A Stylometric Analysis of Stylistic Features: A Quantitative Study on the Authorship of  <i>Lazarillo de Tormes</i>

Abstract The authorship attribution of La vida de Lazarillo de Tormes y de sus fortunas y adversidades (hereafter, Lazarillo de Tormes ), a representative work of Spanish Golden Age literature, has been a subject of ongoing scholarly debate. Previous studies have approached this issue from various perspectives, including historical documentation, the author’s biography, and analyses of lexicon, narrative techniques, and thematic content. Based on these analyses, Juan de Ortega, Alfonso de Valdés, and Diego Hurtado de Mendoza have been proposed as the most likely authors of this anonymous work. By examining the known classical texts of these authors, a cluster analysis of their habitual use of function words can be conducted. The vectors generated by these particles may systematically reflect the writing habits of the authors, thereby facilitating effective authorship identification in classical Spanish literature. In this study, we perform a cluster analysis using vectors of characteristic function words based on 61 samples of approximately 3,000 words from the three aforementioned writers, along with the Lazarillo de Tormes . The results indicate that the writing habits of the author of Lazarillo de Tormes are closer to those of Alfonso de Valdés. This finding offers a new avenue for further in-depth research into the authorship of Lazarillo de Tormes .