• Henry Hopwood-Phillips

Keeping It Roman: How Roman Architecture Survived Its Medieval Competitors



The Great Palace of Constantinople is usually advertised as the last great hurrah of late antique architecture. Indeed, its style – which celebrated peristyle courts and large sprawling complexes (replete with colourful marbles, realistic mosaics and glittery apses) – is typically considered dead-in-the water at roughly the same time as the emperors stopped collecting statuary i.e. mid-sixth century (though quite how the Great Palace’s mosaics, dated to the same century, fit into such a scheme is slightly bewildering unless they were considered self-consciously conservative or even passé during their tessellation). Notions that conveniently slot within the neat chronology of “Late Antiquity,” which grins at the prospect of a curtain fall at the birth of Mohammad (AD 571). According to this line of thought, the Dark Age is presumed to have hit the Eastern Roman Empire like a brick to the head. And with its blood: columns, capitals and entablatures tumbled from its repertoire. This model (which applies labels such as “exhausted” to a “departing” classicism) generously allows for a last flourish; yes, the Romans gave us Hagia Sophia and the cross-in-square types as a belated farewell gift before presumably hopping on a boat to Valinor. This dubious blueprint claims the classical age left a vacuum in its wake that R. Krautheimer reckoned the West – with its block-shaped palaces – filled.[1] While T. Mathews believed Arabs (who gave the Romans their medieval ground plans and façade articulations) played the same role.[2] In other words, both subscribe to a fashionable consensus that pitches post-Roman pluralism against a lumbering, outdated and monolithic Roman tradition.


In reality, however, the Eastern Romans boasted a continuous architectural history in which they were the ones who were imitated. There was no awful caesura but a clear line of Roman development; an impressive architectural hegemony that threw shade on every post-Roman actor by perpetuating a late antique tradition that lay parallel to Arab and Western strains (which contained weaker and simpler echoes of the same motifs). A classic Byzantine move was to roof a peristyle court and turn it into a large reception hall, often with a transverse entrance hall that converted the model into a T-plan. This can be seen most obviously at the Bishop’s Palace at Miletus and the “Byzantine Palace” of Ephesus where a domed reception hall led on to a long transverse entrance hall. Indeed, some of the traits usually associated with Arabic architecture such as the horseshoe arch are really Roman archetypes that are well attested in late antique Anatolia, as are facades with arcaded decoration. Romans rarely robbed from Franks or Arabs. Instead, they looked to their own monastic scene. Early monasteries often had large courtyards that gave onto tall, arcaded facades, indeed many repeated the combination of transverse entrance hall and longitudinal reception hall instead of the earlier narthex-cum-basilica types. The parallels aren’t incidental. Macrina, sister of Basil of Caesarea and Gregory of Nyssa, turned their ancestral estate into one of the first monastic foundations in Cappadocia.[3] A precedent followed by countless donors.


In other words, as the empire’s urbanising impetus sputtered to a halt (and landed aristos took over from city-based elites), Roman architecture took on its own ruralised aesthetic that unashamedly lifted rural, monastic and militarised Cappadocian motifs. Today these are best attested along the south coast of Asia Minor (Cilicia, Lycia and Caria) where inaccessible topography has preserved several examples. Most are farmhouses but some include porticoes in their courtyards. The living quarters were always on the upper floor and had a view across the courtyard. Similar tall houses with low courtyards exist in the late antique villages of the Syrian limestone massif, too. Many Roman buildings took on a defensive guise. Sometimes necessarily, other times less so (in a sense necessity turned into fashion). Tower houses sprouted that still survive in locations such as Niketiaton in Bithynia, Syllaion in Pamphylia and Tekfur Saray in Constantinople. Walls climbed higher before ceding to windows, doors became narrower and thicker. Only a few of these towers can be said to have served much of a practical purpose as military installations. Instead, they mimic the houses of Anatolia’s interior where outcrops of civilisation needed to be fortified. Thanks to Arab raids these defensive towers took root all over Anatolia. Today, the Maeander valley provides several good examples. Back in the capital, continuity was a still a feature in a city where elites (though they had shifted from mainly Illyrian to Anatolian) were a cultural constant. Indeed, any disruption suffered in Constantinople was rarely stylistic. Instead, it was almost always due to the fact there was a demand that grand spaces be (eventually) devoted to God rather than Man. This had a cultural impetus behind it (devoting wealth and beauty to God was a natural Byzantine impulse) and a material one as these huge spaces were probably too large to be maintained in later times when downsizing was the rule and there were fewer bezants circulating – in many cases only the Church or the imperial family could afford outrageous to maintain outrageous expense accounts. Hence the conversion of the large domed palace hall of Antiochus, which came out of the seventh century as the church of St Euphemia[4]; hence the Mangana’s antique palace getting a makeover as a baptistery[5]; hence the “rotunda” – the largest domed hall of the city, which may have been the centre-piece of Arcadia’s palace – being converted into a cistern, with part of it later serving as Romanos I’s Myrelaion monastery.


Perhaps the strongest argument against Latin or Arab influences on Roman architecture is the fact that on the rare occasion a building did echo either of these peoples’ styles it was deemed worthy of comment. Theophanes Continuatus contains an account of the palace of Bryas being modelled on an Arab equivalent that John the Grammarian encountered on an embassy, for instance.[6] And in the early thirteenth century, Nikolaos Mesarites claimed the Hall of Mouchroutas was the work of a “Persian hand,” too.[7]



[1] R. Krautheimer, Early Christian and Byzantine Architecture (1986), 449-450.

[2] T. Mathews & A. C. Daskalakis-Mathews, “Islamic-Style Mansions in Byzantine Cappadocia and the Development of the Inverted T-Plan,” JSAH 57 (1997) 294-315.

[3] W. Jaeger, J. P. Cavaranos & V. W. Callahan, Gregorii Nysseni Opera ascetica (186), 377.

[4] J. Bardill, The Palace of Lausus and Nearby Monuments in Constantinople (1997)

[5] R. Demangel & E. Mamboury, Le Quartier des Manganes et la premiere region de Constantinople (1939)

[6] Cyril Mango, The Empire of the New Rome (1994)

[7] A. Magdalino, Tradition and Transformation in Medieval Byzantium (1978) 101-115.

247 views