Testing Rhythmic Stylometry with a Corpus of Lope de Vega’s Securely Attributed Comedias

Authors

DOI:

https://doi.org/10.24197/6dkssk45

Keywords:

rhythmic stylometry, authorship attribution, Lope de Vega, Spanish Golden Age theatre, machine learning

Abstract

This article evaluates the discriminative capacity of rhythmic stylometry in a controlled setting for authorship attribution in Spanish Golden Age theatre. Using a binary corpus of securely attributed comedias by Lope de Vega contrasted with a large set of non-Lope plays, works are represented through frequencies of stress-based rhythmic patterns obtained via automatic scansion. Supervised classification models are applied, and performance is shown to concentrate in a window of frequent patterns, approximately between 100 and 200, reaching F1 values close to 95%. Although the results are promising, error analysis reveals recurrent false negatives, including two autograph plays, indicating room for improvement and recommending interpretive caution when applying the method to disputed cases.

Downloads

Download data is not yet available.

Author Biography

  • Álvaro Cuéllar González, Autonomous University of Barcelona

    Álvaro Cuéllar is a postdoctoral researcher at the Autonomous University of Barcelona, ​​working in digital humanities applied to Golden Age theater. His research focuses on computational stylometry and authorship attribution, combining statistical methods and machine learning to analyze lexical, metrical, and stylistic features in large theatrical corpora. He also develops and applies artificial intelligence techniques for the automatic transcription and processing of printed and manuscript texts, aiming to expand and improve the materials available for research. Since 2017, he has co-led the project "Stylometry Applied to Golden Age Theater," an international collaborative initiative working with a corpus of approximately 3,000 plays and bringing together more than 120 researchers. He is the creator of TEXORO, one of the leading textual databases for the study of Golden Age theater (40 million words), used by specialists. She has presented her findings at conferences and congresses in Europe, Asia, and the Americas, and her work is part of numerous research projects funded in various countries. Her output includes monographs, articles, and widely used digital resources, with over 500 academic citations.

References

Antonucci, Fausta (2020). “La polimetría en La dama boba: funciones poéticas y dramáticas”. En Javier Espejo Surós y Carlos Mata Induráin (eds.). Preludio a «La dama boba» de Lope de Vega (historia y crítica). Pamplona: Servicio de Publicaciones de la Universidad de Navarra, pp. 111-129.

Artelope (2011-2026). Base de datos y argumentos del Teatro de Lope de Vega. Joan Oleza et al. (dirs.). Universitat de València, https://artelope.uv.es [10/01/2026].

Cuéllar, Álvaro (2022). Propuestas de autoría y datación en el teatro de Lope de Vega a la luz de la Inteligencia Artificial. Tesis doctoral (PhD). University of Kentucky. DOI: https://doi.org/10.13023/etd.2022.428.

Cuéllar, Álvaro (2023). “Cronología y estilometría: datación automática de comedias de Lope de Vega”. Anuario Lope de Vega. Texto, literatura, cultura, 29, pp. 97-130. DOI: https://doi.org/10.5565/rev/anuariolopedevega.483.

Cuéllar, Álvaro (2024). “Stylometry and Spanish Golden Age Theatre: An Evaluation of Authorship Attribution in a Control Group of One Hundred Undisputed Plays”. En Robert Hesselbach, Juan Calvo Tello, Ulrike Henny-Krahmer, Christof Schöch y Daniel Schlör (eds.). Digital Stylistics in Romance Studies and Beyond. Heidelberg: Heidelberg University Publishing, pp. 101-117. DOI: https://doi.org/10.17885/heiup.1157.c19368.

Cuéllar, Álvaro (2025a). “From Atoms to Waves: Rhythmic Stylometry for Authorship Studies of Early Modern Spanish Theatre”. Janus. Estudios sobre el Siglo de Oro, 14. DOI: https://doi.org/10.17979/janus.2025.14.12653.

Cuéllar, Álvaro (2025b). Rhythmic Corpus of 100 Spanish Golden Age Plays [dataset]. Zenodo. DOI: https://doi.org/10.5281/zenodo.17537343.

Kroll, Simon (2022). Sonido y afecto en Calderón. Un estudio de las asonancias. Kassel: Edition Reichenberger.

Kroll, Simon y Fernando Sanz-Lázaro (2022). “Romances teatrales entre Mira de Amescua, Calderón y Lope: ritmo, asonancia y cuestiones de autoría”. Revista de Humanidades Digitales, 7.1, pp. 1-18. DOI: https://doi.org/10.5944/rhd.vol.7.2022.31620.

Kroll, Simon y Fernando Sanz-Lázaro (2023). “Ritmo e inteligencia artificial: nuevas perspectivas sobre teatro lopesco desde las humanidades digitales”. Anuario Lope de Vega. Texto, literatura, cultura, 29, pp. 351-375. DOI: https://doi.org/10.5565/rev/anuariolopedevega.491.

Kroll, Simon y Fernando Sanz-Lázaro (2024). Versos y estructuras teatrales áureos: A corpus of Spanish plays with dramatic and metric information. PHAIDRA (Universidad de Viena). DOI: https://doi.org/10.25365/phaidra.471.

Marco, Guillermo, Javier de la Rosa, Julio Gonzalo, Salvador Ros y Elena González-Blanco (2021). “Automated metric analysis of Spanish poetry: Two complementary approaches”. IEEE Access, 9, pp. 51734-51746. DOI: https://doi.org/10.1109/ACCESS.2021.3069635.

Marco, Guillermo y Julio Gonzalo (2021). “Escansión automática de poesía española sin silabación”. Procesamiento del Lenguaje Natural, 66, pp. 77-87. DOI: https://doi.org/10.26342/2021-66-6.

Morley, S. Griswold y Courtney Bruerton (1968). Cronología de las comedias de Lope de Vega. Con un examen de las atribuciones dudosas, basado todo ello en un estudio de versificación estrófica. Madrid: Gredos.

Navarro-Colorado, Borja (2018). “A metrical scansion system for fixed-metre Spanish poetry”. Digital Scholarship in the Humanities, 33.1, pp. 112-127. DOI: https://doi.org/10.1093/llc/fqx009.

Pedregosa, Fabian et al. (2011). “Scikit-learn: Machine Learning in Python”. Journal of Machine Learning Research, 12, pp. 2825-2830.

Presotto, Marco (2000). Le commedie autografe di Lope de Vega: catalogo e studio. Kassel: Reichenberger.

Sanz-Lázaro, Fernando (2022). “Planteamientos digitales del drama aurisecular: automatización del análisis métrico de obras teatrales”. En Íñigo González Cabeza, Érica Redruello Vidal y Raquel de la Varga Llamazares (eds.). La escritura en el tiempo: Investigaciones en torno a la literatura hispánica. León: Universidad de León, pp. 109-128.

Sanz-Lázaro, Fernando (2023a). Ritmo y estructura de la comedia áurea: Posibilidades y límites del análisis digital automático del teatro español clásico (tesis doctoral, Universität Wien). DOI: https://doi.org/10.25365/thesis.75102.

Sanz-Lázaro, Fernando (2023b). “Del fonema al verso: una caja de herramientas digitales de escansión teatral”. Revista de Humanidades Digitales, 8, pp. 74-89. DOI: https://doi.org/10.5944/rhd.vol.8.2023.37830.

Sanz-Lázaro, Fernando (2024). “libEscansión: A recursive precedence approach to metrical scansion”. Digital Humanities Quarterly, 18.3, https://github.com/fsanzl/libescansion [10/03/2026].

Vega García-Luengos, Germán (2021). “Las comedias de Lope de Vega: confirmaciones de autoría y nuevas atribuciones desde la estilometría (I)”. Talía. Revista de estudios teatrales, 3, pp. 91-108. DOI: https://doi.org/10.5209/tret.74625.

Downloads

Published

2026-07-22

Issue

Section

ARTICLES

How to Cite

Testing Rhythmic Stylometry with a Corpus of Lope de Vega’s Securely Attributed Comedias. (2026). Castilla. Estudios De Literatura, 17, 114-135. https://doi.org/10.24197/6dkssk45