The wheel of time: The computational stylometry comparison of Robert Jordan and Brandon Sanderson
This study investigates a closed set of candidates, focusing on a comparative stylometric analysis between the late Robert Jordan, author of The Wheel of Time series, and his successor, Brandon Sanderson, who completed the final three books. Three key research questions are addressed in this study. The first explores whether Sanderson, as the continuator of Jordan’s legacy, sought to emulate his predecessor’s writing style or allowed his own stylistic tendencies to surface while building on Jordan’s material. The second investigates claims found in online forums regarding the authorship of the final book’s epilogue and explores whether it was written by Jordan, Sanderson, or a combination of both. Lastly, the third research question examines other supplementary linguistic characteristics. Using ‘stylo’ package in R to investigate word and character frequencies on 2-gram and 3-gram levels, the final three Wheel of Time books were compared with the twelve books of the same saga authored by Jordan and four of Sanderson’s earlier works. Initial principal components analysis and hierarchical cluster analysis experiments found that the books in question were authored by Sanderson, indicating that he did not act as a ghostwriter but retained his authorial signature. The results for the last book’s epilogue were mixed, with the majority of tests attributing it to Sanderson. The additional linguistic elements examined produced mixed results, with specific lower-level linguistic features showing Sanderson’s voice interwoven with Jordan’s authorial voice. Empirical testing, such as the approach taken in this study, is crucial for refining methods and improving analysis reliability.
Publication Details
- Journal
- Aristotle University of Thessaloniki
- Published
- 2025-01-01
- DOI
- https://doi.org/10.26262/heal.auth.ir.365036
- Primary Topic
- Authorship Attribution and Profiling
- Type
- article
- Field-Weighted Citation Impact
- 0.00