The speakers speak online.
Abstract:
This talk presents a novel computational study of the gracioso archetype (servant figures whose metatheatrical and humorous interventions both lighten and critically frame the drama) in Calderonian plays. Drawing on a richly annotated XML‑TEI corpus from Calderón Drama Corpus (CalDraCor), we implement two complementary experiments. First, we apply social network analysis to extract character interaction networks from dialogue and scene co‑occurrence data, computing different centrality degrees to assess the strategic positioning of graciosos within the play’s social fabric. Second, we develop a supervised classification model that distinguishes graciosos from other characters, leveraging lexical and stylistic features to capture distinctive patterns. By integrating quantitative network metrics with machine‑learning‐based identification, this talk offers a holistic perspective on how the gracioso functions narratively and structurally in Calderón’s plays. Expected outcomes include insights into stylistic markers, and quantitative evidence of the gracioso’s role as a critical intermediary and social connector. Our interdisciplinary methodology advances both literary scholarship and digital humanities practice, providing a replicable toolkit for analyzing character archetypes and their dramatic significance across theatrical corpora. This methodological framework can be adapted to other early modern repertoires, facilitating comparative studies of dramatic characterization and genre evolution.
Short bio:
Antonio Rojas Castro is a Research Assistant at the University of Tübingen and the Freie Universität Berlin. He holds a Ph.D. in Humanities from the Universitat Pompeu Fabra, where he specialized in digital scholarly editions. His current research applies network analysis to theater and develops open‑source, reproducible methodologies that bridge literary studies and open‑science principles.
Allison Keith is a Doctoral Candidate at the University of Stuttgart. Her project focuses on utilizing machine learning and natural language processing techniques to conduct quantitative analysis on theatrical works.