Social Networking in the Age of Austen, Trollope and Dickens
by Nick Obourn
Mr. Darcy and Elizabeth Bennet on Facebook? With social networking the hot topic of the day, a computer science grad student, his advisor and a literature professor teamed up to analyze social interactions in 19th century British novels.
|A diagram of social networking in Jane Austen’s Mansfield Park created by software developed by David Elson, a Ph.D. candidate in computational linguistics who is interested in using technology to shed light on narrative|
David Elson, a Ph.D. candidate in computational linguistics and a longtime film buff, has long been interested in the intersection of narrative and computers.
“I started thinking about how storytelling works as a language, with a syntax that we pick up as children and that differs from culture to culture,” says Elson. “We are starting to be able to write programs that can learn a language by reading a lot of it—why can’t a program learn the meanings of stories the same way?”
This line of thinking led Elson, who is part of the Natural Language Processing Group in the computer science department at the engineering school, to create a computer program in 2009 that could “read” for dialogue in digitally scanned novels and create social networking maps.
Diagrams of the social networks resemble connected thought bubbles, where the size of the bubble denotes the amount of dialogue spoken by a particular character. “Work in this field—digital humanities—focused on the word level— how often a single word appears over the centuries, for example,” says Elson. “I wanted to look more broadly at social interaction that takes place through quoted speech.”
Elson and his advisor, Kathleen McKeown, the Henry and Gertrude Rothschild Professor of Computer Science, knew they needed help from a literature professor to see if the program worked. With a little social networking of their own, they enlisted Nicholas Dames, Theodore Kahan Associate Professor in the Humanities, an expert in the Victorian era.
Luckily, 19th century literature proved to be perfect for the project since large numbers of books from the period are out of copyright and have been digitized. “For a literary scholar, it’s like a room full of new toys to play with, and no one ‘owns’ those toys,” says Dames. “Then the question is: What are we going to do with these things?”
The answer was to use Elson’s program to try to analyze a longstanding literary theory that Victorian novels set in the city have more characters, looser social networks and less dialogue than those with country settings.
Dames cites the work of Raymond Williams, a Welsh-born Cambridge University literature professor who pioneered the idea. “He developed a series of extremely persuasive arguments about how, in the 19th century, novelists began to imagine urban social interaction as fundamentally different—more dispersed, accidental and fleeting—than the kinds of social interactions found in village or rural settings,” says Dames. “His arguments were based on very elegant readings of a select few authors—most notably Austen and Dickens—and they quickly became fairly standard, canonical theories.”
In contrast, the team worked with 60 19th century novels, totaling more than 10 million words. The list included books by Charles Dickens, Jane Austen, Charlotte and Emily Brontë, Sir Arthur Conan Doyle, George Eliot, Anthony Trollope and Thomas Hardy.
Contrary to Williams’ hypothesis, the analysis found that social networks in rural and urban setting were about the same size and had the same levels of interaction. This past summer, the team presented their findings at the annual conference of the Association for Computational Linguistics in Uppsala, Sweden, where it received the award for best student paper.
Heartened by the positive response, they have decided to expand their efforts and graph how social networks change in novels as the plot unfolds. This would be “a way of thinking about how plot generates or retards or shapes social connections,” says Dames.
Elson, who has a job lined up at Google after he graduates in May, is very interested in the potential of bridging literature and technology.
“We’ll soon be able to trace the spread of ideas from culture to culture and put common assumptions about our heritage to the test,” he says. “We’re not losing the close read—we’re gaining the power to instantly get a sense of where a text or an idea fits in the big picture. We’re just starting to peek at what will be a new expanse of scholarship possibilities.”