Contested Sharing
Image of Mary Theotokos above Koranic passages on rondels in apse/ mihrab of Hagia Sophia in Istanbul, ca. 1935. Church from 537-1453; mosque from 1453-1934 and since 2020; Museum 1935-2020. These decorations reflect all stages.
While most analyses of shared sacred space focus on brief moments of time in single shrines, in the Antagonistic Tolerance” (AT) project we have instead focused on trajectories of change at multiple sites in Bulgaria, India, Portugal, Turkey, Mexico and Peru. The model that we have developed explains long-term patterns of relationship between members of groups which identify themselves and each other as Self and Other communities, differentiated primarily on the basis of religion, living intermingled but rarely intermarrying. The religious distinction is often accompanied by other differences, such as in naming, kinship, diet and perhaps preferred methods of gaining a livelihood. Their separate identities change, but slowly – we refer to them as viscous rather than simply fluid.
In this model, which was inspired by the early 20th century work of F. W. Hasluck (1929), contestation develops in a region in which one religious tradition is dominant when a community identified with a differing religion arrives via trade, or indigenous development. The AT model posits that in such situations, there is “tolerance” in the sense used by John Locke, of enduring the presence of the other but not embracing it, so long as one group is clearly dominant over others. Dominance of one group over the other is indicated in part by control of the primary identity of major religious sites. However, when existing dominance is threatened, violence generally results, and violence often accompanies the transformation of sites, which may happen when one group replaces the dominance previously held by another. The processes involved are long-term, though transformations may take place in short periods. Thus the model is one of punctuated equilibrium.
Hasluck’s model was developed to explain events and transformations in the late Ottoman Empire. In developing the AT model, we found Hasluck’s formulations helpful for explaining social processes far from the Ottoman Empire, in central India from the “medieval” period there through colonialism and in the contemporary Republic of India, and in the former Yugoslavia since the 15th century and especially in the 19th and 20th centuries. Since beginning our comparative research in 2008, we have found patterns of transitions of religious sites linked to changing patterns of dominance of co-resident communities identified with different religions have been congruent with the AT model.
For example, in Mértola, Portugal, Roman paganism was supplanted by Roman Christianity, which was supplanted by later forms of “paleochristianity,” which were in turn supplanted by Islam, which was ultimately supplanted by Roman Catholicism. These stages may be seen in situ at what had been the site of the Roman forum, the fountain of which became a Roman Christian baptistery associated with a Christian church, which was supplanted in the 12th century by a mosque; the mosque is still standing but was transformed into a Catholic church in the 13th century and is still used as such. Nevertheless, the location of the church’s altar was contested over centuries, first being placed on the north wall and thus not that of the mihrab of the mosque (east wall), only to be placed a generation later in front of the mihrab, and a generation again back on the north, later again to the east, and finally (at least thus far) to the north; it is noteworthy that in a paleochristian church, the alter would also be on the east but in Roman Catholic ones in Portugal, the alter is generally the north. The interpretation of local scholars is that these movements of the altar manifested contestation between local sentiment (for the east, thus mihrab and also paleochristian altar wall) and the centralized tenets of the Catholic church.
While recognizing the transitions from one dominant form to another, the Mértola researchers have stressed continuity, and that even the “reconquest” transition of the town was not violent. By 2009, the mihrab had been exposed by archeologists, and while the local priest was willing enough to talk about the previous status of the church as a mosque, he objected strongly when in the course of an “Islamic Festival” in the town in May, which was sponsored by the local authorities, the Muslim call to prayer was given in front of the church, demanding that it be done farther away.
It is important to note that often, members of a subordinated community may visit religious sites claimed by the dominant one, and even perform some observances there. Syncretism may arise from such sharing, even though dominance of one group over the other is clear. To give another example from our research, local Muslims visit the Bulgarian Orthodox Church of St. Nikolas in Smolyan, Bulgaria in order to observe the vigil of the saint, but must remain in the choir gallery, upstairs, not on the main floor, near the icon. Thus they share the space of the church but from a subordinated position (structurally the same position in the church where the women’s section of a Balkans mosque would be located). Further, decorations on the iconostasis of this church include deep-carved images of an Orthodox priest holding a cross over a cringing Ottoman officer, and of a Bulgarian peasant shooting a Turkish soldier. Domination is clear – this is an Orthodox church, and Christian Bulgarians control it. Yet so is syncretism: Muslims are holding the vigil of St. Nikola, and using their own prayer beads as they do so.
Mértola, Portugal: Parish church, 12th century mosque, church since 13th century; Our Lady of Snows (Nossa Senhora das Neves) altar in front of mihrab of the mosque; and close ups. Mihrab uncovered in 1950s. Photo 2009.
While this Bulgarian example is from a formerly Ottoman region, the iconostasis carvings showing hostility to Muslims were made during the communist period, when religious differences should not have mattered; and the present-day segregation of Muslims in the gallery of the church takes place in the context of a European Union pledged, at least officially, to equality.
The contexts of these examples from our own research are very different from those of the Ottoman period in which Hasluck worked, yet the model is still accurate. Other documented interactions that are congruent with the AT model include: between Catholics and Protestants in shared/ contested churches in 16th-17th century German states; between Pagans and Christians in late antiquity Athens, a trajectory carried into the Ottoman and post-Ottoman worlds (the Parthenon having been a temple to Athena for 800 years, then a church dedicated to Mary theotokis for another 800, then a mosque for 400, then a smaller mosque for another 150 before being preserved as the site of the Athena temple as “a supreme model of eternal glory”); between Western and Eastern Christians in the 12th-Century Levant; between peoples holding native religious beliefs, Hellenistic ones and later, Christian ones in Roman Egypt, and one leading scholar has referred to “the fundamental interdependence of violence and tolerance in the Middle Ages” in France (to quote David Nirenberg). What is important is that we are analyzing a pattern that shows remarkable similarity in widely differing historical, social and religious contexts, of trajectories of transformations through time of particular sites, as revealed by comparing such trajectories at widely varying sites. In addition, we have found that the AT model is very useful for understanding the development of areas of larger scale than a single stratified site: of settlements, towns and cities, to colonies, and even the shifting borders of imperial expansion and contraction.
© 2024, by Robert M. Hayden