When linguistics meets computer science: stylometry and professional discourse

Abstract

In an 1887 article, The Characteristic Curves of Composition, published in the journal ‘Science’ and resulting from discussions with the logician Augustus de Morgan, Mendenhall (1888) asserted that the length of words was a characteristic capable of distinguishing authors of a literary text. This study is often considered one of the founders of stylometry as a discipline. This paper analyses the development of stylometry and its use as a computerised analytical tool and explores its potential as a way of identifying authors of online professional communication by the vocabulary and style they use. The objective of this paper is to explain the concept of stylometry as an academic methodology, how it has been adapted to computers and how it is used in online investigation of author and narrative identity. The methodology is based on secondary research to explain how stylometry can be used in author definition and attribution and identification of texts. It goes on to analyse the types of stylometric entities and examines the role of computers in stylometry and its application to professional discourse. The study concludes that although the use of computers is an important quantitative tool in stylometric research, in the end it is human judgement that counts.