I’ve recently returned from a fantastic workshop at the American Academy in Rome, organised by Ekaterina Nechaeva. Ekaterina, a EURIAS fellow at the Collegium Helveticum, works on a decidedly understudied topic, refugees from the late Roman empire. The workshop in Rome looked at emigration in a wider sense, for Ekaterina had managed to assemble a group of scholars ranging from ancient historians to historians of the 19th century. To keep this wide chronological scope manageable, Ekaterina asked us to focus on what emigration meant to ‘sending societies’, rather than to the societies that received émigrés. For example, did governments of states from which people migrated try to encourage or prevent emigration? What effect did emigration have on those who stayed behind? How did emigration shape public opinion in ‘sending societies’?

When Ekaterina asked me to participate, I was at first a little sceptical on what the project could contribute. Exile is of course not the same as emigration, and, in addition, late antique clerics were rarely sent, or went, to outside the Roman empire. But giving a paper in Rome is too good an opportunity to miss, and I also thought this would be a great occasion to put to work our database and see whether it could come up with a few statistics that would help answering some of Ekaterina’s questions. The database now contains a wealth of material that can be explored quantitatively (some figures below). More importantly, over the last two months or so, our IT technicians have made some great progress with the database’s functionality. If anything, the workshop was an opportunity to showcase this work. For my presentation I began with demonstrating a few of the most striking visualisations that our database can do, which I will also cover in this blogpost. A second blog addressing more specifically the attitudes towards clerical exile in ‘sending societies’ can be found here.

Before the talk…many thanks to Ekaterina for the invite, and for taking this picture!

 As mentioned before on this blog, we are of course recording not just an exiled cleric’s personal circumstances, but also the individuals and groups they were in contact with. One of our main aims is to visualise and analyse these connections, to test the hypothesis that these (at times forced) interactions disseminated ideas, practices and types of behaviour. The image below shows all connections that we have recorded so far; a ‘global’ network of clerical exile in late antiquity. There are clearly some distinctive clusters of connections, which can be inspected further, such, as, for example, the social network of John Chrysostom, circled in red (who, as is well known, was in contact with over 200 people while in exile)


As ever, it is of course important to remember that the network graph does not show all social connections created through clerical exile, but only those that are remembered in our sources. What we are actually mapping in our database is the textual commemoration of exile. We cannot assume that exile cases that just appear as a single dot in this network did not generate social relationships (and that these clerics were hence solitary or isolated figures). We must always assume that for some reason the recording of social relationships (and the subsequent transmission of these records) may have been more important for some exiled clerics and for some periods of time than for others. This is important to bear in mind always when applying social network analysis to historical data, although as I tried to explain in my presentation in Rome as well, applying network analysis to our data can help us to understand patterns of discourses as well. More on this in the next blog post.

The global network graph shows quite nicely as well that, even though this is a network that spans c. 300 years of history, there were connections that bridged generations. This is above all the case where clerical exile generated rival successions of bishops and distinctive lines of authority which then generated even more exile cases down the years. This was the case, for example, in sixth-century Egypt, as this table from Stephen Davis’ The Early Coptic Papacy, adorned with my additional comments, shows.

In 6th century Egypt theological controversy (pro & anti-chalcedonian) led to the splintering of the church, but it wasn’t just theological controversy that did this: the problem was hugely acerbated by the exiling of bishops which created schisms in Alexandria, but also by the arrival of exiled clerics in Egypt (Severus of Antioch, Julian of Halicarnassus), which brought new ideas and followings. Davis’ table neatly (and very helpfully!) visualises these different groups as four distinct blocks, each with their own list of bishops and traditions. As I showed in Rome, the clerical exile database is able to visualise this complex situation differently, as network graphs, which instantly shows how much these different groups were in contact with each other (Severus of Antioch and Julian of Halicarnassus were famously living at the same monastery outside Alexandria). This is perhaps a helpful reminder that their respective ideas and teachings, even though often opposing, may have been developed in dialogue with each other. Importantly our network visualisations classify relationship types in much detail (e.g. whether someone was a companion, host or correspondent of an exiled cleric). Another category, not shown here, is relationship ‘quality’: whether someone was supportive of an exiled cleric or in conflict with him. This allows users to understand the nature, and perhaps also the effect, of different kinds of networks an exiled cleric was part of.

[Please note that the arrows all point in the wrong direction. The visualisation of different types of relationships through different colours is also not fully realised yet. These are just two of the many fiddly things we still need to sort out; the main reason why the database is not online yet].

Another general feature of the database that I demonstrated in Rome was the possibility to visualise the spatial aspects of clerical exile. This is important because exile divided people spatially, but it also brought people closer together. That is impressively demonstrated in the map below (where possible, we use the fantastic Pleiades Project data). The interactive map function of the database allows exploring the geography of exile in a number of ways. For example, users will be able to visualise locations as clusters (e.g. how many locations connected to clerical exile were in Egypt) or as points, select certain time spans, or explore locations according to so-called location ‘activities’, i.e. whether they were places of destination, arrival, places of return or places of exiled clerics’ contacts. The map here shows the departure and destination locations of exiled clerics in the years 355 to 362, between the council of Milan and the death of Constantius II, and the connections between them. During these years more than 50 bishops and clerics were exiled. The graph shows the immense mobility in the Mediterranean created by clerical exile in these years. In fact, these seven years see the most intense contacts between the Eastern and Western Mediterranean created by clerical exile (and exile generally) in the entire period that we are studying. They are more intense than the contacts generated in the twenty years after the council of Nicaea. This is also the last time clerics were exiled between East and West. After the death of Constantius and up to the time of Justinian no exiled cleric was apparently sent across the Mediterranean anymore. It can hence be argued that these seven years were a crucial window in terms of spatial mobility of theological ideas, and in this case specifically Nicene ideas (although of course ideas may have travelled through different routes afterwards, such as through trade, council attendance, pilgrimage or letter writing).
The database also allows users to create ‘ego-maps’ that track the movement of and locations connected with individual exiled clerics. The map below shows the ‘ego-map’ of Lucifer of Cagliari, exiled after the council of Milan in 355. The red arrows are actually my own subsequent manipulation of the powerpoint slide, but even without these it is easy to follow how Lucifer moved, because each location is assigned a colour that corresponds to a certain ‘activity’ (departure, arrival, return, location of previous office, and, importantly, locations of Lucifer’s contacts). Interestingly, for Lucifer, we can see some frenetic criss-crossing of the Mediterranean and he seems to have mobilised much contact as well. There are an astonishing 11 locations connected to his exile. This mobility is quite comparable to other Western bishops exiled in this period as well, who rarely seem to have stayed in one place (Eusebius of Vercelli, Hilary of Poitiers, Dionysius of Milan).

 We can assume then that Lucifer had ample opportunity to forge new relationships while in exile. However, when we think about the spatial aspects of exile and the opportunities of making contacts, it is important to understand the nature of locations as well: did exiled clerics reside in cities or on islands, in quarries or in oases? How well connected where these locations to other locations? Our map feature contains a layer of roads (again provided by the Pelagios Project) that will help users understand socio-geographical dynamics, but the database also contains much extra information about locations. The graph below is an example of another feature of the database, the ability to visualise data as charts, and shows the distribution of types of exile arrival places (we distinguish between ‘destination’, where clerics were sent, and ‘arrival’, where they actually went to, for these may differ). Interestingly, ‘city’ is the most popular type of exile arrival place.

After all these fancy forms of visualisations, it should perhaps be made clear that all our data will be able to be accessed in conventional forms too, through tabular listings of results. There is much information about each exile case, including biographical information, information on legal circumstances, writings in exile, doctrinal activities and so on (all database categories are listed here). Below you can see the top of the table listing all information the database contains on Arius’ exile.

We hope, however, that our users will be tempted to try out all the other features of the database too, be this displaying data as charts, maps or network graphs (you can see in the image above that users will be able to get to Arius’ network graph and also to his ‘ego-map’ from this tabular result). Network graphs, in particular, we believe have the potential to generate new insights into the effects of clerical exile, as I have already discussed in my contribution to Clerical Exile in Late Antiquity. Yet, as I have also already discussed on this blog, the biggest benefit of recording data on social connections for this project has been that we start to see people whose life was affected by clerical exile beyond the exiled cleric himself. That, as I hope to show soon in the next blog post, also allows us – at least to some extent – to say something about Ekaterina’s ‘sending societies’.