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

This study investigates the stylistic multiverse of Eileen Chang, a distinguished Chinese author and translator, by analyzing her original English writings, self-translations, and translations of others' works through a multifactorial design incorporating translation status (original vs. translated) and translation type (self-translation vs. regular translation). Employing multivariate statistical methods—principal component analysis (PCA) and linear discriminant analysis (LDA)—alongside micro-analysis of the most frequent words, the research reveals distinct patterns of stylistic convergence and divergence shaped by Chang's literary ambition, translation philosophy, and patronage. The findings demonstrate that Chang's original writings emphasize narrative specificity and intimate character engagement, her self-translations show linguistic fluency and adaptability, and her regular translations reflect broader thematic exploration and stylistic adjustment to source texts and audiences. Methodologically, the study highlights the advantages of multifactorial and multivariate approaches in author-translator style research, offering deeper insights and methodological triangulation beyond traditional corpus analyses.