Towards the Dissolution of Religious Violence in Late Antiquity
The traditional narrative traces an ascendant line of religious violence in late antiquity. The support given to Christianity by Constantine leads to other religions being progressively driven out of society.
The process is reflected in an increasingly strict legislation, culminating in the outlawing of pagan acts by Theodosius I (391-2 ce) and accompanied by extrajudicial attacks by Christian groups on cult places. The fractious nature of Christianity, in turn, entails deep conflict between different groups claiming orthodoxy, resulting in numerous clashes, which are exploited by bishops to establish their own authority. The process reaches a climax in the reign of Justinian (527-65 ce), whose legislation seeks to eliminate all dissent. With its double emphasis on legislation and violence, this narrative sees the identification of church and state as a legacy of late antiquity - a knot that only modernity would disentangle.The sources provide, at first glance, much confirmation of this narrative. One regularly encounters stories of violent destructions and clashes, seemingly backing up the idea of widespread religious violence. Yet, scholars recently have grown sensitive to the limits of the evidence. First, the limited number of instances of religious violence are highlighted. For example, ancient accounts of destructions of temples, such as that of the Serapeion in Alexandria (391 ce), of Marnas in Gaza as reported in the life of Porphyry (possibly early fifth century), or the ones alleged by Libanius in his oration For the Temples (381-92 ce), are not very numerous. Moreover, the archaeological evidence does not bear out a picture of widespread destruction of places of worship. This does not mean that no destruction took place, but the act may be rarer than Libanius' highly evocative image of monks descending as locusts on the Syrian countryside to destroy temples might suggest.[1023] Wolfram Kinzig has made a similar argument for anti-Jewish violence.
He counts twenty violent episodes from the fourth to the sixth centuries ce, of which fourteen can be reckoned to be certain. Most of these took place in Antioch, Syria and Palestine. He points out that if one adheres to the lachrymose view of Jewish history in late antiquity, this may be sufficient proof of extensive anti-Jewish violence, but one may also understand it in the light of a more complex view of Jewish-Christian relations, marked by fascination and rejection, tension and cohabitation.[1024] Depending on the narrative one adheres to, then, the attested instances of violence may be the tip of iceberg or isolated cases of escalation. We should therefore reflect as much on the narrative as on the evidence.Second, almost all our evidence is of a literary nature and thus shaped by rhetoric and tropes. Pretending to report violence, our sources produce particular representations that conform to particular stereotypes and serve particular purposes. Hagiography, for example, is full of stereotypes about the anti-pagan struggle, of which the destruction of idols and temples is one. These cannot be taken at face value. Indeed, contrary to the impression that his own writings might generate, there is only one certain case of temple destruction by the famously aggressive Egyptian abbot Shenoute (d. 465 / 466 ce).[1025] Rhetoric does not occur just in Christian texts. In his discourse For the Temples, Libanius consciously creates a picture of widespread destruction of temples by monks, fully knowing that the act was not allowed by law and that monks had a rather dubious reputation, also in Christian circles. As I have argued elsewhere, this text is not the eloquent testimony to pagan concerns about temple destruction for which it has been held, but a clever and subtle defence of elite interests by creating a picture of extensive illegal activity by monks.[1026] If we do not contextualise each text carefully, we risk mistaking rhetoric for reality.
Increasingly aware of the limits of the traditional narrative and of the evidence, scholars have been exploring new avenues of research. One way is to embed episodes of violence in a broader social context. Acknowledging the fact that late antiquity was characterised by the coexistence of a variety of religious groups (both within Christianity and without), scholars seek to understand violence as one possible outcome of a wider social dynamic. Options abound: one can relate religious violence to pre-existing social tensions in cities and understand it as grafted onto different and older conflicts; one can study the formation of new identities in late antiquity and stress the tension between ordinary people with their multiple identities (civic, social, and religious) and the various religious groups that sought to promote membership in their group as the single overriding identity; or one can argue that late antiquity sees the continuation of the religious competition that marked the Roman Empire, but now with different groups dominating. All these approaches still start out from the assumption, seemingly borne out by the literary evidence, that religious identities became dominant in late antiquity.11 Yet, as a survey of conflict documented in Egyptian papyri from late antiquity has shown, religion is rarely a factor in the kinds of social conflicts that papyri document.[1027] [1028] This suggests that religion had not yet become a primary identity for many and puts into perspective the numerous literary texts that suggest otherwise. The social landscape of late antiquity is not yet that of the medieval Middle East in which religious and social identity merge.
This brings us to the second approach, namely the contextualisation of religious violence in a wider history of violence. We should start out from the fact that various forms of violence were present in late antique society. Correction of subordinates (slaves, children, women) often included corporal punishment.
Equally, children were regularly beaten in school. The threat with, or the effective application of, punishment was part of a strategy of reform and re-education of offenders. The absence of a monopoly of violence on the part of the Roman state and the relatively thin presence of imperial representatives in the provinces also allowed powerful individuals to pursue their own interests: the sources are full of, undoubtedly exaggerated, criticism of rapacious tax collectors, unjust governors and looting soldiers. These were structural features of life in the later Roman Empire which were often complained about but rarely remedied. If these sorts of violence help to understand why contemporaries had a lower sensitivity for violence than we do, an important parallel for religious violence is the recurrent popular riots. One can cite famous episodes such as the killing of the charioteer in Thessalonica that was the cause for the execution of thousands of citizens on orders of Theodosius I (390 ce), the so-called riot of the statues in Antioch three years earlier, when its inhabitants protested against tax rises, and the long list of popular riots, often taking place in the hippodrome, in Constantinople. The Nika riot (532 ce), which nearly cost Justinian his throne, is the most famous example. If some of these could be religious in motivation, like the anti-pagan riots under Tiberius in 580 ce, they could be triggered equally well by famine or oppressive administrations. If scholarship traditionally considers the elite to be the real instigators of such riots, and in the case of religious violence usually blames bishops, the capacity of the people to be an independent actor should not be underestimated. There is plenty of evidence for popular riots that should be considered spontaneous group actions, like the revolt of the people of Constantinople when the popular bishop John Chrysostom was exiled in 403 ce. Such revolts could be very violent, resulting in lynch justice. Although it has not been exhaustively analysed, there is clear evidence that such popular riots drew on a repertoire of actions, which usually mimicked official public rituals. For example, the parading of heads on spikes, the dragging of a criminal through the streets and his cremation and burial outside the city, and the destruction of pagan or imperial statues can all be paralleled in the official repertoire, for example that of victory celebrations and the damnatio memoriae. Interestingly, as far as I can judge, no typically religious repertoire is found in Late Antiquity as it is later periods of European history: popular riots are not drawing on a repertoire linked to Christian ritual or theology. In this respect, popular ritualised behaviour Christianises at the equally slow pace as official public ritual, which only in the second half of the sixth century acquires an explicitly Christian face.A third context is that of language. It is often assumed that aggressive language leads to violence. On that premise, a high rate of violence is proven by the often highly polemical, not to say insulting tone in which many late antique texts are couched. Such a relation is never straightforward, and definitely not so for antiquity. One should remember that vituperation is an integral part of ancient rhetoric, as any reader of speeches by Cicero and Demosthenes knows: genre and rhetorical rules may shape creation as well as reception of speeches. Societies at large but also specific subfields have particular metaphors that seemingly unconsciously dominate language: modern medicine, for example, often draws on the language of warfare to describe the battle against bacteria and viruses. Within Christianity, in turn, agonistic and military language abounds to describe spiritual progress. If it is obviously possible that such language reverts to its original meaning and becomes a trigger for violence, there is no reason to assume it always does and that military metaphor is, in itself, a cause of violence.
Indeed, violent language can coexist with other discourses: as we shall see, late ancient Christianity also emphasised peacefulness when it came to actions. Finally, in some circumstances the use of violent language could be a way to deescalate a real threat of violence, as we shall see below when martyrdom is discussed.[1029] Overall, there has been little socio-linguistic research done on violent language in late antiquity which could shed more light on these issues.These various new avenues in scholarship provide a more profound contextualisation of religious violence and lead, in effect, to a dissolution of that very concept. Once freed of the idea of an all-encompassing religious violence that is born in late antiquity, we may be able to return to the evidence and reassess it with fresh eyes. Indeed, notwithstanding all the nuances we have just added, late antique sources clearly are preoccupied with violence. I do not wish to argue that late antique sources are more or less fascinated by it than classical sources. Rather, I suggest the late antique perception of violence has two specific characteristics. The first is the creation of a number of highly charged categories to describe reality, in particular ‘martyrdom', ‘persecution' and ‘sacrifice'. I propose that Christian sources identify these as specifically religious, or more appropriately, sacred forms of violence, which are polemically ascribed to one's enemies (a proper Christian does not create martyrs, persecute, or sacrifice). Although ‘martyrdom' and ‘persecution' are regularly used as objective descriptors in a scholarly vocabulary, we should recognise that they are deeply relational, in that they describe what one's enemies do. Given the high degree of conflict in late antiquity, it should not surprise us that we see accusations of this type of violence proliferate. Second, these accusations play out against a background of the moral condemnation of violence. Violence was problematic and this can be noticed even in reports that seek to justify it. This does not mean that Christians did not engage in violence (man does contravene moral injunctions), but their justificatory strategies play out against a background of moral condemnation. I shall discuss these two elements in turn, each time beginning with a famous episode.