several complexes in the same settlement. The same workshops that decorated the façades also decorated the interior spaces, with certain nuances used as the signature of workshops and stamps of the patrons of the distinctive settlement. For instance, the complexes at Açıksaray feature a framed pair of horseshoe-shaped arches (Figs 4–5), and those at Çanlı Kilise exhibit the rows of gabled horseshoe-shaped arches (Fig. 8), while Eski Gümüş features deeply carved elongated arcading (Fig. 7).Footnote ²³

Fig. 7: Eski Gümüş, the main façade (photo: author)

Fig. 8: Çanlı Kilise, Area 7, the probable main façade (photo: author)

The craftsmen who carved the façades and the rest of the complexes were most likely trained stonemasons. Engraved imitations of structurally unnecessary elements, such as impost blocks, show that the craftsmen were familiar with masonry techniques. Indeed, masonry architecture had a long tradition in central Anatolia.Footnote ²⁴ Ousterhout even claims that medieval masonry architecture was more refined in Cappadocia than in the Capital. Nevertheless, he notes that very few examples from the tenth and eleventh centuries have survived in central Anatolia.Footnote ²⁵ On the other hand, brick was a feature more closely associated with Constantinople, and it appeared rarely, mostly for decorative purposes, in central Anatolia.Footnote ²⁶ Still, carvers of the façades of the courtyard complexes must have been familiar with the techniques and vocabulary of traditional masonry, brick, and hybrid structures; and they must frequently have encountered ancient rock-cut tombs in the area. This spectrum of available vocabulary in situ is reflected in various ways in the decoration of the rock-cut façades and interiors, though preferably in a symbolic manner. For instance, while red lines imitating stone courses were painted in the monolithic vaults, the paintings mentioned above of zigzag motifs on the façades recall brick courses’ decorative use.Footnote ²⁷

Inspirations and origins

The uniqueness of the courtyard complexes’ rock-cut façades have prompted scholars to look at comparisons outside Cappadocian and Byzantine architecture: examples cited include façades of secular and religious architecture of the pagan, Christian and Islamic worlds from various locations across and beyond the Mediterranean, stretching from Persia and Transcaucasia to the Iberian Peninsula. The examples are often chronologically distant too, and further confusion arises from the varying materials and different construction techniques used. Frequently referenced examples in this respect include, among others, the Sassanian Taq-I Kisra palace at Ctesiphon and the Great Mosque of Cordoba.Footnote ²⁸

Thomas Mathews and Annie Christine Daskalakis-Mathews were the first to note the secular character of the Cappadocian courtyard complexes in general. While pointing to the common use of the inverted T-plan and the horseshoe-shaped arch, they claim similarities among the Islamic palaces’ architecture, upper-class houses, and the Cappadocian courtyard complexes, which they even name ‘Islamic-style mansions’. They argue for a shared lifestyle between the elites of neighbouring cultures whose distinct status was reflected in houses along similar lines, regardless of religion and ideology.Footnote ²⁹

Nevertheless, it is important to remember that the use of the horseshoe-shaped arch in the composition of monumental rock-cut façades can, on rare occasions, also be attested in a probable religious context, such as Ala Kilise located in the Ihlara (Peristrema) Valley, in Cappadocia (Figs 2 and 9). Indeed, the design of the façade of Ala Kilise bears a striking resemblance to the stylistic approach favouring the total composition exemplified in the courtyard complex in Açıksaray Area 5 (Figs 4–5).Footnote ³⁰ Kalas points to the seemingly exceptional position of Ala Kilise, where a church dominates the space behind the decorated façade. She claims that ‘all of the monumental carved façades recorded thus far in Byzantine Cappadocia’ were found in the elite and domestic setting of the courtyard complexes. Having in mind the horseshoe-shaped arches on the façade of Ala Kilise, Kalas warns that ‘[i]slamicizing elements in a monument's features does not necessarily indicate a secular function’.Footnote ³¹ Indeed, in Cappadocia, in contrast to the scarcity of monumental façades that adorned religious architecture, the use of horseshoe-shaped arches was not restricted to the secular sphere: such arches decorated the interiors of many churches. They were even used in the plan and elevation of the apses.Footnote ³² Likewise, Ousterhout highlights the ‘blind arcades with horseshoe-shaped arches’ as ‘the norm’, which appeared ‘in varying levels of complexity throughout Cappadocia’.Footnote ³³ On the other hand, Mathews and Daskalakis-Mathews accuse earlier scholars of ignoring ‘an important clue’, namely the horseshoe-shaped arch, despite its frequent appearance in Cappadocian structures, due to its association with Islamic architecture.Footnote ³⁴ It can be concluded that such decorative elements as horseshoe-shaped arches would not have been so laden with ideological meanings as we often think of them today and that the term ‘islamicizing’ alone needs to be reconsidered.Footnote ³⁵

Fig. 9: Ala Kilise at Ihlara Valley (Peristrema), the main façade (photo: Robert Ousterhout)

By contrast, and with more acuity, Lyn Rodley suggests looking at local examples such as Ala Kilise, a masonry church, on Ali Suması Dağı in the neighbouring Lycaonia rather than ‘exotic [Near Eastern] sources’ for comparison.Footnote ³⁶ Although there is indeed an occasional use of similar elements, such as the horseshoe-shaped arch and blind recessed arches or arcades, elsewhere in central Anatolia, none of the existing built structures seems to offer a superimposed façade arrangement similar to that of the courtyard complexes. While Rodley, still admitting the likelihood of a ‘loosely defined’ vocabulary that Cappadocian façades and Near Eastern examples may share, points to the common legacy of the Hellenistic world,Footnote ³⁷ Nicole Thierry stresses that the Hellenistic heritage was variously translated depending on the region and materials: for instance, she stresses that in medieval Georgia, the arch was ‘stretched’, while in Cappadocia it was ‘multiplied’.Footnote ³⁸ However, in Cappadocia, in some cases, as in Eski Gümüş, one notices façade arrangements reminiscent of Thierry's ‘Georgian’ type (Fig. 7). From this, it may be deduced that Cappadocia, being outside the major centres but on the main road that connected them, was more receptive and less selective in this sense. Above all, the nature of rock-cut architecture freed craftsmen from the material and structural concerns associated with masonry and brick architecture, enabling them to create designs based primarily on formal visual appreciation.Footnote ³⁹

However, regarding the readiness to borrow, Ousterhout represents another point of view. He asserts that for ‘the provincial elite of Cappadocia … the cosmopolitan court culture of Constantinople would have been the most immediate source of inspiration.’Footnote ⁴⁰ Unfortunately, the only surviving secular example from Constantinople is the façade of the late thirteenth-century so-called Tekfur Palace. Indeed, although a later building, its superimposed façade arrangement recalls those of Cappadocian courtyard complexes.Footnote ⁴¹ Nevertheless, it should be noted that instead of horseshoe-shaped arches that were extensively used on Cappadocian façades, semi-circular arches were used on the façade of the Tekfur Palace.

Concluding remarks: patronage and monumentality

The varying state of the evidence, coupled with the Cappadocian patrons’ readiness to borrow and the carvers’ inventiveness in adapting the various vocabulary into the unique setting of the rock-cut architecture, may explain the difficulty of defining stylistically conclusive types for the Cappadocian façades. It is unlikely that the question of origin will ever be answered with certainty: it is likely that throughout the Mediterranean and beyond, there was a mutual, multi-faceted and continuous interchange and transmissions of ideas.Footnote ⁴² Indeed, the various Mediterranean and neighbouring cultures in question have intermingled so profoundly and for so long that any attempt to investigate the origin and transformation of a common particular architectural element, such as the horseshoe-shaped or the pointed arch, cannot be easy. Neither has the issue been free of scholarly preconceptions rooted in a dichotomy between East and West.Footnote ⁴³

For Cappadocian façades, the primary question to ask concerns neither style nor origin, but patronage and monumentality. Who were the patrons and what were the motivations for the patronage of monumental architecture in this rural domestic setting? The dating of courtyard complexes to the tenth to eleventh centuries corresponds to the brief period of security and prosperity between the Arab attacks of the eighth and ninth centuries and Seljuk arrival after 1071.Footnote ⁴⁴ Cappadocia, a frontier zone for most of the period, was one of the regions where military aristocracy originated.Footnote ⁴⁵ Accordingly, the majority of these complexes most likely belonged to the military aristocracy which owned extensive land and dominated the region during the tenth and eleventh centuries.Footnote ⁴⁶ Settlements near medieval military installations, such as Çanlı Kilise and Selime-Yaprakhisar, suggest an even more direct association, while the settlement at Açıksaray, with its number of large stables, might have supplied the army with horses (Fig. 2).Footnote ⁴⁷

Although tending to identify most of the Cappadocian rock-cut structures as monastic, when it comes to their façades, Spiro Kostof is very clear:

This [the rock-cut façade of Sümbüllü Kilise in Ihlara (Peristrema) Valley, in Cappadocia] is no product of rustic imagination. The monk who envisaged it could not have been innocent of monumental architecture. Indeed, there can be little doubt that the original progenitors of this and all the other Cappadocian frontispieces to monasteries are Late Antique façades to palaces, formal fountains ‘nymphaea’, and theater stages.Footnote ⁴⁸

The fact that the ‘false’ façades of courtyard complexes were not required to be in accordance with the interior and the structure recalls indeed the façades of monuments of late antiquity, which were maintained as witnesses of the cities’ past greatness, while their interiors were of less concern and were divided into small houses, or even ransacked.Footnote ⁴⁹

Ousterhout claims that ‘[b]uilt architecture, accorded higher prestige, was the referent of the elaborate rock-cut forms’.Footnote ⁵⁰ It was through the façades that the expression of the status of the patrons spoke loudest, and so it is no surprise that the sculpted façades of the Cappadocian mansions recall the common ‘language of power’ that has been applied here and there across the Mediterranean at least since the Roman period.Footnote ⁵¹ While the ‘false’ monumentality might have allowed the complexes to dominate their rural environment, details inspired by a wide range of sources also allowed the competitive patrons to differentiate among themselves.

To conclude: rock-cut façades in Byzantine Cappadocia, not only in terms of appearance but also in terms of material, were humble reflections of the on-going contest for grandeur imitating imperial level that can be traced – if not earlier – back to that between the Romans and the Sasanians, then between the Sasanians and the Byzantines, and finally between the Byzantines and the Caliphate.Footnote ⁵² When all else might fail, one statement can be made with some certainty: that the ambitious patrons of rock-cut courtyard complexes pursued the ‘illusion of power’ provided by ‘false’ monumentality, and that what remains of the secular medieval architecture in Cappadocia is to a large extent but this illusory palimpsest.

Fatma Gül Öztürk Büke is Associate Professor at Çankaya University. She studied architecture at the Universität Stuttgart and received her PhD in Architectural History from the Middle East Technical University (Ankara) in 2010. She has been a postdoctoral fellow at the University of Pennsylvania (2013–14). Her research interests include Byzantine Cappadocia, rock-cut architecture, housing and settlement in the Mediterranean.

Footnotes

This is an extended and revised version of a paper titled ‘Rock-cut façades in Byzantine Cappadocia’, read at the symposium ‘From Constantinople to Cappadocia’ at the University of Pennsylvania Museum of Archaeology and Anthropology (April 2014). I would like to thank Robert Ousterhout, the symposium organizer, for his invitation. Thanks also go to Suna Güven for her input in earlier versions of this paper. I wish to thank the Republic of Turkey's Ministry of Culture and Tourism General Directorate of Cultural Heritage and Museums for granting me a work permit for the Scientific Research Project to survey rock-cut façades in Nevşehir, Cappadocia, in 2020 and 2021. Special thanks go to Çankaya University for the financial support of the Scientific Research Project in 2021.

References

1 Ousterhout, R. G., Visualizing Community: art, material culture, and settlement in Byzantine Cappadocia (Washington, DC 2017) 23–4Google Scholar.

2 Mathews, T. F. and Daskalakis-Mathews, A.-C., ‘Islamic-style mansions in Byzantine Cappadocia and the development of the inverted T-plan’, The Journal of the Society of Architectural Historians 56.3 (1997) 294–315 (295)CrossRef Google Scholar. For an overview on the state of evidence for the ‘Byzantine House’ in general, see S. Ćurčić, ‘Houses in the Byzantine world’, in D. Papanikola-Bakirtzi (ed.), Everyday Life in Byzantium (Athens 2002), 228–38; and K. Dark (ed.), Secular Buildings and the Archaeology of Everyday Life in the Byzantine Empire (Oxford 2004).

3 Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, 299.

4 Rodley, L., Cave Monasteries of Byzantine Cappadocia (repr. Cambridge 2010)Google Scholar, was the first to offer a comprehensive architectural study, classifying the buildings as ‘courtyard monasteries’. For a critique of the common opinion that Cappadocia was a monastic centre, see Ousterhout, R. G., ‘Questioning the architectural evidence: Cappadocian monasticism’, in Mullett, M. and Kirby, A. (eds), Work and Worship at the Theotokos Evergetis 1050–1200 (Belfast 1997), 420–31Google Scholar; Kalas, V., ‘Early explorations of Cappadocia and the monastic myth’, Byzantine and Modern Greek Studies 28 (2004) 101–19CrossRef Google Scholar. See also Kalas, V., ‘Challenging the sacred landscape of Byzantine Cappadocia’, in Walker, A. and Luyster, A. (eds), Negotiating Secular and Sacred in Medieval Art (Aldershot 2009), 147–73Google Scholar; R. G. Ousterhout, A Byzantine Settlement in Cappadocia (rev. 2nd edn. Washington, DC 2011), 206–12, and Visualizing Community, 6–9. For the secular identification of the complexes and their dating, see e.g. Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’; Ousterhout, ‘Cappadocian monasticism’; V. Kalas, ‘Rock-cut architecture of the Peristrema valley: society and settlement in Byzantine Cappadocia’, PhD thesis, New York University, 2000; F. Tütüncü, ‘The land of beautiful horses: stables in middle Byzantine settlements of Cappadocia’, MA thesis, Bilkent University, 2008; F. G. Öztürk, ‘Negotiating between the independent and groups of courtyard complexes in Cappadocia’, in A. Brown and A. Leach (eds), Proceedings of the Society of Architectural Historians, Australia and New Zealand: Open 30 (Gold Coast 2013), 2.837–49.

5 For instance, while Peker classifies Mavrucandere, a rock-cut settlement in Eastern Cappadocia, as a medieval ‘agricultural village’, she uses the absence of decorated façades – as seen in Açıksaray, Çanlı Kilise or Selime – as the main argument. She denies the possibility that the ‘secular halls and rooms’ found in the settlement belonged to a ‘courtyard complex for rural elites’, due to the lack of decorated façades. Peker, N., ‘Agricultural production and installations in Byzantine Cappadocia: a case study focusing on Mavrucandere’, Byzantine and Modern Greek Studies 44.1 (2020) 40–61 (n. 32)CrossRef Google Scholar.

6 Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, esp. 299–300; Rodley, Cave Monasteries, 11, 236–7; Ousterhout, A Byzantine Settlement, 167; Ousterhout, Visualizing Community, 351–3; V. Kalas, ‘Rock cut façades from Byzantine Cappadocia’, in M. Parise et al. (eds), Cappadocia Hypogea: proceedings of international congress of speleology in artificial cavities (Istanbul 2017) 40–5 (43–4).

7 Kostof, S., Caves of God: Cappadocia and its Churches (Oxford 1989), 65Google Scholar; Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, 299; Ousterhout, Visualizing Community, 353. F. dell'Aquila and B. Polimeni, ‘Cave facades of Cappadocian churches: morphological analysis and excavation techniques’, in C. Crescenzi and R. Caprara (eds), The Rupestrian Settlements in the Circum-Mediterranean Area (Florence 2012), 179–88 (179), point to ‘the grandiose architecture of the façades’ of ‘cave-monasteries’ and ‘cave-churches’ in Cappadocia. However, the examples they refer to as monasteries or churches either have simple façades or are secular establishments (such as Açıksaray), mistakenly defined as monasteries or churches by the authors.

8 Exceptionally, in the Soğanlı (Soandos) Valley in Cappadocia are a few churches carved into isolated cones, while in the exterior, the tops of the cones were formed into pinnacle domes. Three churches in the north of the valley are referred to in the literature as the Kubbeli (Domed) Churches; see Ousterhout, Visualizing Community, 306–7. One of the outstanding examples of a three-dimensionally formed exterior of rock-cut architecture in Anatolia is the cruciform church at Kilistra in Lycaonia; see F. G. Öztürk, ‘Rock-cut architecture’, in P. Niewöhner (ed.), The archaeology of Byzantine Anatolia: from the end of Late Antiquity until the coming of the Turks (New York 2017), 148–59.

9 Ousterhout, Visualizing Community, 279, 351.

10 See Kostof, Caves of God, 69; Ousterhout, Visualizing Community, 484.

11 Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, 300; Ousterhout, A Byzantine Settlement, 145–9; Öztürk, ‘Negotiating’, 843.

12 For the concern for security that, among other factors, shaped the plan and elevation of the Seljuk inns, see Yavuz, A. T., ‘The concepts that shape Anatolian Seljuq caravanserais’, Muqarnas 14 (1997) 80–95 (84)Google Scholar.

13 Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, 299; Kalas, ‘Rock cut façades’, 40, 43–4; Ousterhout, Visualizing Community, 352.

14 For a discussion on the façades of the Byzantine built churches, see E. Tok, ‘Türkiye'deki orta ve son Bizans dönemi kiliselerinde cephe düzeni’, MA thesis, Ege University (Izmir), 1997; Ousterhout, R. G., Master Builders of Byzantium (Philadelphia 2008), 194–200Google Scholar; Ousterhout, A Byzantine Settlement, 84–5. The Early Byzantine masonry houses near Silifke in Turkey and the Late Byzantine masonry houses at Mystras in Greece are rare examples of still standing Byzantine secular architecture, and so of the surviving façades. For Silifke, see G. Varinlioğlu, ‘Rural landscape and built environment at the end of antiquity: limestone villages of southeastern Isauria, PhD thesis, University of Pennsylvania, 2008; I. Eichner, Frühbyzantinische Wohnhäuser in Kilikien. Baugeschichtliche Untersuchung zu den Wohnformen in der Region um Seleukia am Kalykadnos (Tübingen 2011). For Mystras see Orlandos, A. K., ‘Quelques notes complémentaires sur les maisons paléologuiennes de Mistra’, Art et société à Byzance sous les Paléologues (Venice 1971), 73–82Google Scholar.

15 The author conducted a comparative architectural study of 43 courtyard complexes including groups of complexes at Çanlı Kilise, at Selime-Yaprakhisar, and at Açıksaray, and ten isolated examples at Direkli Kilise, Karanlık Kale, Eski Gümüş, Soğanlı Han, Erdemli, Şahinefendi, Aynalı Kilise, Hallaç, Kılıçlar and Bezir Hane. See F. G. Öztürk, ‘A comparative architectural investigation of the middle Byzantine courtyard complexes in Açıksaray-Cappadocia: questions of monastic and secular settlement’, PhD thesis, Middle East Technical University (Ankara), 2010; Öztürk, ‘Negotiating’.

16 Kostof, Caves of God, 65; Kalas, ‘sacred landscape’, 168; Kalas, ‘Rock cut façades’, 42–3.

17 For the Açıksaray group see Öztürk, ‘A comparative architectural investigation’, 157–97; F. G. Öztürk, ‘Açıksaray “open palace”: a Byzantine rock-cut settlement in Cappadocia’, Byzantinische Zeitschrift 107/2 (2014) 127–52; Rodley, Cave monasteries, 121–50.

18 V. Kalas, ‘Cappadocia's rock-cut courtyard complexes: a case study for domestic architecture in Byzantium’, in L. Lavan, L. Özgenel and A. Sarantis (eds), Housing in Late Antiquity: from palaces to shops (Leiden 2007), 393–414 (403–4); Kalas, ‘Rock cut façades’, 43–4.

19 Ousterhout, Visualizing Community, 353.

20 L. Rodley and N. Thierry, ‘Cappadocia’, Grove Art Online (2003) (https://www.oxfordartonline.com, retrieved 2021-02-17).

21 Nevertheless, the scarcity of sculptural ornaments on the façades cannot be explained with the nature of the rock-cut architecture alone. Ramsay, W. M. and Bell, G. L., The Thousand and One Churches (reprint, London 2012), 319CrossRef Google Scholar, having in mind masonry churches at Karadağ, write that ‘colour, and not the plastic arts, was counted on to adorn these buildings.’ Likewise, Redford, S., Landscape and the State in Medieval Anatolia: Seljuk gardens and pavilions of Alanya, Turkey (Oxford 2000), 89Google Scholar, underlines that ‘painted plaster imitations of more costly marble panelling’ was an ‘established Byzantine practice’. He points to the common use of colour red, zigzag and chequerboard patterns on wall paintings of Byzantine rock-cut architecture in Cappadocia and Seljuk buildings, and to the association of these paintings with military and elite settings: S. Redford, ‘Flags of the Seljuk sultanate of Anatolia: visual and textual evidence’, in N. Vryzidis (ed.), The Hidden Life of Textiles in the Medieval and Early Modern Mediterranean: contexts and cross-cultural encounters in the Islamic, Latinate and Eastern Cristian worlds (Turnhout 2020), 67–82 (69–70, 72).

22 For the Eski Gümüş complex, see Rodley, Cave Monasteries, 103–18; Öztürk, Comparative Architectural Investigation, 137–9.

23 Sometimes it is difficult to decide whether a façade is the main façade or the interior wall of a destroyed vestibule that has not survived. This situation poses a problem, especially in the Çanlı Kilise, where the complexes are in poor condition.

24 Ousterhout, Visualizing Community, 353.

25 Ousterhout, A Byzantine Settlement, 10–11, 81–2.

26 The masonry church with brick decoration at Çanlı Kilise, Cappadocia, is an example. Its details indicate Cappadocian and Constantinopolitan characteristics and probably a collaboration of craftsmen; see R. Krautheimer and S. Ćurčić, Early Christian and Byzantine Architecture, rev 4th edn (Harmondsworth 1986), 398–400; Ousterhout, A Byzantine Settlement, 25, 84–5; Ramsay and Bell, Thousand and One Churches, 446–7.

27 See Ousterhout, A Byzantine Settlement, 168, 176. For a brief discussion on the brick and stone architecture and the transmission of decorative vocabulary from one to another construction medium, including rock-cut architecture, see Ramsay and Bell, Thousand and One Churches, 446–56. In this respect, see also Trkulja, who points to the nature of the material which affects the degree of abstraction, the ornament on a brick façade being more abstracted than those carved into a cut stone façade; J. Trkulja, ‘Divine revelation performed: symbolic and spatial aspects in the decoration of Byzantine churches’, in A. Lidov (ed.), Spatial Icons: performativity in Byzantium and medieval Russia (Moscow 2011), 213–46 (220). See also n. 21 above.

28 For several comparisons for the Cappadocian façades, see e.g. G. de. Jerphanion, Une nouvelle province de l'art byzantine. Les églises rupestres de Cappadoce, 2 vols (Paris 1925–42), vol. 1, 44–5; Kostof, Caves of God, 69–75; Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, 299; Kalas, ‘Domestic architecture’, 404; Kalas, ‘Rock cut façades’, 44; Rodley, Cave Monasteries, 236–7; Ousterhout, A Byzantine Settlement, 169–70, and Visualizing Community, 351–3; Cooper, E. and Decker, M. J., Life and Society in Byzantine Cappadocia (New York 2012), 206–8CrossRef Google Scholar; Ramsay and Bell, Thousand and One Churches, 449.

29 Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, esp. 300, 309–10; see also Mathews, T. F., Byzantium: from antiquity to the Renaissance (New Haven 2010) 91–2Google Scholar.

30 For Ala Kilise, see Thierry, N. and Thierry, M., Nouvelles églises rupestres de Cappadoce. Région du Hasan Dağı (Paris 1963) 193–200Google Scholar; V. Kalas, ‘Middle Byzantine art and architecture in Cappadocia: the Ala Kilise in the Peristrema Valley’, in J. Alchermes, H. Evans, and T. Thomas (eds), Anathemata Eortika: studies in Honor of Thomas F. Mathews (Mainz 2009), 184–94; Rodley, Cave Monasteries, 119–20.

31 Kalas, ‘Ala Kilise’, 193.

32 See Kostof, Caves of God, 70; Ramsay and Bell, Thousand and One Churches, 316–17.

33 Ousterhout, A Byzantine Settlement, 176–7.

34 Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, 300. The discussion on the origin and transformation of the horseshoe-shaped arch is a complex issue and goes beyond this article's scope. Existence of horseshoe-shaped arch in Europe was the subject of debate between ‘easterners’ and ‘westerners’ in the first quarter of the twentieth century. The debate focused on the origin of the horseshoe-shaped arch where ‘Visigothic origin’ was put against ‘Moorish origin’. E. T. Dewald, ‘The appearance of the horseshoe arch in Western Europe’, American Journal of Archaeology 26.3 (1922) 316–37. Dewald claims the debate was ‘[…] a matter of patriotism rather than of archaeology’. Dewald, ‘the horseshoe arch’, 316. Dewald proposes a third possibility: that the horseshoe arch, ‘originally oriental’ was introduced to Europe before the coming of the Moors to Spain, wherever there was the influence of the East, especially of Asia Minor and Syria. Dewald, ‘the horseshoe arch’, 317. Likewise, Ramsay and Bell point out the appearance of the horseshoe-shaped arch in countries whose architectural tradition derives entirely or partly from a common ‘Asiatic source’. Ramsay and Bell, Thousand and One Churches, 316–17. As for the Cappadocian examples, dating the Cappadocian examples mistakenly into the fourth century, Texier and Pullan claim that the use of the horseshoe-shaped arch in Cappadocian rock-cut façades was before Islam. Texier, C. and Pullan, R. P., Byzantine Architecture: Illustrated by Examples of Edifices Erected in the East During the Earliest Ages of Christianity, with Historical and Archaeological Descriptions (London 1864) 4, 40Google Scholar.

35 For a general discussion on the validity of designation ‘Islamic Art’ and ‘Islamic Architecture’ see, O. Grabar, The Formation of Islamic Art (rev 2nd edn. New Haven 1987) 1–18; and Rabbat, N., ‘What is Islamic architecture anyway?’, Journal of Art Historiography 6 (2012) 1–15Google Scholar.

36 Rodley, Cave Monasteries, 237.

37 Op.cit., 236–7; see also Kostof, Caves of God, 69; and Ousterhout, A Byzantine Settlement, 174.

38 N. Thierry, La Cappadoce de l'Antiquité au Moyen Âge, Bibliothèque de l'Antiquité tardive 4 (Turnhout 2002) 101.

39 Kostof, Caves of God, 45; Ousterhout, ‘Ecumenical character’, 219–20; Ousterhout, Visualizing Community, 23–4, 486; Kalas, ‘Rock cut façades’, 45. For more on the nature of rock-cut architecture, see F. G. Öztürk, Kapadokya'da Dünden Bugüne Kaya Oymacılığı / Rock Carving in Cappadocia From Past to Present (Istanbul 2009); Öztürk, ‘Rock-cut architecture’; Rodley, Cave Monasteries, 224–5.

40 Ousterhout, A Byzantine Settlement, 174.

41 Kalas, Peristrema Valley, 114–15; Ousterhout, A Byzantine Settlement, 169; Ousterhout, Visualizing Community, 351.

42 For critical approaches to the issue of architectural influences between the East and the West during the medieval period, see e.g. Howard, D., ‘Venice and Islam in the Middle Ages: some observations on the question of architectural influence’, Architectural History 34 (1991) 59–74CrossRef Google Scholar; Draper, P., ‘Islam and the West: the early use of the pointed arch revisited’, Architectural History 48 (2005) 1–20CrossRef Google Scholar. See also note 28 above.

43 See notes 34 and 42 above.

44 Rodley, Cave Monasteries, 2–5. See also note 4 above.

45 Kazhdan, A. and Epstein, A. W., Change in Byzantine Culture in the Eleventh and Twelfth Centuries (Berkeley 1985) 63Google Scholar.

46 The first intensive study to make this suggestion was by Ousterhout, who surveyed Çanlı Kilise in western Cappadocia between 1994 and 1997. See Ousterhout, A Byzantine Settlement and Ousterhout, Visualizing Community, esp. 275–77. See also Kalas, ‘Rock cut façades’, 45. For historical and administrative changes of the tenth and eleventh centuries see Hild, F. and Restle, M., Kappadokien (Kappadokia, Charsianon, Sebasteia und Lykandos) (Vienna 1981) 70–105Google Scholar. For the Byzantine aristocracy in general, see Ostrogorsky, G., ‘Observations on the aristocracy in Byzantium’, Dumbarton Oaks Papers 25 (1971) 3–32CrossRef Google Scholar. For the Cappadocian aristocracy and wealthy landowners during the tenth and eleventh centuries, see M. Kaplan, ‘Les grands proprietaires de Cappadoce (VI–XI siècles)’, in C. D. Fonseca (ed.), Le aree omogenee della civilta rupestre nell'ambito dell'Impero Bizantino: La Cappadocia (Galatina 1981) 125–58; and Cheynet, J. -C., ‘L'aristocratie cappadocienne aux Xe et XIe siècles’, Dossiers d'Archéologie 283 (2003) 42–50Google Scholar.

47 For the strategic position and proposed military association of Çanlı Kilise, see Ousterhout, A Byzantine Settlement, esp. 7–9, 172–3, 182–4; of Selime-Yaprakhisar, see Kalas, Peristrema Valley, esp. 156–9; of Açıksaray, see A. Grishin, ‘Açık Saray and medieval military campaigns’, in L. Rasmussen, V. Spear, D. Tillotson and J. H. Tillotson (eds), Our Medieval Heritage: Essays in Honour of John Tillotson for his 60th Birthday (Cardiff 2002) 164–71; Öztürk, Comparative Architectural Investigation, 157–97; Öztürk, ‘Açıksaray “open palace”’; and Rodley, Cave Monasteries, 121–50, esp. 150. See Tütüncü, The Land of Beautiful Horses, for a proposed link between horse-breeding and the military context in Cappadocia. See Redford, ‘Flags of the Seljuk sultanate’, for a proposed link between red painted wall decorations found in Açıksaray and military context.

48 Kostof, Caves of God, 69.

49 See Ellis, S., ‘The end of the Roman house’, American Journal of Archaeology 92 (1988) 565–76 (567)CrossRef Google Scholar, and ‘Early Byzantine housing’, in K. Dark (ed.), Secular Buildings and the Archaeology of Everyday Life in the Byzantine Empire (Oxford 2004) 37–52 (48).

50 Ousterhout, A Byzantine Settlement, 176.

51 See Ousterhout, ‘Ecumenical character’, 214–18; Ousterhout, Visualizing Community, esp. 279, 352; see also e.g. Mathews and Daskalakis-Mathews, ‘Islamic-style mansions’, 299; Kalas, ‘Sacred landscape’, 165, 168–9; Kalas, ‘Rock cut façades’, 43, 45; and Rodley, Cave Monasteries, 237. Redford especially emphasizes the twelfth and thirteenth centuries as an era of ‘culture contact between Islamic and Christian societies’ in the eastern Mediterranean, which allowed a ‘transmission of visual language of power and privilege’; Redford, ‘Flags of the Seljuk sultanate’, 73.

52 For a discussion on architectural ‘exchange’ between Roman, Byzantine, Sasanian and Arabic palaces, see N. Westbrook, ‘An exchange between East and West emulations and borrowings in Roman, Byzantine, Sasanian and Arabic palaces, from the third to tenth centuries’, in A. Brown and A. Leach (eds), Proceedings of the Society of Architectural Historians, Australia and New Zealand: Open 30 (Gold Coast 2013), 1.365–74.