Building a Database
This project used Gephi v.0.8.2, an open-source social networking program, to analyze one of Foigny’s cartularies, Paris BN Lat 18374, available in an edited and summarized form created by Edouard de Barthélemy, one of the descendants of the monastery’s founding bishop, in 1879. The cartulary contains approximately 700 entries from the monastery’s founding in 1121 to approximately 1300, although there are a half-dozen or so later entries from around 1400.
First, the relevant information had to be extracted from the cartulary. Only about half (342) of the cartulary entries were used. The temporal scope was limited to the original pre-1300 entries, and only entries which related to the donation, confirmation, or sale of items or rights were included. The remaining entries, which dealt with subjects such as judgments rendered in favor Foigny or equal exchanges of parcels of land with other monasteries, were excluded because they were functionally ambiguous with respect to the nature of the relationship.
The result was a database of those who provided support for the abbey, both those who engaged with the abbey directly and their facilitators – bishops, clerks, witnesses, and suzerain lords who served as guarantors. The 342 charters used contained a total of 374 unique actors associated with 188 separate toponyms, mostly from the immediate vicinity.
The primary difficulties in the creation of the database stemmed from the source text. De Barthélemy summarized, rather than translated, most of the original Latin of the manuscript in French, and simply transcribed the others. In doing so, he standardized neither personal names nor toponyms, leaving a great deal of work to be done on this score. Additionally, de Barthélemy often used deliberately anachronized or ambiguous language, leaving substantial uncertainty as to the precise nature of an interaction. Comparison of de Barthélemy’s work to another extant Foigny charter from the twelfth century, Paris BN Lat 18373, available in digital facsimile online, shows that this ambiguity is most likely not inherent to the original.
Of secondary note is the weighting of relationships between actors. Various schemas were tested to assign importance to interactions, but all were discarded as too problematic due to the ambiguity and inconsistency noted above and the inability to create suitably discrete categories. The one piece of data about a relationship which was preserved was its directionality, that is, in terms of an individual, who was providing support for whom.
The data from the cartulary was analyzed in two ways: spatially and relationally. The spatial analysis aggregated individuals based on the toponym with which they were associated, plotted those toponyms based on latitude and longitude, and then marked the network of interactions between those points. This analysis showed the geographic scope of Foigny’s friendship network and, when broken down by modularity (i.e. sub-networks determined by computer algorithm) the geographic reach of each those networks. The relational analysis dealt with interactions between groups centered on individual actors. Since the data were aggregated over approximately 200 years, the relational analysis could tell little about the day-to-day interactions of Foigny’s entire friendship network, but by looking at the central figures within each sub-network, it revealed the truly important actors.