Was there a Christian precedent for the Great Mosque of Cordoba?

The cliché of the replacement of sacred spaces, which has arisen without empirical basis, vanishes when it has been excavated.


Fernando Arce Sainz
Tomás Navarro Tomás Library (CCHS-CSIC)


Patio of the Orange Trees of the Mosque of Cordoba. Wikimedia Commons.

Spanish version

From the information available (documentary and archaeological) it can be affirmed that the Great Mosque of Córdoba was built in the time and at the instigation of Abderramán I (756-788). It is not clear, within these dates, when the project began. Some of the accounts that allude to the origin of the building mention the year 170 hegira (784-785 AD), although the texts that proclaim this date, apart from being late (from the 12th century onwards), often contain discredited information from other sources. It is highly unlikely, as these sources claim, that the work was executed in a single year.

The mosque is not a direct and immediate consequence of the conquest. It appears fifty years after the conquest. In the mid-8th century, Abderraman Ibn Muawiya, grandson of the Umayyad caliph Hisham, managed to escape Abbasid violence and establish an independent political-territorial entity in the Iberian Peninsula based on the legitimacy of his belonging to the lineage of the conquerors of al-Andalus. The emergence of the independent emirate signified the start of the Umayyad dynasty in al-Andalus. Following tried-and-tested models, the nascent dynasty promoted monumental spaces from which it projected its power and ideology to society. One of these was Córdoba’s Great Mosque, promoted by the ruler, in which the community of believers, headed by the sultan, gathered for collective prayer. Until the arrival of Abderraman in Córdoba, communal prayer did not take place in a building, a mosque, but in musallas, open-air spaces roughly marked out on the ground where the believers congregated. The open, unobstructed esplanades used as musallas could only be found outside the limits of the city, which was full of physical obstacles.

The building of the Great Mosque opposite the musallas had a major impact on the urban fabric. As it was to be built inside the old Roman walls, it was necessary first of all to prepare a huge site in an area of the city that had been urbanised for centuries. Why was this setting chosen? What had to be removed? This is where an old discussion about the mosque’s immediate past comes into play. Francisco Javier Simonet, a 19th-century Arabist, speaks of a Christian presence in the form of a church that had to be removed to make way for the mosque. Simonet relies on a Muslim literary tradition according to which, when the city was conquered, all the churches were demolished except one, where the Christians were forced to make room for Muslim prayer. Decades later, according to this account, when Abderraman I decided to undertake the construction of the Great Mosque, negotiations were entered into with the Christians to obtain their half of the church. The latter were to leave the site after receiving financial compensation and permission to restore the churches that had been destroyed. One of the versions of this literary tradition mentions the church of Sant B.n.y.n.t, a name traditionally identified with San Vicente, as the scene of the events.

Francisco Javier Simonet (1829-1897​ ). Wikimedia Commons.

A historical myth was thus forged that has made the Mosque of Cordoba the focus of historiographical, political and ideological dispute and battles over its heritage. Recent years have seen a qualitative and quantitative leap in activity on the part of defenders of the Christian past of the mosque . The idea that there was a solitary church (the basilica of San Vicente) seems to have been abandoned. Now there is talk of a whole episcopal quarter made up of various buildings, religious as well as residential and service buildings. Was the Visigothic episcopal complex on the site where the Great Mosque was erected?

An affirmative answer would reinforce the more conservative reading that the Muslim implantation was a symbolic and material expulsion of the defeated and subdued local powers, who were forced to leave their iconic spaces to be used by the conquerors. Without going any further, the case of Cordoba and the supposed basilica of San Vicente has given rise to a persistent historiographical cliché which consists of affirming that pre-Islamic cathedrals were expropriated in order to erect the main mosques of the new Islamic medinas on top of them.

Let us now assess the origins of the mosque by analysing the available documentary and archaeological sources of information. On the whole, the documentary dossier is rather confusing. Apart from the famous literary tradition that the church was first shared and then bought out, there are others that, unjustifiably, are not usually discussed. Taking all the Arabic documentary evidence together brings irreconcilable results. Some authors say that Abderraman I built his mosque on top of an earlier mosque dating back to the time of the conquest. Others say that there was a church but that it was not being shared. The fate of the church of San Vicente is unclear; one source says that the Christian temple that was demolished was not on the site of the future prayer hall but in what became the courtyard of the mosque. If we order the sources for the origin of the mosque chronologically, we find that it is the later ones that mention the presence of churches. Authors from the 10th and 11th centuries such as al-Razi, Ibn Hayyan (who claims to have taken his information from the former) and Ibn al-Qutiya present scenarios of Muslim exclusivity: the emir built the mosque on top of an earlier one erected by the conquerors. It is from the 12th century onwards (Ibn Idari) that the idea of a shared temple appears in what seems to be a clear adaptation, in Córdoba, of a literary tradition whose setting was Damascus.

Roof of the mosque of Cordoba, with the transept of the cathedral in the center. Wikimedia Commons.

As far as the archaeological record is concerned, there have not been many excavations in the mosque, although there have been enough to rule out the possibility that this urban sector had a religious building that needed to be removed prior to the construction of the mosque. This is true for both churches and mosques, since, let us not forget, there are accounts that speak of Christian temples while others allude to mosques. The most extensive and ambitious excavation campaigns were carried out in the 1930s, promoted by Manuel Gómez-Moreno and directed by Félix Hernández. The results were completely demystifying since, according to their own testimonies, nothing was found that could be assimilated to the expected basilica. The most recent levels, those that had to be razed to make way for the site, revealed architectural structures of far from monumental size which, furthermore, did not define liturgical spaces that were even minimally coherent. It was necessary to go down to the deepest, Roman levels, buried by the previous ones, to find remains of a certain monumentality and quality of execution. Some of these can be seen through a glass pane in the floor of the mosque, some three metres below it. The visitor is told that they are part of the Basilica of St. Vincent. False. It is a domestic, residential area, erected around the 4th century.

Location of the structures found by Félix Hernández. Drawing based on the information of Félix Hernández revised by Pedro Marfil and Antonio Fernández-Puertas, published in José Manuel Bermúdez, “El atrium del complejo episcopal cordubensis. Una propuesta sobre la funcionalidad de las estructuras
tardoantiguas del patio de la mezquita de Córdoba”, Romula, 9 (2010), pp. 315-341.

Excavations that took place some time later (in the 1990s and in 2017) reveal the same sequence, while still failing to provide evidence of a razed cult centre. It is no surprise that no demolished churches have been found under the Cordoban Mosque. The cliché of the supplanting of sacred spaces (from cathedrals to central mosques), which arose without any empirical basis, vanishes when these spaces are excavated. In Cordoba, but also in Zaragoza and Toledo, under whose medieval cathedrals we find Islamic oratories, if we dig deeper, Visigothic cathedrals are nowhere to be found.

When Abderraman I, the independent emir of al-Andalus, began to rule in Córdoba, he inherited a city that had been conquered. The area in which the central mosque was to be built already had a strong Muslim presence: a powerful fortified building next to the wall, the Alcazar, which was the residence of the governors who had succeeded each other over the previous fifty years. Excavations of the mosque have provided interesting information about what was going on in this part of the city in the first half of the 8th century. A cobblestone floor was excavated in the courtyard and a large number of Muslim lead coins (known as feluses), a type of currency that was used on a daily basis in small transactions, were found. Chronologically, all the feluses correspond to the emirate dependent on Damascus (during the first half of the 8th century). They are coins lost by people who moved and interacted with their money in this part of the city. At the gate of San Esteban, Félix Hernández excavated a cesspit, sealed by the construction of the mosque, which yielded ceramic materials typical of domestic environments that presented the first novelties with respect to late Roman local pottery tradition. Things were undoubtedly changing in the city and in the lives of its residents.

Arches of the mosque of Cordoba. Wikimedia Commons.

This budding medina would take a leap forward with the emergence of the emirate, starting with the construction of the Great Mosque. How could the Cordovan churches have been affected by this new scenario? Not at all. In the area of the mosque there was no church to destroy, either before or after the mosque was built, so there is nothing to say about this type of activity in this part of the city. Let us look instead at where we have material evidence of pre-Islamic churches that began a new phase in their lives after the capture of the city. Today we have firm archaeological evidence for only three of the late antique and early medieval churches in Cordoba. All three are located in Cercadilla, a suburb in the west of the city, and functioned without interruption after the conquest. One of these churches, identified with the basilica of San Acisclo in the Arab-Christian sources, was certainly active until the beginning of the 11th century.

What we have is a model of implantation rather than displacement. The previous inhabitants of cities where the newcomers settle remained where they were because the elites who dominated and structured local society had made a pact with the incoming power. The city and its inhabitants, redefined its topography, but not in an arbitrary or conflictive manner. Muslim rulers naturally imposed their own interests, but did not contest the symbolic spaces of those with whom they had made the pact. The legitimacy of the agreement, backed by force, allowed the Muslims to act decisively in certain parts of the city, as can be seen in the southern part, where first the citadel and later the mosque appeared. At the same time, other parts of the city continue to maintain unchallenged their previous (Christian) religious references. This is the case of Cercadilla, where the churches are neither destroyed nor challenged by placing mosques nearby. The first archaeologically accredited mosque in this suburb appears in the 10th century and is a consequence of a context (the explosive growth of the city in the Caliphate period) that has nothing to do with what was happening in the 8th century.

This model of implantation is not unique to al-Andalus. We find it in other territories of conquest. The Great Mosque of Damascus, the central mosque of the Umayyad Caliphs, stands within the confines of a colossal pagan temple, with no remains of intervening churches. The Dome of the Rock and al-Aqsa Mosque in Jerusalem stand on the Temple Esplanade, an urban wasteland deliberately maintained under Byzantine rule as a humiliation to the Jewish community. In Amman, one of the hills that once housed a cluster of Roman ruins saw the rise of a large Umayyad palace complex, a mosque and a market.

The establishment of these important mosques undoubtedly represented a solution to the continuity of the history of the cities, but it was not done at the cost of provoking forced displacements but with the intention of generating new poles of urban development that would be Muslim in their own right. From this perspective, let us think again of the historiographical cliché about cathedrals and central mosques: should we imagine that in each and every one of the cities of al-Andalus that continued to have bishoprics (around twenty) the bishops had to pack their bags, leave their cathedral behind and move to another church? For the time being, we have no documentary or archaeological evidence to support such moves.

The Mihrab of the mosque of Cordoba. Fototeca del Patimonio Histórico.

Let us finish by asking about the building that has been hovering over these lines for some time. Where was the Cordoban cathedral? As there is no evidence of relocation (there are no churches under the mosque, let alone an episcopal complex) we have to think that it was always in the same place, from pre-Islamic times until at least the end of the 10th century, when we know for sure that the bishopric of Cordoba still existed (from the colophon of the Biblia Hispalense, written by the then bishop of Cordoba, Juan, in 988). A collection of literary works by Christian writers from Cordoba in the mid-9th century (Eulogio, Álvaro and Sansón) provides a great deal of news about current affairs of the time. We see many people passing through and many religious establishments in the city and its environs. One of them must necessarily be the cathedral. The problem is that none of the three writers uses the term cathedral to refer to a specific church. Among all the churches mentioned, by logic and casuistry, the cathedral has to be one of those located in the urban environment of Cordoba: San Acisclo, San Zoilo, Los Tres Santos. All these religious centres are located in different suburbs – outside the limits marked by the Roman walls – which does not mean that they do not form an effective part of the city of Cordoba. Within its walls, by the way, we find no information about the presence of churches. Neither in the texts from the Visigothic period nor in the Arab-Christian texts are there any references to churches within the walls, either present or past. As for the proposals made from the archaeological record (concerning the basilica of San Vicente in the mosque and the church of the Convent of Santa Clara), they have no firm evidential basis.

Among the candidates for a cathedral, San Acisclo is perhaps the most likely: there are bishops buried in its necropolis (there is a tombstone of Bishop Lampadius who died in 549 and a ring of Bishop Samson, undated); the Arab-Christian textual information shows a strong link between this church and episcopal authority. Whether or not San Acisclo was the cathedral, the important thing is that the city where the Muslims arrived had, outside the old Roman enclosure, its main religious buildings. This circumstance, a consequence of previous historical dynamics that are common to Hispanic post-classical urban centres, worked in favour of those who came, allowing them to occupy spaces that were previously central, which had lost their dynamism but which offered scenarios in which to develop new monumental discourses. The real supplanting would come later, when the medieval Christian powers conquered the great medinas of al-Andalus: the central mosques were transformed into cathedrals, the citadels into Christian fortress-palaces, the neighbourhood mosques into parishes, the souks into markets.


Further reading:


Descubre más desde Al-Andalus y la Historia

Suscríbete y recibe las últimas entradas en tu correo electrónico.