Linguistic Evidence of Ageing in the Pratchett Canon

Abstract

A corpus of literary works authored by Sir Terry Pratchett is analyzed from the perspective of linguistic variables which from the literature on natural ageing, one might expect to show effects of gradual change and periods without change. Aspects of lexical complexity exhibit trends that diverge from patterns associated healthy ageing. Background: Past study of linguistic healthy ageing has analyzed corpora of individual professional writers over time and cross-sectional corpora constructed with writing samples from individuals in distinct age groups. Among other effects, negative linear correlations have been found between the use of both first person singular and first person plural pronouns and age and between past-tense verb forms and age; positive linear correlations have been found between cognitive complexity features and age. Main goal: This study seeks to contribute to understanding of whether Alzheimer’s disease is accompanied by the same, accelerated or distinct patterns of change in linguistic features associated with healthy ageing. Method: A corpus of works published by Sir Terry Pratchett is analyzed with respect to pronoun use and linguistic signals of cognitive complexity, such as embedding words and lexical variety, testing correlations between those quantities and author age. Results: The Pratchett corpus exhibits effects in the opposite direction of those associated with healthy ageing for: first person pronoun use and long words. Lexical variety strongly diminishes over the corpus. Conclusions: Linguistic data produced by individuals diagnosed with Alzheimer’s disease appears to contain signals of the fact, when analyzed post hoc. This suggests further work to identify inflection points and to study further features that may be indicative as the linguistic data is produced.