The interest in social networks extracted from literary texts dates back to, at least, the 1990s (Schweizer and Schnegg 1998). Starting with Franco Moretti’s exploratory network analysis of Shakespeare’s , the interest is especially focusing on character networks in dramatic texts (Moretti 2011). At the same time, we have seen the emergence of larger corpora for English (), French (), German () and Swedish drama (), all encoded in TEI.

In our talk we will describe the building of a Russian drama corpus (RusDraCor), along with an aggregated German drama corpus (GerDraCor), both spanning roughly 200 years from the 1730s to the 1930s. They will be presented via a shared eXistdb-driven interface with the aid of respective JavaScript libraries, allowing for the direct visualisation and analysis of inherent structures of the character networks.

Since the dramatis personae, the list of persons usually given at the beginning of a play, is often incomplete (and sometimes non-existent), transforming them into a castList will not convey a complete picture of what we might call the social life on stage. To address this shortcoming, we generate a list of occurring characters in the particDesc section of the TEI header (leaving the original texts untouched). Assigning IDs to every character, including incognito side characters (the likes of A voice) and persons appearing in groups (All), is based on the automatic assignation of IDs, which in some cases is ambiguous. For data correction, we use an innovative gamification approach (Göbel and Meiners 2016).

Our network-related findings will allow us to study the structural evolution of drama (Trilcke et al. 2015, Fischer et al. 2017) and convey a much more complete picture of the social tectonics in dramatic texts, especially by involving hitherto neglected side characters and person groups.