Introduction I. Students and Teachers Study Pendulums II. What is a Research Lesson? III. Types of Research Lessons IV. The Impact of Research Lessons
	1. Individual Professional Development 2. Spread of New Content and Approaches 3. Connect one's own practice to broader goals 4. Push for coherence 5. Create Demand 6. Shape National Policy 7. Honor the Central Role of Classroom Teaching
V. Research Lessons: What Are the Supporting Conditions?
	1. A shared, frugal curriculum 2. Established collaboration 3. Self-critical Reflection 4. Stability of Educational Policy
VI. Epilogue References

Introduction How have Japanese teachers succeeded in shifting from "teaching as telling" to "teaching for understanding," in elementary science? As we have investigated this question over the past three years, Japanese teachers have repeatedly pointed to the impact of "research lessons" (kenkyuu jugyou), a practice they see as central to individual, schoolwide, and even national improvement of teaching. Let's "visit" a research lesson. I. Students and Teachers Study Pendulums Working in pairs, forty Japanese fifth-graders weight small wire pendulums with clay and "race" them, trying to figure out which of three variables suggested by the class -- the length of the wire, the clay's weight on the pendulum, or the angle of release -- affect the pendulum's cycle time. Noisily intent on their investigations, students pay little attention to the tape recorder that their teacher holds in his hand, or to the more than 20 observing teachers who walk around the classroom snapping flash pictures, recording the lesson on videotape, and taking detailed notes on the students' work. After the 45-minute lesson, the observing teachers move to another room to discuss the lesson they have just seen. Mr. Ohara, who taught today's research lesson, begins by explaining his aspirations for the lesson: to see whether students would demonstrate scientific thinking by "untangling the three variables from one another to study them one at a time." A lively debate ensues. Several teachers argue that it would have been better to tell students to control variables, since few did so spontaneously; other teachers disagree. Teachers also disagree about whether using a stopwatch or comparing two pendulums side-by-side is a better way for fifth graders to measure the impact of the three variables. Mr. Otaki avoided using a stopwatch, he explains, because students take differences of just a hundredth of a second very seriously, and draw erroneous conclusions about a variable's effect. Other teachers counter that fifth-graders are old enough to discuss and understand measurement error. As the two-hour colloquium on the lesson draws to a close, teachers offer their opinions on how Mr. Ohara should structure the next day's lesson, in which students will report and discuss the results of their (often uncontrolled) experiments. Once again, tomorrow's lesson will be observed, recorded, and discussed by teachers from within and outside the school.

II. What is a Research Lesson? Research lessons are actual classroom lessons, taught to one's own students, with several special features:
1.	Research lessons are observed by other teachers. Some research lessons include only the faculty within a school, or the school's faculty with a few invited outside commentators; others are open to all teachers from a district, town, region, or even the whole of Japan.
2.	Research lessons are planned for a long time, almost always collaboratively. For example, in one school we studied, the four third-grade teachers met regularly for several months to discuss how to promote students' "initiative" in the study of science. Collectively, they decided it was important for students (not just teachers) to ask productive scientific questions, and they came up with teaching strategies in service of this goal. By watching and discussing each other's lessons, they progressively honed these strategies. Finally, one of the teachers presented their new approach to the entire faculty as a research lesson, while the other third-grade teachers acted as "co-hosts," recording the lesson, preparing written background materials on their discussions, presenting the highlights of their months of work together, and so forth.
3.	Research lessons are designed to bring to life in a lesson a particular goal or vision of education. Though the goal or vision is chosen by the faculty, often it is an issue of lively national interest. Examples of goals from the research lessons we observed included helping students to "take initiative as learners," "be active problem-solvers," "be active problem-seekers" "develop scientific ways of thinking," and "develop their individuality" -- all issues central to national educational debate. The research lesson is not so much a finished product that is expected to be used in toto elsewhere as an example of a goal or vision of education in action that individual educators will draw on as appropriate to their own philosophy and setting.
4.	Research lessons are recorded. Usually teachers record these lessons in multiple ways, including one or more videotapes, one or more audiotapes, narrative and/or checklist observations, and copies of student work. Recording is focused on particular issues of interest to the instructing teacher: for example, we observed lessons in which the teacher asked colleagues to tally the number of students who volunteered their ideas during whole-class discussion, to record the discussion in each small group, or to transcribe all comments made by three selected children (one very shy, one outspoken, and one very knowledgeable in science).
5.	Research lessons are discussed. The faculty, sometimes joined by outside educators, discuss the research lesson during a colloquium or panel discussion. Typically, such a gathering begins with presentations by the teachers who planned and taught the lesson, followed by free or structured discussion in which teachers who observed the lesson comment on its strengths and weaknesses, and ask questions; often an invited outside educator or researcher also comments on the lesson.

III. Types of Research Lessons The most common research lesson is the "in-school research lesson" (kounai kenkyuu jugyou); these take place regularly at ordinary elementary schools throughout Japan. As one Japanese elementary teacher told us: "The research lesson system is valued very highly by Japanese teachers. You find it even in very isolated mountain schools, where there are fewer than 20 students. You won't find a school without them. That's one reason why the education throughout Japan is fairly standard, whether you're talking about Tokyo schools or the remotest mountain school." Another teacher said: "Why do we do research lessons? I don't think there are any laws [requiring it]. But if we didn't do research lessons, we wouldn't be teachers." Teachers decide the theme and frequency of research lessons; schools we studied held them at frequencies varying from monthly to several times a year. A second type of research lesson is the public research lesson (koukai kenkyuu jugyou or gakushuu kenkyuu happyoukai). These research lessons are open to teachers from outside of the school; invitations may be sent to educators in the local district, the region, or even the whole of Japan. When schools receive grants to develop some part of their educational program -- such as computer instruction or international education -- they are often expected to culminate their work in a public research lesson. Research lessons are a popular place to see new subjects or approaches in action. For example, when a new subject -- life environment studies -- replaced science and social studies for first and second graders, teachers flocked to research lessons at schools that had pioneered the new subject. Perhaps the largest and best-known public research lessons are those conducted several times a year at national elementary schools, the 73 selective-admission public schools throughout Japan where new educational approaches often originate. When we emerged from a Tokyo subway station in 1996 to attend a research lesson at a national elementary school, the broad walkways leading to the school were jammed with educators from all over Japan, in a scene reminiscent of the huge crowds that pay homage at shrines on New Year's Day. The elementary school attracted nearly 5000 educators over its two days of research lessons. As lessons went on throughout the school, dozens of teachers crowded inside each classroom, and dozens more looked in from the hallways, through large sliding windows opened to afford a view of the classroom. Background materials on lesson goals, philosophy, and unit context were distributed to all visitors. During panel discussions following the lessons, visiting teachers had a chance to question the teachers about their lessons, to exchange views, and to hear the teachers' own assessments of what went well and poorly. Research lessons also occur in many other contexts. For example, the annual conference of Japan's Elementary Science Education Association rotates yearly to different regions of Japan, where it takes place mainly in a cluster of elementary schools. The thousand or so teachers attending the conference spend most of their time observing and discussing the research lessons at the schools; only at the end of the conference do participants assemble in a single location for a plenary session. Research lessons are also central to the work of many teachers' study circles and of school districts' professional development (for example, the required professional development that is often provided during the first, fifth, and tenth years of teaching).

IV. The Impact of Research Lessons Based on our interviews with more than 70 Japanese elementary educators and our observations of about 30 research lessons, we speculate on eight ways in which research lessons contribute to the improvement of Japanese education. 1. Individual Professional Development Japanese teachers mention many effects of research lessons on their own professional development. When teachers conduct such lessons, they receive feedback on their teaching, as several teachers describe: "As a brand new teacher, my colleagues who saw my research lesson told me I talked too fast. They were right. My students were having a hard time keeping up with what I said, and I didn't even know it!" I was told after a research lesson that I talked too loudly and scared the children. I had never taught first-grade before, just upper grades, and I didn't realize how big my voice sounded to young children. They weren't used to a male voice. A teacher who saw my research lesson commented that it was taking me a lot of time to write on the blackboard each word of every student comment--that I should just write brief phrases instead. The data gathered by one's colleagues during a research lesson provides an external reference point on one's practice. In one research lesson, an observing teacher told her colleague: "Only 47% of the children spoke up today during your science lesson. To increase participation, you might have quickly polled all students, especially since you already had their names on magnets." Another teacher talked about the value of seeing one's own teaching through the eyes of others: "Research lessons help you see your teaching from various points of view. ... A lesson is like a swiftly flowing river; when you're teaching you must make judgments instantly. When you do a research lesson, your colleagues write down your words and the students' words. Your real profile as a teacher is revealed to you for the first time." When we asked how they had shifted from lecture-centered to student-centered science, teachers often mentioned specific techniques picked up at research lessons that had helped them make this transition: "I've learned a lot from other people ['s research lessons]. For example, to write on chart paper rather than the blackboard. That way, you can save it as a record. You can pull it out at the beginning of the next lesson and the image of the prior lesson comes to mind immediately. Typically... there are some kids who have a hard time remembering. But if you have the poster paper, everyone can remember. You can also pull out the charts to show the path of learning over the year; you can reflect on the path of learning." Another example of something I learned from seeing a research lesson is finding out how teachers deal with certain common problems in the classroom, such as how to get a debate going when there's just one point of view held by most of the children in the class. For example, if there's just one child holding the "B" point of view, and the rest of the class holds the "A" point of view, the child holding "B" may feel bad if you stimulate a debate between views A and B. The "B" child may feel alone, and want to switch to be with the majority. That's a kind of torture for children. One thing many teachers will do in that situation is to take the "B" point of view themselves. But then the teacher is talking a lot, instead of the students. What I learned is that you can ask children how sure they are of the viewpoint they espouse. Are they 100% sure, or 80% sure, or half sure? Then you can ask what their doubts are about the idea, and have a debate between people who do and don't have doubts of a certain kind. ...That's a technique that I learned from research lessons that I apply in my classroom lessons when there's not enough difference of opinion to sustain a debate. Another example of something I learned from research lessons is to use magnets with children's names on the blackboard. For example, you can...use them to keep track of how children's opinions about something changed over the course of a lesson or unit. Or you can put the name-magnets next to ideas, and children can look at them and be conscious of their own ideas. In addition to seeing research lessons as a source of feedback and of new techniques, teachers described influences on their philosophy of teaching. A teacher who saw research lessons as part of his participation in a teachers' research group recalls: "[Before I joined the teachers' research group], I had always seen education as teachers giving knowledge to children, as a top-down process. Through my work with the elementary science research group, I came to see education not as giving knowledge to children but as giving them opportunities to build their own knowledge. Initially, that was not what I believed. Even when I saw it in practice, I couldn't believe in it at first. When I first saw lessons in which children were building their own knowledge, I thought 'Is this kind of instruction really OK? It takes so much time.' But then I began to realize that if children don't experience something, they don't understand it. They can memorize it but when the time comes to use it, they can't."

2. Spread of New Content and Approaches When a new topic -- such as solar energy -- is added to the curriculum, it often becomes a popular focus for research lessons. Research lessons give teachers the opportunity to ask questions about the new topic. In the discussion following a fourth-grade research lesson on solar energy, a teacher asked: "I want to know whether the three conditions the children described -- 'to put the battery closer to the light source,' 'to make the light stronger,' and 'to gather the light' -- would all be considered the same thing by scientists. They don't seem the same to me. But I want to ask the teachers who know science whether scientists would regard them as the same thing. Research lessons may also give teachers the chance to make sense, collectively, of topics or approaches newly added to the national curriculum." After the same solar energy research lesson, another teacher commented: "I haven't taught fourth graders for a while, so I have no idea how and why solar batteries were added to the curriculum. I'm only guessing that including solar batteries reflects adults' hope that children will become ... interested in solar energy and thereby help Japan. Science education specialists might be concerned about children using the proper vocabulary or setting up certain experimental conditions, but if the goal of including solar batteries in the curriculum is to get children interested in the fact that electric current can be changed by light, then Mr. Hori's lesson fulfilled that. So I'd really like to know the reason why solar batteries were included as a new curriculum material for fourth graders." In other words, teachers had the benefit of colleagues' thinking as they sought to understand new science content added to the curriculum, and to understand why it had been added. When we asked principals how they helped teachers shift to "life environment studies," the subject that replaced primary science and social studies, many, like this assistant principal, mentioned the importance of research lessons: "The way to improve life environment studies is to see many good actual examples. We can do that by going to lots of schools that are doing presentations and research lessons on life environment studies. Many people from this school have gone. Each school has its own way of approaching the new subject. Some are appropriate for your school, some aren't. What works elsewhere might not work at your school because the children are different. So you need to see lots of examples."

3. Connect one's own practice to broader goals In recent years, as concerns that Japan's students are passive, unimaginative test-takers have dominated the Japanese press, national educational guidelines have increasingly emphasized student qualities such as "initiative," "autonomy," "desire to learn" and "active problem-solving." The qualities actively discussed and advocated at the national level often find their way into the goals chosen by school faculties for their research lessons. For example, in a school that had chosen student "initiative" as its research goal, third-grade teachers who had formerly started the science unit on sound by asking students: "What is the connection between sound and vibration?" redesigned the unit so that it began by having students build musical instruments. Their intent was to provoke students to ask about the connection between sound and vibration, rather than have teachers introduce the question. In another school, teachers who had formerly taught about levers using small desk-top models decided to introduce the topic by challenging students to lift 40-kilogram sacks of sand; teachers made poles and ropes available, and students used classroom furniture as fulcrums. Teachers talked about the effect of the materials on their goal of building students' desire to learn: "How can you discover the beauty of a lever if you're using it to lift something you could lift easily with your bare hands?" Research lessons provide an opportunity for teachers to discuss big ideas currently shaping national educational debate, to think these through, and to bring them to life in an actual classroom lesson. The impact of research lessons in connecting teachers with practice outside of their school is reflected in the comments of teachers who said they attend national school research lessons to "See where Japanese education is going" and "To find out what's new." Teachers also reported that research lessons connected them with teachers within their schools. A teacher who had just completed a research lesson commented: "The research lesson is not over yet. It's not a one-time lesson; rather, it gives me a chance to continue consulting with other teachers. For example, I may say to other teachers, 'I want to ask you about my last lesson you saw. . . ' Then, the other teachers can provide me with concrete suggestions and advice because they have seen at least one lesson I conducted. We teachers can better connect with each other in this way."

4. Push for coherence A fourth impact of research lessons may be to support teachers as they try to reconcile competing goals or visions of education. The following discussion occurred after the pendulum research lesson: Host Teacher: We have the feeling that recently in science education the process has been over-emphasized, and the results and conclusions underemphasized. We feel that the conclusions -- what you might want to call children's knowledge -- have been underemphasized of late. Why is a lesson good simply because children are active? Visiting Teacher: If children are making connections with daily life, then that's science. [Reads a quote to that effect from the national science Guidelines.] Host Teacher: Not just any kind of experience qualifies as science. If children leave here thinking that weight makes a difference in pendulum swing, then there's something wrong with the scientific process that's going on here. Visiting Teacher: Do you call it scientific reasoning if they get the right answer, but not if they don't? When does it suddenly become unscientific thinking? In this conversation, two views of science education are bumping up against each other. Is it more important to have students gain the factual knowledge that weight does not influence pendulum cycles? Or to be active, interested scientific experimenters? Both seem to be worthy goals; the research lesson system may increase the likelihood that such opposing views of education will bump up against each other, so that neither is neglected. In the discussion following a research lesson on solar batteries, several teachers suggested that the teacher who taught the lesson should have used the students' words, rather than his own words, in the lesson summary. As one teacher said, "I felt sorry for the students when the teacher concluded the lesson with his own summary statement." Another agreed that the teacher had "forcibly" pushed students' results into his own summarizing statements. Yet other teachers disagreed: "I don't agree with several teachers who think that students' ideas were somehow stifled by the teacher's summary. As someone who doesn't know much about electricity, I found the teacher's summary helpful. Students who, like me, have limited knowledge about solar cells may have found the teacher's statement helpful, after hearing such a wide variety of opinions." As recent battles over both reading and mathematics attest, U.S. education is often plagued by pendulum swings between different educational approaches. How often do U.S. teachers have opportunities for conversations like the one above -- in which Japanese teachers debated the importance of scientific knowledge versus process, using as shared data a lesson they had all watched? Research lessons bring together teachers from the whole spectrum of viewpoints to plan, view, and discuss lessons. It seems likely that, the more frequently different educational philosophies come into contact around a shared, concrete lesson, the more likely teachers are to notice and incorporate the strengths of different approaches -- such as the emphasis on scientific content and inquiry. An American teacher who saw our videotape of a Japanese research lesson commented: "How different American mathematics education might be if we saw each other's lessons and found out what other teachers actually meant by terms like 'constructivism.'"

5. Create Demand Richard Elmore has made the case that education in the U.S. suffers not from a low supply of good educational programs, but a low demand for those programs. Demand occurs when educators want to improve their practice. Research lessons may be seen as a way of creating demand. One Japanese teacher recalled how, early in her career, she burst out into tears after seeing a wonderful research lesson by her fellow first-grade teacher: "I felt so sorry for my own students. I thought their lives would have been so much better if they'd been in the other teacher's class. You realize you have had a big impact on your students. You see how authoritarian teachers have very quiet classes. Teachers who value students' ideas have very active classes. You see how teachers are creating a class, not just teaching a lesson. The teacher's way of speaking and the teacher's way of getting angry are all passed on to the students." Several principals expressed the view that research lessons build momentum for improvement much more effectively than direct leadership by the principal (see also Bjork, unpublished). Although himself a science expert, one principal explained that he did not provide concrete advice to his faculty, but instead relied on research lessons to stimulate demand for improvement among teachers: "It is necessary for teachers themselves to think about how to teach science, to tell their ideas frankly to other teachers, to get ideas from other teachers, and to improve themselves. The teachers in this school don't know much about science, but with their own knowledge, they will express their opinions as to what kind of lessons they want to do and what kind of teaching materials they want to develop.... Since there isn't a science specialist here, they don't know at all whether their ideas are good or bad. They come to me, but I try not to interject my own ideas. So who can advise them? Since this school will be ...the site for the National Science Teachers' Association conference, teachers in Tokyo will assist this school, because they want the research lessons at the Tokyo conference to be successful. Members of the Science Teachers' Association in Tokyo want to assist us. Our teachers can discuss with them how to design the flow of the lessons, and what kinds of teaching materials should be developed. Based upon their exchange of opinions, our teachers will redesign lesson plans ... and then, they will conduct the research lessons. Our teachers and those teachers who assist them ...will improve themselves together. That is how we work together." Research lessons may create a "demand" to improve practice because they are viewed by other teachers. In the words of one second-grade teacher with 20 years of teaching experience: "Even if teachers do not think hard about the lessons they teach daily from the textbook, for research lessons they must really rethink the fundamental issues."

6. Shape National Policy We found evidence that research lessons sometimes influence national educational policy. Solar energy, for example, entered the national Course of Study after individual classroom teachers pioneered research lessons on the topic. These lessons spread among teachers through the research lesson system, and were noticed by members of the national curriculum committee. A second route of policy influence is the outside commentators invited to research lessons; often these individuals have been active in the development of a new content area or approach. Commentators are often classroom teachers, but may also include principals, district resource teachers, university professors, and policymakers. When they attend research lessons, they have a chance to see how students and teachers are grappling with new subject matter, or with vague new national goals such as "initiative" and "autonomy." For example, one invited commentator at the solar energy lesson was an elementary principal who had served on the Ministry of Education committee that added solar energy to the national curriculum. At research lessons, he could see how this new content area was actually brought to life in the classroom, hear teachers' questions and concerns, and see how students grappled with the new content. He could share this information with individuals in a position to shape curriculum and textbooks, and he could spread word of exemplary techniques. Well-known teachers and principals may be invited to dozens of research lessons every year as commentators. They see how new approaches and topics are being implemented and understood in many different schools across Japan. In effect, this amounts to a system of "formative research" in which policy can be informed by actual classroom education.

7. Honor the Central Role of Classroom Teaching Finally, research lessons honor the central role of teachers. Japan's national educational guidelines are remarkably terse, underscoring the idea that policy is created in the classroom, not on paper. The entire Japanese Course of Study for Elementary Schools spans just 122 pages of a half-letter-size booklet. The additional volume provided for each subject area is, like the Course of Study itself, small and thin, and does not specify the particular teaching materials to be used. (The volume for all of elementary science, for example, covers 116 half letter-size pages.) The changes made to these documents (about once a decade) are often brief, abstract descriptions of new goals: "autonomy," "initiative", "desire to learn" "problem-solving capacity." When we first began our research, we found these vague goals -- provided without accompanying concrete examples -- frustrating. Yet they may reflect an underlying assumption that policymakers cannot define good classroom practice; rather, research lessons provide a systematic way for teachers to bring policy to life, thoughtfully and collaboratively, in the classroom. Research lessons also provide a way for Japanese classroom teachers to rise to national stature while remaining in the classroom. Though teachers do not receive increased salary or position by conducting research lessons, they do, in some cases, become known throughout Japan, often publishing books and articles about their lessons. As we have interviewed teachers in various regions of Japan about the influences on their science teaching, we've had the odd experience of hearing them talk about teachers whose lessons we have also observed: "I don't know him, but I saw his research lesson nine years ago, and I realized I had seen a real student discussion for the first time," said one Nagoya educator, about a Tokyo teacher whose lessons we had both seen, albeit eight years apart. The research lesson system provides a route to become nationally known that does not lead inexorably out of the classroom.

V. Research Lessons: What Are the Supporting Conditions? We suspect that several features of the Japanese educational landscape may provide essential support for research lessons: 1. A shared, frugal curriculum By U.S. and world standards, the Japanese curriculum is very spare. The TIMSS (Third International Mathematics and Science Study) documents, for example, that Japanese eighth grade science textbooks cover just eight topics, compared to an average of more than 65 for U.S. eighth grade textbooks (Schmidt et al, 1997). Japanese textbooks are brief, and Japanese students spend substantial time on each of the small number of topics they study. For example, Japanese fifth-graders are expected to spend 10 science periods studying the function of weight, although there are just two pieces of knowledge that they are expected to take away: that a bigger and faster object impacts another object more; and that the round-trip time of a pendulum is affected only by its length, not by weight. Since Japanese teachers have a relatively large number of class periods to help students master a relatively small amount of science content, teachers can devote time to studying the most effective ways to present it, rather than to wading through massive textbooks to figure out what's really important to teach (Lewis & Tsuchida, 1997, 1998; Stigler & Hiebert, 1997).

2. Established collaboration Japanese teachers routinely collaborate in planning lessons, in planning 30 days or so per year of schoolwide activities, on many schoolwide committees, and in taking care of each other's classes, since substitutes are not hired for short-term absences (Bjork, unpublished; Sato, 1996; Sato & McLaughlin, 1992; Rohlen & LeTendre, 1996; Shimahara & Sakai, 1995). Accounts of Japanese elementary school life suggest that collaboration among students is emphasized and competition avoided (e.g., Lewis, 1995); this may be true for teachers as well. Electing a "teacher of the year" is, for example, an American practice that surprises many Japanese teachers who visit the U.S.. The oft-noted finding that Japanese attribute success to hard work rather than ability (Stevenson & Stigler, 1992) may apply to teachers as well as to students, as teachers are expected to improve themselves through collaborative study of lessons: "Our textbooks are very thin with few explanations....Teachers have to fill the blanks between the lines in the textbook. That is why we have to study about lessons. . . Unless you improve your own skills, you can't do a good lesson even with a good lesson plan or good textbooks. Precisely because of this belief, we all do open lessons and try to improve our teaching skills. If you isolate yourself and do whatever you wish to do, I don't think you can ever conduct good lessons." Routine collaboration may also shape attitudes toward borrowing, as two Japanese teachers suggest: "Even if you copy someone else or are copied by someone else, I don't think anything can be absolutely the same. So, I think it is all right to copy others." "If you shoot for originality too early in your development as a teacher, you're likely to fail. Initially, you must take a lot from others. But ultimately, to move to a higher level of teaching, your lesson must become your own original thing, not simply imitation of others. But it's through imitating others' lessons you create your own authentic way of teaching." It is not the case (despite accounts to the contrary) that Japanese elementary teachers have more time for collaboration than their U.S. counterparts; daily time with students is comparable or longer in Japan (see Lewis, 1995). However, general support for teachers and for their professional development activities may be greater in Japan (U. S. Department of Education, 1985).

3. Self-critical Reflection Within Japanese schools and perhaps within Japanese culture more widely, hansei -- self-critical reflection -- is emphasized and esteemed (Lewis, 1995; Rohlen, 1976). Both teachers and students set goals for self-improvement in a "quest for character improvement [that] is close to being a national religion" (Lewis, 1995; Rohlen, 1976, p.128). De-emphasis of external evaluations (merit reviews, checklist evaluations, etc.) of teachers may create safety to reveal one's weaknesses (Bjork, unpublished). Self-critique may have a decidedly different emotional meaning when it is established and valued as it seems to be in Japan; identifying one's shortcomings and gracefully accepting criticism may be ways of showing competence, not failures to be avoided. Nor is critique typically focused on a single individual; collaborative planning of research lessons means that criticism is generally shared with several colleagues. 4. Stability of Educational Policy Although some Japanese educators complain that Japanese education is slow to change, (Shimahara & Sakai, 1995; Horio & Platzer, 1988), overall stability may make it easier to concentrate on policy changes that do occur. The comments of a Ministry of Education official suggest a surprisingly long timetable for change: "We change the Course of Study about every ten years. But the truth is that ten years is too short a time to change classroom education. If we greatly changed the Course of Study every ten years, teachers would be turning their heads this way and that so often that their necks would break. So we make major changes in the Course of Study only every twenty years or so, and in between it's just fine tuning."

VI. Epilogue On Day Two of the research lesson, Mr. Ohara begins science by asking students to report the results of the previous day's experiments. As students volunteer their results, he records them on the blackboard, and then regards the findings with a puzzled expression: "From these results here, I can't say at all what we found -- if we found that [variable] A, B, or C, was important. Here it says A alone; here it says C alone.. ... What should we do?.... Different groups found different results." Students comment that some students changed weight at the same time as length, and several students offer the opinion that everything but the variable under study needs to be kept the same. Students then suggest crossing out the experiments that didn't meet this criterion. When this is done, a pattern suddenly emerges: the properly controlled experiments show that the length of the pendulum, but not weight, was important. As students see that the controlled experiments gave clear results on two of the variables, the feeling of "aha" in the classroom -- not just among students, but among the observing teachers -- is almost palpable. For us as observers, the second day's lesson was stunning. Believers though we were in the power of student-centered instruction, we never imagined that the sloppy experiments of the prior day could be salvaged, let alone turned into such a powerful "aha." Though much remains to be learned about the nature and impact of research lessons in Japan, we feel no doubt about its dramatic impact on us: Mr. Ohara's lesson pushes us to think, in ways large and small, about the nature of good teaching, about how good practices are honed and spread, and about how teachers can be recognized and supported as they reinvent policy in the classroom. Copyright held by Catherine Lewis For further information contact: Catherine Lewis Mills College Women's Leadership Institute 5000 MacArthur Blvd. Oakland CA 94613 Email: c_lewis@post.harvard.edu

References Bjork, C. (unpublished). Cultural foundations for professional growth: The influence of community in Japanese schools. Horio, T., & Platzer, S. (1988). Educational thought and ideology in modern Japan. Tokyo: University of Tokyo Press. Ito, Y. (1994). Kyoshi bunka gakkobunka no nichibei hikaku (Teacher culture, school culture: A Japan-U.S. comparison) In Inagaki, T., & Kudomi, Y. (Eds.), Nihon no kyoshi bunka (The culture of teachers in Japan) (pp. 140-156). Tokyo: University of Tokyo Press. Lewis, C. (1995) Educating Hearts and Minds: Reflections on Japanese Preschool and Elementary Education (New York: Cambridge University Press). Lewis, C. & Tsuchida, I. (1997). Planned Educational Change in Japan: The Shift to Student-Centered Elementary Science. Journal of Educational Policy. 12, No. 5, 313-331. Lewis, C. & Tsuchida, I. (1998). The Basics in Japan: The Three C's. Educational Leadership 55:6, 32-37. Rohlen, T. (1976, Spring) The promise of adulthood in Japanese spiritualism. Daedalus, 125-143. Rohlen, T. (1983). Japan's High Schools. (Berkeley, Ca.: University of California Press). Rohlen, T. & LeTendre, G. (1996). Teaching and learning in Japan. New York: Cambridge University Press. Sato, N. In Rohlen, T. & LeTendre, G. (1996). Teaching and learning in Japan. New York: Cambridge University Press. Sato, N., & McLaughlin, M. W. (1992, January). Context matters: Teaching in Japan and in the United States. Phi Delta Kappan, 359-366. Schmidt, W.H., McKnight, C.C., & Raizen, S.A. (1977). A Splintered Vision: An Investigation of U.S. Science and Mathematics Education (Boston: Kluwer Academic Publishers). Shimahara, N., & Sakai, A. (1995). Learning to Teach in Two Cultures. New York: Garland. Stevenson, H. & Stigler, J. (1992). The Learning Gap. New York: Summit. Stigler, J. & Hiebert, J. (1997). Understanding and Improving Mathematics Instruction: An Overview of the TIMSS Video Study. Phi Delta Kappan, 79:1, 14-21.

Copyright (c) 2000, Child Research Net, All rights reserved.