Tracking the evolution of literary style via Dirichlet–multinomial change point regression

Gordon J. Ross

Journal of the Royal Statistical Society Series A, 2020, vol. 183, issue 1, 149-167

Abstract: It is typical in stylometry to assume that authors have a unique writing style which is common to all their published writings and is constant over time. Based on this assumption, statistical techniques can be used to answer literary questions, such as authorship attribution, in a quantitative manner. However, the claim that authors have a constant literary style has not received much investigation or validation. We propose a collection of statistical models based on Dirichlet–multinomial change point regression which can capture the evolution of writing style over time, including both gradual changes in style as the author matures, and abrupt changes which can be caused by extreme events in the author's life. To illustrate our framework, we study the literary output of the celebrated British author Sir Terry Pratchett, who was tragically diagnosed with Alzheimer's disease during the last years of his life. Contrary to the usual assumptions made in stylometry, we find evidence of both gradual changes in style over his lifetime, and an abrupt change which corresponds to his Alzheimer's diagnosis. We also investigate the published writings of Agatha Christie, who is also rumoured to have suffered from Alheizmer's disease towards the end of her life, and find evidence of gradual drift, but no corresponding abrupt change. The implications for stylometry and authorship attribution are discussed.

Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1111/rssa.12492

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bla:jorssa:v:183:y:2020:i:1:p:149-167

Ordering information: This journal article can be ordered from
http://ordering.onli ... 1111/(ISSN)1467-985X

Access Statistics for this article

Journal of the Royal Statistical Society Series A is currently edited by A. Chevalier and L. Sharples

More articles in Journal of the Royal Statistical Society Series A from Royal Statistical Society Contact information at EDIRC.
Bibliographic data for series maintained by Wiley Content Delivery ().