[Humanist] 23.561 more on the stylometry of Christie's dementia

Humanist Discussion Group willard.mccarty at mccarty.org.uk
Tue Jan 12 07:35:23 CET 2010

                 Humanist Discussion Group, Vol. 23, No. 561.
         Centre for Computing in the Humanities, King's College London
                Submit to: humanist at lists.digitalhumanities.org

        Date: Mon, 11 Jan 2010 11:33:21 -0800
        From: "Dr. Katherine D. Harris" <katherine.harris at sjsu.edu>
        Subject: Using Digital Tools to Discover Author's Dementia
        In-Reply-To: <37b656b91001111130m4ac8449dleccd30d6eda3bf50 at mail.gmail.com>

[Here follows a second, somewhat different notice of the attention given to the Lancashire and Hirst project. --WM]

Dear All,

Here's an interesting tidbit about the use of digital tools to attribute
dementia and Alzheimer's to Agatha Christie's later novels. This research
has been circulating all over Canadian and British news outlets as well as
through health organizations.  The primary author's poster & paper are
available here (scroll down some): http://www.chass.utoronto.ca/~ian/


Did Agatha Christie, who wrote several dozen mystery novels during her
53-year career, suffer from Alzheimer's-related dementia? Though some of her
biographers have suspected as much, actual evidence was advanced in March by
a research team led by Ian Lancashire and Graeme Hirst, professors at the
University of Toronto, in a paper called "Vocabulary Changes in Agatha
Christie's Mysteries as an Indication of Dementia."

The professors digitized 14 Christie novels (and included two more available
in the Gutenberg online text archive), and then, with the aid of
textual-analysis software, analyzed them for "vocabulary size and richness,"
an increase in repeated phrases (like "all sorts of") and an uptick in
indefinite words ("anything," "something") — linguistic indicators of the
cognitive deficits typical of Alzheimer's disease. The results were
statistically significant; Christie's lexicon decreased with age, while both
the number of vague words she employed and phrases she repeated increased.
Her penultimate novel, "Elephants Can Remember," exhibits a "staggering drop
in vocabulary" — of 31 percent — when compared with "Destination Unknown," a
novel she wrote 18 years earlier. For Agatha Christie fans, the findings may
be proof of a truth they have long recognized: the author's final two books,
written in her early 80s, do not hold up against her earlier ones.
Christie's body of work lends itself to such analysis because it spans the
bulk of an adult life, from age 28, when Christie wrote her first novel, to
age 82, when she wrote her last. Still, Hirst cautions, "the question is not
early style versus late style, but the late style of someone who is elderly
but healthy versus the late style of someone who is elderly but not
cognitively healthy." To contextualize their evidence, Lancashire and Hirst
plan to analyze the work of P.D. James, a still-healthy writer who has
continued to publish into her 80s, as well as the writings of authors like
Ross Macdonald who are known to have had Alzheimer's. AMANDA FORTINI


Dr. Katherine D. Harris
Editor, Forget Me Not Hypertextual Archive

Assistant Professor
Department of English & Comparative Literature
San Jose State University
One Washington Square
San Jose, CA 95192-0090

Email: katherine.harris at sjsu.edu
Phone: 408.924.4475

More information about the Humanist mailing list