Attributing the Authorship of the Henry VI Plays by Word Adjacency

Abstract

This article reports the invention of a new means of authorship attribution based on the relative distances (measured in intervening words) between functions words in the texts being analysed. The data about distances are stored in Markov chains (with each function word represented as a node and the summed distance between a pair of words being represented as a weighted edge between their respective nodes) called Word Adjacency Networks (WANs). The method is applied to the problem of determining the authorship of the three plays called Henry VI Parts One, Two, and Three that are traditionally attributed to William Shakespeare, and we find that Christopher Marlowe had a hand in each of them.