In the past few years, I have applied automated authorship attribution analysis to several disputed or partially disputed texts within the corpus of Greek drama. Using methods ranging from visualisation to advanced Machine Learning–Artificial Intelligence models, I have explored their potential to unravel questions of authorship in texts such as Rhesus (attributed to Euripides), Prometheus Bound (attributed to Aeschylus), the problematic ending of Aeschylus’ Seven Against Thebes, and a controversial debate scene in Euripides’ Electra.

In this talk, I will present primarily published work on the subject, focusing on the most comprehensive and promising Machine Learning models I have employed. I will delve into the methodology, results, trends, challenges, and far-reaching implications of this approach, as well as highlight unexpected benefits emerging from this form of textual analysis. Additionally, I will address the impact of some of these re-attributions on contemporary theatre practice.

As Michel Foucault insightfully observed, “an author’s name is not simply an element in a discourse… [I]t performs a certain role with regard to narrative discourse, assuring a classificatory function. Such a name permits one to group together a certain number of texts, define them, differentiate them from, and contrast them to others.” This raises a critical question: How willing or prepared are theatre practitioners to detach a work, long attributed to a canonical master, from its traditional place in the celebrated canon and stage it as the creation of an anonymous author?

Poster
Download below