Beyond the Lecture: Eric Mazur's Educational Conversion

Photos: Mauro Gottardo

"Buongiorno," Eric Mazur says, as we sit down at a table in the historic Caffè Pedrocchi, a fixture of the city's intellectual life for nearly two centuries. We order a cup of the traditional recipe, prepared with coffee, mint and cocoa. Everything in Padova seems to carry the weight of history, and Mazur, who speaks Italian and recalls a year he spent teaching at the University of Genoa, appears perfectly at ease among these layers of memory.

There is, however, a certain irony in his visit. In a few hours he will speak in the Aula Magna of a University founded in 1222, more than eight centuries ago. Nearby stands the imposing chair from which Galileo taught, elevated high above the students so that everyone could see and hear the master. For centuries that image embodied the ideal of education: knowledge flowing from the professor to the audience.

But Mazur has spent much of his career arguing that this image, powerful as it may be, is fundamentally flawed. His lecture is titled Confessions of a Converted Lecturer, organised within the ALMA network. The language is almost religious — and maybe not by accident. 

As a child in the Netherlands, Mazur had fallen in love with stars and galaxies after receiving a book from his grandfather. So, by eight years old, he decided he wanted to become an astronomer. But when he entered the University of Leiden — one of Europe's historic centres of astronomical research — that passion quickly evaporated. "All of a sudden, the beauty disappeared. It was all formalism and equations, writing on the board, memorising and making calculations." Within weeks he switched to physics.

The turning point

Years later, after joining the faculty at Harvard in the early 1980s, he unknowingly reproduced the same educational model. "I started doing to my students exactly what had happened to me," he says. They liked him: his evaluations were excellent, exam results were good. By every conventional measure, he appeared to be a successful teacher. "I thought I was doing a really, really good job, but that turned out to be a complete illusion."

After some seven years at Harvard, Mazur read an article in a physics journal claiming that students who take an introductory physics course but do not continue in the subject learn absolutely nothing from it. The authors — Halloun and Hestenes, at Arizona State University — had devised a simple test: not standard problems, but questions about everyday situations. They gave it at the start of the semester and at the end, and the results were statistically identical. The course had made no measurable difference.

Mazur was sceptical, he assumed his own students would do better. The moment that changed everything came when he administered the same test to his class. "Within thirty seconds, one student raised her hand and asked: 'Professor Mazur, should I answer according to what you taught us, or according to the way I normally think about these things?'"

The question shook him. "How could that be? There's only one universe. There's only one physics." He had suddenly realised that many students had learned to reproduce information without changing the way they understood the world. "I was focusing on transferring knowledge rather than helping students understand," he says. The distinction became the foundation of everything that followed. "You don't learn by listening, you learn by doing. Memorising and regurgitating are not the same thing as learning."

What Chaos Can Teach

The decisive breakthrough came almost by accident. One day, while explaining a concept to a class of 250 students, he noticed complete confusion on their faces. He explained the problem once. Then a second time, using a different approach. The confusion only deepened. In frustration, he asked the students to discuss the problem among themselves. What happened next surprised him: "Complete chaos," he says with a smile. "Everybody started talking." Within two minutes, many had reached understanding. "You could literally see the 'aha' moments on their faces."

That apparent disorder became the foundation of what would later be known as Peer Instruction — the educational method that made Mazur one of the most influential voices in higher education reform. Instead of treating students as passive recipients of knowledge, the approach asks them to confront a conceptual question, commit to an answer, discuss it with their peers, and then reconsider their position. The teacher becomes less a source of answers than a designer of questions and a guide through uncertainty. "Often professors have forgotten why a problem is so difficult. A student who has only recently learned it still remembers where the obstacles are." That is what psychologist Steven Pinker has called the curse of knowledge: "Once you know something really well, it doesn't become easier to explain. It becomes harder."

The insight led him to abandon what he calls "teaching by telling" in favour of "teaching by questioning". "Socrates and Galileo were doing it long before us: learning through questions, discussion and exchange." There is something fitting about making that observation in Padova. Galileo's Discorsi e Dimostrazioni Matematiche, structured as precisely such a dialogue, were published in 1638 in Leiden, the city where Mazur studied. Two cities, one method.

A lesson in the Aula Magna

A few hours later, in the Aula Magna itself, Mazur demonstrates what he means. Speaking to an audience largely composed of academics and teachers, he does not simply explain his ideas — he enacts them, inviting the room to do exactly what he asks of his students: stop listening passively and start thinking.

"Knowledge is not something that can simply be transmitted," he tells them. "It has to be constructed in the learner's mind." During a conventional lecture, he argues, students spend most of their time copying notes while deferring the harder work of understanding. "If I think about my own experience, the physics I know was not learned in the classroom but afterwards, by continuing to work on it and revisiting it. Often we learn in spite of the method."

Education, he explains, consists of two distinct steps. The first is the transfer of information. The second — and far more demanding — is making sense of that information. Traditional teaching devotes most of its energy to the first step and leaves the second largely to students themselves. This is why Mazur became one of the early advocates of the flipped classroom, in which students acquire information before class through readings, videos or digital resources, while class time is reserved for discussion, questions and conceptual challenge. "We give students responsibility for the easier part — information transfer. Teachers can then use precious class time to help them assimilate it."

To make the point concrete, he involves the audience in a live exercise. A question is posed; participants vote; they discuss their answers in small groups, then vote again. Many change their minds after talking to colleagues. The room fills with conversation, laughter and debate. For Mazur, that is not a distraction. It is learning.

Beyond Covid and artificial intelligence

Thirty-five years have passed since Mazur made his switch and higher education, by his own assessment, has changed very little. Two disruptions have since forced the question into the open, though with markedly different results.

The first was Covid. Remote teaching, he argues, simply exposed problems that had always been there, making them impossible to ignore. But when the pandemic ended, almost everyone went straight back to the lecture. "A missed opportunity," he says.

The second disruption is artificial intelligence, and Mazur believes this one will not pass so quietly. The question it poses is fundamental: why teach a student to solve a problem, if AI can do it faster, better, and more cheaply? In the United States, the effects are already visible: hiring of recent graduates is falling sharply, as companies find it more efficient to use AI than to hire an inexperienced graduate. 

"AI makes good-enough cognition cheap," he says, adding that students should be allowed to use AI, including during coursework and exams. "What it cannot provide is judgement, responsibility, and the ability to evaluate its own output. A human being can take responsibility for a decision. AI cannot." The task of education, he argues, is to cultivate precisely those capacities: how to use AI intelligently, how to interrogate it critically, how to assume responsibility for whatever emerges from it. "If we keep teaching the same things in the same way, we will produce graduates who are inferior to a tool that costs pennies per query."

There is a question that follows Mazur wherever he makes this argument. The old lecture system was so ineffective? Didn’t it literally send us to the Moon? "That question contains a hidden assumption," he says. "The lecture did not take us to the Moon. Smart, determined people did — people who managed to learn despite the lecture, not because of it." He invokes Einstein, who once observed that education is what remains after everything that has been learned is forgotten. "Most of what you are told evaporates. What endures is built through your own thinking, your own effort to make sense of things."

And then there is the waste — the students who never got the chance to find out. When Mazur studied physics at Leiden, 120 students enrolled in the first year. After one year, eleven remained. A hundred and nine had left for medicine, or law, or something else. "Were they incapable of physics? We have no idea. They were simply turned off — the way I was turned off by astronomy at seventeen."

Back at the Pedrocchi, the coffee cups are nearly empty. The conversation turns to what, if anything, should be preserved from the old way. A lecture, Mazur concedes, can inspire; it can create the desire to know. But the primary vehicle for learning, he insists, is doing: guided, active, engaged work. As the World Cup is about to start, Eric Mazur explains it in football terms: the best coaches do not sit their team down and demonstrate technique while everyone watches. "They put the ball at your feet, watch you fail, and help you improve." He smiles. "I see my role more as a coach than a professor. More guide on the side than sage on the stage."

Eric Mazur is Balkanski Professor of Physics and Applied Physics at Harvard University, where he has taught since 1984. He is the author of Peer Instruction: A User's Manual (Prentice Hall, 1997) and founder of Perusall, a social learning platform used by more than five million students worldwide.