We probably wouldn't welcome yet another book giving us the bad news that American students do not achieve as well as Asian students. But The Learning Gap: Why Our Schools Are Failing and What We Can Learn from Japanese and Chinese Education (New York: Summit Books, 1992) by Harold W. Stevenson and James W. Stigler does far more than that: It challenges us to take a new look at some education practices that we have come to take for granted.

Stevenson and Stigler have been studying elementary schools in Minneapolis; metropolitan Chicago; Sendai, Japan; Taipei, Taiwan; and Beijing for over IO years. Besides testing first and fifth graders and some kindergarten children, they have interviewed teachers, students and parents and spent many hours observing classes. Among the many fascinating things they've found are at least two radical differences in the way Asian and American teachers conduct and plan lessons.

The classes in the Asian schools that Stevenson and Stigler observed were typically much larger than we tend to think classes should be -- 38 to 50 students. And they were not grouped by ability -- all the elementary school classes contained kids who were achieving at every level. But, like American teachers, the Japanese and Chinese teachers did a lot of teaching to the whole class. The difference, Stevenson and Stigler say, is that Asian teachers' classes were more like lessons than lectures. The Asian teachers didn't try to cover a lot of material. A math class, for example, might be organized around solving a single problem. As a result, the teachers had time to engage the entire group of 40 or so students in finding a satisfactory answer.

In the course of this search, Asian teachers tried to get students to explain their reasoning and to evaluate their own answers and the answers other students gave. They also encouraged students to see that there were many different ways of solving a problem -- some of them quicker or more elegant than others but all productive. By the end of the lesson, they could be pretty sure that everybody in their mixed-ability class understood the solution -- and the various ways of getting to it -- well enough to be able to build on it in future lessons.

American teachers, Stevenson and Stigler found, were much more likely to try to cover lots of territory, and they were much more answer-oriented. A teacher would ask a question, try to elicit the "right" answer from a student and then move quickly on. As a result, American kids, unlike Asian ones, did not get experience with the many ways of solving a problem. Lots of them probably concluded that a single "right" answer was all that was important anyway.

Why didn't American teachers challenge students on their answers and encourage other students to do so? Partly because they needed to get through a great deal of material. But the fear of humiliating students also played a big part. What if someone gave a wrong answer in discussion or up at the chalkboard? This did not seem to be a problem for the Asian teachers or students Stevenson and Stigler observed. Of course the kids were used to this way of conducting the class, but Stevenson and Stigler also noticed that Asian teachers did not convey the idea that there was anything shameful about getting a wrong answer. They considered it a part of the learning process and so, apparently, did the students.

What accounts for these very big differences in classes that, on the surface, might seem to be similar? One of the big reasons is the difference in the way American and Asian teachers prepare their lessons. American teachers who take pride in their teaching are likely to think of a lesson as something personal. It is their own way of handling the material. The examples and questions are ones they have selected for their students. And their ideal might be a lesson that is fresh and new every time. But this is a standard that few people can live up to day after day and week after week.

The Asian teachers whom Stevenson and Stigler observed were more likely to work with other teachers -- in part because they had more time to do so -- to develop or revise an approach to a topic and to craft questions and examples. By pooling their experience and knowledge, they created common lessons that were of consistently high quality.

This may sound limited and limiting to people who consider preparing a lesson analogous to standing in front of a canvas on which they paint; it may sound like turning teaching into a rote exercise. But you could also compare the Asian teachers' approach to the way a number of different pianists would approach a concerto:

They might have different interpretations and different styles, but they would all be playing the same notes. If teaching is a performing art, as Asian teachers seem to think it is, teachers don't have to worry about composing a new concerto -- or painting a new picture -- for every class. They can practice and perfect ones they develop together with other teachers. They can rethink the questions that led to deep silence instead of a lively discussion and they can discard examples that failed last year or yesterday and think up new ones. And as they collaborate with other teachers, they can break out of the isolation in which they now work.

It's worth our thinking about.