Detecting Collaborations in Text Comparing the Authors' Rhetorical Language Choices in The Federalist Papers

Authors: Collins J.¹; Kaufer D.²; Vlachos P.²; Butler B.³; Ishizaki S.⁴

Source: Computers and the Humanities, Volume 38, Number 1, February 2004 , pp. 15-36(22)

Publisher: Springer

Buy & download fulltext article:

Price: $47.00 plus tax (Refund Policy)

Abstract:

In author attribution studies function words or lexical measures are often used to differentiate the authors' textual fingerprints. These studies can be thought of as quantifying the texts, representing the text with measured variables that stand for specific textual features. The resulting quantifications, while proven useful for statistically differentiating among the texts, bear no resemblance to the understanding a human reader – even an astute one – would develop while reading the texts. In this paper we present an attribution study that, instead, characterizes the texts according to the representational language choices of the authors, similar to a way we believe close human readers come to know a text and distinguish its rhetorical purpose. From our automated quantification of The Federalist papers, it is clear why human readers find it impossible to distinguish the authorship of the disputed papers. Our findings suggest that changes occur in the processes of rhetorical invention when undertaken in collaborative situations. This points to a need to re-evaluate the premise of autonomous authorship that has informed attribution studies of The Federalist case.

Keywords: authorship attribution; collaboration; federalist papers; statistics

Document Type: Research article

DOI: http://dx.doi.org/10.1023/B:CHUM.0000009291.06947.52

Affiliations: 1: Carnegie Mellon University, USA, Email: jeff.collins@acm.org 2: Carnegie Mellon University, USA 3: University of Pittsburgh, USA 4: Pittsburgh, Pennsylvania, USA

Publication date: 2004-02-01