Unraveling Eileen Chang’s stylistic multiverse: insights from multivariate analysis with multifactorial design

This study employs multivariate analysis within a multifactorial design to investigate the author-translator style of distinguished Chinese author Eileen Chang. Using principal component analysis (PCA) and linear discriminant analysis (LDA), it examines Chang’s stylistic multiverse across three literary types: original writing, self-translation, and regular translation of others’ works. The findings reveal diverse patterns of stylistic convergence and divergence among these literary types, analyzed using the metrics of the most frequent words: 200-MFW, 20-MFW, and 5-MFW. These patterns are shaped by the perspectives introduced by two explanatory factors, translation status (original versus translated) and translation type (self-translation versus regular translation), and influenced by Chang’s literary ambition, translation philosophy, and patronage. The study contributes methodologically to author-translator style research by demonstrating the effectiveness of multifactorial and multivariate approaches, which not only provide profound insights into datasets but facilitate methodological triangulation.