Teaching dwarfs every other profession that requires a college degree. Nationwide, 3.7 million schoolteachers serve grades K–12—more than all the doctors, lawyers, and engineers in the country combined. Teacher shortages, once chronic, abated during the recession, when layoffs were widespread, but will soon return with a vengeance. Fully half of all teachers are Baby Boomers on the brink of retirement. Among novice teachers, who constitute an increasingly large proportion of the remaining workforce, between 40 and 50 percent typically quit within just five years, citing job dissatisfaction or more-alluring prospects. Given this drain at both ends of the teaching pipeline, schools will likely need to hire more than 3 million new teachers by 2020. That is an enormous talent hole to fill.

Yet the United States has, if anything, too many teacher-training programs. Each year, some 1,400 of them indiscriminately churn out twice as many graduates as schools can use. Program quality varies widely, so many would-be teachers don’t suit schools’ needs. In a scathing 2006 report, Arthur Levine, a former president of Columbia University’s Teachers College, accused many education schools of being little more than a “cash cow” for their hosting institution. Among the problems he highlighted were exceedingly “low admission standards,” a “curriculum in disarray,” and faculties “disconnected” from the realities of the classroom.

Once hired, many teachers are left to sink or swim. In recent years, several states have adopted controversial accountability measures, known as “value added” metrics, with a view toward winnowing out poor performers who haven’t produced student improvement on standardized tests; helping teachers hone their craft has seldom made it onto the agenda. But perhaps we’re finally ready to focus attention on the far bigger and more important question of how to attract and retain the top teachers we want.

This spring, the Obama administration announced plans to begin rating teacher-training programs. Consensus on what makes an effective teacher, however, remains elusive. Student achievement does not correlate strongly with teachers’ years of experience in the classroom (beyond the initial few) or with the caliber of their preparation—whether they have acquired certification, earned a master’s degree in education, or aced state licensing exams. Even particular personality traits, such as an extroverted willingness to ham it up in the classroom, appear irrelevant. The conundrum doesn’t daunt Elizabeth Green, a co-founder of GothamSchools (a news Web site originally devoted to covering New York City schools that has recently expanded to other cities and been rechristened Chalkbeat). Her book, Building a Better Teacher, couldn’t be better timed.

At the heart of Green’s exploration is a powerfully simple idea: that teaching is not some mystical talent but a set of best practices that can be codified and learned through extensive hands-on coaching, self-scrutiny, and collaboration. Yet her account suggests that implementing this vision may entail a bigger transformation than she quite realizes.

Green begins by profiling an array of educators who have been inspired by Deborah Ball, now the dean of the University of Michigan School of Education. In the early 1980s, she was a charismatic math teacher in East Lansing, Michigan, who developed a successful approach to teaching even very young children sophisticated concepts in math. Instead of relying on rote memorization or repetitive skills practice, Ball shepherded children through in-depth discussions of a single mathematical conjecture—for example, do two odd integers always add up to an even number? The students, steered along by their teacher, deliberated together to derive proofs for their various hypotheses. Some of the most exhilarating parts of Green’s book are the detailed descriptions of precisely how, and why, these lessons succeed. Ball helped other teachers adopt her techniques not through the usual education-school lectures, but through rigorous apprenticeship: mutual observation of lessons, followed by intensive dissection of what worked and what didn’t.

Green likens the approach to the Japanese practice of jugyokenkyu. “Lesson study” is the main form of teacher training in Japan, where colleagues routinely sit in on one another’s classes and then scrutinize a single session for hours, extracting general guidance for future instruction. Japan substantially outperforms America in math on international tests, and Green clearly believes jugyokenkyu is a crucial factor in the country’s success. She recounts how some of Ball’s ideas were adopted by the state of California in the mid-1980s but never had a chance to catch on: Teachers were expected to absorb the new policies, outlined in a state “manifesto,” and then revamp lesson plans on their own, with little or no training or ongoing support. Some educators didn’t even see the guidelines—all but ensuring the reforms would fail. The rollout of the Common Core State Standards appears to be replicating this dispiriting pattern in many places.

At first, Green decides that Teach for America and some charter-school leaders are now following in Ball’s and Japan’s footsteps—albeit with plenty of stumbling. She focuses on Doug Lemov, an entrepreneurial-minded educator who started a charter school in Boston in the mid-1990s and later became a managing director and teacher trainer with the Uncommon Schools charter network. As part of his job, he began compiling an inventory of effective teaching techniques. The taxonomy became a book, Teach Like a Champion, and a cause célèbre within the charter movement; videos of sample lessons circulated like samizdat literature. There’s technique No. 2, “Right Is Right”: teachers refuse to accept students’ half-baked responses to questions and insist on well-formulated, and eventually correct, replies. Technique No. 32 is “SLANT,” which stands for “Sit up, Listen, Ask and answer questions, Nod your head, and Track the speaker,” a formula for eliciting attention from students. But the motions of following a lesson, Green soon discovers, aren’t necessarily a sign of genuine engagement.

The taxonomy includes plenty of useful, even commonsense, advice. Yet Green reveals how, in practice, Lemov’s early acolytes in the charter world became obsessed with a disciplinary approach that dictated no talking in hallways, silent lunches, and skyrocketing suspensions for even minor infractions. What at first appeared to be a huge success—Lemov’s school initially posted impressive test scores—turns out to be a more complicated story. Green finds that out of some 55 students who started at the school in seventh grade, only 11 made it to their senior year, an astounding rate of attrition. A later class began with 100 sixth-grade students and was winnowed to 30 by graduation.

Japanese “lesson study,” she observes, was premised on the notion that “children needed structured opportunities to talk in order to learn.” Lemov banked on a rather different principle: that “learning first required the foundational ability to be quiet and listen.” As Green concludes, Lemov had built a vocabulary that Deborah Ball might admire for describing precisely what teachers should do in the classroom, but applied it to “a sort of teaching that she didnt do.” Green ends up saluting Ball and Japan for getting the balance between classroom discipline and student engagement right.

But Green’s account cries out for a look at the bigger picture. She is absolutely correct about the importance of self-critical reflection and collaboration. What she is not the first, or I’m sure the last, to miss are the structural obstacles to importing such an apprentice-style ethos into American teachers’ experience. As it happens, an administrator introduced lesson study as part of the staff’s professional development at a school where I’ve worked. There was just one problem: we teachers—juggling tutoring before and after school, supervising clubs, or coaching sports—had only one period a week to meet as a group. It would be generous to say lesson study didn’t work; it never got off the ground. There typically isn’t time in American teachers’ workdays for this kind of collaborative enterprise.

That lack of time is an American anomaly, and it is key. Since 2000, the Organization for Economic Cooperation and Development has been overseeing tests of 15-year-olds every three years among its members. The PISA exams, as they’re called, show that American students’ performance is barely above average in reading and trails substantially in math. The tests also record other information about classroom instruction around the world, and American researchers, policy makers, and pundits have pored over the results for clues to improving our schools. For example, the United States falls roughly in the middle of the pack when it comes to class size. Countries with far larger classes than we have, such as South Korea, outperform us. So do countries, like Finland, with smaller ones. Not surprisingly, some reformers have concluded that reduced class size isn’t the secret to student success.

But class size is a crude measure of a more important, encompassing concept that is worth attending to: teacher workload. How much time do teachers spend on classroom instruction, and how much time do they have outside of class to devote to the other considerable, less visible aspects of the job: lesson planning, paper grading, conferring with students, calling parents, meeting with colleagues to discuss methods and goals. Here, the PISA results are not ambiguous. Every single country that outperforms us has significantly smaller teacher workloads. Indeed, on the scale of time devoted by teachers to in-class instruction annually, the United States is off the charts. We spend far more hours in the classroom on average, twice and nearly three times more in some cases, than teachers in any other OECD country save Chile. Finnish high-school teachers, for example, clock 553 hours in the classroom each year. In Japan, home of jugyokenkyu, that number is 500. In the U.S., it’s 1,051. (Figures for elementary and middle school show roughly the same skew.)

In practice, this means that most teachers in this country have zero time to work together on new pedagogical approaches and share feedback in the way Green advocates in her book. They rarely have an opportunity to watch other teachers teach, the single best kind of training, in my experience; they’re too busy in their own classrooms (not to mention outside them).

A big problem with American education, in other words, is how we conceive of the job. Green is right: there’s much about teaching that isn’t instinctive, and as her book usefully shows, learning how to perfect the art is demanding. It is high time to correct a common misimpression: teaching isn’t the relatively leisurely occupation many people imagine, enviously invoking a nine-to-three school day and long summer vacations, which in reality seldom exist. We think of no other white-collar profession in terms of a single dimension of job performance. We don’t, for example, regard lawyers as “working” only during the hours they’re actually presenting a case before a judge; we recognize the amount of preparation and subsequent review that goes into such moments. If teaching is such a plum post, we might ask ourselves why attrition rates are so high.

In closing, Green decides to teach a lesson herself and is thrilled to find that it goes well, thanks to so many of the techniques she learned in her reporting—and, it’s worth noting, thanks to plenty of planning. She recounts spending hours getting ready for this one lesson, selecting readings, conferring with a seasoned teacher, and rehearsing how she would present the material to the class. All this, and she wasn’t grading a single paper or speaking to parents or meeting individually with students. Such work constitutes a large portion of what teachers do each day. It’s why the job, done right, is so hard and burns teachers out so fast.

The goal isn’t to lighten teachers’ load but to redistribute it. At one point, Deborah Ball remarks that what she loves about teaching is that it is so hard—by which she means intellectually challenging and rewarding. Teaching is all-consuming, and that absorption is part of the joy of the job. But if teaching is to be a profession of the mind (as well as of the heart) that retains top talent and delivers results on the same level that other countries boast, the people who spend hours with our children in the classroom also need what they currently don’t get: the hours with peers and mentors that are essential to improving their craft.