Is the Future of English Bad Science?


Last week, I enthused about the growing field of “psychonarratology,” which uses methodologies from cognitive psychology and neuroscience to interpret literature. (I, for one, would be interested in learning what tangible effects Mr. Darcy’s motivations have on me.) But even the most promising rose has its prickle: psychonarratology isn’t always about borrowing methodologies. Sometimes it’s the findings from psychology and neuroscience experiments that make their way into literary scholarship.

A 2010 New York Times article reported that psychologists who study theory of mind, or the ability to ascribe mental states to oneself and others, have argued that people can easily track three different mental states at once (e.g., “Peter said that Paul believed that Mary liked chocolate”). But comprehension of these sorts of sentences falls precipitously once people are required to keep track of four or five mental states (“Peter said that Paul believed that Suzanne mistakenly thought Gillian commented that Mary liked chocolate”). This, according to Lisa Zunshine, an 18th-century British literature scholar at the University of Kentucky, is what makes reading the works of Jane Austen and Virginia Woolf so challenging; their novels require the reader to track as many as six different mental states.

Zunshine is using scientific evidence to advance her claim just as one might use historical or biographical evidence, and to be clear, there’s nothing wrong with this. The problem is, extending laboratory findings willy-nilly to other tasks and contexts, particularly those that differ considerably from the experimental tasks and contexts, can be a dangerous enterprise. (Note that scientists and science writers, under the guise of making laboratory results seem more relevant, can be guilty of this too.)

Psychologists have studied theory of mind for decades and are still trying to grapple with its complexity. For instance, for many years it was assumed that children developed a theory of mind right around their fourth birthday. In a typical task, a child watches as two puppets hide a marble. One of the puppets then goes offstage, and the second puppet moves the marble. When the first puppet returns, the child is asked where the returning puppet will look for the marble. Children younger than four tend to incorrectly point to the location where the marble actually is, not where the puppet would expect it to be.

But, oddly enough, children behave much differently in a modified version of the task: when simply instructed to watch as the first puppet returns in search of the marble, even one-year-old infants will express surprise when the returning puppet looks for the marble in its new location. This suggests that the infants were capable of forming expectations about the puppets’ false beliefs, which is to say they demonstrated theory of mind.

Renée Baillargeon, a developmental psychologist at the University of Illinois, has argued that tasks that require children to express their knowledge about others’ beliefs explicitly (as in the first task) are qualitatively different from those that test this knowledge implicitly (as in the second). And subtle task differences can produce drastically different results in adults too. Take another hotly debated topic in psychology, “audience design,” or the extent to which people take their audience’s knowledge into account when speaking. Here again, the literature can best be described as “conflicted”: sometimes people take their audience’s knowledge into account, sometimes they don’t, and the details are devilish.

With this in mind, the ability to successfully answer questions about a sentence like Peter said that Paul believed that Suzanne mistakenly thought Gillian commented that Mary liked chocolate and the ability to appreciate characters’ motivations in a long, slowly unfolding novel may not be equivalent. To her credit, Zunshine has begun working with researchers at Haskins Laboratory to empirically evaluate her hypotheses about theory of mind on more complicated texts.

Of course, not all scholars interested in psychonarratology need take this route. Literary scholars have long borrowed ideas from psychology—Freudian analyses come to mind—in a nonscientific but productive way. But when such borrowings start to sound themselves like scientific findings, perhaps the onus is on the scholar to make sure we know she knows we know they’re not.


Permission required for reprinting, reproducing, or other uses.

Jessica Love holds a doctorate in cognitive psychology and edits Kellogg Insight at Northwestern University’s Kellogg School of Management.


Please enter a valid email address
That address is already in use
The security code entered was incorrect
Thanks for signing up