Archaeology Online

The Online Hub For All Things Archaeological

The Fate of Temples in Late Antiquity – H.Cassidy (2013)

The Fate of Temples in Late Antiquity

Henry Cassidy

3rd year BA Student University College London

The Emperor Theodosius II (AD 408-450) issued an edict in AD 435 requiring the destruction of any pagan temples and shrines still remaining intact, and the purification of their sites by the setting up of a cross.[1] The Codex Theodosianus includes many other laws prohibiting acts of Paganism with the aim of speeding up the process of Christianisation throughout the Empire. These official records, along with the several testimonies of temple destruction found in the writings of Eusebius, Julian, Gregory of Nazianzus or Libanius might lead the archaeologist or historian to believe that, during the 4th century, traditional polytheism was triumphantly eradicated by Christianity and the temples were either razed to the ground or left to decay as a scar in the late antique landscape. However, recent research on the phenomenon of temple conversion, focussing on the abundant archaeological remains, has produced a far more complex picture. It is the purpose of this paper to provide some insight into the current literature on the topic of temple conversion in late antiquity. I shall begin by presenting the typology for (i) secular conversions and (ii) temple churches which have recently been proposed. Secondly I shall turn to assess the fate of temples in Athens and Rome, in order to illustrate how socio-political factors affected the process of Christianisation in these two regions. Finally, I shall approach the temple conversions at Aphrodisias and Athens as examples of spolia, and discuss how these examples may help us to elucidate this peculiarly late antique phenomenon.

Categories of Temple Conversion

In order to understand and to quantify a complex process such as temple conversion, a methodology must be utilised to classify the different types of extant evidence. In this section two broad categories of temple conversions proposed by Talloen and Vercauteren and Bayliss will be used to illustrate the possible fates of temples, and to set out the terms used for further analysis in the remainder of this paper.[2]


Secular Conversions

The secular category includes those temples that were preserved for use in a different non-pagan function such as the Temple of Artemis and Hadrian on the Curetes Street at Ephesus. The shrine is thought to have been restored in the 4th century AD, seemingly adopting a new role celebrating Theodosius and his father as new city founders. This theory is evidenced by the new architectural frieze sculpture, depicting Ephesian foundation legends such as the wild boar hunt of Androklos alongside an image of the Emperor Theodosius.[3] The second group of secular conversions were those temples which were simply left standing. It has been argued by Saradi-Mendelovici that in some cases Christians had a positive attitude to pagan monuments and preserved them for their aesthetic value.[4] For example, a law passed in AD 382 orders that a temple in Osrhoene be kept open:

By the authority of the public council we decree that the temple shall continually be open that was formerly dedicated to the assemblage of throngs of people and now also is for the common use of the people, and in which images are reported to have been placed which must be measured by the value of their art rather than by their divinity.[5]

Those emperors that had previously ordered the closure of sanctuaries and the abolition of sacrifice simultaneously appreciated the artistic value of the temples and allowed the pagan past to remain as a feature in the landscape. An example of this in Asia Minor is the Corinthian prostylos temple at Patara in Lycia, which appears to have been vacated and subsequently preserved perhaps as a civic ornament. Alchermes suggests that this imperial policy “hoped to maintain not only the monuments of the past, but also the civic spirit that helped to produce them”.[6]

Religious Conversions – Temple Churches

The second category of temple conversion includes those temples that have been transformed in some way into religious buildings. According the new methodology developed by Bayliss, religious conversions can be distinguished by two subcategories: direct and indirect conversions.  I will begin by looking at an indirect conversion, a term used to describe “a church that in terms of construction, orientation, extent or ground plan was influenced by the non-existent remains of a pre-existing temple.”[7] An example of an indirect conversion is the church of Mary at Ephesus built within the temenos of the Olympieion. In this case, the appropriation of the Olympieion temple included a temenos area flanked by a stoa, this standing masonry was incorporated into the fabric of the church. Indirect conversions exemplify the practice of exploiting the structural capacities of the pre-existing temple remains as far as possible.[8] Furthermore, the appropriation of the whole temenos, as in the conversion of the Olympieion at Ephesus, reflects the changing function of the ritual space. That is to say the temple was large enough to act as domus dei, but the whole temenos-precinct was required to fulfil the needs of the domus ecclesiae. Bayliss’ second subcategory is a direct conversion, one he describes as “a church that preserved some in situ remains of a pre-existing temple within its fabric.”[9] The most common example of a direct transformation is the so-called ‘cella-church’. The conversion of the Parthenon is an example of a cella-church, in which the main structural implications are the construction of an apse, the modification to the internal colonnade and the opening of three doorways in order to convert the western chamber into a narthex (Fig. 1).[10] It is interesting here to note that the majority of the architectural sculptures: the frieze, the metopes and the pediments were left relatively untouched. In fact Ward-Perkins points out that “an ancient Greek transported forward in time, who visited the cathedral of medieval Athens, would have no difficulty in recognising it as the temple of Athena.”[11]

Fig. 1. Athens, Parthenon: plan of cella-conversion (after Bayliss 2004, 153)

Fig. 1. Athens, Parthenon: plan of cella-conversion (after Bayliss 2004, 153)

A rather less common but far more dramatic direct conversion is the Temple of Aphrodite at Aphrodisias, thought to have been transformed during the reign on Leo I (AD 457-474). This is an example of what Bayliss terms an inverted transformation (Fig. 2), that is to say the columns of the long peristasis stayed in their original place acting as a partition between the central nave and the aisles, while the columns from the shorter sides of the peristasis were relocated to construct the outer walls of the church. In contrast to the more discreet cella-church exemplified by the Parthenon, the inverted temple was a drastic piece of construction leaving behind an enormous and perhaps unrecognisable church in comparison to the preceding temple.[12]

Fig. 2. Aphrodisias: plan of inverted temple transformation (after Bayliss 2004, 166)

Fig. 2. Aphrodisias: plan of inverted temple transformation (after Bayliss 2004, 166)

Christian Perceptions and Regional Variation

The data compiled by Bayliss suggests that the phenomenon of indirect temple conversion took place between the 4th and 7th centuries, however direct conversions only appear to have occurred from the middle of the 5th century.[13] Unfortunately, the majority of the temple conversions cannot be dated stratigraphically and rely on an independent piece of datable evidence, often a coin or historical source. However, the assumed model of temple conversion has been given by Bayliss as “use – abandonment – reuse”. This section will explore the reasons for this hiatus, and what these might tell us about Christian perceptions of temples and inter-regional differences in the process of Christianisation. The evidence suggests that Christians considered a building such as the Temple of Hadrian and Artemis at Ephesus to be apt for conversion into a building extolling the virtues of the Christian emperor in the 4th century. On the other hand the Temple of Aphrodite, at Aphrodisias was not converted until the mid-5th century. The explanation for this gap is perhaps that pagan sites were thought to have been the homes of evil spirits, and therefore it was ideologically easier to convert them for use in a more secular function. The idea that these ritual spaces were polluted is attested in tales of saints’ spiritual fights against the demons inhabiting temples.[14] Additionally, the practical requirements of a church as compared to a temple must be understood, as Ward-Perkins points out “with their small cellae [pagan temples] would seldom make satisfactory congregational churches.”[15] This would suggest that the initial reaction of early Christians was to avoid pagan temples at all costs and perhaps let them fall into ruin as a sign of the death of an ousted religion. On the other hand, it is conceivable that temple conversion in itself would have been the most powerful means of spiritual cleansing. Furthermore, it cannot be denied that the appropriation and conversion of a pagan space would be a powerful statement of intent to speed up conversion throughout the empire.

However, talking of a movement ‘throughout the empire’ may seem to overlook a key factor in the phenomenon of temple conversion and late antique urbanism more generally. As Bayliss put it, “The religious climate varied greatly between cities and probably hinged to a large degree on the respective dispositions of the individuals wielding local power, whether it was the bishop, a governor or prefect, or the local elites.”[16] The fact that regionalism played a key part in the rate and nature of temple conversion is illustrated by two extreme examples: Athens and Rome. The situation at Athens, in which the three big examples of temple conversion: the Hephaisteion, the Parthenon and the Erectheion were each converted in the 6th or even 7th centuries, is a testimony to the relative autonomy of the city in late antiquity. The tenacity of traditional polytheism in Athens is generally put down to the support from the academic elite, centred on the heads of the re-established academy, a post that was filled from Plutarch (AD 350-432) to Damascius (AD 458-538), the latter of which was in office until AD 529.[17] However, the influence of the Neo-Platonists was not only exerted on Athenian education; inscriptions tell us that the philosopher Proclus (AD 412-485) was consulted on civic matters and left a large part of his estate to the city. Furthermore, magistrates and officials in charge of Athens are thought to have acted as benefactors to the academy and seem to have had no problem conducting pleasant affable relations with the pagan elite.[18] The evidence suggests a similarly late date for the conversion of temples in Rome, the earliest of which is the Pantheon, which was converted into the church of S. Maria ad Martyres in 608/15.[19] As with Athens, those wielding the power in Rome were pagan. And the history of local aristocrats, whose fortunes were invested in temple estates, expressing their wealth through honouring gods publicly was a deeply entrenched part of society.[20] One further reason why the fates of temples in Rome and perhaps the west in general seems to have been drawn out in this manner is the influence of Constantine’s behaviour, which perhaps set a model for further acts of religious conversion. A clear distinction can be draw between the first Christian emperor’s treatment of Rome in AD 312, and his treatment of Jerusalem in AD 324. In Rome he took power and built an impressive set of churches but did not touch the pagan sanctuaries. Instead, Constantine chose to build either on religiously insignificant land, as in the case of the Lateran, or on shrines of martyrs, as in the case of St Peter’s. However in Jerusalem, Eusebius describes how “Constantine commanded the materials of the idol temple, and the soil itself, to be removed at a distance.”[21] After ostentatiously having the temple destroyed, he built the Church of the Holy Sepulchre at the location. The example set by Constantine, which seems to foretell the religious conflict, described in the ancient sources, mainly taking place in the diocese of Oriens, might go some way to explaining the comparatively slow development of temple conversion in the west.[22]

Spolia: The Past in the Present

This final section will analyse two direct temple conversions with specific reference to their use of spolia, a phenomenon specifically associated with late antiquity and the Middle Ages. The appropriation and subsequent reuse of architectural material has been variously described in previous scholarship, some see it in terms of decline and inferiority to the past, some see it in terms of expediency and pragmatism, while others see it as a conscious choice as part of a new aesthetic. I will argue that two very different temple conversions – dated contemporaneously- were conceived by a mix of pragmatism and conscious monumental continuity. The Parthenon is thought to have been converted into a cella-church in the late 6th century. If this conversion had taken place in the 580s, during which the city was devastated by Avar and Slav raids, it is conceivable that the bishops of Athens were forced to move the sacred space up on top of the acropolis and convert the Parthenon with great haste, in order to house the congregation.[23] This emergency would explain the minimal changes made to the exterior of the temple and its construction. Even if the date was rather earlier, and the city was at peace, the reasons for the very minimal temple conversion may well have also been pragmatic. It was perhaps deemed unnecessary to dismantle and relocate parts of the temple, the cella was already unusually large and could be converted into a church with few structural changes (Fig. 3).

Fig. 3. East end of the Parthenon drawn as a church (after M. Korres)

Fig. 3. East end of the Parthenon drawn as a church (after M. Korres)

With this in mind we may turn to the Aphroditeion, a temple with a rather small cella that was drastically converted and enlarged. Perhaps this type of conversion was practically necessary, for if it was converted in the same manner as the Parthenon, the resultant church would be a mere 25m long and unsuitable for a cathedral.[24] Thus we can see that in both cases, the appropriation and reuse of temples can be seen as the necessary reaction to circumstances in these late antique cities. On the other hand, there is also much to say about spolia as a key feature of the late antique aesthetic. In his contribution to Age of Spirituality: A Symposium, Brown asserts, “the art of late antiquity has provided a vivid and ready communicable paradigm of the central problem of the period – the relation between change and continuity in late antique civilisation.”[25] The phenomenon of temple conversion can be seen as an example of this paradigm. The landscape of Athens and Aphrodisias was undoubtedly changed by the construction of these two churches, however one can perceive a sense of continuity in the reuse of material inextricably linked with the pagan past. The preservation of the Parthenon’s architectural sculptures within the frame of the building linked the present with the golden age of Periclean Athens, and all the cultural references that came with it. Similarly, at Aphrodisias the retention of a monumental centrepiece would preserve familiarity with the urban landscape of the past.[26] For these two cities, in which there existed a pagan presence all the way up to the early 6th century[27], Christianisation of the built environment was necessary and inevitable. But it did not simply displace the physical manifestations paganism; rather it incorporated them into something new.


As Christianity spread throughout the Roman Empire during the 4th century the pagan temples and shrines, which had played a key role in civic life for centuries, gave way to the construction of churches. However, the archaeology of these monuments attests to a complex process and one that calls into question the received wisdom of the literary sources. In reality, the fate of temples in late antiquity did not consist solely of violent destruction or conversion into a church. Rather, examples from the cities of Asia Minor present two alternatives: re-use as a secular building as in the case of the Temple of Hadrian and Artemis at Ephesus, and re-use as a monument to the past as in the case of the temple at Patara. The sense of civic pride that placed such great importance on these buildings and fostered their preservation is perhaps a facet peculiar to the cities of Asia Minor. There was after all a long history of euergetism in these cities stretching back to the Hellenistic period and picked up during the Hadrianic and Antonine periods.[28] For this reason we might expect the elite of these cities to keep these traditions alive, opting for re-use rather than destruction.

Additionally, one must not forget that non-ideological and practical reasons may have determined the fate of temples in late antiquity. Some temples were most probably used as quarries for the construction of churches, and in some cases part of the temple structure was used in the construction of a church. The church of Mary at Ephesus is an example of the latter; the construction of a church within the temenos of the Olympieion does not suggest triumphal destruction but rather a pragmatic use of material and space.

A precise chronology for temple use and re-use in the examples presented is not possible. However, three stages can be delineated: temples were converted into secular or monumental buildings in the late 4th and early 5th centuries, indirect temple conversions occurred from the first half of the 5th century onwards, finally direct conversions from temple to church begun in the late 5th and early 6th centuries. The relatively late dates for direct conversions such as the Temple of Aphrodite at Aphrodisias (c. mid-late 5th Century) and the Parthenon (c. late 6th Century) suggest that Christians had significant reservations about shifting sacred space during the 4th and 5th centuries. Furthermore, the re-use of architectural members and decorative friezes in these newly constructed churches demonstrates the importance of pagan heritage in the late antique cityscape.

I hope to have shown that temple conversion, and more generally the material remains of Christianisation cannot be viewed simply as the next logical step, the usurping of power from paganism and initiation of Christianity into the built environment. Rather, the fate of temples, and all the complexities and categories involved are a microcosm of the process of Christianisation itself with all its spatial and temporal variations.


Alchermes, 1994, ‘Spolia in the Roman Cities of the Late Empire: Legislative Rationales and Architectural Reuse’, Dumbarton Oaks Papers(48), pp167-178

Bayliss, 2004, ‘Provincial Cilicia and the Archaeology of Temple Conversion’, BAR International Series(1281), Oxford

Bayliss, 2005, ‘From Temple to Church: Converting Christianity to Paganism in Late Antiquity’, Minerva, pp16-18

Brown, 1980, ‘Art and Society in Late Antiquity’, in K. Weitzmann (ed), Age of Spirituality: A Symposium, New York, pp17-29

Cameron and S. Hall, 1999, Life of Constantine by Eusebius of Caesarea (trans.), Oxford

Frantz, 1965, ‘From Paganism to Christianity in the Temples of Athens’, Dumbarton Oaks Papers(19), pp187-205.

Pharr, 1952, The Theodosian Code and Novels and the Sirmondian Constitutions(trans.), Princeton

Salzman, 1999, ‘The Christianization of Sacred Time and Sacred Space in Late Antique Rome’, in W.V. Harris (ed), The Transformation of Urbs Romana in Late Antiquity(Journal of Roman Archaeology Supplementary Series 33), pp123-134

Saradi-Mendelovici, 1990, ‘Christians Attitudes to Pagan Monuments in Late Antiquity and their Legacy in Later Byzantine Centuries’, Dumbarton Oaks Papers(44), pp47-61

Scherrer, 1995, ‘The City of Ephesos: From the Roman Period to Late Antiquity’, in

Koester (ed), Ephesos: Metropolis of Asia: An Interdisciplinary Approach to its Archaeology, Religion and Culture(Harvard Theological Studies 41), Valley Forge, p1-25

R.R.R. Smith, 1990, ‘Late Roman Philosopher Portraits from Aphrodisias’, Journal of Roman Studies (80), pp127–55

Strong, 1976, Roman Art, New Haven

Talloen and L. Vercauteren, 2011, ‘The Fate of Temples in Late Antique Anatolia’, in L. Lavan and M. Mulryan (eds), The Archaeology of Late Antique Paganism, Leiden, pp347-389

Ward-Perkins, 2003, ‘Reconfiguring Sacred Space: From Pagan Shrines to Christian Churches’, in G. Brands and H.G. Severin (eds), Die spätantike Stadt und ihre Christianisierung, Wiesbaden, pp285-290

Ward-Perkins, 1999, ‘Re-using the Architectural Legacy of the Past, Entre Idéologie et Pragmatism’, in G.P. Brogiolo and B. Ward-Perkins (eds), The Idea and Ideal of the Town in Late Antiquity and the Early Middle Ages, Leiden, pp225-245

[1] C.Th.16.10.25 (trans. Pharr 1952)

[2] Talloen and Vercauteren 2011, 358; Bayliss 2004, passim

[3] Scherrer 2004, 21

[4] Saradi-Mendelovici 1990, 47-61

[5] C.Th.16.10.8 (trans. Pharr 1952)

[6] Alchermes 1994, 168

[7] Bayliss 2004, 7

[8] Bayliss 2005, 18

[9] Bayliss 2004, 7

[10] Bayliss 2004, 36

[11] Ward-Perkins 1999, 236

[12] Bayliss 2004, 41

[13] Bayliss 2004, 56

[14] Saradi-Mendelovici 1990, 54

[15] Ward-Perkins 2003, 286

[16] Bayliss 2004, 27

[17] Frantz 1965, 191

[18] Frantz 1965, 192

[19] Ward-Perkins 2003, 287

[20] Salzman 1999, 132

[21] Eus. Life. 27 (trans. Cameron and Hall 1999)

[22] Ward-Perkins 2003, 289

[23] Ward-Perkins 1999, 240

[24] Ward-Perkins 1999, 239

[25] Brown 1980, 18

[26] Ward-Perkins 2003, 290

[27] Smith 1990, 127 on late paganism in Aphrodisias

[28] Strong 1976, 214-17


%d bloggers like this: