The interpretation of zeta test results: a supplement
Pervez Rizvi
Abstract
Abstract Zeta is a mathematical method which has been used to try to solve authorship attribution problems in both the early modern and modern periods. David Hoover, one of its leading practitioners, has recently made two important claims (Hoover, Zeta revisited, Digital Scholarship in the Humanities, print publication forthcoming). Given his status as an expert on Zeta, these claims are likely to be treated by researchers as authoritative. Most Zeta graphs show a good separation between the base and counter segments. Hoover claims that this is an important finding, rather than, as Pervez Rizvi argued, a mechanical consequence of the method (Rizvi, The interpretation of zeta test results. Digital Scholarship in the Humanities, 34(2): 401–18). Drawing a bisector line is a common technique for interpreting a Zeta graph and, in practice, it produces the same attributions as the ones that Hoover’s less mathematical approach produces. Rizvi argued that this is an unsound technique, an argument that Hoover has now rejected. This paper elaborates Rizvi’s arguments and presents the results of new experiments to show that both of Hoover’s claims are incorrect and should not be accepted.