In his estimation, literature shows that not all knowledge is quantifiable. There is always something new in a literary work--a new interpretation, a new insight. In a sense, the literary work itself changes over time, he says.

"There are certain things we can't prove, but we still want to say them," Chaouli says, "and literature allows us to do that. It's always going to be in the in-between area between science and opinion. We shouldn't give it up because of this, but enjoy it."

Teaching Ambiguity

Conveying these ideas to his students hasn't always been so easy, however.

Chaouli recalls that when he first started teaching at Harvard he would begin a class by presenting one interpretation of a work of literature and then follow it with another. His students would often cross out their notes on the first interpretation, thinking it was proved wrong by the second.

"They didn't understand," Chaouli remembers, "that the second interpretation lives off the first. Literature doesn't drive to one point that you can write down and say on an exam."

There is still, of course, certain information that must be known to study literature. "You need to know what happens on a structural level to understand the excesses to that structure that authors use," Chaouli says.

Still, what is truly learned from the study of literature is not something that can be written on a syllabus. It is a side-effect, a moment when something clicks.

Chaouli's critics have not taken kindly to his ideas. A recent letter to the Times Literary Supplement attacked him for allowing theories that have been discarded by other disciplines, such as Freudian analysis, to be considered legitimate in the study of literature. By that reasoning, the letter claimed, the flat earth theory could also be useful to literary scholars.

But, Chaouli counters, the flat earth theory is useful to literary scholars.

Such a critique misunderstands his arguments, he says.

For the purposes of literature, certain questions can be posed more shrewdly or more interestingly with recourse to terms that are no longer acceptable in scientific disciplines.

Literature shouldn't be mistaken for an exact description of the human condition or an agent for promoting morality, he adds.

"Literature departments shouldn't think of themselves as quasi-religious institutions," Chaouli says.

He doesn't deny that there are moral aspects to literature, but he says that there is no way of knowing what effect a piece of literature will have on a reader. There is no reason to assume that literature will raise the moral standards of its readers.

For some critics, such a view is unacceptable. They claim that literature departments cannot survive in a university setting unless they develop clear standards for students and a system of positive knowledge. The anxious and uncomfortably ambiguous state of literary studies needs to be resolved, they claim.

"That's fine," Chaouli says, "but such a system doesnt exist yet."

For now, then, Michel Chaouli will continue urging his students and fellow scholars to embrace rather than lament the so-called crisis in literary studies.