The Death of the Lecture


As anyone who’s been a student knows, if you put a professor in front of a classroom and let him talk, a good amount of information will bounce about the room, vibrate a few ear drums, and even inspire a few notes. Little of it actually sinks into grey matter. But when students are thrown an actual problem to solve among themselves, then the real learning starts.

That’s why Eric Mazur wants to end the lecture as we know it.

Mazur, the area dean of applied physics at Harvard University, hopes to change the role of the professor to that of guide, the role of the classroom to a place where students learn together, completing projects. “Once you hook the students with a problem that needs to be solved, the students take ownership of the problem solving, and it’s no longer a chore, no longer disconnected with the real world,” says Mazur.

Problem-Solving Approach

Though the idea might seem all too radical to many math and physics professors, who are themselves successful products of traditional education, engineering classrooms have to a large degree already made the shift. “In engineering education, things are very different—there’s a lot of project-based learning, team-based learning,” says Mazur. In other disciplines, the methods of evaluation are as much a problem as what goes on in the classroom. “Study habits, and approaches to learning, are driven by assessment.” Rote memorization becomes favored over problem-solving. Engineers, though, are more often graded on the outcomes of their projects.

Eric Mazur, the area dean of applied physics at Harvard University, believes that students learn better when they solve a problem as a team than when they listen to the drone of a professor and memorizing facts.

The traditionalist may balk at throwing out all fact-based learning and memorization. Is there not some foundational core of math and science that all engineers should know? “What’ does it mean to really learn calculus?” asks Mazur. “There’s much evidence that graduates don’t know calculus yet. They know how to find the derivative of a polynomial, but I would argue that the current system doesn’t do a good job at all. Take a line out of a contour, change a ball into a brick, they’ll understand, but if I change the situation more radically, suddenly they’re frozen.  I would say yes, if you do it all project-based, students may not be exposed to parts of the curriculum that they are now—yes, we will graduate engineers that are not exposed to some things that more traditional engineers would say ‘What!  You did not cover this?’ But what matters is what is learned. Exposure is meaningless if it’s not real learning.”

Re-Awakening Curiosity

Whatever can be learned by rote can likely be learned, if not now, someday soon, by a computer. “Maybe the historical approach works if you want to only design bridges as they’ve been designed until now, with exactly the same construction. But my hunch is that those jobs will be taken over by automation—computers will take those jobs away,” says Mazur. “We want students to be able to solve problems we can’t yet solve, design bridges that we have not yet seen, design gadgets we have not yet designed.”

Of course if we really want students to learn through problem-solving, classrooms will have to change at every level of education. “I think you see that the learning at a very young age takes place very differently—not by asking how but by asking why—and driving parents and teachers crazy. After they get to middle school, all that curiosity has been erased, probably because people discourage the question ‘why?’” says Mazur.

“I consider that my most important task—reawakening their curiosity.”

Michael Abrams is an independent writer.

We want students to be able to solve problems we can’t yet solve, design bridges that we have not yet seen, design gadgets we have not yet designed.

Eric Mazur, the area dean
of applied physics,
Harvard University


September 2012

by Michael Abrams,