LANGUAGE CHANGE WITH AGING IN PELHAM GRENVILLE WODEHOUSE AND GEORGE BERNARD SHAW
Paul E Cotter, Catherine Wilkinson, Michelle D. Canavan, Shaun T. O’Keeffe
Journal of the American Geriatrics Society · 2011 · 인용 3
To the Editor: Linguistic examination of literary output has been used for posthumous detective work, with works by Iris Murdoch,1 Agatha Christie,2 and King James I3 showing an exaggerated decline in vocabulary size and syntax from earlier to later works consistent with the early effects of their later dementia. Changes in language use with age were investigated in the works of two authors who remained prolific and successful into advanced old age: Pelham Grenville (PG) Wodehouse (1881–1975) and George Bernard (GB) Shaw (1856–1950). The text of 20 novels by Wodehouse (10 Wooster and Jeeves books published between 1934 and 1974 and 10 Blandings Castle books published between 1915 and 1969) were converted to digital format using commercial software and corrected manually.
The text of 21 introductions or prefaces Shaw wrote to his plays after 1906 (when he was 50 years old) that contained at least 1,500 words were downloaded from the Project Gutenberg Web sites. Language measures that could be automated were examined using Concordance 3.3 (Watt 2009) and WordSmith 5.0 (Scott 2010). The effects of age on linguistic markers in Shaw and Wodehouse were examined separately.
The primary outcomes of interest were the linear relationships between the authors' ages at publication and the language measures. Lexical measures made were proportion of unique words, proportion of vague words, and repetition of “content” words. Unique words (types) as a proportion of total words provide a simple measure of writing style and, in a long sample of writing, a measure of the size of the writer's vocabulary.4 Type numbers at 45,000 words were used to examine Wodehouse's vocabulary size and type numbers at 1,500 words to examine potential changes in Shaw's writing style.
The frequency of the vague, indefinite words “thing,”“anything,” and “something” was examined in all texts.2 Repetition of the 20 most commonly used “content” words (including nouns, verbs, adjectives, and most adverbs but excluding “function” words such as prepositions, pronouns, conjunctions, and articles) within 20 subsequent words was also examined. Mean clauses per sentence (utterance) (MCU), calculated by averaging the total number of clauses (including main, embedded, and subordinate clauses) in a number of sentences, is the most useful and reproducible measure of age-related and individual differences in linguistic complexity and one that correlates strongly with other measures.5 MCU was calculated for each Wodehouse book for the first 20 complete narrative sentences from each of the first, sixth, tenth, and last chapters in each book and for each Shaw preface for the first 80 complete narrative sentences. Twenty-sentence samples from both authors were separated, coded, and reassembled in random order for analysis by a single rater.
This study of works, extending to their early 90s, by Wodehouse and Shaw did not show any decline in language production with age. There were no significant changes in indefinite word use or in repetition in either author (Table 1). Wodehouse's vocabulary size and Shaw's writing style remained unchanged with age.
MCU was unchanged in Shaw and rose slightly but significantly with age in Wodehouse. In contrast, cross-sectional and longitudinal studies in “normal” healthy aging consistently show an age-related decline in language production from the mid-70s, including reduced syntactic complexity and a slight decrease in vocabulary size.6–8 Wodehouse and Shaw share a number of characteristics that make them particularly attractive subjects for examining language changes with optimal aging; their high level of productivity into old age provides a substantial body of work for analysis, and their stylistic consistency reduces the possibility that apparent age-related changes might be due to changes in approach. They had healthy lifestyles for their time—relatively so in the case of Wodehouse, who smoked a pipe and was fond of martinis, and perhaps excessively so in the case of Shaw who was a vegetarian, teetotaler, and nonsmoker—and remained physically and mentally well until close to death.9,10 They are, of course, atypical examples of cognitive aging, but examining how exceptional individuals differ from those with more-usual aging patterns may assist in understanding the intellectual potential of older people and how that potential can be achieved.
Many cognitive changes seen in usual aging result from potentially modifiable factors such as clinical or subclinical disease, lifestyle, and disuse. Wodehouse and Shaw shared a dedication to writing they maintained to the end.9,10 Practice or repetition are critical to the development and maintenance of skills in general,11 and experimental and observational studies of older people confirm that this is true for performance on cognitive tasks.7 This study supports the idea that the maxim “use it or lose it” extends to written language production. It remains unclear whether this may be achieved without regular production of a novel or play in later life.
Conflict of Interest: There are no conflicts of interest related to this study. Author Contributions: PEC, CW, and MC performed manual correction of the texts. STOK analyzed the data.
All authors contributed to writing the paper. Sponsor's Role: This study had no sponsor.