Quantitative Authorship Attribution and the Bakhtin Circle: The Problem of Authorship in Intellectual Communities

One of the central issues in Bakhtin studies is the Bakhtin’s authorship question, which concerns the attribution of texts signed by V.N. Voloshinov, P.N. Medvedev and I.I. Kanaev to M.M. Bakhtin. The problem does not involve isolated cases but rather a significant body of works – ranging from short reviews and essays to major “programmatic” texts. In this study, we address the problem of establishing the authorship of these disputed texts using the Delta method, which compares intertextual distances calculated from the distribution of the most frequent words. Under this approach, works by the same author tend to exhibit the smallest distance values. Revious researchers have suggested that the similarity between the disputed texts and Bakhtin’s undisputed writings might be explained by their belonging to the same intellectual circle. We test this idea empirically and conclude that the influence of a scholarly school does not erase the author’s stylistic “fingerprint”: even within close intellectual communities, stable individual stylistic traits persist. We test the existence of a “school effect” using materials from four intellectual communities: the Geneva School of Linguistics, the Moscow Phonological School, Jena Romanticism, and OPOJAZ. The research corpora include reliably attributed texts. Applying this method to the disputed texts of the Bakhtin Circle demonstrates that the resulting distance map and nearest-neighbor topology provide further grounds for including several key disputed works in M.M. Bakhtin’s corpus.